Methods Kaplan

1
METHODS in CLINICAL
CHEMISTRY
An accessory work to the 5th edition of
Kaplan and Pesces : Clinical Chemistry:

Theory, Analysis, Correlation*
Published by Pesce Kaplan Publishers 2009
Editors
Peter E. Hickman, MB BS, PhD, FRCPA
Methods Editor
Associate Professor
Australian National University Medical School;
Director of Chemical Pathology
The Canberra Hospital
Australian Capital Territory, Australia
Gus Koerbin, BAppSci, AFAIM

Methods Associate Editor
Principal Scientist ACT Pathology
Adjunct Professional Associate
University of Canberra
ACT Pathology
The Canberra Hospital,
Garran, Australian Capital Territory, Australia
A work of 144 Methods of Analysis describing current methodology.
*Published by Mosby, an affiliate of Elsevier; 2009

2
3
METHODS in CLINICAL
CHEMISTRY

Volume I
4
Pesce Kaplan Publishers

Methods Clinical Chemistry
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Permissions may be sought directly
from Pesce Kpalan Publishers: phone: (1) 212 864 4403 or 858 278 4587 (US); e-mail:
http://www.pescekaplan.com/contact_us.htm#Feedback%20Form.
Notice
Knowledge and best practice in this field are constantly changing. As new research and experience broaden
our knowledge, changes in (i) on state-of-the-art methodology and measurement technique, (ii) effect of
interferences, or (iii) interpretation of results may become necessary or appropriate. Readers are advised
to check the most current information in the relevant scientific literature. It is the responsibility of
the clinical laboratory practitioner, relying on their own experience and knowledge of the patient, to
determine the best procedure to take all appropriate safety precautions. To the fullest extent of the law,
neither the Publisher nor the Editors assumes any liability for any injury and/or damage to persons or
property arising out of or related to any use of the material contained in this book.
The Publisher
Previous copyrighted printed editions: 1987, electronic editions 1996, 2003
Library of Congress Cataloging-in-Publication Data

Methods Clinical chemistry [edited by] Lawrence A. Kaplan, Amadeo J. Pesce.
Bound version and electronic version.
Printed in the United States of America

5
Contributors to Methods of Analysis

Zakaria Ahmed, PhD Department of Pathology
Clinical Chemist Johns Hopkins School of Medicine
Department of Pathology and Laboratory Baltimore, Maryland
Medicine
Rochester General Hospital Paul F. Coleman, PhD
Rochester, New York Research Fellow
Infectious Disease Core R&D
Hassan M. E. Azzazy, PhD, DABCC, FACB Abbott Diagnostics
Chairman and Associate Professor Abbott Park, Illinois
Department of Chemistry
The American University in Cairo Joe DAgostino, BSc, Grad Dip FMI,
Cairo, Egypt MAACB
Senior Scientist
Tony Badrick, BAppSc, BSc, BA, MLitSt Clinical Biochemistry Unit
(Math), MBA, PhD, FAIMS, FAACB, FQSA, Alfred Pathology Service
FAIM, FACB, FRCPA (Hon) Alfred Hospital, Melbourne, Australia
Executive ManagerLaboratories
Sullivan Nicolades Pathology Sheila Dawling, PhD, CChem, FRSC
Taringa, Queensland, Australia Associate Professor of Pathology
Director, Toxicology
John Beilby, BSc(Hons), PhD, FAACB, TDM Laboratory,
MHGSA, ARCPA Associate Director
Principal Scientist Clinical Chemistry
PathWest The Vanderbilt Clinic
Nedlands, Western Australia, Australia Nashville, Tennessee
Marion Black, BSc(Hons), Dip. Ed, MAACB Joris R. Delanghe

Senior Scientist Professor of Clinical Chemistry
Clinical Biochemistry, Alfred Pathology Department of Clinical Chemistry
Service Ghent University Hospital
Victoria, Australia Gent, Belgium
John R. Burnett, MB ChB, MD, PhD, Goce Dimeski, BSc

FRCPA, FAHA Supervising Scientist
Clinical Professor Pathology QLD
PathWest Laboratory Medicine Princess Alexandra Hospital
Royal Perth Hospital Woolloongabba, Brisbane, Australia
Perth, Western Australia, Australia
Angela Ferguson, PhD
Kevin Carpenter, PhD, FHGSA Clinical Chemistry Fellow
Principal Scientist and Head of Washington University School of
Department Medicine
NSW Biochemical Genetics Service Department of Pathology and Immunology
The Childrens Hospital at Westmead St. Louis, Missouri
Westmead, New South Wales, Australia
Michael J. Figursk, PhD
Kee Cheung, BSc(Hons), PhD, GradCert. Research Associate
Mgt. University of Pennsylvania
Manager Hospital of University of Pennsylvania
Pathology QueenslandPrincess Alexandra Philadelphia, Pennsylvania
Hospital
Woolloongabba, Queensland, Australia John Galligan
Supervising Scientist
William Clarke, PhD, MBA, DABCC Pathology Queensland Central Laboratory
Director, TDM and Toxicology, Director, Royal Brisbane Hospital
CPOCT Herston, Queensland, Australia
6
Atholl Johnston, BSc, MSc, PhD,

Karen Golemboski, PhD, MT(ASCP) FBPharmacolS, FRCPath
Program Director and Chair Professor of Clinical Pharmacology
Clinical Laboratory Science William Harvey Research Institute
Bellarmine University (School of Medicine and Dentistry)
Louisville, Kentucky Queen Mary
University of London
Ronda F. Greaves, BSc, Grad Dip Ed, London, United Kingdom
MAppSc, PhD, MAACB
Senior Scientist Graham Jones, MBBS, DPhil, FRCPA, FAACB
Complex Biochemistry Department Staff Specialist in Chemical Pathology
The Royal Childrens Hospital Department of Chemical Pathology
RCH Laboratory Services St. Vincents Hospital
Parkville, Victoria, Australia Darlinghurst, New South Wales,
Australia
Kathryn Green, BSc(Hons), MSc(Med)
Senior Scientist Saeed A. Jortani, PhD, DABCC, FACB
NSW Biochemical Genetics Service Associate Professor of Pathology and
The Childrens Hospital at Westmead Laboratory Medicine
Westmead, New South Wales, Australia University of Louisville
School of Medicine
Elizabeth M. Hall, BSc, MSc, FRCPath Louisville, Kentucky
Principal Clinical Scientist
Department of Clinical Biochemistry Lawrence A. Kaplan, PhD, DABCC
East Kent Hospitals University New York, New York
NHS Trust
Kent and Canterbury Hospital Steven C. Kazmierczak, PhD, DABCC
Canterbury, Kent, United Kingdom Professor of Pathology
Director of Clinical Chemistry and
Peter E. Hickman, MB BS, PhD, MPH, Toxicology
MAACB, FRCPA Oregon Health and Science University
Associate Professor Department of Pathology
Australian National University Medical Portland, Oregon
School;
Director of Chemical Pathology Sandra Klingberg, B App Sci
The Canberra Hospital Supervising Scientist, Protein
Australian Capital Territory, Australia Laboratory
Pathology Queensland, Central
Gregory A. Hobbs, PhD, DABCC Laboratory
Clinical Laboratory Science Royal Brisbane Hospital
Bellarmine University Herston, Queensland, Australia
Louisville, Kentucky
David W. Holt, BSc, PhD, DSc(Med), Principal Scientist ACT Pathology
CSci, EurClin Chem, FESC, FRCPath Adjunct Professional Associate
Professor of Bioanalytics University of Canberra
ASI, Ltd ACT Pathology
London, United Kingdom The Canberra Hospital
Garran, Australian Capital Territory,
David G. Hughes, BAppSci, Grad Dip Sci Australia
Scientist
Clinical Chemistry, ACT Pathology Magdalena Korecka, PhD
The Canberra Hospital Senior Research Investigator
Garran, Australian Capital Territory, School of Medicine
Australia University of Pennsylvania
Philadelphia, Pennsylvania
Mind Jin, PhD
Temple University
7
William J. Korzun, PhD, DABCC, MT(ASCP) Brett McWhinney, BSc, MSc, MBA, MPhil
Associate Professor Supervising Scientist
Department of Clinical Laboratory HPLC Section, Department of Chemical
Sciences Pathology
Virginia Commonwealth University Royal Brisbane Hospital
Richmond, Virginia Herston, Queensland, Australia
Edmund Lamb, PhD, FRCPath Danni L. Meany, PhD

Clinical Scientist (Biochemistry) and Clinical Chemistry Fellow
Head of Department Johns Hopkins Medical Institutions
Department of Clinical Biochemistry Baltimore, Maryland
East Kent Hospitals
NHS Trust James J. Miller, PhD, DABCC, FACB
Kent and Canterbury Hospital Professor
Canterbury, Kent, United Kingdom University of Louisville
School of Medicine
Stanley S. Levinson, PhD, DABCC Department of Pathology and Laboratory
Professor of Pathology and Laboratory Medicine
Medicine Louisville, Kentucky
University of Louisville
Director of Clinical Chemistry and Michael Milone, MD, PhD
Immunochemistry Assistant Professor of Pathology and
Department of Veteran Affairs Medical Laboratory Medicine
Center Associate Director, Toxicology
Louisville, Kentucky Laboratory
School of Medicine
Barry Lewis, MD, FRCPA, FHGSA University of Pennsylvania
Head, Department of Clinical Philadelphia, Pennsylvania
Biochemistry
PathWest Laboratory Medicine WA Gerald J. Mizejewski, BS, MS, PhD
Princess Margaret Hospital Senior Research Scientist
Perth, Western Australia, Australia Wadsworth Center
New York State Department of Health
Jinong Li, PhD Albany, New York
Clinical Chemistry Fellow
Johns Hopkins Medical Institutions Scott A. Muerhoff, PhD
Baltimore, Maryland Volwiler Research Fellow
Infectious Diseases Research and
Greg Maine, PhD Development
Manager, Global Scientific Affairs Abbott Diagnostics
Associate Research Fellow Abbott Laboratories
Abbott Laboratories Abbott Park, Illinois
Abbott Park, Illinois
Anthony O. Okorodudu, PhD, MBA
Christopher R. McCudden, PhD, DABCC, Professor
NRCC Director, Clinical Chemistry Division
Assistant Professor UTMB/CMC Outreach Laboratory Services
Department of Pathology and Laboratory Department of Pathology
Medicine University of Texas Medical Branch
School of Medicine Galveston, Texas
University of North Carolina
Chapel Hill, North Carolina
Denise A. McKeown, MSci, AMRSC

Senior Analyst
St Georges, University of London
Analytical Unit
Department of Cardiac and Vascular
Sciences
London, United Kingdom
8
Peter OLeary, BSc, MAACB, AFACHSE, Terry Pry, PhD

ARCPA, PhD Retired-Manager Scientific Affairs,
Adjunct Professor, School of Public Asia-Pacific
Health (Curtin) Abbott Diagnostic Division
Adjunct Associate Professor Abbott Laboratories
School of Womens & Infants Health Auckland, New Zealand
(UWA),
Director, Office of Population Health Kishor Raja, BSc(Hons), MSc, PhD, CSci
Genomics Principal Clinical Scientist/Honorary
Public Health Division Senior Lecturer
Health Department of Western Australia Clinical Biochemistry Department
Perth, Western Australia, Australia Kings College Hospital
Matthew T. Olson, MD
House Officer, Department of Pathology Jordan Reynolds, MD
Johns Hopkins Medical Institutions Resident Physician
Baltimore, Maryland University of Cincinnati
Department of Pathology and Laboratory
Felix O. Omoruyi, PhD Medicine
Fellow Cincinnati, Ohio
Department of Pathology
Clinical Chemistry Ken Robertson, BSc, AAIMS
University of Texas Medical Branch Senior Scientist in Charge (Research)
Galveston, Texas PathWest Laboratory Medicine
Royal Perth Hospital
Mauro Panteghini, MD Wellington St. Perth, Western
Professor of Clinical Biochemistry and Australia, Australia
Clinical Molecular Biology
University of Milan Medical School Andrea M. Rose, PhD, MBA
Laboratorio Analisi Senior Clinical Support Consultant
Milan, Italy Roche Diagnostics
Indianapolis, Indiana
Gerardo Perotta, MPA
Interim-Coordinator, Pathology Enrico Rossi, PhD, MAACB
Education Research Biochemist
Department of Pathology and Laboratory PathWest Laboratory Medicine
Medicine Nedlands, Western Australia, Australia
University of Cincinnati College of
Medicine Randal J. Schneider, MS, PhD
Cincinnati, Ohio Director of Clinical Chemistry and
Toxicology
Michael A. Pesce, PhD ProHealth Care Laboratories
Professor Emeritus of Pathology and Waukesha, Wisconsin
Cell Biology
Columbia University Medical Center Les Shaw, BS, PhD
New York Presbyterian Hospital Professor
New York, New York University of Pennsylvania
School of Medicine
Julia M. Potter, B Med Sc(Hons), MB BS, Philadelphia, Pennsylvania
PhD, FRCPA
Professor of Pathology Run Zhang Shi, PhD
Australian National University Medical Instructor
School Assistant Director of Clinical
Executive Director, ACT Pathology Chemistry and Immunology
The Canberra Hospital Department of Pathology
Australia Stanford University
Stanford, California
9
Ravinder Jit Singh, PhD John G. Toffaletti, PhD

Co-Director, Endocrine Laboratory Professor in Pathology
Organization Clinical Laboratories and Department of
Mayo Clinic Pathology
Rochester, Minnesota Duke University Medical Center
Durham, North Carolina
Patricia Slev, PhD
Assistant Professor of Pathology Susan Vickery, MSc, PhD
(Clinical) Senior Clinical Scientist
University of Utah, East Kent Hospitals University
Medical Director NHS Trust
Serologic Hepatitis and Retrovirus Kent and Canterbury Hospital
Laboratory Canterbury, Kent, United Kingdom
ARUP Laboratories
Salt Lake City, Utah Ping Wang, PhD, DABCC
Medical Director of Clinical Chemistry
Ramasamyiyer Swaminathan, MBBS, MSc, The Methodist Hospital
PhD, FRCPath Houston, Texas
Professor and Head of Department of
Chemical Pathology Gregory Ward, BSc(Hons), MSc, MAACB,
St. Thomas Hospital FAACB
London, United Kingdom Head, Biochemistry and Endocrinology
Sullivan Nicolades Pathology (Sonic
Danyal B. Syed, BSc, MA, PhD, C(ASCP), Healthcare)
CC(NRCC), DABCC, FACB Brisbane, Queensland, Australia
Laboratory Director
William F. Ryan Community Health Center Alan H. B. Wu, PhD
New York, New York Professor, Laboratory Medicine
University of California, San Francisco
Danyel H. Tacker, PhD, FACB Clinical Chemistry and Toxicology
Clinical Chemist Laboratories
Ochsner Medical CenterNew Orleans San Francisco General Hospital
New Orleans, Louisiana San Francisco, California
Jillian R. Tate, BSc(Hons), MSc Odette Youdell, BAppSci(Hons), MAACB

Senior Scientist Senior Scientist
Pathology Queensland Clinical Biochemistry
Chemical Pathology Department Alfred Pathology Service
Royal Brisbane and Womens Hospital Melbourne, Victoria
Brisbane, Queensland, Australia Australia
10
Foreword
Foreword for the 2009 edition of Methods in Clinical Chemistry
In the mid-1980s we perceived a need for an extensive, up-to-date, compilation of methods available for
use in clinical chemistry laboratories. To meet this need, we published in 1987 Methods in Clinical
Chemistry with C.V. Mosby. This volume provided not only a review of extant methodologies, but also a
critique of each method. This enabled the authors, when appropriate, to suggest one technique as a
recommended method. Since its initial publication, Methods in Clinical Chemistry has been repeatedly
updated and made available in electronic form (CD-ROM or Internet) by Pesce Kaplan Publishers.
Like the previous version, this edition is published in parallel with the current edition of our textbook,
Clinical Chemistry: Theory, Analysis and Correlation (5th edition; Elsevier, 2010). The editors of this
work, Peter Hickman and Gus Koerbin, have assembled an international group of expert clinical chemists
from the United State, Europe, and Australia/New Zealand. We have retained the scope of previous
editions, including:
144 revised method reviews of available technologies for the analysis of each analyte,
a critique of each methodology,
analytical quality goals (when available),
recent references,
a suggested procedure for manual methods.
This edition will be available in both electronic (CD-ROM) and printed (two volumes) formats. It is our
hope that this edition will be widely used and vigorously reviewed by its users. Using new software
technology, we will provide a mechanism for input from readers for future versions of this edition. We
have created an Internet site (http://www.pescekaplan.com/) where individuals can publicly post their
comments. Eventually, the Editors will redact the suggestions into changes incorporated into Methods in
Clinical Chemistry.
11
Preface for the 2009 edition of Methods in Clinical Chemistry
It was certainly an honor and privilege to be invited by Larry Kaplan and Amadeo Pesce to undertake the
editorial role for the 2009 edition of Methods in Clinical Chemistry. The challenge for us was to provide a
text that was contemporary, detailed but readable, and of paramount importance, a useful resource and
laboratory tool for both students and professionals as we near the end of the first decade of this
millennium.
Much in clinical chemistry has changed since the first edition was published in 1987: the breadth and
depth of our knowledge base, the development of laboratory technology, and the widespread use of the
internet and digital technology. Review of each chapter and final publication for this edition of Methods
in Clinical Chemistry has been vastly different from the original compilation, with Internet and digital
technology significantly streamlining the process. In 2003, the methods portion of the text was removed
from the 4th edition of Kaplan and Pesces Clinical Chemistry Theory, Analysis, Correlation and provided
as a CD-ROM. This edition includes 131 new and revised methods plus 13 older methods that were not
revised. Each method contains a critique of alternate methodologies; contemporary and historical,
analytical quality goals, and performance data. Methods in Clinical Chemistry is not only offered in a CD-
ROM/DVD format, but also in a two volume printed format. In addition, the methods will be accessible to
purchasers of the 5th edition of Kaplan and Pesces Clinical Chemistry Theory, Analysis, Correlation via
the publishers (Elsevier) Internet site, Evolve.
We have been very fortunate to collaborate with an outstanding team of 76 authors from all over the
world. Our clinical chemistry experts reside in 16 states of the United States, most states of Australia, and
in New Zealand, Egypt, Italy, Belgium and England, making this a truly international project. Without the
contributions of these industry leaders, this volume could not have been completed and to the team we say
a heartfelt thank you. It is somewhat clich to say that it takes a cast of thousands to produce a body of
work such as this, but without these authors and the help of the following individuals, we would not have
succeeded: Professor Julia Potter, Executive Director of ACT Pathology, Canberra, Australia for
supporting our undertaking of this project, Priscilla Delatorre for her patience, along with her formatting
and word processing skills, Nataliya Polyakov of the College of American Pathologists Surveys team for
her assistance in accessing proficiency data (usually at very short notice), Elseviers Ellen Wurm-Cutter
for her valued advice and particularly for her patience and tenacity in keeping us on schedule and to her
editorial assistant, Jennifer Hermes, for her behind-the-scenes contribution. No list of acknowledgements
would be complete without a mention of the respected and appreciated guidance, praise, constructive
criticisms and most of all encouragement that Larry Kaplan and Amadeo Pesce have provided throughout
this process, which commenced in early 2007.
We trust that this edition will be widely used and reviewed by you the reader. We invite you to use the
internet site (http://www.pescekaplan.com/) created to accept your valued input, comments and
suggestions so that future editions of Methods in Clinical Chemistry can respond to the ever changing
needs of the laboratory professional into the second decade of this millennium and beyond.
We intend to continue with this project, progressively updating methods and releasing a new version as
dictated by progress in the field.
12
We also intend to include more international contributors, with the hope that it would become the world
methods reference. Finding appropriate authors from diverse countries has proven to be a challenge and
we ask for recommendations from our colleagues.
.
Peter Hickman
Gus Koerbin
Australia, 2009
13
Methods
1. 25-OH-Vitamin D 24. Barbiturates* Page 205
Ravinder Jit Singh Page 17 25. Bence Jones Protein
2. 1-Antitrypsin Stanley S. Levinson Page 219
John Beilby Page 28 26. Benzodiazepines* Page 232
3. Acetaminophen 27. Beta-hCG (beta-human chorionic
Gus Koerbin, gonadotropin)
David G. Hughes, James J. Miller Page 251
Julia M. Potter Page 33 28. Beta-2-Microglobulin
4. Adrenocorticotropic Hormone (ACTH) James J. Miller Page 257
Hassan M. E. Azzazy Page 43 29. Bilirubin
5. Alanine Aminotransferase R. Swaminathan Page 261
James J. Miller Page 46 30. Blood Gas Analysis and Oxygen
6. Albumin Saturation
Kee Cheung Page 52 Goce Dimeski Page 275
7. Albumin in Urine 31. C-Reactive Protein (CRP)
Graham Jones Page 61 Odette Youdell Page 288
8. Alcohol 32. Calcium
Ping Wang Page 67 Randal J. Schneider Page 296
9. Aldolase* Page 81 33. Cancer Antigen 125 (CA 125)
10. Aldosterone Hassan M. E. Azzazy Page 304
Hassan M. E. Azzazy Page 85 34. Carbamazepine
11. Alkaline PhosphataseTotal Gus Koerbin,
Danyal B. Syed Page 90 Julia M. Potter Page 307
12. Alpha-Fetoprotein 35. Carbohydrate Antigen 15-3 (CA 15-3)
Gerald J. Mizejewski Page 101 Gregory A. Hobbs Page 317
13. Aluminum (Aluminium) 36. Carbohydrate Antigen 19-9 (CA 19-9)
Tony Badrick Page 117 Hassan M. E. Azzazy Page 320
14. Amino Acid Screening 37. Carbon Dioxide and Bicarbonate
Kevin Carpenter Page 124 William J. Korzun Page 323
15. Ammonia 38. Carcinoembryonic Antigens (CEA)
Elizabeth M. Hall Page 128 Gregory A. Hobbs Page 327
16. Amniotic Fluid Phospholipids 39. CatecholaminesPlasma
(AFPL) LS Ratio and PG Brett McWhinney Page 337
Hassan M. E. Azzazy Page 132 40. CatecholaminesUrine
17. Amylase Brett McWhinney Page 344
Ming Jin Page 150 41. Cerebrospinal Fluid (CSF) Protein
18. Angiotensin Converting Enzyme Quantitation
(ACE) Danyel H. Tacker,
Hassan M. E. Azzazy Page 157 Anthony O. Okorodudu Page 363
19. Anion Gap 42. Ceruloplasmin
Tony Badrick Page 162 Ahmed Zakaria Page 372
20. Anticonvulsive Drugs* Page 165 43. Chloride
21. Apolipoproteins A-1 and B William J. Korzun Page 377
Jillian R. Tate Page 181 44. Cholesterol
22. Aspartate Aminotransferase John R. Burnett,
James J. Miller Page 190 Ken Robertson Page 382
23. B-Type Natriuretic Peptide, 45. Cholinesterase
Danyal B. Syed Page 393
NT-proBNP, and ProBNP
Alan H.B. Wu Page 197 46. CMV (Cytomegalovirus)
Terry Pry, Greg Maine Page 405
14
47. Copper The following analytes are in Volume

K. Raja, R. Swaminathan Page 410 II
48. Cortisol
R. Swaminathan Page 420 70. Glucose
49. Creatine Kinase John Beilby Page 651
Mauro Panteghini Page 430 71. Glycated Hemoglobin
50. Creatine Kinase Isoenzymes Andrea M. Rose Page 662
Mauro Panteghini Page 436 72. Haptoglobin
51. Creatinine Joris R. Delanghe Page 670
Edmund J. Lamb Page 440 73. Hepatitis B
52. Cyclosporin (Cyclosporine A) Paul Coleman Page 674
David W. Holt, 74. Hepatitis C Virus
Denise A. McKeown, A. Scott Muerhoff Page 680
Atholl Johnston Page 452
75. High-Density Lipoprotein (HDL)
53. D-Xylose* Page 470 Cholesterol
54. Dehydroepiandrosterone and its John R. Burnett,
sulfate (DHEA and DHEA-S) Ken Robertson Page 686
Gus Koerbin Page 481 76. Holotranscobalamin
55. Digoxin and Digitoxin Marion Black Page 698
Randal J. Schneider Page 489 77. Homocysteine
56. Drug Screen Sheila Dawling Page 703
Christopher R. McCudden Page 495 78. Homovanillic Acid
57. Estradiol Lawrence A. Kaplan Page 718
Greg Ward Page 516 79. Human Immunodeficiency Virus (HIV)
58. Estriol* Page 520 Patricia Slev Page 728
59. Ethylene Glycol 80. Immunoelectrophoresis
Ping Wang Page 535 Stanley S. Levinson Page 740
60. Fecal Electrolytes and Osmolality 81. Immunoglobulin Quantitation
Felix O. Omoruyi, Karen Golemboski Page 753
Anthony O. Okorodudu Page 544 82. Insulin and C-Peptide
61. Fecal Fat and Fat Absorption Steven C. Kazmierczak Page 762
Lawrence A. Kaplan Page 547 83. Ionized Calcium
62. Fecal Occult Blood John G. Toffaletti Page 769
R. Swaminathan Page 559 84. Iron and Iron-Binding Capacity
63. Ferritin Gerardo Perrotta,
Hassan M. E. Azzazy Page 567 Jordan Reynolds Page 781
64. Folic Acid 85. Ketones
Sheila Dawling Page 573 Lawrence A. Kaplan Page 785
65. Follicle-Stimulating Hormone (FSH) 86. Lactate Dehydrogenase and Lactate
Angela Ferguson Page 586 Dehydrogenase Isoenzymes
66. Free Thyroxine and Free Mauro Panteghini Page 793
Triiodothyronine 87. Lactic Acid
Greg Ward Page 589 Steven C. Kazmierczak Page 797
67. Gamma-Glutamyl Transferase (GGT) 88. Lead
Danyal B. Syed Page 609 Gus Koerbin Page 803
68. Gastric Fluid Analysis* Page 618 89. Lipase
69. Gentamicin and Other Ming Jin Page 814
Aminoglycosides 90. Lipoprotein (a)
Danni L. Meany, Gregory A. Hobbs Page 820
William Clarke Page 625
91. Lithium
Danni L. Meany,
15
92. Luteinizing Hormone 115. Pyruvic Acid

Gregory A. Hobbs Page 828 Steven C. Kazmierczak Page 1019
93. Lysozyme* Page 834 116. Renin
94. Magnesium Greg Ward Page 1025
Steven C. Kazmierczak Page 840 117. Rheumatoid Factor
95. Maternal Fetal Screening Terry Pry Page 1030
Peter OLeary, 118. Rubella
Barry Lewis Page 846 Terry Pry, Greg Maine Page 1035
96. MetanephrinesUrine 119. Salicylates
Brett McWhinney Page 856 Gus Koerbin,
97. Methotrexate Julia M. Potter Page 1043
Michael A. Pesce Page 865 120. Serum Protein Electrophoresis
98. Methylmalonic Acid Sandra Klingberg Page 1052
Kevin Carpenter, 121. Sirolimus
Kathryn Green Page 875 Magdalena Korecka,
99. Mycophenolic Acid Michael Milone,
Michal J. Figurski, Leslie M. Shaw Page 1062
Magdalena Korecka, 122. Sodium and Potassium
Leslie M. Shaw Page 880 William J. Korzun Page 1070
100. Myoglobin 123. Steroid Hormone
Alan H.B. Wu Page 892 Receptors* Page 1078
101. Opiates* Page 899 124. Sweat Electrolytes: The Sweat Test
102. Organic Acid Screening Ronda F. Greaves Page 1104
Kevin Carpenter Page 912 125. T3 Uptake
103. Osmolality Run Zhang Shi Page 1115
Goce Dimeski Page 916 126. Tacrolimus
104. Oximetry Michael C. Milone,
Goce Dimeski Page 921 Michal Figurski,
105. Parathyroid Hormone (Parathyrin) Magda Korecka,
(PTH) Leslie M. J. Shaw Page 1120
Susan Vickery, 127. Testosterone
Edmund J. Lamb Page 934 Greg Ward, Gus Koerbin,
106. Phenylalanine* Page 939 Peter E. Hickman Page 1127
107. Phenytoin 128. Theophylline and Caffeine
Gus Koerbin Page 946 Saeed A. Jortani Page 1140
108. Phosphorus and Phosphate 129. Thyroglobulin (Tg)
Steven C. Kazmierczak Page 957 Run Zhang Shi Page 1151
109. Plasma Free Metanephrines 130. Thyroid Autoantibodies
Brett McWhinney Page 964 Run Zhang Shi Page 1158
110. Porphobilinogen Screening and 131. Thyroid-Stimulating Hormone (TSH)
Quantitation Greg Ward Page 1168
Enrico Rossi Page 974 132. Thyroxine (Total)
111. Procainamide and N- Greg Ward Page 1175
Acetylprocainamide 133. Total Serum Protein
Matthew T. Olson, Kee Cheung Page 1182
William Clarke Page 982 134. Transferrin and Carbohydrate-
112. Progesterone Deficient Transferrin
John Galligan Page 991 Sandra Klingberg Page 1191
113. Prolactin 135. Transthyretin (Prealbumin)
Sheila Dawling Page 996 Danyel H. Tacker,
114. Prostate Specific Antigen (PSA) Anthony O. Okorodudu Page 1201
Hassan M. E. Azzazy Page 1013 136. Tricyclic Antidepressants
Jinong Li, William Clarke Page 1206
16
137. Triglycerides 141. Urine Porphyrin Quantitation

John R. Burnett, Enrico Rossi Page 1258
Ken Robertson Page 1213 142. Urine Protein, Total
138. Troponins Susan Vickery,
Jillian R. Tate, Edmund J. Lamb Page 1264
Mauro Panteghini Page 1224 143. Vitamin B12
139. Urea Joe DAgostino Page 1270
Elizabeth M. Hall Page 1246 144. Zinc
140. Uric Acid Tony Badrick Page 1276
Elizabeth M. Hall Page 1252
*Updated in the last edition.
17
25-OH-Vitamin D
25-OH-Vitamin D
Ravinder Jit Singh
Clinical significance: Refer to Chapter 33, Bone Disease, in the 5th edition of Clinical Chemistry: Theory, Analysis,
Correlation.
Chemical class: Steroid
Cholecalciferols Molecular Merck Molecular
formula Index mass
25-Hydroxycholecalciferol C27H44O2 1610 400.62
(25-OH-vitamin D3, calcifediol)
25-Hydroxycholecalciferol C28H46O2 1610 412.62
(25-OH-vitamin D2, ergocalciferol)
and LC-MS/MS methods for 25-OH-D have been

Principles of Analysis and Current Usage reported by several investigators [7-9]. These studies
Vitamin D deficiency is widely recognized, and various report reasonable correlations but with significant
treatment options are being proposed in the literature. differences, the reasons for which are not transparent
Vitamin D (written without the subscript) will refer to both or well understood. Automated or manual competitive
vitamin D2 and vitamin D3. Vitamin D itself is immunoassays are known to have less specificity for
biologically inert and is activated in the body through two low-molecular-weight compounds, and immunoassays
specific hydroxylation reactions (Figures 1 and 2). Most for 25-OH-D are no exception.
evidence indicates that both natural vitamin D3 and
The first useful techniques for measurement of 25-
synthetic vitamin D2 are metabolized by the same enzyme
OH-vitamin D were competitive protein-binding
systems so that ingested vitamin D2 is also converted first (CPB) assays (Table 1, Method 1) [10-13]. These
to 25-OH-vitamin D2 and then to 1,25-(OH)2-vitamin D2. procedures exploit the natural serum vitamin D
From the standpoint of assay development, it is critical that binding protein (DBP), which has a high affinity (5
any vitamin D2 metabolites present in serum be included 108 M-1) for 25-OH-vitamin D3. The protein from
in the total assayed fraction. It should be noted that in many species has similar binding characteristics,
addition to 25-OH-vitamin D3 and 1,25-dihydroxyvitamin widening the available sources of binding reagent. The
D3, other hydroxylated metabolites of vitamin D are serum does not need to be from vitamin Ddeficient
animals because the endogenous vitamin D
known to circulate. These include 24,25-dihydroxyvitamin metabolites occupy only a small number of the
D3, 25,26-dihydroxyvitamin D3 , and 1,24,25-
binding sites.
trihydroxyvitamin D3.
DBP also binds other circulating vitamin D
The use of methods for measuring these metabolites has metabolites, including 24,25-(OH)2-vitamin D and
been restricted to research, and the usefulness of the 25,26-(OH)2-vitamin D. Because of the lack of
measurements is unknown; thus they will not be specificity of the binding protein and the existence of
considered in this discussion. Analysis of 25-OH-vitamin
nonspecific lipid interference in serum, measurement
D in circulation has been reported to be the best marker for of 25-OH-vitamin D concentrations directly in crude
determining vitamin D deficiency [1-6]. serum extracts using CPB assays results in higher
values than those observed when chromatography is
Various methods are available for measuring circulating used before assay. 25-OH-vitamin D can be readily
concentrations of 25-OH-D. Current methods include separated from other metabolites by chromatography
HPLC, RIA with low throughputtohigh throughput
on a minicolumn of silica gel. Thus in the case of 25-
automated chemiluminescence immunoassays, and liquid OH-vitamin D assay, if a rapid index of overall
chromatographytandem mass spectrometry (LC-MS/MS). vitamin D status is desired, the use of the CPB assay
These methods have already aroused controversy [7-9].
with preliminary Sep-Pak chromatography is
Correlation and agreement studies between immunoassays preferable.
i
Vitamin D In addition to CPB assays, methods based on high-
Previous and current authors of this method: performance liquid chromatography (HPLC-UV) and
First edition: Not done detection by ultraviolet absorbance have been used
Methods edition: Thomas L. Clemens successfully to measure 25-OH-vitamin D in serum
Second edition: Not updated (Table 1, Method 2). Because of the limited sensitivity
Third edition: Not updated of conventional flow-through ultraviolet detectors
Fourth edition: Thomas L. Clemens (low-nanogram range), these methods generally
Fifth edition: Ravinder Jit Singh require greater sample volumes than CPB assays do,
but this in turn entails rigorous prepurification
18
25-OH-Vitamin D
procedures to remove the larger amounts of contaminating All the methodologies described above are being used
lipids and proteins. In the earlier versions of HPLC in various clinical laboratories. It is acknowledged that
procedures, serum was equilibrated with tracer quantities there are challenges in all of these methods, and high-
of 3H-25-OH-vitamin D3 to determine the recovery of level technical expertise is required to perform the
analysis in clinical laboratories. The College of
sterol during extraction and chromatography. The 25-OH-
American Pathologists (CAP) and the United
vitamin D was prefractionated either by minicolumn
Kingdombased DEQAS (Vitamin D External Quality
chromatography or by HPLC. Final resolution and
Assessment Scheme) surveys provide independent
quantitation of the sterol is achieved by HPLC on
approaches to monitor the performance of laboratories
microparticulate columns of silica or C18 bonded silica.
that use various methods for testing of 25-OH-D
Detection is by ultraviolet spectroscopy at 254 nm. The (Table 2). The survey feedback does not assess the
concentration of sterol is then determined by relating the accuracy of 25-OH-D measurements by laboratories
peak area of unknown to a calibration curve generated by but scores laboratories for agreement within the group
HPLC of known amounts of authentic 25-OH-vitamin D. using a particular method. Recent CAP data (CAP
The HPLC methods for 25-OH-vitamin D, though precise survey, 2007 Ligands Special) indicate that clinical
and accurate, are more time consuming. When the HPLC laboratories using chemiluminescence immunoassays
method is used, automation of sample injection and can report a result ranging from 41 to 96 g/L for a
column elute collection is desirable. In recent versions of survey sample with a value of 75 g/L determined by
HPLC-UV methods, non-isotopic internal standards are LC-MS/MS (BGS-04 in Figure 8). There could be
used, and above-mentioned limitations have been many reasons for these variations, including drifts in
addressed. the reagents being manufactured, but there is a clear
and urgent need for harmonization and
Although more laborious and expensive, the HPLC assays standardization. Considerable variation is observed in
offer certain advantages over the CPB methods [14-20]. results observed for the same sample, as demonstrated
For example, it is possible, with the appropriate columns, in DEQAS data (Table 2).
to separate the natural form of 25-OH-vitamin D3 from 25-
OH-vitamin D2, a synthetic form, and thus gain a measure NIST is developing quality-control materials (human
of dietary versus endogenously derived vitamin D3. serum, SRM 972) that will contain 25-OH-D2, 25-OH-
D3, and the metabolite 3-epi-25-OH-D at 4 different
Moreover, so that 25-OH-vitamin D is completely resolved concentrations as characterized by LC-MS/MS. SRM
from other lipids, the HPLC methods are less susceptible is especially important for assays for which the cross-
than the CPB methods to nonspecific interference. For reactivity with these metabolites is not well defined
these reasons, the HPLC methods are increasingly [9].
employed to measure 25-OH-vitamin D (Figures 3 and 4).
LC-MS/MS is becoming the technique of choice for
A specific radioimmunoassay for 25-OH-vitamin D has various reference laboratories. Laboratories which use
been developed, and results indicate that this approach in-house LC-MS/MS have responsibility for many
allows direct measurement of 25-OH-vitamin D in serum steps of the assay. The LC-MS/MS technology for
extracts (Table 1, Method 3). Although it has been testing of human samples is not approved by the U.S.
suggested that 25-OH-vitamin D can be reliably measured Food and Drug Administration (FDA), and
in ethanol extracts of serum without preliminary manufacturers of LC-MS/MS instrumentation are not
chromatography, comparisons of assays with and without responsible for troubleshooting the assays.
chromatography have clearly shown that direct assays Laboratories performing 25-OH-D testing by LC-
produce overestimation of the 25-OH-vitamin D MS/MS technology have differences in their standard
concentration (Figure 5) [21, 22]. operating procedures, and thus inter-laboratory CVs
are in the range of 20%. The preparation of the
Implementation of LC-MS/MS (Table 1, Method 4) has reagents required for in-house LC-MS/MS assays is
revolutionized the use of mass spectrometry in clinical conducted by individual laboratories under their
laboratories. In large reference labs in the United States, institutionally regulated standard procedures. The
serum 25-OH vitamin D is now measured routinely using complexity of the LC-MS/MS technology in its
LC-MS/MS, the gold standard (Figures 6 and 7) [23, 24]. present form demands a robust, fully automated
platform that can meet the need for throughput,
Reference and Preferred Methods precision, and accurate testing of vitamin D and
It is critical to have a reference method for 25-OH vitamin metabolites. Multiplexed immunoassays may have the
D against which commercial manufactures and clinical potential of achieving accuracy and precision for
labs can standardize their methods for better patient care. multiple vitamin D metabolites.
In recent years, reference procedures for cholesterol have
been developed and are available as a service for
Specimen
commercial manufacturers and clinical laboratories, and Whole or heparinized blood should be collected and
this has made significant impact on the clinical practice. At the serum or plasma frozen at 80C. 25-OH-vitamin
present, there exist no reference procedures for D is stable when stored frozen at 80C.
measurement of 25-OH vitamin D.
19
25-OH-Vitamin D
Serum 25-OH-Vitamin D Reference Intervals calcium and phosphate by affecting their renal tubular
Recent publications have proposed that a serum 25-OH- reabsorption.
vitamin D concentration < 30 ng/mL be used as a cut-point
to define vitamin D deficiency [1, 5]. Latest research has
Reference Intervals for 25-OH vitamin D
shown that deficiency of vitamin D may be associated with
susceptibility to various diseases, including cancers. For
25-OH Vit D
prevention of these diseases, there are many individual
concentration Clinical state
recommendations in the literature for minimum levels of
<10 ng/mL Severe deficiency*
circulating 25-OH-vitamin D, but no consensus or
10 ng/mL-25 ng/mL Mild to moderate
evidence-based-medicine guidelines have been established
25OHD deficiency**
to help patients and physicians. Since 2004, the Mayo
25 ng/mL-80 ng/mL Optimum 25OHD
Clinic has defined 25-OH-vitamin D deficiency in patients
levels
(using an LC-MS/MS method) based on the criteria below.
80 ng/mL Toxicity possible
We have observed that 8.5% of the U.S. patients (n =
40,000) have < 10 ng/mL in winter, which drops to 4% in
* Could be associated with ostemalacia or rickets
summer. Optimum levels were present in 60% and 73% of
**May be associated with increased risk of
the population in winter and
osteoporosis or secondary
summer, respectively.
hyperparathyroidism
Optimum levels in the normal population
Characteristic seasonal fluctuations are seen for serum 25-
80 ng/mL is the lowest reported level associated
OH-vitamin D concentration. These changes reflect the
with toxicity in patients without primary
amount of sunlight to which a person is exposed.
hyperparathyroidism and with normal
Concentrations of 25-OH-vitamin D are highest in late
renal function
summer and lowest in spring.
Interpretation
Upon entering the bloodstream, vitamin D is bound by a Of the circulating vitamin D metabolites, 25-OH-
specific transport protein, vitamin Dbinding protein vitamin D is the most abundant form and has the
(DBP). In the liver, vitamin D is hydroxylated at the longest half-life (approximately 1 to 2 weeks). Its
carbon-25 position, giving rise to 25-OH-vitamin D3, the concentration serves as the best index of skin synthesis
and dietary intake of vitamin D. Nutritional vitamin D
most abundant circulating form of the vitamin. The final
deficiency, rickets or osteomalacia, is associated with
hydroxylation step is catalyzed in the kidney by a 1--
chronic low 25-OH-vitamin D levels, and subjects
hydroxylase enzyme, resulting in the production of the
who are intoxicated with vitamin D have
biologically active form, 1,25-(OH)2-vitamin D. The
concentrations above 200 ng/mL. Chronic therapy
activity of the renal -hydroxylase is under tight control, with some anticonvulsants decreases 25-OH-vitamin
so the production of 1,25-(OH)2-vitamin D remains D concentrations by induction of hepatic clearance.
constant over a wide range of substrate (25-OH-vitamin D) Abnormally low concentrations of the metabolite
concentrations. The main regulators of 1--hydroxylase (resulting from malabsorption) are also observed in
activity are calcium, parathyroid hormone, and phosphate. patients with inflammatory bowel disease, bowel
Low serum calcium stimulates the secretion of parathyroid resection, or biliary cirrhosis (Table 3) [3].
hormone, which acts to increase the conversion of 25-OH-
vitamin D to 1,25-(OH)2-vitamin D. Hypophosphatemia Performance Goals
25-OH-vitamin D is ordered and interpreted clinically
also stimulates conversion of 25-OH-vitamin D to 1,25-
in light of calcium balance and homeostasis. The
(OH)2-vitamin D, but this process does not require
quality and performance goals for calcium analysis in
parathyroid hormone. the clinical labs have been revolutionized. Calcium
measurement is performed on highly automated
Vitamin D, through its active form 1,25-(OH)2-vitamin D, instruments and has a precision of less than 1%
has its most important effects on the intestine, where it coefficient of variation (CV) for day-to-day
stimulates intestinal calcium and phosphate transport. It is performance. There is no reason to accept inferior
believed that 1,25-(OH)2-vitamin D acts at the intestinal performance of the 25-OH-vitamin D assays in the
brush-border membrane, altering the properties of the cells 21st century. For better patient care, the goal should be
to allow greater permeability to calcium and phosphate not only to have an accurate 25-OH-D value but also
ions. The enhanced absorption of these ions raises their precision for 25-OH-D testing, with a CV < 1%. We
concentration in blood to the levels necessary to permit would propose the following other goals for the 25-
normal skeletal mineralization. OH-vitamin D assay.
Vitamin D also acts directly on bone and kidney. In bone, Desirable interassay precision would be a 1% CV, but
1,25-(OH)2-vitamin D causes bone-mineral resorption by the current technologies used for 25-OH-vitamin D
analysis cannot achieve this. A CV < 10% is
increasing osteoclastic resorption but probably does not
acceptable, but most current assays do not perform to
play a direct role in bone mineralization. In the kidney,
even this level (see Table 2). Current external quality-
1,25-(OH)2-vitamin D decreases the excretion of both
20
25-OH-Vitamin D
assessment programs indicate that accuracy is also a hydroxyvitamin D3 using high-pressure liquid
significant problem. Currently, commercial assays for 25- chromatography. Anal Biochem 1971;86:298.
OH-vitamin D do not perform to a satisfactory standard. 15 Jones G. Assay of vitamin D2 and D3 and 25-
hydroxyvitamins D2 and D3 in human plasma
References by high-performance liquid chromatography.
Clin Chem 1978;24:287-298.
1 Cannell J, Hollis B, Zasloff M, Heaney R. 16 Preece MA, ORiordan JLH, Lawson DEM,
Diagnosis and treatment of vitamin D deficiency. Kodicek E. An assay for 25-
Expert Opin Pharmacother 2008;9:107-118. hydroxycholecalciferol and 25-
2 Wicherts IS, van Schoor NM, Boeke AJP, Visser hydroxyergocalciferol in serum. Clin Chim
M, Deeg DJH, Smit J, Knol DL, Lips P. Vitamin Acta 1971;54:235-243.
D status predicts physical performance and its 17 Neyestani T R, Gharavi A, Kalayi A.
decline in older persons. J Clin Endocrinol Metab Determination of serum 25-hydroxy
2007;92:2058-2065. cholecalciferol using high-performance liquid
3 Fraser DR. The regulation of vitamin D chromatography: a reliable tool for
metabolism. Physiol Rev 1980;60:551-613. assessment of vitamin D status. Int J Vitam
4 Schmidt-Gayk H, Bouillon R, Roth HJ. Nutr Res 2007;77:341-346.
Measurement of vitamin D and its metabolites 18 Takeuchia A, Ishidaa Y, Sekimotoa H,
(calcidiol and calcitriol) and their clinical Masudaa S, Okanoa T, Nishiyamab S,
significance. Scand J Clin Lab Invest Matsudab I, Kobayashia T. Simplified
1997;227:35-45. method for the determination of 25-hydroxy
5 Holick MF. Vitamin D deficiency. Engl J Med and 1 alpha,25-dihydroxy metabolites of
2007;357:266-281. vitamins D2 and D3 in human plasma -
6 Zerwekh JE. Blood biomarkers of vitamin D Application to nutritional studies. J
status. Am J Clin Nutr 2008;87:1087S-1091S. Chromatogr B Biomed Appl 1997;691:313 -
7 Carter GD, Carter R, Jones J, Berry J. How 319.
accurate are assays for 25-hydroxyvitamin D? 19 Lensmeyer GL, Wiebe DA, Binkley N,
Data from the international vitamin D external Dresner MK. HPLC method for 25-
quality assessment scheme. Clin Chem hydroxyvitamin D measurement: comparison
2004;50:2195-2197. with contemporary assays. Clin Chem
8 Binkley N, Krueger D, Cowgill CS, Plum L, Lake 2006;52:1120-1126.
E, Hansen KE et al. Assay variation confounds 20 Lensmeyer G, Wiebe D, Binkley N, Drezner
the diagnosis of hypovitaminosis D: a call for M. The authors of the article cited above
standardization. J Clin Endocrinol Metab respond. Clin Chem 2006;52:2305-2306.
2004;89:3152. 21 Hollis BW. Comparison of commercially
9 Souberbielle JC, Fayol V, Sault C, Lawson-Body available 125I-based RIA methods for the
E, Kahan A, Cormier C. Assay-specific decision determination of circulating 25-hydroxy
limits for two new automated parathyroid vitamin D. Clin Chem 2000;46:1657-1661.
hormone and 25-hydroxyvitamin D assays. Clin 22 Bouillon R. Radiochemical assays for
Chem 2005;51:395-400. vitamin D metabolites: technical possibilities
10 Belsey R, DeLuca HF, Potts JT Jr. Competitive and clinical applications. J Steroid Biochem
protein-binding assay for vitamin D and 25-OH- 1983;19:921-927.
vitamin D. J Clin Endocrinol Metab 1971;33:554- 23 Higashi T, Awada D, Shimada K.
557. Simultaneous determination of 25-
11 Bouillon R, Van Baelen H, DeMoore P. hydroxyvitamin D2 and 25-hydroxyvitamin
Comparative study of the affinity of serum D3 in human plasma by liquid
vitamin D-binding protein. J Steroid Biochem chromatography-tandem mass spectrometry
1980;13:1029-1034. employing derivatization with a Cookson-
12 Haddad JG, Chyu KJ. Competitive protein- type reagent. Biol Pharm Bull 2001;24:738-
binding radioassay for 25-hydroxycholecalciferol. 743.
J Clin Endocrinol Metab 1971;33:992-995. 24 Singh RJ, Taylor RL, Reddy GS, Hollis BW,
13 Haddad JG, Min C, Walgate J, Hahn J. Grebe SK. C-3 epimers can account for a
Competition by 24,25-dihydroxycholecalciferol in significant proportion of total circulating 25-
the competitive protein-binding radioassay of 25- hydroxyvitamin D in infants, complicating
hydroxycholecalciferol. J Clin Endocrinol Metab accurate measurement and interpretation of
1976;43:712-771. vitamin D status. J Clin Endocrinol Metab
14 Eisman JD, Shephard RM, DeLuca HF. 2006;91:3055-3061.
Determination of 25-hydroxyvitamin D2 and 25-
21
25-OH-Vitamin D
Table 1: Methods for 25-OH-Vitamin D Assay

Analyte: 25-OH-vitamin D
Method 1: Competitive protein-binding assay (CPB); employs serum or tissue-binding protein, quantitative
Principle of analysis: Competition between 25-OH-vitamin D in serum sample extract and 3H-labeled analog for
binding to tissue or serum D-binding protein; separation of bound and free labeled ligand by treatment with
dextran-coated charcoal
Comments: Preliminary chromatography essential for accurate measurement at high concentrations of 25-OH-
vitamin D
Method 2: High-performance liquid chromatography (HPLC); chromatographic, quantitative
Principle of analysis: Serum extracts are resolved on silica columns, and the concentration of 25-OH-vitamin D
is calculated by peak area (A254) integration.
Comments: Accuracy depends on efficiency of separation of 25-OH-vitamin D from interfering peaks.
Method 3: RIA and CLIA
Principle of analysis: Competition between 25-OH-vitamin D in serum sample extract and 125I-labeled or
acridinium esterlabeled analog for binding to antibody. In competitive RIA, the signal is monitored by counting
radioactivity, and assay is mostly manual. In comparison, in automated CLIA, the signal is chemiluminescence,
with moderate throughput.
Comments: Expensive; accuracy depends on standardization against gold-standard reference assay, precision
depends on consistent manufacturing and least lot-to-lot variations
Method 4: Liquid chromatographymass spectrometry (LC-MS/MS)
Principle of analysis: Non-isotopically labeled internal standard is used to account for variable recovery. Tandem
mass spec is not only sensitive but also provides specificity and does not require extensive chromatography, as
required in HPLC. Concentration of 25-OH-vitamin D is calculated by ratio of peak area/height of analyte/internal
standard.
Comments: Initially requires expensive capital investment
Table 2: Performance of 25-OH-Vitamin D Methods in Various Laboratories as

Reported to DEQAS
Sample Method n Method Mean SD CV
Automated IDS EIA 13 44.7 10.7 23.9
Chromatographic ligand-binding assay 1 48 0 0
DiaSorin liaison 37 40.7 6.4 15.8
DiaSorin liaison, total 8 41.2 5.7 13.9
DiaSorin RIA 50 44.1 6.4 14.5
HPLC 10 48.5 16.6 34.3
IDS EIA 56 49.3 9.4 19
IDS RIA 31 45.2 4.5 10
LC-MS 20 50.8 9.3 18.2
Unknown 2 57 12.5 22
All Methods 228 45.4 7.5 16.6
n, Number of laboratories using a particular method.

22
25-OH-Vitamin D
Table 3: Vitamin D Metabolite Concentrations in Disorders of Calcium Homeostasis

Condition: Nutritional rickets or osteomalacia Condition: Hypophosphatemic rickets
25-OH-D: Low 25-OH-D: Normal
1,25-(OH)2-D: Low/low-normal 1,25-(OH)2-D: Low/normal
Condition: Vitamin D-dependent rickets Condition: Primary hyperparathyroidism
a. Type I 25-OH-D: Normal
b. Type II 1,25-(OH)2-D: High/normal
25-OH-D: Condition: Pseudohyperparathyroidism
a. Normal 25-OH-D: Normal
b. Normal 1,25-(OH)2-D: Low/normal
1,25-(OH)2-D:
Condition: Sarcoidosis with hypercalcemia
a. Low 25-OH-D: Normal
b. High 1,25-(OH)2-D: High
Condition: Vitamin D intoxication
25-OH-D: High
1,25-(OH)2-D: Low/normal
Figures
Figure 1: Pathways in the metabolism of vitamin D3. Vitamin D3 ingested in diet or made in skin accumulates in liver,
where it is converted to 25-OH-vitamin D3, the major circulating form. Subsequent metabolism of 25-OH-vitamin D3 in
kidney gives rise to 1,25-(OH)2-vitamin D3, the hormonal form.
23
25-OH-Vitamin D
Figure 2: Pathways in metabolism of vitamin D2. Vitamin D2 ingested in diet or made in skin is accumulated in liver
where it is converted to 25-OH-vitamin D3, the major circulating form. Subsequent metabolism of 25-OH-vitamin D2 in
kidney gives rise to 1,25-(OH)2-vitamin D2, the hormonal form.
Figure 3: Representative HPLC-UV chromatograms.
(A), late-eluting peaks; (B), calibrator in extracted serum; (C), sample from patient with low 25(OH)D3 treated with
vitamin D2; (D), sample from patient with high concentrations of 25(OH)D3. Int. Std., internal standard; mAU,
milliabsorbance units.
24
25-OH-Vitamin D
Figure 4: HPLC-UV Chromatograms showing separation of 25(OH)-epi-D3 a recently discovered metabolite of Vit-
D. (A), elution of 25(OH)-epi-D3 (not extracted); (B), a serum extract containing 25(OH)D3 (40 g/L) and 25(OH)-epi-D3
(30 g/L); and (C), a serum extract containing 25(OH)D3 (69 g/L), 25(OH)-epi-D3 (15 g/L), and 25(OH)D2 (12 g/L).
(26)
Figure 5: Comparison of the proposed HPLC method with the Diasorin RIA [21]
25
25-OH-Vitamin D
Figure 6: 25-OH-Vitamin D Metabolites. LC-MS/MS Chromatograms-MRM ion peaks for 25-(OH)-D3 and 25-(OH)-
D2.
Figure 7: Comparison of the proposed HPLC method with LC-MS/MS. [19]

26
25-OH-Vitamin D
Figure 8: CAP survey data for 2007 on commonly used immunoassays for
25-OH-D.
The ranges of results for survey materials BGS-01 through BGS-04 are shown for laboratories using either RIA (solid
lines, n = 16) or automated chemiluminescent immunoassays (broken lines, n = 18). Single lab LC-MS/MS 25-OH-D test
results are shown (closed circles) for comparison. CLIA, chemiluminescence immunoassay.
Procedure: 25-OH-Vitamin D-Commercial Assays.

Follow the latest kit inserts supplied with the reagents
and the instrumentation.
25-OH-Vitamin D -HPLC-UV Method [19]
reservoir contained water. The solvent evaporator was a
Reagents Turbo VapTM LV (Caliper Life Sciences). Temperature
The vitamin D metabolites 25(OH)D3 and 25(OH)D2 can was set at 35C, nitrogen flow was adjusted to 10 psi on
be obtained from Fluka Chemicals. ACS reagent-grade the instrument gauge, and the typical drying time setting
acetonitrile (CH3CN), and ethyl acetate were obtained was 25 min. The HPLC unit was an integrated system
from Fisher. Methanol (HPLC grade) was obtained from with a UV3000 detector set at 275 nm, a P4000 pump set
Mallinckrodt Chemicals. Ultrapure water (18.2 M /cm) at 1.2 mL/min, an AS2000 autosampler, and an
was obtained from a MilliQ water purification system SCM1000 solvent system, all from Thermo Separation
(Millipore). The precipitation reagent contained the Products. A silica-saturator column (250 4.6 mm [i.d.]
internal standard laurophenone (400 g/L) in CH3CN stainless steel column; Alltech) packed with ICN silica
and was stored in an amber bottle. Strata-X (surface- gel (particle size, 63 to 100 m; MP Biochemicals) was
modified styrene-divinylbenzene resin) 60-mg (1 mL) installed in the oven between the pump and injector and
extraction cartridges were from Phenomenex. An is necessary here to prevent deterioration of the
automated extraction instrument, the Gilson ASPEC analytical column. The guard column (12.5 4.6 mm
XL4 (Gilson Instruments), consisted of a 4-syringe [i.d.]) and analytical column (250 4.6 mm [i.d.]), both
pump module and a 4-needle sampler module with four containing 5-m Stable Bond Cyanopropyl (SB-CN),
2-way solvent ports. Areas in the sampler racks were were from Agilent Technologies. All columns were
defined as the sample zone, reagent zone, result zone, operated at 50C. The methanolwater (67:33 by
and a disposable extraction column (DEC) zone. volume) used as mobile phase was filtered and degassed.
Acetonitrile was delivered via solvent ports.
Acetonitrilewater (35:65 by volume) was stored and Individual calibrator stock solutions (40 mg/L) of each
delivered from tubes within the reagent zone. The main metabolite are prepared in ethanol, and the concentration
was verified on a Beckman DU 7500 spectrophotometer,
27
25-OH-Vitamin D
using molar absorptivities at 265 nm (1-cm pathlength) 25-OH-Vitamin D LC-MS/MS Method [24]
of 19,400 and 18,300 for 25(OH)D2 and 25(OH)D3,
respectively. From these primary stocks, dilute, For the standard 25-OH-D method, online extraction and
combined-stock solutions of the compounds at 10,000 HPLC chromatography of the supernatants were
g/L each in ethanol can be prepared, which are stable performed using a TX4 Turbo Flow system (Cohesive
for at least 1 year at 20C. Multiple working calibrators Technologies, Franklin, MA) with 1.0 50 mm Cyclone
in the range of 5 to 200 g/L for each of the 2 vitamin D extraction columns and 3.3 cm 4.6 mm, 3-m LC-18
metabolites were combined in a drug-free serum pool. (Supelco, St. Louis, MO) analytical columns. After
The concentrations of endogenous 25(OH)D3 and online extraction, the analytes were eluted onto the
25(OH)D2 present in the pool were taken into account analytical column for 90 sec with a mobile phase of
when assigning the final concentration to the calibrator. 39.5% vol/vol methanol, 0.005% vol/vol formic acid.
Serum calibrators were stored frozen at 20C in 10-mL There was a step gradient to 87% vol/vol methanol,
glass vials sealed with Teflon-lined caps (QuorpakTM; 0.005% vol/vol formic acid for the analytical column.
Fisher Scientific) and were stable for at least 6 months. The analytes then entered an API 4000 triple-quadrupole
Controls were prepared and used in the same manner. mass spectrometer (ABI-Sciex, Toronto, Canada) and
Commercial lyophilized serum controls were custom- were ionized in an atmospheric-pressure chemical-
prepared for us by Utak Laboratories, Inc. Reconstituted ionization source and detected by multiple reaction
Utak controls and thawed calibrators/controls were monitoring of the following ion pairs: m/z 413.0/395.3
stable for at least 1 month stored at 4C. for 25-OH-D2, m/z 401.4/383.3 for 25-OH-D3, and m/z
407.4/389.5 for 25-OH-D3-d6. The raw signals of 25-
Procedure OH-D2 and 25-OH-D3 in the calibrators, controls, and
To prepare samples, we dispensed 2 mL of precipitation samples were normalized to their respective internal
reagent with internal standard into a 13 100 mm standard 25-OH-D3-d6 signals, and concentrations in the
disposable glass test tube; we then added 1.0 mL of samples and controls were calculated off the normalized
serum (calibrator, control, or patient sample) to the tube six-point calibration curves (0 to 200 ng/mL (0 to 500
without mixing of contents to avoid balling of the nmol/L). Samples with concentrations that exceeded the
protein. The tube was allowed to sit for 5 min at room highest calibrator were diluted and run again. The total
temperature, after which it was vortex-mixed for 10 sec 25-OH-D concentrations of each control and sample
to obtain a flocculent precipitate. After another 5-min were calculated by summing the measured values of 25-
wait, the tube was vortex-mixed and centrifuged at 2000 OH-D2 and 25-OH-D3.
g for 10 min. The clear supernatant was decanted into a
10 75 mm disposable glass test tube, which was then For separation of epimers, the standard LC-18 column
transferred to the sample zone of the ASPEC XL4 and was replaced with a longer 5-dinitrobenzoyl-(R)-
protected from exposure to natural sunlight to prevent phenylglycine column (Chirex-PGLY and DNB 250
degradation of analytes. The extraction conditions are 4.6 mm; Phenomenex, Torrance, CA), and 100 L of the
defined in detail in ref 21. The XL4 processed 4 samples supernatant was injected. The step gradient extends only
simultaneously and unattended in ~ 15 min. The unit up to 67% vol/vol methanol, 0.005% vol/vol formic acid
sequentially conditioned the Strata-X cartridge in the at an analytical column flow rate of 0.9 mL/min. The
DEC zone with 2.0 mL of CH3CN followed by 2.0 mL mass spectrometer settings remained unchanged. The
of 35:65 CH3CNwater; added 1.0 mL of water to each concentrations of 25-OH-D2, 25-OH-D3, and 25-OH-D
extract; transferred 3.5 mL of extract mixture to the were calculated as above. The concentrations of any
DEC; rinsed the DEC with 2.0 mL of 35:65 CH3CN detected C-3 epimers of 25-OH-D2 or 25-OH-D3 were
water; and eluted the Strata-X cartridge in the DEC zone also calculated off the normalized 25-OH-D2 and 25-
with 2.0 mL of CH3CN. The eluate was dried at 35C OH-D3 calibration curves, and the total 3-epi-25-OH-D
under a stream of nitrogen; the dry extract was then concentration is the sum of 3-epi-25-OH-D2 and 3-epi-
reconstituted with 150 L of ethyl acetateCH3CN (5:95 25-OH-D3 concentrations. Intraassay CVs were 3.8%,
by volume) and vortex-mixed for 5 sec. Water (110 L) 2.4%, and 4.7% at 25, 54, and 140 ng/mL, respectively.
was then added to the tube, and the contents were Interassay CVs were 6.4%, 6.8%, and 5.0% at 24, 52,
vortex-mixed for 5 sec. The sample was centrifuged at and 140 ng/mL respectively.
2000 g for 10 min to settle the precipitate. The clear
liquid was transferred to a glass microvial insert
positioned in an amber-colored vial. The sample was
capped and placed in the autosampler unit of the HPLC.
The extract was stable for at least 3 days at room
temperature. The processor software calculated relative
retention time for peak identification and peak-height
ratio for quantification.
28
1-Antitrypsin
1-Antitrypsin
John Beilby
Names: 1-Antitrypsin, 1-protease inhibitor

Clinical significance: 1-Antitrypsin deficiency associated with lung and liver disease. Refer
to Chapter 52, Diseases of Genetic Origin, in the 5th edition of Clinical Chemistry: Theory, Analysis,
Correlation.
Molecular mass: 52,000 D
Chemical class: Glycoprotein
Most common allelic types: PiM, PiS, PiZ
Principles of Analysis and Current Usage 1-antitrypsin antibodies and 1-antitrypsin in the
The clinical importance of the measurement of 1- sample. The amount of light scattering is dependent
antitrypsin serum levels is to detect people with deficient upon the 1-antitrypsin concentration in the sample and
states. Severe deficiency (OMIM #107400) is clinically can be quantified by comparison with calibrators of
under-recognized [1,2] and is associated with a known 1-antitrypsin concentration.
substantially increased risk for the development of
pulmonary emphysema by the 3rd and 4th decades of Serum Electrophoresis
life. It is also associated with the risk of hepatic disease, Occasionally, 1-antitrypsin deficiency can be detected
cutaneous panniculitis, arterial aneurysm, bronchiectasis, by close examination of the 1-globulin band on serum
and renal disease [3]. 1-Antitrypsin is responsible for electrophoresis gels. Since 1-antitrypsin composes
approximately 90% of the trypsin-inhibitory capacity of approximately 90% of the 1-globulin band, a severe
serum [4,5]. It is a member of the serine protease deficiency results in the almost complete absence of 1-
inhibitor or serpin superfamily [6] and is also known as globulins. It should be noted that the densitometer scan
1-protease inhibitor because of its ability to inhibit a might give a normal value for the 1-globulins, even
broad range of protease enzymes, including trypsin, though visual inspection clearly shows an absence of the
chymotrypsin, pancreatic elastase, neutral proteases of band. Other 1-proteins such as 1-lipoprotein may stain
polymorphonuclear leukocytes and macrophages, and a to give a diffuse background, thus accounting for the
number of other enzymes [7]. It is a globular discrepancy between the visual inspection and the scan
glycoprotein that is found in the 1 region of an agarose results.
electrophoresis pattern at pH 8.6. It has a molecular mass
of 52 kDa and consists of a 418 amino acid single Phenotyping
polypeptide chain with four carbohydrate side chains. 1- 1-Antitrypsin deficiency is inherited as an autosomal
Antitrypsin is found in serum and in a number of body co-dominant disorder, with more than 100 alleles
fluids such as saliva, tears, lymph, semen, cervical identified [8]. The 12.2 kb gene is located on the long
mucus, synovial fluid, and human milk [7]. The half-life arm of chromosome 14 (14q31-32.3) and consists of
of exogenous 1-antitrypsin in serum is approximately a seven exons and six introns. Different phenotypes are
week, with catabolism taking place in the liver. classified by a coding system in which the inherited
alleles are usually letters that denote the migration of the
Quantitation molecule in an isoelectric pH gradient from A (for
1-Antitrypsin concentrations in serum are measured by anodal variants) to Z (for slower migrating variants). The
immunochemical techniques. In the past, electro- MM phenotype indicates individuals who are
immunodiffusion and radial immunodiffusion techniques homozygous for the normal allele, and ZZ indicates that
were used. Today, automated immunoturbidimetric and they are homozygous for the Z allele. The Z variant
immunonephelometric methods are widely used in consists of a glutamine substitution for a lysine residue
laboratories. These quantitative methods determine 1- at codon 342, causing an 1-antitrypsin-deficient state
antitrypsin concentrations in human serum or plasma by and a dysfunctional protein [9]. The S variant is due to a
measuring the absorbance change due to light scattering substitution of valine for glutamine at codon 264,
caused by insoluble complexes that form between anti- causing intracellular degradation of the protein [10].
Low serum concentrations of 1-antitrypsin must be
i confirmed by Pi (Protease Inhibitor) phenotyping [11].
Alpha1-antitrypsin
Previous and current authors of this method:
First edition: Not done The gold standard for the identification of 1-antitrypsin
Methods edition: Gayle Jackson variants is the phenotyping of serum samples by
Second edition: Not updated isoelectric focusing on thin-layer gels in a pH gradient
Third edition: Steven C. Kazmierczak [11]. This can be done on polyacrylamide or agarose
Fourth edition: Gayle Jackson gels. In this procedure, a stable, stationary pH gradient is
Fifth edition: John Beilby obtained by the use of ampholytes which are polyamino-
29
1-Antitrypsin
polycarboxylic acids. Since 1-antitrypsin proteins have Reference and Preferred Methods
isoelectric points in the pH region from 4.2 to 4.65, There is no reference method for the determination of
ampholines with a pH range of 4 to 5 are used. This serum 1-antitrypsin levels. Automated
technique separates the various isoforms of 1- immunoturbidimetric and immunonephelometric
antitrypsin based on their migration in a pH gradient. methods are currently used with the
Each isoform migrates to the position within the pH immunoturbidimetric group having the best imprecision
gradient at which the overall net charge of the protein is (see performance goals below). Reference material for
zero. The separated protein bands are visualized by a serum proteins, CRM 470 was released in 1993 and has
protein stain such as Coomassie blue R-250 [12]. Acid- resulted in an improvement in between-laboratory
starch gel electrophoresis followed by crossed antigen- variability of serum protein results [19].
antibody electrophoresis [13] has been used in the past
(Table 1, Method 6). Specimen
Isoelectric focusing is a technically challenging test and Serum samples are preferred, but plasma samples are
requires a highly skilled scientist to perform the test and acceptable for 1-antitrypsin concentration
interpret the results. Samples of known Pi types are not determination. Samples may be stored at 4C for up to 7
readily available and can only be found by testing many days prior to testing. For long-term storage or for
samples. In addition, severe deficiency is relatively samples suspected of a severe deficiency of 1-
uncommon. Screening for heterozygous 1-antitrypsin antitrypsin, storage at 70C is recommended.
deficiency or for intermediate deficiencies does not seem
worthwhile, because these phenotypes do not correlate Serum or plasma samples may be used for phenotyping.
statistically to respiratory problems [14]. Samples with 1-antitrypsin concentrations of 50% of
normal or above may be stored on a long-term basis at
The medically important variants are those associated 70C. Storage at 20C is acceptable for 1 or 2 weeks.
with 1-antitrypsin deficiencynamely, the S and Z Unfortunately, samples with severe 1-antitrypsin
variants and the rare null (non-production) variant. deficiency cannot prevent normal activation of intrinsic
Approximately 6% of people of northern European proteolytic enzymes and may become degraded rapidly,
descent carry the S gene, and 3% to 4% carry the Z even if stored at less than 20C. Bacterial
variant [15]. contamination or improper storage results in altered
migration rates of the bands with eventual loss of
Genotyping banding [20]. Samples suspected of a severe deficiency
Genotyping can largely replace the isoelectric focusing should be phenotyped as soon as possible after
technique. The molecular assay is readily automated collection.
using real time PCR, and the interpretation is
straightforward, unlike the isoelectric focusing method. Interferences
The PCR assay detects the most common disease- Some work has been published on assay interferences
associated alleles, Z and S. An algorithm for for 1-antitrypsin concentration determination. However,
quantitation, genotyping, and phenotyping of 1- the results appear to be variable, depending on the
antitrypsin has been described [16]. In this approach, 1- technique used and the manufacturer of the
antitrypsin was quantitated and genotyped when the instrumentation [21]. To identify the interferences
serum concentrations were less than 70 mg/dL (0.7g/L). associated with each method, the user should refer to the
If the results were concordant, they were released with manufacturers kit insert.
an interpretative comment. If the quantitation and
genotype results were discrepant, isoelectric focusing 1-Antitrypsin Reference Interval
was performed, and all results were reported with an In 1995, fifteen professional societies and twelve
interpretative comment. Snyder et al. [16] demonstrated diagnostic companies agreed on a procedure for uniform
that 2% of all results were discordant. In another study, reporting of protein reference intervals [19]. The 1-
all samples without an observed S or Z variant and a antitrypsin reference interval was reported as 90 to 200
serum concentration of less than 100 mg/dL (1.0g/L) mg/dL (0.9 to 2.0g/L) for subjects with the MM
were phenotyped [17]. Using this technique, Bornhorst genotype. Age- and gender-related differences in
et al. [18] identified 6.5% of patients at risk of reference interval have been reported. Following birth,
developing 1-antitrypsin-related symptoms owing to concentrations fall during the first 6 months but rise to
the presence of rare deficiency alleles not detected by the adult concentrations by 1 year of age [22]. Females have
genotyping assay. A variation on this approach was higher concentrations than men across all ages [23].
genotyping by direct sequencing all patient samples with
1-antitrypsinconcentrations less than 100 mg/dL Interpretation
(1.0g/L) [18]. This approach identified all deficient A number of factors have been reported to increase 1-
alleles and two new null alleles. Genotyping is quick and antitrypsin concentration, including inflammatory
easy to perform but needs to be complemented by either disorders, malignancy, trauma, increases in estrogen
phenotyping or direct sequencing so that disease concentrations with puberty, pregnancy, or the use of the
associated variants are not missed. oral contraceptive pill [23]. However, values rarely
increase more than fourfold. These factors can cause
considerable overlap in measured concentration between
30
1-Antitrypsin
mildly and moderately 1-antitrypsin-deficient subjects immunoturbidimetric technique is within the analytical
and concentrations in normal subjects [7]. imprecision specification if biological variability defines
required performance.
Severe 1-antitrypsin deficiency is a genetic disorder
that affects about 1 in 2000 to 5000 individuals [1]. The References
World Health Organization guidelines recommend
screening all patients with chronic obstructive 1. Stoller JK, Aboussouan LS. Alpha1-antitrypsin
pulmonary disease and adults and children with asthma deficiency. Lancet 2005; 365: 2225-2236.
[24]. Any 1-antitrypsin concentration less than 100 2. Laurell CB, Eriksson, S. The electrophoretic
mg/dL (1.0g/L) should be genotyped or phenotyped to alpha 1-globulin pattern of serum in alpha 1
detect 1-antitrypsin variants. In the few cases where the antitrypsin deficiency. Scand J Clin Lab Invest
genotype does not agree with the 1-antitrypsin 1963; 25: 132-140.
concentration, the sample should be further investigated 3. Mulgrew AT, Taggart CC, McElvaney NG.
by phenotyping or DNA sequencing. Serum Alpha-1-antitrypsin deficiency: current
concentrations of 1-antitrypsin less than a protective concepts. Lung 2007; 185: 191-201.
threshold of 50mg/dL (0.5 g/L) by nephelometry are 4. Crystal RG. The alpha 1-antitrypsin gene and
associated with an increasing risk of emphysema [25]. its deficiency states. Trends Genet 1989; 5:
411-417.
5. Lisowska-Myjak B. AAT as a diagnostic tool.
Allelic types that result in deficiencies of serum 1-
Clin Chim Acta 2005; 352: 1-13.
antitrypsin include PiS (mean plasma concentration,
6. Potempa J, Korzus E, Travis J. The serpin
60% to 70% of normal), PiZ (mean plasma
superfamily of proteinase inhibitors: structure,
concentration, 10% to 15% of normal), and Pi null [26].
function, and regulation. J Biol Chem 1994;
The ANT concentration in severely deficient subjects
269: 15957-15960.
rises only slightly in response to inflammatory disorders,
7. Morse JO. alpha1-antitrypsin deficiency (first
malignancy, trauma, and increases in estrogen
of two parts). N Engl J Med 1978; 299: 1045-
concentration.
1048.
8. DeMeo DL, Silverman EK. Alpha1-antitrypsin
The prevalence of the various genotypes are
deficiency. 2: genetic aspects of alpha(1)-
approximately MM 91%, MS 6.1%, MZ 2.7%, SS 0.1%,
antitrypsin deficiency: phenotypes and genetic
SZ 0.1%, ZZ 0.02% and null-null. The risk of
modifiers of emphysema risk. Thorax 2004; 59:
emphysema is 20% to 50% for the SS genotype, 80% to
259-264.
100% for SZ, and 100% for the null-null genotype [1].
9. Yoshida A, Lieberman J, Gaidulis L, Ewing C.
The null alleles do not produce 1-antitrypsin transcript,
Molecular abnormality of human alpha1-
produce truncated protein, or produce unstable proteins
antitrypsin variant (Pi-ZZ) associated with
that are essentially completely degraded before
plasma activity deficiency. Proc Natl Acad Sci
secretion. The risk of emphysema is greatly increased in
USA 1976; 73: 1324-1328.
a person with a severe 1-antitrypsin deficiency if they 10. Owen MC, Carrell RW, Brennan SO. The
smoke cigarettes. abnormality of the S variant of human alpha1-
antitrypsin. Biochim Biophys Acta 1976; 453:
1-Antitrypsin Performance Goals 257-261.
Survey data from the 2007 College of American 11. Stoller JK, Snider GL, Brantly ML, Fallat RJ,
Pathologists Participant Summary Report show that 73% Stockley RA, Turino GM et al. [American
of laboratories use immunonephelometric techniques, Thoracic Society/European Respiratory Society
and 27% use immunoturbidimetric techniques. The Statement: Standards for the diagnosis and
imprecision of the methods as determined by the management of individuals with alpha1-
coefficient of variation (CV%) at a mean 1-antitrypsin antitrypsin deficiency]. Pneumologie 2005; 59:
concentration of 53.9 mg/dL (0.54g/L) ranged from 36-68.
4.3% to 6.7% for immunonephelometric techniques and 12. Kueppers F. Determination of alpha1-
from 2.6% to 5.8% for immunoturbidimetric techniques. antitrypsin phenotypes by isoelectric focusing
in polyacrylamide gels. J Lab Clin Med 1976;
Acceptable performance criteria (CLIA-88) for 88: 151-155.
measurement of 1-antitrypsin require that laboratories 13. Lieberman J, Gaidulis L, Garoutte B, Mittman
be accurate to within 3SD of the peer group mean. The C. Identification and characteristics of the
intra-individual variation of 1-antitrypsin in blood in common alpha 1 -antitrypsin phenotypes. Chest
healthy adults has been determined to be approximately 1972; 62: 557-564.
5.9%. Desirable specifications for analytical imprecision 14. Morse JO, Lebowitz MD, Knudson RJ,
derived from studies of biological variation indicate an Burrows B. Relation of protease inhibitor
assay imprecision of no greater than 3.0% and a total phenotypes to obstructive lung diseases in a
error of no greater than 9.2% [27]. The community. N Engl J Med 1977; 296: 1190-
immunonephelometric and immunoturbidimetric 1194.
techniques are both within the desired performance 15. de Serres FJ. Worldwide racial and ethnic
criteria as defined by CLIA-88. However, only one distribution of alpha1-antitrypsin deficiency:
31
1-Antitrypsin
summary of an analysis of published genetic 1-Antitrypsin: Methods Summary Table

epidemiologic surveys. Chest 2002; 122: 1818- Method 1: Electrophoresis; quantitative; estimation of
1829. 1-globulins by physical separation
16. Snyder MR, Katzmann JA, Butz ML, Wiley C, Principle of analysis: Proteins separate based
Yang P, Dawson DB et al. Diagnosis of alpha1- on class (albumin, gamma globulins)
antitrypsin deficiency: An algorithm of Comments: Serum; good screen for severe 1-
quantification, genotyping, and phenotyping.
antitrypsin deficiency. The 1-globulin band is
Clin Chem 2006; 52: 2236-2242.
absent on the electrophoretogram.
17. Bornhorst JA, Procter M, Meadows C,
Method 2: Electroimmunodiffusion (EID); quantitative;
Ashwood ER, Mao R. Evaluation of an
quantitation by size of immunoprecipitate in gel
integrative diagnostic algorithm for the
Principle of analysis: Electrophoresis of 1-
identification of people at risk for alpha1-
antitrypsin into antibody-containing agarose
antitrypsin deficiency. Am J Clin Pathol 2007;
gel; height of gel pattern (rocket) proportional
128: 482-490.
18. Prins J, van der Meijden BB, Kraaijenhagen RJ, to 1-antitrypsin concentration
Wielders JP. Inherited Chronic Obstructive Comments: Serum; kits not available
Pulmonary Disease: New Selective-Sequencing commercially; mainly of historical interest
Workup for 1-Antitrypsin Deficiency Method 3: Radial immunodiffusion; quantitative;
Identifies two Previously Unidentified Null quantitation by immunoprecipitate in gel
Alleles. Clin Chem 2008; 54:101-107. Principle of analysis: 1-Antitrypsin diffuses
19. Dati F, Johnson AM, Whicher JT. The existing into gel containing antibody, forming a ring-
interim consensus reference ranges and the shaped immunoprecipitate; diameter of ring
future approach. Clin Chem Lab Med 2001; 39: proportional to concentration
1134-1136. Comments: Serum; commercial plates readily
20. Ritchie RF, Smith R. Immunofixation. II. available, not widely used in diagnostic
Application to typing of alpha1-antitrypsin at laboratories
acid pH. Clin Chem 1976; 22: 1735-1737. Method 4: Nephelometry or immunoturbidimetry;
21. Bossuyt X, Blanckaert N. Evaluation of quantitative; quantitation by immunoprecipitate
interferences in rate and fixed-time in solution
nephelometric assays of specific serum Principle of analysis: Reaction of 1-
proteins. Clin Chem 1999; 45: 62-67. antitrypsin with its specific antibody results in
22. Ward AM, White PA, Wild G. Reference immunoprecipitate, which has light-scattering
ranges for serum alpha 1-antitrypsin. Arch Dis properties; amount of light scatter proportional
Child 1985; 60: 261-262. to 1- antitrypsin concentration
23. Ritchie RF, Palomaki GE, Neveux LM, Comments: Serum; commercial kits readily
Navolotskaia O, Ledue TB, Craig WY. available
Reference distributions for the positive acute Method 5: Electrophoresis of serum on acid starch gel,
phase serum proteins, alpha1-acid glycoprotein followed by antigen-antibody crossed
(orosomucoid), alpha1-antitrypsin, and electrophoresis; phenotyping; physical
haptoglobin: a practical, simple, and clinically separation of 1-antitrypsin variants followed
relevant approach in a large cohort. J Clin Lab by immunoprecipitin reaction in agarose
Anal 2000; 14: 284-292. Principle of analysis: Two-step procedure :
24. Alpha 1-antitrypsin deficiency: memorandum 1. Separation of 1-antitrypsin
from a WHO meeting. Bull World Health variants on acid-starch gel by electrophoresis
Organ 1997; 75: 397-415. 2. A second electrophoresis causes the
25. American Thoracic Society/European separated variants to migrate perpendicularly to
Respiratory Society Statement: Standards for the first separation into agarose containing
the Diagnosis and Management of Individuals antibody to 1-antitrypsin; precipitin peaks
with Alpha-1 Antitrypsin Deficiency. Am J form; the second electrophoresis is necessary
Respir Crit Care Med 2003; 168: 818-900. for separation of the Z and S variants
26. Eriksson S, Elzouki AN. Alpha 1-antitrypsin Comments: Serum; original, standard method
deficiency. Baillieres Clin Gastroenterol 1998; for phenotyping; time consuming (1 days) and
12: 257-273. technically difficult; mainly of historical
27. Ricos C, Alvarez V, Cava F, Garcia-Lario JV, interest
Hernandez A, Jimenez CV et al. Current
databases on biological variation: pros, cons
and progress. Scand J Clin Lab Invest 1999; 59:
491-500.
32
1-Antitrypsin
1-Antitrypsin: Methods Summary Table (cont)

Method 6: Isoelectric focusing; separation of 1-
antitrypsin variants on agarose or
polyacrylamide gel based on surface-property
charge
Principle of analysis: 1-antitrypsin variants
are electrophoresed on polyacrylamide or
agarose gels containing ampholines with a pH
range of 4 to 5. The variants migrate under the
electric field until they reach a pH equal to their
isoelectric point.
Method 7: Genotyping for the separation of 1-
antitrypsin variants
Principle of analysis: DNA extracted from
whole blood is subject to real-time PCR
amplifying the S and Z alleles in separate tubes.
The different primer design and detection
probes have been described in several published
articles [16, 17].
33
Acetaminophen
Acetaminophen
Gus Koerbin, David G Hughes and Julia M. Potter
Name: Acetaminophen, N-acetyl-p-aminophenol, paracetamol

Clinical significance: Refer to Chapter 55, Toxicology, in the 5th edition of Clinical Chemistry:
Theory, Analysis, Correlation.
Molecular formula: C8H9NO2

Molecular weight: 151.16 D
Merck Index: 39
Chemical class: p-Aminophenol derivative
Structure:
Principles of Analysis and Current Usage acetaminophen with o-cresol in ammonium hydroxide
Acetaminophen did not gain much attention for therapeutic produces an indigo product (indophenol blue) which
use until 1948 when Brodie and Axelrod [1] identified it as indicates the presence of acetaminophen.
the active metabolite of phenacetin and acetanilide. It is
now widely used as an analgesic and antipyretic drug. It is Acetaminophen can be detected by silica gel thin-layer
freely available as an over-the-counter preparation in drug chromatography (TLC) as a quenching spot under long
stores and supermarkets and is frequently used in suicide ultraviolet radiation. In the TLC screen,
attempts. acetaminophen has an Rf of 0.41.
The current routine methods for clinical analysis of
acetaminophen in the United States, as shown in 2007 EMIT methodology has been adapted [3] to produce a
CAP data, are those for automated platforms. These are rapid screening test of serum acetaminophen. In this
either enzymatic methods with color development or procedure, enzyme-labeled glucose-6-phosphate
immunoassays. Previous methods, which are summarized dehydrogenase (G6PDH) acetaminophen and NAD are
briefly below, including those based on the original diluted with Tris buffer. Serum samples are also
method of Brodie and Axelrod [1] (Table 1, Method 1), are diluted with this same buffer. The diluted serum
now superseded for routine clinical analysis and are of samples are mixed with equal amounts of diluted,
historical interest only. More detail is available in the labeled acetaminophen/NAD solution and incubated at
previous edition of this text. This includes high- room temperature for 10 to 15 min. Following
performance liquid chromatography (HPLC), gas incubation, 1.0 L of the reaction mixture is placed
chromatography (GC), and spectroscopic analysis, which onto filter paper, dried, and any observed fluorescence
still have roles in industry and have been reviewed recently indicates the presence of acetaminophen.
[2].
A qualitative point-of-care test for urine has been
Qualitative Analysis incorporated into the Triage Tox Drug Screen panel
A spot test for the detection of acetaminophen (Biosite, Ltd., UK), using the competitive
(paracetamol) in urine is available (Table 1, Method 2). immunoassay ASCEND Multi-immunoassay
The reaction of the conjugated metabolites of technology. The sample comes in contact with
lyophilized reagents, that then is applied to a solid
i
Acetaminophen phase containing immobilized antibodies. The method
Previous authors of this method: incorporates preset threshold concentrations for each
First edition: Joseph Svirbely drug. In the absence of drug or in the presence of drug
Methods edition: Joseph Svirbely in quantities less than the threshold concentration, no
Second edition: Joseph Svirbely red-colored bar is produced. Samples containing the
Third edition: Helen M. Dodds, Julia M. Potter drug at or above the threshold concentration cause the
Fourth edition: Helen M. Dodds, Julia M. Potter appearance of that bar. The quoted detection
Fifth edition: Gus Koerbin, David G. Hughes, Julia M. concentration is 5mg/L. The urine samples require no
Potter pretreatment.
34
Acetaminophen
Quantitative Analysis determination of acetaminophen in serum (Table 1,

Colorimetric Methods Method 5b). Serum samples were diluted in phosphate
The earliest methods were colorimetric techniques based buffer and after 2 min, 5 mL ethyl acetate was added,
on the hydrolysis of acetaminophen to p-aminophenol with followed by mixing and centrifugation. The organic
addition of -naphthol for color development, which was phase was evaporated under nitrogen, reconstituted in
then measured spectrophotometrically. Applications for methanol, and the sample scanned immediately
either serum or urine were developed [4-6]. However, between 300 nm and 216 nm.
although the p-aminophenol methods were rapid and
simple, they measured the sulfate and glucosamine High-Performance Liquid Chromatography
metabolites of acetaminophen, as well as parent A variety of HPLC assays have been developed for the
acetaminophen. The metabolites are clinically unimportant, analysis of acetaminophen, such as normal phase, ion-
so these methods were not of great clinical use. exchange, and reversed-phase techniques.
An alternative colorimetric technique utilized ring nitration Most HPLC methods assaying acetaminophen use
of the acetaminophen with nitrous acid [7]. This was reversed-phase liquid chromatography on a C18
initially used in quantification in the pharmaceutical column with an extraction procedure prior to injection.
industry and with a modification is known as the Glynn A direct-injection technique was described by Manno
and Kendall method [8] (Table 1, Method 3a). This method [20] in 1981. Equal volumes of sample and internal
is subject to significant interference from salicylic acid, standard (beta-hydroxyethyltheophylline) were mixed
and many method modifications have attempted to reduce and injected directly onto a Bondapak C18 column.
this, ranging from mathematical corrections for known
The mobile phase of sodium acetate and acetonitrile
salicylate concentration [9] to changes in the reaction
eluted acetaminophen at a flow rate of 2 mL/min with
conditions [10-14].
detection at 254 nm (0.01 AUFS). Both sample and
internal standard eluted within 10 min. Carryover
These p-aminophenol conversion methods are rapid and
using this injection technique was not reported.
offer a degree of simplicity. However, by using harsh
chemical treatment and heat with no extraction step, the
In 1982, Kinberger [21] published a method
method will measure sulfate and glucuronide conjugates of
describing the simultaneous determination of serum
acetaminophen, thus rendering the assay nonspecific.
acetaminophen, theophylline, and salicylate. The
sample and internal standard (8-chlorotheophylline)
Longlands and Weiner [14], in 1982, reported similar
were deproteinated with trichloroacetic acid and
modifications (Table 1, Method 3b). The addition of
extracted with methylene chloride and isopropanol.
hydrochloric acid was omitted and trichloroacetic acid
The aqueous phase was discarded, and the supernatant
substituted. A reduction in nitrite concentration was used,
was evaporated under air. The supernatant was
with a nitration reaction time of 2 min, and the absorbance
reconstituted with mobile phase (methanol/acetate
read at 450 nm.
buffer), centrifuged, and an aliquot injected for
analysis. The samples were analyzed using a C18
Spectrophotometric Method
Most spectrophotometric methods require an extraction reversed-phase column with a flow rate of 1.0 mL/min
procedure followed by a chemical reaction to convert monitored at 280 nm.
acetaminophen into a light-absorbing compound.
The methods developed by Starkey [22] and Sood [23]
Liu and Oka [15] utilized ferric reduction following used 10-octadecylsilane-coated silica columns,
extraction of acetaminophen, (Table 1, Method 4), but protected with guard columns of the same material.
although such methods are reliable and low cost, there is Both assays involved a direct-injection technique.
potential interference from phenolic hydroxyl compounds. Samples were mixed with internal standard, vortexed,
More complex processes such as differential extinction, as and centrifuged. In Starkeys method, proteins are not
described by Knepil [16] and modified by Routh [17], are precipitated with ZnSO4 upon addition of the internal
based on measuring the absorbance of an alkaline solution standard (benzoic acid). The mobile phase used was
read against the acidic solution of identical concentration sodium acetate and methanol, with the column eluate
of the sample at 290 nm (Table 1, Method 5a). This monitored at 249 nm. Soods method utilized 3-
wavelength corresponds to the isosbestic point of a acetaminophen as the internal standard, with a mobile
salicylate and allows quantitation of acetaminophen phase of potassium phosphate monobasic and
without interference from salicylates. methanol at a flow rate of 2.0 mL/min and measured at
254 nm. In both assays, acetaminophen and internal
Derivative spectroscopy is a technique where the rate of standard were eluted in less than 10 min.
change of absorbance with wavelength is measured. First-
derivative spectroscopy denotes this measurement in units An HPLC method suitable for routine use in clinical
of absorbance per nanometer of wavelength versus laboratories was described by Kamali and Herd [24] in
wavelength. Second-derivative spectroscopy is the 1990. The assay is based on ion-pair extraction that
derivative of the first-derivative spectrum in units of eliminates the acetaminophen conjugates, glucuronide
absorbance per nanometer squared [18]. Dingeon [19] and sulfate, preventing interference with
described a second-derivative spectroscopy method for the acetaminophen. The isocratic mobile phase consisted
35
Acetaminophen
of methanol/1% acetic acid, containing pyridine and derivatization to the trimethylsilyl

tetrabutylammonium dihydrogen phosphate (TBA) and derivative, the extract was injected onto an OV-17
potassium sulfate at a flow rate of 1.2 mL/min. Internal column (80 to 100 mesh) at 200C.
standard (-hydroxyethyl theophylline) was added to the
plasma or urine samples and treated with TBA and Thomas and Coldwell [29] modified Prescotts
ammonium sulfate. The mixture was vortexed and method by using a more powerful silylating reagent
acetaminophen extracted with equal volumes of (Regisil, bis[trimethylsilyl]trifluoroacetamide) and a
chloroform and isopropanol. Following centrifugation, the more selective extractant (diethyl ether). They also
organic layer was transferred and evaporated to dryness substituted an OV-1 column for Prescotts OV-17
under nitrogen. The sample was reconstituted with mobile column.
phase and injected for analysis at 254 nm.
A simple and reproducible method was reported by
El-Mouelhi and Buszewski [25], also in 1990, described Thoma [30] (Table 1, Method 10). Solid ammonium
the use of solid-phase extraction in measuring urine sulfate was added to an aliquot of both serum sample
acetaminophen and its metabolites. and internal standard (2-acetamidophenol) and
vortexed. Methylene chloride was added to each
Immunoassay Methods sample, which was then mixed and centrifuged. The
The enzymatic immunoassay (EMIT) procedure is a organic layer was decanted and acetic anhydride and
homogenous enzyme immunoassay based on competitive pyridine added. The extracts were evaporated and
binding, in which drug in the sample competes with reconstituted in methylene chloride prior to injection.
enzyme-linked drug for antibody to acetaminophen. When
the drug-linked enzyme binds to the antibody, its activity is Hugget [31] described a rapid GC method in which the
greatly reduced. Drug concentration in the sample can extraction and derivatization of acetaminophen is
therefore be related to enzyme activity. The enzyme label carried out in one step. Sample, internal standard (N-
is G6PDH, the substrate for the enzyme and NAD are butyryl-p-aminophenol), phosphate buffer, and
included in the reaction mixture, and conversion of NAD acetylating reagent (acetic anhydride, N-
to NADH is measured at 340 nm (Table 1, Method 7). methylimidazole, and chloroform) were mixed and
centrifuged. An aliquot of the chloroform layer was
Analysis using fluorescence polarization immunoassay injected onto a 3% (w/w) C87 hydrocarbon (Apolane-
(FPIA) is similar to the enzyme-linked assays. Drug in the 87) on Chromosorb W HP (100 to 120 mesh) at
sample competes with tracer (fluoroscein)-labeled drug for 235C. The specimen, together with a quality-control
binding to acetaminophen antibody. The change in sample, can be analyzed in duplicate within 20 min.
polarization of the fluorescent radiation emitted by the Acetaminophen is detected using a flame ionization
tracer is measured and is inversely proportional to the detector with the limit of measurement of the assay
amount of acetaminophen in the original sample (Table 1, being 10 mg/L. This method has few potential sources
Method 8). of interference.
Enzymatic Methods
Price [26] used a commercially available kit which is based Reference and Preferred Methods
on the enzymatic degradation of acetaminophen by a There is no reference method for acetaminophen
bacterial arylacylamide amidohydrolase and where the estimation.
product is measured colorimetrically (Table 1, Method 9).
The method is similar to the colorimetric methods where The spot test is an inexpensive screen for
acetaminophen is converted to p-aminophenol by acid and acetaminophen in urine. The spot test will become
heat. In this assay, paracetamol is converted enzymatically positive in patients with greater than 2 mg/L in the
to 4-aminophenol, then coupled with o-cresol in serum. However, these methods are inherently
ammoniacal cupric sulfate to produce an indophenol dye imprecise. Ingestion of phenacetin and benorilate, as
which is measured at 615 nm. well as acetaminophen, may result in a positive spot
test. With the automation of acetaminophen assays on
In 1983, Hallworth [27] modified and adapted the manual laboratory instruments and very acceptable turnaround
assay for a centrifugal analyzer procedure. times, there is little place for these qualitative
methods.
Following an overdose, when it is likely that

Gas Chromatography potentially toxic acetaminophen metabolites will be
With most published GC methods, the sample must first be present, treatment with the antidote, N-acetylcysteine,
extracted and derivatized to be made volatile at should be commenced as soon as possible. In this
temperatures inside the GC column. setting, the assay method chosen for acetaminophen
must be rapid, accurate, and precise. Consideration
Prescott [28] published a GC method utilizing flame must also be given to the size and manpower of the
ionization detection. The sample containing internal laboratory and the available instrumentation.
standard (p-chloroacetanilide) was extracted into ethyl
acetate and the extract evaporated. After reconstitution in
36
Acetaminophen
According to survey data from the 2007 College of In a recent study by Polson [34], the effect of
American Pathologists (CAP) Participant Summary hyperbilirubinemia on six commonly performed
Report, only enzymatic and immunological methods are assays representing 70% of the methods chosen by
currently used for measurement of acetaminophen. All of U.S. laboratories, according to the CAP, were studied.
these methods are acceptable, so acetaminophen assays are Immunoassays that do not require a color reaction are
determined by the choice of instrumentation in the less prone to interference from endogenous
laboratory. substances.
HPLC could be considered a candidate reference method This study also suggested that products and
of analysis because of its excellent sensitivity, precision, byproducts created in colorimetric assays may show
and freedom from interference, but it is not often used as a absorbance at wavelengths near to the primary or
routine method in clinical laboratories. Although most secondary analytical wavelengths used, causing a
HPLC methods use spectrophotometric determination, positive interference. However, not all samples
some methods employ electrochemical detection to showing hyperbilirubinemia exhibited these
improve the sensitivity of the assay but are subjected to characteristics, suggesting interference from non-
electrical interferences and may involve lengthy, complex, bilirubin substances.
sample extractions.
Acetaminophen Toxicity
Gas chromatography techniques were widely used for The clinical value of measurement of acetaminophen
quantitation of acetaminophen because of their selectivity, lies in the detection and diagnosis of overdose (see
small sample volumes, and the ability to measure other Interpretation). As such, it is not appropriate to quote
drugs simultaneously. Combined gas an optimal therapeutic range. However, standard
chromatography/mass spectroscopy has also been utilized therapeutic doses of acetaminophen for analgesia and
in assaying for acetaminophen [31] and shows enhanced antipyresis give plasma concentrations of the order of
sensitivity. Both methods are time consuming and labor 10 to 20 g/mL (65 to 130 mol/L).
intensive and not recommended for routine use at this time.
Interpretation
Specimen The mechanism of action of acetaminophen is due to
Serum is recommended as the sample of choice. Plasma its inhibition of one of the cyclooxygenase enzymes,
collected into EDTA, heparin, oxalate, or citrate can be COX-3 [36,37]. By inhibiting COX-3 in the
used with most enzymatic and immunological methods. No hypothalamus, prostaglandin production is reduced,
special handling of specimen is required. Qualitative and if the temperature is raised (i.e., there is fever), the
screening for acetaminophen in urine is not recommended temperature center will reset, and the temperature will
(see above). Sample stability at standard therapeutic doses fall to normal. Likewise, COX-3 inhibition decreases
of acetaminophen for analgesia and antipyresis of 10 to 20 transmission in pain fibers in the spinal cord, and the
g/mL (65 to 130 mol/L) is at least 4 days (Koerbin sensation of pain is decreased. Paracetamol has no
unpublished). significant action on COX-1 and COX-2. This
contrasts with inhibition of cyclooxygenases 1 and 2,
Interferences through which the nonsteroidal antiinflammatory
Some of the automated immunoassay procedures have drugs (e.g., aspirin) operate, and which also have
been reported to suffer negative bias from hemolysis and analgesic and some antipyretic properties.
icterus and positive interference from lipemia [33]. The Acetaminophen therefore has neither the
nitration method is susceptible to interferences from antiinflammatory nor the antithrombotic properties of
salicylate [9] and phenolic acids if patients are uremic [10]. aspirin but is without the same gastrointestinal side
effects or association with Reye Syndrome.
Bilirubin has been shown to interfere with a variety of
colorimetric assays [34,35]. In the GDS Diagnostics The primary clinical use of acetaminophen
enzymatic assay, based on enzymatic conversion of measurement is in the assessment of possible
acetaminophen to p-aminophenol, with chromogen overdose. At usual analgesic doses in an adult (500 to
activator o-cresol forming a blue-colored indophenol, an 1000 mg), the elimination half-life is 2 to 3 hours. At
increase in background absorbance at 600 nm caused by these doses, acetaminophen is cleared by the liver, the
the presence of bilirubin appeared to cause a false increase majority being conjugated with sulfate and
in the reported acetaminophen concentration. The glucuronide. In acetaminophen overdose, the
magnitude of the false increase was directly proportional to conjugation pathways are saturated, and the excess
the bilirubin concentration. Electronic correction for drug is metabolized by cytochrome P450. The
absorbance at a second wavelength did not adequately resultant intermediate metabolites, such as the minor
compensate for the interference, which also suggests a alkylating metabolite N-acetyl-p-benzo-quinone imine
nonspectral mechanism. Although the false detection of (NAPQI) are highly reactive, and although they are
acetaminophen in hyperbilirubinemic specimens was reduced and detoxified by reacting with hepatic
consistent and reproducible, the accuracy of actual glutathione, the capacity of this system is rapidly
acetaminophen measurements between 50 and 150 mg/L exceeded, and tissue necrosis results. In addition, free-
did not seem to be affected [35]. radical production is thought to contribute to the
pathology. Patients with preexisting liver disease or
37
Acetaminophen
severe nutritional deficiencies may be at greater risk of although most effective in the first 8 hours post-
acetaminophen toxicity if their hepatic stores of ingestion, treatment with NAC provides significant
glutathione are depleted. Patients in whom cytochrome- benefit regardless of timing or degree of hepatic injury
P450 activity has been induced (e.g., with the use of [41,42].
phenobarbital) are also at a greater risk of toxicity.
It is not useful to measure the serum concentration
The consequent hepatotoxicity may not become apparent earlier than 4 hours after ingestion of the
for 2 to 3 days, when nonspecific symptoms develop acetaminophen overdose; absorption is likely to be
(gastrointestinal), and liver function tests deteriorate. continuing, and the distribution phase will be
Prothrombin time (PT) including INR, is the most sensitive incomplete. The predictive nomograms are based on
test and may be significantly abnormal 18 hours after the understanding that in the normal liver following
acetaminophen overdose. ingestion of a therapeutic dose of acetaminophen, its
half-life is less than 4 hours. In acetaminophen
The most practical indication of the likelihood of hepatotoxicity, the half-life increases [38], and the
hepatotoxicity arising following acetaminophen overdose observed incidence of hepatotoxicity is related to
is the concentration of acetaminophen in plasma/serum at a serum concentrations [39]. Furthermore, early
known time after its ingestion. Treatment of acute treatment with NAC (within 8 hours of ingestion) is
overdose is guided by the use of nomograms. These more effective than later treatment [43].
nomograms are based on the observations of Prescott
[38,39] and Rumack [40] and their successful treatment
within 12 hours of ingestion of acetaminophen using N-
acetylcysteine (NAC). Evidence now suggests that
Nomogram Interpretation
A Data uninterpretable if sample taken within 4 hours of ingestion. Repeat collection is recommended. Note the
minimal single hepatotoxic dose of paracetamol is 7.5 g in an adult (150 mg/kg), and if there is suspicion of a large
overdose, then treatment with NAC is recommended immediately.
B Liver damage highly likely. Treatment with NAC is recommended.
C Liver damage possible, especially in high-risk patients. These patients should be considered for treatment with
NAC and should be reviewed by a senior clinician.
D Severe liver damage unlikely. If there is doubt about the timing of ingestion or the need for treatment, treat with
NAC.
E Severe liver damage is still possible if large doses of paracetamol have been ingested. These patients should be
considered for treatment with NAC and should be reviewed by a senior clinician.
(With permission of Dr. G. Jones, St. Vincents Hospital, Sydney, Australia.)
38
Acetaminophen
NAC may act as a donor of sulfhydryl groups (cf. 5 Welch RM, Conney AH. Simple method for
glutathione), as well as acting as a radical scavenger. The the quantitative determination of N-acetyl-p-
critical factors in the efficient diagnosis and treatment of aminophenol (ACAP) in urine. Clin Chem
acetaminophen poisoning are: 1965;11:1064-7.
(1) time and frequency of ingestion 6 Meola JM. Emergency determination of
(2) time of blood sampling following ingestion acetaminophen. Clin Chem 1978;24:1642-3.
(3) concentration of acetaminophen 7 Chafetz L, Daly RE, Schriftman H, Lomner JJ.
(4) time of administration of NAC Selective colorimetric determination of
It should be noted that the advice and use of these acetaminophen. J Pharm Sci 1971;60:463-6.
predictive nomograms is valid only for a single ingestion 8 Glynn JP, Kendal SE. Paracetamol
(not serial episodes) and does not cover ingestion of slow- measurement. Lancet 1975;i:1147-8.
release formulations of acetaminophen [44,45]. 9 Mace PF, Walker G. Salicylate interferences
with plasma paracetamol method. Lancet.
Acetaminophen Performance Goals 1976;2:1362
With the occurrence of repeated ingestion, the serum 10 Bailey DN. Colorimetry of serum
acetaminophen concentration is of little use other than to acetaminophen (paracetamol) in uraemia. Clin
confirm ingestion. Clinical treatment decisions may then Chem 1982;28:187-90.
be based on estimations of intake. 11 Archer CT, Richardson RA. An improved
colorimetric method for the determination of
Survey data from the 2007 CAP Participant Summary plasma paracetamol. Ann Clin Biochem
Report shows imprecision values (% coefficient of 1980;17:45-6.
variation [CV]) for measurement of acetaminophen range 12 Weiner K. Paracetamol estimation: comparison
from 1.9% to 7.6% at concentrations of approximately 150 of a quick colorimetric method with a standard
g/mL and from 1.9% to 7.3% at about 90 g/mL for spectrophotometric method. Ann Clin Biochem
assays with > 20 participants. 1977;14:55-8.
13 Barker DE, Jacobs AG. Paracetamol
Acetaminophen is not regulated under the Clinical estimation: a new approach to reducing
Laboratory Improvement Amendments of 1988 (CLIA-88) salicylate interference. Ann Clin Biochem
for proficiency testing, but acceptable performance criteria 1982;19:120-4.
for measurement of acetaminophen require that 14 Longlands MG, Weiner K. Minimisation of
laboratories be accurate to within 10% of the peer-group salicylate interference in the Glynn and Kendal
mean. Analytical performance goals set by the Royal paracetamol procedure. Ann Clin Biochem
College of Pathologists of Australasia (RCPA) are 3 1982;19:187-190.
g/mL up to 30 g/mL and 10% > 30 g/mL (20 15 Liu TZ, Oka KH. Spectrophotometric screening
mol/L) up to 200 mol/L and 10% > 200 mol/L. method for acetaminophen in serum and
plasma. Clin Chem 1980;26:69-71.
The 2007 CAP data show that there is a lack of 16 Knepil J. A sensitive, specific methods for
harmonization between commonly used acetaminophen measuring N-acetyl-p-aminophenol
assays, and there are differences in concentration between (paracetamol) in blood. Clin Chim Acta
these assays of greater than 25%. These differences will 1974;52:369-72.
affect distribution of patients between B and C categories 17 Routh JI, Shane NA, Arredondo EG, Paul WD.
of potential treatment. Determination of N-acetyl-p-aminophenol in
References 18 OHaver TC. Potential clinical applications of
1 Brodie BB, Axelrod J. The estimation of derivative and wavelength-modulation
acetanilide and its metabolic products, aniline, N- spectrometry. Clin Chem 1979;25:1548-53.
acetyl-p-aminophenol pre-aminophenol in 19 Dingeon B, Charvin MA, Quenard MT, Thome
biological fluids and tissues. J Pharmacol Exp Ther H. Multi-wavelength analyses of second-
1948;94:22-8. derivative spectra for rapid determination of
2 Bosch ME, Sanchez AJ, Rojas FS, Ojeda CB. acetaminophen in serum. Clin Chem
Determination of paracetamol: historical evolution. 1988;34:1119-21.
J Pharm Biomed Anal 2006;42:292-321. 20 Manno BR, Manno JE, Dempsey CA, Wood
3 Liu TZ. A simple homogenous EIA-based MA. A high-pressure liquid chromatographic
fluorescence spot test on filter paper for rapid method for the determination of N-acetyl-p-
screening of serum acetaminophen. Clin Chem aminophenol (acetaminophen) in serum or
1984;30:1025. plasma using a direct-injection technique. J
4 Lester L, Greenberg LA. The metabolic rate of Anal Toxicol 1981;5:24-8.
acetanilid and other aniline derivatives. J Pharmacol 21 Kinberger B, Holmen A. Simultaneous
Exp Ther 1947;90:68-75. determination of acetaminophen, theophylline
39
Acetaminophen
and salicylate in serum by HPLC. J Chromatogr 34 Polson J, Wians FH Jr, Orsulak P, Fuller D,
Biomed Appl 1982;229:492-7. Murray NG, Koff JM et al. False-positive
22 Starkey BJ, Loscombe SM, Smith JM. Paracetamol acetaminophen concentrations in patients with
(acetaminophen) analysis by HPLC. Ther Drug liver injury. Clin Chim Acta 2008;391:24-30.
Monit 1986;8:78-84. 35 Bertholf RL, Johannsen LM, Bazooband A,
23 Sood SP, Green VI. Routine methods in toxicology Mansouri V. False-positive acetaminophen
and therapeutic drug monitoring by HPLC. I. Rapid results in a hyperbilirubinemic patient. Clin
method for determination of acetaminophen in Chem 2003;49:695-8.
plasma, including a STAT procedure. Ther Drug 36 Chandrasekharan NV, Dai H, Roos KL,
Monit 1987;9:248-54. Evanson NK, Tomsik J, Elton TS, Simmons
24 Kamali F, Herd B. Liquid-liquid extraction and DL. COX-3, a cyclooxygenase-1 variant
analysis of paracetamol (acetaminophen) and its inhibited by acetaminophen and other
major metabolites in biological fluids by reverse- analgesic/antipyretic drugs: cloning, structure,
phase ion-pair chromatography. J Chromatogr and expression. Proc Natl Acad Sci USA
Biomed Appl 1990;530:222-5. 200;99:13926-31.
25 el-Mouelhi M, Buszewski B. Application of solid- 37 Graham GG, Scott KF. Mechanism of action of
phase extraction to the isolation and determination paracetamol. Am J Ther. 2005;12:46-55.
of paracetamol and its metabolites. J Pharm Biomed 38 Prescott LF, Wright N, Roscoe P, Brown SS.
Anal 1990;8:651-3. Plasma paracetamol half-life and hepatic
26 Price CP, Hammond PM, Scawen MD. Evaluation necrosis in patients with paracetamol overdose.
of an enzymic procedure for the measurement of Lancet 1971;1:519-22.
acetaminophen Clin Chem 1983;29:358-61. 39 Prescott LF, Illingworth RN, Critchley JAJH,
27 Hallworth MJ. Enzymic method for acetaminophen Stewart MJ, Adam RD, Proudfoot AT.
adapted to a centrifugal analyzer. Clin Chem Intravenous N-acetylcysteine: the treatment of
1983;29:2123- 4. choice for paracetamol poisoning. Br Med J
28 Prescott LF. The gas-liquid chromatographic 1979;2:1097-1100.
estimation of phenacetin and paracetamol in plasma 40 Rumack B, Peterson RC, Koch GG, Amara IA.
and urine. J Pharm Pharmacol 1971;23:111-5. Acetaminophen overdose. Arch Intern Med
29 Thomas BH, Coldwell BB. Estimation of 1981;141:380-385.
phenacetin and paracetamol in plasma and urine by 41 Harrison PM, Keays R, Bray GP, Alexander
gas-liquid chromatography. J Pharm Pharmacol GJ, Williams R. Improved outcome of
1972;24:243. paracetamol-induced fulminant hepatic failure
30 Thoma JJ, McCoy M, Ewald T, Myers N. by late administration of acetylcysteine. Lancet.
Acetaminophen: an improved gas chromatographic 1990;335:1572-3.
assay. J Anal Toxicol 1978;2:226-8. 42 Rowden AK, Norvell J, Eldridge DL, Kirk
31 Huggett A, Andrews P, Flanagan RF. Rapid MA. Acetaminophen poisoning. Clin Lab Med
micromethod for the measurement of paracetamol 2006;26:49-65.
in blood plasma or serum using gas-liquid 43 Kerr F, Dawson A, Whyte IM, Buckley N,
chromatography with flame-ionisation detection. J Murray L, Graudins A et al. The Australasian
Chromatogr 1981;209:67-76. Clinical Toxicology Investigators
32 Garland WA, Hsiao KC, Pantuck EJ, Conney AH. Collaboration randomized trial of different
Quantitative determination of phenacetin and its loading infusion rates of N-acetylcysteine. Ann
metabolite acetaminophen by GLCchemical Emerg Med 2005;45:402-8.
ionization mass spectrometry. J Pharm Sci 44 Graudins A, Aaron CK, Linden CH. Overdose
1977;66:340-4. of extended-release acetaminophen. N Engl J
33 Ryder KW, Trundle DS, Bode MA, Cole RE, Med 1995;333:196.
Moorhead WR, Glick MR. Effects of hemolysis, 45 Temple AR, Mrazik TJ. More on extended-
lipemia, and icterus on automated immunoassays. release acetaminophen. N Engl J Med
Clin Chem 1991;37:1134-5. 1995;333:1508-9.
40
Acetaminophen
Tables
Table 1: Acetaminophen Methods Summary

Method 1: Azo dye formation (Brodie and Axelrod)[1]; spectrophotometric
Principle of analysis: Sample extracted into diethyl ether and back-extracted into alkali. Heat hydrolysis to p-
aminophenol that is diazotized with ammonium sulfamate and coupled with -naphthol. This red-violet color is
measured at 510 nm.
Comments: Sulfate and glucuronide conjugates render assay nonspecific; Amax = 510nm
Method 2: Spot test; qualitative, colorimetric
Principle of analysis: o-Cresol + acetaminophen indole phenol blue
Comments: Screening test; highly sensitive and relatively specific
Method 3: Nitro dye formation
a. Spectrophotometric
Principle of analysis: Samples are deproteinized with TCA, the supernatant mixed with 6NHCl and nitric acid to
form a nitrous acid derivative that is then neutralized with ammonium sulfamate. NaOH is added and the yellow
color measured at 430nm.
Comments: Salicylate and phenolic acids interfere; Amax = 430 nm
b. Spectrophotometric
Principle of analysis: A modified procedure of Glynn and Kendal [7]; trichloroacetic acid was substituted for
hydrochloric acid; a reduced nitrite concentration and a nitration reaction time of 2 min
Comments: Linear in range of 0 to 500 mg/L; HPLC more sensitive and specific; Amax = 450 nm
Method 4: Ferric reduction; spectrophotometric (simple extinction measurement)
Principle of analysis: Measures acetaminophen based on the reduction of ferric 2,4,6-tris(2-pyridyl)-S-triazine
(TPTZ)
Comments: Phenolic hydroxyl groups interfere, and TPTZ is light sensitive; Amax = 593 nm
Method 5: Spectrophotometry
a. Differential extinction measurement
Principle of analysis: Measures acetaminophen at differential absorbance peak of 290 nm, avoiding interference
with salicylate
Comments: Reading at isosbestic point of salicylate removes interference from this compound; requires narrow
band-pass spectrophotometer
b. Second derivative
Principle of analysis: Rate of change of absorbance with wavelength is measured in units of absorbance per
nanometer2
Comments: Some drugs lead to spectral interferences; Amax = 300 nm; Amin = 216 nm
Method 6: High-performance liquid chromatography (HPLC), reversed-phase; chromatographic separation
Principle of analysis: Packing octadecylsilane-bonded silica; organic solvent extraction, protein precipitation or
direct injection; mobile phase is dilute acetic acid/methanol/ethyl acetate; detection at 254 nm
Comments: Rapid and sensitive (to 1 g/mL), or improved to 1 ng/mL with more lengthy and labor intensive
extractions; ability to quantitate other drugs simultaneously
Method 7: Enzyme multiplied immunoassay (EMIT); competitive binding
Principle of analysis: Drug in patient sample competes with drug-enzyme complex for limited amount of
antibody; enzyme activity related to drug level
Comments: Available on stat basis; linear in range 10 to 200 mg/L
Method 8: Fluorescence polarization (FPIA); competitive binding
Principle of analysis: Drug in patient sample competes with fluorescein tracer for limited amount of antibody;
polarization inversely related to drug level
Comments: Requires dedicated instrument, most commonly used; linear in range 10 to 200 mg/L
Method 9: Enzymatic techniques; spectrophotometric
Principle of analysis: Acetaminophen converted to aminophenol by bacterial arylacylamidase; indophenol dye
complex formed by the reaction of aminophenol with o-cresol and ammoniacal cupric sulfate
Comments: Adaptable to automation; available on stat basis; linear in range 0 to 377.5 mg/L; Amax = 615 nm
Method 10: Gas chromatography; chromatographic separation
Principle of analysis: Extracted into methylene chloride and ammonium sulfate; converted to trimethylsilyl
derivative; injected isothermically onto column and flame ionization detectors
Comments: Replaced by HPLC and immunoassay; time consuming and labor intensive
41
Acetaminophen
Figures
Figure 1: Acetaminophen HPLC Chromatogram
Chromatogram of HPLC analysis for acetaminophen:

A, Spiked plasma: (1) acetaminophen (100 mg/L); (2) theophylline (20 mg/L); (3) -8-hydroxyethyltheophylline (3 mg/L)
(internal standard).
B, Serum patient sample: (1) acetaminophen (30 mg/L); (3) -8-hydroxyethyltheophylline (3 mg/L) (internal standard).
Procedure: High-Performance Liquid Chromatography

Principle Waters Radial Compression Module, model RCM-100
Acetaminophen is separated by reversed-phase high- Waters Radial-PAK cartridge, 8 NV C 18 4 (Nova-Pak)
performance liquid chromatography and quantitated by part #86342
measurement of peak heights relative to the internal Waters Guard-Pak Cartridge Nova-Pak C 18 part #
standard. Serum is deproteinized and extracted with 15220
chloroform; detection is by ultraviolet spectroscopy at Waters Variable Wavelength UV Detector model 481
270 nm. Hewlett-Packard Integrator model 3394
Specimen Sample Preparation:
Serum 1. Pipet 100 L of sample, control, or standards.
Reagents and Materials 2. Add 50 L of 1 M hydrochloric acid to each
1. HPLC-grade isopropanol and acetonitrile, tube.
glacial acetic acid, and chloroform. 3. Add 1 mL of the internal standard solution.
2. Stock internal standard (300 mg/L). Place 4. Vortex mix and centrifuge to 1500 rpm.
150 mg of -8-hydroxyethyltheophylline in a 5. Decant the organic layer, and place tubes in a
500-mL volumetric flask, and dilute with heating block at 60C under a gentle stream of air.
chloroform. This standard is stable indefinitely 6. Add 200 L of the mobile phase to each tube
if stored away from light. and vortex.
3. Working internal standard (3 mg/L). Place 7. Inject 50 L of sample into the HPLC system,
5 mL of stock internal standard in a 500-mL using the following chromatographic parameters:
volumetric flask with 50 mL of isopropanol, a. Chart speed: 0.5 cm/min
and make to the mark with chloroform. This is b. Ultraviolet sensitivity: 0.1 AUFS
stable indefinitely if stored away from light. (absorbance units at full scale) at 270 nm
4. Standards. Acetaminophen c. Column temperature ambient
5. Mobile phase. Add 1.36 g of sodium acetate d. Flow rate: 2.0 mL/min
and 35 mL of 20% (v/v) acetic acid to 890 mL e. Run time: Approximately 7 min
deionized water in a 1-L volumetric flask, and Elution Order; Approximate Retention Time (min)
swirl to dissolve. Filter the aqueous buffer Theobromine 2.33
using a nitrocellulose filter (such as Millipore) Acetaminophen 3.01
of 0.5-mm pore size. Filter 20 mL of Paraxanthine 3.52
isopropanol and 55 mL acetonitrile to the Theophylline 3.74
reservoir. (pH 3.8) Internal standard 4.34
Assay The retention times of caffeine and salicylate depend on
Equipment: the pH of the mobile phase.
Waters pump, model 6000A The assay is linear to at least 500 mg/L, with a lower
Waters autosampler, WISP model 712
42
Acetaminophen
limit of detection of 1 mg/L. In each sample, the peak height of acetaminophen is

A typical chromatogram is shown in Acetaminophen: compared to the internal standard from which peak
Figure 1. height ratios are calculated. Each unknown sample is
then compared to a calibrator peak height ratio of a
known concentration. The calculation is as follows:
Calculations
[Acetaminophen] = [CAL] peak height unknown/peak height of internal standard

peak height calibrator/peak height of internal standard
where [Acetaminophen] = drug concentration (mg/L),

[CAL] = calibrator concentration (mg/L)
Notes
1. When a new column is installed, check that
adequate separation of theobromine and
acetaminophen is obtained. If the separation is
inadequate, decrease the percentage of
acetonitrile in the mobile phase.
2. Salicylate may interfere, depending on the pH
of the mobile phase. To increase the retention
time of salicylate, decrease the pH. To decrease
the retention time of salicylate, increase the pH.
43
Adrenocorticotropic Hormone (ACTH)

Hassan M.E. Azzazy
Name: Adrenocorticotropic hormone (ACTH)

Clinical Significance: Refer to Chapter 51 Adrenal Hormones and Hypertension, in the 5th edition of
Clinical Chemistry: Theory, Analysis, Correlation
Molecular mass: ~4500 D
Chemical class: Polypeptide (39 amino
acids)
Principles of Analysis and Current Usage al. [7] evaluated an IRMA developed for the
ACTH is a 39-amino-acid peptide, produced in the measurement of ACTH in human plasma. The assay
anterior pituitary as part of a large precursor protein employed goat polyclonal antibodies specific for ACTH
called pro-opiomelanocortin (POMC), which contains 26-39 as capture antibodies and 125I-labeled monoclonal
the sequence for several other biologically active antibodies specific for ACTH 1-17. Solid-phase
peptides as well. After secretion, POMC is rapidly separation of the immune complex was achieved using
cleaved to its component peptides. The major stimulus mouse anti-goat antibody-coated polystyrene beads. The
for POMC (and hence ACTH) release is assay had a detection limit of 1.7 ng/L and total
hypocortisolemia. imprecision of < 10% at ACTH concentrations between
9 and 801 ng/L. A time-resolved immunofluorometric
ACTH concentration can be determined by bioassay, assay for measurement of ACTH in unextracted plasma
receptor assays, or immunoassays. Bioassays are based has also been reported [8].
on the ability of ACTH to stimulate steroidogenesis or Automated non-isotopic ACTH immunoassays with
depletion of ascorbic acid in isolated adrenocortical cells short incubation times have recently been developed.
or in adrenal glands of hypophysectomized animals [1]. The Nichols Advantage ACTH assay utilizes acridinium
Bioassays are labor intensive and expensive. Receptor esterlabeled mouse monoclonal antibody that binds to
assays measure only the biologically active fraction of the C-terminal region of ACTH and a biotinylated goat
ACTH [2]. These assays utilize solubilized binding polyclonal antibody that binds to the N-terminal region.
proteins obtained from normal or neoplastic After incubation with the patients plasma, streptavidin-
adrenocortical cells. Receptor assays suffer from several coated magnetic beads are used to separate the immune
drawbacks, including lability of receptor preparations complex. Interassay imprecision at mean ACTH values
and technical complexity. of 17.7 and 731 ng/L were 12% and 6.1%, respectively
[9]. A second assay, the DPC IMMULITE ACTH assay,
Several radioimmunoassays (RIA) were initially employs monoclonal murine antibodies coated on
developed for ACTH measurement. Many of the RIA polystyrene beads and alkaline phosphataselabeled
methods, however, required a pre-assay extraction step polyclonal rabbit anti-ACTH antibodies. This assay also
and extended incubation time because they employed uses a chemiluminescent substrate. Interassay
antibodies with low avidity, and the circulating imprecision of this assay at mean ACTH concentration
concentration of ACTH in healthy subjects is low [3,4]. of 15.7 and 811 ng/L were 11% and 2.8%, respectively.
Additionally, a single-step electrochemiluminescence
Sandwich immunoradiometric assays (IRMA) have been
immunoassay has been developed (Elecsys ACTH
shown to have several advantages over the competitive
assay). The assay utilizes two murine monoclonal
immunoassays, including improved sensitivity,
antibodies. One of the antibodies is labeled with
precision, and shorter incubation time [5,6]. Rosano et
ruthenium (detector), and the other is a biotinylated
antibody (capture) that binds to microparticles coated
with streptavidin. The reported assay detection limit is
0.5 ng/L, and the within-run and between-run
imprecision (at ACTH concentration 12 to 971 ng/L)
i
Adrenocorticotropic Hormone (ACTH) were 1.9% and 5%, respectively [10].
First edition: Not done Reference and Preferred Methods
Methods edition: Barbara M. Goldsmith There is no reference method for ACTH. Both isotopic
Second edition: Not updated and non-isotopic ACTH immunoassays are reported to
Third edition: Not updated have comparable performance. Comparing results from
Fourth edition: Not updated different ACTH immunoassays is possible only if
Fifth edition: Hassan M.E. Azzazy calibrators are clearly defined. Two standards are
44
currently available from the National Institute of Reference Intervals

Biological Standards and Control (UK, purified human
ACTH 1-39; MRC 74/555; 6.2 IU/25 g) and the Method Specimen Reference Range
National Pituitary Agency (Baltimore, MD; synthetic (pg/mL)
a
ACTH 1-39; 4.71 IU/50 g). IRMA ACTH assays are RIA EDTA plasma
b
Newborn (1 day) 10-185
Adult (8:00 am) <120
generally preferred over RIA for investigating b
Adult (4:00-8:00 pm) <85
hypothalamic-pituitary-adrenal disorders. Because RIA a
[Endocrine Sciences. Pediatric
assays utilize polyclonal antibodies and will be able to Laboratory Services.
recognize both fragments and whole-molecule ACTH, Tarzana, CA: Endocrine
Sciences; 1988.]
they are recommended for investigating ectopic ACTH b
[Vanderbilt Pathology
production by tumors, because tumors may secrete Laboratory Services. Test
fragments as well as intact ACTH. Catalog. Nashville, TN:
Vanderbilt Pathology
Laboratory Services; 1989.]
IRMA EDTA plasma Adult (7:00-10:00 am) 9-52
Specimen [Nichols Institute Reference
ACTH is unstable in blood, easily oxidized, strongly Laboratories. Test Catalog.
adsorbs to glass surfaces, and can be rapidly degraded by San Juan Capistrano, CA:
Nichols Institute Research
plasma proteases into immunoreactive fragments. Laboratories; 1993.]
Therefore, proper care must be taken in collection, RIA Amniotic fluid 10-18 weeks: 209
handling, transportation, and storage of specimens. 26-30 weeks: 430
Preanalytical variables such as time of day at which 35-36 weeks: 162
sample is collected, stress from a poorly performed
[Tulchinsky D, Ryan KJ, eds.
venipuncture, and/or prior administration of cortisol Maternal-Fetal
must be taken into consideration. Samples for ACTH Endocrinology. Philadelphia:
tests must be drawn prior to glucocorticoid Saunders; 1980.]
administration. Morning (6:00 to 10:00 am) specimens Automated EDTA plasma Adult (AM draws) 10-60
chemiluminesc [Mayo Clinic. Test Catalog.
are preferred. Blood is usually drawn in pre-chilled ent immuno- Rochester, MN: Mayo
polystyrene EDTA tubes. Plasma is separated metric assay Clinic; 2005.]
immediately from cells in a refrigerated centrifuge and
frozen within 15 minutes. Plasma is stored frozen at
20C or preferably at 70C in new plastic vials. For Interpretation
long-term storage, aprotonin (500 kU/mL) should be Determination of ACTH in plasma is the best test to
added. Mercaptoethanol may be added to specimens to distinguish between primary and secondary adrenal
protect ACTH against oxidation. Prior to assay, thawed insufficiency. Measurement of plasma ACTH
plasma must be centrifuged to remove any fibrin clots concentration is used to assess Cushings disease,
that may interfere with the assay. adrenal tumors, ectopic ACTH-producing tumors,
Addisons disease, Nelsons syndrome, adrenal tumors,
Interferences and hypopituitarism. ACTH secretion is pulsatile and
ACTH concentration follows a nycthemeral rhythm, increases in cases of hypocortisolemia when the pituitary
with highest levels observed between 6:00 and 8:00 a.m. and hypothalamus are intact.
and lowest levels between 9:00 and 10:00 p.m. For
proper monitoring, samples should be collected at the In primary adrenal insufficiency, ACTH levels usually
same time each day (early morning preferred). exceed 100 pg/mL by RIA. It should be noted, however,
Pregnancy, menstruation, and stress may increase ACTH that ACTH level alone cannot be used to differentiate
secretion. Aminoglutethimide, amphetamine, levodopa, between control subjects and patients with central
metoclopramide, metyrapone, pyrogens, RU-486, hypoadrenalism. One study reported identical ACTH
vasopressin, hypoglycemia, and insulin have been levels in control subjects (4 to 81 pg/mL) and patients
reported to increase plasma ACTH levels. Depressed with confirmed pituitary disease (8 to 75 pg/mL) [12].
ACTH levels may be observed following administration
of dexamethasone or other corticosteroids or collection In Addisons disease (primary adrenal insufficiency),
of specimens in tubes containing heparin. elevated ACTH levels (>1000 pg/mL) are typical.
ACTH levels are also elevated in congenital adrenal
Heterophilic antibodies in serum of humans routinely hyperplasia, pituitary-dependent Cushings disease,
exposed to animals or animal serum products can react ectopic ACTH-producing tumors, and Nelsons
with the immunoglobulins included in ACTH syndrome. ACTH determinations can also help to
immunoassays, causing interference [11]. identify the cause of cortisol hypersecretion in Cushings
syndrome. ACTH levels are typically low when this is
due to lesions or hyperplasia of the adrenal cortex, and
high in cases of ectopic ACTH production or
45
hypersecretion of ACTH by the pituitary. Low levels are 6 Zahradnik R, Brennan G, Hutchison JS, Odell
observed when adrenal insufficiency is secondary to WD. Immunoradiometric assay of corticotropin
pituitary dysfunction, adrenal carcinoma, adenoma, and with use of avidin-biotin separation. Clin Chem
hypopituitarism. 1989;35:804-7.
7 Rosano TG, Demers LM, Hillam R, Dybas MT,
Because the major stimulus to ACTH production is Leinung M. Clinical and analytical evaluation
hypocortisolemia, interpretation of ACTH concentration of an immunoradiometric assay for
requires concurrent measurement of cortisol to be made. corticotropin. Clin Chem 1995;41:1022-7.
High levels of ACTH and cortisol are observed in 8 Dobson S, White A, Hoadley M, Lovgren T,
Cushings disease and ectopic ACTH secretion. Low Ratcliffe J. Measurement of corticotropin in
levels of both hormones are observed in hypopituitarism. unextracted plasma: comparison of a time-
Elevated levels of cortisol with low ACTH may indicate resolved immunofluorometric assay and an
adrenal tumor. Elevated ACTH and low cortisol levels immunoradiometric assay, with use of the same
are observed in Addisons disease. monoclonal antibodies. Clin Chem
1987;33:1747-51.
ACTH in amniotic fluid may be of fetal origin, and low 9 Vogeser M, Engelhardt D, Jacob K.
levels may be observed in anencephalic fetuses; high Comparison of two automated
levels may indicate immaturity of hypothalamic adrenocorticotropic hormone assays. Clin Chem
regulation. Dexamethasone suppression testing may be 2000;46:1998-2000.
useful for distinguishing causes of increased ACTH 10 Verschraegen I, Anckaert E, Schiettecatte J,
levels. High-dose dexamethasone suppresses ACTH and Mees M, Garrido A, Hermsen D et al.
cortisol secretion in Cushings disease but has no effect Multicenter evaluation of a rapid
in cases of adrenal adenomas, adrenal carcinomas, and electrochemiluminescent adrenocortico-tropic
ectopic ACTH producing tumors. hormone (ACTH) immunoassay. Clin Chim
Acta 2007;380:75-80.
Performance Goals 11 Boscato LM, Stuart MC. Heterophilic
The current Clinical Laboratory Improvement antibodies: a problem for all immuno-assays.
Amendments (CLIA) performance goal for measurement Clin Chem 1988:34:27- 33.
of ACTH is for laboratories to be within 3 standard 12 Oelkers W, Diederich S, Bahr V. Diagnosis and
deviations of the peer-group mean. According to the therapy surveillance in Addisons disease: rapid
2007 College of American Pathologists (CAP) Survey, adrenocortico-tropin (ACTH) test and
coefficients of variation (CV) for all ACTH methods at measurement of plasma ACTH, renin activity,
mean values of 168.6 pg/mL (SD 20.6) and 311.3 pg/mL and aldosterone. J Clin Endocrinol Metab
(SD 45.8) were 12.2% and 14.7%, respectively. 1992;75:259-64.
References
1 Sayers G. Bioassay of ACTH using isolated

cortex cells. Applications: structure activity
relationship for ACTH and analogues, assay of
corticotrophin-releasing factor, and assay of
plasma ACTH. Ann NY Acad. Sci
1977;297:220-41.
2 Lefkowitz RJ, Roth J, Pastan I. Radio-receptor
assay of adrenocorticotropic hormone: new
approach to assay of polypeptide hormones in
plasma. Science 1970;170:633-5.
3 Gutkowska J, Julesz J, St-Louis J, Genest J.
Radioimmunoassay of corticotropin from
4 Arts CJ, Koppeschaar HP, Veeman W, Thijssen
JH. A direct radioimmunoassay for the
determination of adrenocortico-tropic hormone
(ACTH) and a clinical evaluation. Ann Clin
Biochem 1985; 22:247-56.
5 Dobson SH, Gray C, Smith H, Baker T,
Ratcliffe JG, White A. Selection and
optimization of monoclonal antibodies for a
two-site immunoradiometric assay for ACTH. J
Immunol Methods 1986;88:83-90.
46
Alanine Aminotransferase
Robert L. Murray
Name: Alanine aminotransferase, ALT, L-alanine:2-oxoglutarate

aminotransferase, serum glutamate pyruvate transaminase, SGPT
Clinical significance: Refer to Chapter 27, Liver Function, in the 4th Edition of Clinical Chemistry:
Enzyme number: EC 2.6.1.2
Molecular mass: approximately 101,000 daltons
Chemical class: enzyme, protein
NRSCL reference
Method: NCCLS RS4-A
Principles of Analysis and Current Usage adenine dinucleotide (NADH). The disappearance of NADH
Alanine aminotransferase (ALT) catalyzes the transfer of is monitored spectrophotometrically (at 340 nm). Preferably,
an amino group between L-alanine and L-glutamate; the the absorbance change should be monitored continuously
corresponding ketoacids in this process are - rather than by readings at several time points or only the end-
point.
ketoglutarate and pyruvate (Figure 1). In vivo, this
reaction proceeds to the right to provide a source of
Reference and Preferred Methods
nitrogen for the urea cycle. The pyruvate thus generated is
There have been many minor modifications in the enzymatic
available for entry into the citric acid cycle, whereas the
technique since its introduction by Wrblewski and LaDue [3-
glutamate is deaminated (catalyzed by glutamate
7]. In addition, the specifications listed in the reference
dehydrogenase), yielding ammonia and -ketoglutarate.
method published by the International Federation of Clinical
Chemistry (IFCC) have been modified over the years, the
The reaction is reversible, with the chemical equilibrium
most recent modification in 2002 being to optimize conditions
favoring the formation of alanine and -ketoglutarate. for 37C [7,8]. The main use of the IFCC reference method is
Because these products are relatively difficult to assay, to assay calibrators for use in routine methods. This allows
however, analytical techniques typically force the reverse routine methods to be traceable to the IFCC reference method,
reaction, allowing quantitation of pyruvate. Two methods with the goal of reducing the biases between methods.
of ALT analysis have enjoyed wide popularity for routine
clinical use: the Reitman-Frankel method [1], which In the most recent IFCC reference method (2002) [7], 0.20 mL
involves the measurement of ALT activity by conversion of serum is preincubated for 5 minutes in 2.00 mL of a
of the reaction product, pyruvate, to its hydrazone (Table
mixture that contains all reactants except -ketoglutarate.
1: Method 1); and the Wroblewski method [2] (Table 1:
During this preincubation period, the added lactate
Method 2), in which the ALT reaction is coupled to a
dehydrogenase (LD) rapidly converts the endogenous
lactate dehydrogenase (LD) reaction. This is the most
pyruvate in the serum to lactate, and the pyridoxal phosphate
common method in use today, and the former method is
cofactor joins with any inactive apoenzyme to form an
of historical interest only.
increased amount of active ALT. With the addition of 0.20 mL
of -ketoglutarate, the primary reaction is initiated, and the
In the LD coupled method, the pyruvate product of the
concentrations shown in Table 2 are reached, exclusive of the
ALT reaction is reduced to lactate by nicotinamide
small increases caused by the presence of endogenous material
in the serum. After steady state is reached, the rate of NADH
oxidation is monitored repeatedly at 339 nm. The rate of
iALT
change in absorbance is corrected for a reagent blank.
First edition: Robert L. Murray Most reference methods, including the current IFCC
Methods edition: Robert L. Murray procedure [7], have included pyridoxal phosphate. The need
for addition of this cofactor has been widely debated. The
Second edition: Robert L. Murray
essential nature of the cofactor has long been recognized, but
Third edition: Steven C. Kazmierczak because it is usually present in human serum in adequate
Fourth edition: Steven C. Kazmierczak amounts, many investigators do not add this component to the
Fifth edition: James J. Miller reaction mixture. In the occasional patient with severe vitamin
47
B deficiency, this could lead to a serious underestimation [9]. A marked decrease in ALT activity is seen following
of AST activity. freeze/thaw cycles 10]. ALT has been found to be stable in
whole blood for up to 24 hours [11]. Urine has little or no
The presence of aminotransferases in the reagents is activity and is not recommended for analysis.
possible if the LD is not carefully prepared. Good-quality
enzymes will not pose a problem; in any event, a blank
determination will identify the problem. The presence of Interferences
pyruvate is a potential source of error, since in the There is significant ALT activity in erythrocytes and
presence of endogenous LD, pyruvate will be converted significant hemolysis (>300 mg/dL hemoglobin) may
to lactate, with simultaneous consumption of NADH. This artifactually increase apparent ALT activity. Icteric (bilirubin
problem is circumvented by the addition of a large excess <40 mg/dL) and lipemic (triglycerides <3000 mg/dL)
of LD, so that endogenous pyruvate is converted during specimens generally do not interfere with measurement of
the preincubation period, eliminating interference during ALT [12]. Metronidazole (Flagyl) may interfere with ALT
the measurement period. methods because of its relatively high concentration and
absorbance near 340 nm [13].
Some older reference methods based on the Wrblewski
coupled enzymatic method used phosphate buffer, which Alanine Aminotransferase Reference Interval
retards the recombination of added pyridoxal phosphate When analyzed at 37C by methods employing activation with
with the apoenzyme. If a large amount of the inactive PP, the normal adult reference interval is approximately 8 to
enzyme is present, a falsely low activity will be observed. 47 U/L [14]. However, it must be noted that ALT activities are
Activation of apoenzyme is more efficient in Tris buffer. age dependent. Healthy newborns have been reported to show
However, NADH is somewhat less stable in Tris buffer an upper reference interval of up to double the adult level.
than it is in phosphate buffer. For this reason, the Tris These values decline to adult levels by approximately 3
concentration is kept relatively low at 100 mmol/L. months of age. This increased activity has been attributed to
seepage from the neonates immature hepatocytes, which have
Most routine methods today use the Wrblewski coupled more permeable membranes. Men have been reported to show
enzymatic method and may be traceable to the current higher ALT values than women. Upper reference limits in
IFCC reference method. The American Association for individuals 10 years of age are approximately half of those
Clinical Chemistry proposed a method [3] for the small seen in individuals at 40 years of age [15]. ALT activity peaks
clinical chemistry laboratory that differs from the IFCC in at approximately the fourth to fifth decade of life and then
that (1) a single reagent is used to avoid a two-step gradually declines.
addition procedure, (2) the reaction is read after 150
seconds for the following 180 seconds, and (3) pyridoxal Diurnal variations in ALT have been observed in both healthy
phosphate is not added. individuals and those with cirrhosis. Up to 45% variation may
be seen, with higher values being observed in the afternoon
The drawbacks of the dinitrophenylhydrazine method are [16]. Other factors that have been reported to affect ALT
that (1) the pyruvate produced by the reaction results in include African-American race (15% higher than Caucasians),
feedback inhibition of ALT, and thus specimens body mass index (40 to 50% higher with high body mass
exhibiting high activity are spuriously lowered; and (2) index), and exercise (20% lower in those who exercise) [17].
any ketone in serum can react, though most do not result Ingestion of food causes no changes in measured ALT
in an absorbance change in the region measured. activity.
However, acetoacetic acid and hydroxybutyric acid, both
components of ketosis, do cause false elevations. Interpretation
In contrast to aspartate aminotransferase (AST), which is
The U.S. National Institute of Standards and Technology found in both the cytoplasm and mitochondria, ALT is found
(NIST) Standard Reference Material No. 909b is a exclusively in the cytoplasm. The tissue distribution of ALT
lyophilized human serum preparation intended for use in and the ratio of ALT tissue activity to ALT plasma activity are
evaluating the accuracy of routine methods. It is available presented in Table 3. Based on activity per gram of wet tissue,
to manufacturers and laboratories for the validation of liver has the greatest amount of enzyme activity, with kidney
ALT methods. being the next most active tissue. Liver disease, in particular
hepatocyte necrosis, is the most important cause of increased
Specimen ALT activity. Because serum activities of ALT are unusually
Serum is the preferred specimen. Oxalate, heparin, and sensitive to liver damage, increases in ALT readily occur
citrate do not inhibit the enzymatic activity but may following moderate to excessive use of alcohol or following
introduce slight turbidity. Hemolyzed specimens should exposure to a variety of hepatotoxic agents. ALT is often used
be avoided, since erythrocytes contain three to five times as part of a battery of enzymes to establish the presence and
more ALT activity than is found in serum. ALT is stable extent of liver damage. The half-life of ALT is approximately
in serum for 3 days at room temperature or 1 week at 4C 47 10 hours [18].
48
ALT is usually higher than AST in most types of liver catalytic concentrations of alanine aminotransferase.
disease in which the activity of both enzymes is Clin Chem Lab Med 2002; 40: 718-24.
predominantly from the hepatocyte cytosol. When liver 8 Bergmeyer HU, Horder M, Rej R. International
necrosis is substantial, as in individuals with alcoholic Federation of Clinical Chemistry (IFCC). Approved
and viral hepatitis, mitochondrial AST is also released recommendation on IFCC methods for the
into the blood, and AST activity is usually higher than measurement of catalytic concentrations of enzymes.
ALT. The ratio of AST to alanine aminotransferase Part 3. IFCC method for alanine aminotransferase. J
(ALT), sometimes called the De Ritis Ratio, is often used Clin Chem Clin Biochem 1986; 24: 481-95.
to evaluate alcoholic liver disease [19] and severity of 9 Heins M, Heil W, Withold W. Storage of serum or
liver disease in viral hepatitis [20,21]. This ratio is only whole blood samples? Effects of time and
pertinent in isolated liver disease when comorbidities that temperature on 22 serum analytes. Eur J Clin Chem
increase AST are not present. Clin Biochem 1995; 33: 231-238.
10 DiMagno EP, Corle D, O'Brien JF, Masnyk IJ, Go
Alanine Aminotransferase Performance Goals VL, Aamodt R. Effect of long-term freezer storage,
The current target for total error in ALT measurements is thawing, and refreezing on selected constituents of
< 20% (CLIA). Biological variation data suggest that in serum. Mayo Clin Proc 1989; 64: 1226-1234.
patients with stable ALT activities, total error of < 30% is 11 Ono T, Kitaguchi K, Takehara M, Shiiba M, Hayami
required for optimum use [22]. The interlaboratory K. Serum-constituents analyses: effect of duration
coefficients of variation listed for various ALT methods and temperature of storage of clotted blood. Clin
in the 2007 College of American Pathologists Participant Chem 1981; 27: 35-38.
Summary Report (C-B) are all < 5%, indicating that this 12 McEnroe RJ, Burritt MF, Powers DM, Rheinheimer
target is easily met using current ALT methods. DW, Wallace BH. Eds. CLSI: Interference Testing in
Clinical Chemistry, Approved Guielines Second
References Edition CLSI document EP7-A2, 2005 Accessed 19
1 Reitman S, Frankel S. A colorimetric method for Jan 2009
the determination of serum glutamic oxalacetic 13 Karlsen RL, Kristiansen G, Solberg JH. Effects of
and glutamic pyruvic transaminases. Am J Clin metronidazole (Flagyl) on the determination of serum
Pathol 1957; 28: 56-63. ASAT on the SMA 12/60 Auto Analyser. Scand J
2 Wrblewski F, LaDue JS. Serum glutamic- Clin Lab Invest 1983; 43: 175-177.
pyruvic transaminase in cardiac and hepatic 14 Stromme JH, Rustad P, Steensland H, Theodorsen L,
disease. Proc Soc Exp Biol Med 1956; 91: 569- Urdal P. Reference intervals for eight enzymes in
571. blood of adult females and males measured in
3 Butler TJ, Klotzsch SG, Osberg IM. Alanine am- accordance with the International Federation of
inotransferase, ALT provisional. In Faulkner Clinical Chemistry reference system at 37oC: part of
WR, Meites, S, editors: Selected methods of the Nordic Reference Interval Project. Scand J Clin
clinical chemistry. Washington D.C.: American Lab Invest 2004; 64: 371-384.
Association for Clinical Chemistry; 1982. p. 69- 15 Soldin SJ, Hicks JM, editors. Pediatric reference
73. ranges. Washington, D.C.: AACC Press: 1995.
4 Wilkinson JH, Baron DN, Moss DW, Walker 16 Cordoba J, O'Riordan K, Dupuis J, Borensztajin J,
PG. Standardization of clinical enzyme assays: a Blei AT. Diurnal variation of serum alanine
reference method for aspartate and alanine transaminase activity in chronic liver disease.
transaminases. J Clin Pathol 1972; 25: 940-944. Hepatol 1998; 28: 1724-1725.
5 Committee on Enzymes of the Scandinavian 17 Dufour DR. Laboratory guidelines for screening,
Society for Clinical Chemistry and Clinical diagnosis and monitoring of hepatic injury.
Pathology. Recommended methods for the Washington, D.C.: National Academy of Clinical
determination of four enzymes in blood. Scand J Biochemistry; 2000.
Clin Lab Invest 1974; 33: 291-305. 18 Price CP, Alberti KGMM. Biochemical Assessment
6 Enzyme Commission of the German Society for of Liver Function. In: Wright R, Alberti KGMM,
Clinical Chemistry. Recommendations of the Karran S, Millward-Sadler GH, editors. Liver and
German Society for Clinical Chemistry. Z Klin biliary diseasepathophysiology, diagnosis, man-
Chem Klin Biochem 1972; 10: 281-91. agement. London: W.B. Saunders; 1979. p. 381-416.
7 Schumann G, Bonora R, Ceriotti F, Ferard G, 19 Majhi S, Baral N, Lamsal M, Mehta KD. De Ritis
Ferrero CA, Franck PFH, et al. IFCC primary ratio as diagnostic marker of alcoholic liver disease.
reference procedures for measurement of Nepal Med Coll J 2006; 8: 40-42.
catalytic activity concentrations of enzymes at 20 Giannini E, Risso D, Botta F, Chiarbonello B, Fasoli
37C. International Federation of Clinical A, Malfatti F, et al. Validity and clinical utility of the
Chemistry and Laboratory Medicine. Part 4. aspartate aminotransferase-alanine aminotransferase
Reference procedure for the measurement of ratio in assessing disease severity and prognosis in
49
patients with hepatitis C virus-related chronic chronic hepatitis C patients in clinical practice. J Clin
liver disease. Arch Intern Med 2003; 163: 218- Gastroenterol 2006; 40: 521-527.
224. 22 Fraser CG. Biological variation: from principles to
21 Giannini EG, Zaman A, Ceppa P, Mastracci L, practice. Washington, DC: AACC Press; 2001, p.
Risso D, Testa R. A simple approach to 140.
noninvasively identifying significant fibrosis in
Tables
Table 1: Methods of Alanine Aminotransferase (ALT) Analysis
Method 1: Dinitrophenylhydrazine coupling (colorimetric) (Reitman and Frankel [1]); quantitative
Principle of analysis: click here
Comments: Serum
Method 2: Enzymatic (ultraviolet monitoring) (Wrblewski and LaDue [2]); quantitative
Principle of analysis*: UV monitoring of NADH disappearance at 340 nm:
Usage: Serum; most frequently employed procedure
*Ala, Alanine; -KG, -ketoglutarate; Gl, glutamate; Lac, lactate; NAD , nicotinamide adenine dinucleotide;
+
NADH, reduced nicotinamide adenine dinucleotide; Pyr, pyruvate.
Table 2: Conditions of the 2002 IFCC ALT Reference Method
Component/Condition Concentration/Value
L-Alanine (mmol/L) 500
2-Oxoglutarate (mmol/L) 15
Buffer & concentration (mmol/L) Tris 100
pH 7.15
Pyridoxal phosphate (mmol/L) 0.1
NADH (mmol/L) 0.18
LDH (U/L) 1700
Volume fraction (v/v) 0.0833
Temperature (C) 37.0
Wave length (nm) 339
Band width (nm) 2
Light path (mm) 10
Incubation time (s) 300
Delay time (s) 90
Measurement interval (s) 180
Readings (measurement points) 6
Table 3: Tissue Distribution of ALT in Normal Human Adult Tissues

Tissue U 10-3 g of Wet Ratio of Activity in
Tissue Homogenate Tissue to That in Serum
Heart 7.1 444
Liver 44 2750
Skeletal muscle 4.8 300
Kidney 19 1188
Pancreas 2 125
Spleen 1.2 75
Lung 0.7 44
Serum 0.016 1
From Wrblewski F: Adv Clin Chem 1:313-351, 1958.
50
Figures
Figure 1: Amino transfer catalyzed by ALT.
Figure 2: Reaction of pyruvate with dinitrophenylhydrazine, as in Method 1.
Figure 3: Conversion of pyruvate to lactate with lactate dehydrogenase, as in Method 2.

51
Procedure: Kinetic Analysis of ALT [3] phosphate (Solution 3), 0.2 mL of NADH (solution
Principle 4), and 0.1 mL of the enzymes (Solution 5). Mix
The rate of the reaction in which an amino thoroughly, and store in a dark bottle. Stable for 1
group is transferred from L-alanine to 2-oxoglutarate day at 2C to 8C.
to form pyruvate and L-glutamate catalyzed by the
action of ALT is monitored by use of an indicator 7. Start reagent solution. Dissolve 407 mg of
reaction. The pyruvate formed is converted to lactate, 2-oxoglutaric acid, disodium salt in 10.0 mL of
and monitoring the absorbance change at 339 nm of distilled water. Stable for 1 week at 2 to 8 C.
the consumed NADH follows the reaction. Assay
Reagents Equipment: Spectrophotometer with 2nm
band pass at 339 nm with a constant temperature
1. Tris, L-alanine buffer (121.1 mmol/L Tris, cuvette capable of maintaining a constant
630 mmol/L L-alanine, pH 7.15). Dissolve 1.47 g of temperature with less than 0.1C fluctuation. A
tris(hydroxymethyl)aminomethane and 5.61 g of L- recording spectrophotometer is preferable.
alanine (free acid) in 80 mL of distilled water. Adjust 1. Add 2 mL of reaction solution and 0.2 mL
to pH 7.15 at 37C with 1 mol/L HCl (approximately of serum to the cuvette.
8.0 mL). Allow the solution to cool to the calibration 2. Mix and allow to stand for 5 minutes.
temperature, and bring to a volume of 100 mL in a 3. Add 0.2 mL of start reagent solution.
volumetric flask. This is Solution 1, stable for 3 5. Mix, wait 90 s, and record the change in
months at 2C to 8C. Check for bacterial growth. absorbance for an additional 180 s. If no
recorder is available, record at least every 30
2. Tris/hydrochloric acid buffer (121.2 s.
mmol/L Tris, pH 7.15). Dissolve 1.47 g of 6. A reagent blank is run by replacing the
tris(hydroxymethyl)aminomethane in 80 mL of sample with 9 g/L sodium chloride in step 1.
distilled water. Adjust to pH 7.15 at 37C with 1 7. A sample blank should be checked by
mol/L HCl. Allow the solution to cool to the replacing the start reagent in step 3 with 9
calibration temperature, and bring to volume of 100 g/L sodium chloride. The sample blank is
mL in a volumetric flask. This is Solution 2, stable not included in the calculation of AST
for 3 months at 2C to 8C. Check weekly for activity, but a rate > 1% of the ALT activity
bacterial growth. indicates that the material is not suitable as a
calibrator. For the reagent blank of the
3. Pyridoxal phosphate solution (6.3 mmol/L sample blank, replace both the start reagent
pyridoxal phosphate). Dissolve 16.7 mg of and the sample with 9 g/L sodium chloride.
pyridoxal phosphate in Solution 2, and bring to 10 Calculations
mL volume. This is Solution 3. Stable for 1 week at If the change in absorbance per second
2C to 8C when stored in a dark bottle. (A/t) is greater than 0.0025 per second, dilute the
sample 5- to 10-fold with 9 g/L NaCl, and repeat the
4. Reduced nicotinamide adenine measurement.
dinucleotide (11.34 mmol/L NADH). Dissolve 16.1 Correct for the blank reaction using the
mg of the disodium salt of NADH (or an equivalent following schema:
amount correcting for water of hydration) in 2.0 mL
of Solution 2. This is Solution 4, stable for 1 week at (A/t)A = Measured reaction rate with serum in
2C to 8C when stored in a dark bottle. step 1
(A/t)B = Measured reaction rate with NaCl in
Diluent for reagent enzymes. Dissolve 1.2 g bovine place of serum in step 1
serum albumin and 0.9 g NaCl in 100 mL of water.
Stable at least 1 month at 2C to 8 C. A/tALT = (A/t)A (A/t)B and
5. Lactate dehydrogenase (3.57 mkat/L or ALT activity = 1905 x A/tALT kat/L

214,000 U/L). Dilute the enzyme in Diluent for
Reagent Enzymes. This is Solution 5, stable for at ALT activity in kat/L can be converted to U/L by
least 2 days at 4C. multiplying by 60.
6. Reaction solution. Mix 10 mL of Tris-
aspartate (Solution 1) with 0.2 mL of pyridoxal
52
Albumin
Albumin
Kee Cheung
Name: Albumin
Clinical significance: Refer to Chapter 31, Liver Function, in the 5th Edition of Clinical Chemistry:
Theory, Analysis, Correlation
Molecular weight: 66,248 D
Merck Index: 203
Half-life 15-19 days
Chemical class: Protein
Principles of Analysis and Current Usage automated [6]. However, tryptophan-content methods
Albumin is the most abundant circulating plasma protein have never come into common use because of the ease
i and specificity of the dye-binding methods for
(40 to 60 % of the total) . It has diverse functions, playing
important roles in the maintenance of colloid osmotic albumin.
pressure of blood, in transport of various ions, acids, and
hormones, and in nutrition. It is a globular protein with a Serum albumin can be quantitated by electrophoretic
molecular weight of approximately 66,000 D and is unique techniques (Table 1, Method 3). The major classes of
among major plasma proteins in containing no serum protein are separated by serum protein
carbohydrate in its single polypeptide chain of 580 amino electrophoresis on a medium such as agarose. The
acids. It has a relatively low content of tryptophan and is separated fractions are stained, and the percentage of
an anion at pH 7.4 [1]. These properties have been each fraction present in the samples is determined by
exploited in the estimation of albumin in body fluids such densitometric analysis. The concentration of albumin
as plasma, CSF, interstitial fluid, urine, and amniotic fluid. is calculated by multiplication of the concentration of
total protein in the sample by the percentage of the
The earliest techniques for albumin analysis were based on albumin band. This method of albumin determination
acid and salt precipitation of the protein (Table 1, Method is labor intensive. In addition, there is no dye that has
1) [2]. After precipitation with ammonium sulfate, the been demonstrated to bind to all serum proteins
albumin in the supernatant was measured by total nitrogen equally or to have a binding affinity that is linear with
analysis or by biuret reaction. Modern salt fractionation the concentration of all serum proteins [7]. As a result,
techniques are cumbersome because of the many steps electrophoretic procedures tend to overestimate
required in the procedure and are not very specific because albumin, which is the best binder of the staining dyes.
of protein interactions [3]. Since they are not easily Another problem relates to lateral diffusion of
automated, the precipitation methods are of historical relatively small proteins, including albumin, on
interest only. support media.
The measurement of albumin can also be performed by a Immunological methods (Table 1, Methods 4a and 4b)
direct determination of globulin based on tryptophan used for serum albumin quantitation include radial
content and calculation of the albumin content by immunodiffusion (RID) and electroimmunodiffusion
subtraction of globulin from total protein (Table 1, Method (EID), in which albumin either passively diffuses
2). The tryptophan content of human serum albumin is (RID) or is electrophoresed (EID) onto a stationary
0.2%, compared with 2% to 3% for the various globulins phase such as agarose that contains antibodies to
[4]. Such a large difference has been used as a means for albumin. The precipitin lines formed by the reaction
albumin determination. In a one-reagent system, glyoxylic between albumin and the antibody can be fixed and
stained. In RID, the diameter (or square of the
acid in the presence of Ca2+ in an acid medium condenses
diameter) of the precipitin ring formed is proportional
with the tryptophan residues in globulins to produce a
to the albumin concentration. For EID, the height of
purple color that is measured at 540 nm [5]. The method
the rocket precipitin line is related to albumin
needs to be standardized with serum to compensate for the
concentration.
albumin interference in the reaction. In this method, free
tryptophan does not interfere. This method has been
The reaction between albumin and an anti-albumin
i
antibody can be monitored by turbidimetric or
Albumin nephelometric means (Table 1, Methods 4c and 4d).
Previous and current authors of this method: The antibody-albumin complexes that form either
First edition: Steven Gendler block (turbidimetry) or scatter (nephelometry) incident
Methods edition: Steven Gendler light, and this can be related to the albumin
Second edition: Steven Gendler concentration. Automated nephelometric and
Third edition: Kee Cheung, Peter E. Hickman turbidimetric techniques have been adapted for
Fourth edition: Not updated automated analysis as well [7,8]. However, few
Fifth edition: Kee Cheung
53
Albumin
laboratories are currently reporting the use of [23]. Details of urine albumin methods are covered in
immunological techniques for the quantitation of albumin a separate methods section.
in serum, although they are considered to be the most
specific methods. Electrophoretic and immunochemical Reference and Preferred Methods
methods are generally used for body fluids other than The IFCC Committee on plasma proteins (C-PP)
serum and plasma, in which albumin concentration is currently recommends optimized immuno-
normally low (e.g., urine and cerebrospinal fluid [CSF]). turbidimetry/immunonephelometry as the reference
method for albumin [24], and the reference material is
The most widely used methods for the analysis of serum BCR-470, serum proteins.
albumin are dye-binding procedures (Table 1, Method 5).
Albumin has the ability to bind a wide variety of organic The BCG and BCP methods are currently the
anions, including complex dye molecules. Dye-binding preferred methods for serum and plasma albumin
techniques are based on a shift in the absorption maximum determination in clinical laboratories. The BCG
of the dye when bound to albumin. The shift in the method is also adapted in the dry chemistry
absorption maximum allows the resulting color to be measurement of serum albumin. Multilayered
measured in the presence of excess dye, which in concert analytical element is coated on a clear polyester slide,
with the high-binding affinity of albumin allows all of the and the amount of colored complex formed is
albumin molecules to take part in the reaction. A wide measured by reflectance spectrophotometry.
variety of dyes has been employed for the measurement of
albumin, including methyl orange, 2-(4- The specificity of the BCG method is relatively good,
hydroxyazobenzene) benzoic acid (HABA), bromcresol as shown by correlation with salt fractionation when
green (3,3,5-tetrabromo-m-cresolsulfonphthalein, or sera from normal patients were analyzed [25].
BCG) and bromcresol purple (5,5-dibromo-o-
cresolsulfonphthalein, or BCP). BCG and BCP are most Probably the most promising adaptation of the BCG
commonly used nowadays; the others have fallen into reaction for albumin analysis utilizes fast reaction
disuse because of undesirable interferences. Because of the readings. Gustafsson [26] reported that measuring the
nonspecificity of the dye-binding techniques, it had been absorbance of the BCG-protein complex at 629 nm at
recommended that these methods should be used for a time shortly after mixing improves the specificity of
screening purposes only [9]. However, subsequent the assay. Interference by other proteins such as
improvements in the reagent composition and assay ceruloplasmin and orosomucoid becomes significant at
conditions have allowed the BCG and BCP methods to times greater than 5 minutes. When BCG was
become routine for the assay of albumin in serum and compared to EID (x axis), a regression equation of y =
plasma. Their simplicity and low cost make these two dye- 0.98x + 1.83 was obtained.
binding assays the predominant means for serum and
plasma albumin determination in modern laboratories. Extrapolation does not appear to be necessary to
Dye-binding procedures, however, have very limited improve specificity. Non-extrapolated readings taken
application in the determination of albumin in urine or CSF at 10 to 30 seconds on a spectrophotometer [27,28], at
because of low protein concentration and high 10 s on a reaction-rate analyzer or discrete multi-
concentrations of interfering substances. channel system [29,30], at 0.5 to 6 s on a centrifugal
analyzer [31,32], or at 20 to 30 s on a continuous-flow
According to the 2007 College of American Pathologists analyzer [33-35] improve the specificity of the BCG
Participant Summary Report, the BCG and BCP reaction. The concentration of BCG is a factor in the
procedures are used in approximately equal proportions by linearity of the reaction and is optimal at 60 mol/L
laboratories (43% and 57%, respectively). A handful of [30].
laboratories report using nephelometric measurement of
albumin. New methods of albumin determination include The BCP method has been gaining popularity as a
spectrofluorimetry [10], near-infrared spectroscopy [11], more specific method. It is considered to be superior
and total internal reflected resonance light scattering [12]. to BCG, which overestimates serum albumin,
particularly in patients with low concentrations of
CSF albumin is readily measured directly by albumin who may also have increased globulins or
immunoturbidimetric methods of EID, RID, or monoclonal bands [36]. It seems to have overcome
immunonephelometry. All these procedures are available most disadvantages seen in other dye-binding
from commercial vendors and function in the range of methods. It compares well with EID (x axis), giving a
about 200 mg/L. Immunonephelometric methods have linear regression equation of y = 0.95x + 1.72 when
been automated on a variety of nephelometers. analysis is done on a population selected to provide a
maximum range of albumin, bilirubin, and globulin
Estimation of urine albumin is important in assessing concentrations. Sera with lipemia, hemolysis, and
developing renal dysfunction. Urine albumin is present in drugs were also part of this tested population. BCP
much lower concentration (<30 mg/L) than plasma does not bind to non-albumin proteins, although
albumin, and more sensitive techniques are required for its ceruloplasmin and orosomucoid have not been tested
measurement, such as radioimmunoassay (RIA) [13-15], specifically. This specificity is seen without the need
enzyme immunoassay [16-19], and immunoturbidimetric of fast reaction readings. Human serum albumin Cohn
[20,21], immunonephelometric [22], and RID techniques fraction V must be used as a standard, since bovine
54
Albumin
and equine serum albumindye complexes do not absorb Albumin has been found to be lower in the last two
as strongly at 600 nm as human albumindye complexes trimesters of pregnancy [47]. Long-term fasting or
[37]. The BCP dye has been adapted for use in continuous- food restriction can cause a decrease in serum
flow analyzers, centrifugal analyzers, and random access albumin.
analyzers [37,38]. Interpretation
Albumin is synthesized by the hepatic parenchymal
Disadvantages of the BCP methods include (1) the cells at a rate that is dependent on colloidal osmotic
technique reportedly underestimates albumin in cases of pressure and dietary protein intake [48]. The rate of
renal insufficiency [39] and obstructive jaundice [40], (2) albumin synthesis is also subject to feedback
BCPalbumin dye complex has a molar absorptivity about regulation determined by the plasma albumin
half of that of BCGalbumin complex, thus greatly concentration. The half-life of albumin has been
reducing the sensitivity and precision of the BCP methods; estimated at 15 to 19 days. Traces of albumin can be
and (3) the BCP methods cannot be conveniently found in almost all extracellular body fluids. Little is
standardized and monitored with nonhuman fluids, since lost from the body by excretion. It is catabolized in
BCP does not bind strongly to bovine albumin. various tissues, where it is taken up by cells by
pinocytosis. Its constituent amino acids are released by
Methods for urine albumin are covered in a separate intracellular proteolysis and returned to the body pool.
chapter.
Albumin has a vast capacity for ligand binding, owing
Specimen to the large number of charges on each molecule, as
Serum is the specimen of choice, but heparinized plasma well as the very large number of molecules available.
can also be used if precautions are taken to prevent heparin This binding property of albumin is related to its
interferences [41]. Fasting is not required, although it may function of transporting molecules such as bilirubin
be desirable, since marked lipemia interferes in the BCG and free fatty acids [49]. Albumin also binds many
assay. Venostasis should be avoided when collecting hormones such as thyroxine, triiodothyronine, cortisol,
samples; hemoconcentration increases the apparent and aldosterone. It thus acts as a reservoir in which
concentrations of albumin and other plasma proteins. these physiologically potent compounds are stored in
an active form but from which they can be readily
Interferences mobilized. This binding property extends to xenobiotic
Bilirubin, slight to moderate lipemia, and salicylate do not compounds such as drugs. Albumin binds salicylate,
interfere with BCG methods. Hemoglobin decreases the valproate, phenytoin, warfarin, phenylbutazone,
apparent albumin concentration by 1 g/L for each 100 g/L clofibrate, and many other drugs. Therefore, low
added. Blanking does not correct this interference, and the albumin values may explain drug toxicity at
negative bias is therefore caused by interference with the apparently low concentrations when only the total
dye binding rather than hemoglobin color. For the BCP concentration of drug in serum is measured. In
method, a blank correction is required on icteric sera and hypoalbuminemic sera, the free (active) form of the
on grossly hemolyzed and grossly lipemic sera to correct drug may be far greater than in sera with normal
for an underestimation of albumin caused by these agents. albumin levels because of the large difference in
Heparin causes a positive interference with BCP and BCG albumin-binding capacity between the different sera.
methods. This interference can be eliminated by the The increased amount of free drug may thus cause a
addition of hexadimethrine bromide to a concentration of toxic response, a phenomenon observed in
50 mg/L in the BCP reagent [42]. hypoalbuminemic patients on dialysis who are treated
with phenytoin. Toxic effects of phenytoin may be
observed at serum concentrations considered
Albumin Reference Interval therapeutic. Some 40% of serum calcium is also bound
The existing interim consensus reference interval for to albumin. A summary of the major causes of
albumin in serum on the basis of the IFCC/BCR/CAP changes in albumin concentration in serum are listed
reference material CRM 470 and valid for adult Caucasian in Table 2.
persons and adolescents is 35 to 52 g/L [43]. Albumin
concentrations reach these levels around 20 to 30 weeks of Hypoalbuminemia may also result in underestimation
gestation and remain relatively constant until at least 20 of the true anion gap in the interpretation of acid-base
years of age and then slowly decrease with age. abnormalities [50].
Concentrations are, however, significantly lower in
preterm infants [44]. In addition to its role as a binding and transport
molecule, albumin plays a large role in nutritional
Serum albumin values are lower in the recumbent (lying status. It is argued that the protein is constructed so
down) patient compared to ambulant ones. Some that it is readily metabolized and contains all the
laboratories take note of the difference between inpatients essential amino acids. In starvation, plasma
and outpatients. For example, Walmsley and White use a concentration of albumin decreases considerably more
reference interval of 30 to 50 g/L for inpatients and 37 to than the levels of gamma globulins. Very low
52 g/L for outpatients [45]. A difference of 5 g/L between concentrations are observed in malnutrition,
recumbent and ambulant values has been used [46]. particularly in kwashiorkor (protein- and calorie-
deficient diets).
55
Albumin
imprecision of no greater than 1.6% and a total error

Another major role of albumin is that of maintaining of no greater than 3.9% [55]. The two dye-binding
colloid osmotic pressure. It accounts for 75% of the colloid methods for albumin determination are therefore
osmotic pressure of plasma. When albumin levels are within the desired performance criteria as defined by
significantly decreased to about 20 g/L, edema is often CLIA-88 and are within (BCP) or just marginally
observed because of the consequent decrease of colloid outside the biological variation limits (BCG).
osmotic pressure. Decreases in serum albumin may occur
in response to many pathological events, thus changes in References
serum albumin are very nonspecific. 1 Peters T Jr. Serum albumin. In: Putnam FW,
editor. The plasma proteins, vol. 1. New
Plasma albumin measurement, although important for York: Academic Press; 1975. p. 133-181.
management and follow-up, has very little value in clinical 2 Peters T Jr. Serum albumin. Adv Clin Chem
diagnosis. Hyperalbuminemia is usually attributable to 1970; 13: 37-111.
dehydration or hemoconcentration. Hypoalbuminemia is 3 Henry RJ. Proteins. In: Henry RJ. Clinical
usually the result of (1) hemodilution, (2) a rate of chemistry: principles and technics. New
synthesis less than the rate of albumin loss, (3) diseases York: Harper & Row Publishers; 1964. p.
that cause a large albumin loss from urine, skin, or 199-226.
intestine, or (4) increased catabolism observed in fevers, 4 Block RJ, Weiss KW, Carroll DB, editors.
untreated diabetes mellitus, and hyperthyroidism. Amino acid handbook. Springfield IL:
Charles C Thomas; 1956. p. 252.
Decreased synthesis may be because of malnutrition, 5 Goldenberg H, Drewes PA. Direct
malabsorption, or an inability of the liver to synthesize photometric determination of globulin in
albumin because of diseases such as acute or chronic serum. Clin Chem 1971; 17: 358-362.
hepatitis. Low plasma albumin concentration may be a 6 Savory J, Heintges MG, Sobel RE.
result of large losses in the case of diseases such as Automated procedure for simultaneously
nephrotic syndrome, protein-losing enteropathy, exudative measuring total globulin and total protein in
skin lesions, or burns. Burns in particular may be serum. Clin Chem 1971; 17: 301-306.
associated with severe albumin loss. 7 Blom M, Hjrne N. Immunochemical
determination of serum albumin with a
In chronic liver disease, albumin is a good indicator of centrifugal analyser. Clin Chem 1975; 21:
prognosis and has been used in the Child-Pugh scoring 195-198.
system [51]. Currently the MELD (model for end-stage 8 Dito WR. Rapid immunonephelometric
liver disease) score [52] has taken over, and that does not quantitation of eleven serum proteins by
include albumin. This has been challenged recently by centrifugal fast analyser. Am J Clin Pathol
Brown et al. [53], who argue that the Child-Pugh score is a 1979; 71: 301-308.
better predictor than the MELD score of survival at 9 International Federation of Clinical
specific time points, since albumin level is the most useful Chemistry, Newsletter 1976 Mar 5;13.
predictor of survival. 10 Jiang CQ, Luo L. Spectrofluorimetric
determination of human serum albumin using
There are more than 20 inherited variants of albumin a doxycycline-europium probe. Anal Chim
which are not associated with disease. This includes Acta 2004; 506: 171-175.
bisalbuminemias in which two chemical types of albumin 11 Kasemsumran SY, Du YP, Murayama K,
are present. Congenital absence of albumin, or Huehne M, Ozaki Y. Near-infrared
analbuminemia, is asymptomatic except for occasional spectroscopic determination of human serum
slight edema. albumin, gamma-globulin, and glucose in a
control serum solution with searching
Albumin Performance Goals combination moving window partial least
Survey data from the 2007 College of American squares. Anal Chim Acta 2004; 512: 223-230.
Pathologists Participant Summary Report shows 12 Feng P, Huang CZ, Li YF. Direct
imprecision values (% coefficient of variation) for quantification of human serum albumin in
measurement of albumin at a concentration of 3.5 g/dL to human blood serum without separation of
range from approximately 2.0% to 7.5% for the BCG gamma-globulin by the total internal reflected
procedure and from approximately 1.5% to 2.5% for resonance light scattering of thorium-sodium
methods employing BCP. dodecylbenzene sulfonate at
water/tetrachloromethane interface. Anal
Acceptable performance criteria (Clinical Laboratory Biochem 2002; 308: 83-89.
Improvement Amendments [CLIA]-88) for measurement 13 Keen H, Chlouverakis C. An immunoassay
of albumin require that laboratories be accurate to within method for urinary albumin at low
10% of the peer group mean. The intraindividual variation concentrations. Lancet 1983; 2: 913-916.
of albumin in blood in healthy adults studied for up to 5 14 Woo J, Floyd M, Cannon DC, Kahan B.
months has been determined to be less than 3.0% [54]. Radioimmunoassay for urinary albumin. Clin
Desirable specifications for analytical imprecision derived Chem 1978; 24: 1464-1467.
from studies of biological variation indicate an assay
56
Albumin
15 Jury DR, Speed JF, Dunn RJ. Urinary albumin 32 King SW, Cross RE, Savory J. Improved
radioimmunoassay using a solid phase second specificity of the bromocresol green (BCG)
antibody and solid phase iodination. Clin Chim dye method for serum albumin using an early
Acta 1985; 148: 63-67. (6 seconds) absorbance reading. Clin Chem
16 Fielding BA, Price DA, Houlton CA. Enzyme 1977; 23: 1136.
immunoassay for urinary albumin. Clin Chem 33 Cederblad G, Hickey BE, Hollender A,
1983; 29: 355-357. Akerlund G. Improved continuous-flow
17 Puri A, Casburn-Budd R, Eisen V, Slater JDH. (SMAC) determination of serum albumin.
Simpler measurement of albumin in urine or Clin Chem 1978; 24: 1191-1193.
plasma. Clin Chem 1985; 31: 1241-1242 34 Plesher CJ. Continuous-flow analysis of
serum albumin by use of the immediate
18 Feldt-Rasmussen B, Dinesen B, Deekert M.
reaction with bromcresol green. Clin Chem
Enzyme immunoassay: an improved
1978; 24: 2036-2039.
determination of urinary albumin in diabetes with
35 von Schenck H, Rebel K. Improved
incipient nephropathy. Scan J Clin Lab Invest.
continuous-flow determination of albumin
1985; 45: 539-544.
with bromcresol green. Clin Chem 1982; 28:
19 Townsend JC. A competitive immunoenzymatic
1408-1409.
assay for albumin in urine. Clin Chem 1986; 32:
36 Duly EB, Grimason S, Grimason P, Barnes
1372-1374.
G, Trinick TR. Measurement of serum
20 Harmoienen A, Vuorinen P, Jokela H.
albumin by capillary zone electrophoresis,
Turbidimetric measurement of microalbuminuria.
bromocresol green, bromocresol purple, and
Clin Chim Acta 1987; 166: 85-89.
immunoassay methods. J Clin Pathol 2003;
21 Lloyd DR, Hindle EJ, Marples J, Gatt JA. Urinary
56: 780-781.
albumin measurement by immunoturbidimetry.
37 Pinnel AE, Northam BE. New automated
Ann Clin Biochem 1987; 24: 209-210.
dye-binding method for serum albumin
22 Stamp RJ. Measurement of albumin in urine by
determination with bromcresol purple. Clin
end-point immunonephelometry. Ann Clin
Chem 1978; 24: 80-86.
Biochem 1988; 25: 442-443.
38 Haythorn P, Sheehan M. Improved
23 Watts GF, Bennett JE, Rowe DJ, Morris RW,
centrifugal analyzer assay of albumin. Clin
Gatling W, Shaw KM, Polak A. Assessment of
Chem 1979; 25: 194.
immunological methods for determining low
39 Maguire GA, Price CP. Bromocresol purple
concentrations of albumin in urine. Clin Chem
method for serum albumin gives falsely low
1986; 32: 1544-1548.
values in patients with renal insufficiency.
24 Blirup-Jensen S. Protein standardization III:
Clin Chim Acta 1986; 155: 83-88.
method optimization. Basic
40 Bush V, Reed RG. Bromocresol purple dye-
principles for quantitative determination of human
binding methods underestimate albumin that
serum proteins on
is carrying covalently bound bilirubin. Clin
automated instruments based on turbidimetry or
Chem 1987; 33: 821-823.
nephelometry. Clin Chem
41 Hallbach J, Hoffmann GE, Guder WG.
Lab Med 2001; 39:1098-1109.
Overestimation of albumin in heparinized
25 Doumas BT, Biggs HG. Determination of serum
plasma. Clin Chem 1991; 37: 566-568.
albumin. Stand Methods Clin Chem 1972; 7: 175-
42 Duggan J, Duggan PF. Albumin by
188.
bromcresol green: a case of laboratory
26 Gustafsson JEC. Improved specificity of serum
conservatism. Clin Chem 1982; 28: 1407-
albumin determination and estimation of acute
1408.
phase reactants by use of the bromocresol green
43 Dati F, Johnson AM, Whicher JT. The
reaction. Clin Chem 1976; 22: 616-622.
existing interim consensus reference ranges
27 Webster D. The immediate reaction between
and the future approach. Clin Chem Lab Med
bromocresol green and serum as a measure of
2001: 39: 1134-1136.
albumin content. Clin Chem 1977; 23: 663-665.
44 Cartlidge PHT, Rutter N. Serum albumin
28 Corcoran RM, Durnan SM. Albumin
concentrations and oedema in the newborn.
determination by a modified bromocresol green
Arch Dis Child 1986; 61: 657-660.
method. Clin Chem 1977; 23: 765-766.
45 Walmsley RN, White GH. A guide to
29 Gustafsson JEC. Automated serum albumin
diagnostic clinical chemistry. Victoria,
determination of plasma albumin by the
Australia: Blackwell Scientific Publications;
bromocresol green dye-binding method. Clin
1983. p. 242-247.
Chem 1978; 24: 369-373.
46 Jacobs DS, Kasten BL Jr, Demott WR,
30 Robertson WS. Optimizing determination of
Wolfsen WL. Laboratory test handbook with
plasma albumin by the bromocresol green dye-
DRG index. St Louis: Mosby Lexi-comp;
binding method. Clin Chem 1981; 27: 144-146.
1984. p. 32-33.
31 Izquierdo JM. Immediate reaction (0.5s) of serum
47 Tietz NW, editor. Clinical guide to laboratory
albumin with BCG in Cobas BioAnalyzer. Clin
tests. 3rd ed. Philadelphia: W.B. Saunders;
Chem 1977; 23: 1136.
1995.
57
Albumin
48 Rothschild MA, Oratz M, Schreiber SS. Albumin predict survival in patients with end-stage
synthesis. 1. N Engl J Med 1972; 286: 748-57. liver disease. Hepatology 2001; 33: 464-470.
AND 53 Brown DB, Fundakowski CE, Lisker-
Rothschild MA, Oratz M, Schreiber SS. Albumin Melman M, Crippin JS, Pilgram TK,
synthesis (second of two parts). N Engl J Med Chapman W et al. Comparison of MELD and
1972; 286: 816-21. Child-Pugh scores to predict survival after
49 Natelson S, Natelson EA. Principles of applied chemoembolization for hepatocellular
clinical chemistry: chemical background and carcinoma. J Vas Interventional Radiology
medical applications. In: Plasma Proteins in 2004; 15: 1209-1218.
Nutrition and Transport, vol 3. New York: 54 Fraser CG, Cummings ST, Wilkinson SP,
Plenum Press; 1980. p. 43-74. Neville RG, Knox JD, Ho O et al. Biological
50 Durward A, Mayer A, Skellett S, Taylor D, Hanna variability of 26 clinical chemistry analytes in
S, Tibby SM., Murdoch IA. Hypoalbuminaemia elderly people. Clin Chem 1989; 35: 783-786.
in critically ill children: incidence, prognosis, and 55 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
influence on the anion gap. Arch Dis Child 2003; Hernandez A, Jimenez CV et al. Current
88: 419-422. databases on biologic variation: pros, cons
51 Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni and progress, Scand. J Clin Lab Invest 1999;
MC, Williams R. Transection of the oesophagus 59: 491-500.
for bleeding oesophageal varices. Brit J Surg
1973; 60: 646-649.
52 Kamath PS, Wiesner RH, Malinchoc M, Kremers
W, Therneau TM, Kosberg CL et al. A model to
58
Albumin
Tables
Table 1: Albumin Methods Summary
Method 1: Precipitation; quantitative
a. Salt fractionation
b. Solvent fractionation
c. Acid fractionation
Principle of analysis: Changes of net charge of protein result in precipitation
Comments: Historical; still used in manufacturing albumin; manual
Method 2: Tryptophan content; quantitative
Principle of analysis: Glyoxylic acid + tryptophan in globulin Purple chromogen (Amax, 540 nm); Total
protein globulin = albumin.
Comments: Correlates well with electrophoresis but requires total protein measurement; manual and automated
Method 3: Electrophoresis; quantitative
a. Moving boundary
b. Cellulose acetate
c. Cellulose acetate with elution of peak
Principle of analysis: Albumin is separated from other proteins in electrical field; percent staining of albumin
fraction multiplied by total protein value
Comments: Very labor intensive, but if albumin is eluted for measurement, very accurate; manual and automated
Method 4: Immunochemical
a. Electroimmunoassay; quantitative
Principle of Analysis: Protein migrates in electrical field through medium containing a specific antibody
Comments: Candidate reference method; somewhat labor intensive; manual
b. Radial immunodiffusion; quantitative
Principle of analysis: Protein diffuses through medium containing specific antibody
Comments: Reference method; very long incubation time; manual
c. Turbidimetry; quantitative
Principle of analysis: Antigenantibody complexes decrease light transmission more than free antigen
Comments: Reagent cost high; manual or automated
d. Nephelometry; quantitative
Principle of analysis: Antigenantibody complexes scatter light more than free antigen
Comments: Reagent cost high; automated
e. Radioimmunoassay; quantitative
Principle of analysis: Radiolabeled albumin competes with test albumin for limited amount of antibody
Comments: Primarily used for urine
f. Enzyme immunoassay; quantitative
Principle of analysis: Sandwich assay uses antibody bound to surface and peroxidase-labeled antibody
Comments: New
Method 5: Dye binding
a. Methyl orange; quantitative
Principle of analysis: Albumin binds to dye and changes spectral profile of dye
Comments: Nonspecific for albumin; manual or automated
b. HABA (2-[4-hydroxyazobenzene]benzoic acid); quantitative
Principle of Analysis: Albumin binds to dye and changes spectral profile of dye
Comments: Specific for albumin; poor sensitivity; many drug interferences; manual or automated
c. BCG (bromcresol green); quantitative
Principle of analysis: Albumin binds to dye and changes spectral profile of dye; Amax, 628 nm
Comments: Nonspecific for albumin if absorbance reading taken after 30 s; manual or automated; most often
used method
d. BCP (bromcresol purple); quantitative
Principle of analysis: Albumin binds to dye and changes spectral profile of dye; Amax, 603 nm
Comments: Specific for albumin; low sensitivity; albumins from animal sources do not bind equivalently to
human albumin; manual or automated
Method 6: Dye binding; semiquantitative
Principle of analysis: Bromphenol blue in test strip changes color from yellow to blue in presence of albumin
Comments: Nonspecific; most sensitive with albumin; most commonly used test for urine protein
59
Albumin
Table 2: Causes of Changes in Serum Albumin Concentration

Hyperalbuminemia Hypoalbuminemia
Dehydration Pregnancy
Normal Variant Reduced synthesis
- inadequate diet
- malabsorption
- liver disease
- hereditary
Increased loss
- nephrotic syndrome
- severe bleeding
- burns
Increased metabolism
- hyperthyroidism
- sepsis
- burns
- malignancy
Table 3 Albumin Reaction Conditions Using BCG

Condition Requirement
Temperature Room temperature
pH 4.20 0.05
Final concentration of reagent components BCG: 150 mmol/L
Sodium succinate buffer: 75 mmol/L
Sodium azide: 1.5 mmol/L
Brij-35: 1.2 g/L
Fraction of sample volume 0.004
Sample Serum, heparinized plasma
Linearity 10 to 60 g/L
Reaction time 30 s
Major interferences Gross hemolysis, gross lipemia
Precision (between day) CV: 2.5%-5.9 %
Figures
A. Albumin-BCG Absorption Spectra.

Dashed line, Bromcresol green reagent versus water B. Albumin-BCP Absorption Spectra.
blank; dotted dashed line, Bromcresol green reagent and Absorption spectra of bromcresol purple reagent and
albumin versus water blank; solid line, Bromcresol albumin. Dashed line, Bromcresol purple reagent versus
green reagent and albumin versus reagent blank. water blank; Dotted dashed line, Bromcresol purple
reagent and albumin versus water blank; Solid line,
Bromcresol purple reagent and albumin versus reagent
blank.
60
Albumin
Procedure: Measurement of Albumin Using

Bromcresol Green Procedure
1. Set spectrophotometer to zero absorbance at
Principle 628 nm with water. Read absorbance of BCG
BCG complexes with albumin, resulting in the dye reagent. It should be about 0.150 A. Reset
having a spectral shift. The presence of albumin absorbance to zero again against the BCG
increases the absorbance at 628 nm, which is determined reagent.
spectrophotometrically (Albumin Table: Reaction 2. Prepare a set of tubes labeled for standard,
Conditions) control, and unknown sample.
3. Add 5.0 mL BCG reagent to each tube.
Specimen
Serum, heparinized plasma 4. Add 20 L of appropriate specimen to each
tube, one at a time, and mix without delay.
Reagents and Materials 5. Record absorbance at 628 nm at 30 3 s after
1. BCG reagent. Dissolve 0.105 g of BCG (or the addition. Observe the same time intervals
0.108 g of BCG sodium salt), 8.850 g of for all samples.
succinic acid, 0.100 g sodium azide, and 4 mL
Brij-35 (polyoxyethylene lauryl ether, 300 g/L) Calculation
in about 950 mL of distilled water. Adjust pH Albumin in the sample (g/L) = AU / AS CS
to 4.15 to 4.25 with 6 N NaOH. Make up final where Cs is the concentration of albumin in the standard
volume to 1 L with water. Store in a tightly in g/L.
closed polyethylene bottle. The reagent is stable
at room temperature for at least 6 months. Linearity: The method is linear from 10 to 60 g/L.
2. BCG blank solution. Prepare exactly as for
BCG reagent, but add no BCG. Interferences: Hyperbilirubinemia or hemolysis does not
3. Standardization. Highly purified human serum interfere. Marked lipemia causes positive interference,
albumin fraction V can be used, which has been which can be corrected as follows:
confirmed for purity by electrophoresis and
protein concentration determined by a correctly Set spectrophotometer to zero absorbance at 628 nm
standardized and well-controlled biuret method. with BCG blank solution; add 20 L of sample to 5.0
mL of BCG blank solution; read absorbance of the
Assay
serumblank mixture (Ab); subtract Ab from absorbance
Equipment: Any spectrophotometer may be used if it
has a band pass of less than 8 nm, an accurate of serumreagent mixture, and calculate with this
wavelength scale, and acceptable photometric accuracy corrected Au.
and linearity.
61
Albumin in Urine
Albumin in Urine
Graham Jones
Name: Albumin in urine

Clinical Significance: A marker of renal damage in diabetes and other conditions. Also a marker of
risk for cardiovascular disease. Refer to Chapter 30, Renal Function, in the 5th edition of Clinical
Chemistry: Theory, Analysis, Correlation.
Common name: Microalbuminuria
Measurand concentrations found in healthy individuals, often with
Human albumin is a single polypeptide chain of 585 amino limits of detection below 5 mg/L, and are able to
acids with 17 internal disulphide bonds but without separate patients with normal albumin excretion from
i those with increased excretion. In addition, these
carbohydrate side chains . Albumin in the circulation has
microheterogeneity due to structural flexibility, ligand assays have precision characteristics able to meet
binding, and other factors. The transition of albumin from biological variation criteria with claimed coefficient of
the serum to the urine has the potential to markedly variation (CV) values for total precision of < 5%. For
increase the structural variability by fragmentation, internal determination of the albumin-creatinine ratio, there is
cleavage, oxidation, or selective tubular resorption, and also the need to measure urine creatinine
multiple forms of albumin have been identified in urine concentrations. Because the presence of noncreatinine
[1,2]. This variability in albumin has the potential to lead chromogens in urine is much less than in plasma,
to standardization difficulties with assays and is known to routine creatinine assays are generally acceptable for
produce different results for individual patients using this purpose. If a timed sample is received for
different assays [3,4]. The development of assays for calculation of the albumin excretion rate, the urine
albumin fragments detectable by HPLC but not by volume must be measured using appropriate scales or
standard immunoassays, the so-called immunochemical a volumetric flask. Given the high prevalence of
non-reactive albumin, has raised the possibility of diabetes, most laboratories need a method with high
improved sensitivity for detection of early renal damage capacity to meet the clinical needs.
[2]. There is a need to identify a measurand which is the
most clinically relevant and analytically suitable to provide There are a number of point-of-care methods available
standardized assays for urine albumin [5]. The structure of for urine albumin measurement using different assay
albumin in urine and its possible effect on various assays formats and technologies. These may be
has been the subject of an extensive review [6]. semiquantitative or quantitative, and some also
measure creatinine to allow calculation of the
Principles of Analysis and Current Usage (see Table 1) albumin-creatinine ratio. Point-of-care testing is of
Assays for albumin in urine can be divided into three main particular use in the clinical setting where a rapid
categories: routine laboratory assays, point-of-care assays, return of laboratory results is difficult [7], although in
and other reference or developmental assays. Owing to the any setting, the provision of a result for use within the
issues mentioned above with regard to variability in the same medical consultation as the sample collection
structure of urine albumin, there are some systematic and can be beneficial. Since positive results can be referred
patient-specific differences between results from different to a laboratory for further analysis, a point-of-care test
assays, depending on the measurement technology and the used as a screening test should have sufficient
antibody specificity. sensitivity to avoid missing positive results. The use of
a high-quality quantitative point-of-care analyzer may
The vast majority of routine laboratories use avoid the need for referral of positive samples. Below
immunoassays to quantitate albumin in urine. These may are examples of point-of-care devices for
be structured as nephelometric or turbidimetric measurement of urine albumin.
homogenous immunoassays or heterogenous competitive The Siemens DCA Vantage analyzer (previously
or noncompetitive immunoassays and may use monoclonal known as the DCA 2000) uses a monoclonal antibody
or polyclonal antibodies. These assays are generally agglutination technique for albumin, with a
purchased from diagnostic companies and are available for simultaneous chemical creatinine assay. These tests
use on high-volume chemistry or immunoassay analyzers, are performed on a single 40 L sample; results are
as well as in manual assay formats such as enzyme-linked available in 7 minutes, using a disposable cartridge
immunosorbent assay (ELISA) or radioimmunoassay. and a portable analyzer. The system achieves good
These commercial assays generally provide sufficient precision, with CV < 5% for albumin and < 3% for
sensitivity to measure urine albumin down to creatinine, with quantitation of albumin down to 5
mg/L in a laboratory evaluation [8] with similar
performance when used in remote locations [7].
i
Albumin in Urine The Haemocue instrument is small, portable analyzer
New method which uses disposable cuvettes preloaded with
Fifth edition: Graham Jones
62
Albumin in Urine
reagents. The system uses immunoturbidimetry and performance on a routine chemistry or immunoassay
produces a result in 90 seconds from 18 L of sample. The analyzer. The required key performance characteristics
reporting range is 5 to 150 mg/L; however, precision was are described further below.
poorer than seen in laboratory methods at all Specimen
concentrations tested, with within-run CVs of between The amount of albumin in urine can be expressed in a
10% and 13% for patient samples with albumin number of different formats which require different
concentrations above 20 g/L [9,10], although overall a samples [6]. These reporting formats include the
good correlation with routine laboratory methods has been following:
achieved [10].
Albumin Excretion Rate (AER)
The Siemens Clinitek system provides semiquantitative Albumin excretion rate is commonly expressed as mg
results for both albumin concentration and the per 24 hours or micrograms per minute. The former
albumin/creatinine ratio, using a regent strip with dye- requires a 24-hour sample, whereas the latter may use
binding techniques. The strips may be read in a small a 24-hour sample or other timed period, such as an
reader device, and results higher than 20 mg/L are reported overnight collection. Both samples require close
as positive. Precision for the system cannot be easily attention to start and finish times, as well as avoiding
determined, because the results are reported in large over- and under-collection from other causes. AER is
increments. A number of interferences are listed in the considered to be the gold standard but is not generally
product information, including hematuria, soaps, dyes, and recommended for routine use because of these
some drugs, as well as high levels of urine protein. An collection difficulties.
evaluation of patient samples showed that approximately
12% of patients with laboratory urine albumin results Urine Albumin/Creatinine Ratio (ACR)
below 20 mg/L were falsely reported as elevated and 11% Expressing the albumin concentration in a spot sample
of samples with low-positive lab results (20 to 55 g/L) as a ratio to urine creatinine is a method to reduce the
were reported as negative [11]. A higher false-positive rate effect of patient hydration on the albumin
has been described in children [12], and it has been concentration. ACR is measured in a spot sample,
recommended that low-positive results be confirmed with preferably a first morning sample; a random sample is
laboratory testing [10]. acceptable, but daily activity may lead to false-
positive results. The units are mg/g creatinine or
The Roche Micral reagent strips are dipped into a urine mg/mmol creatinine. The use of creatinine to correct
sample and the resultant immunologically-mediated color for hydration also adds an influence of muscle mass to
formation visually compared with a semiquantitative chart. the result with larger people, who produce more
The method has demonstrated acceptable between-user creatinine, giving lower results for the same albumin
correlation [13].The lowest positive result is a color excretion. This is seen with the different decision
intensity associated with a nominal value of 20 mg/L. This points for males and females recommended by some
decision point has been shown to have a high false-positive bodies. The ACR is recommended by the American
rate in two studies, indicating the need for laboratory Diabetes Association (ADA) for urine albumin testing
follow-up if available [14,15]. [20]. These recommendations also indicate that results
from at least 2 out of 3 samples over a 6-month period
Measurement of urine albumin by HPLC detects different are used to confirm significant changes in albumin
fragments of albumin, compared to immunoassay as excretion status.
described above. This leads to higher results, especially in
the low range, with consequent higher detection rates for Urine Albumin Concentration (UAC)
microalbuminuria when standard decision points are used The UAC is reported as mg/L or the equivalent g/mL
[16,17]. More recent work has indicated possible co- and is measured in a spot samplelike the ACR,
elution of other proteins with albumin leading to preferably a first morning sample, but a random
overestimation of albumin by this method [18]. The sample is acceptable. Some studies have shown
differences between HPLC and immunoassay highlights minimal difference between the sensitivity of ACR
the need for agreed reference methods and materials. and UAC for increased AER, so some authors
recommend the use of UAC for general purposes,
Reference and Preferred Methods because this removes the requirement for creatinine
There are currently no reference methods or reference measurement [21].
materials for urine albumin listed on the Joint Committee
for Traceability in Laboratory Medicine (JCTLM) database No preservatives are usually required for urine
[19]. In the absence of specific reference materials for albumin collections, and manufacturers
urine albumin, most manufacturers reference their assays recommendations should be consulted if a preservative
to human serum albuminfor example, using CRM470. is required. The sample may be stored at room
temperature for up to 7 days and 1 month at 4C to
The preferred methods for routine use are immunoassays 8C, according to World Health Organization (WHO)
for urine albumin with additional measurement of urine guidelines [22]. It is possible that bacterial
creatinine to allow calculation of the albumin/creatinine contamination may affect stability at room
ratio. For many routine laboratories, this type of temperature, so earlier cooling may be preferred.
technology has the advantages of high throughput and Storage at 20C causes breakdown to fragments
63
Albumin in Urine
which are measured in some assays but not others, but this The response to the finding of an elevated urine
is not seen at 80C, and long-term storage is possible at albumin should be increased attention to risk factors,
this temperature [3,23]. with the aim of reducing the risk of further damage to
the kidneys or other organs.
Interferences
Biological causes for increased urine albumin excretion Performance Goals
other than kidney damage include fever, exercise, heart Like many other analytes in urine, the concentration of
failure, marked hyperglycemia, and hypertension [20]. albumin may vary considerably from day to day. A
Additionally, collection shortly after ejaculation may CV for within-subject biological variation of 36% is
elevate results owing to the albumin content of semen. If listed on the Biological Variation database on the
these causes are identified, repeat testing at an appropriate Westgard website [30], although a recent review has
time may be indicated to further evaluate positive results. shown marked variation in the estimates for this
parameter, with the central tertile for all studies
Homogenous immunoassays such as turbidimetry are at showing a range of 28% to 47% [6]. Thus assays with
risk of producing falsely low results due to a prozone values for total analytical CVs below 7% will meet
effect [24]. Routine procedures to identify this problem optimal precision requirements of less than a quarter
may include (1) measurement of all samples neat and in of the within-subject variation for most estimates of
dilution to confirm linear dilution, (2) measurement neat this parameter. Given that urine albumin/creatinine
and with additional albumin added to confirm complete ratios continue to provide information down to 10
recovery, or (3) testing for high total protein with a mg/g (1.1 mg/mmol), the ability to measure albumin
dipstick to identify samples where excess albumin is likely down below 5 mg/L is an advantage when measuring
[25]. dilute samples. These criteria can be met by most
laboratory-based immunoassays, including
Attention should also be given to the possibility of carry- immunoturbidimetric assays that may be run on
over effects when serum and urine are run on the same routine chemistry analyzers. By contrast, most point-
analyzer, given the > 1000-fold difference in albumin of-care analyzers, particularly the semiquantitative
concentrations in the two sample types. methods, are unable to provide good analytical
performance near upper limit of normal, although they
Reference Intervals are clearly able to identify higher levels of albumin
Decision points for interpretation of urine albumin are not within the microalbuminuria range.
based on population reference intervals but rather on
outcome-based consensus decision points. Different References
professional bodies have made slightly different 1 Candiano G, Musante L, Bruschi M, Petretto
recommendations, and Table 2 below is based on the data A, Santucci L, Del Boccio P et al. Repetitive
from the American Diabetes Association nephrology fragmentation products of albumin and alpha-
guidelines [26]. Laboratories are encouraged to adopt their 1-antitrypsin in glomerular disease associated
national guidelines where these are available. Of note, the with nephrotic syndrome. J Am Soc Nephrol
decision point of 30 mg/g creatinine is also recommended 2006:17:3139-3148.
as an indication of renal damage in the nondiabetic 2 Osicka TM, Comper WD. Characterization of
population [27]. immunochemically nonreactive urinary
albumin. Clin Chem 2004;50:2286-2291.
*Based on ADA Nephropathy guidelines [26]. 3 Svridov D, Drake SK, Hortin GL. Reactivity
**Albumin concentration decision point from KDOQI of urinary albumin (microalbumin) assays
[27]. with fragmented or modified albumin. Clin
Chem 2008;54:61-68.
Interpretation 4 Comper WD, Jerums G, Osika TM.
Urine albumin is primarily measured as a marker of the Differences in urinary albumin detected by
risk of development of renal damage in diabetic patients. It four immunoassays and high-performance
is now becoming established as a marker for renal damage liquid chromatography. Clin Biochem
in nondiabetic patients, owing to vascular disease 2004;37:105-111.
associated with hypertension, elevated lipids, and other 5 Becker GJ. Which albumin should we
standard risk factors. An elevated urine albumin is also an measure? Kidney Int 2004;66(suppl 92):S16-
established marker of cardiovascular risk in the diabetic 17.
and nondiabetic populations [28], and this risk may extend 6 Miller WG, Bruns DE, Hortin GL et al.
down to results within the currently accepted normal Current issues in measurement and reporting
range [29]. The benefit of urine albumin in the of urinary albumin. Clin Chem 2009;55:24-
microalbumin range compared to measurement of total 38.
protein is the increased sensitivity provided by albumin. 7 Shephard MDS, Gill JP. An innovative
Once a result is in the macroalbumin range, the Australian point-of-care model for urine
significance is the same as frank proteinuria. albumin/creatinine ratio testing that supports
diabetes management in indigenous medical
64
Albumin in Urine
services and has international application. Ann <http://www.bipm.org/en/committees/jc/jctl

Clin Biochem 2005;42:208-215. m/> Accessed 05.28.2008.
8 Parsons MP, Newman DJ, Newall RG, Price CP. 20 American Diabetes Association. Standards of
Validation of a point-of-care assay for the urinary medical care in diabetes. Diabetes Care
albumin/creatinine ratio. Clin Chem 1999;45:414- 2008;31(suppl 1):S12-S54
417. 21 Gansevoort RT, Verhave JC, Hillege HL,
9 VonSchenckH.Validationofalbumindetermined Burgerhof JGM, Bakker SJL, De Zeeuw D,
in urine with the HemoCue pointofcare De Jong PE. The validity of screening based
analyzer. Scand J Clin Lab Invest 2003;63:119 on spot morning urine samples to detect
126. subjects with microalbuminuria in the general
10 Sarafidis PA, Riehle J, Bogojevic Z, Basta E, population. Kidney Int 2005;67:S28-S35.
Chugh A, Bakris GL. A comparative evaluation 22 World Health Organization. Use of
of various methods for microalbuminuria anticoagulants in diagnostic laboratory
screening. Am J Nephrol 2008;28:324-329. investigations. Available at
11 Pugia MJ, Lott JA, Luke KE, Shihabi ZK, Wians <http://whqlibdoc.who.int/hq/2000/WHO_DI
FH, Phillips L. Comparison of instrument read L_00.4.pdf>
dipsticks for albumin and creatinine in urine with 23 Parekh RS, Kao WH, Meoni LA, Ipp E,
visual results and quantitative methods. J Clin Lab Kimmel PL, La Page J et al. Family
Analysis 1998;12:280-284. Investigation of Nephropathy and Diabetes
12 Meinhardt U, Ammann RA, Fluck C, Diem P, Research Group. Reliability of urinary
Mullis PE. Microalbuminuria in diabetes mellitus. albumin, total protein, and creatinine assays
Efficacy of a new screening method in after prolonged storage: the family
comparison with timed overnight urine collection. investigation of nephropathy and diabetes.
J Diabetes Complications 2003;17:254-257. Clin J Am Soc Nephrol 2007;2:1156-1162.
13 Mogensen CE, Viberti GC, Peheim E et al. 24 Jury DR, Mikkelsen DJ, Dunn PJ. Prozone
Multicenter evaluation of the Micral Test- II test effect and the immunoturbidimetric
strip, an immunological rapid test for the measurement of albumin in urine. Clin Chem
detection of microalbuminuria. Diabetes Care 1990;36:1518-1519.
1997;20:1642-1646. 25 Bakker AJ, Bierma-Ram A, Keidel H,
14 Parikh CR, Fischer MJ, Estacio R, Schrier RW. Syperda H, Zijlstra A. (Micro)albuminuria:
Rapid microalbuminuria screening in type 2 antigen excess detection in the Roche
diabetes mellitus: simplified approach with Micral Modular analyser. Clin Chem 2005;51:1070-
test strips and specific gravity. Nephrol Dial 1071.
Transplant 2004;19:1881-1885. 26 Molitch ME, DeFronzo RA, Franz MJ, Keane
15 Incerti J, Zelmaovitz T, Camargo JL, Gross JL, de WF, Mogensen CE, Parving HH, Steffes
Azevedo MJ. Evaluation of tests for MW. American Diabetes Association.
microalbuminuria screening in patients with Nephropathy in diabetes. Diabetes Care
diabetes. Nephrol Dial Transplant 2005:20:2402- 2004;27(Suppl 1):S79-83.
2407. 27 Levey AS, Eckardt KU, Tsukamoto Y, Levin
16 Brinkman JW, Bakker SJ, Gansevoort RT, A, Coresh J, Rossert J et al. Definition and
Hillege HL, Kema IP, Gans RO et al. Which classification of chronic kidney disease: a
method for quantifying urinary albumin excretion position statement from Kidney Disease:
gives what outcome? A comparison of Improving Global Outcomes (KDIGO).
immunonephelometry with HPLC. Kidney Int Kidney Int 2005;67:2089-2100.
2004;66:S69-S75 28 Sarnak MJ, Levey AS, Schoolwerth AC,
17 Polkinhorne KR, Su Q, Chadban SJ, Shaw JE, Coresh J, Culleton B, Hamm LL et al. Kidney
Zimmet PZ, Atkins RC. Population prevalence of disease as a risk factor for development of
albuminuria in the Australian Diabetes, Obesity, cardiovascular disease: American Heart
and Lifestyle (AusDiab) Study: Association scientific statement. Circulation.
immunonephelometry compared with high- 2003;108:2154-2169.
performance liquid chromatography. Am J 29 Xu J, Knowler WC, Devereux RB, Yeh J,
Kidney Dis 2006:47:604-613. Umans JG, Begum M et al. Albuminuria
18 Denis Sviridov D, Meilinger B, Drake SK, Hoehn within the normal range and risk of
GT, Hortin GL. Coelution of other proteins with cardiovascular disease and death in American
albumin during size-exclusion HPLC: Indians: the Strong Heart Study. Am J Kid
implications for analysis of urinary albumin. Clin Dis 2007;49:208-216.
Chem 2006;52:389-397. 30 Ricos C. Biological variation database.
19 Joint Committee for Traceability in Laboratory Available at <www.westgard.com> Accessed
Medicine website. Available at 05.02.2008.
65
Albumin in Urine
Table 1: Albumin in Urine Methods Summary Table

Method1:Immunoturbidimetry
Sensitivity(mg/L):Typically<10
Principle:Antialbuminantibodiesreactwithalbumininthesampletoscatterlight,reducingtransmittedlight.
Absorbancechangecanbemeasuredatarangeofwavelengthsinthevisiblerange.Ahomogenous
immunoassay.
Usage:WidelyusedonroutinechemistryanalyzersandinHaemocuepointofcaredevice
Comments:Prozoneeffectmustbeconsidered
Method2:Immunonephelometry
Sensitivity(mg/L):Typically<10
Principle:Antialbuminantibodiesreactwithalbumininthesampletoscatterlight.Lightisdetectedatanangle
totheincidentlight.
Usage:Incommonuse
Comments:Requiresspecificanalyzerwithnephelometriccapacity
Method3:Radioimmunoassay
Sensitivity(mg/L):<5
Principle:Competitiveimmunoassay
Usage:Uncommoninroutineuse.
Comments:Thefirstmeasurementsystemforurinealbumininthemicroalbuminuricrange
Method4:HPLC(AusAMtechnologies)
Sensitivity(mg/L):<5
Principle:ZorbaxpreparativeGF250HPLCcolumn
Usage:Uncommoninroutineuse
Comments:Detectsbothimmunoandnonimmunoreactiveintactalbuminwithpossibledetectionofother
proteins
Method5:Semiquantitativedipstickdyebinding(Clinitek)
Sensitivity(mg/L):Approximately20
Principle:Bindingtohighaffinitysulfonephthaleindyewithcolorchange.Quantitationbycomparisonwith
colorchartorClinitekReaderusingreflectometry.Colourchartis10,30,80,150mg/L.Creatininecanalsobe
measuredonthesamesystem.
Usage:Pointofcaredevice
Comments:Semiquantitativeonly.Visiblehemoglobinormyoglobininsamplecanaffectresult.Coloreddrugs
ordyesmaymasktrueresponse.Lowpositiveresultsrequireconfirmation.
Method6:Semiquantitativeimmunologicaldipstick(Micral)
Sensitivity(mg/L):Approximately20
Principle:Detectionofalbuminwithalbuminenzymecomplexwithsubstratetoformcoloredpad.Comparison
withcolorchartat0,10,20,50,100mg/L.
Usage:Commoninpointofcaresetting
Comments:Usefulscreeningtest.Noequipmentrequired.Timeofexposetourineiscritical.Lowpositive
resultsrequireconfirmation.
66
Albumin in Urine
TABLE 2: Decision Points for Interpretation of Urine Albumin.*

Spot Samples Timed Samples
Albumin Albumin/Creatinine Ratio Albumin Excretion Rate
Concentration
**
mg/L mg/g or mg/mol mg/24 hours g/min
g/mg
Normal <30 <30 <2.5 (male) <30 <20
<3.5 (female)
Microalbuminuria 30-299 2.5-29 (male) 30 -299 20-199
3.5-29
(female)
Macroalbuminuria 300 30 300 200
67
Alcohol
Alcohol
Ping Wang
Name: Alcohol
Molecular formula: Ethanol Methanol Isopropanol
C2H5OH CH3OH C3H7OH
Molecular mass: 46.07 32.04 60.09
Merck Index: (11th edition) 3716, 5868, 5096
Chemical class: Alcohol
Principles of Analysis and Current Usage

Ethanol, because of its extensive availability in such Early chemical methods for determination of ethanol
products as beverages and medicines, is the most were often based on the oxidation of ethanol by
frequently encountered toxic substance. Detection and potassium dichromate or other oxidizing agents in a
quantitation is often required to determine legal strongly acid medium. Reduction of the dichromate
impairment, evaluate patients suspected of poisoning, results in a color change that can be measured to
determine eligibility for organ transplantation, and assess monitor the reaction. A variation of this chemical
compliance with dependency treatment programs. method was introduced by Widmark. His method was
based on the simultaneous distillation and quantitative
Alcohol analysis is frequently requested of a laboratory oxidation of ethanol by dichromate. A microdiffusion
providing medical and/or forensic services. The clinical method, based on essentially the same principle, has
laboratory must provide rapid and reliable results for been developed for commercial use. In this procedure,
patients in life-threatening situations. The forensic chromic acid reagent, contained within a sheet of glass
laboratory, while less constrained by turnaround time fiber paper, is reduced to blue-colored chromic oxide
requirements, is expected to provide results that are by ethanol. Heating the sample at 80C to 120C
defensible in a court of law. Today, many laboratories releases ethanol into the glass fiber sheet that is placed
performing alcohol measurements are expected to meet directly above the sample according to the following
both medical and legal requirements. reaction.
The term alcohol is often used to refer specifically to Equation 1

ethanol but can also include the other monohydroxy 2K2Cr2O2 (orange-yellow) + 10H2SO4 + 3C2H5OH
alcohols that may be ingested, methanol and isopropanol. 2Cr2(SO4)4 (blue-green) + 2K2SO4 + 3CH3COOH +
Therefore, in the clinical setting, an alcohol or volatile
11H O + 4H+
2
testing procedure should detect all three substances. If a
method is used which measures only ethanol, results may
Most of these chemical methods are considered
be reported with respect to ethanol, and those served by
obsolete because of their poor specificity, tediousness,
the laboratory must be aware of this limitation.
and lack of adaptation to automation [1]. Furthermore,
Techniques for the determination of the alcohols include
it is important to remember that the chemical assays
assays that are nonspecific and semiquantitative (osmolar
are not specific for ethanol but detect volatile reducing
and diffusion methods) and assays that are specific and
agents. Despite the above limitations, this type of
quantitative (enzymatic and chromatographic procedures)
assay does have the advantage of being amenable to a
(Table 1). (In this chapter, if the method does not
wide variety of specimen types, including all body
distinguish among ethanol, methanol, and isopropanol, it
fluids and tissues.
will be identified as an alcohol procedure. If the technique
is specific for one of the alcohols, it will be identified as a
The presence of alcohol in serum leads to an increase
method for the particular substance detected [e.g.,
in the osmolality of the serum when the osmolality is
ethanol].)
measured by use of the freezing-point depression
technique (Table 1, Method 2). Osmometers that use
i
Alcohol the property of vapor-pressure depression to measure
Previous and current authors of this method: osmolality cannot be used for the purpose of
First edition: Timothy J. Schroeder estimating alcohol concentrations, because alcohols
Methods edition: Timothy J. Schroeder are volatile and contribute significantly to the vapor
Second edition: Timothy J. Schroeder pressure above a solution. This phenomenon will
Third edition: K. Michael Parker result in a falsely low measurement for serum
Fourth edition: K. Michael Parker osmolality.
Fifth edition: Ping Wang
68
Alcohol
In the presence of alcohol, there will be an increased gap with respect to the type of alcohol detected. Thus the
between the measured (by freezing-point depression) and osmolal gap can help reveal the presence of alcohols
the calculated osmolality values. This osmolal gap other than ethanol if the results of enzymatic ethanol
correlates reasonably well with blood alcohol measurements are available.
concentrations, because serum osmolality increases by
0.22 mOsm/kg H2O for each 1 mg/dL of ethanol. Using Enzymatic methods have become the most common
the osmolal gap, Equation 2 can then be used to estimate means for measurement of ethanol. Alcohol
a serum ethanol concentration: dehydrogenase (ADH), the enzyme employed in these
assays, is specific for ethanol and does not react with
Ethanol (mg/dL) = osmolal gap 4.6 acetone. However, ADH shows slight cross reactivity
with other alcohols. Relative to ethanol, the cross
Ethanol can also be estimated according to another reactivities are 6% for isopropanol, 1% for n-propanol,
variation of this equation as follows: 3% for methanol, and 4% for ethylene glycol [4].
Although the high specificity of ADH for ethanol
(Measured osmolality 290 mOsm/kg) 42.4 = ensures a small probability of interference, enzymatic
ethanol (mg/L) assays can be the source of misleading results in cases
where intoxication results from the ingestion of other
The factor 42.4 is the product of the molecular mass of alcohols [5]. Comparison of results from enzymatic
ethanol (46.07 mg/mmol) and the correction for the assays with those from less specific assays (chemical
volume fraction of water in serum (0.92). As a variation or osmometric) can help in identifying the presence of
on this principle, the osmolar gap, which is the difference these or other alcohols. The enzymatic ethanol assay is
between the measured and calculated osmolality, can be based on the oxidation of alcohol to acetaldehyde with
used to estimate the concentration of an alcohol. The concomitant reduction of NAD+ to NADH (Equation
difference between the measured osmolality and the 3). The NADH that is produced may be measured
calculated osmolality is determined using the following directly at 340 nm, or may be coupled to alternative
equation: detection schemes involving electrochemical,
fluorometry, and colorimetric techniques. In one such
Calculated osmolality = 2[Na+(mmol/L)] + secondary reaction the NADH is coupled to a
glucose(mg/dL)/18 + BUN(mg/dL)/2.8 diaphorasechromogen system producing a red-
or colored solution that can be measured at 500 nm.
Calculated osmolality = 2[Na+ (mmol/L)] + glucose
(mmol/L) + urea (mmol/L) Equation 3:
This difference is normally 10 mOsm/kg. However, if a NAD+ + alcohol alcohol dehydrogenase
low-molecular-mass toxin such as an alcohol is present, NADH(340nm) + acetaldehyde + H+
the osmolar gap will be increased. The serum
concentration of the particular alcohol can be estimated Another variation on the enzymatic scheme is a
when only one alcohol is present using the osmolar gap fluorometric technique. The technique is termed
and the following relationships: radiative energy attenuation, and it measures the
degree of inhibition of the fluorescence of fluorescein
Each 1 mg/dL ethanol = 0.22 osmolality increase dye resulting from the production of a colored product
Each 1 mg/dL methanol = 0.34 osmolality increase [6]. In the original version of this assay, the initial
Each 1 mg/dL isopropanol = 0.17 osmolality increase reaction of ADH with ethanol is coupled with a second
reaction between NADH and the tetrazolium salt,
To increase the specificity of the method, Pappas et al. [2] iodonitrotetrazolium violet dye (INT). This additional
proposed determining the ratio of the estimated ethanol reaction, catalyzed by diaphorase, results in the
from the osmolar gap to the ethanol concentration reoxidation of NADH to NAD along with the
measured by a specific ethanol procedure. The ratio of generation of a red colored formazan-INT. This
estimated ethanol to measured ethanol was helpful in product has an absorbance peak at 492 nm, which
evaluating acutely intoxicated patients for the presence of overlaps the excitation and emission spectral profile of
ethanol and/or another low-molecular-mass volatile. fluorescein included in the reaction mixture. The
Discrepancies in the ratio reliably predicted the presence decrease in fluorescence intensity is inversely related
of another volatile. to ethanol concentrations present in the sample. This
assay has since been reformulated to eliminate a time-
However, use of the osmolal gap to estimate ethanol consuming probe wash step by replacing the INT with
concentrations may result in an overestimation of alcohol a thiazolyl blue dye (MTT). The mechanism behind
concentrations by up to 30% [3]. This discrepancy has the assay is unchanged with the reduced MTT yielding
been attributed to the atypical osmotic behavior of a purple color with an absorbance at 565 nm [7].
ethanol, which can alter the degree of dissociation of
solutes within the specimen. Regardless of these
shortcomings, osmolal gap evaluations can be helpful in
the diagnosis of acutely intoxicated patients in emergency
situations. In addition, the osmolal gap is nonselective
69
Alcohol
Equation 4 Instrumentation that measures breath alcohol can be

either qualitative or quantitative. Hand-held devices
are usually qualitative in design and are used to detect
the presence of alcohol above a certain level or give a
semiquantitative estimate of the blood alcohol
concentration. Hence these devices are readily used
There is good agreement between enzymatic methods and for roadside screening of suspected drinking drivers,
chromatographic methods [8]. However, some enzymatic in emergency medicine, and in fitness-for-duty testing.
methods can give falsely increased ethanol concentrations The principle behind these hand-held devices is
with samples from patients with increased levels of serum usually based on an electrochemical detection scheme.
lactic acid and lactate dehydrogenase [9]. Although the Ethanol from the breath is oxidized by a fuel cell that
enzymatic methods are not absolutely specific for ethanol, causes the production of free electrons. The electric
these methods usually meet the requirements for accuracy, current generated by the free electrons is directly
precision, and reliability. The need for rapid analysis in proportional to the amount of ethanol oxidized by the
emergency settings has led to the demand for more point- fuel cell [4].
of-care instrumentation. New instrumentation has been
developed that is portable, accurate, and uses specimens The most common quantitative method for detection
such as breath and saliva that are less invasive than of ethanol in breath is the use of an infrared detection
traditional venipuncture. In addition, point-of-care testing system. These devices are used in the measurement of
is increasingly being used in fitness-for-duty and post- evidential breath alcohol levels. The results from such
accident testing and in compliance monitoring at a test can then be used as evidence in a court of law.
detoxification and correctional centers. Measuring the infrared energy at the beginning and
end of the sample chamber quantitates the ethanol in a
Saliva is an appropriate specimen for the analysis of breath sample. This is because the amount of ethanol
ethanol, since the amount of ethanol reaching the saliva present in the breath sample is proportional to the
should be equivalent to that in plasma or serum. This is amount of infrared energy lost to absorption. The
because ethanol is of low molecular mass, is highly water wavelengths utilized are 3.4 and/or 9.5 microns,
soluble, and does not bind to blood proteins and therefore corresponding to the CH and CO vibrational
freely diffuses into saliva. A number of point-of-care stretching in the ethanol molecule, respectively.
saliva ethanol testing kits are commercially available. Selectivity for ethanol is enhanced by the use of more
They provide either qualitative or quantitative results, and than one wavelength. The risk of misreporting an
the majority employ the same enzymatic processes interfering substance as ethanol was further reduced
described above. With one such test system, saliva is by the introduction of an instrument that makes use of
collected on a sterile cotton swab provided with the kit, five different infrared wavelengths to identify and
and the swab is inserted into the test cartridge. The saliva quantitate ethanol [4]. Gas chromatography with flame
must then completely fill the reaction capillary tube in ionization detection (GC-FID) is considered to be the
order for the internal quality assurance device to indicate reference method [11].
that a proper collection was obtained. In a process that
takes approximately 2 minutes, any ethanol in the saliva Chromatography is not as popular as enzymatic
will react with the enzymes in the tube to form a colored methods because of the need for specific expertise for
endpoint. The tube is fitted with a linear scale in the range performing the analysis and the concern that the
of 0 to 150 mg/dL or 0 to 350 mg/dL, depending on the purchase and maintenance of sophisticated
kit [10]. One limitation of this type of device, however, is instrumentation dedicated only to a single class of
that 20 minutes must have elapsed between the cessation analytes is not cost effective. However, gas
of drinking and the collection of the saliva, or an chromatography offers the distinct advantage of
overestimation of ethanol may result. In addition, some simultaneous detection of other alcohols and volatiles,
cross reactivity and response occurs with n-propanol and such as acetone, methanol, and isopropanol.
isopropanol. For the most part, these devices have been
thoroughly tested and found to be reasonably accurate, Direct or head-space injection techniques are
convenient, fast and free from such interferences as employed for GC analysis of volatiles. The direct
methanol, ethylene glycol, acetone, and methyl ethyl injection of blood or serum requires a thorough
ketone [10]. manual washing of the syringe after each injection.
The addition of an autosampler to the GC system
Breath is also an appropriate noninvasive specimen for the improves throughput and assay precision (CVs) while
analysis of ethanol. A small amount of unchanged ethanol decreasing chances of carryover through the use of
is expired with every breath, and this amount is sophisticated washing routines. Head-space
proportional to the amount of ethanol in whole blood. procedures inject into the GC system a vapor sample
This is based on the tenet that ethanol in alveolar capillary removed from a confined space above blood in a
blood promptly equilibrates with alveolar air in a ratio of closed container. Head-space analysis prevents
2100:1 (blood:breath). In other words, 2100 mL of contamination of the column and injector. Increased
expiratory air will contain the same amount of ethanol as sensitivity of head-space analysis can be gained by the
1 mL of whole blood [10]. addition of a salt, such as sodium chloride, to the
specimen. The addition of salt results in an increase in
70
Alcohol
the concentration of the volatile substance in the vapor dehydrogenase. The 340 nm absorbance change
phase. Both injection techniques require the sample to be associated with generation of NADH in the second
diluted with an aqueous solution containing an internal reaction is proportional to the methanol concentration.
standard to normalize variation among specimens. This assay can be adapted to automatic enzymatic
platforms. Because high concentrations of ethanol
Reference and Preferred Methods (usually > 200 mg/dL) may cause false biases in the
The reference method for alcohol testing is gas methanol enzymatic assay, the test result should be
chromatography [11]. The choice of method for ethanol interpreted together with an ethanol test result and
determination depends on several factors, such as clinical presentation. Dipstick methods using
medicolegal requirements, available instrumentation, enzymatic systems with color-generating indicator
testing personnel expertise, and turnaround-time reactions have been useful in identifying ethanol at
limitations. In many states there are specifications high concentrations (>100 mg/dL) [13,14].
defining the methods that must be used when performing
medicolegal testing. Gas chromatography remains the reference method for
alcohol testing. Proficiency testing surveys indicate
Diffusion methods have largely been replaced by more that it remains one of the commonly used methods.
specific and quantitative procedures such as the enzymatic Although it is specific for ethanol, the other
or gas chromatographic techniques. Diffusion methods monohydroxy alcohols can be specifically measured in
may still be usedonly because of their simplicity, the same analytical run. Therefore, it is often the
reagent stability (more than 1 year), and low costin preferred method in clinical situations. With regard to
laboratories providing very basic stat screening for sample introduction, the direct-injection method
volatile compounds. A test can be set up in a few minutes, provides results (approximately 10 min analysis time)
and a qualitative result can be obtained in half an hour. A more rapidly than the head-space technique, and
true quantitative transfer of the alcohol by diffusion consequently it is often more suited to the demands of
requires at least 18 hours of incubation time at room emergency testing. On the other hand, when whole
temperature. This makes the method unsuitable as a blood is the specimen, problems with inlet and column
quantitative test. Since the technique detects all volatile fouling and syringe plugging can rapidly develop with
oxidizable substances, any of the simple alcohols, as well the direct-injection technique. Therefore, for
as compounds such as ketones, would be detected. processing a large volume of specimens, as in a
forensic setting, the head-space method is usually
Indirect methods based on osmometry should be used in preferred. Combination of an enzymatic method for
very limited emergency situations when more specific screening followed by confirmation by gas
methods are not available. Any condition that changes the chromatography provides the most accurate approach
osmolality affects the estimation of alcohol. Significant for forensic testing.
errors can occur in patients with diabetic hyperglycemia,
uremia, or hypernatremia. Substantially higher results are Specimen
often obtained for the estimated alcohol concentration Specimens for the measurement of ethanol include
based on osmometry than with those measured by a whole blood (venous or capillary), plasma, serum,
specific method [12]. The use of the estimated ethanol to breath, saliva, and urine. For enzymatic methodology,
measured ethanol ratio (described above) may alert one to whole blood should be deproteinated prior to analysis
the possibility of ingestion of another volatile substance. according to the manufacturers instructions.
Anticoagulants do not interfere with the enzymatic or
Enzymatic methods are widely used, according to gas chromatographic ethanol methods. When gas
proficiency testing surveys. Most laboratories have chromatography is used, any tissue or body fluid may
automated chemistry instrumentation which can easily be used as specimen. Regardless of what specimen is
and conveniently be set up to perform enzymatic ethanol used, the samples must be well sealed to prevent loss
measurements. Most analyzers are capable of directly of volatiles. Other mechanisms of ethanol loss include
measuring absorbance of NADH at 340 nm, and no metabolism by microorganisms and temperature-
coupled indicator reaction is necessary. This approach, dependent oxidation to acetaldehyde. The addition of
when used on automated instruments, yields the best sodium fluoride and refrigeration will effectively stop
precision of any of the ethanol methods. The radiative the loss of ethanol by these two processes [15].
energy attenuation method provides rapid, reliable results
using instrumentation found in many laboratories and Contamination by microorganisms may also result in
most toxicology sections. Since these enzymatic methods false apparent increases in ethanol concentration.
have been designed to quantitate ethanol, other alcohols Urine specimens contaminated with yeast and
will not be detected. Emergency drug testing will require containing modest amounts of glucose have been
additional methods to detect and quantitate methanol and shown to produce sizeable amounts of ethanol by the
isopropanol. Ingestion of these alcohols occurs with process of fermentation [16]; for this reason,
sufficient frequency to require that toxicology laboratories specimens must be kept well stoppered and preferably
have the capacity to measure them. A quick enzymatic refrigerated to prevent loss of ethanol. It has been
assay for methanol is available for quantitative detection. shown that sealed samples of whole blood or whole
This method uses alcohol oxidase to oxidize methanol to blood plus fluoride can be stored at 0C to 3C or at
formaldehyde, which is then further oxidized by alcohol room temperature (22C to 29C) without significant
71
Alcohol
loss of ethanol content over a 14-day period [17]. Several Interpretation

microorganisms are capable of converting glucose to There are both clinical and forensic applications for
ethanolfor example, Candida albicans. The presence of the analysis of ethanol. Alcohol is a primary
these organisms along with glycosuria and prolonged depressant of the central nervous system. If alcohol is
storage at room temperature may lead to falsely increased taken in sufficient quantity, death can result from
ethanol results [18]. Therefore, urine specimens must be irreversible depression of respiration. It is also
handled in such a way to inhibit microbial production of important to distinguish altered mental status due to
ethanol. A 1% solution of phenylmercuric nitrate can be ethanol intoxication from other causes, especially in
added to the urine collection container to inhibit any cases where head trauma, low blood glucose
microorganisms which may be present in the urine. concentrations, or seizure are complicating the clinical
diagnosis. Because of the serious consequences of
Interferences ethanol intoxication, rapid analysis is required for
The use of alcohol-containing swabs to clean the initiation of appropriate therapy. Forensically, alcohol
venipuncture site has generally been discouraged, but is monitored for the purpose of workplace drug
McIvor and Cosbey [19] have reported that alcohol swabs testing, post-accident investigations, driving
led to minimal interference as measured by head-space impairment, and postmortem examinations. As the
GC, which was unlikely to significantly affect the results. medicolegal ramifications of ethanol intoxication are
Specimens with increased lactate and lactate serious, the analysis must be accurate and follow strict
dehydrogenase concentrations can give falsely increased guidelines to ensure legal acceptance.
ethanol results with some enzymatic assays.
Anticoagulants do not interfere with the enzymatic or GC Medicolegal interpretation of ethanol results is usually
procedures. For gas chromatography, any tissue or body based on state laws. Historically, 0.1 g/dL (100
fluid may be used. Some state regulations require that mg/dL; 21.7 mmol/L) has been used as the legal limit
ethanol analysis be performed on unclotted samples. for blood ethanol concentration (BEC). The trend now
Senkowski and Thompson [20] reported that analysis is toward lower BEC limits. Most states now use a
following homogenization of the clot produced results BEC limit between 0.05 and 0.10 g/dL, with 0.08 g/dL
comparable to those obtained with unclotted specimens. being the most common limit in use. In the United
States, drivers under age 21 are held to stricter
Alcohol Reference Interval standards under zero-tolerance laws, with a BEC limit
Units used for reporting ethanol levels in biological between 0 and 0.02 g/dL. In Europe, 0.05 g/dL has
specimens depend on the purpose in which the ethanol been proposed as the goal for the BEC limit. Sweden
value will be used. Clinical laboratories generally utilize has lowered the limit further to 0.021 g/dL [22].
SI units for reporting, which express ethanol in terms of
the number of molecular mass units of ethanol Endogenous ethanol production accounts for a BEC
(millimoles) per unit volume. However, if an ethanol less than 0.3 mg/dL. Following acute ingestion, BEC
result is to be used in court proceedings, state law usually usually peaks at 0.5 to 2 hours (fasting) or 1 to 6 hours
mandates the units used for reporting. Additionally, in the (nonfasting). BEC < 50 mg/dL may not be
United States, most statutes require that ethanol accompanied by obvious behavioral effects. As BEC
concentrations be expressed in terms of mass per unit increases beyond 50 mg/dL, central nervous system
volume of whole blood [21]. Law enforcement agencies in effects become more apparent, progressing from
other countries may have different requirements. euphoria, excitement, and confusion to stupor. At a
BEC > 150 mg/dL, most individuals are obviously
This can cause some confusion if the specimen utilized for drunk. BEC > 350 mg/dL may lead to coma and death.
analysis was plasma or serum, because these specimens Chronic alcohol users may tolerate greater BEC.
have a higher water content than whole blood and will Alcohol clearance is primarily by hepatic metabolism.
therefore contain more ethanol per unit volume than Between concentrations of 20 and 300 mg/dL, ethanol
whole blood. Because of the potential for error when metabolism follows zero-order kinetics at a rate of
making conversions of ethanol content between specimen approximately 10 mL of ethanol metabolized per hour,
types, some states have amended their statutes to include corresponding to a decrease in blood ethanol of about
per se limits for different specimen types. Ethanol, as well 20 mg/dL/h [23]. However, alcohol metabolism varies
as other volatiles, methanol, and n-propanol, are normally significantly between individuals, with alcoholics
not detectable in blood or tissues. A blood ethanol showing up to a twofold increase. Gastric alcohol
concentration of 300 mg/dL (65.1 mmol/L) or greater may dehydrogenase activity is less in women and alcoholic
be associated with coma, whereas values greater than 400 men than in nonalcoholic men, leading to increased
mg/dL (86.8 mmol/L) have been reported to be fatal. ethanol bioavailability and BEC [24]. Medications
Individuals who are chronic users of ethanol can show such as aspirin [25] and H2-receptor antagonists
even higher concentrations. In the postabsorptive phase, [26,27] inhibit gastric alcohol dehydrogenase, thereby
ethanol is distributed approximately as follows compared increasing the amount of ingested ethanol reaching the
to blood (1.0): urine (1.3), saliva (1.12), breast milk (1.1), systemic circulation. Ascorbic acid has been reported
CSF (1.14), brain (0.8), fat (0.02), and breath (1/2100). to increase ethanol clearance in healthy individuals
Plasma or serum ethanol is about 1.16 times that of whole [28].
blood, depending on the hematocrit. Methanol and isopropanol are usually not detected in
body fluids. Methanol concentrations greater than 20
72
Alcohol
mg/dL are often toxic. Patients with significant methanol and alcohol determination, J Chromatogr
concentrations should be treated with ethanol or 4-methyl 1992: 580: 161-190.
pyrazole to inhibit the production of the toxic metabolites 12 Snyder H, Williams D, Zink B,Reilly K.
formaldehyde and formic acid. Toxicity with isopropanol Accuracy of blood ethanol determination
usually develops at concentrations greater than 150 using serum osmolality, J Emerg Med 1992:
mg/dL. Acetone is produced as a consequence of 10: 129-133.
isopropanol metabolism and may be followed in addition 13 Christopher TA, Zeccardi JA. Evaluation of
to isopropanol concentration when monitoring the the Q.E.D. saliva alcohol test: A new, rapid,
clearance of the alcohol. accurate device for measuring ethanol in
saliva. Ann Emerg Med 1992: 21: 1135-
Alcohol Performance Goals 1137.
Survey data from the 2007 College of American 14 Schwartz RH, ODonnell RM, Thorne MM,
Pathologists (CAP) Participant Summary Report show Getson PR, Hicks JM. Evaluation of
that imprecision values (% coefficient of variation [CV]) colorimetric dipstick test to detect alcohol in
for ethanol determinations range from approximately 4% saliva: a pilot study. Ann. Emerg Med 1989:
to 9% at a mean ethanol concentration of 40 mg/dL and 18: 1001-1003.
CVs of 5% to 16% at a mean ethanol concentration of 24 15 Ellenhorn MJ, Schonwald S, Ordog G,
mg/dL. Precision for laboratories measuring ethanol by Wasserberger J, editors. Ellenhorns medical
gas chromatography is similar to laboratories that employ toxicology: diagnosis and treatment of human
enzymatic oxidation methods. Acceptable performance poisoning. Baltimore, MD, 1997, Williams &
criteria (CLIA-88) for measurement of ethanol require Wilkins, p. 793.
that laboratories be accurate to within 25% of the peer- 16 Lough PS, Fehn R. Efficacy of 1% sodium
group mean. For ethanol-free whole blood or serum fluoride as a preservative in urine samples
specimens, CAP has adopted an acceptable result range of containing glucose and Candida albicans. J
0 to 9 mg/dL. Forensic Sci 1993: 38: 266-271.
17 Winek CL, Paul LJ. Effect of short-term
References storage conditions on alcohol concentrations
1 Garriott JC, editor: Medicolegal aspects of in blood from living human subjects, Clin
alcohol, ed. 3, Tucson, AZ, 1996, Lawyers & Chem 1983: 29: 1959-1960.
Judges Publishing Co., Inc. 18 Alexander WD, Wills PD, Eldred N, Gower
2 Pappas AA, Gadsden RH, Taylor EH. Serum R. Urinary ethanol levels and diabetes,
osmolality in acute intoxication: A prospective Lancet 1981: 1(2):789.
clinical study. Am J Clin Pathol 1985: 84: 74-79. 19 McIvor RA, Cosbey SH. Effect of using
3 Bhagat CI, Beilby JP, Garcia-Webb P, Dusci LJ. alcoholic and non-alcoholic skin cleansing
Errors in estimating ethanol concentration in swabs when sampling blood for alcohol
plasma by using the osmolal gap, Clin Chem estimation using gas chromatography, Br J
1985; 31:647-648. Clin. Pract 1990: 44: 235-236.
4 Jones AW, Pounder DJ. Measuring blood alcohol 20 Senkowski CM, Thompson KA. The
concentrations for clinical and forensic purposes. accuracy of blood alcohol analysis using
In: Drug Abuse Handbook, Karch, S.B., editor, head-space gas chromatography when
Boca Raton, FL, 1998, CRC Press. performed on clotted samples, J Forensic Sci
5 Garriott JC, editor: Medicolegal aspects of 1990: 35: 176-180.
alcohol, ed. 3, Tucson, AZ, 1996, Lawyers & 21 Garriott JC, editor: Medicolegal aspects of
Judges Publishing Co., Inc. p. 221. alcohol, ed. 3, Tucson, AZ, 1996, Lawyers &
6 Yost DA, Boehnlein L, Shaffer M. A novel assay Judges Publishing Co., Inc., p. 268.
to determine ethanol in whole blood on the 22 Jones AW. Limits of detection and
Abbott TDX, Clin Chem 1984: 30: 1029A. quantitation of ethanol in specimens of whole
7 Garriott JC, editor: Medicolegal aspects of blood from drinking drivers analyzed by
alcohol, ed. 3, Tucson, AZ, 1996, Lawyers & head-space gas chromatography. J Forensic
Judges Publishing Co., Inc. p. 224. Sci 1991: 36: 1277-1279.
8 Jortani SA, Poklis A. Emit\R ETS\R plus ethyl 23 Gershman H, Steeper J. Rate of clearance of
alcohol assay for the determination of ethanol in ethanol from the blood of intoxicated patients
human serum and urine. J Anal Toxicol 1992: in the emergency department, J Emerg Med
16: 368-371. 1991: 9: 307-311.
9 Badcock NR, OReilly DA. False-positive 24 Frezza M, diPadova C, Pozzato G, Terpin M,
EMIT\R-st\T ethanol screen with post-mortem Baraona E, Lieber CS. Higher blood alcohol
infant plasma, Clin Chem 1992: 38: 434. levels in women: the role of decreased gastric
10 Jones AW. Measuring ethanol in saliva with the alcohol dehydrogenase activity and first-pass
QED? enzymatic test device: Comparison of metabolism. N Engl J Med 1990: 322: 95-99.
results with blood- and breath-alcohol 25 Roine R, Gentry RT, Hernandez-Munoz R,
concentrations. J Anal Toxicol 1995: 19: 169-74. Baraona E, Lieber CS. Aspirin increases
11 Tagliaro F, Lubli G, Ghielmi S, Franchi D, blood alcohol concentrations in humans after
Marigo M. Chromatographic methods for blood
73
Alcohol
ingestion of ethanol. JAMA 1990: 264: 2406-

2408.
26 Caballeria J, Baraona E, Rodamilans M, Lieber
CS. Effects of cimetidine on gastric alcohol
dehydrogenase activity and blood ethanol levels.
Gastroenterol 1989: 96: 388-392.
27 DiPadova C, Roine R, Frezza M, Gentry RT,
Baraona E, Lieber CS. Effects of ranitidine on
blood alcohol levels after ethanol ingestion:
comparison with other H2-receptor antagonists.
JAMA 1992: 267: 83-86.
28 Chen MF, Boyce HW Jr, Hsu JM. Effect of
ascorbic acid on plasma alcohol clearance. J Am
Coll Nutr 1990: 9: 185-189.
29 Watts MT, McDonald OL. The effect of sodium
chloride concentration, water content, and
protein on the gas chromatographic head-space
analysis of ethanol in plasma. Am J Clin Pathol
1990: 93: 357-362.
74
Alcohol
Table 1: Alcohol Methods Summary

Methods for All Alcohols:
Method 1: Distillation-oxidation; colorimetric
Principle: Alcohol diffuses into gas phase and reacts with oxidizing agent, changing its color:
2 K2Cr2O7 + 10 H2SO4 + 3C2H5OH
(yellow-orange)
2 Cr2(SO4)4 + 2 K2SO4 + 3 CH3COOH +11 H2O + 4 H+
(blue-green)
Usage: Stat or routine; all body fluids; tissue
Comments: Nonspecific; gives reaction with all volatiles
Method 2: Osmometry; freezing-point depression
Principle: Alcohol in high concentration increases serum osmolality: (a) difference between normal and measured value is
proportional to alcohol levels, or (b) difference between measured and calculated value (osmolar gap) proportional to alcohol
concentration
Usage: Stat; serum
Comments: Nonspecific; measured osmolality increased by non-alcohols
Methods for a Specific Kind of Alcohol:

Method 3: Enzymatic:
(a) Spectrophotometric
Ethanol:
_
Principle: NAD+ + alcohol alcohol dehydrogenase NADH + H+ + acetaldehyde
Usage: Stat or routine; serum; urine
Comment: Specific for ethanol; other alcohols not readily measured
Methanol:
alcohol oxidase
Principle: Methanol + O2 formaldehyde + H2O2
H O + formaldehyde + NAD+ formaldehyde dehydrogenaseformic acid +
2 2
NADH + H+
Comment: Specific for methanol; high ethanol concentration causes positive bias; alcohol oxidase is very labile and needs to
be prepared fresh before testing
(b) Colorimetric, dipstick
Principle:
alcohol dehydrogenase
NAD+ + alcohol NADH + H+ + acetaldehyde
diaphorase
NADH + iodonitrotetrazolium dye
NAD+ + reduced iodonitrotetrazolium dye (insoluble)
(colored)
(c) Fluorometric
alcohol dehydrogenase
Principle: NAD+ + alcohol NADH + H+ + acetaldehyde
diaphorase
NADH + tetrazolium dye
NAD+ + reduced monotetrazolium dye (insoluble)
(fluorescein fluorescence decreased)
The reduced dye is colored, decreasing the amount of light reaching the fluorescein dye reagent and
therefore attenuating its fluorescence
Usage: Stat or routine; serum
Method 4: Gas chromatography; flame ionization detection and quantitation
Principle: Alcohol separates on chromatographic column; identification of specific alcohol by retention time; quantitation by
comparison of peak height to an internal standard. Direct injection of sample fluid onto column or head-space analysis, where
the airspace above the sample is allowed to come to equilibrium, and the air sample is chromatographed.
Usage: Stat or routine; all body fluids
Comments: Specific for all alcohols
Method 5: Infrared spectrophotometry
Principle: The amount of infrared energy lost due to absorption is proportional to the alcohol concentration. Two
wavelengths used to monitor for interference.
Usage: Forensic and compliance; breath analysis
Comments: Qualitative or quantitative analysis
75
Alcohol
Table 2: Comparison of Assay Conditions for Ethanol

Parameter Enzyme Reaction Gas Chromatography
Temperature 37C 75C
pH 8.8 7.35
Final concentration of Reduced nicotinamide adenine n-propanol: 1.2 g/L
reagent components dinucleotide: 0.6 mmol/mL (20 mmol/L)
Alcohol dehydrogenase: 50 U/mL
Tetrasodium pyrophosphate: 0.075 mol/L
Sodium semicarbazide: 0.075 mol/L
Glycine: 0.022 mol/L
Fraction of sample volume Approximately 0.008 0.33
Sample volume 0.5 mL 0.2 mL: 0.5 mL injection
Linearity 0-600 mg/dL (130 mmol/L) 0-1000 mg/dL (217 mmol/L)
Precision* 4% (automated analyzers) 7% (automated analyzers)
Time for analysis 10 min (auto analyzers) 10 min (auto analyzers)
Interferences Enzyme inhibitors; some alcohols None known
Specimen Serum, plasma, urine Whole blood, serum, plasma, urine
*Coefficient of variation; obtained from College of American Pathologists toxicology survey data
Table 3: Units for Expressing Ethanol Concentrations and Common Conversion

Factors
% g/100 %
Units g/L (wt/vol) mL mg% mg/dL ppm g/mL (wt/wt) mg/g mmol/L
g/L 1 0.1 0.1 100 100 1000 1000 0.0948 0.948 21.71
%
(wt/vol) 10 1 1 1000 1000 1E+4 1E+4 0.948 9.48 217.1
g/100
mL 10 1 1 1000 1000 1E+4 1E+4 0.948 9.48 217.1
mg% 0.01 1E-3 1E-3 1 1 10 10 9.48E-4 9.48E-3 0.2171
mg/dL 0.01 1E-3 1E-3 1 1 10 10 9.48E-4 9.48E-3 0.2171
ppm 1E-3 1E-4 1E-4 0.1 0.1 1 1 9.48E-05 9.48E-4 0.02171
g/mL 1E-3 1E-4 1E-4 0.1 0.1 1 1 9.48E-05 9.48E-4 0.02171
%
(wt/wt) 10.55 1.055 1.055 1055 1055 10548 10548 1 10 229
mg/g 1.055 10.55 10.55 105.5 105.5 1055 1055 0.1 1 22.9
mmol/ 4.61E-
L 0.0461 4.61E-3 3 4.61 4.61 46.06 46.06 4.367E-3 4.367E-2 1
76
Alcohol
Figure 1: Photograph of Conway diffusion dish showing sample well, A, and inner well, B,
containing reacted dichromate solution (blue-green).
_____________________________________________________________________
Figure 2: Chromatogram of alcohol standards obtained by direct injection of the sample into
the gas chromatograph.
Retention time in minutes is recorded next to peak. Alcohols are eluted in the following
sequence: 1.545, methanol; 1.825, acetone; 2.195, ethanol; 2.880, isopropanol; 3.970, n-propanol
(internal standard).
_____________________________________________________________________
77
Alcohol
Procedure: Gas Chromatography of Alcohols Alcohol Retention Time

Table 2 compares the assay conditions for the Methanol 1.545
two most commonly used methods. The reactions for Acetone 1.825
the enzyme method are given in Table 1. Actual Ethanol 2.195
parameters for the enzyme procedure may vary, Isopropanol 2.880
depending on the manufacturer of the reagents and the n-Propanol 3.970
instrument used in the analysis. (internal standard)
See Alcohol: Figure 2 for sample chromatogram.
Principle
The following direct-injection gas 2. Determine the peak height (in millimeters) for
chromatographic procedure can be used to separate the ethanol (E) and internal standard (IS) peaks
volatile compounds such as ethanol, methanol, on each chromatogram.
isopropanol, and acetone. After sample injection into the 3. Calculate peak height ratios, E/IS.
heated injection port, the volatiles are separated on a 4. Calculate ethanol in samples as follows:
Carbowax column, detected with a flame ionization Ethanol (mg/dL) = ____E/IS (sample)____ 79 mg/dL
detector, and quantitated by comparison of the peak E/IS (calibration standard)
height ratio of the sample to internal standard, relative to 5. If an electronic integrator is used, calculate
the peak height ratio of the calibration standard to ethanol in samples as follows:
internal standard. Ethanol (mg/dL) = Factor E/IS (unknown)
where Factor = _______79 mg/dL______
Specimen E/IS (calibration standard)
Serum, plasma or whole blood may be Notes
analyzed. 1. Direct injection of the sample into the gas
chromatograph may result in inlet and column
Reagents and Materials contamination. As an alternative, analysis of
Equipment: Gas chromatograph equipped with the head-space vapor above a specimen may be
a flame ionization detector and a 6 ft 2 mm (I.D.) performed. The following procedure can be
column packed with 60/80 Carbopack B/5% Carbowax followed to prepare the head-space samples.
20M. a. Place 1.0 g of sodium chloride in a
glass screw-capped tube (screw caps
Reagents should contain Teflon-lined silicone
1. Internal standard: Dilute 75 L n-propanol to septa) for each calibration standard,
50 mL with deionized water. Stable when assay control, or patient blood.
stored in a refrigerator for up to 6 months. b. Add 0.5 mL of calibration standard,
2. Calibration standard: Prepare a calibration control, or sample to each tube.
standard of 79 mg/dL (17.1 mmol/L) by c. Add 0.5 mL of internal standard to
diluting 1 mL of 100% ethanol (anhydrous each tube, seal, and mix well.
from a sealed bottle) to 1 L with deionized d. Place tubes at 38C for 45 min.
water. Stable when tightly sealed and stored in e. Inject 0.5 mL of head-space vapor into
refrigerator for up to 1 month. Other dilutions the gas chromatograph.
should be prepared to verify linearity of the The same chromatographic conditions and
method (0 to 500 mg/dL; 0 to 108.5 mmol/L). calculation procedure can be used for the head-
space method as with the direct-injection
Conditions: Chromatographic parameters: injection technique. Preparation of calibration
temperature, 225C; column temperature, 75C; detector standard(s) in a matrix equivalent to that in the
temperature, 250C; carrier gas (helium) flow rate, 30 samples, particularly with respect to protein,
mL/min; detector gas (hydrogen) flow rate, 30 mL/min; has been suggested to reduce partitioning
(air) flow rate, 400 mL/min. differences of volatiles between standard(s) and
samples [29].
Assay 2. Isopropanol, methanol, and acetone can also be
1. Combine 50 L of calibration standard, assay quantitated by this procedure. (refer to
control, or patient blood (serum) sample and retention times in Alcohol: Figure 2.) The same
100 L of internal standard in an appropriate- peak height ratio method with n-propanol as
size glass vial. Cap the vial and mix. the internal standard is used for the calculation.
2. Sequentially inject 0.5 L from each vial into
the chromatograph. Chromatographic run time Concentration =
is approximately 6 min. Peak height ratio (unknown) Standard concentration
Peak height ratio (standard)
Calculation
1. Identify the ethanol peak using the retention
time:
78
Alcohol
Alcohols Gas Chromatography: Direct Injection 3. Methanol calibration standard (158 mg/dL)
Important Procedural Note: Do not use injection Same as #2 above, using chromatoquality methanol.
syringes that have been exposed to methanol as a 4. Alcohol Working Mixed Standard: Combine
washing solvent. Use deionized water when rinsing 100 L each of ethanol, isopropanol, and methanol
all syringes. standards with 100 L of internal standard. Prepare
fresh as needed. (Total volume: 400 L.)
Principle 5. Ethanol linearity check standardsTo
Most volatile compounds such as ethanol, respective 100 mL volumetric flasks, add 0.1, 0.2, 0.3,
methanol, and isopropanol will separate from 0.4, 0.5, 0.6, 0.7, and 0.8 mL of 100% chromatoquality
biological fluids in the heated injection port of ethanol, representing concentrations of 79, 158, 237,
a gas chromatograph. These compounds will 316, 395, 474, 553, and 632 mg/dL of ethanol,
then be separated from each other by the GC respectively. Dilute to mark with distilled water. Store at
column and can be identified and quantitated room temperature. Stable for 2 months. To be used for
by comparing their peak height ratio to a monthly instrument linearity check.
known internal standard. C. 12 75 mm glass culture tubes.
Specimens of Choice D. 100 L Eppendorf pipet.
A. For nonlegal/nonforensic quantitative uses: 0.5 E. Utak Volatiles Control Serum.
mL of serum collected in a red-top tube; or plasma F. Hamilton 7002 NWG syringe.
collected in either a heparinized or EDTA tube. Instruments and Parameters
B. For nonlegal/nonforensic qualitative uses: 1 A. The parameters given here are for a Shimadzu
mL urine from spot aliquot collection with no Mini-2 gas chromatograph.
preservatives added.
C. For legal/forensic* quantitative use: 1.0 mL Detector: Flame ionization
whole blood collected in a gray-top NaF tube. Column: 5% Carbowax 20M on 60/80 Carbopack B
D. For legal/forensic* qualitative uses: 1 mL urine (Supleco); 6 ft, 2 mm ID
from spot aliquot collection with no preservatives Temperatures: Col. 90C; Det. 240C; Inj. 240C.
added. Attenuation: 32
All specimens may be transported at Range: 10
room temperature. For nonlegal cases, serum or plasma Nitrogen flow: 50 mL/min
should be separated from RBCs and may be stored at Gas pressures at tank: air, 20; nitrogen, 50; hydrogen, 30
room temperature until time of testing (up to 3 days) mm Hg
then stored frozen at 20C. For legal/forensic whole
blood and all urine specimens, store at 2C to 8C prior B. Model 3390 Hewlett Packard GC Terminal
to testing.
*Note: For legal/forensic uses, specimen must Procedure
be accompanied by a Chain of Custody Form. A. Nonforensic analytical runs will consist of an
Indications internal control, a calibration standard if required, and
As an aid in assessing ethanol or other alcohol patient specimens.
abuse which may relate to acute or chronic clinical 1. Assess standardization by running a single
problems (i.e., coma, hepatic damage, drug/alcohol control at the beginning of each shift or after any
interactions, psychiatric disorders, acidosis); to assess instrument changes (i.e., septum change, gas tank
ethanol abuse for forensic purposes, including employee change, etc.).
monitoring and postmortem studies. 2. If control is within acceptable range, proceed to
Reagents and Materials step 4.
A. n-Propanol internal standard. Using a 1.0 mL 3. If control is out of acceptable range, repeat the
serological pipet, deliver 0.4 mL of n-propanol to a 100 control. If it is still unacceptable, perform a
mL volumetric flask. Dilute to mark with distilled H2O. recalibration.
Store in 100-mL reagent bottle at room temp. Stable for 4. Run the patient sample. If patient sample is
2 months. positive for any alcohol, prepare a second internal
B. Alcohol working standards. standard and sample mixture, and reinject for a duplicate
Note: For routine instrument calibration, a working analysis. For quantitative analyses (i.e., serum or blood)
mixed standard will be prepared from individual average the duplicate values for a reportable
standards. concentration. If patient chromatogram is negative, a
1. Ethanol standard (150 mg/dL)aqueous single analysis is all that is necessary.
Setpoint standard (Stephens Scientific Co.). Store 5. Run any other appropriate standard (i.e.,
refrigerated at 2C to 8C. Stability as indicated by methanol or isopropanol) if patient specimen indicates
manufacturer. need.
2. Isopropanol calibration standard (157 B. Forensic analytical runs will consist of a
mg/dL)Using a 0.2-mL serological pipet, deliver 0.2 calibration standard, an internal control, and the patient
mL of 100% chromatoquality isopropanol (2-propanol) specimens. Note that a standard and control must be
to a 100-mL volumetric flask. Dilute to mark with analyzed with each specimen or batch of specimens.
distilled water. Store at room temperature. Stable for 2 1. Assess standardization by running the 150
months. mg/dL setpoint calibrator. If the calibrator result is
79
Alcohol
between 142 and 158 mg/dL, proceed by running a Concentration (mg/dL) = PHR of 158 methanol std
control. If the calibrator is outside these limits, perform 3. Isopropanol
a recalibration. a. Measure peak heights as above in a for
2. Run the single internal control. Assess its patient, control, and 157 mg/dL isopropanol standard.
acceptability (will have been determined previously). b. Calculate PHR as above in a. Compute
3. Run the patient/subject specimen as with concentration of isopropanol in patient and control:
nonforensic runs, analyzing all positive samples in PHR of patient or control 157
duplicate. Concentration (mg/dL)
4. Indicate type of specimen (i.e., whole blood, = PHR of 157 isopropanol std
urine) in the report by appending computer comments C. Acetonereport only internally as being
Test performed on whole blood, or Test performed positive. Do not quantitate or officially report.
on urine. D. Concentration conversions (For use when
C. Sample Preparation converting to other concentration units):
1. Add 100 L (with a 100-L Eppendorf pipet) 1. To convert from mg/dL to g/mL, multiply by
of internal standard solution to a glass culture tube. 10.
2. Add 100 L (with a 100-L Eppendorf pipet) 2. To convert from mg/dL to grams %, divide by
of patient specimen, control, or appropriate standard to 1000.
the tube and mix by agitation. Analyze immediately E. An acceptable run includes control values
(within 1 min of preparation). within acceptable limits and patient duplicate analyses
3. Inject 1.0 L of the mix into the GC. (Note: within 10% of the duplicate mean. If outside of 10%,
Injection amounts may vary with time.) Peak height of analyze a third sample preparation, and recalculate from
internal standard should be a minimum of full-scale the two closest values.
response. Use syringe designated for alcohols; do not
rinse with methanol. Interpretation
4. Immediately press [START] on the integrator. A. For ethanol, 100 mg/dL denotes legally
5. Proceed to Calculations. intoxicated. Greater than 400 mg/dL is considered
critical. Critical values for methanol and isopropanol are
Calculations and Quality Control 30 mg/dL and 200 mg/dL, respectively. Reportable
A. Integrated calculations concentrations are up to 1000 mg/dL for each alcohol.
Note: For all positive patient samples, duplicate Call all critical values to the caregiver.
analyses are performed. Average these duplicate results B. For alcohol concentrations above 632 mg/dL,
for the reportable quantitative result for serum and/or dilute the patient specimen 1:1 (50 L plus 50 L) with
blood assays. distilled H2O, add 100 L internal standard, and
1. The integrator prints direct alcohol analyze. Multiply calculated mg/dL result by 2.
concentrations in mg/dL, as well as retention times for C. Run a complete ethanol curve to establish
the alcohols of interest. linearity monthly, when changing GC columns, or when
2. If only qualitative analysis is required (i.e., other changes dictate.
urine), compare retention times of patient sample peaks D. If methanol is detected in the patient sample,
to those of the known control. Urine concentrations at or run a blank (distilled water) taken through the
greater than 10 mg/dL for any of the alcohols will be procedure, then rerun the patient. This is to rule out false
reported as Positive. Concentrations below 10 mg/dL positives due to methanol from wash bottle. If methanol
will be reported as None detected. Urine results will appears in the blank, change water rinse.
not be reported quantitatively. E. Acetone will most frequently be found in
B. Manual calculations (to be used whenever patients who are acidotic or who have ingested
direct instrument integration is unavailable). isopropanol where acetone is produced as a metabolite.
1. Ethanol Acetone detection is noted but not officially reported.
(a) Measure the peak height of the n-propanol I.S. F. False-positive isopropanol levels may be a
and of the ethanol in millimeters. result of improper blood drawing technique (i.e., using
(b) Compute the peak height ratio (PHR) for the an alcohol swab). A good indication that a serum
150 mg/dL standard, control, and patient: isopropanol is due to swab alcohol is that there will not
peak height of ethanol be any acetone in the sample. Also, if available, confirm
PHR = peak height of I.S. the presence of isopropanol by analyzing the patients
Compute the concentration of ethanol in the urine. A negative corresponding urine is also indicative
patient and control: of a false-positive serum result.
PHR of patient or control 150 G. The current method sensitivity for each alcohol
Concentration (mg/dL) = PHR of 150 ethanol std is 10 mg/dL. Negative results should be reported as < 10
2. Methanol mg/dL (<10).
a. Measure peak heights as above in a for H. Because of the relatively high threshold
patient, control, and 158 mg/dL methanol standard. concentrations for reporting alcohols (10 mg/dL), other
b. Calculate PHR as above in a. Compute volatiles, although they may appear in the
concentration of methanol in the patient and control: chromatographic field, will generally not be detected.
PHR of patient or control 158 Even so, if clinical histories indicate specific nonalcohol
volatiles as being possible poisons, the area supervisor
80
Alcohol
should be consulted. This would include compounds B. Add 0.5 mL of potassium dichromate
such as acetonitrile, ethyl acetate, chlorinated to the center well with a 1 mL
hydrocarbons, and formaldehyde. serological pipet.
C. Place the cover on the dish, and allow
Alcohols, Serum, Plasma, Whole Blood, Quantitative it to stand at room temperature for 2 h.
1. Process a Utak control with a known ethanol Read promptly.
concentration at the beginning of each shift for each D. Observe the color produced.
staff member who will be performing ethanol Calculations (Observations) and Quality Control
concentration measurements on that shift. A. The dichromate reagent will change
2. To a 12 75 mm tube, add 100 L of serum, color in 15 min to 2 h if ethanol or
plasma, or whole blood. other volatile reducing substances are
3. Add 100 L of n-propanol internal standard. present, with the concentration of
4. Inject 1 L into gas chromatograph. substance being proportional to time
5. Integrator will directly report ethanol of color change. Color change is from
concentrations in mg/dL. yellow, to green, to blue.
B. A yellow nonreactive color
Alcohol Spot Test (Qualitative) indicates no alcohols above the
Principle procedure threshold cutoff of 20
In a Conway microdiffusion dish, alcohols will mg/dL.
diffuse out of an aqueous specimen, be trapped, and C. A green to blue color is indicative of
form a color complex with potassium dichromate. the presence of:
Specimen Ethanol
Serum (1 mL) drawn in a red-top tube with or Isopropanol
without separator gel, whole blood drawn in a Methanol
heparinized or EDTA tube, or urine (1 mL) spot aliquot. Other reducing substances
Indications D. For all positive samples, proceed
To be used as a general adjunct alcohol with confirmatory testing using gas
screening procedure. To qualitatively assess the chromatography.
presence of ethanol, methanol, or isopropanol. E. An acceptable run includes a green-to-
Reagents and Materials blue color formation with the positive
All chemicals used should be analytical reagent control and no green-to-blue color
grade. with the negative control.
A. Conway microdiffusion dish. Interpretation and Notes
B. Potassium dichromate. A. A positive finding indicates the
Into a 1-L flask, dissolve 1.0 g of presence of a reducing substance, such
potassium dichromate in 500 mL of as alcohol, but may not be indicative
water. Add 0.1 g of silver nitrate to of toxicity or impairment.
the solution. Carefully, add 500 mL of B. Sensitivity: 20 mg/dL; report
concentrated sulfuric acid. Stir to mix, negatives as None detected.
transfer to an amber reagent bottle, C. Specificity: Nonspecific method;
and store at room temperature. Stable unable to identify specific alcohol
for 1 year. present; will react with other reducing
C. Negative control. substances.
Distilled water blank.
D. Positive control (790 g/mL).
To a 100-mL volumetric flask, add 0.1
mL of absolute 100% ethanol. Fill to
mark with distilled water. Transfer to
a reagent bottle. Store at room
temperature. Stable for 3 months.
E. 500-L Eppendorf pipet.
F. 1-mL serological pipet.
Instrument and Parameters
None
Procedure
Note: With reach run analyze a positive and
negative control.
A. Cover the bottom of the middle ring of
the microdiffusion dish with blood,
serum, or urine by adding 0.5 mL of
sample with an Eppendorf pipet.
81
Aldolase
Aldolase
Stephen M. Gendler
Name: Aldolase, ALS,

fructose-1,6-diphosphate:d-glyceraldehyde-3-phosphate lyase
Clinical significance: Refer to Chapter 36, Cardiac and muscle disease, in the 5th Edition of Clinical
Molecular mass: approximately 160,000 daltons
Chemical class: enzyme protein
Known isoenzymes: subunits A, B, and C in tetrameric form
Biochemical reaction:
Principles of Analysis and Current Usage formation or degradation of NADH (methods 1 and 2,
Aldolase (ALS) catalyzes the reversible biochemical Aldolase Table: Comparison of methods for serum
reaction shown above. It is present as a tetramer aldolase (ALS) analysis). Coupled enzyme systems that
composed of two of the three known subunits designated result in the appearance of NADH have been shown to
A, B, and C. When analyzed as homotetramers, aldolase be subject to large errors caused by the presence of
A, B, and C have similar pH-activity profiles, endogenous interfering enzymes such as glycerol
MichaelisMenten constants (Km) for fructose dehydrogenase (GDH, glycerol-3-phosphate:NAD 2-
diphosphate (FDP), and molecular weights. Using FDP oxidoreductase, EC 1.1.1.8) or TPI. Thus the most
as substrate, the pH optima in Tris-malonate buffer [1] satisfactory design for a coupled enzymatic system is as
and in collidine buffer [2] range from pH 7.0 to 8.0. The follows:
Michaelis constants with FDP as substrate range from FDP ALS DAP + GAP
10-6 to 3 10-6 M. The molecular weights for the GAP TPI DAP
homomeric enzymes range from about 145,000 to
175,000 daltons [1]. The A homomer is found in high 2 DAP + 2 NADH + 2 H+ GDH
concentration in skeletal muscle. The B homomer 2 glycerol 1-phosphate + 2 NAD+
predominates in liver, and the C homomer predominates
in brain and other tissues. The primary isoenzyme in In this scheme interferences can be eliminated in the
normal serum is the A homomer. The hybrid isoenzyme preincubation phase. Interferences can be demonstrated
composed of three A subunits and one C subunit is by a decrease in A340 after the addition of the indicator
present in a somewhat lower concentration [3]. In enzymes and sample and before the addition of FDP.
aldolase analyses in current use, enzymatic activity is These interferences have not been characterized but may
determined by use of FDP as the substrate, since be endogenous substrates and dehydrogenases in serum
aldolase cleaves FDP more rapidly than other substrates that utilize NADH.[7] DAP or GAP contamination of
[1]. For analytical purposes the aldolase reaction is the FDP preparation can also be demonstrated by a
driven to the formation of trioses by use of hydrazine, a decrease in A340 after the addition of indicator enzymes
trapping agent for the triosephosphates [2,4,5]. The and FDP and before the addition of serum sample [8].
aldolase reaction can also be shifted to trioses by Proposed by Beisenherz et al [6] in 1953, this coupled
removal of the glyceraldehyde phosphate (GAP) with the reaction has been optimized by Ludvigsen [9], Pinto et al
enzyme triose-phosphate isomerase (TPI, D- [7], and Bergmeyer and Bernt [10] and adapted to the
glyceraldehyde-3-phosphate:ketol isomerase, EC kinetic measurement of enzyme activity [11]. This
5.3.1.1). This enzyme catalyzes the conversion of GAP method has been adapted to kit form and is
to dihydroxyacetone phosphate (DAP) [6]. The commercially available (Boehringer Mannheim,
ultraviolet spectrophotometric analyses use two coupled Calbiochem, and other companies).
enzyme indicator reactions, with the concomitant
Radioimmunoassays (RIA) for the homotetramers of
i aldolase A [12] and aldolase C [13] have also been
Aldolase
reported. RIA is not discussed further because these
assays are not widely used. In addition, the clinical
First edition: Steven M Gendler
utility of these assays is as yet unproved.
Methods edition: Steven M Gendler
Second edition: Not updated
Third edition: Not updated
Fourth edition: Steven M Gendler
Fifth edition: Not updated
82
Aldolase
Reference and Preferred Methods between survivors and nonsurvivors [22]. In contrast, in
A colorimetric method was published as a selected a study performed at a university hospital, the test was
method by the American Association for Clinical found to be worthless for the purposes ordered, including
Chemistry (AACC) in 1961 [14]. Because of the many diagnosis of polymyositis or other inflammatory
manipulations and the time required, this method is not myopathies and determination of whether an elevated
recommended. creatine kinase (CK) activity was caused by skeletal
muscle damage or myocardial damage [23]. These
The various modifications of the coupled enzymatic authors observed that the serum aldolase added little
techniques are difficult to compare, since the methods information when CK data were available and
have not been compared by one laboratory in a recommended the deletion of this test from the hospital
controlled situation (Table 2). The Bergmeyer and Bernt laboratory test repertoire. Other authors have expressed a
[10] method has been developed as a kit. Day-to-day similar view [20].
imprecision claims of 2.7% in the normal range and
1.9% in the abnormal range have been reported. For Aldolase Performance Goals
reasons of ease of use, good reproducibility, and Evaluation studies indicate total imprecision of
adaptability to semiautomation the method of Bergmeyer approximately 10% to 15% for aldolase activities in the
and Bernt is the method of choice. normal range and approximately 5% in the abnormal
range when the enzyme is measured using the method of
Specimen Bergmeyer and Bernt.
The anticoagulants oxalate, citrate, fluoride, heparin, and
EDTA reportedly have no effect on aldolase activity References
[15]. Whether serum or plasma is used, it is essential to 1 Penhoet EE, Kochman M, Rutter WJ.
remove the specimen from contact with the red blood Molecular and catalytic properties of aldolase
cells and platelets. In general, plasma is preferred over C. Biochemistry 1969; 8:4396-4402.
serum because of the release of aldolase from platelets 2 Bruns FH. Bestimmung und Eigenschaften der
during the clotting process. If refrigerated specimens are Serumaldolase. Biochem Z 1954; 325: 156-162.
not removed from the clot for 24 hours, the aldolase 3 Tzvetanova E. Aldolase isoenzymes in patients
level will be 12% to 46% greater than that of a properly with progressive muscle dystrophy and in
treated specimen [16]. The stability of the enzyme at human fetuses. Clin Chem 1971; 17: 926-930.
room temperature is reported as 5 hours to 2 days, 4 Sibley JA, Lehninger AL. Determination of
depending on the study. Boric acid powder, added aldolase in animal tissues. J Biol Chem 1949;
directly to serum to achieve a concentration of 25 g/L, 177: 859-872.
extends the stability of the enzyme at room temperature 5 Fleischer GA. Aldolase. In American
to 6 days [17]. The enzyme is stable for at least 5 days Association of Clinical Chemists: Standard
when kept refrigerated at 4 C.[18] The activity of methods of clinical chemistry, New York, 1961,
aldolase remains unchanged for at least 6 months if the Academic Press, Inc., vol. 3, pp. 14-22.
specimen is kept at -20 C [19]. 6 Beisenherz G, Boltze HJ, Bucher T et al.
Diphosphofructose-Aldolase,
Interferences Phosphoglyceraldehyde-Dehydrogenase,
Aldolase is present in high concentrations in Milchsaure-Dehydrogenase, Glycerophosphat-
erythrocytes and platelets. Because the concentration of Dehydrogenase and Pyruvat-Kinase aus
aldolase in red blood cells is 10 times that in serum, Kaninchenmuskulatur in einem Arbeitsgang. Z
hemolysis is a cause for rejection of the specimen [18]. Naturforsch 1953; 8b: 555-577.
7 Pinto PVC, Kaplan A, Van Dreal PA. Aldolase.
Aldolase Reference Interval II. Spectrophotometric determination using an
Aldolase exhibits sex- and age-related differences. At ultraviolet procedure. Clin Chem 1969; 15: 349-
birth, activities are approximately twice those seen in 360.
adults. Physical activity also influences aldolase 8 Bruns FH, Bergmeyer H-U. Fructose-1,6-
activities with physically active individuals showing diphosphate aldolase. In Bergmeyer, H.-U.,
higher activities. Reference intervals for healthy adults editor: Methods of enzymatic analysis, ed. 2,
are as follows: New York, 1974, Academic Press, Inc., vol. 2,
Males 2.61 to 5.71 U/L [16] pp. 724-731.
Females 1.98 to 5.54 U/L 9 Ludvigsen B. DPNH method for the estimation
of serum aldolase activity. J Lab Clin Med
Interpretation 1963; 61: 329-337.
It has been suggested that aldolase is useful in 10 Bergmeyer H-U, Bernt E. Fructose-1,6-
differentiating muscle disease from other diseases [20]. diphosphate aldolase, UV assay, manual
Aldolase, however, is present in many tissues [21] and as method. In Bergmeyer, H.-U., editor: Methods
a result has not been found to be specific for muscle of enzymatic analysis, ed. 2, New York, 1974,
disorders. In a multivariate analysis of plasma-enzyme Academic Press, Inc., pp. 1100-1105.
profiles in severe head injury, aldolase in conjunction 11 Harjanne A. The kinetic measurement of serum
with aspartate aminotransferase, creatine kinase, and Aldolase. Clin Chim Acta 1979; 92: 311-313.
lactate dehydrogenase can be used to discriminate
83
Aldolase
12 Asaka M, Nagase K, Miyozaki T et al. Tables

Radioimmunoassay for human aldolase A. Clin
Chim Acta 1981; 117: 289-296. Table 1: Methods of serum aldolase (ALS) analysis
13 Wilson VJC, Thompson RJ. Human brain
aldolase C4 isoenzyme: purification,
radioimmunoassay, and distribution in human Method 1: NADH formation; end-point
tissues. Ann Clin Biochem 1980; 17: 114-121. spectrophotometric
14 Aldolase. In American Association of Clinical Principle of analysis:
Chemists: Standard methods of clinical FDP DAP + GAP
chemistry, New York, 1961, Academic Press, GAP TPI DAP
Inc., vol. 3, pp. 14-22.
15 Ladenson JH. Nonanalytical sources of GAP + NAD+ GAPDH
+ AsO2
variation in clinical chemistry results. In
Sonnenwirth, A.C., and Jarett, L., editors: 3-Phosphoglycerate + NADH + H
Gradwohls clinical laboratory methods and Comments: Manual; subject to large
diagnosis, ed. 8, St. Louis, 1980, The C.V. interferences
Mosby Co., vol. 1, pp. 153-154. Method 2: NADH degradation
16 Sweetin JC, Thomson WHS. Revised normal Principle of analysis:
ranges for six serum enzymes: further statistical FDP DAP + GAP
analysis and the effects of different treatments GAP TPI DAP
of blood specimens. Clin Chim Acta 1973; 48:
49-63. 2 DAP + 2 NADH + 2 H+ GDH
17 Beardslee R, Owers P. Stabilization by boric 2 Glycerol-1-phosphate + 2 NAD+
acid of aldolase activity at room temperature. Comments: Manual or automated; accurate,
Clin Chem 1976; 22: 1543-1545 (letter). precise, flexible, and easy to use
18 Demetriou JA, Drewes PA, Gin JB. Enzymes. *DAP, Dihydroxyacetone phosphate; FDP, fructose
In Henry, R.J., Cannon, D.C., and Winkelman, diphosphate; GAP, glyceraldehyde phosphate; GAPDH,
J.W., editors: Clinical chemistry: principles and glyceraldehyde phosphate dehydrogenase; GDH,
techniques, Hagerstown, MD., 1974, Harper & glycerol-3-phosphate:NAD 2-oxidoreductase; NAD+,
Row, Publishers, Inc., pp. 971-974. nicotinamide adenine dinucleotide; NADH, reduced
19 Kaldor J, Schiavone DJ. Automated procedure NAD; TPI, triosephosphate isomerase.
for serum aldolase estimation. Clin Chem 1968;
14: 735-739.
20 Jacobs DS, Masten BL Jr, Demott WR,
Wolfson WL. Laboratory test handbook with Table 2: Reaction conditions for aldolase (ALS)
DRG index, Mosby/Lexi-comp, St. Louis, analysis*
1984, The C.V. Mosby Co., pp. 33-34. Temperature: 37 C
21 Kaplan LA, Pesce AJ. Clinical chemistry: pH: 7.4
theory, analysis, and correlation. St. Louis, Final concentration of reagent components:
1984, The C.V. Mosby Co., p. 956. Collidine: 50 mmol/L
22 Bourguigrat A, Ferard G, Jung G et al. Iodoacetate: 0.27 mmol/L
Multivariate analysis of plasma enzyme profiles FDP: 3.6 mmol/L
in severe head injury. Clin Chem 1983; 29: 107- NADH: 2.7 mmol/L
109. GDH: >3 U/mL
23 Velazquez FR, Ng RH, Statland BE. Clinical TPI: >15 U/mL
utility of serum aldolase measurements. Clin LD: >9 U/mL
Chem 1985; 31: 1004. Sample volume: 200 L
24 Bergmeyer H-U. Neue Werte fur die molaren Fraction of sample volume: 0.07
Extinktion-Koeffizienten von NADH and Linearity: 27.5 U/L
NADHP zum Gebrauch im Routine- Precision (%CV): 2.7%12.4%
Laboratorium. Z Klin Chem Z Klin Biochem *From Bergmeyer, H.-V., and Bernt, E.: In Bergmeyer,
1975; 13: 507-508. H.-V., editor: Methods of enzymatic analysis,
ed. 2, New York, 1974, Academic Press, Inc.
Sources of analytical error: any specimen
with visible hemolysis will give elevated results
because of release of aldolase from cells.
Serum stored on the clot gives similar results.
LD, Lactate dehydrogenase.
84
Aldolase
Procedure: Coupled Assay of Bergmeyer and Bernt Calculations

[10] With use of the molar absorptivity () of NADH as
6300 L cm
Principle mole
FDP ALS GAP + DAP
The formula for calculation of enzyme activity is as
GAP TPI DAP
follows:
2 DAP + 2 NADH + 2 H+ GDH
2 glycerol-1-phosphate + 2 NAD+
Lactate dehydrogenase (LD) is included in the
reagent to remove the interference of endogenous
pyruvate. The oxidation of NADH is followed by
absorbance measurement at 340 nm.
where A = difference between A1 and A2 over analysis
Reagents time, T minutes
1. Buffer/substrate solution. Dissolve 220 mg of = molar absorptivity of NADH [24]
Na3FDP 8H2O or 370 mg of tricyclohexylammonium
10 6 = factor to convert concentration to micromoles
FDP 10H2O and 6.2 mg of iodoacetic acid in 90 mL of per liter
distilled water. Add 0.75 mL of collidine, mix, and For the assay described, the values are as follows:
adjust the pH to 7.4 with approximately 0.6 mL of 5 M
HCl. Dilute with distilled water to 100 mL. Final U/L = _A_ 2.76 mL _1__ 106 = A 55
concentrations are 55 mmol collidine buffer per liter at 20 min 0.2 mL 6300 2
pH 7.4, 0.3 mmol iodoacetate per liter, and 4 mmol FDP
per liter. This solution is stable for 4 weeks at 2 to 8 C. The factor of 2 indicates that for every molecule
2. NADH solution. Dissolve 25 mg of of FDP catalyzed, two molecules of NADH are
Na2NADH and 20 mg of NaHCO3 in 2 mL of distilled
converted to NAD+.
water. Final concentration is 15 mmol/L. This is stable
for 4 weeks at 2 to 8 C. Interferences
3. GDH/TPI/LD suspension. Stock solutions of Hemolysis is cause for rejection.[18]
the enzymes, as suspensions in ammonium sulfate, are Also, refrigerated serum not removed from the clot will
available from Boehringer Mannheim Biochemicals have an aldolase level 12% to 46% greater than that of a
(Indianapolis, IN). Using 3.2 mol of ammonium sulfate properly treated specimen.[16]
per liter, mix enzyme solutions so that final
concentrations are >75 U/mL of GDH per milliliter (25
C) (source: rabbit muscle), >500 U of TPI per milliliter
(25 C) (source: rabbit muscle), >233 U of LD per
milliliter (25 C) (source: rabbit muscle). This is stable
at 2 to 8 C for about 1 year. Suspend thoroughly before
use.
Assay
Equipment: 37 C constant-temperature
incubator and spectrophotometer (band pass <10 nm)
capable of reading at 340 nm.
1. Place 2.50 mL of buffer/substrate solution, 0.05
mL of NADH solution, 0.01 mL of
GDH/TPI/LD suspension, and 0.20 mL of
serum or plasma in a 3.0 mL cuvette.
2. Mix well, and incubate at 37 C for about 5
minutes.
3. Read the initial A340 (A1).
4. After exactly 20 minutes at 37 C, read the final
A340 (A2).
5. If the A (A1 - A2) is greater than 0.500, dilute
specimen five- to tenfold with isotonic saline,
and repeat the analysis using the same volumes
of reagents and specimen.
85
Aldosterone
Aldosterone
Hassan M.E. Azzazy
Name: Aldosterone
Clinical significance: Refer to Chapter 51, Adrenal Hormones and Hypertension, in the 5th edition of
Clinical Chemistry: Theory, Analysis, Correlation.
Molecular mass: 360.4 D
Chemical class: Mineralocorticoid
Principles of Analysis and Current Usage chromatography, or prefractionation step, which are
Aldosterone is the most important human relatively complex and time consuming (Table 1,
mineralocorticoid; it is secreted by the zona glomerulosa of Methods 1 and 2) [7]. Average salivary aldosterone
the adrenal cortex and stimulates sodium transport across values are almost a third of those in plasma, and an
cell membranes in the distal renal tubules. Aldosterone RIA was also developed for aldosterone determination
plays a major role in homeostasis of sodium and potassium in saliva [8,9].
and maintenance of arterial blood pressure in the sodium-
depleted state. The renin-angiotensin system regulates Aldosterone measurement in urine is performed using
aldosterone secretion. Low sodium concentration or low the same testing methodologies, but a preassay acid-
blood volume causes release of renin from kidney cells, hydrolysis step is required. This is because the term
which mediates release of aldosterone. urinary aldosterone actually refers to the 18-
glucuronide conjugate of aldosterone (10% of all
Aldosterone exists in picomolar concentrations in serum, urinary aldosterone metabolites), and acid hydrolysis
and thus sensitive assays are necessary for its is required to form free aldosterone. Only 2% of
measurement. Several methods are available for aldosterone is excreted in the free form [4].
determination of aldosterone in blood and urine (Table 1).
These include immunoassays, isotope-dilution gas Enzyme immunoassays are available for aldosterone,
chromatography/mass spectrometry ID-(GC/MS) [1], high- including competitive-binding assays to monoclonal
performance liquid chromatography-tandem mass antibodies and direct-detection assays using antibodies
spectrometry (HPLC-MS/MS) [2], and liquid conjugated to horseradish peroxidase, with the
chromatography-tandem mass spectrometry (LC-MS/MS) detection limits being in the picomolar range [9,10]. A
[3]. competitive-binding chemiluminescence immunoassay
has been developed. The assay utilizes a biotinylated
Radioimmunoassay (RIA) has been the most common mouse monoclonal anti-aldosterone antibody as the
assay for aldosterone measurement since the early 1970s capture reagent and acridinium ester-labeled
[4]. The antibodies used were typically polyclonal, raised aldosterone as a tracer (Table 1, Method 3) [11].
in rabbits and generated against an aldosterone-3-(O-
carboxymethyloxime) bovineserum albumin conjugate Liquid chromatographytandem mass spectrometry
[3]. Commercial immunoassays use different approaches to (LC-MS/MS) is currently being used for aldosterone
separate bound from free aldosterone: a solid-phase first determination in both blood and urine. Current LC-
antibody, a solid-phase second antibody, an accelerated MS/MS methods use atmospheric-pressure chemical
liquid-phase second antibody, dextran-coated charcoal, or a ionization or photospray ionization along with
polyethylene glycol precipitant. RIA methods with and multiple steroid profiling [12,13]. A high-performance
without extraction steps are being employed [5,6]. The liquid chromatographyatmospheric pressure chemical
extraction of aldosterone prior to analysis may be done into ionizationtandem mass spectrometry (HPLC-APCI-
dichloromethane, which improves the specificity of the MS/MS) method was proposed as a reference method
assay, since aldosterone concentrations in blood are in the for aldosterone quantification in serum and plasma
picomolar range. Many patients being investigated have (Table 1, Method 4). In this method, extraction was
some degree of renal failure, and patients with chronic performed using dichloromethane/diethyl ether
renal failure have polar aldosterone metabolites present in containing flumethasone as internal standard. A
high concentrations in plasma; these cross-react with phenyl column was used for chromatography, and the
aldosterone antibodies. Thus solvent extraction helps to mobile phase was 50 mM ammonium formate (pH
increase assay specificity by removing these metabolites. 7.1)/methanol (50/50, v/v). The reported accuracy of
Currently a large portion of RIAs for aldosterone this method ranged from 93.1% to 98.9% [2]. In other
measurement do not involve an extraction, LC-MS/MS methods, deuterated aldosterone
(aldosterone-d6) is added to samples as an internal
control. Aldosterone-d6 and endogenous aldosterone
i
Aldosterone are extracted from serum or plasma using a Strata X
New method: cartridge, and the eluate is dried and then reconstituted
Fifth edition: Hassan M.E. Azzazy with 70/30 methanol/water containing estriol before
analysis on LC-MS/MS [13]. Recently, an LC-MS/MS
86
Aldosterone
method for aldosterone measurement with electrospray assay. Results of urinary aldosterone depend on the
ionization has been developed, with a lower limit of amount of sodium in the diet, where levels are lower
detection of 30 pmol/L and interassay coefficients of with greater intake of sodium. For adults with normal,
variation (CVs) between 4.3% and 7.5% at aldosterone low-sodium, and high-sodium diets, reported urinary
concentrations of 97 to 993 pmol/L (Table 1, Method 5) aldosterone levels are 2 to 21, 17 to 44, and 0 to 14
[12]. g/day, respectively. If the 24-hr urinary sodium
excretion is > 200 mEq/day, urinary aldosterone
Reference and Preferred Methods should be < 10 g [16,17].
ID-GC/MS is the reference method for determination of
aldosterone concentration. Although this technique
provides reliable results, methods utilizing GC/MS require
a labor-intensive and time-consuming sample Plasma/Serum Aldosterone Reference Intervals
derivatization step. Liquid chromatography tandem MS (RIA)
(LC-MS/MS) has recently been proposed to replace Age Position Reference SI Units
GC/MS to circumvent the laborious derivatization [12]. Interval (pmol/L)
Immunoassays are, however, favored for measuring (ng/dL)
aldosterone in large numbers of small-volume samples. 1-12 5-90 139-2493
Immunoassays may be problematic unless extracted months
because of low aldosterone concentration and cross- 1-2 years 7-54 194-1496
reactivity. 2-10 Supine 3-35 83-970
years
Specimen 2-10 *Upright 5-80 139-2216
Serum, plasma, and urine specimens can be used. For years
serum, venous blood is to be collected in a glass tube with 10-15 Supine 2-22 55-609
no additives; plasma can be collected in a glass or plastic years
tube with the anticoagulants heparin or EDTA. Serum or 10-15 Upright 4-48 111-1330
plasma should be stored frozen and are stable for up to years
years at 20C in an airtight container. >15 years Supine 3-16 83-443
>15 years Upright 7-30 194-831
For urine analysis, 24-hour urine collection is required, and Adrenal 200-800 5540-
the specimen should be refrigerated during collection. vein 22,160
Acetic acid (50% preferred) is added after completion of Data from Endocrine Sciences. Tarzana, CA: Pediatric
collection to achieve a pH of 2.0 to 4.0. Other allowed Laboratory Services; 1992.
preservatives include thymol and boric acid. Strong * Patient should be seated or standing for at least 2
mineral acids should not be used. Samples may be stored hours prior to collection of upright specimen.
frozen.
Interpretation
Elevated concentrations of aldosterone are observed in
Interferences primary aldosteronism due to aldosterone-secreting
Cross-reactivity with other steroids such as 5- adenomas (Conns syndrome). Primary aldosteronism
dihydroaldosterone and corticosterone depends on is characterized by suppressed renin activity and
antibody specificity, which varies in different cannot be stimulated by either sodium restriction or
immunoassays. Grossly hemolyzed or lipemic samples or treatment with a diuretic, or by demonstrating lack of
contamination of the sample or tube with 125I or other suppression of aldosterone following saline infusion or
radioisotopes would interfere with RIA. Also, administration of a mineralocorticoid. Pseudoprimary
pharmacological interference was observed in women aldosteronism is due to bilateral adrenal hyperplasia.
taking drospirenone, a synthetic progestin with anti- Sampling of adrenal venous renin and aldosterone can
mineralocorticoid activity and aldosterone-receptor be used to differentiate between adenomas and
antagonism. This drug was reported to interfere with hyperplasia as the cause of hyperaldosteronism. In
laboratory screening and confirmatory testing for hyperplasia, both adrenals secrete high levels of
diagnosing primary aldosteronism [15]. Angiotensin, aldosterone. Secondary aldosteronism may be
estrogens, laxatives, oral contraceptives, sodium observed in laxative abuse, cardiac failure, diuretic
restriction, and thiazide diuretics lead to elevated abuse, and Bartters syndrome (a rare, inherited defect
aldosterone concentration. Aminoglutethimide, ACE in the thick ascending limb of the loop of Henle).
inhibitors such as captopril, lisinopril, deoxycorticosterone,
prolonged heparin therapy, and saline decrease aldosterone Aldosterone concentration is important in the
concentration. assessment of persons with primary hypertension.
Primary hyperaldosteronism has recently become
accepted as a more frequent cause of hypertension
Reference Intervals [18]. On the other hand, low aldosterone concentration
Normal value ranges vary among different laboratories with hypertension is associated with a number of
using different assays. It is recommended that each conditions such as Turners syndrome, diabetes
laboratory establish reference ranges for their particular mellitus, acute alcoholic intoxication, and excess
87
Aldosterone
secretion of deoxycorticosterone. Low concentrations 4 Jaffe BM, Behrman HR, eds. Methods of
without hypertension are observed in Addisons disease Hormone Radioimmunoassay. New York:
(primary adrenal insufficiency) and in the syndrome of Academic Press; 1979.
hypoaldosteronism due to renin deficiency. 5 Bayard F, Beitins IZ, Kowarski A, Migeon
CJ. Measurement of aldosterone secretion
Measurement of aldosterone concentration by itself is of rate by radio-immunoassay. J Clin Endocrinol
little value, and irregularities in blood pressure and/or Metab 1970;31:507-11.
disorders of sodium and potassium homeostasis are best 6 Jowett TP, Slater JD, Piyasena RD, Ekins RP.
investigated by assessing the integrity of the renin- Radioimmunoassay of aldosterone in plasma
angiotensin-aldosterone axis [11]. This is usually done by and urine: validation of a novel separation
measuring renin and aldosterone concentrations. Although technique and a rapid urine assay. Clin Sci
aldosterone measurements alone are difficult to interpret, Mol Med 1973;45:607-23.
they may be useful for assessing response of the adrenal 7 Stowasser M, Gordon RD. Aldosterone
cortex to stimulation. The aldosterone-to-renin ratio (ARR) assays: an urgent need for improvement. Clin
has been proposed as a screening tool for primary Chem 2006;52:1640-2.
aldosteronism. The clinical utility of the ARR remains 8 Atherden SM, Corrie JE, Jones DB, Al-
questionable because of many factors that can affect ARR Dujaili EA, Edwards CR. Development and
results. These include posture, time of blood collection, application of a direct radioimmunoassay for
and use of antihypertensive medication. A number of ARR aldosterone in saliva. Steroids 1985;46:845-
cutoff values have been published, owing to differences in 55.
populations studied, collection methods, and hormone 9 Hubl W, Taubert H, Freymann E, Hofmann
assays used. There is no agreed-upon ARR cutoff value, F, Meissner D, Garten CD et al. A simple
and it is unknown which population should be screened solid-phase enzyme immunoassay for
[19,20]. The urinary aldosterone-to-active-renin ratio has aldosterone in plasma and saliva. Exp Clin
been suggested as the best independent predictor of cure of Endocrinol 1983;82:188-93.
hypertension after adrenalectomy in patients with 10 Hanquez C, Rajkowski KM, Desfosses B,
aldosterone-producing adenomas [21]. Cittanova N. A competitive microtitre plate
enzyme immunoassay for plasma aldosterone
Performance Goals using a monoclonal antibody. J Steroid
Clinical Laboratory Improvement Amendments acceptable Biochem 1988;31:939-45.
performance criteria (CLIA 88) for measurement of 11 Schirpenbach C, Seiler L, Maser-Gluth C,
aldosterone require that laboratories be accurate to within Beuschlein F, Reincke M, Bidlingmaier M.
3 SD of the peer-group mean. Survey data from the Automated chemiluminescence-immunoassay
College of American Pathologists 2007 participant for aldosterone during dynamic testing:
summary report show imprecision values (% CV) for comparison to radioimmunoassays with and
extracted methods of 14.9% at a mean concentration of without extraction steps. Clin Chem
50.2 g aldosterone/L and 20.3% at a mean concentration 2006;52:1749-55.
of 14.8 g/L [22]. 12 Turpeinen U, Hmlinen E, Stenman UH.
Determination of aldosterone in serum by
Within-subject and between subject biological variation liquid chromatography-tandem mass
were 29.4% and 40.1% (serum) and 32.6% and 39% spectrometry. J Chromatogr B Analyt
(urine), respectively. Desirable specifications for analytical Technol Biomed Life Sci 2008;862:113-8.
imprecision derived from studies of biological variation 13 Mayo Medical Laboratories Test Catalog.
indicate an assay imprecision of no greater than 14.7% Available at
(serum) and 16.3% (urine) and a bias of 12.4% (serum) and <http://www.mayomedicallaboratories.com/te
12.7% (urine) [23]. st-catalog/>
14 Perschel FH, Schemer R, Seiler L, Reincke
References M, Deinum J, Maser-Gluth C et al. Rapid
1 Stckl D, Reinauer H, Thienpont LM, De screening test for primary
Leenheer AP. Determination of aldosterone in hyperaldosteronism: ratio of plasma
human serum by isotope dilution gas aldosterone to renin concentration determined
chromatography/mass spectrometry using a new by fully automated chemiluminesc-ence
heptafluorobutyryl derivative. Biol Mass immunoassays. Clin Chem 2004;50:1650-5.
Spectrom 1991;20:657-64. 15 Pizzolo F, Pavan C, Corrocher R, Olivieri O.
2 Fredline VF, Taylor PJ, Dodds HM, Johnson AG. Laboratory diagnosis of primary
A reference method for the analysis of aldosterone aldosteronism and drospirenone-
in blood by high-performance liquid ethinylestradiol therapy. Am J Hypertens
chromatographyatmospheric pressure chemical 2007;20:1334-7.
ionization-tandem mass spectrometry. Anal 16 Bravo EL. Primary aldosteronism. Issues in
Biochem 1997;252:308-313. diagnosis and management. Endocrinol
3 Cawood ML. Measurement of aldosterone in Metab Clin North Am 1994;23:271-83.
blood. Methods Mol Biol 2006;324:177-85. 17 Young WF Jr. Pheochromocytoma and
primary aldosteronism: diagnostic
88
Aldosterone
approaches. Endo-crinol Metab Clin North Am 21 Mourad JJ, Girerd X, Milliez P, Lopez-Sublet
1997;26:801-27. M, Lejeune S, Safar ME. Urinary
18 Mulatero P, Dluhy RG, Giacchetti G, Boscaro M, aldosterone-to-active-renin ratio: a useful tool
Veglio F, Stewart PM. Diagnosis of primary for predicting resolution of hypertension after
aldosteronism: from screening to subtype adrenalectomy in patients with aldosterone-
differentiation. Trends Endocrinol Metab producing adenomas. Am J Hypertens
2005;16:114-9. 2008;2:742-7.
19 Giacchetti G, Ronconi V, Lucarelli G, Boscaro M, 22 College of American Pathologists. 2007
Mantero F. Analysis of screening and Survey Participant Summary Report.
confirmatory tests in the diagnosis of primary Northfield, IL: CAP; 2007.
aldosteronism: need for a standardized protocol. J 23 Rics C, Alvarez V, Cava F, Garca-Lario JV,
Hypertens 2006;24:737-45. Hernndez A, Jimnez CV et al. Current
20 Jansen PM, Boomsma F, van den Meiracker AH. databases on biologic variation: pros, cons
Aldosterone-to-renin ratio as a screening test for and progress. Scand J Clin Lab Invest
primary aldosteronism: the Dutch ARRAT Study. 1999;59:491-500.
Neth J Med 2008;66:220-8.
Table 1: Characteristics of Selected Aldosterone Assays

*Method 1: DSL Active Aldosterone (non-extraction)
Sample (volume): Serum/plasma (100 L)
Detection: Polyclonal antibody, 125I tracer
Range (ng/L): 2-1600
Intraassay variability (%): 3.6-8.3
Interassay variability (%): 7.3-10.4
Reference interval (ng/L): 30-340 (serum); 30-220 (plasma)
*Method 2: Adaltis Aldosterone Maia (non-extraction)
Detection: Polyclonal antibody, 125I tracer
Range (ng/L): 6-2500
Reference interval (ng/L): 70-350 (plasma)
*Method 3: Nichols Advantage Aldosterone [14]
Detection: Monoclonal antibody, chemiluminescence
Range (ng/L): 15-1200
Reference interval (ng/L): 38-313 (serum)
Method 4: High-performance liquid chromatography atmospheric pressure chemical ionization
tandem mass spectrometry (HPLC-APCI-MS/MS) [2]
Sample: Serum/plasma
Limits of detection and quantification (pg/mL): 10 and 15
Linear range (pg/mL): 15-500
Assay imprecisions at 15, 20, 150, and 450 pg/mL were 18.5%, 8.8%, 10.6%, and 9.5%
Method 5: Liquid chromatographytandem mass spectrometry (LC-MS/MS) [12]
Sample (volume): Serum (500 L)
Lower limit of detection (pmol/L): 30
Linear range (pmol/L): 60-3000
* Commercially available immunoassays.
89
Aldosterone
90
Alkaline Phosphatase - Total

Danyal B. Syed
Name: Alkaline phosphatase, ALP, AP, ortho-phosphoric monoester
phosphohydrolase (alkaline optimum)
Clinical significance: Refer to Chapter 31, Liver Function, in the 5th edition of Clinical Chemistry:
Molecular weight: Varies with tissue source of enzyme, ranges from 70,000 to 120,000 D
Chemical class: Enzyme, protein
Known enzyme forms: Bone, liver, placenta, placental-like, intestine, kidney, neutrophil, high
molecular weight, Regan (fetal), Nagao, Kasahara, hepatoma
Number of subunits: Two
Chromosomal location: Tissue-specific ALP = 2q37.1 [1] Tissue nonspecific ALP = 1p34-36
[2]
Number of mutations : Tissue nonspecific ALP (TNSALP) as of January 7, 2008 = 194 [3]
Number of polymorphisms: 12 [3]
R2-OH + R1-O-P-O3- ALP, pH>9 R1-OH + R2-HPO4-
Net reaction is hydrolysis if R2OH is H2O
Principles of Analysis and Current Usage and Zn2+ ions for stability and maximum activity. It is
The group of nonspecific phosphatases that catalyze the inhibited by Ca2+ and inorganic phosphate,
reaction shown above are collectively known as alkaline monoethanolamine, Be2+, chelating agents of bivalent
i metal ions (such as EDTA, oxalate, citrate, cysteine, and
phosphatase (ALP)
. Phosphatases transfer a phosphate moiety from one histidine), non-alkaline pH, phenylalanine, tryptophan,
group to a second, forming an alcohol and a second L-homoarginine, urea, iodoacetamide, and high
phosphate compound. When water is the phosphate concentrations of Zn2+.
acceptor, inorganic orthophosphate is formed.
Reaction rates are dependent on the incubation
The earliest assays for total serum ALP measured the temperature, the tissue source of the enzyme, and the
release of inorganic phosphate. Newer ALP methods use type of substrate and buffer used. ALP is denatured
self-indicating, chromogenic substrates and rate- slowly at temperatures above 40C. All isoenzymes of
enhancing buffers, leading to significant improvements ALP except placental and germ cell ALPs lose their
in analytical sensitivity and precision. The older ALP activity rapidly at temperatures above 60C. Since the
assays are now of historical interest only [4-9]. These physiological substrates of ALP are unknown, most of
have been extensively reviewed by McComb et al. [10], the assays used in clinical laboratories for ALP employ
and more recently Millan [11] has discussed (in a p-nitrophenyl phosphate (pNPP) as the substrate. The
general manner) various methodologies available for Km value of ALP for pNPP is dependent upon a variety
assaying ALP activity. These include spectrophotometric of factors, including the source of the enzyme, pH and
(including inhibition-based), amperometric, nature of the buffer, temperature, and ionic strength. At
potentiometric, electrophoretic, and high-performance an alkaline pH, pNPP is colorless, but the reaction
liquid chromatography (HPLC) assays. The reader is product p-nitrophenol (pNP) is intensely yellow with a
also referred to the earlier edition of this book for critical molar absorptivity of about 18,450 Lmol-1 cm-1 at
review of the procedures that were employed in the 403 nm (Figure 1).
clinical laboratory in the past.
The optimal reaction pH is between 9 and 10.5 and Transphosphorylating buffers greatly increase the rate of
varies with the buffer and substrate. ALP requires Mg2+ reaction compared to the barbital, carbonate, or glycine
buffers used in the older methods [4-6]. The most
commonly used transphosphorylating buffers are 2-
i
Alkaline Phosphatase amino-2-methyl-1-propanol (AMP), diethanolamine
Previous and current authors of this method: (DEA), Tris, and N-methyl-D-glucamine (MEG) [12].
First edition: William C. Wenger, John A. Lott Mannitol also acts as a transphosphorylating phosphate
Methods edition: Steven C. Kazmierczak, John A. Lott acceptor and is often used as an accelerator in some
Second edition: Steven C. Kazmierczak, John A. Lott methods for ALP. In U.S. patent filing 6713275 [13],
Third edition: Julie Raymond-Habecker, John A. Lott Weisheit, Ralph and Treiber,Wolfgang describe a
Fourth edition: Julie Raymond-Habecker, John A. Lott method for the determination of ALP activity using
Fifth edition: Danyal B. Syed pNPP as substrate but at a wavelength of 450 10 nm
91
instead of 410 5 nm in combination with a rate blank have measurable traceability. Traceability in itself is the
procedure to eliminate hemoglobin interference, which most important means of achieving standardization in
also absorbs around 415 nm. the clinical laboratory field, with the goal of having
comparable results regardless of the method, the
Lack of agreement between widely used ALP methods is measurement procedure (test kit) and the location of the
common [14,15]. While this might be due to different laboratory. International Standards Organization (ISO)
temperature conditions in different assays, it was [21], the European Union [22], and CLSI all have joined
observed that AMP activates intestinal ALP more than it together to achieve this goal. Currently there is no IFCC
does liver or bone ALP. Such findings have clinical primary method available for ALP. However, most of
implications when patients have a high serum intestinal the instruments and kit manufacturers have incorporated
ALP, such as patients on chronic hemodialysis or those the 37C temperature for the ALP activity measurement
with liver cirrhosis [15]. in their procedures as proposed. Some of the instrument
and ALP kit manufacturers, on their own, have already
Current Assay Methods for Measurement of ALP begun using the calibrators and reference materials to
The current methods being used in clinical laboratories standardize ALP assays. Schiele et al. [23] reported an
around the world are all spectroscopic methods; they IFCC-certified procedure for preparing a lyophilized
include both colorimetric and fluorometric assays. The preparation of ALP that can be used as an enzyme
colorimetric methods using p-NPP as substrate are reference substance. The enzyme was partially purified
widely used in the clinical laboratory, and the from pig kidney to an activity of 400 U/mg of protein
fluorometric methods using 4-methyllumbelliferyl while largely free from contaminating enzyme activities.
phosphate [16], FDP (3,6 fluorescein diphosphate) [17], The ALP was lyophilized in a matrix also containing
and other such compounds as substrates have gained bovine serum albumin, MgCl2, ZnCl2, and NaCl. Vial-
popularity in research and development, especially in to-vial catalytic concentration variability of the final
ALP-conjugated ELISA assays and in the detection of product was < 1% and the predicted annual loss of
immobilized amplified products in one phase system activity was < 0.01% at 20C and < 0.04% at 4C. A
(DIAPOPS) [18]. patent has been recently granted for a tissue-nonspecific
alkaline phosphatase conjugate for use as a suitable
There has not been much change in the current assay reference or standard ALP material [24]. In this
methods for the measurement of ALP in the past 3 preferred embodiment, a tissue-nonspecific ALP (tns-
decades. The latest emphasis is on the standardization AP), which can be obtained by recombinant expression
and traceability of the assays. McComb and Bowers [19] of nucleic acid coding for the AP in a prokaryotic cell,
suggested that the variation in the reported incompatible preferably Escherichia coli, is used as an unglycosylated
numerical data for ALP found in large interlaboratory tns-AP to produce the conjugate, with the dextran having
surveys could be reduced from a coefficient of variation a molecular weight of 10 to 500 kDa. The claimed
(CV) of 25% to 30% to a CV of as low as 2% by advantages of this invention are its reproducibility, better
expressing all numerical values on a single scale: the stability, and absence of risk of infection from the blood-
International Clinical Enzyme Scale (ICES). The ICES borne pathogens.
for ALP requires a well-defined reference system that
relates the IFCC Reference Method for ALP to Reference and Preferred Methods
numerous stable primary and secondary ALP reference Currently there is no formal reference method for the
materials. Thirteen years later at an IFCC General determination of ALP activity. The kinetic pNPP method
Conference in Sevilla, Spain, it was decided to establish of Bowers and McComb [25,26] is a candidate reference
a global reference system for the measurement of method for the Clinical Laboratory Standards Institute
catalytic concentrations of enzymes comprising the Expert Panel on Enzymes (CLSI/EPE). It has also been
following elements [20]: adopted by the World Health Organization for use as
Reference Measurement Procedures: Use the Guidelines on Standard Operating Procedures for
existing 30C IFCC reference method as a basis for Clinical Chemistry: Alkaline Phosphatase p-
developing a set of standard operating procedures Nitrophenol Method [27]. The method offers the
(SOPs) for a reaction temperature of 37C. convenience and sensitivity of pNPP in AMP buffer, and
Network of Reference Laboratories: A group of bilirubin does not interfere. With careful work, within-
reference laboratories, including manufacturers day CVs of 3% to 6% have been reported [26].
laboratories, which will provide necessary expertise
and equipment to conduct measurements following The specific parameters of the assay are summarized in
the SOPs to a high metrological level. Table 1. The Scandinavian Society for Clinical
Reference Materials: The existing BCR reference Chemistry and Clinical Physiology [28] and The German
materials available from International Reference Society for Clinical Chemistry [29] have similar
Materials and Methods (IRRMEurope) are to be methods. The American Association for Clinical
recertified by the network reference laboratories Chemistry (AACC) reference method optimizes all
according to the cooperation contract between IFCC reaction conditions, including temperature, pH, reagent
and IRMM. concentrations, and sample volume fraction [30].
The main objective of introducing reference systems for Most of the clinical laboratories in the developed world
the measurement of analytes in a clinical laboratory is to are using automated procedures with reagents,
92
calibrators, and quality controls (in some cases) supplied significant concentrations of monoethanolamine (MEA),
and reaction conditions set by the manufacturers of the a potent inhibitor of ALP. Solutions of DEA can
instruments, thereby eliminating most sources of errors. deteriorate during storage, with the concomitant
formation of MEA. AMP also contains ALP inhibitors
The measured activity of ALP depends on the volume such as diamines and 5-amino-3-aza-2,2,5-
fraction of serum. Increases in activity, corrected for trimethylhexanol that inactivate the enzyme by binding
2+
dilution, have been observed when the serum fraction Zn . Some commercial pNPP substrate preparations
was decreased from 0.04 to 0.02; no further increases contain excessive amounts of pNP or inorganic
were seen below the 0.02 volume fraction. The cause of phosphate, or both, with the former producing high
this effect is unknown, but it may result from the blank absorbance and the latter inhibiting ALP. See also
dissociation of ALP aggregates at the greater dilutions. under Interpretation for interferences due to drugs.
The German Society for Clinical Chemistry proposed a Alkaline Phosphatase, Total - Reference Intervals
reference method for ALP in serum and plasma at 37C The serum reference intervals for healthy persons,
[31]. The method differs in that methylglucamine, a determined at 30C by the Bowers and McComb
substance with moderate phosphate-acceptor properties, method, are shown below [26,34]. Each laboratory must
is used as a buffer. Methylglucamine was proposed by
establish its own reference ranges either by direct data
Chromy et al. [32] and recommended by the Italian
collection or by performing correlation studies and then
Society [33] because of inhibitors in amino alcohol
buffers and the absence of same in methylglucamine. verifying the adopted reference ranges.
The reaction is started with the substrate, and linear
reaction rates prevail during prolonged measurements at ALP Reference Intervals [26]
37C. When compared to DEA, methylglucamine has a Group, ALP, Group, ALP,
favorable pK value (9.63 at 37C), has a low viscosity at Age in Years Up to U/L Age in Years
optimal buffer concentrations, and has approximately Females: Males:
equal reactivity to all human ALP isoenzymes. pNPP is Newborns 250 Newborns
used as the substrate because of the rapid enzymatic 1 to 9 350 1 to 19
hydrolysis, the high absorptivity of pNP at a pH of 10.0, 10 to 14 280 10 to 14
and a negligible isoenzyme bias. A reagent blank must 15 to 19 150 15 to 19
be used, because p-nitrophenol is generated 20 to 24 85 20 to 24
spontaneously at 37C. 25 to 34 85 25 to 34
35 to 44 95 35 to 44
Specimen 45 to 54 100 45 to 54
Blood should be drawn after a fast of at least 8 hours. 55 to 64 110 55 to 64
Serum and heparinized plasma give the same results. 65 to 74 145 65 to 74
Slight hemolysis is tolerable, but gross hemolysis should 75+ 165 75+
be avoided. Certain sample storage conditions tend to
increase serum ALP. There is a significant increase in Note: The values at 30C can be converted to values at
activity after warming of previously refrigerated or 37C by multiplying the result with a factor of
frozen sera. The ALP activity in fresh serum increases 1.344.
by up to 2% in 6 hours at 25C. Increases of up to 30% Physiological factors causing increased ALP activities
of ALP activity occur after frozen serum is thawed and include age, gender, race, body mass, food intake,
in lyophilized specimens after reconstitution. These smoking, and pregnancy.
increases may be due to the release of ALP from
complexes with lipoproteins or because the non- Interpretation
complexed enzyme has greater activity. It is best to The human ALPs (hALP) are found anchored on the cell
analyze ALP specimens the same day they are drawn. membrane by glycosylphosphatidylinositol. They are
Specimen can be drawn either in a plain red-top tube or released in the serum by the action of specific
speckled-red-top tube with gel as serum separator, phospholipases [35,36]. There are four isozymes: (1)
employing normal phlebotomy procedures. To avoid placental AP or hPLAP (human placental AP), (2) germ
contamination, do not draw tubes containing cell AP (GCAP or PLAP-like), (3) intestinal AP (IAP),
anticoagulants before the red-top/speckled-red-top tube. and (4) tissue-nonspecific AP (TNAP). Of these four,
PLAP and GCAP are the most heat stable at 65C, and
Interferences the bone AP component of TnAP the least. The
ALP is inhibited by metal-complexing anticoagulants. homology between PLAP and GCAP are about 98%,
All of EDTA, oxalate, and citrate inhibit the enzyme by whereas intestinal AP and TnAP exhibit 88% and 56%
complexing Mg2+ and Zn2+ and should not be used. homologies, respectively [37]. In nonsmoking healthy
Hemolyzed and lipemic specimens should be rejected if individuals, the PLAP and GCAP represent less than 1%
they have a high background absorbance. Bilirubin at of total AP activity in the serum. Smoking causes
concentrations of up to 20 mg/dL does not interfere. elevated levels of PLAP, which return to normal range
after 1 to 2 months of smoking cessation [38-40].
The purity of transphosphorylating buffer is of great Ectopically expressed PLAP has been associated with
importance. Sources of DEA have been found to contain cancer of the ovary, testis, lung, and colorectal tract [41].
93
The PLAP-like enzyme (GCAP), on the other hand, has Any drug that is hepatotoxic or induces cholestasis will
been associated with testicular cancers, seminoma, increase serum ALP, sometimes dramatically. Young
choriocarcinoma, and embryonal carcinoma. Between [50] lists approximately 250 drugs that increase serum
25% and 66% of patients with ovarian cancer and 22% ALP. Both increases and decreases of plasma ALP are
to 89% of patients with testicular carcinoma were found important clinically. Table 3 lists some of the disorders
to have elevated serum levels of PLAP and PLAP-like and diseases associated with abnormal serum activities
enzyme [38,42]. With 98% homology, these two of ALP [51]. ALP is a sensitive indicator of liver
isozymes are difficult to assay. Specific immunoassays obstruction; however, ALP is also increased in bone
have been developed for placental-like alkaline diseases. Frequently, ALP is measured along with other
phosphatase [43], which could be employed to monitor analytes considered more specific for liver diseases, such
tumor regression. as 5-nucleotidase or -glutamyl transferase, to permit
differentiation between liver and bone as the source of
Serum ALP activity primarily reflects changes in bone an increased serum ALP [51]. ALP is often interpreted
and liver function, even though higher ALP activities as abnormal, particularly in children and in geriatric
can be found in other organs, as depicted in Table 2. One patients, because of the use of inappropriate reference
fourth of all normal persons have low serum activities of intervals by a laboratory.
intestinal ALP at all times. Individuals with blood types
B and O exhibit increases in serum activity of intestinal Alkaline Phosphatase Performance Goals
ALP approximately 2 hours after eating a fatty meal. Acceptable performance (accuracy) for alkaline
Mutations in the TnAP gene have been associated with phosphatase assays according to the Clinical Laboratory
hypophosphatasia, a rare inherited disorder manifesting Improvement Amendments of 1988 (CLIA-88) is the
itself in poor bone mineralization and even death at a target value 30%, which is also the acceptability
very young age. The status of the fetus for risk of having criteria used by the College of American Pathologists
missense mutations leading to severe hypophosphatasia (CAP) in its surveys. A laboratory must score 80% in
can be assessed by: (i) by measuring the ALP activity in each testing event for each analyte and overall in all
the amniotic fluid supernatant and fetal serum [44], (ii) analytes [52].
by using the allele specific oligonucleotide mis-sense
mutational (previously known) probes on DNA extracted Within-day precision of the automated methods ranges
from the amniocytes (amniotic fluid cells) in the ALP from 2% to 6%. Long-term precision within most
gene and look for the same mutations [45]. laboratories is also acceptable. Over a 3-year period, a
Electrophoretic technique is most commonly used for CV of about 5% is typical for procedures using pNPP.
differentiating among the ALP isozymes and isoforms, The inter-laboratory precision of the methods is much
but assays based on heat stability of these enzymes are poorer. In a CAP interlaboratory proficiency survey in
also employed. As stated earlier, serum ALP levels are the past, ALP was among the least precise of the
affected by many drugs, physical conditions, herbal common enzyme tests [53,54]. Reported interlaboratory
medicines, food intake, smoking, alcohol intake, and CVs ranged from 17% to 28%. The factors contributing
pregnancy. Some 1000 drugs, herbs and physiological to poor interlaboratory precision are impure reagents,
conditions that affect ALP activity have been slight variations in the reaction temperature, subtle
documented [46]. The effects of some of the most enzyme activation and inactivation phenomena, and
common substances are described below: variations in conditions of specimen storage. In
comparison, the CV ranges exhibited by commonly used
Clofibrate lowers serum ALP activity, reducing all ALP instruments in a recent CAP participant summary report
fractions except liver; this may be caused by an [55], as depicted in Table 4, show that the
increased biliary clearance of ALP. Azathioprine lowers interlaboratory CVs within the same instrument group
ALP by an unknown mechanism. Estrogens, alone or in have improved from 17% to 28% to < 2% to 13.7%.
combination with androgens, depress ALP activity, Even among the peer groups, the interlaboratory CVs
whereas in other studies, estrogens and androgens exhibit improvement. With efforts focused on reagent
increased serum ALP activity [47,48]. The decrease seen quality by the manufacturers, specified assay conditions,
with estrogen therapy may result from a lowered rate of and traceability to ERMs, all methods will correlate
bone turnover. well, and interlaboratory CVs will be reduced further to
2% as envisioned. Each laboratory must establish its
Verapamil, widely used in the treatment of cardiac own acceptability criteria for intra-daily and inter-daily
arrhythmias, angina pectoris, and essential hypertension, variations based on the instrument and quality-control
increases serum ALP in hypertensive patients [49]. materials.
When verapamil was administered to hypertensive
patients for 2 months, total ALP activity increased References
significantly from 166 to 191 U/L (P = 0.004), and an 1 Mornet E, Stura S, Lia-Baldini A-S, Stigbrand
increase in bone ALP in serum was subsequently T, Mnez A, Le Du M-H. Structural evidence
observed. These results suggest that verapamil may for a functional role of human tissue
affect bone metabolism that is secondary to the nonspecific alkaline phosphatase in bone
enhancement of parathyroid hormone secretion. mineralization. J Biol Chem. 2001;276:31171-
31178. Available at
94
http://www.jbc.org/cgi/content/full/276/33/311 isoenzymes and buffer system. Clin Chem.

71 1990;36:2012-2013.
2. Greenberg CR, Evans JA, McKendry-Smith S, 16 Cornish CJ, Neale FC, Posen S. Automated
Redekopp S, Haworth JC, Mulivor R, fluorometric alkaline phosphatase microassay
Chordiket BN. Infantile hypophosphatasia: with 4-methylumbelliferyl-phosphate as a
localization within chromosome Region substrate. Am J Clin Pathol. 1970;53:68-76.
1p36.1-34 and prenatal diagnosis using linked 17 AnaSpec SensoLyte FDP Alkline
DNA markers. Am J Hum Genet. 1990:46:286- Phosphatase Assay Kit, ANASPEC, 2149
292. OToole Avenue, San Jose, CA 95131.
3 Human Molecular Genetics Laboratory and Available at http://www.anaspec.com
SESEP Laboratory website, Etienne Mornet, 18 Nunc, Thermo Fischer Scientific. Alkaline
ed. The tissue-nonspecific phosphatase gene phosphatase (AP). Available at
mutations database (more than two screens): http://www.nuncbrand.com/NAG/DP0010.htm
(ALPL; MIM 171760). Updated January 2008. 19 McComb RB, Bowers GN. Alkaline phospha-
Available at tase and the international clinical enzyme scale:
http://www.sesep.uvsq.fr/database_hypo/Mutati Am J Clin Pathol. 1985;84:67-73.
on.html 20 Siekmann L, Department of Clinical
4 Shinowara G, Jones LM, Reinhart HL. Biochemistry, University of Bonn, Germany,
Estimation of serum inorganic phosphate and IFCC.org. Reference systems in clinical
acid and alkaline phosphatase activity. J biochemistry. Available at
Biol Chem. 1942;142:921-933. http://www.ifcc.org/ejifcc/vol13no3/130301002
5 Kind PRN, King EJ. Estimation of plasma .htm
phosphatase by determination of hydrolysed 21 International Standards Organization (ISO).
phenol with amino-antipyrine. J Clin Pathol. Org ISO 17511:2003, in-vitro diagnostic
1954;7:322-326. medical devices, measurement of quantities in
6 Bessey O, Lowry OH, Brock MJ. Method for samples of biological origin, metrological
the determination of alkaline phosphatase with traceability of values assigned to calibrators and
five cubic millimeters of serum. J Biol Chem. control materials, 2003-08-22, PDF 427 KB.
1946;164:321-329. Available at
7 King EJ, Armstrong AR. A convenient method http://www.iso.org./iso/iso_catalogue/catalogue
for determining serum and bile phosphatase _detail.htm?csnumber=30716
activity. Can Med Assoc J. 1934;31:376-381. 22 Directive 98/70/EC of the European Parliament
8 Togari A, Sakai J, Matsumoto S, Nagatsu T. and the Council of 27 October, 1998 on in-vitro
Highly sensitive assay for alkaline and acid diagnostic medical devices.
phosphatase activity by high-performance 23 Schiele F, Muller J, Colinet E, Siest G.
liquid chromatography with electrochemical Certification of an enzyme reference material
detection. J Chromatogr. 1987;417:41-46. for alkaline phosphatase (CRM 371). Clin
9 Moss DW. A note on the spectrophotometric Biochem. 1991;24:159-168.
estimation of alkaline phosphatase activity. 24 Schneidinger B, Meier T, Schmuck R, Shao Z
Enzymologia. 1966;31:193-202. (inventors); Roche Diagnostics Operations Inc
10 McComb RB, GN Bowers Jr, Posen S. Alkaline (assignee). Conjugate of a tissue nonspecific
Phosphatase. New York: Plenum; 1979. alkaline phosphatase and dextran process.
11 Millan JL. Mammalian Alkaline Phosphatases: Patent filing on 2003-07-21. Available at
From Biology to Applications in Medicine & http.www.patentsorm.us/patents/6905689-
Biotechnology. Weinheim, Germany: Wiley- description.html
VCH; 2006. 25 Bowers GN,Jr, McComb RB. Measurement of
12 Lewandrowski K, Lee-Lewandrowski E, total alkaline phosphatase activity in human
Bowers GN Jr, McComb RB. Investigation of serum. Clin Chem. 1975;21:1988-1995.
N-methyl-D-glucamine buffer for assay of 26 Bowers GN Jr, McComb RB. Alkaline
alkaline phosphatase in serum. Clin Chem. phosphatase, total activity in human serum. In:
1992;38:2286-2294. Faulkner WH, Meites S, eds. Selected Methods
13 Weisheit R, Treiber W. Method for determining for the Small Clinical Chemistry Laboratory.
alkaline phosphatase. Patent filing date 2001, Washington, DC: American Association for
April 6. Available at Clinical Chemistry; 1982:79-83.
www.freepatentsonline.com/6713275.html 27 World Health Organization Regional Office for
14 Gorus FK, DePree L. Effects of isoenzyme Southeast Asia, Blood Safety and Clinical
composition on Kodak Ektachem test results for Technology. Guidelines on standard operating
creatine kinase, lactate dehydrogenase, alkaline procedures for clinical chemistry, alkaline
phosphatase and amylase. Clin Chem. phosphatase, p-nitrophenol method. Available
1990;36:685-686. at
15 VanHoof VO, Van Campenhout CM, DeBroe http://www.searo.Who.int/EN/Section10/Sectio
ME, Lepoutre LG. Variations in measures n17/Section53/Section481_481_1761.htm
alkaline phosphatase activity: influence of
95
28 Scandinavian Society for Clinical Chemistry placenta and their occurrence in human cancer.
and Clinical Physiology. Recommended Cancer Res. 1976;36:2268-2273.
methods for the determination of four enzymes 42 Weissbach L, Bussar Maattz R, Mann K, The
in blood. Scand J Clin Lab Invest. 1974;33:291- value of tumor markers in testicular seminomas.
306. Results of prospective multicenter study. Eur
29 German Society for Clinical Chemistry. Urol. 1997;32:16-22.
Recommendations of the Enzyme Commission. 43 Stinghen ST, Moura JF, Zancanella P, Rodrigues
Z Klin Chem Klin Biochem. 1972;10:281-291. GA, Pianovski MA, Lalli E et al. Specific
30 Tietz NW, Burtis CA, Duncan P, Ervin K, immunoassays for placental alkaline
Petitclerc CJ, Rinker AD et al. A reference phosphatase as a tumor marker. J Biomed
method for measurement of alkaline Biotech. 2006;Article ID 56087:1-8.
phosphatase activity in human serum. Clin 44 Warren RC, Mckenzie CF, Rodeck CH,
Chem. 1983;29:751-761. Moscoso G, Brock DJ, Baron L. First-trimester
31 Working Group on Enzymes. Proposal of diagnosis of hypophosphatasia with a
standard methods for the determination of monoclonal antibody to the liver/bone/kidney
enzyme catalytic concentrations in serum and isoenzyme of alkaline phosphatase. Lancet.
plasma at 37C. I. Alkaline phosphatase. Eur J 1985;2(8460):856-858.
Clin Chem Clin Biochem. 1992;30:247-256. 45 Henthorn PS, Whyte MP. Infantile
32 Chromy V, Zahradnicek L, Voznicek J. Use of hypophosphatasia: successful prenatal
N-methyl-D-glucamine as buffer in assessment by testing TnALP isoenzyme gene
determination of serum alkaline phosphatase mutations. Prenat Diagn. 1995;15:1001-1006.
activity. Clin Chem. 1981;27:1729-1732. 46 Youngs Effect online. Available at
33 Ceriotti G, Bonvicini P, Certiotti F, Franzini C, http://www.fxol.org
Prencipe I, Spandrio L. Valutazione di un 47 Nanji AA. Decreased activity of commonly
metdod per la determinazione dellattivit della measured serum enzymes: causes and clinical
fosfatasi alcalina (ALP) con il tampone N- significance. Am J Med Technol. 1983;49:241-
metilgucammina (MEG). Proposta del suo uso 245.
come metodo raccomandato. Giorn It Clin Clin. 48 Posen S, Doherty E. The measurement of serum
1984;9:167-181. alkaline phosphatase in clinical medicine. Adv
34 Munan L, Kelly A, Petitclerc C, Billon B. Atlas Clin Chem. 1981;22:163-254.
of Blood Data. Prepared by the Epidemiology 49 Sjd'n G, Rosenqvist M, Kriegholm E,
Laboratory and the Laboratory of Clinical Nordenenstrm J, Bjrkhem I. Verapamil
Biochemistry, University of Sherbrooke, increases serum alkaline phosphatase in
Quebec, 1980. hypertensive patients. J Intern Med. 1990; 228:
35 Lehto MT, Sharon FJ. Proximity of the protein 339-342.
moiety of GPI-anchored protein to the 50 Young DS. Effects of Drugs on Clinical
membrane surface: a FRET study. Laboratory Tests. 3rd ed. Washington, DC:
Biochemistry. 2002;41:8368-8376. AACC Press; 1990:3-193-25.
36 Low MG, Prasad ARS. A phospholipase D 51 Lott JA, Wolf PL. Clinical Enzymology: A
specific for the phosphatidylinositol anchor of Case-Oriented Approach. Chicago: Year Book
cell-surface proteins is abundant in plasma. Medical; 1986:57-74.
Proc Nat Acad Sci USA. 1988;85:980-984. 52 U.S. Department of Health and Human Services.
37 Le Du MH, Stigbrand T, Taussig MJ, Menez A, Medicare, Medicaid and
Stura EA, Crystal structure of alkaline CLIA Programs: regulations implementing the
phosphatase from human placenta at 1.8 A Clinical Laboratory
resolution, implication for a substrate Improvement Amendments of 1988 (CLIA). Final
specificity. J Biol Chem. 2001;276:9158 -9165. rule. Fed Regist.
38 Muensch HA, Maslow WC, Azama F, Bertrand 1992;57:7002-186.
M, Dewhurst P, Hartman B. Placental-like 53 Lott JA, Massion CG. Interlaboratory quality
alkaline phosphatase: re-evaluation of the tumor control of enzyme analyses: the CAP
marker with exclusion of smokers. Cancer. experience. In: Hamburger HA, ed. Clinical
1986; 58: 1689-1694. and Analytical Concepts in Enzymology. Skoki,
39 Hirano K, Matsumoto H, Tanaka T, Hayashi, IL: College of American Pathologists;
Y, Lino S, Domar U, Stigbrand T. Specific 1983:233-256.
assays for human alkaline phosphatase 54 Lott JA, Tholen DW, Massion CG. Survey of
isozymes. Clin Chim Acta. 1987;166:265-273. serum enzyme analyses: human tissues as a
40 Williams GH, McLaughlin PJ, Johnson PM. source of enzymes. Arch Pathol Lab Med.
Tissue origin of serum placental-like alkaline 1984;108:949-953.
phosphatase in cigarette smokers. Clin Chim 55 College of American Pathologists. 2007
Acta. 1986;155:329-333. Surveys: Chemistry/Therapeutic Drug
41 Fishman L, Miyayama H, Driscoll SG, Monitoring (C-A) Participant Summary Report.
Fishman WH. Developmental phase-specific Northfield, IL: CAP; 2007:7-8.
alkaline phosphatase isoenzymes of human
96
Tables
Table 1: Comparison of Original Reaction

Table 2: Mean ALP Activity in Tissue Specimens
Conditions for Alkaline Phosphatase (ALP) Analysis*
(AMP Buffer System) [25]
Condition: Temperature Tissue Activity
Manual reference (Bowers and McComb) [25]: 30C (molsec-1 kg-1 of wet tissue)
German [29]: 25C Placenta 600
Scandinavian [28]: 37C Adrenal 500
AACC [30]: 30C Liver 210
Condition: pH Bone 125
Manual reference (Bowers and McComb) Spleen 125
[25]: 10.5 Lung 110
German [29]: 9.8 Intestine (whole) 80
Scandinavian [28]: 9.8 Kidney 71
AACC [26]: 10.4 Prostate 55
Condition: Final concentration of reagents Brain 35
Manual reference (Bowers and McComb) [25]: Thyroid 30
pNPP: 16 mmol/L Myocardium 25
MAP: 1.0 mol/L
Mg2+: 1.0 mmol/L
German [29]:
pNPP: 10 mmol/L
DEA: 1.0 mol/L
Mg2+: 0.5 mmol/L
Scandinavian [28]:
pNPP: 10 mmol/L
DEA: 1.0 mol/L
Mg2+: 0.5 mmol/L
AACC [26]:
pNPP: 16 mmol/L
MAP: 0.35 mol/L
Mg2+: 2.0 mmol/L
Condition: Fraction of sample volume
Manual reference (Bowers and McComb)
[25]: 0.0164
German [29]: 0.009
Scandinavian [28]: 0.009
AACC [26]: 0.196
Condition: Linearity (approximate)
[25]: To 500 U/L
German [29]: To 500 U/L
Scandinavian [28]: Linear for 10 minutes or
up to 1500 U/L
AACC [26]: To 900 U/L
Condition: Precision (in reference interval)
[25]: 3% to 6%
German [29]: 5%
Scandinavian [28]: 5%
AACC [26]: 3% to 5%
AACC, American Association for Clinical Chemistry;
German, German Society for Clinical Chemistry;
Scandinavian, Scandinavian Society for Clinical
Chemistry and Clinical Physiology.
DEA, Diethanolamine; MAP, 2-methyl-2-amino-1-
propanol; pNPP, p-nitrophenyl phosphate.
*Major interferences: EDTA, citrate, oxalate, inorganic
phosphate, calcium, and ammonium sulfate.
It has been reported that a contaminant in some lots of
DEA causes significant loss of ALP activity [25].
97
Thyroiditis
Malignancy
Table 3: List of Disorders That Usually Result in Bile duct adenocarcinoma
Abnormal Serum Activities Cervical cancer
(Adapted from Reference 51) Gynecological neoplasms
Causes of Increased Serum ALP Hodgkins disease
Bone disorders Liver, primary & secondary
Fractures Lung, ectopic production
Osteomalacia Lymphoma, leukemia
Pagets disease Maxillary carcinoma
Osteoblastic lesions Multiple myeloma
Osteomyelitis Osteosarcoma
Liver diseases Ovarian cancer
Biliary atresia Pancreatic cancer
Cholestasis Prostatic cancer
Cirrhosis Renal
Fatty liver, acute Testicular
Hepatitis Urinary bladder cancer
Portal hypertension Drugs and poisons
Primary biliary cirrhosis Alcohol
Kidney disease Cholestatogenic drugs
Chronic renal failure Cimetidine
Obstruction of urinary tract Furosemide
Renal disease Halothane
Renal infarction Hepatotoxic drugs
Renal vascular hypertension Papaverine
Gastrointestinal Phenobarbital
Crohns disease Phenytoin
Duodenal ulcer
Pancreatitis, acute Miscellaneous
Peptic ulcer Cystic fibrosis
Small bowel infarction Immunoglobulin complexed ALP
Bowel ulceration Infectious mononucleosis
Splenic infarct Pulmonary infarction
Ulcerative colitis Rheumatoid arthritis
Heart disease Systemic infection
Acute myocardial infarction Unexplained, transient increase
Congestive heart failure Causes of decreased serum ALP
Hematological disorders Drugs
Pernicious anemia Clofibrate
Sickle cell anemia Sulfonamides
Thalassemia Other drugs
Endocrine disorders Miscellaneous
Acromegaly Hypophosphatasia
Diabetes mellitus Malabsorption
Diabetes, uncontrolled Malnutrition
Hyperthyroidism Zinc deficiency
Hypothyroidism
Menopause
98
Table 4: CV Ranges of Currently Used Major Instruments Systems

From a 2007 College of American Pathologists (CAP) Survey Report [55]
Instruments Manufacturer # of Labs Sp # Range of Means SD range CV range

And Reagent Supplier (T) Within the Group
Abbott all instruments (37C) 200 SP # 1 374.8 U/L 6.1 1.6

(Architect, Aeroset) 200 SP # 2 35.9 U/L 1.5 4.3
201 SP # 3 139.3 U/L 2.8 2.0
Bayer (Siemens) (37C) 183 SP #1 370.6 U/L 10.7 2.9

(All Instrument models) 182 SP #2 34.9 U/L 2.3 6.6
184 SP #3 136.9 U/L 4.5 3.3
Beckman (37C) 1378 SP # 1 380.0 388.6 U/L 7.1 12.0 1.9 3.1
(All Instrument models) 1381 SP #2 30.8 31.3 U/L 1.6 2.4 5.3 7.8
1376 SP #3 137.9 139.3 U/L 3.1 5.0 2.2 3.6
Dade Behring (37C) 1357 SP #1 294.9 U/L 12.1 4.1

Dimension 1371 SP #2 39.5 U/L 5.3 13.3
1371 SP #3 116.9 U/L 7.0 6.0
Olympus (37C) 267 SP #1 357.3 U/L 15.8 4.4

268 SP #2 31.8 U/L 2.0 6.2
267 SP #3 131.0 U/L 5.7 4.3
Roche (37C) 789 SP #1 310.9 330.8 U/L 9.7 14.3 3.0 4.3
795 SP #2 28.7 35.3 U/L 1.1 3.9 2.9 13.7
794 SP #3 119.6 131.5 U/L 3.8 6.0 3.2 - 4.9
Vitros Systems (37C) 822 SP #1 323.8 338.1 U/L 10.211.8 3.1 3.5
821 SP #2 48.7 51.4 U/L 2.0 -2.2 4.1 4.3
818 SP #3 133.7 142.5 U/L 5.0 5.6 3.7 3.9
All Instruments All Models

4996 SP #1 294.9 - 388.6 U/L 6.115.8 1.64.4
5018 SP #2 28.7 50.4 U/L 1.5 - 5.3 2.9 -13.7
5011 SP #3 116.9 142.5 U/L 2.8 - 7.0 2.0 6.0
Note: For the sake of brevity, the data has been combined to reflect peer-group performance at three levels of
concentration (three CAP Proficiency Samples, CHM-03 to CHM-05). Conclusions drawn are exclusively of the author
himself and do not reflect the views of the CAP.
99
Figures storage conditions. Any special precautions

must also be noted.
Figure 1: ALP pNPP absorption spectra.
Assay
Equipment: Any recording spectrophotometer equipped
with a temperature-controlled cell compartment is
suitable for the procedure. The temperature should be
held constant at 37C 0.1C.
1. Add 50 L of specimen to 1.0 mL of
magnesium acetate solution. Thoroughly mix,
and incubate for 5 min at 37C.
2. Prewarm buffered pNPP solution to 37C, and
add 2.0 mL to the incubation mixture from step
1. Agitate thoroughly.
3. Transfer the reaction mixture to a cuvette with a
1-cm light path, and read the absorbance
Visible-spectrum absorbance curves for substrate, p- change versus time at 403 nm for 2 min.
nitrophenyl phosphate, which has a maximum Readings can be taken immediately after
absorbance at 311 nm at 25C in 10 mmol/L NaOH, mixing.
solid curve, and for product, p-nitrophenol, which has a
maximum absorbance at 402.5 nm at 30C in 2-amino-2- Calculations
methyl-1-propanol (pH 10.30, 30C), dot-dashed curve. International units (U) of activity are expressed as
micromoles of p-nitrophenoxide formed per min.
Enzyme concentrations are expressed as international
Procedure: NCCLS/EPE Reference Method units per liter (U/L). U/L can be calculated from the
change in absorbance by the following equation for a 1
Principle cm light path:
The colorless pNPP substrate is converted at alkaline pH
to the yellow pNP. The reaction is followed by U/L = A TV 106
measurement of the increase in absorbance at 403 nm.
t SV
Reagents and Materials where A = change in absorbance for time t, TV is the
1. AMP buffer (1.5 mol/L). Liquefy 2-methyl-2- total volume, SV is the specimen volume, and is the
amino-1-propanol by warming to 30C to 35C. molar absorptivity for p-nitrophenoxide (18.8 103 L
Weigh 135 g of the liquid directly into a 1 L mol-1 cm-1); 106 is the factor to convert the
volumetric flask, add 500 mL of distilled water, concentration to micromoles per liter. For the conditions
and mix. Carefully add 190 mL of 1.0 mol/L above,
HCl to the flask. When the solution has cooled
to room temperature, bring to 1 L with distilled U/L = A/min (3050 L/50 L) 106/18,800
water. Confirm that the pH at 30 C is 10.5. or U/L = A/min 3245
This is stable for 1 month when stored in an The method is linear to approximately 500 U/L.
airtight container at 25C.
Procedure: German Society for Clinical Chemistry
2. Magnesium acetate solution (3 mmol/L). Reference Method
Dissolve 650 mg magnesium acetate 4 H2O in
1 L of water. Stable indefinitely at 4C. Principle
3. p-Nitrophenyl phosphate solution (24.5 Colorless pNPP is converted to the intensely yellow
mmol/L). Dissolve 91 mg of disodium 2- pNP, and the phosphate group is transferred to the buffer
nitrophenyl phosphate 6H2O in 10 mL of to form N-methyl-D-glucamine phosphate. The increase
in absorbance at 405 nm is followed.
AMP buffer. Prepare fresh daily.
4. p-Nitrophenol spectrophotometric standard Specimen
solutions (1 mmol/L). Dissolve 139.1 mg of Refer to NCCLS document for Alkaline Phosphatase,
pNP in 1 L of distilled water. Stable for several Total
months when stored in the dark. To prepare the
working standard solution, add 25 mL of the 1 Reagents and Materials
mmol/L solution to 900 mL of AMP buffer, and 1. N-methyl-D-glucamine (500 mmol/L).
dilute to 1 L with distilled water. The working Dissolve 27.33 g of methylglucamine (560
standard solution is stable for at least 2 months. mmol/L) and 1.15 g of sodium chloride (78.4
Use this solution to confirm the molar mmol/L, pH 10.6 at 20 C or 10.1 at 37C) in
absorptivity of pNP at 403 nm. about 200 mL of water. Adjust pH to 10.5 with
Note: Label all reagents with name of reagents, 1 mol/L HCl, add 30 mg magnesium acetate
date of preparation, date of expiration, and tetrahydrate, and bring to a total volume of 250
mL with distilled water. This is stable for at
100
least 2 months at 20C if kept in a tightly Assay

stoppered flask. Equipment: A spectrometer with a constant temperature
2. p-Nitrophenylphosphate (20 mmol/L). cuvette compartment is necessary. The temperature
Dissolve 4.16 g of disodium p-nitrophenyl should be held constant at 37C 0.1C.
phosphate hexahydrate, and add water to a final 1. Pipet 500 L of N-methyl-D-glucamine into a
volume of exactly 50 mL. Prepared just prior to cuvette, and add 10 L sample.
use; it must be used within 8 hrs if stored at 2. Mix thoroughly and incubate at 37C for at
20C to 25C or at the most within 1 day after least 300 sec or until the mixture has reached
storage at 3C to 5C. this temperature.
3. NaCl, (155 mmol/L). Dissolve 0.9 g of NaCl 3. Add 50 L of pNPP.
and bring to 100 mL with distilled water.
4. Mix and record the increase in absorbance after
60 sec at 405 nm.
Calculations
The reagent blank must be subtracted from the overall
change in absorbance: (A/t)overall- (A/t)blank =
(A/t)corrected. Use the corrected value in the
calculation above.
101
Alpha-Fetoprotein
Alpha-Fetoprotein
Gerald J. Mizejewski
Name: Alpha-fetoprotein, AFP

Clinical significance: Refer to Chapter 44, Pregnancy, in the 5th edition of Clinical
Principles of Analysis and Current Usage of adult patients and in the sera of cancer patients
Introduction and Historical Background (breast cancer, hepatomas, etc.). Truncated forms of
Human alpha-fetoprotein (HAFP) is a tumor-associated HAFP (~50,000 D) have also been detected in cell
fetal mammalian glycoprotein involved with both cultures comprising hepatomas, testicular embryonal
ontogenic and oncogenic growth [1,2]i. The fetal carcinomas, and breast tumors [12-14].
protein is a 69-kDa single-polypeptide chain containing
3% to 5% carbohydrate; it exhibits a triplicate domain B. Biological Roles: Similar to albumin, serum
structure configured by intramolecular loops dictated AFP is known to bind and transport a multitude of
by disulfide bridging [3]. In electrophoretic profile, ligands such as bilirubin, fatty acids, retinoids, steroids,
AFP occupies an alpha-1 anodic position running heavy metals, dyes, flavonoids, phytoestrogens, dioxin,
slightly slower than albumin. and various drugs [13,14]. Other ligands that bind to
AFP (rodent and human) include metabolic stains, L-
HAFP is synthesized in the yolk sac, fetal liver, and tryptophan, warfarin, triazine dyes, phenylbutazone,
gastrointestinal tract during pregnancy but is streptomycin, phenytoin, anilinonaphthalene sulfate,
reexpressed in multiple adult tumors of mixed heavy metals, low-carbon-chain alcohols, and
mesodermal/endodermal origin [4,5]. In the clinical polyunsaturated fatty acids [15]. Although the
laboratory, HAFP has been employed both as a physicochemical and structural properties of this
postoperational tumor marker and as a gestational age- glycoprotein have been described, it has been the in
dependent fetal defect marker demonstrating utility in vitro functional roles that have been extensively
screening for neural tube defects and aneuploidies [6,7]. studied, especially the ligand carrier/transport functions
While maternal serum AFP (MS-AFP) concentrations [12]. In addition, the ability of AFP to modulate native
associated with neural tube fetal defects are elevated, and adaptive immune responses has also been pursued.
the chromosomal disorders are associated with much Since the first reports by Murgita et al. [16,17] in the
lower levels. 1970s, AFP has long been recognized as both a B- and
T- cell immunoregulatory agent. Overall, full-length
HAFP has been found immunosuppressive in both B-
Structural and Functional Aspects of AFP: and T-cell lectin stimulation, although AFP can also
A. Structural Variants: Many molecular variants induce immune cell stimulation under certain
of mammalian AFP have been reported in the scientific conditions [18,19].
literature since the 1970s (Table 2). Some of these
earlier variant forms of AFP were attributed to A multitude of studies have further established AFP as
carbohydrate microheterogeneity and isoforms a regulator of normal and neoplastic growth [20-24]. In
associated with varying isoelectric points [8,9]. Later fact, it is the growth-modulating activity that
reports demonstrated AFP forms that were genetic distinguishes AFP from albumin, the major blood
variants and lectin glycoforms demonstrable by protein carrier/transport molecule of the albuminoid
electrophoretic and chromatographic procedures gene family. The growth regulatory properties of AFP
[10,11]. Still other variants were detected following have aroused investigational interest in studies of
high-pressure liquid chromatography (HPLC) utilizing ontogenetic and oncogenic growth in both cell cultures
lectin, heavy metal, and hydrophobic solid-phase and animal models. A myriad of reports have now
separation methodology [12]. Aberrant forms of HAFP documented that AFP is capable of regulating growth in
have been detected in the reproductive/and urinary tract ovarian, placental, uterine, hepatic, phagocytic, bone
marrow, and lymphatic cells in addition to neoplastic
cells (i.e., MCF-7 and MTW9A breast cancer)
i Alpha-Fetoprotein [18,25,26]. Since the 1990s, AFP is no longer
Previous and current authors of this method: considered only a biomarker for cancer and fetal
First edition: Not done disorders; it is also viewed as a protein associated with
Methods edition: George J. Knight modulating cell proliferation, differentiation,
Second edition: Not updated regeneration, and transformation in both ontogenetic
Third edition: Not updated and oncogenic growth processes. HAFP has further
Fourth edition: Not updated been shown to possess pro-angiogenic properties that
Fifth edition: Gerald J. Mizejewski promote neovascularization and growth in both fetal
102
Alpha-Fetoprotein
and tumor tissues [27,28]. Finally, HAFP has recently same sample will vary, depending on the source of the
been shown to functionally impair dendritic cells, standard used to calibrate the assay. Three reference
causing immune dysfunction and apoptosis of antigen- preparations are available: the World Health
processing cells (APCs) [29]. Organization (WHO) Reference Preparation for human
Measurement of AFP AFP (72-225), the British Standard (72-227), and the
AFP has historically been measured by some form of U.S. National Reference Preparation for AFP in mid-
immunoassay. Although rocket electroimmunodiffusion pregnancy maternal serum [30]. The WHO and British
(Table 1, Method 2) is still occasionally used to standards were derived from the same original cord
measure amniotic fluid AFP, this method has been serum pool and may be considered essentially
largely replaced by more sensitive immunoassays equivalent, whereas the U.S. standard has been
capable of measuring the low ng/mL concentrations calibrated against the WHO standard. Laboratories
required for both maternal serum AFP screening and wishing to determine the relationship of their local unit
cancer diagnosis and monitoring. Competitive to the International Unit (IU) should obtain a reference
radioimmunoassay has been widely used to quantitate preparation for direct calibration of their local standard.
AFP, but the practice now is to measure AFP by non- Interlaboratory comparison should be based only on
isotopic immunoassays that employ enzyme, IU/mL, and all laboratories are encouraged to obtain the
fluorescent, or chemiluminescent labels (Table 1). relationship of their standard to IU either from the
manufacturer of the kit in question, or by direct
Immunoassays employed to measure AFP are of two comparison. Estimates of the relationship of the IU and
major types: (1) competitive and (2) two-site solid- mass units range from about 0.91 to 1.29 ng/IU, but 1.0
phase immunometric. In a typical radioisotopic IU is usually equivalent to about 1.0 ng.
125
competitive type (Table 1, Method 2), purified I- Assay Sensitivity
labeled AFP is mixed with the sample containing AFP Currently, almost all laboratories use competitive RIA
and allowed to compete for a limited amount of AFP or non-isotopic immunometric assays to measure AFP
antibody. in serum or amniotic fluid. Screening for low AFP
levels in fetal Down syndrome requires accurate
In the two-site immunometric assays (Table 1, Methods measurement of AFP down to 10 ng/mL. Cancer
2, 3, and 4), a solid phase such as magnetic particles, screening and monitoring requires assays that can
plastic beads, or microtiter plates is coated with anti- measure values as low as 1 to 5 ng/mL.
AFP antibody and incubated with patient specimen or
standards. AFP present in the sample reacts with the Specimen
anti-AFP antibody and is immobilized on the solid AFP in serum, plasma, or amniotic fluid may be
phase. Unbound components of the specimen are measured by any of the assays previously discussed.
removed by washing of the solid phase. A second anti- Serum is preferred to plasma for AFP analysis,
AFP antibody labeled with an enzyme, radioactive, although some manufacturers permit the use of plasma.
fluorescent, or chemiluminescent label is incubated AFP may also be measured by elution from blood spots
with the solid phase and reacts with AFP bound at a collected on filter paper [31]. Precision of these latter
second epitope site. Polyclonal and monoclonal determinations is usually lower, however, because of
antibodies have been used in tandem for each of these variability in spotting of blood, punch size, elution of
steps. Unreacted label is removed by washing of the AFP, and use of various extraction buffers and is not
solid phase, and any additional reagent that is needed to routinely employed for MS-AFP screening. Serum AFP
generate a signal is added. In the case of a fluorescence is somewhat thermostable, and samples may be shipped
or radioactive label, the bound label can be measured at ambient temperatures after separation from red blood
directly. For enzyme labels, the solid phase must be cells. Specimens derived from sera are stable at room
incubated with substrate before the detection step. For temperature or at 4C for at least 1 week [32] and for
chemiluminescence labels, reagent must be added to several months at 20C [33,34]. To avoid matrix
generate the chemiluminescent signal. In all cases, the effects, it is absolutely essential to dilute AFP
amount of signal generated with the solid phase is specimens with diluent provided by the kit
directly proportional to the AFP concentration of the manufacturer.
standard or unknown.
Storage Stability of AFP
Reference and Preferred Methods Lantz et al. in a previous study evaluated the effect of
Because of the heterogeneity of AFP, there is no different sample collection, storage, and preparation
reference method. techniques on the immunoassay of serum alpha-
fetoprotein (AFP) [35]. Investigators found that
Standardization immediate freezing of serum and subsequent thawing
Accuracy of AFP values is dependent on the source of resulted in a significant increase in several different
the calibrating standard. AFP values provided by analyte levels, but only small changes in AFP levels
reference laboratories or obtained from manufactured were observed. However, there was an effect over time
kits are frequently given in mass units (ng/mL), even on AFP concentrations, dependent on type of storage.
though no pure AFP standard in mass units is available. Frozen (0C) and refrigerated (4C) storage
Consequently, values in mass units obtained on the temperatures were optimal. A change in AFP levels
103
Alpha-Fetoprotein
could also be influenced by the type of centrifugation in the literature (Table 3). The assay interference is
and by nonstable refrigeration frost-free cycles. Thus it above and beyond the sensitivities of epitope cross-
was concluded that different sample collection, storage, reactivity with proteins such as albumin and transferrin,
and preparation techniques should be monitored in as well as bilirubin. Experience indicates that AFP
maternal serum AFP screening programs. In a separate measured using the non-isotopic immunometric assays
study, investigators found no association between a are not affected by hemoglobin up to 10 g/L,
laboratorys AFP test volume and the reliability of triglycerides up to 1250 mg/dL, bilirubin up to 30
reported multiples of the median (MoM), and there was mg/dL, or protein up to 20 g/L. Freeze/thaw cycles also
no difference in AFP medians across geographic do not cause appreciable changes in AFP
medians of the United States [36]. concentrations [45]. With the exception of heterophile
antigens, most if not all todays commercially available
Investigators have further examined the levels of CIAs display no cross-reactivity with these human
second-trimester maternal serum markers used in Down proteins, nor has any been found with anticancer drugs
syndrome screening in relation to the time between such as cyclophosphamide, doxorubicin, cisplatin,
sample collection and arrival at the laboratory [35]. All vincristine, 5-fluorouricil, and mitomycin [46,47].
blood samples were drawn in serum separator tubes, However, drug inference of AFP quantitation in EIAs
centrifuged within 30 min, and stored at 4C until has been reported for some non-cancer drugs, which
shipment by air express. To examine the effect of include both prescribed and over-the-counter
delayed shipment, serum marker levels (expressed as medications for diabetes, hypertension, inflammatory
MoM) were evaluated in the second-trimester samples and rheumatoid disorders, bacterial infections, and
and stratified by the number of days between serum pain/fever symptoms. However, many of these assay
collection and laboratory receipt. It was found that flaws have since been corrected with todays improved
under specified collection and shipment conditions, immunoassay platforms.
second-trimester mean AFP concentrations and degrees
of measurement variance were stable for up to 9 days at To date, most of the drug interference in AFP assays
ambient temperatures. It is recommended to restrict has been eliminated with the advent and use of CIA
testing to within 6 days of draw, with the purpose of methodology. Such was not the case during the 1990s
keeping the shipping delays to a minimum. A when AFP assays were performed on RIA and EIA
comparison of fresh frozen MS-AFP serums to platforms described in previous studies [48]. Results of
proficiency testing materials (PTM) found that the fresh such earlier reports were disturbing in that commonly
frozen serums did not provide a consistently different used nonprescription and specifically prescribed drugs
imprecision or bias than did standard PTMs [37]. produced reductions in AFP assay concentrations
ranging from 10% to 37% per specimen (Table 3). Such
Stability of Newborn Dried Bloodspots: non-cancer medications included injectable insulin
Storage stability of AFP dried bloodspots was preparations whose results may have contributed to the
performed to determine whether AFP was more stable lower MS-AFP concentrations observed in insulin-
in the liquid or dry state as determined by a number of dependent diabetic pregnancies of that time period
authors (Table 4) [31,38-43]. These investigators [49,50]. The antibiotic drugs resulted in a 9% to 11%
determined whether AFP in dried-blood spots was more reduction, antihypertensive drugs (-adrenergic
stable when stored refrigerated or at room temperature blockers) a 10% reduction, analgesic and antipyretic
and compared to liquid cord blood. AFP concentrations drugs (aspirin) a 16% to 24% reduction, and
were found to decrease 8% in cord plasma stored at methyldopa a 37% to 47% reduction (Table 3). Drug
20C but to increase 3% in cord-blood spots stored at interference in clinical tests would not be unique to
4C. In other studies using cord bloods at room AFP quantitation assays, since different drugs bind to
temperature, AFP values for bloodspots (x = 41.5, SD = HAFP at similar known albumin-drug interaction sites.
29.7) and for liquid plasma correlated closely (r = 0.89, Additional confounders may also be present in that
slope = 0.71) for 40 such samples tested. In a separate various pharmaceutical formulations of drugs may
study, it was found that AFP measurement in dried include solubilized fillers (mannose), stabilizers, and
blood spots could be influenced by seasonal variation, matrix components. Some drugs are dissolved in
time in transport, and number of freeze/thaw cycles, aqueous alcohols, solubilizers such as DMSO, and
unlike serum AFP samples [44]. However, the overall other buffers which could cause drug-sera interactions,
assessment of AFP in dried bloodspots revealed that precipitations, or denaturization.
AFP was stable for many months when sealed in plastic Pregnant women taking anticonvulsant drugs such as
bags and frozen at 20C (see Table 4). phenytoin and valproic acid have long been known to
display elevated MS-AFP concentrations coincident
Assay Interferences with a 1% to 2% incidence of spina bifida [51,52].
At the assay level, drug interference may occur to a Although the reasons for this elevation remain unclear,
limited degree if AFP has been quantitated by AFP is known to bind drugs similar to valproic acid in
radioimmunoassay (RIA), enzyme immunoassay (EIA), a fashion resembling that of albumin binding of this
and chemiluminescence assay (CIA). Although little drug [46]. In a further study, an article was reported by
has been reported on AFP-CIAs, AFP-RIA and EIA Einstein et al. [50,53] which linked the presence of
drug assay interference has been previously described protease inhibitors (PIs) with results of low MS-AFP
104
Alpha-Fetoprotein
screening levels in pregnant women infected with the ng/mL or less [62]. In a study of 338 healthy adults by
human immunodeficiency virus (HIV). Attention can Abbott Diagnostics, 99% of AFP values by an enzyme
be directed at the physiological properties of AFP immunoassay were under 8 ng/mL, and 100% were
regarding the PIs and their biochemical substrates. It below 20 ng/mL. Using cutoff values of 5 to 8.0 ng/mL
has been known since the 1980s that human AFP binds thus seems a suitable choice for general cancer
serum and tissue PIs (proteins), as well as small screening. However, in monitoring cancer therapy, a
molecular PIs such as arginine, benzamidine, and series of increasing or decreasing AFP values is of
Trasylol [54-58]. In theory, HAFP might bind to the PIs primary importance, and a value below 8 ng/mL that
such that AFP could serve as a protease decoy docking shows a rise above a basal level can be of prognostic
protein to sequester the inhibitor and thus modulate the significance.
proteolysis substrate rate reactions.
Monoclonal and polyclonal antibodies have been used
Alpha-Fetoprotein Reference Intervals to detect serum AFP associated with specific
AFP produced by the fetus increases steadily during malignancies. The AFP tumor marker is most useful for
pregnancy, reaching a peak concentration of 3 mg/mL monitoring response to therapy and detecting early
in fetal plasma at 12 to 14 weeks of gestation and relapse. AFP, a marker for hepatocellular carcinoma,
declining slowly thereafter to 50 g/mL at birth [59]. has been used for screening in Asia to assess hepatic
Moreover, premature newborns can show AFP levels masses in patients at particular risk for developing
two- to threefold higher than term babies [60]. By the hepatomas. Testing for the beta subunit of human
end of the first year of life, values level off at about 5 to chorionic gonadotropin (beta-hCG) is an integral part of
8 ng/mL, a level that is maintained throughout life in the diagnosis and management of gestational
the absence of certain disease states (see AFP as a trophoblastic disease (i.e., germ cell tumors). Combined
Serum Tumor Marker below) [61]. AFP and beta-hCG testing has been found to be a
useful adjunct in the evaluation and treatment of
Although there is less variability in AFP values nonseminomatous germ cell tumors and in monitoring
measured with modern kits, differences in regional the response to therapy. Combined AFP and beta-hCG
populations and assay methods can exist, and each also may be useful in evaluating potential origins of
prenatal screening laboratory should establish its own poorly differentiated metastatic testicular cancer (see
gestational agespecific median values. The various Tables 8 and 9).
options for obtaining a reliable set of medians
appropriate for your screened population have been AFP-Secreting Tumors:
described above. Once medians are available, it is Levels of serum AFP that exceed those seen in healthy
customary to report the AFP test results as a multiple of adults (8 ng/mL) have been reported in some patients
the median (MoM) to normalize for gestational age. with benign hepatic disorders such as viral hepatitis or
Each laboratory must select a MoM screening cutoff cirrhosis (see Table 9). The first tumors found to
which meets its needs. secrete AFP were the hepatocellular carcinomas
(hepatomas) in both mice and humans [1,2]. Other
The performances of the various present-day kits for tumor types later found to synthesize and secrete AFP
maternal serum AFP have been assessed in various were testicular germ cell tumors (teratomas) and yolk
proficiency testing programs throughout the United sac tumors of the ovary [4,5]. However, the AFP serum
States. Such kits could include Abbott (IMX) AxSYM, level (500 to 900 ng/mL) secreted by germ cell tumors
Beckman Unicel/Access, Bayer-Advia/Centaur and did not approach the serum levels (microgram/mL)
Immulite. The AFP mass measurements among the found earlier in hepatomas. Following detection of
individual kits seem to largely agree, although Bayer- AFP-secreting tumor types, many other cancers have
Centaur MS-AFP values are sometimes slightly higher now been classified in this category (Table 2). Aside
and DPC Immulite and Beckman Unicel values slightly from the hepatoma and reproductive cancers, AFP
lower for some specimens. Kit comparisons have also secretion has been linked to the gastrointestinal cancers
been studied for their performance among amniotic- of endodermal origin, especially stomach and
fluid (AF-AFP) test samples. Again, the amniotic fluid pancreatic tumors. Less frequently, AFP synthesis and
overall kit performance approached that observed with secretion have been associated with tumors of
the maternal serum samples. With Bayer-Centaur and pineal/pituitary cysts, hepatoblastomas,
to a lesser extent Abbott-AxSYM kits, AF-AFP values hemangioendotheliomas, hepatic bile duct carcinomas,
are sometimes higher, whereas Beckman Access and gallbladder carcinomas, epidermoid cysts, granulosa
DPC Immulite values are somewhat lower in all-lab cell tumors, and others (see Tables 8 and 9). AFP serum
comparisons. levels have also been used as a diagnostic aid for the
differential diagnosis of seminoma versus non-
AFP as a Serum Tumor Marker seminomatous germ cell tumors (see above). For
Cancer and Hepatic Disease: example, patients with germ cell tumors displaying
The upper limit of the reference interval for AFP in elevated AFP serum levels are considered to have the
healthy populations has been tentatively established non-seminomatous type [63].
(Table 8). Hunter, using a very sensitive AFP assay,
found that most healthy persons have values of 5 to 8
105
Alpha-Fetoprotein
Interpretation factors that influence AFP biological fluid levels are

AFP produced by the fetus is transferred to the amniotic listed below as grouped items together with their
fluid (AF) by fetal urination, and amniotic fluid levels associated reference citations:
generally parallel those found in fetal sera, although 1. Birth Weight: Newborn AFP serum levels were
concentrations will be approximately 150 times lower. found to be correlated inversely with birth
AFP also appears in maternal serum as pregnancy weight. Fetal serum AFP levels are significantly
proceeds, by placental transfer or by diffusion across higher in babies born at 40 weeks that had a
the fetal membranes, and peaks at approximately 31 to birth weight below the population mean
32 weeks of gestation. In contrast to typical second- compared to those above the mean. It is
trimester levels of 10 to 12 g/mL in amniotic fluid, recommended that AFP blood concentrations
levels in maternal serum during this same period are determined for infants in the first weeks of life
only 30 to 40 ng/mL. In the second trimester, MS-AFP be adjusted for birth weight and considered
increases by about 15% per week, whereas amniotic elevated only after premature (very low) birth
fluid AFP declines by about 13% per week [64]. In weight is ruled out [69-73].
open lesions of the fetus such as spina bifida and 2. Gestational Age: A significant inverse linear
anencephaly, elevated levels of AFP are found in correlation has been reported between newborn
amniotic fluid due to leakage of AFP into the fluid from AFP levels and gestational age. It was
open lesions. concluded that gestational age and birth weight
(size) together play a significant role in
The increased concentration of AFP in amniotic fluid determining newborn AFP concentrations [74-
results in increased MS-AFP levels as well, and 77].
screening of maternal sera for elevated concentration of 3. Parity, Gravity: No direct effect was observed
AFP is widely used to select women at increased risk of between serum AFP and parity and gravity.
carrying fetuses with an open-neural-tube defect [65]. However, women delivered of their second
Decreased MS-AFP levels may be used to select infant showed a significant decrement in mean
women at increased risk of carrying fetuses with birth weight of the newborn [78,79].
chromosomal defects, such as Down syndrome and 4. Race, Ethnicity: No differences were observed
trisomy-18 [66]. Definitive diagnoses of neural tube in serum AFP of oriental, white, or Hispanic
defects in fetuses of women with elevated serum AFP origins, but maternal sera of blacks can average
are then achieved by measurement of the level of AFP 10% higher values for a given gestational age
in amniotic fluid. Mandatory secondary diagnostic tests [78,80].
for confirming affected infants include level-II 5. Maternal Body Weight: A correction formula
ultrasound and measurement of amniotic fluid (~10%) for body weight is required, because
acetylcholinesterase [67]. Interested readers are referred AFP is diluted out in the maternal bloodstream
to the outstanding in-depth review of this application by in pregnant women with increased body weight
Wald and Cuckle [68]. (exclusive of fetal/placental weight).
6. Diabetic Status: The insulin-dependent diabetic
Although AFP can be measured quite reliably, use of patients serum assay result requires a 10%
maternal serum AFP levels to screen women at downward correction in prenatal screening
increased risk of carrying fetuses with neural tube programs. Such patients serums are usually
defects was originally controversial because of concern lower than nondiabetic patients.
over misuse and misinterpretation of AFP test results; 7. Sex of the Fetus/Newborn: Both cord blood and
however, these concerns are no longer valid. Successful newborn serum AFP levels are known to be
screening requires an integrated program involving the considerably higher in male infants than in
laboratory, ultrasonographers, physicians, genetic females at any gestational and postnatal age [81-
counselors, and a program coordinator. In 1984, the 84].
U.S. Food and Drug Administration (FDA) approved a Abnormal AFP Levels During Pregnancy
stringent set of regulations that limited the sale, Serum levels of gestational age-dependent AFP that fall
distribution, and use of kits to centers that could outside the normal limits seen in healthy pregnant
demonstrate proficiency in all aspects of the screening women have been reported for a multitude of
process. Since that time, kits became available without congenital malformations of the fetus/infant. The first
restriction, but it is strongly recommended that a developmental abnormalities found associated with
laboratory measure AFP only as separate parts of a abnormal AFP levels were neural tube and related
prenatal and/or tumor-marker screening program spinal defects (Table 5) [6,64,85]. Later, other types of
because of the potential harm that could result from birth defects found to reflect aberrant/discordant AFP
incorrect interpretation of test results between cancers levels were the chromosomal disorders (trisomies) and
versus pregnant patients. various anatomical abnormalities [7,86]. The MS-AFP
concentrations (100 to 500 ng/mL) associated with fetal
Factors That Influence AFP Levels in Biological defects are abnormally high; the trisomies are
Fluids. accompanied by abnormally low levels. Following the
Demographic/biological factors can also influence AFP detection of AFP-associated birth defects such as spina
concentrations during perinatal and postnatal life. Some bifida and anencephaly, many other congenital
106
Alpha-Fetoprotein
anomalies have now been classified in the elevated equation of the fitted line is then used to calculate the
AFP category [87,88]. Abnormal levels of AFP median for each gestational week. Detailed information
synthesis and secretion have been associated with three on weighted log-linear regression of AFP values has
compartments: fetal serum, amniotic fluid, and been published [66]. Kit manufacturers provide median
maternal serum. Less frequently, AFP synthesis and values in package inserts, and these may be used as a
secretion have been observed in other biological fluids general guide for expected values with that particular
such as urine, cerebrospinal fluid (CSF), placental kit being used. It must be cautioned, however, that
extracts, and anatomical cysts. AFP serum levels have some manufacturers provide medians that are highly
also been used as an ancillary aid in the diagnosis of erratic and deviate significantly from those expected.
erythropoietic disorders (anemias), placental Medians should rise smoothly by 15% to 20% per
disruptions, fetal death, growth restriction/retardation, gestational week in serum and fall smoothly by 11% to
and preterm labor and birth. 14% per gestational week in amniotic fluid. Median
AFP concentrations based on over 10,000 observations
Prenatal screening programs initially consisted only of for both maternal serum and amniotic fluid are
maternal serum (MS) AFP and amniotic fluid AFP, presented in Tables 6 and 7, including median
which included ancillary acetylcholine and fetal concentrations by gestational week. In general, medians
hemoglobin testing. During this period, neural tube established by most laboratories will not differ from
defects were the only fetal disorders for which prenatal these results by more than 10% to 20%. These same
screening was available (Table 4). In 1992, prenatal median values can be used to calculate MoM values to
screening for Down syndrome using three maternal determine the fetuss risk for having Down syndrome.
serum markersAFP, unconjugated estriol (E3), and AFP levels are known to slowly rise in the aged normal
human chorionic gonadotropin (hCG)was reported, population and are usually higher in males than in
which ushered in the period of the triple test [7]. By females.
1996, a fourth marker, dimeric inhibin-A (DIA), was
added to the fetal defect marker (FEDM) program, In summary, AFP is present at very low levels in
launching the quad test. The quad marker screen was healthy, nonpregnant adults (1 to 8 ng/mL).
developed during the early 1990s and by 1996 provided Concentrations are increased in certain neoplasms such
a means to improve the detection rate while lowering as hepatocellular carcinoma [91], germ cell tumors
the false-positive rate in prenatal screening for Down [60,61], and gastric and pancreatic carcinoma, reaching
syndrome (DS) and possibly other trisomies [86,89]. levels of over 1,000,000 ng/mL in some cases. Levels
may also be elevated in diseases of hepatic origin,
Neural Tube Defect Screening including viral hepatitis and cirrhosis. An important
It is almost universally accepted that cutoff values for application of these associations is the monitoring of
neural-tube-defect screening results are expressed as progression or regression of hepatocellular carcinoma
MoM, a convention developed from the original U.K. and germ cell tumors by serial determination of AFP.
collaborative study [90]. The MoM selected as the Rare cases of hereditary persistence of AFP have been
cutoff point for both serum and amniotic fluid ranges reported [92], as well as cases of AFP congenital
from 2.0 to 2.5, and the exact cutoff point used by an deficiency [93].
individual lab is determined by a trade-off between the
detection rate for neural tube defects and the false- Alpha Fetoprotein Performance Goals
positive rate. It is recommended that each laboratory AFP assays, whether commercial kits or produced
establish its own medians for both maternal serum and internally, using in-house components, are among the
amniotic fluid for each gestational week. Ideally, a most stable, specific, and robust immunoassays
minimum of 100 samples per week should be assayed available. The assays employing non-isotopic labels
in the period from 15 to 20 weeks of gestation to have shelf lives ranging from months to years.
establish reliable medians. However, obtaining 100 Between-assay coefficients of variation of 5% to 10%
samples for each gestational age can be difficult are readily achieved by most of the current assays in the
because most specimens are collected at 16 to 18 weeks region of clinical usefulness, that is, in the range of 5 to
of gestation, the optimum time for neural-tube-defect 100 ng/mL, and high concentration samples can be
screening. A more practical approach is to use the reliably tested by dilution.
information that the log of the median versus
gestational age is a straight line between 15 to 22 weeks Between-laboratory precision is excellent when results
of gestation. AFP is measured on 300 to 500 are expressed in the IU. External quality-assessment
consecutively screened patients (the number of schemes in the United States and Britain, based on IU,
observations at each gestational week will correspond obtain between-laboratory CVs of 10% to 15%. The
to the distribution in the screening program), and a College of American Pathologists proficiency surveys
median calculated derived from AFP concentrations for also demonstrate the acceptable interlaboratory
each week of gestation (Tables 6.1 and 6.2). Next, a precision achievable by two-site immunometric assays.
linear regression analysis is performed of the log of the At a level of 21.8 ng/mL, the overall %CV for many
median versus the gestational age (weighted according laboratories using this technology has achieved 6.7%.
to the number of observations used to calculate the At a level of 174.1 ng/mL, the overall %CV for
median for any given gestational age). The resulting laboratories using this technology was 5.67%.
107
Alpha-Fetoprotein
immune response by alpha-fetoprotein on the

References primary and secondary antibody response. J
1 Tatarinov YS. Content of the embryo-specific Exp Med. 1975;141:269-286.
alpha-globulin in the serum of the fetus, 17 Peck AB, Murgita RA, Wigzell H. Cellular
newborn infant and adult man with primary and genetic restrictions in the
liver cancer. Vopr Med Khim. 1965;11:20-24. immunoregulatory activity of alpha-
2 Abelev GI. Alpha-fetoprotein in ontogenesis fetoprotein. I. Selective inhibition of anti-Ia-
and its association with malignant tumors. Adv associated proliferative reactions. J Exp Med.
Cancer Res. 1971;14:295-358. 1978;147:667-683.
3 Luft AJ, Lorscheider FL. Structural analysis of 18 Mizejewski G, Muehlemann M, Dauphinee
human and bovine alpha-fetoprotein by MJ. Update of alpha-fetoprotein growth
electron microscopy, image processing, and inhibitory peptides as chemotherapeutic agents
circular dichroism. Biochemistry. for tumor growth and metastasis.
1983;22:5978-5981. Chemotherapy. 2006;52:83-90.
4 Hasegawa H, Mukojima T, Hattori N, Sano R, 19 Mizejewski GJ, Eisele L, Maccoll R.
Hirato T. Embryonal carcinoma and AFP with Anticancer versus antigrowth activities of
special reference to hepatoblastoma. Gann three analogs of the growth-inhibitory peptide:
Monographs Cancer Res. 1973;14:129-131. relevance to physicochemical properties.
5 Ballas M. Yolk sac carcinoma of the ovary Anticancer Res. 2006;26:3071-3076.
with alpha fetoprotein in serum and ascitic 20 Leffert HL, Sell S. Alpha1-fetoprotein
fluid demonstrated by immuno-osmophoresis. biosynthesis during the growth cycle of
Am J Clin Pathol. 1972;57:511-516. differentiated fetal rat hepatocytes in primary
6 Brock DJ, Sutcliffe RG. Alpha-fetoprotein in monolayer culture. J Cell Biol. 1974;61:823-
the antenatal diagnosis of anencephaly and 829.
spina bifida. Lancet. 1972;2:197-199. 21 Toder V, Blank M, Gold-Gefter L, Nebel L.
7 Haddow JE, Palomaki GE, Knight GJ, The effect of alpha-fetoprotein on the growth
Williams J, Pulkkinen A, Canick JA et al. of placental cells in vitro. Placenta. 1983;4:79-
Prenatal screening for Downs syndrome with 86.
use of maternal serum markers. N Engl J Med. 22 Hamel S, Hoskin DW, Hooper DC, Murgita
1992;327:588-593. RA. Phenotype and function of bone marrow-
8 Smith CJ, Kelleher PC, Belanger L, Dallaire L. derived T and non-T cells activated in vitro by
Reactivity of amniotic fluid alpha-fetoprotein alpha-fetoprotein. In: Mizejewski GJ,
with concanavalin A in diagnosis of neural Jacobson HI, eds. Biological Activities of
tube defects. BMJ. 1979;1:920-921. Alpha-Fetoprotein. Boca Raton, FL: CRC
9 Breborowicz J, Mackiewicz A, Breborowicz Press; 1987:167-177.
D. Microheterogeneity of alpha-fetoprotein in 23 Leal JA, May JV, Keel BA. Human alpha-
patient serum as demonstrated by lectin affino- fetoprotein enhances epidermal growth factor
electrophoresis. Scand J Immunol. 1981; 14: proliferative activity upon porcine granulosa
15-20. cells in monolayer culture. Endocrinology.
10 Taketa K. Characterization of sugar chain 1990;126:669-671.
structures of human alpha-fetoprotein by lectin 24 Mizejewski GJ, Warner AS. Alpha-fetoprotein
affinity electrophoresis. Electrophoresis. can regulate growth in the uterus of the
1998;19:2595-2602. immature and adult ovariectomized mouse. J
11 Taketa K. Multimodal application of lectin Reprod Fertil. 1989;85:177-185.
affinity electrophoresis of alpha-fetoprotein. 25 Mizejewski GJ, MacColl R. Alpha-fetoprotein
Electrophoresis. 1998;19:1774-1779. growth inhibitory peptides: potential leads for
12 Mizejewski GJ. Alpha-fetoprotein structure cancer therapeutics. Mol Cancer Ther.
and function: relevance to isoforms, epitopes, 2003(2):1243-1255.
and conformational variants. Exp Biol Med 26 Muehlemann M, Miller KD, Dauphinee M,
(Maywood). 2001;226:377-408. Mizejewski GJ. Review of growth inhibitory
13 Mizejewski G. New Insights Into AFP peptide as a biotherapeutic agent for tumor
Structure and Function: Potential Biomedical growth, adhesion, and metastasis. Cancer
Applications. Orlando: Academic Press; Metastasis Rev. 2005;24:441-467.
1985:5-34. 27 Liang OD, Korff T, Eckhardt J, Rifaat J, Baal
14 Deutsch HF. Chemistry and biology of alpha- N, Herr F, Preissner KT, Zygmunt M.
fetoprotein. Adv Cancer Res. 1991;56:253- Oncodevelopmental alpha-fetoprotein acts as a
312. selective proangiogenic factor on endothelial
15 Hirano K, Watanabe Y, Adachi T, Ito Y, cell from the fetomaternal unit. J Clin
Sugiura M. Drug-binding properties of human Endocrinol Metab. 2004;89:1415-1422.
alpha-foetoprotein. Biochem J. 1985;231:189- 28 Takahashi Y, Ohta T, Mai M. Angiogenesis of
191. AFP producing gastric carcinoma: correlation
16 Murgita RA, Tomasi TB Jr. Suppression of the with frequent liver metastasis and its inhibition
108
Alpha-Fetoprotein
by anti-AFP antibody. Oncol Rep. Bellisario R. Measurement of serum alpha-

2004;11:809-813. fetoprotein in early infancy: utilization of dried
29 Um SH, Mulhall C, Alisa A, Ives AR, Karani blood specimens. Pediatr Res. 1983;17:47-50.
J, Williams R et al. Alpha-fetoprotein impairs 42 Gonzalez C, Guerrero JM, Elorza FL,
APC function and induces their apoptosis. J Molinero P, Goberna R. Evaluation of
Immunol. 2004;173:1772-1778. maternal serum alpha-foetoprotein assay using
30 Reimer CB, Smith SJ, Wells TW. The U.S. dry blood spot samples. J Clin Chem Clin
National Reference Preparation for alpha- Biochem. 1988;26:79-84.
fetoprotein in mid-pregnancy maternal serum. 43 Tsao D, Hsiao KJ, Wu JC, Chou CK, Lee SD.
Clin Chem. 1982;28:709-716. Two-site enzyme immunoassay for alpha-
31 Mizejewski GJ, Bellisario R, Beblowski DW, fetoprotein in dried-blood samples collected on
Carter TP. Commercial radioimmunoassay kit filter paper. Clin Chem. 1986;32:2079-2082.
for measurement of alpha-fetoprotein adapted
44 Masse J, Summers A, Cherian G, Forest J.
for use with dried blood specimens from
Transportation of maternal serum specimens
newborns. Clin Chem. 1982;28:1207-1210.
for screening for chromosomal aneuploidies:
32 Dale JC, Pruett SK. Phlebotomy: a minimalist
effect of seasonal variations, distance, and
approach. Mayo Clin Proc. 1993;68:249-255.
freezing on the stability of the biological
33 Davies MC. Stability of serum AFP and free
markers. Clin Biochem. 2000;33:273-277.
beta-hCG in clotted blood. Proc ACB Nat
Meeting. 1995:51. 45 DiMagno EP, Corle D, OBrien JF, Masnyk IJ,
34 Wood PJ, Donovan SJ, Humphries T. Primary Go VL, Aamodt R. Effect of long-term freezer
sampling for common endocrine tests: storage, thawing, and refreezing on selected
comparison of results from frozen serum and constituents of serum. Mayo Clin Proc.
primary blood samples stored at 4 degrees C 1989;64:1226-1234.
for two weeks. Proc ACB Nat Meeting. 46 Hirano K, Watanabe Y, Adachi T, Sugiura M.
1994:51. Carrier proteins in human fetal serum:
35 Lambert-Messerlian GM, Eklund EE, Malone bilirubin-binding abilities of albumin, alpha-
FD, Palomaki GE, Canick JA, DAlton ME. fetoprotein and ligandin. Chem Pharm Bull
Stability of first- and second-trimester serum (Tokyo). 1984;32:708-715.
markers after storage and shipment. Prenat 47 Hirano K, Watanabe Y, Adachi T, Ito Y,
Diagn. 2006;26:17-21. Sugiura M. The immunological cross-reaction
36 Eckfeldt JH, Long TA. Influence of laboratory between human alpha-fetoprotein and serum
test volume and geographic location on albumin and its relation with the binding
maternal alpha-fetoprotein results. Information ability. Chem Pharm Bull (Tokyo).
from the College of American Pathologists 1984;32:4979-4985.
Maternal alpha-fetoprotein Survey. Arch 48 Goldstein PJ, Sundaram SG, Manimekalai S.
Pathol Lab Med. 1991;115:647-653. Drug interference on alpha-fetoprotein assays.
37 Schreiber WE, Endres DB, McDowell GA, Md Med J. 1991;40:513-516.
Palomaki GE, Elin RJ, Klee GG, Wang E. 49 Henriques CU, Damm P, Tabor A, Pedersen
Comparison of fresh frozen serum to JF, Molsted-Pedersen L. Decreased alpha-
proficiency testing material in College of fetoprotein in amniotic fluid and maternal
American Pathologists surveys: alpha- serum in diabetic pregnancy. Obstet Gynecol.
fetoprotein, carcinoembryonic antigen, human 1993; 82: 960-964.
chorionic gonadotropin, and prostate-specific 50 Albert TJ, Landon MB, Wheller JJ, Samuels P,
antigen. Arch Pathol Lab Med. 2005;129:331- Cheng RF, Gabbe S. Prenatal detection of fetal
337. anomalies in pregnancies complicated by
38 Grenier A, Belanger L, Laberge C. Alpha1- insulin-dependent diabetes mellitus. Am J
fetoprotein measurement in blood spotted on Obstet Gynecol. 1996;174:1424-1428.
paper: discriminating test for hereditary 51 Guibaud S, Simplot A, Boisson C, Pison H.
tyrosinemia in neonatal mass screening. Clin Prenatal diagnosis of 4 cases of spina bifida in
Chem. 1976;22:1001-1004. mothers treated with valproate. J Genet Hum.
39 Katsumata Y, Oya M, Suzuki O, Kido A. 1987;35:231-235.
Identification of fetal blood stains by 52 Omtzigt JG, Los FJ, Hagenaars AM, Stewart
radioimmunoassay of alpha 1-fetoprotein. Z PA, Sachs ES, Lindhout D. Prenatal diagnosis
Rechtsmed. 1979;82:323-326. of spina bifida aperta after first-trimester
40 Wong PY, Mee AV, Doran TA. Studies of an valproate exposure. Prenat Diagn.
alpha-fetoprotein assay using dry blood-spot 1992;12:893-897.
samples to be used for the detection of fetal 53 Gross S, Castillo W, Crane M, Espinosa B,
neural tube defects. Clin Biochem. Carter S, DeVeaux R, Salafia C. Maternal
1982;15:170-172. serum alpha-fetoprotein and human chorionic
41 Mizejewski GJ, Carter TP, Beblowski DW, gonadotropin levels in women with human
109
Alpha-Fetoprotein
immunodeficiency virus. Am J Obstet measurement: a screening test for Down

Gynecol. 2003;188:1052-1056. syndrome. Lancet. 1984;1:926-929.
54 Baker M. Regulation of Estrogen Binding to 69 Hubinont C, Fisk NM, Nicolini U, Rodeck
Rat Alpha-Fetoprotein. New York: Academic CH, Johnson RD. Fetal alpha-fetoprotein
Press; 1985:79-104. concentration in growth retardation. Br J
55 Baker ME, Vaughn DA, Fanestil DD. Obstet Gynaecol. 1989;96:1233-1234.
Inhibition by protease inhibitors of binding of 70 Librach CL, Hogdall CK, Doran TA. Weights
adrenal and sex steroid hormones. J Supramol of fetuses with autosomal trisomies at
Struct. 1978;9:421-426. termination of pregnancy: an investigation of
56 Baker ME, Fanestil DD. Diethylpyrocarbonate the etiologic factors of low serum alpha-
inhibition of estrogen binding to rat alpha- fetoprotein values. Am J Obstet Gynecol.
fetoprotein: evidence that one or more 1988;158:290-293.
histidine residues regulate estrogen binding. 71 Mizejewski GJ, Bellisario R, Carter TP. Birth
Biochem Biophys Res Commun. 1981;98:976- weight and alpha-fetoprotein in the newborn.
982. Pediatrics. 1984;73:736-737.
57 Baker ME, Frecker DG, Fanestil DD. 72 Di Palma LVM, Coppola BL, Di Cesare M,
Inhibition of estrogen binding to rat alpha- Pipola E, Savarese M. Correlations existing at
fetoprotein by tryptophan p-nitrophenyl esters. birth and during the first days of life between
J Steroid Biochem. 1982;16:503-507. alpha1 fetoprotein and gestation age, weight
58 Baker ME, Sklar DH, Terry LS, Hedges MR. and morbid conditions. Minerva Pediatrica
Diethyl pyrocarbonate, a histidine selective [Italian]. 1978;30:313-325.
reagent, inhibits estrogen binding to receptor 73 Cederqvist LL, Spigelman S, Litwin SD. The
protein in rat uterus cytosol. Biochem Int. fetal serum alpha-fetoprotein and its
1985;11:233-238. relationship to immunoglobulins and birth
59 Gitlin D. Normal biology of alpha-fetoprotein. weight at term. Obstet Gynecol. 1983;61:233-
Ann N Y Acad Sci. 1975;259:7-16. 237.
60. Mizejewski GJ. Levels of alpha-fetoprotein 74 Finnstrom O, Karlsson B, Zetterlund B.
during pregnancy and early infancy in normal Studies on maturity in newborn infants. VIII.
and disease states. Obstet Gynecol Surv. Alpha-foetoprotein and albumin. Acta
2003;58:804-826. Paediatr Scand. 1975;64:409-412.
61 Masseyeff R, Gilli J, Krebs B, Calluaud A, 75 Bugajer-Gleitman HE, Eder G, Lachmann D,
Bonet C. Evolution of alpha-fetoprotein serum Pollak A. Preliminary investigations of the
levels throughout life in humans and rats, and serum alpha1 fetoprotein level in mature and
during pregnancy in the rat. Ann N Y Acad Sci. premature newborn infants [authors
1975;259:17-28. translation]. Wien Klin Wochenschr.
62 Hunter W, Knight GJ. Quality assurance in 1975;87:601-605.
MSAFP screening. Scarborough, ME: 76 Bryan EM, Thorp RW, Nicholson E. Serum
Foundation for Blood Research; 1981:25-34. alphafetoprotein in multiple pregnancy. Biol
63 Wepsic HT. Alpha-Fetoprotein: Its Neonate. 1978;34:40-47.
Quantitation and Relationship to Neoplastic 77 Edeling CJ, Schioler V, Thisted J. Alpha1
Disease. New York: Masson; 1981:115-129. fetoprotein in cord serum correlated to
64 Brock DJ, Bolton AE, Monaghan JM. Prenatal gestational age. Acta Obstet Gynecol Scand.
diagnosis of anencephaly through maternal 1977;56:15-17.
serum-alphafetoprotein measurement. Lancet. 78 Crandall BF, Lebherz TB, Schroth PC,
1973;2:923-924. Matsumoto M. Alpha-fetoprotein
65 Palomaki GE, Hill LE, Knight GJ, Haddow JE, concentrations in maternal serum: relation to
Carpenter M. Second-trimester maternal serum race and body weight. Clin Chem.
alpha-fetoprotein levels in pregnancies 1983;29:531-533.
associated with gastroschisis and omphalocele. 79 Chen RJ, Lin YH, Huang SC. Fetal sex and
Obstet Gynecol. 1988;71:906-909. maternal alpha-fetoprotein concentration at
66 Knight GJ. Serum alpha-fetoprotein screening. late normal singleton pregnancies. Acta Obstet
New York: Wiley-Liss; 1991:496-518. Gynecol Scand. 1994;73:192-194.
67 Johnson AM, Palomaki GE, Haddow JE. 80 Purdie DW, Young JL, Guthrie KA, Picton
Maternal serum alpha-fetoprotein levels in CE. Fetal growth achievement and elevated
pregnancies among black and white women maternal serum alpha-fetoprotein. Br J Obstet
with fetal open spina bifida: a United States Gynaecol. 1983;90:433-436.
collaborative study. Am J Obstet Gynecol. 81 Bansal V, Kumari K, Dixit A, Sahib MK.
1990;162:328-331. Alpha fetoprotein levels in newborn infants
68 Cuckle HS, Wald NJ, Lindenbaum RH. with reference to sex, gestational age and birth
Maternal serum alpha-fetoprotein weight. Indian J Exp Biol. 1989;27:666-667.
110
Alpha-Fetoprotein
82 Obiekwe BC, Malek N, Kitau MJ, Chard T. Suggested Website Locations:

Maternal and fetal alphafetoprotein (AFP)
levels at term. Relation to sex, weight and 1. en.wikipedia.org/wiki/Alpha-fetoprotein
gestation of the infant. Acta Obstet Gynecol
2. www.webmd.com/hw/being_pregnant/hw1663.
Scand. 1985;64:251-253.
asp
83 Sowers SG, Reish RL, Burton BK. Fetal sex-
3. www.medicinenet.com/Alpha-
related differences in maternal serum alpha-
fetoprotein_Blood_Test/article.htm
fetoprotein during the second trimester of
pregnancy. Am J Obstet Gynecol. 4. www.healthline.com/galecontent/alpha-
1983;146:786-789. fetoprotein-afp-test
84 James WH. Maternal alpha-fetoprotein levels 5. www.thedoctorsdoctor.com/labtests/alpha_feto
and sex of fetus. Acta Obstet Gynecol Scand. protein.htm
2006;85:635. 6. www.labcorp.com/datasets/labcorp/html/chapte
85 Leek AE, Ruoss CF, Kitau MJ, Chard T. r/mono/ri0ri019500.htm
Raised alpha-fetoprotein in maternal serum 7. www.pennhealth.com/health_info/Surgery/alph
with anencephalic pregnancy. Lancet. afetoprotein_1.html
1973;2:385. 8. www.ucsfhealth.org/childrens/medical_services
86 Cuckle HS, Holding S, Jones R, Groome NP, /preg/prenaenatal/afp.html
Wallace EM. Combining inhibin A with 9. www.americanpregnancy.org/prenataltesting/af
existing second-trimester markers in maternal p.html
serum screening for Downs syndrome. Prenat 10. www.marchofdimes.com/professionals/681_11
Diagn. 1996;16:1095-1100. 66.asp
87 Thomas RL, Blakemore KJ. Evaluation of 11. pregnancy.about.com/cs/afp/a/afptesting.htm
elevations in maternal serum alpha- 12. health.allrefer.com/health/alpha-fetoprotein-
fetoprotein: a review. Obstet Gynecol Surv. info.html
1990; 45: 269-283. 13. headtotoe.apta.org/topic/medtest/hw1663/result
88 Krause TG, Christens P, Wohlfahrt J, Lei U, s.htm
Westergaard T, Norgaard-Pedersen B, Melbye 14. www.pregnancy-
M. Second-trimester maternal serum alpha- info.net/slpha_feto_protien.html
fetoprotein and risk of adverse pregnancy
15. www.healthopedia.com/alpha-fetoprotein
outcome(1). Obstet Gynecol. 2001;97:277-
282. 16. chorus.rad.mcw.edu/to-go/00434.html
89 Van Lith JM, Pratt JJ, Beekhuis JR, Mantingh 17. www.calgarylabservices.com/LabTests/Alphab
A. Second-trimester maternal serum eticalListing//g/A/Alpha-1-Fetoprotein-
immunoreactive inhibin as a marker for fetal Pregnant.htm
Downs syndrome. Prenat Diagn. 18. www.pregnancy-guide.net/Medical-
1992;12:801-806. Tests/Alpha-Fetoprotein-Testing
90 Wald NJ, Cuckle HS. Open neural-tube 19. www.antibodyshop.com/products/research_pro
defects. In: Wald NJ, ed. Antenatal and ducts/markerker_molecules/alpha_fetoprotei
Neonatal Screening. New York: Oxford n_human
University Press; 1984:25-73. 20. www.obgyn.net/pb/pb.asp?page=/pb/articles/M
91 Mason MD. Tumour markers. In: Horwich A. SAFP_fetal_hydrocephalus
Testicular Cancer: Investigation and
Management. Baltimore: Williams and
Wilkins; 1991:33-50.
92 Schefer H, Mattmann S, Joss RA. Hereditary
persistence of alpha-fetoprotein: case report
and review of the literature. Ann Oncol.
1998;9:667-672.
93 Sharony R, Zadik I, Parvari R. Congenital
deficiency of alpha feto-protein. Eur J Hum
Genet. 2004;12:871-874.
111
Alpha-Fetoprotein
Table 1: Alpha-Fetoprotein Methods Summary Table

Method 1: Electroimmunodiffusion; immunoprecipitation
Principle of analysis: AFP is placed in an electric field, enabling it to migrate into an agarose gel containing
anti-AFP; extent of migration and height of precipitin line are proportional to AFP concentration.
Comments: Lacks sensitivity for serum analysis; historical, amniotic fluid only Method 2:
Radioimmunoassay; competitive binding assay
Principle of analysis: Labeled 125I-AFP competes with sample AFP for binding to anti-AFP antibodies.
Separation of bound from free ligand can be achieved by precipitation with a second antibody or by
immobilization of anti-AFP on a tube or microtiter well.
Comments: Reference method but uses radioisotopes; requires dilution to measure amniotic fluid AFP; serum
or amniotic fluid
Method 3: Immunoradiometric assay
125
Principle of analysis: Sample AFP binds to solid-phase immobilized anti-AFP antibody; I-labeled antibody
binds to immobilized AFP.
Comments: Commercial kits not available in the United States; rarely used; serum or amniotic fluid
Method 4: Immunometric assay
Principle of analysis: Same as for method 3, except the antibody is labeled with an enzyme, fluorescent, or
chemiluminescent label.
Comments: Most common assay type, measures AFP in maternal serum and amniotic fluid; suitable for cancer
monitoring
Table 2: Time Line of Aberrant Forms of Human Alpha-Fetoprotein Truncated Molecules

Resulting From Genetic Variants
Year Observer Observation Ref.
1975 Chandra Cystic fibrosis forms in serum BMJ 1:714
1975 Smith Aberrant form in CF in serum BMJ 2:392
1975 Fitzsimmons Aberrant form in CF in serum BMJ 3:544
1976 Biswas Aberrant form in CF in serum Clin Chem Acta 69:541
1976 Ramirez Aberrant from in CF in serum N Engl J Med 295:1381
1976 Norgaard-Petersen Cationic form of HAFP (electrophoresis) Clin Chem Acta 71:343 (AFP is
anionic)
1978 Marrink Cationic form* in serum Scand J Immunol* 8(Suppl 8):309
1978 Nishi Cationic form* in ELISA Scand J Immunol 8:35305
1982 Morinaga 54 Kd form in testis cancer
1983 Sarcione Breast cancer form in cytosol Cancer Res 43:3739
1983 Naketa Cationic form* in serum Oncodev Biol Med 4:101
1984 Lippes 54 Kd testicular/oviductal form Fertil Steril 59:148
1985 Sarcione 42Kd form in breast cancer cytosols Int J Cancer 35:315
1987 Sarcione Serum form in breast cancer, women Dis Markers 5:75
1987 Mizejewski Bound forms of AFP in serum CRC Press; pp.162
1990 Kronquist Aberrant pH form in Down syndrome Prenat Diagn 10:739
1990 Pavlov Aberrant form in CF in tissue Lab Delo 4:47
1999 Deutsch 52 Kd form in (liver cancer tissue) Tumour Biol 21:267
2000 Vakaria/GJM Conformationally transformed (serum) Breast Can Res Treat 63:41-52
2002 Kubota Avariant forms found in endoderm tissue J Biol Chem 277:27629
2005 Fukasawa 65 kD with missing signal sequence J Soc Gynecol Investig 12:456
(cell culture)
Data was extracted and summarized from References # 1, 4, 6, 7.
*HAFP in its native form in an anionic molecule
CF, Cystic fibrosis; ELISA, enzyme-linked immunosorbant assay.
112
Alpha-Fetoprotein
Table 3.1: Drug Interference of AFP Levels Measured by Radioimmunoassay and Enzyme
Immunoassay* - Insulin Related
Insulin Preparation Injection Type MS-AFP (ng/mL) Type of Immunoassay
Decrease RIA EIA
1. Human Recombinant-I 7%-8% X X
2. Iletin Purified Human 6%-14% X ND
3. Iletin Protamine-Zinc 8%-12% X ND
4. Iletin Purified Bovine 7%-8% X ND
5. Iletin Purified Porcine 6%-8% X X
6. Humulin L-1, R 7%-8% X X
Table 3.2: Drug Interference of AFP Levels Measured by Radioimmunoassay and Enzyme
Immunoassay* - Small Molecule Drug Related
Drug MSAFP (ng/mL) Type of Immunoassay
Decrease RIA EIA
1. Sulfasalazine 11% X ND
2. Erythromycin 9% X X
3. Propranolol 7%-10% X X
4. Methyldopa 37%-46% X X
5. Acetaminophen 16%-18% X X
6. Aspirin 19%-24% X X
EIA, Enzyme immunoassay; ND, not done; RIA, radioimmunoassay. (CIA, Chemiluminescent immunoassay was not
done.) From Goldstein PJ et al. Drug interference on AFP assays. Md Med J. 1991;40:513-516.
Table 4: The Stability of Human AFP Measured in Dried Blood Samples Reported for Years
1979 to 2000
Reference Citation Storage Results and/or Comment
Time Temperature Comment
1. Katsumata Y. Identification of fetal blood stains 1 month 25oC Stable (no
by RIA of alpha-fetoprotein. Z Rechtsmed. degradation)
1979;82:323-326.
2. Mizejewski GJ et al. Commercial RIA kit for 10 months 25oC and 4oC Stable (no
AFP measurement in newborn dried blood spots. degradation)
Clin Chem. 1982;28:1207-1210.
3. Dallaire L et al. Maternal sera AFP measured on 6 months 4oC Stable (no
dried blood spots. Prenat Diagn. 1982;2:265-271. degradation)
4. Wong et al. Studies of AFP measurement in 6 months 4oC Stable (no

dried blood spots for prenatal diagnosis. Clin degradation)
Biochem. 1982;15:170-172.
5. Mizejewski et al. AFP measured in infant dried 10-12 months 25oC and 4oC Stable (no
bloodspots. Pediatric Res. 1983;17:47-50. degradation)
6. Mizejewski et al. Birthweight and AFP in 10 months 25oC and 4oC Stable (no
newborns dried bloodspots. Pediatrics. degradation)
1984;73:736-737.
7. Augier D et al. Result of prenatal screening of Assayed 25oC Stable at time of
AFP in dried blood spots. STAT* testing
J Genet Hum. 1985;33:325-326. (Combined
with
ultrasound)
113
Alpha-Fetoprotein
8. Tsao et al. EIA for AFP on dried blood 1.0 month 37oC Stable (no
specimens. Clin Chem. 1986;32:2079-2082. degradation)
9. Fang M et al. [Measurement of Assayed 4oC Stable at time of
radioimmunoassay of alpha-fetoprotein in dried STAT* (mass 25oC testing
blood samples on paper for mass screening of screening)
hepatocellular carcinoma]. Kaku Igaku. 1986; 23:
1079-82..
10) Gonzalez C et al. Evaluation of MSAFP using 1.0 month 4oC Stable (no
dried blood samples. J Clin Chem Clin Biochem. 25oC degradation)
1988;26:79-84. 37oC
11) Verloes A et al. Non-radioactive assay of AFP Assayed 25oC Stable at time of
from dried bloodspot STAT* testing
Prenat Diagn. 1992;12:1073-1074.
12. Macri JN et al. Prenatal maternal AFP dried Prenatal 4oC Stable upon storage
blood screening. Am J Obstet Gynecol. Evaluation 25oC
1996;174:566-572.
13) Masse J. Transportation of material sera AFP Assayed STAT 0oC and 25oC, Adverse affect of
dried blood specimens. Clin Biochem. for season, seasonal variation summer transport on
2000;33:273-277. delay, and studied stability
freezing
variations
*STAT, Assayed within 1-2 days of arrival.

114
Alpha-Fetoprotein
Table 5: Prenatal Screening Timeline for Alpha-Fetoprotein Employed as a

Biomarker Alone or in Combination with Other Analytes
Year Observation/Event Reference
1972 Elevated AFP in amniotic fluid for neural tube Brock DJH et al: Lancet 2:191-
defects. Indication: potential biomarker 194.
1973 Elevated AFP in maternal serum for neural tube Leek AE et al: Lancet 2:385-386.
defects. Indication: potential biomarker
1980 Early antenatal diagnosis of ventral wall defect using Wald NJ et al: Lancet 1:368.
AFP.
1981 Amniotic fluid acetylcholinesterase diagnoses for Collaboration Study: Lancet
neural tube defects with elevated AFP levels. 2:321-323.
1984 Low maternal serum AFP levels discovered in Merkatz IR et al: Am J Obstet
prenatal Down-syndrome pregnancy samples. Gynecol 148:886-894.
1987 Combination of maternal age and AFP levels useful Cuckle HS et al: Br J Obstet
in Down syndrome pregnancies. Gynecol 94:387.
1989 Screening for Down syndrome using AFP, E3, and Wald NJ: Am J Human Genet
hCG (Triple Biomarker Test). 44:586,
1991 Low MS-AFP in congenital cardiac and Resta RG: Am J Med Genet
diaphragmatic defects. 40:129.
1992 Prenatal screen in maternal serum using multiple Haddow et al: N Engl J Med
markers for fetal distress (AFP, hCG, etc.). 327:588.
1994 Four-marker serum screening for Down syndrome Wald NJ: Prenat Diagn 14:707-
(quad test) using AFP, estriol, hCG, inhibin-A (2nd 716.
trimester)
2004 Combined (sequential) 1st and 2nd trimester Platt LD et al: 104:661-666.
screening for Down syndrome using PAPP-A, B-
hCG, followed by the AFP-triple test.
B-hCG, Beta hCG; estriol, unconjugated estriol; hCG, human chorionic gonadotrophin;
PAPP-A, pregnancy associated placental protein-A.
Table 6.1: Maternal Serum Alpha-Fetoprotein Concentrations by Gestational Week in Maternal Sera and
Amniotic Fluid (Median Values)
Gestation week 14 15 16 17 18 19 20 21 22
Maternal serum 22.2 25.5 29.4 33.6 38.6 44.5 51.0 58.5 67.2
(U/mL)
Amniotic fluid 19.2 17.4 13.8 11.0 8.8 7.0 5.5 4.4 3.5
(kU/mL)
kU/mL = 103 U/mL.
Table 6.2: Multiples of Median AFP Values in Maternal Serum
Gestational No. of Median Multiples of Median (MoM, ng/mL)
Week Samples ng/mL 2.0 MoM 2.5 MoM 3.0 MoM

15 347 14.0 62.8 78.5 94.2
16 412 36.3 72.6 90.8 108.9
17 320 41.9 83.8 104.8 125.7
18 333 48.5 97.0 121.3 145.5
19 201 56.1 112.2 140.3 168.3
20 77 64.8 129.6 162.0 194.4
*Medians are determined based on a weighted linear regression model.14
This laboratorys reportable range for AFP values in maternal serum is dependent on gestational week, as illustrated by
data obtained from 1690 serum samples at three sites. MS-AFP concentrations are reported in ng/mL.
115
Alpha-Fetoprotein
Multiples of Median (MoM, ng/mL) AFP Values in Amniotic Fluid
This laboratorys reportable range for AFP in amniotic fluid samples is dependent on gestational week, as illustrated by
data obtained from 666 amniotic fluid samples at two sites.
Table 7: AFP Values in Amniotic Fluid
Gestational No. of Median* Multiples of Median (MoM, g/mL)
Week Samples g/mL 2.0 MoM 2.5 MoM 3.0 MoM

15 92 17.3 34.6 43.3 51.9
16 138 14.4 28.8 36.0 43.2
17 152 11.9 23.8 29.8 35.7
18 133 9.9 19.8 24.8 29.7
19 103 8.1 16.2 20.3 24.3
20 48 6.7 13.4 16.8 20.1
*Medians are determined based on a weighted linear regression model.14
Note: 1 g/mL = 1000 ng/mL.
AFP concentrations are reported in g/mL
Table 8: AFP Concentrations in Benign and Malignant Disease
0- 10.1- 20.1- 500.0-

Sample No. of 10.0 20.0 500.0 1000.0 > 1000.0
Category Samples ng/mL ng/mL ng/mL ng/mL ng/mL
Apparently
Healthy Subjects 196 189 7 - - -
males 99 96 3 - - -
females 97 93 4 - - -
Testicular Cancer
seminoma 13 12 1 - - -
non-seminoma 62 30 1 23 3 5
Hepatoma
primary 16 6 - 4 - 6
secondary 5 4 1 - - -
Liver Disease
hepatitis 18 9 - 9 - -
cirrhosis 17 10 1 4 1 1
Other Cancer
ovarian 20 20 - - - -
genitourinary 15 15 - - - -
pancreatic 6 6 - - - -
In the above study, 96.4% of apparently healthy subjects had AFP values < 9.6 ng/mL, and 100%
percent of apparently healthy subjects had AFP values < 15.3 ng/mL.
116
Alpha-Fetoprotein
Table 9.1: Examples Using Alpha-Fetoprotein (AFP) as a Biomarker for Both Hepatic Disorders and
Malignant Tumors - Hepatic Disorders*
Liver Condition Number of Range of AFP Median AFP
Patients Concentrations Concentration
(ng/mL) (ng/mL)
A. Subacute hepatic necrosis 75 30 to 3300 900
B. Virus B hepatitis 101 40 to 405 350
C. Chronic active hepatitis 35 10 to 400 225
D. Alcoholic hepatitis without 40 30 to 440 75
cirrhosis
E. Primary biliary cirrhosis 35 10 to 250 50
*The tumors are presented according to tumor type and tissue of origin. The AFP was measured in serum and reported as
nanogram (ng)/mL concentrations.
Table 9.2: Examples Using Alpha-fetoprotein (AFP) as a Biomarker for both hepatic disorders and
malignant tumors - Malignant Tumors*
Tumor Type Tissue of Patient AFP Serum Level
Origin
1. Hepatocellular carcinoma Liver 35 adult males 1000 to 10,000,000
ng/mL
2. Lung tumor Lung 80-year-old male 17,000 ng/mL

3. Embryonal cell carcinoma Testicle 49-year-old male 1,700 ng/mL
4. Hepatoblastoma Liver 3-year-old female 332 ng/mL
5. Yolk sac tumor Ovary 22-year-old female 262 ng/mL
Data were extracted and compiled from the following references:
Wespsic HT. Alpha-fetoprotein: its quantitation and relationship to neoplastic disease. In: Kirkpatrick AM, Nakamura RM,
eds. Alpha-Fetoprotein Laboratory Procedures and Clinical Application. New York: Masson; 1981:115-129.
Taketa K. Multimodel application of lectin affinity electrophoresis of AFP. Electrophoresis. 1998;19:1774-1779.
Mizejewski GJ. Biological role of alpha-fetoprotein in cancer: prospects for anticancer therapy. Exp Rev Anticancer Ther.
2002;2:89-115.
*The tumors are presented according to tumor type and tissue of origin. The AFP was measured in serum and reported as
nanogram (ng)/mL concentrations.
Signifies years of age of male or female patient.

117
Aluminum
Aluminum
Tony Badrick
Name: Aluminum, aluminium

Clinical significance: Refer to Chapter 42, Trace Metals, in the 5th edition of Clinical Chemistry:
Atomic symbol: Al
Atomic weight: 26.98 D
CAS No: 7429-90-5
Chemical class: Metal
GFAAS is the most commonly used technique for the

Principles of Analysis and Current Usage determination of aluminum in serum, plasma, urine,
Aluminum is a periodic system group IIIB metal. It occurs dialysate fluid, and water [2-14]. This is because GFAAS
only in the +3 oxidation state. offers the best combination of sensitivity, simplicity, and
It is the most abundant metal found on earth [1] and is low cost. With this technique, the sample is placed in a
present in water and soil. Natural human exposure is graphite tube or pyrolytic platform, and in successive
unavoidable, with moderate amounts entering the body via steps, the temperature is raised to dry, char, and atomize
the ingestion of food and drink and the inhalation of dust. the sample into the measuring chamber, where absorption
of a specific wavelength is generated by a hollow cathode
Aluminum does not have a known biological role. Despite lamp. The method of standards addition is then used to
the high exposure, relatively little is absorbed, and this is allow matrix-matched calibration. There is usually no
readily excreted via the kidneys. Industrial aluminum preparation for water samples, and only a mixing step with
toxicity is rare. However, in the presence of renal failure, a wetting agent or dilute nitric acid for serum samples.
high levels of aluminum can accumulate in bone and Instrument detection limits extend down to concentrations
tissues. This increased absorption occurs because of (1) of approximately 0.1 to 0.2 mol/L for serum and 0.04 to
oral administration of aluminum hydroxide, used as a 0.12 mol/L for water, dialysate, or urine [15]. Within-run
phosphate binder, and (2) the presence of aluminum in the imprecision for serum and urine are approximately 3% and
water used to prepare dialysis fluid. This retained 5%, with between-run imprecision of the order of 7%.
aluminum can lead to dementia (dialysis encephalopathy),
microcytic anemia, and bone disease (osteitis fibrosa When used as a detector for high-performance liquid
cystica) [1]. chromatography (HPLC), GFAAS can analyze for species
of complexed or bound aluminum which have been
The current methods for the analysis of aluminum include separated into fractions on the chromatography column.
graphite furnace atomic absorption spectrometry
(GFAAS), flame atomic absorption spectrometry (FAAS), The use of inductively coupled plasma (ICP) techniques is
electrothermal atomic absorption spectrometry (ETAAS), now becoming more popular in clinical laboratories. There
neutron activation analysis (NAA), inductively coupled are two major variants, ICPmass spectrometry (ICP-MS)
plasmaatomic emission spectrometry (ICP-AES), and ICPatomic emission spectrometry (ICP-AES). These
inductively coupled plasma-mass spectrometry (ICP-MS), techniques offer rapid, highly sensitive, multi-element
and laser microprobe mass analysis (LAMMA). Front-end determinationsfor example, Al, Co, Cr, Mn, Ni, and Se
separation techniques such as chromatography are [16,17]. However, the equipment comes at a relatively
frequently coupled with analytical methods. high cost, with high facility costs as well (gas piping,
separate space, air-conditioning), and the technique is also
relatively complex, all of which limit its routine use in
many laboratories. Sample preparation is minimal and
i
involves a dilution which also reduces matrix effects that
Aluminum can cause signal instability due to the presence of
Previous and current authors of this method: dissolved solids. ICP-AES has lower sensitivity than
First edition: Not done ETAAS or ICP-MS, but it can handle greater levels of
Methods edition: Michael R. Wills, Sue Brown, John dissolved solids than ICP-MS and is faster than ETAAS.
Savory However, sample volume requirements for ICP-AES will
Second edition: Not updated generally be higher than for the other techniques.
Fourth edition: Not updated The ICP-AES technique, also referred to as ICPoptical
Fifth edition: Tony Badrick emission spectroscopy (ICP-OES), is an excellent
118
Aluminum
alternative to GFAAS for those laboratories possessing the abnormal results due to sample contamination, which can
appropriate instrumentation [18-21]. ICP-AES is a multi- lead to misinterpretation and misdiagnosis.
elemental technique that is relatively free of chemical
interferences. The matrix problems that can exist in atomic Urine specimens need not have preservatives added but
absorption spectrometry (AAS) are minimized in ICP-AES should be refrigerated to prevent bacterial growth [26].
because of the very high excitation temperature of the
sample [11]. The limits of detection for the ICP-AES Sample collection requirements for trace metal analysis are
method have been reported to be about 0.037 and 0.148 available from the International Union of Pure and Applied
mol/L (1 and 4 g/L) of urine and blood, respectively Chemistry [27].
[18]. Two major problems with using the ICP-AES
technique are the intense and broad emission of calcium, Interferences
which increases the aluminum background and can raise There are three general forms of interference in atomic
the detection limit for this element and interference from absorption spectrophotometry: chemical, ionization, and
Titanium [19-21]. matrix. Chemical interference occurs when the metal
cannot be dissociated into free atoms because of certain
Inductively coupled plasmamass spectrometry (ICP-MS) complexes. This is generally overcome by the use of high
is a powerful technique that uses an inductively coupled temperatures, which force dissociation, or by the addition
plasma as an ion source and a mass spectrometer as an ion of another cation that competes with the complexing
analyzer. It can measure the presence of > 75 elements in a interferents.
single scan and can achieve detection limits down to parts
per trillion (ppt) for many elementsconcentrations that Ionization interference occurs when the atom of interest is
are two or three orders of magnitude lower than those excited beyond the ground state and emits rather than
obtained by ICP-AES [22]. ICP-MS is more expensive absorbs. This is temperature dependent and can be reduced
than ICP-AES and requires more highly skilled technical by adding an excess of more easily excited atoms to the
operation. Aluminum concentrations in urine and saliva mix.
were detected down to 0.1 mol/L and in blood serum to
0.004 mol/L, using ICP-MS [23]. Speciation studies have Matrix interference occurs when there is a reaction
employed ICP-MS as a detector for aluminum in tissue between the atom and the matrix in the reaction cell.
fractions separated by size-exclusion chromatography Examples of these reactions include enhanced absorption
(SEC) in femur, kidney, and brain [24]. in the presence of certain solvents and the formation of
solids which produce physical interference in the light
Reference and Preferred Methods path. Zeeman correction and continuum-source
The reference method is AAS (see Method 1 below), but background-correction systems are used to reduce this type
ICP (Method 2) is also acceptable. The choice of method of interference.
for aluminum determination depends upon the equipment
available in the laboratory. At the time of writing this In the case of aluminum estimation by ETAAS, various
review, about half of the participants in the College of methods used to reduce interference have been compared
American Pathologists trace metals QA program were [28]. The finding was that there is no significant difference
using AAS, and approximately half were using ICP-MS. between (1) a Perkin Elmer (PE) Model 3110 AAS
equipped with a longitudinally (end) heated graphite
Specimen atomizer (HGA) and continuum-source (deuterium)
Aluminum is measured in serum, urine, dialysate, water, background correction; (2) a PE Model 4100ZL AAS
and bone [25]. Collection of specimens for serum equipped with a transversely heated graphite atomizer
aluminum analysis can be a complicating factor. Most of (THGA) and longitudinal Zeeman background correction;
the common evacuated blood collection devices used in and (3) a PE Model Z5100 AAS equipped with an HGA
phlebotomy today have rubber stoppers made of aluminum and transverse Zeeman background correction.
silicate. Simple puncture of the rubber stopper for blood
collection is sufficient to contaminate the sample with Aluminum Reference Intervals
aluminum sufficient to produce an abnormal concentration Serum aluminum concentrations have been reported as
of the metal. Typically, blood collected in standard 0.07 to 0.56 mol/L [13]. The upper reference limit for
evacuated blood tubes will be contaminated by 0.74 to aluminum in a healthy, nonexposed population is reported
2.23 mol/L of aluminum. This can be readily as 0.6 mol/L in urine [29].
demonstrated by collecting blood from a normal volunteer
into a standard evacuated phlebotomy tube. Special Conversion factor mol/L = g/L 0.0371.
evacuated blood collection tubes are required for
aluminum testing. These tubes are readily available from Clinical signs of aluminum toxicity occur at concentrations
commercial suppliers and should always be used. Failure > 7.4 mol/L, but levels of > 3.7 mol/L require
to pay attention to this issue can result in the generation of surveillance. A level > 2.2 mol/L warrants attention.
119
Aluminum
Interpretation 5 Bouman AA, Platenkamp AJ, Posma FD.

For more than 25 years, aluminum toxicity has been Determination of aluminium in human tissues by
recognized as a potential hazard to patients with end-stage flameless atomic absorption spectroscopy and
renal disease (ESRD) [30]. The biological effects of comparison of references values. Ann Clin
aluminum toxicity have been of considerable interest, Biochem 1986; 23: 97-101.
particularly in uremic patients, where toxins may be 6 Couri D, Liss L, Ebner K. Determination of
involved in complex interactions with aluminum-mediated aluminium in biological samples.
response mechanisms. Neurotoxicology 1980; 1: 17-24.
7 Gardiner PE, Stoeppler M. Optimisation of the
Aluminum concentrations in blood are not a reliable analytical conditions for the determination of
marker of aluminum absorption or organ load in dialysis aluminium in human blood plasma and serum by
patients, only stainable aluminum at the mineralization graphite furnace atomic absorption spectrometry.
front reflects the histopathological changes observed in Part 2. Assessment of the analytical method. J
bone. In bone, aluminum exerts three general pathological Anal Atom Spectrom 1987; 2: 401-404.
effects: (1) inhibition of hydroxyapatite formation and 8 Gorsky JE, Dietz AA. Determination of
growth; (2) inhibition of bone cell proliferation; and (3) aluminium in biological samples by atomic
suppression of bone cell activity. These pathological absorption spectrophotometry with a graphite
effects of the metal lead to inhibition of bone furnace. Clin Chem 1978; 24: 1485-1490.
mineralization, decreased bone formation, and lower bone 9 Guillard O, Tiphaneau K, Reiss D, Piriou A.
mass. Patients with stainable bone aluminum also have a Improved determination of aluminium in serum
higher volume of lamellar osteoid, a lower volume of by electrothermal atomic absorption spectrometry
woven osteoid, and significantly lower numbers of both and zeeman background correction. Anal Lett
osteoclasts and osteoblasts [31]. 1984; 17: 1593-1605.
10 Keirsse H, Smeyers-Verbeke J, Verbeelen D,
Aluminum Performance Goals Massart DL. Critical study of the speciation of
The analytical performance of serum aluminum aluminium in biological fluids by size-exclusion
estimations in European laboratories as judged by various chromatography and electrothermal atomic
external quality assurance programs in 2002 was poor, absorption spectrometry. Anal Chim Acta 1987;
with up to 26% of enrolled laboratories failing to achieve 196: 103-114.
agreed performance limits [32]. 11 Savory J, Wills MR. Analytical methods for
aluminium measurement. Kidney Int 1986;
Recommended performance goals for serum aluminum 29(Suppl 18): S24-S27.
levels are a total error of 0.186 mol/L (5 g/L) or 12 van der Voet GB, de Haas EJM, de Wolff FA.
20% of the target on an external quality assurance program Monitoring of aluminium in whole blood, plasma,
[33]. These performance goals were set on the basis of serum, and water by a single procedure using
biological variation in normal subjectsthat is, the flameless atomic absorption spectrophotometry. J
variation in serum aluminum levels measured biweekly for Anal Toxicol 1985; 9: 97-100.
30 weeks. 13 Wrobel K, Gonzalez EB, Sanz-Medel AE
Aluminium and silicon speciation in human
References serum by ion-exchange high performance liquid
1 Agency for Toxic Substances and Disease chromatography-electrothermal atomic absorption
Registry, U.S. Department of Health and Human spectrometry and gel electrophoresis. Analyst
Sources. Toxicological Profile for Aluminium. 1995; 120: 809-815.
September 2006. 14 Xu N, Majidi V, Markesbery WR, Ehmann WD.
http://www.atsdr.cdc.gov/toxprofiles/tp22.html. Brain aluminium in Alzheimers disease using an
Accessed 2 July 2007. improved GFAAS method. Neurotoxicology
2 Alder JF, Samuel AJ, West TS. The single 1992; 13: 735-743.
element determination of trace metals in hair by 15 Parkinson IS, Ward MK, Kerr DN. A method for
carbon-furnace atomic absorption spectrometry. the routine determination of aluminium in serum
Anal Chim Acta 1976; 87: 313-321. and water by flameless atomic absorption
3 Alderman FR, Gitelman HJ. Improved spectrometry. Clin Chim Acta 1982; 125: 125-
electrothermal determination of aluminium in 133.
serum by atomic absorption spectroscopy. Clin 16 Bianchi F, Maffini M, Mangia A, Marengo E,
Chem 1980; 26: 258-260. Mucchino C. Experimental design optimization
4 Bettinelli M, Baroni U, Fontana F, Poisetti P. for the ICP-AES determination of Li, Na, K, Al,
Evaluation of the lvov platform and matrix Fe, Mn, Zn in human serum. J Pharmaceutical
modification for the determination of aluminium Biomedical Analysis 2007; 43: 659-665.
in serum. Analyst 1985; 110: 19-22. 17 Dillio S, Violante N,Caimi S, Di Gregorio M,
Petrucci F, Senofonte O. Determination of trace
120
Aluminum
metals in serum by dynamic reaction cell 30 Alfrey AC, LeGendre Gr, Kaehny WD. The
inductively coupled plasma mass spectrometry. dialysis encephalopathy syndrome. Possible
Anal Chim Acta 2005; 573: 432-438. aluminium intoxication. N Engl J Med 1976; 294:
18 Allain P, Mauras Y. Determination of aluminium 184-188.
in blood, urine, and water by inductively coupled 31 Monier-Gaugere MC, Malluche HH. Trends in
plasma emission spectrometry. Anal Chem 1979; renal osteodystrophy: a survey from 1983 to 1995
51: 2089-2091. in a total of 2248 patients. Nephrol Dial
19 Lichte FE, Hopper S, Osborn TW. Determination Transplant 1996; 11: 111-120.
of silicon and aluminium in biological matrices by 32 Taylor A, Angerer J, Claeys F, Kristiansen J,
inductively coupled plasma emission Mazarrasa O, Menditto A, et al. Comparison of
spectrometry. Anal Chem 1980; 52: 120-124. procedures for evaluating laboratory performance
20 Que Hee SS, Boyle JR. Simultaneous multi- in external quality assessment schemes for lead in
elemental analysis of some environmental and blood and Aluminium in serum demonstrates the
biological samples by inductively coupled plasma need for the common quality specifications. Data
atomic emission spectrometry. Anal Chem 1988; Supplement Clinical Chemistry Online
60:1033-1042. (http://www.clinchem.org/content/vol48/issue11/
21 Sanz-Medel A, Roza RR, Alonso RG, Vallina accessed 2007-10-22).
AN. Atomic spectrometric methods (atomic 33 Taylor A, Angerer J, Claeys F, Kristiansen J,
absorption and inductively coupled plasma atomic Mazarrasa O, Menditto A, et al. Comparison of
emission) for the determination of aluminium at procedures for evaluating laboratory performance
the parts per billion level in biological fluids. J in external quality assessment schemes for lead in
Anal Atom Spectrom 1987; 2: 177-184. blood and Aluminium in serum demonstrates the
22 Keeler R. ICP mass spectrometry shows its need for the common quality specifications. Clin
mettle. Res Dev 1991; 33: 44-48. Chem 2002; 48: 2000-2007.
23 Ward NI. Environmental contamination of 34 Standards Australia. Analysis of serum and
aluminium and other elements in North Cornwall plasma for trace elements Determination of
as a result of the Lowermoor water treatment aluminium content Graphite furnace atomic
works incident. In: Vernet J-P, ed. Heavy metals absorption spectrophotometric method (Australian
in the environment. Edinburgh: CEP Consultants, Standard AS 4195.1-1994)
1989;118-121. http://www.saiglobal.com/shop accessed 4 July
24 Owen LMW, Crews HM, Bishop NJ, Massey RC. 2007.
Aluminium uptake from some foods by guinea
pigs and the characterization of aluminium in vivo METHOD 1: ELECTROTHERMAL ATOMIC
intestinal digestion by SEC-ICP-MS. Food ABSORPTION SPECTROSCOPY
Chem.Toxicol 1994; 32: 697-705.
25 Tang S, Parsons PJ, Slavin W. Rapid and reliable The aluminum in serum or plasma is determined by
method for the determination of aluminium in graphite furnace atomic absorption spectrometry. Analysis
bone by electrothermal atomic absorption of aluminum is routinely performed by atomic absorption
spectrometry. Analyst 1996: 121: 195-200. spectrometry with electrothermal atomization. The
26 Bornhorst JA, Hunt JW, Urry FM, McMillin GA. following method is based upon the methods of Standards
Comparison of sample preservation methods for Australia [34]. At the time this section was written, there
clinical trace element analysis by inductively was no other standard method available.
coupled plasma mass spectrometry. Am J Clin
Path 2005; 123: 578-583. There are a range of external quality assurance programs
27 Cornelis R, Heinzow B, Herber RFM, Molin available that have serum aluminum as an analyte. There
Christensen J, Paulsen OM, Sabbioni E et al are also more specialized programs available which also
Sample collection guidelines for trace elements in have water as a matrix [33].
blood and urine. Pure Appl Chem 1995; 67: 1575-
1608. Principle
28 Kruger PC, Parsons PJ. Determination of serum Calibrators, controls, or patient specimens are mixed with
aluminium by electrothermal atomic absorption magnesium nitrate and a surfactant to prepare them for
spectrometry: A comparison between Zeeman and graphite furnace atomic absorption analysis. This matrix
continuum background correction systems. forms a complex with aluminum that has good thermal
Spectrochimica Acta Part B 2007; 62: 288-296. stability; this allows for in situ ashing of the specimen to
29 Valkonen S, Aitio A. Analysis of aluminium in vaporize aqueous and organic background materials. The
serum and urine for the biomonitoring of final step of analysis causes vaporization of aluminum,
occupational exposure. The Science of the Total which absorbs energy at the 309.3-nm line emitted from an
Environment 1997; 199: 103-110. aluminum hollow cathode lamp. Absorbance of energy at
121
Aluminum
this wavelength is specific for aluminum and proportional volumetric flask containing approximately 200 mL of
to its concentration. water and 10 mL of nitric acid . Dilute the contents of the
flask to volume with water. This solution should be used
Instrumentation within 4 hours of preparation.
A good-quality graphite furnace atomic absorption
spectrometer located in a dust-free environment is required
for analysis. The procedure requires that a Lvov platform APPARATUS
be placed in the graphite tube to enhance sensitivity and
accuracy, and the Zeeman background correction is Glassware. Volumetric glassware of grade A or B should
required to reduce background to a minimum. be used throughout. All glassware should be cleaned with
nitric acid and rinsed thoroughly with water before use.
Use of volumetric glassware should comply with AS 2162.
REAGENTS Volumetric flasks should comply with AS 2164. Pipettes
should comply with AS 2166.
General Requirements:
All reagents shall be of analytical reagent grade quality. Plasticware. Polyethylene or PTFE containers are
Grade 1 water, as specified in ISO 3696, is used preferred. Polystyrene containers have been found to be
throughout. The water should contain less than 0.01 mol unsuitable for diluted samples. Plasticware and automatic
of aluminum per liter. sampler cups should be soaked for between 30 min and 1h
in EDTA cleaning solution, followed by rinsing with
Solutions: copious quantities of water. Dry in a clean atmosphere.
EDTA cleaning solution (2% m/V). Dissolve 2 g of Micropipettes and microsample injectors. Micropipettes
ethylenediamine N,N,N tetra-acetic acid disodium salt and microsample injectors should be shown to have an
(EDTA) per 100 mL of water and mix. accuracy of 2% and a precision of better than 2% and to
be chemically inert to the reagents used.
Diluting solution. Prepare 100 mL of a 0.2% (m/V)
solution of surfactant in water. The solution should be Atomic absorption spectrometer. The atomic absorption
shown to contain less than 0.1 mol of aluminum per liter. spectrometer, equipped with background correction and a
NOTE: Triton X-100 has been found to be a suitable graphite furnace atomizer, should be prepared in
surfactant for this application. accordance with AS 2134.2.
Human plasma or serum (for the preparation of Signal recording. The instrument should have facility for
calibration solutions). The serum or plasma should be real-time display. If a chart recorder is used, it should have
shown to have less than 0.5 mol of aluminum per liter. a full-scale response time of less than 0.5 s for a chart
Nitric acid concentrated (p20 1.42 g/mL). width of not less than 250 mm.
Nitric acid cleaning solution. Add 200 mL of nitric acid Purge gas. Argon is recommended for use as a purge gas.
to 800 mL of water and mix.
INSTRUMENT OPERATING PARAMETERS
Standard Solutions:
Analytical parameters. A variety of furnace atomizers are
Stock standard solution (50 mmol/L). Weigh 1.349 available, and each manufacturers instruction manual
0.001g of aluminum wire (>99.9% Al), and transfer to a 1 should provide adequate information for their operation.
L volumetric flask. Add 0.1g of mercurous (ll) nitrate
(Hg[NO3]2) to catalyze dissolution, 5 mL of water and 20 Wavelength. The wavelength 309.3 nm produced by an
mL of nitric acid. When dissolution is complete, dilute to aluminum line source has been used in the development of
volume with water, and store in a polyethylene bottle. this method.
NOTES: Injection Volume. To limit calibration curvature, the

1. This solution has been shown to be stable for at volume injected should be such that the absorbance
least 3 months. difference between the zero concentration (C0) and the
2. Alternative standard solutions may be used in a highest concentration calibration solution (C0) is not less
preparation of this stock solution, provided the than 1.7 times the absorbance difference between the zero
traceability to recognized national standards body concentration (C0) and the mid-range concentration
can be demonstrated. calibration solution (C0). This volume should remain
constant for the calibration solutions and the test samples.
Intermediate standard solution (200 mol/L). Pipette
4.00 mL of stock aluminum solution into a 1 liter
122
Aluminum
Furnace operating parameters. The following (Co to C6) and then two volumes of diluting
information is given as a guide: solution and mix thoroughly.
(i) Dry at approximately 130C. Optimize the
drying temperature and its duration by (b) Adjust the instrument read-out to zero.
observing the droplet of the sample in the
atomizer during the drying cycle. The droplet (c) Inject the injection volume of the diluted Co
should evaporate rapidly but not boil during solution into the furnace atomizer.
this process.
(ii) Ash at approximately 1500C. Optimize the (d) Carry out the dry, ash, atomize, and clean
ashing temperature and its duration so that cycle, and record the absorbance signal (Ao)
aluminum is not volatilized. peak area or maximum peak height.
(iii) Atomize at approximately 2600C. Optimize
the atomization temperature and its duration (e) Repeat the injection, and assess the
so that all aluminum is volatilized (the acceptability of the absorbance signals in
absorption signal should return to the accordance with the procedure outlined in AS
baseline at least 1 s before the end of the 2134.2.
atomization cycle).
(iv) Clean the furnace at the atomizing (f) Repeat steps (b) to (e) for the remaining
temperature or higher for 3 to 5 s with the gas diluted calibrating solutions C1 to C6, and
flow on. record the absorbance signals A1 to A6.
Instrument contamination. Special attention should be Prepare a calibration graph of absorbance versus
paid to graphite components, sampler assemblies, and rinse concentration of aluminum in mol/L.
fluids.
METHOD 2: INDUCTIVELY COUPLED PLASMA
CALIBRATION [18]
Preparation of calibrating solutions. The calibrating Reagents
solutions should be prepared in serum or plasma and in A stock solution of 1000 g/mL of Al is obtained by
accordance with Table 1. dissolving 1.000g of aluminum foil (analytical-reagent
grade, E. Merck) in 20 mL of 1 + 1 sulfuric acid (Carlo-
Erba) and diluting to 1000 mL with ultra pure water and 50
Table 1: Aluminum Concentrations for mL of concentrated nitric acid (E. Merck). An intermediate
Calibration Graph solution of 100 mg/L of Al is obtained by diluting 10 mL
of the above solution to 100 mL with 1 + 20 dilute nitric
Calibration Volume of Volume Added acid.
Solutions Intermediate of Aluminum Standard solutions are obtained by direct dilution of the
Standard Plasma mol intermediate solutions or stock solutions with ultra pure
Solution or water (Milli-Q).
(mL) Serum
Absorption Measurements
mL Serum and urine are analyzed directly after dilution 1 + 1

C0 0 10.0 0.0 with ultra pure water. The calibration graph is constructed
C1 0.05 9.95 1.0 using the aqueous standards, and the instrumental
C2 0.10 9.90 2.0 conditions used are described in Table 1. Further dilution
C3 0.15 9.85 3.0 of dialysis fluids (concentrate diluted 1 + 35) or tap-water
C4 0.20 9.80 4.0 samples is not necessary.
C5 0.25 9.75 5.0
C6 0.30 9.70 6.0 Sample preparation for ICP emission analysis is as
follows. The samples are diluted 1 + 2 for serum and 2 + 3
NOTE: These calibrating solutions may be further for urine with standard 150 g/L Al solution (spike), and
dispensed into plastic tubes in smaller aliquots and stored the concentration of the metal is determined by difference
frozen at 20C for up to 6 months. after analyzing the spiked solutions by reference to a
calibration graph prepared from Al in NaCl (1100 100
Preparation of calibration graph procedure: ppm of Na+) solution. A 5 L aliquot of sample solution is
pipetted onto the graphite rod and vaporized into the
(a) To a series of plastic tubes or sample cups, plasma. If the saline content of the sample is high (e.g.,
add one volume of each calibrating solution dialysis solutions or serum), dilution (1 + 1) with water is
carried out.
123
Aluminum
Special care has to be exercised in order to avoid possible Appendix: Abbreviations Used in the Text
contamination along the three main stages (all sources of
AAS Atomic absorption spectrometry
contamination) of analysis: sampling-storage, pre-
ETAAS Electrothermal atomic absorption spectrometry
treatment and measurement.
FAAS Flame atomic absorption spectrometry
Sampling has to be carried out in a way to ensure that GF Graphite furnace atomic absorption spectrometry
manipulations and material or utensils do not release HGA Heated graphite atomizer
aluminum. Storage of samples should be carried out in ICP-AES Inductively coupled plasmaatomic emission
special plastic vials previously treated with acid or EDTA. spectrometry
We have successfully used polystyrene tubes previously ICP-MS Inductively coupled plasmamass spectrometry
treated with 10% nitric acid for 24 h, followed by rinsing ICP-OES Inductively coupled plasmaoptical emission
with copious amounts of Milli-Q water. Regular testing for spectrometry
possible aluminum release from these tubes is performed LAMMA Laser microprobe mass analysis
for every new batch received. NAA Neutron activation analysis
PE Perkin Elmer
Preparation of the sample, reagents, and the like, is another SEC Size exclusion chromatography
important possible source of contamination. In the THGA Transversely heated graphite atomizer
proposed method, there is only a 1 + 1 dilution with water,
so the only recommendation is to use ultra-pure water
(e.g., Milli-Q water). However, regular determination of
the aluminum content by GFAAS of such water is also
advisable.
Instrumental measurement can also introduce a

contamination risk. It is recommended that all sample
manipulations, including the final measurement, should be
carried out in a Class 100 laminar-air-flow clean bench.
We have verified the importance of such clean-room
conditions for Al concentrations 10 g/L; detection
limits attained are about three times lower than those
obtained out of the clean room, and it was confirmed by
experiment that precision at concentrations of 2 to 5 g/L
is much better with complete clean-room conditions.
124
Amino Acid Screening

Kevin Carpenter
Name: Amino acid screening

Clinical significance: Refer to Chapter 52, Diseases of Genetic Origin, in the 5th edition of Clinical
Principles of Analysis and Current Usage iodide reagent can be used as an overstain after reading the
Inborn errors of metabolism affecting amino acid Ninhydrin stain to locate sulfur-containing amino acids,
metabolism were among the first to be described, and from a which appear white or cream against a pink background.
historical perspective, testing for inborn errors was often
limited to qualitative tests for amino acids in plasma or Location of amino acids by Ninhydrin may be almost
i universal, but the range of techniques available to effect
urine.
While there is no doubt the amino acid disorders are now separation of the amino acids in physiological fluids is wide.
only a small part of the inherited metabolic disease There are two main categories of separation techniques:
spectrum, there is still a place for initial screening tests chromatography and electrophoresis. These may be used
which, although rarely diagnostic, may give insight into the singly in one or two dimensions or in combination as a two-
confirmatory investigations required. Amino acid screens dimensional separation.
are usually run in conjunction with organic acids, and
interpretation of both requires considerable experience and Chromatography of amino acids was originally performed
is usually confined to laboratories with a special interest in on paper, commonly using n-butanol/acetic acid/water
inborn errors of metabolism. solvent systems running overnight [1]. Enhanced separation
may be achieved by running in a second solvent at 90
Quantitative analysis of amino acids in physiological fluids degrees to the original, using t-butanol/methyl ethyl
is a specialist investigation and outside the scope of this ketone/ammonia solvents for the second separation.
chapter, but it is essential to confirm findings from Resolution on paper can be remarkably good, but the
screening tests and to monitor treatment. medium has been largely superseded by the use of
commercially available thin-layer cellulose plates which
One of the reasons amino acid disorders figured so afford excellent and reproducible separations but may
prominently in the early descriptions of inborn errors of require de-salting of urine samples prior to analysis.
metabolism was the availability of a suitable locating Multiple samples per sheet can be applied if only a single
reagent for various chromatographic techniques. Ninhydrin dimensional separation is run.
(triketohydrindene hydrate), has been in use for over 90
years and reacts with primary or secondary amines through a The major disadvantage of chromatographic separation is
complex process to produce Ruhemanns purple. All - the time required to obtain a result. Typical separations are
amino acids will react with Ninhydrin at room temperature, run overnight, and if a rapid result is required, this can be an
and heating to 105C will allow all compounds containing a issue. A more rapid alternative to paper or thin-layer
primary or secondary amine to react. Although the basic chromatography is electrophoresis. Cellulose acetate
reaction results in production of a purple color, there are electrophoresis using a formic acid/acetic acid buffer at pH
many subtle variations in the colors different amino acids 1.9 can give reasonable separation in 20 to 30 minutes at
yield, which may be of value in identifying poorly separated voltages of 200-300 V [2].
compounds. The amino acids proline and hydroxyproline,
for example, react in a different manner to produce a Even better separations can be achieved using paper as the
characteristic yellow color. Ninhydrin staining may be support media and operating at high voltages [3]. This
conveniently achieved for chromatography, simply by method generates considerable heat and requires
running the chromatography plate or paper a second time in instrumentation incorporating a cooling plate and suitable
a solvent containing Ninhydrin. Alternatively, spraying or safeguards against electric shock. However, modern ceramic
dipping in the stain dissolved in acetone can be used for cooling plates linked to a recirculating coolant reservoir all
chromatography or electrophoresis on any support media. but eliminate problems of the paper drying out during the
run, and excellent results are obtained using 2500 V, 20-
Different stains, often used in conjunction with Ninhydrin, minute separation on 265 200 mm paper sheets which can
can enhance detection of specific species, and platonic comfortably fit up to 9 urine samples per sheet.
For all electrophoresis methods, samples are applied close to

i
Amino acid screening the anode end with the dibasic amino acids, ornithine,
Previous and current authors of this method: lysine, and arginine migrating the furthest, and acidic
First edition: Zulfikarali H. Verjee compounds taurine and phosphoethanolamine remaining
Methods edition: Not updated close to the origin. S-sulfocysteine (a marker for sulfite
Second edition: Not updated oxidase deficiency) migrates slightly towards the anode, so
Third edition: Not updated samples are applied a few cm above the anode position to
Fourth edition: Not updated allow for this.
Fifth edition: Kevin Carpenter
125
Electrophoresis does not completely separate amino acids Diet influences the pattern of amino acids seen, with a
with similar isoelectric points, but when used in generalized increase in excretion postprandially and specific
combination with chromatography run at 90 degrees to the increases in carnosine, anserine, and 1-methylhistidine
electrophoresis, the techniques are complimentary, enabling following white poultry meat ingestion. Catabolism will
separation of amino acids neither technique in isolation can result in a decrease in alanine concentration and an increase
achieve [4]. The downside to the enhanced separation is that in -aminoisobutyrate concentration.
only one sample per sheet can be applied, and therefore
throughput is limited. Decreased levels of serine and glutamine may be seen in
bacterially contaminated urine samples.
There are no reference methods for amino acid screening. Interpretation
Interpretation of the profiles obtained in both urine and
The American College of Medical Genetics publishes plasma is a highly skilled job requiring experience of what
Standards and Guidelines for Clinical Genetics Laboratories to expect in normal and disease states, and the information
[5]. The 2006 edition states, Qualitative amino acid given here should be considered a very basic guide only. It
analysis must reliably detect conditions in which there are must also be noted that not all aminoacidurias result in
either gross or modest elevations of specific amino acids in clinical disease [6].
blood and/or urine but warns Qualitative amino acid
analysis by thin-layer chromatography (TLC) is suitable In normal plasma, the most prominent amino acids are
only for the detection of gross abnormalities. As some glutamine and alanine, with lesser amounts of glycine,
disorders may be missed by this method, its use for the valine, lysine, leucine, serine, and threonine usually visible.
purpose of evaluating high-risk patients should be The aromatic amino acids phenylalanine and tyrosine give
discouraged. quite faint bands, as do the remainder of the physiological
amino acids.
However, TLC is widely used for initial amino acid
screening, and provided the limitations are understood, it In normal urine, age is an important factor in the amino acid
can still yield valuable information. For laboratories setting excretion pattern. Babies often show an immaturity of renal
up amino acid screening for the first time, TLC will give reabsorption mechanisms, resulting in an increased
reliable results in most hands, whereas the enhanced excretion of proline, hydroxyproline, glycine, and
resolution achieved by high-voltage electrophoresis requires sometimes cystine and lysine. Taurine is also prominent at
more specialized equipment and considerable skill to get birth but declines rapidly. Generally heavy patterns
consistent results. (nonspecific increases in all amino acids) are seen in the
neonatal period, gradually declining to adult-type patterns
Specimen during childhood and adolescence.
Plasma or urine can be used for amino acid screens, and
many labs request both. This can be useful where increased A normal urine pattern is dominated by glycine, with lesser
concentrations of particular amino acids in urine may be due and similar-intensity bands seen for alanine, serine, and
to their accumulation in blood or as a result of impaired glutamine. Lysine, cystine, and the aromatic amino acids
renal reabsorption. Comparison of the patterns obtained in produce only very faint bands in the healthy individual.
the two samples will reveal the source of the increase.
Plasma alone will not be informative for detection of renal Inborn errors of amino acid metabolism will usually result in
transport defects such as cystinuria or detection of accumulation of the amino acid proximal to the defect in the
generalized aminoaciduria such as is seen in Fanconi synthetic pathway in both plasma and urine. Thus
syndrome. Plasma requires deproteinization prior to phenylketonuria gives elevated phenylalanine, nonketotic
running, and urine samples may need removal of inorganic hyperglycinemia gives elevated glycine, and maple-syrup
salts by ion exchange resin if TLC is used. urine disease shows increases in valine, leucine, and
isoleucine. Urea-cycle defects will all give elevated
Samples should be stored and transported at or below 20C glutamine, but argininosuccinate synthetase deficiency will
to prevent bacterial degradation. Under such conditions, also result in elevated citrulline, and argininosuccinate lyase
most amino acids are stable for many years. Typical deficiency will show the presence of argininosuccinate and
volumes required for chromatographic or electrophoretic its anhydride.
techniques are only a few hundred microliters. Urine
samples should be loaded or diluted according to creatinine In all instances, interpretation should be performed in light
concentration. of the clinical condition of the patient and preferably in
conjunction with organic acid profile results.
Interferences
Ninhydrin will react with a number of exogenous
compounds that may interfere with identification and
interpretation. Penicillin-containing antibiotics produce a Amino Acid Performance Goals
number of Ninhydrin-positive metabolites. Other antibiotics The key determinant of analytical acceptability for amino
may also produce abnormal spots, as will acetaminophen, L- acid screening is the ability to detect inherited metabolic
dopa, & GABA analogues. disease if present. Since the majority of the methods
126
described here allow several samples to be run on a single equilibration before use. Make fresh solutions
plate or sheet, the simplest way to ensure reliability may be every 2 days.
to run one or more positive controls on each plate. However, 5. Commercial amino acid standard mixture.
since this limits the analytes being controlled to those Example: Sigma A2161 diluted 1:8 with ethanol to
elevated in the control sample chosen, it may be useful to give equivalent of 125 mol/L concentrations in
supplement the sample with additions of other key amino plasma. Stored at 80C.
acids at concentrations typical of those found in disease. In Equipment:
this way, one can be sure an elevation in a particular amino Cellulose TLC plates 10 20 cm (Merck # 5730)
acid can be reliably detected, and it also serves as a rough TLC tanks
comparator for grading the extent of the increase in a patient Micropipette for delivery of sample to TLC plate
sample.
Assay
References
1 Smith I, Ersser RS. Aminoacids and Related 1. Samples are deproteinized by mixing with ethanol.
Compounds. In: Smith I and Seakins J W T. In a microcentrifuge tube add 200 L of ethanol to
Chromatographic and Electrophoretic Techniques. 50 L of plasma. Centrifuge for 2 min at 10,000 g.
4 ed. London: William Heinemann Medical Books; 2. Transfer supernatant to a clean tube.
1976. 75-121. 3. Mark a cellulose thin-layer plate with lines 1.5 cm
2 Kohn J. Cellulose Acetate Electrophoresis. In: long and 1 cm apart, 1.5 cm from the bottom edge
Smith I. Chromatographic and Electrophoretic of the plate, using a soft pencil. NOTE: Use gloves
Techniques. 4 ed. London: William Heinemann and take care not to touch the plate; fingerprints
Medical Books Ltd; 1976. 90-137. contain enough amino acids to stain the plate.
3 Beale D, Smith I. High Voltage Paper 4. Apply 15 L of the plasma extract in 3 lots of 5
Electrophoresis. In: Smith I. Chromatographic and L, allowing the extract to completely dry between
Electrophoretic Techniques. 4 ed. London: William applications (warm air may be used to aid drying
Heinemann Medical Books Ltd; 1976. 31-65. between applications). The sample is applied using
4 Shih VE, Mandell R, Sheinhait I. General a 5 L positive-displacement micropipette as a
Metabolic Screening Tests. In: Hommes FA. streak 1-cm long between the lines drawn on the
Techniques in Diagnostic Human Biochemical plate.
Genetics.New York: Wiley-Liss; 1991. 45-68. 5. A maximum of 7 samples can be run on a single
5 American College of Medical Genetics. Standards plate. One track is reserved for the amino acid
and Guidelines for Clinical Genetics Laboratories. standard mixture applied as per plasma samples.
2006 Edition. 6. Place the TLC plate in tank 1, and develop until the
http://www.acmg.net/Pages/ACMG_Activities/stds- solvent reaches the top of the plate (approximately
2002/f.htm Accessed: 2007-08-09 20 to 30 min).
6 Shih VE. Detection of hereditary metabolic 7. Remove the plate, and allow to dry completely in a
disorders involving amino acids and organic acids. fume hood.
Clin Biochem 1991; 24: 301-309. 8. Place the plate in tank 2, and develop until solvent
reaches the top of plate.
Procedure 1: Plasma Amino Acids by Thin-Layer 9. Remove the plate, and allow to dry overnight.
Chromatography Color will develop at room temperature, but if an
Principle urgent result is required, plate may be heated at
Plasma samples are deproteinized by ethanol precipitation 100C to aid drying and development.
and applied to cellulose TLC plates. The plates are run twice 10. Comparison of individual tracks with standard
in an n-butanol/acetone/acetic acid/water solvent, the second mixture allows qualitative scoring of amino acids,
run incorporating Ninhydrin into the solvent to locate the but gross elevations should always be followed up
amino acids. by quantitative analysis in a specialist laboratory.
Reagents
1. Absolute Ethanol Procedure 2: Urine Amino Acids by High-Voltage
2. TLC Solvent 1 (tank 1). In a fume hood, measure Electrophoresis
out: 21 mL of acetone, 21 mL n-butanol, 6 mL of Principle
glacial acetic acid, and 12 mL of water. Pour into a High-voltage electrophoresis (HVE) is a very rapid and
TLC tank and cover with a glass lid. Allow 30 min efficient method for separating small molecules. At pH 1.9,
for equilibration before use. Make fresh solutions the basic amino acids lysine, arginine, and histidine become
every 2 days. positively charged and travel rapidly towards the cathode
3. Ninhydrin 0.2 % w/v. Weigh out 0.5 g of (), whereas the acidic amino acid cysteic acid is negatively
Ninhydrin. Dissolve in 250 mL acetone. Store in a charged and remains near the origin close to the anode (+).
brown stoppered bottle, stable for 1 month. A standardized volume of urine (100 nmol creatinine) is
4. Locating Solvent 2 (tank 2). In a fume hood, applied as a narrow band onto a paper sheet. Electrophoresis
measure out: 21 mL of the 0.2% Ninhydrin is performed for 24 min at 2500 V. Staining the
solution, 21 mL n-butanol, 6 mL of glacial acetic electrophoretogram with Ninhydrin localizes the amino
acid, and 12 mL of water. Pour into a TLC tank, acids. The profiles are interpreted, and then the sheet is
and cover with a glass lid. Allow 30 min for overstained with iodoplatinate reagent that enables
127
visualization of sulfur-containing compounds, including 3. In lead pencil, mark out electrophoresis paper 6 cm
homocystine, methionine, cystine, and S-sulfocysteine. from one end for 9 samples. These should be 1.0
cm long, 1.5 cm from both edges, and 1.0 cm apart.
NOTE: plasma may be run on the same method but must be 4. Apply sample as a narrow band in small volumes,
deproteinized before application to the paper. drying between applications using hot air.
5. When all samples have been applied, carefully wet
Reagents the sheet with buffer, avoiding wetting the origin
1. Formic acid/acetic acid buffer pH 1.9. Add 37.5 directly and allowing buffer to run up to the origin
mL formic acid and 150 mL glacial acetic acid from both sides.
slowly to approx. 1500 mL of Milli-Q water in 6. Gently blot the sheet between a folded sheet of
fume hood. Make up to 2 L with Milli-Q water. filter paper and place on the ceramic cooling plate,
Store in brown glass bottle at room temperature. origin at anode end.
Stable for 1 month. 7. Wet the IEF electrode strips with buffer, and blot
2. Ninhydrin 0.2 % w/v. Weigh out 0.5 g of gently before applying to the sheet 1cm from each
Ninhydrin. Dissolve in 250 mL acetone. Store in a end.
brown stoppered bottle, stable for 1 month. 8. Place the glass plate on the sheet to maintain good
3. Chloroplatinic acid, 1 g/500 mL. Weigh out 1g of contact with cooling plate.
chloroplatinic acid (CAUTION: highly toxic, wear 9. Position IEF electrodes over electrode strips and
mask and gloves). Make up to 500 mL with water. connect electrodes. Fit safety lid.
4. Hydrochloric Acid (HCl) 6M. 10. Turn on power supply, and run at constant voltage
5. Potassium iodide, 167 g/L. Weigh 41.75 g 2500 V (approximately 50-60 mA) for 25 min.
potassium iodide, make up to 250 mL with water; 11. Turn off power supply. Remove sheet and hang to
replace approx. every 12 months. Store in brown dry in oven at 100C for 5 min.
glass bottle.
6. Iodoplatinate (IP) stain. Make up the stain in the Ninhydrin and iodoplatinate staining
fume hood immediately prior to staining. Mix 5 mL
of chloroplatinic acid, 0.5 mL potassium iodide, 1 1. Stain sheet by rapidly dipping through Ninhydrin
mL 6M HCl, and 42.5 mL acetone. reagent in a staining tray. Dry briefly in fume hood
before transferring to oven at 120C for 3 min.
Equipment: 2. Interpretation of Ninhydrin staining requires
REQUIRED: High-voltage electrophoresis equipment experience and knowledge of changing excretion
capable of running at 2500 V, with integrated patterns with age. Follow-up of abnormal patterns
cooling system. A typical system would be: requires quantitation in a specialist centre.
Pharmacia HVE system: 3. If required, following interpretation of the
o Hoefer PS3000 DC power supply Ninhydrin stain, the sheet can be overstained with
o Multiphor II electrophoresis unit iodoplatinate reagent to highlight the presence of
comprising: sulfur-containing amino acids.
o Buffer tank 4. Ensure that the staining dish and paper are
o Ceramic cooling plate thoroughly dry, then add a volume of the IP stain.
o Electrode holder and electrodes 5. Quickly and evenly dip the paper through the stain.
o Safety lid 6. Hold the papers in the fume hood briefly to remove
o Multi Temp III cooling unit. any excess acetone. Paper clips are used to hold the
Electrophoresis paper, Machery Nagel MN 214, papers in a cylindrical shape. These cylinders are
200 265 mm stood on their end in a glass tank in the fume hood
IEF Electrode strips (Pharmacia Cat. # 18-1004- overnight at room temperature. The Ninhydrin
40) positive bands are rapidly decolorized by the acid
Wash bottle with a fine jet present, except for pipecolic and hydroxypipecolic
Large glass dish acid bands that remain fluorescent purple.
Glass plate 195 mm square 7. The iodoplatinate stain will give a pale pink/peach
background with sulfur-containing amino acids,
Blotting paper, 445 570 mm, 135 gsm, APPM
homocystine, methionine, cystathionine, cystine,
Oven set at 100C 2C, for drying and 125C
and S-sulfocysteine, having a cream or white color.
2C for Ninhydrin stain development
8. Several drugs and other compounds will give
Micropipette for delivery of sample cream bands, and interpretation again requires
considerable experience.
Assay
Electrophoresis
1. Calculate volume of urine in L to be applied by

dividing100 by creatinine concentration in mmol/L.
2. Turn on recirculating cooler and set to 4C
128
Ammonia
Ammonia
Elizabeth M. Hall
Name: Ammonia
Clinical Significance: Refer to Chapter 31, Liver Function, in the 5th edition of Clinical Chemistry:
Molecular formula: NH3
Merck Index: 13,492
Chemical class: Amine
Reference method: None
which diffuses across a semipermeable membrane and
i causes a pH change detected by a standard glass
electrode with silver-silver chloride reference electrode
Methods for analysis of blood ammonia can be (Table 1, Method 2) [5]. When adapted to a continuous-
considered under four headings: microdiffusion, ion- flow analyzer, this method is reliable [6,7] but not
selective electrode, ion-exchange, and enzymatic widely available.
methods. The pKa of the ammonium ion is 9.5, therefore
at physiological pH, nearly all ammonia is present as In ion-exchange methods (Table 1, Method 3), ammonia
ammonium ions (NH4+). Both microdiffusion methods is isolated by adsorption onto a strong cation exchange
and ion-selective electrodes rely on alkalinization of the resin. After washing, the ammonia is eluted and
sample to release ammonia gas. measured using the Berthelot reaction. In the Berthelot
reaction, ammonia and phenol form a blue indophenol
Microdiffusion methods were the earliest methods used complex (max 630 nm) in the presence of hypochlorite.
for blood ammonia. Workers such as Conway allowed Sodium nitroferricyanide (Na2(Fe[CN]5NO),
the released ammonia to diffuse into a chamber nitroprusside) acts as a catalyst and increases color
containing hydrochloric acid. The ammonia intensity and assay reproducibility. Microscale methods
concentration was determined by titration with barium were developed to use smaller sample volumes but
hydroxide and a pH indicator [1]. Seligson used remained cumbersome and time consuming [8,9].
nesslerization to measure the ammoniamercuric
chloride and ammonia form a yellow chelate that can be The most common method in use today for measurement
measured at 410 nm [2]. These methods were of plasma ammonia uses glutamate dehydrogenase
cumbersome and required technical skill; in addition, (GLDH, L-glutamate:NAD(P) oxidoreductase
deamination of amino acids occurs during prolonged (deaminating), EC1.4.1.3) with either NADH or
incubation of plasma with alkali, leading to spuriously NADPH (Table 1, Method 4) [6].
high results. However, microdiffusion is the basis of two
dry chemistry methods for ammonia (Table 1, Method NH4+ + 2-oxoglutarate + NADH + H+ _glutamate
1). The sample is added to borate buffer pH 9.2 to
release ammonia gas, which diffuses across a
dehydrogenase glutamate + NAD+ + H2O
semipermeable membrane into a layer containing a pH
indicator [3]. The color change in the pH indicator is The ammonia concentration is proportional to the
measured by reflectance. The dry slide method on Vitros decrease in absorbance at 340 nm. Preincubation of the
analyzers uses bromophenol blue, monitored at 600 nm, sample and substrates before addition of the enzyme
to measure ammonia in plasma samples. The Ammonia avoids interference from other NADH-consuming
Checker is a handheld meter that uses whole blood with reactions [10]. Alternatively, the reaction may be started
bromocresol green monitored at 635 nm, with results by addition of coenzyme [11].
being available in less than 4 minutes [4].
Data from the 2007 College of American Pathologists
For ammonia analysis by ion-selective electrode, plasma (CAP) proficiency testing program indicate that most
is mixed with buffer pH 10.4 to liberate ammonia gas, laboratories in the United States use a GLDH method for
plasma ammonia measurement, with 15% using a Vitros
dry slide method. In contrast, in the United Kingdom,
i
Ammonia 18% of participants in the 2007 Wales External Quality
Previous and current authors of this method: Assessment Scheme (n=202) use the handheld
First edition: Nancy Gau reflectance meter, 10% use a Vitros plasma dry slide
Methods edition: Nancy Gau method, and the majority use a GLDH method (A.
Second edition: Not updated Thomas, personal communication).
Fourth edition: Nancy Gau
Fifth edition: Elizabeth M. Hall
129
Ammonia
Reference and Preferred Methods For longer storage, freezing at 70C is recommended
There is no reference method for the measurement of [16]. Serum is unsuitable because complete clotting
ammonia in biological fluids. takes longer than 15 minutes and because ammonia is
produced during the formation of crosslinks between
Blood ammonia measurement is usually required fibrin molecules [12,18].
urgently for immediate diagnosis and patient
management. Both qualitative screening methods and Interferences
quantitative methods are available. Method choice will Contamination of the sample by ammonia from
depend on workload and patient mix, the speed of exogenous sources or due to endogenous production
analysis required, and the availability of instrumentation during specimen transport and handling are the most
and technical expertise. common causes of inaccurate results. Results may be
affected by contamination of the atmosphere, water, and
The preferred method for measurement of plasma laboratory equipment with ammonia. If capillary
ammonia on todays laboratory automated analyzers is samples are used, care must be taken to decontaminate
either GLDH or the Vitros dry slide method. the skin to avoid contamination with sweat. Smokers
usually have higher plasma ammonia concentrations, and
The handheld reflectance meter has a narrow working it has been recommended that samples be collected by a
range (up to 285 mol/L) but is suitable for screening for non-smoker. Within the laboratory, analysis should not
hyperammonemia by laboratories with small workloads occur in proximity to a system using ammonium buffers,
or at point of care. All abnormal results must be and care should be taken to use deionized water and
confirmed with a quantitative method. Larger exclude detergents. Care must also be taken to avoid loss
laboratories and those serving acute pediatric units must of ammonia from calibrators as ammonia gas.
provide quantitative analysis for the diagnosis and Hemoglobin does not interfere with ammonia
monitoring of hyperammonemia. determinations, but the ammonia concentration within
erythrocytes is about three times the concentration in
Specimen plasma, so hemolyzed samples are unsuitable [16].
Venous samples collected without stasis and avoiding Lipemic samples and those with high bilirubin
hemolysis are preferred for blood ammonia concentrations may be unsuitable for analysis in some
measurement. Capillary specimens produce significantly systems [11].
higher results [6,12] but may be used for rapid screening,
provided suitable reference intervals are available and Endogenous production of ammonia within separated
abnormal results are confirmed by laboratory analysis of plasma occurs due to deamination of amino acids. The
a venous sample. Plasma ammonia concentrations are glutaminase activity of -glutamyl transferase is a major
not significantly affected by diet or fasting [6], although contributor to this effect, so ammonia concentration may
fasting is recommended by some [11]. be artifactually increased in samples with raised -
glutamyl transferase activity [14]. Colonization of the
Plasma from samples collected into either EDTA or laboratory analyzer with urease-positive bacteria has
lithium heparin is suitable for most methods of ammonia also been reported to give spuriously high ammonia
determination. However, heparin can increase results [19].
background absorbance with some buffers [11,13].
Ammonium heparin is not a suitable anticoagulant. An unidentified metabolite of the antibiotic cefotaxime
Ideally, each batch of sample tubes used for ammonia may interfere in the enzymatic assay for ammonia,
sample collection should be checked for contamination giving an ammonia result of less than zero [10].
before use by adding a low standard to five tubes,
mixing well, and measuring the apparent ammonia Reference Intervals and Units
concentration. Tubes are acceptable if apparent ammonia Plasma ammonia concentrations are expressed in
concentration increases by less than 10%. Tubes should mol/L.
be stored in an airtight box [6]. Alternatively, each Conversion factors: ammonia (g/dL) 0.714 =
sample should be accompanied by an empty tube from ammonia (mol/L)
the same batch, which is checked for ammonia
contamination by the analysis of low standard added to Ammonia Reference Intervals [20]
the tube.
Plasma mol/L g/dL
Plasma ammonia concentration increases during storage Premature neonate < 200 < 143
of whole blood and separated plasma, owing to the Term neonate < 100 < 71
metabolic activity of erythrocytes and platelets and Infant/child < 40 < 29
deamination of plasma proteins and amino acids by Adult < 40 < 29
enzymes [14-16]. Samples should be transported to the
laboratory on ice and whole blood analysis or separation
of plasma performed within 15 minutes of venesection The higher ammonia concentrations in neonates are due
[17]. Centrifugation must be with sufficient force to to immaturity.
remove platelets (>2000g) [12]. If analysis is delayed,
separated plasma may be stored at 4C for up to 1 hour.
130
Ammonia
Interpretation References
Circulating ammonia is produced principally from the 1 Conway E. Microdiffusion analysis and
catabolism of amino acids derived from endogenous or volumetric error. 3rd ed. London: Crosby
dietary proteins. The gastrointestinal tract is an Lockwood and Son Ltd; 1950.
important source. Usually, plasma concentrations are 2 Seligson D, Seligson H. A microdiffusion
maintained at low levels by urea synthesis in the liver method for the determination of nitrogen
and by formation of glutamine. Hyperammonemia liberated as ammonia. J Lab Clin Med 1951;
occurs when there is disruption of the urea cycle in the 38: 324-330.
liver. This may be due to (1) inherited defects of the urea 3 Sundberg MW, Becker RW, Esders TW,
cycle enzymes secondary to disruption of liver Figueras J, Goodhue CT. An enzymic
metabolism or (2) bacterial production of ammonia in creatinine assay and a direct ammonia assay in
the gastrointestinal or urinary tract [6,21]. The urea cycle coated thin films. Clin Chem 1983; 29: 645-
defects are each rare disorders but have an estimated 649.
combined incidence of about 1:30,000, with the 4 Diaz J, Tornel PL, Martinez P. Reference
commonest being ornithine transcarbamylase deficiency intervals for blood ammonia in healthy subjects,
(OTC). Some disorders of amino acid, organic acid, and determined by microdiffusion. Clin Chem
fatty acid metabolism may also present with 1995; 41: 1048.
hyperammonemia. In neonates, clinically significant 5 Cooke RJ, Jensen RL. Micromethod for
hyperammonemia may occur secondary to asphyxia, determining plasma ammonia nitrogen with use
infection, and sepsis. Hyperammonemia has been of an ion-selective electrode. Clin Chem 1983;
associated with parenteral nutrition, sodium valproate 29: 867-869.
therapy, acute-onset hepatic failure, and advanced 6 Green A. When and how should we measure
chronic liver disease [6,21]. In Reyes syndrome, acute plasma ammonia? Ann Clin Biochem 1988; 25:
encephalopathy and fatty degeneration of the liver may 199-209.
be associated with a viral illness or salicylate therapy 7 Willems D, Steenssens W. Ammonia
that lead to acute mitochondrial injury [6]. determined in plasma with a selective electrode.
Clin Chem 1988; 34: 2372.
Ammonia is neurotoxic. Hyperammonemia has a range 8 Wu J, Ash KO, Mao E. Modified micro-scale
of presentations from an acute catastrophic illness in enzymatic method for plasma ammonia in
babies to episodes of lethargy and vomiting. newborn and pediatric patients; comparison
Measurement of ammonia should be considered in any with a modified cation-exchange procedure.
neonate with unexplained neurological deterioration, any Clin Chem 1978; 24: 2172-2175.
older patient with unexplained encephalopathy, and 9 Oberholzer VG, Schwarz KB, Smith CH,
children or adults with a history of episodes of vomiting Dietzler DN, Hanna TL. Microscale
and lethargy or of protein avoidance, which may indicate modification of a cation-exchange column
a mild urea-cycle defect. Increased ammonia procedure for plasma ammonia. Clin Chem
concentration should always be confirmed on a second 1976; 22: 1976-1981.
sample to exclude artifactual increases caused by poor 10 Kirk JM. Probable cefotaxime interference in
sample handling. Infection should always be excluded as enzymatic ammonia assay - a cautionary note.
a cause of mild increases in ammonia in neonates. Ann Clin Biochem 1989; 26: 195-196.
Inherited disorders should be excluded in patients with 11 Roche Diagnostics. Ammonia. test information
unexplained hyperammonemia, Reyes-like syndrome, for Cobas Integra 400/700/800. 2000.
unexplained hepatic encephalopathy, or cyclical illness 12 Cowley DM, Nagle BA, Chalmers AH, Sinton
[21]. Increased ammonia concentrations are a TJ. Effects of platelets on collection of
nonspecific finding in liver disease, and measurements specimens for assay of ammonia in plasma.
do not aid with diagnosis, prognosis, or monitoring of Clin Chem 1985; 31: 332-333.
treatment [6]. 13 da Fonseca-Wollheim F, van Dam M.
Interference by heparin in enzymatic
Significant encephalopathy usually develops at determination of plasma ammonia depends on
concentrations above 300 mol/L. Concentrations reagent composition. Clin Chem 1992; 38:
greater than 500 mol/L usually present with coma and 1921-1922.
convulsions and are associated with neonatal-onset 14 da Fonseca-Wollheim F. Deamidation of
inherited metabolic diseases. glutamine by increased plasma gamma-
glutamyltransferase is a source of rapid
Ammonia Performance Goals ammonia formation in blood and plasma
There are no studies of biological variation of ammonia specimens. Clin Chem 1990; 36: 1479-1482.
concentrations on which to base goals for analytical 15 da Fonseca-Wollheim F. Preanalytical increase
performance [22]. However, with a GLDH method, of ammonia in blood specimens from healthy
between-day precision of 3.6% can be achieved at a subjects. Clin Chem 1990; 36: 1483-1487.
concentration of 60 mol/L [11]. 16 Howanitz JH, Howanitz PJ, Skrodzki CA,
Iwanski JA. Influences of specimen processing
and storage conditions on results for plasma
ammonia. Clin Chem 1984; 30: 906-908.
131
Ammonia
17 Losty H. Appendix - Notes on the measurement

of ammonia in blood/plasma. [cited 2007, 20 Green A, Morgan I, Gray J. Neonatology and
September 25]; Available from: Clinical Biochemistry. London: ACB Venture
www.metbio.net Publications; 2003.
18 Mousli S, Wakid NW. Ammonia production 21 Losty H. Guidelines for the investigation of
during clot retraction and its use in assay of hyperammonaemia for inherited metabolic
fibrinoligase. Clin Chem 1977; 23: 1739-1743. disorders. [cited 2007, September 25];
19 Hayes LW, Swanson JR. Ammonia results Available from: www.metbio.net
increased by bacteria in aca needle tubing. Clin 22 Rics C, Garca-Lario J, Alvarez V, Cava F,
Chem 1984; 30: 1882-1883. Domenech M, Hernndez A et al. Biological
variation database and quality specifications for
imprecision, bias and total error. The 2006
update. [cited 2007, June 20]; Available from:
http://www.westgard.com/guest32.htm
Methods of Ammonia Analysis

Method 1: Microdiffusion
Principle of analysis:
NH3 gas released by increasing pH diffuses across semipermeable membrane; detection by color change of pH
indicator
Comments: Automated plasma assay or strip-based whole blood assay
Method 2: Ion-selective electrode; quantitative, EP
Principle of analysis: Diffusion of NH3 at electrode surface; pH change measured potentiometrically
Comments: Automated; good precision and accuracy
Method 3: Ion-exchange; quantitative, end point
Principle of analysis: Adsorption of ammonia followed by colorimetric analysis by Berthelot reaction:
pH 10.5
NH4+ + HOCl H2NCl + H3O+; phenol + H2NCl complex;
OH
complex + phenol indophenol blue (560 nm)
Comments: Manual; good sensitivity and accuracy
Method 4: Enzymatic; quantitative, end point
GLDH
NH4+ + -ketoglutaric acid + NADH NAD+ + glutamic acid
Comments: Automated, good accuracy, rapid
132
Amniotic Fluid Phospholipids (AFPL): L/S Ratio and PG

Hassan M.E. Azzazy i
Name: Amniotic fluid phospholipidsL/S ratio and PG

Clinical significance: Refer to Chapter 44, Pregnancy, in the 5th edition of Clinical Chemistry:
Molecular formula: Depends on fatty acid moieties

Phosphatidylcholine (lecithin) (L) C40H82O9NP (dipalmitoyl-)
Sphingomyelins (S) C39H81O7N2P (palmitoyl-)
Phosphatidylinositol (PI) C40H80O14P (dipalmitoyl-)
Phosphatidylglycerol (PG) C42H83O10P (distearoyl-)
Molecular mass: Dependent on fatty acid moieties
Phosphatidylcholine (lecithin) 752 (dipalmitoyl-)
Sphingomyelin 721 (palmitoyl-)
Phosphatidylinositol about 883
Phosphatidylglycerol 779
Merck Index: 5271 (lecithin), 8590 (sphingomyelins)
Chemical class: Phospholipids
Structures: 1 Phosphoglyceride structure
2 Lecithin (Phosphatidyl choline)
3 Phosphatidyl inositol
4 Phosphatidyl glycerol
5 Sphingomyelin
1 2
3 4
i
Amniotic Fluid Phospholipids
First edition: Not done
Methods edition: Paul T. Russell
Fourth edition: Steven C. Kazmierczak
Fifth edition: Hassan M.E. Azzazy
133
List of Abbreviations agents, such as cupric acetate and sulfuric acid with or
ANS: 8-anilino-1-naphthalenesulfonic acid without dichromate, react with the fatty-acid double
DCF: 2,7-dichlorofluorescein bonds, yielding colored products. The intensity of the
FLM: Fetal lung maturity colored spots depends on the degree of unsaturation of
FSI: Foam stability index the phospholipid constituent fatty acids. Iodine vapor
GLC: Gas-liquid chromatography and molybdate ions also react with the unsaturated bonds
HPLC: High-performance liquid chromatography in the fatty acid moieties of the phospholipids.
LBC: Lamellar body count
LS ratio: lecithin sphingomyelin ratio After the phospholipids are separated by thin-layer
PG: Phosphatidyl glycerol chromatography and identified on the chromatogram by
RDS: Respiratory distress syndrome an indicator, the L/S ratio can be evaluated in a number
TLC: Thin-layer chromatography of ways.
1. The chromatogram can be evaluated by
Principles of Analysis and Current Usage transmission or reflection densitometry.
Tests for the estimation of pulmonary surfactant in Compounds that are separated and charred on
amniotic fluid fall into three general categories: (1) those thin-layer plates are quantitated
that measure the chemical constituents of surfactant, (2) densitometrically by use of a scanner. The L/S
those that measure the physical properties of lung ratio is calculated from the recorder tracings,
surfactant, and (3) those that measure a variety of which are related proportionately to the plate
chemical agents that correlate with stages of fetal concentrations of lecithin and sphingomyelin.
maturation. 2. The phospholipids can be eluted from the
chromatogram and the phospholipid
The analysis of the chemical constituents of surfactant quantitatively estimated by a phosphorus
entails the separation of the individual components of determination. Phosphorus determinations may
the phospholipids present in amniotic fluid. Extraction be performed by several methods [11-15].
and separation, followed by quantitation, permit the 3. One can estimate the phospholipids on the
establishment of relationships such as the lecithin-to- chromatogram by planimetry. The area of each
sphingomyelin (L/S) ratio [1], the phospholipid visualized spot is estimated by either taking the
profile, [2] the fatty acid composition of amniotic fluid product of the length and width of the
lecithin [3], and many others that provide an index of individual spots or using a planimeter. The
fetal lung maturity [4-6]. Thin-layer chromatography estimated area is proportional to the quantity of
(TLC) became the most widely used technique for most material present.
of these analyses after it was published as the method of Gas-liquid chromatography (GLC) (Table 1,
choice by Gluck et al. [1,7,8] in their pioneering work on Method 2) has provided a means to estimate
the L/S ratio. pulmonary surfactant based on its unique fatty
acid composition. GLC involves quantification
The thin-layer chromatography system (Table 1, Method of the palmitic acid concentration of amniotic
1) is extensively used [6] and employs silica-gel fluid, establishment of a ratio of palmitic acid to
stationary phases and mobile phases consisting of either stearic acid, and determination of the relative
chloroform-methanol-water or chloroform-methanol content of palmitic acid contained in lecithin
ammonium hydroxide. TLC separations are based on the isolated from amniotic fluid. The gas
differential adsorption of phospholipids with different chromatography for these fatty acid separations
polarities. Both one-dimensional and two-dimensional has been performed on polyester columns of
TLC systems have been used. The choice is determined diethylene glycol succinate and ethylene glycol
primarily by balancing the need to completely separate succinate, with flame ionization detectors.
the phospholipids of interestlecithin, sphingomyelin, Phosphatidylcholine has been determined as the
and phosphatidylglycerolfrom one another and the diacylglycerol trimethylsilyl ether derivative by
need to separate these compounds from the other gas chromatography using a glass capillary
phospholipid components of amniotic fluid. column [16]. Preliminary TLC is required for
this method and for some others before GLC
A number of techniques can be used for the visualization analysis (Table 1, Method 2) [3].
of the separated phospholipids. One can char organic
materials by spraying the TLC plates with sulfuric or High-performance liquid chromatography (HPLC)
phosphoric acid and then heating to 280C, although (Table 1, Method 3) has been used to separate
variations in charring temperatures can give variable phospholipid extracted from the amniotic fluid. One
results [9]. Alternatively, ammonium sulfate can be method employs diol bonded-phase columns with
incorporated into the silica gel, eliminating the need for gradients (2% to 15% water) of acetonitrile/H2O as the
strong acid sprays during the charring procedure [7].
Lipids are visualized under ultraviolet radiation after mobile phase. Detection and quantitation are often based
spraying with rhodamine B or dichlorofluorescein. upon ultraviolet absorption characteristics of double
Bismuth subnitrate reacts with the choline moiety bonds (203 nm) in the fatty acid moieties of the
contained in lecithin and sphingomyelin [10]. Other phospholipids, though methods that measure saturated
phospholipid palmitate have been published [17].
134
formation of bubbles by other substances in amniotic

In methods proposed to measure specifically the fluid, such as proteins, bile salts, or salts of free fatty
disaturated phosphatidylcholine, total lipids are reacted acids, is eliminated. Pulmonary surfactant forms stable
with osmium tetroxide dissolved in carbon tetrachloride surface films that support bubble stability. Therefore, the
[18]. The disaturated phosphatidylcholine is then test is positive if bubbles persist.
isolated by column chromatography on neutral alumina.
A disadvantage of this method is that unsaturated lipids The foam stability index (FSI) uses a mixture of
are damaged and can not be analyzed further. different volumes of ethanol with 0.5 mL of amniotic
fluid and provides semiquantitative evidence of
All the procedures that measure phospholipid surfactant content. The index is defined as the highest
constituents of amniotic fluid require partial purification ethanol volume fraction of an amniotic fluidethanol
of the compounds before chromatography. Purification mixture that will permit a stable ring of bubbles at the
schemes usually involve one or more of the following meniscus after vigorous shaking. An FSI value of 0.48 is
steps: centrifugation, extraction, acetone precipitation, comparable to an L/S ratio of 2.0 in correlating with fetal
evaporation, and reconstitution. However, there is pulmonary maturity [27].
considerable controversy about the efficacy of the first
three steps listed. Thus a large number of extraction One can also measure surface tension directly [28-30]
schemes using permutations and combinations of these using a surface balance or tensiometer (Table 1, Method
techniques are in use. 5). These measurements give good correlations with
other tests for lung surfactant [31].
Various techniques have been developed for
measurement of lecithin [19-23], sphingomyelin [19,23], The fluorescence polarization assay measures the
and phosphatidylglycerol [24] using either microviscosity of amniotic fluid lipids, which is related
spectrophotometry [20-23] or radiochemical techniques to surface tension (Table 1, Method 6). A fluorescent
[19,24]. Radiochemical techniques, in general, have the probe, when mixed into amniotic fluid, dissolves in the
inherent disadvantages of high cost and waste disposal. hydrocarbon region. Its rotation in this hydrophobic
environment depends on the microviscosity of the fluid.
Enzymatic methods for lecithin, sphingomyelin [23], and The greater the viscosity (that is, the more surfactant
phosphatidylglycerol [25] that have many attractive present), the more effectively opposed is the rotation of
features have been published. Phospholipids are the probe. This results in an increased polarization of the
extracted from amniotic fluid with chloroform-methanol incident fluorescent light.
and, after solvent evaporation, are redissolved into a
zwitterionic detergent. Choline and glycerol are liberated Non-chromatographic methods for assessing surfactant
after treatment with phospholipase and are subsequently activity of amniotic fluid have only been recently
oxidized enzymatically to generate hydrogen peroxide. developed and are rarely used. TLC remains the almost
The hydrogen peroxide reacts with 4-aminoantipyrene to universal technique for amniotic fluid analysis, though
produce an intense red chromogen that can be measured there may be an increase in the use of alternative
at 510 nm. This approach offers the potential for rapid, procedures as their clinical utility is demonstrated. The
specific, and sensitive assays for key phospholipids of foam stability (shake) test is used as a STAT procedure
pulmonary surfactant. when TLC analysis is unavailable (Table 2).
Methods that measure the physical properties of lung One relatively new technique for assessing lung maturity
surfactant have also attracted interest because they is by determination of the number of lamellar body
measure functional lung surfactant. Besides multiple particles in the amniotic fluid. Lamellar bodies are
phospholipid components, surfactant may also include a packages of phospholipids secreted by type II
protein component for functional integrity. These tests pneumocytes and extruded into the lung fluid. The
measure surface tension, stable foam (bubbles), majority of these particles have a volume between 2 and
microviscosity, and other properties [5,6]. Many of these 20 fL. Since platelets roughly have the same size
tests are simpler to perform and quicker to carry out than (diameter 1 to 5 m), lamellar bodies can be counted in
the chemical tests and thus have strong potential for automated blood-counting instrumentation using the
routine laboratory use. To their disadvantage, some of platelet channel. Counting lamellar bodies by using this
these methods require expensive equipment and include type of instrumentation is a rapid, simple, and cost-
techniques that are less familiar to most laboratory effective way of rapidly assessing fetal lung maturity
workers than the chromatographic methods. (FLM) [32-34]. The performance of the lamellar body
count (LBC) and the L/S ratio for prediction of neonatal
One of the physical methods in most common use is the RDS was compared in a meta-analysis [35]. The LBC
foam stability or shake test (Table 1, Method 4). This was found to perform slightly better than L/S ratio, and
test was devised by Clements et al. [26] and calls for the authors concluded that LBC should become the test
serial dilutions of amniotic fluid mixed with equal of first choice in the assessment of FLM. In 2001,
volumes of 95% ethanol to exclude spurious surface- Neerhof et al. [36] published a standardized
active compounds. The tubes are inspected for bubbles methodology for LBC. and maturity was suggested by a
around the meniscus after being shaken. At the count of 50,000/L or more. A count below 15,000/L
recommended final percentage of ethanol (47.5%) the indicates immaturity, and no further testing is required.
135
The LBC is classified as transitional if the count falls Centrifugation speeds of 1000 g for 5 minutes result in a
between 15,000 and 50,000/L. Under these variable loss of both total phospholipid and lamellar-
circumstances, further testing such as TDx FLMII, PG body phospholipid [42]. Some studies have shown that
test, and/or L/S ratio is required [37]. Szallasi et al. [38] the losses are often proportional for lecithin and
compared LBC in amniotic fluid using four different sphingomyelin, so the ratio remains constant (Figure 1).
hematology analyzers. It was concluded that because However, other studies have reported that certain
different hematology analyzers count lamellar bodies centrifugation speeds can affect lecithin more than
differently, it would be necessary to establish analyzer- sphingomyelin and can alter the L/S ratio [39,43-45].
specific LBC clinical decision limits that are confirmed The effect of centrifugal force on the L/S ratio has been
by outcome-based studies. studied by Cherayil et al. [43] and is presented in Figure
1.
Reference and Preferred Methods Centrifugation also results in the loss of
There is no reference method for AFPL. The method of phosphatidylglycerol (PG), so that when small quantities
Gluck et al., either in its original form or in some are present in a sample, losses can influence
modified form, has been the most widely used in-house interpretation of results. Losses probably occur for all
method. Clinical interpretation and predictions have for phospholipids by adsorption to the cells or cellular
the most part used Glucks criteria [7]. A large number debris [46].
of variations of the original procedure are in use today
because virtually every step has more than one Debris can also be removed by filtration, but losses of up
legitimate means for its execution. Other variations have to 90% of the total phospholipid content after filtration
developed because of differences of opinion over the through Whatman No. 1 filter paper have been reported.
need for specific steps in the procedure, such as the Filtration also can lower the L/S through
acetone precipitation step. Accurate clinical correlations disproportionate losses of lecithin [39].
are the only meaningful criteria for methodological
considerations, and so it is incumbent on each laboratory Extraction
to establish its own criteria that correlate with clinical It is important to standardize the proportion of extracting
experience. However, those steps in the generalized solvent to amniotic-fluid sample used in the assay, since
procedure that can drastically compromise the method the relative proportions of chloroform, methanol, and
deserve specific mention. water significantly affect the extraction of lipid [47,48].
One volume of amniotic fluid to one volume of
A commercial one-dimensional TLC kit, Helena Fetal chloroform-methanol (1:1, v/v) is commonly used,
Tek 200, has become available for separating lecithin though chloroform-methanol (2:1, v/v) also gives good
and sphingomyelin; charring is achieved using recovery of phospholipids. After centrifugation to
phosphoric acid and cupric acetate. In addition, lung facilitate phase separation, the lower organic phase is
maturity of unborn infants can be assessed using TDx removed with a Pasteur pipet and evaporated to moist
FLM II fluorescent polarization immunoassay that dryness under a stream of nitrogen. If the sample is taken
measures the ratio of surfactants to albumin in filtered, to complete dryness, the residue will not redissolve
uncentrifuged amniotic fluid. easily in the chloroform-methanol mixture, and losses
can occur. This extraction is the starting point for the
Technical Considerations various assays.
Centrifugation
To remove whole cells and other debris, one usually Two extractions of the amniotic fluid with equal volume
centrifuges amniotic fluid before further processing. This of fluid, methanol, and chloroform are recommended.
procedure has significant influence on the analysis of Although the L/S ratio remains unchanged through a
amniotic fluid phospholipids and the L/S ratio [39]. Yet series of extractions [49], a better recovery of lipid is
this preliminary step has not been standardized. obtained with more than one extraction. Since
Centrifugation speeds of 300 to 500 g for 10 minutes phosphatidyl glycerol (PG) is but a minor constituent of
pellet much of the debris but maintain phospholipid surfactant, complete extraction is an important factor in
losses within reasonable bounds. the detection and quantitation of PG.
Acetone Precipitation
Amniotic fluid contains phospholipids from many
Most naturally occurring lecithins are insoluble in cold
sources. Some of the phospholipid is related to
acetone, and surface-active phospholipids can be
surfactant, but much of it is not. When an amniotic fluid
precipitated by acetone from lung fluid [50]. Glucks
specimen is centrifuged, even at relatively low speed
method used acetone precipitation to effect a crude
(300 g for 10 min), 5% to 10% of the phospholipids are
isolation of surfactant phospholipids [8,50]. This step
found in the pellet [40].
has provoked considerable controversy. A number of
authors claim the step is unnecessary and time
Surfactant phospholipid is present in amniotic fluid in
consuming, but Gluck is adamant that since surface
membranous aggregates called lamellar bodies, and
activity resides solely in the acetone-precipitable
although high-speed centrifugation (33,000 g for 60
material (Figure 2), this step is an essential feature in
minutes or longer) is required to sediment these
separating surface-active phospholipids from a total lipid
structures, some sedimentation occurs during
extract.
centrifugation at only 80 g for 5 minutes [41].
136
From 60% to 90% of the total surface-active lecithins because of the low percentage of PG among total
present in the sample are precipitated with cold acetone. amniotic fluid phospholipids.
This means that 10% to 40% of the desired lecithins may
be lost. If reliable clinical correlations are to be obtained, Methods that use ammonium hydroxide as a constituent
it must be assumed that sphingomyelin is lost in the mobile phase can have run-to-run variations
proportionately, but this may not always be the case traceable to the loss of volatile ammonia from the stock
[51]. Some authors have reported preferential losses of solution [59]. The ammonium hydroxide concentrations
lecithin [39,52] which falsely reduce the L/S ratio. Thus should be within 15% of the optimum concentration, and
an occasional immature L/S is diagnosed when in fact the working solution should be discarded at 1 week.
adequate surfactant is present. Other authors have not
reported losses as serious with the precipitation step Visualization of Spots
[5,53]. Despite the ongoing controversy, this step In the context of the phospholipid methodology for
remains an integral part of many methods in use today, measurement of surfactant, there are some important
but to ensure reproducible results, one must pay strict considerations. The paper by Spillman et al. [54]
attention to the ice-cold temperature conditions specified compared seven of the more commonly used techniques
[8]. Warmer conditions will increase the solubility of for detection of phospholipids after thin-layer
surfactant (saturated) lecithin. chromatography.
Chromatography It is clear that the L/S ratio can be distorted by the

The choice of a chromatographic system for the visualization process. When charring carried out under
separation of phospholipid components of surfactant is a acid conditions at high temperatures is used as the
choice based on many considerations. Ideally, the method of detection, error can be introduced by the
chromatographic system should separate lecithin, difficulty of assessing the completeness of the charring
sphingomyelin, and PG from the other lipids present in process. Similarly, when one accomplishes visualization
amniotic fluid. It is also desirable to run standards in by staining with a lipophilic dye, it is difficult to assess
parallel with the phospholipid extract if a TLC system is the completeness of the color development. Furthermore,
employed. These optimal criteria have been difficult to the mechanism whereby the visualization reagent
attain, and thus many of the TLC systems used for interacts with the lipid can often bias the results by
surfactant assays have not satisfied one or the other of underestimating the amounts of saturated lecithin present
these criteria. The L/S ratio determination, as introduced [54]. Reagents that react with double bonds do not give
by Gluck and co-workers, employed a one-dimensional, information relating to the saturated lipid of interest.
silica-gel, thin-layer system. Subsequently, Gluck et al. Hence as the more saturated surfactant is produced by
[2] introduced a two-dimensional system that offered the the lung and extracted from amniotic fluid, detection by
opportunity to detect other components of the surfactant double-bond reagents deviates more and more from the
matrix (notably phosphatidylinositol and true saturated lipid content. Also, the intensity of the
phosphatidylglycerol). The change was necessary developed spot when treated with reagents that attack
because the one-dimensional system used initially could double bonds (such as cupric acetate and sulfuric acid)
not provide sufficient resolution for the phospholipids of can vary widely (10-fold). One can include ammonium
interest. sulfate in this group. This salt is used as part of the
stationary phase in certain TLC solvent systems to
It was recognized early that L/S ratios of 2:1 as eliminate the need to spray with strong acid [7].
measured by TLC methods correspond to actual L/S
concentration ratios ranging from 4:1 to 6:1 [5,7]. This Molybdate, 8-anilino-1-naphthalenesulfonic acid (ANS),
discrepancy, for the most part, results from unresolved 2,7-dichlorofluorescein (DCF), and bromthymol blue
separations by the TLC system, inadequate visualization stain interact with the lipid independently of the degree
methods or reagents [54], and diffusion of spots, of saturation. The color intensity of the bromthymol
particularly during two-dimensional TLC [55,56]. One- blue, however, fades quickly after the plate is removed
dimensional TLC systems are preferable if they can from the ammonia vapors, and for this reason
effect the separations desired because they are faster, are bromthymol blue is less desirable than the other
less expensive, and permit standards to be run members of this group.
concurrently. But often, one-dimensional systems have Color development with molybdate and cupric acetate
not provided complete resolution of the lipid mixture. In requires heating for development, as the charring
one instance, Painter [57] was able to attain adequate techniques do. Background darkening can occur,
separations on a one-dimensional system, previously especially if a plate has been developed in a solvent
possible only by two-dimensional TLC, by treating the system that contains an unsaturated component that is
plates with cupric chloride, thereby taking advantage of not fully removed from the plate before application of
the amphoteric nature of phosphatidylserine, a the visualization reagent. ANS and DCF lipid complexes
phospholipid frequently not separated in the are revealed under long-wavelength ultraviolet radiation,
chromatography of surfactant. Laboratory conditions and heating is therefore not required for spot
such as humidity [58] can also affect the resolutions of visualization with these reagents. DCF has a higher
TLC separations. In many instances, PG cannot be background than ANS, but both of these reagents have
measured on the same plate used to determine the L/S the advantage of not reacting chemically with lipid, thus
ratio without overloading the plates with L and S allowing recovery of the intact lipid material, if it is so
desired. The fluorescein indicator is less stable than
137
molybdic stain, and plates treated with this indicator

should be protected from light if they are kept for HPLC-based methods have been published to estimate
extended periods for reference purposes. amniotic fluid phospholipids. The method by DCosta et
al. uses an isocratic system and an ultraviolet detector to
If plates are heated for visualization, they should be establish an L/S ratio. L/S values above 7 correlate with
preheated at 110C before they are placed on the 280C pulmonary maturity. The method by Andrews [66]
hot plate [60]. This technique is recommended because utilizes an internal standard so that individual
commercial thin-layer chromatographic plates have a concentrations of all the common phospholipids can be
tendency to shatter. estimated more accurately. Gradient chromatography is
used with an ultraviolet detector. The time for each assay
Lipid staining is facilitated when ethanol is used as is 16 min. This latter method is therefore capable of
solvent for the staining reagent. Reagents dissolved in providing an entire phospholipid profile in a short period
alcohol more easily penetrate the region containing the of time.
phospholipid, resulting in more evenly stained regions.
Exposure of the spot to the staining reagent for 2 min Of the physical methods, the shake test has been widely
ensures uniform penetration. used. The test is very reliable (more than 99%) when the
test is positive. Less than 1% of infants with a positive
Detection and Quantification test result develop respiratory distress syndrome (RDS).
For those laboratories using planimetry for measuring However, approximately 50% of infants with a negative
the L/S ratio, lecithin and sphingomyelin spots remain test result develop RDS [67]. The test is inexpensive and
visible longer with phosphomolybdic acidstannous quick to perform and when done properly is a useful
chloride than with bromthymol blue. Planimetric adjunct to other methods. Its reliability requires an
methods are critically discussed by Whitfield et al [61]. uncontaminated sample of amniotic fluid. Bile salts, salts
of free fatty acids, and proteins can also exhibit surface
A review of other methods to measure the L/S ratio has activity and contribute to the formation of a stable ring
been published by Olson and Graven [62]. An L/S ratio of bubbles. Also, for the test to be reliable, all glassware
of 2.0 obtained by reflectance densitometry is about must be free of soap, serum, or biological fluids. The
equal to an L/S ratio of 4.0 determined by gravimetric sample must be free of blood contamination and
techniques (Figure 3). It is recommended that a synthetic meconium and, if possible, should be uncentrifuged.
lecithin and sphingomyelin control be routinely used on Clements [26] suggested a cautious centrifugation at 500
plates to monitor spot development to ensure that the end g for 5 minutes to remove red blood cells if necessary
point has been reached before chromatograms are but pointed out that a hematocrit of greater than 3%
quantified [54]. negated the result. Also of paramount concern is the care
with which the ethanol solutions are prepared and
The methods that propose measuring specifically the maintained. In the original paper, Clements [26]
disaturated phosphatidylcholine [18] eliminate classified the results as negative, intermediate, and
unsaturation with osmium tetroxide. This method is positive. The shake test is a valuable screening test but
more applicable to research than to the clinical does not supersede the necessity for the more specific
laboratory [55]. Special care must be taken when one is tests when the shake test is negative or inconclusive.
working with osmium tetroxide. Only when foaming persists at dilutions of 1:2 and
greater should results be considered positive [4].
Gas chromatographic methods have been critically
reviewed [4]. Although this methodology is complicated Surface tension measurements [28-30] give good
and time consuming and requires expensive equipment correlations with other tests for respiratory distress
and higher levels of expertise, useful clinical correlations syndrome and lung surfactant [18,26,31]. Methods that
with surfactant levels have been established [3,63,64]. use a tensiometer are relatively rapid but demand
Correlations of palmitic acid concentrations, palmitic meticulous attention to cleaning and maintenance of the
acidstearic acid ratios, fatty acid compositions of equipment. The foam stability index method [28] is
lecithin isolated from amniotic fluid, and lecithin itself simple and rapid and requires no expensive equipment or
[16] with surfactant levels have aided in successful expertise. This test is sensitive to centrifugation and to
prediction of fetal lung status. contaminants in the amniotic fluid, and patients with
hydramnios may have falsely elevated results [31].
High-performance liquid chromatography (HPLC) offers Erythrocyte contamination is not a problem if the cells
many practical advantages over TLC, such as greater are not hemolyzed, but plasma phospholipids do pose a
sensitivity, precision, accuracy, and speed. Its usefulness problem [28]. The final assay mixture is critical to the
has been limited because it is difficult to obtain results, and extremely close attention to this detail is
simultaneous lecithin, sphingomyelin, and phosphatidyl required [5].
glycerol data from an individual run. HPLC analysis of
phospholipids is also limited by the detector, which The fluorescence polarization assay is reported to require
measures fatty acid double bonds (absorption at 203 nm) less time to perform and to give results that are less
and not the specific saturated lecithin of interest. This subject to variation than the L/S ratio [68]. It has become
type of detection also gives a different response to the most common method in use for assessing fetal lung
ultraviolet absorption for the individual phospholipids
[65].
138
maturity. Results obtained with the fluorescence Although the number of laboratories performing the L/S
polarization method are reported in artificial units (mP). ratio test is decreasing, this test is still considered an
important method for assessing FLM. It is generally
Specimen accepted that with L/S ratios of 2 or greater there is
Amniotic fluid obtained by amniocentesis is transported greatly diminished risk for RDS in the newborn. In one
to the laboratory on ice and is promptly centrifuged at study, it was found that if the L/S ratio was less than 1.5,
500 g for 5 minutes. However, specimens for analysis by RD was found in 73% of the infants. If the L/S ratio was
fluorescence polarization should not be centrifuged, 1.5 to 2.0, RD was found in 40% of the infants. If the
because centrifugation falsely decreases results [69]. A L/S ratio was > 2, there was a 2% risk of RD but only a
3.0-mL aliquot of the supernatant is needed for 0.9% risk if the L/S was > 2.5. This relationship does not
extractions. hold true for infants of diabetic mothers. With diabetics,
the risk for respiratory distress of the newborn is still
Samples should be processed as soon as possible after high, even with an L/S greater than 2. Falsely elevated
collection. Untreated amniotic fluid held at room L/S ratios have been reported after intrauterine
temperature can lose both lecithin and sphingomyelin transfusions. In a later study using an L/S ratio cutoff
over a period of 48 hours. According to Wagstaff et al. value of 2.0 to indicate FLM, cited values for sensitivity
[39], after only 4.5 hours, lecithin levels are decreased and specificity range from 80% to 85% [75]. In addition,
about 25%. Decreases of a small order occur with studies also suggest that clinical interpretation of L/S
sphingomyelin over a similar period of time. This ratios may be different between different racial groups;
disproportionate loss of lecithin relative to African-American infants show FLM at L/S ratios of
sphingomyelin with time results in a consistent fall in the greater than 1.2 [76].
L/S ratio. Because there does not seem to be a significant
quantity of phospholipase activity in human amniotic Surfactant that contains PG has better alveolar
fluid, the decrease may be the result of the lipids stabilizing properties than surfactant that has not yet
adhering to cellular debris in the fluid. acquired PG. Thus mature lung function is also
associated with measurable quantities of PG. In the
absence of PG, the risk of respiratory distress is
Centrifugation of amniotic fluid samples is necessary if significant, but development of the disease is not
storage is contemplated. If amniotic fluid is centrifuged inevitable [77,78]. In one study, when PG was present,
at a low speed, such as 500 g for 5 minutes, immediately RD was found only in 8.4% of infants, but even in the
after collection, it can be kept at room temperature for at absence of PG, 20% of infants did not develop RD [79].
least 4 days with no apparent change in the L/S ratio PG measurement proves a useful adjunct to the L/S ratio
[70]. Freezing does not adversely affect the phospholipid determination when dealing with amniotic fluid that is
content [40]. Centrifuged samples stored at 20C are contaminated with blood or from diabetic mothers. PG
stable for prolonged periods without noticeable appears when the L/S ratio exceeds 2, and its presence
alteration of the L/S ratio. Thus if a delay is anticipated, continues to increase in amniotic fluid with continuing
the fluid should be centrifuged and immediately gestation (Figure 6). By contrast, PI
refrigerated or frozen. (phosphatidylinositol) closely parallels the L/S up to the
point of maturity, thereafter falling as the PG continues
Interferences to increase [77]. The presence of PG appears to provide
The fluorescence polarization assay is sensitive to the the final component in the biochemical maturation of the
effects of centrifugation [71], and contamination with surfactant.
small amounts of serum or meconium could significantly
alter the determined value [72,73]. High-density The question of what PG value validly predicts FLM has
lipoproteins contained in amniotic fluid may hinder the been approached both by estimation of PG
reliability of the method [74]. Contamination of amniotic concentrations in amniotic fluid and by determination of
fluid from debris, mucus, bacteria, or cellular material the proportion of phospholipid represented by PG. PG
from the vagina also causes the L/S to be greater than 2.1 mg/L of amniotic fluid or 10 nmol/mL
uninterpretable. [24] correlates with fetal maturity. FLM is likewise
indicated when PG is equal to or exceeds 3% of the total
Amniotic Fluid Phospholipids Reference Interval phospholipids [2].
An L/S ratio more than 2.0 is considered indicative of
fetal maturity. If a spot is seen in the PG area, the test is Gluck [7] and other workers reported that the L/S ratio
considered positive. If no spot is seen, it is considered was a less reliable predictor of FLM in a diabetic than in
negative. Infants of diabetic mothers may be at greater a normal pregnancy. A delay in the L/S ratio maturation
risk of developing respiratory distress despite an L/S until 36 weeks of gestation or later is not unusual.
ratio of more than 2.0. When using the lamellar body Subsequently it was recognized that the L/S ratio may
count for assessing fetal lung maturity, a count of also be unreliable with hypertension, premature rupture
55,000/L in a centrifuged specimen or 60,000/L in an of the membranes, and intrauterine growth retardation.
uncentrifuged specimen typically implies lung maturity To address these conditions, Kulovich et al. [2,80]
[32]. incorporated the pertinent phospholipids into the lung
profile so that as many factors as possible are
Interpretation considered in the prediction of fetal lung status. This
139
approach assumes an identical extraction of the study to validate their reference intervals or compare
phospholipids. It includes an acetone precipitation step their methods to published clinical outcome studies. The
and a two-dimensional thin-layer chromatographic guidelines discouraged reporting separate reference
separation, followed by quantitation by reflectance intervals for diabetic patients. That was based on
densitometry. The unambiguous, simultaneous absence of clinical outcome study with enough RDS
assessment of PG, lecithin, and sphingomyelin is the cases from diabetic mothers to indicate that the L/S test
object of the majority of the chemical methods currently results are unreliable. Also, well-controlled diabetes
in use, although routine measurement of PG in the expected for diabetic pregnant women will not delay
nondiabetic may not be necessary. pulmonary maturation.
An excessive amount of blood in the sample is an References

important source of error in the L/S ratio [81]. 1 Gluck L, Kulovich MV, Borer RC Jr. The
Concentrations of blood greater than 0.05 mL per 3 mL diagnosis of the respiratory distress syndrome
of amniotic fluid pose problems, particularly if the L/S is (RDS) by amniocentesis. Am J Obstet Gynecol
near 2.0. The L/S of blood is near 1.8, and so fluid 1971;109:440-5.
contaminated with blood would falsely lower the L/S. 2 Kulovich MV, Hallman MB, Gluck L. The lung
PG is helpful in cases of blood contamination, because profile. I. Normal pregnancy. Am J Obstet
PG is not found in blood, and its presence therefore Gynecol 1979;135:57-63.
indicates mature lung status. PG may be misleading, 3 Russell PT, Miller WJ, McLain CR. Palmitic
however, in the patient with chorioamnionitis with acid content of amniotic fluid lecithin as an
premature rupture of the membranes [82]. index to fetal lung maturity. Clin Chem
1974;20:1431-4.
Amniotic Fluid Performance Goals 4 Wagstaff TI. The estimation of pulmonary
Standards of laboratory practice for evaluation of FLM surfactant in amniotic fluid. In: Fairweather
have been published that thoroughly covered DVI, Eskes TK, eds. Amniotic Fluid: Research
preanalytical, analytical, and postanalytical and Clinical Application. 2nd ed. Amsterdam:
considerations for FLM testing [83]. The guidelines Excerpta Medica; 1978:347-391.
recommend laboratories that support obstetric patients 5 Freer DE, Statland BE. Measurement of
not to perform L/S testing if they receive < 15 requests amniotic fluid surfactant. Clin Chem
per week and send samples to a reference laboratory. 1981;27:1629-41.
6 Tsao FH, Zachman RD. Prenatal assessment of
The current evaluation criteria for quantitative L/S ratio fetal lung maturations: a critical review of
is for laboratories to be within 3 standard deviations of amniotic fluid phospholipid tests. In: Farrell
the peer-group mean. For PG, laboratories should be at PM, ed. Lung Development: Biological and
90% of participant consensus. In the 2007 College of Clinical Perspectives. Vol II. Neonatal
American Pathologists proficiency survey, the mean, Respiratory Distress. New York: Academic
SD, and coefficient of variation were not calculated for Press; 1982:167-203.
the L/S ratio in-house methods, owing to the very low 7 Gluck L, Kulovich MV.
number of participating laboratories. There was a great Lecithin/sphingomyelin ratios in amniotic fluid
variation in methods and cutoffs (1.9 to 4.1 for in normal and abnormal pregnancy. Am J
maturity). Interpretations, however, were consistent for Obstet Gynecol 1973;115:539-46.
LM-04 and LM-05, but no consensus was achieved for 8 Gluck L, Kulovich MV, Borer RC Jr. Estimates
LM-06. For the Helena Fetal Tek 200 peer group, the of fetal lung maturity. Clin Perinatol
reported sample ID, mean, SD, and CV were: LM-04, 1974;1:125-39.
1.08, 0.09, 8.4%; LM-05, 3.54, 0.71, 20.1%; and LM-06, 9 Mueller RG. Effect of charring temperature on
1.85, 0.29, 15.8%. Again, there was great consistency on observed L/S ratio. Clin Chim Acta
interpretations of LM-04 and LM-05 challenges but no 1982;122:79-83.
consensus on LM-06 [84]. 10 Coch EH, Kessler G, Meyer JS. Rapid thin-
layer chromatographic method for assessing the
Laboratories should offer at least one test that can be lecithin-sphingomyelin ratio in amniotic fluid.
performed in less than 1 hour, such as the fluorescence Clin Chem 1974;20(10):1368-75.
polarization immunoassay on Abbott TDx (TDx/FLM 11 Fiske CH, Subbarow Y. The colorimetric
test), the Irvine Scientific slide test for PG, or the foam determination of phosphorus. J Biol Chem
stability test (a homebrew test or Beckman Lumadex- 1925;66:375-400.
FSI). It was also recommended that laboratories should 12 Chen PS, Toribara TY, Warner H.
perform the LBC test but report its result together with Microdetermination of phosphorus. Anal Chem
one of the recommended rapid tests. Laboratories should 1956;28:1756-8.
offer routine L/S testing once per day, 7 days a week. 13 Bartlett GR. Phosphorus assay in column
FLM results should be communicated rapidly to the chromatography. J Biol Chem 1959;234:466-8.
ordering clinicians, and reports should also include 14 Badham LP, Worth HGJ. Critical assessment of
sample condition (in particular, any observed phospholipid measurement in amniotic fluid.
contamination) and reference information. Laboratories Clin Chem 1975;21:1441-7.
performing L/S ratio should conduct a clinical outcome
140
15 Yee HY, Yee TML, Jackson B. Determination 30 Goldkrand JW, Varki A, McClurg JE. Surface
of phosphatidylcholine in amniotic fluid. tension of amniotic fluid lipid extracts:
Microchem J 1980;25:61-71. prediction of pulmonary maturity. Am J Obstet
16 Lohinger A, Salzer H, Sumbruner G et al. 1977;128:591-8.
Relationships among human amniotic fluid 31 Bichler A, Daxenbichler G, Ortner A, Grill H,
dipalmitoyl lecithin, postpartum respiratory Geir W, Dapunt O. Amniotic fluid surface
compliance, and neonatal respiratory distress tension measurements versus various
syndrome. Clin Chem 1983;29:650-5. phospholipid determinations in the assessment
17 Sax SM, Moore JJ, Oley A, Amenta JS, of fetal lung maturity. Respiration 1979;37:114-
Silverman JA. Liquid-chromatographic 21.
estimation of saturated phospholipid palmitate 32 Ashwood ER, Palmer SE, Taylor JS, Pingree
in amniotic fluid compared with a thin-layer SS. Lamellar body counts for rapid fetal lung
chromatographic method for acetone- maturity testing. Obstet Gynecol 1993;81:619-
precipitated lecithin. Clin Chem 1982;28:2264- 24.
8. 33 Lee IS, Cho YK, Min WK, Kim KS, Mok JE.
18 Mason RJ, Nellenbogen J, Clements JA. Lamellar body count in amniotic fluid as a rapid
Isolation of disaturated phosphatidylcholine screening test for fetal lung maturity. J Perinatol
with osmium tetroxide. J Lipid Res 1996;16:176-80.
1976;17:281-4. 34 DeRoche ME, Ingardia CJ, Guerette PJ, Wu
19 McDonald L, Robin NL, Siegel L. New method AH, LaSala CA, Mandavilli SR. The use of
for determining lecithin and sphingomyelin in lamellar body counts to predict fetal lung
amniotic fluid. Clin Chem 1981;27:410-6. maturity in pregnancies complicated by diabetes
20 Artiss JD, Draisey TF, Thibert RJ, Taylor KE. mellitus. Am J Obstet Gynecol 2002;187:908-
A procedure for the direct determination of 12.
micromolar quantities of lecithin employing 35 Wijinberger LD, Huisjes AJ, Voorbij HA,
enzymes as reagents. Microchem J Franx A, Bruinse HW, Mol BW. The accuracy
1979;24:239-58. of lamellar body count and
21 Artiss JD, Draisey TF, Thibert RJ, Zak B, lecithin/sphingomyelin ratio in the prediction of
Taylor KE. The determination of lecithin and neonatal respiratory distress syndrome: a meta-
total choline-containing phospholipids in analysis. Br J Obstet Gynecol 2001;108:583-8.
amniotic fluid employing enzymes as reagents. 36 Neerhof MG, Donhnal JC, Ashwood ER, Lee
Microchem J 1980;25:153-68. IS, Anceschi MM. Lamellar body counts: a
22 Anaokar S, Garry PJ, Standefer JC. Enzymatic consensus protocol. Obstet Gynecol
assay for lecithin in amniotic fluid. Clin Chem 2001;97:318-20.
1978;25:103-7. 37 Ventolini G, Neiger R, Hood D, Belcastro C.
23 McGowan MW, Artiss JD, Zak B. Enzymatic Update on assessment of fetal lung maturity. J
colorimetry of lecithin and sphingomyelin in Obstet Gynecol 2005;25:535-8.
aqueous solution. Clin Chem 1983;29:1513-7. 38 Szallasi A, Gronowski AM, Eby CS. Lamellar
24 Siegel L, Walker SI, Robin NI. An enzymatic body count in amniotic fluid: a comparative
radiochemical method for determining study of four different hematology analyzers.
phosphatidylglycerol in amniotic fluid. Clin Clin Chem 2003;49:994-7.
Chem 1983;29:782-5. 39 Wagstaff TI, Whyley GA, Freedman G. Factors
25 Artiss JD, McGowan MW, Strandbergh DR, influencing the measurement of the
Epstein E, Zak B. Enzymatic colorimetric lecithin/sphingomyelin ratio in amniotic fluid. J
determination of phosphatidylglycerol in Obstet Gynaecol Br Commonw 1974;81:264-
amniotic fluid. Clin Chem 1984;30:534-7. 77.
26 Clements JA, Platzker ACG, Tierney DF. 40 Van Voorst tot Voorst EJ. Effects of
Assessment of risk of respiratory distress centrifugation, storage, and contamination of
syndrome by a rapid test for surfactant in amniotic fluid on its total phospholipid content.
amniotic fluid. N Engl J Med 1972;286:1077- Clin Chem 1980;26:232-4.
81. 41 Oulton M. The role of centrifugation in the
27 Sher G, Statland BE, Freer DE. Clinical measurement of surfactant in amniotic fluid.
evaluation of the quantitative foam stability Am J Obstet Gynecol 1979;135:337-43.
index test. Obstet Gynecol 1980;55:617-20. 42 Duck-Chong CG, Brown LM, Hensley WJ.
28 Statland BE, Freer DE. Evaluation of two Sedimentation of lung-derived phospholipid
assays of functional surfactant in amniotic fluid, during low-speed centrifugation of amniotic
surface tension lowering ability and the foam fluid. Clin Chem 1971;27:1424-6.
stability index test. Clin Chem 1979;25:1770-3. 43 Cherayil GD, Wilkinson EJ, Borkowf HI.
29 Tiwary CM, Goldkrand JW. Assessment of Amniotic fluid lecithin/sphingomyelin ratio
fetal pulmonary maturity by measurement of changes related to centrifugal force. Obstet
the surface tension of amniotic fluid lipid Gynecol 1977;50:682-8.
extract. Obstet Gynecol 1976;48:191-4.
141
44 Wilkinson EJ, Cherayil GD, Borkowf HI. L/S 59 Glick JR Jr, Crocker CL. Concentration of
ratio and the g-force factor. N Engl J Med ammonium hydroxide is critical in
1977;296:286-7. chromatographic solvents for amniotic fluid
45 Lindback T, Frantz T. Effect of centrifugation phospholipids. Clin Chem 1982;28:1997-8.
on amniotic fluid phospholipid recovery. Acta 60 Gross TL, Wilson MV, Kuhnert PM, Sokol RJ.
Obstet Gynecol Scand 1975;54:101-3. Clinical laboratory determination of
46 Hudson EA, Aimls JG. Amniotic fluid cells and phosphatidylglycerol: one- and two-
the lecithin/sphingomyelin ratio. Obstet dimensional chromatography compared. Clin
Gynecol 1976;49:280-6. Chem1981;24:486-90.
47 Bligh EG, Dyer WJ. A rapid method of total 61 Whitfield CR, Chan WH, Sproule WB, Stewart
lipid extraction and purification. Can J Biochem AD. Amniotic fluid lecithin: sphingomyelin
Physiol 1959;37:911-7. ratio and fetal lung development. Br Med J
48 Bhagwanani SG, Fahmy D, Turnbull AC. Quick 1972;2:85.
determination of amniotic fluid lecithin 62 Olson EB, Graven SN. Comparison of
concentration for prediction of neonatal visualization methods used to measure the
respiratory distress. Lancet 1972;2:66-7. lecithin/sphingomyelin ratio in amniotic fluid.
49 Sarkozi L, Kovacs HN, Fox HA, Kerenyi T. Clin Chem 1974;20:1408-15.
Modified method for estimating the 63 Szalay J, Lukacs E. Palmitic acid concentration
phosphatidylcholine-sphingomyelin ratio in of amniotic fluid in diabetic pregnancy. Arch
amniotic fluid, and its use in the assessment of Gynaekol 1977;222:279-83.
fetal lung maturity. Clin Chem 1972;18:956-60. 64 ONeal GJ, Davies IJ, Siu J. Palmitic-stearic
50 Gluck L, Motoyama EK, Smits HL, Kulovich ratio of amniotic fluid in diabetic and non-
MV. The biochemical development of surface diabetic pregnancies and its relationship to
activity in mammalian lung. Pediatr Res development of respiratory distress syndrome.
1967;1:237-46. Am J Obstet Gynecol 1978;132:519-23.
51 Olson EB, Graven SN, Zachman RD. Amniotic 65 Geurts Van Kessel WSM, Hax WMA, Demel
fluid lecithin to sphingomyelin ratio of 3.5 and RA, DeGhier J. High-performance liquid
fetal pulmonary maturity. Pediatr Res chromatographic separation and direct
1975;9:65-9. ultraviolet detection of phospholipids. Biochim
52 Armstrong D, VanWormer DE. Rapid Biophys Acta 1977;486:524-30.
determination of pulmonary surfactant. Am J 66 Andrews AG. Estimation of amniotic fluid
Obstet Gynecol 1972;114:1083-6. phospholipids by high-performance liquid
53 Parkinson CE, Harvey DR. A comparison chromatography. J Chromatogr 1984;336:139-
between the lecithin/sphingomyelin ratio and 50.
other methods of assessing the presence of fetal 67 Parkinson GE, Harvey D. Amniotic fluid and
lung pulmonary surfactant in amniotic fluid. J pulmonary maturity. In: Sandler M, ed.
Obstet Gynaecol Br Commonw 1973;80:406- Amniotic fluid and its clinical significance.
11. New York: Marcel Dekker; 1981:229-252.
54 Spillman T, Cotton DB, Lynn SC Jr, 68 Gebhardt DOE. Relationship between the
Bretaudiere JP. Influence of phospholipid lecithin/sphingomyelin ratio and the P value of
saturation on classical thin-layer amniotic fluid. Am J Obstet Gynecol
chromatographic detection methods and its 1979;133:937.
effect on amniotic fluid lecithin/sphingomyelin 69 Russell JC, Cooper CM, Ketchum CH, Torday
ratio determinations. Clin Chem 1983;29:250-5. JS, Richardson DK, Holt JA et al. Multicenter
55 Tsai MY. Relative merits of one- and two- evaluation of TDx test for assessing fetal lung
dimensional TLC of phospholipids in amniotic maturity. Clin Chem 1989;35:1005-10.
fluid. Clin Chem 1981;27:1957-8. 70 Whitfield CR, Sproule WB. Fetal lung
56 Cunningham MD, McKean HE, Gillespie DH, maturation, Br J Hosp Med 1974;12:678-90.
Greene JW. Improved prediction of fetal lung 71 Simon NV, Elser RC, Levisky JS, Polk DT.
maturity in diabetic pregnancies: a comparison Effect of centrifugation on fluorescence
of chromatographic methods. Am J Obstet polarization of amniotic fluid. Clin Chem
Gynecol 1982;142:197-204. 1981;27:930-2.
57 Painter PC. Simultaneous measurement of 72 Blumenfeld TA, Stark RI, James LS, George
lecithin, sphingomyelin, phosphatidylglycerol, JD, Dyrenfurth I, Freda VJ et al. Determination
phosphatidylinositol, of fetal lung maturity by fluorescence
phosphatidylethanolamine, and polarization of amniotic fluid. Am J Obstet
phosphatidylserine in amniotic fluid. Clin Chem Gynecol 1978;130:782-7.
1980;26:1147-51. 73 Golde SH, Vogt JF, Gabbe SG, Cabal LA.
58 Tsao FHC. The effect of humidity on the Evaluation of the FELMA microviscosimeter in
determination of lecithin/sphingomyelin ratio predicting fetal lung maturity. Obstet Gynecol
and phosphatidylglycerol by thin-layer 1979;54:639-43.
chromatography. Clin Chim Acta 1982;122:75- 74 Gebhardt DOE. Detection of neonatal
8. respiratory distress on the basis of fluorescence
142
polarization (microviscosity) measurements of 83 Ashwood ER. Standards of laboratory practice

amniotic fluid: a word of caution [letter]. Clin evaluation of fetal lung maturity. Clin Chem
Chem 1980;26:1629. 1997;43:211-4.
75 Dubin SB. Assessment of fetal lung maturity by 84 College of American Pathologists. 2007
laboratory methods. Clin Lab Med Surveys: Lung Maturity. Northfield, IL: CAP;
1992;12:603-20. 2007.
76 Richardson DK, Torday JS. Racial differences 85 Hallman M, Kulovich M, Kirkpatrick E,
in predictive value of the Sugarman RG, Gluck L. Phosphatidylinositol
lecithin/sphingomyelin ratio. Am J Obstet and phosphatidylglycerol in amniotic fluid:
Gynecol 1994;170:1273-8. indices of lung maturity. Am J Obstet Gynecol
77 Hallman M, Feldman BH, Kirkpatrick E, Gluck 1976;125:613-7.
L. Absence of phosphatidylglycerol (PG) in 86 Wagstaff TI, Whyley GA, Freedman G. The
respiratory distress syndrome in the newborn. measurement of the lecithin/sphingomyelin
Pediatr Res 1977;11:714-20. ratio of amniotic fluid after thin-layer
78 Hallman M, Feldman B, Gluck L. RDS: the chromatography. Ann Clin Biochem
absence of phosphatidylglycerol in surfactant 1974a;11:24-7.
[abstract]. Pediatr Res 1975;9:396. 87 Buhi WC, Spellacy WN. Effects of blood or
79 Bent AE, Gray JH, Luther ER, Oulton M, meconium on the determination of the amniotic
Peddle LJ. Phosphatidyl-glycerol determination fluid lecithin/sphingomyelin ratio. Am J Obstet
on amniotic fluid 10,000 g pellet in the Gynecol 1975;121:321-3.
prediction of fetal lung maturity. Am J Obstet 88 Torday J, Carson L, Lawson EE. Saturated
Gynecol 1981;139:259-63. phosphatidylcholine in amniotic fluid and
80 Kulovich MV, Gluck L. The lung profile. II. prediction of the respiratory-distress syndrome.
Complicated pregnancy. Am J Obstet Gynecol N Engl J Med 1979;301:1013-8.
1979;135:64-70. 89 Oulton M, Martin TR, Faulkner GT, Stinson D,
81 Gluck L, Kulovich MV, Borer RC, Keidel WN. Johnson JP. Developmental study of a lamellar
The interpretation and significance of the body fraction isolated from the human amniotic
lecithin/sphingomyelin ratio in amniotic fluid. fluid. Pediatr Res 1980;14:722-8.
Am J Obstet Gynecol 1974;120:142-55. 90 Gerbie MV, Gerbie AB, Boehm J. Diagnosis of
82 Schumacher RE, Parisi VM, Steady HM, Tsao fetal maturity by amniotic fluid phospholipids.
FHC. Bacteria causing false-positive test for Am J Obstet Gynecol 1977;114:1078-82.
phosphatidylglycerol in amniotic fluid. Am J
Obstet Gynecol 1985;151:1067-8.
143
Chromatography
Tables (TLC)
Table 1: Methods of Amniotic Fluid Analysis Assay Ambient Ambient
Chromatographic Analysis of Phospholipids temperature
Method 1: Thin-layer chromatography (TLC)
Type of analysis: Semiquantitative Sample 1 mL 1.0, 0.75, 0.50,
Principle: Silica-gel adsorption volume 0.25, and 0.20
chromatography; visualization of phospholipids mL
by color-forming sprays, or charring
Usage: Most common Fraction of 0.50 (extraction) 0.50
Comment: Can be run as one dimension or sample
two dimensions volume
Method 2: Gas-liquid chromatography (GLC)
Type of analysis: Quantitative Final (Not applicable) 47.5% ethanol
Principle: Analysis of fatty acids on polyester concentration
columns with flame ionization detector of reagents
Usage: Rare
Comment: Measured as methyl ester Interferences Blood, meconium, Blood,
derivatives high-speed meconium,
Method 3: High-performance liquid chromatography centrifugation soap, serum,
(HPLC) before testing biological
Type of analysis: Quantitative fluids, surface
Principle: Separation of L, S, PI, PE, PS, and evaporation
PG by reversed-phase chromatography, from tubes,
detection at 203 nm movement of
Usage: Rare tubes once they
Comment: Ability to quantitate multiple have been
components simultaneously may make this an shaken, high-
important method in the future. speed
centrifugation
Measurement of Surfactant Functional Activity of sample
Method 4: Foam stability (shake test, FSI) before testing
Type of analysis: Qualitative
Principle: Measures presence of surfactant by
its ability to support foam bubbles generated by
shaking amniotic fluid with ethanol
Usage: Common as stat procedure
Comment: Problem of false-negative results
Method 5: Surface tension
Type of analysis: Quantitative
Principle: Presence of surfactant related to
surface-tension properties of amniotic fluid
Usage: Rare
Comment: Dedicated instrument, but
relatively easy to perform
Method 6: Fluorescence polarization
Type of analysis: Quantitative
Principle: Surfactant increases microviscosity
of fluid, which reduces the rotation of dissolved
fluorescent probe, resulting in an increase in
polarization of incident fluorescent light
Usage: Frequently used
Comment: Common as stat procedure;
requires dedicated instrument but relatively
easy to perform
Table 2: Comparison of Assay Conditions for

Amniotic Fluid Analysis
Thin Layer Shake Test

144
Figures
Figure 1. L/S (lecithin/sphingomyelin) ratio related to

centrifugal force. In amniotic fluids with an initial L/S
ratio above 2, L/S ratios decrease with increasing g force
[43].
Figure 2. Surface tensions of purified lecithin from lung

homogenates at different times of gestation in fetal
rabbit. Total lecithin extracted from lung is not surface
active. Acetone-precipitated lecithin was surface active
and present in lungs from fetuses at all gestational ages
examined. NB, Newborn [50].
145
Figure 4. Separation of lecithin and sphingomyelin on

thin-layer strips as described for method to determine
L/S ratio. Pt, Patient sample; STD, L and S standards.
Figure 3. Relationship between densitometry L/S ratios

and those with area measurement and by gravimetry
[81].
Figure 5. Phosphatidyl glycerol (PG) spot on silica gel

thin-layer plate according to method for PG. L, Lecithin;
PI, phosphatidyl inositol, PT, patient; S, sphingomyelin;
STD, standard (left sample, PG; middle sample, PG, L,
and S).
Figure 6. Content of phosphatidyl inositol (PI) (closed

circles) and phosphatidylglycerol (PG) (open circles) in
amniotic fluid during normal gestation. Phospholipids
146
were quantified by measurement of phosphorus (P)

content and expressed as percentages of total lipid
phosphorus. Means standard deviations of three to five
samples are shown for each point [85].
147
2. Pipet 1 mL of the supernatant into duplicate 12-

Procedure: Lecithin-Sphingomyelin Ratio (L/S) by mL conical screw-capped test tubes. Pipet 1 mL into
TLC each of three tubes if PG is also to be determined.
3. To each 1-mL sample of amniotic fluid, add 1
Principle
mL of methanol. Vortex mix. Add 2 mL of chloroform,
Phospholipids are extracted from the amniotic fluid into and vortex mix vigorously.
chloroform-methanol and then precipitated in cold 4. The tubes are centrifuged (500 g for 5 min to
acetone. The precipitated phospholipids are separated by
facilitate phase separations). Three layers should be
thin-layer chromatography, visualized, and quantitated. seen:
An estimate is made of the amount of lecithin and a. Aqueous layer (top)
sphingomyelin present in the amniotic fluid. This is
b. Protein fluff layer (middle)
expressed as a ratio of lecithin to sphingomyelin (L/S) c. Chloroform layer (bottom)
and is used in the evaluation of FLM. 5. Remove as much of the chloroform layer as
Reagents possible using a disposable Pasteur transfer pipet (go
1. Eastman Chromatograph Sheets, 6061 Silica through the top two layers carefully). Place this in a 3-
gel without fluorescent indicator, 20 20 cm. Cut mL conical tube. Evaporate to moist dryness under
nitrogen. (Note: Water bath should be below 40C.)
into 1 13 cm strips. Activate by storing in a desiccator
6. If PG is to be determined, repeat the extraction
over Drierite for 24 hr.
(steps 3 to 5) for the tube for PG.
2. Developing solvent. Pipet into a 50-mL round-
7. Wash down the lipid adhering to the sides of
bottom, capped centrifuge tube 5.0 mL of methanol, 0.8
mL of distilled water, 13.0 mL of chloroform. Vortex each tube with 50 L of chloroform. Take to moist
mix. Keep solution capped when not in use. Prepare dryness again.
fresh daily. 8. Chill the 3-mL tubes with the lipid extract in ice
3. Methanol. Spectral grade for 1 min. With a Pasteur pipet, add 2 drops of ice-cold
4. Chloroform. Spectral grade acetone while swirling the tube in the ice. Eight more
5. Acetone. Reagent grade drops of ice-cold acetone is added, and the tube is iced
6. Bromthymol blue (417 mg/L). A water- for another minute. The acetone is decanted. The
soluble (3,3-dibromothymol sulfone phthalein, sodium precipitate is washed with another 10 drops of ice-cold
salt) available from Aldrich Organic Chemicals acetone while the tube remains iced. The supernatant is
(Milwaukee, WI). Add 200 mg of bromthymol blue to decanted. The precipitate is dried gently under a stream
200 mL of distilled water and 32.0 mL of 1 M NaOH. of nitrogen.
Dilute to 480 mL with distilled water. Add 5.0 g of boric 9. The cold acetone precipitate is dissolved in 6
acid powder (reagent grade) and mix. This solution is L of chloroform. The entire amount is then spotted 2
stable for 3 months. cm from the end of the activated TLC strip. (The PG
7. Nitrogen gas. Compressed, 100% tube is spotted on a plate to be described later.)
8. 1 M NaOH (40 g/L). Place 40 g of NaOH in a 10. Using forceps to handle the strips, place each
1-L volumetric flask, and slowly add about 800 mL of into 1.5 mL of the developing solvent in a 20 150 mm
distilled water to dissolve. When the NaOH is fully test tube. The tube is stoppered with a cork. The strip is
dissolved, and the temperature of the solution is back to chromatographed to a distance of 2 cm from the top of
room temperature, add distilled water to mark. Mix the strip.
thoroughly. Stable for 12 months at room temperature. 11. The strips are removed from the tubes with
9. Standards. The lecithin, sphingomyelin, and forceps, air dried, and then dipped in the bromthymol
phosphatidylglycerol standards can be purchased from blue indicator contained in a 20 150 mm test tube.
Supelco, Inc. (Bellefonte, PA) as chloroform-methanol Excess dye is blotted from the strip. The lecithin and
solutions. The contents of each ampule are quantitatively sphingomyelin spots appear dark yellow (Figure 4).
transferred to a volumetric flask (the ampule must be Visualization is made easier when the strip is briefly
rinsed as well) and then diluted with chloroform- placed in a 20 150 mm test tube containing
methanol (2:1, v/v) to a final concentration of 1 mg/mL. concentrated NH4OH. The ammonia vapors turn the
10. Lecithin. 50 mg is diluted to 50 mL with spots blue.
chloroform-methanol. Calculation
11. Sphingomyelin. 25 mg is diluted to 25 mL 1. Circle each spot along the outermost edges with
with chloroform-methanol. a pencil. Measure maximum width and length of each
12. Phosphatidylglycerol. 10 mg is diluted to 10 spot in millimeters. Using the following calculation:
mL with chloroform-methanol. L/S ratio = _
(length of lecithin spot) (width of lecithin spot)_
Assay
Equipment: centrifuge, 20 50 mm test tubes with cork (length of sphingomyelin spot) (width of sphingomyelin spot)
stoppers, forceps, 12-mL conical screw-capped 2. Average the values from the duplicate samples.
centrifuge tubes, 3-mL conical test tubes. 3. Apply L/S standards daily. Spot 30 L of
This procedure is best performed in duplicate. lecithin standard (10 mg/mL) on top of 15 L of
1. Centrifuge 3 to 5 mL of amniotic fluid at 500 g sphingomyelin standard (100 g/mL) on a fresh strip and
for 5 min at ambient temperature. chromatograph.
4. Running time for L/S, 30 to 45 minutes.
148
Amniotic Fluid Phospholipids Reference Interval Contamination of amniotic fluid with meconium
An L/S ratio more than 2.0 is considered indicative of interferes with the interpretation of chromatograms,
fetal maturity. Infants of diabetic mothers may be at though Gerbie et al. [90] did not consider that its
greater risk of developing respiratory distress despite an presence interfered with the estimation of the L/S ratio.
L/S ratio of more than 2.0. Wagstaff et al. [39] demonstrated a consistent rise in the
L/S ratio with increasing contamination with meconium.
Procedure: Phosphatidylglycerol (PG) by TLC Chromatograms from samples contaminated with
meconium are characterized by a lysolecithin spot
Reagents immediately after the sphingomyelin spot. If
1. Chloroform. Spectral grade chromatographic separation is not good, the lysolecithin
2. Methanol. Spectral grade
spot will come to lie so close to the sphingomyelin spot
3. Glacial acetic acid. Reagent grade that the two may be confused, giving a falsely decreased
4. Solvent system. Chloroform/methanol/acetic L/S ratio [87]. Other authors report falsely high L/S
acid/water (65:25:8:4, v/v) ratios from fluid contamination with meconium. Again,
5. 2,7-Dichlorofluorescein (0.2 g/L) in ethanol. it seems prudent to rely on the presence of PG if samples
Dissolve 2 g of dye into 100 mL of ethanol. This is the
are contaminated with meconium or to assess surfactant
stock solution. Dilute 1:100 in ethanol for working
of the 10,000 g pellet [41,89].
solution. When stored at room temperature, the stock
solution is stable for 1 year and the working solution for Amniotic Fluid Phospholipids Reference Interval
3 months. If a spot is seen in the PG area, the test is considered
positive. If no spot is seen, it is considered negative.
Assay
Equipment: Silica gel 60 thin-layer plates, 5 20 cm (E. Shake Test on Amniotic Fluid: Rapid Test for
Merck, Darmstadt, Germany). Plates are activated at Surfactant
100C for 1 hr before use. Filter-paper-lined thin-layer
chromatography tank.
I. Principles of Test and Clinical Significance
1. Use the third tube prepared in step 2 of the L/S
ratio procedure. Lipid contents are dissolved in 10 L of
chloroform and spotted 2 cm from the bottom and a third A. Principle: This test depends on the
of the way in from the side of a silica-gel (5 20 cm) ability of pulmonary surface active material (surfactant)
TLC plate (activated). to generate stable foam in the presence of 95% ethanol.
It looks at a physical property of the amniotic fluid.
2. Lecithin (20 L of 1 g/L) and PG (20 L of
1 g/L) standards are spotted 2 cm from the bottom and
B. Clinical Significance: This test is a
a third of the way in from the opposite side of the same
rapid assessment of fetal lung maturity. If test result
plate.
indicates mature, RDS should not occur in infants if they
3. The plate is placed in a filter-paper-backed thin-
are delivered within 72 hr after analysis is done.
layer tank previously equilibrated with the solvent
system for at least 1 hr (replace solvent system weekly
II. Sample (Specimen)
or more often if runs become atypical).
4. The solvent system is permitted to run about 10
A. 1.5 mL of unspun amniotic fluid
cm up the plate. The plate is withdrawn from the tank
obtained by amniocentesis.
and air dried for 15 to 20 min.
5. The plate is sprayed with dichlorofluorescein
B. If test is not able to be performed right
reagent, allowed to dry, and viewed under ultraviolet
away, refrigerate sample at 2C to 8C. Then mix
radiation 15 min to 1 hr later (Figure 5).
sample well before pipetting amniotic fluid for shake
6. The visual identification of a PG spot confirms
test.
the presence of PG. This visualization is sensitive to 0.5
g and indicates the presence of PG in amniotic fluid to C. Samples that are bloody, contain
be greater than or equal to 0.5 g/mL. meconium, or are from vaginal pool collections are not
acceptable. Ascitic fluid samples from infants will give
Notes
false-positive results. These are also unacceptable
Blood contamination of amniotic fluid has some effect
specimens.
on the L/S ratio, but there is no agreement about whether
the blood increases or decreases L/S ratio [39,86,87].
III. Reagents
Samples contaminated with blood are still useful in the
prediction of fetal lung maturity if phosphatidyl glycerol
A. Normal saline (0.9% NaCl). Pour fresh
is found. PG is virtually absent from blood, meconium,
aliquot daily.
and vaginal secretions, all of which contain
phospholipids and other components that cause
B. Absolute ethanol (200 proof). Store at
inaccurate L/S ratios [41,87,88]. Alternatively, one can
room temperature.
isolate the 10,000 g pellet from the contaminated fluid
and measure fetal pulmonary surfactant from the
C. 95% ethanol. To prepare 100 mL of
lamellar bodies [41,89].
95% ethanol, add 5 mL distilled H2O to 95 mL absolute
149
ethanol. Keep tightly closed. Stable 1 month at room X. Procedure Notes

temperature. A. Comparable results are obtained using
glass tubes with an inner diameter varying from 8 to 14
IV. Equipment mm. In larger tubes, it is more difficult to detect the fine
A. 5 mL, 1 mL, and 0.5 mL volumetric pipets. bubbles.
B. 13 100 mm glass tubes and B. Evaporation decreases the stability of

corresponding plastic caps. the foam, especially in hot, dry rooms. Clean rubber
stoppers should be used to minimize evaporation.
V. Calibration N/A
C. A precipitate occasionally forms on
VI. Quality Control N/A addition of ethanol to amniotic fluid. When this happens,
the meniscus must be observed carefully to avoid
VII. Procedure and Methodology confusion between the precipitate and the foam.
A. Do not spin sample. Gently invert tube
to suspend particles. D. Ascitic fluid gives positive tests.
B. Label two 13 100 mm test tubes #1 and #2. E. This test should not be used when
C. Set up the tubes in the following manner: meconium or blood is seen in the amniotic fluid (may
produce a false positive).
Tube #1:1.0 mL of unspun fluid
1.0 mL of 95% ethanol
Tube #2:0.5 mL of unspun fluid

0.5 mL of normal saline
1.0 mL of 95% ethanol
Note: Use volumetric pipets.
D. Cap tubes with disposable plastic caps,

and shake vigorously for exactly 15 sec.
E. Let stand in rack for 15 min at room

temperature.
F. Examine the top of the fluid for small,

stable bubbles. Do not move the tubes. View against a
black background if hard to visualize.
G. Interpretation:
If foam persists in tubes 1 and 2 mature

If foam is in tube 1 and not tube 2borderline
No foam in tubes 1 and 2immature
VIII. Calculations (Derivation of Results)
No calculations are necessary.

Report results as immature, borderline, or
mature.
IX. Reference Value
If test result indicates mature, RDS should not

occur in an infant delivered within 72 hours after
analysis is done.
150
Amylase
Amylase
Ming Jin
Name: Amylase, -1,4-glucan 4-glucanohydrolase (AMY)
Clinical significance: Refer to Chapter 34, The Pancreas: Function and Chemical
Pathology, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Enzyme commission number: EC 3.2.1.1
Molecular weight: 54 to 62 kDa
Chemical class: Enzyme
Isoenzymes: At least seven in human tissue; salivary and
pancreatic forms in plasma
i of amylase activity by use of laser nephelometry has

shown improvements in analytical sensitivity and
Amylases are enzymes that degrade complex precision when compared to conventional nephelometry
carbohydrate molecules into smaller components. [2].
Amylase is produced by the exocrine pancreas and the
salivary glands to aid in the digestion of starch. Human Turbidimetric and light-scattering by nephelometry
amylase is -amylase, which has the ability to hydrolyze measurements are carried out either kinetically or after a
-1,4 glucosidic linkages in polysaccharides. The -1,6 fixed time interval, and the change in turbidity or light
linkages at the branch points remain untouched. The scattering is proportional to amylase activity. These
result of -amylase action on polysaccharide is the methods are difficult to standardize because of substrate
formation of dextrans, maltose, and glucose molecules. variation, and in general they lack precision [3].
There are two types of amylases: one is P-type, which is
produced in the pancreas, and the other is S-type, which The iodometric methods are based on the ability of
is present in the salivary gland. iodine to form a vivid blue color after reaction with
starch. The by-products of amylase action may also form
More than 200 methods for amylase measurement have colored substances with iodine, but their absorbance
been devised based upon different principles and maxima are at different wavelengths from the
substrates. The assay conditions must be rigidly adjusted characteristic starch-iodine complex. The methods are
to suit the specific requirements of amylase. The pH carried out by adding iodine color reagent to the
optimum is 6.9 to 7.0, and calcium and chloride ions are substrate-sample mixture after an incubation period. The
required cofactors. A variety of different detection greater the amount of amylase activity, the lighter will
techniques are available to measure the activity of be the color of the final solution.
amylase. The different approaches can be grouped
generally into three categories. The most common approach employs a fixed time
interval [4]. Substrate and sample are mixed together and
Amyloclastic Techniques (Table 1, Methods 1a and 1b) incubated for a fixed period. The iodine color solution is
Amyloclastic methods for amylase assay are based on then added, and a spectrophotometric measurement is
the measurement of rate hydrolysis of starch by amylase. performed. The lower the final absorbance (the greater
Turbidimetric, nephelometric, and iodometric techniques the difference between the absorbance of the blank and
are used to detect the rate of starch hydrolysis. the absorbance of the sample), the higher the activity.
Amyloclastic methods are no longer widely employed,
Turbidimetric technique utilizes the ability of amylase to in part because substrate variability makes
decrease the turbidity of a substrate suspension by standardization difficult.
enzymatic degradation of the substrate into smaller units. Saccharogenic Techniques (Table 1, Methods 2a and 2b)
Nephelometry has also been used for amylase Saccharogenic techniques depend on the measurement of
measurements. Nephelometric techniques measure the monosaccharides or disaccharides liberated by amylases
increase in light scattering due to the hydrolysis of starch reaction with the substrate. The classical amylase
by amylase [1]. These instruments carry out rate method of Somogyi is a saccharogenic technique [5].
measurements over a short time. Recently, measurement Sample and substrate are incubated for 30 minutes, then
a serum blank and the sample reaction tube undergo
measurement for reducing substances. Since the by-
i
Amylase products of amylase action are reducing substances, the
enzymes action is directly proportional to the amount of
First edition: Michael D.D. McNeely reducing substance produced.
Methods edition: Michael D.D. McNeely
More recently, the saccharogenic approach has been
Second edition: Michael D.D. McNeely
Third edition: Steven C. Kazmierczak modified for automated kinetic analysis. In these
Fourth edition: Steven C. Kazmierczak techniques, the maltose hydrolysed by amylase from the
Fifth edition: Ming Jin substrate, such as maltopentaose and maltotetraose, is
151
Amylase
converted into glucose by enzyme, which is included in described at the end of the section, and the detailed
the reagent mixture. The glucose liberated by this reference procedure is in the Procedure Section.
reaction can be measured by several enzymatic glucose
assay techniques [6]. Historically, the amyloclastic methods using starch as a
substrate (e.g., turbidimetric and nephelometry
Chromolytic Techniques (Table 1, Methods 3a, 3b, and techniques) and the iodometric method have been
3c) displaced with the well-defined shorter glucosyl-chain
In recent years, commercial manufacturers have substrates. The use of the defined amylase substrates has
produced a variety of intriguing and convenient amylase improved the reaction stoichiometry and led to more
methods that depend on the liberation of a dye coupled consistent amylase measurement.
to polysaccharides such as amylopectin. Samples are
incubated with the dyed starch substrate, and after a The saccharogenic methods are reliable but time
fixed time, the amount of color of smaller dyed consuming and have high sample blanks and variable
saccharides is measured. substrates. A correction is required for the amount of
glucose originally in the patient sample. One can do this
Dry-film procedures are based on the hydrolysis of the by consuming all the glucose in the patients sample
dyed starch substrate. The released dyed saccharides in before performing the complete analysis or by using a
the spreading layer diffuse into an underlying reagent serum blank. Such blanks may be very high, rendering
layer, where they are measured by reflectance spectrophotometric measurements difficult. In general,
spectrophotometry. Unreacted dyed starch substrate that the reagents for these reactions are very complex and
is present in a spreading layer remains hidden from view therefore rather expensive. The reagents may also lose
(Method 3a). their effectiveness through a compromise among the
many coupled enzyme reactions, which must go on
A different method depends on the release of small, simultaneously. They rarely have true, zero-order
water-soluble fragments in such a way that the color can kinetics. The use of synthetic, defined substrates with the
be measured spectrophotometrically [7,8]. The substrate enzymatic methods allows for reproducible substrates
used in the method is p-nitrophenyl glucosides, such as and reaction products.
p-nitrophenyl-maltoheptaoside (G7PNP). The substrate is
hydrolysed to produce p-nitrophenol by amylase with Chromolytic techniques can be fast, precise, consistent,
coupled reaction of substrate fragments involving - and easy to perform. The methods employing p-
glucosidase. The reaction is monitored by measurement nitrophenyl-modified defined substrates are more
of the liberated p-nitrophenol at 405 nm (Method 3b). sensitive than the saccharogenic methods because of the
This method is also recommended by the International higher molar absorptivity. The p-nitrophenyl
Federation of Clinical Chemistry (IFCC) for amylase maltoheptaoside substrate has been shown to be equally
measurement (see method procedures in Reference and active with both salivary and pancreatic isoenzymes [9]
Preferred Methods). and to correlate very well with a saccharogenic method
with ultraviolet monitoring of the glucose reaction. This
An alternative substrate is 2-chloro-p-nitrophenyl -D- method showed no interference by hemoglobin, lipemia,
maltotrioside (CNP-G3), which can be hydrolysed to 2- or glucose.
chloro-p-nitrophenol (CNP) by amylase. The CNP is
detected spectrophotometrically and is proportional to The IFCC has a recommended method for amylase
the amylase activity in the sample. This method does not measurement by chromolytic technique using the
require glucosidases and is considered direct assay substrate of 4,6-ethylidene(G1)-p-nitrophenyl(G7)--D-
(Method 3c). maltoheptaoside at 30C [10]. Because most of the
procedures on the automated instruments in clinical
Historically the methods with starch as substrate were laboratories are performed at 37C, the amylase value of
used in amylase assay. Now most the methods are used the 30C reference method is not given a high level.
with the well-defined substrates, and these defined IFCC has developed the 37C reference procedures for
substrates have improved the reactive stoichiometry. amylase measurement with the same substrate of 4,6-
Chromolytic techniques use as substrates either p- ethylidene(G1)-p-nitrophenyl(G7)--D-maltoheptaoside
nitrophenyl-maltoheptaoside (G7PNP) or 2-chloro-p- as used in the reference procedure at 30C [11].
nitrophenyl -D-maltotrioside (CNP-G3), and
saccharogenic techniques use as substrates either Specimen
maltopentaose or maltotetraose. These are the current Either serum or heparinized plasma can be used as
methods used for amylase and are widely used on sample [12]. However, one study found that heparinized
automatic analyzers. plasma samples gave significantly higher results than
serum samples when measured using dry-film
Reference and Preferred Methods technology [13]. Whether this effect is due to the unique
A primary reference procedure for amylase measurement properties of the dry-film technology or whether heparin
at 37C was established by IFCC Scientific Division, affects amylase measurements in wet chemistry
Committee on Reference Systems for Enzymes (C-RSE) systems is not known. Because amylase has an absolute
using a chromolytic technique. The reference method is requirement for calcium ions, chelating anticoagulants
such as citrate, oxalate, and EDTA cannot be used to
152
Amylase
collect plasma to be used for amylase measurements. Macroamylase is the complex formed between amylase
Urine specimens with no preservatives in random or and immunoglobulins [20]. Amylase is readily filtered
timed collections are also valid specimens. by the glomerulus and is normally found in urine, but the
large molecular mass (greater than 200 kDa) of
Serum shows no loss of amylase activity for 4 days at macroamylase prevents its filtration through the
room temperature, for 2 weeks at 5C, for 1 year at glomerulus of the kidney and its appearance in urine.
28C, or for 5 years at 75C. Urine specimens should Patients with macroamylasemia usually have increased
be analyzed within 12 hours at room temperature or serum amylase activities and low or normal urine
within 5 days at 5C, and urine should not be frozen amylase activities. No clinical symptoms are associated
[10]. with the macroamylase disorder. The clinical importance
of macroamylase lies in its potential to create confusion
Interferences during the investigation of possible pancreatitis injury,
Amylase assays are generally not prone to interference and some cases of macroamylasemia have been found in
from hemoglobin, bilirubin, or triglycerides. Collection the investigation of pancreatitis injury.
of specimens in tubes containing oxalate, citrate, or
EDTA may result in falsely decreased values due to Numerous nonpancreatic abdominal diseases, such as
chelation of necessary amylase cofactors. appendicitis and peritonitis, can cause hyperamylasemia.
Salivary gland lesions caused by infection, irradiation,
Amylase Reference Interval obstruction, surgery, and tumor can increase the S-type
Reference intervals of amylase differ between the amylase, as well as total amylase level. Also, many
various available assays method because of differences drugs have been reported to cause an increase in serum
in substrates used and reagent preparations [14]. The amylase measurements [21]. The other common cause
reference interval of for hyperamylasemia is renal insufficiency. Patients with
amylase by the IFCC recommended method at 37C is renal insufficiency can show increases in amylase
31 to 107 U/L. activity up to five times the upper reference interval,
especially if the creatinine clearance is less than 50
Blood amylase activity of newborns is approximately mL/min.
18% of adults [12]. Mean serum amylase activity
increases from the neonatal period until adult levels are Amylase Performance Goals
achieved at approximately 3 to 4 years of age. There are The criteria for acceptable performance for amylase
no significant differences between males and females in assay by the Clinical Laboratory Improvement
the serum activity of amylase [12]. Amendments (CLIA) and College of American
Pathologists (CAP) proficiency program is within 30%
Interpretation of the mean value of laboratory peer groups. Amylase
Blood total amylase increases in acute pancreatitis and assays with different methods performed well in the
salivary gland inflammation disorders. Amylase clinical laboratories. The 2007 CAP
measurement is used primarily in the diagnosis of Chemistry/Therapeutic Drug Monitoring Survey (C-A)
pancreatitis. Following the onset of acute pancreatitis, shows that clinical laboratories use different instruments
amylase starts to rise in 5 to 8 hours and returns to and methods for amylase assay at 37C, and these
normal on the third or fourth day. Urine amylase also methods are developed by Abbott, Bayer, Beckman,
increases in acute pancreatitis, and it remains elevated Dade Behring, Roche, and Vitro. The coefficients of
for extended periods (7 to 10 days). variation (CVs) of each of these methods run in variable
numbers of laboratories are from 1.7% to 9.1% at
The lack of specificity of total amylase measurements different means. The average CV of these methods is
has led to the development of other tests (e.g., lipase and 5.4%.
pancreatic-specific amylase) instead of total amylase
activity for the diagnosis of acute abdominal pain. References
Studies have shown that total amylase measurement is 1 Smeaton JR, Marquardt HF. A reaction rate
inferior for the diagnosis of pancreatitis when compared nephelometer for amylase determinations. Clin
to these other markers [15-18]. Reported diagnostic Chem 1974; 20: 896.
efficiencies of amylase for acute pancreatitis range from 2 Liu TZ, Wei JS. Rapid laser nephelometric
54% to 83%, while reported sensitivities and determination of amylase activity in serum and
specificities range from 46% to 90% and 73% to 96%, urine. J Formosan Med Assoc 1991; 90: 217-
respectively [15-17]. Improved methods for 220.
measurement of lipase and pancreatic-specific amylase 3 Zinterhofer L, Wardlaw S, Jatlow P, Seligson
have been developed [19]. If acute pancreatitis is D. Nephelometric determination of pancreatic
suspected, serum amylase should be monitored in enzymes: I. Amylase. Clin Chim Acta 1973; 43:
conjunction with lipase or pancreatic-specific amylase. 5-12.
Serial measurements are generally more helpful than 4 Caraway WT. A stable starch substrate for the
single determinations in helping to establish a correct determination of amylase in serum and other
diagnosis. body fluids. Am J Clin Pathol 1959; 32: 97-99.
153
Amylase
5 Somogyi M. Modifications of two methods for 13 Doumas BT, Hause LL, Simuncak DM.
the assay of amylase. Clin Chem 1960; 6: 23- Differences between values for plasma and
35. serum in tests performed in the Ektachem 700
6 Kaufman RA, Tietz NW. Recent advances in XR analyzer, and evaluation of plasma
measurement of amylase activity: a comparative separator tubes (PST). Clin Chem 1989; 35:
study. Clin Chem 1980; 26: 846-853. 151-153.
7 Lorentz K. -Amylase assay: Current state and 14 Chen CT, Dineen H, Newton JD. Specificity of
future development. J Clin Chem Clin Biochem different substrates used in three amylase
1979; 17: 499-504. assays. Clin Chem 1988; 34: 1363-1364.
8 Okabe H, Uji Y, Netsu K, Noma A. Automated 15 Kazmierczak SC, Van Lente F, Hodges ED.
measurement of amylase with 4- Diagnostic and prognostic utility of
nitrophenylmaltoheptaoside as a substrate and phospholipase A activity in patients with acute
use of a selective amylase inhibitor. Clin Chem pancreatitis: comparison with amylase and
1984; 30: 1219-1222. lipase. Clin Chem 1991; 37: 356-361.
9 Rauscher E, Neumann U, Schaich E, von 16 Van Lente, F, Kazmierczak SC.
Bulow S, Wahlefeld AW. Optimized conditions Immunologically-derived pancreatic amylase,
for determining activity concentration of - pancreatic lipaase, and total amylase compared
amylase in serum with 1,4--D-4- as predictors of pancreatic inflammation. Clin
nitrophenylmaltoheptaoside as substrate. Clin Chem 1989; 35: 1542.
Chem 1983; 31: 14-19. 17 Kazmierczak SC, Van Lente F. Measuring
10 Lorentz K. Approved recommendation on IFCC carboxypeptidase A activity with a centrifugal
methods for the measurement of catalytic analyzer: analytical and clinical considerations.
concentration of enzymes, Part 9: IFCC method Clin Chem 1989; 35: 251-255.
for -amylase (1,4-a-D-glucan4- 18 Lott JA, Lu CJ. Lipase isoforms and amylase
glucanohydrolase, EC 3.2.1.1) Clin Chem Lab isoenzymes: assays and application in the
Med 1998; 36: 185-203. diagnosis of acute pancreatitis. Clin Chem
11 Schumann G, Aoki R, Ferrero CA, Ehlers G, 1991; 37: 361-368.
Feard G, Gella FJ, et al. IFCC primary 19 Apple F, Benson P, Preese L. Lipase and
reference procedures for the measurement of pancreatic amylase activities in tissues and in
catalytic activity concentrations of enzymes at patients with hyperamylasemia. Am J Clin
370C, Part 8: reference procedure for the Pathol 1991; 96: 610-614.
measurement of catalytic concentration of - 20 Fridhandler L, Berk JE. Macroamylasemia. Adv
amylase. Clin Chem Lab Med 2006; 44: 1146- Clin Chem 1978; 20: 267-28.
1155. 21 Young DS, Pestaner LC, Libberman V. Effects
12 Gillard BK, Simbala JA, Goodglick L. of drugs on clinical laboratory tests. Clin Chem
Reference ranges for amylase isoenzymes in 1975; 21: 255D-556D.
serum and plasma of infants and children. Clin
Chem 1983; 29: 1119-1123.
154
Amylase
Tables
Table 1: Methods of Amylase Analysis

Method 1: Amyloclastic
a. Turbidimetric or nephelometric. Decrease in turbidity or scattered light (nephelometry) of a starch solution is
directly related to amylase concentration.
b. Iodometric. Degradation of starch by amylase reduces the reaction of iodine with starch; reduction in iodine-
starch product (Amax = 660 nm) is inversely related to amylase activity.
Comments:
a. Rare; with consistent substrate, automated technique acceptable
b. Rare; can be easily adapted by most laboratories
Method 2: Saccharogenic
Glucose released from substrate is quantitated, usually by an enzymatic glucose assay procedure.
a. Maltopentaose as substrate:
Amylase
maltopentaose -------------------------------------> maltotriose + maltose
-Glucosidase
maltotriose + maltose ------------------------------> 5 glucose
Hexokinase
glucose + ATP --------------------------------------> glucose-6-phosphate + ADP
Glucose-6-Phosphate Dehydrogenase
glucose-6-phosphate + NAD+ --------------------> 6-phosphogluconate + NADH + H+
b. Maltotetraose as substrate:
Amylase
maltotetraose + H2O ------------------------------> 2 maltose
Maltose phosphorylase
maltose + phosphate ------------------------------> glucose + glucose-1-phosphate
-phosphoglucomutase
glucose-1-phosphate ------------------------------> glucose-6-phosphate
Glucose-6-phosphate dehydrogenase
glucose-6-phosphate NAD+ ---------------------> 6-phosphogluconate + NADH + H+
Comments:
Very common; can be easily adapted to many automated, discrete analyzers
Method 3: Chromolytic
a. Liberation of a dye coupled to a polysaccharide to dyed saccharide
Amylase
dyed polysaccharide -------------------> dyed saccharides
b. Liberation of chromogen from soluble, defined substrates (4,6-ethylidene-G7PNP)

Amylase
4,6-ethylidene-G7PNP + H2O -----------------> 4,6-ethylidene-Gx + G(7x) PNP
Glycosidase
G(7x) PNP + (7x) H2O ----------------------> (7x) glucose + PNP
Where 4,6-ethylidene-G7PNP is 4,6-ethylidene(G1)-p-nitrophenyl(G7)--D-maltoheptaoside; PNP is p-

nitrophenol
c. Liberation of chromagen from soluble, defined substrates (CNP-G3)

Amylase
CNP-G3 --------------> CNP + CNP-G2 + maltotriose + glucose
Where CNP-G3 is 2-chloro-p-nitrophenyl -D-maltotrioside
Comments:
a. Common
b. Common; IFCC recommended method
c. Common
155
Amylase
Procedure: IFCC Recommendation for the Determine the catalytic -glucosidase concentration
Measurement of Amylase [11] according to Appendix 1 of the reference article [11].
Reconstitute the lyophilized -glucosidase with a
Principal volume of the Diluent for Reagent
Amylase hydrolyzes 4,6-ethylidene(G1)-p- Enzymes to obtain a catalytic concentration of the
nitrophenyl(G7)--D-maltoheptaoside to different 4,6- reconstituted material of 16.9 mkat/L (1014 kU/L) at
ethylidene--D-glucopyranosyl oligosaccharides and p- 37C.
nitrophenyl--D glucopyranosyl oligosaccharides, and Freeze the enzyme solution in portions of 0.25 mL at
the latter is degraded to p-nitrophenyl by -glycosidase, 25C.
and p-nitrophenyl is detected at 405 nm. The chemical Stability at 25C: at least 6 months.
reaction scheme is in Method 3b of the Amylase
Methods summary table. Reaction Solution (Working Solution)
Mix 25 mL of Solution 2 with 0.25 mL of Solution 3.
Reagents Stability at 2C to 8C: 2 weeks.
1. N-2-hydroxyethylpiperazine-N-ethanesulfonic acid Starting Reagent Solution (Substrate Solution)

[HEPES] (C6H18N2O4S) 1.01 g (31.00 mmol/L) 4,6-ethylidene(G1)-4-
2. 4,6-Ethylidene(G1)-p-nitrophenyl(G7)--D- nitrophenyl(G7)-a-(1>4)-D-maltoheptaoside.
maltoheptaoside (C50H77NO38) 0.310 g (52.10 mmol/L) N-2-hydroxyethylpiperazine-N-
3. -Glucosidase (EC 3.2.1.20) ethanesulfonic acid.
4. Sodium chloride Dissolve in approximately 20 mL of water.
5. Calcium chloride dihydrate Adjust to pH (37C) 7.00 with 0.2 mol/L sodium
6. Sodium hydroxide solution (NaOH) 0.2 mol/L hydroxide solution.
7. Bovine serum albumin, fraction V, Mr = 68000 Transfer to a 25-mL volumetric flask.
8. Aqueous sodium chloride solution, 0.154 mol/L Equilibrate the volumetric flask and water to 20C.
Fill with water (20C) up to the calibration mark of the
Solution 1 volumetric flask.
6.14 g (417.5 mmol/L) calcium chloride, dihydrate. Stability at 2C to 8C: 2 weeks.
Dissolve in approximately 80 mL of water.
Transfer to a 100-mL volumetric flask. Final Concentrations of Reagents for the Measurement
Equilibrate the volumetric flask and water to 20C. of Amylase
Fill with water (20C) up to the calibration mark of the N-2-Hydroxyethylpiperazine-N-ethanesulfonic acid
volumetric flask. 50 mmol/L
Stability at 20C: 3 months. pH (37C) 7.00 0.03
4,6-Ethylidene(G1)-p-nitrophenyl(G7)--D-
Solution 2 maltoheptaoside 5 mmol/L
3.10 g (52.10 mmol/L) N-2-hydroxyethylpiperazine-N- Sodium chloride 70 mmol/L
ethanesulfonic acid. Calcium chloride 1 mmol/L
1.26 g (86.13 mmol/L) sodium chloride. -Glucosidase (37C) 135 kat/L (8100 U/L)
Dissolve in approximately 200 mL of water. Volume fraction of sample 1:31
Add 0.75 mL of Solution 1.
Adjust to pH (37C) 7.00 with 0.2 mol/L sodium Measurement Condition
hydroxide solution. Temperature 37.0C 0.1C
Transfer to a 250-mL volumetric flask. Wavelength 405 nm (1 nm)
Equilibrate the volumetric flask and water to 20C. Bandwidth <2 nm
Fill with water (20C) up to the calibration mark of the Light path length 10.00 0.01 mm
volumetric flask. Incubation time 60s s
Stability at 2C to 8C: 5 weeks. Delay time 180 s
Measurement interval 180 s
Diluent for Reagent Enzymes Reading (measurement points) >6
1.20 g bovine albumin.
0.900 g (154 mmol/L) NaCl. Measurement Procedure
Dissolve in approximately 80 mL of water. 2.000 mL Reaction Solution
Transfer to a 100-mL volumetric flask. Equilibrate to 37.0C
Equilibrate the volumetric flask and water to 20C. 0.080 mL Sample
Fill with water (20C) up to the calibration mark of the Mix thoroughly and incubate for 60 s. At the end of the
volumetric flask. incubation time, the temperature of the solution in the
Stability at 2C to 8C: at least 3 months. cuvette shall have reached 37.0C
Solution 3 0.400 mL Starting Reagent Solution

16.9 mkat/L (1014 kU/L) -glucosidase at 37C. Mix thoroughly, wait 180 s and monitor time and
absorbance for an additional 180 s
156
Amylase
Reagent and sample blank rate are also needed to

determine (see details in reference article [12])
Calculation
Amylase (kat/L) = A/t 1/ L Vt/Vs
Where A/t is absorbance change per second, is
molar absorption coefficient of p-nitrophenyl at 405nm,
L is light path length, Vt is total reaction volume, and Vs
is sample volume.
The amylase catalytic concentration in kat/L can be

converted to U/L by multiplication by the factor 60.
157
Angiotensin-Converting Enzyme (ACE)

Hassan M.E. Azzazy
Name: Angiotensin-converting enzyme (ACE)

Clinical significance: Refer to Chapter 28, Physiology of body water and electrolytes, in 5th edition
of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular mass: 150-180 kDa (somatic isoform); 90-110 kDa (testes isoform)
Chemical class: Metalloprotein
Principles of Analysis and Current Usage (Table 1). Most of these assays are based on the ability
Angiotensin-converting enzyme, or ACE (EC 3.4.15.1), of ACE to hydrolyze hippuryl-L-histidyl-L-leucine
is a zinc-metalloendopeptidase that plays an important (HHL) or furanacryloyl-L-phenylalanylglyclglycine
role in the control of blood pressure in humans through (FAPGG) substrates (Table 1, Methods 1, 3). The HHL
regulation of angiotensin II and bradykinin. ACE cleaves assay is based on the hydrolysis of HHL to hippuric acid
the C-terminal dipeptide of angiotensin I to produce the and HL. The hippuric acid formed is generally extracted
potent vasoconstrictor angiotensin II. ACE also and quantified at 228 nm [7]. The FAPGG assay was
inactivates bradykinin, a potent vasodilator peptide, by originally developed to measure ACE activity in serum
sequential removal of two C-terminal dipeptides [1]. and is based on hydrolysis of FAPGG to FAP and GG
Somatic ACE isoform is attached to the cell membrane [8]. Hydrolysis of FAPGG was assessed by measuring
and contains two highly homologous domains at the N- the decrease in absorbance at 340 nm. In addition, an
and C-terminal regions, each possessing an active site enzyme inhibitor binding assay for ACE has been
[2]. Testes ACE isoform is also membrane bound but described [9] that is based on specific binding of 125I-
contains only a single active site corresponding to the C- labeled ACE inhibitor (an analog of the ACE inhibitor
terminal somatic isoform [3]. Plasma ACE originates enalapril) to the active center of the enzyme (Table 1,
from proteolytic shedding from the cell membrane [4]. Method 2).
ACE is present in high activity at the lung vascular Erickson et al. [10] have reformatted a commercial
endothelial surface. ACE is also found in epithelial cells quantitative spectrophotometric CSF ACE assay for a
of renal proximal tubules, the GI tract, cardiac tissues, 96-well plate. The assay is based on the ability of ACE
and in the brain [5]. The wide tissue distribution of ACE, to hydrolyze the tripeptide N-[3-(2-furyl)acryloyl]-L-
where other elements of the renin-angiotensin system are phenylalanylglycylglycine to furylacryloylphenylalanine
not present, may suggest additional roles for ACE and glycylglycine. The assay, which utilizes only 115 L
besides angiotensin II production and bradykinin of CSF specimen, had a detection limit of 0.4 U/L and
inhibition [6]. upper reference value of 2.5 U/L.
Several analytical methods have been developed for A new assay has recently been described [11] for
determination of ACE activity, including screening of ACE inhibitory activity of potent food
spectrophotometry, fluorometry, radiometric assay, sources and drugs. The assay utilizes a new substrate 3-
radioimmunoassay, enzyme-linked immunosorbent hydroxybutyrylglycyl-glycyl-glycine. The substrate is
assay (ELISA), high-performance liquid cleaved by ACE and aminoacylase into amino acids and
chromatography (HPLC), and capillary electrophoresis 3-hydroxybutyric acid, which is measured
The serum ACE concentration has been reported to Reference and Preferred Methods
depend on the ACE gene insertion/deletion (I/D) There is no reference method for serum ACE. Most
mutation polymorphism. The D/D genotype is associated assays for determining ACE activity in serum employ
with a higher serum ACE concentration and may be HHL and FAPGG as substrates. The kinetic
associated with ischemic heart disease [12]. A PCR spectrophotometric method is probably the most
assay has been developed to detect the insertion itself, common method. Several versions of this method have
which is a 287-bp alu repetitive sequence located in been developed and adapted to various automated
intron 16 of the gene (chromosomal location of ACE instruments.
gene is 17q23) [13]. A real-time PCR assay that utilizes
fluorescent hybridization probes has been developed for
rapid genotyping of ACE gene I/D polymorphisms [14].
i Second edition: Not updated
i Fourth edition: Not updated
Angiotensin-converting enzyme (ACE)
Previous and current authors of this method: Fifth edition: Hassan M.E. Azzazy
Methods edition: Gerald Kessler
158
Specimen and oxalate. Triiodothyronine, acetate, nitrate, fluoride,

Serum is preferred. Blood is drawn from a fasting patient chloride, or bromide may cause elevated values.
in a red-top or a serum gel tube. After centrifugation,
separated serum is stable for 1 week at 4C and can be Reference Intervals
stored at 20C in a plastic vial for 6 months. Reference intervals for healthy individuals vary per the
method used to measure ACE. Furthermore, reported
Interferences reference intervals also varied even when the same
The use of ACE-inhibiting antihypertensive drugs such method (e.g., kinetic spectrophotometry) and essentially
as captopril, cilazapril, enalapril, lisinopril, perindopril, the same reagents were used on different instruments.
propranolol, ramipril, and trandolapril will decrease This underscores the urgent need to standardize different
ACE values. Other factors that will cause lower ACE ACE methods to establish common reference intervals.
results include hemolysis, lipemia, EDTA, heavy metals, Clearly, assay manufacturers can help by developing
appropriate enzyme reference materials.
Method Reference Intervals

Spectrophotometric assay (228 nm) 19 6 (SD) kU/L
Substrate: hippuryl-L-histidyl-L-leucine [15]
Fluorimetric assay 32.2 9.9 (SD) nmol/min/mL
Substrate: hippuryl-L-histidyl-L-leucine and o-
phthaldialdehyde [16]
Continuous spectrophotometric assay (340 nm) <2 yr: 5-83 U/L
Substrate: phenylalanyl-glycyl-glycine [17] 3-7 yr: 8-76 U/L
8-14 yr: 6-89 U/L
>14 yr: 8-52 U/L
Colorimetric assay (505 nm) <1 yr: 10.9-42.1 U/L
Substrate: hippuryl-L-histidyl-L-leucine with 4-amino- 1-2 yr: 9.4-36 U/L
antipyrine [18] 3-4 yr: 7.9-29.8 U/L
4-9 yr: 9.6-35.4 U/L
10-12 yr: 10.3-37 U/L
13-16 yr: 9-33.4 U/L
17-19 yr: 7.2-26.6 U/L
>19 yr: 6.1-21.1 U/L
Interpretation activity in the brain may be caused by a variety of

ACE (dipeptidyl carboxypeptidase; EC 3.4.15.1), also neurological disorders that are reflected by changes in
known as kininase II, catalyzes the conversion of ACE in the CSF [29]. Increased concentrations of ACE
angiotensin I to angiotensin II by cleavage of the C- have been associated with neurosarcoidosis [30] and
terminal dipeptide (histidyl-leucine) and also inactivates viral and bacterial meningitis. Low ACE levels are
bradykinin by cleavage of a dipeptidyl moiety [19]. ACE reported in patients with Alzheimers disease,
is found mainly on the luminal surface of pulmonary Parkinsons disease, and progressive supranuclear palsy
endothelial cells [20] but is also present in the circulation [31].
[21].
Recent evidence indicates that ACE activity in infancy
ACE has a key role in regulation of peripheral blood may contribute to the link between low birth weight and
pressure and is considered a major risk factor for later cardiovascular disease [32]. Because angiotensin II,
cardiovascular disease [22,23]. the main metabolite of ACE, has been shown to have
atherogenic effects [33], increased serum ACE activity
Although increased serum activity of ACE is not specific in infancy may identify individuals with increased
for sarcoidosis, ACE is used almost exclusively for cardiovascular disease later in life.
diagnosing and monitoring disease activity in sarcoidosis
[24]. ACE activity is also elevated in acute and chronic It should be noted that a homolog of ACE, termed
bronchitis, pulmonary fibrosis, rheumatoid arthritis, and ACE2, was discovered in 2000 [34]. It hydrolyzes
Gauchers disease. Elevated ACE activities were angiotensin I to Ang-(1-9) which is converted into Ang-
reported in patients with erectile dysfunction [25]. (1-7) by the action of two enzymes, neutral
Patients with anorexia nervosa showed a low serum endopeptidase and ACE. However, ACE2 releases Ang-
ACE activity that reverted to within the normal range (1-7) more efficiently than its catalysis of Ang-(1-9) by
following weight gain [26]. On the other hand, patients cleavage of Pro-Phe bound in Ang II. Thus Ang-(1-7) is
with advanced lung neoplasms had significantly lower the major biologically active product of ACE2 and is
levels of ACE as compared to controls [27]. considered to be a beneficial peptide of the RAS cascade
in the cardiovascular system. ACE2 has 42% identity
ACE has also been associated with an endogenous renin- with the catalytic domain of ACE, is present in most
angiotensin system in the brain [28]. Changes in ACE cardiovascular-relevant tissues, and is an ectoenzyme as
159
ACE. Unlike ACE, ACE2 has only one catalytic site and assay for angiotensin-converting enzyme in
is insensitive to ACE inhibitors. ACE2 has vasodilatory cerebrospinal fluid. Clin Chem 2003;49:970-2.
and antiproliferative effects of Ang-(1-7) in the heart and 11 Lam le H, Shimamura T, Sakaguchi K, Noguchi
kidney [35]. K, Ishiyama M, Fujimura Y, Ukeda H. Assay of
angiotensin I-converting enzyme-inhibiting
Performance Goals activity based on the detection of 3-
Desirable imprecision for ACE as derived from hydroxybutyric acid. Anal Biochem
biological variation is 0.1% [36]. The current Clinical 2007;364:104-11.
Laboratory Improvement Amendments (CLIA) 12 Lindpaintner K, Pfeffer MA, Kreutz R,
performance goal for measurement of ACE is for Stampfer MJ, Grodstein F, LaMotte F, Buring
laboratories to be within 3 standard deviations of the J, Hennekens CH. A prospective evaluation of
peer-group mean. According to the 2007 College of an angiotensin-converting-enzyme gene
American Pathologists (CAP) Survey, coefficients of polymorphism and the risk of ischemic heart
variation (CVs) for all ACE reagents at mean values of disease. N Engl J Med 1995;332:706-11.
48.6 (SD 6.9) and 99.2 (SD 13.1) U/L were 14.2% and 13 Rigat B, Hubert C, Corvol P, Soubrier F. PCR
13.2%, respectively [37]. Despite the acceptable detection of the insertion/deletion
precision of current ACE testing, standardization is polymorphism of the human angiotensin-
needed to reduce variations among different methods converting enzyme gene (DCP1) (dipeptidyl
and to develop common reference intervals. carboxypeptidase 1). Nucleic Acids Res
1992;20:1433.
References 14 Somogyvari F, Szolnoki Z, Marki-Zay J, Fodor
1 Yang HY, Erds EG, Levin Y. A dipeptidyl L. Real-time PCR assay with fluorescent
carboxypeptidase that converts angiotensin I hybridization probes for exact and rapid
and inactivates bradykinin. Biochim Biophys genotyping of the angiotensin-converting
Acta 1970;214:374-6. enzyme gene insertion/deletion polymorphism.
2 Soubrier F, Alhenc-Gelas F, Hubert C, Clin Chem 2001;47:1728-9.
Allegrini J, John M, Tregear G, Corvol P. Two 15 Ryder KW, Jay SJ, Jackson SA, Hoke SR.
putative active centers in human angiotensin I- Characterization of a spectrophotometric assay
converting enzyme revealed by molecular for angiotensin-converting enzyme. Clin Chem
cloning. Proc Natl Acad Sci USA 1981;27:530-4.
1988;85:9386-90. 16 Friedland J, Silverstein E. A sensitive
3 Ehlers MR, Fox EA, Strydom DJ, Riordan JF. fluorimetric assay for serum angiotensin-
Molecular cloning of human testicular converting enzyme. Am J Clin Pathol
angiotensin-converting enzyme: the testis 1976;66:416-24.
isozyme is identical to the C-terminal half of 17 Holmquist B, Bunning P, Riordan JF. A
endothelial angiotensin-converting enzyme. continuous spectrophotometric assay for
Proc Natl Acad Sci USA 1989;86:7741-5. angiotensin-converting enzyme. Anal Biochem
4 Beldent V, Michaud A, Wei L, Chauvet MT, 1979;95:540-8.
Corvol P. Proteolytic release of human 18 Kasahara Y, Ashihara Y. Colorimetry of
angiotensin-converting enzyme. Localization of angiotensin-I converting enzyme activity in
the cleavage site. J Biol Chem 1993;268:26428- serum. Clin Chem 1981;27:1922-5.
34. 19 Erdos EG. Angiotensin I converting enzyme.
5 Zhuo J, Moeller I, Jenkins T, Chai SY, Allen Circ Res 1975;36:247-55.
AM, Ohishi M, Mendelsohn FA. Mapping 20 Ryan JW, Ryan US, Schultz DR, Whitaker C,
tissue angiotensin-converting enzyme and Chung A. Subcellular localization of pulmonary
angiotensin AT1, AT2 and AT4 receptors. J angiotensin-converting enzyme (kininase II).
Hypertens 1998;16:2027-37. Biochem J 1975;146:497-9.
6 Dzau VJ. Circulating versus local renin- 21 Ondetti MA, Cushman DW. Enzymes of the
angiotensin system in cardiovascular renin-angiotensin system and their inhibitors.
homeostasis. Circulation 1988;77:I4-13. Annu Rev Biochem 1982;51:283-308.
7 Cushman DW, Cheung HS. Spectrophotometric 22 Dahlof B, Devereux RB, Kjeldsen SE, Julius S,
assay and properties of the angiotensin- Beevers G, de Faire U et al. LIFE Study Group.
converting enzyme of rabbit lung. Biochem Cardiovascular morbidity and mortality in the
Pharmacol 1971;20:1637-48. Losartan Intervention For Endpoint reduction in
8 Harjanne A. Automated kinetic determination hypertension study (LIFE): a randomised trial
of angiotensin-converting enzyme in serum. against atenolol. Lancet 2002;359:995-1003.
Clin Chem 1984;30:901-2. 23 Brunner HR, Gavras H. Angiotensin blockade
9 Fyhrquist F, Tikkanen I, Gronhagen-Riska C, for hypertension: a promise fulfilled. Lancet
Hortling L, Hichens M. Inhibitor binding assay 2002;359:990-2.
for angiotensin-converting enzyme. Clin Chem 24 Lieberman J. Enzymes in sarcoidosis.
1984;30:696-700. Angiotensin-converting-enzyme (ACE). Clin
10 Erickson JA, Cousin R, Wu JT, Ashwood ER. Lab Med 1989;9:745-55.
Quantitative spectrophotometric microplate
160
25 Hamed EA, Meki AR, Gaafar AA, Hamed SA. 34 Donoghue M, Hsieh F, Baronas E, Godbout K,
Role of some vasoactive mediators in patients Gosselin M, Stagliano N et al. A novel
with erectile dysfunction: their relationship with angiotensin-converting enzyme-related
angiotensin-converting enzyme and growth carboxypeptidase (ACE2) converts angiotensin
hormone. Int J Impot Res 2003;15:418-25. I to angiotensin 1-9. Circ Res 2000;87:E1-9.
26 Matsubayashi S, Tamai H, Kobayashi N, 35 Burrell LM, Johnston CI, Tikellis C, Cooper
Takaichi Y, Fukata S, Hirota Y et al. ME. ACE2, a new regulator of the renin-
Angiotensin-converting enzyme and anorexia angiotensin system. Trends Endocrinol Metab
nervosa. Horm Metab Res 1988;20:761-4. 2004;15:166-9.
27 Ashutosh K, Keighley JF. Diagnostic value of 36 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
serum angiotensin-converting enzyme activity Hernandez A, Jimenez CV et al. Current
in lung diseases. Thorax 1976;31:552-7. databases on biological variation: pros, cons
28 Phillips MI, Weyhenmeyer J, Felix D, Ganten and progress. Scand J Clin Lab Invest
D, Hoffman WE. Evidence for an endogenous 1999;59:491-500.
brain renin-angiotensin system. Fed Proc 37 College of American Pathologists. 2007 Survey
1979;38:2260-6. Participant Summary Report. Northfield, IL:
29 Schweisfurth H, Schioberg-Schiegnitz S. Assay CAP; 2007.
and biochemical characterization of 38 Shihabi ZK, Scaro J. Liquid-chromatographic
angiotensin-I-converting enzyme in assay of angiotensin-converting enzyme in
cerebrospinal fluid. Enzyme. 1984;32:12-9. serum. Clin Chem 1981;27:1669-71.
30 Jones DB, Mitchell D, Horn DB, Edwards CR. 39 Buttery JE, Chamberlain BR. A scheme for
Cerebrospinal fluid angiotensin-converting determining the correct activity of the kinetic
enzyme levels in the diagnosis of angiotensin-converting enzyme. Clin Chem
neurosarcoidosis. Scot Med J 1991;36:144-5. 1985;31:645-6.
31 Zubenko GS, Volicer L, Direnfeld LK, 40 Alves MF, Araujo MC, Juliano MA, Oliveira
Freeman M, Langlais PJ, Nixon RA. EM, Krieger JE, Casarini DE et al. A
Cerebrospinal fluid levels of angiotensin- continuous fluorescent assay for the
converting enzyme in Alzheimers disease, determination of plasma and tissue angiotensin
Parkinsons disease and progressive I-converting enzyme activity. Braz J Med Biol
supranuclear palsy. Brain Res 1985;328:215-21. Res 2005;38:861-8.
32 Forsyth JS, Reilly J, Fraser CG, Struthers AD. 41 Danilov SM, Deinum J, Balyasnikova IV, Sun
Angiotensin-converting enzyme activity in ZL, Kramers C, Hollak CE, Albrecht RF.
infancy is related to birth weight. Arch Dis Detection of mutated angiotensin I-converting
Child Fetal Neonatal Ed 2004;89:F442-4. enzyme by serum/plasma analysis using a pair
33 Scholkens BA, Landgraf W. ACE inhibition of monoclonal antibodies. Clin Chem
and atherogenesis. Can J Physiol Pharmacol 2005;51:1040-3.
2002;80:354-9.
161
Table 1: Selected ACE Methods.

Method 1. Fluorometry [16]
Principle of analysis: Hydrolysis of L-Hip- L-His-L-Leu (HHL)
Detection: Measurement of the fluorescence of the ortho-phthaldialdehyde-His-Leu adduct
Comment: Hydrolysis of His-Leu by dipeptidases present in specimens may lower ACE activity by
destroying the fluorescence-emitting compound.
Method 2. Enzyme inhibition [9]
Principle of analysis: Binding of a radiolabeled specific inhibitor (compound 351A, an analog of
enalapril) to the active center of ACE
Detection: Count radioactivity
Comment: 125I-labeled ACE inhibitor (a p-hydroxybenzamidine derivative of N-(1-carboxy-3-
phenylpropyl)-L-lysyl-L-proline)
Method 3. Liquid chromatography [38]
Principle of analysis: This assay is based on separation and quantification of hippuric acid (under
conditions of reverse phase) released from HHL by serum ACE.
Detection: The column effluent is monitored at 254 nm.
Comment: This method may be performed on the same column as that used for theophylline assay,
using the same solvents and without an evaporation step.
Method 4. Kinetic spectrophotometry [39]
Principle of Analysis: Hydrolysis of N-[3-(2-furyl)acryloyl]-L-phenylalanyl-
glycylglycine (FAPGG), to N-[3-(2-furyl)acryloyl]-L-phenylalanine (FAP) and glycylglycine (GG)
Detection: Measure change in absorbance at 340 nm (or 345 nm). The change in absorbance is
converted into enzyme activity by a factor derived from the absorbance change produced by hydrolysis
of the substrate or a concentration of 1 mmol/L.
Comment: FAPGG absorbs more strongly than FAP at 340 nm; GG does not absorb at 340 nm.
Method 5. Continuous fluorescent assay [40]
Principle of analysis: Hydrolysis of internally quenched fluorescent peptides (e.g., orthoaminobenzoic
acid-Phe-Arg-Lys-2,4-dinitrophenyl-P-OH).
Detection: Continuous recording of the fluorescence (ex = 320 nm; em = 420 nm)
Comment: ACE cleaves the above peptide substrate at the Arg-Lys bond.
Method 6. Immunocapture enzyme assay (ICEA) [41]
Principle of analysis: ACE is captured by monoclonal antibodies and its activity determined by a
fluorometric method using HHL as a substrate.
Detection: Measurement of the fluorescence of the ortho-phthaldialdehyde-His-Leu adduct
Comment: This method is used for detection of mutated ACE in plasma and may be useful for
differential diagnosis of increased ACE activity.
162
Anion Gap
Anion Gap
Tony Badrick
Name: Anion Gap
Clinical significance: Refer to Chapter 28, Physiology of Body Water and Electrolytes, in the 5th
edition of Clinical Chemistry: Theory, Analysis, Correlation.
Computation formula: Na+ (Cl + CO2) or (Na+ + K+) (Cl + CO2)
Principles of Analysis and Current Usage and has a typical reference interval of 7 to 16 mmol/L if
The anion gap is the difference between the measured using Formula 1, or 10 to 20 mmol/L if potassium is
cations and the measured anions and is an estimate of included in the calculation, as in Formula 2.
unmeasured anions, based on the electroneutrality of
i It can be seen from the above that an increased anion gap
total ions in solution
. The major anions and cations in plasma are: would occur in the following situations: (1) an increase
in the unmeasured anions (protein, sulphate, phosphate,
Cations mEq/L Anions mEq/L organic acids) listed above; (2) the presence of an
abnormal anion such as a drug; and (3) a decrease in the
Sodium 140 Chloride 100 unmeasured cations magnesium and calcium.
Potassium 4 Bicarbonate 27
Calcium 5 Protein 15 The anion gap may be useful clinically in the
Magnesium 1 Phosphate 2 investigation of a metabolic acidosis which can be
Sulphate 1 classified as having either an elevated anion gap or a
Organic acids 5 normal anion gap.
Total 150 150
Although we are actually looking at charge There are no reference or preferred methods; this is a
measurements in mEq/L, these are equivalent to calculation.
millimoles for the anions and cations used in the
calculation. For consistency, this terminology is used Specimen
hereafter. Note that the mEq/L of protein above does not Serum or plasma.
equate to mmol/L of protein.
If all of these anions and cations were measured, there Reference Intervals
would be no gap; however, usually only the major Reference intervals differ for different patient
ionsnamely sodium, chloride, bicarbonate, and populations. The reference interval for ambulatory
potassiumare measured. Calculations of the anion gap patients is 7 to 18 mmol/L and for hospitalized patients,
may be made in two ways: with and without potassium. 6 to 10 mmol/L. However, the reference interval was
Because potassium-concentration changes are relatively reviewed more recently by Paulson et al. [2], who
small compared to the other three analytes, it is often showed that the true mean anion gap was about 7
dropped from calculations [1]. mmol/L rather than the 12 mmol/L commonly cited [3].
In addition, they showed that there was a variance of
The anion gap is calculated as: about 7 mmol/L between instruments, making these
values instrument dependent.
(1) Measured cations (Na) measured anions (Cl + CO2)
or Interpretation
(2) Measured cations (Na + K) measured anions (Cl + There is debate about the value of the anion gap, both
CO2), clinically in the investigation of metabolic acidoses and
as a quality-control tool [1,4,5].
The main clinical utility of the anion gap is in classifying

i
Anion gap metabolic acidosis as either an increased anion gap,
Previous and current authors of this method: associated with a metabolic acidosis, or a normal gap,
First edition: W. Gregory Miller associated with hyperchloremic acidosis [6]. Some
Methods edition: P. Phillip Anderson and W. Gregory believe that the finding of an elevated anion gap is
Miller virtually synonymous with a metabolic acidosis [7] and
Second edition: P. Phillip Anderson and W. Gregory that the determination of an anion gap is mandatory
Miller whenever one is facing a suspected metabolic acidosis of
Third edition: P. Phillip Anderson and W. Gregory undetermined cause [8].
Miller
Fourth edition: P. Phillip Anderson and W. Gregory However, Gabow et al. [5] have shown that only when
Miller the anion gap is greater than 30 mmol/L is an organic
Fifth edition: Tony Badrick
163
Anion Gap
acidosis assured. Even at anion gaps of 20 to 29 mmol/L,

approximately one third of patients did not have an Because of these analytical considerations, the anion gap
organic acidosis. is too insensitive to be clinically useful in the
investigation of metabolic acidosis, and the direct
Using the anion gap as a screen for lactic acidosis has measurement of lactate and/or betahydroxybutyrate are
been shown to be insensitive, with normal anion gaps more useful options to follow. It also has limited value
present in patients with lactates as high as 9.9 mmol/L as a quality-control tool because it is the sum of three (or
[9]. four) measurements, and error in any one of these cannot
be directly identified; an error in two or more may
Low anion gaps usually suggest an analytical failure, cancel each other out. Use of the anion gap cannot
dilution, hypoproteinemia (usually hypoalbuminemia, replace individual quality-control programs for its
since albumin represents approximately 75% of the component analytes.
unmeasured anions in the anion gap), or multiple
myeloma, although the clone of immunoglobulin is References
important because IgG tends to be cationic, whereas IgA 1 Natelson S. On the significance of the
tends to be anionic at pH 7.4 [10]. expression anion gap. Clin Chem 1983; 29:
282-283.
Goldstein et al. [11] have shown that in the majority of 2 Paulson WD, Roberts WL, Lurie AA, Koch
cases, low anion gaps are not reproducible and suggest a DD, Butch AW, Aguanno JA. Wide variation in
spurious result. Samples lose CO2 while waiting to be serum anion gap measurements by chemistry
analyzed, thus the CO2 concentrations fall, and the anion analyzers. Am J Clin Pathol 1998; 110: 735-
gap is increased. Patients who actually have negative or 742.
low anion gaps, such as those with respiratory acidosis, 3 Witte DL, Rodgers JL, Barrett DA. The anion
may appear to have normal anion gaps and thus could be gap: its use in quality control. Clin Chem 1976;
missed. However, since the CO2 decreases only about 4 22: 643-646.
mmol/L in the first hour in an uncapped tube [12], the 4 Badrick T, Hickman PE. The anion gap: a
problem of missing low or negative anion gaps may not reappraisal. Am J Clin Pathol 1992; 98: 249-
be severe. 252.
5 Gabow PA, Kaehny WD, Fennessy PV.
The anion gap is not meant to be a total quality-control Diagnostic importance of an increased serum
program for electrolyte analyses, but rather should be anion gap. N Engl J Med 1980; 303: 854-858.
considered a supplement to a quality-control program for 6 Gabow PA. Fluids and electrolytes: clinical
the individual analyses. In fact, it has been shown that a problems and their solutions. Boston: Little,
combination of simple control rules used with the Brown & Co; 1983; pp. 23-42.
individual components (sodium, chloride, and 7 Emmett ME, Narins RG. Clinical use of the
bicarbonate) is more powerful than using anion gaps anion gap. Medicine 1977; 56: 38-54.
[13]. 8 Jacobsen D, Bredensen JE, Eide L, Ostborg J.
Anion and osmolal gaps in the diagnosis of
Causes of a metabolic acidosis with an elevated anion methanol and ethylene glycol poisoning. Acta
gap include ketoacidosis, lactic acidosis, renal failure Med Scand 1982; 212: 17-20.
(due to retention of phosphoric, sulfuric, and organic 9 Iberti TJ, Leibowitz AB, Papadakos PJ, Fischer
acids) and ingestions of compounds that are metabolized EP. Low sensitivity of the anion gap as a screen
to an organic acid (ethanol, methanol, salicylate, to detect hyperlactemia in critically ill patients.
ethylene glycol). Crit Care Med 1990; 18: 275-277.
10 Moe OW, Fuster D. Clinical acid-base
Metabolic acidosis with a normal anion gap occurs with pathophysiology: disorders of plasma anion
a loss of bicarbonate (diarrhea) or inability to secrete gap. Best Practice &Research Clinical
hydrogen ion through the renal tubules (renal tubular Endocrinology & Metabolism 2003; 17: 559-
acidosis). 574.
11 Goldstein RJ, Lichtenstein, NS, Souder D. The
These various causes of abnormal anion gaps are myth of the low anion gap. JAMA 1980; 96:
summarized in Table 1. 874-876.
12 Gambino RS, Schreiber H. The measurement of
Anion Gap Performance Goals CO2 content with the autoanalyzer. Am J Clin
The imprecision of the anion gap is determined by Pathol 1966; 45: 406-411.
summing the imprecision of the individual components 13 Cembrowski GS, Westgard JO, Kurtycz DF.
of the calculation, that is SD2anion gap = SD2Na + SD2Cl + Use of anion gap of the quality control of
SD2HCO3. Thus the typical value of the SD is electrolyte analysers. Am J Clin Pathol 1983;
approximately 2.7 mmol/L. If the true anion gap is 10 79: 1697-1698.
mmol/L, there is a 95% chance that the calculated value
will lie between 4.6 and 15.4 mmol/L (mean plus or
minus 2 SD).
164
Anion Gap
Table 1: Causes of Changes in the Anion Gap

Causes of Increased Anion Gap Causes of Decreased Anion Gap
Decreased unmeasured cations: Decreased unmeasured anions:
Hypocalcemia Hypoalbuminemia
Hypomagnesemia Hypophosphatemia
Increased unmeasured anions: Increased unmeasured cations:
Associated with metabolic acidosis: Hypercalcemia
Uremia (renal failure) Hypermagnesemia
Ketoacidosis Paraproteins
Lactic acidosis Polyclonal gamma globulins (IgG)
Salicylate poisoning Drugs such as polymyxin B or lithium
Ethanol Underestimation of serum sodium:
Methanol formic acid Hyperproteinemia (myeloma protein)
Ethylene glycol glycolic acid Hypertriglyceridemia (turbidity)
Paraldehyde Overestimation of serum chloride:
Pyroglutamate Bromism
Not necessarily associated with Turbidity (for ferric thiocyanate
metabolic acidosis: method)
Hyperphosphatemia
Hypersulfatemia
Large doses of antibiotics (such as
penicillin and carbenicillin
Presence of IgA myeloma protein
Treatment with lactate, citrate, or
acetate
Increase in net protein charge as in alkalosis
Increase in bromide leading to renal loss of
chloride (pyridostigmine bromide)
165
Anticonvulsant Drugs
i
Paul Salm
Paul J. Taylor
Julia M. Potter
Name: Valproic acid Phenobarbital Phenytoin

Clinical significance: Refer to Chapter 42, Nervous System, in the 4th Edition of Clinical Chemistry:
Molecular formula: C8H16O2 C12H12N2O3 C15H12N2O2
Molecular weight: 144.21D 232.23D 252.26D
Merck Index: 9574 7032 7130
Chemical class: Carboxylic acid Barbiturate Hydantoin
Structure: click here click here click here
do not prepare derivatives of the anticonvulsant drugs. viref

Principles of Analysis and Current Usage 1
However, of those procedures that use derivatization,
Most anticonvulsant drugs have specific ultraviolet
spectral characteristics (Anticonvulsant Drugs: Figure 1) methylation is most commonly used. viiref In the past,
that were exploited in the initial attempts to monitor the column support materials most frequently used were acid-
ii washed silanized materials (e.g., Gas Chrom-Q,
therapeutic levels of these drugs. ref The drugs were
Chromosorb-W, and Supelcoport). Nowadays fused-silica
extracted into an organic solvent and reextracted into
capillary columns are preferred because of their high
aqueous solution, and the absorbance of the solution
temperature stability and inertness, which not only removes
recorded (method 1, Methods Summary Table). By use of
the need for derivatization, but offers high sensitivity,
the appropriate solvents, the drugs could be extracted with
relative specificity. By recording the difference spectra of selectivity, and speed. viiiref In the method described by
the extract at two pH values, one could also increase the Volmut et al. ixref, six antiepileptic drugs may be analyzed
specificity. The difference spectra were obtained by use of a simultaneously by capillary gas chromatography.
dual-beam spectrophotometer and placement of one solution High-performance liquid chromatography assays
into the reference compartment and the other into the usual (HPLC) require some form of sample preparation before
test compartment. separation on a column (method 4, Methods Summary
A second method for measuring anticonvulsant Table). Sample preparation usually involves either organic
drugs involved extraction and separation by thin-layer solvent extraction of the drugs or protein precipitation.
iii
chromatography. ref Quantitation was achieved by either Nearly all methods use reverse-phase chromatography with
x
ultraviolet scanning of the plate or by elution of the drug ultraviolet absorption spectrometry for detection. ref
from the plate and recording of its ultraviolet absorption Immunological procedures (methods 5a to 5i,
(method 2, Methods Summary Table). Methods Summary Table) developed for the measurement
Gas chromatographic techniques (GC) were for of anticonvulsants include radioimmunoassay (RIA),
many years the primary technique used to analyze enzyme immunoassay (EIA), fluorometric enzyme
antiepileptic drugs in biological fluids (method 3, Methods immunoassay (FEIA), fluorescence immunoassay (FIA),
iv
Summary Table). ref In a review by Rambec and Meijer fluorescence polarization immunoassay (FPIA),
vref nephelometric inhibition immunoassay (NIA), cloned
, flame ionization detection (FID) or the more sensitive
enzyme donor immunoassay (CEDIA), and dry-film
nitrogen selective detection (NSD) systems were the most
multilayer immunoassay (DFMI).
widely used. NSD may have considerable advantage over
The RIA procedures (method 5a, Methods
FID in terms of specificity and sensitivity, since almost all
Summary Table) involve the standard competitive binding
antiepileptics except valproic acid contain nitrogen.
procedure in which there is competition between the labeled
However, NSD requires more frequent and careful
drug with that from the sample for a limited amount of
maintenance. Approximately half of the gas xi
chromatographic methods described by Rambec and Meijer antibody. ref
The introduction of the enzyme-multiplied
immunoassay technique (EMIT) into the clinical chemistry
11
This chapter was originally prepared by: laboratory in the 1970s made possible the routine
Paul Salm monitoring of all the anticonvulsant drugs (method 5b,
Paul J. Taylor Methods Summary Table). This homogeneous enzyme
and Julia M. Potter immunoassay is based on the competitive protein-binding
166
technique, with an enzyme as the label and an antibody as polarization (method 5e, Methods Summary Table). The
the binding protein. When the enzyme-labeled drug fluorescent tracer consists of fluorescein covalently attached
becomes bound to an antibody to the drug, the activity of to the drug being analyzed. When excited with polarized
the enzyme is reduced. Free drug in the sample competes light, this tracer will emit light with a degree of polarization
with the enzyme-labeled drug for binding to the antibody, that is inversely proportional to its rate of rotation; the more
decreasing the amount of antibody-bound labeled drug, and slowly the tracer molecule rotates, the greater the
thereby decreases the antibody-induced inactivation of the polarization signal of the emitted light. When a specific
enzyme. The enzyme activity then correlates with the antibody binds to the tracer, its rotation is slowed and the
concentration of the free drug. emitted light increases in polarization. It is a competitive
Fluorometric enzyme immunoassay (FEIA) is a binding assay in which a specific fluorescein-labeled
competitive binding method that is based on the radial anticonvulsant competes with endogenous drug from a
partition immunoassay principle that combines solid phase patient sample for a limited amount of antibody. The
immunological techniques and radial chromatography amount of bound labeled anticonvulsant is inversely
xii
(method 5c, Methods Summary Table). ref In FEIA the proportional to the drug in the patient sample. The intensity
drug of interest in the clinical sample is premixed with of the polarized fluorescence is proportional to the amount
enzyme-labeled conjugate and spotted on to a glass-fiber of antibody-bound labeled anticonvulsant. The greater the
filter paper which contains immobilized antibody at the polarized fluorescence, the smaller the amount of drug in
analysis site. These two constituents then compete for the patient sample.
antibody binding sites. After a suitable incubation period, Another method for the measurement of phenytoin
unbound label is removed from the center of the reaction and phenobarbital employs a labeled prosthetic group. In
zone through radial elution by applying a wash solution. this assay, a prosthetic group necessary for the activity of
This solution contains the substrate for the enzyme that the enzyme is covalently linked to the drug to be measured:
initiates enzymatic activity simultaneously with the wash. for example, phenytoin linked to flavine adenine
The enzymatic rate of the bound fraction at the center of the dinucleotide (FAD) (method 5f, Methods Summary Table).
reaction zone is quantified by front surface fluorescence and The FAD is necessary for the enzyme apoglucose oxidase to
xv
is inversely proportional to the concentration of drug present function. ref If a limited amount of antibody to the drug is
in the sample. xiiiref present, the drug will bind to the antibody. In the presence
The principles of competitive protein binding to of drug in the patient sample, less FAD-labeled drug will
measure levels of these drugs in serum or plasma have also bind to the antibody, making more FAD available to bind to
xiv
been used. ref (method 5d, Methods Summary Table). the apoprotein. The enzyme is then activated, and the
With this method the anticonvulsant drug is labeled with a amount of activity is proportional to the amount of drug in
derivative of the fluorogenic enzyme substrate umbelliferyl- the patient sample. The enzyme converts glucose and O2 to
b-d-galactoside. This fluorogenic drug reagent (FDR) gluconolactone and H2O2. The color reaction is generated
remains nonfluorescent under the conditions of the assay, by use of peroxidase and a dye substrate 3,3' ,5,5'-
until hydrolysis catalyzed by b-alactosidase yields the tetramethylbenzidine, which, when oxidized, forms a blue
fluorescent product. When antibody to the anticonvulsant colored product. In this assay, the patient serum or plasma
drug reacts with the FDR, the FDR is prevented from acting sample is first diluted with distilled water, and a specific
as a substrate for the b-galactoside. Competitive binding volume of the prepared specimen is placed on the reagent
reactions are set up with a constant amount of FDR, a area of the strip. The strip is then placed in the instrument, a
limiting amount of antibody, and a patient sample solid-phase reflectance photometer. The strip color is at 740
containing the anticonvulsant, as follows: nm, and the results are calculated from a calibration curve.
The reaction sequence is as follows:
1. FDR + Antibody + Anticonvulsant
2. AntibodyFDR + Antibodynticonvulsant + FDR +
Anticonvulsant
3. FDR b-Galactosidase Umbelliferone + Galactose
The anticonvulsant in the sample competes with

the FDR for the limited number of antibody-binding sites.
The FDR not bound to antibody is hydrolyzed by -
galactosidase to produce the fluorescent product. Therefore
the fluorescence produced is proportional to the
anticonvulsant concentration in the sample. The intensity of
the fluorescence is related to the sample anticonvulsant drug
concentrations by means of a standard curve.
Fluorescence polarization immunoassay (FPIA)
uses the principle of competitive binding assay and
measures binding of tracer directly by fluorescence
167
A rate nephelometric inhibition immunoassay is fluorescent intensity measured is inversely proportional to

also commonly used (method 5g, Methods Summary the antigen concentration in the serum sample.
Table). In this assay the anticonvulsant is covalently linked
to a protein. Drug from the patients sample competes with
this
protein-bound drug for binding to an antibody. The rate of The ideal analytical procedure for the
xvi
change of light scatter is measured by the instrument. ref anticonvulsant drugs should be as follows:
A homogeneous immunoassay technique, cloned 1. Both accurate and precise, affording the reliable
enzyme donor immunoassay (CEDIA), is a unique method quantitation of the antiepileptic drugs.
that utilizes genetically engineered b-galactosidase enzyme 2. Technically easy, so that little effort is required in
fragments (method 5h, Methods Summary Table). The training to perform the task.
CEDIA principle is based on the use of two inactive b- 3. Rapid, to minimize delay in reporting results to
galactosidase fragments which have been constructed using clinician. Many requests for drug quantitation arise as a
recombinant DNA techniques. One of these is designated result of problems in patient management.
the enzyme donor (ED), a small amino-terminal peptide, 4. Sensitive, to allow analysis for drugs in a
and the other is the enzyme acceptor (EA), a larger protein microscale sample and permit simultaneous analysis of the
xvii
that has residues deleted near the amino terminus. ref The most commonly used drugs. Many patients with epilepsy
EA and ED each on their own show no enzymatic activity; are on multiple-drug regimens.
however, active b-galactosidase forms when the fragments Unfortunately, no method meets all these
recombine spontaneously. The formation of this active requirements, and the decision on which procedure to adopt
enzyme can be measured spectrophotometrically by the depends largely on the laboratory conditions and on
hydrolysis of a chromogenic substrate such as o- personal preferences of those involved.
nitrophenyl-b-d-galactopyranoside. A ligand molecule, such Spectrophotometric methods were initially used for
as a drug or hormone, can be covalently linked to the ED the measurement of anticonvulsant drug concentrations in
xx
peptide forming a EDligand conjugate, such that the body fluids. ref Although differential extraction
enzyme complementation reaction is not affected. Adding a procedures were developed to overcome interference by
ligand-specific antibody to the system will block or inhibit other substances, particularly drugs, the specificity of these
spontaneous assembly of the enzyme by binding to the ED and other wet-chemistry procedures is questionable, and
ligand conjugate and therefore decrease the amount of the possibility of drug interferences still exists. For these
enzyme formed. Active enzyme formation is influenced by reasons, alternative modes of analysis were developed,
the concentration of antibody available for binding to the employing either chromatographic or immunological
EDligand conjugate. In turn the concentration of available procedures (Anticonvulsant Drugs: Comparison of methods
antibody is dependent on the concentration of the analyte in conditions).
the sample that can competitively bind to the antibody. Thin-layer chromatography procedures have been
xviiiref
The amount of enzyme created and the rate of described that semiquantitate or quantitate many
xxi
substrate hydrolysis is directly proportional to the sample anticonvulsant drugs. ref This procedure, because of both
analyte concentration. its lack of specificity and semiquantitative nature, is not a
Another procedure is based on a combination of very popular mode of analysis.
multilayer film technology and competitive immunoassay Gas chromatography (GC) affords reliable analysis
xix
(method 5i, Methods Summary Table). ref A coated for anticonvulsant drugs on microscale samples. The
multilayer film chip is encased in a plastic test module. The procedure permits the simultaneous analysis of all the
serum sample is applied to the topcoat layer in the test anticonvulsant drugs including valproate, and these methods
module. This layer contains buffer components, surfactants are easily automated. This simultaneous determination is an
and other reagents (ie. antibody and labeled hapten drug) as obvious advantage as many patients with epilepsy are on
well as acting as a filter to prevent protein interference with multiple drug regimens. Similar to high performance liquid
the immunoreaction. The sample travels from the topcoat chromatography (HPLC), GC is most useful in laboratories
layer through an iron oxide screen to the signal layer. The with a large workload. Disadvantages of GC include the
iron oxide serves to block excitation of the label outside the requirements of specialized equipment and considerable
signal layer. The signal layer contains a monoclonal expertise. GC is one of the methods of choice for the
xxii
antibody specific for the drug being assayed conjugated analysis of valproic acid ref primarily because the poor
with a fluorescent-labeled hapten in an agarose matrix ultraviolet absorption characteristics of this compound
coated onto a clear polyster film base. The sample antigen prevents its analysis by HPLC. Although steadily losing
displaces the labeled antigen from the binding sites of the ground to both HPLC and some of the competitive binding
antibodies. The unbound labeled antigen then diffuses into assays, GC remains a popular method for anticonvulsant
the upper layers where it cannot be measured. Equilibrium drug analysis.
occurs within 35 min, after which the fluorescent signal The adoption of HPLC for therapeutic drug
from the remaining bound conjugate is measured from the monitoring has been one of the most significant advances in
bottom of the test module by front surface fluorimetry. The
168
the clinical laboratory. The early methods used concentration for phenytoin and carbamazepine. The use of
(spectrophotometric methods, thin-layer chromatography, collection tubes containing gel has been reported to decrease
and gas chromatography) have been largely replaced by the concentrations of some drugs in serum and plasma. Gel
HPLC. tubes should be avoided when collecting specimens for
analysis of anticonvulsants.
Some advantages of HPLC over earlier
techniques: Interferences
1. Analyte volatility and thermal stability, so essential Icteric or hemolyzed specimens have less effect on
for gas chromatography, are not required for this type of the chromatographic methods than on immunoassays.
chromatography.
2. Relatively little sample workup is required before Anticonvulsant Drugs Reference Interval
analysis. The generally accepted therapeutic ranges for the
3. Characteristically, methods involving HPLC major anticonvulsant drugs which are subject to therapeutic
require only a short analysis time. drug monitoring are as follows:
4. Derivatization of the analyte is not required.
5. For laboratories with a large workload, the cost per Carbamazepine 412 mg/L (1751 mol/L)
analysis is very low because reagent costs are low. Ethosuximide 40100 mg/L (280460 mol/L)
6. HPLC affords good sensitivity. The sensitivity Phenobarbital 1530 mg/L (65170 mol/L)
varies with the drug, but reliable quantitation at drug Phenytoin 1020 mg/L (4080 mol/L)
concentrations as low as 1 g/mL can often be achieved. Primidone 512 mg/L (2355 mol/L)
7. HPLC methods are readily automated. Routine Valproate 50100 mg/L (350700 mol/L)
analyses therefore require a minimum of technician time, a
major factor in arriving at the low cost per analysis. Interpretation
8. HPLC makes possible the simultaneous analysis of Therapeutic drug monitoring of anticonvulsants is
most anticonvulsant drugs in a microsample. the best established of any medical therapy. The prevention
of any single seizure is important, as each episode is
Some limitations of HPLC in drug analysis are as potentially a life-threatening event. Epilepsy is a general
follows: term for a group of disorders in which there are recurrent
1. Spectrophotometric, fluorometric, and seizures. The causes of epilepsy are numerous. In many
amperometric detectors are commonly used. Therefore the patients, a particular pathological brain lesion is not
analyte in question must either absorb light, fluoresce, or be identified (i.e., the seizures are idiopathic). The major
electrochemically active. known causes of epilepsy include head injury,
2. Equipment is expensive. Investment of capital is cerebrovascular accidents (strokes), tumors (both primary
only warranted if the laboratory has a substantial work load. and secondary), infections (meningitis and encephalitis),
3. Although HPLC analysis of valproic acid has been electrolyte disturbances, and alcohol and other drugs. The
described these procedures require post column reactions or description of a seizure determines the category or type of
epilepsy (Anticonvulsant drugs: Seizure Table). Such
precolumn derivatization.xxiiiref The methods of choice for
definition is important, for different forms of epilepsy
the analysis of valproic acid are gas chromatography,
respond to different anticonvulsant drugs. It should not be
enzyme immunoassay, or fluorescence polarization
overlooked that the treatment of some secondary seizures
immunoassay.
(e.g., hypoglycemia, electrolyte disturbances, fever) is
primarily the treatment and correction of the underlying
Many antiepileptic drugs can be determined by
xxiv problem.
radioimmunoassay. ref Although extremely sensitive,
The major drugs which are used currently in the
these techniques have the disadvantages of not allowing for treatment of epilepsy can be divided into three groups with
the simultaneous assay of drugs in patients on multiple-drug xxv
ref All alter
regimens and their relatively long analysis time. They regard to their mechanisms of action.
require the availability of a liquid or gamma-ray scintillation neuronal conductance of ions, particularly sodium and
counter and availability and stability of suitable antisera. calcium. Phenytoin and carbamazepine inhibit high-
frequency, repetitive neuronal firing. Phenobarbital and
Specimen valproic acid enhance the synaptic inhibition associated with
Plasma or serum may be used for the gas gamma-aminobutyric acid (GABA) as do the
chromatographic or high-performance liquid benzodiazepines (e.g., diazepam and clonazepam).
chromatographic methods. Serum is the preferred specimen Ethosuximide alters calcium currents in the thalamus.
for the immunoassays. Sera containing the anticonvulsants Plasma/serum reference intervals for all these drugs, apart
discussed in this chapter can be stored at room temperature from the benzodiazepines, have been established to assist
for several hours. If frozen at -20 C , samples containing with their clinical use in the control of epilepsy. In the case
these drugs are stable for at least 1 year. Saliva may also be of benzodiazepines, there is as yet insufficient data to
used for the measurement of some anticonvulsants. The warrant their routine monitoring. In the past few years the
saliva concentration provides an estimate of the free drug practice of monotherapy with anticonvulsant drugs has
169
become increasingly accepted as desirable, i.e., a concerted The plasma therapeutic reference intervals shown
attempt should be made to control an individual patients are total concentrations. If a drug is highly protein bound, a
seizures with a single drug rather than multiple decrease in binding protein may potentially increase the
anticonvulsants. Patients on monotherapy tend to have a unbound moiety of drug. If in addition, there is rate-limited
lower incidence of side effects and drug interactions, as well clearance of that drug, then the unbound moiety will
as better compliance overall. However, if it becomes clear increase and there will be a high likelihood of clinical
that in an individual patient, control is not obtained with a toxicity. Such is the case with phenytoin. At the same time
single agent, then combination therapy is utilized. The the total plasma concentration may remain within the
combination will comprise drugs from groups with different reference interval. In these circumstances, some
mechanisms of action. practitioners would also measure the free phenytoin
The wide acceptance of the role of therapeutic drug concentration, utilizing equilibrium dialysis or
monitoring of anticonvulsants is due not only to the ultracentrifugation.
importance of good clinical control of epilepsy, but to the
widely varying pharmacokinetics of the drugs between Anticonvulsant Drugs Performance Goals
individuals. The major pharmacokinetic parameters are Survey data from the 2002 College of American
summarized in the Anticonvulsant drugs Table: Pathologists Participant Summary Report shows
Pharmacokinetic parameters. The timing of the sample to be imprecision values (% coefficient of variation) for
assayed in relation to the dose is important. The most valproate, phenobarbital, and phenytoin to range from 3.5%
appropriate sample for routine monitoring is the trough to 7.5%, 3.0% to 11.0%, and 3.0% to 10.0%, respectively,
concentration, i.e., that immediately prior to the next dose of in samples with drug concentrations within the therapeutic
the drug. When introducing a drug or adjusting the dose, it range.xxviref Acceptable performance criteria (CLIA-88)
is not generally useful to measure serum concentrations for measurement of carbamazepine, phenytoin, primidone,
until the patient has reached steady state. From a practical and valproic acid require that laboratories be accurate to
point of view, the steady state may be considered as being within 25% of the peer group mean, while phenobarbital
approached at a time interval equivalent to five times the and ethosuximide results must be within 20% of the peer
elimination half-life, e.g., in the case of primidone group mean.
approximately 2 days, for phenytoin more than 5 days, and
for phenobarbital more than 2 weeks.
170
Tables
Anticonvulsant Drugs Methods Summary Table
Method 1: Extraction and ultraviolet spectroscopy
Principle of analysis: Anticonvulsant extracted free of other ultraviolet compounds; ultraviolet spectrum or the
ultraviolet difference spectra are specific for the drug
Comments: Interference with drugs and endogenous compounds
Principle of analysis: Extracted drug separated by TLC and its Rf noted; drug quantitatively eluted and quantified by
ultraviolet spectroscopy
Comments: Difficult to separate all anticonvulsants; technically difficult to achieve reproducibility
Method 3: Gas chromatography (GC)
Principle of analysis: C18 solid phase extraction; capillary column; flame ionization detection
Comments: Can resolve all anticonvulsants simultaneously; requires simple extraction
Method 4: High performance liquid chromatography (HPLC)
Principle of analysis: Organic solvent extraction of drugs or protein precipitation of sample; anticonvulsants are
separated by reversed-phase chromatography and monitored by ultraviolet spectroscopy
Comments: Can resolve all anticonvulsants except valproate simultaneously; commonly used
Method 5: Competitive-binding assays
a. Radioimmunoassay (RIA)
Principle of analysis: Competitive binding of radioactive ligand; radio-labeled hapten competes with unknown for
antibody binding site
Comments: All problems of radioactive usage; slower than other immunoassays
b. Enzyme-multiplied immunoassay technique (EMIT)
Principle of analysis: Competitive binding with drug attached to enzyme
Comments: Can measure all drugs, but each must be done separately; commonly used
c. Fluorometric enzyme immunoassay (FEIA)
Principle of analysis: Combination of competitive binding with enzyme-labeled conjugate for antibody binding site (on
a solid-phase matrix) and radial chromatography
Comments: Measures all anticonvulsants except valproate and ethosuximide; use routine clinical chemistry analyzer
(e.g., Baxter Stratus); commonly used
d. Substrate-labeled fluorescence immunoassay (SLFIA)
Principle of analysis: Competitive binding substrate label; substrate competes with unknown antigen for antibody
binding site
Comments: Fluorescence interferes; not widely used
e. Fluorescence polarization immunoassay (FPIA)
Principle of analysis: Competitive binding with fluorescein-labeled drug; fluorescent drug competes with unknown for
antibody site
Comments: Each drug assayed individually; requires specialized instrument; most commonly used
f. Ames test strip
Principle of analysis: Antibody binding to drug labeled with prosthetic group prevents glucose oxidase activity;
competition with endogenous drug allows prosthetic group to activate enzyme; glucose oxidase is coupled to
colorimetric reaction
Comments: Designed for inexpensive instrument; reflectance spectrophotometer; rapid test
g. Rate nephelometric inhibition immunoassay
Principle of analysis: Competitive binding for hapten; hapten-protein competes with unknown for antibody binding site
Comments: Each drug individually assayed; requires specialized instrument
h. Cloned enzyme donor immunoassay (CEDIA)
Principle of analysis: Competitive binding with enzyme donor(ED)ligand conjugate for antibody binding site;
endogenous drug modulates enzymatic activity by influencing free EDligand conjugate available for complementation
Comments: Measures phenobarbital and phenytoin; each drug assayed individually; can use routine clinical chemistry
analyzers (e.g., BM Hitachi system); newer technology; not commonly used at present
i. Dry film multilayer immunoassay (OPUS)
Principle of analysis: Combination of competitive binding with a fluorescent labeled hapten in an agarose matrix and
multilayer film technology; labeled antigen competes with sample antigen for antibody binding site
Comments: Each drug assayed individually; requires specialized instrument; newer technology; not widely used at
present
171
Anticonvulsant Drugs: Comparison of method conditions.

Parameter: High-performance liquid chromatography
Temperature: Ambient
Sample volume: 100 L
Linearity (g/mL):
Phenytoin: 5.060.0
Phenobarbital: 7.590.0
Primidone: 2.530.0
Ethosuximide: 25.0300.0
Carbamazepine: 2.530.0
Valproic acid: ---
Precision (CV): 27
Parameter: Gas chromatography
Temperature: 60250 C
Linearity (g/mL):
Phenytoin: 10.0100.0
Primidone: 10.0100.0
Valproic acid: 10.0100.0
Precision (CV): 18
Parameter: Enzyme-multiplied immunoassay technique
Temperature: 30 C
Sample volume: 50 L
Linearity (g/mL):
Phenytoin: 2.540.0
Primidone: 2.520.0
Precision (CV): 612*
Parameter: Fluorescence polarization
Temperature: 37 C
Sample volume: 50 L
Linearity (g/mL):
Phenytoin: 2.540.0
Primidone: 2.024.0
Precision (CV): 49*
Parameter: Fluorometric enzyme immunoassay
Temperature: 3740 C
Linearity (g/mL):
Phenytoin: 2.540.0
Primidone:
Ethosuximide:
Valproic acid:
Precision (CV): 812*
*1990 College of American Pathologists Therapeutic Drug Monitoring Series 2 Surveys.
172
Anticonvulsant Drugs Table: Classification of epileptic seizures*

Type Characteristics
I. Partial (focal, local)
Electroencephalography (EEG) may reveal localized abnormality over seizure focus.
A. Simple partial Consciousness is not impaired. Includes Jacksonian motor epilepsy (confined to a
single muscle group or limb).
B. Complex partial Consciousness is impaired. Most commonly arise in the temporal lobe. Include many
different symptoms and behavior patterns.
C. Secondarily generalized Both (A) and (B) may spread to involve both hemispheres, with convulsions of all
limbs and loss of consciousness.
II. Generalized seizures (examples) Simultaneous initiation of epileptic activity in both hemispheres.
A. Absence Brief impairment of consciousness. Motor activity may not occur.
B. Myoclonic Isolated muscle jerks.
C. Clonic Rhythmic contractions, loss of consciousness.
D. Tonic-clonic (grand mal) Major convulsions, loss of consciousness, autonomic activity and prolonged
depression of central nervous system function.
* Modified from Commission on Classification and Terminology of the Internation League
against Epilepsy. Epilepsia 1981; 22:489-501.
Anticonvulsant Drugs Table, Pharmacokinetic properties.

Anticonvulsant: Carbamazepine
Volume of distribution (L/kg): 1.2
Percent protein binding: 75
Elimination half-life (h): 1217
Comments: Induction of own and other anticonvulsants metabolism; metabolite, carbamazepine-10,11-epoxide, may
contribute to adverse effects
Anticonvulsant: Ethosuximide
Percent protein binding: <10
Comments:
Anticonvulsant: Phenobarbital
Comments: Predominantly hepatic clearance, approximately 25% renal; significant induction of hepatic cytochrome
P450 with increased clearance of many drugs, including other anticonvulsants
Anticonvulsant: Phenytoin
Volume of distribution (L/kg): 0.40.8
Comments: Dose-dependent kinetics with saturable hepatic metabolism; induction as for phenobarbital; protein binding
important: NB hypoalbuminemia (e.g., renal disease)
Anticonvulsant: Primidone
Percent protein binding: <20
Comments: Metabolized to phenobarbital, but anticonvulsant in its own right; both primidone and phenobarbital should
be monitored concurrently
Anticonvulsant: Valproate
Comments: Protein binding as for phenytoin; in addition, binding saturable at high concentrations, unbound valproate
concentrations increasing proportionately more than total serum concentration
173
Figures
Anticonvulsant drugs: Figure 1
Anticonvulsant Drugs: ultraviolet absorption spectra of 5 drugs in acetonitrile. Solid line carbamazepine (10g/mL); thick dashed
line, ethosuximide (100g/mL); dashed and dotted line, phenytoin (100g/mL); regular dashed, phenobarbital (100g/mL);
dotted, primidone, (100g/mL)

Permission for electronic not granted; see Methods in Clinical Chemistry Pesce AJ and Kaplan LA Eds Mosby St Louis 1987.
Assay scheme of a typical CEDIA assay.

Permission for electronic not granted; see Methods in Clinical Chemistry Pesce AJ and Kaplan LA Eds Mosby St Louis 1987.
Chromatograms obtained at 204 nm using tolybarb and methsuximide as the internal standards from serum standard A, level III
concentrations.
174
Chromatogram of a serum sample spiked with 50 g/ml VPA and ET, and 10 g/ml PB, PR, CZ and PT, with two internal
standards (CA and MPPH) at 20 g/ml. The serum and column were acidified prior to the extration with 0.5 M HCl. Injected
volume, 2 l of extract.
a. 10 mmol/L potassium phosphate buffer

Procedure: High Performance Liquid Chromatography
xxvii (pH 6.3) is prepared by dissolving 2.72 g
Analysis of Anticonvulsant Drugs ref
of KH2PO4 in 1900 mL of distilled water.
Principle The pH is adjusted to 6.3 with about 2 mL
Five anticonvulsant drugs (primidone, of 1 M NaOH, then the volume brought to
phenobarbital, carbamazepine, phenytoin and ethosuximide) 2 L.
are separated by reversed-phase liquid chromatography and b. 1 mmol/L potassium phosphate buffer
quantitated by measurement of peak height relative to an (pH 6.8) is prepared by dissolving 13.6 g
internal standard. Tolybarb is used as an internal standard of KH2PO4 in 90 mL of distilled water.
for the quantitation of primidone, phenobarbital, The pH is adjusted to 6.8 with about 3 mL
carbamazepine, and phenytoin. Methsuximide is used as the of 10 M NaOH, then the volume is
internal standard for the quantitation of ethosuximide as it brought to 100 mL.
has a similar volatility when evaporating the extract to 4. Stock internal standards are prepared by
dryness. dissolving 30 mg of tolybarb and 20 mg of
Specimen methsuximide in 20 mL of methanol.
Serum or plasma. 5. Solvent extraction solution is prepared by
combining 500 L of the stock internal standard
Reagents and Materials solution with 500 mL of 5% isopropanol in
1. HPLC-grade acetonitrile, methanol, isopropanol chloroform.
and chloroform can be purchased from Fisher 6. Mobile phase is prepared by combining 1300 mL
Scientific (Winnipeg, Manitoba, Canada). of phosphate buffer (pH 6.3) with 350 mL of
2. Standards. The pure drugs can be purchased from methanol and 350 mL of acetonitrile. The mixture
Sigma Chemical Co., St. Louis, MO (phenytoin), is degassed by stirring in a magnetic mixer for 30
Aldrich Chemical Co., Milwaukee, WI (tolybarb, min, then filtered by suction through a 0.45-m
5-ethyl-5-p-tolybarbituric acid), BDH Chemicals, pore filter (Millipore Corp., Bedford, MA). The
Toronto, Ontario, Canada, (phenobarbital), Ayerst apparent pH of the mobile phase is 7.0.
Laboratories, Motreal, Quebec, Canada
(primidone), Ciba-Geigy Canada Ltd., Assay
Mississauga, Ontario (carbamazepine), and Parke 1. Equipment. The liquid chromatographic system
Davis Canada Inc., Scarborough, Ontario consists of a Series 2 pump, a model LC-75
(ethosuximide and methsuximide). variable wavelength detector, a R-100 strip chart
recorder (all from PerkinElmer Corp., Norwalk,
3. Phosphate buffer.
175
CT), and a model 7105 injector (Rheodyne Inc., All six drugs (valproic acid, ethosuximide,
Cotati, CA). A 5-m particle size, 250 4.6 mm phenobarbital, primidone, carbamazepine, and phenytoin)
Supelcosil LC-1 analytical column and a 5-m are extracted from acidified serum samples by a simple
particle size, 20 4.6 mm Supelguard column solid-phase (C18) extraction process. Prior to evaporation,
(Supelco, Bellefonte, PA) are used to effect the basic methanol is added to the eluates from extraction
drug separations. columns to convert acid drugs to their potassium salts, in
2. Sample preparation. order to reduce the loss of volatile drugs. The residue is then
a. Add 100 L of 1 mmol/L phosphate buffer (pH redissolved in acidic methanol and 2 L is injected into the
6.8) and 1.5 mL of solvent extracting solution gas chromatograph equipped with a split capillary inlet
(containing internal standard) to 100 L of serum system and a flame ionization detector.
or plasma. Specimen
b. Vortex mix the sample for 30 sec and centrifuge Serum.
briefly to separate the two phases.
c. Aspirate the aqueous phase (top layer) to waste Reagents and Materials
and transfer the organic phase (lower layer) to a 1. Six antiepileptic drugs (SPOFA, Prague,
clean tube. Czechoslovakia and VEB Arzneimittelwerk,
d. The organic layer is evaporated at room Dresden, Germany).
temperature with a dried air stream. The samples 2. Stock standard solution of the six anticonvulsant
are removed from the air stream within 2 min of drugs is prepared in analytical reagent grade
reaching dryness to avoid loss of ethosuximide and methanol to a concentration of 1 g/L.
its internal standard due to their high volatility. 3. Stock internal standard is prepared by dissolving
e. Reconstitute the residue with 100 L of mobile 40 mg of MPPH [5-(4-methylphenyl)-5-
phase. phenylhydantoin] in 20 mL of methanol.
3. Chromatographic parameters. 4. 0.5 M HCl solution in water. Dilute 4.2 mL of
a. Flow rate 2.0 mL/min. concentrated HCl to 100 mL with H2O.
b. Wavelength 204 nm. 5. 0.05 M KOH in methanol. Add 0.281 g of KOH
c. Sensitivity 0.08 AUFS(absorbance units at full to 80 mL of methanol and bring to 100 mL with
scale). methanol.
d. Column temperature ambient. 6. 0.05 M HCl in methanol. Dilute 42 L of
e. Injection volume 4 L. concentrated HCl to 100 mL of methanol.
f. Run time 7 min. Assay
Equipment: A HP-5790A gas chromatograph
A typical chromatogram is shown in the Figure 4: (Hewlett-Packard, Avondale, PA) equipped with a split
Anticonvulsant Drugs HPLC Elution Pattern. injection port (ratio 1:20) using an empty glass liner and
flame ionisation detection. A fused-silica capillary column
Calculations
packed with cross-linked 5% phenylmethylsilicone gum
In each sample, peak heights (PH) of each drug are
phase HP-5 (Hewlett-Packard), 25 m 0.20 mm ID, 0.33
compared to the internal standard from which peak height
ratios (PHR) are calculated. Each unknown sample PHR is m film thickness is used to separate the drugs of interest.
compared to a calibrator PHR of known concentration. The Sample preparation
calculation of drug concentration is as follows. 1. Add 100 L of 0.5 M HCl and 10 L of internal
[DRUG] = [CAL] PH of sample/ PH of internal standard standard solution (MPPH) to 500 L of serum and
PH of calibrator/ PH of internal thoroughly vortex mix.
standard
2. Precondition the extraction column (a 2 mL
where [DRUG] = drug concentration (mg/L) polyethylene syringe packed with 200 mg of
[CAL] = calibrator concentration (mg/L) Silipor C18) with a 1 mL wash of methanol
followed by a 1 mL wash of 0.5 M HCl.
3. Pour mixture (from step 1) onto the column and
Bibliography allow the mixture to flow through. Then rinse the
See popups in the text column with two 1-mL volumes of H2O and dry by
applying a vacuum.
Procedure: Antiepileptic Drugs Including Valproate by
xxviii
ref 4. The drugs are then eluted from the column with 1
Capillary Gas Chromatography mL of methanol. A 50- L aliquot of 0.05 M KOH
Principle is added to the methanol solution. The solvent is
then evaporated under nitrogen at 40 C in a
176
heating block. The residue is reconstituted in 50 L Specimens may be transported at room temperature.
of 0.05 M HCl in methanol. Separate serum component and store at room
5. Chromatographic parameters: temperature, up to 24 h, until time of analysis, or
a. Injection temperature 240 C prepare ultrafiltrate (Section VI.A), then freeze at -
b. Detector temperature 300 C 20C until time of testing, marking the sample as
c. Dual ramp column temperature program such.
(60250 C) (NOTE: Typical laboratory room temperature ranges
d. Nitrogen inlet pressure 100 kPa between 22 and 25 C. If temperature is >25 C,
e. Injection volume 2 L separated serum should be refrigerated at 28 C.
Grossly hemolyzed, lipemic, or icteric specimens are
Calculation unacceptable samples for routine analysis.
1. A single-point calibration:
III. Indications
Peak Height Ratio (PHR) = Peak Height of Drug
For the monitoring of phenytoin for therapeutic drug
Peak Height of Internal Standard
monitoring (TDM) purposes, as an aid in setting the
dosing regimen, and to assess toxicity, when total
[Drug] = PHR of Drug [Standard]
phenytoin drug analysis does not provide adequate
PHR of Standard
information.
where [Drug] = concentration of drug (mg/L)
[Standard] = concentration of standard (mg/L) IV. Reagents and Material
A. Bio-Rad Control Materials (Level 1 and 2) Store
at 4 C. Stable as labeled.
Bibliography B. 100-L and 1-mL Eppendorf pipets.
See popups in text C. Amicon CentriFree Micropartition Filters
(Amicon).
Free Phenytoin
Abbott TDx/FLx D. Sample Cartridges (Abbott).
E. Sample Cuvettes (Abbott).
I. Principle F. TDx Reagent Pack Adapter for FLx
The concentration of free, or unbound, phenytoin in G. Calibration and Routine Carousels (Abbott).
samples can be determined using fluorescence H. Buffer (Abbott); store at room temperature.
polarization immunoassay. After a filtration step, a The following Abbott Laboratories Phenytoin Assay
clear supernatant containing the drug is analyzed. Kit Materials:
Unlabeled drug in the sample competes with the tracer I. Phenytoin Reagent Pack* (Batch Assay).
labeled with fluorescein for limited antibody sites. J. Free Phenytoin Calibrators*.
The tracer, when excited by linearly polarized light,
emits fluorescence with a degree of polarization
*Store at 4 C. Stable for 1 month after opening.
inversely related to its rate of rotation. If the patient
contains high concentrations of the drug, the
polarization value is low. If the patient sample V. Instrument and parameters.
contains a low concentration of the drug, the
polarization value is high. Abbott TDx/FLx; however, any of the automated drug
analyzers which have a linear range to 0.5 g /mL are
acceptable
II. Specimens of Choice VI. Procedure

Serum (2 mL) drawn in a red-top Vacutainer tube
without separator gel. (NOTE: Gel tubes are not Each routine run will generally consist of Bio-
recommended, because of the potential to decrease Rad Level 1 and 2, along with patients,
drug concentration during storage. This is negated processed in singlecate.
somewhat by always aliquoting serum into a separate
pour-off tube immediately.) Recommended drawing Controls and patients require Amicon filtering;
time is the trough level, immediately prior to the next calibrators do not.
scheduled dose of drug, or during the acute phase of
toxicity.
177
A. Pretreatment (for Control and Patient The memory circuit board (Board
Specimens) #2) is replaced
Assay control values fall outside of
When conducting ultrafiltration prior to the acceptable range specified in the
measurement of free phenytoin, check that the specific assay section and the
room temperature at that time does not exceed control is not suspect.
25 C. If the temperature is >25 C, contact the Assays may require recalibration (when
area supervisor. If the sample is ultrafiltered at indicated by unacceptable control data)
room temperature of 3037 C protein binding whenever:
will be affected and a different reference range A new lot number of reagent is used
may need to be applied:
A new lot of buffer is used
12 g/mL at 25 C. Any dispense component is replaced
1.53 g/mL at 37 C Any instrument calibration
procedure is performed
1. Free phenytoin calibrators, controls, and A wide variation in room
patient specimens should be mixed by temperature is experienced
gentle inversion. Avoid foaming; if this
The TDx/FLx System stores a calibration
should occur, allow sample to sit until
curve for each assay. The random access
foam dissipates.
calibration curves are separate and distinct
2. Number an Amicon fFilter for each from the calibration curves stored for the
specimen to be tested and place in a batch assays. Therefore, prior to running
suitable rack. A maximum of 20 an assay, the instrument must be
specimens can be run in one batch. calibrated for that assay in the mode of
3. Pipet 1 mL of control or patient sample to operation being run.
be assayed into its corresponding Amicon Notes and Interpretations
filter.
A. Optimal therapeutic concentrations of free
4. Cap the Amicon filter sample resevoir. phenytoin will generally be between 1 and 2
5. Place filter into a fixed-angle centrifuge g/ml. Frequency and severity of toxic effects
and centrifuge samples for 25 min. due to drug therapy will increase as the free
6. After centrifugation, transfer 100 L of phenytoin level exceeds 3.0 g/mL. Results
ultrafiltrate into appropriate sample wells may be reported up to 10 g/mL (refer to
and place in a sample carousel. "Alert" values information.)
B. Specificity: No known interfering substances.
B. Calibration (no Amicon filtering required)
C. Sensitivity: 0.5 g/mL; report negatives as <0.5
g/mL.
1. Performing the Calibration
IX. reference
Assays should be calibrated whenever:
A new assay is introduced Abbott Laboratories' TDx and TDx/FLx Operations Manual.
An assay activation (new reagent
pool) is issued
i
This chapter was originally prepared by:
Paul Salm
Paul J. Taylor
and Julia M. Potter
ii
Plaa, G.L. and Hine, C.H.:
A method for the simultaneous determination of phenobarbital and diphenylhydantoin in blood, J. Lab. Clin. Med. 47:649-
657, 1956.
Svensmark, O., and Kristensen, P.:
Determination of diphenylhydantoin and phenobarbital in small amounts of serum, J. Lab. Clin. Med. 61:501-507, 1963.
iii
Pippenger, C.E., Scott, J.E., and Gillen, H.W.:

Thin-layer chromatography of anticonvulsant drugs, Clin. Chem. 15:255-260, 1969.
178
Huisman, J.W.:
The estimation of some important anticonvulsant drugs in serum, Clin. Chim. Acta 13:323-328, 1966.
iv
Rambeck, B., and Meijer, J.W.A.:
Gas chromatographic methods for the determination of antiepileptic drugs: A systematic review, Ther. Drug Monit. 2:385-
396, 1980.
v
396, 1980.
vi
Meijer, J.W.A.:
Simultaneous quantitative determination of anti-epileptic drugs, including carbamazepine, in body fluids, Epilepsia
12:341-352, 1971.
Gardner-Thorpe, C., Parsonage, M.J.H., Smethurst, P.F., and Toothill, C.:
A comprehensive gas chromatographic scheme for the estimation of antiepileptic drugs, Clin. Chim. Acta 36:223-230,
1972.
Toseland, P.A., Albani, M., and Gauchel, F.D.:
Organic nitrogen selective detector used in gas-chromatographic determination of some anticonvulsant and barbiturate
drugs in plasma and tissues, Clin. Chem. 21:98-103, 1975.
Rambeck, B., Meijer, J.W.A.:
Comprehensive method for the determination of anti-epileptic drugs using a novel extraction technique and a fully
automated gas chromatograph, Arzneim. Forsch. 29:99-103, 1979.
vii
396, 1980.
viii
Plotczyk, L.L.:
Application of fused-silica capillary gas chromatography to the analysis of underivatized drugs, J. Chromatog. 240:349-
360, 1982.
ix
Volmut, J., Matisova, E., and Pham Thi Ha.: Simultaneous determination of six antiepileptic drugs by capillary gas
chromatography, J. Chromatogr 527:428-435, 1990.
x
Meatherall, R., and Ford, D.:
Isocratic liquid chromatographic determination of theophylline, acetaminophen, chloramphenicol, caffeine,
anticonvulsants, and barbiturates in serum, Ther. Drug Monit. 10:101-115, 1988.
Riedmann, M., Rambeck, B., and Meijer, J.W.A.:
Quantitative simultaneous determination of eight common antiepileptic drugs and metabolites by liquid chromatography,
Ther. Drug Monit. 3:397-413, 1981.
Soldin, S.J., and Gilbert Hill, J.:
Rapid micromethod for measuring anticonvulsant drugs in serum by high-performance liquid chromatography, Clin.
Chem. 22:856-859, 1976.
Christofides, J.A., and Fry, D.E.:
Measurement of anticonvulsants in serum by reverse-phase ion-pair liquid chromatography, Clin. Chem. 26:499-501,
1980.
Kabra, P.M., Koo, H.Y., and Marton, L.J.:
Simultaneous liquid-chromatographic determination of 12 common sedatives and hypnotics in serum, Clin. Chem.
24:657-662, 1978.
Kabra, P.M., Stafford, B.E., and Marton, L.J.:
Simultaneous measurement of phenobarbital, phenytoin, primidone, ethosuximide, and carbamazepine in serum by high-
pressure liquid chromatography, Clin. Chem. 23:1284-1288, 1977.
Ou, C.N., and Rognerud, C.L.:
Simultaneous measurement of ethosuximide, primidone, phenobarbital, phenytoin, carbamazepine, and their bioactive
metabolites by liquid chromatography, Clin. Chem. 30:1667-1670, 1984.
Adams, R.F., and Vandemark, F.L.:
Simultaneous high-pressure liquid-chromatographic determination of some anticonvulsants in serum, 22:25-31, 1976.
Szabo, G.K., and Browne, T.R.:
179
Improved isocratic liquid-chromatographic simultaneous measurement of phenytoin, phenobarbital, primidone,

carbamazepine, ethosuximide, and N-desmethylmethsuximide in serum, 28:100-104, 1982.
xi
Cook, C.E., Christensen. H.D., Amerson, E.W., et al.: Radioimmunoassay of anticonvulsant drugs: phenytoin, phenobarbital,
and primidone. In Kellaway, P., and Petersen, I.S., editors: Quantitative analytical studies in epilepsy, New York, 1976, Raven
Press.
xii
Price, C.P., and Newman, D.J. editors:
Radial partition immunoassay. In Principles and practice of immunoassay, New York, 1991, Stockton Press.
xiii
Geigel, J.L., Brotherton, M.M., Cronin, P., et al:
Radial partition immunoassay, Clin. Chem. 28:1894-1898, 1982.
xiv
Wong, R.C., Burd, J.F., and Carrico, R.J., et al:

Substrate-labeled fluorescent immunoassay for phenytoin in human serum, Clin. Chem. 25:686-691, 1979.
xv
Tybach, R.:
Adaptation of prosthatic-group label homogenenous immunoassay to reagent strip format, Clin. Chem. 27:1499-1504,
1981.
xvi
Nishikawa, T., Kubo, H., and Saito, M.:
Competitive nephelometric immunoassay method for antiepileptic drugs in patient blood, J. Immunol. Methods 29:85-89,
1979.
xvii
Henderson, D.R., Friedman, S.B., Harris, J.D., et al.:
CEDIA, a new homogeneous immunoassay system, Clin. Chem. 32:1637-1641, 1986.
xviii
Engel, W.D., Khanna, P.L.:
CEDIA in vitro diagnostics with a novel homogeneous immunoassay technique, J. Immunol. Methods 150:99-102, 1992.
xix
Jandreski, M.A., Shah, J.C., Gardincius, J., and Bermes Jr., E.W.: Clinical exaluation of five therapeutic drugs using dry
film multilayer technology on the OPUS immunoassay system, J. Clin. Lab. Anal. 5:415-421, 1991.
xx
Plaa, G.L. and Hine, C.H.:
A method for the simultaneous determination of phenobarbital and diphenylhydantoin in blood, J. Lab. Clin. Med. 47:649-
657, 1956.
Svensmark, O., and Kristensen, P.:
Determination of diphenylhydantoin and phenobarbital in small amounts of serum, J. Lab. Clin. Med. 61:501-507, 1963.
xxi
Comparison of method conditions).

Pippenger, C.E., Scott, J.E., and Gillen, H.W.:
Thin-layer chromatography of anticonvulsant drugs, Clin. Chem. 15:255-260, 1969.
Huisman, J.W.:
The estimation of some important anticonvulsant drugs in serum, Clin. Chim. Acta 13:323-328, 1966.
xxii
Willox, S., and Foote, S.E.:

Simple method for measuring valproate (Epilim) in biological fluids, J. Chromatogr. 151:67-70, 1978.
Jakobs, C., Bohasch, M., and Hanefeld, F.:
New direct micromethod for determination of valproic acid in serum by gas chromatography, J. Chromatogr. 146:494-
497, 1978.
Berry, D.J., and Clarke, L.A.:
Determination of valproic acid (dipropylacetic acid) in plasma by gas-liquid chromatography, J. Chromatogr. 156:301-
307, 1978.
Freeman, D.J., and Rawall, N.:
Extraction of underivatized valproic acid from serum before gas chromatography, Clin. Chem. 26:674-675, 1980.
xxiii
Fairinotte R. Pfaff, MC and Manhuzier G. Simultaneous determination of phenobarbital and valproic acid using high
performance liquid chromatography. Ann. Biol. Clin. 36:347-353, 1978. Gupta RN, Keane PM, and Gupta ML, Valproic acid in
plasma as determined by liquid chromatography. Clin Chem 25:1984-1985, 1979.
xxiv
Cook, C.E., Christensen. H.D., Amerson, E.W., et al.: Radioimmunoassay of anticonvulsant drugs: phenytoin, phenobarbital, and
primidone, In Kellaway, P., and Petersen, I.S., editors: Quantitative analytical studies in epilepsy, New York, 1976, Raven Press.
xxv
Albers, G.A., and Peroutka, S.J.:
Neurologic disorders. In K.L. Melmon, H.F. Morrelli, B.B. Hoffman, D.W. Nierenberg, editors: Melmon and Morrellis
clinical pharmacology: Basic principles in therapeutics, ed. 3, New York, 1992, McGraw-Hill Inc. Health Professions Division.
180
xxvi
All conclusions and interpretations in the Infobase with respect to the College of American Pathologists data base are those of
the author of this section or the editors of the Infobase, and not those of the College.
xxvii
Meatherall, R., and Ford, D.:
Isocratic liquid chromatographic determination of theophylline, acetaminophen, chloramphenicol, caffeine,
anticonvulsants, and barbiturates in serum, Ther. Drug Monit. 10:101-115, 1988.
xxviii
Volmut, J., Matisova, E., and Pham Thi Ha.:
Simultaneous determination of six antiepileptic drugs by capillary gas chromatography, J. Chromatogr. 547:428-435,
1990.
181
Apolipoproteins A-1 and B

Jillian R. Tate
Name: Apolipoproteins A-1 (apo A-1) and B-100 (apo B)

Clinical significance: Predictors of cardiovascular disease risk. Refer to Chapter 37, Coronary Artery
Disease: Lipid Metabolism, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular mass: 28,300 D (apo A-1); 549,000 D (apo B-100)
Chemical class: Proteins
Reference method: HPLC-MS (apo A-1); no reference method available for apo B
Principles of Analysis and Current Usagei then specific polyclonal goat or sheep anti-apo A-1 or
Apo A-1 and B-100 are measured by immunochemical apo B antiserum (reagent 2) is added to the cuvette, and
methods. The original measurements were by gel the reaction proceeds, forming insoluble antigen-
techniques, including radial immunodiffusion (RID) and antibody complexes. The resultant turbidity is measured
rocket or electroimmunoassay (EIA), and by in a photometric system as an increase in absorption at
radioimmunoassay (RIA). Whereas the gel methods 340 nm over a defined period of time with subtraction of
were difficult to automate and used large amounts of the sample blank reading. Very small changes in
apolipoprotein antiserum, the advantage of RIA was its turbidity against a higher background scatter are
increased analytical sensitivity, precision, and lesser use precisely measured using the photometer of modern
of antiserum. RIA, however, is difficult to automate, and analyzers (Tables 1a and 1b, Methods 1 through 3).
reagents have a limited shelf life [1-3].
Immunonephelometric Methods
More recent methods include enzyme-linked In end-point INA methods for apo A-1 and B-100,
immunosorbent assay (ELISA), particle-concentration diluted sample is added to buffer containing a
fluorescence immunoassay (PCFIA), and precipitation enhancer (e.g., PEG), then specific
immunonephelometric (INA) and immunoturbidimetric polyclonal rabbit anti-apo A-1 or apo B antiserum is
(ITA) assays, all of which use mono-specific antibodies added to the cuvette, and the reaction proceeds, forming
in the liquid phase. Dedicated protein analyzers using soluble antigen-antibody complexes. The immuno-
end-point or rate INA or centrifugal analyzers using ITA particles scatter light from an infrared LED, and after a
methods became available in the 1980s and 1990s in fixed time the scattered light, which is measured at an
routine clinical laboratories. In todays laboratories, angle to the incident beam of light, is converted into
high-throughput, fully automated chemistry analyzers protein concentration using a standard curve and V-Lin
measure a range of proteins by ITA or particle-enhanced Integral modeling (Tables 1a and 1b, Method 4).
ITA, including the measurement of apo A-1 and apo B-
100. The precision of the tests performed on automated In peak-rate INA methods for apo A-1 and apo B-100,
analyzers is high. By ITA methods, total imprecision is the antigen-antibody reaction is monitored continuously
between 1.6% and 3.5% CV in the apo A-1 measuring as the reagents (buffer and polyclonal goat anti-apo A-1
range 47 to 209 mg/dL (0.47 to 2.09 g/L) and 1.1% to or apo B antibody) and diluted sample are mixed in the
6.3% CV in the apo B measuring range 43 to 150 mg/dL cuvette. The immune complexes scatter light from a
(0.43 to 1.50 g/L). By INA, total imprecision is between laser beam (670 nm), and the rate of change in
3.3% and 6.0% CV in the apo A-1 measuring range 58 to absorbance measured at an angle of 90 degrees over a
167 mg/dL (0.58 to 1.67 g/L), and 2.4% to 6.0% CV in fixed time is determined. Apo A-1 and apo B
the apo B measuring range 58 to 186 mg/dL (0.58 to concentration are determined by comparison against a
1.86 g/L) [1,4]. calibration curve (Tables 1a and 1b, Method 5).
Immunoturbidimetric Assays
In end-point ITA methods for apo A-1 and B-100,
diluted sample is mixed and incubated with reagent 1 Reference and Preferred Methods
containing a polymer (e.g., polyethylene glycol [PEG]), Reference Materials and Methods
The candidate primary reference measurement procedure
for apo A-1 quantification is an enzymatic hydrolysis
ii
Apolipoproteins A-1 and B isotope dilutionmass spectrometric method [5]. The
Previous and current authors of this method: primary standard used is a hydrolyzed peptide unique to
First edition: Not done apo A-1 that was purified by reversed-phase HPLC and
Methods edition: Suman T. Patel, Howard A.I. Newman its mass determined by amino acid analysis. The method
Second edition: Not updated was validated by assessing the concentration of apo A-1
Third edition: Not updated in the European Community Bureau of Reference (BCR)
Fourth edition: Not updated lyophilized Certified Reference Material (CRM 393).
Fifth edition: Jillian R. Tate For apo B-100, no primary reference method is
available.
182
Use of reference materials in a physical state equivalent Until recently, the large-scale introduction of
to that of patient samples is necessary to reduce the apolipoprotein measurement was hampered by the lack
variation in values among laboratories and between of reference materials and methods, which resulted in the
different measurement systems. Primary standards for use of different calibration methods and a lack of
apo A-1 and B-100 (i.e., pure proteins) are not suitable comparability of results. It was soon realized that the
for manufacturers assay systems because they show preparation, selection, and calibration of appropriate
instability and/or matrix effects. Pure apo B-100 is serum-based reference materials were key to obtaining
particularly unsuited as a primary standard because of international standardization and harmonization of
self-association, sensitivity to oxidation, and irreversible immunoassays for measurements of apo A-1 and B-100
matrix aggregation. Therefore, secondary matrix- [10-13] and would enable the establishment of
matched, serum-based materials have been developed for standardized reference values and decision cutoff limits
apo A-1 and B. Value assignment of these materials, in different ethnic populations (Tables 2a and 2b). Apo
SP1-01 for apo A-1 and SP3-07 for apo B-100, was by A-1 material SP1-01 and apo B-100 material SP3-07
the use of consensus methods, RIA for apo A-1 and INA (more recently SP3-08) are used to transfer assigned
for apo B, and purified materials, BCR-393 for apo A-1 values to manufacturers calibrators [6,7].
and LDL freshly prepared by ultracentrifugation (density
1.030-1.050 kg/L) for apo B [6,7]. An apo A-1 value of Through the use of IFCC standardized methods traceable
1.50 g/L was assigned to SP1-01, and an apo B value of to SP1-01 and SP3-07 and the availability of these
1.22 g/L was assigned to SP3-07. These reference certified reference materials that can be used by
materials have become the WHO-IFCC International diagnostic manufacturers to value-assign their assay
Reference Reagents for apo A-1 and apo B-100. calibrators, the value distributions of apo A-1 and apo B
in various populations have been described. These
Current preferred methods are end-point and rate include studies in the United States (NHANES III and
immunonephelometric assays available on dedicated Framingham Offspring studies) and in Finland, Italy, and
protein analyzers or end-point immunoturbidimetric Sweden [14-19]. From this work, clinically meaningful
assays available on automated chemistry analyzers. cut-points for risk stratification have been established in
various populations (Tables 2a and 2b). The studies
Specimen showed the percentile distribution of apo A-1 levels was
The optimum apolipoprotein specimen is a fresh serum nearly identical between populations, whereas apo B-100
sample and is the preferred sample type in INA (Tables levels were slightly higher in the Finnish and Swedish
1a and 1b, Methods 4 and 5). Plasma-heparin and populations, compared with the American and Italian
plasma-EDTA may be suitable in ITA methods, populations. The higher apo B-100 levels in the
depending on the specific assay to be used (Tables 1a Scandinavian groups were matched by higher cholesterol
and 1b, Methods 1 through 3). It is reported that apo A-1 levels.
and B remain stable for approximately a week at 2C to
8C, about 6 months at 20C, and for over 5 years Similar to the serum lipids, apolipoproteins are
when stored at 50C to 80C [8]. In another study, influenced by biological variation. Whereas there are
apo A-1 was shown to rise slightly in some serum marked differences in high-density lipoprotein
specimens stored at 4C after a couple of weeks, cholesterol (HDL-cholesterol) and low-density
whereas apo B was stable for at least 4 weeks [9]. lipoprotein cholesterol (LDL-cholesterol) between
Commercial assays make specific recommendations for fasting and nonfasting patients, apo A-1 and B-100
sample type and specimen stability, which have been concentration in lipoproteins is more stable than the
tested for these assays. lipids concentration, which may vary according to
presence of metabolic disease and diet. In healthy
Interferences subjects within-individual variation for apo A-1 and B is
Similar to ITA and INA methods for other specific approximately 6% to 7% [8]. Similar to lipids and
proteins, apo A-1 and B measurement is affected in lipoproteins, in severe illness the apolipoproteins may
particular by turbidity from large triglyceride particles in decrease substantially in concentration [20]. During
grossly lipemic samples. These contribute to excess light pregnancy, apo A-1 and B levels increase by 30% and
scattering that is not due to apo A-1 or apo B immune 60%, respectively [21].
complexes. Other particulate matter present in samples
(e.g., fibrin, cellular debris, etc.) can also cause excess Apo A-1 is important in lipoprotein transport and has
scatter unrelated to the analyte concentration. other physiological functions, including (1) structural
Commercial assays make specific recommendations for protein for chylomicrons and HDL; (2) cofactor activity
levels above which interference may occur. In general, in the enzyme lecithin:cholesterol acyl transferase
triglyceride interference is reported to occur at a (LCAT), which esterifies free cholesterol with fatty
concentration of 500 to 1,000 mg/dL and above. Other acids from phospholipids; (3) ability to remove excess
interferents include icterus (bilirubin > 30 to 60 mg/dL), cholesterol from tissues; and (4) ligand for HDL-
hemolysis (hemoglobin > 500 to 1,000 mg/dL), and receptor. These processes are important for the anti-
paraproteins, especially monoclonal IgM, may affect apo atherogenic reverse cholesterol transport pathway back
A-1 and apo B results by both ITA and INA methods. to the liver. In addition, apo A-1 seems to play a role in
prostacyclin stabilization, leading to improved
Reference Intervals and Interpretation vasodilation and the inhibition of platelet aggregation,
183
these functions playing an important role in the methods. Elevated apo B-100 concentration occurs in
protection against development of atherosclerosis. familial hypercholesterolemia, FCHL, familial
hyperapobetalipoproteinemia in which LDL-cholesterol
Elevated apo A-1 concentration occurs in familial levels are below the decision limit for risk of CHD, but
hyperalphalipoproteinemia, familial cholesteryl ester plasma apo B-100 and LDL-apo B-100 concentrations
transfer protein deficiency, which results in a doubling are elevated [25], in individuals with elevated Lp(a),
of apo A-1 concentration over normal, exercise, liver nephrotic syndrome, hypothyroidism, and biliary
disease, moderate alcohol intake, estrogens and obstruction, and is associated with increased risk for
pregnancy, and is associated with decreased risk for CHD and atherosclerosis. Decreased apo B-100
coronary heart disease (CHD) and atherosclerosis. concentration is found in abetalipoproteinemia due to an
Measurement of apo A-1 enables identification of absence of apo B in the intestine and liver, familial
patients with certain heritable abnormalities of hypobetalipoproteinemia in which plasma apo B-100 is
lipoprotein metabolism not detectable using other low but not as low as in homozygous
methods. Decreased apo A-1 concentration occurs in abetalipoproteinemia and is 25% to 50% of normal
familial hypoalphalipoproteinemia, familial combined levels in heterozygotes, in chronic anemia, liver disease,
hyperlipidemia (FCHL), fish-eye disease, which is due and neuromuscular degeneration.
to LCAT deficiency and apo A-1 concentrations 15% to
30% of normal, Tangier disease, cholestasis, acute Clinical Significance of Apolipoprotein A-1 and B
hepatitis, sepsis, and in insulin-treated diabetics. There is a considerable evidence base linking high levels
Decreased apo A-1 is generally associated with of certain plasma lipoproteins to accelerated
increased risk of CHD and atherosclerosis. Generally, atherogenesis and CHD. Apo A-1, the major protein in
the apo A-1 variants are associated with low HDL- HDL, and apo B-100, the major protein in LDL, serve as
cholesterol levels and are incomplete forms of the important predictors of cardiovascular disease risk. The
protein. Tangier disease, for example, is a rare non-HDL lipoproteins, VLDL, IDL, and LDL that
autosomal-recessive disorder with very low apo A-1, contain apo B-100, have been identified as the vehicles
total cholesterol, and HDL-cholesterol levels. Plasma in which cholesterol is transported into the artery wall
apo A-1 levels are 1% to 3% of normal in homozygotes and are regarded as atherogenic particles. Apo B-100
and half that of normal in heterozygotes [22]. In contrast, interacts with the mucopolysaccharide, chondroitin
the mutation called A-1Milano with low levels of apo A-1 sulfate B, at the artery wall and probably leads to in-situ
and HDL-cholesterol is associated with longevity, and degradation of the lipoproteins. Concentrations of apo B-
carriers rarely have coronary heart disease [23,24]. 100 and non-HDL cholesterol are closely correlated
[26].
Apo B is the main major protein moiety component of
all lipoproteins other than HDL, that is, LDL, Whereas plasma apo B-100 reflects LDL primarily, a
intermediate-density lipoprotein (IDL), very-low-density proportion of non-HDL cholesterol is found in VLDL
lipoprotein (VLDL) classes, and also chylomicrons. Apo and IDL. Measurement of apo B-100 reflects the number
B accounts for approximately 95% of the total protein of particles presenting cholesterol to the arterial wall and
content of LDL. It is important for the secretion of may give a better prediction of risk of vascular events
triglyceride-rich lipoproteins both from the liver and the than measurement of total cholesterol alone. LDL (and
intestine and is essential to the formation and release of thus the usual lipid parameters) is not a uniform
lipoproteins into the plasma. Apo B-100 acts as a lipoprotein but a very heterogeneous macromolecule.
structural protein for chylomicrons, VLDL, IDL, and Although LDL particles contain one molecule of apo B,
LDL, with a single molecule of apo B present in each they can vary in their cholesterol content by almost 25%,
particle. Lipoprotein (a) (Lp[a]) is also an apo B- resulting in large, buoyant particles enriched in
containing lipoprotein and will contribute to total apo B- cholesterol and small, dense LDLs that contain less
100 concentration, the amount depending on the Lp(a) cholesterol [27]. Small LDL particles can enter the
concentration (average apo B-100 is 5 mg/dL for normal arterial wall more easily than larger LDLs, they have a
Lp[a] levels). Unlike the smaller apolipoproteins, such as greater affinity for the glycoproteins of the arterial
apo A-1 and apo C, very-high-molecular-weight intima, and they are more susceptible to oxidation
apolipoproteins such as apo B-100 do not migrate from leading to enhanced macrophage uptake and foam-cell
one particle to another but remain bound during formation. Hence, by determining LDL-cholesterol, it is
assembly, secretion, and catabolism of the lipoprotein not possible to make assumptions regarding the total
particles. Apo B-100 interacts with the B/E (LDL) LDL mass or the total number of LDL particles.
receptors of peripheral cells, thereby functioning in the By contrast, the measurement of apo B-100 provides
recognition of cellular receptors for the catabolism of information regarding the number of LDL particles and
LDL. Synthesis of apo B-100 is regulated by steroid the relative density of this lipoprotein. Such information
hormone receptors and thyroid hormones and also by is useful, since small and dense LDLs (higher apo B
lipid-lowering drugs such as HMG-CoA reductase concentration) are more atherogenic than buoyant ones
inhibitors. (lower apo B concentration). In hypertriglyceridemic
patients with impaired glucose tolerance and type II
Measurement of apo B also enables identification of diabetes, and where total cholesterol is similar to the
patients with certain heritable abnormalities of non-diabetic population, apo B-100 can differentiate
lipoprotein metabolism not detectable using other
184
small, dense LDL enriched with cholesterol ester from treatment. The AMORIS study, with a mean follow-up
less atherogenic particles [27]. period of 66.8 months for 98,722 males and 64.4 months
for 76,831 women, demonstrated that the apo B/apo A-1
Apo B/apo A-1 Ratio ratio was significantly more informative of risk than the
Diagnostic Relevance of Apolipoprotein Measurements conventional cholesterol parameters [34]. Adding further
The Adult Treatment Program recommends LDL- weight to the predictive power of the apo B/apo A-1
cholesterol as the most clinically useful lipid parameter ratio is the data from the INTERHEART study, which
to assess risk of vascular disease and to guide treatment involved 30,000 subjects from 52 countries and showed
therapy. An alternate paradigm, based on atherogenic that the ratio accounted for approximately 50% of
lipoprotein particles, suggests that rather than equating cardiovascular events [39]. The apo B/apo A-1 ratio, or
risk due to LDL and the plasma cholesterol so-called Cholesterol Balance, indicates an increased
concentration of LDL-cholesterol, or to VLDL and risk of vascular events at values > 0.9 for men and > 0.8
triglyceride concentration, the lipoprotein atherogenic for women.
burden is best represented by total plasma apo B.
Current apo B assays measure the sum of apo B-100 and Apolipoprotein Performance Goals
apo B-48 lipoprotein particles, with approximately 90% Standardization Efforts Early standardization efforts
of plasma apo B derived from LDL and IDL particles occurred in the 1980s. Initially the Centers for Disease
and the remainder largely from VLDL. Chylomicrons Control and Prevention (CDC), in collaboration with the
and their remnants contribute minimally to total apo B. International Union of Immunological Societies,
undertook an apolipoprotein survey of 28 participating
The quantitative measurement of apo A-1 and apo B-100 laboratories and found that comparability of
is therefore an alternative to conventional lipid and measurements was 13% CV at 112.4 mg/dL apo A-1 and
lipoprotein analysis. Increasing evidence has 27% CV at 58.9 mg/dL apo B-100, respectively. Sources
accumulated linking apolipoproteins to atherosclerosis, of variation were different calibrators, matrix interaction,
indicating that their measurement is as reliable as and lack of assay linearity and parallelism [43].
measurements of lipoproteins for coronary artery disease
(CAD) risk assessment and response to lipid-lowering In a further international standardization study
therapy. Elevated concentrations of apo B-100 may be implemented by the International Federation of Clinical
found in normolipidemic patients with early CAD, even Chemistry and Laboratory Medicine (IFCC) and
when levels of total and LDL-cholesterol are normal. coordinated by the Northwest Lipid Research
Apo B-100 measurement may be more relevant to CAD Laboratories (NWLRL) in collaboration with diagnostic
risk assessment, because the amount of apo B-100 per manufacturers and research institutes, candidate
LDL particle is relatively constant, whereas that of reference materials and serum pools for apo A-1 and apo
cholesterol is variable [1,28]. Numerous prospective B-100 were tested for analytical performance,
epidemiological studies have shown apo B to be superior commutability of material, and comparability of method
to LDL-cholesterol in predicting the risk of comparison values before and after common calibration.
cardiovascular events and disease progression and in Whereas apo B-100 concentrations before the uniform
determining the adequacy of LDL-cholesterol lowering calibration showed a large among-assay variation
therapy [29-37]. (>19%) due to different assay standardization, this was
significantly reduced to a mean of 6% after uniform
Apo B seems to be useful for assessment of the calibration. Of the 26 reference materials for apo A-1
effectiveness of pharmacological treatment with lipid- and apo B proposed by the manufacturers, 15 for apo A-
lowering drugs. Statins lower both LDL-cholesterol and 1 and 11 for apo B, two lyophilized preparations for apo
apo B. However, there is faster removal of LDL- A-1 and two liquid-stabilized preparations for apo B
cholesterol from the circulation than number of LDL were selected for further evaluation in Phase 2 of the
particles, with triglyceride replacing some of the study [44]. Using these materials as the common
cholesterol in LDL particles. Use of apo B as a marker calibrator, results again showed that by uniform
for LDL particle number in the assessment of persons at calibration of 37 test systems, harmonization of results
a moderate risk of vascular events (i.e., at LDL- could be accomplished. Among-systems variation in apo
cholesterol concentrations below the 50th percentile) A-l levels improved from 9% to 5.4% using the common
may be a better discriminator than LDL-cholesterol [38]. calibrator and from ~ 20% to 7% for apo B. The
Whereas the relationship between risk and LDL- candidate reference materials, SP1 and SP3, gave
cholesterol concentration flattens out at lower levels, analytical characteristics closest to that of the serum
there is a linear relationship between risk and apo B pools, and there were no significant matrix effects in the
from the lowest to highest decile of apo-B concentration tested assays. Following recalibration, 50 fresh-frozen
according to the results of the AMORIS and serum samples covering a wide range of apo A-1 and
INTERHEART epidemiological health studies [34,39]. apo B-100 concentrations were analyzed to test for
comparability between various immunoassays. The
The ratio of apo B/apo A-1 allows those subjects with results obtained by the diagnostic companies were
and without CAD to be distinguished [25,40-42]. A high excellent, with an interlaboratory variation for results of
ratio of apo B to apo A-1 can be used as a marker of 2% to 6% for apo A-1 and 3% to 7% for apo B [6,7].
increased risk for CAD and a decreasing ratio as an
indicator of lowering of the risk following lipid-lowering
185
Between-Method Result Comparability Federation of Clinical Chemistry

The desirable quality specifications for precision, bias, standardization project for measurements of
and total allowable error for apo A-1 and apo B apolipoproteins A-I and B. IV. Comparability
measurement are an imprecision (CV) of 3.3% and 3.5% of apolipoprotein B values by use of
(based on 50% of the intraindividual biological international reference material. Clin Chem.
variation), a bias of 3.7% and 6.0%, and a total error of 1994;40:586-592.
9.1 % and 11.6% for apo A-1 and apo B, respectively 8. Marcovina SM, Gaur VP, Albers JJ. Biological
[8,45]. According to the 2007 American College of variability of cholesterol, triglyceride, low- and
Pathologists (CAP) survey of apolipoproteins A-1 and B, high-density lipoprotein cholesterol, lipoprotein
the number of laboratories measuring these proteins is (a), and apolipoproteins A-I and B. Clin Chem.
small compared with the number measuring lipids and 1994;40:574-578,
lipoproteins. Method principles are ITA and INA in 9. Riepponen P, Marniemi J, Rautaoja T.
approximately equal numbers. The measurement Immunoturbidimetric determination of
variation for apo A-1, at a mean level of 128.0 mg/dL, apolipoproteins A-1 and B in serum. Scan J
was 6.6% CV (range: 5.0% to 6.7% CV for all method Clin Lab Invest. 1987;47:739-744.
principles and 115 participating laboratories). For apo B, 10. Albers JJ, Marcovina SM. Standardization of
at a mean level of 71.1 mg/dL, the variation was 7.7% apolipoprotein B and A-I measurements. Clin
CV (range: 3.8% to 7.6% CV for 112 laboratories; Chem. 1989;35:1357-1361.
another 10 labs used Roche Cobas Integra, with a 10.6% 11. Albers JJ, Marcovina SM. Lipoprotein(a)
CV at a mean apo B concentration of 66.5 mg/dL). quantification: comparison of methods and
According to the CAP survey, not all apo A-1 and apo B strategies for standardization. Curr Opin
methods showed acceptable precision. Although current Lipidol. 1994;5:417-421.
assays are traceable to the certified secondary reference 12. Marcovina SM, Albers JJ. Apolipoprotein
materials for apolipoproteins, bias may contribute to assays: standardization and quality control.
concentration differences between methodsfor Stand J Clin Lab Invest. 1990;50(suppl
example, apo A-1 for the same sample as above ranged 198):58-65.
between 124.5 and 134.9 mg/dL (mean 128.0 mg/dL) 13. Dati F. Standardization of commercial assays
and apo B between 66.5 and 73.0 mg/dL (mean 71.1 for serum Apo A-I and Apo B: A consensus
mg/dL). procedure for the calibration of reference
material. Stand J Clin Lab Invest.
References 1990;50(suppl 198):73-79.
1. Bhatnagar D, Durrington PN. Measurement and 14. Bachorik PS, Lovejoy KL, Carroll MD,
clinical significance of apolipoproteins A-1 and Johnson CL. Apolipoprotein B and A-I
B. In: Rifai N, Warnick GR, Dominiczak MH, distributions in the United States, 1988-1991:
eds. Handbook of Lipoprotein Testing, 2nd ed. results of the National Health and Nutrition
Washington, DC: AACC Press; 2000:287-310. Examination Survey III (NHANES III). Clin
2. Bachorik PS, Kwiterovich PO Jr. Chem. 1997;43:2364-2378.
Apolipoprotein measurements in clinical 15. Contois JH, McNamara JR, Lammi-Keefe CJ,
biochemistry and their utility vis--vis Wilson PWF, Massov T, Schaefer EJ.
conventional assays. Clin Chim Acta. Reference intervals for plasma apolipoprotein
1988;178:1-34. A-I determined with a standardized commercial
3. Laker MF, Evans K. Analysis of immunoturbidimetric assay: results from the
apolipoproteins. Ann Clin Biochem. 1996;33:5- Framingham Offspring Study. Clin Chem.
22. 1996;42:507-514.
4. Labeur C, Sheperd J, Rosseneu M. 16. Contois JH, McNamara JR, Lammi-Keefe CJ,
Immunological assays of apolipoproteins in Wilson PWF, Massov T, Schaefer EJ.
plasma: methods and instrumentations. Clin Reference intervals for plasma apolipoprotein B
Chem. 1990;36:591-597. determined with a standardized commercial
5. Barr JR, Maggio VL, Patterson DG Jr, Cooper immunoturbidimetric assay: results from the
GR, Henderson LO, Turner WE et al. Isotope Framingham Offspring Study. Clin Chem.
dilution-mass spectrometric quantification of 1996;42:515-523.
specific proteins: model application with 17. Leino A, Impivaara O, Kaitsaari M, Jrvisalo J.
apolipoprotein A-I. Clin Chem 1996; 42: 1676- Serum concentrations of apolipoprotein A-I,
1682. apolipoprotein B, and lipoprotein(a) in a
6. Marcovina SM, Albers JJ, Henderson LO, population sample. Clin Chem. 1995;41:1633-
Hannon WH. International Federation of 1636.
Clinical Chemistry standardization project for 18. Graziani MS, Zanolla L, Righetti G, Marchetti
measurement of apolipoproteins A-I and B. III. C, Mocarelli P, Marcovina SM. Plasma
Comparability of apolipoprotein A-I values by apolipoproteins A-I and B in survivors of
use of international reference material. Clin myocardial infarction and in a control group.
Chem. 1993;39:773-781. Clin Chem. 1998;44:134-140.
7. Marcovina SM, Albers JJ, Kennedy H, Mei JM, 19. Jungner I, Marcovina SM, Walldius G, Holme
Henderson LO, Hannon WH. International I, Kolar W, Steiner E. Apolipoprotein B and A-
186
I values in 147576 Swedish males and females, the Qubec Cardiovascular Study. Circulation.
standardized according to the World Health 1996;94:273-8.
Organization-International Federation of 33. St-Pierre AC, Cantin B, Dagenais GR,
Clinical Chemistry First International Maurige P, Bernard PM, Desprs JP,
Reference Materials. Clin Chem. 1998;44:1641- Lamarche B. Low-density lipoprotein
1649. subfractions and the long-term risk of ischemic
20. Cooper GR, Myers GL, Smith SJ, Sampson EJ. heart disease in men: 13-year follow-up data
Standardization of lipid, lipoprotein, and from the Qubec Cardiovascular Study.
apolipoprotein measurements. Clin Chem. Arterioscler Thromb Vasc Biol. 2005;25:553-
1988;34:B95-105. 559.
21. Evans K, Laker MF. Intra-individual factors 34. Walldius G, Jungner I, Holme I, Aastveit AH,
affecting lipid, lipoprotein and apolipoprotein Kolar W, Steiner E. High apolipoprotein B, low
measurement: a review. Ann Clin Biochem. apolipoprotein A-1, and improvement in the
1995;32:261-280. prediction of fatal myocardial infarction
22. Francis GA, Knopp RH, Oram JF. Defective (AMORIS study): a prospective study. Lancet.
removal of cellular cholesterol and 2001;358:2026-2033.
phospholipids by apolipoprotein A-I in Tangier 35. Talmud PJ, Hawe E, Miller GJ, Humphries SE.
disease. J Clin Invest. 1995;96:78-87. Non-fasting apolipoprotein B and triglyceride
23. Franceschini G, Sirtori M, Gianfranceschi G, levels as a useful predictor of coronary heart
Sirtori CR. Relation between the HDL disease risk in middle-aged UK men.
apoproteins and AI isoproteins in subjects with Arterioscler Thromb Vasc Biol. 2002;22:1918-
the A-I Milano abnormality. Metabolism. 1923.
1981;30:502-509. 36. Jiang R, Schulze MB, Li T, Rifai N, Stampfer
24. Zambon A, Brown BG, Deeb SS, Brunzell JD. MJ, Rimm EB, Hu FB. Non-HDL cholesterol
Genetics of apolipoprotein B and and apolipoprotein B predict cardiovascular
apolipoprotein AI and premature coronary disease events among men with type 2 diabetes.
artery disease. J Intern Med. 2006;259:473-480. Diabetes Care. 2004;27:1991-1997.
25. Sniderman AD. Apolipoprotein B and 37. Thompson A, Danesh J. Associations between
apolipoprotein A-I as predictors of coronary apolipoprotein B, apolipoprotein AI, the
artery disease. Can J Cardiol. 1988;4:24A-30A. apolipoprotein B/AI ratio and coronary heart
26. Pischon T, Girman CJ, Sacks FM, Rifai N, disease: a literature-based meta-analysis of
Stampfer MJ, Rimm EB. Non-high density prospective studies. J Int Med. 2006;259:481-
lipoprotein cholesterol and apolipoprotein B in 492.
the prediction of coronary heart disease in men. 38. Mora S, Szklo M, Otvos JD, Greenland P, Psaty
Circulation. 2005;112:3375-3383. BM, Goff DC Jr et al. LDL particle subclasses,
27. Barter PJ, Ballantyne CM, Carmena R, Cabezas LDL particle size, and carotid atherosclerosis in
M, Chapman MJ, Couture P et al. Apo B versus the Multi-Ethnic Study of Atherosclerosis
cholesterol in estimating cardiovascular risk (MESA). Atherosclerosis. 2007;192:211-217.
and in guiding therapy: report of the thirty- 39. Yusuf S, Hawken S, Ounpuu S et al. Effect of
person/ten-country panel. J Int Med. potentially modifiable risk factors associated
2006;259:247-58. with myocardial infarction in 52 countries (the
28. Sniderman AD, Pederson T, Kjekshus J. Putting INTERHEART study): case-control study.
low-density lipoproteins at center stage in Lancet. 2004;364:937-952.
atherogenesis. Am J Cardiol. 1997;79:64-67. 40. Cannon CP, Braunwald E, McCabe CH, Rader
29. Stampfer MJ, Sacks FM, Salvini S, Willett WC, DJ, Rouleau JL, Belder R et al. Intensive versus
Hennekens CH. A prospective study of moderate lipid lowering with statins after acute
cholesterol, apolipoproteins, and the risk of coronary syndromes. N Engl J Med.
myocardial infarction. N Engl J Med. 2004;350:1495-1504.
1991;325:373-381. 41. Walldius G, Jungner I. The apoB/apo A-I ratio:
30. Coleman MP, Key TJA, Wang DY, Hermon C, a strong, new risk factor for cardiovascular
Fentiman IS, Allen DS et al. A prospective disease and a target for lipid-lowering therapy:
study of obesity, lipids, apolipoproteins, and a review of the evidence. J Int Med.
ischaemic heart disease in women. 2006;259:493-519.
Atherosclerosis. 1992;92:177-185. 42. Sniderman AD, Jungner I, Holme I, Aastveit A,
31. Sigurdsson G, Baldursdottir A, Sigvaldason H, Walldius G. Errors that result from using the
Agnarsson U, Thorgeirsson G, Sigfusson N. TC/HDL C ratio rather than the apoB/apo A-I
Predictive value of apolipoproteins in a ratio to identify the lipoprotein-related risk of
prospective study of coronary artery disease in vascular disease. J Int Med. 2006;259:455-461.
men. Am J Cardiol. 1992;69:1251-1254. 43. Smith SJ, Cooper GR, Henderson LO, Hannon
32. Lamarche B, Moorjani S, Lupien PJ, Cantin B, WH and the international collaborative study on
Bernard P-M, Gilles R et al. Apolipoprotein A-I standardization of apolipoproteins A-I and B.
and B levels and the risk of ischemic heart Part 1. Evaluation of a lyophilized candidate
disease during a five-year follow-up of men in
187
reference and calibration material. Clin Chem. 45. Fraser CG. Biological variation from principles
1987;33:2240-2249. to practice. Washington, DC: AACC Press;
44. Albers JJ, Marcovina SM, Kennedy H. 2001:139-141.
International Federation of Clinical Chemistry
standardization project for measurement of Acknowledgement
apolipoproteins A-I and B. part 2. evaluation Data on current external quality assessment were kindly
and selection of candidate reference materials. provided by Sharon Burr from the American College of
Clin Chem. 1992;38:658-662. Pathologists.
Tables
Table 1a: Apo A-1 Methods Summary Table
Method 1: End-point ITA with sample blanking and using polyclonal goat anti-human apo A-1 antibody
Principle of analysis: Light scattering by antigen/antibody complexes increases turbidity, which is
measured over a defined time interval
Comments: Serum and plasma (Li heparin, Na heparin, EDTA); avoid interference due to grossly
lipemic samples and paraproteins.
Method 2: End-point ITA with sample blanking and using polyclonal goat anti-human apo A-1 antibody
Principle of analysis: Same as Method 1
Comments: Serum and plasma-heparin; avoid interference due to grossly lipemic samples.
Method 3: End-point ITA with sample blanking and using polyclonal sheep anti-human apo A-1 antibody
Comments: Serum and plasma (Li, Na, NH4+-heparin, EDTA); avoid interference due to grossly
Method 4: End-point INA with sample blanking and using polyclonal rabbit anti-human apo A-1 antibody
Principle of analysis: Antigen/antibody complexes cause light scattering, with the scattered-light
intensity measured at an angle to the incident beam of light over a defined time interval
Comments: Serum is preferred; avoid interference due to particulate matter in the sample and from
grossly lipemic samples.
Method 5: Rate INA with sample blanking and using polyclonal goat anti-human apo A-1 antibody
Principle of analysis: Measures the rate of increase in light scattered by antigen/antibody complexes
after sample blanking check. Scattered-light intensity is measured at an angle to the incident beam of
light.
Comments: Serum is preferred; avoid interference due to particulate matter in the sample, grossly
lipemic samples, and IgM paraproteins.
188
Table 1b: Apo B-100 Methods Summary Table
Method 1: End-point ITA with sample blanking and using polyclonal goat anti-human apo B-100 antibody
Comments: Serum and plasma (Li heparin, Na heparin, EDTA); avoid interference due to grossly
Method 2: End-point ITA with sample blanking and using polyclonal goat anti-human apo B-100 antibody
Comments: Serum and plasma-heparin; EDTA plasma NOT RECOMMENDED; sample stable when
stored at 18C to 28C 1 day; 2C to 8C 3 days; 70C 1 year, but storage at 20C NOT
RECOMMENDED; avoid interference due to grossly lipemic samples.
Method 3: End-point ITA with sample blanking and using polyclonal sheep anti-human apo B-100 antibody
Comments: Serum and plasma (Li, Na, NH4+-heparin, EDTA); avoid interference due to grossly
Method 4: End-point INA with sample blanking and using polyclonal rabbit anti-human apo B-100 antibody
Principle of analysis: Antigen/antibody complexes cause light scattering with the scattered light
intensity measured at an angle to the incident beam of light over a defined time interval
Comments: Serum is preferred; avoid interference due to particulate matter in the sample and from
grossly lipemic samples.
Method 5: Rate INA with sample blanking and using polyclonal goat anti-human apo B-100 antibody
Principle of analysis: Measures the rate of increase in light scattered by antigen/antibody complexes
after sample blanking check. Scattered light intensity is measured at an angle to the incident beam of
light
Comments: Serum is preferred; avoid interference due to particulate matter in the sample, grossly
lipemic samples, and IgM paraproteins.
Avoid interference due to grossly lipemic samples and paraproteins.
189
Table 2a:Apo A-1 Reference Intervals and Cut-points
Country Method Age (y) Apo A-1 (mean SD) in mg/dL

(10th percentile)
Male n Female n
Finland [17] ITA Orion kit 27-67 136 22 286 158 29 289
(Kone Olli CD
analyzer) (109) (125)
Northern Italy Rate INA 49.8 16.0 141 21 298 156 23 315
[18] (Beckman Array
360) (116) (127)
Sweden [19] ITA Orion kit M: 46.8 13.2 136 22 83,112 151 24 64,46
(Kone C/D 4
analyzer) F: 49.5 15.2 (110) (122)
US ITA Orion kit M: 51.9 10.2 134 23 1,879 154 28 1,939

Framingham (Abbott Spectrum
[15,16] CCX analyzer) F: 51.3 10.6 (107) (122)
US NHANES RID and later rate 20 136 22 6,000 151 23 6,000

III [14] INA (Beckman
Protein System) (111) (120)
Table 2b:Apo B Reference Intervals and Cut-points
Country Method Age (y) Apo B-100 (mean SD) in mg/dL

(75th percentile)
Male n Female n
Finland [17] ITA Orion kit 27-67 121 32 286 109 32 289
(Kone Olli CD
analyzer) (141) (129)
Northern Italy Rate INA 49.8 16.0 111 29 298 110 35 315
[18] (Beckman Array
360) (129) (131)
Sweden [19] ITA Orion kit M: 46.8 13.2 131 35 83,112 122 36 64,464
(Kone C/D
analyzer) F: 49.5 15.2 (153) (142)
US ITA Orion kit M: 51.9 10.2 103 24 1,880 96 26 1,944

Framingham (Abbott Spectrum
[15,16] CCX analyzer) F: 51.3 10.6 (118) (111)
US NHANES RID and later rate 20 107 25 6,000 103 28 6,000

III [14] INA (Beckman
Protein System) (122) (119)
190
Aspartate Aminotransferase
James J. Miller
Name: Aspartate aminotransferase, AST, L-aspartate:2-oxoglutarate

aminotransferase, formerly serum glutamate oxaloacetic transaminase (SGOT)
Molecular weight: Cytoplasmic: 93,000 D; mitochondrial: 90,400 D
Biochemical reaction: Amino transfer catalyzed by AST (see Figures at end of text)
Methods: CLSI RS2-A
great excess when the patient specimen has low AST

Principles of Analysis and Current Usage i activity), little error is introduced.
Aspartate aminotransferase (AST) catalyzes the transfer of
an amino group from specific amino acids (L-glutamate or In the second approach, first described by Karmen [2],
oxaloacetate is again the product quantitated. Malate
L-aspartate) to specific ketoacids (-ketoglutarate or
dehydrogenase (MDH) is used to convert the oxaloacetate
oxaloacetate) (see biochemical reaction in Figures at end
to malate (Table 1, Method 2). The decrease in absorbance
of text). Although at physiological pH, the reaction is
at 340 nm, caused by NADH (reduced nicotinamide
energetically favored toward the formation of L-aspartate
adenine dinucleotide) consumption in the second reaction,
and -ketoglutarate (to the left, as in the equation), in vivo
is used to follow the course of the AST reaction. More
the reaction goes to the right to provide a source of
than 80% of reporting survey laboratories use the Karmen
nitrogen for the urea cycle. The glutamate thus produced is
reaction.
deaminated by glutamate dehydrogenase, resulting in
ammonia and regeneration of -ketoglutarate. A number of photometric techniques have been described,
all of which followed a similar principlethe coupling of
Two methods of AST analysis have enjoyed the widest an aryldiazonium salt to the active methylene of
popularity. The coupling of oxaloacetate with 2,4- oxaloacetate (Table 1, Method 3) [3]. In all cases, the aryl
dinitrophenylhydrazine (DNPH) to produce a blue substitute (Ar) is a dye with absorbance characteristics that
hydrazone (Method 1, AST Methods Summary Table) was change with the formation of the azo bridge to
introduced in 1957 [1] and gained popularity because of its oxaloacetate. The reaction is thus monitored by
simplicity and relative accuracy. In this method, as in all measurement of the increase in absorbance of the
others, it is necessary to force the AST reaction toward the diazonium reaction product.
energetically less favored products, of which only
oxaloacetate is easily quantitated. Thus L-aspartate and - A colorimetric method for AST is also available. This
ketoglutarate are added in excess. After incubation, the method is employed in one of the dry-film assays available
reaction is stopped with the addition of DNPH, which for AST measurement (Table 1, Method 4). In this
reacts with both -ketoglutarate and oxaloacetate. After procedure, oxaloacetate formed in the deamination of L-
condensation, the pH is made alkaline, and the absorbance aspartate is converted to pyruvate and carbon dioxide by
at 505 nm is measured. Because the absorptivity of the oxaloacetate decarboxylase. The pyruvate is next oxidized
oxaloacetate hydrazone is much greater than the by pyruvate oxidase to acetyl phosphate and hydrogen
corresponding hydrazone of -ketoglutarate (present in peroxide. The final indicator reaction step involves the
oxidation of a leuco dye by peroxidase to produce a
iAST colored dye. Samples that contain increased concentrations
of pyruvate will result in substrate depletion and must be
Previous and current authors of this method: diluted. The thin-film assays are also interfered with by the
First edition: Robert L. Murray presence of increased protein concentrations, especially
Methods edition: Robert L. Murray samples with increased immunoglobulin concentrations
Second edition: Robert L. Murray such as those found in patients with multiple myeloma [4].
Third edition: Steven C. Kazmierczak
Fourth edition: Steven C. Kazmierczak Identification and quantitation of AST isoenzymes has
been reported [5-7]. The catalytic and physicochemical
Fifth edition: James J. Miller
properties of mitochondrial and soluble (cytoplasmic) AST
191
are compared in Table 3 below. Measurement of AST the preincubation period, eliminating interference during
isoenzymes has not become routine. However, several the measurement period.
reports have suggested clinical utility in the differential
diagnosis of liver disease and for timing of acute Some older reference methods based on the Karmen
myocardial infarctions [8-11]. coupled enzymatic method used phosphate buffer, which
retards the recombination of added pyridoxal phosphate
Reference and Preferred Methods with the apoenzyme. If a large amount of the inactive
There have been many minor modifications in the enzyme is present, a falsely low activity will be observed.
enzymatic technique since its introduction by Karmen Activation of apoenzyme is more efficient in Tris buffer.
[2,12]. In addition, the specifications listed in the reference However, NADH is somewhat less stable in Tris buffer
method published by the International Federation of than it is in phosphate buffer. For this reason, the Tris
Clinical Chemistry (IFCC) have been modified over the concentration is kept relatively low at 80 mmol/L.
years, the most recent modification in 2002 being to
optimize conditions for 37C [13-15]. The main use of the Most routine methods today use the Karmen coupled
IFCC reference method is to assay calibrators for use in enzymatic method and may be traceable to the current
routine methods. This allows routine methods to be IFCC reference method.
traceable to the IFCC reference method, with the goal of The American Association for Clinical Chemistry
reducing the biases between methods. proposed a method for the small clinical chemistry
laboratory that differs from the IFCC in that (1) a single
In the most recent IFCC procedure (2002) [15], 0.20 mL of reagent is used to avoid a two-step addition procedure, (2)
serum is preincubated for 5 minutes in 2.00 mL of a the reaction is read after 150 seconds for the following 180
mixture that contains all reactants except -ketoglutarate. seconds, and (3) pyridoxal phosphate is not added [17].
During this preincubation period, the added lactate
dehydrogenase (LD) rapidly converts the endogenous The drawbacks of the dinitrophenylhydrazine method are
pyruvate in the serum to lactate, and the pyridoxal that (1) the oxaloacetate produced by the reaction results in
phosphate cofactor joins with any inactive apoenzyme to feedback inhibition of AST, and thus specimens exhibiting
form an increased amount of active AST. With the high activity are spuriously lowered; and (2) any ketone in
addition of 0.20 mL of -ketoglutarate, the primary serum can react, though most do not result in an
reaction is initiated, and the concentrations shown in Table absorbance change in the region measured. However,
2 are reached, exclusive of the small increases caused by acetoacetic acid and hydroxybutyric acid, both components
the presence of endogenous material in the serum. After of ketosis, do cause false elevations. The diazonium
steady state is reached, the rate of NADH oxidation is method similarly is limited by nonspecificity in that any
monitored repeatedly at 339 nm. The rate of change in activated methylene group will react. This is seen in
absorbance is corrected for a reagent blank. patients on certain drugs and those with renal failure.
Most reference methods, including the current IFCC The U.S. National Institute of Standards and Technology
procedure [15], have included pyridoxal phosphate. The (NIST) Standard Reference Material No. 909b is a
need for addition of this cofactor has been widely debated. lyophilized human serum preparation intended for use in
The essential nature of the cofactor has long been evaluating the accuracy of reference methods. It is
recognized, but because it is usually present in human available to manufacturers and laboratories for the
serum in adequate amounts, many investigators do not add validation of AST methods.
this component to the reaction mixture. In the occasional
patient with severe vitamin B deficiency, this could lead to Specimen
a serious underestimation of AST activity. Addition of Serum or plasma may be used. Heparin, oxalate, EDTA,
pyridoxal phosphate results in an increased activity, and citrate do not cause enzyme inhibition [18].
reportedly ranging up to 50% [16]. The higher values Anticoagulants with ammonium as the cation should be
might be expected in severely ill patients, especially if avoided to reduce the possibility of error. Because of the
elevated AST activity is present, and in patients suffering high activity of AST in red blood cells, hemolyzed
from pyridoxal phosphate deficiency. samples are unacceptable. For the colorimetric thin-film
procedure, serum or heparinized plasma only are
The presence of aminotransferases in the reagents is acceptable specimens.
possible if the LD is not carefully prepared. Good-quality
enzymes will not pose a problem; in any event, a blank Interferences
determination will identify the problem. The presence of Potential sources of error in the enzymatic method include
pyruvate is a potential source of error, since in the endogenous serum glutamate dehydrogenase,
presence of endogenous LD, pyruvate will be converted to aminotransferases (in the reagent preparation), very high
lactate with simultaneous consumption of NADH. This serum pyruvate concentration, or the use of phosphate
problem is circumvented by the addition of a large excess buffer. Glutamate dehydrogenase is present in some
of LD, so that endogenous pyruvate is converted during pathological sera, particularly in serum from patients with
192
liver disease. This enzyme introduces error by catalyzing Summary Report (C-B) are all <5%, indicating that this
the formation of glutamate with NH4+ to form - target is easily met using current AST methods.
ketoglutarate, with reduction of NADH to NAD+. Since
this source of error is possible only when ammonium ions References
are present, the IFCC method eliminates all deliberate 1 Reitman S, Frankel S. A colorimetric method for
addition of these. (Older methods sometimes included the determination of serum glutamic oxalacetic
ammonium sulfate in the formulation.) and glutamic pyruvic transaminase. Am J Clin
Pathol 1957; 28: 56-63.
Aspartate Aminotransferase Reference Interval 2 Karmen A. A note on the spectrophotometric
IFCC, with pyridoxal phosphate, 37C assay of glutamic oxalacetic transaminase in
human blood serum, Appendix: p. 131-3 in
Adults [19]: M: 14-45 U/L Karmen A, Wroblewski F, LaDue JS.
F: 13-37 U/L Transaminase activity in human blood. J Clin
Invest 1955; 34: 126-33.
Infants: AST activity in neonates and infants is 3 Sax SM, Moore JJ. Determination of glutamic
approximately twice that in adults, but these decline to oxalacetic transaminase activity by coupling of
adult levels by approximately 6 months of age [20,21]. oxalacetate with diazonium salts. Clin Chem
1967; 13: 175-85.
Interpretation 4 Tang M, Sullivan M, Gibson D, Truskolawski C.
There are two forms of ASTmitochondrial and soluble Kinetic error on Kodak Ektachem: A clue in
forms [9,12]. The isoenzymes have significant similarities diagnosis of myeloma. Clin Chem 1994; 40: 166.
in amino acid sequence (approximately 50%) and in 5 Bodansky O, Schwartz MK, Nisselbaum JS.
active-site positions. Each enzyme comprises two identical Isozymes of aspartate aminotransferase in tissues
subunits. The molecular weights of these dimers are and blood of man. Adv Enzyme Regul 1966; 4:
93,000 for the soluble and 90,400 for the mitochondrial 299-315.
isoenzyme. A summary of the physical and catalytic 6 Sampson EJ, Hannon WH, McKneally SS,
properties of these isoenzymes is presented in Table 3. The McKenzie C, Miller SA, Whitner VS, Burtis CA.
mitochondrial form of the enzyme accounts for only about Column chromatography and immunoassay
12% of the total activity in normal serum. compared for measuring the isoenzymes of
aspartate aminotransferase in serum. Clin Chem
The tissue distribution of total AST and the ratio of the 1979; 25: 1691-6.
activity of AST in each tissue to the AST activity in serum 7 Rej R, Bretaudiere J, Graffunder N. Measurement
are presented in Table 4 [22,23]. Per gram of wet tissue, of aspartate aminotransferase isoenzymes: six
heart, liver, skeletal muscle, and kidney have similar procedures compared. Clin Chem 1981; 27: 535-
amounts of AST. While enzyme release is not specific, 42.
high enzyme activity does indicate tissue damage. The 8 Nalpas B, Vassault A, Le Guillou A, Lesgourgues
mitochondrial form represents 81% of the total AST B, Ferry N, Lacour B, Berthelot P. Serum activity
present in human liver [24]. of mitochondrial aspartate aminotransferase: a
sensitive marker of alcoholism with or without
Alanine aminotransferase (ALT) activity is usually higher alcoholic hepatitis. Hepatology 1984; 4: 893-6.
than AST in most types of liver disease, in which the 9 Panteghini M. Aspartate aminotransferase
activity of both enzymes is predominantly from the isoenzymes. Clin Biochem 1990; 23: 311-9.
hepatocyte cytosol. When liver necrosis is substantial, as 10 Panteghini M, Pagani F, Cuccia C. Effects of
in individuals with alcoholic and viral hepatitis, therapeutic coronary reperfusion on aspartate
mitochondrial AST is also released into the blood, and aminotransferase isoenzymes in sera of patients
AST activity is usually higher than ALT. The ratio of AST with acute myocardial infarction. Clin Chem
to ALT, sometimes called the De Ritis Ratio, is often used 1989; 35: 909-12.
to evaluate alcoholic liver disease [25] and severity of liver 11 Yoneda K, Katayama Y, Koike T, Tanimizu I. A
disease in viral hepatitis [26,27]. This ratio is only homogeneous assay system of aspartate
pertinent in isolated liver disease when comorbidities that aminotransferase iso-enzymes using proteases
increase AST activity are not present. and application for clinical evaluation of
myocardial infarction. J Clin Lab Anal 1992; 6:
Aspartate Aminotransferase Performance Goals 362-7.
The current target for total error in AST measurements is 12 Rej R. Measurement of aminotransferases: Part 1:
20% (CLIA). Biological variation data suggest that in Aspartate aminotransferase. Crit Rev Clin Lab Sci
patients with stable AST activities, a total error of <15% is 1984; 21: 99-186.
required for optimum use [28]. The interlaboratory 13 Bergmeyer HU, Bowers GN Jr., Horder M, Moss
coefficients of variation listed for various AST methods in DW. Provisional recommendations on IFCC
the 2007 College of American Pathologists Participant methods for all measurements of catalytic
193
concentrations of enzymes, part 2, IFCC method in blood of adult females and males measured in
for aspartate aminotransferase. Clin Chem 1977; accordance with the International Federation of
23: 887-99. Clinical Chemistry reference system at 37C: part
14 Bergmeyer HU, Horder M, Rej R. Approved of the Nordic Reference Interval Project. Scand J
recomendation (1985) on IFCC methods for the Clin Lab Invest 2004; 64: 371-84.
measurement of catalytic concentrations of 20 Meites S, editor. Pediatric clinical chemistry.
enzymes, Part 2. IFCC method for aspartate Washington, D.C.: American Association for
aminotransferase (L-aspartate: 2-oxoglutarate Clinical Chemistry: 1981. p. 117-20.
aminotransferase, EC 2.6.1.1). J Clin Chem Clin 21 Soldin SJ, Hicks JM, editors. Pediatric reference
Biochem 1986; 24: 497-510. ranges. Washington, D.C.: AACC Press: 1995.
15 Schumann G, Bonora R, Ceriotti F, Ferard G, 22 Wroblewski F. Serum transaminase activity. Adv
Ferrero CA, Franck PF, et al. IFCC primary Clin Chem 1958; 1: 313-51.
reference procedures for the measurement of 23 Whitby LG, Percy-Robb IW, Smith AF. Lecture
catalytic activity concentrations of enzymes at notes on clinical chemistry. 3rd ed. London:
37C, part 5, reference procedure for the Blackwell Scientific Publications; 1984. p. 141.
measurement of catalytic concentration of 24 Rej R. Aspartate aminotransferase activity and
aspartate-aminotransferase. Clin Chem Lab Med isoenzyme proportions in human liver tissues.
2002; 40: 725-33. Clin Chem 1978; 24: 1971-9.
16 Bruns DE, Savory J, Titheradge AC, Cross RE, 25 Majhi S, Baral N, Lamsal M, Mehta KD. De Ritis
Wills MR. Evaluation of the IFCC-recommended ratio as diagnostic marker of alcoholic liver
procedure for serum aspartate aminotransferase as disease. Nepal Med Coll J 2006; 8: 40-2.
modified for use with a centrifugal analyzer, Clin 26 Giannini E, Risso D, Botta F, Chiarbonello B,
Chem 1981; 27: 156-159. Fasoli A, Malfatti F, et al. Validity and clinical
17 Butler TJ, Civantos F, Steele BW, Airan RS, utility of the aspartate aminotransferase-alanine
Klotzsch SG. Aspartate aminotransferase, AST aminotransferase ratio in assessing disease
(provisional). In: Faulkner WR, Meites S, editors. severity and prognosis in patients with hepatitis C
Selected methods of clinical chemistry. virus-related chronic liver disease. Arch Intern
Washington, D.C.: American Association for Med 2003; 163: 218-24.
Clinical Chemistry; 1982. p. 101-7. 27 Giannini EG, Zaman A, Ceppa P, Mastracci L,
18 Demetriou JA, Drewes PA, Gin JN. Enzymes. In: Risso D, Testa R.. A simple approach to
Henry RJ, Cannon DC, Winkelman JE, editors. noninvasively identifying significant fibrosis in
Clinical chemistry: principles and technics. 2nd ed. chronic hepatitis C patients in clinical practice. J
Hagerstown, MD: Harper & Row; 1974. p. 815- Clin Gastroenterol 2006; 40: 521-7.
1001. 28 Fraser CG. Biological variation: from principles
19 Stromme JH, Rustad P, Steenland H, Theodorsen to practice. Washington, DC: AACC Press; 2001,
L, Urdal P. Reference intervals for eight enzymes p. 140.
Table 1: Methods of Aspartate Aminotransferase (AST) Analysis

Method 1: Dinitrophenylhydrazine coupling (colorimetric) (Reitman and Frankel [1]); quantitative
Principle of analysis*: click here
Comments: Serum
Method 2: Enzymatic (ultraviolet monitoring) (Karmen [2]); quantitative
Comments: Serum
Method 3: Diazonium dye coupling (colorimetric); quantitative
Comments: Serum
Method 4: Enzymatic leuco dye (visible); quantitative
_
Aspartate + -Ketoglutarate ASTOxaloacetate + Glutamate
Oxaloacetate Oxaloacetate decarboxylasePyruvate + CO2
_
Pyruvate + Phosphate + O2 Pyruvate oxidase H2O2 + Acetyl phosphate
_
H2O2 + Leuco dye Peroxidase Colored dye
Comments: Infrequently used *Asp, Aspartic acid; -KG, -ketoglutaric acid; Gl, glutamic acid; Oa, oxaloacetic acid
194
Table 2: Conditions of the 2002 IFCC AST Reference Method

Component/Condition Concentration/Value
L-aspartate (mmol/L) 240
2-Oxoglutarate (mmol/L) 12
Buffer & concentration (mmol/L) Tris 80
pH 7.65
Pyridoxal phosphate (mmol/L) 0.1
NADH (mmol/L) 0.18
MDH (U/L) 600
LDH (U/L) 900
Volume fraction (v/v) 0.0833
Temperature (C) 37.0
Wave length (nm) 339
Band width (nm) 2
Light path (mm) 10
Incubation time (s) 300
Delay time (s) 90
Measurement interval (s) 180
Readings (measurement points) 6
Table 3: Properties of AST Isoenzymes

Mitochondrial Soluble
Catalytic
Km of oxaloacetate 9.0 10-6 3.7 10-5
Km of glutamate 3.8 10-2 1.0 10-2
Km of oxoglutarate 5.6 10-3 1.3 10-4
Km of L-aspartate 4.0 10-4 2.0 10-3
pH optimum Approximately 7.0 Approximately 8.0
Physicochemical
Molecular weight 90,400 D 93,000 D
Isoelectric point >9.0 5.15 to 5.80
Modified from Rej R: Aspartate aminotransferase, Crit Rev Clin Lab Sci 21:99-186, 1984.
Table 4: Tissue Distribution of AST in Normal Human Adult Tissues

Ratio of activity
-3
U 10 /g in tissue
Tissue wet tissue homogenate to that in serum
Heart 156 7800
Liver 142 7100
Skeletal muscle 99 4950
Kidney 91 4550
Pancreas 28 1400
Spleen 14 700
Lung 10 500
Plasma 0.02 1
From Wrblewski F: Serum transaminase activity, Adv Clin Chem 1:313-351, 1958.
Figures
195
Figure 1: Amino transfer catalyzed by AST.
Figure 2: Reaction of oxaloacetate with dinitrophenylhydrazine as in Method 1.
Figure 3: Conversion of oxaloacetate to malate with malate dehydrogenase as in Method 2.
Figure 4: Diazonium dye coupling to oxaloacetate as in Method 3.
Procedure: IFCC Recommended Procedure[15] Principle

In the presence of aspartate aminotransferase, L-
196
aspartate and 2-oxoglutarate exchange an amino group to (Solution 1) with 0.2 mL of pyridoxal phosphate (solution
form oxaloacetate and L-glutamate. The rate of this 3), 0.2 mL of NADH (Solution 4), and 0.1 mL of the
reaction is monitored by use of an indicator reaction in enzymes (Solution 5). Mix thoroughly and in a dark bottle.
which the oxaloacetate formed is converted to malate by Stable for 1 day at 2C to 8C.
an excess of malate dehydrogenase. The decrease in
absorbance at 339 nm is monitored as the NADH is 7. Start reagent solution. Dissolve 326 mg of 2-
consumed. The IFCC reference method is primarily used oxoglutaric acid, disodium salt in 10.0 mL of distilled
to assay calibrators to allow traceability of routine methods water. Stable for 1 week at 2C to 8C.
to the reference method.
Assay
Reagents Equipment: Spectrophotometer with 2 nm band
1. Tris-aspartate buffer (96.92 mmol/L Tris, pass at 339 nm with a constant temperature cuvette capable
302.4 mmol/L-aspartic acid, pH 7.65). In a 100 mL of maintaining a constant temperature with less than 0.1C
volumetric flask, dissolve 1.17 g of fluctuation. A recording spectrophotometer is preferable.
Tris(hydroxymethyl)aminomethane, 4.02 g of L-aspartic 1. Add 2 mL of reaction solution and 0.2 mL of
acid, and 0.052 g sodium azide in 80 mL of distilled water. serum to the cuvette.
Adjust to pH 7.65 with 2 mol/L NaOH. Make up to 100 2. Mix and allow to stand for 5 minutes.
mL with distilled water. Stable for 3 months at 2C to 8C.
3. Add 0.2 mL of start reagent solution
This is Solution 1.
5. Mix, wait 90 s, and record the change in
2. Tris-HCl buffer (96.92 mmol/L Tris, pH 7.65). absorbance for an additional 180 s. If no recorder
In a 100 mL volumetric flask, dissolve 1.17 g of is available, record at least every 30 s.
tris(hydroxymethyl)aminomethane and 0.052 g sodium 6. A reagent blank is run by replacing the sample
azide in 80 mL of water, adjust to pH 7.65 with 1 mol/L with 9 g/L sodium chloride in step 1.
HCl, and bring to 100 mL volume with water. Stable for 3 7. A sample blank should be checked by replacing
months at 2C to 8C. This is Solution 2. the start reagent in step 3 with 9 g/L sodium
chloride. The sample blank is not included in the
3. Pyridoxal phosphate, 6.3 mmol/L. Dissolve calculation of AST activity, but a rate >1% of the
16.7 mg of pyridoxal- AST activity indicates that the material is not
5-phosphoric acid monohydrate in 10.0 mL of the Tris- suitable as a calibrator. For the reagent blank of
HCl buffer. Store in a dark bottle at 2C to 8C. Stable for the sample blank, replace both the start reagent
1 week. This is Solution 3. and the sample with 9 g/L sodium chloride.
Calculations
4. NADH, 11.34 mmol/L. Dissolve 16 mg of - If the change in absorbance per second (A/t) is
nicotinamide adenine dinucleotide reduced from disodium greater than 0.0022 per second, dilute the sample 5- to 10-
salt in 2.0 mL Tris-HCl buffer. Stable for 1 week at 2C to fold with 9 g/L NaCl, and repeat the measurement.
8C. This is Solution 4. Correct for the blank reaction using the following
schema:
Diluent for reagent enzymes. Dissolve 1.2 g bovine
serum albumin and 0.9 g NaCl in 100 mL of water. Stable
(A/t)A = Measured reaction rate with serum in step 1
at least 1 month at 2C to 8C.
(A/t)B = Measured reaction rate with NaCl in place of
serum in step 1
5. Malate dehydrogenase/lactate dehydrogenase
(1.26 mkat/L [75.6 kU/L] and 1.89 mkat/L [113.4 kU/L]
at 37C, respectively). Separately dilute stock MDH and A/tAST = (A/t)A - (A/t)B and
LDH to twice these activities with diluent for reagent
enzymes, and mix equal volumes of the two enzyme AST activity = 1905 x A/tAST kat/L
solutions. Stable for 2 days at 2C to 8C. This is Solution
5. AST activity in kat/L can be converted to U/L by
multiplying by 60.
6. Reaction solution. Mix 10 mL of Tris-aspartate
197
B-Type Natriuretic Peptide, NT-proBNP and ProBNP

Alan H.B. Wu i
Name: B-type natriuretic peptide, NT-proBNP, and proBNP

Clinical significance: See Chapter 28, Physiology of Body Water and Electrolytes, and Chapter 36,
Cardiac and Muscle Disease, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular mass:
Chemical class: Peptide hormone, prohormone metabolite and prohormone
Structure: (See Figure below)
The enzymatic cleavage of proBNP into BNP (active hormone, 32 amino acids) and NT-proBNP
(inactive peptide, 76 amino acids) by corin. Used with permission from Omland T, Hall C: N-terminal
pro-B-type natriuretic peptide. In Wu AHB: Cardiac Markers, ed 2, Totowa, NJ, 2003, Humana Press.
Figure 1
The enzymatic cleavage of proBNP into BNP (active hormone, 32 amino acids) and NT-proBNP
(inactive peptide, 76 amino acids) by corin. Used with permission from Omland T, Hall C: N-
terminal pro-B-type natriuretic peptide. In: Wu AHB: Cardiac Markers, ed. 2, Totowa, NJ, 2003,
Humana Press.
Principles of Analysis and Current Usage natriuretic peptide, and urodilantin. Pro-BNP is a 108
B-type natriuretic peptide (BNP) is a member of a family amino acid precursor protein of BNP found in the
of natriuretic peptides that include atrial natriuretic ventricles and atria of the heart. The gene for proBNP is
peptide (ANP), C-type natriuretic peptide, D-type found on chromosome 1 [1]. In situations of volume
i
New method
Fifth edition: Alan H.B. Wu
198
overload and myocardial muscle stretching, proBNP is need to measure both assays on the same patient. For
upregulated at the genomic level. A small amount of more information, see the Interpretation section.
BNP is found in cytoplasmic granules. Upon stimulation
for production and release, ProBNP is cleaved by corin Currently there are no commercial assays approved by
to equimolar amounts of the biologically active BNP the U.S. Food and Drug Administration (FDA) for
hormone and the biologically inactive amino-terminal proBNP. However, one company (Bio-Rad Laboratories)
proBNP (NT-proBNP) (Figure 1). ProBNP is also has produced an assay that is currently undergoing
released into the circulation as the unprocessed precursor clinical evaluations [7]. The specificity of this assay is
protein. When the effects of this peptide on endothelial dependent on the use of a monoclonal antibody directed
and vascular smooth muscle cells [2] and cardiac towards the hinge-76 region of the peptide, an epitope
fibroblasts and myocytes [3] were evaluated, this peptide present in proBNP and absent in both BNP and NT-
did not appear to have any biological activity. proBNP (Table 1). The second antibody is unspecified.
At this point, it is unclear whether or not assays for
BNP is a 32amino acid peptide hormone with a proBNP will be superior, equivalent, or provide
molecular weight of 3472 daltons. The molecule consists supplementary information to that currently provided by
of an N-terminal tail of 9 amino acids, a C-terminal tail BNP and/or NT-proBNP. There are likely differences in
of 6 amino acids, and a 17-member ring that is closed by the release of proBNP from patients with heart failure
a disulfide bond between two cysteine residues. Eleven for example, there are genetic polymorphisms in the
of the amino acids found in the ring are conserved across gene encoding corin [8]. These and other questions will
the family of natriuretic peptides. BNP binds to be addressed in upcoming clinical studies.
natriuretic peptide receptors A and C, causing the loss of
water and electrolytes, vasodilation, and inhibition of the Reference and Preferred Methods
renin/aldosterone axis. BNP is cleared from the There are no reference materials or reference methods
circulation by neutral endopeptidases. The first two currently available for BNP, NT-proBNP or proBNP.
amino acids from the N-terminal sequence (serine and For BNP, the International Federation of Clinical
proline) are removed by dipeptidyl peptidase IV very Chemistry Committee on the Standardization of Markers
soon after BNP is released into the circulation [4]. BNP of Cardiac Damage has undertaken the responsibility of
has an estimated half-life of about 20 minutes in humans proposing a secondary reference standard from candidate
[5]. The amino-terminal proBNP (NT-proBNP) is a 76 materials for future standardization consideration by
amino acid, biologically inactive peptide with a manufacturers. From there, development of a primary
molecular weight of 8460 daltons (Figure 1). Although standard would be possible. There is no current effort for
not exactly known, NT-proBNP has a much longer half- developing a reference method for BNP or a reference
life than BNP, estimated to be between 1 and 2 hours standard or method for NT-proBNP or proBNP. Due to
[5]. This peptide exists as a straight chain and does not matrix effects, a reference method will likely require
contain any disulfide linkages. mass spectrometry. Reference methods for NT-proBNP
and proBNP are less of an issue, owing to the ownership
All assays for BNP, NT-proBNP, and proBNP are of the intellectual property rights for clinical use of these
measured by dual-site sandwich immunoassays. There is assays by Roche and Bio-Rad, respectively (see previous
no standardization between manufacturers for BNP with section).
regard to results or the antibodies used in the assays. The
Beckman BNP assay uses the same antibodies and is All clinical assays for BNP and NT-proBNP are two-site
standardized to the Inverness (Biosite) Medical Triage sandwich immunoassays and are acceptable methods.
point-of-care BNP assay, and the Siemens assay uses the The use of two antibodies improves the specificity of the
same antibodies as the Shionogi assay (Table 1). Most assay and affords freedom from cross-reactivity from
manufacturers have assigned a cutoff value of 100 ng/L other structurally related peptides. Radioimmunoassay
as the cutoff for diagnosis of heart failure. Figure 1 techniques (e.g., Shionoria BNP) are acceptable for
shows the epitopes for the capture and detection research purposes but not for routine clinical practice.
antibodies for some of the commercially-approved BNP
tests [6]. For NT-proBNP, both the assay and antibodies Interferences
are standardized to the Roche Diagnostics Elecsys As with other immunoassays, the specificity of the
platforms. This assay uses one antibody directed towards antibodies used in the tests determine the specificity of
the N-terminus of the peptide and the other towards the the assay and freedom from interferences. There are
central portion (Table 1). The intellectual property rights three classes of compounds that have been studied with
for use of NT-proBNP belong to Roche. As a condition reference to BNP and NT-proBNP: structurally similar
for acquiring a license for clinical use, all manufacturers peptides, precursor proteins, and other neurohormones;
of NT-proBNP assays must use the Roche antibodies drugs used in heart failure therapy; and general
and calibrators. photometric interferents. The peptides and proteins that
have been demonstrated to have no interference in BNP
There have been many studies that have documented the and NT-proBNP assays include renin, aldosterone,
clinical equivalence of BNP and NT-proBNP for patients angiotensin I, II, and III, endothelin, adrenomedullin,
with heart failure. Although there are minor differences and vasopressin. Other natriuretic peptides that have
in the performances of these tests, there is no current significant homology in amino acid sequences found in
199
the ring structure (ANP, preproANP, CNP, and relative composition of blood with regard to the various
urodilantin) also do not exhibit cross-reactivity. Selected BNP, NT-proBNP, proBNP, and glycosylated forms
antibodies towards the non-homologous amino acids in present in the blood of patients with heart failure is not
the ring structure are essential in ensuring the required known. Therefore, comparing results of different
assay specificity. It is also important to demonstrate that markers and different assays of the same marker will be
antibodies used in the BNP assays do not exhibit any difficult.
cross-reactivity with NT-proBNP and vice versa.
Because there are no amino acid sequences that are in Specimen
common to these peptides (Figure 1), cross-reactivity is The in vitro stability of BNP and NT-proBNP has been
not expected. Cardiac drugs that show no interference in extensively studied. BNP is labile and must be collected
BNP/NT-proBNP assays include -blockers, into tubes containing EDTA and tested ideally within 4
angiotensin-converting enzyme inhibitors, angiotensin hours of collection. Serum, citrated and heparinized
inhibitors, diuretics, digoxin and digitoxin, plasma cannot be used. BNP can be stored for up to 24
antiarrhythmics, calcium channel blockers, statins and hours at 2C to 8C for some assays. BNP is degraded
other lipid lowering agents, platelet inhibitors, and other by contact activation of the blood coagulation system
noncardiac medications, including some analgesics and [13] and can be stabilized by storage under frozen
antibiotics [9]. conditions or with the addition of kallikrein protease
inhibitors [14]. In contrast, NT-proBNP is much more
All immunoassays for BNP and NT-proBNP are stable than BNP and can be kept at refrigerated
heterogeneous (i.e., use a separation step between bound temperatures for 3 days. NT-proBNP can be tested in
and free antibodies), so there are no problems with serum and plasma collected in either heparin or EDTA
photometric interferents (i.e., hemoglobin, bilirubin, or [15]. When EDTA plasma is used, NT-proBNP
lipemia). One study showed that a wide hematocrit range concentration is about 10% lower than for serum or
(28% to 58%), which can affect blood glucose lithium heparin.
measurements, did not affect whole blood NT-proBNP
results using a point-of-care device [10]. There have BNP, NT-proBNP, and ProBNP Reference Intervals
been no documented reports of interference due to the Reference intervals for BNP and NT-proBNP were
presence of human anti-mouse antibodies (HAMA) or established by the first manufacturers of FDA-approved
heterophile antibodies. The abnormal antibodies have assays (Inverness and Roche, respectively). For BNP, a
been shown to produce false positive results for other single cutoff of 100 ng/L was determined and validated
sandwich-type immunoassays for other cardiac and [16]. Commercial BNP assays developed and released
noncardiac markers such as troponin and human after the FDA clearance of the Inverness BNP assay
chorionic gonadotrophin (hCG), where low have configured their assay to maintain the use of this
concentrations have considerable clinical significance cutoff concentration. For NT-proBNP, the manufacturer
(acute coronary syndromes and trophoblastic tumors, (Roche Diagnostics) established two cutoff
etc.). The absence of these reports does not infer concentrations: <125 ng/L for patients under 75 years of
freedom from these interferences. However, the clinical age, and <450 ng/L for patients 75 years and older. All
significance of the presence of HAMA and heterophile other NT-proBNP assays make use of these limits as
antibodies is diminished for BNP and NT-proBNP well. Values for BNP and NT-proBNP in healthy
assays; mild false-positive results do not have significant subjects are considerably lower than these limits and
clinical implications. show significant variance by age and gender. These
subgroups and lower cutoff concentrations have not been
The cross-reactivity of assays for BNP and NT-proBNP incorporated into routine clinical practice. Figure 2
towards proBNP is an important consideration, since it is shows the relationship between age and gender for BNP
now known that proBNP also circulates in the blood of [17] and NT-proBNP [18]. The results show that males
patients with heart failure. Because proBNP is the have lower BNP/NT-proBNP concentrations than
precursor protein to both BNP and NT-proBNP, the females, and younger subjects have lower BNP/NT-
epitopes used to raise antibodies for these latter two proBNP concentrations than older subjects. The reason
peptides are also present in the proBNP peptide for these differences are not fully known. For gender,
structure. ProBNP also exists in the blood of heart some have suggested that the use of hormone
failure patients as an O-linked glycoprotein [11]. There replacement therapy (HRT) in postmenopausal women
are 7 potential sites for glycosylation from the center of may result in higher BNP concentrations in this
the molecule (from the non-BNP amino acid sequences). population. Maffei et al. showed that treatment with
There have been three studies that have tested the cross- estrogen therapy caused BNP to increase from 12.6 to
reactivity of proBNP towards commercial BNP and NT- 19.8 ng/L after 3 months [19].
proBNP assays [2,3,12]. For BNP, the cross-reactivity
ranged from 17% to 38% for glycosylated proBNP and Decreasing renal function with age may be partly
from 5% to 14% for non-glycosylated proBNP [12]. For responsible for the increase in BNP/NT-proBNP. In the
NT-proBNP, the cross-reactivity ranged from 29% to absence of heart failure, the concentration of BNP/NT-
249% for non-glycosylated proBNP but with no cross- proBNP increases in patients with a decreasing
reactivity towards glycosylated proBNP [3,12]. The glomerular filtration rate [20]. It is also possible that
200
apparently healthy elderly subjects may have declining groups to advocate the use of two BNP cutoffs: 100 ng/L
cardiac function not detectable by current noninvasive for ruling out and 400 ng/L for ruling in heart failure
techniques. There is controversy with regard to which [27]. In the PRIDE study, a NT-proBNP cutoff of 450
marker, BNP or NT-proBNP, is more influenced by and 900 ng/L for patients <50 and 50 years,
renal dysfunction. Vickery et al. suggested that NT- respectively, produced a clinical sensitivity of 93% and
proBNP is more influenced than BNP in patients with 91% and clinical specificity of 95% and 80%,
declining renal function [21]. However, deFilippi et al. respectively. The AUC-ROC was 0.94. These studies
suggested that while both of these markers are have also shown that BNP/NT-proBNP correlates with
equivalent predictors of decompensated heart failure in the severity of heart failure, as assessed subjectively by
patients with kidney disease, NT-proBNP was a better criteria established by the New York Heart Association.
predictor of mortality than BNP (odds ratio of 4.5 vs.
2.6, respectively) [22]. There is currently no data on the Increased concentrations of BNP and NT-proBNP have
influence of declining renal function on measurements of been associated with risk for cardiovascular death and
proBNP. Given that this marker is the same size as the risk stratification for patients with heart failure [28],
combination of BNP and NT-proBNP, it is likely that acute coronary syndromes [29], stable heart disease [30],
this peptide will be retained to a greater extent in and apparently healthy subjects without a history of
patients with reduced glomerular filtration rates. heart disease [31]. For patients with acute coronary
syndromes, the combination of BNP/NT-proBNP with
The body mass index is another variable that influences other markers such as cardiac troponin and C-reactive
the reference intervals for BNP and NT-proBNP. Obese protein further improves the odds ratio for prediction of
patients have lower BNP and NT-proBNP adverse events [32]. The acute release of BNP and/or
concentrations than lean patients [23,24]. There have NT-proBNP appears to arise from the small cytoplasmic
been several theories as to the mechanism for these pool and not from upregulation at the genomic level. The
observations. Natriuretic peptide receptors are found in success of these studies has led some investigators to
high concentrations in adipose tissues, suggesting that suggest a role for BNP/NT-proBNP for screening
BNP is degraded. However, this does not account for the patients for heart failure who are asymptomatic [33].
obesity effect seen for NT-proBNP, which is not However, screening programs with natriuretic peptide
degraded by these receptors [23]. Others have suggested markers have not been adopted by national and
that BNP may be a causative factor in development of international clinical guidelines and therefore are not
obesity [24]. The reduced clinical utility of BNP and currently used for this purpose.
NT-proBNP in obese patients has prompted a search for
new serum biomarkers for heart failure that can There have been a few studies directly comparing the
complement the use of these peptides. clinical utility of BNP to NT-proBNP for use in patients
with heart failure. While there are minor differences in
Interpretation performance, most studies have shown that the two
Results for BNP and NT-proBNP assays are used as an markers can be used interchangeably [34-36]. There
aid in the diagnosis and management of patients with would be no benefit in a laboratory offering both tests.
heart failure. The most documented usage is for patients Given that the markers are different, results from one
who present with acute symptoms of heart failure such assay cannot be directly converted to the other.
as dyspnea and shortness of breath. Two widely quoted Therefore, both assays should not be used on the same
studies are the Breathing Not Properly (BNP) Trial for patient.
BNP [25] and the Pro-BNP Investigation of Dyspnea in
the Emergency Department (PRIDE) Study for NT- BNP, NT-proBNP, and ProBNP Performance Goals
proBNP [26]. These studies were conducted on patients The biological variability of BNP and NT-proBNP has
presenting to the emergency department with symptoms been extensively studied in healthy subjects and patients
suggestive of heart failure. BNP and NT-proBNP testing with stable heart failure [37-39]. Results vary according
was conducted on blood collected at admission. Results to the time interval between serial collections. For short-
were compared to a final assessment conducted by two term (day-to-day) variation, BNP and NT-proBNP have
attending physicians, using results of all diagnostic intraindividual imprecision (CVI) of 20% to 25%. The
procedures (symptoms, electrocardiography, analytical imprecision of analytical assays for BNP and
echocardiography, chest x-ray, clinical notes, response to NT-proBNP (CVA) range from 3% to 5% for automated
therapy, etc.), excluding the results of BNP or NT- immunoassays and 10% to 15% for point-of-care testing
proBNP. A third physician was asked to adjudicate assays and manual assays. Using 1/2CVI as the criteria,
discordances in results between the first two attendants. all manufacturers of BNP and NT-proBNP assays are
The major objective of these studies was to determine able to meet imprecision goals based on biological
the clinical performance of these assays. Using a cutoff variability. The CVI is also used to compute the relative
of 100 ng/L in the BNP Trial, BNP produced a clinical change value (RCV)the difference in test results from
sensitivity and clinical specificity for diagnosis of heart serial collections that is considered statistically
failure of 90% and 76%, respectively, with an area under significant. With 95% confidence, the RCVs for BNP
the receiver-operating characteristic curve (AUC-ROC) and NT-proBNP range from 30% to 40% [39]. For
of 0.91. Improved specificity can be achieved by patients admitted to a hospital with decompensated heart
increasing the cutoff to 400 ng/L, prompting some
201
failure, most studies have shown that the BNP and/or evaluation of a new point-of-care test for the
NT-proBNP decreases by more than the RCV when determination of NT-proBNP in whole blood.
patients are successfully treated over the period of Clin Chem Lab Med 2006;44:1269-77.
hospital admission (compensation of heart failure), 11 Schellenberger U, ORear J, Guzzetta A, Jue
suggesting that these tests are useful for short-term RA, Protter AA, Pollitt NS. The precursor to B-
monitoring. For longer-term (week-to-week) variation, type natriuretic peptide is an O-linked
BNP and NT-proBNP have CVIs of 35% to 50% and glycoprotein. Arch Biochem Biophys
RCVs of 50% to 70%. Studies on the use of serial BNP 2006;456:160-6.
and NT-proBNP measurements for monitoring the 12 Luckenbill KN, Christenson RH, Jaffe AS,
success of drug therapy on an outpatient basis do not Mair J, Ordonez-Llanos J, Pagani F, et al.Cross-
usually exceed these RCV limits. This puts into question reactivity of BNP, NT-proBNP, and proBNP in
the value of BNP and NT-proBNP testing for long-term commercial BNP and NT-proBNP assays:
monitoring of patients treated with chronic heart failure. preliminary observations from the IFCC
Committee for Standardization of Markers of
The College of American Pathologists offers an external Cardiac Damage. Clin Chem 2008; 54: 619-21.
QC program for BNP and NT-proBNP. As of 2007, 13 Shimizu H, Aono K, Masuta K, Asada H,
there were about 2300 participants in the BNP survey Misaki A, Teroaka H. Degradation of human
and 700 for NT-proBNP. brain natriuretic (BNP) by contact activation of
blood coagulation system. Clin Chim Acta
References 2001;305:181-6.
1 Hall C. Essential biochemistry and physiology 14 Belenky A, Smith A, Zhang B, Lin S, Depres
of (NT-pro) BNP. Eur J Heart Failure N, Wu AHB et al. The effect of class-specific
2004;6:257-60. protease inhibitors on the stabilization of B-
2 Liang F, ORear J, Schellenberger U, Tai L, type natriuretic peptide. Clin Chim Acta
Lasecki M, Schreiner GF et al. Evidence for 2004;340:163-72.
functional heterogeneity of circulating B-type 15 Yeo KT, Wu AH, Apple FS, Kroll MH,
natriuretic peptide. J Am Coll Cardiol Christenson RH, Lewandrowski KB et al.
2007;49:1071-8. Multicenter evaluation of the Roche NT-
3 Heublein DM, Huntley BK, Boerriger G, proBNP assay and comparison to the Biosite
Cataliotti A, Sandberg SM, Redfield MM et al. Triage BNP assay. Chim Chim Acta
Immunoreactivity and guanosine 3,5-cyclic 2003;338:107-15.
monophosphate activating actions of various 16 Wieczorek SJ, Wu AHB, Christenson R,
molecular forms of human B-type natriuretic Krishnaswamy P, Gottlieb S, Rosano T et al. A
peptide. Hyertension 2007;49:1114-9. rapid B-type natriuretic peptide assay
4 Shimizu H, Masuta K, Aono K, Asada H, accurately diagnoses left ventricular
Sasakura K, Tamaki M et al. Molecular forms dysfunction and heart failure: A multicenter
of human brain natriuretic peptide in plasma. evaluation. Am Heart J 2002;144:934-9.
Clin Chim Acta 2002;316:129-35. 17 Redfield MM, Rodeheffer RJ, Jacobsen SJ,
5 Mair J. Friedl W, Thomas S, Puschendorf B. Mahoney DW, Bailey KR, Burnett JC Jr.
Natriuretic peptides in assessment of left- Plasma brain natriuretic peptide concentration:
ventricular dysfunction. Scand J Clin Lab impact of age and gender. J Am Coll Cardiol
Invest 1999;59(Suppl 230):132-42. 2002;40:976-82.
6 Wu AHB. Plasma BNP/NT-proBNP assays: 18 Hess G, Runke S, Zdunek D, Hitzler WE.
what do they target and what else might they Reference interval determination for N-
recognize? In: Heart Failure Clinics. terminal-B-type natriuretic peptide (NT-
Natriuretic peptides. Mills RM, Tang WHW, proBNP): a study in blood donors. Clin Chim
Burnett JC, Jr., Eds., Elsevier Inc., Acta 2005;360:187-93.
Philadelphia. 2006:291-8. 19 Maffei S, Del Ry S, Prontera C, Clerico A.
7 Giuliani I, Rieunier F, Larue C, Delangneau JF, Increase in circulating levels of cardiac
Granier C, Pau B et al. Assay for measurement natriuretic peptides after hormone replacement
of intact B-type natriuretic peptide prohormone therapy in postmenopausal women. Clin Sci
in blood. Clin Chem 2006;52:1054-61. 2001;101:447-53.
8 Dries DL, Victor RG, Rame JE, Cooper RS, 20 McCullough PA, Duc P, Omland T, McCord J,
Wu X Zhu X et al. Corin gene minor allele Nowak RM Hollander JE et al., for the BNP
defined by 2 missense mutations is common in Multinational Study Investigators. B-Type
blacks and associated with high blood pressure natriuretic peptide and renal function in the
and hypertension. Circulation 2005;112:2403- diagnosis of heart failure: an analysis from the
10. Breathing Not Properly Multinational Study.
9 Package inserts. Biosite Triage Assay. San Am J Kid Diseases 2003,41:571-9.
Diego, CA. 21 Vickery S, Price CP, John RI, Abbas NA,
10 Zugck C, Nelles M, Katus HA, Collinson PO, Webb MC, Kempson ME et al. B-type
Gaze DC, Dikkeschei B et al. Multicentre natriuretic peptide (BNP) and amino-terminal
202
proBNP in patients with CKD: relationship to peptide levels and the risk of cardiovascular
renal function and left ventricular hypertrophy. events and death. N Engl J Med 2004;350:655-
Am J Kidney Dis 2005;46:610-20. 63.
22 deFilippi CR, Seliger SL, Maynard S, 32 Sabatine MS, Morrow DA, de Lemos JA,
Christenson RH. Impact of renal disease on Gibson CM, Murphy SA, Rifai N et al.
natriuretic peptide testing for diagnosing Multimarker approach to risk stratification in
decompensated heart failure and predicting non-ST elevation acute coronary syndromes:
mortality. Clin Chem 2007;53:1511-9. simultaneous assessment of troponin I, C-
23 Wang TJ, Larson MG, Levy D, Benjamin EJ, reactive protein, and B-type natriuretic peptide.
Leip EP, Wilson PWF et al. Impact of obesity Circulation 2002;105:1760-3.
on plasma natriuretic peptide levels. Circulation 33 Hobbs FD, Davis RC, Roalfe AK, Hare R,
2004;109:594-600. Davies MK, Kenkre JE. Reliability of N-
24 Bayes-Genis A, Lloyd-Jones DM, van terminal pro-brain natriuretic peptide assay in
Kimmenade RJJ, Lainchbury JG, Richards AM, diagnosis of heart failure: cohort study in
Ordonez-Llanos J et al. Effect of body mass representative and high risk community
index on diagnostic and prognostic usefulness populations. BMJ 2002;324:1498.
of amino-terminal pro-brain natriuretic peptide 34 Hammerer-Lercher A, Neubauer E, Muller S,
in patients with acute dyspnea. Arch Intern Med Pachinger O, Puschendorf B, Mair J. Head-to-
2007;167:400-7. head comparison of N-terminal pro-brain
25 Maisel AS, Krishnaswamy P, Nowak RM, natriuretic peptide, brain natriuretic peptide and
McCord J, Hollander JE, Duc P et al, for the N-terminal pro-atrial natriuretic peptide in
BNP Multinational Study Investigators. Rapid diagnosing left ventricular dysfunction. Clin
measurement of B-type natriruetic peptide in Chim Acta 2001;310:193-7.
the emergency diagnosis of heart failure. N 35 Tsutamoto T, Sakai H, Ishikawa C, Fujii M,
Engl J Med 2002;347:161-7. Tanaka T, Yamamoto T et al. Direct
26 Januzzi JL Jr., Camargo CA, Anwaruddin S, comparison of transcardiac difference between
Baggish AL, Chen AA, Krauser DG et al. The brain natriuretic peptide (BNP) and N-terminal
N-terminal pro-BNP investigation of dyspnea in pro-BNP in patients with chronic heart failure.
the emergency department (PRIDE) study. Am Eur J Heart Failure 2007;9:667-73.
J Cardiol 2005;95:948-54. 36 Clerico A, Fontana M, Zyw L, Passino C,
27 Silver MA, Maisel A, Yancy CW. BNP Emdin M. Comparison of the diagnostic
Consensus Panel 2004: a clinical approach for accuracy of brain natriuretic peptide (BNP) and
the diagnostic, prognostic, screening, treatment the N-terminal part of the propeptide of BNP
monitoring, and therapeutic roles of natriuretic immunoassays in chronic and acute heart
peptides in cardiovascular disease. Cong Heart failure: a systematic review. Clin Chem
Fail 2004;10 (suppl 3):1-30. 2007;53:813-22.
28 Doust JA, Pietrzak E, Dodson A, Glasziou P. 37 Wu AHB, Smith AC, Mather JF, Duncan B,
How well does B-type natriuretic peptide White CM, Ahlberg A et al. Biological
predict death and cardiac events in patients with variation for NT-pro- and B-type natriuretic
heart failure: systematic review. BMJ peptides and implications for therapeutic
2005;330:625. monitoring of patients with congestive heart
29 Morrow DA, de Lemos JA, Blazing MA, failure. Am J Cardiol 2003,92:628
Sabatine MS, Murphy SA, Jarolim P et al. 38 Bruins S, Fokkema MR, Pomer JWP,
Prognostic value of serial B-type natriuretic DeJongste MJL, van der Dijs FPL, van den
peptide testing during follow-up of patients Ouweland JMW et al. High intraindividual
with unstable coronary artery disease. JAMA variation of B-type natriuretic peptide (BNP)
2005;294:2866-71. and B-type proBNP in patients with stable
30 Bibbins-Domingo K, Gupta R, Na B, Wu AHB, chronic heart failure. Clin Chem 2004;50:2052-
Schiller NB, Whooley MA. N-terminal pro-B- 8.
type natriuretic peptide and incident heart 39 Wu AHB. Serial testing of B-type natriuretic
failure in ambulatory patients with coronary peptide and NTpro-BNP for monitoring therapy
disease: data from the Heart and Soul Study. of heart failure: the role of biologic variation in
JAMA 2007; 297:169-176. the interpretation of results. Am Heart J
31 Wang TJ, Larson MG, Levy D, Benjamin E, 2006;152:828-34.
Leip E, Omland T et al. Plasma natriuretic
203
Table 1: Commercial Natriuretic Peptide Assay Characteristicsa
Assay Antigen Capture Ab Detection Ab
Abbott AxSYM BNP BNP a.a. 1-32 Scios (ring) Shionogi

(C-terminus)
Siemens Centaur BNP BNP a.a. 1-32 Shionogi Shionogi (ring)

(C-terminus)
Inverness Triage BNP a.a. 1-32 Scios (ring) Inverness

(Beckman Access) BNP (not characterized)
Shionogi BNP BNP a.a. 1-32 Shionogi Shionogi (ring)

(C-terminus)
Roche Elecsys NT-proBNP 1-76 Roche Roche

(Dade Dimension, Response (N-terminus) (central)
Biomedical)
NT-proBNP
Bio-Rad ProBNP 1-108 Hinge 76 region BNP (unspecified

(not FDA approved) (between BNP region)
and NT-proBNP)
a
Adapted from Apple FS. Personal communication.
Table 2: Methods of BNP, NT-proBNP, and ProBNP Analysis

Method 1: BNP, automated
Principle of analysis: Immunoassay
Comments: No standardization among automated BNP testing platforms. A diagnostic cutoff of 100 ng/L has
been established for all commercial BNP assays.
Method 2: BNP, POC

Principle of analysis: Immunochromatography coupled with immunoassay
Comments: Inverness (Biosite) Triage assay standardized to the Beckman Access assay.
Method 3: NT-proBNP, automated

Comments: All NT-proBNP assays standardized to the Roche Elecsys platforms.
Method 4: NT-proBNP, POC

Principle of analysis: Immunochromatography coupled with immunoassay
Comments: All NT-proBNP assays standardized to the Roche Elecsys platform.
Method 5: Pro-BNP
Comments: No commercial assay available. One manufacturer has developed a prototype assay.
204
BNP (Inverness Triage) NT-proBNP (Roche)
70 70
60 60
50 50
40 40
30 30
20 20
10 10
0 0
45-54 55-64 65-74 75-83 18-29 30-39 40-49 50-59 60-69
Age, years
Fig. 2. Reference intervals for BNP (Inverness Medical, n=2042) and NT-proBNP (Roche Diagnostics, n=1980) by age
and gender. Results in ng/L for both tests. Data from Redfield et al. [17] and Hess et al. [18] Aqua is male, blue is female
205
Barbiturates
Barbiturates
i
Steven C. Kazmierczak
Name: Amobarbital Butabarbital sodium Hexobarbital

Molecular formula: C11H18N2O3 C10H15N2NaCO3 C12H16N2O2

Molecular weight: 226.27 D 234.23 D 236.26 D
Merck Index: 592 1471 4598
Chemical class: Barbiturate, organic heterocyclic
Structures: See Figures
Name: Pentobarbital sodium Phenobarbital Secobarbital sodium

Clinical significance: click here for all
Molecular formula: C11H17N2NaO3 C12H12N2O3 C12H17N2NaO3
Merck Index: 6998 7109 8268
Chemical class: Barbiturate, organic heterocyclic
Structures: See Figures
Principles of Analysis and Current Usage i Qualitative

Barbiturates are a group of drugs classified as hypnotics Qualitative analytical procedures for barbiturates using a
and are prescribed for a number of conditions. They act mercury complex colorimetric test [1] or absorbance
as sedatives and are used to help control seizure spectroscopy [2] are available and provide rapid analysis
disorders. Most commonly, phenobarbital is taken for for barbiturates in most body fluids or from tissue
the control of epilepsy, and its determination, as part of sources [3]. However, the specificity of the colorimetric
therapeutic drug monitoring, is usually the most mercury complex test is poor and the spectrophotometric
frequently performed analysis for this group of tests require sophisticated spectrophotometric equipment
compounds. Recently, the use of high-dosage which may not be available in modern clinical
pentobarbital has become important in the therapy of laboratories (Table 1, Methods 1, 2, and 3).
head trauma. However, barbiturates are also abused
substances; thus barbiturate determination in the Ultraviolet absorbance spectroscopy may be used to both
overdose patient is one of the major components of the qualitatively identify as well as quantify barbiturates.
toxicological drug screen. Both qualitative and Following extraction, an absorbance spectrum is
quantitative tests are performed by laboratories. The obtained from 230 to 340 nm (Figure 1). The
qualitative tests are selected for their rapid performance characteristic spectrum of each barbiturate (Barbiturates:
in drug screening and usually react with any drug of the Figure 2) allows for the qualitative identification. A
barbiturate class. Most qualitative analyses are variation of this method, using differential spectroscopy,
performed on urine or gastric contents, since these body allows for quantitation of the barbiturates [4] (Table
fluids are best for drug screening. In contrast, 1,method 3).
quantitative analyses are performed on blood, since
blood concentrations correspond most closely with Chromatographic procedures are also useful for the
pharmacological effect. identification of barbiturates; these include thin-layer
chromatography and gas chromatography. Thin-layer
chromatography (Table 1, method 4) is usually
ii
Barbiturates performed on urine that has been acidified and extracted
Previous and current authors of this method: into an organic phase. One places the sample on a thin-
First edition: Amadeo J. Pesce layer chromatographic plate and identifies the position of
Methods edition: Amadeo J. Pesce the barbiturate after chromatography by developing a
Second edition: Amadeo J. Pesce color reaction with one of several sprays, such as
Third edition: Mark W. Linder potassium permanganate, mercuric sulfate, or
Fourth edition: Steven C. Kazmierczak diphenylcarbazone [5]. Alternatively, a commercial thin-
Fifth edition: Not updated layer method is available for the identification of several
barbiturates in urine which requires minimum
206
Barbiturates
manipulation of the specimen. The Rf value and the secobarbital is conjugated to the enzyme glucose-6-
color are the specific criteria for identification. phosphate dehydrogenase (G6PDH). The presence of
anti-barbiturate antibodies in the reaction mixture
Gas chromatography detection (Table 1, method 5a) is suppresses G6PDH activity. Barbiturate introduced to
performed by the extraction of the acidified urine the reaction in the patient specimen competes for the
specimen with toluene or other selected organic solvents limited number of antibodies resulting in a dose-
and the separation of the extracted analyte. An OV- dependent increase in G6PDH activity monitored by the
series column with a flame ionization detector is the production of NADH at 340 nm [11].
detection system most commonly used as part of an
overall screening procedure for drugs including Additional qualitative tests for barbiturates in urine and
barbiturates [6]. Derivatization (methylation) improves serum based on competitive binding immunoassays are
chromatographic behavior of these relatively nonvolatile available in formats applicable to both automated
drugs. Figure 3 demonstrates the difference in retention analysis and formats which do not require
time and peak shape between chromatography of a instrumentation. Other common automated procedures in
mixture of derivatized versus underivatized barbiturates. addition to EMIT, include the FPIA procedure and a
The relative retention time is specific for each barbital procedure based on the kinetic interaction of
derivative and is used for presumptive identification. microparticles in solution (KIMS) [12]. In the latter
method, absence of drug in the patient specimen, drug-
A dedicated HPLC system for drug detection has also microparticle conjugates are aggregated into a lattice
been developed (Table 1, method 5b). Both serum and through the interaction of anti-barbiturate antibodies
urine can be analyzed with minimal sample preparation. leading to increased absorbance of incident light. Drug
Following the addition of internal standard to the (barbiturate) in the sample competes for drug-
specimen, a sample is automatically injected into the microparticle conjugate for a limited number of
system. The effluent is scanned by ultraviolet light in antibodies in the reaction mixture. Binding of antibody
the region of 190 to 300 nm and the combination of to drug in the sample inhibits microparticle lattice
retention time and ultraviolet absorbance characteristics formation (nephelometric inhibition) and diminishes the
allow the identification of approximately 400 different increasing absorbance in proportion to the concentration
compounds. Unfortunately, the sensitivity of the system of drug in the sample.
for detecting barbiturates in urine is unacceptable. Also,
the system is subject to interference and Instrument-independent methods have also become very
misidentification of drugs caused by co-eluting popular for point-of-care use [13]. These devices allow
compounds [7, 8]. for visual detection of barbiturates and other abused
drugs [14]. One device utilizes colloid gold particles
Alternatively, the extract, as described in Method 5a, can conjugated to the drug(s) of interest and monoclonal
be analyzed with a gas chromatograph-mass antibodies directed toward these drugs. Addition of
spectrometer (GC/MS) and identified by the pattern of specimen containing drug(s) results in competition of
mass fragmentation (Table 1, method 6). Solid-phase antibody for free drug(s) in urine and drug bound to gold
extraction methods have advantages over the common colloid. If the specimen contains drug(s) above the
liquid/liquid extractions and coupled to capillary gas detection limit of the system, some of the drug(s) bound
chromatography provide for a rapid and convenient to the gold colloid particles will not be bound by
method for the qualitative identification of barbiturates antibody. The reaction mixture is next transferred to a
in urine [9]. nylon membrane with monoclonal antibodies bound to
the membrane. Specimens that have no drug(s) present
Competitive binding immunoassays (Table 1, method 7) will have no colloid-bound drugs not bound to antibody
for the detection of drugs of abuse in urine have and thus pass through the membrane. Specimens, which
advanced in popularity because of the ease of use and have drug(s) present above the detection limit of the
the rapidity of the analysis. With respect to barbiturates, system will have some colloid-bound drugs not linked to
these assays rely on the cross-reactivity of the antibody and thus be able to bind to antibody on the
structurally similar barbiturates with an antibody nylon membrane. Binding of the gold colloid to the
developed against a specific barbiturate, commonly nylon membrane results in the appearance of a colored
secobarbital. These assays are available for automation bar which appears in the area containing antibody
on a variety of random access analyzers and provide for directed toward the drug of interest.
a rapid screen for the presence of barbiturates. These
assays have found wide acceptance in both forensic and These assays may be employed outside of the typical
emergency toxicology. Because of the nature of these laboratory environment. Although these methods have
assays, it is highly recommended that positive results be proven useful, the user must comprehend the limitations
confirmed by a more specific assay such as gas of the analyses and results should be confirmed by a
chromatography/mass spectrometry [10]. method which does not rely on competitive binding
immunoassay.
The enzyme-multiplied immunoassay technique (EMIT) Quantitative
for barbiturate in serum and/or urine is a homogeneous Techniques of quantitation include gas chromatography,
enzyme immunoassay (EIA). The barbiturate high performance liquid chromatography (HPLC),
ultraviolet spectroscopy, and a variety of competitive
207
Barbiturates
binding immunoassays. The various immunoassays are The presence of barbiturates should be confirmed by
the same in the respect that specificity of the analysis is thin-layer chromatography or the preferred method, gas-
based on the binding specificity of the antibody for the liquid chromatography with mass spectrometric
drug. The unique aspect of the various methods detection.
available is with respect to the measured signal Quantitative
generated by the presence of drug in the patient sample For the analysis of mixtures of barbiturates, only the
(Table 1, method 8). gas/liquid and high performance liquid chromatography
procedures are applicable because they provide for
The fluorescent polarization immunoassay (FPIA), in separation of the barbiturates prior to quantitation.
which fluorescein-labeled phenobarbital competes with
phenobarbital in the patient specimen for a limited Phenobarbital is the most commonly quantitated
amount of antibody, is commonly used. The amount of barbiturate in serum. Several immunoassays are
bound labeled phenobarbital is inversely proportional to available with applications compatible with most testing
the phenobarbital in the patient sample. The intensity of formats. Quantitation of the other barbiturates by
the polarized fluorescence is proportional to the amount immunoassay is limited by availability of
of antibody-bound labeled phenobarbital. The polarized immunoassays. For pentobarbital, both gas
fluorescence is inversely proportional to the chromatography and HPLC are acceptable.
concentration of phenobarbital in the patient specimen.
When individual barbiturates are known to be present,
In general, quantitative immunoassays are used only the ultraviolet procedure yields accurate results.
when the specific drug is known, such as phenobarbital.
In these procedures drug in the specimen competes with A comparison of the gas liquid chromatography, EMIT,
labeled drug (method dependent) for a limited number of and fluorescence polarization is given in Table 2.
antibody binding sites. Increases in specimen
barbiturate concentration leads to measurement of either Specimen
a proportional increase or decrease in the assay signal. It The specimen can be blood, plasma, serum, urine, or
should be noted that the antibodies employed in these tissue extract, depending on the calibration of the
assays have a high degree of specificity for the intended method. The drug is stable in any of these specimens for
barbiturate, most commonly phenobarbital. 1 week. However, most laboratories refrigerate or freeze
samples until analysis.
HPLC has been adapted for the measurement of the
barbiturates, but the most common procedure quantifies Interferences
phenobarbital in the presence of other antiepileptic The presence of detergents and liquid soap have been
drugs, such as primidone and phenytoin. The reported to increase the barbiturates when measured
chromatographic system uses a 30 0.4 cm Porasil using FPIA [16].
column. The drug is detected with an ultraviolet
spectrophotometer, which monitors the absorbance at Barbiturates Therapeutic Concentrations
254 nm [15]. Therapeutic concentrations in serum
1. Amobarbital Up to 8g/mL (<35 mol/L)
Reference and Preferred Methods 2. Butabarbital Up to 8 g/mL (<34 mol/L)
Qualitative 3. Pentobarbital Up to 4g/mL (<16 mol/L)
Qualitative analysis of barbiturates is of clinical value in 4. Phenobarbital 15 to 49 g/mL (6472 mol/L)
screening for overdose or abuse. The identification of 5. Secobarbital Up to 6 g/mL (<23 mol/L)
barbiturates requires two independent methods, a class-
specific screening assay followed by a more specific
Barbiturates Performance Goals
method for identification and confirmation. In the case
For qualitative analysis of barbiturates, the evaluation
of an overdose, urine is the fluid of choice for screening.
limits (CLIA-88) requires 80% participant consensus.
In this circumstance, a competitive binding
immunoassay with general specificity for the barbiturate
Interpretation
class and good sensitivity is the best choice for a rapid
The barbiturates belong to the sedative hypnotic class of
screen. The immunoassay and thin-layer
drugs, and they function as central nervous system
chromatography methods are the most commonly
depressants. Barbiturates have been used as hypnotics
utilized procedures and offer comparable clinical utility.
for more than 80 years. They reversibly depress the
Following are the typical lower limits of detection
activity of all excitable tissues, inhibiting synaptic
employed for the barbiturate immunoassay screens in
responses.
urine:
There are many chemical derivatives of barbituric acid.
Secobarbital 200 ng/mL
Those with hypnotic action have amyl of alkyl groups
Pentobarbital 500 ng/mL
substituted at the C-5 position. The barbiturate
Butabarbital 1000 ng/mL
derivatives are termed long acting, intermediate acting,
Phenobarbital 6000 ng/mL
short acting, and ultrashort acting; the designation is
related to lipid solubility and time required to induce
208
Barbiturates
sleep. A classification summary of some barbiturates is estimation, and identification of barbiturates in

presented follows: biological material. Biochemistry 1956; 63:207-
13
Action Drug 4 Schumann GB, Lauenstein K, LeFever D,
Long Phenobarbital Henry JB. Ultraviolet spectrophotometric
Intermediate Amobarbital, butabarbital analysis of barbiturates. Am J Clin Pathol 1976;
Short Hexobarbital, pentobarbital, secobarbital 66:823-30
Ultrashort Sodium pentothal 5 Ganshirt H. Pharmaceutical products. In Stahl,
E., editor: Thin-layer chromatography: a
laboratory handbook, New York, 1962
Academic Press, 318-319.
The short-acting barbiturates are highly lipid soluble and
6 Sunshine I. Methodology for analytical
rapidly pass through the bloodbrain barrier. The
toxicology. Cleveland, 1975, CRC Press, Inc.
molecule is a weak acid, and the carbonyl group at the
34-44.
C-2 position tautomerizes between the lactam and lactim
7 Adams AK, Essien H, Binder SR.
configurations. The pI for barbituric acid is 7.4.
Identification of drugs in physiological fluids
following on-line liquid chromatographic
The most commonly prescribed barbiturate is
purification and analysis. Ann Biol Clin
phenobarbital, which is used in the therapy of seizure
1991;49:291-7
disorders.
8 Binder SR, Adams A., Regalia M, Essien H,
Rosenblum R. Standardization of multi-
Pentobarbital is used in very high concentrations in
wavelength UV detector for liquid
severe trauma to reduce cerebral edema. The mode of
chromatography-based toxicological analysis. J
action is not known, but the barbiturates in high
Chromatogr 1991;551:449-59
concentrations reduce central nervous system blood flow
9 Pocci R, Dixit V, Dixit VM. Solid-phase
and metabolism and can reduce oxygen requirements by
extraction and GC/MS confirmation of
50%.
barbiturates from human urine. J Anal Toxicol
1992;16:45-7
The barbiturates can be habit forming, and patients can
10 Urine testing for drugs of abuse, National
develop tolerance to the drug. Withdrawal symptoms
institute on drug abuse (NIDA) Research
can occur in abusers of the drug, and seizure is a
Monograph 73, 1986.
common sign of withdrawal. The removal of barbiturate
11 Ollerich M: Enzyme immunoassays in clinical
from a patient can be enhanced when the urine is made
chemistry: Present status and trends. J Clin
alkaline. However, the principal route of elimination of
Chem Clin Biochem. 1980;18:197-206.
barbiturates is metabolism in the liver. Several
12 Adler FL, Liu CT. Detection of morphine by
mechanisms take place: oxidation of the substituent
hemagglutinationinhibition. J Immunol
groups at the C-5 position, N-dealkylation, and
1971;106:1684-5
destruction of the barbiturate ring. The barbiturates
13 Biosite Triage and Roche Abutrack.
modulate cytochrome P-450 dependent metabolism of
14 Wu AHB, Wong SS, Johnson KG, Callies J,
other drugs either by inhibiting the biotransformation
Shu DX, Dunn WE, Wong SH. Evaluation of
thus increasing the serum half-life of the drug, or by
the Triage System for emergency drugs-of-
enhancing metabolism of certain drugs through
abuse testing in urine J Anal Toxicol
induction of specific enzymes of the cytochrome P-450
1993;17:241-5
system.
15 Kabra PM. Simultaneous liquid-
chromatographic determinations of 12 common
References
sedatives and hypnotics in serum Clin Chem.
1 Curry AS. Rapid quantitative barbiturate
1978;24:657-62,
estimation, Br Med J 1964; 1:354-5
16 Schwarzhoff R, Cody JT. The effects of
2 Walker JT, Fisher RS, and McHugh JJ.
adultering agents on FPIA analysis of urine for
Qualitative estimation of barbiturates in blood
drugs of abuse. J Anal Toxicol 1993;17:14-7
by ultraviolet spectrophotometry, Am J Clin
17 Data taken from College of American
Pathol 1948; 18:451-61
Pathologists (CAP) Profiency testing survey,
3 Broughton PMG. A rapid ultraviolet
1993 specimens Z-11, Z-12 and Z-13.
spectrophotometric method for the detection,
209
Barbiturates
Tables
Table 1: Methods of Barbiturate Analysis

Qualitative
Method 1. Extraction and colorimetric reaction
Principle: Barbiturate extracted from slghtly acidic solution and complexed with Hg2+; reacts with
diphenylthiocarbazone to form an orange colored product
Usage: Blood, plasma, serum, urine, gastric contents
Comments: Nonspecific for barbiturates
Method 2. Extraction and ultraviolet spectroscopy
Principle: Barbiturate extracted free of other ultraviolet-absorbing compounds; ultraviolet spectrum is
characteristic and specific for each drug
Usage: Blood, plasma, serum, urine, gastric juice, tissue
Comments: Original specific method; some interferences; can be quantitative; not recommended for mixtures
Method 3. Ultraviolet difference spectroscopy
Principle: As in Method 2, extracted barbiturate is divided into aliquots and difference between spectra at pH
greater than 13 and at pH 10.5 recorded
Usage: Blood, serum
Comments: More specific than ultraviolet spectroscopy; requires recording dual beam spectrophotometer
Method 4. Thin-layer chromatography
Principle: Extracted barbiturate chromatographed, its Rf factor (relative mobility factor) and reaction with stain
recorded; Rf factor and color reactions presumptive for barbiturates
Comments: Used as either qualitative or confirmatory test; can separate mixtures of barbiturates
Method 5a. Gas chromatography
Principle: Extracted barbiturate chromatographed and its retention time recorded; parent drug and possible
degradation products have specific retention times
Comments: Not quantitative; degradation of some barbiturates; not a gaussian elution pattern; can separate
mixtures
Method 5b. High-performance liquid chromatography
Principle: Extracted barbiturate chromatographed and its retention time recorded; detection by ultraviolet
spectra; parent drug and possible degradation products have specific retention times
Comments: Not quantitative; can separate mixtures
Method 6. Gas chromatographymass spectrometry
Principle: Steps identical to those for method 5; mass spectrophotometric detector can identify specific
barbiturates and resolve mixtures
Comments: Improves identification of simple gas chromatographic method
Method 7. Competitive binding immunoassays
Principle: Competition between labeled drug (method dependent) and drug in the sample for a limited number of
antibodies in the reaction mixture
Usage: Urine; serum (limited use)
Comments: Positive results corresponds with an assay signal for the sample which exceeds the signal for a
known concentration calibrant (cutoff)
Quantitativeall barbiturates
Method 8. Ultraviolet spectroscopy (similar to method 2)
Principle: Extracted barbiturate has characteristic absorbance spectrum, with amount of absorbance proportional
to concentration; some identification depends on spectral characteristics
Comments: Original method, but subject to interference by endogenous compounds and other drugs; does not
work well on mixtures
Method 9. Enzyme Immunoassay
Principles: EMITRelief from inhibition of phenobarbital labeled G6PDH; G6PDH- dependent production of
NADH is the measured signal; the signal is proportional to sample phenobarbital concentration
Radial partition immunoassayPhenobarbital in the patient specimen and alkaline phosphatase (Alk
phos) labeled phenobarbital compete for binding to immobilized anti-phenobarbital antibodies; Alk phos
dependent production of a fluorescent product is the measured signal; the signal is inversely proportional to
sample phenobarbital concentration
Usage: EMITSerum
210
Barbiturates
RPIASerum
Comments: EMIT most commonly used method; applications available for a wide variety of automated
analyzers
RPIA common; limited instrument options
Method 10. Fluorescence Polarization
Principle: Competitive binding with fluorescence polarization; bound labeled drug has higher polarized
fluorescence than free; amount of polarized fluorescence inversely proportional to the concentration of the drug
Usage: Serum, plasma
Comments: Commonly used procedure; lowest coefficient of variation
Method 11. Turbidimetric inhibition (TI)
Principle: Inhibitions of insoluble particlephenobarbitalantibody aggregates; the signal (rate of particle
aggregation) is inversely proportional to the sample phenobarbital concentration
Usage: Serum, plasma
Comments: Available as liquid reagents
Method 12. Nephelometric inhibition
Principle: Competitive inhibition of precipitation; signal is inversely related to sample phenobarbital
Usage: Serum
Comments: Liquid reagents, limited instrumentation compatibility
Table 2: Accuracy and Precision Comparison of Competitive binding Immunoassays for Quantitative Analysis of
phenobarbital in serum [17].
Quantitative Analysis of phenobarbital in serum. 1

Assay Weighed in value Mean measured value S.D. C.V.
6.8 mg/L
FPIA 6.14 0.37 6.0%
EMITa 6.60 0.31 4.7%
RPIA 7.31 0.66 9.1%
TI 7.22 0.63 8.7%
EMITb 6.46 0.41 6.4%
26.4 mg/L
FPIA 23.75 1.00 4.2%
EMITa 24.87 1.06 4.3%
RPIA 28.53 2.62 9.2%
TI 24.43 1.39 5.7%
EMITb 24.54 1.70 6.9%
75.6 mg/L
FPIA 63.21 3.36 5.3%
EMITa 62.15 3.22 5.2%
RPIA 72.07 8.06 11.2%
TI 62.93 3.70 5.9%
EMITb 67.74 5.27 7.8%
aDuPont reagents. bSyva reagents.
1
Data taken from College of American Pathologists (CAP) Profiency testing survey, 1993 specimens Z-11, Z-12 and Z-13.
211
Barbiturates
Table 3: Comparison of Assay Conditions for Serum Phenobarbital

Enzyme-multiplied Fluorescence
Gas-liquid immunoassay polarization
chromatography technique immunoassay
Sample size (mL) 1 0.05 0.02
Sensitivity (mg/L) 2 5 0.34
Coefficient of 8.5 8.2 5.0
variation (%)
Interference Phenobarbital and Some barbitals Some barbit-
mephobarbital not urates
separated
Time 1h 15 to 30 min 20 min
Figures
Figure 1: Barbiturates
Absorption spectra of a barbitone solution, 13.5 mg/mL: in 0.45 M NaOH, solid tracing, and in borate buffer (pH 10.5),
dashed tracing. (From Broughton, P.M.G.: Biochem. J. 1956; 63: 207-213.)
212
Barbiturates
Ultraviolet absorption spectra of barbiturates in methanol, solid; in 0.1 M NaOH, dashes; and in 0.1 M HCl, dots and
dashes. A, Phenobarbital; B, butabarbital; C, secobarbital.
213
Barbiturates
Gas chromatographic tracing of barbiturates separated as underivatized (above) and derivatized (below) form. For
derivatization, procedure outlined in manual method was used. Mixed barbiturate standard was 10 mg of each barbiturate
dissolved in 10 mL of CH3OH. 1, Butabarbital; 2, amobarbital; 3, pentobarbital; 4, secobarbital; 5, phenobarbital; 6,
phenytoin. Comparative testing was performed with the use of Hewlett-Packard 583A with a flame ionization detector and
a 2 mm inner-diameter glass column 6 feet long packed with 3% OV-17 on Chromosorb W (HP) of 800/100 mesh.
Nitrogen was used as a carrier gas at a flow rate of 30 mL/min. Temperature programming was performed from 115 to
270 C at a 15 C/min increase after an initial hold time of 0.5 min. (Courtesy Hassan, F.M., and Fahey, C.: University of
Cincinnati.)
214
Barbiturates
Elution pattern of barbiturate standards after HPLC. 1, Methyprylon; 2, phenobarbital; 3, butabarbital, 4, butalbital; 5,
pentobarbital; 6, amobarbital; 7, glutethimide; 8, secobarbital; 9, nitrazepam internal standard; 10, methaqualone.
the dissolved standard. This is mixed and frozen as 1-
Procedure: Gas Chromatography mL aliquots, which are stable for 1 year. The resultant
calibration mixture contains 10 g/mL (approximately
Principle
The barbiturates are extracted from acidified serum with 45 mol/L) of each of the barbiturates.
an organic solvent. They are then back-extracted into Assay
tetramethylammonium hydroxide. The extracts are Equipment: Gas chromatographic equipment,
introduced into a gas chromatograph, and the methylated including a flame ionization detector; a 6-foot glass
derivatives of the drugs are formed on the column. The column, 2 mm inner diameter, packed with 3% OV-1 on
gas chromatograph detects, separates, and quantitates the 80/100 mesh Chromasorb W (HP).
drugs of interest, using aprobarbital as an internal
standard.
Sample preparation
Specimen
a. To a 10 mL screw-capped tube, add 1.0 mL of
Reagents and Materials patient serum (or serum calibrator), 0.5 mL of
1. Toluene. Chromatoquality. This is stable for 2 1M H3PO4, 100 L of aprobarbital internal
years at room temperature. standard, and 5 mL of toluene.
2. Methanol. Chromatoquality. This is stable for b. Shake vigorously for 1 min and centrifuge.
2 years at room temperature.
c. Transfer upper toluene layer to a 5 mL conical
3. Tetramethylammonium hydroxide. This is
centrifuge tube.
stable for 2 years at room temperature.
4. Barbiturates. Obtained in pure form from d. Add 12 L of tetra methylammonium
manufacturer. These are stable for 2 years at room hydroxide, shake for 1 min, and centrifuge.
temperature. Discard the upper layer.
5. 1 M H3PO4. Add 6 mL of concentrated e. Inject 2 L of lower layer into gas
H3PO4 (phosphoric acid) to 80 mL of H2O. This is chromatograph. Start programmer as the
solvent peak is eluted.
stable for 1 year at room temperature.
6. Working aprobarbital, internal standard,
Chromatographic parameters
1000 g/mL (4.76 mol/mL). Add 50 mg of each of
aprobarbital, butabarbital, pentobarbital, and secobarbital a. Column program, 90 to 200 C at 8 C/min
in 50 mL of methanol. To 99 mL of serum add 1 mL of b. Injection port temperature, 275 C
215
Barbiturates
c. Detector temperature, 280 C 3. Phosphate buffer (pH 4.4, 0.15 mmol/L).

d. Nitrogen flow, 60 mL/min Prepare 1 mol/L KH2PO4 by dissolving 13.6 g
e. Attenuation, 16, range 10-12 of KH2PO4 in 100 mL of distilled water.
f. Recorder, 1 mV and 0.5 inches/min Prepare 0.9 mol/L phosphoric acid by adding
4. Sample run time is approximately 20 min. 6.125 mL of phosphoric acid to distilled water.
Bring the volume to 100 mL.
5. Run known quality-control specimens with
each run. 4. Add 300 L of 1 mol/L KH2PO4 to 1800 mL
of distilled water. Using a pH meter, adjust the
pH to 4.4 with 0.9 mol/L phosphoric acid
Calculations (approximately 50 L).
1. Measure retention times of each peak, and 5. Acetonitrile: phosphate buffer (mobile
compare to the serum calibrator standards. phase). Add 190 mL of acetonitrile to 810 mL
2. To calculate each corresponding drug: of the phosphate buffer; filter through a 0.2 m
pore filter and degas under vacuum for 5 min.
Peak height of drug = Ratio of drug to aprobarbital 6. Internal standard (nitrazepam). Stock
Peak height of internal standard solution 1 mg/mL. Dissolve 10 mg nitrazepam
in methanol, and bring to 10 mL with methanol.
Divide peak height ratios of unknowns by peak Store at 4 C. Stable for 1 year.
height ratios of the calibration standard. 7. Working internal standard solution, 0.01
Multiply resulting ratio by 10 g/mL: mg/mL. Dilute 1 mL of working standard to
100 mL with sodium acetate buffer. Store at 4
Ratio of unknown 10 g/mL = Unknown (g/mL) C. Stable for 6 months.
Ratio of calibrated material 8. Calibration standards. Dissolve the desired
barbiturate in methanol, and dilute to a
concentration of 5 g/mL in serum. For
Procedure: High-Performance Liquid methyprylon and phenobarbital, dilute to a final
Chromatography concentration of 20 g/mL. Stable for 3 months
Principle at 4 C.Assay
For the specific quantitative analysis of any of the Equipment: high performance liquid chromatography
following barbiturates and drugs that are extracted equipment such as Waters HPLC System including
together: injector column heater, pump, detector, recorder, and a
15 cm C18 Bondapak Column; Vac Elut extraction
Barbiturates Other drugs apparatus; and 1 mL C18 Bond Elut solid-phase
Amobarbital extraction columns.
Glutethimide 1. Sample preparation
Butabarbital Methaqualone
a. Set up a Bond Elut column for each
Butalbital Methyprylon
sample to be analyzed. With vacuum
Phenobarbital
on, prime columns twice with
Pentobarbital
methanol and twice with distilled
Secobarbital
water.
The above compounds are isolated by use of a b. Turn off vacuum.
C18 bonded-phase column and analyzed by HPLC with c. To the respective columns, add 0.5 mL
spectrophotometric detection at 214 nm, with of working internal standard followed
comparison of retention times and peak height ratios of by 0.5 mL of appropriate sample.
unknowns against known standards and controls. d. Turn the vacuum on to pull samples
through the columns.
Specimen e. Immediately wash columns twice with
Serum, plasma distilled water.
f. Turn off vacuum.
Reagents and Materials
1. Methanol (HPLC grade). This is stable for 2 g. Wipe needle tips inside Vac Elut, and
years at room temperature. position labeled collection tubes in
rack.
2. Sodium acetate buffer (pH 4.5, 0.1 mol/L).
Dissolve 8.2 g of sodium acetate in distilled h. Add 100 L of methanol; wait 20 sec.
water and bring to 1000 mL with distilled i. Turn on vacuum; then turn off after
water. Adjust the pH to 4.5 with acetic acid if methanol passes through.
necessary. Store at room temperature in brown j. Repeat with another 100 L of
bottle. Stable for 2 years; check monthly for methanol.
growth.
216
Barbiturates
k. Remove collection tubes, and with a d. A glutethimide metabolite has the

disposable glass pipet, transfer the same retention time as butalbital.
contents of a collection tube to a larger e. Hemolyzed, lipemic, and icteric
12 75 mm glass tube. samples do not interfere with the
l. Inject 20 L into the HPLC system. analysis.
2. Chromatographic parameters
a. Flow rate, 1.5 mL/min Pentobarbital, Quantitative by Gas Chromatography
b. Column temperature, 60 C
c. Detector AUFS, 0.1 (absorbance units I. Principle
at full scale)
d. Detector wavelength 214 nm Pentobarbital is isolated from acidified
e. Chart speed, 0.5 cm/min biological fluids by liquidliquid solvent
extraction, and, after concentration by
Calculations evaporation, can be separated from other
A sample chromatogram of a series of barbiturate barbiturates by a gas chromatographic capillary
standards is presented in Barbiturates: Figure 4. column, detected utilizing a nitrogen
phosphorus detector, and quantitated by its peak
Calculate the peak height ratio of drug of interest to height ratio to known internal standard and
internal standard. subsequent calculation from known calibrator
standards.
Peak height (mm) of drug
Peak height (mm) of internal standard = Peak height II. Specimens of Choice
ratio
Preferred specimen is serum 1 mL (minimum
Compare elution times of standards to the elution times volume of 0.5 mL), collected in a red-top tube
of all unknown peaks of interest. See Barbiturates: with no separator gel. Plasma, collected in
Figure 4 for elution order of the various drugs. either heparinized or EDTA tubes, is also
acceptable. Specimens may be icteric or
Concentration unknown (g/mL) = slightly hemolyzed, but should not be grossly
hemolyzed or lipemic. Specimen should be
Peak height ratio Concentration of
obtained during the acute phase of toxicity or,
in patient or control appropriate drug
for TDM purposes, as a trough level drawn
Peak height ratio in standard (g/mL)
of calibration standard immediately prior to the dose of drug, or during
IV infusion.
Notes
1. The following limits of sensitivity for each to III. Indications
the barbiturates have been observed:
Compound Limit (g/mL) To document overdose or monitor therapeutic
Amobarbital 1 dosing of pentobarbital.
Butabarbital 1
Butalbital 1 IV. Reagents and Materials
Phenobarbital 5
Pentobarbital 1 A. Aprobarbital internal standard (5
mg/100 mL). To a 100 ml volumetric
Secobarbital 1
flask add 5 mg of pure aprobarbital
Glutethimide 1 (Applied Science Laboratories).
Methaqualone 1 Dissolve in deionized water and dilute
Methyprylon 5 to mark with the same. Store in 100
mL reagent bottle at 4 C. Stable for 3
Specific negatives are reported either as none months.
detected,or less than an appropriate limit of B. Stock Pentobarbital Standard (1
quantitation. mg/mL). To a 10 mL volumetric flask
2. Specificity add 10 mg of accurately weighed
a. Phenytoin is known to have the same Pentobarbital (Applied Science
retention time as amobarbital. Laboratories). Dilute to mark with
b. Clonazepam is known to have the methanol. Transfer to two 5 mL drug
same retention time as methaqualone. standard storage vials and refrigerate
at 4 C. Stable for 1 year.
c. Carbamazepine has a similar retention
time to secobarbital. C. Working 20 mg/mL Pentobarbital
Calibration Standard. To a 100 mL
217
Barbiturates
volumetric flask, add approximately N. Ethyl acetate (Chromatoquality)

50 mL of Utak Negative Serum Pool. O. Methanol (Chromatoquality)
With a 2 mL volumetric pipet add 2
mL of stock pentobarbital standard. V. Instruments and Parameters
Mix well while adding additional
Negative Pool to the mark. Aliquot 1 A. Hewlett Packard 5970 Gas
mL portions into 12 75 plastic tubes. Chromatograph Column: Ultra-1 (Hewlett Packard Co.)
Cap and store frozen at -20 C. Stable Temperatures: Column temperature program 120 to 300
for 1 year. C at 15 C/min after a 0.5 min hold; detector temperature
D. Working Pentobarbital Calibration 250 C; injector temperature 260 C.
Standards (for the establishment of a Attenuation: 4 (Note: May need to be
full linear curve). To 10 mL 16 100 optimized/adjusted with each run)
glass tubes add the following amounts Carrier gas: helium
of stock pentobarbital standard and Gas pressures at tank: air, 20; helium, 40; hydrogen, 30
Utak Negative Serum Pool, resulting mm Hg
in the concentrations noted: Gas flows (mL/minute): air, 100; hydrogen, 3; helium,
25
B. Gas Chromatograph Method:
Standard Concentration PENTOSR2
(mg/mL) L of Pool SerumL C. Heating block: 60 C with nitrogen gas
of Stock Pentobarbital attachment
D. Extraction rocker
0 500 VI. 0 Procedure
Note: Each run will consist of a control, a 20
5 497.5 2.5mg/mL calibration standard, and patient
serum analyzed in singlecate.
10 495 5
A. Sample Preparation and Entering Run Identifiers
20 490 10
1. To a 10 mL 16 100 glass tube add 0.5
mL of control, frozen calibration standard, or patient
50 475 25 a 0.5 mL Eppendorf pipet.
serum with
2. With a 300 L Eppendorf pipet add
80 460 300 L40of aprobarbital internal standard to each tube.
Vortex mix 5 sec.
100 450 503. With a 1 mL serological pipet add 0.1
mL of 0.5 N HCl to each tube, then vortex mix.
4. With a 10 mL serological pipet add 5
Prepare fresh as needed and proceed to Section VI.A.2. mL of ethyl acetate. Cap tubes and extract on rocker for
5 min.
D. 16 100 mm glass culture tubes with 5. Centrifuge samples for 3 min (Note:
caps Break any emulsions, which occur with a glass stirring
E. Utak Negative Serum Pool (Utak rod, and recentrifuge).
Associates) 6. With a glass Pasteur pipet transfer
F. Utak Pentobarbital Control Serum upper solvent layers to premarked and preheated 16
(Utak Associates) 100 glass tubes. Evaporate to dryness at 60 C under
G. 300 L Eppendorf pipet nitrogen.
H. 0.5 mL Eppendorf Pipet 7. Reconstitute dried vials in 100 L of
methanol, then vortex for 10 sec.
I. 1 10 mL serological pipets
8. Inject 1 L of reconstituted 20 g/mL
J. Glass disposable pasteur pipets standard solution into the "B" detector (NPD) of the gas
K. Hamilton 10 and 50 L syringe chromatograph, then puch the GC START button.
L. 0.5 N HCl - To a 1 L volumetric flask
add approximately 600 ml of VII. Calculations and Quality Control
deionized water. Carefully add 42.8
mL of concentrated HCl. Slowly A. Integrated Results
dilute to mark with deionized water.
Transfer to a 1 L reagent bottle. Store
The 5890, after performing a calibration, will print
at room temperature. Stable for 1
control and patient data directly, in g/ml concentration
year.
units. When integrated, test results are questioned, or
M. 5 mL glass Reacti-vials when calibration cannot adequately be performed,
218
Barbiturates
proceed with manual calculation. 1. Optimal serum concentrations are specific and
relative to the drug's use:
B. Manual Calculation 3040 g/mL optimal during treatment for
intracranial pressure
1. Observe the peak height (PH) amounts noted 13 g/mL for conventional anesthetic uses
for pentobarbital and internal standard on the hardcopy
report. Compute the peak height ratio (PHR) for the 20 Toxic effects may be noted at concentrations
g/mL standard, the Utak Control, and patient greater than 5 g/mL, and life-threatening
chromatograms: levels are generally greater than 10 g/mL.
PHR = Peak height of pentobarbital 2. Run a complete pentobarbital curve to establish

Peak Height of Internal Std. linearity monthly, when changing GC columns,
or when other changes dictate (i.e., detector
repair).
Note: If printed peak heights are not available, manual
peak heights should be measured in millimeters.
3. For concentration above 100 g/mL, dilute the
patient specimen 1:1 (250 L plus 250 L) with
2. Compute the concentration of Pentobarbital in
Utak Negative Pool, add internal standard, and
the control and patient:
reanalyze. Multiply calculated g/mL result by
2.
Concentration (g/ml) =PHR of Control or Patient 20
PHR of 20 g/ml Standard 4. The current method sensitivity for pentobarbital
is 1 g/mL. Negative results should be reported
C. An acceptable run includes control values as "less than 1 g/mL."
within acceptable limits (i.e., refer to "Internal QC"
Program). 5. No known interfering compounds.
XI. references
Heinemeyer, G., Ther. Drug Monit. 8:145-150,
1986.
Notes and Interpretation

219
Bence Jones Protein
Bence Jones Protein

Stanley S. Levinson
Name: Bence Jones protein

Clinical significance: Refer to Chapter 53, Neoplasia, in the 5th edition of Clinical Chemistry: Theory,
Analysis, Correlation
Merck Index: 1031
i trichloroacetic or sulfosalicylic acid or a protein-binding
More than a century after Dr. Henry Bence Jones received dye such as Coomassie blue or Ponceau S (Table 1,
a peculiar urine specimen from Dr. Thomas Watson, Method 2). The degree to which these confirmatory
Edelman and Gally showed that the Bence Jones protein methods measure FLC is unclear. The strips mainly
(BJP) was identical to the free light chain (FLC) of IgG measure albumin, and it is widely known that a mildly
[1,2]. The name Bence Jones protein is now commonly positive or negative result by dip strip analysis, followed
used to describe monoclonal FLC in urine and will be used by a grossly positive confirmatory result, may indicate the
here [3,4]. The urinary substance in the peculiar urine presence of an unsuspected BJP. In actual fact, this type of
specimen had the property of precipitating when heated to discrepancy cannot be relied on, because most patients
a temperature between 45C and 60C and redissolving with renal disease are spilling albumin along with BJP into
when boiled (Table 1, Method 1). This property, which did the urine and test positive by both approaches. It is
not show good sensitivity for BJP, was used as the basis important to follow up all cases of proteinuria with
for the Bence Jones heat test, which served as the test for electrophoretic analysis [5,8]. Except for the biuret method
this urinary protein in clinical laboratories until the early that reacts with the peptide bonds in all proteins, most
1970s. methods detect albumin more effectively than globulins,
and as a result, these methods are poorly accurate. Also,
Although some continue to use high-resolution protein the calibrators may contain only albumin or an array of
electrophoresis (HRPE) (Table 1, Method 3) as a screen proteins different from those found in samples producing
for BJP, it is now recognized that immunofixation more variability.
electrophoresis (IFE) (Table 1, Method 5) is the most
sensitive and specific method for screening and Since some samples may contain large amounts of protein
identification of BJP. This is the method recommended by with very little BJP, whereas others may contain large
the College of American Pathologists (CAP) Diagnostic amounts of BJP with little other protein, measurement of
Resource Committee [5] and the International Federation urinary protein is not necessary prior to specifically
of Clinical Chemistry (IFCC) Committee on Plasma assaying for BJP and may be misleading. What is
Proteins [5,6] (see Table 1, Method 5). A new, enhanced important is to maximize the sensitivity of detecting very
nephelometric/turbidimetric method for identifying free low concentrations of BJP on IFE by concentrating all
light chains in serum has recently been developed and is in samples > 100-fold for the initial screen [5].
a continued process of being tested (Table 1, Method 6)
[7]. Identification of BJP by Urine HRPE (UPE) and IFE (See
Immunoelectrophoresis Method)
Reference and Preferred Methods UPE and IFE should be performed on all samples
Reference Methods suspected of Bence Jones proteinuria (BJPuria) (see Table
There are no reference methods for any of the techniques 1, Methods 3 and 5). These initially should be performed
or procedures discussed in this paper. on samples concentrated approximately 100- to 150-fold
and directly applied to the gel without any dilution,
Preferred Methods regardless of the amount of protein. If the gel is
Measurement of Total Protein overloaded and the pattern is confusing, it may be
Most laboratories screen urine for total protein with a dip necessary to rerun the assay after making dilutions with
strip impregnated with a pH indicator dye, followed by a saline. The sensitivity of UPE for BJP on concentrated
more specific confirmatory test such as precipitation with samples is in the order of 10 mg/L [6], but IFE is about
fivefold more sensitive [9]. Nevertheless, UPE is very
i helpful for identifying severe prozoning of very large BJP
Bence Jones Protein that may occur with IFE, so the two techniques
Previous and current authors of this method: supplement one another and should be run together, as will
First edition: Gayle Birkbeck be discussed below [10]. Unless there is very strong
Methods edition: Gayle Jackson reason to suspect a BJP, for initial screening only, kappa
Second edition: Not updated and lambda antisera against FLC and light chain bound to
Third edition: Not updated heavy chain (intact) (F & B) should be run (antisera
Fourth edition: Gayle Jackson against IgG, IgA, IgM, and free-light-chain-only antisera
Fifth edition: Stanley Levinson
220
Bence Jones Protein
is not needed).The older immunoelectrophoresis technique The CAP committee recommends a 24-hour urine
is far less sensitive and should not be used for identifying collection [5,17], whereas the IFCC Committee
BJP in urine [5,6] (Table 1, Method 4) and is no longer recommends a second morning-voided specimen, because
used. 24-hour collections are cumbersome [6]. Either collection
is probably adequate for an initial screen, but if the
If the initial screen with F & B kappa and lambda indicates analysis is negative for a BJP and disease is still suspected
that there is a BJP, the assay should be repeated using or to better rule out amyloid AL, a repeat 24-hour
antisera against IgG, IgA, and IgM to ensure that the band collection should be performed. Random collections are
is not an intact immunoglobulin. As a cost cutting measure not sufficient to rule out BJPuria. Urine specimens stored
and because of its low avidity compared to antibody at 4C for 5 to 7 days remain stable for testing;
against F & B light chain, the IFCC Committee nevertheless, the IFCC Committee recommends that if a
recommends fixation only with F & B antisera and not 24-hour collection is used, 1g sodium azide be added to
concomitantly with antisera against free light chain only, the container [6]. The urine specimen should be
unless it is suspected that a BJP is co-migrating with or concentrated > 100-fold before electrophoresis for reasons
near an intact monoclonal protein [6]. that will be discuss below.
Identification of Light Chains by Immunonephelometry or Interferences

Turbidimetry Free hemoglobin, lysozyme and some beta-migrating
Some laboratories measure kappa and lambda in proteins may be identified as a paraprotein on UPE but
unconcentrated urine by routine nephelometric and will not react on IFE.
turbidimetric methods, using antibody against F & B Characteristics
kappa and lambda (Table 1, Method 7) [11]. This Pathophysiological Characteristics of Free Light
approach can be used as a screen. If the kappa-to-lambda Chains
(ratio is increased or decreased, the urine can then be Importance of Identification
assayed by IFE after 100-fold concentration to see if there Identification of and quantification of monoclonal FLC is
is a BJP present. If the kappa and lambda are both important for the following reasons:
undetectable or the ratio is within the reference interval, it 1. In overt hematological malignancy, such as
is very unlikely that there is a BJP present [11]. The major multiple myeloma and Waldenstrms
problem with this approach is that because of poor macroglobulinemia, they portend a greater threat
methodological sensitivity, a majority of urines show only of kidney disease and poorer prognosis,
a kappa or lambda that is undetectable while the other especially lambda free light chains.
light chain type is detectable, so a useful ratio cannot 2. In about 10% of myeloma cases, only an FLC is
be calculated (i.e., kappa < 1.5 mg/dL but lambda = 5 secreted, and its concentration can be used to help
mg/dL or lambda < 5 mg/dL but kappa = 10 mg/dL). gauge the effect of therapy and disease
Generally, this approach can eliminate about 30% of the progression.
specimens from the need for IFE. 3. In cases of amyloidosis not associated with an
overt hematological malignancy, where there is
Recently an enhanced nephelometric assay using not a clear overproduction of plasma cells or B-
polyclonal antibodies against FLCs attached to latex lymphocytes, identification of a monoclonal FLC
particles was developed. It is commercially available and is presumptive evidence for amyloidosis AL, and
has been and continues to be widely tested (Freelite from failure to identify a monoclonal FLC suggests a
The Binding Site Inc, San Diego, CA 92121) (Table 1, different form of amyloid.
Method 6). Studies suggest this approach is not very
useful for assaying FLC in urine [12], but because of its Normal Physiology and Characteristics of Free Light
great sensitivity (down to 1mg/L) [13], this method is Chains
useful for identifying FLC in serum, where it shows In some cases, BJP may exhibit deletions or other peculiar
approximately 150-fold greater sensitivity than IFE modifications, owing to the malignant nature of the cells
[7,14,15]. Practical problems associated with this method producing them. However, the characteristics of BJP are
include inaccuracy in recovery with dilution, inaccuracy usually very similar to normal polyclonal FLC, with the
due to polymeric aggregates, concerns of incomplete major difference being that BJP are produced by a single
reaction between antibodies and structurally peculiar BJP, clone that produces either kappa or lambda free light
and imprecision between different instruments [16]. This chains but not both, whereas normal FLCs are produced
method appears to be particularly useful for following by multiple clones, including both kappa and lambda.
patients with nonsecretory myeloma and replacing UPE
for periodic monitoring of most patients with BJP to Polyclonal FLCs are synthesized in excess of heavy
follow the progress of their tumor load. It may also be chains, and normally very small amounts of polyclonal
useful to help rule out amyloidosis AL when urine IFE is FLCs are found in serum and spill over into the urine
negative before proceeding with testing for other types of [8,18]. The free light chains are peptides of about 22 kDa.
amyloidosis (Figure 1). They may exist as monomers or as dimers with a
molecular weight of about 44 kDa [18]. The dimers may
Specimen be covalently disulfide-linked or noncovalently linked
[18]. Normally, more kappa than lambda chains are
synthesized, and since light chains measured in serum are
221
Bence Jones Protein
most commonly those in complete (intact) Multiple Myeloma, Waldenstrms Macroglobulinemia,

immunoglobulin molecules (bound light chainsB) and Other Overt Hematological Malignant Proliferations
(mostly IgG), normal serum / ratios are about 2 to 1 [8]. [22]
Studies that measure only FLC and not intact light chains Most frequently, BJP is seen in urine from patients with
suggest the normal polyclonal free light chain / ratio in multiple myeloma, which is a malignant proliferation of
serum is lower than the intact ratio (between 0.3 and 1.6), plasma cells in bone derived from a single abnormal clone
either because of the synthesis and secretion of more (plasma cell dyscrasia). Table 2 illustrates criteria that
lambda than kappa chains or because of preferential define myeloma. Besides multiple myeloma, BJP may be
filtration of -monomers and dissociable dimers through associated with Waldenstrms macroglobulinemia (in
the glomerulus, or both [7,19,20]. The / ratio in urine is which small lymphocytes and plasmacytoid lymphocytes
higher, about 3 to 1 [7,19,20]. in the bone marrow secrete IgM), with B cell lymphomas,
and with leukemia, especially advanced chronic
The structure of light chains characteristically shows a 107 lymphocytic leukemia, where identification probably adds
N-terminal half and a kappa C-terminal half, with 107 little to the diagnosis or predicted outcome.
residues or a lambda C-terminal with 105 residues [3,4].
Amino acid sequence and immunochemical studies In most cases of myeloma and in Waldenstrms, an intact
indicate that the variable region consists of three regions monoclonal protein is found in sera, and it can be used as
of extensive sequence variability (hypervariable region), the main noninvasive marker for following tumor burden.
and regions of lesser variability that provide structural In about 10% of myeloma cases, a BJP is found in the
framework (framework regions, FR). Homologies in absence of an intact monoclonal protein. In these cases,
amino acid sequence, especially those in the first 22 estimation of the monoclonal FLC level is the best
lambda residues and first 23 kappa residues of the FR1 noninvasive marker for disease progression.
region, chemically define four subgroups of kappa and six
of lambda chains [4]. Monoclonal Gammopathy of Undetermined Significance
(MGUS) (See Method 4, Immunoelectrophoresis)
Clearance In some cases, low concentrations of BJP are seen in the
The primary means for elimination of proteins with urine of patients without apparent disease [23,24].
molecular weights below 60 kDa, including FLCs, is by However, most persons with BJP develop overt disease
filtration through the glomerulus, followed by reabsorption within 8 to 20 years [25]. It is therefore recommended that
and catabolism by the cells of the proximal tubules. Since such patients be kept under observation indefinitely. It is
2 to 3 g of protein is filtered per day and only about 75 to far more unusual to find MGUS in the form of BJP in
150 mg/L is excreted, most of the protein is reabsorbed. urine than as intact immunoglobulins in serum. This may
Normally, FLCs account for about 40% of the be due to the toxicity of the BJP, as compared to intact
immunoglobulin in urine [8]. Persons with impaired monoclonal proteins that may persist in serum for a
tubular reabsorption show an increased concentration of considerable period of time after its secretion without
polyclonal free light chains in urine [4,8]. causing harm [25].
Characteristics and Excretion of BJP in Disease Amyloidosis

As would be expected on the basis of normal FLC Many proteins have a tendency to aggregate. This
structure, most commonly -BJP appear as disulfide- tendency seems to be especially great when proteins are
linked covalent dimers, and -BJP as noncovalent dimers, denatured or in an abnormal conformation. In such cases,
stable monomers, or mixtures of the two [4,21]. BJP may aggregation of the abnormal proteins may take on an
also be found as smaller fragments with molecular weights amorphous form that deposits in tissue. When this
< 22 kDa and as anomalously larger forms > 45 kDa. proteinaceous material is composed of organized fibrils
Those with molecular weights of approximately 55 kDa that are arranged in an antiparallel conformation and -
may contain covalently linked carbohydrates such as sialic pleated sheet structure when examined by x-ray
acid which are not normally associated with light chains diffraction, the deposits are called amyloid [26]. In
[9]. Less commonly, BJP have been identified as amyloidosis AL, the fibers are composed of FLC [27,28].
trimolecular complexes, tetramers, and even higher As compared to myeloma, where -type chains are seen
polymer forms [8,14]. more frequently, chains occur more often in
amyloidosis. Systemic amyloidosis comprises a
BJP usually appears in the urine because of overflow heterogeneous group of diseases that have in common the
conditions in which the concentration of FLCs in the renal extracellular deposition in tissue of a hyaline-like material
tubules is greater than the amount that can be reabsorbed, which when stained with Congo red, exhibits apple-green
either because of high concentrations in plasma or because birefringence under the polarizing light microscope. This
of damage to the kidney. BJP may cause tubular disease in appearance under polarized light is the most common
the form of cast nephropathy or Fanconis syndrome [8], method used to identify amyloid. Although the electron
or they may cause amyloidosis AL, usually leading to microscope is more sensitive and is employed for
glomerular failure. In early cases of tubular disease, serum suspected cases when light microscopy is nondiagnostic
BUN and creatinine may not be elevated, but proteinuria [26], it usually adds little to a carefully performed Congo
may appear. red stain [8].
222
Bence Jones Protein
The cerebral amyloidosis of Alzheimers disease is a case Eighty-six percent of patients with AL show a BJP in
where amyloidosis is primarily found in the brain due to urine [34]. Thus the importance of identifying a BJP
the accumulation of beta-amyloid from a normal associated with amyloidosis cannot be overemphasized. In
transmembrane protein-amyloid beta-precursor protein. BJP-negative cases of AL amyloidosis, although the bone
Prion disease appears to be the case where abnormal marrow may show a normal or only slightly elevated
proteins (proteinaceous infectious particles) that cause number of plasma cells, a clonal predominance of either -
encephalopathy can actually induce the unfolding of other or -type plasma cells can often be identified by
proteins [29], although the amorphous form associated immunohistochemical staining of a biopsy [30]. Together,
with prion disease is different from that of amyloidosis. these two techniques can allow for presumptive
identification of amyloidosis AL in most cases. Figure 1
Types of Systemic Amyloidosis shows a diagnostic algorithm for identifying the type of
Table 3 shows the most common types of amyloidosis, the amyloidosis [30].
causative protein, and clinical features [30]. Amyloid AL
is the most common. It affects the kidney, heart, Interpretation
gastrointestinal tract, liver, tongue, nerves, skin, lungs, HRPE and IFE Identification of BJP
adrenals, and thyroid, with involvement of the latter three Figure 2 shows HRPE UPE, illustrating two gel strips with
organs often going unrecognized [30]. high-concentration BJPs. Usually BJP migrate in the beta
to gamma region. However, as illustrated in Figure 2,
Diagnosis of Systemic Amyloidosis where a BJP can be seen in the beta-1 region on track 1,
Because of the variety and widespread organ involvement because of their small size, they may migrate anywhere
with diverse symptoms, amyloidosis may imitate other between the alpha and cathodal end of the gamma region.
syndromes, so it is often undiagnosed or misdiagnosed; These paraproteins should be definitively identified by IFE
when diagnosed, it is usually well advanced. About 10% to be certain they are monoclonal BJP and not intact
of patients with myeloma develop amyloidosis [30]. In the immunoglobulins or some other protein such as lysozyme
absence of myeloma or Waldenstrms, the disease may (see Figure 4).
be suspected in presentations of peripheral nervous system
disease (especially carpal tunnel syndrome), macroglossia, Figure 3 shows a typical IFE from the sample seen on
or an echocardiogram that shows restrictive track 4 from Figure 2, where a lambda BJP is confirmed,
cardiomyopathy with a granular sparkling appearance since there is no band seen with IgG, IgA, IgM, or kappa
[31]. It may be identified on a biopsy done to investigate IFE. Figure 4 shows a lysozyme band on urine UPE that
kidney or liver involvement. In the case of amyloid AL, could be confused with a cathodally migrating BJP. IFE
the first clue may be a urine electrophoresis that shows a for immunoglobulins (not shown) was negative, and
BJP in a patient with chronic renal disease. A rectal or fat- lysozyme was confirmed by IFE, using antibody against
pad biopsy with Congo red staining will be positive about lysozyme as seen in Figure 4.
85% of the time for all types of amyloidosis [32,33]. If
suspicion is high and a fat-pad biopsy is negative, the Problems of Overloaded Gels Secondary to 100-fold
affected organ should be biopsied [30]. Concentration of Urine
Figure 5 shows an overloaded gel. One difficulty with
Misdiagnosis of Systemic Amyloidosis performing IFE when the concentration of an
Identification of amyloid deposits does not define the type immunoglobulin is very high is that it may be in the zone
of disease. One study showed that 34 of 350 patients of antigen excess (see Method 4, Immunoelectrophoresis),
(9.7%) were misdiagnosed as AL, when they actually causing prozoning when its specific antisera is applied to
manifested other types [34]. Subsequent correct the gel. In the mild case, this may cause the band to look
identification showed 18 patients were AFib, 13 patients like a donut, as seen on the strip under F & B kappa 0
were ATTR, 2 patients were AApoA-I, and 1 patient was (urine as collected, not concentrated or diluted). In Figure
ALys. Some misdiagnosis occurred because a small serum 5, as the sample is diluted, the antigen and antibody
MGUS (often < 10 g/L) was seen in about a third of the concentrations become more equivalent, and the band
misdiagnosed patients, erroneously leading clinicians to appears more and more discrete, as seen with the 1:10
conclude the condition was amyloidosis AL. dilution, with the faded area disappearing. In the figure,
the band on the initial UPE appears so large that it could
Another reason for misdiagnosis is that although be mistaken for being polyclonal, but it is clearly
immunohistochemical staining of effected tissue almost monoclonal. Thus a BJP is clearly apparent from analysis
always identifies amyloidosis AA, monoclonal light chains of the 100-fold concentrate by IFE, since the dense area on
indicative of AL are identified only about 30% of the time UPE combined with extensive prozoning when fixed with
[34], and amyloid deposits are only found in bone marrow F & B kappa, but not lambda, implicate a -BJP. For
about 30% of the time. Moreover, in amyloidosis that is demonstration, the IFE was repeated with dilutions. Notice
unassociated with overt B-cell dysplasia, plasma cells in that antibody against free kappa does not show a reaction
the bone marrow are usually within the normal range or with the 100-fold concentrate. This occurs because of its
only slightly elevated (< 5% to 10%). As a result, bone lower avidity and is the reason the IFCC Committee on
marrow biopsy is usually nondiagnostic. Plasma Proteins does not recommend using free antibody
for fixation except when F & B antibody is used
Importance of BJP Identification in Correct Amyloid concomitantly [6].
Diagnosis
223
Bence Jones Protein
Very Low Concentration BJP although sometimes one to three bands will be seen.
The reason it is necessary to perform IFE on very Although a ladder pattern per se does not reflect an
concentrated urine is to maximize the sensitivity for abnormal condition, it may obscure a low concentration
detecting low concentrations of BJP that are so important BJP migrating coincidental with the pattern. In such a
for identifying amyloidosis AL. Figure 6 illustrates four case, the presence of a tiny amount of BJP may be
cases in which the BJP concentration is too low to be seen overlooked. Polyclonal multiband patterns may also
with UPE, where only albumin is seen, but a BJP is interfere with the interpretation of IFE if the denser band
identified by the more sensitive IFE. If lesser concentrated is mistakenly classified as a BJP. These ladder patterns are
samples were used, some BJP may not have been most commonly seen in samples from patients with
identified by either technique. Before these low proteinuria that is unrelated to BJPuria. Sometimes they
concentration BJPs were identified, these patients were are seen in very concentrated samples from normal
thought to have nephropathy due to hypertension and persons. Since IFE is not a quantitative technique with a
diabetes. The correct diagnosis of amyloidosis was made well-defined lower reference limit for detection, systems
only after Congo red staining of a biopsy, based on the with sufficient separation power to discern these ladder
finding of a BJP in urine. patterns may offer little advantage over otherwise sensitive
HRPE IFE systems that do not.
Use of Antibody Against FLC Only
Antibody against free light chain only is useful for Quantification of FLCs from Urine by UPE and
identifying a BJP when it is migrating nearly in the same Densitometry
place as an intact monoclonal protein, as illustrated in Quantification of BJP by electrophoresis is fraught with
Figure 7A. It is also effective in ruling out a BJP when an problems for the following reasons:
intact monoclonal immunoglobulin without an apparent 1. Except for biuret, most urine total protein dye
BJP is present in urine, as illustrated in Figure 7B. Figure assays react 20% to 40% less with
7A shows a kappa BJP migrating very close to an intact immunoglobulins than with albumin, and it is
monoclonal IgG-kappa that would react with the F & B unclear to what degree FLCs are measured.
kappa antibody and with the free-kappa-only antibody. On 2. BJP react poorly with most immunometric
the other hand, in Figure 7B, where there seems to be an quantification assays [38], so the common
intact IgG-kappa and no BJP, no reaction with the approach for measuring them has been by
antibody against free light chain only is seen. The densitometry, and the densitometer relies on the
antibody against free light chain only was diluted over a total protein concentration, so it must be semi-
range of concentrations to make sure a band was not quantitative at best.
masked (no reaction) by prozoning due to the lower 3. BJP often migrate very close to or coincidental
avidity of the free-only antibody. with an intact immunoglobulin (Figure 7A) or
superimposed upon normal proteins that migrate
Problem of Polyclonal Multiband Patterns in the beta region of the gel (e.g., on transferrin
Polyclonal FLCs may show diffuse patterns or multiple that has leaked into the urine due to renal
banding patterns (ladder pattern) overlying diffuse patterns disease).
upon IFE and isoelectric focusing. The ladder pattern,
observed as two to seven equally spaced bands, is more These features may make quantification difficult and in
often associated with kappa but may also be found on some cases impossible. Nevertheless, sufficient
analysis of lambda light chains and can be seen in most assessment of the amount can usually be made either
urine specimens if the sample is highly concentrated [35- quantitatively or qualitatively, to be useful if needed for
37]. Figure 8 illustrates this type of pattern in samples prognosis or to follow therapy. Although the exact amount
fixed for kappa and lambda in comparison with a sample determined by testing may vary from the actual amount of
containing a true BJP. Multiple banding patterns usually BJP, it will generally be clear from visual evaluation of the
appear as two to five bands, with one being denser than density of the band by UPE and IFE, whether the amount
the others, largely depending on the resolution of the is very large or not [38].
agarose gel system. It was demonstrated that this pattern is
a property of normal free light chains by inducing the Analysis of FLC in Serum by Nephelometry or
transient appearance of multiple kappa bands with arginine Turbidimetry (Table 1, Method 6)
infusion in normal healthy volunteers [35]. The multiple Studies have shown that the greater sensitivity of these
banding pattern appears to be largely a product of charge methods allows more complete identification of FLC in
differences [36], possibly due to the four kappa and six the serum of persons with nonsecretory myeloma [13] and
lambda structural framework regions in the Fab regions amyloidosis with negative IFE results than do
[3,9]. electrophoretic analysis of serum and urine. It also appears
to be a more accurate marker of complete remission [15].
Owing to poorer resolving power, IFE devices from some It has been suggested that when a monoclonal protein is
manufacturers do not appear to have the ability to identified in serum by HRPE, it is not necessary to
routinely distinguish this pattern. With some systems, perform a urine IFE study but to assay the same serum
electrophoresis of polyclonal free light chains in high using the FLC immunoassay [39]. If the serum shows a
concentration will often show ladder band patterns with normal FLC ratio, it is unlikely there is a monoclonal
five to seven bands. Patterns on other systems will more FLC. If the FLC / ratio is abnormal, an FLC is likely,
often show diffuse staining zones without ladder patterns,
224
Bence Jones Protein
and the urine would then be tested by IFE for chain, so that F & B antisera will show a precipitant ring,
confirmation. This would eliminate the need to but free-light-chain-only antisera will not (see Figure 7).
concentrate and assay many urines by IFE, thereby For this reason, it is advisable to check each lot of new
simplifying the process. It is important to remember that free-light-chain antisera carefully against urines with
the serum FLC assay measures polyclonal FLCs and is not known BJP and against normal serum that does not
specific for monoclonal FLCs. An abnormal ratio is contain significant amounts of free light chain before use.
considered presumptive identification of monoclonal Control sera with exact cutoff limits are available for
FLCs. Thus the method is apt to produce inaccurate results methods that quantify kappa and lambda concentrations
when a monoclonal gammopathy is concomitant with a (Table 1, Methods 6 and 7). Controls are supplied by the
polyclonal increase that is not uncommon, since the same company that makes the kit.
ratio is likely to be within the normal range.
Performance Goals
Also, greater sensitivity is a two-edged sword when used As with most testing, performance goals are best assessed
for general screening; it identifies more true positives but by proficiency testing. The CAP provides testing
usually more false positives as well. Moreover, assessments for monoclonal free light chains in urine
identification of low-level monoclonal proteins is (Bence Jones Protein specimens). These require qualitative
problematic, since the great majority are found in older identification by IFE as positive or negative (Table 1,
people who will not develop disease from it in their Method 5) and quantitative testing by immune-
lifetime (MGUS) [40]. Even in patients with higher levels onephelometry or turbidimetry (Table 1, Method 7).
of monoclonal proteins that are inadvertently identified in
serum, therapy is often withheld until symptoms develop Performance is judged acceptable on the basis of the
[22]. Therefore, more sensitive screening methods, apt to following:
identify more patients for questionable follow-up, may not 1. Bands on IFE should be sharp (Table 1, Method
be desirable. 5) and meet all the QC requirements listed above.
2. The IFE findings from the CAP surveys should
This method is most useful for following therapy in cases be in accord with the majority of laboratories in
of nonsecretory myeloma and most cases of amyloidosis the CAP survey, as listed on the CAP critique.
AL once the diagnosis is established. Even if accuracy is These results are listed qualitatively as kappa or
not exact and varies between instruments and between lambda free light chain present or BJP absent.
people [16], changes with therapy should have clinical 3. Results of urine immunonephelometry or
value. The change in magnitude should be related to the turbidimetry (Table 1, Method 7) should be in
initial value in the same person, and results should be agreement with the quantitative mean obtained by
superior to the present densitometric scans or visual the laboratories participating in the survey and
evaluation from electrophoretic techniques discussed within the limits of acceptability provided in the
above. That is to say, once a BJP has been identified in survey critique.
urine, a baseline for free light chain can be established by
this test in serum. It is then much simpler to follow the There is as yet no proficiency testing for the ultrasensitive
progress of the disease by monitoring the FLC in serum serum FLC method (Table 1, Method 6). Laboratories
than to continue to attempt to follow the BJP in using this test should install their own proficiency testing.
concentrated urine by densitometry. In some cases of One way to accomplish this is to obtain the serum of a
proven amyloidosis AL, IFE on urine is not sensitive patient with a BJP in urine and assay the serum using the
enough to identify a BJP in urine or a free light chain in serum FLC method.
serum, but the enhanced assay shows an elevated FLC Since QC sera is available from the manufacturer of the
[39,41]. In such cases, progress of the disease can continue kit, a positive result from a patient with a known BJP in
to be followed by the serum FLC assay without the need urine, along with daily QC results that are within the
for IFE on urine. It may also be useful for following (1) manufacturers control range, should ensure proper
early remission after therapy, since the FLC reflects the performance.
status earlier than intact monoclonal proteins because of References
more rapid turnover [41], and (2) in cases of documented 1 Bence Jones H. Papers on chemical pathology:
amyloidosis but a negative BJP in urine, where a normal lecture III. Lancet 1847; ii: 88-92.
/ ratio would strongly suggest testing for types of 2 Edelman GM, Gally JA. The nature of Bence
amyloidosis other than AL (Figure 1). Jones proteins: chemical similarities to
polypeptide chains of myeloma globulins and
Quality Control and Performance Goals normal -globulins. J Exp Med 1962; 116: 207-
Quality Control (QC) 227.
Since the results of IFE (Table 1, Method 5) are 3 Solomon A, Frangione B, Franlin EC. Bence
interpretive, exact control limits are not defined. Most Jones proteins and light chains of immuno-
electrophoretic plates contain wells at the bottom of the globulins. J Clin Invest 1982; 70: 453-460.
plates for control sera that qualitatively indicate whether 4 Solomon A. Light chains of immunoglobulins:
the reaction has occurred: a precipitant ring indicates a structural-genetic correlates. Blood 1986; 68:
positive reaction (see Figure 3). Most commercial controls 603-610.
contain intact, bound light chain but not unbound free light
225
Bence Jones Protein
5 Keren DF, Alexanian R, Goeken JA, Gorevic PD, polymerism, mechanisms of renal handling and
Kyle RA, Tomar RH. Guidelines for clinical and clinical significance. Scan J Clin Lab Invest
laboratory evaluation patients with monoclonal 1981; 41(Suppl 157): 15-83.
gammopathies. Arch Pathol Lab Med 1999; 123: 19 Abe M, Goto T, Kosaka M, Wolfenbarger D,
106-107. Weiss DT, Solomon A. Differences in kappa to
6 Graziani M, Merlini G, Petrini C. IFCC lambda (kappa:lambda) ratios of serum and
Committee on Plasma Proteins; SIBioC Study urinary free light chains. Clin Exp Immunol
Group on Proteins. Guidelines for the analysis of 1998; 111: 457-462.
Bence Jones protein. Clin Chem Lab Med. 2003; 20 Ambrosino DM, Kanchana MV, Delaney NR,
41: 338-346. Fineberg RW. Human B cells secrete
7 Katzmann JA, Clark RJ, Abraham RS, Bryant S, predominantly l L chains in the absence of H
Lymp JF, Bradwell AR, Kyle RA. Serum chain expression. J Immunol 1990; 146: 599-602.
reference intervals and diagnostic ranges for free 21 Solomon A, Weiss DT. Serologically defined V
kappa and free lambda immunoglobulin light region subgroups of human light chains. J
chains: relative sensitivity for detection of Immunol 1987; 139: 824-830.
monoclonal light chains.Clin Chem 2002; 48: 22 Alexanian R, Weber D, Liu F. Differential
1437-1444. diagnosis of monoclonal gammopathies. Arch
8 Levinson SS, Keren DF. Free light chains of Pathol Lab Med 1999; 123: 108-113
immunoglobulins: clinical laboratory analysis: 23 Pezzoli A, Pascali E. Bence Jones protein in
critical review. Clin Chem 1994; 40: 1869-1878. multiple sclerosis. Clin Chem 1987; 33: 1923-
9 Levinson SS, Elin RJ, Yam L. Light chain 1924.
proteinuria and lysozymuria in a patient with 24 Pascali E, Pezzoli A. The clinical significance of
acute monocytic leukemia. Clin Chem 2002; 48: pure Bence Jones proteinuria at low
1131-1132. concentration. Am J Clin Pathol 1989; 91: 473-
10 Levinson SS. Urine protein electrophoresis and 475.
immunofixation electrophoresis supplement one 25 Kyle RA, Greipp PR. Idiopathic Bence Jones
another in characterizing proteinuria. Ann Clin proteinuria. N Engl J Med 1982; 306: 564-567.
Lab Sci 2000; 30: 79-84. 26 Frangione B. Systemic and cerebral amyloidosis.
11 Levinson SS. An algorithmic approach using Annal Med 1989; 21: 69-72.
kappa/lambda ratios to improve the diagnostic 27 Glenner GG, Terry W, Harada M, Ibersky C,
accuracy of urine protein electrophoresis and to Page D. Amyloid fibril proteins : proof of
reduce the volume required for homology with immunoglobulin light chains by
immunoelectrophoresis. Clin Chim Acta. 1997; sequence analysis. Science 1971; 172: 1150-
262: 121-130. 1151.
12 Le Bricon T, Bengoufa D, Benlakehal M, 28 Ibersky C, Ein D, Page DL, Harada M, Glenner
Bousquet B, Erlich D. Urinary free light chain GG. Immunochemical cross-reaction of human
analysis by the Freelite immunoassay: a amyloid proteins with human immunoglobulin
preliminary study in multiple myeloma. Clin light polypeptide chains. J Immunol 1972; 108:
Biochem. 2002; 35: 565-567. 486-493.
13 Drayson M, Tang LX, Drew R, Mead GP, Carr- 29 Picken MM. The changing concept of amyloid.
Smith H, Bradwell AR. Serum free light chain Arch Pathol Lab Med 2001; 12: 538-43.
measurements for identifying and monitoring 30 Falk RH, Comenzo RL, Skinner M. The systemic
patients with nonsecretory multiple myeloma. amyloidoses. New Engl J Med 1997; 337: 898-
Blood 2001; 97: 2900-2902. 908.
14 Abraham RS, Katzmann JA, Clark RJ, Bradwell 31 Kushwaha SS, Fallon JT, Fuster V. Restrictive
AR, Kyle RA, Gertz MA. Quantitative analysis of cardiomyopathy. N Engl J 1997; 336: 267-275.
serum free light chains. A new marker for the 32 Westermark P, Stenkvist B. A new method for
diagnostic evaluation of primary systemic the diagnosis of systemic amyloid. Arch Int Med
amyloidosis. Am J Clin Pathol 2003; 119: 274- 1973; 132: 522-523.
278. 33 Westermark P, Benson L, Juul J, Sletten K. Use
15 Bradwell AR, Carr-Smith HD, Mead GP, Harvey of subcutaneous abdominal fat biopsy specimen
TC, Drayson MT. Serum test for assessment of fo detailed typing of amyloid fibril protein-AL by
patients with Bence Jones myeloma. Lancet. amino acid sequence analysis. J Clin Pathol 1989;
2003; 361: 489-491. 42: 817-819.
16 Tate JR, Gill D, Cobcroft R, Hickman PE. 34 Lachmann HJ, Booth DR, Booth SE, Bybee A,
Practical considerations for the measurement of Gilbertson JA, Gillmore JD et al. Misdiagnosis of
free light chains in serum. Clin Chem. 2003; 49: hereditary amyloidosis as AL (primary)
1252-1257. amyloidosis. N Engl J Med. 2002; 346: 1786-
17 Kyle RA. Sequence of testing for monoclonal 1791.
gammopathies. Arch Pathol Lab Med 1999; 123: 35 MacNamara EM, Aguzzi F, Petrini C, Higginson
114-118 J, Gasparro C, Bergami MR, et al. Restricted
18 Slling K. Free light chains of immunoglobulins: electrophoretic heterogeneity of immunoglobulin
studies of radioimmunoassay of normal values, light chains in urine: a cause for confusion with
226
Bence Jones Protein
Bence Jones protein. Clin Chem 1991; 37: 1570- serum immunofixation and free light chain
1574. assays. Mayo Clin Proc 2006; 81: 1575-1578.
36 Harrison HH . The"ladder light chain"or pseudo- 40 Kyle RA, Therneau TM, Rajkumar SV, Larson
oligoclonal pattern in urinary immunofixation DR, Plevak MF, Offord JR et al. Prevalence of
electrophoresis (IFE) studies: a distinct IFE monoclonal gammopathy of undetermined
pattern and an explanatory hypothesis relating it significance.Engl J Med 2002; 346: 564-569.
to free polyclonal light chains. Clin Chem 1991; 41 Lachmann HJ, Gallimore R, Gillmore JD, Carr-
37: 1559-1564. Smith HD, Bradwell AR, Pepys MB, Hawkins
37 Hess PP, Mastropoulo W, Thompson GD, PN. Outcome in systemic AL amyloidosis in
Levinson SS Interference by polyclonal free light relation to changes in concentration of circulating
chains with identification of Bence Jones free immunoglobulin light chains following
proteins. Clin Chem 1993; 39: 1734-1738. chemotherapy. Br J Haematol. 2003; 122: 78-84.
38 Levinson SS. Studies of bence jones proteins by 42 Durie BGM, Salmon SE. Multiple myeloma,
immunonephelometry. Ann Clin Lab Sci 1992; macroglobulinemia and monoclonal
22: 100-109. gammopathies. Recent Adv Hematol 1977; 2
:243-61.
39 Katzmann JA, Dispenzieri A, Kyle RA, Snyder 43http://www.multiplemyeloma.org/about_myeloma/2.05.
MR, Plevak MF, Larson DR et al. Elimination of php
the need for urine studies in the screening
algorithm for monoclonal gammopathies by using
Tables
Table 1: Methods of Bence Jones Protein Analysis
Method 1: Heat precipitation of urine; qualitative.
Principle of analysis: Bence Jones proteins precipitate from acidic solutions heated to 45C to 60C and
redissolve on boiling.
Comments: Of historical interest only.
Method 2: Total urinary protein analysis; quantitative
Principle of analysis: Measurement of total urinary protein excretion by dip stick, dye binding, or precipitation.
Very insensitive and nonspecific.
Comments: Usually a measure of albumin, since renal disease usually accompanies BJPuria so that most of the
protein is albumin. The dip stick only measures albumin. Not helpful for identifying BJP.
Method 3: Protein electrophoresis by HRPE after 100 concentration of urine; qualitative
Principle of analysis: Bence Jones proteins separate from other proteins and may be identified as a band in the
gamma region.
Comments: Common; not sufficiently sensitive and cannot identify BJP migrating in the beta region of the gel,
since nonspecific proteinuria often produces other proteins that migrate in this region. Nevertheless, this
procedure is useful when run concomitantly with IFE, because it helps in identifying prozoning that may occur
when very large BJP are present.
Normal reference range: no paraproteins
Method 4: Immunoelectrophoresis of urine; qualitative
Principle of analysis: Bence Jones proteins separated from other proteins by electrophoresis; react with
antibodies to light chains to form visible precipitate.
Comments: No longer recommended and rarely used.
Method 5: Immunofixation after 100 concentration of urine; qualitative
Principle of analysis: Similar to Method 4.
Comments: This is the most common and most sensitive of the electrophoretic procedures and the recommended
approach.
Normal reference range: No kappa or lambda FLC.
Method 6. Analysis of serum / ratio by an enhanced nephelometric or turbidimetric assay: quantitative for any single
patient but not between patients
Principle of analysis: Immunoassay using highly specific polyclonal antibodies for FLC only.
Comments: Proven to be useful as a marker for following nonsecretory myeloma and patients with proven BJP,
because it is much more exact and simple to run than attempting to quantify BJP by HRPE or IFE. Not yet
recommended for screening.
Normal reference range: Free kappa 3.3 to 19.4 mg/L; free lambda 5.7 to 26.3 mg/L, / ratio 0.26 to 1.65.
Method 7. Analysis of urine / ratios using standard immunonephelometric/turbidimetric methods for F & B FLC on
unconcentrated urine.
Comments: Very insensitive but can reduce the need for IFE on concentrated specimens by about 30%.
Normal reference range: / ratio normal for serum or kappa and lambda undetectable.
227
Bence Jones Protein
Figure Legends
Figure 1. Algorithm for identifying the type of amyloidosis (see Table 3 for definition of types).
228
Bence Jones Protein
Figure 2. Example of high-concentration Bence Jones proteins

(BJP) on agarose HRPE UPE after 100 concentration of urine.
1, 2, 3, 4 indicate tracks with different urine samples. Arrows
indicate the positions of the BJP. The region of migration for
common serum protein fractions is indicated. The direction of
migration is towards the anode.
Figure 3. Typical IFE showing definitive

identification of BJP from sample from Figure 2,
track 4. The wells at the bottom of the gel are
control wells showing a circular precipitate. F &
B antisera was used for fixation of light chains.
(See Chapter 6, Electrophoresis for
methodological details.)
Figure 4. Identification of a paraprotein as lysozyme.

Compare this with Figure 3, showing a BJP.
229
Bence Jones Protein
Figure 5. Overloaded gel. 100 concentrated urine was diluted as indicated for IFE. F & B indicates
antisera against free and bound light chain, and F indicates antisera against free light chain
only. For other details, see Figure 3.
Figure 6. Identification of very low concentration BJP by IFE. Arrows point to the BJP bands. Atisera
against F & B light chains used. For more details, see Figure 3.
230
Bence Jones Protein
Figure 7. Usefulness of antisera against free light chain only. A, Above, a kappa BJP migrating close to an intact
monoclonal protein (IgG-kappa). B, below, BJP is apparently ruled out by no reaction with free-light-chain-only
antisera over a range of dilutions. This suggests the reaction by F & B antisera is due to the intact IgG-kappa. F &
B indicates antisera against free and bound light chain and Free or F indicates antisera against free light chain
only.
231
Bence Jones Protein
Figure 8. Normal
multiple banding
(ladder) patterns.
Samples 1 and 3
show ladder
patterns. Sample
2 shows a true
lambda BJP.
Sample 4 shows
the most common
pattern seen with
normal samples
(no staining).
Arrows to the left
indicate the
positions of the
multiple bands.
Arrow to the
right, BJP.
Antisera against
free and bound
light chain used.
Other details
same as in Figure
3.
Table 2: Characteristics for Multiple Myeloma [42,43]

> 30% plasma cells in bone marrow (normal <5%).
Monoclonal protein in serum> 35 g/L IgG or > 20 g/L IgA.
>10% plasma cells in with other evidence: significant monoclonal protein in serum (usually IgG > 20
g/L or IgA >10 g/L or urine BJP, lytic bone lesions, or hypogammaglobulinemia (uninvolved
immunoglobulins).
> 1 g of monoclonal free light chain (BJP) in urine per 24 hours.
Table 3: More Common Systemic Amyloidosis

Type Fibril Composition Clinical Features
AL Immunoglobulin FLC Cardiomyopathy, hepatomegaly, proteinuria, renal
insufficiency, macroglossia, orthostasis,
autonomic and peripheral neuropathy, ecchymoses.
AA Amyloid A protein Underlying inflammatory disorder,

proteinuria, renal insufficiency,
hepatosplenomegaly, macroglossia, orthostasis.
ATTR Transthyretin Midlife onset of (prealbumin)

(prealbumin) autonomic and peripheral,
neuropathy, cardiomyopathy, vitreous opacities.
Other familial types:

AFib Fibrinogen A Neuropathy, hypertension
(Most common)
ALys Lysozyme Neuropathy, hepatomegaly

AApo-A-I Apolipoprotein A-I Polyneuropathy, neuropathy
232
Benzodiazepines
Benzodiazepines
Harold W. Peel
Name: Benzodiazepines
Molecular formula: see Benzodiazepine Table: Drugs
Molecular mass: seeBenzodiazepine Table: Drugs
Merck Index: seeBenzodiazepine Table: Drugs
Chemical class: Benzodiazepines
this method was applicable only to blood concentrations

Principles of Analysis and Current Usage above 0.3 g/mL.[3] De Silva and Puglisi in 1970 used
The benzod1azepines are a group of nitrogen-containing the more stable nickel electron-capture detector (ECD)
compounds that are related by a common 1,4- and the polysiloxane polymer stationary phases, OV-1
benzodiazepine structure (Benzodiazepine Table: Drugs and OV-17, to analyze diazepam, medazepam, and their
). Early analytical methods for the benzodiazepines metabolites in blood and urine and improved the
required only the measurement of a single drug or, at detection limits to about 0.04 g/mL.[4]
most, two or three additional compounds. However,
because of the large number of drugs and metabolites in A 1971 study by Zingales described a GLC method for
the benzodiazepine class, the number of reported chlordiazepoxide that resolved the compound from its
methods is now quite large, involving many procedural metabolites desmethylchlordiazepoxide and
approaches (Benzodiazepines Methods Summary Table) demoxepam.[5] Whereas most earlier reports described
such as gas-liquid chromatography (GLC), high- the use of primarily diethyl ether, benzene, or
performance liquid chromatography (HPLC), chloroform as the extraction solvent, Zingales
spectrofluorescence, ultraviolet spectrophotometry (UV), recommended a solvent of n-heptane containing 1.5%
polarography, thin-layer chromatography (TLC), and the isoamyl alcohol. The method could detect plasma
immunoassay procedure of EMIT. The procedures concentrations of chlordiazepoxide of 0.015 g/mL.
developed for whole blood are usually appropriate for
plasma or serum analysis. A 1976 report by De Silva et al. described a
Gas-Liquid Chromatography (GLC) Procedures comprehensive extraction scheme for many
(see methods 1ac, Benzodiazepines Methods Summary benzodiazepine drugs that included diazepam,
Table) chlordiazepoxide, flurazepam, nitrazepam, clonazepam,
bromazepam, oxazepam, lorazepam, and some of their
The early GLC methods for chlordiazepoxide and metabolites.[6] Blood was extracted by different
diazepam required the hydrolysis of the blood extracts procedures depending on the extractability of the
with 6 M hydrochloric acid to form benzophenone specific benzodiazepines from blood, at pH 9 or 12, by
products, which are more easily chromatographed than use of diethyl ether or the solvent mixture
other products. In 1964, De Silva[1] described a GLC benzene:methyl chloride (9:1). Derivatization by
method that used a tritium foil electron-capture detector methylation or silylation was employed to reduce
and a 2% Carbowax 20M column for the analysis of adsorption or decomposition effects, with the
diazepam. After an ether extraction of blood, diazepam compounds detected as intact 1,4-benzodiazepine or
and desmethyldiazepam were hydrolyzed to form 2- benzodiazepine-2-one. The method was reported as
methylamino-5-chlorobenzophenone (MACB) and 2- simple and readily automated for large-scale clinical
aminochlorobenzophenone (ACB). The method was procedures. Blood concentrations, after a single oral
therapeutic dose of diazepam, bromazepam, clonazepam,
sensitive to 0.02 g/mL. The method was modified by
De Silva et al.[2] in 1966, correcting the problem of or flunitrazepam, were detected from 1.0 to 10.0 ng/mL.
rapid column deterioration by using a more stable
column consisting of 2% Carbowax 20M and A rapid microprocedure for diazepam and
terephthalic acid. These two methods were also desmethyldiazepam that used a plasma volume of 10 to
applicable to chlordiazepoxide analysis, by analysis of 100 L was reported by Rutherford[7] in 1977. Plasma
its benzophenone, ACB. Foster and Frings utilized the was extracted directly with n-butyl acetate that contained
flame ionization detector with a 3% SE-30 column, but prazepam as the internal standard. After thorough
mixing and centrifugation, a 2 L injection was made
1
onto a 3% OV-7 column, equipped with a 63Ni ECD
This chapter was originally prepared by: detector.
Harold W. Peel
233
Benzodiazepines
In a procedure using flame ionization, Baselt et al[8] therapeutic concentrations of flurazepam and its
quantified the thermolabile drugs chlordiazepoxide and metabolites.[15] Methods for directly determining
oxazepam by chromatographing their decomposition lorazepam in plasma were reported by de Groot et al[16]
products on a 2% OV-17 column, after direct extraction and later by Greenblatt et al[17] Although both methods
of 4 mL of blood with n-butyl chloride. The procedure used a similar 3% OV-1 column, Greenblatt et al[17]
was limited to overdose cases involving diazepam, were able to improve the sensitivity to a range of 1 to 3
desmethyldiazepam, flurazepam, and ng/mL, probably because of the selection of an
desalkylflurazepam at concentrations of 0.2 g/mL or extraction solution consisting of benzene containing
greater. 1.5% isoamyl alcohol. Following an analytical approach
similar to that used for lorazepam, Greenblatt et al[18]
A combined GLC-ECD and HPLC procedure for analyzed plasma volumes of 0.5 to 2.0 mL for the potent
common benzodiazepines in blood or plasma was benzodiazepines triazolam and alprazolam. The
described by Peat and Kopjak.[9] They reported the sensitivity was reported to be 0.25 ng/mL. Jochemsen
chromatographic properties of six parent and Breimer[19] reported extracting 2 mL plasma with
benzodiazepines and some metabolites, including GLC the solvent pentane:dichloromethane (4:3) and injecting
retention data on a 3% OV-17 column and HPLC data the residue with a solid injection system onto a SCOT
on a C18 column. HPLC was used to confirm the column consisting of a 0.5% PE-21 and 3% OV-17. The
presence of the benzodiazepines and to quantify sensitivity was 0.5 ng/mL.
chlordiazepoxide and desmethylchlordiazepoxide. More
recently, Levine et al. used this approach in their studies The use of fused silica capillary columns offers excellent
on the stability of benzodiazepines in stored potential for improved GLC analysis of benzodiazepines.
samples.[10] They found that chlordiazepoxide, In particular, the nonpolar DB-1 columns have excellent
desmethyldiazepam, and desmethylchlordiazepoxide characteristics for broad screening with good baseline
were not stable beyond a few days, unless refrigerated or stability. Coupled with the 63Ni electron-capture
frozen. Wallace et al., who also used HPLC in similar detector, this system gives highly resolved peaks for
combination with GLC-ECD, were able to resolve very low amounts of drug. Anderson and Stafford[20]
diazepam and nordiazepam more successfully using an have recently shown the advantages of using a capillary
SP-2250 DB stationary liquid phase.[11] In 1979, Kelly column coated with a nonpolar phase and a flame
et al. reported a sensitive method (from 0.05 g/mL), ionization detector for routine drug screening.
using a 1 mL blood sample, for measuring diazepam,
chlordiazepoxide, oxazepam, flurazepam, and A summary of some of the gas chromatographic
chlorazepate.[12] The procedure used an extraction procedures used for benzodiazepine analysis is presented
solvent consisting of toluene, hexane, and isoamyl in Benzodiazepines Table: Summary of GLC and HPLC
alcohol (78:20:2) and a 3% OV-17 column. Oxazepam methods.
chromatographed reproducibly as its quinazoline
rearrangement product. Chlordiazepoxide was the only
drug examined that was not well quantified by this High-Performance Liquid Chromatography
procedure because the thermal degradation products for Procedures
the parent drug, desmethylchlordiazepoxide and (method 2, Benzodiazepines Methods Summary Table)
demoxepam, both had retention times similar to that of
nordiazepam. The HPLC procedures, which involve direct extraction
and chromatography of the intact drug, may be
In 1980 Peel and Perrigo,[13] in reviewing the previous preferable for thermolabile drugs such as
methodology, reported a broad screening procedure chlordiazepoxide. Both normal- and reversed-phase
requiring 1 mL of blood or serum for 22 compounds chromatography systems have been reported.
consisting of parent benzodiazepines, some of their
metabolites, and benzophenone hydrolysis products. A One technique[21] employed a silica column, Partisil 10,
choice of two columns (10% OV-1, 3% OV-17) with with a mobile phase of n-heptane:isopropanol:methanol
for the quantitative analysis of diazepam and
detection by a 63Ni detector was offered for general desmethyldiazepam. The sensitivity was reported to be
screening or quantification. This system overcame many
0.25 g/mL, using a 2 mL sample of blood.
of the problems of close retention times reported in
previous reports. An optional step that formed the
Greizerstein and Wojtowicz[22] reported a micromethod
benzophenone product provided confirmation using gas
for chlordiazepoxide and desmethylchlordiazepoxide
chromatography or thin-layer chromatography.
that required 50 L of blood to be extracted with
Concentrations of flurazepam and its blood metabolites
were reported in an earlier method that used a 3% QF-1 subsequent analysis on a Bondapak C18 column with a
column and electron-capture detector and had sufficient solvent-gradient program of aqueous potassium
sensitivity to detect levels from 0.005 to 0.01 phosphate and methanol. It was reported that as little as
g/mL.[14] Other investigators used a column of 1% 0.1 g/mL of the compounds in a 50 L sample was
SP-1000, developed by a temperature program and routinely detected.
employing a nitrogen selective detector to analyze
234
Benzodiazepines
Oxazepam, diazepam, and nordiazepam were analyzed acid:pentanesulfonic acid. The methylation step was
using[23] a 2 mL blood sample that was extracted, necessary because carbamazepine had the same retention
purified by back-extraction, and subsequently separated time as clonazepam. The retention times of 45 other
by a reversed-phase column, Partisil-10, using a mobile common basic drugs, along with four benzodiazepines
phase of acetonitrile and 0.1 M aqueous sodium acetate and their metabolites, were included.
(pH 4.6). The sensitivity of this method was 0.3 g/mL
for oxazepam and desmethyldiazepam and 0.4 g/mL Wong[30] described a single-step extraction with
for diazepam. Brodie et al[24] proposed a less toluene and a choice of either a Bondapak C18 or
complicated extraction procedure for benzodiazepine Bondapak phenyl column to separate 15
analysis in 2 mL plasma samples. The method analyzed benzodiazepine types of compounds and metabolites.
diazepam and desmethyldiazepam with a sensitivity of The mobile phase found to be particularly good for
10 and 2 ng/mL, respectively, using a reversed-phase benzodiazepines was methanol:water (65:35), as
C18 Bondapak column with an elution solvent of originally used by Brodie.[24] The method normally
methanol and water. requiring 0.5 mL of blood was sensitive from 0.05 to 0.1
g/mL and was reportedly applicable for triazolam
Ascalone[25] recommended reversed-phase HPLC as the detection with a 2 mL sample.
most suitable method for determining chlordiazepoxide
and its metabolites in body fluids because of its A summary of some of the high-performance liquid
specificity, sensitivity, simplicity, and speed. The chromatographic procedures used for benzodiazepine
method used a LiChrosorb RP-18 column with a analysis is presented in Benzodiazepines Table:
precolumn and solvent of acetonitrile and 0.1% Summary of GLC and HPLC methods.
ammonium carbonate to separate chlordiazepoxide,
desmethylchlordiazepoxide, demoxepam, and UV and Fluorometric Procedures
desmethyldiazepam. Ultraviolet spectrophotometric (UV) and fluorometric
methods (see method 3, Benzodiazepines Methods
In developing a method for determining therapeutic Summary Table) have limited use for the analysis of
levels of diazepam and its metabolites in serum, urine, benzodiazepines because of their lack of specificity in
and saliva, Tjaden et al[26] achieved a separation of nine discriminating between various parent compounds and
benzodiazepines within 12 min using a high- metabolites. In addition, UV is not sufficiently sensitive
performance column of methyl-silica and a 50% aqueous for analysis of many of the newer, more potent drugs.
methanol solvent. The UV methods described by Walberg[31] for
chlordiazepoxide and diazepam both followed similar
By employing a solvent extraction of plasma with extraction procedures on 2 mL of serum. Selective
benzene:isoamyl alcohol as in previous GLC procedures, reextraction with 0.5 N sulfuric acid separated
Divoll et al.[27] separated chlordiazepoxide and its two chlordiazepoxide. The concentration for
metabolites using a C18 Bondapak column with a chlordiazepoxide was determined by the following
difference: A245 nm A 290 nm and A 240 nm A 290
mobile phase of methanol, acetonitrile, and aqueous
sodium acetate. A sensitivity of 0.05 g/mL was nm for diazepam. The limit of detection was reported to
achieved. Chlordesmethyldiazepam was used as the be about 3 g/mL. In a procedure designed for screening
internal standard. and semiquantitative analyses of total
benzodiazepines, Valentour et al.[32] hydrolyzed
Sutheimer and Sunshine[28] described a broad method extracts from blood to produce the benzophenones,
for the analysis of chlordiazepoxide, which were then converted to the fluorescent product 9-
desmethylchlordiazepoxide, flurazepam, acridanone. With an excitation wavelength of 270 nm,
desalkylflurazepam, diazepam, and desmethyldiazepam the test samples were scanned from 400 to 480 nm,
in a 1 mL sample of blood, plasma, urine, bile, or gastric giving the procedure a reported sensitivity of 0.025
content. Clonazepam was used as an internal standard g/mL for diazepam, desmethyldiazepam, oxazepam,
for quantification and was contained in the extraction chlordiazepoxide, demoxepam, and desalkylflurazepam.
solvent of n-butyl chloride. The evaporated extract was However, the sensitivity for flurazepam and
reconstituted in the elution solvent, acetonitrile, and desmethylchlordiazepoxide was only 0.25 g/mL.
0.015 M, pH 3.3 phosphate buffer and separated on an
Altex Ultrasphere-ODS (octadecyl sulfate) column. The Thin-layer chromatography (TLC) (see method 4,
reported sensitivity was generally 0.01 g/mL. Benzodiazepines Methods Summary Table) has been
used to separate and detect benzodiazepines, either as the
In an attempt to develop a procedure that could be used intact compound or as the benzophenone products.
for clonazepam, as well as for a wide variety of other Reviews by Hailey[33] and Clifford and Franklin
drugs requested in therapeutic monitoring, Shaw et Smyth[34] describe various solvent systems and special
al[29] methylated extracts from serum. The derivatized detection methods for TLC.
products were chromatographed on a C18 reversed-
phase column using a complex solvent mixture of Chlordiazepoxide and its metabolites were assayed by
acetonitrile:methanol:diethylamine:phosphoric Straughn et al[35] after separation by thin-layer
235
Benzodiazepines
chromatography, following extraction from 2 mL of of urine to convert glucuronide metabolites to more

serum. The Quanta-Gram LQD plates were exposed to readily detectable immunoreactive compounds improves
red fuming nitric acid and then heated to 100 C to the sensitivity of some methods.
produce intensely fluorescing benzodiazepine spots that
were measured using a spectrodensitometer equipped The homogeneous enzyme-multiplied immunotechnique
with a fluorescent attachment. The sensitivity was about (EMIT) (see method 6, Benzodiazepines Methods
0.1 g. In a similar approach, van der Merwe and Summary Table) for benzodiazepine urine screens was
Steyn[36] used silica gel 60 plates and exposed them to reported by Haden et al.[40] This direct method of
hydrogen chloride gas followed by radiation at 245 nm analysis detected several benzodiazepines and their
for 45 min. This procedure converted diazepam to metabolites with an assay response equivalent to 0.5 g
fluorescent compounds that were measured of oxazepam per milliliter. Poklis[41] evaluated the
quantitatively on a spectrofluorometer equipped with a EMIT urine assay and found no false-positive results,
scanning TLC attachment (exc, 360 nm; em, 460 nm). but about 10% inconclusive or false-negative results.
Using reconstituted organic extracts of blood, bile, or
Palermo and Poklis[37] included TLC data for tissue, Slightom[42] adapted this commercially available
flurazepam, hydroxyethylflurazepam, oxazepam, kit to toxicological screening applications.
diazepam, and chlordiazepoxide in their evaluation of
TLC method for flurazepam in urine. The systems, Reference and Preferred Methods
which were all developed on silica gel GF plates, were The number of benzodiazepines and their metabolites,
as follows: their variable chemical characteristics, and the wide
range of concentrations encountered in therapeutic and
System A: ethylacetate:ethanol:n-butanol:ammonia (70:35:5:1). toxicological analyses require that the clinical laboratory
System B: benzene:chloroform:acetone (40:40:20). use a flexible method that is useful for qualitative
System C: benzene:methanol:acetic acid (90:10:10). screening as well as for quantification purposes.
System D: ethanol:acetic acid:water (50:30:20). Methods involving ultraviolet spectrophotometry are
System E: benzene:dioxane:ethanol:ammonia (50:40:5:5). generally limited by insufficient sensitivity and
System F: ethyl acetate:methanol:ammonia (85:10:15). specificity. For example some drugs that may be
extracted in a similar fashion to and exhibit a maximum
Peel and Perrigo[13] described the TLC characteristics absorbance wavelength in nanometers in the same region
(Rf, visualization) of nine parent benzodiazepines and
as diazepam (240), or desmethyldiazepam (245),
their metabolites and benzophenone products in three
include methapyrilene (238), flurazepam (239),
TLC systems, as a means to confirm GLC identification.
The limit of sensitivity for most of the TLC methods desmethyldiazepam (239), amitriptyline (240),
reported was about 1 to 2 g. lorazepam (240), prazepam (240), and imipramine
(247). The conventional ultraviolet methods for
Using differential pulse polarography (see DPP, method
5, Benzodiazepines Methods Summary Table) as an benzodiazepines are limited to the microgram-per-
approach for selective screening of total benzodiazepine milliliter level because the extinction coefficients are of
concentration in blood, Brooks et al[38] assayed blood the order of 104 L mol-1 cm-1.
for chlordiazepoxide, diazepam, chlorazepate,
exazepam, and their common metabolites before The pulse polarographic procedure, as reported by
quantification by more specific methods. This technique Brooks et al,[38] was more sensitive (0.25 to 0.5 g/mL)
is possible because the output signal in differential pulse than ultraviolet spectrophotometry but was limited in
polarography is much stronger than the signal of specificity. Resolution of oxazepam from diazepam or
conventional direct current polarography. The desmethyldiazepam was difficult because of the poor
benzodiazepine analysis is based on the reduction of the separation of the azomethine peaks. Although useful for
4,5-azomethine group. Reduction peak potentials (volts) single-drug monitoring, this dedicated instrumentation is
for chlordiazepoxide, flurazepam, and diazepam are primarily useful for research applications.
0.655, 0.635, and 0.715, respectively.
Chlordiazepoxide and its desmethyl metabolite are also For large qualitative drug screening programs of urine,
characterized by the N4-oxide reduction peak at 0.335 TLC and homogeneous assay (EMIT) procedures are
V. The DPP method was adapted to direct analysis of most useful. The degree of sensitivity of the TLC
diluted urine or gastric content. procedure is quite dependent on the visualization
technique, with some benzodiazepines fluorescing when
Immunoassay Procedures as little as 0.2 g of drug is present. Combining TLC
A wide variety of immunoassay procedures including techniques with fluorodensitometry for quantification of
EIA, ELISA, FPIA, and agglutination or kinetic drugs and their metabolites is possible but not efficient
interaction of microparticles are the most common for routine or clinical stat analyses since the procedures
methods in use today. These methods demonstrate for the chromophore development usually require 45 to
variable degrees of cross reactivity to benzodiazepines 60 min.
so that the different methods will not all recognize all
benzodiazepines and their metabolites.[39] Hydrolysis
236
Benzodiazepines
The EMIT procedure is sensitive (0.5 g/mL) and very

useful for efficiently screening large numbers of urine The extremely thermolabile nature of chlordiazepoxide
samples for the benzodiazepine class of drugs. It has the and the decline of detector response to this drug limit the
distinct advantage of being very fast, since there are no use of GLC as a method of choice for this particular
extraction procedures involved. However, EMIT is drug. However, by using an internal standard and
limited as a quantitative method because it gives only a developing the calibration graph during each
total measurement of combined parent drug and chromatographic session, one can obtain very good
metabolites. There are no reports of EMIT being used precision. The flexibility of being able to increase the
for detecting the more potent alprazolam and triazolam temperature to quantify later-eluting drugs, or to apply
drugs. temperature programming for selective isolation of
specific peaks, makes GLC the single method of choice
HPLC is often preferred to GLC techniques for for this large class of drugs. Changing operating
benzodiazepine analysis because of the milder column parameters is not as easily accomplished with HPLC.
conditions. However, chlordiazepoxide is probably the
only benzodiazepine drug that may be adversely affected A comparison of the EMIT, GLC, and HPLC procedures
by packed column gas chromatography. Although some is found in Benzodiazepine Table: Comparison of
specific HPLC methods using ultraviolet detectors have methods.
reported a sensitivity as low as 2 ng/mL, sensitivity in
the range of 10 ng/mL is more common. The modified Specimen
method of Sutheimer and Sunshine[28] can be Serum or plasma are acceptable samples for analysis.
recommended because it includes sufficient range and Whole blood (ante mortem or post mortem) is also
sensitivity, which are very useful in the toxicology acceptable for most of the procedures. Storing samples at
laboratory. 4 C or frozen is generally sufficient to ensure stability
for benzodiazepine analysis, though Levine et al[10]
A distinct advantage of GLC procedures is that the recommend that analysis for chlordiazepoxide be
conditions for a broad qualitative screen are also suitable performed as soon as possible. There is no indication of
for virtually all quantitative analyses. The technique interferences incurred by commonly used anticoagulants
does not require individual adjustments to accommodate or preservatives.
the various benzodiazepines. For those situations where Urine specimens can be analyzed, but quantification is
a project will require several quantitative analyses for a not usually useful in judging toxic situations. Since the
later-eluting benzodiazepine, the temperature can be benzodiazepines are mostly present as glucuronide
increased over a minimum equilibration time, and so the conjugates in the urine, conventional glucuronidase
total analysis time is shorter. This type of flexibility for hydrolysis at pH 5 is used to free the bound compounds.
broad screens and adaptability for quantitative
procedures is more easily accomplished by use of gas Interferences
chromatography. Concentrations in patients taking benzodiazepines tend
to be lower in smokers than in nonsmokers. The
The GLC procedures using flame ionization detectors presence or addition of beta-glucuronidase to urine
have sufficient sensitivity for many of the specimens can result in increased sensitivity of
benzodiazepines that are present at higher concentrations immunoassay methods.
in the blood (that is, above 0.2 g/mL) but would not be
applicable to the newer, more potent drugs even at toxic Benzodiazepine Therapeutic Concentration Ranges
doses. All the benzodiazepines have therapeutic concentrations
in serum ranging from 0.1 to 1500 ng/mL, depending on
Gas-liquid chromatography using a 63Ni electron- the particular drug. The concentration values presented
capture detector is the most versatile technique, with in Benzodiazepines Table: Pharmacological and
sufficient sensitivity to be of use for broad therapeutic concentration data can be used only as a
benzodiazepine analysis. The GLC methods that general guide because the effects of these drugs vary
hydrolyze the extracted benzodiazepines to their considerably because of the occurrence of active
benzophenones are usually sufficiently sensitive for metabolites related to the therapeutic regime (single or
chronic dose schedule). Unlike many other drug
therapeutic concentrations (0.02 g/mL), but one can
overdoses, plasma concentrations of the benzodiazepines
presume only a general identification because one
are often not related to or predictive of clinical
benzophenone can result from several parent
outcome.[43]
compounds. The easily prepared benzophenone may be a
useful corroborative procedure to use in some cases after
initial identification. Careful selection of the extraction Interpretation
solvent, buffer pH, and injection solvent can overcome The benzodiazepine group of drugs comprises a
most of the problems of the various drugs, including the relatively large number of compounds that are in
poor recovery of less than 45% (in the case of prominent use as anticonvulsants, muscle relaxants,
oxazepam) and the thermal decomposition, that were hypnotics, and anxiolytics because of their clinical
encountered in many of the earlier methods. effectiveness and relatively low toxicity
(Benzodiazepines Table: Pharmacological and reference
237
Benzodiazepines
interval data). All the benzodiazepines undergo hepatic liquid chromatography, J Chromatogr
biodegradation by dealkylation, hydroxylation, and other 1971;61:237-52.
pathways to produce numerous metabolites that are 6 De Silva JA., Berkersky I, Puglisi CV. et al.
significant because of their biological activity Determination of 1,4-benzodiazepines and -
(Benzodiazepines Table: Benzodiazepines with active diazepin-2-ones in blood by electron-capture
metabolites) or their extended excretion rate. Some gas-liquid chromatography. Anal Chem
metabolites are common to more than one parent drug. 1976;48:10-19.
For example, desmethyldiazepam and oxazepam are 7 Rutherford DM. Rapid micro-method for the
metabolites that occur in the blood after administration measurement of diazepam and
of chlordiazepoxide, diazepam, or chlorazepate. desmethyldiazepam in blood by gas-liquid
Extensively bound to serum albumin, benzodiazepines chromatography. J Chromatogr 1977;137:439-
are excreted in the urine primarily as glucuronide 48.
conjugates of hydroxylated metabolites and related 8 Baselt RC, Stewart CB, Franch SJ.
compounds produced in the liver. Only negligible Toxicological determination of benzodiazepines
amounts of unchanged drug are found in urine. in biological fluids and tissues by flame-
The benzodiazepines have emerged as drugs ionization gas chromatography. J Anal Toxicol
with a useful range of clinical applications and with 1977;1:10-13.
relatively few undesired side effects. However, because 9 Peat MA., Kopjak L. The screening and
of their widespread use, a significant number of cases quantitation of diazepam, flurazepam,
involving intoxication or undesirable episodes do occur. chlordiazepoxide and their metabolites by
The most common adverse effect that does occur is an electron-capture gas chromatography and high
excessive degree of sedation. Toxic concentrations may pressure liquid chromatography. J Forensic Sci
be manifested as drowsiness, or coma. Acute overdoses 1979;24:46-54.
of benzodiazepines taken alone are seldom fatal, but, 10 Levine B, Blanke RV, Valentour JC. Post-
because of their depressant effects, they can be very mortem stability of benzodiazepines in blood
dangerous when ingested with other central nervous and tissues. J Forensic Sci 1983;28:102-15.
system depressants, such as alcohol and barbiturates. 11 Wallace JE, Schwertner HA, Schimek EL.
Patients with plasma benzodiazepine concentrations 2 to Analysis for diazepam and nordiazepam by
10 times above the therapeutic concentrations will electron-capture gas chromatography and by
generally exhibit toxic symptoms. Long-term usage has liquid chromatography. Clin Chem
resulted in cases of benzodiazepine withdrawal 1979;25:1296-1300.
syndrome, as reported by Greenblatt and Shader.[44] 12 Kelly RC, Anthony RM, Krent L, Thompson
Abrupt cessation of these drugs can result in symptoms WL, Sunshine O. Toxicological determination
similar to those associated with the withdrawal of other of benzodiazepines in serum: methods and
central nervous system depressants. Chlordiazepoxide concentrations associated with high-dose
has been used as a substitute during the withdrawal intravenous therapy with diazepam. Clin
process. In updating earlier reports on benzodiazepine Toxicol 1979;14:445-57.
overdose by Greenblatt et al[45] and Finkle et al,[46] 13 Peel HW, Perrigo BJ. Toxicological analysis of
again affirms that serious toxicities from these drugs benzodiazepine-type compounds in postmortem
alone are rare and deaths almost negligible. blood by gas chromatography. J Anal Toxicol
1980;4:105-113.
References 14 De Silva JA, Puglisi CV, Brooks MA, Hackman
1 De Silva JA, Schwartz MA, Stefanovic V, et al. MR. Determination of flurazepam (Dalmane)
Determination of diazepam (Valium) in blood and its major metabolites in blood by electron-
by gas chromatography. Anal Chem capture gas-liquid chromatography and in urine
1964;36:2099-105. by differential pulse polarography. J
2 De Silva JA, Koechlin BA., Bader G. Blood Chromatogr 1974;99:461-83.
level distribution patterns of diazepam and its 15 Ferrara SD, Tedeschi L, Marigo M, Castagna F.
major metabolite in man. J Pharmacol Sci Concentrations of phenobarbital, flurazepam
1966;55:692-6,. metabolites in autopsy cases. J Forensic Sci
3 Foster LB, Frings CS. Determination of 1979;24:61-9.
diazepam (Valium) concentrations in serum by 16 De Groot G, Maes RA, Lemmens HH.
gas-liquid chromatography, Clin Chem Determination of lorazepam in plasma by
1970;16:177-9. electron capture G.L.C. Arch Toxicol
4 De Silva JAF, Puglisi, CV. Determination of 1976;35:229-34,.
medazepam (Nobrium), diazepam (Valium) and 17 Greenblatt DJ, Franke K, Shader RI. Analysis
their major biotransformation products in blood of lorazepam and its glucuronide metabolite by
and urine by electron capture gas-liquid electron capture gas-liquid chromatography. J
chromatography. Anal Chem 1970;42:1725-36. Chromatogr 1978;146:311-20
5 Zingales IA. Determination of chlordiazepoxide 18 Greenblatt DJ, Divol M, Moschitto LJ, Shader
plasma concentrations by electron capture gas- RI. Electron capture gas chromatographic
analysis of the triazolo-benzodiazepines
238
Benzodiazepines
alprazalom and triazolam. J Chromatogr 33 Hailey DM. Chromatography of the 1,4-

1981;225:202-7. benzodiazepines. J Chromatogr 1974;98:527-
19 Jochemsen R, Breimer DD. Assay of triazolam 69.
in plasma by capillary gas chromatography. J 34 Clifford JM, Franklin Smyth W. The
Chromatogr 1981;223:438-44,. determination of some 1,4-benzodiazepines and
20 Anderson WH, Stafford DT. Applications of their metabolites in body fluids, a review.
capillary gas chromatography in routine Analyst 1974;99:241-72.
toxicological analysis. J High Resolut 35 Straughn JL, Cathcart-Rake WF, Shoeman DW,
Chromatogr Commun 1983;6:247-54. Azarnoff DL. Quantitation of chlordiazepoxide
21 Bugge A. Quantitative high-performance liquid and its major metabolites in biological fluids by
chromatography of diazepam and n- thin-layer chromatography. J Chromatogr
desmethyldiazepam in blood. J Chromatogr 1978;146:473-80
1976;128:111-16. 36 Van der Merwe PJ, Steyn JM. Thin-layer
22 Greizerstein HB, Wojtowicz C. Simultaneous chromatographic method for determination of
determination of chlordiazepoxide and its n- diazepam and its major metabolite, n-
desmethyl metabolite in 50-microL blood desmethyldiazepam, in human serum. J
samples by high pressure liquid Chromatogr 1978;148:549-52.
chromatography. Anal Chem 1977;49:2235-6. 37 Palermo SF, Poklis A. Evaluation of thin-layer
23 Kabra PM, Stevens GL, Marton LJ. High chromatographic methods for the determination
pressure liquid chromatographic analysis of of flurazepam and n-1-hydroxymethyl
diazepam, oxazepam, and n-desmethyldiazepam flurazepam. Can Soc Forensic Sci J 1977;10:77-
in human blood. J Chromatogr 1978;150:355- 82.
60,. 38 Brooks MA, DArconte L, Hackman MR, De
24 Brodie RR, Chasseaud LF, Taylor T. High- Silva J.A. Toxicologic analysis of 1,4-
performance liquid chromat-ographic benzodiazepines in differential pulse
determination of benzodiazepines in human polarography, J. Anal. Toxicol. 1977;1:179-83.
plasma. J Chromatogr 1978;150:361-6. 39 Drummer OH. Methods for measurement of
25 Ascalone V. Determination of chlordiazepoxide benzodiazepines in biological samples. J
and its metabolites in human plasma by Chromatogr B Biomed Sci App 1998;713:201-
reversed-phase high-performance liquid 25.
chromato-graphy.J Chromatogr 1980;181:141- 40 Haden BH, McNeil KG, Huber NA. An EMIT
6. assay for benzodiazepines in urine. Clin Chem
26 Tjaden VR, Meeles MT, Thys CP, Van der 1976;22:1200.
Kaay M. Determination of some 41 Poklis A. An evaluation of EMIT-d.a.u.
benzodiazepines and metabolites in serum, benzodiazepine metabolite assay for urine drug
urine and saliva by high-performance liquid screening J Anal Toxicol 1981;5:174-6.
chromato-graphy. J Chromatogr 1980;181:227- 42 Slightom EL. The analysis of drugs in blood,
41. bile and tissue with an indirect homogeneous
27 Divoll M, Greenblatt DJ, Shader RI. Liquid enzyme immunoassay. J Forensic Sci
chromatographic determine-ation of 1978;23:292-303.
chlordiazepoxide and metabolites in plasma. 43 Divoll M, Greenblatt DJ, Lacasse Y, Shader RI.
Pharmacology 1982;24:261-6. Benzodiazepine overdosage: plasma
28 Sutheimer C, Sunshine I. Benzodiazepines by concentrations and clinical outcome.
reverse-phase HPLC. In Sunshine I, Jatlow PI. Psychopharmacology 1981;73:381-3.
editors: Methodology for analytical toxicology. 44 Greenblatt DJ, Shader RI. Dependence,
Boca Raton, FL: CRC Press; 1982; p 25-30. tolerance and addiction to benzodiazepines:
(vol. 2) clinical and pharmacokinetic considerations.
29 Shaw W, Long C, McHan J. An HPLC method Drug Metab. Rev 1978;8:12-8
for analysis of clonazepam in serum, J Anal 45 Greenblatt DJ, Allen MD, Noel BJ, Shader RI.
Toxicol 1983;7:119-22,. Acute overdosage with benzodiazepine
30 Wong AS. An evaluation of HPLC for the derivatives. Clin Pharmacol Ther 1977;21:497-
screening and quantitation of benzodiazepines 514.
and acetaminophen in post mortem blood. J 46 Finkle BS, McCloskey KL, Goodman L.S.
Anal Toxicol 1983;7:33-9. Diazepam and drug-associated deaths. JAMA
31 Walberg CB. Diazepam (Valium). In Sunshine 1979;242:429-34
I, Jatlow PI. editors: Methodology for analytical
toxicology. Boca Raton, FL: CRC Press; 1976.
p. 76-9, p. 119-21.
32 Valentour JC, Montforte JR, Lorenzo B,
Sunshine I. Fluorometric screening method for
detecting benzodiazepines in blood and urine.
Clin Chem 1975;21:1976-9.
239
Benzodiazepines
Tables
Benzodiazepine drugs
Merck Molecular Molecular Structure
Drug Index formula weight R1 R2 R3 R4
Bromazepam 1357 C14H10BrN3O 316.2 H H 1
2 (N)* Br
General Structure of 1,4-Benzodiazepine
Clonazepam 2352 C15H10C1N3O3 315.2 H H C1 NO2
Diazepam 2967 C16H13ClN2O 284.7 CH3 H H C1
Flurazepam 4100 C21H23ClFN3O 387.9 C2H4N(C2H5)2 H F C1
Halzepam 4472 C17H15ClN2O16 298.8 C2H5 H H C1
Lorazepam 5400 C15H10ClN2O2 312.2 H OH C1 C1
Oxazepam 6799 C15H11ClN2O2 286.7 H OH H C1
Temazepam 8976 C16H13ClN2O2 300.7 CH3 OH H C1
Chlordiazaepoxide
2049 C16H14ClN3O3 299.8 click here
Clobazam 2325 C16H13ClN2O2 300.7 click here
Medazepam 5609 C16H15ClN2 270.8 click here
Alprazolam 303 C17H13ClN4 308.8 click here
Triazolam 9418 C17H12Cl2N4 343.2 click here
Chemical structures go here for: 1,4-Benzodiazepine; Clobazam; Alprazolam; Chlordiazepoxide; Medazepam; and
Triazolam
*C-5 pyridyl group.
240
Benzodiazepines
Methods of benzodiazepine analysis

Method 1: Gas-liquid chromatography (GLC); chromatographic separation, quantitative
a. Derivative formation, electron-capture detector
b. Direct GLC flame ionization detector
c. Direct GLC electron-capture detector
a. Extraction of alkalinized blood, formation of benzophenone by acid hydrolysis, and analysis by GLC
with electron-capture detector
b. Extraction of alkalinized sample, evaporation, reconstitution, and GLC analysis
c. Extraction of alkalinized sample, evaporation, reconstitution, and GLC analysis on a 10% OV-1
column with a 63Ni electron-capture detector
Comments:
a. Single drug determination; not specific, but can be corroborative if linked to other methods
b. Screens and quantification in toxicological cases; limited sensitivity, 0.2 g/mL
c. Method of choice; most flexible for screening and quantification; sensitivity of 0.01 to 0.5 g/mL
Method 2: High-performance liquid chromatography; chromatographic separation, quantitative
Principle of analysis: Extraction of alkalinized sample, evaporation, reconstitution, separation, usually on a
reversed-phase column, and detection at 254 nm
Comments: In common use for general quantification; particularly useful in chlordiazepoxide analysis;
sensitivity about 0.01 g/mL
Method 3: Spectrophotometric; ultraviolet, fluorescent, qualitative
Principle of analysis: Extraction into solvent, separate appropriate fractions, measure ultraviolet absorbance
Comments: Single drug determinations; both have limited specificity; ultraviolet methods not sensitive
Method 4: Thin-layer chromatography; chromatographic separation, qualitative
Principle of analysis: Extracted drugs separated on silica gel and detected fluorometrically
Comments: Drug screens; sensitivity of 0.2 g/mL; not useful for stat work
Method 5: Polarographic; differential pulse polarography (DPP), quantitative
Principle of analysis: Extraction of alkalinized blood and DPP analysis for identification
Comments: Drug screens; sensitivity of 0.25 to 0.50 g/mL
Method 6: Immunoassay; EMIT (enzyme-multiplied immunoassay technique), qualitative
Principle of analysis: Competitive binding assay; drug attached to enzyme glucose-6-phosphate dehydrogenase
and unknown compete for antibody; cross-reactivity for antibody allows for reaction with many benzodiazepines
Comments: Drug screens, normally using urine; variable sensitivity to 0.5 g/mL equivalent oxazepam
241
Benzodiazepines
Summary of GLC and HPLC methods for benzodiazepines

Gas-liquid chromatographySource: De Silva et al., flurazepam, lorazepam, nitrazepam,
1964 [1] oxazepam, and metabolites
Column: 2% Carbowax 20M Sensitivity: 1 to 10 ng/mLSource: De Groot
Temperature (F): 190 et al., 1976 [16]
Column: 3% OV-1
Detector: 3H, electron-capture detector
Temperature (F): 240
(ECD)
Internal standard: Detector: 3H, ECD
Compounds measured: MACB (diazepam); Internal standard: Griseofulvin
ACB (desmethyldiazepam) Compounds measured: Lorazepam
Sensitivity: 20 ng/mLSource: De Silva et al., Sensitivity: 10 ng/mL
1966 [2] Source: Rutherford, 1977 [7]
Column: 2% Carbowax 20M terephthalic Column: 3% OV-7
acid Temperature (F): 250
Temperature (F): 215 Detector: 63Ni, ECD
Detector: 3H, ECD Internal standard: Prazepam
Internal standard: Compounds measured: Diazepam,
Compounds measured: MACB (diazepam); desmethyldiazepam
ACB (desmethyldiazepam) Sensitivity: 20 ng/mLSource: Greenblatt et
Sensitivity: 20 ng/mLSource: Foster and al., 1978 [17]
Fringes, 1970 [3] Column: 3% OV-17
Column: 3% SE-30 Temperature (F): 280
Temperature (F): 205 Detector: 63Ni, ECD
Detector: Flame ionization detector Internal standard: Oxazepam
Internal standard: Compounds measured: Lorazepam
Compounds measured: Diazepam Sensitivity: 1 to 3 ng/mLSource: Kelly et al.,
Sensitivity: 330 ng/mL 1979 [12]
Source: De Silva and Puglisi, 1970 [4] Column: 3% OV-17
Column: 3% OV-17 Temperature (F): 265
Detector: 63Ni, ECD
Detector: 63Ni, ECD Internal standard: Medazepam
Internal standard: Compounds measured: Chlorazepate,
Compounds measured: Diazepam, chlordiazepoxide, diazepam, flurazepam,
medazepam, and metabolites and metabolites
Sensitivity: 4 ng/mL Sensitivity: 5 ng/mL
Source: Zingales, 1971 [5] Source: Peat and Kopjak, 1979 [9]
Column: 2% OV-17 Column: 3% OV-17
Temperature (F): 275 Temperature (F): 240
Detector: ECD
Detector: 63Ni, ECD
Internal standard:
Internal standard: Flunitrazepam
Compounds measured: Chlordiazepoxide
Compounds measured: Diazepam,
and metabolites
flurazepam, and metabolites
Sensitivity: 15 ng/mLSource: De Silva et al.,
Sensitivity: 100 ng/mLSource: Ferrara et al.,
1974 [14]
1979 [15]
Column: 3% QF-1
Column: 1% SP-1000
Detector: 63Ni, ECD Detector: Nitrogen
Internal standard: Diazepam Internal standard: Diazepam
Compounds measured: Flurazepam and Compounds measured: Flurazepam and
metabolites metabolites
Sensitivity: 5 ng/mLSource: De Silva et al., Sensitivity: 10 ng/mL
1976 [6] Source: Wallace et al., 1979 [11]
Column: 3% OV-17 Column: 3% SP-2250-DB
Temperature (F): Various isothermal Temperature (F): 235
Detector: 63Ni, ECD Detector: 63Ni, ECD
Internal standard: Internal standard: Prazepam
Compounds measured: Bromazepam, Compounds measured: Diazepam,
clonazepam, chlordiazepoxide, diazepam,
242
Benzodiazepines
desmethyldiazepam desmethyldiazepam
Sensitivity: 10 ng/mL Sensitivity: 10 ng/mL
Source: Peel and Perrigo, 1980 [13] Source: Ascalone, 1980 [25]
Column:10% OV-1 or 3% OV-17 Column: LiChrosorb RP-18
Temperature (F): 230 Solvent: Acetonitrile, 0.1% ammonium
Detector: 63Ni, ECD carbonate (31:69)
Internal standard: Prazepam or chloroquin Detector: 260 nm (variable)
Compounds measured: Chlordiazepoxide, Internal standard: Nitrazepam
clonazepam, diazepam, flurazepam, lorazepam, Compounds measured: Chlordiazepoxide
nitrazepam, oxazepam, prazepam, 5 Sensitivity: 30 ng/mL
metabolites, 9 benzophenones Source: Tjaden et al., 1980 [26]
Sensitivity: 10 to 100 ng/mL Column: Methyl silica
Source: Greenblatt et al., 1981 [18] Solvent: 50% methanol
Column: 1% OV-17 Detector: 254 nm (fixed)
Temperature (F): 290 Internal standard:
Detector: ECD Compounds measured: Bromazepam,
Internal standard: U-31485 clonazepam, chlordiazepoxide, diazepam,
(triazalobenzodiazepine) flurazepam, lorazepam, nitrazepam, oxazepam,
Compounds measured: Triazolam, alprazolam prazepam, and metabolites
Sensitivity: 0.25 ng/mL Sensitivity: 1 ng/mL
Source: Jochemsen and Breimer, 1981 [19] Source: Divoll et al., 1982 [27]
Column: SCOT, 0.5% PPE-21 and 3% OV-17 Column: Bondapak C18
Temperature (F): 250 Solvent: 450 mL of methanol and acetonitrile
Detector: 63Ni, ECD (50:50), 550 mL of water, 1 mL of sodium
Internal standard: Clonazepam acetate
Compounds measured: Triazolam Detector: 254 nm (fixed)
Sensitivity: 0.05 ng/mL Internal standard: Chlordesmethyldiazepam
High-performance liquid chromatography Compounds measured: Chlordiazepoxide and
Source: Bugge, 1976 [21] metabolites
Column: Partisil 10 Sensitivity: 50 ng/mL
Solvent: n-Heptane, isopropanol, methanol Source: Sutheimer and Sunshine, 1982 [28]
(40:10:1) Column: Ultrasphere-ODS
Detector: 232 nm (variable) Solvent: 0.015 M, pH 3.3 phosphate buffer,
Internal standard: acetonitrile
Compounds measured: Diazepam, Detector: 254 nm (fixed)
desmethyldiazepam Internal standard: Clonazepam
Sensitivity: 25 ng/mL Compounds measured: Chlordiazepoxide,
Source: Greizerstein and Wojtowicz, 1977 [22] diazepam, and metabolites
Column: Bondapak C18 Sensitivity: 10 ng/mL
Source: Shaw et al., 1983 [29]
Solvent: 1 mM KH2PO4 (pH 8), methanol
Column: C18, reversed phase
(60:40, programmed)
Solvent: 800 mL of methanol, 1200 mL of
Detector: 254 nm (fixed)
water, 4 mL of diethylamine, 1 mL of
Internal standard: Chlorpromazine
phosphoric acid, 1.5 g of pentanesulfonic acid,
Compounds measured: Chlordiazepoxide and
pH to 6.5
metabolites
Detector: 308 nm (variable)
Sensitivity: 100 ng/mL
Internal standard: Desalkylflurazepam
Source: Kabra et al., 1978 [23]
Compounds measured: Clonazepam
Column: Partisil 10
Solvent: 0.01 M acetonitrile, sodium acetate
Source: Wong, 1983 [30]
(pH 4.6) (35:65)
Column: Bondapak C18
Detector: 240 nm (variable)
Internal standard: Prazepam Solvent: Methanol, water (65:35)
Compounds measured: Oxazepam, diazepam, Detector: 254 nm (variable)
desmethyldiazepam Internal standard: Prazepam
Sensitivity: 300 ng/mL Compounds measured: Chlordiazepoxide,
Source: Brodie et al., 1978 [24] clonazepam, diazepam, demoxepam,
Column: Bondapak C18 desalkylflurazepam,
desmethylchlordiazepoxide,
Solvent: Methanol, water (65:35)
desmethyldiazepam, lorazepam, oxazepam,
Detector: 254 nm (fixed)
nitrazepam, triazolam, and other metabolites
Internal standard: Prazepam
Compounds measured: Diazepam,
243
Benzodiazepines
Comparison of methods for benzodiazepine analysis

Parameter: Sample volume EMIT*: Tris HCl: 0.55 mol/L with
Urine: surfactant
TLC: 5 mL Serum:
EMIT*: 50 L GC:
Serum: HPLC: Mobile phase: 30%;
GC: 1 mL acetonitrile: 70%; 0.5 mol/L phosphate buffer
HPLC: 0.5 mL EMIT*: Tris HCl: 0.55 mol/L with
EMIT*: 50 L surfactant
Parameter: Fraction of sample volume Parameter: Sensitivity
Urine: Urine:
TLC: 1.0 TLC: 1 g/mL
EMIT*: 0.0016 EMIT*: 0.3 g of oxazepam/mL,
Serum: varies from 2 to 3 g/mL for other
GC: 1.0 benzodiazepines
HPLC: 1.0 Serum:
EMIT*: 0.0061 GC: 50 ng/mL for diazepam and
Parameter: Temperature clonazepam
Urine: HPLC: 0.5 g/mL
TLC: Ambient EMIT*: 0.3 g of diazepam/mL,
EMIT*: 30 C varies from 2 to 5 g/mL for other
Serum: benzodiazepines and metabolites
GC: 230 C Parameter: Linearity
HPLC: 60 C Urine:
EMIT*: 30 C TLC:
Parameter: pH EMIT*: 0.3 to 10 g/mL for
Urine: oxazepam varies for other benzodiazepines
TLC: Serum:
EMIT*: 8.0 GC: 100 to 1000 ng/mL
Serum: HPLC: 0.5 to 10 g/mL
GC:
EMIT*: 0.3 to 2.0 g of
HPLC: 4.4
diazepam/mL; varies for other benzodiazepines
EMIT*: 8.0
*Enzyme-multiplied immunoassay technique
Parameter: Final concentration of reagents
Method presented in text.
Urine:
TLC:
Retention data on 3% OV-17

Compound Reaction index Retention time (min)
Oxazepam 2805 3.40
Lorazepam 2925 4.26
Diazepam 2940 4.57
Desalkylflurazepam 3030 5.72
Desalkyldiazepam 3085 6.47
Chlordiazepoxide I 3125 6.85
Prazepam 3160 7.46
Flurazepam 3245 9.16
Hydroxyethylflurazepam 3255 10.08
Nitrazepam 3490 15.75
Clonazepam 3540 19.24
Chlordiazepoxide II 3575 19.31
Desmethylchlordiazepoxide 3770
Retention data on 10% OV-1

244
Benzodiazepines
Compound Retention index Retention time (min)

Medazepam 2215*
Halazepam 2270*
Oxazepam 2305 3.65
Lorazepam 2380 4.45
Diazepam 2405 4.73
Desalkylflurazepam 2430 4.96
Desmethyldiazepam 2490 5.62
Chlordiazepoxide I 2500 5.70
Clobazam 2535*
Temazepam 2575*
Bromazepam 2575*
Prazepam 2630 8.43
Hydroxyethylflurazepam 2655 8.90
Flurazepam 2775 12.08
Clonazepam 2860*
Alprazolam 2920*
Triazolam 3075*
Desmethylchlordiazepoxide 3100*
*Calculated retention index.
Pharmacological and reference interval data for commonly used benzodiazepines

Therapeutic Toxic
Daily Half- concentration concentration
Compound dose (mg) life (hours) (ng/mL) (g/mL)
Alprazolam 0.53 1220 222
Chlordiazepoxide 15100 627 5001600 >3000
Chlorazepam 810 1850 730 >70
Diazepam 240 2137 301500 >3000
Desalkylflurazepam 1530 47100 30110 >500
(flurazepam)
Lorazepam 26 1015 20240 >300
Oxazepam 3060 411 1001500 >500
Triazolam 0.1250.5 2.55 0.18
Benzodiazepines with active metabolites

Parent Metabolite
Chlorazepate Desmethylchlorazepate
Chlordiazepoxide Desmethylchlordiazepoxide
Demoxepam
Clobazam Desmethylclobazam
Diazepam Desmethyldiazepam
Hydroxydiazepam
Oxazepam
Flurazepam Desalkylflurazepam
Medazepam Desmethylmedazepam
Diazepam
Desmethyldiazepam
Prazepam Desmethyldiazepam
245
Benzodiazepines
Figures
Benzodiazepines: Figure 1
Standard (calibration) curves for some

common benzodiazepines using the GLC
(gas-liquid chromatography) method
described in text: response ratio (peak area
analyte divided by peak area of internal
standard) versus concentration of analyte.
Typical GLC chromatograms of

common benzodiazepines on either
10% OV-1 (left) or 3% OV-17
(right).
246
Benzodiazepines
A, Typical HPLC chromatogram of benzodiazepine standards: 1,demoxepam; 2, nitrazepam (internal standard); 3,

clonazepam; 4, norchlordiazepoxide; 5, nordiazepam; 6, chlordiazepoxide; 7, diazepam. B, Patient sample analyzed for
benzodiazepines by HPLC: 1, demoxepam; 2, unknown; 3, internal standard; 4, norchlordiazepoxide; 5, chlordiazepoxide.
Extraction scheme for

isolation of amphoteric,
basic, and neutral drugs for
TLC (thin-layer
chromatography) analysis.
247
Benzodiazepines
remove the n-butyl chloride layer

Procedure: Quantitative Benzodiazepine Analysis in carefully into a 5 mL evaporation vial.
Blood by Gas Chromatography e. Evaporate the extract to dryness under
Principle air or nitrogen, and take up the residue
A general gas chromatographic procedure that employs a by vortex mixing the vial with 0.1 mL
of injection solvent.
10% OV-column and 63Ni detector is used. The
2. Chromatographic parameters:
benzodiazepine and metabolites are extracted from 1 mL
of buffered blood or serum into n-butyl chloride, a. Temperature: oven 230 C; injection
concentrated, and analyzed on a gas chromatograph.[13] port 250 C; detector 300 C
b. Carrier gas: argon/methane (95:5) at
Reagents 50 mL/min
All reagents are stored in glass containers. 3. Inject 2 L of sample or standard to begin
1. Glass distilled solvents. n-Butyl chloride; analysis.
injection solventbenzene/acetone/methanol (80:15:5).
2. Internal standard solution (200 mg/L). Calculations
Twenty mg of prazepam is dissolved in 100 mL of A calibration graph is prepared for each benzodiazepine
absolute ethanol and diluted with water (1:100) just or its metabolite using triplicate analysis for each
before use. If prazepam is not available as a standard or concentration. A plot of the ratio of area of
is interfered with by another peak in the chromatogram, benzodiazepine to area of internal standard times 100
a solution of chloroquine diphosphate in water, 0.02 versus blood concentration is prepared. Linear regression
mg(base)/mL, is acceptable. When stored in glass and analysis is carried out as follows:
refrigerated, the standard is stable for about a year.
3. Buffer solution pH 10, 0.05 M. (Potassium (Area benzodiazepine/Area internal standard) 100 =
carbonatepotassium boratepotassium hydroxide.
4. Stock solution of appropriate benzodiazepine m (conc.) + b or y = mx + b
standard (250 mg/L, as base). Bromazepam,
clonazepam, chlordiazepoxide, diazepam, flurazepam, Graphical representation of standard curves for some
lorazepam, and oxazepam are dissolved in absolute common benzodiazepines is shown in Benzodiazepines:
ethanol. The stock solutions are stable for about 6 Figure 1. A carefully prepared calibration graph is
months or longer if stored in amber bottles under accurate for many months but should be replaced if a
refrigeration. The chlordiazepoxide solution is the least new column is used or if quality control tests indicate
stable and should be checked after 3 to 4 weeks for changes.
decomposition.
5. Benzodiazepine standard curve. To aliquots Notes
of blood or serum known to be free of the drug, add 10, 1. The retention data for the benzodiazepines and
20, 30, and 40 L of stock solution, and bring to 10 mL some metabolites are presented in
volume with blood or urine. If lower concentrations are Benzodiazepines Tables: Retention data on 3%
necessary for the more potent drugs (such as triazolam), OV-17 and Retention data on 10% OV-1. The
further dilution can be made. The controls should be retention time of the chlordiazepoxide
stored in a frozen condition. decomposition products are described as
chlordiazepoxide I and chlordiazepoxide II. In
Assay the conditions shown, the chlordiazepoxide
Equipment: Gas chromatograph equipped with a 63Ni peak (II-RI 2800) is predominant over its
electron-capture detector, molecular sieve trap on carrier degradation products (99:1).
gas line, injection port equipped with septum swinger, 2. The procedure is applicable to most
and electronic integrator. The column is a 10% OV-1 on benzodiazepines and their metabolites.
Chromosorb G/HP (80/100), glass (3 feet, 4 mm inner However, since chlordiazepoxide is
diameter). thermolabile and its analysis is greatly affected
1. Sample preparation: by a variable detector response (that is, not
a. Pipet 0.5 mL of buffer solution and 0.5 similarly exhibited by the internal standard), the
mL of internal standard into a 10 mL procedure is somewhat limited for this drug. If
centrifuge tube. a calibration graph for quantification is
prepared at the time the chlordiazepoxide
b. Add dropwise 1 mL of blood or serum,
sample is analyzed, the precision is similar to
and mix.
that of other benzodiazepines. One can increase
c. Add 4 mL of n-butyl chloride, and the column temperature to 260 C for ideal
shake for 5 min. quantification of chlorodiazepoxide and other
d. Centrifuge the mixture for 5 min at compounds that are eluted after prazepam in
2000 rpm (approximately 500 g), and
248
Benzodiazepines
order to reduce analysis time and improve peak from the serum by adsorption onto a Bond Elut silica gel
shape. column. The analytes are then eluted from the Bond Elut
3. The limit of sensitivity is 10 to 50 ng/mL for column with methanol and injected into the HPLC
diazepam, clonazepam, and system where separation of the benzodiazepines occurs.
desalkylflurazepam; 50 to 100 ng/mL for The concentration of each drug is then quantitated
flurazepam, lorazepam, and through the use of a proportional calculation.
desmethyldiazepam; and 100 to 150 ng/mL for
oxazepam. The limit for chlordiazepoxide is Reagents
generally sufficient for concentrations 1. Stock calibration standard. Place 10 mg of
encountered in therapy. each of the following in a volumetric flask: demoxepam,
nordiazepam, diazepam, norchlordiazepoxide,
4. If the blood sample or reagents are stored in
chlordiazepoxide. Dissolve the drugs in methanol and
plastic tubes before analysis, a phthalate ester
bring the solution to a final volume of 10 mL. Stable for
(retention index, 2500) can be extracted into the
2 months at 4 C stored in the dark.
sample and interfere with desmethyldiazepam
2. Stock internal standard. Dissolve 10 mg of
(retention index, 2490). One can overcome the
nitrazepam in methanol. Bring to a final volume of 10
problem of concomitant drug therapy and drugs
mL with methanol. Stable for 2 months at 4 C stored in
with similar retention indices by reviewing lists
the dark.
of drug-retention indices published for the
3. Plasma working standards. Obtain plasma
similar SE-30 column.
from the blood bank and filter through Whatman no. 50
5. To use the procedure for quantification of the paper. Add 50, 100, 200, 500, and 100 L of the stock
very potent drug triazolam, one should extract 3 calibration standard to 100 mL aliquots of the filtered
mL of serum or blood twice with 5 mL of n- plasma. This yields standards of 0.5, 1.0, 2.0, 5.0, and
butyl chloride and then clarify it by passing
10.0 g/mL. Aliquot the pooled plasma standards into 2
through Na2SO4 before evaporation. The GLC
mL portions and store frozen at -20 C. Stable for 6
conditions are changed to the following: initial months.
temperature 270, increase by 8/min; final 4. Working sodium carbonate buffer, 0.01
temperature 290 for 16 min. mol/L. Dissolve 1.06 g of Na2CO3 in 80 mL of distilled
Benzodiazepines Therapeutic Concentration Range water and bring to a final volume of 100 mL with
click here distilled water. Stable for 6 months at room temperature.
5. Acetonitrile. HPLC grade.
Procedure: Qualitative Benzodiazepine Analysis 6. Working phosphate buffer, 192 mol/L (pH
Amadeo J. Pesce 4.4). Dilute 300 L of 1 M potassium phosphate and 50
L of 0.9 M phosphoric acid to 1800 mL with distilled
Gas chromatographic screen. A qualitative screen for water. The pH must be measured and adjusted with
the numerous benzodiazepines and their metabolites can small amounts of phosphoric acid. Proper pH value is
be efficiently carried out by gas chromatography using a critical. Stable for 3 months at 4 to 8 C.
63Ni detector. Although the 10% OV-1 column is 7. Phosphoric acid (0.9 mol/L). Dilute 6.125 mL
suitable for quantification and general qualitative of 14.7 M phosphoric acid to 100 mL with distilled
analysis, the commonly used 3% OV-17 column is also water. Stable for 1 year at room temperature.
very useful for screening. The instrument conditions are 8. Potassium phosphate (1 mol/L). Dissolve
as follows: 183 0.2 cm inner diameter glass column 13.6 g of KH2PO4 in 80 mL of distilled water. Bring to
containing 3% OV-17 on Chromosorb W-HP, 80 to 100 a final volume of 100 mL with distilled water. Stable for
mesh, at 260 C. The retention data for some of the 1 year at room temperature.
benzodiazepines are included in Benzodiazepines
Tables: Retention data on 3% OV-17 and Retention data Assay
on 10% OV-1. Following the extraction procedure Equipment: The HPLC equipment includes a solvent
previously described and using either of these two delivery system, absorbance detector, and a recorder.
columns, one can attain a sensitive screen (10 to 100 Other equipment include a water bath (60 C), Bond Elut
ng/mL). A typical gas chromatogram for each column is extraction columns, and a Vac Elut processing station.
shown in Benzodiazepines: Figure 2.
1. Insert the needed number of Bond Elut columns
Benzodiazepines Reference Interval into the Vac Elut holder, and after turning the
click here vacuum on, prime each column twice with
methanol and then twice with distilled water.
Procedure: Serum Benzodiazepines by HPLC Turn the vacuum off.
Principle 2. Into separate 12 75 mm disposable test tube
After the addition of buffer and internal standard add the following:
(nitrazepam) to the sample, the benzodiazepines a. 500 L of sample or control.
(diazepam, nordiazepam, chlordiazepoxide, b. 500 L of working internal standard.
norchlordiazepoxide, and demoxepam) are separated
249
Benzodiazepines
c. 500 L of sodium carbonate buffer. they are present in high therapeutic or toxic
Vortex mix for 10 sec and pipet onto the Bond concentrations in blood, urine, or stomach aspirate. The
Elut column. limiting factor for benzodiazepine detection is the
3. Turn vacuum on, and draw sample through the sensitivity of the conventional TLC sprays used during
column. Then rinse the column twice with the drug screen. The following procedure has been found
distilled water. useful for screening basic and neutral drugs, including
4. Remove lid of the Vac Elut processing chamber benzodiazepines. The schematic procedure is shown in
and wipe off tips inside. Label the appropriate Benzodiazepines: Figure 4.
number of 10 75 mm test tubes, and place in Reagents
metal rack. Put rack inside, and replace lid. 1. Concentrated ammonium hydroxide (15.1
5. Add 100 L of methanol to each column, wait mol/L).
20 sec, and turn vacuum on. Repeat one more 2. n-Butyl chloride (reagent grade).
time. 3. 0.2 N H2SO4 (0.1 mol/L). Slowly add 5.5 mL
6. Take eluate and inject into HPLC. of concentrated sulfuric acid to approximately 800 mL
7. The HPLC parameters are as follows: of distilled water in 1 L volumetric flask. Mix, and allow
a. Column: Bondapak C18 to cool to room temperature. Bring to mark with distilled
water, and mix well. Stable at room temperature for 1
b. Mobile phase: year.
(1) 300 mL of acetonitrile 4. Sodium carbonate (solid). Reagent grade.
(2) 700 mL of phosphate buffer 5. Chloroform and isopropanol. Reagent grade.
Filter each using an organic filter, mix, and 6. TLC sprays.
degas the solution for 15 min.
c. Pump: Assay
Equipment: Silica TLC plates, spraying equipment, TLC
(1) Flow rate = 1.0 mL/min chromatographic tanks, ultraviolet lamp, and oven (80
(2) Pressure limit = 4000 psi C).
d. Detector: 1. Mix blood, urine, filtered stomach aspirate, or
(1) Model 441 absorbance detector set at tissue homogenate (5 mL) with ammonium
254 nm hydroxide (0.1 mL) in a screw-capped test tube
(2) Sensitivity = 0.02 AUFS (absorbance and extract twice with n-butyl chloride (5 mL
units at full scale) portions) for 3 min. Separate the n-butyl
e. Recorder: chart speed = 0.5 inch/min chloride extracts from the aqueous layer using a
8. Elution of all the peaks requires approximately disposable pipet, combine, and centrifuge.
25 min. An example of chromatograms for 2. Bring the aqueous layer to pH 8.5 with excess
standards and for a patient with an overdose of sodium carbonate and extract 5 min with 60 L
benzodiazepines is seen in Benzodiazepines: of a 1:1 mixture of chloroform/isopropanol.
Figure 3. (This step is optional if hydrolyzed urine is to
be screened for morphine or other amphoteric
Calculations drugs.)
1. Determine all relative retention times of 3. Extract the combined organic extracts for 3 min
standards, controls, and patients. with 0.2 N sulfuric acid (5 mL) in a screw-
2. Calculate the peak height of each capped test tube. Remove the n-butyl chloride
benzodiazepine (benzo) and for the internal layer (containing neutral drugs) using a
standard (IS), nitrazepam. Determine the peak disposable pipet, and evaporate to dryness
height ratio: under a stream of nitrogen.
Peak height ratio (PHR) = Peak height of benzo
4. Bubble air or nitrogen through the sulfuric acid
Peak height of 1S extract for about 1 min to remove the residual
n-butyl chloride. This acid extract may be
Concentration of unknown = screened by ultraviolet spectrophotometry
before you proceed.
Peak height ratio of unknown benzo Concentr of benzo in standard
PHR of that benzo standard 5. Add 0.8 mL of ammonium hydroxide to 4.5 mL
of the sulfuric acid extract in a 5 mL centrifuge
tube. Shake the basic solution for 2 min and
Benzodiazepines Therapeutic Concentration Range centrifuge. The chloroform extract contains
basic drugs.
Procedure: TLC Drug Screen Including
6. Screen the three extracts containing basic drugs,
Benzodiazepines
neutral drugs, and amphoteric drugs by thin-
Principle layer chromatography. The TLC system is a
Some of the benzodiazepines can be detected by a common system for use in broad drug screens.
general drug screen using thin-layer chromatography if
250
Benzodiazepines
TLC system: The practice of sequential spraying and viewing is very

Plates: Silica gel G or silica gel GF, 250 m useful in differentiating compounds with similar
Developing solvent: retention values that may belong to different classes of
Ethylacetate/methanol/ammonia (17:2:1). A drugs. The Rf values of a series of benzodiazepines are
small beaker of fresh ammonia is placed in the tank. listed in Benzodiazepines Table: Relative migration
Visualization sequence: values in TLC system tested in text.
1. Observe under ultraviolet lamp.
2. Spray plate with 50 g of Flu-ram Benzodiazepines Therapeutic Concentration Range
(fluorescamine) per liter of acetone; observe under click here
ultraviolet lamp.
3. Spray plate with 0.1% ninhydrin in
acetone; heat at 80 C for 5 min and observe.
4. Observe under ultraviolet lamp.
5. Spray with potassium iodoplatinate.
251
Beta-hCG (Beta-human Chorionic Gonadotropin)

James J. Miller
Name: Human chorionic gonadotropin, hCG, HCG, -hCG

Clinical significance: Refer to Chapter 44, Pregnancy, in the 5th edition of Clinical Chemistry: Theory,
Analysis, Correlation.
Chemical class: glycoprotein, hormone
Principles of Analysis and Current Usage i antibodies directed toward this amino acid region [4,5].
Current two-site immunometric sandwich assays using
Human chorionic gonadotropin (hCG) is a glycoprotein monoclonal antibodies have decreased the cross-reactivity
composed of two noncovalently linked polypeptides: the to about 0.1% [6]. Besides the heterogeneity described
alpha and beta subunits. The individual subunits lack above, hCG and subunits may dissociate and the
biological activity but become active when associated to subunit may be cleaved (nicked) in various positions,
form the intact complex. For clarity and for historical including removal of the CTP [4]. A hyperglycosylated
reasons, we will use hCG as the abbreviation for this form of hCG has been described [7]. In addition, during
hormone rather than HCG or -hCG. The designation the lifetime of hCG molecules in the circulation [4], they
-hCG is used to represent specifically the beta subunit of lose N-acetylneuraminic acid and possibly other sugars
hCG, not the entire hormone. The term human, though from the carbohydrate side chains [8]. Thus although the
usually superfluous in a clinical setting, is in common use major form of hCG in early pregnancy is the intact ,
with chorionic gonadotropin and with growth hormone and dimer, a variety of forms are present, and the assortment of
is indicated with a lowercase h. All other hormone these forms changes throughout pregnancy.
references are in uppercase letters and do not include a
species designation. Since it is the beta chain that specifies the biological
activity of hCG, many current assays for hCG are designed
The polypeptide chains contain various amounts of to specifically detect this portion of the hCG molecule.
carbohydrate moieties, including D-galactose, D-mannose, Thus many commercial assays are known as -hCG assays
and N-acetylneuraminic acid, the last of which is essential even though they may be specific for the intact hCG
for the biological activity of the hormone [1]. The -hCG molecule, the -hCG subunit, or both. Most hCG assays
subunit is essentially identical to the alpha chain of several detect the various altered forms of hCG to some extent.
other pituitary polypeptide hormones, such as thyroid- Because of differences in antibody specificity, it is
stimulating hormone (TSH), follicle-stimulating hormone possible to obtain different results from the same samples
(FSH), and luteinizing hormone (LH). It is the individual with different hCG kits, even if they are standardized with
beta subunit that gives each of these hormones their the same standard material.
specific biological characteristics.
The hormone hCG is synthesized by the placenta, and it
There is also a great degree of similarity between the beta appears in urine and serum relatively soon after
subunit of hCG and the beta subunit of LH. In fact, LH and implantation of the developing embryo; the presence of
hCG bind to the same receptor, the LHCG receptor [2]. In this hormone serves as the basis for pregnancy testing. In
the past, using polyclonal antibodies in competitive an acute care setting, knowledge of pregnancy is used in
immunoassays, this similarity of structure led to cross- deciding whether to submit or exclude a patient for
reactivity between hCG and LH. The carboxyterminal 24 abdominal roentgenograms, drug treatment, or surgery and
amino acid peptide (CTP) of the -hCG chain is absent in to rule out an ectopic or molar pregnancy. Qualitative
LH [3], and many current -hCG assays use monoclonal pregnancy tests have thus been developed for use in acute
care situations to provide this information accurately
within 5 to 15 minutes of sample receipt.
i Beta-hCG (Beta-human Chorionic Gonadotrophin)
In the urine, a significant portion of hCG is present as a
Previous and current authors of this method: metabolic fragment commonly referred to as the -core
First edition: Lawrence A. Kaplan fragment (hCGcf) [9,10]. The hCGcf consists of two
Methods edition: David C. Hohnadel, Lawrence A. Kaplan polypeptide chains derived from the subunit of hCG,
Second edition: David C. Hohnadel, Lawrence A. Kaplan which are joined by disulfide bridges. This fragment still
Third edition: Steven C. Kazmerczak retains the unique immunological determinants found in
Fourth edition: Steven C. Kazmierczak both intact hCG and the subunit. Urine of pregnant
Fifth edition: James J. Miller women contains large quantities of hCGcf, while it is
252
essentially undetectable in serum. Some urine pregnancy the hCG concentration can be obtained (up to 1000
tests detect hCGcf and may cross-react with the mIU/mL) by measuring the time (in seconds) from sample
corresponding metabolite of LH, the LH -core fragment or reagent addition to the appearance of a positive test
(LHcf) [11]. This is rarely a problem, however, since the result [15].
urinary concentration of hCGcf would typically be much
greater than the urinary concentration of LHcf in early Quantitative Assays
pregnancy. Quantitative hCG assays are usually used only for serum
(or plasma) samples. In the past, radioimmunoassays
Quantitative hCG measurements have many other (RIAs) had been used for measurement of hCG levels
applications besides detection of routine pregnancy and (Table 1, Method 1). RIAs were typical competitive
estimation of gestational age [12]. In threatened or binding assays in which the hCG in the sample competed
spontaneous abortions, lower than expected or declining with hCG labeled with radioactive iodine for binding sites
hCG concentrations may be found, whereas hCG is often on an hCG antibody. Both solid-phase and double-
elevated in non-uterine pregnancies (ectopic, molar), in antibody procedures were used. Some quantitative RIA
trophoblastic growths (choriocarcinomas), and in the procedures were modified to shorten the reaction time and
presence of certain cancers, such as seminomas, teratomas, allow the procedure to be used as a qualitative, stat
hepatoblastomas, and bronchiogenic carcinomas. In these pregnancy test. However, with the appearance of good
patients with ectopic hCG-secreting tumors or qualitative and quantitative immunometric sandwich
trophoblastic diseases, serum hCG concentrations are assays, the use of RIA for hCG testing has virtually
followed to assess the success of surgical or drug therapy. disappeared.
Measurements of hCG are also useful in monitoring
hormone-treated or subfertile patients when ovulation is Virtually all current quantitative hCG assays utilize the
induced. two-site sandwich immunometric assay technique (Table
1, Method 2). The capture antibody (monoclonal or
The need for hCG analysis has produced a wide variety of polyclonal) may be bound to a microparticle of glass,
methods. These procedures differ in their sensitivity and plastic, or magnetic material. This reacts with the hCG in
specificity, speed and ease of analysis, and cost. The most the sample. Then a labeled detection antibody (monoclonal
important considerations for use of any stat assay for hCG or polyclonal) binds to the hCG to form a sandwich.
are the sensitivity of the assay and the time necessary to After the reaction is complete, the unbound labeled anti-
report a result. hCG molecules are washed away. The amount of label
captured in the sandwich is measured and is directly
Qualitative Assays proportional to the hCG concentration. This is illustrated in
Qualitative tests for serum and urine hCG are a widely Figure 1-A. The label may be an enzyme, such as
divergent group of assays with many different physical horseradish peroxidase or alkaline phosphatase, or (more
forms. Most are variants of the sandwich immunometric commonly today) a chemiluminescent molecule. These
assay (Table 1, Method 2). A capture antibody to intact assays are sensitive to approximately 0.5 mIU/mL.
hCG or the -hCG subunit is bound to a solid surface. The
surface can be a membrane, a plastic bead, a plastic Reference and Preferred Methods
dipstick or paddle, or a coated tube. After incubation with At this time there, is no reference method for hCG.
the sample (serum or urine), any hCG in the sample will be Theoretically, a reference method should measure not only
bound to the surface antibody. A second enzyme- or color- intact hCG but also all of its partially degraded forms on
tagged detection antibody to intact hCG or the -hCG an equimolar basis. This is unlikely to be possible with an
subunit is added. For enzyme-tagged detection, a substrate immunoassay but could be possible with liquid
for the enzyme reaction is added, and development of a chromatographicmass spectrometric techniques.
color is determined visually. Color-tagged detection is
directly visible. Only samples with hCG present can form The preferred method depends on whether a qualitative or
the sandwich (first antibodyhCG enzyme linked or quantitative assay is needed and whether the assay needs to
colored second antibody) and ultimately any color. The be used for urine, serum, or both. In addition, the method
antibodies may be monoclonal or polyclonal. of choice may depend on a balance between the need for a
rapid result and the need for an accurate result. A false
In most cases, these assays have been designed to detect - negative can result in delayed recognition of an ectopic or
hCG-intact or free subunits and may also detect the molar pregnancy or a fetus being subjected to x-irradiation
hCGcf. They react within a few minutes at sensitivities of or threatened by a surgical procedure. On the other hand, a
10 to 25 mIU/mL [13,14]. This is an area of testing false-positive pregnancy test in cases of acute lower
currently undergoing constant improvement and change. abdominal pain in women may cause helpful diagnostic
Although these assays are qualitative in nature, it has been procedures and pharmacological therapies not to be used.
shown that with serum as specimen, an approximation of
Based on 2008 College of American Pathologists (CAP)
253
proficiency testing survey participant summaries, over 30 Any random urine sample is adequate, but first morning
professional (not over-the-counter), qualitative urine hCG voids have slightly higher hCG concentrations. Samples
devices are used by laboratories in the United States. Many containing particulate matter should be centrifuged before
of these are also approved for qualitative serum hCG analysis. Urine samples can be stored for up to 8 hours at
analysis. Most of these have analytical sensitivities room temperature or refrigerated for up to 72 hours. Urine
between 20 and 25 mIU/mL. For pregnancy tests in samples should not be frozen.
physician-office laboratories, emergency rooms, or other
clinical areas such as radiology, any of these kits should be Serum/Plasma
acceptable. For hospital laboratories serving emergency For most methods, serum or plasma may be used, although
rooms, it may be preferable to use one of the devices that some methods may be affected by chelating anticoagulants
can also be used with serum samples. such as EDTA. Remove the serum from the clot as soon as
possible. Use of serum or plasma separator tubes is
Standardization acceptable. Most methods recommend analysis as soon as
The first and second World Health Organization (WHO) possible, probably because of gradual nicking of hCG, but
International Standards (1st IS and 2nd IS) were prepared some methods allow samples to be at room temperature for
in 1938 and 1964, respectively. Both were crude hCG up to 8 hours. If analysis cannot be performed within a
preparations. They were replaced by the 1st International short time, refrigeration for 48 to 72 hours is acceptable, or
Reference Preparation (1st IRP; hCG preparation 75/735) the sample can be frozen.
in 1978. The 1st IRP was prepared from highly purified
hCG and in 1986 was renamed the 3rd International Interferences
Standard (3rd IS; hCG preparation 75/537). In 1999, a The most common interference in hCG assays is due to
second batch of the 1st IRP was prepared and released as interfering antibodies. The reagent excess conditions of
the 4th IS (hCG preparation 75/589). All of these hCG two-site immunometric assays, which makes them rapid,
standards were purified from crude extracts of urine from also introduces interference by antibodies present in some
pregnant women [16,17] and contain approximately 9% patients serum or plasma. These interfering antibodies are
nicked hCG [16]. All current commercial hCG assays use of various types, including human anti-mouse antibodies
calibrators traceable to either the 3rd or 4th IS. (HAMA), other anti-animal antibodies, heterophile
antibodies, and rheumatoid factor [20-23]. When present,
On a recent (2008) CAP proficiency testing challenge, such interfering antibodies may bind together the capture
there was a difference of 40% to 50% in hCG measured on antibody and detection antibody in the absence of the
the same sample between the methods detecting the lowest analyte. This is illustrated in Figure 1-B. All
and highest concentrations (e.g., 724 versus 1092 immunometric assays include blocking agents such as
mIU/mL). Even after removing the lowest and highest preimmune serum, gamma globulin, or other proprietary
methods, there was 25% to 30% difference between the products to minimize the effects of interfering antibodies.
second lowest and the second highest methods (e.g., 729 Rarely, patients with high concentrations of interfering
versus 915 mIU/mL). Clearly, additional efforts to antibodies may have false-positive or falsely increased
standardize hCG methods are needed. Toward that end, a results in immunometric assays, including hCG assays.
new preparation, the 1st Reference Reagent (RR; hCG This type of interference, though rare, has caused
preparation 99/688), was recently prepared from a fresh unnecessary chemotherapy and surgery [24-26].
urine extract and is in the process of being released. This
new preparation was further purified to remove nicked Because of the similarity in structure, LH could cross-react
hCG [18]. It remains to be seen whether this new standard in some hCG assays. However, because these are two-site
will improve the agreement between assays. Even when a immunometric assays, cross-reactivity is quite low (about
pure hCG-IS is available, it may not be possible to 0.1 %), and LH would have to be 10,000 mIU/mL to give
improve the harmonization between assays. The various an apparent hCG of 10 mIU/mL. LH could only be that
methods do not actually use the IS for calibrating the assay elevated in a patient with an LH-secreting tumor. The
but have their working calibrator assayed against the IS highest reported LH in a patient with an LH-secreting
that is, the working calibrator nominal concentration is tumor is 707 mIU/mL [27]. Thus it is unlikely that LH
traceable to the IS. The working calibrator may not be and could cause a false elevation of hCG due to cross-
is probably not as pure as the IS. Given this fact, the reactivity.
variability of hCG forms in different samples, and
differences in specificity of the antibodies used in different Depending on the assay design, some immunometric
assays, it may not be possible for the various assays to assays are prone to interference due to antigen excess,
agree [19]. whereby high concentrations of the analyte may give
falsely low results. This type of interference is called the
Specimen hook effect. In some immunometric assays, the capture
Urine antibody, sample, and detection antibody are added
Urine samples should be collected into a clean container. simultaneously. These assays are called one-step assays. In
254
other assays, only the capture antibody and sample are presence of hCG-producing tissue. A negative result rules
incubated first, followed by a wash step. Then the out all cases of pregnancy except those seen immediately
detection antibody is added, followed by a second (1 to 2 days) after conception. If the result is positive and
incubation and wash. These assays are called two-step abdominal pain or vaginal bleeding is present, there is the
assays. One-step assays are more rapid but more prone to possibility of ectopic (non-uterine) pregnancy.
the hook effect. The possibility of a hook effect is
eliminated in two-step assays. In practice, most hCG In ectopic pregnancies, hCG values are lower than in
assays use the more rapid one-step approach and avoid the uterine pregnancies, but ectopic pregnancies are difficult to
hook effect by setting an appropriate upper limit of the distinguish on the basis of hCG values alone because of
analytical range. the large overlap of hCG values with hCG values seen in
normal pregnancies. The rate of increase of hCG is more
In general, hemolysis, icterus and lipemia (turbidity) do helpful. The rise in hCG in 2 days varies somewhat from
not interfere with hCG assays, but product labeling should 53% to 228%. A rise slower than 53% in 2 days is
be checked, since these may interfere with some detection associated with a higher risk of ectopic pregnancy or
systems. spontaneous abortion [34].
Beta-hCG Reference Interval The diagnostic sensitivity of an hCG that is apparently low
The absolute amount of hCG present in the serum during for gestational age can be increased when it is combined
pregnancy varies greatly with gestational age and between with ultrasonography. If the hCG is greater than 6000 to
patients. Following is a table of approximate values for the 6500 mIU/mL, ultrasonography will usually demonstrate
first trimester (ranges will vary by method). an intrauterine pregnancy; if no sac is present, an ectopic
pregnancy is 94% probable [35]. For hCG values less than
Serum hCG Levels With Gestational Age 6000 mIU/mL, in nonemergency cases, sequential
Gestational Age hCG (mIU/mL) sampling over several days or weeks can be used to predict
(Weeks) when ultrasonography will prove diagnostically useful.
0.2 1 5 50
When serum hCG is used as a tumor marker, the results
12 50 500 are used to monitor or stage the tumor [33,36]. In these
23 100 5000 cases, progression or regression of the cancer will be
34 500 10,000 monitored by changes in serum hCG levels. With the
45 1,000 50,000 current analytical sensitivity of quantitative hCG methods,
56 10,000 100,000 a serum value of less than 2 mIU/mL would indicate that
68 15,000 200,000 no hCG-secreting tumor was present.
9 13 10,000 100,000
hCG Performance Goals
Interpretation In laboratory practice, hCG is an analyte regulated for
During the first 8 weeks of pregnancy, hCG doubles proficiency testing. Acceptable performance criteria
(100% increase) approximately every 2 days. require reporting laboratories to be within 3 SD of the
Concentrations begin to decline after the eighth week, mean of the peer group laboratories. On a recent CAP
reaching a plateau 15% to 20% of peak concentration proficiency-testing participant summary, the median
around the 18th week until term [28]. In twin pregnancies coefficient of variation (CV) was 5.6%. This represents a
at same stage of gestation, hCG concentrations are total allowable error of approximately 17%, Suggested
approximately twice those seen in singleton pregnancies medical decision limits are 25 mIU/mL and 10,000
during weeks 15 to 22 [29]. Healthy, nonpregnant, mIU/mL.
premenopausal females have hCG concentrations of < 5
mIU/mL [30]. Some postmenopausal females have hCG References
concentrations as high as 13 mIU/mL [30,31]. Healthy 1. Ross GT. Clinical relevance of research on the
males have low circulating concentrations of hCG of < 1 structure of human chorionic gonadotropin. Am J
mIU/mL [32]. Obstet Gynecol 1977;129:795-805.
2. Ji I, Zeng H, Ji TH. Receptor activation of and
Higher concentrations of hCG than normal may be signal generation by the
associated with twin pregnancies, molar pregnancies, and lutropin/choriogonadotropin receptor.
choriocarcinoma. Elevated values of hCG in males have Cooperation of Asp397 of the receptor and alpha
been associated with testicular tumors (seminomas, Lys91 of the hormone. J Biol Chem
teratomas) and ectopic hCG-producing tumors [33]. 1993;268:22971-4.
3. Maston GA, Ruvolo M. Chorionic gonadotropin
A qualitative serum hCG assay that becomes positive at 15 has a recent origin within primates and an
to 25 mIU/mL can be used to diagnose pregnancy or the evolutionary history of selection. Mol Biol Evol
255
2002;19:320-35. and its subunits. Bull World Health Organ

4. Cole LA. Immunoassay of human chorionic 1976;54:463-72.
gonadotropin, its free subunits, and metabolites. 18. Birken S, Berger P, Bidart J-M, Weber M,
Clin Chem 1997;43:2233-43. Bristow A, Norman R et al. Preparation and
5. Cole LA, Abbott JT, Armstrong EG. Comment on characterization of new WHO reference reagents
the nature and specificity of Bayer Corporation for human chorionic gonadotropin and
and Chiron Diagnostics hCG immunoassays. Clin metabolites. Clin Chem 2003;49:144-54.
Chem 1998;44:2001-2. 19. Ekins R. Immunoassay standardization. Scand J
6. Cole LA, Sutton JM, Higgins TN, Cembrowski Clin Lab Invest Suppl. 1991;205:33-46.
GS. Between-method variation in human 20. Kricka LJ, Schmerfeld-Pruss D, Senior M,
chorionic gonadotropin test results. Clin Chem Goodman DB, Kaladas P. Interference by human
2004;50:874-82. anti-mouse antibody in two-site immunoassays.
7. OConnor JF, Ellish N, Kakuma T, Schlatterer J, Clin Chem 1990;36:892-4.
Kovalevskaya G. Accuracy of home pregnancy 21. Ward G, McKinnon L, Badrick T, Hickman PE,
tests at the time of missed menses. Prenat Diagn Heterophilic antibodies remain a problem for the
1998;18:1232-40. immunoassay laboratory. Am J Clin Pathol
8. Sutton JM. Charge variants in serum and urine 1997;108:417-21.
hCG. Clin Chim Acta 2004;341:199-203. 22. Kricka LJ. Human anti-animal antibody
9. Kato Y, Braunstein GD. Beta-core fragment is a interferences in immunological assays. Clin
major form of immunoreactive urinary chorionic Chem 1999;45:942-56.
gonadotropin in human pregnancy. J Clin 23. Levinson SS, Miller JJ. Towards a better
Endocrinol Metab 1988;66:1197-201. understanding of heterophile (and the like)
10. Wehmann RE, Blithe DL, Flack MR, Nisula BC. antibody interference with modern
Metabolic clearance rate and urinary clearance of immunoassays. Clin Chim Acta 2002;325:1-15.
purified beta-core. J Clin Endocrinol Metab 24. Cole LA. Phantom hCG and phantom
1989;69:510-7. choriocarcinoma. Gynecol Oncol 1998;71:325-9.
11. Iles RK, Javid MK, Gunn LK, Chard T. Cross- 25. Cole LA, Rinne KM, Shahabi S, Omrani A.
reaction with luteinizing hormone beta-core is False-positive hCG assay results leading to
responsible for the age-dependent increase of unnecessary surgery and chemotherapy and
immunoreactive beta-core fragment of human needless occurrences of diabetes and coma. Clin
chorionic gonadotropin in women with Chem 1999;45:313-4.
nonmalignant conditions. Clin Chem 26. Rotmensch S, Cole LA. False diagnosis and
1999;45:532-8. needless therapy of presumed malignant disease
12. Tyrey L. Human chorionic gonadotropin assays in women with false-positive human chorionic
and their uses. Obstet Gynecol Clin North Am gonadotropin concentrations. Lancet
1988;15:457-75. 2000;355:712-5.
13. O'Connor RE, Bibro CM, Pegg PJ, Bouzoukis JK. 27. Hirshberg B, Conn PM, Uwaifo GI, Blauer KL,
The comparative sensitivity and specificity of Clark BD, Nieman LK. Ectopic luteinizing
serum and urine hCG determinations in the ED. hormone secretion and anovulation. N Engl J
Am J Emerg Med 1993;11:434-6. Med 2003;348:312-7.
14. Cole LA, Sutton-Riley JM, Khanlian SA, 28. Braunstein GD, Rasor J, Danzer H, Adler D,
Borkovskaya M, Rayburn BB, Rayburn WF. Wade ME. Serum human chorionic gonadotropin
Sensitivity of over-the-counter pregnancy tests: levels throughout normal pregnancy. Am J Obstet
comparison of utility and marketing messages. J Gynecol 1976;126:678-81.
Am Pharm Assoc 2005;45:608-15. 29. Wald N, Cuckle H, Wu TS, George L. Maternal
15. Mishalani SH, Seliktar J, Braunstein GD. Four serum unconjugated oestriol and human chorionic
rapid serum-urine combination assays of gonadotrophin levels in twin pregnancies:
choriogonadotropin (hCG) compared and implications for screening for Downs syndrome.
assessed for their utility in qualitative Br J Obstet Gynaecol 1991;98:905-8.
determination of hCG. Clin Chem 1994;40:1944- 30. Snyder JA, Haymond S, Parvin CA, Gronowski
9. AM, Grenache DG. Diagnostic considerations in
16. Birken S, Gawinowicz MA, Kardana A, Cole LA. the measurement of human chorionic
The heterogeneity of human chorionic gonadotropin in aging women. Clin Chem
gonadotropin (hCG). II. Characteristics and 2005;51:1830-5.
origins of nicks in hCG reference standards. 31. Gronowski AM, Fantz CR, Parvin CA, Sokoll LJ,
Endocrinology 1991;129:1551-8. Wiley CL, Wener MH, Grenache DG. Use of
17. Canfield RE, Ross GT. A new reference serum FSH to identify perimenopausal women
preparation of human chorionic gonadotrophin with pituitary hCG. Clin Chem 2008;54:652-6.
256
32. Lee CL, Hes R, Shephert JH, Hudson CN, Chard Tables
T. The purification and development of a
radioimmunoassay for beta-core fragment of Table 1: Methods for hCG Measurements
human chorionic gonadotrophin in urine:
applications as a marker of gynecological cancer Method 1: Radioimmunoassay (RIA)
in premenopausal women. J Endocrinol Type of analysis: Competitive
1991;130:481-9. Principle: Radiolabeled (radioactive iodine,
33. Stenman UH, Alfthan H, Hotakainen K. Human 125I) hCG competes with sample analyte for
chorionic gonadotropin in cancer. Clin Biochem binding to anti-hCG. Increased hCG in sample,
2004;37:549-61. decreased bound radioactivity.
34. Barnhart KT, Sammel MD, Rinaudo PF, Zhou L, Usage: Rarely used today
Hummel AC, Guo W. Symptomatic patients with Comments: Of historical interest
an early viable intrauterine pregnancy: hCG Method 2: Immunometric assay
curves redefined. Obstet Gynecol 2004;104:50-5. Type of analysis: Two-site sandwich assay
35. Kadar N, Caldwell BV, Romero R. A method of Principle: Labeled anti-hCG reacts with sample
screening for ectopic pregnancy and its hCG bound to solid-phase anti-hCG. Amount of
implications. Obstet Gynecol 1981;58:162-6. bound label directly proportional to amount of
36. Rule AH, DeLellis RA, Papish SW. Alpha- hCG in sample.
fetoprotein and beta-chorionic gonadotropin in Usage: Most frequently used assay
the diagnosis and management of testicular germ Comments: One or both antibodies are usually
cell tumors. J Clin Immunoassay 1983;6:213-20. monoclonal. Reagent excess conditions make this
method rapid but subject to rare false-positive
results due to heterophile antibodies in patients
samples. Sensitivity of 0.1 to 1 mIU/mL. Assay
time 15 to 20 minutes.
257
Figures
Beta-hCG: Figure 1
Immunometric Assay. A, Incubation of immobilized capture antibody, sample containing hCG, and labeled antibody results in
a complex (sandwich). Label captured is proportional to hCG concentration. B, Interfering antibody (heterophile, HAMA,
etc.) forms the sandwich in the absence of hCG, causing a false-positive or falsely elevated result.
Beta-hCG: Figure 2
Mean serum hCG levels throughout normal pregnancy. Arithmetical scale used on ordinate. Bars represent 1 standard error of
the mean [28].
258
Beta-2-Microglobulin
James J Miller
Name: Beta-2-microglobulin, 2M, B2M, thymotaxin

Clinical significance: see Interpretation,
Chemical class: Protein, light-chain, of the MHC class I antigens
i Blood:
Beta-2-microglobulin (B2M) is the light chain of the Serum is the preferred specimen. If possible, lipemic
major histocompatibility complex (MHC) class I antigen specimens should be avoided. Samples should be
and is found on the surface of all human cells expressing analyzed fresh or refrigerated for up to 3 days. Samples
this antigen. It is noncovalently bound to the MHC class may be frozen until assayed.
I heavy chain and is shed into the circulation [1,2]. B2M
has a molecular mass of 11.8 kDa, with 100 amino acids Urine:
in a single polypeptide chain with one disulfide bond. It B2M is unstable in acid urine [14]. Random urine
was first isolated from urine by Berggard and Bearn in samples should be adjusted to a pH of 7 immediately
1968 [3]. B2M was found to be identical to a T-cell after collection; 24-hour urine collections should have a
precursor chemotactic factor in the thymus called neutral buffer added before collection, and the collection
thymotaxin [4], and in 1999, it was found to be an should be refrigerated during collection. Aliquots of
apoptosis-inducing factor [5]. random or 24-hour urines should be centrifuged before
analysis.
Early assays for B2M used radial immunodiffusion [6]
(Table 1, Method 1) and immunoprecipitation methods CSF should be centrifuged and assayed immediately or
[7]. These methods lacked sufficient sensitivity for stored refrigerated for <3 days. Samples may be frozen
measurements in serum and required concentration of until assayed.
urine before measurement. Sensitive and specific
radioimmunoassays were developed in the early 1980s Interferences
[8]. In the late 1980s and 1990s, latex agglutination Lipemia may interfere with B2M measurements in some
(Table 1, Method 2) [9], turbidimetric (Table 1, Method nephelometric and turbidimetric methods [15].
3) [10], nephelometric (Table 1, Method 4) [11], and
nonisotopic immunometric [12,13] assays (Table 1, Beta-2-Microglobulin Reference Interval
Method 5) were developed.
Serum: 0.11-0.23 mg/dL (1.1-2.3 mg/L) [8]
Reference and Preferred Methods Urine: 0.04-0.36 mg/d (0.4-3.6 mg/d) [8]
There is no reference method for B2M. Clearance: 8-130 L/min [8] =
(B2MUr V 1000 L/mL)/(B2MS t),
Radial immunodiffusion and immunoprecipitation where B2MUr = urine B2M in mg/dL;
methods are inexpensive and relatively simple to B2MS = serum B2M in mg/dL;
perform, but they are labor intensive and not suitable for V = urine volume in mL;
rapid analysis. Latex agglutination methods are rapid but and t = duration of collection in min.
imprecise. Radioimmunoassay methods are sensitive, CSF: 0.06-0.22 mg/dL (0.6-2.2 mg/L) [16]
specific, and relatively inexpensive to perform, but they
are rarely used today because of the short reagent shelf Interpretation
life, the hazards of radioactivity, and the expense of Because of its small size, B2M is most used for
radioactive waste disposal. Currently, most laboratories assessing renal function, especially for differentiating
use nephelometric or turbidimetric assays because they glomerular and tubular proteinuria. High serum B2M
are faster and readily automated procedures. concentrations are associated with low glomerular
filtration rate, inflammation, and various neoplastic
Specimen conditions, especially those involving B lymphocytes
(e.g., multiple myeloma). With normal glomerular
i
Beta-2 microglobulin function, even high serum B2M concentrations do not
Previous and current authors of this method: usually lead to elevated urine B2M levels because of its
First edition: Not done catabolism in the proximal tubules. On the other hand,
Methods edition: Donald T. Forman elevated urinary B2M concentration is found in acute
Second edition: Not updated tubular necrosis and failing renal transplants [8].
Third edition: Donald T. Forman
Fourth edition: Not updated B2M has not proven to be a useful tumor marker, but it
Fifth edition: James J. Miller does have prognostic value for patients with multiple
myeloma [17]. A role for B2M as a prognostic marker in
259
human immunodeficiency virus (HIV) infection has 11 Viedma JA, Pacheco S, Albaladejo MD.
been explored [2]. More recently, B2M has been found Determination of 2-microglobulin in serum by
to be the cause of dialysis-related amyloidosis, first a microparticle-enhanced nephelometric
discovered by Assenat et al. [18] in 1980. immunoassay. Clin Chem 1992; 38: 2464-8.
12 Tienhaara A, Eskola JU, Manto V. Double
In acute leukemia and lymphoma with central nervous monoclonal time-resolved immunofluorometric
system involvement, the concentration of B2M is assay of 2-microglobulin in serum. Clin Chem
increased in cerebrospinal fluid [19]. 1990 ; 36: 1961-4.
13 Kandoussi A, Cachera C, Pagniez D, Saile R,
Beta-2-Microglobulin Performance Goals Tacquet A. Quantification of 2-microglobulin
B2M survey data from the 2007 College of American and albuminin plasma and peritoneal dialysis
Pathologists Participant Summary Report shows median fluid by a nonbompetitive immunoenzymo-
interlaboratory imprecision values (% coefficient of metric assay. Clin Chem 1993; 39: 93-6.
variation) of 6% to 8%. Based on biological variability, 14 Donaldson M D, Chambers RE, Woolridge
desirable imprecision should be <3.0%; desirable bias MW, Whicher JT. Stability of alpha-1-
should be <4.1%; and desirable total error should be microglobulin, beta-2-microglobulin, and
<9.0% [20]. retinol binding protein in urine. Clin Chim Acta
1989; 179: 73-7.
References 15 IMMAGE Immunochemistry Systems Beta-2-
1 Harris HW, Gill TJ III. Expression of class I Microglobulin product labelling. Beckman
transplantation antigens. Transplantation 1986; Coulter, Inc. 2001.
42: 109-17. 16 Lutz CT, Cornell SH, Goeken JA.
2 Miyata T, Jadoul M, Kurokawa K, Van Establishment of a reference interval for beta-2-
Ypersele de Strihou C. Beta-2 microglobulin in microglobulin in cerebrospinal fluid with use of
renal disease. J Am Soc Nephrol 1998; 9: 1723- two commercial assays. Clin Chem 1991; 37:
35. 104-7.
3 Berggard I, Bearn AG. Isolation and properties 17 Greipp PR, San Miguel J, Durie BG, Crowley
of a low molecular weight beta2-globulin JJ, Barlogie B, Blade J, et al. International
occurring in human biological fluids. J Biol staging system for multiple myeloma. J Clin
Chem 1968; 243: 4095-4103. Oncol 2005; 23: 3412-20.
4 Dargemont C, Dunon D, Deugnier MA, 18 Assenat H, Calemard E, Charra B, Laurent G,
Denoyelle M, Girault JM, Lederer F, et al. Terrat JC, Vanel T. Hemodialysis: carpal tunnel
Thymotaxin, a chemotactic protein, is identical syndrome and amyloid substance. Nouv Presse
to beta 2-microglobulin. Science. 1989; 246: Med 1980; 9: 1715.
803-6. 19 Clausen N, Ibsen KK. Central nervous system
5 Mori M, Terui Y, Ikeda M, Mishima Y, relapse surveillance by serial beta-2-
Tomizuka H, Ikeda M, Itoh T, et al. Beta(2)- microglobulin measurements in childhood acute
microglobulin identified as an apoptosis- lymphoblastic leukemia. Acta Paediatr Scand
inducing factor and its characterization. Blood 1984; 73: 848-54.
1999; 94: 2744-53. 20 Fraser CG. Biological variation: from principles
6 Petherson PA, Evrin PE, Berggaard I. to practice. Washington, DC: AACC Press;
Differentiation of glomerular, tubular and 2001, p. 140.
normal proteinuria: Differentiations of urinary
excretion of 2-microglobulin, albumin, and
total protein. J Clin Invest 1969; 48: 1189-98.
7 Hemmingsen L, Skaarup P. The 24 hour
excretion of plasma proteins in the urine from
apparently health subjects. Scand J Clin Lab
Invest 1975; 35: 347-53.
8 Woo J, Floyd M, Longley MA, Cannon DC. 2-
Microglobulin clearance as measured by radio-
immunoassay Clin Chem 1980; 26: 1193-7.
9 Itoh Y, Enomoto H, Kawai T. A latex
agglutination test for urinary beta-2-micro-
globulin. Contrib Nephrol 1988; 68: 172-8.
10 Rifai N, Morales A. Immunoturbidimetry of 2-
microglobulin in serum. Clin Chem 1989; 35:
1996-7.
260
Table 1: Beta-2-Microglobulin: Methods Summary Table
Method 1: Radial immunodiffusion (RID); quantitative

Principle of analysis: Diffusion of B2M through medium containing anti-B2M antibody
Usage: Seldom used today
Comments: Serum, plasma, manual; requires very long incubation time
Method 2: Latex agglutination

Principle of analysis: Agglutination of latex particles coated with anti-B2M antibodies due to presence of B2M
results in formation of visible aggregates; size of aggregates related to B2M concentrations
Usage: Seldom used today
Comments: Semiquantitative; may be useful as rapid screen
Method 3: Immunoturbidimetry
Principle of analysis: Reaction of B2M with anti-B2M antibody results in formation of antigen-antibody
complexes which absorb light
Usage: Common quantitative method for B2M
Comments: Sensitive, specific; automated; can be implemented on most current random access chemistry
analyzers
Method 4: Immunonephelometry
Principle of analysis: Reaction of B2M with anti-B2M antibody results in formation of antigen-antibody
complexes which scatter light
Usage: Most common quantitative method for B2M in use
Comments: Sensitive, specific; automated; requires specialized instrumentation (nephelometer).
Method 5: Immunometric assay

Principle of analysis: Two anti-B2M antibodies are used. One is a capture antibody immobilized on a
microparticle; the other detection antibody is labeled with an enzyme or a fluorescent or chemiluminescent
compound. After incubation and removal of unbound reagents by washing, B2M in the sample sandwiches the
capture and detection antibody, and the captured label is proportional to the B2M concentration.
Usage: Not commonly used for B2M
Comments: Sensitive, specific; automated; requires specialized instrumentation; more expensive than
nephelometry and turbidimetry.
261
Bilirubin
Bilirubin
R Swaminathan
Name: Bilirubin
Structure: See figure 1
Principles of Analysis and Current Usage i or conjugated, bilirubin, whereas in the presence of a
Bilirubin was first demonstrated to be present in normal solvent, both forms of bilirubin readily react. The
serum by van den Bergh and Snapper [1]. They found bilirubin that requires solvent to react is termed indirect
that bilirubin in normal serum reacted with Ehrlichs or unconjugated bilirubin. However, it has been known
diazo reagent (diazotized sulfanilic acid) only when for some time that a portion of the unconjugated
alcohol was added. It was later observed that pigment in bilirubin is available for reaction in the absence of
human bile reacted with the diazo reagent without the solvent and that the extent of the reaction is dependent
addition of alcohol [2]. The pigment that reacted in the on the time and temperature of the reaction and on the
absence of alcohol was termed direct. The pigment final concentration of reagents [3]. Decreasing the time
that required the presence of alcohol was termed the of reaction to minimize reactivity of the unconjugated
indirect bilirubin fraction. The response of serum to bilirubin in the absence of solvent can increase the
van den Berghs test, with or without the presence of specificity of the conjugated (direct) bilirubin assay. A
alcohol, has been the basis for several classifications of third bilirubin fraction has also been described [4-7].
jaundice. This fraction, called delta () bilirubin, is a covalent
conjugate between albumin and bilirubin. The -
It is now clear that indirect bilirubin is unconjugated bilirubin fraction was identified by high-performance
bilirubin bound to albumin en route to the liver from the liquid chromatography (HPLC) and made available to
reticuloendothelial system, where it is formed. The the routine clinical laboratory as a result of the dry-film
unconjugated bilirubin is a nonpolar molecule and is not technology used for measuring bilirubin. The -bilirubin
soluble in water. Consequently it will react with diazo fraction reacts like conjugated bilirubin in the diazo
reagent only in the presence of an agent (historically reaction and is usually present in very small amounts in
called an accelerator) such as alcohol, in which both the blood. However, when the levels of conjugated
bilirubin and the diazo reagent are soluble. (The alcohol bilirubin increase, so does the concentration of -
also enhances the intensity of the color formed.) Because bilirubin. The -bilirubin fraction does not react
of the nonpolar nature of the unconjugated bilirubin and quantitatively in most current diazo reactions [8].
its strong affinity to albumin, unconjugated bilirubin is Modifications in diazo concentration and time of
not filtered at the glomerulus and is not normally found reaction are necessary to fully react the -fraction.
in urine.
Qualitative analysis of serum bilirubin as indirect or
In contrast, bilirubin conjugated with glucuronate is a direct, according to the type of van den Bergh reaction,
polar and water-soluble compound that exists in plasma has long been replaced by quantitative determinations of
loosely bound to albumin. When blood levels of the amount of conjugated bilirubin and total bilirubin, in
conjugated bilirubin are high, it is filtered at the which the difference between the total and direct
glomerulus and excreted in the urine. reaction is assumed to represent the indirect or
unconjugated bilirubin.
Conjugated and unconjugated bilirubin fractions have
historically been differentiated by the time of the Methods for Total Bilirubin
reaction and solubility of the fractions. Bilirubin that Diazo Reagent Methods
reacts quickly in the absence of solvent is called direct, Commonly used procedures for measurement of
bilirubin and its fractions are based on the diazo reaction
i
Bilirubin [9,10]. In this reaction, bilirubin reacts with diazotized
Previous and current authors of this method: sulfanilic acid (diazo reagent) to form a colored
First edition: John E. Sherwin, Ronald Overholte chromophore (Figure 1). The diazotized sulfanilic acid
Methods edition: John E. Sherwin, Ronald Overholte reacts at the central methylene carbon of bilirubin to split
Second edition: John E. Sherwin, Ronald Overholte the molecule, forming a mixture of two isomers of
Third edition: Steven C. Kazmierczak azodipyrroles (azobilirubin). The reaction can be carried
Fourth edition: John E. Sherwin, Charles Thompson out at various pH; in the original method of Malloy-
Fifth edition: R. Swaminathan Evelyn (Table 1, Method 1), the procedure was
performed at pH 1.2. At this pH, the azobilirubin is red
262
Bilirubin
purple in color with an absorption maximum of azobilirubin, which is then measured by reflectance at
approximately 560 nm (Figure 2) . The rate of reaction 540 nm and 460 nm. The measurement at 460 nm is used
of the diazo reagent with the bilirubin varies for the for any spectral interference [19].
different forms of bilirubin. Conjugated bilirubin reacts
very rapidly [11] and -bilirubin reacts more slowly Spectrophotometric Methods
[12]. The rate of reaction of unconjugated bilirubin is pH Bilirubin can also be measured by direct
dependent [13]. To accelerate the reaction at pH 6, a spectrophotometry (Table 1, Method 5) [14]. This can be
compound usually called the accelerator is added. In the done with or without prior dilution of the sample. This
original Malloy-Evelyn method, methanol was used as a method is applicable to samples from neonates up to 3
solubilizer of the unconjugated fraction, but many other months of age, because carotenoids which are present in
chemicals have been employed, such as dimethyl older infants and adults also absorb in the 454 nm
sulfoxide and nonionic detergents. The Malloy-Evelyn region. It is, however, necessary to correct for the
method is seldom used in practice now because of presence of oxyhemoglobin (oxyHb), because oxyHb
several problems. These include matrix effects, protein also absorbs at 454 nm. This is accomplished by
precipitation by methanol (causing turbidity) [14], subtraction of the absorbance at 540 nm, since oxyHb
negative interference from hemoglobin, and relatively absorbs equally at each of these wavelengths [20]. This
long reaction time [15]. Other compounds used as method requires either a spectrophotometer or a
accelerator include caffeine, sodium benzoate, bichromatic photometer (bilirubinometer) capable of
dyphylline, sodium dodecyl sulphate, and surfactants measurements at 454 nm and 540 nm. Direct
[16]. spectrophotometry can also be performed with or
without dilution of serum with a suitable buffer [21].
The Jendrassik-Grof modification (Table 1, Method 2) is When using this method, it is assumed that only
carried out at a pH near 6.5, but the absorbance of the unconjugated bilirubin is present in the sample and that
reaction is measured after alkalinization of the reaction the only interfering substance is hemoglobin.
solution to pH 13 [17,18]. At this pH, the absorption
spectrum of the azobilirubin is shifted to a more intense Instruments dedicated for this purpose
blue color measured at 600 nm. The original Jendrassik- bilirubinometersare used in neonatal units, because
Grof method employed sodium benzoatecaffeine to they are convenient and relatively easy to use. However,
solubilize unconjugated bilirubin for the measurement of blood samples need to be centrifuged and serum/plasma
total bilirubin. introduced into the instrument. This is an additional step
for clinical staff to perform. With the availability of
Diazo methods have been successfully automated. whole-blood bilirubin measurements and transcutaneous
However, the modified Jendrassik and Grof method bilirubinometers, this type of meter is less often used at
requires the use of four separate reagents, whereas most present. Although results from a bilirubinometer
commonly used analyzers are designed to use two (UniStat) showed good correlation with routine methods
reagents only. Furthermore, the azosulfanilic acid and [22], in samples with a significant amount of -bilirubin,
ascorbic acid reagents are only stable for a short period the results are unreliable because this species of bilirubin
of timeusually a day. In an attempt to improve the has much higher absorbance than unconjugated bilirubin
assay, hydroxylamine salt was substituted for ascorbic [12]. Methods in which predilution is used are affected
acid (U.S. Patent No. 2569,721). Others have used by protein concentration of the sample and the strength
different diazo reagents such as 2,4 dichlorophenyl or of the pH of the buffer/diluent. Use of diluents
2,5 dichlorophenyl diazonium salts to help with containing caffeine reduces this matrix-related
automation [14]. In the Ektachem dry chemistry system interference [23]. Predilution of samples with Tris buffer
for total bilirubin, a very stable diazonium salt 4N- before spectrophotometric measurements is used in some
carboxymethylsulphanyl-benzene-dazonium automated instruments for the measurement of neonatal
hexafluorophosphate is used [19]. Another common bilirubin (e.g., Abbott Aerostat/Architect).
diazonium salt used in many automated analyzers is 2,5-
dichlorophenyldiazonium tetrafluoroborate (DPD), The direct spectrophotometric method has now been
which reacts very rapidly with bilirubin under acidic incorporated into blood-gas analyzers to measure
conditions (Table 1, Method 3). In the Roche Modular bilirubin in whole blood. In the Radiometer
analyzer, DPD in strong acidic medium (pH 1.0 to 2.0) is (Copenhagen, Denmark) ABL735 blood-gas analyzer,
used together with a solubilizing agent to accelerate the multiple absorbance measurements are made on whole
reaction and avoid protein precipitation. In the Abbott blood between 478 and 672 nm over 128 wavelengths.
Aeroset/Architect, 2,4-dichloroaniline is used as the Red cells are hemolyzed by ultrasonic vibration and the
diazo reagent. absorbance of the hemolysate measured accurately.
Because erythrocytes do not contain any bilirubin,
The dry-film technology method uses analytical reagents plasma bilirubin concentration is calculated using a
layered onto a clear polyester film. In the total bilirubin dilution factor based on the hematocrit to correct for
method, which is a modification of the Jendrassik-Grof dilution of plasma by intracellular fluid. This bilirubin
diazo reaction, dyphylline and a surfactant are used to value is then corrected for hemoglobin pigments. This
dissociate the unconjugated bilirubin from albumin analyzer has the facility to correct for either adult or fetal
(Table 1, Method 4). Conjugated, unconjugated and - hemoglobin, which have different absorption spectra
bilirubin react with diazo reagent to produce [24]. A recent multicenter evaluation of this instrument
263
Bilirubin
found good agreement with laboratory-based The 2007 College of American Pathologists (CAP)
measurements [25]. This method has also been found to comprehensive chemistry quality assurance surveys [34]
give acceptable results for specimens from adult patients showed that 60% of the participants used a Jendrassik-
as well [24]. Direct spectrophotometry methods are rapid Grof procedure (53% of these employed a sample
and convenient to use for point-of-care testing. blank), 13% used a diazonium saltbased method, 16%
used a dry-film assay, 3.8% used an oxidation method,
One of the methods employed in the dry-chemistry and less than 1% used the bilirubin oxidase method [35].
system (Ektachem, Johnson & Johnson) is reflectance
spectrophotometry (Table 1, Method 6). In this method, Other Methods
serum is exposed to a binding agent at pH 8.0, and the A method based on the oxidation of bilirubin to
light reflected is measured at two wavelengths to allow biliverdin by potassium ferricyanide was described by
for the different absorbance maxima of the unconjugated OLeary et al, [36], and this has been adapted for
(420 to 425 nm) and conjugated bilirubin (460 to 465 automated analyzers (Table 1, Method 9). The decrease
nm). This allows the simultaneous measurement of in absorbance at 480 nm is measured with blanking at
conjugated and unconjugated species [26]. 600 nm. This method is not affected by hemolysis, and
the reagents are stable. Nearly 25% of the laboratories in
Direct spectrophotometry is the principle behind the the United Kingdom are using this method routinely.
transcutaneous bilirubinometers (Table 1, Method 7). Results by this method compare well with those by
Several noninvasive transcutaneous bilirubinometers are Jendrassik-Grof methods [37].
now available for the measurement of bilirubin in the
neonate. In the original devices, the yellowness of the Several other methods for measuring total bilirubin have
skin was assessed by measuring the optical density of been described. A fluorimetric method based on the
blue and green light reflected from the skin. These fluorescence of bilirubin bound to protein or detergent
devices were influenced by skin pigmentation [27]. [38], an electrochemical technique [39], and a
BiliCheck (SpectRx Inc., Norcross, GA, USA) and chemiluminescence method based on the reaction of
JM103 (Minolta AirShields) are two commercially bilirubin and peroxynitrite [40] have all been described.
available bilirubinometers. In the BiliCheck meter, white However, none of these have been found to be useful in
light is transmitted onto the skin, and reflected light is routine clinical laboratory practice. An amperometric
analyzed. Absorption from interfering substances method using bilirubin-imprinted poly(methacrylic acid-
(hemoglobin, melanin, and dermal thickness) is isolated to-ethyl glycol dimethylacrylate) film as a sensing
mathematically from light absorption by bilirubin in the electrode has been recently described [41].
capillary bed. Several studies have shown that these
bilirubinometers give results which are in agreement Measurement of Bilirubin Fractions
with laboratory results, especially when the bilirubin
concentration is less than 240 mol/L(14 mg/dL) [28]. Liquid Chromatography
These devices, however, are no substitute for laboratory Several high-performance liquid chromatography
measurement of serum bilirubin in deciding treatment (HPLC) methods have been developed to separate and
options, but they are useful in identifying infants who quantitate the four bilirubin fractions, namely,
require serum bilirubin determination [29]. unconjugated (-bilirubin), monoconjugated (-
bilirubin), diconjugated (-bilirubin), and -bilirubin (a
Enzymatic Methods fraction that is strongly bound to proteins). These
An enzymatic method for determination of bilirubin has methods are also capable of detecting photoisomers
been developed using bilirubin oxidase (EC 1.3.35) which are produced during phototherapy [42]. It has
isolated from Myrothecium verrucaria MT-1 (Table 1, been shown that in sera from patients with cholestatic
Method 8) [30]. The enzyme catalyzes the oxidation of jaundice that the major fraction is -bilirubin (37%) and
bilirubin to biliverdin. The reaction is monitored by di- and monoconjugated forms account for 24% and
measurement of the decrease in absorbance at 405 to 460 13%, respectively. The remaining 27% is unconjugated
nm. Sodium dodecyl sulfate (SDS) and/or sodium bilirubin [43]. Initially, normal HPLC was used to
cholate are included in the total bilirubin assay to separate bilirubin fractions after protein precipitation
dissociate bilirubin from albumin and accelerate the followed by extraction [44]. Later, reverse-phase HPLC
reaction. was used after precipitation of globulins by sodium
sulphate [45], and this separated bilirubin into the four
Enzymatic methods has been automated [31], and the fractions. Adachio et al. [42] have described a reverse-
results of the enzymatic methods are in good agreement phase HPLC method using a MicroMax RP-30 column
with those by the Jendrassik-Grof method. However, without protein precipitation. A more recent direct
differences were observed in samples with high injection method using a Shodex Asshipak GS-320 HQ
direct/conjugated bilirubin concentration, and this can be column was described with good results [46]. A method
reduced by measuring the absorbance at 425 nm instead for measuring -bilirubin using anion exchange
of 465 nm [32]. chromatography has also been reported [47]. The
precision and accuracy of the HPLC methods are not
Hemoglobin was initially thought to cause negative adequate enough to consider this to be a reference
interference, but the inclusion of EDTA in the regent method. Calibration of HPLC methods is a problem. The
reduced this effect [33]. commonly used calibrant with unconjugated bilirubin is
264
Bilirubin
unsatisfactory because the molar absorptivity of the four recommended to reduce the reaction of unconjugated
fractions may not be identical [45]. bilirubin [11]. However, this will also reduce the
reactivity of conjugated and delta fractions. By carefully
Spectrophotometric Methods selecting the reaction conditions, it is possible to
The conjugated and unconjugated bilirubin fractions can measure direct bilirubin, and this correlates well with the
also be estimated by a dry-film spectrophotometric HPLC method.
technique [26]. In the method used in the dry-chemistry
system, conjugated and unconjugated bilirubin are Other Methods
measured simultaneously. Samples are treated with A thin-layer chromatography [56] and a capillary
caffeine, sodium benzoate, and surfactant to dissociate electrophoresis method [57] to separate and quantitate
unconjugated bilirubin from albumin. Unconjugated and bilirubin fractions have also been described. None of
conjugated bilirubin then bind to a cationic polymer these are clinically useful, because the identification of
called a mordant. This interaction of the bilirubin with all bilirubin fractions do not add any diagnostic value.
the mordant causes a spectral shift such that both species
of bilirubin have similar molar absorptivity at 400 nm, Free (Unbound) Unconjugated Bilirubin
but at 460 nm, unconjugated bilirubin has a higher molar The fraction of unconjugated bilirubin which is not
absorptivity. The concentration of conjugated and bound to albumin is negligible in adults, but may be high
unconjugated bilirubin are determined by measuring the in neonates. This fraction is implicated in the
reflectance at these two wavelengths. The concentration pathogenesis of kernicterus [58]. Based on the principle
of -bilirubin is calculated as the difference between that free bilirubin is rapidly oxidized by horseradish
total bilirubin (by the diazo dry-film method) and the peroxidase, whereas the albumin-bound fraction is
sum of conjugated and unconjugated bilirubin by the protected from this reaction [59], a method for
spectral shift method [48]. measuring free bilirubin has been described. Two
commercial analyzers are available to measure this
Enzymatic Methods fraction. These are the UB analyzer (Arrows Co Ltd.,
Enzymatic methods using bilirubin oxidase have also Osaka, Japan)[60] and the FloPro analyzer (Global FIA,
been used for the measurement of bilirubin fractions. Fox Island, WA) [61]. In the UB analyzer, the sample is
Direct bilirubin reacts at pH 3.7, and total bilirubin at pH diluted 42-fold, whereas in the FloPro analyzer, the
7.2 [33]. In the absence of detergents, conjugated sample is diluted only 2-fold. A recent study showed that
bilirubin and -bilirubin are oxidized by bilirubin the use of assay buffer without chloride gave lower
oxidase at acidic pH, and unconjugated bilirubin is values in the UB analyzer [62].
oxidized at alkaline pH. By altering the pH and by
addition of detergent, all three fractions can be Other methods of measuring free bilirubin include
measured. Total bilirubin is measured at pH 10.0 in the Sephadex column chromatography, where the unbound
presence of SDS, direct bilirubin (conjugated and delta) bilirubin is retained, while the bound bilirubin is not
is measured at pH 3.7, and conjugated bilirubin at pH [63]. Assays based on the fluorescence of albumin-
10.0 in the absence of SDS. This method compares well bound bilirubin [64] and on the quenching of albumin
with results from an HPLC method [49,50]. In another fluorescence by bilirubin binding [65] have also been
approach to measure conjugated bilirubin, the reaction of described to measure bilirubin binding capacity.
bilirubin oxidase with -bilirubin and unconjugated
bilirubin is reduced by the addition of sodium fluoride Urine bilirubin measurements are most often made on
and N-acetylcysteine [51]. Morimoto et al. [52] fresh specimens using dipsticks (Chemstrip or Multistix)
measured the decrease in absorbance at 450 nm caused impregnated with direct-reacting diazo reagent, such as
by the oxidation of conjugated bilirubin by bilirubin 2,4-dichloroaniline diazonium salts. The Ictotest uses a
oxidase in the presence of methylene phosphonic acids, similar chemical reaction as the bilirubin pad on the
EDTA, manganese, and 4-hydroxy-2,2,6,6,- multi-reagent dipstick, but is presented in a tablet
tetramethylpiperidine 1-oxyl. It has been reported that up format. The active reagent is p-nitrobenzene diazonium
to 5% of unconjugated bilirubin may be oxidized in p-toluenesulfonate . With a sensitivity of about 10 mg/L,
these enzymatic methods [53], and others have reported it is two to four times more sensitive than the dipstick
differences between diazo methods and enzymatic procedure. Dipstick methods are specific but subject to
methods in the measurement of conjugated bilirubin interference. Medications such as phenazopyridine,
[54]. which color the urine, can cause false-positive results.
Other compounds reported to cause interference are
Diazo Methods aminosalicylic cid, ascorbic acid, chlorhexidine, and
In the absence of an accelerator (such as caffeine chlorpromazine and its metabolites [66].
reagent), bilirubin fractions other than the unconjugated
fraction react with diazo reagents. This fraction is called Reference and Preferred Methods
direct bilirubin. However, it is now known that some A modified Jendrassik-Grof procedure that employs
unconjugated bilirubin and -bilirubin react under these caffeine-benzoate reagent to solubilize all the various
conditions [12,48]. Even at low pH, a small proportion bilirubin fractions with excellent recovery has been
of unconjugated bilirubin will react with the diazo recommended as a reference method [11].
reagent, especially if the sodium nitrite concentration is
high [55]. Preincubation of serum with HCl has been
265
Bilirubin
Because the direct and total bilirubin methods have ascribed to the ability of the dry-film methods to
limited specificity, the true concentrations and differentiate between -bilirubin and bilirubin
proportions of conjugated and unconjugated bilirubin in glucuronide, both of which react in diazo procedures but
serum have not been known with any great certitude. to different extents. There is very little interference from
The development of HPLC procedures [44] that can hemoglobin [35]. Carotene does not interfere in the dry-
physically separate and quantitate each fraction has led film assay. It continues to be the only method used in the
to a better understanding of the composition of bilirubin routine clinical laboratory that measures the bilirubin
in normal blood. The HPLC results show that in healthy fractions (conjugated, unconjugated, and -bilirubin)
subjects, conjugated bilirubin is less than 5% of the total rather than classifying them as direct or indirect reacting
and there is very little -bilirubin. Comparison with groups.
HPLC methods show that most diazo methods greatly
overestimate the concentration of conjugated fraction. In Transcutaneous methods for measurement of bilirubin
addition, the HPLC analysis indicates the presence in have been developed that are not affected by skin color
serum of non-bilirubin compounds that are also diazo and show good agreement with bilirubin measured in
reacting. Certain compounds present in urine, such as serum [67]. However, one must be aware that
mesobilifuscin and uroerythrin, are known to interfere transcutaneous and blood bilirubin measurements do not
with urinary bilirubin analysis, and it is presumed that measure the same parameter. Transcutaneous methods
these or similar compounds exist in plasma. measure the amount of bilirubin that has moved from the
serum into tissue. Whether or not the transcutaneous
These studies indicate that chemical assays that give a bilirubin measurements are indicative of the bilirubin
low proportion (15% or less) of direct-reacting bilirubin that is available to move into brain tissue is still
in healthy persons probably give the most accurate unanswered.
results. For a particular diazo method, the amount of
direct-reacting bilirubin apparent in normal serum Perhaps a more significant problem than choosing
depends on the time and temperature of the reaction. between the available methods is the problem of
choosing an appropriate calibrator. Many apparent
The primary advantages of the Jendrassik-Grof inaccuracies and differences between methods are a
procedures include the greater sensitivity and precision result of differences in standardization.
(especially at low bilirubin concentrations) as a result of Recommendations for a uniform bilirubin standard are
the higher molar absorptivity of the azobilirubin under available, as are serum-based calibrated bilirubin
alkaline conditions (approximately 7 104 L mol1 standards. However, concerned laboratories should
check the accuracy of serum-based calibrators by using
cm 1), relative insensitivity to hemoglobin interference, pure bilirubin standards available from the National
rapid color development, and a large linear range (up to Institute for Standards and Technology (Standard
1.7 absorbance units). Jendrassik-Grof procedures often Reference Material No. 916a). Several methods are
use ascorbic acid to minimize hemoglobin interference, available for preparing pure bilirubin standards, and
though a sample blank is also used to minimize the these are discussed below. Bilirubin conjugates have
effect of endogenous color. been synthesized and are available commercially
(Ultimate D, Beckman, Brea, CA). There is evidence
The use of bilirubinometers to measure total bilirubin is that these conjugates react similarly to glucuronide
useful only in the neonatal population because of the conjugates in diazo reactions and thus could be used for
presence of carotenoid compounds in adult serum that more accurate standardization of direct bilirubin
can cause strong positive interference. These instruments analyses [4].
are simple to use, however, require a small sample
volume, and have good accuracy and precision if Specimen
maintained properly. The recent introduction of blood- Total bilirubin determinations using a diazo method
gas analyzers with bilirubin measurement by direct require either serum or plasma. Bilirubin is very
spectrophotometry makes the measurement of bilirubin sensitive to and destroyed by light and heat, and
even easier, and some studies suggest that these specimens should be protected from ambient light prior
instruments can be used in older children and adults to and during analysis. Concentrations may decrease by
[24]. 30% to 50% per hour if exposed to direct sunlight. If
separated and stored in the dark, serum or plasma is
There is no definitive method for bilirubin. At this time, stable for 2 days at room temperature, 4 days at 4C and
there seems to be no clear-cut reason to choose between at 20C for an indefinite period. Repeated freezing and
the diazo, enzymatic, spectrophotometry, or OLeary thawing has no significant effect.
methods. Since all have acceptable precision and are
adapted to many automated instruments, the choice will Conjugated bilirubin may be determined in either serum
depend on the available instrument or individual or plasma, but serum is the usual sample. Spinal fluid
laboratory preference. samples can be analyzed for both total and direct
bilirubin. Urine samples can also be analyzed by direct
The dry-film procedure for total, conjugated, and diazo methods, since the protein-bound unconjugated
unconjugated bilirubin analysis has compared well with bilirubin does not cross the glomerular barrier [68].
diazo procedures. The bias between the dry-film and
Jendrassik-Grof methods for direct bilirubin has been
266
Bilirubin
Conjugated bilirubin levels up to 3.5 mol/L (2 mg/L)

Interferences are found in infants by 1 month of age, and conjugated
Diazo methods are prone to interference from turbidity, bilirubin remains at this level thereafter. Figure 3
hemoglobin, indican, metal ions, and some drugs such as summarizes serum bilirubin concentrations as a function
L-dopa [14]. Although metal ions, especially zinc, can of age and sex [80].
cause significant interference in clinical practice, it is
thought to be of little importance [11]. Interference from Serum bilirubin shows diurnal variation, with evening
turbidity and lipemia is reduced when caffeine or other values being approximately 30% higher than in the
surfactants are used [23]. Interference from high protein morning [82]. During pregnancy, serum unconjugated
concentration (due to paraproteinemia) has been reported bilirubin concentrations are about 55% lower compared
to cause spurious elevation with the Roche automated to nonpregnant values. Some of this change is due to
total bilirubin method and is thought to be due to decreased albumin concentration during pregnancy [83].
precipitate formation [69]. A small seasonal variation (lower in summer) has also
been described [84]. In breastfed infants, the serum
Indicans, which are present in high concentration in bilirubin concentration is higher (up to 253 mol/L)
uremia, can react with diazo reagents and give compared to bottle fed infants (up to 212 mol/L) [85].
spuriously high results. Methods using dichlorophenyl Fasting tends to increase serum bilirubin; after 48 hours
diazonium salts are most affected, and diazotized of fasting, a 240% rise in bilirubin is observed.
sulfanilic acid methods are least effected [70].
Interference from indicans can be reduced by shortening
the reaction time [71,72]. Interpretation
The most common cause of an elevated bilirubin
Hemolysis can cause interference in the diazo methods concentration in the population is Gilberts syndrome, a
by several possible mechanisms: interfering with the benign condition affecting up to 5% of the population
diazo reaction, spectral interference [73], and destruction and due to reduced hepatic glucuronyl transferase
of bilirubin prior to its reaction [74]. The degree of activity as a result of a mutation in the bilirubin-UDP-
interference depends on the formulation of the bilirubin glucuronosyltransferase (UGT1A1) gene. Total bilirubin
reagent [75], and the interference can be negative or concentration is usually < 50 mol/L in this condition
positive [76]. In neonates, the commonest method of and is predominantly unconjugated. Fasting tends to
blood collection for bilirubin is by capillary heel-prick, increase the bilirubin concentration in Gilberts
and this method is associated with some degree of syndrome [86].
hemolysis [77]. The mean serum free hemoglobin
concentration in capillary heel-prick samples was found Jaundice is common in the newborn and is most
to be 1.62 g/L (range < 0.5 to 6.00 g/L), and this can commonly due to elevation in unconjugated bilirubin as
cause significant negative interference when bilirubin is a result of immaturity of the liver (physiological
measured by methods using 2,5-dichlorophenyl jaundice) or due to hemolytic disease of the newborn.
diazonium tetrafluoroborate (DPD) [78]. Serum unconjugated bilirubin may reach more than 400
mol/L in hemolytic disease.
The presence of lipemia has been shown to increase
measured bilirubin concentrations. Turbid serum creates In the adult, increased bilirubin and jaundice can arise
artifacts in the performance of spectrophotometers and due to increased production (prehepatic), diseases of the
should not be directly analyzed. The use of gel-barrier liver (hepatic), or obstruction to bile flow (posthepatic).
tubes for collection of specimen showed no significant In prehepatic jaundice, the bilirubin is predominantly
change in bilirubin concentrations after 1 week [79]. unconjugated, and the concentration rarely exceeds 100
mol/L. In obstructive jaundice, the bilirubin is
Bilirubin Reference Interval predominantly conjugated bilirubin, and concentrations
Reference interval for bilirubin varies with the method. in excess of 700 mol/L can be seen. In hepatic jaundice,
For most diazo-based methods, the reference interval for there is a mixture of conjugated and unconjugated
total bilirubin concentration in healthy adults is up to 26 bilirubin.
mol/L (15 mg/L), with a median of 12 mol/L (7.0
mg/L). In both sexes, the median concentration rises Although it has been possible to measure -bilirubin, the
after puberty. After puberty, males have both higher methodology is limited to HPLC, dry-chemistry, or
median values and a more pronounced skew to high enzymatic methods and therefore not available in all
levels than females do [8]. The observation that men clinical laboratories. There has been a suggestion that
have a higher serum value than women has been measurement of -bilirubin may be helpful in monitoring
confirmed, but the difference is probably not clinically biliary drainage. An increase in the percentage of -
significant. In addition, significant racial differences bilirubin is an indication of good drainage [43].
have been reported [80]. The racial differences were Measurement of -bilirubin and conjugated bilirubin has
more pronounced between the female groups than the been reported to be useful in the early detection of
male. Concentrations decrease slightly during the sixth rejection after liver transplantation [87].
through ninth decade of life [81].
Apart from its use in the diagnosis of liver and biliary
diseases, serum bilirubin has also been suggested to be a
267
Bilirubin
marker of coronary heart disease. Bilirubin via its 12 Doumas BT, Wu TW, Jendrzejczak B. Delta
antioxidant potential is inversely associated with risk of bilirubin: absorption spectra, molar
coronary artery disease [88]. absorptivity, and reactivity in the diazo
reaction. Clin Chem 1987;33:769-774.
13 Landis JB, Pardue HL. Kinetics of the reactions
Bilirubin Performance Goals of unconjugated and conjugated bilirubins with
The current target for acceptable performance for total p-diazobenzenesulfonic acid. Clin Chem
bilirubin measurements as mandated by The Clinical 1978;24:1690-1699.
Laboratory Improvement Amendments [89] is 0.7 14 Westwood A. The analysis of bilirubin in
mol/L (4.0 mg/L) or 20% of the peer-group mean serum. Ann Clin Biochem 1991;28:119-130.
value, whichever is greater. The recommended target, 15 Doumas BT, Perry BW, Sasse EA, Straumfjord
however, is 10% [90]. The interlaboratory coefficient of JA. Standardization in bilirubin assays:
variation (CV) is generally 6% to 11% at bilirubin levels evaluation of selected methods and stability of
near normal ranges (approximately 10 mg/L) and 3% to bilirubin solutions. Clin Chem 1973;19:984-
7% CV at bilirubin concentrations of 45 to 50 mg/L [91]. 993.
In healthy adults, bilirubin demonstrated a within-subject 16 Novros JS, Koch TR, Knoblock EC. Improved
variation of approximately 17% when measured over a method for accurate quantitation of total and
20-week period [92]. conjugated bilirubin in serum. Clin Chem
1979;25:1891-1899.
17 Jendrassik L, Grof P. Vereinfachte
References photometrische Methoden zur Bestimmung des
1 van den Bergh AAH, Snapper J. Die Blutbilirubin. Biochem 1938;297:81-89.
Farbenstoffe des Blutserums. Dtsch Arch Klin 18 Koch TR, Doumas DT, Elser RC, Straumfjord
Med 1913;110:540-541. JV. Bilirubin, total and conjugated, modified
2 van den Bergh AAH, Mller P. ber eine Jendrassik-Grof method. In: Faulkner WR,
direkte und eine indirekte Diazoreaktion auf Meites S, eds. Selected Methods of Clinical
Bilirubin. Biochem 1916;77:90-103. Chemistry. Vol. 9. Washington, DC: AACC
3 Winkelman J, Cannon DC, Jacobs SL. Press; 1982:113-118.
Bilirubin. In: Henry RJ, ed. Clinical Chemistry: 19 Dappen GM, Sundberg MW, Wu TW, Babb
Principles and Technics. 2nd ed. New York: BE, Schaeffer JR. A diazo-based dry film for
Harper & Row; 1974:1042-1079. determination of total bilirubin in serum. Clin
4 Doumas BT, Wu TW, Jenrzejezak B. The Chem 1983;29:37-41.
reaction of bilirubin firmly-bound to protein (- 20 Levofff AH, Westphal MC, Finklea JF.
bilirubin) with the diazo reagent. Clin Chem Evaluation of direct reacting spectrophotometer
1984;30:971. for neonatal bilirubinometry. Am J Clin Pathol
5 Brett EM, Hicks JM, Powers DM, Rand RN. 1970:54:562-565.
Delta bilirubin in paediatric patients: 21 Hertz H, Dybkaer R, Lauritzen M. Direct
correlations with age and disease. Clin Chem spectrometric determination of the
1984;30:1561-1564. concentration of bilirubins in serum. Scand J
6 Ou CN, Gilman GE, Buffone GJ. Evaluation of Clin Lab Invest 1974;33:215-230.
the Kodak Ektachem clinical chemistry slide for 22 Barko HA, Jackson GL, Engle WD. Evaluation
the measurement of bilirubin in newborns. Clin of a point-of-care direct spectrophotometric
Chim Acta 1984;40:167-172. method for measurement of total serum
7 Lauff JJ, Kasper ME, Wu TW, Ambrose RT. bilirubin in term and near-term neonates. J
Isolation and preliminary characterization of a Perinatol 2006;26:100-105.
serum bilirubin fraction firmly bound to protein. 23 Vink KL, Schuurman W, van Gansewinkel R.
Clin Chem 1982;28:629-637. Use of the caffeine reagent in direct
8 Mullon CJ-P., Langer, R. Determination of spectrophotometry of bilirubin. Clin Chem
conjugated and total bilirubin in serum of 1986;32:1389-1393.
neonates, with use of bilirubin oxidase. Clin 24 Peake M, Mazzachi B, Fudge A, Bais R.
Chem 1987;33:1822-1825. Bilirubin measured on a blood gas analyser: a
9 Malloy, HT, Evelyn KA. The determination of suitable alternative for near-patient assessment
bilirubin with the photoelectric colorimeter. J of neonatal jaundice? Ann Clin Biochem
Biol Chem 1937;119:481-490. 2001;38:533-540.
10 Doumas BT, Wu TW. The measurement of 25 Borgard JP, Szymanowicz A, Pellae I, Szmidt-
bilirubin fractions in serum. Adjid V, Rota M. Determination of total
Crit Rev Clin Lab Sci. 1991;28:415-45. bilirubin in whole blood from neonates: results
11 Doumas BT, Kwok-Cheung PP, Perry BW, from a French multicenter study. Clin Chem
Jendrzejczak B, McComb RB, Schaffer R, Lab Med 2006;44:1103-1110.
Hause LL. Candidate reference method for 26 Sundberg MW, Lauff JJ, Weiss JS, Dappen
determination of total bilirubin in serum: GM, Wu TW, Spayd RW, Seligson D.
development and validation. Clin Chem Estimation of unconjugated, conjugated, and
1985;31:1779-1789. "delta" bilirubin fractions in serum by use of
268
Bilirubin
two coated thin films. Clin Chem 41 Syu MJ, Chiu TC, Lai CY, Chang YS.
1984;30:1314-1317. Amperometric detection of bilirubin from a
27 Bertini G, Rubaltelli FF. Non-invasive micro-sensing electrode with a synthetic
bilirubinometry in neonatal jaundice. Semin bilirubin imprinted poly (MAA-co-EGDMA)
Neonatol 2002;7:129-33. film. Biosens Bioelectron 2006;15:22:550-557.
28 Leite MG, Granato Vde A, Facchini FP, Marba 42 Adachi Y, Inufusa H, Yamashita M, Kambe A,
ST. Comparison of transcutaneous and plasma Yamazaki K, Sawada Y, Yamamoto T. Clinical
bilirubin measurement. J Pediatr (Rio J) application of serum bilirubin fractionation by
2007;83:283-286. simplified liquid chromatography. Clin Chem
29 Boo NY, Ishak S. Prediction of severe 1988;34:385-388.
hyperbilirubinaemia using the BiliCheck 43 Kozaki N, Shimizu S, Higashijima H, Kuroki
transcutaneous bilirubinometer. J Paediatr Child S, Yamashita H, Yamaguchi K et al.
Health 2007;43:297-302. Significance of serum delta-bilirubin in patients
30 Osaki S, Anderson S. Enzymatic determination with obstructive jaundice. J Surg Res
of bilirubin. Clin Chem 1984;30:971. 1998;79:61-65.
31 Heinemann G, Vogt W. The determination of 44 Blanckaert N, Kabra PM, Farina FA, Stafford
bilirubin with a new enzymatic method (Dri- BE, Marton LJ, Schmid R. Measurement of
STAT bilirubin) using the Hitachi 704 selective bilirubin and its monoconjugates and
analyzer. J Clin Chem Clin Biochem diconjugates in human serum by alkaline
1988;26:391-397. methanolysis and high-performance liquid
32 Perry B, Doumas BT, Buffone G, Glick M, Ou chromatography. J Lab Clin Med 1980;96:198-
CN, Ryder K. Measurement of total bilirubin by 212.
use of bilirubin oxidase. Clin Chem 45 Lauff JJ, Kasper ME, Ambrose RT.
1986;32:329-332. Quantitative liquid-chromatographic estimation
33 Otsuji S, Mizuno K, Ito S, Kawahara S, Kai M. of bilirubin species in pathological serum. Clin
A new enzymatic approach for estimating total Chem 1983;29:800-805.
and direct bilirubin. Clin Biochem 1988;21:33- 46 Osawa S, Sugo S, Yoshida T, Yamaoka T,
38. Nomura F. An assay for separating and
34 College of American Pathologists website. quantifying four bilirubin fractions in untreated
Available at human serum using isocratic high-performance
<http://www.cap.org/apps/cap.portal> Accessed liquid chromatography. Clin Chim Acta
2008.04.15. 2006;366:146-155.
35 Wu TW, Dapper, GM, Powers DM Lo DH, 47 Seligson D, Seligson H, Wu TW. An anion-
Rand RN, Spayd RW. The Kodak Ektachem exchange chromatographic method for
clinical chemistry slide for measurement of measuring bilirubin covalently bound to
bilirubin in newborns: principles and albumin. Clin Chem 1985;31:1317-1321.
performance. Clin Chem 1982;28:2366-2372. 48 Wu TW. Bilirubin analysis: the state of the art
36 O'Leary N, Pembroke A, Duggan PF. A robust and future prospects. Clin Biochem
procedure for the automated measurement of 1984;17;221-229.
total serum bilirubin using potassium 49 Kosaka A, Yamamoto C, Morishita Y, Nakane
ferricyanide. Ann Clin Biochem 1993;30:175- K. Enzymatic determination of bilirubin
179. fractions in serum. Clin Biochem 1987;20:451-
37 Sturm E, Albrecht-Groos R, Seyfarth M. 458.
Evaluation of new oxidation methods for the 50 Doumas BT, Perry B, Jendrzejczak B, Davis L.
measurement of bilirubin on the Aeroset Measurement of direct bilirubin by use of
clinical chemistry analyzer and comparison bilirubin oxidase. Clin Chem 1987;33:1349-
with methods on the Hitachi 717. Clin Lab 1353.
2002;48:401-405. 51 Kurosaka K, Senba S, Tsubota H, Kondo H. A
38 Cashore WJ, Oh W, Blumberg WE, Eisinger J, new enzymatic assay for selectively measuring
Lamola AA. Rapid fluorometric assay of conjugated bilirubin concentration in serum
bilirubin and bilirubin binding capacity in blood with use of bilirubin oxidase. Clin Chim Acta
of jaundiced neonates: comparisons with other 1998;269:125-136.
methods. Pediatrics 1980;66:411-416. 52 Morimoto Y, Ishihara T, Takayama M, Kaito
39 Koch TR, Akingbe OO. Feasibility of M, Adachi Y. Novel assay for measuring serum
measuring free and total bilirubin conjugated bilirubin and its clinical relevance. J
electrochemically in serum. Clin Chem Clin Lab Anal 2000;14:27-31.
1981;27:1295-1299. 53 Doumas BT, Yein F, Perry B, Jendrzejczak B,
40 Chao Lua, Jin-Ming Lina, Carmen W, Huie B. Kessner A. Determination of the sum of
Determination of total bilirubin in human serum bilirubin sugar conjugates in plasma by
by chemiluminescence from the reaction of bilirubin oxidase. Clin Chem 1999;45:1255-
bilirubin and peroxynitrite. Talanta 1260.
2004;63,333-337. 54 Nakayama K. Differences between enzymatic
and diazo methods for measuring direct
269
Bilirubin
bilirubin. Eur J Clin Chem Clin Biochem 68 Wilding P, Zilva JA, Wilde CE. Transport of
1995;33:513-517. specimens for clinical chemistry analysis. Ann
55 Chan KM, Scott MG, Wu TW, Clouse RE, Clin Biochem 1977;14:301-306.
Calvin DR, Koenig J et al. Inaccurate values 69 Pantanowitz L, Horowitz GL, Upalakalin JN,
for direct bilirubin with some commonly used Beckwith BA. Artifactual hyperbilirubinemia
direct bilirubin procedures. Clin Chem due to paraprotein interference. Arch Pathol
1985;31:1560-1563. Lab Med 2003;127:55-59.
56 Sieg A, Knig R, Ullrich D, Fevery J. 70 Poon R, Hinberg IH. Indican interference with
Subfractionation of serum bilirubin by alkaline six commercial procedures for measuring total
methanolysis and thin-layer chromatography. bilirubin. Clin Chem 1985;31:92-94.
An aid in the differential diagnosis of icteric 71 McPhaul L, Kershaw M, Tilque D, Eckfeldt JH.
diseases. J Hepatol 1990;11:159-164. A 2,4-dichlorophenyl diazonium-based method
57 Wu N, Sweedler JV, Lin M. Enhanced for total bilirubin without interference from
separation and detection of serum bilirubin indican in uremic sera. Clin Chem
species by capillary electrophoresis using a 1985;31:1229-1231.
mixed anionic surfactant-protein buffer system 72 Henrich LM, Bruns DE, Haverstick DM,
with laser-induced fluorescence detection. J Reynolds VG, Boyd JC. Indican interference in
Chromatogr B Biomed Appl 1994;654:185-191. bilirubin assays: a classical solution still
58 Amin SB, Ahlfors C, Orlando MS, Dalzell LE, applies. Clin Chem 2007;53:359-361.
Merle KS, Guillet R. Bilirubin and serial 73 Mori L. Modified Jendrassik-Grof method for
auditory brainstem responses in premature bilirubins adapted to the Abbott Bichromatic
infants. Pediatrics 2001;107:664-670. Analyzer. Clin Chem 1978;24:1841-1845.
59 Jacobsen J, Wennberg RP. Determination of 74 Shull BC, Lees H, Li PK. Mechanism of
unbound bilirubin in the serum of newborns. interference by hemoglobin in the determination
Clin Chem 1974;20:783-789. of total bilirubin. II. Method of Jendrassik-Grof.
60 Shimabuku R, Nakamura H. Total and unbound Clin Chem 1980;26:26-29.
bilirubin determination using an automated 75 van der Woerd-de Lange JA, Guder WG,
peroxidase micromethod. Kobe J Med Sci Schleicher E, Paetzke I, Schleithoff M, Wieland
1982;28:91-104. OH. Studies on the interference by haemoglobin
61 Ahlfors CE, Marshall GD, Wolcott DK, Olson in the determination of bilirubin. J Clin Chem
DC, Van Overmeire B et al. Measurement of Clin Biochem 1983;21:437-443.
unbound bilirubin by the peroxidase test using 76 Brady J, OLeary N. Interference due to
Zone Fluidics. Clin Chim Acta 2006;365:78-85. haemolysis in routine photometric analysis: a
62 Ahlfors CE, Vreman HJ, Wong RJ, Bender GJ, survey. Ann Clin Biochem 1998;35:128-134.
Oh W, Morris BH et al. Phototherapy 77 Meites S, Lin SS, Thompson C. Studies on the
Subcommittee; National Institute of Child quality of specimens obtained by skin puncture
Health and Development (NICHD) Neonatal of children. I. Tendency to hemolysis and
Research Network. Effects of sample dilution, hemoglobin and tissue fluid as contaminants.
peroxidase concentration, and chloride ion on Clin Chem 1981;27:875-878.
the measurement of unbound bilirubin in 78 Algeciras-Schimnich A, Cook WJ, Milz TC,
premature newborns. Clin Biochem Saenger AK, Karon BS. Evaluation of
2007;40:261-267. hemoglobin interference in capillary heel-stick
63 Kapitulnik J, Blondheim SH, Kaufmann NA. samples collected for determination of neonatal
Sephadex absorption of bilirubin from neonatal bilirubin. Clin Biochem 2007;40:1311-1316.
and adult serum. Clin Chem 1972;18:43-47. 79 Hein M, Heil W, Withold W. Storage of serum
64 Lamola AA, Eisinger J, Blumberg WE, Patel or whole blood samples? Effects of time and
SC, Flores J. Fluorometric study of the partition temperature on 22 serum analytes. Eur J Clin
of bilirubin among blood components: basis for Chem Clin Biochem 1995;33:231-238.
rapid microassays of bilirubin and bilirubin 80 Werner M, Toos RE, Hultin JV, Mellecker J.
binding capacity in whole blood. Anal Biochem Influence of sex and age on the normal range of
1979;15:100:25-42. eleven serum constituents. Clin Chem
65 Levine RL. Fluorescence-quenching studies of 1970;8:105-115.
the binding of bilirubin to albumin. Clin Chem 81 Tietz NW, Shuey DF, Wekstein DR. Laboratory
1977;23:2292-2301. values in fit aging individuals, sexagenarians
66 Youngs Effects Online. Available at through centenarians. Clin Chem 1992;38:1167-
<https://www.fxol.org/aaccweb/> 1185.
67 Rubaltelli FF, Gourley GR, Loskamp N Modi 82 Pocock SJ, Ashby D, Shaper AG, Walker M,
N, Roth-Kleiner M, Sendert P et al. Broughton PM. Diurnal variations in serum
Transcutaneous bilirubin measurement: a biochemical and haematological measurements.
multicenter evaluation of a new device. J Clin Pathol 1989;42:172-179.
Pediatrics 2001;107:1264-1271. 83 Lockith G. Handbook of Diagnostic
Biochemistry and Hematology in Normal
270
Bilirubin
Pregnancy. Boca Raton, FL: CRC Press; 88 Mayer M. Association of serum bilirubin
2000:47. concentration with risk of coronary artery
84 Letellier G, Desjarlais F. Study of seasonal disease. Clin Chem 2000;46:1723-1727.
variations for eighteen biochemical parameters 89 Clinical Laboratory Improvement Amendments.
over a four-year period. Clin Biochem Available at <http://www.cms.hhs.gov/clia/>
1982;15:206-211. Accessed 16.04.2008.
85 Soldin SJ, Brugnara C, Hicks JM. Pediatric 90 Lo SF, Doumas BT, Ashwood ER. College of
Reference Ranges. 3rd ed. Washington, DC: American Pathologists.
AACC Press; 1999:45. Bilirubin proficiency testing using specimens
86 Ashraf W, van Someren N, Quigley EM, containing unconjugated
Saboor SA, Farrow LJ. Gilberts syndrome and bilirubin and human serum: results of a College
Ramadan: exacerbation of unconjugated of American Pathologists study. Arch Pathol
hyperbilirubinemia by religious fasting. J Clin Lab Med. 2004 ;128:1219-23.
Gastroenterol. 1994 ;19:122-4. 91 College of American Pathologists.
87 Wu TW, Levy GA, Yiu S, Au JX, Greig PD, Comprehensive Chemistry Survey Report C-B,
Strasberg SM et al. Delta and conjugated 2007. Washington, DC: CAP; 2007.
bilirubin as complementary markers of early 92 Fraser CG. Biological variation in clinical
rejection in liver-transplant recipients. Clin chemistry, Arch Pathol Lab Med 1992;116:916-
Chem 1990;36:9-14. 23.
Table 1: Methods for Total Bilirubin Analysis

Method Principle Usage Comments
Diazo Methods
Method 1 Mallory-Evelyn Reaction of diazotized sulfanilic Of historical Risk of protein
acid at pH 1-2 interest precipitation
Method 2 Jendrassik-Grof Reaction as above, but readings Most common
taken at alkaline pH method
Other Diazo Methods
Method 3 Dichlorophenyl- Rapid reaction with bilirubin Used in Roche
diazonium- under acidic conditions modular
tetrafluoroborate analyzer
(DPD)
Method 4 Dyphylline Similar to Method 2, Used in dry-
absorbance measurement at chemistry
540 nm system
Direct
Spectrophotometric
Methods
Method 5 Bilirubinometer Direct absorbance reading of Useful POCT Interference from
plasma at 454 nm device oxyHb
Method 6 Dry chemistry Reflectance spectrophotometry
Method 7 Transcutaneous Reflectance spectrophotometry Useful POCT May not be
bilirubinometer device applicable in
subjects with
dark skin
Oxidation
Method 8 Enzymatic Decrease in absorbance Used in
following oxidation of automated
bilirubin by the enzyme analyzers
bilirubin oxidase
Method 9 Chemical oxidation As above but oxidation using Used in
potassium ferricyanide automated
analyzers
271
Bilirubin
Figures
Figure 1
Formation of diazotized sulfanilic acid and its reaction with esterified and nonesterified forms of bilirubin to form
azobilirubin derivatives. Me, CH3 group; R, CH=CH2 group.
Figure 2
Spectra of alkaline azobilirubin chromophores. Dashed

curve, reagent versus water; dashed-dotted curve, sample
(serum containing 65 mg/L total bilirubin) reaction versus
water; solid curve, sample reaction versus reagent blank.
272
Bilirubin
Figure 3
Figure 3
Serum total bilirubin concentrations in percentiles. In both sexes, median concentration rises after puberty, falls during
third decade, and thereafter remains stable. Notice that similar shifts are more pronounced at 90th and 95th percentile
levels and that 5th and 10th percentile levels hardly change. After puberty, males have both higher median values and a
more pronounced skew at high values than females do. (From [80] Werner M et al. Z Klin Chem 1970;8:105-115.)
Preparation of Bilirubin Standard The following procedure must be carried out in a well-
darkened room:
Pure bilirubin can be obtained from several commercial Stock Bilirubin Standard (200 mg/L) Weigh about 20
sources. However, the best source is the National mg of bilirubin SRM916a to the nearest 0.01 mg in a
Institute for Standards and Technology (Standard plastic weighing boat, and transfer to the bottom of a
Reference Material No. 916a). It should be used as 100-mL volumetric flask using 1 mL of dimethyl
described below to calibrate serum that is used as the sulfoxide (DMSO). Disperse the bilirubin by gently
reference or standard in the total bilirubin procedure. swirling the contents of the flask. Wash the plastic dish
Thus the commercially available reference pools or and the neck of the flask with 2.0 mL of 0.1 M sodium
laboratory-prepared pools can be checked for the carbonate solution (solution 10). Swirl the flask gently
accuracy of the stated bilirubin values. If the procedure until a crystal-clear red-orange solution is obtained.
is carefully carried out, one should use the value Make up to 100 mL with the BSA diluent. Wrap the
obtained by the validation method rather than the flask in aluminium foil to protect from light. This
manufacturers stated value if there is a substantial standard is stable for at least 6 months at 70C, but
difference between the two. deteriorates at 20C.
273
Bilirubin
Inspect an aliquot of the stock standard after Reagents

centrifugation at 1500 g for 10 minutes. If there is any 1. Sulfanilic acid (26.2 mmol/L in 0.18 mol/L
orange sediment visible, discard the stock and prepare it HCl). Add 5.0 g of sulfanilic acid to about 500 mL of
again. distilled water, followed by 15 mL of concentrated
hydrochloric acid (approximately 12 mol/L), and make
Analyze the stock standard solution in triplicate using up to 1 L with water. Store at room temperature.
the method described below. Omit the addition of 2. Sodium nitrite stock (0.725 mol/L). Dissolve
ascorbic acid before the tartrate, since there is no 0.5g sodium nitrite (NaNO2) in distilled water, and
hemoglobin. Analyze the BSA diluent in duplicate as a dilute to 100 mL. Prepare fresh every 2 weeks.
test, and subtract its absorbance from that of the stock 3. Caffeine-benzoate reagent. Dissolve 56 g of
standard. Calculate the molar absorptivity of the anhydrous sodium acetate, 56 g of sodium benzoate, 1 g
azopigments at 598 nm. Acceptable limits for the molar of disodium ethylenediamine tetra acetic acid (EDTA)
absorptivity are from 74,750 to 78,380. If the molar and 37 g of caffeine in about 700 mL of distilled water.
absorptivity falls outside these limits, prepare a new Make up to 1 L. This reagent is stable at room
stock standard. temperature for at least 6 months.
4. Hydrochloric acid 0.05 mol/L.
Working Standards: Dilute the stock standard with BSA 5. Diazo reagent. Mix 20 mL of the sulfanilic
diluent to give bilirubin concentrations of 5, 10, 15, 20, acid with 0.5 mL of sodium nitrite solution. Prepare
50, and 100 mg/L. Dispense these and the stock fresh daily and store at 4C.
standards into polypropylene tubes in suitable aliquots 6. Ascorbic acid solution (40 g/L). Dissolve 200
and store at 70C. These standards are stable for at mg of ascorbic acid in 5 mL of water. Prepare fresh daily
least 6 months. and store at 4C.
7. Alkaline tartrate solution. Dissolve 75 g of
This standard is suitable for methods using caffeine or sodium hydroxide and 320 g of sodium potassium
dyphylline as accelerator but is not suitable for direct tartrate tetrahydrate in about 700 mL of water. Allow the
spectrophotometric methods or diazo methods that use solution to cool, and make up to 1 L. If the solution is
other accelerators. BSA (Cohen fraction V, 40g/L) turbid, filter. This reagent is stable for 6 months at room
solution is an acceptable diluent for the stock calibrant of temperature.
unconjugated bilirubin. 8. Serum-based calibrant and controls.
9. Bovine serum albumin (BSA) diluent (40
An alternative method for preparation of stock standard g/L). Dissolve 10.0 g of BSA in 200 mL of Tris buffer,
in protein matrix is to dissolve pure bilirubin using adjust the pH to 7.35 0.05 if necessary, and make up to
sodium hydroxide. Weigh out approximately 22 to 24 250 mL with Tris buffer. Store at 4C for 2 to 3 days or
mg of bilirubin in a tared 50-mL beaker. Add 4 mL of at 20C (in aliquots) for long-term storage. This diluent
buffered NaOH, pH 11.3. {To prepare, dilute 1 mL of 5 is required for the preparation of standards.
N NaOH to 100 mL with 0.067 M pH 7.4 Srensens 10. Sodium carbonate (0.1mol/L). Dissolve 1.06 g
phosphate buffer. [This buffer is made by mixing 41.3 of anhydrous sodium carbonate, and make up to 100 mL.
mL of 0.067 M monopotassium phosphate (9.073 g
KH2PO4/L) with 51.7 mL of 0.067 M disodium Assay
phosphate (11.87 g Na2HPO4 2H2O/L)]}. Stir gently Equipment: spectrophotometer, with band pass 8 nm,
with a glass rod, but do not form foam or bubbles. The capable of reading at 598 nm. For accurate results, the
bilirubin will dissolve completely in 2 to 4 minutes to analysis should be performed in subdued light and away
form a clear, red-orange solution. Add 40 mL of 40 g/L from windows.
bovine serum albumin solution dissolved in 0.067 M pH 1. Pipette 4.0 mL of caffeine reagent into a series
7.4 Srensens phosphate buffer and stir. An alternative of glass tubes.
to the 40 g/L albumin solution is Plasmonate (Cutter 2. Pipette 0.5 mL of sample (standard, serum, or
Laboratories, Berkeley, CA). Completely transfer the QC), mix, and allow to stand for 10 min.
solution to a 100-mL volumetric flask with the aid of 3. Add 1.0 mL of freshly prepared diazo reagent,
additional albuminpH 7.4 Srensens phosphate buffer, mix immediately, and let stand for 10 min (add
and bring to volume with this solution. the diazo reagent at timed intervals, and
observe the same interval for the addition of
Procedure: Jendrassik-Grof Method for Total alkaline tartrate and for the absorbance
Bilirubin measurement).
4. Add 0.1 mL of ascorbic acid solution and mix.
Principle 5. Add 3.0 mL of alkaline tartrate, mix
In this method, the specimen is added to a caffeine thoroughly, and let stand for 10 min before
reagent and then diazotized sulfanilic acid solution. measuring the absorbance.
After a period of incubation, ascorbic acid and alkaline 6. Sample blanks. Follow the procedure as above
tartrate are added. During the incubation period, from 1 to 5, but substitute sulfanilic acid
bilirubin reacts with the diazo reagent to form solution instead of diazo reagent.
azobilirubin. Alkalinization of the mixture produces a 7. Reagent blank. Prepare a regent blank by
blue-green solution which is measured at 598 nm. mixing 20 mL of caffeine reagent, 2.5 mL of
274
Bilirubin
water, 5 mL of diazo reagent, and 15 mL of 1. Add 0.25 mL of sample to 1.0 mL of 0.05

alkaline tartrate. A reagent blank is required to mol/L HCL, and incubate for 5 min. Add 0.5
check for photometric drift during the assay. mL of diazo reagent an incubate for 10 min.
8. Absorbance. With the reagent blank, set the 2. Add 0.1 mL of ascorbic acid, followed by 1.5
instrument to zero absorbance at 598 nm. mL of alkaline tartrate and 2.0 mL of caffeine
Measure the absorbance of all the sample reagent.
blanks and of the BSA diluent (analyzed as test) 3. Set up a sample blank exactly as the test, except
and then the absorbance of all the tests. For substitute sulfanilic acid for diazo reagent.
each test sample, subtract the absorbance of the 4. Measure the absorbance of test and blank
corresponding blank for the test absorbance. against water at 598 nm, and subtract
Correct the absorbance of the standards by absorbance of the blank from the test.
subtracting that of the BSA diluent. 5. Calculate the concentration of conjugated
9. Using the standard curve or the regression bilirubin in the sample against a standard of
equation of the standards, calculate the unconjugated bilirubin analyzed by the total
concentration of total bilirubin in each of the bilirubin method (dilute the sample with
unknowns. caffeine regent instead of HCl, and add the HCl
after the alkaline tartrate).
Calculation
Notes
Notes 1. A commercially available conjugate of bilirubin
1. Volume of samples and reagents can be reduced (Ultimate D; SmithKline-Beckman, Brea, CA) can
to half of what is described without be used to standardize this reaction.
compromising the precision of the assay. 2. The dilution of the sample with HCl for 5
2. The large excess of sulfanilic acid in relation to minutes keeps the unconjugated bilirubin from
sodium nitrite is necessary for obtaining a fast reacting.
reaction rate in the second stage of the reaction. 3. Ascorbic acid destroys the excess diazo reagent.
Otherwise, the diazo reagent will react with
Procedure for Direct Bilirubin unconjugated bilirubin when the alkaline tartrate is
Principle added.
In the direct bilirubin method, the sample is acidified 4. Hemoglobin may interfere with this assay, more
with hydrochloric acid and then mixed with diazo than in the total bilirubin method. The negative bias
reagent to produce azobilirubin. Only conjugated due to hemoglobin is caused by (1) oxidation of
bilirubin reacts with diazo reagent in the absence of conjugated bilirubin to biliverdin, which does not
accelerator caffeine-benzoate. Addition of ascorbic acid react with diazo reagent, and (2) destruction of the
stops the reaction, and after addition of alkaline tartrate azo pigment by hydrogen peroxide produced when
followed by caffeine reagent, the absorbance is hemoglobin is converted to methemoglobin.
measured at 598 nm. The tartrate reagent provides an 5. Under the conditions of the assay described
alkaline pH to produce the intense blue color of here, there is minimal reaction of the unconjugated
azobilirubin. bilirubin with the diazo reagent.
Reagents Calculation
See Total Bilirubin Procedure: Reagents. The blank-corrected absorbance of cuvette C, the
absorbance of the standard, and the concentration of the
Assay standard are used to determine the concentration of
Equipment: See the total bilirubin assay. conjugated bilirubin in the sample (see total bilirubin
procedure).
275
Blood-Gas Analysis and Oxygen Saturation

Goce Dimeski
Name: Partial pressure of oxygen (PO2), carbon dioxide (PCO2), and the pH of whole
blood
Clinical significance: Refer to Chapter 29, Acid-Base Control and Acid-Base Disorders, in the 5th

i chloride. The calomel electrode consists of a glass tube
Blood-gas analyses are usually performed to assess the containing a platinum wire that extends into a calomel
acid-base status and respiratory status of the patient. It paste of mercurous chloride (Hg2Cl2) and mercury (Hg)
includes measurement of blood pH, PCO2, and PO2 and and a solution of potassium chloride (KCl) at a constant
may include any or all of the calculated parameters of concentration. One end of the glass tube usually contains
oxygen content, oxygen saturation, total CO2, a cotton wick to keep the calomel paste in contact with
the platinum wire and allow the system to be permeated
bicarbonate, and base excess. These calculated with the KCl solution that is inside the glass tube and
parameters are discussed in Chapter 25 of Clinical surrounding it. The KCl solution is usually a saturated
Chemistry: Theory, Correlation, Analysis and are not solution, although some pH electrode systems use lower
considered further in this chapter. concentrations of KClfor example, the Radiometer
ABL uses a 20% KCl solution. The silversilver chloride
Analysis of pH electrode contains a silver wire coated with silver
Although pH is not a gas, it is commonly analyzed chloride and an ion-permeable polymer saturated with
concomitantly with PCO2 and PO2 in blood specimens KCl. The KCl block in the reference chamber allows for
so that patient acid-base status can be assessed. the concentration of Cl- to be maintained constant in the
Maintenance of the blood pH is partly dependent on the solution, ensuring the reference sensor maintains a
respiratory processes that eliminate CO2 from the blood constant electrical potential.
and take up O2. In addition, the calculation of
A liquid junction is present between the reference
bicarbonate, total CO2, and base excess requires
electrode and blood specimen being analyzed for its pH.
knowledge of the blood pH. Generally the reference electrode has a permeable
membrane through which the KCl solution inside the
The components of the blood pH-measuring system are electrode escapes, coming into contact with the
the glass electrode, reference electrode, and the liquid specimen and thus forming the junction. The KCl
junction between the two electrodes [1]. The hydrogen solution is at a high concentration so that the difference
ions in the blood sample exchange with metallic ions in in ionic composition of the blood specimen does not
the glass membrane of the glass electrode, creating a change the constant potential on the reference electrode.
potential difference across the membrane that is
proportional to the hydrogen-ion concentration in the A representation of the entire pH-measuring system is
blood specimen. A solution of constant pH inside the shown in Figure 2. The only variation in potential in the
electrode is in contact with the glass membrane. A system is at the interface of the blood specimen being
silversilver chloride wire that is connected to a analyzed and the glass electrode, so that any voltage
voltmeter lies inside the solution of constant pH and change measured by the voltmeter is attributable to the
measures the potential difference created by the pH of that blood specimen.
hydrogen-ion concentration in the blood specimen, as
shown in Figure 1. Analysis of PCO2
PCO2 is measured by isolating a glass electrode in a
The potential difference measured by the glass electrode
is assessed against the reference electrode (Figure 1). weak bicarbonate buffer and separating this buffer from
The reference electrode, which is also connected to the the blood sample by a gas-permeable membrane. The
voltmeter, is usually a calomel electrode or silversilver membrane is permeable to the carbon dioxide (CO2) in
the specimen. The CO2 diffuses into a bicarbonate buffer
i solution inside the electrode, and the following reaction
Blood-Gas Analysis and Oxygen Saturation occurs:
First edition: John E. Sherwin, Berndt B. Bruegger
CO2 + H2O H2CO3 H+ + HCO3
Methods edition: John E. Sherwin, Berndt B. Bruegger
Second edition: John E. Sherwin, Berndt B. Bruegger
Third edition: John E. Sherwin The change in hydrogen concentration in the buffer is
Fourth edition: John E. Sherwin, Christopher Tennant then measured inside the PCO2 electrode by the same
Fifth edition: Goce Dimeski type of components found in the pH electrode. A glass
276
electrode inside the PCO2 electrode develops a potential flow rate of the gas, and humidification is not a
consideration. This technique results in a positive bias of
difference because of a change in concentration of H+
about 5% to 8% for both PCO2 and PO2 as compared to
ions, while a reference electrode also inside the PCO2
the gas calibration technique, and it is important that the
electrode has a constant potential that is the standard to
results from instruments using different calibration
which the change in potential at the glass electrode is
techniques be identified. If multiple instruments are
compared. The glass electrode and reference electrode
used, the calibrations can be cross-correlated to
are connected to a voltmeter, as shown in Figure 3. The
minimize this difference.
gas-permeable membrane of the PCO2 electrode
excludes hydrogen ions in solution, so the pH of the Noninvasive In-Vivo Blood-Gas Analysis
specimen will not affect the observed pH change inside In the past few years, noninvasive means of measuring
the PCO2 electrode induced by the diffusion of the CO2 PO2 and PCO2 through the skin have been developed,
from the specimen into the electrode. evaluated, and found to be useful [5]. These
transcutaneous monitors allow continual assessment of
Analysis of PO2 the blood-gas status of a person. This is especially
The measurement of PO2 relies on the electrochemical important for monitoring neonates suffering from
measurement of the oxygen that diffuses across a gas- respiratory distress. Too low an O2 content can lead to
permeable membrane into an electrolyte buffer that is brain damage in a neonate, whereas too high a level can
within a Clark electrode [2]. The gas-permeable lead to blindness through damage of the retinal
membrane allows oxygen but not ions to cross into the capillaries by oxygen.
electrode. Inside the electrode is an electrolyte solution
consisting of potassium chloride and phosphate buffer, The transcutaneous measurement of PCO2 and PO2
or buffered potassium hydroxide, a silversilver chloride currently relies on miniaturized electrodes of the same
anode, and a platinum cathode. The cathode is held at a basic design as the electrodes of blood-gas analyzers.
constant potential with respect to the anode so that when These transcutaneous, noninvasive measurements are
oxygen diffuses into the electrode, the oxygen is reduced complicated by the presence of the skin, which acts as a
at the cathode as follows: second gas-permeable membrane.
O2 + 2H+ + 4e H2O2 + 2e 2OH There is a small amount of diffusion of gas from the
capillary bed in the skin toward the surface of the skin. If
the skin is heated so that local blood flow is maximally
The reduction of oxygen causes a current to flow increased, the PO2 at the surface of the skin correlates
between the cathode and anode. The current is measured
well with the arterial PO2. Temperatures of 42C to 43C
and is proportional to the amount of oxygen diffusing
into the electrode from the analyzed specimen. Figure 4 have been found to be best for maximal blood flow
illustrates the components of the PO2 electrode. The while minimizing the risk of burns, so a major feature of
the transcutaneous PO2 (tcPO2) electrode is a heating
electrolyte solution buffers against the accumulation of
element maintaining the 42C to 43C temperature. The
peroxide (H O ) and hydroxide (OH), as well as
2 2 tcPO2 electrode uses the same concept for measuring
serving as the conduction medium for the diffusing PO2 as the Clark polarographic PO2 electrode described
oxygen and the current. In contrast to the pH and PCO2
earlier. A membrane, through which PO2 diffuses, lies
electrodes, the oxygen electrode is a destructive,
nonequilibrium electrode. This has led to controversy between the skin and the electrode, and a heating device
concerning the most accurate way to calibrate a PO2 is present inside the electrode (Figure 5). A sealing ring
around the outer edge of the electrode needs to be
electrode [3,4]. Calibration can be achieved either with a present to prevent nonregulated diffusion of atmospheric
flowing gas or a single equilibrated liquid sample. oxygen to the electrode. Because the PO2 electrode
The majority of the older models of blood-gas analyzers consumes oxygen during measurement, the area of the
use certified gases saturated with water for calibration. cathode is kept as small as possible, minimizing O2
This technique has the advantage of providing a constant consumption by the electrode so that measurements will
and highly reproducible calibration mixture. correlate with arterial PO2 values.
Nevertheless, it suffers from the fact that it is a gas
rather than a liquid and therefore is not equivalent to the The transcutaneous PCO2 (tcPCO2) electrode is based on
patient samples being analyzed. It is essential that the the same principles as the in-vitro PCO2 electrode
gas-flow rate through the measuring chamber be strictly
controlled. Variations will result in electrode cooling and discussed previously, except that a thermostatted internal
incomplete humidification, both of which result in heater has been added. The tcPO2 electrode is sealed
significant calibration errors. The new-model analyzers from the atmosphere so that only CO2 from the skin is
use the second calibration technique, tonometered (gas- detected. The skin is heated by the electrode for maximal
buffered) liquid for calibration. The advantages of this capillary flow. However, increased temperature also
technique are that it is similar to the unknown samples increases skin metabolism, which increases the PCO2.
being analyzed, the calibration is independent of the
277
Additionally, the tcPCO2 electrode does not consume the Equation 3: The Siemens (formerly Bayer) analyzers use
CO2 as the tcPCO2 electrode consumed O2, so the PCO2 the relationship described by Kelman [8] and Thomas
[9]:
is higher at the electrode than at the area of the capillary
closest to the surface of the skin. The tcPCO2 electrode
system uses a correction factor to compensate for this
increase in PCO2 resulting from the electrode
arrangement. The tcPCO2 electrode uses two gas
mixtures to calibrate CO2, in contrast to the tcPO2
electrode which uses surrounding atmospheric oxygen. N = pO2 10[0.48(pH(37) -7.4) 0.0013BE(B)]
In-vivo pH monitoring is being developed but at this

time is still invasive. A pH electrode probe can be Equation 4: O2 saturation = ((O2Hb/(O2Hb + HHb))
100
inserted into muscle tissue to measure the interstitial
fluid pH. Some assumptions in using the data derived are
that interstitial fluid pH reflects the status of the blood
pH and that the insertion of the pH electrode probe does
not cause local tissue changes that alter the pH.
Analysis of Oxygen Saturation (sO2) The O2 saturation is often referred to as the fractional
The oxygen saturation is a ratio of the amount of sO2 (SaO2), since it is determined from the two
hemoglobin (Hb) bound to oxygen to the total amount of hemoglobin fractions. This is the most accurate method
hemoglobin able to bind oxygen, expressed as for sO2 estimation because it is a direct measure of the
percentage. Oxygen saturation can be accurately parameters that affect sO2. CO-oximetry is considered
calculated by use of equations 1 to 3 below, directly the gold standard for oxygen saturation measurement.
determined from the measured oxyhemoglobin (O2Hb) The benchtop blood-gas instruments with oximeters
and reduced hemoglobin (HHb) (equation 4) fractions or calculate oxygen saturation using this formula.
determined by pulse oximetry.
Pulse oximetry is a third method for sO2 estimation. The
Estimated oxygen saturation does not account for saturation calculated from pulse oximetry is a functional
variations in 2,3 DPG levels, or the presence of saturation, SpO2. It is the amount of HbO2 compared
dysfunctional hemoglobins (COHb, MetHb). These with the sum of O2Hb and HHb only. These systems do
equations assume no presence of dysfunctional not measure COHb and metHb, so the SpO2 is falsely
hemoglobins. If any of these conditions change, the elevated in the presence of dysfunctional hemoglobins.
estimated sO2 will be inaccurate. Oxygen saturation can In the presence of normal SpO2 with cyanosis, CO-
be calculated from the measured parameters PO2 and pH oximetry must be used to determine the saturation status
on the basis of standard oxygen-dissociation curves [6]. and presence of any dysfunctional hemoglobin.
Equation 1: sO2 = {[(PO2 + 150 PO2)-1 23 400] +1}-1 Calibration of Gas Electrodes
Gases used to calibrate blood-gas analyzers are bubbled
The Severinghaus [7] equation assumes the following through humidifiers at 37C to saturate the gases with
parameters are normal: pH = 7.40, PCO2 = 40 mm Hg water vapor. The reason for saturating the gases with
and 2,3-diphosphoglycerate = 5 mmol/L. water is to eliminate the problem of maintaining a totally
dry gas for calibrating the blood-gas instruments.
Equation 2: The Radiometer ABL uses the following Some residual moisture will remain in the measuring
Hill equation to determine oxygen saturation from the chamber and contaminate a dry gas. The pressure of
measured pH and O2: water vapor will reduce the partial pressures of the other
gases in the mixture and introduce a variable error. For
instance, at 37C, a water-saturated gas will have a
partial pressure of 47 mm Hg, contributed by the water
vapor. So to produce a reproducible gas mixture of
known partial pressure of O2 and CO2, calibration
in which, requires a fixed partial pressure of water vapor. Since the
specimen to be analyzed is aqueous, this is best achieved
with a water-saturated gas. The cartridge-based
analyzers contain all the calibrants, including gases, in
Found in the equation is a Bohr factor of 0.48 and a the single cartridge. This eliminates the need to have
standardized P50 of 26.6 mm Hg. separate gas bottles, and the gases are water saturated.
Modern Instruments
In the past, instrument design led to a variety of required
sample volumes for blood-gas analysis. All the modern,
278
commonly used blood-gas analyzers have the capability electrode technology as conventional instruments and
of analyzing microsamples of < 100 L to measure pH, are designed so that non-laboratory staff can use them.
PO2, and PCO2 and usually the same volume to measure Most of these have been designated as moderately
all of the on-board analytes. The other range of analytes complex under the Clinical Laboratory Improvement
available on many blood-gas instruments include Amendments of 1988 (CLIA 88) rules. These new
sodium, potassium, chloride, ionized calcium, ionized instruments utilize microelectrode technology. This
magnesium, lithium, lactic acid, glucose, urea, and permits individual electrode packets to contain not only
creatinine (all by analyte-specific electrodes) and the analytic electrode but also the reference electrodes
oximeters to measure hemoglobin fractions. The actual and calibration solutions. The calibration and quality
analyses that are available depend on the electrode control of these electrode assemblies is similar to that
configuration adapted for each instrument. Most used to measure electrolytes using dry-film technology.
instruments currently available automatically aspirate the
sample, thus reducing errors resulting from manual A new technique for measurement of blood gases at the
introduction of the sample. Therefore in total, < 1.0 mL bedside has recently been described. It is based on dry
of whole-blood sample is required in a syringe to chemistry and optical technology [10]. The measurement
complete all the analytes and allow for dead volume. principles of this technology include three-wavelength
infrared spectroscopy of dissolved CO for
2
Most blood-gas instruments are now microprocessor determination of PCO , of O induced changes in the
controlled so that calibration occurs automatically at 2 2
preset intervals, and both the measured parameters and decay time of phosphorescence, and of the absorbance
the calculated parameters are displayed when analysis is change spectral change of an azo-dye color indicator for
complete. Maintenance has been greatly simplified by pH.
the use of microprocessors to monitor all system
parameters and indicate potential malfunctions. The
basic principles of electrode operation are unchanged Reference and Preferred Methods
from the original equipment, and all available Although there are no available reference methods for
instruments use equivalent electrodes. Nonetheless, the pH, PO2, or PCO2 estimation, currently available blood-
electrode configuration within the sample compartment gas analyzers provide acceptable methods for blood-gas
can significantly affect both the sample volume and the estimation.
analysis time. Thus there can be statistical differences in
measured parameters between instruments produced by However, attention has now been focused on techniques
different manufacturers, as well as between different that eliminate the need to collect a blood specimen, that
models produced by the same manufacturer. A review of is, the use of noninvasive instruments that measure PO2
the 2007 blood-gas proficiency surveys of the College of and PCO2 [11]. No doubt if noninvasive methods can
American Pathologists shows similar precisions for the provide pH measurement and improved correlation with
analyzers included in the survey. The pH coefficients of conventional blood-gas analyzer methods and cause no
variation (CVs) across the different concentration levels patient harm (e.g., skin burns) [12], they would become
for about 95% of the analyzers are 0.1%. The remaining the preferred methods.
5% of the analyzers have CVs between 0.1 and 0.5%.
The PCO2 CVs are slightly higher at the low end (PCO2
< 20 mm Hg), ~ 5%, and decrease down to ~ 3% with Specimen
values at the high end (PCO2 ~ 60 mm Hg). The greatest The skill with which arterial specimens need to be drawn
variation in both medians and CVs is seen with the PO2. and the potential harm to the patient limits the personnel
For samples with PO2 values < 100 mm Hg, the CV for who can collect such specimens. Thus the capillary
PO2 with some analyzers is as high as 11.2%. Samples specimen is frequently utilized as a compromise,
with PO2 values > 100 mm Hg had much lower CVs, providing an arterialized specimen that is easily
averaging < 5%the predominant reason being the collected. A capillary specimen can be drawn by a
quality of samples used in the survey (aqueous and competent phlebotomist under conditions that produce
poorly buffered). Thus even small exposure to air during pH, PCO2, and PO2 values closely paralleling arterial
the sample aspiration/presentation process to the blood. Conditions to arterialize capillary blood involve
analyzers has the potential to expose the sample to air warming the limb for several minutes at about 45C
and significantly change the PO2 concentration. immediately before the specimen is collected. The
Therefore, the differences seen can be ascribed to sample warming dilates the capillaries, and the increased blood
aspiration/presentation technique, the different methods flow in that area decreases the accumulation of
of gas calibration, or the juxtaposition of the electrodes metabolic products of tissue respiration and ensures
to each other. more blood of an arterial than a venous composition.
Alternatively, a histamine cream can be used to produce
Blood-gas analyzers are also available for use in point- local hyperemia so that an arterialized specimen can be
of-care sites (e.g., surgery sites, intensive care, collected.
emergency departments, etc.), as bench tops, the
traditional blood-gas analyzer, or as portable or hand- Blood-gas measurements must be performed on whole
held systems with disposable single-use or multiple-use blood that is unclotted and well mixed. An anticoagulant
cartridges. These instruments use the same fundamental
279
is used to inactivate the clotting mechanisms so that the 13.7 mm Hg [16]. Glass syringes have been preferred in
sample remains unclotted in the syringe. Whole blood is the past for the following reasons:
collected in heparinized capillary tubes which may 1. The barrel has minimal friction with the syringe
contain a small metal mixing wire (flea), which is added wall, and the pulsating arterial pressure is
to facilitate mixing with the magnet provided. These are clearly visible as the blood fills the syringe, and
then sealed at one or both ends. The capillaries are thus the nature of the sample is clearly
placed into a labeled test tube or plastic bag and identified to the clinician.
delivered to the laboratory in crushed ice or wrapped in a 2. There is seldom a need to pull back on the
cold pack to prevent deterioration of the sample. Since barrel, a maneuver that can cause air bubbles to
excess heparin can alter the pH of the specimen, sample enter the sample around the barrel.
tubes containing heparin should be properly filled to 3. Small air bubbles adhere tenaciously to the side
avoid heparin effect. of a plastic syringe, making it difficult to expel
the air from the samples.
The time necessary for completion of the test should not Newer plastic syringes overcome many of these
exceed 15 minutes at room temperature (~21C). This concerns and when properly used are as good as glass
includes drawing of sample, delivery to the laboratory, syringes.
log-in, analysis, and reporting of results. Samples for
blood-gas analysis are stable for a maximum of 45 Temperature During Transport and Storage
minutes if properly stored in crushed ice [13]. If this Blood is a living tissue in which oxygen continues to be
maximum time limit is adhered to, no difference is consumed and carbon dioxide continues to be produced,
observed between specimens collected in plastic and even after the blood is drawn into a syringe. Table 1
glass syringes. Samples placed in crushed ice or ice shows the changes in blood-gas parameters at different
slush (ice/water) must be properly sealed and clean of temperatures. A sample at 37C shows changes in all
any exterior blood, particularly around the base of the three parameters at a faster rate than a sample held at
cap, to prevent leakage and contamination of the ice 4C. The Clinical and Laboratory Standards Institute
slush. Ideally, samples should be placed in a biohazard (CLSI, formerly NCCLS)-approved guideline # C27-A
bag before placing in crushed ice. Alternatively, samples 1993 cautions against placing plastic blood-gas syringes
placed in biohazard bags can be transported wrapped in in ice/water as this will tend to raise the PO2 by drawing
cooled (~4C) gel packs. O2 through the plastic wall of the syringe. Therefore, the
time between placing the specimen in ice/water and the
Syringes have been developed that allow air from the
analysis should be as short as possible in order to
sample to be expelled through the cap. The cap contains
maintain the O2 content of the sample. If the sample will
absorbent material which is the same as that used in
autoventing syringes. Once the blood wets the material be brought to the work station within a few minutes of
in the cap, it is air sealed. Such caps prevent the need to draw, no ice is needed.
expel air, minimizing exposure to blood by collectors
and preventing soiling/contaminating of the bench or Pneumatic Tube Transport
other surfaces. Other syringes contain a metal ball The transport of samples for blood-gas analysis by
(Radiometer Pico Syringes for the ABL 800FLEX pneumatic tube systems (PTS) does not alter PCO2 and
system) in the barrel. The syringe is placed on the pH values but does have significant effects on PO2
system platform, which contains a magnet to spin the values. The PO2 changes are due to the presence of
ball to create a uniformly mixed sample. microbubbles in the sample combined with pressure
effects from the delivery system. Samples for blood-gas
Glass Versus Plastic Syringes analysis should be free of air bubbles, and if transported
Early studies indicated that plastic substances may via PTS it is best to use a pressure-sealed container to
absorb so much oxygen that the use of plastic syringes avoid artifacts in the PO2 [17,18].
for blood-gas samples would be ill advised. More recent
studies have not substantiated these observations [14]. In Interferences
essence, pH and PCO2 values are not affected, whereas Oxalates, EDTA, and citrates are not acceptable for
blood-gas samples because they significantly alter the
PO2 values in excess of 400 mm Hg drop more rapidly
blood sample [19]. Heparin is the anticoagulant used in
for samples in plastic than in glass syringes. The rate of syringes. When blood-gas analyzers were only used to
O2 diffusion is also dependent on the temperature of the measure blood-gas parameters, the heparin
sample and the time required before analysis is concentration was high, > 100 IU/mL. With the
completed, with diffusion increasing as both variables introduction of electrolytes, namely ionized calcium,
increase. Smeenk et al. [15] studied patients receiving heparin caused ionized calcium binding and false
100% oxygen with initial PO2 of ~ 500 mm Hg and decreases. Syringe manufacturers
found PO2 decreased after 30 min in both an ice bath responded by either decreasing the heparin concentration
(~11.2 mm Hg) and at room temperature (~40.5 mm Hg) down to ~ 7 IU/mL, which increased the frequency of
[15]. Arterialized samples with PO2 of ~ 96 mm Hg clot formation, or titrated or electrolyte-balanced the
stored in plastic syringes for 30 minutes at 0C to 4C or heparin to minimize ionized calcium binding while
at room temperature both increased the PO2 by 11.9 to keeping the heparin concentration at high levels, ~ 35
IU/mL, to prevent clot formation. However, too much
280
heparin may affect several parameters. The pH of See Chapter 29, Acid-Base Control and Acid-Base
sodium heparin solutions is approximately 7.0, and the Disorders, in Clinical Chemistry: Theory, Analysis,
PCO2 and PO2 of heparin approach room-air values. It Correlation, 5th edition.
has been shown that 0.05 mL of sodium heparin (1000
units/mL or 10 mg/mL) will adequately anticoagulate 1 Blood-Gas Performance Goals
mL of blood, whereas up to 0.1 mL per 1 mL of blood Under CLIA 88 regulations, blood-gas analysis must
will not affect pH, PCO2, or PO2 values. If more than 50 include at least one sample of control material each
eight hours and one sample of calibration or control
L of solution remains in a 1-mL syringe and a minimal material each time patients [samples] are tested unless
specimen (0.2 mL) is collected, a negative error of about automated calibration internally verifies calibration at
5 mm Hg of PCO2 is observed. When a 5-mL syringe is least every 30 minutes (Federal Register, Vol. 57[40],
washed with sodium heparin and then ejected, the dead February 28, 1992, p.7168). The same publication states
space of the syringe will contain approximately 0.15 to (p. 7158) that CLIA rules have also established
0.25 mL of sodium heparin. Thus 2 to 4 mL of blood minimum criteria for acceptable performance for blood-
will theoretically contain at least 0.05 mL of heparin per gas analyses on proficiency sample testing. These are as
1 mL of blood. follows:
PCO2Target value + 5 mm Hg or + 8% (whichever is
Room air contains a PCO2 of essentially zero and a PO2 greater)
of approximately 150 mm Hg. Air bubbles that mix with PO2 Target value + 3 S.D.
a blood sample will result in gas equilibration between pH Target value + 0.04
the air and the blood. Air bubbles may thus lower the
PCO2 values of the blood sample and cause the PO2 to References
approach 150 mm Hg. The greater the amount of air 1 Clark LC. Monitor and control of blood and
mixed with a blood sample, the greater the error. The tissue oxygen tensions. Trans Am Soc Artif
syringe must be immediately sealed with a cap after the Intern Organs 1956;2:41-48.
sample is obtained. It is recommended that any sample 2 Evans NTS, Naylor PFD. The systemic oxygen
obtained with more than minor air bubbles should have supply to the surface of human skin. Respir
the PO2 results carefully interpreted and either withheld Physiol 1967;2:21-27.
or discarded. As a general guide, if the PO2 value is > 3 Fleischer WR, Hartmann AE. Blood Gases and
110 mm Hg and the patient is not oxygen supplemented, Their Measurement. Parts I and II. Chicago:
air contamination should be considered. American Society of Clinical Pathologists; 1976.
4 Thimnig R. A Practical Course in the Techniques,
Blood-Gas Reference Interval (at 37C) Applications, and Maintenance of the Radiometer
The unit commonly used to express partial pressure of Acid-Base Laboratory. Cleveland: The London
blood gases is millimeters of mercury (mm Hg) or its Co; 1977.
equivalent, the torr. However, in the International 5 Huch R, Huch A. Continuous Transcutaneous
System of Units, the kilopascal (kPa) is the Blood Gas Monitoring. New York: Marcel
recommended unit of expression, though it is not yet Dekker; 1983.
widely used. One millimeter of mercury, or 1 torr, is 6 Rem J, Siggaard-Andersen O, Norgaard-Pedersen
equal to 0.133 kPa, or divide mm Hg by 7.5 to convert to B, Storensen S. Hemoglobin pigments:
kPa. photometer for oxygen saturation,
carboxyhemoglobin, and methemoglobin in
The three types of blood specimens assayed for blood- capillary blood. Clin Chim Acta 1972;42:101-8.
gas content are arterial, venous, and capillary. The above 7 Severinghaus JW. Simple, accurate equations for
outline compares the normal ranges for arterial and human blood O2 dissociation computation. J Appl
venous blood-gas parameters. Arterial blood is not so Physiol 1979;46:599-602.
affected by the metabolic activity of the area being 8 Kelman GR. Digital computer subroutine for the
sampled as venous blood is. The slightly lower pH and conversion of oxygen tension into saturation. J
PO2 values and the slightly higher PCO2 in the venous Appl Physiol 1966;21:1375-6.
blood reflect the changes induced by tissue metabolic 9 Thomas LJ. Algorithm for selected blood acid-
activity, which releases acids and uses oxygen. Arterial base and blood gas calculations. J Appl Physiol
blood specimens reflect the state of pulmonary activity, 1972;33:154-8.
which introduces oxygen to the blood and releases 10 Boalth N, Wandrup J, Larsson L, Frischauf PA,
carbon dioxide from the blood to the air spaces of the Lundsgaard FC, Andersen WL et al. Blood gases
lungs. Although the usually small differences in pH and oximetry: calibration-free new dry-chemistry
between arterial and venous blood specimens allow the and optical technology for near-patient testing.
venous specimens to be of value in interpreting the acid- Clin Chim Acta 2001;307:225-233.
base status of the patient, arterial blood is the specimen 11 Huch R, Lubbers DW, Huch A. Reliability of
of choice to understand the dynamics of the bodys arterial PO2 in newborn infants. Arch Dis Child
respiratory system. 1974;49:213-218.
12 Parker SM, Gibson G J. Evaluation of a
Interpretation transcutaneous carbon dioxide monitor
281
(TOSCA) in adult patients in routine 19 Shapiro BA, Harrison RA, Walton JR. In:
respiratory practice. Respir Med 2007;101:261- Clinical Application of Blood Gases. 2nd ed.
264. Chicago: Year Book Medical Publishers; 1977.
13 Mller-Plathe O, Heyduck S. Stability of blood 20 Fleisher WR, Gambino SR. Blood pH, pCO2, and
gases, electrolytes and haemoglobin in Oxygen Saturation. Chicago: American Society
heparinized whole blood samples: influence of of Clinical Pathologists; 1976.
the type of syringe. Eur J Clin Chem Clin 21 Lehmann H, Huntsman RG. Mans
Biochem 1992;30:349-55. Haemoglobins. Philadelphia: Lippincott; 1974.
14 Evers W, Racz GB. A comparative study of 22 Bunn HF, Forget BG, Ranney HM. Human
plastic (polypropylene) and glass syringes in Hemoglobins. Philadelphia: Saunders; 1977.
blood gas analysis. Anesth Analg 1972;50:92-7. 23 Siggaard-Andersen O, Wimberley PD, Gothgen I,
15 Smeenk F, Janssen J, Arends B, Harff G, van den Siggaard-Andersen M. A mathematical model of
Bosch J, Schonberger J, Postmus PE. Effects of the hemoglobin-oxygen dissociation curve of
four different methods of sampling arterial blood human blood and of the oxygen partial pressure
and storage time on gas tensions and shunt as a function of temperature. Clin Chem
calculation in the 100% oxygen test. Eur Respir J 1984;30:1646-51.
1997;10:910913. 24 Van Holde KE. Physical Biochemistry.
16 Knowles TP, Mullin RA, Hunter JA, Douce FH. Englewood Cliffs, NJ: Prentice Hall; 1971.
Effects of syringe material, sample storage time, 25 Bellingham AJ. The red cell in adaptation to
and temperature on blood gas and oxygen anaemic hypoxia. Clin Haematol 1974;3:577.
saturation in arterialized human blood samples. 26 Nelson MG, Savage GA, Cooke PJ, Lappin TR.
Respir Care 2006;51:732-6. Determination of the oxygen dissociation curve
17 Zaman Z, Demedts M. Blood gas analysis: POCT and P50 of whole blood. Am J Clin Pathol
versus central laboratory on samples sent by a
1981;75:395-9.
pneumatic tube system. Clin Chim Acta
2001;307:101-106.
18 Collinson PO, John CM, Gaze DC, Ferrigan LF,
Cramp DG. Changes in blood gas samples
produced by a pneumatic tube system. J Clin Path
2002;55:105-107.
Table
Table 1: Changes in Blood-Gas Parameters at Different Temperatures*
Parameter 37C 4C
pH 0.01/10 min 0.001/10 min
PCO2 1 mm Hg/10 min 0.1 mm Hg/10 min
PO2 33 mm Hg/10 min 3 mm Hg/10 min
*Approximate changes with time and temperature after sample is drawn into syringe.
A temperature of 37C implies that the blood remains at body temperature in the syringe.
Figures
Figure 1
282
Schema of a pH electrode and calomel reference electrode. A potential develops across pH-sensitive
glass membrane of pH electrode because of difference in pH of outer sample and inner electrode
buffer solution. Calomel electrode is in electrical contact with sample and pH electrode through a KCl
salt bridge and acts as a reference against the potential developed at pH electrode.
Figure 2
Schema of contact sequence of specimen and glass electrode and reference electrode. One area of
variable potential is at area of contact between specimen and pH-sensitive glass and is dependent on
pH of specimen.
283
Figure 3
Schema of PCO2 electrode. CO2 diffuses across the outer membrane, which is not permeable to

HCO3 or H+, into a layer of bicarbonate solution. Reaction of diffused CO2 and H2O to form
carbonic acid changes the pH of bicarbonate solution, which is detected by inner pH electrode.
284
Figure 4
Schema of Clark polarographic electrode. A constant voltage is generated between platinum cathode
and silver anode. Oxygen molecules that diffuse past outer membrane into a layer of electrolyte
solution in contact with platinum cathode are reduced at cathode, producing a current that can be
measured and related to amount of diffusing oxygen.
285
Figure 5
Schema of a transcutaneous PO2 electrode. Oxygen diffuses from a heated area of skin across oxygen-
permeable membrane into electrolyte solution, which is in contact with platinum cathode and silver
anode that are generating a constant voltage. Oxygen molecules are reduced at cathode, producing a
current which can be measured and related to amount of diffusing oxygen.
286
Figure 6
Oxygen-dissociation curve in middle is developed under following conditions: pH is 7.40, PCO2 is 40

mm Hg, and temperature is 37C. P50 is derived from that curve. Also indicated are factors that shift
curve to right and left.
P50 Oxygen-Dissociation Curve torr for adults and about 21 torr for newborns, with
standard conditions of pH 7.4, PCO2 40 torr, and 37C.
The term P50 is defined as the partial pressure of
oxygen (PO2) at which hemoglobin is 50% saturated The relationship between PO2 and O2 saturation follows
[3,19,20]. A graph of the percent oxygen saturation a sigmoidal curve (Figure 6). The curve can be
versus PO2 is referred to as the Hill plot or as the expressed in mathematical terms, and this expression is
the Hill equation [23,24]:
oxygen-dissociation curve (Figure 6). A shift of the
curve to the right represents a decrease in the affinity of
hemoglobin for oxygen and an increase in P50. The P50
value increases because a greater O2 partial pressure is
required to saturate 50% of the hemoglobin. This shift to
the right can be caused by an increase in PCO2,
temperature, or 2,3-diphosphoglycerate (DPG)
concentration or by a decrease in pH. The effect of pH
and PCO2 on the oxygen affinity of hemoglobin is in which K is the equilibrium constant and n equals 2.6.
known as the Bohr effect. A shift of the curve to the left The n factor, sometimes termed the interaction
represents an increase in the affinity of hemoglobin for coefficient, derives from the cooperative oxygen-binding
oxygen and a decrease in P50; that is, a smaller O2 partial process of the four sites on the hemoglobin molecules.
Because one or more molecules of O2 are bound to a
pressure is required to saturate 50% of the hemoglobin.
The shift to the left can be caused by an increase in pH hemoglobin molecule, a greater affinity for oxygen
or a decrease in PCO2, temperature, or DPG develops at the other sites on the hemoglobin molecule.
If there were no cooperative binding, n would be equal
concentration. Varying the DPG from 1 to 24 mmol/L to 1, and the oxygen-dissociation curve would not be
changes the P50 from 17 to 44 torr. The change in P50 is sigmoidal. This is the case for myoglobin. The value of n
affected more by acidotic conditions than by alkalotic has been derived empirically.
conditions. When the temperature is increased from
37C to 41C, the P50 is raised from 27 to 34 torr. The The Hill equation can be rearranged as follows:
influence of PCO2 is relatively small. Some abnormal
hemoglobins have been found to have an increased
affinity for oxygen, with decreased P50 values ranging
from 12 to 18 torr. Polycythemia generally develops
with this condition to compensate for the reduced
amount of oxygen reaching the tissues. Other abnormal
hemoglobins have increased P50 values, some reaching
40 torr, and a mild anemia is often associated with this
condition [21,22]. The normal P50 is approximately 27
287
dissociation curve is constructed. The P50 value is

determined directly from the curve. The different
instruments use different procedures to handle the
control of temperature, pH, and PCO2 of the blood
sample and the measurement of pH, PCO2, and
carboxyhemoglobin. The measurement of P50 is not
ordinarily performed in routine clinical laboratories
because of its limited clinical utility. The predominant
use of P50 is in (1) indicating the presence of abnormal
hemoglobin that affects the oxygen transport mechanism
and (2) as an indirect measure of the 2,3 DPG
concentration. It is also useful to indicate changes in pH,
PCO2, and temperature [23].
Analyzing a specimen for O2 saturation and PO2 and

substituting those parameters into equation 6, one can
solve for K. Once K is known, the PO2 at which O2
saturation is 50% can also be determined from equation
6, by substituting in the K factor and 50% for oxygen
saturation. In using equation 6, one assumption that is
made is that n is independent of other variables.
However, n is not always constant, and this difference is
particularly evident with hemoglobin mutants [25].
A number of methods are available that do not depend

on making the assumption of the constancy of n [26].
These are important for analyzing blood from patients
with multiple hemoglobin types, such as newborns with
hemoglobins F and A. One method involves
equilibrating aliquots of a blood sample with gases of
different oxygen composition, as can be done with
tonometry. The aliquots of blood are analyzed for PO2,
pH, PCO2, and oxygen saturation. After correcting the
pH values to a pH of 7.4 and the PCO2 values to 40 torr,
one plots the oxygen-saturation values and
corresponding corrected PO2 values as in Figure 6 to
form an oxygen-dissociation curve. The corrections
allow comparison with normal oxygen-dissociation
curves. From the curve, a P50 value can be determined.
Another method, which has been automated on a number

of instruments, involves continuous monitoring of blood-
gas parameters while O2 content of the blood sample is
changed. Although the process and measured parameters
will differ on the different instruments available, a
general description can be presented. An aliquot of blood
is first deoxygenated and is then subjected to increasing
partial pressures of oxygen. During the diffusion
process, the PO2 and oxygen saturation can be measured.
As more oxygen diffuses into the blood sample, the
parameters are measured continually, and an oxygen-
288
C-Reactive Protein (CRP)

Odette Youdell
Name: C-reactive protein, CRP, acute-phase reactant

Clinical significance: Refer to Chapter 36, Cardiac and Muscle Disease, and to Chapter 37, Coronary artery
disease: lipid metabolism, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation

Principles of Analysis and Current Usagei of inflammation produced a positive result. As a

In 1930, Tillet and Francis [1] described a substance in consequence, the use of CRP for monitoring disease
the sera of patients with pneumococcal pneumonia activity declined significantly, being replaced by tests
which closely paralleled the clinical course of the for other acute-phase proteins and measurement of the
infection. It was noted that in the presence of calcium erythrocyte sedimentation rate (ESR).
ion, this protein could form a complex with
pneumococcal Cpolysaccharide in a flocculation Early methods for CRP analysis included radial
reaction [2]. It was from this reaction that the term C- immunodiffusion and latex agglutination (Table 1,
reactive protein (CRP) was derived. In the decades to Method 1) and have been described as insensitive,
follow, CRP has become one of the most popular clinical semiquantitative, and subjective in assessment [10].
laboratory tests for the evaluation of patients with Further advances in assay and antibody development
various inflammatory disorders [3]. have allowed for more sensitive and quantitative
methods.
CRP is synthesized in the liver and is under the direct
transcriptional control of interleukin 6 (IL-6) and Quantitative Assays
indirect control of IL-1, tumor necrosis factor and Since the late 1970s, purification of CRP by affinity
other cytokines [4]. It is a 1.18 kDa protein which chromatography enabled the development of antibodies
consists of five identical, noncovalently bonded that could be used for quantitative methods [11].
subunits, each approximately 23 kDa and consisting of
206 amino acids. It has been shown to bind to bacteria The major methods reported in a 2007 College of
and tissue membranes and enhance phagocytosis. In American Pathologists (CAP) Participation Survey
addition, CRP can activate the classical complement (1490 participants across 19 different platforms) were
pathway [5]. immunoturbidimetry, (77%), nephelometry (13%),
Today, circulating CRP concentrations previously enzyme immunoassay (9%), and chemiluminescence
considered normal have been found to provide (<1%).
valuable information with regard to cardiovascular risk
stratification for future coronary events [6-9]. As a result Immunoturbidimetric procedures for CRP are currently
of this new application, high-sensitivity CRP (hsCRP) the most popular method for CRP analysis (Table 1,
assays have been developed to quantify concentrations Method 2). Along with immunonephelometric methods
below the limit of conventional assays. (Table 1, Method 3) they allow for CRP analysis which
is sensitive (0.4 mg/L), automated, and rapid [4]. Results
Qualitative/Semiquantitative Assays for the nephelometric and turbidimetric methods
Until the late 1970s, CRP was measured using correlate fairly well with other techniques, with intra-
qualitative or semiquantitative methods. These methods assay biases generally being less than 20% [12].
required precipitation or agglutination of latex particles
by CRP to form large, visible aggregates which were Enzyme immunoassay (Table 1, Method 4) is another
then graded on a scale of +1 to +4. This limited the use quantitative method for CRP. Like the other
of CRP as a differential diagnostic test in that any degree immunochemical methods, enzyme immunoassay assays
for CRP are automated, rapid, and show good sensitivity
and specificity.
i
C-Reactive Protein
Previous and current authors of this method: Other assay techniques for quantitative CRP
First edition: Not done measurement which may still be used by some
Methods edition: Not done laboratories include immunofluorescence assays, latex
Second edition: Not done agglutination (Table 1, Method 5), radial
Third edition: Not done immunodiffusion (RID), and radioimmunoassay.
Fifth edition: Odette Youdell
289
High-Sensitivity/Ultrasensitive Quantitative CRP high sensitivity and a wide measuring range. A recent
Assays study evaluating concentrations spanning 0.15 and 290
The concentration of CRP measured to assess mg/L between two particle-enhanced immuno-
cardiovascular risk is much lower than that measured in turbidimetry based assays concluded that both needed
acute inflammation. Manufacturers have responded to improvements to meet high-sensitivity criteria [19].
this challenge by developing high-sensitivity or ultra-
sensitive CRP methods. The most common methods Specimen
utilize particle-enhanced turbidimetry or nephelometry. CRP is a robust analyte. Common laboratory variables
With these techniques, monoclonal or polyclonal such as specimen type, delays in specimen processing,
antibodies are adsorbed onto latex particles in order to and storage temperature have little effect on
enhance the measurement signal at low concentrations. concentration. Many studies report little or no difference
In a comparison study of nine such assays, the limit of between serum and plasma values, with a negative bias
detection ranged from 0.02 to 0.32 mg/L [13]. up to 16% with EDTA blood [10,17,20,21]. Specimen-
Given that CRP results are interpreted in the context of processing delays of up to 6 hours and varying storage
internationally established decision limits, ideally all conditions ranging from 70C to 21C for up to 3
commercial assays should give equivalent results. weeks had little effect on CRP concentration [22-24]. It
Quality-control surveys in Western Europe and the is also stable for up to 7 freeze/thaw cycles [24,25].
United States during the late 1980s and early 1990s
showed that reported concentration for some proteins No apparent difference has been described between
varied by as much as 100%, depending on the calibrator fasting and nonfasting specimens [3]. In a study of 13
used [14]. The use of a single international reference healthy subjects with hourly samples over a period of 24
material by all manufacturers should reduce the hours, no diurnal variation was reported [26]. However,
variability to a large degree. in a study looking at diurnal, seasonal, and blood-
processing patterns, Rudnicka et al. [27] reported both a
There are two widely recognized standards for CRP. The diurnal pattern and borderline evidence of seasonality
first is the World Health Organization (WHO) 1st (higher in winter months). They concluded, however,
International Reference Preparation for C-Reactive that the combined effects of variation on CRP, was less
Protein (85/506) introduced in 1986, with a CRP than 1%.
concentration of 98 mg/L, and the second is the Certified
Reference Material 470 (CRM 470) introduced in 1993- Serum and CSF are the matrices primarily used in
94, with a CRP concentration of 39.2 mg/L. The second diagnosis. Studies suggest that consumption of CRP in
reference material is sometimes also referred to as the CSF may result in falsely decreased values [28]. CRP
Reference Preparation for Proteins in Human Serum has also been detected in synovial, amniotic, pleural,
(RPPHS). CRP values in CRM 470 were assigned from ascitic, and blister fluids [10]. As CRP does not cross the
WHO IRP 85/506 and so are directly traceable to the placental membrane, a raised CRP in a newborn whose
WHO material [14,15]. mother has presented with prolonged rupture of amniotic
membranes and/or amnionitis should be evaluated for
A number of comparison studies [13,16-18] neonatal bacterial infection [29].
demonstrated significant differences between hsCRP
methods. Since the target value 39.2 is much higher than Interferences
nominated decision limits used for cardiovascular risk An interference study undertaken on three automated
stratification, the reliability of using a diluted form of the hsCRP immunoassays highlighted the need to confirm
CRM 470 to further minimize these differences is manufactures claims regarding hemolysis, icterus, and/or
currently being investigated [15]. lipemia. Based on 10% of the target value, moderate
hemolysis (above 4g/L) was found to interfere with two
Reference and Preferred Method of the assays tested, and one of the assays was affected
There is no reference method for the measurement of by triglyceride concentrations above 2.5 g/L, which was
CRP. Laboratories choosing a method must first lower than the manufacturers claims of 7.4 g/L. None of
determine the clinical application and consider the the assays were affected by bilirubin (at concentrations
advantages and disadvantages of high sensitivity versus up to 60 mg/dL) [18].
a wide measuring range in the context of laboratory
consolidation and automation. Specimens with high titers of rheumatoid factor have the
potential to interfere with immunoassays for CRP, but
Radioimmunoassay (RIA) methods are capable of most reagent manufacturers incorporate dithiothreitol to
detecting as little as 1 to 5 ng/mL of CRP. Although RIA destroy rheumatoid factor by reducing disulfide bonds.
is sensitive and also relatively inexpensive, it is not
commonly employed because of the availability of other In inflammatory states, CRP can increase dramatically to
faster automated procedures. very high concentrations from baseline values, resulting
in falsely low results due to antigen excess. This prozone
To accommodate requirements for cardiovascular risk effect may be seen for single-antibody nephelometric
prediction and acute-phase concentrations, and turbidimetric methods [13]. A number of sites have
manufacturers have developed assays to provide both reported evidence of prozone effects at varying CRP
290
concentrations ranging from as little as < 50 mg/L to 510 mg/L (average risk), and > 3.0 mg/L (high risk). An
mg/L [13,17,28]. In one study, four of the nine assays individuals CRP should be measured twice, either
tested were affected [28]. fasting or not fasting, with the average expressed in
mg/L. If the hsCRP concentration is 10.0 mg/L, then
Reference Interval the test should be repeated and the person examined for
Baseline CRP concentrations in newborns are very low. sources of infection or inflammation [44].
Increases due to infection or other inflammatory
processes may only be detectable with hsCRP assays Studies carried out to determine the usefulness of CRP
[30]. as an independent predictor of cerebrovascular events
and to evaluate prognosis after stroke conclude that at
In the absence of acute inflammation, healthy adults present, there is insufficient evidence to support testing
usually have CRP concentrations less than 10 mg/L, [45,46].
although studies have found mean concentrations tend to
increase slightly in women during late pregnancy [14] CRP Performance Goals
and with increasing age [31-33]. The CAP performance criteria set for CRP is 3 standard
deviations of the target mean, which is determined by
Rifai and Ridker [34] proposed an algorithm for risk each analyzers peer-group mean. For the 2007
assessment in both genders established by retrospective Immunology Survey (CRP-01, CRP-02), 90% of the
epidemiological studies of coronary heart disease in assays are either turbidimetric (77% across 13 platforms)
conjunction with LDL-cholesterol decision limits or nephelometric (13% across 3 platforms). Overall the
recommended by the National Cholesterol Education turbidimetric assays perform marginally better, with
Program (NCEP) (Figure 1). imprecision values (% coefficient of variation [CV])
ranging from 4.3% to 8.2% at a value around 20 mg/L
Interpretation versus nephelometric, 4.9% to 10%. At a value around
A variety of stimuli, including infection, major trauma, 40 mg/L, CVs are 3.4% to 8.6% versus 4.5% to 8.5%,
surgery, and chronic inflammatory conditions, can result respectively.
in a marked increase in CRP production (Table 2) [35].
Following stimuli, concentrations of CRP rise within The performance criteria set by CAP for hsCRP testing
hours, doubling 5 to 8 hours thereafter [25]. It is is 30% of the target value as determined by each
generally accepted that mild inflammation and viral analyzers peer-group mean. For the 2007 Cardiac Risk
infections cause elevation of CRP concentration within Survey (hsCRP-01, hsCRP-02), at a mean of 1.28 mg/L,
the 10 to 40 mg/L range, whereas active inflammation CVs ranged from 4.5% to 29.5%, compared with a mean
and bacterial infection result in concentrations of 40 to of 3.02 mg/L CVs ranging from 2.2% to 13.8%. The
200 mg/L. Concentrations over 200 mg/L are found in exact analytical performance criteria required for
severe bacterial infections and burns [36]. CRP has a accurate classification into risk categories have not yet
half-life of 19 hours [38]. been established. However, it has been suggested that
the within-run precision of hsCRP assays should not
Population studies indicate no major racial or ethnic exceed 10% at a concentration of 0.2 mg/L [14].
differences exist [31,32,34]. Higher CRP concentrations
have been found among obese individuals [38,39,40]. References
Obesity is associated with low-grade inflammation. It is 1 Tillet WS, Francis T Jr. Serological reactions in
thought IL-6 production by adipose tissue induces pneumonia with a non-protein somatic fraction
hepatic CRP synthesis and hence promotes the onset of of pneumococcus. J Exp Med 1930;52:561-571.
cardiovascular complications [39]. Smokers have higher 2 Abernathy TJ, Avery OT. The occurrence
CRP concentrations than those who have never smoked during acute infection of a protein not normally
[13,31]. present in the blood. I. Distribution of the
reactive protein in patients sera and the effect
For both sexes, the 50th percentile of CRP measured in of calcium on the flocculation reaction with C-
various study populations was about 1.5 mg/L (Table 3), polysaccharide of pneumococcus. J Exp Med
hence gender-specific cutoff values are not required for 1941;73:173-181.
risk stratification [33,34]. 3 Hart WR. C-reactive protein: the best
laboratory indicator available for monitoring
There is consensus that baseline CRP concentrations disease activity. Cleve Clin J Med 1989;56:126-
significantly increase following hormone replacement 130.
treatment [41,42]. One study suggests that this increase 4 Gill CW, Bush WS, Burleigh WM, Fischer CL.
may be related to metabolic hepatic activation and not to An evaluation of a C-reactive protein assay
an acute-phase response [43]. using a rate immunonephelometric procedure.
Am J Clin Pathol 1981;75:50-55.
In the context of cardiovascular risk prediction, the 5 Castell JV, Gomez-Lechon MJ, David M, Fabra
American Heart Association (AHA) and the Centers for R, Trullenque R, Heinrich PC. Acute-phase
Disease Control and Prevention (CDC) recommend three response of human hepatocytes: regulation of
decision limits be used: < 1.0 mg/L (low risk), 1.0 to 3
291
acute-phase protein synthesis by interleukin-6. sensitivity C-reactive protein assays and

Hepatology 1990;12:1179-1186. concordance in cardiovascular risk
6 Black S, Kushner I, Samols D. C-reactive stratification. Clin Chim Acta 2004;347:71-79.
protein. J Bio Chem 2004;279:47:48487-48490. 19 iman AR, Kme T, Ta G, Akan P, Tuncel P.
7 Ridker PM, Hennekens CH, Burins JE, Rifai N. Comparison and evaluation of two C-reactive
C-reactive protein and other markers of protein assays based on particle-enhanced
inflammation in the prediction of immunoturbidimetry. J Clin Lab Analysis
cardiovascular disease in women. N Engl J Med 2007;21:71-76.
2000;342:836-843. 20 Nordin G, Samuelsson I, Andersson B. C-
8 Ridker PM, Cushman M, Stampher MJ, Tracy reactive protein: the difference between
RP, Hennekens CH. Inflammation, aspirin, and quantitation in serum and EDTA plasma. Scand
the risk of cardiovascular disease in apparently J Clin Lab Invest 1996;56:123-127.
healthy men. N Engl J Med 1997;336:973-979. 21 Dupuy AM, Michon AL, Badiou S, Cristol JP.
9 Sattar N, Murray H, McConnachie A, Blauw Adaptation and evaluation of the Randox Full-
GJ, Bollen E, Buckley BM et al. C-reactive Range CRP Assay on the Olympus AU2700.
protein and prediction of coronary heart disease J Clin Lab Analysis 2007;21:34-39.
and global vascular events in the Prospective 22 Hartweg J, Gunter M, Perera R, Farmer A, Cull
Study of Pravastatin in the Elderly at Risk C, Schalkwijk C et al. Stability of soluble
(PROSPER). Circulation 2007;115:981-989. adhesion molecules, selectins, and C-reactive
10 Ledue TB, Rifai N. High-sensitivity protein at various temperatures: implications for
immunoassays for C-reactive protein: promises epidemiological and large-scale clinical studies.
and pitfalls. Clin Chem Lab Med Clin Chem 2007;53:1858-1860.
2001;39:1171-1176. 23 Eijsden M, van der Wal MF, Hornstra G,
11 Volanakis JE, Clements W, Schroehenloher R. Bonsel GJ. Can whole-blood samples be stored
C-reactive protein: purification by affinity over 24 hours without compromising stability
chromatography and physicochemical of C-reactive protein, retinol, ferritin, folic acid,
characterization. J Immunol Methods and fatty acids in epidemiologic research? Clin
1978;23:285-295. Chem 2005;51:230-232.
12 Schwartz MA, Schifreen RS, Gorman E, Tuhy 24 Macy EM, Hayes TE, Tracy RP. Variability in
PM, Bienvenu J, Warkentin DL. Development the measurement of C-reactive protein in
and performance of a fully automated method healthy subjects: implications for reference
for assay of C-reactive protein in the aca intervals and epidemiological applications. Clin
Discrete Clinical Analyzer. Clin Chem Chem 1997;43:1:52-58.
1988;34:8:1646. 25 Aziz N, Fahey JL, Detels R, Butch AW.
13 Roberts WL, Moulton L, Law TC, Farrow G, Analytical performance of a highly sensitive C-
Cooper-Anderson M, Savory J, Rifai N. reactive protein-based immunoassay and the
Evaluation of nine automated high-sensitivity effects of laboratory variables on levels of
C-reactive protein methods: implications for protein in blood. Clin Diagn Lab Immunol
clinical and epidemiological applications. Part 2003;652-657.
2. Clin Chem 2001;47:3:418-425. 26 Meier-Ewert HK, Ridker PM, Rifai N, Price N,
14 Whicher J. BCR/IFCC Reference material for Dinges DF, Mullington JM. Absence of diurnal
plasma proteins (CRM 470). Clin Biochem variation of C-reactive protein concentrations in
1998;31:6:459-465. healthy human subjects. Clin Chem
15 Kimberly MM, Vesper HW, Caudill SP, 2001;47:3:426-430.
Cooper GR, Rafai N, Dati F, Myers GL. 27 Rudnicka AR, Rumley A, Lowe GDO, Strachan
Standardization of immunoassays for DP. Diurnal, seasonal and blood-processing
measurement of high-sensitivity C-reactive patterns in levels of circulating fibrinogen,
protein. Phase I: evaluation of secondary fibrin D-dimer, C-reactive protein, tissue
reference Materials. Clin Chem 2003;49:4:611- plasminogen activator, and von Willebrand
616. factor in a 45-year-old population. Circulation
16 Roberts WL, Sedrick R, Moulton L, Spencer A, 2007;115:996-1003.
Rifai N. Evaluation of four automated high- 28 Sindic CJM, Collet-Casart D, Depre A, Laterre
sensitivity C-reactive protein methods: ED, Masson PL. C-reactive protein in serum
implications for clinical and epidemiological and cerebrospinal fluid in various neurological
applications. Clin Chem 2000;46:4:461-468. disorders. J Neurol Sci 1984;63:339-344.
17 Roberts WL, Schwarz EL, Ayanian S, Rifai N. 29 Salzer HR, Genger H, Muhar U, Lischka A,
Performance characteristics of a point-of-care Schatten C, Pollak A. C-reactive protein: an
C-reactive protein assay. Clin Chim Acta early marker for the neonatal bacterial infection
2001;314:255-259. due to prolonged rupture of amniotic
18 McWhorter VC, Ford LC, Butch AW. membranes and/or amnionitis. Acta Obstet
Analytical Performance of the Synchron Gynecol Scand 1987;66:365-367.
LX20 Pro, BN II and IMMAGE high-
292
30 Wasunna A, Whitelaw A, Gallimore R, 43 Silvestri A, Gebara O, Vitale C, Wajngarten M,

Hawkins PN, Pepys MB. C-reactive protein and Leonardo F, Ramires JAF et al. Increased levels
bacterial infection in preterm infants. Eur J of C-reactive protein after oral hormone
Pediatr 1990;149:424-427. replacement therapy may not be related to an
31 Ford ES, Giles WH, Myers GL, Mannino DM. increased inflammatory response. Circulation
Population distribution of high-sensitivity C- 2003;107:3165-3169.
reactive protein among U.S. men: findings from 44 Pearson TA, Mensah GA, Alexander RW,
the National Health and Nutrition Examination Anderson JL, Cannon RO, Criqui M et al.
Survey 1999-2000. Clin Chem 2003;49:4:686- Markers of inflammation and cardiovascular
690. disease: application to clinical and public health
32 Imhof A, Frohlich M, Loewel H, Helbecque N, practice. A statement for healthcare
Woodward M, Amouyel P et al. Distribution of professionals from the Centers for Disease
C-reactive protein measured by high-sensitivity Control and Prevention and the American Heart
assays in apparently healthy men and women Association. Circulation 2003;107:499-511.
from different populations in Europe. Clin 45 Curb JD, Abbott RD, Rodriguez BL, Sakkinen
Chem 2003;49:4:669-672. P, Popper JS, Yano K, Tracy RP. C-reactive
33 Hung J, Knuiman MW, Divitini ML, Davis T, protein and the future risk of thromboembolic
Beilby JP. Prevalence and risk factor correlates stroke in healthy men. Circulation
of elevated C-reactive protein in an adult 2003;107:2016-2020.
Australian population. Am J Cardiol 46 Di Napoli M, Schwaninger M, Cappelli R,
2008;101:193-198. Ceccarelli E, Di Gianfilippo G, Donati C et al.
34 Rifai N, Ridker PM. Population distributions of Evaluation of C-reactive protein measurement
C-reactive protein in apparently healthy men for assessing the risk and prognosis in ischemic
and women in the United States: implication for stroke: a statement for health care professionals
clinical interpretation. Clin Chem from the CRP Pooling Project members. Stroke
2003;49:4:666-672. 2005;36:1316-1329.
35 Pepys MB, Hirschfield GM. C-reactive protein:
a critical update. J Clin Invest 2003;111:1805-
1812.
36 Clyne B, Olshaker JS. The C-reactive protein. J
Emerg Med 1999;17:6:1019-1025.
37 Vigushin DM, Pepys MB, Hawkins PN.
Metabolic and scintigraphic studies of
radioiodinated human C-reactive protein in
health and disease. J Clin Invest 1993;91:1351-
1357.
38 Fogarty AW, Glancy C, Jones S, Lewis SA,
McKeever TM, Britton JR. A prospective study
of weight change and systemic inflammation
over 9 years. Am J Clin Nutr 2008;87:30-35.
39 Bastard JP, Maaachi M, Lagathu C, Kim MJ,
Carson M, Vidal H et al. Recent advances in the
relationship between obesity, inflammation, and
insulin resistance. Eur Cytokine Netw
2006;17:4-12.
40 Bahceci M, Gokalp D, Bahceci S, Tuzcu A,
Atmaca S, Arikan S. The correlation between
adiposity and adiponectin, tumor necrosis factor
, interlukin-6 and high-sensitivity C-reactive
protein levels. Is adipocyte size associated with
inflammation in adults? J Endocrinol Invest
2007;30:210-214.
41 Ridker PM, Hennekens CH, Rifai N, Buring JE,
Manson JE. Hormone replacement therapy and
increased plasma concentration of C-reactive
protein. Circulation 1999;100:713-716.
42 Cushman M, Legault C, Barrett-Connor E,
Stefanick ML, Kessler C, Judd HL et al. Effect
of postmenopausal hormones on inflammation-
sensitive proteins: the Postmenopausal
Estrogen/Progestin Interventions (PEPI) study.
Circulation 1999;100:717-722.
293
Table 1: Methods of Analysis for C-Reactive Protein
Method 1: Single radial immunodiffusion (RID); quantitative

Principle of analysis: Diffusion of protein through medium containing specific antibody
Usage: Seldom used
Comments: Serum, plasma, manual; requires very long incubation time
Method 2: Immunoturbidimetry
Principle of analysis: Reaction of CRP with anti-CRP antibody results in formation of antigen-antibody
complexes which absorb light
Usage: Most common quantitative method for CRP
Comments: Sensitive, specific; automated; requires specialized instrumentation
Method 3: Immunonephelometry
Principle of analysis: Reaction of CRP with anti-CRP antibody results in formation of antigen-antibody
complexes which scatter light
Usage: Very common quantitative method for CRP in use
Comments: Sensitive, specific; automated; requires specialized instrumentation.
Method 4: Florescence-labeled ELISA

Principle of analysis: CRP binds to enzyme-labeled antibodies and to antibodies bound to microparticles;
enzyme in complex releases fluorescent 4-methylumbelliferone, which is measured.
MEIA monoclonal Abs coat microparticle
MicroparticleAb + 2M + AbALP
Microparticle Ab2MAbALP + 2M +
AbALP + 2MAbALP
1. Wash step
2. 4-MU product substrate
3. Quench
Measure 4-MU product fluorescence at 448 nm
Usage: Very common quantitative method for CRP
Comments: Serum, plasma, urine; semiautomated technique; sensitive, specific; automated; requires specialized
instrumentation

Principle of analysis: Agglutination of latex particles coated with anti-CRP antibodies due to presence of CRP
results in formation of visible aggregates; size of aggregates related to CRP concentrations
Usage: Commonly used as semiquantitative procedure; also can be used to report quantitative results
Comments: Semiquantitative; point-of-care test kit has been developed; useful as rapid screen
294
Table 2: CRP Responses in Disease.
Major CRP Acute-Phase Response

Infections Bacterial
Systemic/severe fungal,
mycobacterial, viral
Allergic complications of infection Rheumatic fever

Erythema nodosum
Inflammatory disease Rheumatoid arthritis

Juvenile chronic arthritis
Ankylosing spondylitis
Psoriatic arthritis
Systemic vasculitis
Polymyalgia rheumatica
Reiters disease
Crohns Disease
Familial Mediterranean fever
Necrosis Myocardial infarction

Tumor embolization
Acute pancreatitis
Trauma Surgery
Burns
Fractures
Malignancy Lymphoma
Carcinoma
Sarcoma
Modest or Absent CRP Acute-Phase Response Systemic lupus erythematosus

Scleroderma
Dermatomyositis
Ulcerative colitis
Leukemia
Graft-versus-host disease
From Clyne B, Olshaker JS. The C-reactive protein. J Emerg Med 1999;17(6):1019-1025 [36].
Table 3: Population Distribution of C-Reactive Protein in Men and Women by Percentile.

5th 10th 25th 50th 75th 90th 95th
Women* 0.19 0.29 0.61 1.52 3.48 6.61 9.14
Men 0.28 0.40 0.80 1.50 3.20 6.05 8.55

*
Women were not taking hormone replacement therapy.
From Rifai N, Ridker PM. Population distributions of C-reactive protein in apparently healthy men and
women in the United States: implications for clinical interpretation. Clin Chem 2003;49(4):666-672 [34].
295
Figure 1: Proposed algorithm for risk assessment of coronary heart disease in men and
women. (From Rifai N, Ridker PM. Population distributions of C-reactive protein in
apparently healthy men and women in the United States: implication for clinical
interpretation. Clin Chem 2003;49[4]:666-672 [34].)
4
>3
(mg/L)
2 1-3
CRP
<1.0
0
>1600 1300-1600 <1300
LDL-Cholesterol (mg/L)
296
Calcium
Calcium
Randal J. Schneider
Name: Calcium
Clinical significance: Refer to Chapter 33, Bone Disease, in the 5th edition of Clinical Chemistry:
Atomic symbol: Ca
Atomic mass: 40.08 D
Merck Index: 1613
Chemical class: Alkaline earth element
Number of forms: Free Ca2+, protein bound, or in inorganic complexes
NRSCL Definitive and
Reference Methods: NCCLS RS9-P
i
Within the circulation, calcium is found in several A sensitive fluorescent method for the microanalysis of
different forms, with 46% free, 32% bound to albumin, serum or urine is available [3]. In one method, the
8% bound to globulins, and 14% associated in freely calcium is added to an alkaline solution containing
diffusible calcium complexes. All calcium is ionized no calcein (Table 1, Method 3). The calcium combines with
matter what it is bound to, but it may not be dialyzable the calcein, which at high pH fluoresces at 520 nm
or completely reactive with a chosen chromogen when (excitation at 490 nm). The fluorescent complex is
complexed to another compound. Ion-specific electrode titrated with ethylene glycol-bis(-aminoethylether)-
techniques for measuring the activity of the unbound N,N-tetraacetic acid (EGTA), which binds the calcium
fraction are available and are discussed elsewhere. from the complex with a resulting decrease in
fluorescence. When the baseline fluorescence is
The oldest procedures used for the clinical determination achieved, the volume of EGTA required to titrate the
of total calcium involved the quantitative precipitation of complex is directly proportional to the concentration of
calcium with an excess of anions from oxalic, calcium. It has been reported that magnesium and
chloranilic, or naphthylhydroxamic acids (Table 1, phosphate do not interfere with this method.
Method 1) [1,2].
Direct spectrophotometric measurements of calcium in
Methods using chloranilate or naphthylhydroxamate ions serum or urine are also based upon formation of color
to precipitate calcium were based upon quantification of complexes between calcium and organic molecules.
the color of the precipitating ion. An excess of Examples of compounds that give colored reaction
chloranilate was added to react with the serum calcium. products with calcium are (1) glyoxal-bis(2-
The calcium chloranilate complex was precipitated and hydroxyanil), (2) alizarin, (3) chlorophosphonazo III, (4)
redissolved in alkaline solutions of methylthymol blue, (5) o-cresolphthalein complexone,
ethylenediaminetetraacetic acid (EDTA). The and (6) arsenazo III. Of these calcium-complexing,
absorbance of the intense red-purple color of the colorimetric reagents, the o-cresolphthalein and arsenazo
liberated chloranilate acid (Amax,520 nm) was III dye are the most commonly used for routine calcium
proportional to the concentration of calcium in the analysis. Approximately 45% of the participants in the
specimen (Table 1, Method 2). Similarly, 2007 College of American Pathologists (CAP)
naphthylhydroxamate has been used to precipitate Participant Summary Reports indicated that they used
calcium. The precipitate was dissolved in alkaline the o-cresolphthalein method, while approximately 30%
EDTA, and when acid ferric nitrate was added, a red- used the arsenazo III dye. The remaining laboratories
orange color was produced. These precipitation methods reported using an ion-selective electrode.
are rarely used today, because they are time consuming
and insensitive. The reaction of calcium with o-cresolphthalein produces
a red complex (quantitated at 570 to 575 nm) at pH 10 to
12 (Table 1, Method 4). The spectrum of the complex
i formed with calcium is given in Figure 1. The reaction
Calcium
product is stabilized by the addition of KCN (potassium
cyanide), which also acts to eliminate interference from
First edition: E. Christis Farrell
heavy metals, and interference by magnesium ions is
Methods edition: E. Christis Farrell
eliminated by the addition of 8-hydroxyquinoline.
Second edition: E. Christis Farrell
Similarly, assays using the arsenazo III dye involve a
reaction with dye and calcium to produce a color
complex (blue-purple) that can be measured
Fifth edition: Randal J. Schneider
spectrophotometrically. In this reaction under acidic
297
Calcium
conditions, the color complex is measured at 650 nm, include atomic absorption spectroscopy, induction
and like the o-cresolphthalein reaction is proportional to coupled plasma atomic emission spectroscopy, ion
the calcium concentration in the sample. chromatography, and isotope-dilution mass
spectroscopy.
Dry-slide technology uses the arsenazo III dye in a dry-
reagent layered coating system. When patient fluid is The preferred clinical method for total calcium
applied to the slide, the arsenazo dye migrates from the determinations is atomic absorption spectroscopy,
bottom layer to a mordant layer and is trapped. The although this method is usually limited to large reference
calcium in the sample migrates to the mordant layer, laboratories for reasons mentioned below.
complexing with the dye and changing its color. The Acceptable methods include the dye-complex methods
absorption is recorded by reflectance photometry at 680 based on cresolphthalein and arsenazo III or indirect
nm [4]. potentiometric methods. Proficiency results for 2007
indicate all three methods to be satisfactory for reporting
Calcium present in protein or inorganic complexes is clinical calcium results.
detectable by flame atomic absorption only when steps
are taken to dissociate the calcium from these Additional methods listed in Table 1 are either not
complexes. Acid is used for the dissociation of protein- practical for most clinical laboratories or do not meet
Clinical Laboratory Improvement Amendments (CLIA)
bound calcium, and lanthanum (La2+) or strontium
requirements for total allowable error.
(Sr2+) ions are added to displace Ca2+ from phosphate, The older oxalate precipitation and titration methods are
oxalate, citrate, and other complexes. The La2+ or Sr2+ no longer in use for routine calcium analysis because
is added to the sample diluent. Serum is diluted 1:50 they were quite laborious and imprecise.
with 1% La2+, whereas urine is diluted 1:50 with 5%
The most commonly used reference method for accurate
La2+ in order to overcome the higher phosphate
calcium measurement is atomic absorption. Although it
concentration found in urine. Protein is precipitated by
is not the definitive method, atomic absorption is more
treatment of the sample with acid. Precipitation results in
commonly available and is thus often used as a reference
a 2% to 3% reduction in sample volume when the
method for comparison studies of new methods. The
precipitate is removed. One can mathematically correct
precision of atomic absorption over the range of
for the reduced volume. Acid treatment without protein
physiological concentrations is of the order of 2% to 3%.
precipitation requires increased maintenance of the
Atomic absorption methods are very sensitive, requiring
burner head to prevent accumulation of debris.
10 to 20 L of sample for analysis. The method has been
semiautomated with automatic sample aspiration after
A relatively recent technique for measurement of total
dilution with semiautomatic diluters.
calcium in serum and urine is with the use of indirect
potentiometry (Table 1, Method 7). This method utilizes
The o-cresolphthalein methods have been adapted to a
a calcium ionselective electrode in conjunction with a
wide variety of automated analyzers, with a resulting
sodium reference electrode. Sample is mixed with a
high precision. A significant problem with this technique
buffered solution before its interaction with the calcium
is the requirement for good temperature control during
electrode. Interaction of calcium with the selective
the analysis, since the chromogens molar absorptivity is
membrane results in changes in electrical potential that
temperature dependent.
are compared with the sodium reference electrode.
On the basis of precision and accuracy, the preferred
In atomic absorption, interference by magnesium and
clinical method is certainly atomic absorption. The
other elements is reduced by the use of narrow band-
major drawbacks of atomic absorption are the high level
pass, diffraction-grating spectrophotometers for specific
of maintenance and care the equipment requires and the
isolation of the atomic absorption line of calcium (422.7
limited throughput of specimens for high-volume
nm). Interference by sodium is eliminated by the
laboratories.
addition of physiological concentrations of sodium ions
to calcium standards. Atomic absorption is not
Specimen
frequently used for routine analysis, probably because it
Serum or heparinized plasma is separated from cells as
has rarely been automated for high sample throughput.
rapidly as possible to avoid the uptake of calcium by
This method has been approved as a reference method
erythrocytes. Stasis changes concentration of protein-
for measurement of serum calcium [5].
bound Ca2+, and concentrations of free Ca2+ will be
changed by pH shifts.
The definitive method for calcium measurements is
isotope-dilution mass spectroscopy (Table 1, Method 8)
Serum or heparinized plasma samples may be stored at
[6]. This method, available in only a few institutions, is
room temperature for up to 8 hours at 4C for 1 day or
the accuracy standard against which all methods should
frozen for up to 1 year. Calcium in specimens collected
be compared.
and stored in gel-separator tubes did not show any
difference compared with specimens collected and
stored in tubes not containing gel [8].
Currently there are 5 reported candidate reference
methods for measuring total calcium. These methods
298
Calcium
Urine calcium can be kept dissolved by addition of 10 bound fraction. Although the concentration of ionized
mL of 6 M HCl to the collection container before a 24- calcium may vary during alkalosis and acidosis, the total
hour specimen is collected. Urine should be well mixed calcium concentration is relatively unaffected.
during the collection period. A major cause of hypercalcemia is malignancy, with
increases observed in bone diseases such as multiple
Interferences myeloma and with metastatic tumors with secondary
Most o-cresolphthalein methods using direct analysis bony deposits. Only 10% to 15% of such metastatic
after dilution appear to be sensitive to interferences from cancers result in hypercalcemia.
lipemia, hemoglobin, and myeloma protein [9, 10-12].
Atomic absorption is not affected by important Critical values indicating the possibility of morbidity are
interferences from common compounds such as less than 6.0 mg/dL (1.5 mmol/L) and greater than 14.0
hemoglobin, bilirubin, or lipemia. However, calcium- mg/dL (3.5 mmol/L).
chelating compounds interfere with this method. The
potentiometric method for total calcium is affected by A list of diseases causing hypercalcemia and
bromide, which causes a positive bias. hypocalcemia is presented in the following boxes:
Calcium Reference Intervals Causes of Hypercalcemia

Many sources report serum calcium levels of 90 to 110 Hyperparathyroidism
mg/L for disease-free persons (normals). Newer Thyrotoxicosis
methods show lower intervals of about 80 to 105 mg/L Addisons disease
There is no clinically significant difference in serum Withdrawal of steroids
calcium concentrations between men and women. Urine Tumors
calcium concentration varies considerably and is only Vitamin D and vitamin A intoxication
meaningful if the patient is kept on a low-calcium, Sarcoidosis
neutral-ash diet for 3 days before collection. A low Idiopathic hypercalcemia of infancy
output is of the order of 50 to 100 mg/day, with the Immobilization
average being 100 to 300 mg/day. Values are lower in Subcutaneous fat necrosis in infants
persons older than 70 years. Diurnal variation in serum Thiazide diuretics
calcium has been described, with lowest concentrations Milk-alkali syndrome
seen at approximately 8:00 am and highest values Benign familial hypercalcemia
occurring in early afternoon [13].
Population Serum Ca UrineCa
mg/L (mmol/L) mg/24h(mmol/24hr) Causes of Hypocalcemia
Adults Vitamin D
Atomic absorption 80105 (2.02.6) Men<275 (<6.87)
Women <250 (<6.25) Decreased solar exposure and endogenous synthesis
Hypercalcemic >300 (>7.50) Decreased intestinal intake (malabsorption, dietary
Cresolphthalein deficiency)
complexone 80105 (2.02.6) Average diet 100250 Altered hepatic metabolism of vitamin D (hepatic
(2.56.25)
Pediatric (by atomic absorption)
disease, anticonvulsants)
Premature infants 60100 (1.5-2.5) Decreased renal synthesis of calcitriol (vitamin D
Full-term infants 73120 (1.8-3.0) dependency, renal failure)
1 to 2 years 100120 (2.5-3.0) Parathyroid
Hypoparathyroidism (primary and secondary)
Interpretation Pseudohypoparathyroidism
Total serum calcium concentration is influenced by the Calcitonin
serum albumin concentration, position of the body when Calcitonin or mithramycin infusion
the calcium is drawn, and tourniquet stasis. The ionized Calcium
calcium is not influenced by these factors and is the Intestinal malabsorption
better indicator of the hypercalcemic or hypocalcemic Acute pancreatitis
state. It may be calculated from total protein albumin Infusion of agents complexing calcium
and calcium values, but direct measurement is Alkalosis decreasing ionized calcium
preferable. Magnesium
Magnesium deficiency
The amount of albumin-bound calcium present is Phosphorus
dependent on the albumin concentration. A decrease of Renal failure
serum albumin of 1 g/L will decrease the total serum Phosphate infusion
calcium by about 0.08 mg/L. Cows milk formulas
Posture changes the serum calcium concentration; going
from an upright to a recumbent posture causes a decrease
of approximately 4% (range 2% to 7%). Calcium Performance Goals
Survey data from the 2007 CAP Participant Summary
The fraction of calcium bound to protein in serum is pH- Report show imprecision values (% coefficient of
dependent. Acidosis increases the ionized fraction variation [CV]) for measurement of calcium with a
(decreases binding), whereas alkalosis increases the
299
Calcium
concentration of 10.57 mg/dL to range from 12 Ladenson JH, McDonald JM, Goren M.
approximately 1.4% to 2.9% for the arsenazo III dye and Multiple myeloma and hypercalcemia [case
cresolphthalein complexone methods. The indirect conference]. Clin Chem 1979;25:1821-1825.
potentiometric technique shows imprecision values that 13 Herfarth K, Schmidy-Gayk H, Graf S, Maier A.
range from 1.1% to 2.4%. Acceptable performance Circadian rhythm and pulsatility of parathyroid
criteria (CLIA 88) for measurement of calcium require hormone secretion in man. Clin Endocrinol
that laboratories be accurate to within 1.0 mg/dL of the 1992;37:511-519.
peer-group mean. 14 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
The intraindividual variation of calcium in blood has databases on biologic variation: pros, cons and
been determined to be approximately less than 2.0%. progress. Scand J Clin Lab Invest 1999;59:491-
Desirable specifications for analytical imprecision 500.
derived from studies of biological variation indicate an
assay imprecision of no greater than 1.0% and a total
error or no greater than approximately 2.5% [14].
References
1 Weissman N, Pileggi VJ. Inorganic ions. In:
Henry RJ, Cannon DC, Winkelman JW, eds.
Clinical Chemistry: Principles and Technics.
2nd ed. New York: Harper & Row; 1974:646-
653.
2 Clark EP, Collips JB. A study of the Tisdall
method for the determination of blood serum
calcium with suggested modification. J Biol
Chem 1925;63:461-464.
3 Jackson JE, Breem M, Cheng C. Fluorometric
titration of calcium. J Lab Clin Med
1962;60:700-708.
4 Shirey TL. Development of a layered-coating
technology for clinical chemistry. Clin Biochem
1983;16:147-155.
5 National Committee for Clinical Laboratory
Standards (NCCLS). RS9-Calcium: proposed
summary of methods and materials credentialed
by the NRSCL council. Wayne, PA: NCCLS;
1989.
6 Cali JP, Mandel J, Moore L, Young DS. U.S.
Department of Commerce National Bureau of
Standards. A reference method for the
determination of calcium in serum. NPS Spec.
Pub. 260-36. Washington, DC: U.S.
Government Printing Office; 1972.
7 Brett EM, Hicks JM. Total calcium
measurement in serum from neonates:
limitations of current methods. Clin Chem
1981;27:1733-1737.
8 Hein M, Heil W, Wihold W. Storage of serum
or whole blood samples? Effects of time and
temperature on 22 serum analytes. Eur J Clin
Chem Clin Biochem 1995;33:231-238.
9 Leidtke RJ, Kroon G, Batjer JD. Centrifugal
analysis with automated sequential reagent
addition: measurement of serum calcium. Clin
Chem 1981;27:2025-2028.
10 Porter WH, Carrol JR, Roberts RE. Hemoglobin
interference with the DuPont automatic clinical
analyzer procedure for calcium. Clin Chem
1977;23:2145-2147.
11 Hass RG, Mushel S. Modified Dupont aca
calcium method for hemolyzed specimens. Am
J Clin Pathol 1982;77:216-219.
300
Calcium
Table 1: Methods of Calcium Analysis

Method 1: Precipitation by oxalate and redox titration
Ca2+ + oxalate Ca oxalate (ppt)
Ca oxalate (ppt) + H2SO4 oxalate + CaSO4
_
2KMnO4 + 5 oxalate + 2H2SO4 70C K2SO4 + 2MnSO4 + 10CO2 + 8H2O
Comments: Historical; initial reference method
Method 2: Precipitation by colored anions; spectrophotometric
Ca2+ + chloranilate Ca-chloranilate (ppt)
_
Ca-chloranilate (ppt) + EDTA OH Ca-EDTA + chloranilic acid (purple)
Comments: Historical; labor intensive, imprecise, many other dyes available
Method 3: Titration of fluorescent Ca2+ complex
Ca2+ + calcein Ca-calcein (fluorescent)
EGTA + Ca-calcein Ca+-EGTA + calcein (decreased fluorescence)
Comments: Stat or small labs; small sample size, dedicated instrument
Method 4: Spectrophotometric measurement of Ca2+ complexes
a. Direct
b. Dialysis
_ -
a. Ca2+ + o-cresolphthalein OH red complex (520 nm)
_
b. Ca2+ complex + H+ dialysisCa2+ in recipient stream
Ca2+ detected as in (a)
Comments:
a. Most common; early adapted to a variety of automated instruments; positive bias compared to atomic
absorption
b. On Technicon AutoAnalyzer
Method 5: Flame emission
_ _
Ca2+ heat Ca0 heat Ca* Ca0 + photon 2e
Comments: Historical; poor sensitivity
Method 6: Atomic absorption
_
Ca2+ 2e- Ca0
Photon + Ca0 Ca*
Comments: Reference method; excellent accuracy and sensitivity
Method 7: Indirect potentiometry
Principle of analysis: Calcium ionselective electrode utilized in conjunction with sodium reference electrode;
changes in electrical potential due to calcium referenced to sodium electrode
Comments: Generally shows better accuracy compared with dye-binding techniques, increasing usage
Method 8: Isotope-dilution mass spectrometry
Principle of analysis: Ca and known amount of Ca isotope; isolate Ca2+, and record ratio of two isotopes on
mass spectrometer
Comments: Definitive method; available in reference centers only
Ca*, Calcium atom in excited state; Ca0, calcium atom in ground state; EGTA, ethylene glyco-bis(-aminoethylether)-
N,N1-tetraacetic acid; ppt, precipitate.
301
Calcium
Table 2: Reaction Conditions for Calcium Analysis

Condition: Temperature
o-Cresolphthalein spectrophotometric*: 25C to 37C
Atomic absorption: 2300C
Condition: Sample volume
o-Cresolphthalein spectrophotometric*: 20 L
Atomic absorption: 50 L
Condition: Volume fraction
o-Cresolphthalein spectrophotometric*: 0.01
Atomic absorption: 0.02
Condition: Final concentration
o-Cresolphthalein spectrophotometric*:
o-Cresolphthalein: 0.06 mmol/L
8-Hydroxyquinoline: 9.6 mmol/L
Diethylamine: 193.4 mmol/L
Potassium cyanide: 3.8 mmol/L
Atomic absorption:
Lanthanum diluent: 1 g/L (serum); 5 g/L (urine)
Condition: Wavelength
o-Cresolphthalein spectrophotometric*: 575 nm
Atomic absorption: 422.7 nm
Condition: Reaction mode
o-Cresolphthalein spectrophotometric*: End-point
Atomic absorption: End-point
Condition: Time of reaction
o-Cresolphthalein spectrophotometric*: 30 sec
Atomic absorption: 12 sec
Condition: Linearity
o-Cresolphthalein spectrophotometric*: 150 mg/L
Atomic absorption: Highest standard, usually 120 mg/L
Condition: Interferences
o-Cresolphthalein spectrophotometric*: Gross lipemia
Atomic absorption: Calcium chelators
*From Baginski ES, Marie SS, Alcock NW et al. Calcium in biological fluids. In: Faulkner WR, Meites S, eds. Selected
Methods of Clinical Chemistry. Vol 9. Washington, DC: American Association for Clinical Chemistry Press; 1982.
302
Calcium
Figures
Figure 1: Absorption spectra of o-cresolphthalein reagent.

A, Complexed with calcium standard sample. Dashed line, reagent versus water; dotted dashed line, reagent and sample
versus water; solid line, reagent and sample versus reagent.
B, With various types of serum samples. Reaction was performed at pH 10.4 with 25 L of serum and either 2.00 mL of
blank base solution or reagent. a, Absorbance spectrum obtained with either clean hemolyzed, lipemic, or icteric serum and
the reagent; b, reagent blank; c, lipemic (creamy) serum and base solution; d, hemolyzed (deep red) serum; e, base-solution
curve for icteric (bilirubin level of 240 mg/L) serum and base solution; f, clear serum and base solution. Measured
maximum was 560 to 568 nm.
Procedure: Calcium by Atomic Absorption slowly and cautiously under hood while swirling the
Spectrophotometry flask. Dilute to 1 L with deionized water. Store in dark at
room temperature. Stable for 6 months.
Principle 2. Working lanthanum diluent for serum, 1 g/L
Calcium determinations by atomic absorption (6.28 mmol/L). Dilute 20 mL of stock lanthanum to 1 L
spectroscopy are based on the fact that atoms of an with deionized water. Store in dark bottle at room
element in the ground, or unexcited, state absorb light temperature. Stable for 6 months.
of the same wavelength as that emitted by the element in 3. Working lanthanum diluent for urine, 5 g/L
the excited state. Each element has its own characteristic (31.4 mmol/L). Dilute 100 mL of stock lanthanum to 1
absorption or resonance lines, and no two elements are L with deionized water. Store in dark bottle at room
known to have an identical resonance line. temperature. Stable for 6 months.
4. Stock calcium standard, 1000 mg/L (25
Calcium is determined in serum or urine after the fluid is mmol/L). Dissolve 250 mg of reagent-grade CaCO3 in
diluted sufficiently with lanthanum oxide solution to
3 mL of 1 mol/L HCl. Dilute to 100 mL with deionized
avoid interference by chelating substances, especially
water. Stable 6 months at room temperature in a closed
phosphate ions.
container.
Reagents 5. Calcium blank solution (140 mmol/L Na, 5
Use specially cleaned atomic-absorption glassware and mmol/L K, 0.1 mmol/L phosphate). Place 8.2 g of
deionized water (that is, free of calcium). NaCl, 0.373 g of KCl, and 0.014 g of Na2HPO4 into a 1-
1. Stock lanthanum, 50 g/L (314 mmol/L). L volumetric flask. Dissolve salts in about 250 mL of
Weigh 58.64 g of lanthanum oxide (La2O3), place in a deionized water, and dilute to mark with deionized
1-L flask, and add about 50 mL of deionized water to water. Stable for 1 month at refrigerator temperatures.
wet the powder. Add 250 mL of concentrated HCl
303
Calcium
Working Standards
Dilute to Volume With
Ca (mg/L) Stock Ca (mL) Calcium Blank Solution (mL)
50 5 100
100 10 100
120 12 100
150 15 100
200 20 100
6. Place instrument back in concentration mode.
Store at room temperature in glass flasks. Do 7. Place dirty burner head in 5% FL-70 to soak
not pipet from flasks. Pour aliquots daily into separate overnight.
containers.
Assay
Equipment: PerkinElmer Atomic Absorption
Spectrophotometer Model 460 (or other similar
instrument) with a single-slot burner head and calcium-
magnesium hollow-cathode lamp. Follow each
manufacturers directions for maintenance of burner,
optimization of lamp and flame by peaking the flame,
and airacetylene fuel mixture, as well as for the actual
analytical procedure.
Notes
1. The dilution (1:50) of serum or urine prevents
protein interference. The dilution can be
performed accurately and conveniently with a
semiautomatic dilutor (such as Micromedic
Systems, Inc., Philadelphia, PA).
2. Sera left overnight in plastic sample cups may
give low calcium values.
3. There is no appreciable variation in the
patients calcium levels throughout the day if
exercise is avoided, and there is no difference
between fasting and nonfasting serum calcium
levels.
4. There is no difference between venous and
capillary serum calcium levels.
5. Serum or plasma should not be allowed to
remain in prolonged contact with the
erythrocytes, since with time, the cells become
permeable to calcium.
6. Cerebrospinal fluid (CSF) can be analyzed for
calcium using the procedure for serum analysis.
Since CSF concentrations may be lower than
serum concentrations, use of a slightly lower
standard is advisable.
Daily maintenance to be performed before

instrument use:
1. Turn flame on, and aspirate 5% FL-70 (a
detergent supplied by Fisher Scientific Co.,
Fairlawn, NJ; 50-F-105) from probe for 20 min.
2. Turn flame off, and let cool.
3. Remove burner head, and pour 100 mL of
distilled water down neck of chamber (where
head was, water will come out waste line).
4. Position clean burner head, securing pin in
place and slipping wires over hooks on either
side of chamber; tighten burner head collar.
5. Set instrument for maximum absorbance
(peaking the flame), following
manufacturers instructions.
304
Cancer Antigen 125 (CA 125)

Hassan M.E. Azzazy
Name: Cancer antigen 125, CA 125

Molecular mass: >200,000 D
Refer to Chapter 53, Neoplasia, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Principles of Analysis and Current Usage i Reference Intervals
The first CA 125 assay was developed by Centocor Inc. The upper limit of CA 125 in a healthy population is 35
and approved by the U.S. Food and Drug Administration kU/L [11,13]. The reference interval of 0-35 kU/L is the
(FDA) for monitoring women with primary epithelial same for most CA 125 assays; however, results generated
invasive ovarian cancer. The Centocor CA 125 assay is a
sandwich immunoradiometric assay (IRMA) for by different assays are not interchangeable [4]. Therefore,
quantitative determination of CA 125 in serum; it became the same method must be used for serial monitoring of
commercially available in 1983. The assay utilizes OC 125 patients, and laboratory reports should indicate the method
murine monoclonal antibody developed from cell line used.
OVCA 433, which was derived from an ovarian serous
carcinoma [1]. The OC 125 antibody was immobilized on Age, race, and menopause were found to affect CA 125
polystyrene beads and used to capture the CA 125 antigen. concentrations in healthy individuals. CA 125
125
I-Labeled OC 125 was used as the detector [2,3]. The concentrations appear to decline with aging and the
second generation of this assay (CA 125 II) utilizes OC menopause. Menstrual cycle variations were also observed
125 antibody for capture and M11 monoclonal antibodies with increases in the follicular phase. A 20% to 50%
for detection of CA 125 antigen; the two antibodies bind variation with race in postmenopausal women has been
to non-overlapping epitopes of CA 125 [4,5]. It was reported with CA 125 II measurement (second-generation
reported that the first-generation CA 125 assay can give assay), with concentrations in Caucasian women higher
discordant [6] and even discrepant [7] results. On the other than in African and Asian women [14]. Increased
hand, marker values obtained with second-generation concentrations were detected in 28% of women with non-
assays are reported to be more in agreement [8,9]. gynecological cancers, in 5% of those with benign
diseases, and in 1% to 2% of normal, healthy individuals
In addition to IRMA, enzyme immunoassays (EIA), [2,15,16].
luminometric immunoassays (LIA), and microparticle
enzyme immunoassays (MEIA) were also developed for In a conventional laboratory setting, imprecision of CA
detection of the CA 125 antigen [4]. Currently, several CA 125 assays can contribute significantly to variations of
125 immunoassays are available on automated open results. The degree of result variability is directly related
random access analyzers, many of which are FDA to the absolute CA 125 value, and the relative variability is
approved [10]. fairly constant, with a median of 1.5% to 2.5% of the CA
125 concentration [17].
There is no reference method for CA 125. CA 125 II RIA Interpretation
is considered the preferred method. CA 125 is used as a marker of ovarian and endometrial
carcinomas. It is also increased in some gastrointestinal
Specimen tumors and lung and breast carcinomas. Several expert
Serum is preferred, although it is possible to use plasma in groups and organizations have developed guidelines for
some assays. Samples can be stored prior to testing at 4C the clinical utility of CA 125 as a tumor marker for
for 1 to 5 days or for longer periods (2 weeks to 3 months) ovarian cancer. These include the National Institutes of
at 20C, depending upon which assay is used. Long-term Health (NIH) [18], National Comprehensive Cancer
storage should be at 70C [11]. Network (NCCN) [19], the European Group on Tumor
Markers (EGTM) [20], European Society for Medical
Interferences Oncology (ESMO) [21], and the National Academy for
Heparin and oxalate were found to interfere with CA 125 Clinical Biochemistry (NACB) [11].
measurements [12]. As is the case with other
immunoassays, the presence of heterophilic antibodies in CA 125 concentrations exceed 35 kU/L in 80% of women
the sample (e.g., following diagnostic or therapeutic use) with epithelial ovarian cancer. Its level is elevated in 50%
could interfere with the CA 125 immunoassay [11]. of patients with stage I ovarian carcinoma, in 90% of
patients with stage II disease, and in over 90% of those
i with stage III and IV disease; CA 125 correlates with
CA 125 tumor size and staging [15].
New method:
305
Einhorn et al. [22] studied 100 patients who were cancer. For these patients, the use of radiological imaging
investigated for palpable adnexal masses using diagnostic techniques and/or monitoring other tumor markers is
laparotomy, and 23 of these patients were found to have necessary.
malignant disease. Using a cutoff of 35 kU/L as a
predictive value for malignancy, CA 125 was found to
have 78% sensitivity, 95% specificity, 82% PPV, and 91% Performance Goals
NPV. However, owing to lack of specificity and For CA 125 antigen, the within- and between-subject
sensitivity for a single marker measurement, none of the coefficients of variation (CVs) were 24.7% and 54.6%,
above expert groups recommended CA 125 for screening respectively. The desirable analytical quality
asymptomatic women. specifications for imprecision, bias, and total error (as
derived from biological variation) for serum CA 125 were
Based on current evidence, CA 125 is more widely 12.4%, 15.0%, and 35.4%, respectively [31].
accepted and recommended as an adjunct for differential
diagnosis of suspicious pelvic masses, particularly in The EGTM indicate that assays for tumor markers,
postmenopausal women [18,20,23]. In postmenopausal including CA 125, should demonstrate interassay
women, CA 125 concentration > 95 kU/L can discriminate variability under 10% and intraassay variability less than
benign from malignant pelvic masses with a positive 5% [32]. None of the current CA 125 assays meet their
predictive value of 95% [2]. criteria. The current Clinical Laboratory Improvement
Amendments (CLIA) performance goal for measurement
Elevated CA 125 concentration (>35kU/L) after debulking of CA 125 is for laboratories to be within 3 standard
surgery and chemotherapy indicates the likelihood of a deviations of the peer-group mean. According to the 2007
residual disease. Although elevated postoperative College of American Pathologists (CAP) Survey, CV for
concentrations predict relapse, negative values do not all CA 125 assays at a mean value of 107.3 ng/L (SD 14.5)
exclude disease presence. was 13.5%.
CA 125 may also have a role in monitoring chemotherapy. References

Serial measurements of CA 125 correlated with clinical 1 Bast RC, Feeney M, Lazarus H, Nadler LM,
disease outcome in 89% of 531 patients [24]. There is a Colvin RB, Knapp RC. Reactivity of a
general consensus among current guidelines to use CA monoclonal antibody with human ovarian
125 to monitor response to chemotherapy. The carcinoma. J Clin Invest 1981;68:1331-1337.
Gynecologic Cancer Intergroup (GCIC) defines response 2 Bast RC, Xu FJ, Yu YH, Barnhill S, Zhang Z,
as a 50% reduction in CA 125 concentration compared to Mills GB. CA 125: the past and the future. Int J
that in a pretreatment sample [25,26]. Response should be Biol Markers 1998;13:179-187.
confirmed and sustained for at least 20 days. Patients can 3 Bast RC, Klug TL, St John E, Jenison E, Niloff
only be evaluated if they have a CA 125 concentration in a JM, Lazarus H et al. A radioimmunoassay using a
pretreatment sample that is twice the upper reference limit monoclonal antibody to monitor the course of
and taken 2 weeks prior to initiation of therapy. Additional epithelial ovarian cancer. N Engl J Med
samples are recommended at 2 to 4 weeks during 1983;309:883-887.
treatment and at intervals of 2 to 3 weeks during follow- 4 Davelaar EM, van Kamp GJ, Verstraeten RA,
up. The same method should be used to monitor the Kenemans P. Comparison of seven
patient throughout, and patients who receive immunoassays for the quantification of CA 125
immunotherapy (mouse antibodies) cannot be evaluated. antigen in serum. Clin Chem 1998;44:1417-1422.
Serial measurement of CA 125 to aid in monitoring 5 Sakahara H, Kousaka T, Hattori N, Tomida K,
therapeutic response is approved by the FDA. Yao Z, Hosono M et al. Dissociation in serum
CA125 concentrations measured by different
Pre- and postoperative CA 125 concentrations may be of monoclonal antibodies. Gynecol Oncol
prognostic significance [27-29]. After primary surgery and 1994;52:301-305.
chemotherapy, persistent elevations of CA 125 6 van Kamp GJ, Verstraeten AA, Kenemans P.
concentrations were associated with poor prognosis. Discordant serum CA 125 values in commercial
Patients who had preoperative CA 125 concentrations > 65 immunoassays. Eur J Obstet Gynecol Reprod
kU/L were reported to have a lower 5-year survival rate Biol 1993;49:99-103.
and a 6.37-fold risk of death as compared to patients who 7 Kenemans P, Bon GG, Kessler AC, Verstraeten
had CA 125 levels < 65 kU/L [24]. The half-life of CA AA, van Kamp GJ. Multicenter technical and
125 antigen was reported to have an additional prognostic clinical evaluation of a fully automated enzyme
value. A half-life of < 20 days was associated with immunoassay for CA 125. Clin Chem
improved survival as compared to half-life > 20 days [30]. 1992;38:1466-1471.
Normalization of CA 125 levels after three cycles of 8 Bonfrer JM, Baan AW, Jansen E, Lentfer D,
combination therapy also correlates with improved Kenemans P. Technical evaluation of three
survival. second generation CA 125 assays. Eur J Clin
Chem Clin Biochem 1994;32:201-207.
It should be noted that CA 125 concentration is not
elevated in 10% to 20% of patients with advanced ovarian
306
9 Kenemans P, Verstraeten AA, van Kamp GJ, von serum CA 125 levels in patients with primary
Mensdorff-Pouilly S. The second-generation CA epithelial ovarian cancer. Obstet Gynecol
125 assays. Ann Med 1995;27:107-113. 1986;67:414-416.
10 U.S. Food and Drug Administration website. 23 Chan DW, Shih LM, Sokoll LJ, Bast RC, Jr.
<http://www.accessdata.fda.gov/scripts/cdrh/cfdo National Academy of Clinical Biochemistry
cs/cfCLIA/search.cfm> Guidelines for the use of tumor markers in
11 National Academy of Clinical Biochemistry: ovarian cancer: practice guidelines and
Practice Guidelines and Recommendations for recommendations for use of tumor markers in the
Use of Tumor Markers. In: The Clinic, 2006. clinic, 2006. Available at
Available at <http://www.aacc.org/AACC/members/nacb/LM
<http://www.aacc.org/AACC/members/nacb/LM PG/OnlineGuide/DraftGuidelines/TumorMarkers
PG/OnlineGuide/DraftGuidelines/TumorMarkers /> Accessed 10.7.2007.
/> Accessed 10.7.2007. 24 Meyer T, Rustin GJ. Role of tumour markers in
12 Klug TL, Bast RC Jr, Niloff JM, Knapp RC, monitoring epithelial ovarian cancer. Br J Cancer
Zurawski VR Jr. Monoclonal antibody immuno- 2000;82:1535-1538.
radiometric assay for an antigenic determinant 25 Rustin GJ. Can we now agree to use the same
(CA 125) associated with human epithelial definition to measure response according to CA-
ovarian carcinomas. Cancer Res 1984;44:1048- 125? J Clin Oncol 2004;22:4035-4036.
1053. 26 Rustin GJ, Quinn M, Thigpen T, du Bois A,
13 Duffy MJ, Bonfrer JM, Kulpa J, Rustin GJ, Pujade-Lauraine E, Jakobsen A et al. Re: New
Soletormos G, Torre GC et al. CA 125 in ovarian guidelines to evaluate the response to treatment in
cancer: European Group on Tumor Markers solid tumors (ovarian cancer). J Natl Cancer Inst
guidelines for clinical use. Int J Gynecol Cancer 2004;96:487-488.
2005;15:679-691. 27 Sjovall K, Nilsson B, Einhorn N. The
14 Pauler DK, Menon U, McIntosh M, Symecko significance of serum CA 125 elevation in
HL, Skates SJ, Jacobs IJ. Factors influencing malignant and nonmalignant diseases. Gynecol
serum CA125II levels in healthy postmenopausal Oncol 2002;85:175-178.
women. Cancer Epidemiol Biomarkers Prev 28 Cooper BC, Sood AK, Davis CS, Ritchie JM,
2001;10:489-493. Sorosky JI, Anderson B, Buller RE. Preoperative
15 Jacobs I, Bast RC Jr. The CA 125 tumour- CA 125 levels: an independent prognostic factor
associated antigen: a review of the literature. for epithelial ovarian cancer. Obstet Gynecol
Hum Reprod 1989;4:1-12. 2002;100:59-64.
16 Fleisher M, Dinistrian A, Sturgeon C, Lamerz R, 29 Gadducci A, Cosio S, Fanucchi A, Negri S,
Witliff J. Practice guidelines and Cristofani R, Genazzani AR. The predictive and
recommendations for use of tumor markers in the prognostic value of serum CA 125 half-life
clinic. In: Diamandis EP, Fritsche H, Lilja H, during paclitaxel/platinum-based chemotherapy
Chan DW, Schwartz M, eds. Tumor Markers: in patients with advanced ovarian carcinoma.
Physiology, Pathobiology, Technology, and Gynecol Oncol 2004;93:131-136.
Clinical Applications. Washington, DC: AACC 30 Verheijen RH, von Mensdorff-Pouilly S, van
Press; 2002:33-63. Kamp GJ, Kenemans P. CA 125: fundamental
17 Tso E, Elson P, Vanlente F, Markman M. The and clinical aspects. Semin Cancer Biol
real-life variability of CA-125 in ovarian 1999;9:117-124.
cancer patients. Gynecol Oncol 2006;103:141- 31 Westgard QC website.
144. <http://www.westgard.com/biodatabase1.htm>
18 National Institutes of Health Consensus 32 European Group on Tumor Markers website.
Conference. NIH Consensus Development Panel <http://egtm.eu/1280.html>
on Ovarian Cancer. Ovarian cancer: screening,
treatment, and follow-up. JAMA 1995;273:491-
497.
19 Ozols RF. Update of the NCCN ovarian cancer
practice guidelines. Oncology (Williston Park)
1997;11:95-105.
20 Bonfrer J, Duffy M, Radke M, Segurado O, Torre
G, Van Dalen A et al. European Group on Tumor
Markers. Tumor markers in gynaecological
cancers: EGTM recommendations. Anticancer
Res 1999;19:2807-2810.
21 ESMO minimum clinical recommendations for
diagnosis, treatment and follow-up of ovarian
cancer. Ann Oncol 2001;12:1205-1207.
22 Einhorn N, Bast RC Jr, Knapp RC, Tjernberg B,
Zurawski VR Jr. Preoperative evaluation of
307
Carbamazepine
Carbamazepine
Gus Koerbin and Julia M. Potter
Name: Carbamazepine, 5-carbamoyl-5H-dibenz[b,f]azepine

Clinical Significance: Refer to Chapter 47, Nervous System, in the 5th edition of Clinical
Molecular formula: C15H12N2O
Merck Index: 1758
Chemical class: Tricyclic iminodibenzene
Carbamazepine
CONH2
Structure:
After TLC separations, fluorescence can be induced by

Principles of Analysis and Current Usage treatment of the chromatogram with 70% perchloric acid
Carbamazepine is the generic name for 5-carbamoyl-5H- and heat. An excitation wavelength of 358 nm and an
dibenz[b,f]azepinei. It is a derivative of the tricyclic emission wavelength of 498 nm were used by
iminodibenzyl nucleus (Figure 1). The drug is used to Christiansen [4]. Meilink [5] reported a technique using
treat epilepsy. A metabolite, carbamazepine 10,11- ammonium cerium(IV) sulfate and phosphoric acid
epoxide (CBZ-E), found in the plasma, is reagent to induce fluorescence. For quantitative
pharmacologically active and is increased with ultraviolet spectrophotometric assays, extracted samples
concurrent use of other anticonvulsive medication [1]. were measured at the absorbance of the secondary peak
wavelength, 290 nm, rather than at the maximum
Methods for the measurement of carbamazepine in absorbance of 215 to 220 nm, to eliminate major
biological fluids include spectrophotometry with interferences from other commonly administered
ultraviolet radiation [2], visible light [3], fluorometry, anticonvulsants. The visible-light spectrophotometric
often combined with thin-layer chromatography [4-6], method of Herrman [3] extracted carbamazepine with
gas-liquid chromatography methods [7], and high- dichloroethane at a neutral pH. The extract was then
performance liquid chromatography methods [8-13]. treated with sodium nitrite and nitric acid to produce a
soluble yellow compound whose absorbance at 400 nm
Thin-layer chromatography (TLC) (Table 1, Method 1) was proportional to the level of carbamazepine.
quantitative ultraviolet spectrophotometric assays (Table Spectrophotometric scans of carbamazepine and
1, Method 2) and visible-light spectrophotometric carbamazepine epoxide are shown in Figure 2.
methods (Table 1, Method 3) are of historical interest.
Immunoassay techniques are most commonly used today
for the quantitative measurement of carbamazepine. The
i
Carbamazepine previously under Anticonvulsants homogeneous enzyme-multiplied immunoassay
Previous and current authors of this method: technique (EMIT) (Table 1, Method 4) employs a
First edition: Steven J. Soldin derivative of carbamazepine. In this derivative, the
Methods edition: Elizabeth B. Solow carbamazepine is attached to the enzyme glucose-6-
Second edition: Steven J. Soldin phosphate dehydrogenase. When antibody to the drug
Third edition: Paul Salm, Paul J. Taylor, Julia M. binds to the conjugate, the enzymatic activity is
Potter inhibited. Drug present in the patient sample competes
Fourth edition: Paul Salm, Paul J. Taylor, Julia M. with the labeled drug for the antibody binding sites. The
Potter more drug present in the sample, the less antibody is
Fifth edition: Gus Koerbin and Julia M. Potter
308
Carbamazepine
available to react with the drug-enzyme complex, and

the greater the enzymatic activity observed [9,14]. A Reference and Preferred Methods
homogeneous substrate-labeled fluorescence assay There is no reference method for carbamazepine.
(SLFIA) procedure (Table 1, Method 5) [15-17], a drug Ultraviolet spectrophotometry is not currently used for
derivatized with the fluor umbelliferone, attached to the the analysis of carbamazepine, since it is not specific for
drug through a galactoside linkage, is used as the the drug, and the extraction procedures are labor
indicator molecule. The drug derivative binds to an intensive. Colorimetric procedures are not recommended
antibody specific for the drug. In the presence of drug in for the same reason. Fluorometric determination of
the patient sample, the galactoside derivative of the drug carbamazepine and its metabolites by TLC are also labor
is not bound by the antibody and is free to be cleaved by intensive and subject to considerable imprecision. The
the enzyme beta-galactosidase, present in the imprecision is greater than by other techniques, and the
homogeneous reaction mixture, to release umbelliferone. level of technical expertise necessary is greater than can
The fluorescent umbelliferone formed is readily be easily achieved by most laboratories.
measured by fluorometry. In the absence of drug, the
antibody binds to the labeled drug and prevents Immunoassay procedures are very easy to perform and
cleavage, resulting in low fluorescence. In the are well suited for measurement of carbamazepine in the
fluorescence polarization assay, FPIA, a fluorescent drug therapeutic range. Immunoassay procedures are the
derivative (usually fluorescein labeled) is bound by current methods of choice for laboratories performing
antibody (Table 1, Method 6). The bound drugfluor has therapeutic drug monitoring, but consideration should be
restricted movement, allowing incident polarized light to given to assays that measure the metabolite CBZ-E as
be re-emitted as highly polarized fluorescence. In the well. According to survey data from the 2007 College of
presence of drug in the patient sample, more of the drug- American Pathologists Participant Summary Report,
fluor remains unbound, allowing more rapid movement only immunoassays are currently used for measurement
and resulting in the emission of fluorescent light that is of carbamazepine. All of these methods are acceptable,
less polarized. The decrease in polarized fluorescence so methods for carbamazepine assay are determined by
can be used to quantitate the concentration of the choice of instrumentation in the laboratory.
carbamazepine.
Although it has adequate sensitivity, gas-liquid
Gas-liquid chromatography has been widely used for chromatography is not free from problems. The majority
carbamazepine determination (Table 1, Method 7). Early of these problems are the result of decomposition
methods used derivatization techniques (Table 1, products from high assay temperatures or
Method 7a). More recent procedures, which use chromatographic column conditions (Figure 3A).
deactivated column packings, do not require Derivatization of the compound can minimize these
derivatization of the drug to achieve good resolution. difficulties [19,20,22]. The addition of an internal
Most GLC methods measure only the parent drug [18- standard with a related chemical structure enhances the
26], although the procedure of Morselli et al. [26] accuracy of measurement. Figure 3B illustrates the
measured both the parent drug and its epoxy metabolite. structure of possible standards. Newer technology and
To increase sensitivity further, the combination of GLC deactivating column packings allows carbamazepine
and mass spectrometry has been used. determinations without derivatization and minimizes
breakdown products [25]. Using GLC in conjunction
High-performance liquid chromatography (HPLC) with mass spectrometry detection, Palmer et al. [27]
(Table 1, Method 8) is becoming increasingly popular obtained low-nanogram-range sensitivity. Other drugs
for the measurement of both carbamazepine and the can be analyzed simultaneously, but the epoxide
epoxide, since the current state of column technology metabolite of carbamazepine is unstable, and its
has alleviated the problem of thermal instability. measurement is unpredictable with either the flame
Detection is by ultraviolet spectrophotometry (212 and ionization detector or the mass spectrometer.
254 nm), and several internal standards, such as
dihydrocarbamazepine and cyheptamide, have been used
High-performance liquid chromatography seems to be
to improve analysis. Rambeck described a liquid
the procedure that offers the most analytically accurate
chromatographic method for quantitating antiepileptic
estimation of carbamazepine, since it measures both
drugs, including carbamazepine and its epoxide [12],
parent compound and the 10,11-epoxide. The precision
using a reversed-phase column requiring an analysis
of HPLC assays is 4% to 5% in the therapeutic range. A
time of 22 min. Eichelbaum and Bertilsson [10]
study of Mihaly et al. [9] showed a better correlation of
presented a high-performance liquid chromatography
dosage with the serum 10,11-epoxide concentrations
method using a Carbowax 400-Corasil column and
than with the parent carbamazepine. The minimum
monitoring the eluent absorbance at a nonspecific
detectable quantity by HPLC for both the parent and the
wavelength of 254 nm. They reported that a minimum of
metabolite is 0.8 ng [13].
1 mL of plasma was required to measure the drug and its
metabolite concurrently. A high-performance liquid
Most reports demonstrate good correlations between the
chromatography method was developed by Sawchuk and
assay systems EMIT, TDx, GLC, and HPLC for serum
Cartier [13] for measuring both carbamazepine and the
values. This is true where the major component present
epoxide simultaneously in a single extract using
in the patient samples is carbamazepine. The correlation
cyheptamide as an internal standard.
may not be as good in the presence of metabolites.
309
Carbamazepine
Active CBZ-E is not routinely monitored, partly because major site of action for the biotransformation enzymes.
an HPLC method is needed to measure the metabolite The biologically active metabolite carbamazepine 10,11-
[28]. However, for laboratories without HPLC epoxide (Figure 3A), is found in substantial
technology, immunoassays will give adequate results for concentration in plasma. It has been shown to have both
most patients. The extent of cross-reactivity of CBZ-E anticonvulsant activity [40] and antineuralgic properties
with common immunoassays has been demonstrated by [41]. The effectiveness and toxicity of carbamazepine
Hermida and Tutor [29]. may best be correlated with the combined total plasma
levels of the parent drug and its major metabolite.
Specimen
Either serum or plasma samples are acceptable for The apparent plasma half-life of carbamazepine has been
analysis. There are no known interferences from reported to range from 8 to 60 hours. The plasma half-
commonly used anticoagulants, commercial vacuum life of 10,11-epoxide is shorter than that of the parent
tubes, or tubes with inert polymer gel barrier for serum compound. Protein binding is reported to average 70% to
separation. The free fraction has been shown to be 80% for the parent drug and 50% to 60% for the 10,11-
higher in plasma (26%) than serum (19%) [30]. epoxide.
If a sample cannot be analyzed the same day as the blood
was drawn from the patient, short-term storage at room Toxic effects are frequent with concentrations over 12
temperature or normal refrigeration is satisfactory. When mg/L. Cases of overingestion of carbamazepine and a
the sample is frozen at 10C, no deterioration has been few postmortem cases have been reported [32].
observed during a 6-month period. Specimens stored on Interaction with erythromycin was shown to increase
gel-barrier tubes showed a decrease compared to carbamazepine levels in epileptic children [42] and to
specimens not stored on gel [31]. have associated toxicity.
Interferences McKauge, Tyrer, and Eadie [34] demonstrated a drug

None of the commercially available immunoassays are interaction involving valproic acid and carbamazepine.
specific for the parent compound, and all cross-react to a The co-administration of valproic acid and
greater or lesser degree with the 10,11-epoxide. The carbamazepine resulted in elevated plasma
carbamazepine 10,11-epoxide cross-reacts with the concentrations of the epoxide while leaving plasma
currently available polyclonal antibodies. The carbamazepine concentration unaltered. Levy et al. [43]
immunoassays also cross-react with the tricyclic have also reported a drug interaction involving valproic
antidepressants; however, since these drugs are present acid and carbamazepine in both man and rhesus monkey.
in much lower concentrations than carbamazepine is, In man, they reported, carbamazepine levels were
this does not appear to be of analytical significance. reduced, and the ratio of 10,11-epoxide to
Hemolysis and lipemia have been shown to interfere in carbamazepine was increased. With the potential for
some immunoassay procedures. interaction with other drugs, measurement of both the
parent drug and its metabolite are indicated. Post et al.
Carbamazepine Reference Interval [44,45] reported that levels of 10,11-epoxide in both
The effective plasma carbamazepine level is 4 to 12 plasma and CSF were better correlated with the dose of
mg/L (16.9 to 50.8 mol/L). Some patients seem to carbamazepine than were the levels of carbamazepine
benefit from lower levels [32,33]. Furthermore, the itself.
evaluation is complicated by the presence of 10,11-
epoxide and drug interactions. McKauge et al. [34] Carbamazepine Performance Goals
reported that over the usual concentration range of Survey data from the 2007 College of American
plasma carbamazepine in 295 subjects, values Pathologists Participant Summary Report shows
encountered for the epoxide level averaged 20% to 25% imprecision values (% coefficient of variation [CV]) for
of the parent drug levels. Hundt et al. [35] summarized 8 measurement of acetaminophen range from 3.9% to
years of monitoring carbamazepine and its metabolites 9.5% at concentrations of approximately 6.5 g/mL and
and encouraged the routine assessment of these levels. from 3.2% to 9.6% at about 13.5 g/mL for assays with
The relationship of age, dose, and associated therapy > 20 participants. Acceptable performance criteria for
complicates interpretations [36,37]. measurement of carbamazepine (CLIA 88) requires that
laboratories be accurate to within 25% of the peer-
Interpretation group mean.
Carbamazepine has been widely used for the treatment
of certain types of epilepsy, trigeminal neuralgia, and Desirable specifications for analytical imprecision
other clinical ailments, including depression and mania. derived from studies of biological variation indicate an
The pharmacological properties and the therapeutic use assay imprecision of no greater than 0.5 g/mL and 0.77
of carbamazepine have been reviewed extensively mg/mL at decision levels of 8.0 and 12.0 g/mL [46].
[38,39].
The 2007 CAP data show that there is a lack of
The double bond between positions 10 and 11 in the harmonization between commonly used carbamazepine
tricyclic nucleus is somewhat unstable and provides the
310
Carbamazepine
assays, and there are differences in concentration 15 Jolley ME, Stroupe SD, Wang CH, Panas HN,
between these assays of greater than 25%. Keegan CL, Schmidt RL Schwenzer KS.
Fluorescence polarization immunoassay. I.
References Monitoring aminoglycoside antibiotics in serum
1 Potter JM, Donnelly A. Carbamazepine-10,11- and plasma. Clin Chem 1981;27:1190-7.
epoxide in therapeutic drug monitoring. Ther 16 Popelka SR, Miller DM, Holen JT, Kelso DM.
Drug Monit. 1998;20:652-7. Fluorescence polarization immunoassay. II.
2 Morselli PL, Gerna M, Garattini S. Analyzer for rapid precise measurement of
Carbamazepine plasma and tissue levels in the fluorescence polarization with use of disposable
rat. Biochem Pharmacol 1971;20:2043-7. cuvettes. Clin Chem 1981;27:1198-1201.
3 Herrman B. Tegretol G32883: Determination in 17 Loomis KF, Frye RM. Evaluation of the Abbott
Serum, Plasma, and CSF. Ardsley, NY: Geigy TD for stat measurement of phenobarbital,
Pharmaceuticals; 1966. phenytoin, carbamazepine, and theophylline.
4 Christiansen J. Assay of carbamazepine and Am J Clin Pathol 1983;80:686-91.
metabolites in plasma by quantitative thin-layer 18 Larsen NE, Wendelboe J, Bohn L.
chromatography. In: Meijer, JWA, Meinardi H, Determination of carbamazepine (Tegretol) by
Gardner-Thorpe C, van der Kleijn E, eds. gas chromatography. Twelfth meeting of the
Methods of Analysis of Antiepileptic Drugs. Scandinavian Society for Clinical Physiology,
Amsterdam: Excerpta Medica; 1973:87-90. Otaniemi, Finland, August 11-15, 1969. Scand J
5 Meilink JW. Fluorometric assay of Clin Lab Invest 1969;110(suppl):35.
carbamazepine and its metabolites in blood. 19 Kupferberg HJ. GLC determination of
Pharm Weekbl 1974;109:22-50. carbamazepine in plasma. J Pharm Sci
6 Faber DB, Man in TVeld WA. A thin-layer 1972;61:284-6.
chromatographic method for determining 20 Roger JC, Rodgers G, Soo A. Simultaneous
carbamazepine in blood. J Chromatogr determination of carbamazepine (Tegretol)
1974;93:238-42. and other anticonvulsants in human plasma by
7 Rambeck B, Meijer JWA. Gas chromatographic gas-liquid chromatography. Clin Chem
methods for the determination of antiepileptic 1973;19:590-2.
drugs: a systematic review. Ther Drug Monit 21 Friel P, Green JR. Quantitative assay for
1980;2:385-96. carbamazepine serum levels by gas-liquid
8 Kutt H, Kutt HP. Carbamazepine: chemistry chromatography. Clin Chim Acta 1973;43:69-
and methods of determination. In: Woodbury 72.
DM, Penry JK, Pippinger CE, eds. Antiepileptic 22 Perchalski RJ, Wilder BJ. Rapid gas-liquid
Drugs 2nd ed. New York: Raven Press; chromatographic determination of
1982:453-63. carbamazepine in plasma. Clin Chem
9 Mihaly GW, Phillips JA, Louis WJ, Vajda FJ. 1974;20:492-3.
Measurement of carbamazepine and its epoxide 23 Gerardin A, Abadie F, Laffont J. GLC
metabolite by high-performance liquid determination of carbamazepine suitable for
chromatography, and a comparison of assay pharmacokinetic studies. J Pharm Sci
techniques for the analysis of carbamazepine. 1975;64:1940-2.
Clin Chem 1977;23:2283-7. 24 Sheehan M, Beam RE. GLC determination of
10 Eichelbaum M, Bertilsson L. Determination of underivatized carbamazepine in whole blood. J
carbamazepine and its epoxide metabolite in Pharm Sci 1975;64:2004-6.
plasma by high-speed liquid chromatography. J 25 Bius DL, Teague BL, Dudley KH. Gas
Chromatogr 1975;103:135-40. chromatographic determination of
11 Kabra PM, Morton LJ. Determination of carbamazepine in human plasma. Ther Drug
carbamazepine in blood or plasma by high- Monit 1979;1:525-44.
pressure liquid chromatography. Clin Chem 26 Morselli PL, Biandrate P, Frigerio A. Gas
1976;22:1070-1072. chromatographic determination of
12 Rambeck B, Riedmann M, Meijer JW. carbamazepine and carbamazepine 10,11-
Systematic method of development in liquid epoxide in human body fluids. In: Meijer JWA,
chromatography applied to the determination of Meinardi H, Gardner-Thorpe C, van der Kleihn
antiepileptic drugs. Ther Drug Monit E, eds. Methods of Analysis of Antiepileptic
1981;3:377-95. Drugs. Amsterdam: Excerpta Medica;
13 Sawchuk RJ, Cartier LL. Simultaneous liquid- 1978:169-75.
chromatographic determination of 27 Palmer L, Bertilsson L, Collste P, Rawlins M.
carbamazepine and its epoxide metabolite in Quantitative determination of carbamazepine in
plasma. Clin Chem 1982;28:2127-30. plasma by mass fragmentography. Clin
14 Sun L, Szafir I. Comparison of enzyme Pharmacol Ther 1973;15:827-32.
immunoassay and gas chromatography for 28 Warner A, Privitera M, Bates D. Standards of
determination of carbamazepine and laboratory practice: antiepileptic drug
ethosuximide in human serum. Clin Chem monitoring. Clin Chem 1998; 44:1085-95.
1997;23:1753-6.
311
Carbamazepine
29 Hermida J, Tutor JC. How suitable are currently Antiepileptic Drugs. 2nd ed. New York: Raven
used carbamazepine immunoassays for Press; 1982:465-82.
quantifying carbamazepine-10,11-epoxide in 39 Faigle JW, Feldmann KF. Carbamazepine:
serum samples? Ther Drug Monit 2003;25:384- biotransformation. In: Woodbury DM, Penry
8. JK, Pippinger CE eds. Antiepileptic Drugs. 2nd
30 Ohshima T, Hasegawa T, Johno I, Kitazawa S. ed. New York: Raven Press; 1982:483-95.
Variations in protein binding of drugs in plasma 40 Tomson T, Almkvist O, Nilsson BY, Svensson
and serum. Clin Chem 1989;35:1722-5. JO, Bertilsson L. Carbamazepine-10,11-epoxide
31 Dasgupta A, Dean R, Saldana S, Kinnaman G, in epilepsy. A pilot study. Arch Neurol
McLawhorn RW. Absorption of therapeutic 1990;47:888-92.
drugs by barrier gels in serum separator blood 41 Moosa RS, McFadyan ML, Miller R, Rubin J.
collection tubes. Am J Clin Pathol Carbamazepine and its metabolites in
1994;101:456-61. neuralgias: concentration-effect relations. Eur J
32 Baselt RC. Carbamazepine. In: Baselt RC, ed. Clin Pharmacol 1993;45:297-301.
Disposition of Toxic Drugs and Chemicals in 42 Hedrick R, Williams F, Morin R, Lamb WA,
Man. 2nd ed. Davis, CA: Biomedical Cate JC. Carbamazepine-erythromycin
Publications; 1982:113-7. interaction leading to carbamazepine toxicity in
33 Kutt H. Clinical pharmacology of four epileptic children. Ther Drug Monit
carbamazepine. In: Pippinger CE, Penry JK, 1983;5:405-7.
Kutt H, eds. Antiepileptic Drugs: Quantitative 43 Levy RH, Moreland TA, Morselli PL, Guyot
Analysis and Interpretations. New York: Raven M, Brachet-Liermain A, Loiseau P.
Press; 1978:297-305. Carbamazepine/valproic acid interaction in man
34 McKauge L, Tyrer JH, Eadie MJ. Factors and rhesus monkey. Epilepsia 1984;25:338-45.
influencing simultaneous concentrations of 44 Post RM, Uhde TW, Ballenger JC, Chatterji
carbamazepine and its epoxide in plasma. Ther GC, Greene RF, Bunney WE Jr.
Drug Monit 1981;3:63-70. Carbamazepine and its 10,11-epoxide
35 Hundt HKL, Aucamp AK, Muller FO, Potgieter metabolite in plasma and CSF. Arch Gen
MA. Carbamazepine and its major metabolites Psychiatr 1983;40:673-6.
in plasma: a summary of eight years of 45 Post RM, Ballenger JC, Uhde TW, Smith C,
therapeutic drug monitoring. Ther Drug Monit Rubinow DR, Bunney WE Jr. Efficacy of
1983;5:427-35. carbamazepine in manic-depressive illness:
36 Battino D, Bossi L, Croci D, Francaschetti S, implication for underlying mechanisms. In: Post
Gomeni C, Moise, A, Vitali A. Carbamazepine RM, Ballenger JC eds. Neurobiology of mood
plasma levels in children and adults: influence disorders. Baltimore: William & Wilkins; 1984.
of age, dose, and associated therapy. Ther Drug 46 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
Monit 1980;2:315-22. Hermandez A, Jimunez CV et al. Current
37 Paxton JW, Aman MG, Werry JS. Fluctuations databases on biologic variation: pros, cons and
in salivary carbamazepine and carbamazepine progress. Scand J Clin Lab Invest 1999;59:491-
10,11-epoxide concentrations during the day in 500.
epileptic children. Epilepsia 1983;24:716-24.
38 Morselli PL, Bossi L. Carbamazepine:
absorption, distribution, and excretion. In:
Woodbury DM, Penry JK, Pippinger CE eds.
312
Carbamazepine
Tables
Table 1: Methods of Carbamazepine Analysis
Method 1: Thin-layer chromatography; qualitative
Principle of analysis: Drug extracted from matrix, chromatographed by TLC, and detected by reaction with
perchloric acid and heat
Comments: Rare; not as accurate as other methods
Method 2: Spectrophotometric; quantitative
Principle of analysis: Drug extracted from matrix, quantified by its ultraviolet absorption at specific
wavelengths, such as 290 or 255 nm
Comments: Historic; interference from other compounds
Method 3: Colorimetric; quantitative
Principle of analysis: Drug extracted from matrix, reacted with nitrous acid to make a yellow nitrogen derivative
measured at 400 nm
Comments: Rare; nonspecific colorimetric reaction
Method 4: Enzyme multiplied immunoassay technique (EMIT); quantitative
Principle of analysis: Patient drug competes for antibody with enzyme-labeled drug; enzyme activity
proportional to concentration
Comments: Common; simple; however, does not measure epoxide separately; hemolysis a potential problem
Method 5: Substrate-labeled fluorescence immunoassay (SLFIA); quantitative
Principle of analysis: Patient drug competes for antibody with drug-galactoside derivative; unbound drug
derivative hydrolyzed to fluor by exogenous galactosidase
Comments: Uncommon; as above
Method 6: Fluorescence polarization immunoassay; quantitative
Principle of analysis: Patient drug competes for antibody with drug fluorescein derivative; unbound drug has
lower polarized fluorescence than bound drug
Comments: Common
Method 7: Gas chromatography; quantitative
a. Gas-liquid derivatized
b. Gas-liquid underivatized
a. Drug extracted into organic solvent, converted to a derivative, detected by flame ionization or
nitrogen detector
b. As above, special deactivated columns that do not react with drug are used; has been used with a mass
spectrograph detector
Comments:
a. Uncommon; labor intensive; drug and metabolite subject to thermal decomposition
b. Uncommon; labor intensive; metabolite subject to thermal decomposition
Method 8: High-performance liquid chromatography (HPLC); quantitative
Principle of analysis: Drug extracted, separated by chromatography, detected by ultraviolet absorbance
Comments: Common; detects parent drug and metabolite; method of choice
Figure 1: Carbamazepine and its precursors.

313
Carbamazepine
Figure 2: Spectrophotometric scans

A, Carbamazepine. B, Carbamazepine 10,11-epoxide. C, Dihydrocarbamazepine, internal standard.
Figure 3: A, Carbamazepine and its metabolites and breakdown products. B, Structures of potential
internal standards.
314
Carbamazepine
Figure 4: Chromatograms from HPLC instrumentation.

A, Plasma or serum blank. B, Plasma plus internal standard (IS). C, Drug-free plasma spiked with
carbamazepine (CBZ), 8 mg/L, and its 10,11-epoxide, 2 mg/L. D, Patient sample with both metabolite
and parent drug. Internal standard is at 6.6 min, carbamazepine is at 5.5 min, and its epoxide is at 2.6
min.
Figure 5: Chromatograms from GLC instrumentation.

A, Plasma blank. B, Plasma plus internal standard. C, Drug-free plasma spiked with carbamazepine, 4
mg/L. D, Patient sample with carbamazepine therapy. Internal standard is at 2.5 min, and the
carbamazepine is at 2.9 min. FID, Flame ionization detector.
315
Carbamazepine
prevent protein precipitation. This standard contains 4

Procedure: High-Performance Liquid g of carbamazepine and 1 g of epoxide per milliliter
Chromatography of plasma. Aliquot 1 mL volumes of the standards to
polystyrene tubes, and seal tightly for storage. Stable for
Principle
at least 6 months when frozen at 20C.
The method described below is based on an alkaline
8. Extraction solvent with
solvent extraction with an analog dihydrocarbamazepine
dihydrocarbamazepine internal standard (4.20
included as an internal standard. The residue is
reconstituted in the mobile phase for HPLC analysis. mol/L). Prepare benzene/ethyl acetate, 4:1 (v/v), to
Detection and quantitation are achieved by ultraviolet contain 1 mg of dihydrocarbamazepine per liter. Deliver
1 mL of the stock dihydrocarbamazepine to a 1-L
spectroscopy at 212 nm [8]. The lower limit of detection
volumetric flask. Fill to the mark with benzene/ethyl
for the drug or its metabolite is 0.1 g/mL without
modification of procedure. (See HPLC chromatograms, acetate, and mix well. This solvent now contains 1 g of
Figure 4.) the internal standard per milliliter. When stored at room
temperature, this compound is stable for several months.
Reagents 9. Control serum or plasma. Prepare drug-free
1. Saturated trisodium phosphate buffer. Add plasma or serum with carbamazepine and its epoxide so
190 g of hydrated trisodium phosphate (Na3PO4 that the concentrations are 10 mg/L (42.3 mol/L) for
12H2O) to 500 mL of distilled water. Heat to dissolve the parent drug and 1.5 mg/L (6.0 mol/L) for the
epoxide metabolite. Aliquot 1-mL volumes into
the crystals. Cool to room temperature. Excess crystals polystyrene tubes and seal. Stable for at least 6 months
form, showing that the solution is saturated. Store at when frozen at 20C.
room temperature for up to 2 years.
2. Benzene/ethyl acetate, 4:1 (NOTE: When Assay
using benzene, follow usual safety precautions and work Equipment: A high-performance liquid chromatograph.
under fume hood.) The instrument should be equipped with a variable
3. Methanol. HPLC grade. Filter through wavelength ultraviolet spectrophotometer. The column
Millipore, type FH filter, with vacuum. should be a reversed-phase octadecyldimethylsilyl
4. Stock carbamazepine standard, 2 g/L (8.46 (ODS) on 5 m spherical silica particles.
mmol/L). Dissolve 10 mg of carbamazepine in 5 mL of 1. Label appropriate number of 5-mL, glass-
methanol. Stable for at least 6 months when stored stoppered, conical tubes for standards, control,
tightly sealed at normal refrigeration temperature of 4C. and unknowns.
5. Stock carbamazepine 10,11-epoxide, 1 g/L
2. To each tube, add 0.5 mL of the respective
(4.0 mmol/L). Dissolve 5 mg of dihydrocarbamazepine
plasma or serum samples.
in 5 mL of methanol. Stability is the same as described
above for the parent drug carbamazepine. 3. Add 250 L of saturated trisodium phosphate
6. Stock dihydrocarbamazepine, 1 g/L (4.20 buffer solution.
mmol/L). Dissolve 5 mg of dihydrocarbamazepine in 5 4. Add 2 mL of the extraction solvent containing
mL of methanol. Stable for at least 12 months when internal standard.
stored at 4C. 5. Mix with vortex-mixing device, and centrifuge at
7. Carbamazepine and epoxide plasma 500 g for 5 min.
standards. 6. Remove solvent layer to a conical 3 mL tube, and
evaporate to dryness at 40C to 60C with a flow
Carbamazepine (33.9 mol/L) and epoxide (8.0 of nitrogen.
mol/L). To a 50-mL volumetric flask, deliver drug-free 7. Dissolve residue in an appropriate volume of
plasma to approximately half-full level. Deliver 200 L diluent. For HPLC automatic injection system,
of the stock carbamazepine standard and 100 L of the add 200 L of the mobile phase methanol/water,
stock epoxide standard to the flask. Add each solution 45:55, v/v, and transfer to vials for 0.1 mL
slowly and carefully, with mixing, to prevent volume. Using HPLC automatic sampling
precipitation of protein or coagulation, which would system, inject 10 L (equivalent to 0.025 mL of
hinder solubility of the compounds. Add a minute drop serum or plasma).
of caprylic alcohol to prevent foam. Fill to the 50 mL 8. HPLC parameters:
mark with plasma, and mix thoroughly. This standard Temperature. Ambient temperatures for column and
contains 8 g of carbamazepine and 2 g of the epoxide mobile phase are satisfactory, but if temperature control
per milliliter of plasma. devices are available, the column oven temperature is
40C, and the mobile phase reagent temperature is 30C.
Carbamazepine (16.9 mol/L) and epoxide (4.0 Mobile phase. Methanol/water, 45:55, v/v
mol/L). Similarly to drug-free plasma, in a 50-mL Flow rate. 2.5 mL/min
volumetric flask, deliver 100 L of the stock Ultraviolet wavelength detector. 212 nm
carbamazepine and 50 L of the epoxide. Fill to the Attenuation of integrator. Set so that 200 ng of
mark with plasma, and mix thoroughly. Use the carbamazepine gives a peak two thirds the full scale on
precautions described above to prevent foam and to the chart recorder.
316
Carbamazepine
Chart speed. 0.5 cm/min 7. Carbamazepine and epoxide plasma

Carbamazepine: Figure 3 shows the absorbance standards. 33.9 and 8.0 mol, respectively, and 2, 16.9,
spectra for carbamazepine 10,11-epoxide and and 4.0 mol, respectively.
dihydrocarbamazepine (internal standard). Examples of 8. Extraction solvent with internal standard.
typical chromatograms are illustrated in Carbamazepine: 9. Control serum or plasma.
Figure 4.
Assay
Calculations Equipment: Gas chromatograph equipped with a
1. Measure the peak height (in millimeters) of the hydrogen flame detector. The columns are 6 ft 2 mm
drug, carbamazepine 10,11-epoxide, and internal (inner diameter) silanized glass. The column packing is
standard. 2% SP-2510 DA on 100/1209 mesh Supelcoport.
2. Compute the ratio of the peak height (mm) of the The carrier gas is nitrogen, and gas flow rates are
drug and metabolite versus the peak height of the as follows: nitrogen, 40 mL/min; hydrogen, 40
internal standard for the standards, control, and mL/min; and air, 250 mL/min.
unknowns. The injection port temperature is 250C; the
detector temperature is 300C.
Peak height ratio = Peak height (mm) of analyte
The column oven temperature is programmed as
Peak height (mm) of internal standard
follows: The initial temperature is 200C with
zero hold time; the rate of rise is 4C/min; the
3. Plot standard curve from the peak height ratios of
final temperature is 260C.
standards. See Carbamazepine: Figure 5 for
1. Label an appropriate number of 5-mL, glass-
representative illustration of plot for standard
stoppered, conical tubes for standards, control, and
curve. Calculate the amount of drug or
unknowns.
metabolite by comparing the peak height ratios
2. To each tube, add 0.5 mL of the respective
on the appropriate standard curve.
plasma or serum samples.
Notes 3. Add 250 L of saturated trisodium phosphate
1. Absolute recovery was shown to average 95% for buffer solution.
carbamazepine and 83% for its 10,11-epoxide. 4. Add 2 mL of the extraction solvent containing
2. Reproducibility (CV) was 3.6% within-day for internal standard.
carbamazepine and 1.2% for the epoxide. 5. Mix with vortex-mixing device, and centrifuge
Between-day reproducibility was shown to be at 500 g for 5 min.
4% CV for carbamazepine and 6.3% CV for the 6. Remove solvent layer to a conical 3-mL tube
epoxide. (Reliance Glassworks, Inc., Bensenville, IL 60106,
catalog no. R-2827-200), and evaporate to dryness at
3. Although both the HPLC and GLC (below)
40C to 60C with a flow of nitrogen. Reconstitute in
procedures are described using the
100 L of methanol, and inject 2 L.
dihydrocarbamazepine as the internal standard,
7. Chromatograms are illustrated in
any one of the three compounds illustrated in
Carbamazepine: Figure 6 for GLC.
Carbamazepine: Figure 2B are acceptable and are
eluted without problems with either the HPLC or Calculations
the GLC instruments. 1. Measure the peak height (mm) of the drug,
10,11-epoxide, and internal standard.
Procedure: Gas-Liquid Chromatography 2. Compute the ratio of the peak height of the drug
and metabolite versus the peak height of the internal
Principle standard for the standards, control, and unknowns.
The method described below is based on an alkaline
solvent extraction with an analog dihydrocarbamazepine Peak height ratio = Peak height (mm) of analyte
included as an internal standard. The residue is Peak height (mm) of internal standard
reconstituted in methanol for flame ionization gas-liquid
chromatography. (See GLC chromatograms, Figure 5.) 3. Plot standard curve from the peak height ratios
Reagents of standards. See Carbamazepine: Figure 5 for
1. Saturated trisodium phosphate buffer. representative illustration of plot for standard curve.
2. Benzene/ethyl acetate, 4:1 Calculate the amount of drug or metabolite by
3. Methanol. comparing the peak height ratios on the appropriate
4. Stock carbamazepine standard, 2 g/L (8.46 standard curve.
mmol/L). Notes
5. Stock carbamazepine 10,11-epoxide, 1 g/L 1. Absolute recovery was shown to be 89% for
(4.0 mmol/L). carbamazepine with this GLC procedure.
6. Stock dihydrocarbamazepine, 1 g/L (4.20 2. Reproducibility was shown to have a CV of
mmol/L). 1.9% within-day and a CV of 4.0% between days.
317
CA 15-3
CA 15-3
Gregory A. Hobbs i
Name: CA 15-3
Clinical significance: Post-surgical monitoring of recurrence of breast cancer. Refer to Chapter 53,
Neoplasia, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation
Molecular weight: 300,000 D to 400,000 D, depending on extent of glycosylation

Specimen
Principles of Analysis and Current Usage Serum (collected in SST tubes) or plasma (collected in
CA 15-3 is a glycoprotein encoded by the MUC1 gene. EDTA or heparinized tubes) are the only acceptable
It is present in healthy breast tissue, but at much higher specimens.
concentrations, and with altered glycosylation in breast
carcinoma [1]. CA 15-3 is determined by immunoassay. Interferences
Different assay designs are used, but there are two basic Patients who have received mouse monoclonal
approaches [2]. Sandwich-type assays utilize two antibodies for diagnosis or therapy may have either
antibodies, usually 115D8 as the capture antibody and falsely elevated or falsely depressed values for CA 15-3.
DF3 as the signal antibody. 115D8 binds to the Likewise, patients who have been routinely exposed to
carbohydrate portion of the molecule. DF3 binds to an animals or animal serum products can develop
amino-acid sequence near the C terminus (sequence heterophilic antibodies which can interfere with reagent
DTRPAPGS)[2]. The Siemens IMMULITE, however, immunoglobulins.
uses Ma 695 as the capture antibody and Ma 552 as the
signal antibody. The competitive-type assays use a Reference Interval
single antibody. ADVIA Centaur, for example, uses a Among healthy, female study subjects (N=396), 99%
purified antigen attached to a paramagnetic particle as a had assay values of < 31.3 U/mL [6].
capture method and the B27.29 as the signal antibody.
B27.29 binds to the amino-acid sequence SAPDTRPA, Interpretation
thus distinguishing it from DF3. Assay designs include CA 15-3 is most commonly used to assess posttreatment
enzyme-linked immunosorbent assay (ELISA) [3], progress in breast cancer patients. It is not intended to be
microparticle enzyme immunoassay (MEIA)[4,5], and used as a screening tool. Some workers claim that CA
chemiluminescent microparticle immunoassay [4], 15-3 concentrations have prognostic value [10,11].
among others.
CA 15-3 Performance Goals
Reference and Preferred Methods When validating CA 15-3 assay, method performance
There is no reference method for CA 15-3. Centocor, generally should meet or exceed claims of the
Inc. developed the first radioimmunoassay, and it has manufacturer. In particular, total imprecision (day to
served as the comparison standard in a number of studies day) should be evaluated and physicians informed of
[7]. The assay is still available through Fujirebio what constitutes a significant deviation from previous
Diagnostics [8]. However, this method is no longer the values (>2SD above a mean). If validating an assay to
preferred method. The preferred methods are replace an existing assay, a correlation with the existing
immunoassays with various optical detection systems. method versus the new method, using patient samples,
They include those mentioned above and listed in Table should be performed. A correlation coefficient r > 0.90
1. As an example, the assay on the Abbott Laboratories can be expected, though slopes may vary with method
ARCHITECT instrument is discussed in detail at the end [2]. Differences in slope, indicating possible bias, must
of this methods chapter [6]. be communicated to clinicians who are following
patients long term. Bias must be taken into account in
The performance of many of the above methods have the evaluation of such patients with respect to what
been evaluated and compared [9]. Though the methods constitutes a significant shift in CA 15-3 values.
were largely comparable, significant deviations among
methods for some samples were noted. The authors References
recommended redetermination of the baseline when 1 Duffy MJ, Shering F, Sherry F, McDermott E,
changing methods. OHiggins N. CA 15-3: a prognostic marker in
breast cancer. Int J Biol Markers 2000;15:330-
3.
2 Klee GG, Schreiber WE. MUC1 gene-derived
glycoprotein assays for monitoring breast
i
CA 15-3
New method:
Fifth edition: Gregory A. Hobbs
318
2
cancer (CA 15-3, CA 27.29): are they

measuring the same antigen? Arch Pathol Lab
Med 2004;128:1131-5. Procedure [6]
3 MP Biomedicals. Breast Cancer Antigen CA
15-3 Enzyme Immunoassay Test Kit. MP Principle
Biomedicals, Orangeburg, NY 10962. The ARCHITECT CA 15-3 assay is a two-step
4 Abbott Laboratories Diagnostics Division. IMx immunoassay. In the first step, the sample, wash buffer,
System CA 15-3. May 2007. Abbott and paramagnetic microparticles coated with the
Laboratories, Abbott Park, IL 60064. monoclonal antibody 115D8 are combined. CA 15-3
5 Abbott Laboratories Diagnostic Division. present in the sample binds to the 115D8-coated
Abbott AxSYM System Tumor Markers CS 15- microparticles. After washing, anti-CA 15-3 antibody
3. January 2007. Abbott Laboratories, Abbott conjugated to acridinium is added. Pre-trigger (1.32%
Park, IL 60064. H2O2) and trigger (0.35 N NaOH) solutions are then
6 Abbott Laboratories Diagnostic Division. added. The resulting chemiluminescent reaction is
ARCHITECT System CA 15-3 September measured in relative light units (RLU). There is a direct
2004. Abbott Laboratories, Abbott Park, IL relationship between the amount of CA 15-3 in the
60064. sample and the RLU detected by the optical system.
7 Chell CD, Morris DL, Kish L, Goldblatt J,
Neaman I, Allard JW et al. Mulitcenter Materials Provided
evaluation of Bayer Immuno-I CA 15-3 assay. 2K44 ARCHITECT CA 15-3 reagent kit
Clin Chem 1998;44:765-72.
8 Fujirebio Diagnostics, Inc. CA 15-3 Materials Required But Not Provided
Radioimmunoassay. October 2002. Fujirebio ARCHITECT i System, 3K50 ARCHITECT assay CD-
Diagnostics, Malvern, PA 19365. ROM, CA 15-3 controls, CA 15-3 calibrators, multi-
9 Slev PR, Rawlins ML, Roberts WL. assay manual diluent, pre-trigger solution, trigger
Performance characteristics of seven automated solution, wash buffer, reaction vessels, sample cups,
CA 15-3 assays. Am J Clin Pathol septum, replacement caps, pipettes or pipette tips.
2006;125:752-7.
10 Park BW, Oh JW, Kim JH, Kim KS, Kim JH, Test Procedure
Lee KS. Perioperative CA 15-3 and CEA serum 1. Invert the microparticle bottle 30 times before
levels as predictor for breast cancer outcomes. loading reagent on ARCHITECT. Visually
Ann Oncol 2008;19:675-81. inspect the bottle to make sure microparticles
11 Al-azawi D, Kelly G, Myers E, McDermott are resuspended. If not, continue to invert until
EW, Hill AD, Duffy MJ, Higgins NO. CA 15-3 there is complete resuspension of
is predictive of response and disease recurrence microparticles.
following treatment in locally advanced breast 2. Remove and discard the cap. Wearing clean
cancer. BMC Cancer 2006;6:220. gloves, remove a septum from the bag.
12 Harris L, Fritsche H, Mennel R, Norton L, Squeeze the septum in half to confirm that the
Ravdin P, Taube S et al. American Society of slits are open Carefully snap the septum onto
Clinical Oncology 2007 update of the top of the bottle.
recommendations for the use of tumor markers 3. Order tests.
in breast cancer. J Clin Oncol 2007;25:5287- 4. Load the ARCHITECT CA 15-3 reagent kit on
5312. the ARCHITECT i System. Verify that all
necessary assay reagents are present. Ensure
that septa are present on all reagent bottles.
Table 1: Methods for the Measurement of CA 15-3 5. If using primary or aliquot tubes, use the
Ref. sample gauge to ensure sufficient patient
Int. specimen is present.
Vendor Instruments Principle (U/mL) 6. ARCHITECT CA 15-3 calibrators and controls
ADVIA competitive 035 should be mixed by gentle inversion prior to
Centaur immunometric 035 use.
Siemens IMMULITE 7. When calibrating, hold the bottles vertically,
and dispense 7 drops of each calibrator
Roche Elecsys immunometric 034 solution.
IMx immunometric 031.3 8. When adding controls, hold the bottles
Abbott AxSYM immunometric 031.3 vertically, and dispense 4 drops of each control.
ARCHITECT immunometric 031.3 9. Load samples.
Ortho Vitros ECi immunometric 035 10. Press RUN. The ARCHITECT performs the
following functions:
MP ELISA Plate
Moves the sample to the aspiration point
Biochemicals reader immunometric 035
Loads the reaction vessel (RV) into the
Fujirebio immunometric 0 5 process path
Aspirates and loads the sample into the RV
319
3
Advances the reaction vessel one position The calibration range for the CA 15-3 assay on the
Aspirates wash buffer and microparticles and ARCHITECT is 0-800 U/mL. Specimens exceeding 800
transfers them into the RV U/mL are flagged with the code >800.0 and may be
Mixes, incubates, and washes the reaction diluted using either the Automated Dilution Protocol or
mixture the Manual Dilution Procedure.
Adds conjugate to the RV 1. Automated Dilution Protocol
Mixes, incubates, and washes the reaction The system performs a 1:5 dilution of the
mixture specimen and automatically calculates the concentration
Adds pre-trigger and trigger solutions of the sample before dilution and reports the result.
Measures chemiluminescent emission to 2. Manual Dilution Procedure
determine the quantity of CA 15-3 The suggested dilution for the ARCHITECT
Aspirates contents of RV to liquid waste and CA 15-3 assay is 1:5.
unloads RV to solid waste For a 1:5 dilution, add 100 L patient
Calculates the result specimen to 400 L ARCHITECT I Multi-
11. Calibration Assay Manual Diluent. The operator must enter the
Calibrators A, B, C, D, E, and F are tested in dilution factor in either the patient or control order
duplicate. No further calibration is necessary unless screen. All assays selected for that order will be diluted.
controls are out of range or a reagent kit with a new lot The dilution factor will be used by the system to
number is used. automatically calculate the concentration of the sample
12. Quality controls before dilution. The diluted result should read greater
Quality control should be performed once than 30 U/mL.
every 24 hours or according to laboratory procedure, if
that is more frequent. Interpretation
13. Calculation When monitoring breast cancer patients, changes
The ARCHITECT CA 15-3 assay utilizes a observed in serial CA 15-3 assay values should be
four-parameter logistic curve fit data-reduction method evaluated in conjunction with other clinical methods
to generate a calibration curve. used for monitoring breast cancer. The American
Society for Clinical Oncology has recommended the
Assay Range following [12]:
CA 15-3 is not recommended for use as a screening test for breast cancer.
Screening
CA 15-3 is not recommended for diagnosing or staging of breast cancer
Diagnosis/Staging
CA 15-3 can be used in conjunction with diagnostic imaging, history, and physical exam
for monitoring patients with metastatic disease during active therapy.
Present data are insufficient to recommend the use of CA 15-3 alone for monitoring
response to treatment.
Monitoring response to
therapy When there is no readily measurable disease, increasing levels of CA 15-3 may be used to
indicate treatment failure.
Use caution while monitoring during the first 4 to 6 weeks of a new therapy to avoid
spurious early rises.
CA 15-3 is not recommended for monitoring patients for recurrence after primary breast
Surveillance cancer therapy.
320
Carbohydrate Antigen 19-9 (CA19-9)

Hassan M.E. Azzazy
Name: Carbohydrate antigen 19-9 (CA19-9)
Clinical significance: Refer to Chapter 53, Neoplasia, in 5th edition of Clinical Chemistry: Theory,
Molecular mass: 200,000 to 1,000,000 D
Chemical class: Glycolipid
Principles of Analysis and Current Usagei Specimen

CA19-9 is a sialylated derivative of the Lewisa blood Different manufacturers may recommend either serum or
group antigen (sialylated lacto-N-fucopentaose II plasma for their individual assays. Serum samples can be
ganglioside; denoted Lexa) [1]. It is synthesized by stored at 2C to 8C for 24 hours or frozen at 20C or
normal human pancreatic and biliary ductular cells, as lower for long-term storage [6]. Plain and gel-containing
well as by gastric, colonic, and endometrial epithelia, tubes may be used for measuring CA19-9 in sera stored
and exists in serum as a mucin. at 4C for no more than 24 hours from the time of blood
CA19-9 was originally identified by a monoclonal collection. Use of sera stored in tubes containing
antibody produced from hybridomas obtained from a thixotropic gel separator for > 24 hours is not
mouse immunized with the human colon cancer cell line recommended [7].
SW 1116 [2]. In 1983, an immunoradiometric assay
(IRMA) for CA19-9 determination was developed which Interferences
used the anti-CA19-9 antibody as both the capture and No chemical interferences or in-vivo effects have been
signaling antibody. The assays calibration curve covers reported to influence CA19-9 measurements. However,
the range from 0 to 120 kU/L [3]. In 1991, a immunoassay results may be subject to interference from
chemiluminescent enzyme immunoassay was developed heterophilic antibodies which are present in 30% to 40%
to quantify several tumor markers including CA19-9 [4]. of patients and arise as a result of contacting the same
Currently, several automated CA19-9 assays are animal species used for production of diagnostic
available: ARCHITECT 12000 (Abbott Diagnostics), antibodies. Human anti-mouse antibodies (HAMA) have
ADVIA Centaur (Bayer Diagnostics), UniCel Dxl 800 become a major source of interference since the use of
(Beckman Coulter), IMMULITE 2000 (Diagnostics murine monoclonal antibodies to deliver
Products), and Elecsys E170 (Roche Diagnostics). immunoscintigraphic or chemotherapeutic agents to
ARCHITECT 12000 and ADVIA Centaur assays tumors. Immunoassays, especially single-step
employ the same CA19-9 antibodies. immunometric ones, also suffer from hook effect that is
characterized by the generation of falsely low results in
Although measurement of CA19-9 is currently available samples containing super-high concentrations of the
on several routine analytical systems, a recent study by analyte (beyond the upper limit of the assay).
Passerini et al. concluded that the same analytical system
should be used during tumor follow-up to eliminate the Reference Intervals
possibility of method-induced variation [5]. The most commonly used cutoff value for CA19-9 is <
37 kU/L. CA19-9 levels were measured in 20,517
Reference and Preferred Methods patients, including 90 with pancreatic cancer, 70 with
There is no reference method for CA19-9. Several benign pancreatic disease, 152 with biliary cancer, 170
automated non-isotopic CA19-9 immunoassays are with benign biliary disease, and 20,035 asymptomatic
currently available that have acceptable analytic controls. In asymptomatic individuals, the mean CA19-9
performance. There appears to be no reason to prefer one level was 9.42 9.95 U/mL, whereas levels > 37 U/mL
assay over another, and selection is therefore based were found optimal for discriminating pancreatic cancer
ultimately on the type of the analyzer available in the from benign pancreatic disease (sensitivity 77%,
laboratory. specificity 87%) [8]. A recent study reported that CA19-
9 results of healthy volunteers were < 37 kU/L as
CA19-9 results of a given specimen vary when measured assayed using five automated CA19-9 immunoassays
with assays from different manufacturers, owing to [9].
differences in methods, reagents, and antibodies used.
Efforts to harmonize assays, including the availability of Interpretation
an international reference material, are needed. CA19-9 is a marker of pancreatic and colorectal
carcinoma [10]. It is not a tumor-specific marker but a
i
tumor-associated marker. CA19-9 is elevated in all
CA19-9 forms of gastrointestinal cancers and other
New method adenocarcinomas. It is the most frequently elevated
Fifth edition: Hassan M.E. Azzazy marker in pancreatic cancer, but its concentration does
321
not correlate to tumor mass. It is useful for monitoring patients by using a CA19-9 cutoff 150 kU/L, and the
pancreatic and colorectal cancers, and elevated levels use of serum CA19-9 levels would increase the
can indicate recurrence 1 to 7 months prior to detection efficiency of laparoscopic staging in patients with
by radiographs or other clinical findings [6,11]. Elevated suspected pancreatic malignancy.
levels have also been reported in acute and chronic CA19-9 is basically ineffective for screening
pancreatitis and cirrhosis. asymptomatic populations [13]. In one study, C19-9
CA19-9 is the most commonly used serum marker for levels were measured in an asymptomatic Korean
pancreatic cancer and has good utility for confirming population (n = 70,940). Only four cases of pancreatic
diagnosis of symptomatic patients [12,13]. CA19-9 has a cancer were detected, along with 159 false positives,
specificity range of 68% to 91% (median 82%) and yielding a positive predictive value of 0.9% [22].
sensitivity of 70% to 90% (median 79%) [12,14].
Serum CA19-9 concentration provides an independent
The diagnostic usefulness of serum CA19-9 suffers from prognostic value for overall patient survival. In a cohort
a number of limitations. First, the sensitivity of CA19-9 study of pancreatic cancer patients undergoing resection
is significantly low (55%) in pancreatic ductal (n = 347), the median survival time for patients with the
adenocarcinoma patients with small curative tumors (<3 same tumor stage was significantly longer in patients
cm) [11], and almost half of patients (48.4%) with small whose CA19-9 levels normalize after resection as
pancreatic cancer do not show elevated CA19-9 levels compared to those whose CA19-9 did not return to
[15]. Second, CA19-9 antigen secretion depends on normal levels [23]. In cases of colorectal cancer, C19-9
Lewisa antigen status. The interpretation of CA19-9 has a lower sensitivity than carcinoembryonic antigen
levels may be difficult because ~ 7% of the population (CEA). However, a correlation between elevated
are Lewis-negative and have no detectable levels of preoperative CA19-9 levels and adverse patient outcome
CA19-9 regardless of the tumor burden. Third, CA19-9 has been found. Based on the available data,
is elevated in non-pancreatic gastrointestinal measurement of CA19-9 cannot be recommended for
malignancies and in a variety of benign diseases such as patients with colorectal cancer, but it may provide some
biliary obstruction, cholecystitis, cholangitis, hepatic independent prognostic information [24].
cirrhosis, and acute and chronic pancreatitis. Based on
the previous limitations, CA19-9 has a limited role in the Performance Goals
diagnosis of pancreatic cancer and should be used to For CA19-9, the within- and between-subject biological
complement accurate pancreatic imaging procedures variations were 16.0% and 102.0 %, respectively. The
such as computed tomography (CT) and endoscopic desirable analytical quality specifications for total error,
ultrasonography (EUS) [16,17]. imprecision, and bias, derived from biological variation,
were 39%, 8%, and 25.8%, respectively [25]. A recent
CA19-9 measurement has a documented role in study investigated the effect of intraindividual biological
monitoring gemcitabine chemotherapy for pancreatic variation on C19-9 serial results. Using four blood
cancer. A decline of > 20% in CA19-9 levels (as samples, each collected at a 2-week interval from each
compared to baseline) after 8 weeks of therapy has been healthy volunteer (n = 49; ages 18 to 60 years; 25 males
shown to be a better indicator of response and survival and 24 females), the calculated intraindividual and inte-
than CT imaging [18,19]. Serial CA19-9 measurements rindividual coefficients of variation (CVs) for CA19-9
have greater sensitivity and specificity for predicting (as measured by an immunoluminometric assay on the
response than individual measurements. One study ARCHITECT analyzer) were 27.2% and 62.2%,
showed that serial falls in CA19-9 had a sensitivity of respectively [26].
67% for predicting partial response, and serial elevations
had a sensitivity of 86% for predicting progressive The recommendations of the European Group on Tumor
disease [20]. Another study concluded that kinetics of Markers and National Academy of Clinical
CA19-9 serum concentration serve as an early indicator Biochemistry guidelines indicate that assays for tumor
of response to gemcitabine chemotherapy in patients markers, including CA19-9, should demonstrate
with inoperable pancreatic adenocarcinoma [18]. The interassay variability under 10% and intraassay
U.S. Food and Drug Administration (FDA) has approved variability less than 5% [27,28]. The current Clinical
CA19-9 tests as aids in the management of patients with Laboratory Improvement Amendments (CLIA)
confirmed pancreatic cancer and serial monitoring of performance goal for measurement of CA19-9 is for
their response to therapy and disease progression. laboratories to be within 3 standard deviations of the
However, tests should only be used in patients with peer-group mean. According to the 2007 College of
CA19-9 values above the cutoff at the time of diagnosis. American Pathologists (CAP) Survey, CV for all CA19-
9 methods at mean values of 72 U/mL (SD 24.5) and
One study investigated the role of CA19-9 114.9 U/mL (SD 41.99) were 34% and 36.5%,
measurements in enhancing the effectiveness of staging respectively. Current CA19-9 assays should be
laparoscopy in patients with suspected pancreatic standardized to reduce the large discrepancies commonly
malignancy [21]. The study included 159 patients who observed among different methods.
had CT-predicted resectable disease and who had References
undergone laparoscopic staging. The study determined 1 Magnani JL, Nilsson B, Brockhaus M, Zopf D,
that laparoscopy could have been avoided in 40% of Steplewski Z, Koprowski H, Ginsburg V. A
322
monoclonal antibody-defined antigen associated 16 European Group on Tumor Markers. Tumor

with gastrointestinal cancer is a ganglioside markers in gastrointestinal cancers: EGTM
containing sialylated lacto-N-fucopentaose II. J recommendations. Anticancer Res
Biol Chem 1982;257:14365-9. 1999;19:2811-5.
2 Koprowski H, Steplewski Z, Mitchell K, Herlyn 17 DiMagno EP, Reber HA, Tempero MA.
M, Herlyn D, Fuhrer P. Colorectal carcinoma American Gastroenterological Association.
antigens detected by hybridoma antibodies. AGA technical review on the epidemiology,
Somatic Cell Genet 1979;5:957-71. diagnosis, and treatment of pancreatic ductal
3 Del Villano BC, Brennan S, Brock P, Bucher C, adenocarcinoma. Gastroenterology
Liu V, McClure M et al. Radioimmunometric 1999;117:1464-84.
assay for a monoclonal antibody-defined tumor 18 Ziske C, Schlie C, Gorschluter M, Glasmacher
marker, CA19-9. Clin Chem 1983;29:549-52. A, Mey U, Strehl J et al. Prognostic value of
4 Nishizono I, Iida S, Suzuki N, Kawada H, CA19-9 levels in patients with inoperable
Murakami H, Ashihara Y, Okada M. Rapid and adenocarcinoma of the pancreas treated with
sensitive chemiluminescent enzyme gemcitabine. Br J Cancer 2003;89:1413-7.
immunoassay for measuring tumor markers. 19 Halm U, Schumann T, Schiefke I, Witzigmann
Clin Chem 1991;37:1639-44. H, Mossner J, Keim V. Decrease of CA19-9
5 Passerini R, Riggio D, Salvatici M, Zorzino L, during chemotherapy with gemcitabine predicts
Radice D, Sandri MT. Interchangeability of survival time in patients with advanced
measurements of CA19-9 in serum with four pancreatic cancer. Br J Cancer 2000;82:1013-6.
frequently used assays: an update. Clin Chem 20 Gogas H, Lofts FJ, Evans TR, Daryanani S,
Lab Med 2007;45:928-9. Mansi JL. Are serial measurements of CA19-9
6 Wu AHB, ed. Tietz Clinical Guide to useful in predicting response to chemotherapy
Laboratory Tests. 4th ed. Philadelphia: in patients with inoperable adenocarcinoma of
Saunders; 2006. the pancreas? Br J Cancer 1998;77:325-8.
7 Banfi G, Parma P, Pontillo M. Stability of 21 Connor S, Bosonnet L, Alexakis N, Raraty M,
tumor markers CA19.9, CA 125, and CA 15.3 Ghaneh P, Sutton R, Neoptolemos JP. Serum
in serum obtained from plain tubes and tubes CA19-9 measurement increases the
containing thixotropic gel separator. Clin Chem effectiveness of staging laparoscopy in patients
1997;43:2430-2431. with suspected pancreatic malignancy. Dig Surg
8 Kim HJ, Kim MH, Myung SJ, Lim BC, Park 2005;22:80-5.
ET, Yoo KS et al. A new strategy for the 22 Kim JE, Lee KT, Lee JK, Paik SW, Rhee JC,
application of CA19-9 in the differentiation of Choi KW. Clinical usefulness of carbohydrate
pancreaticobiliary cancer: analysis using a antigen 19-9 as a screening test for pancreatic
receiver operating characteristic curve. Am J cancer in an asymptomatic population. J
Gastroenterol 1999;94:1941-6. Gastroenterol Hepatol 2004;19:182-6.
9 Laulu SL, Roberts WL. Performance 23 Safi F, Schlosser W, Falkenreck S, Beger HG.
characteristics of five automated CA19-9 CA19-9 serum course and prognosis of pan-
assays. Am J Clin Pathol 2007;127:436-40. creatic cancer. Int J Pancreatol 1996;20:155-61.
10 Lamrez R. CA19-9: GICA (gastrointestinal 24 Duffy MJ, van Dalen A, Haglund C, Hansson
cancer antigen). In: Sell S, ed. Serological L, Klapdor R, Lamerz R et al. Clinical utility of
Cancer Markers. Totowa, NJ: Humana Press; biochemical markers in colorectal cancer:
1992:309-339. European Group on Tumour Markers (EGTM)
11 Steinberg W. The clinical utility of the CA19-9 guidelines. Eur J Cancer 2003;39:718-27.
tumor-associated antigen. Am J Gastroenterol 25 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
1990;85:350-5. Hernandez A, Jimenez CV et al. Current
12 Okusaka T, Yamada T, Maekawa M. Serum databases on biological variation: pros, cons
tumor markers for pancreatic cancer: the dawn and progress. Scand J Clin Lab Invest
of new era? J Pancreas 2006;7:332-6. 1999;59:491-500.
13 Freelove R, Walling AD. Pancreatic cancer: 26 Erden G, Barazi AO, Tezcan G, Yildirimkaya
diagnosis and management. Am Fam Physician MM. Biological variation and reference change
2006;73:485-92. values of CA19-9, CEA, AFP in serum of
14 Goonetilleke KS, Siriwardena AK. Systematic healthy individuals. Scand J Clin Lab Invest
review of carbohydrate antigen (CA19-9) as a 2008;68:212-8.
biochemical marker in the diagnosis of 27 European Group on Tumor Markers website.
pancreatic cancer. Eur J Surg Oncol <http://egtm.eu/1280.html>
2007;33:266-70. 28 National Academy of Clinical Biochemistry:
15 Egawa S, Takeda K, Fukuyama S, Motoi F, Practice Guidelines and Recommendations for
Sunamura M, Matsuno S. Clinicopathological Use of Tumor Markers in the Clinic, 2005.
aspects of small pancreatic cancer. Pancreas Available at
2004;28:235-40. http://www.aacc.org/SiteCollectionDocuments/
NACB/LMPG/tumor/chp2_qreqs.pdf
323
Carbon Dioxide and Bicarbonate

William J. Korzun
Name: CO2, total CO2 content
Clinical significance: Refer to Chapter 29, Acid-Base Control and Acid-Base Disorders, in the 5th
Molecular formula: CO2 or O=C=O
Merck Index: 1793
Chemical class: Gas, organic acid in soluble phase
i titrated with a solution of NaOH [2] (Table 1, Method

2).
Total carbon dioxide in serum or plasma exists in three
major chemical forms: dissolved CO2 (3%), carbamino According to the 2007 College of American Pathologists
(CAP) Participant Summary, approximately 64% of
derivatives of plasma protein (33%), and bicarbonate
participating laboratories measure total carbon dioxide
(HCO3) anion (64%) [1]. Other quantitatively minor
by an enzymatic/spectrophotometric assay. This method
forms are carbonic acid (H2CO3) and carbonate ions
has been adapted to many automated analyzers. All CO2
(CO3). Several procedures that quantitate total carbon
forms are quantitatively converted to HCO3 by adding
dioxide content in serum or plasma involve acidification
alkali to the serum. The bicarbonate is then
of the sample to convert all carbon dioxide forms to CO2
enzymatically converted to oxaloacetic acid, which is
gas and measurement of the amount of gas formed. A measured by an NADH-consumption reaction that is
second category of procedures utilizes the enzymatic quantitated spectrophotometrically as follows (Table 1,
reaction of bicarbonate with phosphoenolpyruvate Method 4):
carboxylase coupled to a spectrophotometric indicator
reaction. Most assays which specifically measure
bicarbonate have reagents at an alkaline pH. This HCO3 + phosphoenolpyruvate PEPC
converts other forms of CO2 to bicarbonate, so these oxaloacetate + Pi
assays are also measuring total CO2.
Oxaloacetate + NADH + H+ MDH malate + NAD+
The historical reference method used the Natelson
Microgasometer to manometrically measure CO2 gas PEPC is phosphoenolpyruvate carboxylase, Pi is
liberated after acidification of serum. The CO2 gas was inorganic phosphate, and MDH is malate
then absorbed by alkali, and the residual pressure due to dehydrogenase. The decrease in absorption at 340 nm is
gases other than CO2 was measured. The difference is related to the serum concentration of CO2.
the specific CO2 content of the serum sample (Table 1,
Approximately 36% of laboratories participating in CAP
Method 1). An alternative reference method has been proficiency surveys employ a method that uses a PCO2
described by the IFCC Scientific Division Working electrode to quantitate the CO2 gas produced by
Group on Selective Electrodes, utilizing a semi-
acidification of the sample. Here again, the CO2 gas
automated extraction/titration device. Equal volumes of
specimen and 0.1 mol/L lactic acid are mixed in a diffuses across a silicon-rubber membrane and changes
temperature-controlled extraction vessel. The CO2 the pH of a bicarbonate electrode buffer. The pH change
is detected by a glass pH electrode within the PCO2
liberated from the specimen is then swept by a stream of
electrode assembly (Table 1, Method 3).
CO2-free air into a titration vessel containing a solution
of BaCl2 and a pH electrode. The CO2 reacts with Ba2+
In another method, total CO2 is calculated during a
and H2O to form BaCO3, releasing 2 moles of H+ for
each mole of CO2. The H+ ions are volumetrically blood-gas analysis (Table 1, Method 5). Using the
measured pH and PCO2 values, one can estimate the total
CO2 using the following equation:
i
Total CO2 (mEq/L) = PCO2 + [HCO3]
First edition: William J. Korzun, W. Gregory Miller
Methods edition: William J. Korzun, W. Gregory Miller
Second edition: William J. Korzun, W. Gregory Miller where (0.0301), the Bunson coefficient, is the solubility
Third edition: William J. Korzun, W. Gregory Miller coefficient of CO2, and the bicarbonate concentration
Fourth edition: William J. Korzun, W. Gregory Miller
Fifth edition: William J. Korzun
324
can be estimated from the Henderson-Hasselbalch

equation: Carbon Dioxide Reference Interval
[HCO3] = PCO2 (antilog [pH pKa]) Cord blood [6]

where pKa = 6.10. Although this calculation can provide Arterial 1329 mmol/L
Venous 1528 mmol/L
only an indirect, unmeasured estimate of total CO2, it
Serum or plasma [7]
correlates well with measured total CO2 values. < 1 month 1727 mmol/L
16 months 1729 mmol/L
Reference and Preferred Methods 6 mo1 yr 1829 mmol/l
Although there are no reference methods for the > 1 year 2131 mmol/L
determination of bicarbonate or CO2 concentration, all of
the common automated CO2/bicarbonate methods Interpretation
produce accurate and precise results when calibrated and Measurement of total serum CO2 is useful in
used according to the manufacturers instructions. determining the acid-base status of a patient. The total
Typical within-lab precision for automated analyzers CO2 concentration is depressed in metabolic acidosis.
using acidification/pH electrode or enzymatic methods is
The degree of depression depends on the degree of initial
2% to 4% coefficient of variation (CV). Methods which
acidosis and subsequent compensatory mechanisms. In
can be calibrated using primary aqueous standards of
respiratory acidosis, the total CO2 is mildly increased
sodium carbonate are suitable as accuracy comparison
methods to validate performance of methods which use because of an increased PCO2. However, if renal
serum-based secondary calibration materials. compensation occurs, the bicarbonate levels rise, and
total CO2 also increases.
Specimen
Either serum or heparinized plasma may be used. Other Patients with a respiratory alkalosis initially exhibit only
anticoagulants cannot be used, because they disturb the small decreases in total CO2. As renal compensation
equilibrium between erythrocyte and plasma CO2. They occurs, the levels can become moderately depressed.
also lower the pH of the sample and accelerate the loss Metabolic alkalosis results in an increase in bicarbonate
of CO2 to ambient air. Whole blood cannot be used levels and thus total CO2. The increase in total CO2 can
because the heme-bound CO2 and carbamino-bound become much larger as a result of respiratory
CO2 vary with the hematocrit and oxygen saturation and compensation.
produce variable results. Samples collected in tubes
Although the physician can use total CO2 data to
containing separator gel give the same results as tubes
without separator gel. diagnose and monitor a patients acid-base status, in
acute cases, this information is much less valuable than a
The major errors in total CO2 measurement are complete blood-gas analysis. Total CO2 analysis is
associated with sample handling. Samples should ideally performed frequently because of historical rather than
be handled anaerobically, centrifuged at 37C, and clinical factors. Total CO2 was part of a battery of tests
stored tightly stoppered before analysis. Once separated available on the original Technicon SMA instruments.
from erythrocytes and kept stoppered, serum or plasma Physicians are accustomed to having this test available,
total CO2 content is stable for several days at 4C. and laboratories are used to providing it, though its
Centrifugation at room temperature produces a decrease clinical utility is limited to occasional monitoring of
in CO2 of approximately 0.5 mmol/L [3]. This is the patients in renal failure or acute acidosis. The continued
availability of total CO2 as a component of routine
result of the shift of CO2 into the erythrocyte as the pH
profiles may be questioned in the future.
decreases as the temperature of the blood drops from
37C to 25C. Loss of CO2 from uncapped tubes is
Carbon Dioxide Performance Goals
approximately 4 mmol/L after 1 hour [4]. CO2 loss can Total CO2 content is not regulated under the Clinical
be prevented by raising the pH of the sample to a range Laboratory Improvement Amendments of 1988 (CLIA
of 9.0 to 9.3 by addition of NH4OH. Layering with oil 88). Carbon dioxide is difficult to include in
should not be performed because of the high solubility of proficiency-testing material because of its volatility.
CO2 in oil. When included as part of a special diluent for freeze-
dried materials, there may be substantial variability
Interferences between labs due to loss of CO2 during handling of the
In general, most of the enzymatic total CO2 assays are material.
not affected by bilirubin concentrations up to 20 mg/dL
and hemoglobin up to 500 mg/dL [5]. The presence of Within-subject variation has been reported to be less
high triglyceride concentrations may cause light than 6% over periods greater than 1 month [8]. Precision
scattering and cause a negative bias in some enzymatic goals have been defined as a maximum standard
assays. deviation of approximately 0.5 mmol/L at a decision
325
threshold of 20 mmol/L and approximately 0.7 at a

decision threshold of 30 mmol/L [9].
References
1 Davenport HW. The ABC of Acid-Base
Chemistry. 5th ed. Chicago: University of
Chicago Press; 1969:34.
2 Burnett RW, Covington AK, Fogh-Anderson N,
Kulpmann WR, Lewenstam A, Maas AHJ et al.
IFCC reference measurement procedure for
substance concentration determination of total
carbon dioxide in blood, plasma or serum. Clin
Chem Lab Med 2001;39:283-289.
3 Astrup P. A simple electrometric technique for
the determination of carbon dioxide tension in
blood and plasma, total content of carbon
dioxide in plasma, and bicarbonate content in
separated plasma at a fixed carbon dioxide
tension. Scand J Clin Lab Invest 1956;8:33-43.
4 Gambino SR, Schreiber H. The measurement of
CO2 content with the autoanalyzer. Am J Clin
Pathol 1966;45:406-411.
5 Young DS. Effects of Preanalytical Variables
on Clinical Laboratory Tests. 2nd ed.
Washington, DC: AACC press; 1997:3-793-
82.
6 Perkins SL, Livesey JF, Belcher J. Reference
intervals for 21 clinical chemistry analytes in
arterial and venous umbilical cord blood. Clin
Chem 1993;39:1041-1044.
7 Greeley C, Snell J, Colaco A, Beatey J, Bailey
J, Bjorn S et al. Pediatric reference ranges for
electrolytes and creatinine. Clin Chem
1993;39:1172.
8 Morris HA, Wishart JM, Horowitz M, Need
AG, Nordin BE. The reproducibility of bone-
related biochemical variables in post
menopausal women. Ann Clin Biochem
1990;27:562-568.
9 Fraser CG. Biological variation in clinical
chemistry. An update: collated data, 1988-1991.
Arch Path Lab Med 1992;116:916-923.
326
Tables
Table 1: Carbon Dioxide Methods

Method 1: Gas release; manometric or thermometric
Principle of Analysis: O2 is released from samples after addition of acid, and the gas pressure is measured
manometrically or by thermal conductivity.
Comments: Rarely used; the reference method, used for calibration purposes
Method 2: Gas release; titrimetric
Principle of Analysis: CO2 + acid CO2 gas
+ H2O + Ba+2 BaCO3 + 2 H+
CO2 gas
2 H2 O
2 H+ + 2 OH-
Comments: Rarely used; potential reference method
Method 3: CO2 electrode; ion-selective (pH) electrode
Principle of Analysis: After acidification of sample, CO2 diffuses across a silicone membrane into dilute
bicarbonate buffer. The resulting pH change, as measured by a pH electrode, is related to total CO2 content.
Comments: Common; same method by which PCO2 is measured in blood-gas analyzers
Method 4: Enzymatic; spectrophotometric
_
Principle of Analysis: HCO + PEP PEPC OX + P
3 i
OX + NADH + H+ MDHMAL + NAD+ ( A, 340 nm

Comments: Common; used on discrete analyzers such as Du Pont aca and Kodak Ektachem
Method 5: Calculation; ion-selective (pH and PCO2) electrode
Principle of Analysis: Total CO2 calculated from data obtained in blood-gas analysis; total CO2 = PCO2 +
[HCO3]
Comments: Common; although only an estimate, compares well with measured values
MAL, Malate; MDH, malate dehydrogenase; NADH, dihydronicotinamide adenine dinucleotide; OX, oxaloacetate; Pi ,
phosphate; PEP, phosphoenolpyruvate; PEPC, phosphoenolpyruvate carboxylase.
Figures
Carbon Dioxide: Figure 1
Schema of PCO2 electrode assembly.

327
Carcinoembryonic Antigens (CEA)

Gregory A. Hobbs
Name: Carcinoembryonic antigens (CEA)
Clinical significance Refer to Chapter 53, Neoplasia, in the 5th edition of Clinical
Molecular mass: Colon CEA, 160,000 to 180,000 D
Ovarian CEA, 200,000 D
Breast CEA, 90,000 to 105,000 D
i
In 1969, a radioimmunoassay (RIA) was developed
Cancer-specific antigens of the digestive tract were to measure perchloric acid (PCA)-extracted CEA in
first discovered by Gold and Freedman in 1965 [1- sera obtained from patients with adenocarcinoma
2]. They derive their name from the discovery that of the colon [10]. RIA used to investigate CEA-
they occur in the fetal gut as well as in neoplastic reacting molecules soon showed that many such
tissues [2]. The first CEA-reacting molecule that molecules could be identified from fetal gut
was extensively characterized by this group was extracts obtained at different stages in utero. RIA
obtained from an ovarian carcinoma [3]. Later of saline extracts from patients with 20 primary
reports indicated that colon CEA showed a lower adenocarcinomas of the colon showed 8 to 12 CEA
molecular mass range (160,000 to 180,000) than varieties by isoelectric focusing [11]. It is now
the ovarian-derived molecule [4]. known that three of these bands are cancer specific,
and four belong to a family of normal colon cross-
Although the collective term for these antigens, reacting antigens (NCA) which may also be
CEA, has been used since their discovery, it has produced in excess during carcinogenesis. Other
long been known that CEA is a heterogeneous CEA-reacting molecules are bound to serum
group of molecules [5]. This is why not all CEA- lipoproteins, membranes, and immune complexes.
containing sera from cancer patients produce linear A number of assays have used PCA or heat or both
dilution curves in parallel with CEA standard to extract serum glycoproteins, including CEA, into
curves [6], nor will the results obtained from one the fluid phase while precipitating 90% to 98% of
CEA-ligand assay be identical with another. the other serum proteins (globulins and albumins)
[12,13]. Both methods inactivate proteolytic
Two similar high-molecular-mass glyco-proteins enzymes, cause antigenic loss of different CEA
have been examined extensively. CA 125 was molecules, and remove proteins that would
shown to better distinguish progression, regression, otherwise cause nonspecific protein binding.
or stabilization in patients with epithelial ovarian Measurement of CEA using radioimmunoassay
cancer than CEA, suggesting a significant antigenic utilizes a monoclonal antibody complexed with a
difference between the two markers [7]. This was plastic bead that reacts with the patient sample,
shown directly with in-vitro interference studies in binding CEA. Simultaneously, a second
which anti-CA 125 did not bind CEA above monoclonal antibody, specific for a different site
background [8]. CA 19-9, also of interest in from the first, binds the CEA-Ab-bead complex.
colorectal cancer, has been purified and studied in The second antibody is labeled with 125I. After a
detail [9]. It was shown to have a different amino washing step, the bound counts are determined in a
acid and carbohydrate composition from CEA, gamma counter. The bound counts are proportional
though its molecular mass was close to 200,000, to the plasma concentration of CEA, based on a
depending on the method used to measure it. standard curve.
i By the early 1980s, RIA began to be displaced as

Carcino-embryonic antigen (CEA)
the method of choice by enzyme immunoassay
(EIA). In this method, an anti-CEA antibody is
irreversibly bound to a solid phase. These bound
Methods book: Allyn H. Rule
antibodies selectively remove CEA from the fluid
Second edition: Allyn H. Rule
phase. However, other serum proteins will also
Third edition: Gregory A. Hobbs
attach to the positively charged immunoglobulins
Fourth edition: Gregory A. Hobbs
or the negatively charged glycoproteins because of
nonspecific charges or hydrophobic interactions.
328
These nonspecific proteins must be removed by a above EIA in that it uses as a substrate a phosphate
series of washes before addition of a second anti- ester of adamantyl dioxetane which undergoes
CEA antibody, to which is attached a label. The hydrolysis in the presence of the sandwich-
second antibody binds to additional antigenic associated alkaline phosphatase to produce an
determinants on CEA to form a sandwich. unstable intermediate. This intermediate degrades
Nonspecific proteins are again removed by several with the concomitant production of light, detected
washes. In some assays, both CEA and the labeled by a luminometer. This continuous production of
antibody can be added simultaneously in the first light allows for multiple readings, which
step before washing. At this point, reactions can be potentially improves precision. The direct
measured for color development. The color results immunometric method differs in that (1) the
from the addition of a dye substrate in a second or trapping antibody is bound to a magnetic particle.
third step before spectrophotometric determination The bound CEA and complete sandwiches are
of color. The color developed is directly separated from unbound material by physical
proportional to CEA content. The methods differ separation of the particles in a magnetic field, and
only in the enzyme employed, with the former (2) the second antibody is polyclonal and tagged
using alkaline phosphatase with p-nitrophenyl with acridinium. Thus no enzyme reaction is
phosphate as a substrate and the latter horseradish required.
peroxidase with o-phenylenediamine as a substrate.
EIA was an improvement over RIA in a number of There is no reference method for CEA.
ways. The newer method replaced the time-
consuming PCA extraction with heat [14], thus Since cancer patients may be monitored over long
eliminating the problems associated with removing periods of time, considerable thought must be
perchloric acid. A further improvement was the given before a laboratory changes to an alternative
introduction of monoclonal antibodies against CEA kit or method. Because of the heterogeneity of
[15], which dramatically improved the specificity CEA, it is doubtful that all kits measure the same
of the assay, eliminating the need for an extraction entity, so the laboratory must work closely with
step (current RIA also uses a monoclonal physicians ordering the test, justifying the change
antibody). The monoclonal EIA was particularly with improved performance and/or reduced cost.
less susceptible than (polyclonal) RIA to false Careful correlations should be run between the
positives (non-malignancies) due to liver disease. current method and the proposed one, and patient
Conversely, an assay employing a combination of samples should be measured and reported with both
polyclonal and monoclonal antibodies against CEA methods for a mutually agreeable period of time.
was unable to improve discrimination between
malignant and nonmalignant causes of elevated The 2007 College of American Pathologists (CAP)
CEA [16]. Moreover, analytical problems such as ligand general-participant summary report showed
the hook effect have not disappeared with the that the chemiluminescence assays are the most
advent of monoclonal antibodies [15,17,18]. popular technique for measurement of CEA. These
techniques accounted for approximately 90% of all
Currently, the most popular method for the assays. As was suggested above, there are
determination of CEA is an automated appreciable differences among methods as regards
modification of the two-step, monoclonal EIA accuracy (mean) and precision (coefficient of
method. In this approach, the trapping antibody is variation, or CV). Differences in mean
attached to a microparticle, which is retained on a measurements probably reflect antibody
glass-fiber filter in the washing steps. Detection is differences. This is likely because although
not by color but by the production of a fluorescent antibodies differ among manufacturers, the
product, which further improves the precision of calibrators are all traceable, directly or indirectly,
the assay at low concentration. Because it is to the original Gold and Freedman isolate of CEA.
automated, such run-to-run factors as incubation However, this is not necessarily true of CAP survey
time and temperature and washing steps are more material. This material may possess a different set
consistent. This has resulted in improved analytical of epitopes which will vary in affinity for different
performance [19]. This method is referred to as antibodies. Thus the CEA measurements in a given
microparticle enzyme immunoassay (MEIA). A sample may differ considerably among methods
digest of the above methods, with their underlying employing antibodies of various specificities.
principles is presented in Table 1.
Specimen
Methods using chemiluminescence as a means of At one time, EDTA plasma was the preferred
detection, either in an enzyme immunoassay or in a sample for CEA analysis because of interference
direct immunometric assay [20,21], have also been from plasmin, a calcium-dependent proteolytic
developed. This EIA approach differs from the enzyme. However, more recent methods can
329
measure CEA with equal accuracy in plasma or be useful as a mass screening tool [34], though one
serum. For consistency of results, it is preferred Chinese study recommends it for persons over the
that the same matrix be used for the one patient. age of 50 in their population [35]. Currently, the
CEA is stable for up to 1 week when stored at 4C most useful application of CEA is in the detection
[22]. of liver metastases from colorectal cancer [36].
Interferences The use of CEA ligand assays in monitoring the

Specimens with moderate hemolysis or lipemia or therapy of other types of cancer is a more dubious
bilirubin concentrations up to 30 mg/dL typically practice. Though CEA is effective at distinguishing
do not interfere with most CEA methods. The patients with various cancers from disease-free
presence of human anti-mouse antibodies may controls, it is not selective for various tissues [37]
result in falsely increased or decreased values. and shows poor diagnostic sensitivity for cancers
Freeze/thaw cycles have been found to not affect other than colon cancer [28]. Measurement of CEA
CEA concentrations [23]. cannot be the sole factor used for the diagnosis of
disease; it must be used together with clinical
CEA Reference Interval findings.
A value of 2.5 ng/mL has been used to distinguish
healthy patients from diseased [24]. Commonly, Carcinoembryonic Antigen Performance Goals
the reference interval for smokers is 0 to 5 ng/mL, At present, CEA is not regulated for proficiency
though this may vary considerably from study to testing. The current target for acceptable
study [17,25]. Sensitivity, specificity, and performance is that laboratories be within 3
predictive value in selected studies shows various standard deviations of the peer-group mean. The
diagnostic sensitivities, specificities, and predictive CV of most methods is less than 8% at a CEA
values that have been published [26]. It is clear that concentration of approximately 12 ng/mL. One
selection of method and cutoff concentrations study found the intraindividual variation of CEA in
profoundly influence the diagnostic power of CEA healthy individuals to be approximately 10% [38].
immunoassay.
References
Interpretation 1 Gold P, Freedman SO. Demonstration of
Unfortunately, CEA has proven to not be specific tumor-specific antigens in human colonic
for colon cancer in numerous clinical trials carcinomata by immunologic and
[24,27,28] using CEA-RIA. CEA has found its absorption techniques. J Exp Med
most important use in the detection of recurrence of 1965;121:439-462.
colorectal cancer after resection. One study [29] 2 Gold P, Freedman SO. Specific
showed a significant increase in 5-year survival carcinoembryonic antigens of the human
rates when resections were directed by elevated digestive system. J Exp Med
CEA as opposed to other indicators. Others have 1965;122:467-481.
found CEA to be elevated before clinical diagnosis 3 Krupey J, Gold P, Freedman SO.
of recurrence, especially if a critical value of 7.5 Physicochemical studies of the
ng/mL was used [30]. A study comparing CEA carcinoembryonic antigen of the human
with computerized tomography (CT) reported that digestive system. J Exp Med
CEA provided the first indication of recurrence in 1968;128:387-398.
78% of patients, and 100% of recurrences could be 4 Krupey J, Wilson T, Freedman SO, Gold
detected with a combination of CEA, symptoms, P. The preparation of purified
and physical examination [31]. They concluded carcinoembryonic antigen of the human
that CEA remains the most accurate indicator of digestive system from large quantities of
recurrence. The clinical usefulness of preoperative tumor tissue. Immunochemistry
CT scans may be enhanced if they are used only 1972;9:617-22.
after a CEA value of 6.1 ng/mL [32]. Though 5 Rule AH, Ball HO, Rajarajan B, Diaz S.
monoclonal antibodies have improved the An immunochemical comparison of colon,
specificity of CEA for carcinomas, one report [33] ovarian, and breast carcinoembryonic
has criticized the use of CEA in monitoring antigens (CEA). Presented at the XII
patients with resected colon cancer. These workers annual meeting of the International
argue that monitoring patients with CEA (and Society for Oncodevelopmental Biology
CEA-directed surgery) does not result in and Medicine, Houston, Texas, Oct. 4,
significantly improved survival and thus does not 1984.
justify the large (overall) expense of CEA. This 6 Vrba R, Alpert E, Isselbacher KJ.
would seem to be more an indictment of the Immunological heterogeneity of serum
treatment for colorectal cancer, which is poor, than carcinoembryonic antigens.
for any means of detection. CEA is not thought to Immunochemistry 1976;13:87.
330
7 Bast RC, Klug TL, St. John E, Jenison E, 19 Chen I-W, Barnes E, Sperling M.
Niloff JM, Lazarus H et al. Evaluation of a fully automated
Radioimmunoassay using a monoclonal immunoassay system (IMx) for
antibody to monitor the course of carcinoembryonic antigen (CEA). Clin
epithelial ovarian cancer. N Engl J Med Chem 1993;39:1187.
1983;309:883-887. 20 Fritsche HA, Teodoro C, Veloso C,
8 Bast RC, Feeney M, Lazarus H, Nadler Geitner A. Evaluation of an automated
LM, Colvin RB, Knapp RC. Reactivity of chemiluminescence method (ACS:180)
a monoclonal antibody with human for carcinoembryonic antigen (CEA). Clin
ovarian carcinoma. J Clin Invest Chem 1993;39:1192.
1981;68:1331-1337. 21 Smith C, Schwartz MK, Schwartz D.
9 Klug L, Ledonne NC, Greber TF, Evaluation of carcinoembryonic antigen
Zurawski VR. Purification and (CEA assayed on the Ciba Corning
composition of a novel gastrointestinal ACS:180 chemiluminescent
tumor-associated glycoprotein expressing immunoanalyzer. Clin Chem
sialylated lacto-N-fucopentaose II (CA 19- 1993;39:1189.
9). Cancer Res 1988;48:1505-1511. 22 Dale JC, Pruett SK. Phlebotomy: a
10 Thompson DMP, Krupey J, Freedman SO, minimalist approach. Mayo Clin Proc
Gold P. The radioimmunoassay of 1993;68:249-255.
circulating carcinoembryonic antigens of 23 DiMagno EP, Corle D, O'Brien JF,
the human digestive system. Proc Natl Masnyk IJ, Go VL, Aamodt R. Effect of
Acad Sci USA 1969;64:161-167. long-term freezer storage, thawing, and
11 Rule AH, Kirch ME. Gene activation of refreezing on selected constituents of
molecules with carcinoembryonic antigen serum. Mayo Clin Proc 1989;64:1226-34.
determinants in fetal development and in 24 Hansen HJ, Snyder JJ, Miller E,
adenocarcinoma of the colon. Cancer Res Vandevoorde JP, Miller ON, Hines LR,
1976;36:3503-3509. Burns JJ. Carcinoembryonic antigen
12 Lo Gerfo P, Krupey J, Hansen JH. (CEA) assay: a laboratory adjunct in the
Demonstration of an antigen common to diagnosis and management of cancer.
several varieties of neoplasia. N Engl J Hum Pathol 1974;5:139-47.
Med 1971;285:138. 25 Verdi CJ, Ahmann FR, Schifman RB,
13 Tomita JT, Kim YD, Schenck JR et al. Elvick AL, Ahmann ME, Marx PC.
Ligand assay of CEA: solid-phase Comparative evaluation of serum CA 195
immunoassays for CEA. J Clin and carcinoembryonic antigen in
Immunoassay 1971;7:138. metastatic carcinoma. Cancer
14 Smith AM, Macdonald DJ. Enzyme 1993;71:3625-32.
immunoassay for carcinoembryonic 26 Rule, AH. Carcinoembryonic antigen. In
antigen. Clin Chem 1983;29:2019-2021. Pesce AJ, Kaplan LA, eds. Methods in
15 Hedin A, Carlsson L, Berglund A, Clinical Chemistry. St Louis: Mosby;
Hammarstom S. A monoclonal antibody- 1987:710-712
enzyme immunoassay for serum 27 National Cancer Institute/American
carcinoembryonic antigen with increased Cancer Society. Joint investigation: a
specificity for carcinomas. Proc Natl Acad collaborative study of a test for
Sci USA 1983;80:3470-3474. carcinoembryonic antigen in the sera of
16 Buchegger F, Phan M, Rivier D, Carrel S, patients with carcinoma of the colon and
Accolla RS, Mach J-P. Monoclonal rectum. Can Med Assoc J 1972;107:25.
antibodies against carcinoembryonic 28 Laurence DJ, Stevens U, Bettelheim R,
antigen (CEA) used in a solid-phase Darcy D, Leese C, Turberville C et al.
enzyme immunoassay. First clinical Role of plasma carcinoembryonic antigen
results. J Immunol Methods 1982;49:129- in diagnosis of gastrointestinal, mammary,
139. and bronchial carcinoma. Br Med J
17 Liu Y-S V, Tobias RJ, Zurawski VR. A 1972;3:605-9.
more specific, simpler radioimmunoassay 29 Minton JP, Hoehn JL, Gerber DM. Results
for carcinoembryonic antigen with the use of a 400-patient carcinoembryonic antigen
of monoclonal antibodies. Clin Chem second-look colorectal cancer study.
1985;31:191-195. Cancer 1985;55:1284.
18 Hobbs GA, Hess PP, Watson M, Valdes 30 Carlsson U, Stewnius J, Ekelund G,
R. A hook effect detected in a two-step Leandoer L, Nosslin B. Is CEA analysis of
immunoassay. J Clin Ligand Assay value in screening for recurrences after
1995;18:200-204.
331
surgery for colorectal carcinoma? Dis 35 Colorectal Cancer Cooperating Group.

Colon Rectum 1983;26:369-73. Beijing Society of Gastroenterology.
31 Sardi A, Agnone CM, Nieroda CA, Comparative analysis of patients with
Mojzisik C, Hinkle G, Ferrara P et al. clinically diagnosed colorectal cancer and
Radioimmunoguided surgery in recurrent those detected by mass screening. Chin J
colorectal cancer: the role of Dig Dis 2005;6:26-30.
carcinoembryonic antigen, computerized 36 Duffy MJ. Carcinoembryonic antigen as a
tomography, and physical examination. marker for colorectal cancer: is it
South Med J 1989;82:1235-44. clinically useful? Clin Chem 2001;47:624-
32 Holt AD, Kim JT, Murrell Z, Huynh R, 630.
Stamos MJ, Kumar RR. The role of 37 Chevinsky AH. CEA in tumors of other
carcinoembryonic antigen as a predictor of than colorectal origin. Semin Surg Oncol
the need for preoperative computed 1991;7:162.
tomography in colon cancer patients. Am 38 Sltormos G, Schiler V, Nielsen D,
Surg 2006;72:897-901. Skovsgaard T, Dombernowsky P.
33 Moertel CG, Fleming TR, Macdonald JS, Interpretation of results for tumor markers
Haller DG, Laurie JA, Tangen C. An on the basis of analytical imprecision and
evaluation of the carcinoembryonic biological variation. Clin Chem
antigen (CEA) test for monitoring patients 1993;39:2077-83.
with resected colon cancer. JAMA 39 Primus FJ, Freeman, JW, Goldenberg DM.
1993;270:943-7. Immunological heterogeneity of
34 Tumor Marker Expert Panel. Clinical carcinoembryonic antigen: purification
practice guidelines for the use of tumor from meconium of an antigen related to
markers in breast and colorectal cancer. J carcinoembryonic antigen. Cancer Res
Clin Oncol 1996;14:2843-2877. 1983;43:679.
332
Table 1: Methods of Analysis for Carcinoembryonic Antigens (CEA)

Method 1: Indirect radioimmunoassay (RIA); competitive-binding assay
a. Acid extraction
b. Acid removal
1. Dialysis
2. Chromatography
3. Ultrafiltration
a. Plasma is extracted to remove 95% of all other proteins and release CEA from
complexes. After centrifugation and removal of all acid, antibody first reacts with cold CEA for 30
minutes at 45C and then with 125I-CEA for 30 minutes at 45C. Heat speeds the reaction; sequential
addition gives better low-end sensitivity.
b.
1. Dialysis allows acid to diffuse totally out of cellophane pores after many changes of
water and a final buffer step.
2. CEA is separated from acid based on molecular mass differences; CEA comes out in
first fraction.
3. CEA is retained on top of filter or inside of bag while acid is removed. Serum CEA
sticks to filter.
Comments:
a. Historical; more CEA is available to measure. Assay is sensitive to pH, ions, and
nonspecific protein quench.
b.
1. Method is cheap but tedious and labor intensive.
2. Columns are moderately expensive and must be carefully washed, or they can
become contaminated.
3. Expensive equipment; CEA values are significantly lower.
Method 2: Indirect RIA (heat extraction); competitive binding
Principle of analysis: Heat (70C) causes most serum proteins to precipitate out; after centrifugation,
assay performed on supernatant as in Method 1.
Comments: Historical; complex CEA is not released to supernatant.
Method 3: Direct RIA assay; competitive binding
Principle of analysis: Plasma is directly placed into distilled water with antibody, and assay proceeds
similarly to that in Method 1.
Comments: Historical; used to monitor patients with CEA 20 ng/mL, owing to lack of sensitivity in
low range; sensitive to ion, pH, and protein effects.
Method 4: Solid-phase extraction; sandwich immunoassay
a. One-step assay
b. Two-step assay
Principle of analysis: Antigen binds to a monoclonal anti-CEA fixed on a bead, microparticle, or
solid surface. A second anti-CEA labeled with 125I (RIA) or conjugated with an enzyme (EIA).
a. Trapping antibody, sample, and signal antibody together at the same time. One wash step
removes unbound conjugate.
b. Two wash steps; first to remove unbound material from solid phase, then a second wash to
remove unbound conjugated anti-CEA.
Comments: Plasma or serum; most widely used. Extraction step eliminated. Highly sensitive but still
susceptible to interferences. EIA incubation time consuming.
a. Timing of incubation important in EIA
b. First wash reduces potential for nonspecific interference. Most systems semi-automated;
washes and incubations automated.
333
Procedure: Direct Radioimmunoassay of CEA further absorption.
(This is a homebrew method described in the earlier Add 50 L of the plasma fraction of the A-positive
edition of Methods in Clinical Chemistry and is to be blood per milliliter of the above-treated antiserum,
used as an example only.) incubate overnight at 0C to 4C and for 30 minutes at
Principle 37C. Centrifuge at 29,000 g for 20 minutes at 0C to
This is a competitive binding assay using 125I- 4C. Save the supernatant for further absorption.
labeled CEA and a second antibody to separate
antibody-bound labeled CEA from unbound. The CEA Prepare a normal tissue absorbent by homogenizing
molecules are not extracted before analysis. fresh human liver, lung, spleen, and colon. Mince each
Reagents tissue, suspend in saline, and wash free of hemoglobin
1. Water. Use deionized distilled water (DW) or by repeated centrifugation. Homogenize the tissue with 2
distilled water that is both prefiltered and postfiltered. mL of saline for each gram of tissue. Centrifuge the
2. CEA-free plasma (NHP). Obtain plasma suspension at 29,000 g. The resulting pellet will appear
samples from healthy persons, assay, and pool those to be a greenish gel. Combine equal volumes of the
with less than 2 ng of CEA/mL. Alternatively, outdated preparations and add to an equal volume of the
citrated plasma from A-positive blood donors may be previously absorbed antiserum. Incubate for 30 min at
used. This should be extensively dialyzed at 0C to 4C 37C and for 60 min at room temperature. Centrifuge at
against many changes of 0.05 M NaCl during a 24-hr 29,000 g for 30 min, and discard the pellet. Repeat the
period, centrifuged at 29,000 g for 30 min to remove centrifugation procedure. The absorbed antiserum is
aggregated proteins, and tested for the presence of CEA. stored at 20C. Stable for at least 3 years.
Only those plasmas with values less than 2 ng/mL 4. Radiolabeled CEA. 125I-CEA may be
should be used. The NHP is frozen in 40-mL aliquots purchased from Roche Diagnostics, Nutley, NJ.
and then thawed before use, and the unused reagent is 5. CEA standards. Controls and secondary
discarded. antibody may be purchased separately from Roche
3. Antibody to CEA. Obtain a commercially Diagnostics.
available antiserum, such as the one sold by Dako. These 6. Goat anti-rabbit serum antisera. Can be
preparations usually have antibodies to other proteins purchased from many commercial suppliers.
and should be absorbed before use.
Alternatively, prepare antisera from CEA Anti-CEA antibody titration
purified from an adenocarcinoma of the colon. Mark 16 100 mm glass tubes in duplicate as
Homogenize the carefully trimmed cancer in 2 mL of follows:
0.15 M NaCl for each milligram of tumor. Remove the 1. Total counts: TC
insoluble material by centrifugation at 29,000 g for 30 2. Nonspecific binding: NSB CEA
minutes. Further purify the CEA by adding an equal antisera diluted 100-fold with distilled
volume of 1.2 M perchloric acid to the supernatant to water
precipitate proteins. Centrifuge the mixture at 29,000 g 3. 100 L
for 30 min at 4C. Dialyze the resultant supernatant 4. 50 L
extensively to remove the perchloric acid. Purify the 5. 25 L
antigen further by sepharose-6B column 6. 10 L
chromatography. The CEA fractions are concentrated by CEA antisera diluted 2000-fold with distilled
lyophilization and stored at 20C. Elicit an antibody water
response in rabbits using four biweekly injections of 10
7. 100 L
to 100 g of protein with Freunds adjuvant. Test for
8. 50 L
antibody response by use of Ouchterlony plates.
9. 25 L
Both commercial and individually prepared antisera 10. 10 L
should be absorbed with normal human tissues (such as CEA antisera diluted 40,000 fold with distilled
lung, liver, and spleen) as described below before use. water
11. 100 L
Obtain fresh red blood cells of the A-positive blood type, 12. 50 L
and wash thoroughly in 0.15 M saline. Do not refrigerate 13. 25 L
before use. Decomplement the antisera by heating to 14. 10 L
56C for 10 min in a water bath before absorption.
Dilute the packed red blood cells to 30% volume to 1. Each tube, except for TC, contains 5.0 mL of
volume with the antisera. Incubate for 60 min at room distilled water and 25 L of dialyzed CEA-
temperature, centrifuge at 1500 g for 30 min at room free normal human plasma (NHP) before anti-
temperature, and remove the supernatant fraction for CEA addition: TC contains 0.5 mL of distilled
334
water but no NHP. Vortex mix the tubes after each b. To all tubes except TC, add 5 mL of
addition. distilled water. Add 500 L of distilled water to
2. Add antibody diluted in distilled water to TC tube.
each tube (except for NSB) as indicated c. To all standards, including NSB and I0, to
above, and incubate for 30 min in a water all control tubes, and to samples that do not contain
bath set at 45C. plasma or serum, add 25 L of normal human
3. Remove tubes, add 50,000 counts of 125I- plasma (NHP).
CEA in 50 L, and again incubate for 30 NOTE: Since CEA tends to adhere to the sides of
min at 45C. glass tubes, it is essential to complete steps 1b and
4. Remove from water bath, add 1 mL of 1c before adding standard or samples to tubes.
appropriate diluted second solution of goat d. To standard tubes (tubes 4 through 8), add
anti-rabbit antibody, and let tubes sit for the indicated volumes of standard containing 125
15 min at room temperature before ng of CEA/mL.
centrifugation at 1500 to 2000 g for 15 5 L for I5 = 0.625 ng/mL
min. 10 L for I10 = 1.25 ng/mL
5. Decant supernatants and count tubes in a 25 L for I25 = 3.12 ng/mL
gamma-ray scintillation counter at 80% 50 L for I50 = 6.25 ng/mL
efficiency. 100 L for I100 = 12.5 ng/mL
Calculation e. To control tubes, add the appropriate
Standard curves are constructed and read as cpm amount of in-house CEA preparation to yield
versus ng of CEA or B/B0 versus ng of CEA. If final concentrations of approximately 1, 2, 4, and 8
ng/mL.
one is comparing different monoclonal or organ- f. Samples: Add appropriate amounts of
specific anti-CEA reagents, the latter calculation sample (10 to 100 L) so that results fall within the
should be made. If possible, choose the antibody standard curve. For samples above 500 ng of
concentration for future assays that yields 35% to CEA/mL, dilute 10-fold with distilled water.
50% binding or at least 10,000 counts of 125I- 2. To all tubes except TC and NSB, add 50
CEA. L of diluted CEA antiserum. Vortex mix.
3. Incubate tubes in a water bath at 45C for
If monoclonal or in-house polyclonal antibodies 30 min.
are being assessed, they should be absorbed with 4. To all tubes, add 50 L of diluted 125I-
normal spleen, liver, and lung at 30% w/v for 30 to CEA. Vortex mix.
60 min at 37C and overnight at 0C to 4C as 5. Return to water bath for 30 min.
previously described to remove antibodies to cross- 6. To all tubes except TC, add 1 mL of
reacting antigens found on normal neutrophils and second goat/anti-rabbit antibody suspension. Make
normal colon. If all or part of the CEA reactivity is sure it is thoroughly mixed into the reaction
lost after absorption, the antibodies were not mixture by vortex mixing.
cancer-specific reagents suitable for clinical use. 7. Let stand at room temperature for 15 min,
Stock anti-CEA at 1:100 (etc.) can be made in and then centrifuge for 15 min at 1500 to 2000 g at
the same normal serum of the species used to raise a temperature of 4C to 25C.
anti-CEA antibodies and frozen at 20C. These 8. Decant carefully, and count in gamma-ray
titrations and direct CEA procedures have been scintillation counter.
used to measure CEA obtained from tumors, Calculations
meconium, fetal guts, or ascites fluids [11,39]. Prepare standard curve on semilogarithmic graph
Assay paper using counts per minute (cpm) on the vertical
Equipment: a gamma-ray scintillation counter and axis and nanograms of CEA (ng of CEA) on the
refrigerated centrifuge capable of 1500 to 2000 g horizontal logarithmic axis. Read unknown counts
force. per minute off of the standard curve, and multiply
1. Setup by correct dilution factor to obtain results, in ng of
a. Label 16 100 mm glass test tubes in CEA/mL.
duplicate as follows: Notes
(1) Total counts: TC 1. The direct assay of CEA is exceptionally
(2) Nonspecific binding: NSB sensitive to ions, pH, colloidal metals, bacterial
(3) Maximum binding: I0 products, and organic molecules. The water should
(4-8) Standards: I5, I10, I25, I50, and be filtered before use, since bacteria can grow in
I100 distilled-water systems. Additionally, improperly
(9-12) Controls: 5, 10, 25, and 50 L of deionized tanks may leak resins. In troubleshooting
a CEA sample diluted in distilled water to contain problems with the CEA assay, one must thoroughly
approximately 800 ng/mL. check the water as a possible cause of problems.
(13, etc.) Samples: 1, 2, . . .
335
2. Many problems arise in the control of previously analyzed CEA-containing plasma

CEA values. Calibrators must be independent of samples. Then 100, 50, 20, and 6 L of these
the current laboratory methods, whether they are standards are delivered into the appropriate assay
internally devised or obtained from an external tubes along with the appropriate amounts of CEA-
commercial source. Additionally, since cancer free plasma normally required for that assay, that
cures are considered to be obtained after a 5-year is, 0.5 mL of plasma for perchloric acid extraction
interval, standards must be available in sufficient or 50 to 200 L for various direct or unextracted
quantity to ensure long-term quality control. This is CEA ligand assays. The small amount of extra
particularly true if one contemplates changing CEA volume (water) does not alter the CEA measured
methodologies, manufacturers, or source materials. by any of the currently available commercial CEA
Because of the heterogeneity of CEA-reacting assays, nor does it affect the protein matrix (ions,
molecules, it is well known that CEA values pH, or salt) if deionized distilled water is used as
obtained from one method cannot be obtained with diluent.
a second method, and a correction factor cannot
be applied between different standards or different My second recommendation is that the reference
assays. Above all, one should not obtain reference CEA preparation prepared in the laboratory be
CEA from the same company from which one experimentally tested by a reference laboratory
obtains reagents, because the ranges of the whose CEA standard has been experimentally
reference material will be established by the same calibrated against early Phil Gold and Roche-CEA
assay used to evaluate the reagents. It will be preparations from the past 17 years. If the
impossible to determine if assay results have been laboratory chooses to purchase reference controls
drifting over the years. in quantity, these also should be sent out to a well-
standardized, qualified laboratory.
How should one proceed? Obtain your own
hospital or laboratory standards. First realize that Finally, standardized CEA reference serum should
CEA has traditionally been quantitated in be run once monthly with the assay of your choice
immunochemical reactivity in nanograms of CEA or at any time when CEA values (or curves) seem
per milliliter. This has no reality in terms of to drift. Sufficient standards should be obtained to
nanograms of CEA per milligrams of protein. The last for a 6-year period. However, in the fifth year,
carbohydrate-to-protein ratio may vary from 1:1 to new reference control materials should be
1:4 in different preparations. CEA purified from calibrated against old standards so that continuity
different tumors will have different immunological in monitoring patient samples is ensured.
reactivities because of the different number of 3. In patients with bladder cancer, CEA-
CEA-reacting antigenic sites: that is, ovarian CEA reacting substances are found in urine. If an RIA is
> colon CEA > breast CEA. However, in terms of used to measure urinary CEA, 25 L of fresh
serum CEA: colon CEA > breast CEA > ovarian normal urine must be added to each point of the
CEA. Additionally, spiking commercially obtained standard curve while 25 L of the urine obtained
CEA into CEA-free plasma will give spurious from the cancer patient and 25 L of NHP must be
effects because of the variability of these in the unknown sample or samples.
preparations, lyophilization aggregation, 4. Ascites fluids that contain heparin may
nonspecific heterologous reactions, and deleterious nonspecifically quench the CEA-RIA, even after
matrix effects [5]. excessive dialysis, and thus their use is not advised.
However, if an ascites fluid comes from a patient
Our first recommendation is that the investigator with exceptionally high CEA and is needed for
obtain a fluid containing an exceptionally high research purposes (or CEA standard), heparin can
amount of CEA from a patient with colon cancer. be removed using Sephadex G-100 column
Dilute this in distilled water to achieve a final chromatography. CEA is eluted in the void volume
concentration that falls below the high end of the while heparin (9000 to 20,000 D) is retained in the
standard curve with the addition of the appropriate stationary phase.
amount of CEA-free plasma required for that assay. 5. If ascites, pleural effusion fluids, or spinal
This latter addition controls nonspecific protein cord fluid is tested for CEA content, the 25 L of
quench. CEA at 0.5, 2.0, 4.0, and 8.0 ng/mL should NHP may be omitted from the assay procedure. For
be used for assessment of low-end sensitivity. very dilute CEA samples, for example, those from
These CEA values should parallel the standard peritoneal washes, purification steps, and so on, the
curve. In our laboratory, our group uses an ascites addition of 25 L of NHP in the assay is very
fluid that contains 16,000 ng of CEA/mL, as necessary. If doubt exists, add the NHP. Do not
measured by perchloric acid extraction. Aliquots of extract CEA from dilute solutions. Enzyme
0.5 mL of 1:100 (perchloric acid assay) and 1:200 immunoassays are not advised for research uses,
(direct assay) dilutions are frozen and used monthly that is, purification of CEA, because of their
for monitoring of CEA assays, along with five limited specificities and high expense.
336
6. The direct method is sensitive to high salt 7. Normal CEA-free plasma (NHP) is used
or protein levels, acid, base, and different ions. If to investigate nonspecific binding and to coat the
dilutions of CEA-containing samples, used for inner surface of glass tubes. Previous directions
research purposes, do not yield linear curves that must be carefully followed because all the citrate
parallel the standard curve, they should be dialyzed must be removed before the CEA assay. The use of
against dilute saline (0.03 M) or distilled water. It A-positive plasma is suggested so that only anti-B
is also important to dilute protein samples that have plasma heterospecific antibodies will be present.
more than 100 to 120 mg of protein/mL because These are not likely to interact with plasma
nonspecific protein binding will result. samples, since more than 90% of the people on this
continent have A-positive or O-positive plasma.
Thus the various assays for CEA serve many
clinical and research purposes. With good quality
control, both the RIA and EIA measurement of
CEA yield dependable clinical data.
337
Catecholamines (Plasma)
Brett McWhinney
Name: Catecholamines: epinephrine (E), adrenaline, adrenalin norepinephrine (NE),
noradrenaline, levarterenol dopamine (DA), dihydroxyphenylalanine (DOPA),
dihydroxylphenylactic acid (DOPAc), dihydroxyphenylglycol (UI)(PG)
Clinical significance: Refer to Chapter 36, Cardiac and Muscle Disease, in the 5th edition of Clinical
Epinephrine Norepinephrine Dopamine
Structural formula: See below See below See below
Molecular formula: C9H13NO3 C8H11NO3 C8H11NO2
Molecular mass: 183.20 D 169.18 D 153.18 D
Merck Index: 3543 6504 3422
Chemical class: Catechol derivatives
Dihydroxypheny- Dihydroxyphenyl- Dihydroxyphenyl-
alanine acetic Acid glycol
Molecular formula: C9H8O4 C8H5O4 C8H7O7
Molecular mass: 197.2 D 168.1 D 170.2 D
Merck Index:
Chemical class: hydroxyphenyl
Norepinephrine Epinephrine Dopamine

i with 1 (E) or 2 (NE) moles of ethylenediamine, with the
elimination of water and hydrogen ions to form
The earliest chemical methods for the analysis of plasma fluorescent products. Because the two fluorescent
catecholamines were fluorometric, involving the products derived from E and NE are different, E and NE
oxidation and reduction of the catecholamines to levels can be estimated by measurement of the
fluorescent products. The two most widely used fluorescence of the final reaction at two different
fluorometric assays were the trihydroxyindole (THI) and wavelengths, 510 nm and 580 nm, after excitation at 420
ethylenediamine (EDA) methods [1,2,3]. In the THI nm. The fluorescence of the product derived from
method (Table 1, Method 1), epinephrine (E) and epinephrine is half the fluorescence derived from the
norepinephrine (NE) are oxidized by a suitable oxidizing norepinephrine product at 510 nm, whereas at 580 nm
agent such as ferricyanide, and an intramolecular the ratio is reversed; simultaneous equations can be
cyclization then occurs, forming the corresponding established to estimate E and NE levels.
adrenochrome. At an alkaline pH, the adrenochromes
rearrange to form adrenolutin and noradrenolutin. The Both the THI and EDA methods require some sample
latter two compounds fluoresce at 505 nm after purification to separate the catecholamines from
excitation at 400 nm, allowing quantitation of total interfering compounds. The purification steps typically
catecholamines (NE + E). include alumina and cation-exchange column
In the ethylenediamine method (Table 1, Method 2), E chromatography and often include a liquid-liquid
and NE are oxidized, and the oxidation products react extraction.
A method used frequently to quantitate plasma
catecholamines is the radioenzymatic assay (Table 1,
i
Catecholamines (Plasma) Method 3). The radioenzymatic assay is based on the
Previous and current authors of this method: method of Peuler and Johnson [4]. This rather laborious
First edition: Steven J. Soldin procedure radiolabels the plasma catecholamines by
Methods edition: Lawrence E. Webb, Lawrence A. methylating them in the following reaction:
Kaplan
Second edition: Not updated 3H-SAM + epinephrine/norepinephrine
Third edition: Not updated COMT3H-metanephrine/3H-normetanephrine
Fourth edition: Theodore Dashman, Stanley Samuels 37C, 60'
Fifth edition: Brett McWhinney
338
to selectively extract the vanillin and increase the

3H-SAM is [3H-methyl]-S-adenosyl-L-methionine, and specificity of the assay. Highly purified COMT
COMT is catechol-O-methyltransferase (EC 2.1.1.6). preparations are not readily available, and inhibitory
substances present in the enzyme preparations can also
The catecholamines are extracted by a liquid-liquid interfere in the methylation reaction. These inhibitors in
extraction, and the methylated derivatives both plasma and enzyme cause a large variability in the
(metanephrine, normetanephrine, and 3- production of the 3H-labeled O-methylated products.
methoxytyramine [from dopamine]), labeled and Suggestions for removing these inhibitors include the
unlabeled, are separated by thin-layer chromatography addition of EGTA to bind inhibitory Ca2+ in plasma, the
on silica gel. The metanephrine, normetanephrine, and 3-
precipitation of EGTA-Ca2+ and protein from plasma
methoxytyramine zones are identified by their Rf values
before analysis using HClO4, and the more extensive
and are scraped off and placed in vials. They are then
purification of COMT [13,14]. More purified COMT
eluted from the silica gel, and the 3-methoxytyramine
preparations contain a smaller quantity of interfering
eluate is counted directly. The metanephrine fractions
enzymes such as L-aromatic amino acid decarboxylase
are then oxidized to 3H-vanillin, which is extracted into [15] and monoamine oxidase [13].
scintillation fluid and counted. To assess recovery of the
final 3H-vanillin from each sample and to assess any Additional problems with the radioenzymatic assay can
interference in the methylation reaction, each sample result from the use of inappropriate Michaelis-Menten
must be processed in duplicate, one sample containing conditions in the critical transmethylation step [14]. If an
added norepinephrine and epinephrine internal standards insufficient concentration of methyl donor (3H-SAM) is
(usually 100 pg each), while the other is unspiked. The present in the reaction mixture, inaccurate data can
internal standard is used to assess the extent of the result. This problem is exacerbated by the addition of
enzymatic reaction and correct for interferences. internal standard, as is recommended in the
commercially available assay. This only lowers the ratio
A method that is becoming the standard for
of 3H-SAM to catecholamines and does not allow proper
catecholamine analysis employs high-performance liquid
comparison of reaction rate in the unspiked and spiked
chromatography (HPLC) with electrochemical detection
(internal standard) samples [14,16]. A 10-fold excess of
(Table 1, Method 4) [5,6,7]. HPLC procedures use a
SAM to catecholamine concentration in the reaction
fairly rapid and simple batch type of alumina extraction
mixture is suggested [14]. The large variability in the
step to partially purify the catecholamines [8]. Cation-
amount of labeled added internal standard recovered
exchange chromatography on Dowex 50 and selective
prompted one group to suggest that an internal standard
absorption on boric acid gels [7] have also been used to
not be run on each patient, but that instead a mean net
partially purify the catecholamines [9]. The alumina
standard value obtained from many patients be assumed
HClO4 extract is analyzed directly, usually by reversed-
[17]. This procedure, however, only circumvents the real
phase HPLC (most often, C18 columns), with the mobile problems of individual interferences, which should be
phase usually consisting of an acid buffer (phosphate, corrected. The use of a partial purification of the
phosphate-citrate, or monochloroacetic acid), small catecholamines by an alumina extraction before the
amounts of a polar organic solvent (methanol), and a methylation reaction [14], the use of purified COMT
paired-ion agent (such as heptane- or octanesulfanate). preparation [12,14], and the use of higher levels of SAM
Cation-exchange stationary phases have also been in the reaction mixture [14] seem to be more appropriate
employed in HPLC analysis [10,11]. The radioenzymatic suggestions.
and HPLC methods are the most commonly used
methods for the measurement of plasma catecholamines. The HPLC procedures also have deficiencies. The
review of methods of plasma catecholamine analysis by
Hjemdal [15] showed that although the various HPLC
Reference and Preferred Methods techniques seemed to be appropriate, with generally
There is no current reference method for the analysis of excellent correlation between HPLC and radioenzymatic
plasma catecholamines; however, HPLC with procedures, the ability of individual laboratories to
electrochemical detection appears to give the best recover added analyte or to demonstrate good precision
specificity and sensitivity for the low levels seen in was limited. The fluorometric trihydroxyindole
patients. The fluorometric methods lack sensitivity for procedure was inferior to the HPLC assays. The
plasma catecholamines. The purification steps required technology of the HPLC electrochemical procedures has
to remove interfering compounds only add to the advanced dramatically in recent years. Improvements in
tediousness of this method. The reported precision of the column technology, better control of column
radioenzymatic procedures seems acceptable, and the temperature, and the introduction of the triple-electrode
assay has the capability of reasonable accuracy and coulombic detection system has enhanced the
sensitivity. The major problems with the assay derive quantitation of NE, E, DA, and other catecholamine
from material present in the plasma interfering in the metabolites [7,15]. Thus an HPLC electrochemical
methylation reaction [12]. One of the reasons for procedure is the recommended method for measurement
introducing the final step, oxidizing the radiolabeled of plasma catecholamines.
metanephrine and normetanephrine to 3H-vanillin, was
339
Specimen
Sample collection and storage must be rigorously Caution: Collection of blood from children is
controlled. Heparinized plasma is the preferred sample. particularly difficult because of their normal fear of
During transport from patient to the laboratory, blood venipuncture and their restlessness; for this reason,
samples should be maintained at 4C. After removal of catecholamine levels from excited children may be
the cellular matter, plasma is immediately analyzed or misleading.
stored by freezing at 70C to 80C. However, if the
sample is to be forwarded to a distant location, it should Plasma is separated within 30 min of sample collection
be frozen and shipped in solid CO2. by centrifugation (3000 g, 20 min, 4C). Aliquots (1.2
mL) are quickly transferred to self-capped centrifuge
Specimen Collection tubes, and 10 mg of sodium metabisulfite is added. The
During collection, the patient should be supine and plasma is immediately analyzed or is stored at 70C to
motionless because movement, even from supine to 80C.
erect, may result in a two- to threefold variation in
plasma catecholamine levels. Also, patients should not Interferences
drink coffee for 24 hours prior to blood collection. The fluorometric methods are not sufficiently specific,
Caffeic acid from coffee may interfere with especially with the number of commonly occurring
catecholamine resolution on HPLC. Smokers should be interfering drugs. These include methyldopa, ampicillin,
warned not to smoke for a minimum of 1 to 2 hours vitamin B complexes, and caffeine. Nicotine does not
before blood collection because of the stimulatory interfere directly with measurement of dopamine,
effects of nicotine. epinephrine, or norepinephrine by HPLC, but smoking
within 2 hours of blood collection may stimulate
To ensure a stable metabolic status of the patient during catecholamine release, and an artifactually increased
blood collection for catecholamine analysis, a two-step result may be obtained.
procedure is recommended.
1. Insert a venous catheter percutaneously in the Plasma Catecholamine Reference Intervals
forearm or antecubital region of the arm. The patient is Using the HPLC-EC method described previously [18],
then requested to lie quietly for 20 to 30 min. (Table 1, Method 2), the following levels of plasma
2. Blood is drawn into a prechilled (4C), catecholamines were found in healthy individuals.
heparinized 5-mL Vacutainer. The filled tube is
immediately transported to the laboratory at 4C.
Normotensive
Norepinephrine <30400 pg/mL (<0.1772.36 nmol/L)
Epinephrine <30100 pg/mL (<0.1640.546 nmol/L)
Hypertensive
Norepinephrine <30700 pg/mL (<0.1774.14 nmol/L)
Epinephrine <30120 pg/mL (<0.1640.655 nmol/L)
Using the method described by Holmes et al. [6], the following levels of plasma catecholamines were found in healthy
individuals.
Catecholamine pg/ml nmol/L

Norepinephrine 348 10 2.06 0.06
Epinephrine 48 2 0.26 0.01
Dopamine 15 2 0.10 0.01
Dihydroxyphenylalanine 1596 47 8.10 0.24
Dihydroxyphenyl acetic acid 1304 59 6.07 0.35
Dihydroxyphenylglycol 1168 55 6.87 0.32
Interpretation only very recent catecholamine synthesis, plasma

The catecholamines are produced by the adrenal medulla catecholamine levels can be within the reference interval
(mostly E) and peripheral nerves (NE and dopamine). in cases of tumors whose catecholamine excretion is
The plasma levels of the analytes are usually increased highly episodic. Plasma catecholamine measurements
in persons with hypertension. Plasma levels are even should be performed for those patients with borderline
higher in patients with pheochromocytomas. Since the elevations (above their reference intervals) of urinary
plasma catecholamine assay is very difficult to perform HMMA (vanillylmandelic acid or VMA),
properly, it is not advocated by most clinicians as the metanephrines, or catecholamines.
primary diagnostic test for pheochromocytomas. In
addition, because plasma catecholamine levels reflect
340
A useful challenge test employing plasma norepinephrine by reverse-phase liquid

catecholamines is the clonidine suppression test. chromatography. Anal Chem 1981;53:156-159.
Clonidine normally will cause a suppression of plasma 9 Frayn KN, Maycock PF. Sample preparation
catecholamine levels. Failure of the plasma with ion-exchange resin before liquid
catecholamine levels to decrease after a dose of chromatographic determination of plasma
clonidine is highly suggestive of a pheochromocytoma. catecholamines. Clin Chem 1983;29:1426-
1428.
10 Allenmark S, Hedman L, Sderberg, A.
Plasma Catecholamine Performance Goals Microanalysis of catecholamines in human
The interlaboratory precision and accuracy for plasma by high-performance liquid
laboratories employing reversed-phase HPLC is chromatography with amperometric detection
generally poor, with an overall % coefficient of variation as compared with a radioenzymatic method.
(CV) precision of < 20%. The most reliable HPLC Microchem J 1980;25:567-575.
analyses appear to be those employing cation-exchange 11 Eriksson BM, Persson BA. Determination of
stationary phases (overall % CV precision of < 10%), a catecholamines in rat heart tissue and plasma
method not widely used in the United States, where C18 samples by liquid chromatography with
reversed-phase seems most popular. In this study by electrochemical detection. J Chromatogr
Hjemdahl, the overall interlaboratory precision and 1982;228:143-154.
accuracy of the laboratories using the COMT procedure 12 Hussain MN, Benedict CR. Radioenzymatic
was similar to that achieved by the laboratories using an microassay for simultaneous estimations of
HPLC method. In a comparison between an HPLC dopamine, norepinephrine, and epinephrine in
procedure using a C18 column and a radioenzymatic plasma, urine, and tissues. Clin Chem
1985;31:1861-1864.
assay [19], a definite bias was shown to exist; the
13 Ben-Jonathan N, Porder NC. A sensitive
radioenzymatic results were higher than the results
radioenzymatic assay for dopamine,
obtained by HPLC.
norepinephrine and epinephrine in plasma and
tissues. Endocrinology 1970;98:1497-1507.
References
14 Mason L, Weinkove C. Kinetics of enzymatic
1 Manger, WM, Steinsland, OS, Nahas, GG,
O-methylation: its value in assays for
Wakim KG, Dufton S. Comparison of improved
catecholamines in plasma. Clin Chem
fluorometric methods used to quantitate plasma
1984;30:24-27.
catecholamines. Clin Chem 1969;15:1101-
15 Hjemdahl P. Interlaboratory comparison of
1123.
plasma catecholamine determinations using
2 Martin LE, Harrison C. An automated method
several different assays. Acta Physiol Scand
for determination of noradrenaline and
1984;527:43-54.
adrenaline in tissues and biological fluids. Anal
16 Johnson GA, Kupiecki RM, Baker CA. Single
Biochem 1968;23:529-545.
isotope derivative (radioenzymatic) methods in
3 Victorio JK, Baukal A, Wolff FW. New
the measurement of catecholamines.
automated fluorometric methods for estimation
Metabolism 29(suppl)1980;1106-1113.
of small amounts of adrenaline and
17 Ellis JP Jr, Burns JW. Modified handling of
noradrenaline. Anal Biochem 1968;23:513-528.
internal catecholamine standards in
4 Peuler JD, Johnson GA. Simultaneous single
radioenzymic assays. Clin Chem 1983;29:144-
isotope radioenzymatic assay of plasma
147.
norepinephrine, epinephrine, and dopamine.
18 Kaplan, LA. Catecholamines, plasma. In: Pesce
Life Sci 1977;21:625-636.
AJ, Kaplan LA, eds. Methods in Clinical
5 Premel-Cabic A, Allain P. Simultaneous liquid
Chemistry. 4th ed. St Louis: Mosby; 1987:944.
chromatographic determination of norepi-
19 Causon RC, Brown MJ, Boulous PM, Perret D.
nephrine, 3,4-dihydroxyphenylethyleneglycol,
Analytical differences in measurement of
3-droxyphenylalanine and 3,4-dihydroxy-
plasma catecholamines [letter]. Clin Chem
phenylacetic acid. J Chrom 1988;434:187-190.
1983;29:735-737.
6 Holmes C, Eisenhofer G, Godstein DS.
20 The procedure is based on the method described
Improved assay for plasma dihydroxy-
in a personal communication with C. Holmes
phenylacetic acid and other catechols using
and G. Eisenhofer and Reference 6.
high performance liquid chromatography with
electrochemical detection. J Chrom
1994;653:131-138.
7 Imai Y, Ito S, Maruta K, Fujita K. Simultaneous
determination of catecholamines and serotonin
by liquid chromatography after treatment with
boric acid gel. Clin Chem 1988;34:528-530.
8 Davis GC, Kissinger PT. Strategies for
determination of serum or plasma
341
Tables Method 3: Radioenzymatic; isotopic-enzymatic

Principle of analysis: Epinephrine and
Table 1: Methods of Plasma Catecholamine Analysis norepinephrine are enzymatically converted to
Method 1: Trihydroxyindole (THI); fluorometric their corresponding tritiated methylated
Principle of analysis: Partially purified derivatives (metanephrines). These are
epinephrine and norepinephrine are oxidized to extracted and separated by thin-layer
corresponding adrenochromes which, at chromatography (TLC). The individual
alkaline pH, rearrange to adrenolutin and compounds are extracted from the stationary
noradrenolutin. Quantitation of fluorescence is
phase of the TLC plate and converted to 3H-
at 505 nm after excitation at 400 nm.
vanillin, and the tritium label is measured.
Comments: Historical; measures total
Radioactivity present is directly proportional to
catecholamines (E and NE); requires one or
concentration of E and NE.
more purification steps
Comments: Laborious method; difficult to
Method 2: Ethylenediamine (EDA); fluorometric
control variation in extent of reaction; problems
Principle of analysis: Partially purified
with purity of enzyme reagent.
epinephrine and norepinephrine are oxidized to Method 4: High-performance liquid chromatography
products that are allowed to react with 1 (E) or (HPLC); chromatographic
2 (NE) moles of ethylenediamine to form Principle of analysis: Catecholamines are
fluorescent derivatives. Quantitation of extracted from plasma by alumina and are
fluorescence of E (510 nm) and NE (580 nm) is directly analyzed by ion-paired, reversed-phase
accomplished after excitation at 420 nm; using HPLC. Detection and quantitation are by an
differential equations, one can individually electrochemical detector.
quantitate E and NE. Comments: Commonly used; requires a fair
Comments: Historical; can quantitate E and amount of technical expertise for reliable
NE; requires one or more purification steps analysis.
Procedure: Catecholamine Analysis in Plasma by 900 mL of type I water. Adjust pH to 3.4, and dilute to
High-Performance Liquid Chromatography with 1000 mL. Add 30 mL acetonitrile to the buffer; mix. The
Electrochemical Detection [20] mobile phase is filtered and degassed by filtering
through a 0.22 m microfiltration system (Kimble Glass
Principle Inc., Vineland, NJ). Prepare mobile phase daily.
Plasma catecholamines, partially purified by an alumina 5. Elution solution (ES), phosphoric acid (40
extraction, are fractionated by HPLC on C18 reversed- mmol/L) plus acetic acid (200 mmol/L), 1:4, v/v.
phase columns and quantitated by electrochemical Phosphoric acid (40 mmol/L). Prepare an intermediate
detection. stock solution of 100 mmol/L phosphoric acid by
Reagents (Stock Standard Solutions Table above) diluting 3.4 mL phosphoric acid (85%) to 500 mL type I
1. Acid-washed alumina. Brockman activity 1, water. Prepare 40 mmol/L phosphoric acid by diluting 4
60-325 mesh, obtained from Fisher Scientific Co., mL of 100 mmol/L phosphoric acid to 100 mL with type
Pittsburgh, PA, stored at 90C and at time of use, cooled I water. Stable 1 year at room temperature if covered.
to ambient temperature.
2. Extraction buffer, Tris buffer (1 mol/L, pH Acetic acid (200 mmol/L). Prepare an intermediate stock
8.6) containing EDTA (53.7 mmol/L). Weigh 121 g solution of 1 M acetic acid by diluting 5.75 mL glacial
of Tris base and 20 mg EDTA-Na2, and dissolve in acetic acid (99.5%) to 500 mL with type I water. Prepare
approximately 800 mL type I water. Adjust to pH 8.6 200 mmol/L acetic acid by diluting 20 mL of the 1 M
with HCl, and dilute to 1000 mL. Stable for 2 months at acetic acid to 100 mL with type I water. Stable 1 year at
4C to 8C. room temperature if covered.
3. Wash buffer, sodium bicarbonate (200
mmol/L) containing EDTA (26.8 mol/L) and To prepare the eluting solvent, add 40 mL of 40 mmol/L
sodium metabisulfite (26.3 mol/L). Weigh 1.6 g phosphoric acid to 160 mL of 200 mmol/L acetic acid.
sodium bicarbonate, 10 mg of EDTA-Na2, and 5 mg Prepare daily.
6. Stock Standards
sodium metabisulfite; dissolve in approximately 80 mL Norepinephrine (NE) (1 mg/mL). Dissolve 19.0 mg
of type I water, and dilute to 100 mL. Prepare fresh norepinephrine bitartrate per 10 mL of elution solution.
daily.
4. Mobile phase, monobasic phosphate (100 Epinephrine (E) (1 mg/mL). Dissolve 18.9 mg
mmol/L, pH 3.4) containing EDTA (134.3 mol/L), 1- epinephrine bitartrate per 10 mL of elution solution.
octanesulfonic acid (184.9 mol), and acetonitrile
(3%). Place 13.8 g sodium monobasic phosphate, 50 mg Dopamine (DA) (1 mg/mL). Dissolve 12.4 mg dopamine
EDTA-Na2, and 40 mg sodium 1-octanesulfonic acid in hydrochloride per 10 mL of elution solution.
a 1-L volumetric flask and dissolve in approximately
342
Dihydroxyphenylalanine (DOPA) (1 mg/mL). Dissolve (Models 5021 and 5011; ESA, Inc.). The first electrode
10 mg dihydroxyphenylalanine free base in 10 mL of in the series was set at 300 mV, the second at 160 mV,
elution solution. and the third at 350 mV.
Data System: Millennium 2.1 data system (Waters,
Dihydroxylphenylglycol (DHPG) (1 mg/mL). Dissolve Milford, MA).
10 mg dihydroxylphenylglycol free base in 10 mL of Procedure
elution solution.
Blanks and standards for plasma specimens
Dihydroxyphenylacetic acid, (DOPAC) (1 mg/mL). 1. BlankTo a polypropylene centrifuge tube add
Dissolve 10 mg dihydroxyphenylacetic acid free acid in 1.0 mL type I water and 10.0 L of DHBA (internal
10 mL of elution solution. standard, IS).
7. Internal standard. Dihydroxybenzylamine 2. Recovery standardsTo a polypropylene
(DHBA) (1 mg/mL). Dissolve 14.1 mg centrifuge tube, add 1 mL type I water and 90 L of test
dihydroxybenzylamine hydrobromide per 10 mL of standard solution.
elution solution. 3. Extraction standardsTo a separate
8. Working standards (100 ng/mL). Dilute polypropylene centrifuge tube, add 1 mL of type I water
separately 0.1 mL of each of the stock standards of the and 90 L of test standard solution.
standard solution to 10 mL with elution solution. Then
dilute each solution 1:100 with ES. Aliquot 1.0 mL of
these solutions (100 ng/mL) in polypropylene tubes, and Plasma
store at 70C to 80C. 1. Centrifuge fresh or thawed plasma (3000 g,
9. Combined standard 10 min, 4C) Transfer 1.0 mL to a 2.0-mL
Amine mixture: Add 100 L of each of the NE, polypropylene centrifuge tube.
EPI, DA working standards and 100 L of elution 2. Add 20 L internal standard to each plasma
solution to a 500 to 1000 L polypropylene tube. The specimen.
concentration of each component of the mixture is 25 3. Add 10 mg alumina from the volumetric spatula
ng/mL. to blanks, extraction standards, QC sample, and test
plasma.
Test standard. To a separate tube, add 100 L of the 4. Add 400 L of extraction buffer, pH 8.6, to
amine mixture plus 100 L each of the working each tube.
5. Mix contents on a rotary or orbital shaker for 20
standards of DOPA, DOPAC, and DHPG, 200 L of
to 30 min.
internal standard, and 300 L of elution solution. Total
6. Centrifuge in a microcentrifuge for 1 minute.
volume of combined standard solution is 900 L. When 7. Aspirate supernatant, but leave a few mL of
90 l of the extract of the test standard is analyzed by the solution above the pellet.
HPLC system, this is equivalent to injecting 250 pg of 8. Add 1.0 mL of wash buffer to the alumina
NE, EPI, and DA and 1000 pg of DOPA, DOPAC, and pellet. Wash by inversion with wrist action, centrifuge as
DHPG. in step 6 above, and aspirate as in step 7 above. (Wash 1)
Assay 9. Wash the pellet with wash buffer as in step 8.
Equipment: (Wash 2)
10. Wash the pellet a third time with wash buffer,
Column: Axxion Chromatography ODS 5 m reverse
but at this stage, attempt to remove the total volume of
phase C18 4.6 mm 25 cm (obtained from Thompson
supernatant. (Wash 3)
Inst. Co., Springfield, VA). Column temperature (16C
11. Add elution solution (100 L) to the lumina
to 17C) was controlled using a refrigerated water
pellet. Vortex for 2 minutes.
circulator (Fisher Scientific, Pittsburgh, PA). The
12. Centrifuge for 1 minute as described in step 6
coolant (1:1 mixture of Prestone antifreeze and Type I
above. Transfer the eluent to the centrifuge tube with a
water) circulates through a glass jacket (Column
Engineering, Ontario, CA) surrounding the column. 0.2-m filter insert (Gelmar Instrument Co., Ann Arbor,
Pump: Model 510 dual-head pump (Waters, Milford, MI).
MA). 13. Repeat step 11 using 50 L elution buffer.
Repeat step 12, and combine supernatants.
Pre-Column Filter: with a C18 insert (Waters,
14. Centrifuge combined eluents at 16,000 g for 1
Milford, MA).
minute. Discard filter. The sample is now ready for
Injector: Model 7161 injector (Rheodyne, Cotita, CA).
chromatography. Either begin the analysis or store
Centrifuge: Eppendorf Model 5415C microcentrifuge
eluates at 70C overnight. Attempt to complete the
(Brinkman Instruments, Westbury, NY) with fixed-angle
analysis within a working day.
rotor to accept 1.5-mL centrifuge tubes.
Calibrated Spatula: Volumetric spatula, 10 mg (The 15. Inject 90 L of each sample eluate or combined
Nest Group, Southboro, MA). standard onto the HPLC column for analysis.
Detector: Amperiometric or coulombic electrochemical Calculations
detector. The procedure described here utilized a Calculate using the method of area ratios, that is, the
coulombic electrochemical detector (Model 5100A, peak area of the catecholamine to the peak area of the
ESA, Inc. Chelmsford, MA) with triple electrode system
343
internal standard. A standard curve for each component Catecholamine ratio (Ru) = ___Unknown area___
is obtained from peak-area ratios. Internal standard area
Area ratio (R): Catecholamine (pg/mL plasma) =

Standard ratio (Rs) = Recovery standard area Ru Concentration of standard
Internal standard area Rs
Recovery (%) = RE 100% = % recovery
Extraction ratio (RE) = Extracted standard area Rs
Internal standard area
temperature. When thawed, immediately process each

Notes QC sample as if it were a patient sample. QC samples
are included in each HPLC run.
Limit of Detection and Reportable (Linear) Range:
Catecholamine Blanks are included in the analysis to be certain that the
Limit of Linear Range (pg) HPLC system is operating satisfactorily. If peaks other
than the internal standard are observed, stop analysis and
Detection (pg) check system for contamination. Determine recovery
Dihydroxyphenyl glycol (DHPG) 20 each day of analysis. If recovery is less than established
1002000 value, do not report data; check reagents and system for
Norepinephrine (NE) 20 deterioration.
125500 Notes
Dihydroxyphenylalanine (DOPA) 20 a. If an autosampler is available, it is suggested
1001000 that about 30 samples should be the maximum number
Epinephrine (EPI) 20 of samples per run. The suggested sample sequence is:
50250 First and last injection is the reagent blank. Second and
Dopamine (DA) 25 penultimate injection is the recovery standard. Third and
50500 third-from-last injection, the extraction standard. Fourth
Dihydroxyphenylacetic acid (DOPAC) 50 and fourth-from-last injection the QC sample. Add
502000 plasma sample in order of priority between the fourth
and fourth-from-last QC samples.
Precision: b. If an autosampler is not available, it is
Within-run: A CV of about 5% is achievable. suggested that 8 samples be the maximum number of
Between-run: A CV of about 5% is achievable. patient samples in one run. The suggested sample
sequence is: first, reagent blank; second, recovery
standard; third, extraction standard; and fourth, QC
Quality Control: sample. Add plasma samples after the QC sample in
Plasma: Obtain outdated plasma from the blood bank, order of priority.
and store 1.2-mL aliquots containing 10 mg sodium
metabisulfate at 80C. At the time of analysis, remove
two tubes from storage, and thaw to ambient
344
Catecholamines (Urine)
Brett McWhinney
Name: Epinephrine, E, 4-[1-hydroxy-2-(methylamino)ethyl]-1,2-benzenediol, adrenaline

Norepinephrine, NE, 4-(2-amino-1-hydroxyethyl)-1,2-benzenediol, noradrenaline
Dopamine, DA, 4-(2-aminoethyl)-1,2-benzenediol
Epinephrine Norepinephrine Dopamine

Structural formula: See below
Molecular formula: C9H13NO3 C8H11NO3 C8H11NO2
Molecular mass: 183.20 169.18 153.18
Merck Index: 3567 6536 3428
Chemical class: Amino acid metabolite, dihydroxyphenyl amino derivative
i relatively independent of the diet. Ingestion of bananas

Chemically, catechols are benzene rings with two or synthetic vanilla has been reported to increase the
adjacent hydroxy groups. In catecholamines, the amine excretion of conjugated catecholamines [4]. For these
function is in a side chain, as shown in Figure 1 for reasons, the hydrolysis step is usually omitted, and only
norepinephrine, epinephrine, and dopamine. These are the free catecholamines are measured.
the three catecholamines most commonly measured in
urine. In neutral or alkaline solutions, the dihydroxy All urine catecholamine methods require careful sample
group is easily oxidized, with the loss of two electrons purification because of the many closely related amines
and two protons as shown in Figure 2. and catechols found in urine. Currently, ion-exchange
chromatography, adsorption chromatography, or a
Catecholamines are found in urine both in the free and combination of both of these techniques is most often
conjugated form, primarily as sulfates. The excretion used to isolate the catecholamines. For ion-exchange
rate of conjugated urinary norepinephrine and purification, the urine sample is adjusted to a neutral pH
epinephrine is about three times that of the free and passed through a cation-exchange resin. The amines
catecholamines [1,2]. The conjugates can be hydrolyzed are positively charged and are retained by the resin.
by being boiled in acid before sample extraction and Amino acids such as 3,4-dihydroxyphenylalanine (dopa)
included in the total catecholamines measured. Most are not retained, since at a pH above 3, they are
studies on the clinical significance of catecholamines uncharged zwitterions. The amines are then eluted with
have been done with free catecholamine measurements, strong acid or with a solution of high ionic strength.
since these are more closely related to the tumor load of Additional specificity can be conferred by eluting with
pheochromocytoma than measurements of conjugated boric acid, which forms a complex with cis-diol groups
catecholamines [3]. Since dietary catechols are as shown in Figure 3. The complexed catecholamines in
extensively conjugated before reaching the systemic the eluate are stable for several days [5]. Alumina is used
circulation, free catecholamine measurements are also to separate catechols from noncatechols by adsorption of
the former onto alumina at pH 8.6. The catechols are
i
eluted with acid. A small amount of alumina is mixed
Catecholamines (Urine) with the sample and buffer, the supernatant is removed
Previous and current authors of this method: by aspiration, and the catecholamines are eluted with
First edition: Not done only 200 to 300 L of acid, providing a convenient
Methods edition: Not done preconcentration of the catecholamines before
Second edition: Not done quantitation. Purification on a column of immobilized
Third edition: Not done boric acid offers advantages of rapidity, specificity, and
Fourth edition: Lawrence E. Webb high recovery [6,7], and this technique may become
345
more common in the future now that pre-prepared representation of the reaction sequence is shown in
columns are commercially available. Figure 6. The catecholamines are isolated by a cation-
exchange column or alumina [14,15,16]. The eluate is
Low detection limits are an absolute necessity for mixed with ethylenediamine, the condensation products
catecholamine analyses, since samples may have are extracted with isobutanol, and the fluorescence is
concentrations as low as 1 ng/mL. The sensitivity of measured (Table 1, Method 1b) using 435 nm as the
fluorometric or electrochemical detectors is required. exciting wavelength and 560 nm as the emitting
Ultraviolet detection, though used in early high- radiation.
performance liquid chromatographic (HPLC) methods
[8], is not sufficiently sensitive. HPLC methods can be divided into HPLC-F and HPLC-
EC procedures according to whether fluorometric (F) or
Current methods of analysis can be conveniently electrochemical (EC) detection is used (Table 1,
classified according to their capabilities of separating Methods 2a-2c). The catecholamines are isolated and
norepinephrine, epinephrine, and dopamine from each concentrated by ion exchange preceded or followed by
other in the purified sample. The older methods do not alumina adsorption chromatography. The HPLC
make this separation and measure only the total separation of norepinephrine, epinephrine, and dopamine
catecholamines, or else they differentiate norepinephrine is done by ion-exchange or reversed-phase ion-pair
and epinephrine indirectly. Norepinephrine and chromatography. The microparticulate silica stationary
epinephrine are distinguished either by their differing phase is bonded with sulfonic acid groups for ion
rates of oxidation at pH 3 and pH 6 or by their differing exchange or an alkyl organic group for reversed-phase
fluorescence intensities at two different excitation and separations. In the reversed-phase system, an ion-pair
emission wavelengths. The latter procedure is easier to reagent, such as the sodium salt of heptylsulfonic acid, is
do, but the former provides better discrimination [5]. added to the mobile phase. There are several theories of
HPLC methods, on the other hand, separate the separation mechanism [17,18]. The simplest
norepinephrine, epinephrine, and dopamine approach for interpretation of the reversed-phase
chromatographically and measure each of these chromatography is to liken the column to a dynamic ion
compounds as it appears in the chromatographic eluate. exchanger. The hydrophobic part of the ion-pair reagent
is attracted to the reversed-phase column in a dynamic
The older methods (Table 1, Method 1) depend on the exchange process. The column then presents negatively
conversion of norepinephrine and epinephrine into charged sites to the analyte and acts as a cation-exchange
derivatives called fluorophores which emit a column.
characteristic fluorescence greater than the native
fluorescence of the parent amine. The most commonly Adequate sensitivity can be obtained with HPLC-F
used of these methods is the trihydroxyindole (THI) methods using native fluorescence (200 nm excitation,
method [1,5,9,10], so called because the fluorophore 300 nm emission or 280 nm excitation, 320 nm
contains three hydroxyl groups and an indole ring. The emission) [19] (Table 1, Method 2a). Fluorophores can
reaction sequence and associated chemical structures for be formed by precolumn [20] or postcolumn [21,22]
epinephrine are shown in Figure 4. After isolation, derivatization, though the derivatization step is not
usually by a cation-exchange column, norepinephrine required for urine assay applications. The THI reaction
and epinephrine are oxidized with potassium (Figure 6) has been used in a postcolumn derivatization
ferricyanide to noradrenochrome and adrenochrome, system for the detection of norepinephrine and
respectively. These rearrange in the alkaline medium to epinephrine by monitoring of fluorescence at 485 nm,
form adrenolutins. Ascorbic acid must be added and the with excitation at 395 nm [21]. Derivatization with o-
adrenolutin fluorescence measured quickly (405 nm phthalaldehyde and 2-mercaptoethanol (Figure 7) has
excitation, 495 nm emission), because these derivatives been applied to the analysis of norepinephrine,
are easily oxidized. The method is summarized in Table dopamine, and other biogenic amines (340 nm
1, Method 1a. Dopamine is not included in the reaction. excitation, 480 nm emission) [20], but secondary amines
This catecholamine can be detected separately using such as epinephrine do not react.
periodate as the oxidant [11,12]. More recently,
norepinephrine, epinephrine, and dopamine have been Electrochemical detectors are most commonly operated
measured simultaneously using iodine as an oxidizing in the amperometric mode, which means that the
agent; differential analysis of these three catecholamines, electrode is maintained at a constant operating potential
however, requires measurement of fluorescence at three [17]. This potential forces oxidation of the ortho-
different wavelengths (excitation/emission: 310/365, dihydroxyl group, yielding an o-quinone (Table 1,
386/480, 410/500 nm) after oxidation at two different Method 2c). The electrons from this reaction pass into
pHs [7]. the electrode, and the resulting current is measured as
each oxidizable analyte passes the detector. Lower
In the ethylenediamine (EDA) method (Table 1, Method detection limits can be attained than with HPLC-F
1b), the adrenochrome formed from epinephrine is methods that measure native fluorescence.
condensed with 1 mole of ethylenediamine to form a
fluorophore [13], as shown in Figure 5. The Reference and Preferred Methods
condensation of norepinephrine with ethylenediamine is There is no established reference method at this time.
more complex and less well understood. One Selection of catecholamine methods requires a good
346
understanding of the advantages of the available sample determined by HPLC [24], in agreement with an earlier
purification techniques. Adsorption of catechols on study [28].
alumina must be done at pH 8.6, and catecholamines are
very susceptible to oxidation at this pH. With ion- The method of choice for the routine measurement of
exchange methods, the catecholamines are adsorbed at a catecholamines in urine appears to be one of the HPLC
neutral or acidic pH, at which they are relatively stable. procedures. The THI method is not recommended for
With the weak carboxylic acid ion-exchange resin and differentiation and quantitation of norepinephrine and
boric acid eluent, recoveries are consistently greater than epinephrine, but the THI method can be used to
90%. The same column can be used to separate both determine the sum of these two catecholamines. If
catecholamines and metanephrines. The metanephrines possible, medications should be discontinued during the
are eluted with ammonia after elution of the urine collection period when the THI method is used.
catecholamines.
Both ion-exchange and reversed-phase HPLC columns
Neither the ion-exchange nor the alumina methods are have been used successfully; the reversed-phase columns
completely specific for catecholamines. For the HPLC with ion-pair reagents have the advantage of predictable
methods, which depend on detection of the underivatized and convenient adjustment of retention times by simply
catecholamines, ion exchange is usually used with varying the concentrations of ion-pair reagent and
alumina to obtain the required specificity and sensitivity organic modifier.
[19]. Ion exchange has been used alone [23,24], but the
dilution of the sample by boric acid eluent has made it There is a good correlation between electrochemical and
difficult to obtain good precision for epinephrine fluorescence detection methods [19]; the choice between
concentration within the healthy reference interval. them is a matter of personal preference and availability
Where HPLC equipment is not available, the of the technique in a particular laboratory. If plasma
trihydroxyindole (THI) method is most commonly used catecholamines are also determined, electrochemical
and is the method of choice. The EDA method is used in detection is advantageous because it has sufficient
some of the older plasma catecholamine assays, but it is sensitivity to measure down to the low picogram per mL
generally agreed that it lacks the specificity required for concentration required for this assay. Fluorometric
urine assays [14,16,25,26]. The sensitivity of the assay is detectors are simple to set up and operate, and freedom
difficult to describe because it is dependent on the from baseline noise and small interfering peaks is easily
fluorometer used and on sample elution volumes. With obtained.
filter fluorometers, as little as 0.01 g of norepinephrine
can be detected at a sensitivity adequate for human urine The methods presented here use fluorometric and
[5]. The THI method presented here is commercially electrochemical detection. Both detectors may be used
available in a convenient kit form that provides the simultaneously. The excitation and emission
carboxylic acid resin in the form of prefilled wavelengths for the fluorescence method are optimized
minicolumns. The differentiation of norepinephrine and for a Kratos Model 970 fluorometer with a deuterium
epinephrine is not presented here; this separation and lamp. Fluorometers with xenon lamps may also be used
quantitation should be performed by HPLC methods. with 285 nm excitation [19].
References may be consulted for the details of other
differential quantitation methods [5,9,25]. Specimen
A 24-hour urine specimen is required. A 1983 study [29]
HPLC methods offer many advantages over the has confirmed that catecholamines deteriorate rapidly in
traditional methods, which do not employ this untreated urine at room temperature and that the simplest
technology. Derivatization and the strict time limitations effective treatment is acidification with hydrochloric
of the associated reactions are not required. Sensitivity acid. Refrigeration offers an extra measure of security,
can be extended to levels lower than 1 ng/mL of urine, if though this precaution may be impossible in the hospital
required. Norepinephrine, epinephrine, and dopamine room. Once collected and properly preserved, urine may
are easily separated from each other and from be frozen.
interferences, including drugs such as -methyldopa
(Aldomet), isoproterenol, quinidine, and propranolol It is essential that the urine be kept acidic during the
[17,19,24,27]. entire collection. At least 10 to 15 mL of 6 mol/L HCl
should be added to the collection vessel before the start
The THI and HPLC methods correlated fairly well, and of the collection period, with the amount of acid added
values from normal urine determined by the two adjusted according to the expected urine volume. The
methods were in close agreement [19,24,27]. Correlation addition of 4 mL of 5 mol/L sulfuric acid per 400 mL of
of the THI and HPLC data was not so good for urines urine has been shown to be adequate [15]. Boric acid
from hypertensive patients who had stopped taking should not be used as a preservative. When the
drugs. This finding indicated that minor amounts of collection is completed, the pH should be less than 3.
drugs or drug metabolites may interfere with the THI
analysis [24]. When norepinephrine and epinephrine Guidelines for Collection of Urine HMMA (VMA),
values for normal urines were calculated from the THI HVA, Catecholamines, and Metanephrines
data by solution of a simultaneous equation, the THI Urine catecholamines and metabolites are unstable at a
epinephrine values were twice the epinephrine values pH above 3.0, so it is necessary to add HCl prior to
347
starting the urine collection. However, since excess HCl method, when used with a carboxylic acid ion-exchange
will interfere with the analysis of VMA and HVA (not resin and boric acid eluent, has generally good accuracy,
cats and mets) the urine collection should be commenced though drugs such as -methyldopa (Aldomet),
with a minimum of acid. This is particularly important in isoproterenol, and quinidine interfere [5]. Substances in
the case of pediatric patients being screened for urine may cause a quenching of fluorescence, which also
neuroblastoma, because low 24-hour urine volumes can interferes.
be expected. The following is a list of recommendations
for the amount of acid required per volume of urine: It should be recognized that a particular drug may
interfere in an individual HPLC system [28]. However,
an interfering substance can often be detected simply
Expected Urine Volume Add from the appearance of the HPLC chromatogram [27],
whereas the recognition of interferences in the THI
<100 mL 0.5 mL 6 N HCl methods is much more difficult and time consuming
100250 mL 3 mL 6 N HCl or [30].
1.5 mL conc HCl
250750 mL 6 mL 6 N HCl or Urine Catecholamines Reference Intervals
3 mL conc HCl Reference interval values should be established by each
7501250 mL 10 mL 6 N HCl or individual laboratory. Urinary dopamine excretions of
5 mL conc HCl 100 and 115 g/24 hr have been reported as upper limits
>1250 mL (most adults) 15 mL 6 N HCl or of normal [6,7]. Reference intervals established with the
10 mL conc HCl THI method have been generally found to agree with
those determined more recently by HPLC methods
[19,27]. Moyer et al. [27], in an extensive population
Interferences study using an HPLC method, found no significant age
The presence of catechols in bananas may cause falsely or sex differences in adult normal values, but excretion
increased concentrations in individuals eating bananas rates for children were much lower than in adults. Earlier
immediately prior to or during urine collection. The THI studies [31,32,33] based on the THI method also found
lower ranges for children, as follows:
Total* Urine Catecholamines Reference Intervals
Age (years) Norepinephrine Epinephrine
Newborn 0.573.3 g/day 0-0.77 g/day
1 5.415.9 0.14.3
15 8.130.8 0.89.1
615 19.071.1 1.310.5
>15 34.487.0 3.513.2
*Total urine catecholamines is the sum of free and conjugated catecholamines in urine.
However, excretion rates expressed in terms of excretion per kilogram of body weight were found to be higher than in
adults [32].
The reference intervals for free catecholamines in adults, as measured by HPLC, are as follows:
Free (Unconjugated) Urine Catecholamines Reference Intervals

(g/24 h) (g/24 h) (g/24 h)
1580 020 65400
In a population study using an HPLC method with electrochemical detection, no significant age or sex differences were
evident in adult reference intervals [27]. Excretion rates for children were much lower than in adults.
Interpretation the nerve endings at the synapse, with only small

Circulating catecholamines are derived from either the amounts appearing in peripheral blood. Most (80%) of
adrenal medulla or peripheral nerves [34]. The adrenal the circulating norepinephrine and epinephrine and
medulla is the source of almost all the plasma essentially all the dopamine are excreted in the urine as
epinephrine, along with small amounts of the sulfate (predominantly) or glucuronide conjugates
norepinephrine. Norepinephrine and dopamine are [35].
released from the axonic terminals of adrenergic nerves
as a neurotransmitter. Catecholamines synthesized in the The most frequent use for urine catecholamine analysis
brain do not cross the blood-brain barrier. is the diagnosis of pheochromocytomas.
Norepinephrine is rapidly retaken up or metabolized by Pheochromocytomas are tumors derived from
348
chromaffin cells distributed throughout the body. Most measurement of plasma free metanephrines would be
of these cells and thus most (90%) of the recommended [39].
pheochromocytomas are located in the adrenal medulla.
Most of the extra-adrenal tumors are located in the Urine Catecholamine Performance Goals
abdomen. Only 10% of pheochromocytomas are Studies suggest that the precision of the THI method
bilateral, except in cases of familial (coefficient of variation [CV] 10% to 15%) is less than
pheochromocytomas, when the rate exceeds 50% [36]. that of the HPLC method [19,24,27]. Long-term
precisions (%CV) for norepinephrine and epinephrine
Although pheochromocytomas are relatively uncommon measured in urine by HPLC were found to be 3.3% at a
(only 1 in 200 hypersensitive patients), it is important to concentration of 48 g/L and 5.3% at a concentration of
detect these tumors. Removal of a pheochromocytoma 13 g/L, respectively. This better precision for the
will usually cure uncontrollable and sometimes fatal HPLC method can be attributed to at least two factors:
hypertension. In addition, some pheochromocytomas are (1) the better sensitivity of the HPLC method and (2) the
malignant (less than 5%), but they metastasize slowly. addition of an internal standard to each urine processed
The diagnosis of a pheochromocytoma can alert the by the HPLC method. The internal standard is used to
physician to other associated endocrine disturbances that correct for variation in catecholamine recovery.
are part of a familial multiple endocrine neoplasia
syndrome (MEN II) [36]. References
1 Crout JR. Catecholamines in urine. In: Seligson
A urinary catecholamine analysis is usually requested for D, ed. Standard Methods of Clinical Chemistry.
patients whose hypertension is poorly controlled by the Vol 3. New York: Academic Press; 1961:62-80.
usual drug and diet regimens. The existence of a 2 Khane Z, Esser AH, Kline NS, Vestergaard P.
pheochromocytoma as an organic cause of the Estimation of conjugated epinephrine and
hypertension must be ruled out for the reasons listed norepinephrine in urine. J Lab Clin Med
above. The hypertension in these patients is usually 1967;69:1042-1050.
continuous but may be intermittent. Widely fluctuating 3 Causon RC, Brown MJ. Catecholamine
or paroxysmal blood pressure changes are noted in three- measurements in phaeochromocytoma: a
quarters of the patients known to have review. Ann Clin Biochem 1982;19:396-404.
pheochromocytomas [36]. This finding is considered to 4 Crout JC, Sjoerdsma A. The clinical and
reflect the intermittent release of catecholamines from laboratory significance of serotonin and
the tumors themselves. Frequently associated clinical catecholamines in bananas. N Engl J Med
symptoms, including headache, excessive perspiration, 1959;261:23-26.
and palpitations, also tend to occur as paroxysmal 5 Sandhu RS, Freed RM. Catecholamines and
events, lasting from minutes to days. The presence of all associated metabolites in human urine. In:
these symptoms has a clinical diagnostic sensitivity for Seligson D, ed. Standard Methods of Clinical
pheochromocytoma of 90.3% and a specificity of 93.8%. Chemistry. Vol 7. New York: Academic Press;
The absence of all four signs has reported a negative 1972:231-246.
predictive value of 99.9% [37]. 6 Speek AJ, Odink J, Schrijver J, Schreurs WHP.
High-performance liquid chromatography
It is because of the paroxysmal nature of catecholamine determination of urinary free catecholamines
release that a 24-hour urine collection is considered the with electrochemical detection after
preferred specimen, since the catecholamine excretion is prepurification on immobilized boric acid. Clin
averaged over a long period. Shorter timed collections Chim Acta 1963;128:103-113.
(12 hours or random [spot] urines) have been 7 Oka K, Sekiya M, Osada H, Fujita K, Kato T,
suggested, for which the catecholamine excretion rate Nagatsu T. Simultaneous fluorometry of urinary
would be normalized to creatinine excretion (milligrams dopamine, norepinephrine, and epinephrine
of catecholamine per gram of creatinine). However, this compared with liquid chromatography with
technique is not widely used. Greater than 90% of electrochemical detection. Clin Chem
patients with pheochromocytoma will have a urine 1982;28:646-649.
excretion of catecholamines more than twice the upper 8 Mell LD, Gustafson AB. Urinary free
limit of normal [38]. Most others will require norepinephrine and dopamine determined by
consecutive urine analyses to demonstrate a pattern of reverse-phase high-pressure liquid
elevated catecholamine excretion. The measurement of chromatography. Clin Chem 1977;23:473-476.
at least one other indicator of catecholamine synthesis 9 Catecholamines by Column Test: Instruction
and excretion is advocated. The most frequently Manual. Bio-Rad Laboratories, Richmond, CA
measured analyte is HMMA or vanillylmandelic acid 94804.
(VMA). The repeated elevation of several parameters of 10 Sobel C, Henry R. Determination of
catecholamine excretion can be used as biochemical catecholamines (adrenalin and noradrenalin) in
evidence for the presence of a pheochromocytoma. urine and tissue. Am J Clin Pathol 1957;27:240-
Rarely, the measurement of the urinary catecholamines 245.
and their metabolites will not establish a clear diagnosis 11 Anton AH, Sayre DF. The distribution of
of pheochromocytoma, although the clinical indications dopamine and DOPA in various animals and a
are very suggestive of the disease. In these cases, the method for their determination in diverse
349
biological material. J Pharmacol Exp Ther 25 Anton AH, Sayre DF. A study of the factors
1964;145:326-366. affecting the aluminum oxide-trihydroxyindole
12 Routh JI, Bannow RE, Fincham RW, Stoll JL. procedure for the analysis of catecholamines. J
Excretion of L-dopa and its metabolites in urine Pharmacol Exp Ther 1962;138:360-378.
of Parkinsons disease patients receiving L- 26 Viktora JK, Baukal A, Wolff FW. New
dopa therapy. Clin Chem 1971;17:867-871. automated fluorometric method for estimation
13 Natelson S, Lugovoy J, Pincus J. A new of small amounts of adrenalin and noradrenalin.
fluorometric method for the determination of Anal Biochem 1968;23:513-528.
epinephrine. Arch Biochem 1949;23:157-158. 27 Moyer TP, Jiang NS, Tyce GM, Sheps SG.
14 Natelson S. Techniques of Clinical Chemistry. Analysis for urinary catecholamines by liquid
3rd ed. Springfield, IL: Charles C Thomas; chromatography with amperometric detection:
1971:237-244. methodology and clinical interpretation of
15 Weil-Malherbe H. The chemical estimation of results. Clin Chem 1979;25:256-263.
catecholamines and their metabolites in body 28 Jones TA, Walwick ER. Interference of Tylenol
fluids and tissue extracts. Methods Biochem with liquid chromatography of urinary
Anal 1971;Suppl:119-152. catecholamines. Clin Chem 1981;27:1951.
16 Manger WM, Steinsland OS, Nahas GG, 29 Giles HG, Meggiorini S. Stability of
Wakim KG, Dufton S. Comparison of improved catecholamines in urine. Clin Chem
fluorometric methods used to quantitate plasma 1983;29:595.
catecholamines. Clin Chem 1969;15:1101- 30 Jacobs SL, Sobel C, Henry RJ. Specificity of
1123. the trihydroxyindole method for determination
17 Urinary Catecholamines by LCEC. LCEC of urinary catecholamines. J Clin Endocrinol
Application Note No. 15. Bioanalytical Systems Metab1961;21:305-314.
Inc., Purdue Research Park, West Lafayette, IN 31 Grenburg RE, Gardner LE. The excretion of
47906. free catecholamines by newborn infants. J Clin
18 Krstulovic AM, Dziedzic SW, Bertani-Dziedzic Endocrinol Metab 1960;20:1207-1214.
L, DiRico DE. Plasma catecholamines in 32 Voorhees ML. Urinary catecholamine excretion
hypertension and pheochromocytoma by healthy children. 1. Daily excretion of
determined using ion-pair reversed-phase dopamine, norepinephrine, epinephrine, and 3-
chromatography with amperometric detection: methyl-4-hydroxymandelic acid. Pediatrics
investigation of the separation mechanism and 1967;39:252-257.
clinical methodology. J Chromatogr 33 Meites S. Pediatric Clinical Chemistry.
1981;217:523-537. Washington, DC: American Association for
19 Anderson GM, Young JG, Jatlow PE, Cohen Clinical Chemistry Press; 1977:66-67.
DJ. Urinary free catecholamines determined by 34 Krakoff LR. Pheochromocytoma and clinical
liquid chromatography-fluorometry. Clin Chem hypertension: pathogenesis and diagnosis.
1981;27:2060-2063. Cardiovasc Res Rep 1982;3:199-206.
20 Davis TP, Gehrke CW, Gehrke CW Jr, 35 Kuchel O, Buu NT, Hamet P, Larochelle P,
Cunningham TD, Kuo KC, Gerhardt KO et al. Bourque M, Genest J. Essential hypertension
High-performance liquid chromatographic with low conjugated catecholamines imitates
analysis of biogenic amines in biological pheochromocytoma. Hypertension 1981;3:347-
materials as o-phthalaldehyde derivatives. J 355.
Chromatogr 1979;162:293-310. 36 Cryer PE. Diseases of the adrenal medullae and
21 Yamatodani A, Wada H. Automated analysis sympathetic nervous system. In: Felig P, Baxter
for plasma epinephrine and norepinephrine by JD, Broadus AE, Frohman LA, eds.
liquid chromatography, including a sample Endocrinology and Metabolism. New York:
cleanup procedure. Clin Chem 1981;27:1983- McGraw-Hill; 1981:529-545.
1987. 37 Gambino SR. Laboratory diagnosis of
22 Yui Y, Fujita T, Yamamoto T, Itokawa Y, pheochromocytoma. Lab Rep Physicians
Kawai C. Liquid-chromatographic 1982;4:75-79.
determination of norepinephrine and 38 Sjoerdsma A, Engelman K, Waldman TA et al.
epinephrine in human plasma. Clin Chem Pheochromocytoma: current concepts of
1980;26:194-196. diagnosis and treatment. Ann Intern Med
23 Jackman GP. Differential assay for urinary 1966;65:1302-1326.
catecholamines by use of liquid 39 Eisenhofer G. Biochemical diagnosis of
chromatography with fluorescence detection. pheochromocytoma: is it time to switch to
Clin Chem 1981;27:1202-1204. plasma-free metanephrines? J Clin Endocrinol
24 Schleicher ED, Kees FK, Wieland OH. Metab 2003;88:550-552.
Analysis of total urinary catecholamines by
liquid chromatography: methodology, routine
experience and clinical interpretations of
results. Clin Chim Acta 1983;129:295-302.
350
Tables
Table 1: Methods of Urine Catecholamine Analysis
Method 1: Fluorescence derivatives; total (NE, E) or NE, E individually without prior chromatographic separation
a. Trihydroxyindole (THI)
b. Ethylenediamine (EDA)
Principle of analysis: Formation of derivatives, which are quantitated fluorometrically. Differential analysis of
NE and E by oxidation at pH 3 and pH 6 or by fluorescence at two different excitation and emission wavelengths.
a. Isolation by weak cation-exchange resin, followed by oxidation to fluorescent adrenolutins:
NE, E _ferricyanideadrenochromes _ascorbic acidadrenolutins
alkaline pH
pH 6.5
405 nm excitation, 495 nm emission (436 nm excitation and 520 nm emission for differential assay)
b. Isolation by adsorption on alumina or by cation-exchange chromatography. Oxidation to
adrenochromes followed by condensation with EDA to form fluorophores, which are extracted into isobutanol.
NE, E ___ adrenochromes ____EDA fluorophores

pH 6 condensation
Comments: Does not include dopamine

a. Common when HPLC not available; generally good accuracy but some drugs interfere
b. Historical; limited specificity
Method 2: High-performance liquid chromatography (HPLC); HPLC separation and quantitation of NE, E, and DA
a. Native fluorescence (HPLC-F)
b. HPLC with fluorescent derivatives
c. Electrochemical (HPLC-EC)
Principle of analysis: Isolation by cation-exchange chromatography and alumina adsorption; HPLC separation
by cation-exchange or ion-pair chromatography
a. Monitor eluate at 200 nm excitation, 300 nm emission or 280 nm excitation, 320 nm emission
b. Precolumn or postcolumn derivatization, such as THI formation (method 1a)
c. With amperometric detector, monitor eluate at 0.65 volts and detect current from the reaction (Figure
2)
Comments: Good sample cleanup required
a. Common where HPLC available; good sensitivity; excellent specificity
b. Rare; excellent sensitivity; not required for urines
c. Common where HPLC available; excellent sensitivity; good specificity
DA, Dopamine; E, epinephrine; NE, norepinephrine.

351
Table 2: Reaction Conditions for Catecholamine Analysis THI Method
Temperature 20C to 25C

pH 6.5 0.02
Final concentration of reagent ZnSO4: 0.29 mmol/L
components K3FE(CN)6: 0.26 mmol/L
Sodium ascorbate: 3.6 mmol/L
NaOH: 0.69 mol/L
Phosphate buffer: 0.17 mol/L
Volume of sample 5 mL
Sample Urine
Linearity (depending on At least 0 to 300 g/L
fluorometer used)
Time of reaction Add ferricyanide: 0 min
Add alkaline ascorbate: 2 min
Read fluorescence within 20 min
Excitation wavelength 405 nm
Emission wavelength 495 nm
Major interferences Drugs, such as -methyldopa (Aldomet),
isoproterenol, quinidine
Precision (between-day) Normal X = 30 g/L, CV = 15%
Abnormal X = 250 g/L, CV = 8%
CV, Coefficient of variation.
Table 3: Chromatographic Conditions for Urine Catecholamine Analysis by HPLC

Comment: Ambient
Precision (with fluorometric method) (between-day): Norepinephrine
Low normal (X = 15 ng/mL), CV = 8.1%
Abnormal (X = 499 ng/mL), CV = 1.8%
Condition: Flow rate
Comment: 2.5 mL/min
Precision (with fluorometric method) (between-day): Epinephrine
Normal (X = 16 ng/mL), CV = 16.9%
Abnormal (X = 497 nm/mL), CV = 1.9%
Condition: Excitation wavelength
Comment: 200 nm
Precision (with fluorometric method) (between-day): Dopamine
Low normal (X = 36 ng/mL), CV = 11.9%
Abnormal (X = 505 ng/mL), CV = 3.1%
Condition: Emission wavelength
Comment: 300 nm
Condition: Electrode
Comment: Glassy carbon
Condition: Applied potential
Comment: 0.65 volts versus Ag/AgCl electrode
Condition: Sample
Comment: Urine
Comment: 0-80 ng/mL
Condition: Detectability limit
Comment: 1 ng/mL (lower with electrochemical detector)
Condition: Interferences
Comment: No interferences from drugs such as -methyldopa (Aldomet), isoproterenol, or quinidine;
methyltyrosine can interfere with NE, whereas L-dopa medication is broken down into dopamine
CV, Coefficient of variation; NE, norepinephrine and epinephrine.

352
Table 4: Excretion Rates (Range in g/24 hr) of Free Catecholamines for Children 0 to 15 Years of Age
Age (years)
Catecholamine 0-1 1-2 2-4 4-7 7-10 10-15
Norepinephrine 0*-10 1-17 4-29 8-45 13-65 15-80
Epinephrine 0-2.5 0-3.5 0-6.0 0.2-10 0.5-14 0.5-20
Dopamine 0-85 10-140 40-260 65-400 65-400 65-400
*0 implies not detected.
Figures
Figure 1: Three catecholamines

commonly measured in urine.
______________________________________________________________________________
Figure 2: Oxidation of o-dihydroxy groups of the

catechol core, with formation of quinone derivative.
__________________________________________________________________________________
353
Figure 3: Formation of cis-diol complexes

between boric acid and catecholamines.
__________________________________________________________________
Figure 4: Sequence of reactions

converting catecholamines to
trihydroxyindole derivatives, the
adrenolutins.
In this example, epinephrine is
first converted to adrenochrome
and then oxidized to adrenolutin.
354
Figure 5: Sequence of reactions

converting epinephrine to a
fluorescent condensation
product with ethylenediamine.
Figure 6: Sequence of reactions converting norepinephrine to a fluorescent condensation product with

ethylenediamine.
____________________________________________________________________________
355
Figure 7: Reaction of o-phthalaldehyde with norepinephrine and 2-mercaptoethanol to form a fluorescent isoindole
product.
Only primary amines, such as norepinephrine and dopamine, will undergo this reaction. (Modified from Davis TP et al.
High-performance liquid chromatographic analysis of biogenic amines in biological materials as o-phthalaldehyde
derivatives. J Chromatogr 1979;162:293-310.)
Figure 8: Example of standard curve for trihydroxyindole fluorescence assay for catecholamines.
Corrected (relative, Rel.) fluorescence at 495 nm (excitation at 405 nm) versus concentration of catecholamine.
Figure 9: Fluorescence spectra of trihydroxyindole derivatives of catecholamines.

A, Norepinephrine; B, epinephrine; C, a typical urine.
A, Activation spectrum; B, blank; F, fluorescence spectrum; IS, internal standard S, standard; U, urine. (From Jacobs SL et
al. Specificity of the trihydroxyindole method for determination of urinary catecholamines. J Clin Endocrinol Metab
1961;21:305-314.)
356
Figure 10: Excitation spectra for norepinephrine.

A, epinephrine, B, and dopamine, C, in 0.1 M perchloric acid.
Figure 11: Example of chromatographic pattern of

catecholamine standard (40 ng/mL): relative detector
response versus reaction time (in min).
DA, Dopamine; DHBA, dihydroxybenzylamine; E,
epinephrine; ECD, electrochemical response; FLUOR,
fluorescence response; NE, norepinephrine. Vertical
arrows on horizontal axes, Time of injection; numbers on
chromatogram, retention times of corresponding peaks.
357
Figure 12: Example of

chromatographic pattern for a
patient sample. Abbreviations as for
Figure 11.
Figure 13: Example of

standard curve for
fluorescence HPLC assay for
catecholamines.
Peak area ratio
(analyte/DHBA) versus
concentration (20, 40, 80, and
200 ng/mL standards). ( )
and , Norepinephrine; (- - - - -
) and , epinephrine; ()

and , dopamine.
358
Procedure: Trihydroxyindole Method 9. Ascorbic acid. Weigh out 100 mg quantities

and store at room temperature in tightly sealed vials.
Principle Stable 1 month.
The urine sample is adjusted to pH 6.5 and poured 10. Ascorbic acid solution, 20 mg/mL (0.11
through a column filled with a weak cation-exchange mol/L). Prepare immediately before use by dissolving
resin. After the catecholamines are eluted with boric 100 mg of ascorbic acid in 5 mL of water.
acid, norepinephrine and epinephrine are oxidized by 11. Alkaline ascorbic solution (4.1 mol/L NaOH,
ferricyanide in the presence of zinc ions to form 21 mmol/L sodium ascorbate). Prepare immediately
adrenochromes. In alkaline medium, the adrenochromes before use by mixing 18 mL of 5 mol/L NaOH with 4
rearrange to form adrenolutins. Ascorbic acid is added to mL of 20 mg/mL ascorbic acid.
destroy excess oxidizing agent and to prevent further 12. Standard curve. Should be obtained with each
oxidation of the adrenolutins. The fluorescence of the set of determinations.
adrenolutins is measured at 495 nm, with excitation at Normal urine. Use a urine known from a
405 nm. Quenching in samples prepared from urine may previous run to have low blank and test readings with the
reduce a fluorometric reading relative to that obtained in fluorometer used for the assay. Stable 4 months at 4C at
an aqueous standard [1,5]. Standards are therefore pH 3 0.1.
prepared in a urine matrix to make their fluorescence Stock standard (norepinephrine 100 mg/L).
characteristics as close as possible to those of the Dissolve 60.8 mg of norepinephrine hydrochloride in
specimens (Table 2). 500 mL volumetrically with 0.1 mol/L HCl. Stable at
least 6 months at 4C.
Reagents
1 mg/L standard. Dilute 1 mL stock to 100 mL
The procedure is based on Catecholamines by Column
volumetrically with 0.1 mol/L HCl. Prepare on day of
Test Kit (Bio-Rad Laboratories, Richmond, CA 94804),
run.
which includes columns and all reagents. Instructions are
For the curve, run 0, 100, 200, 300 g/L
included here for those who prefer to prepare their own
standards by mixing the 1 mg/L standard and normal
reagents. All water used must be deionized, and reagents
urine as follows:
should be reagent grade.
1. Disodium ethylenediaminetetraacetic acid
Standard mL of 1 mg/ mL of mL of
(Na2EDTA), 1 g/L (2.7 mmol/L). Dissolve 1 g of
(g/L) standard normal urine
Na2EDTA in 1 L of water. Stable 1 year at room 0 0 5
temperature. 100 0.50 4.5
2. Sodium hydroxide (5 mol/L, 20%). Dissolve 200 1.00 4.0
20 g of NaOH pellets in 100 mL of water. Stable 1 year 300 1.50 3.5
at room temperature.
3. Sodium hydroxide (0.5 mol/L). Dilute 5 mL Prepare these standards immediately before the analysis.
of 5 mol/L NaOH to 50 mL with water. Stable 1 year at Linearity of the standard curve will exceed the
room temperature. concentration of the highest standard; the linearity
4. Hydrochloric acid (0.1 mol/L). Dilute 0.85 should be checked with the fluorometer used for the
mL of concentrated HCl to 100 mL with water. Stable 1 assay.
year at room temperature. 13. Controls. Controls are available commercially,
5. Boric acid, 40 g/L (0.65 mol/L). Dissolve 40 g or they may be prepared from urine characteristic of the
of boric acid in 1 L of water. The crystal form dissolves specimens to be analyzed. To 100 mL of urine that has
more readily than the powder form. Stable 1 year at been acidified to a pH less than 3, add 1 mL of 5 g/L
room temperature. sodium metabisulfite and 1 mL of 50 g/L Na2EDTA. If
6. Phosphate buffer (1.0 mol/L, pH 6.5). particulate matter is present, filter under vacuum through
Dissolve 6.38 g of Na2HPO4 and 7.52 g of KH2PO4 in coarse filter paper. Add norepinephrine standard for
approximately 75 mL of water with warming. Cool. controls in the abnormal range. Mix well. Place aliquots
Adjust pH to 6.5 0.02. Make volume up to 100 mL in scintillation vials. Store in freezer (20C). Stable for
with water. Stable 1 month at 4C. Discard if solid about 6 months.
material forms. 14. Columns. These are filled with the weak
7. Zinc sulfate (ZnSO47H2O), 2.5 mg/mL (8.7 carboxylic acid ion-exchange resin Bio-Rex 70 and are
mmol/L). Add 10 mL of water to 25 mg of included in the Bio-Rad kit. Store in refrigerator and use
ZnSO47H2O and mix well. Stable 1 week at 4C. before expiration date. Columns can be prepared directly
from the Bio-Rex 70 resin itself [5,17,19]. However, the
8. Potassium ferricyanide, 2.5 mg/mL (7.6
mesh size of the bulk Bio-Rex 70 is not so closely
mmol/L). Add 10 mL of water to 25 mg of K3Fe(CN)6
controlled as it apparently is for the prepacked columns,
and mix well. Stable 1 week at 4C. If crystals form and good flow characteristics through the columns may
under refrigeration, allow to stand at room temperature be more difficult to obtain than with the prepacked
until crystals dissolve. If the solution turns from pale columns.
yellow to green, it should be discarded.
Assay
359
Equipment: (1) Fluorometer, either a filter type such as 1.0 mL of alkaline ascorbate reagent to the first
Turner Model 111 (G.K. Turner Associates, Inc., Palo tube, mix, and then continue adding 1.0 mL of
Alto, CA 94303) or a diffraction grating instrument; (2) alkaline ascorbate reagent to each of the
stopwatch. remaining tubes in order at 15-sec intervals.
1. Set fluorometer at 405 nm excitation 13. Read all tubes on fluorometer within 20 min.
wavelength and 495 nm emission wavelength
with appropriate settings or filters. Calculations
2. Pipet 5 mL of urine into a 50-mL beaker. Calculate the net fluorescence reading for each unknown
(Excessively turbid urines should be filtered or and standard by subtracting the blank reading from the
centrifuged first.) Prepare each standard in a corresponding test reading. Let
separate 50-mL beaker.
RF100 = F100 F0
3. To each beaker add 15 mL of 1 g/L Na2EDTA
RF200 = F200 F0
and mix well.
4. Adjust the pH of the contents of each of the RF300 = F300 F0
beakers to 6.5 0.2 by adding 0.5 mol/L NaOH
dropwise and monitoring the pH with a pH where RF100, RF200, and RF300 are the relative
meter. fluorescence readings of the 100, 200, and 300 g/L
5. Prepare the resin columns, one for each sample standards; F100, F200, and F300 are the fluorescence
and one for each standard. Shake or rotate readings of these standards; and F0 is the fluorescence
gently to resuspend the resin, and then place
each column upright to allow the resin to settle. reading of the 0 g/L standard.
Remove the caps, snap off the tips, and allow Prepare a standard curve by plotting RF100,
the columns to drain. RF200, and RF300 versus concentration. Use the curve
6. Pour each sample and standard into a separate and the net fluorescence reading of the unknown sample
column. Allow to drain completely, and discard to find its concentration in micrograms per liter of urine.
the eluate. Wash each column with 2 full A typical standard curve is given in Figure 8. Then:
reservoirs (20 mL) of water, and discard the g of catecholamines per 24 h = g/L
eluate. 24-hr volume in liters
7. Place the column in a clean, appropriately
Interferences
labeled 16 150 mm test tube, add 10.0 mL of
Drugs such as -methyldopa (Aldomet), isoproterenol,
40 g/L boric acid to each column, collect each
and quinidine interfere. A high blank reading indicates
eluate in its test tube, and mix.
possible interfering substances.
NOTE: At this point, the samples are stable for
Spectrophotofluorometric examination of activation and
at least 96 hr if refrigerated.
fluorescence spectra may be required to reveal
8. Pipet from each tube two separate 3.5-mL occasional instances of falsely high results caused by
aliquots into two separate 16 100 mm test interferences [24]. The fluorescence spectrum of the
tubes, one appropriately labeled blank and the trihydroxyindoles is given in Figure 9.
other test. To each tube add 1.0 mL of
phosphate buffer and mix.
9. Prepare the alkaline ascorbate solution, and add Catecholamines-Urine Reference Interval
1.0 mL of it to each tube labeled blank. Mix click here
well. Procedure: High-Performance Liquid
10. To each blank tube, add 0.2 mL of the zinc Chromatography
sulfate solution and 0.2 mL of the K3Fe(CN)6
Principle
solution. Mix well.
The urine sample is adjusted to pH 6.5 and poured
NOTE: The reaction must be halted
through a column filled with a weak cation-exchange
immediately by quick and thorough mixing, or the
resin. After the catecholamines are eluted with
blanks will have a high value. The reaction is stopped
ammonium sulfate, they are absorbed on alumina at pH
when the yellow of the K3Fe(CN)6 disappears. The
8.6. The catechols are eluted with acid and separated on
solutions must be clear. The following steps (steps 11 a reversed-phase column, using an ion-pair reagent in the
and 12) must be done in carefully timed sequence mobile phase. The eluted peaks are detected by
without interruption. The additions at 15-sec intervals measurement of fluorescence at 300 nm after excitation
allow 8 samples to be run as a batch in 2 min. at 200 nm or by measurement of the current produced by
11. To each of the tubes labeled test: from step 8, oxidation of the o-dihydroxy group (Table 3). The
add 0.2 mL of the zinc sulfate solution and 0.2 excitation maximum at 280 nm (Figure 10) may also be
mL of the K3Fe(CN)6 solution in sequence at used with emission at 320 nm.
15-sec intervals, mixing well after the addition
of the second solution. Reagents
The reagents are stable for a year unless otherwise noted.
12. Precisely 2 min after the addition of zinc sulfate All water used must be deionized, and reagents should
and K3Fe(CN)6 solutions to the first tube, add be reagent grade.
360
1. Norepinephrine hydrochloride (Sigma then at 200C for 2 hr. Store in a tightly sealed bottle in
Chemical Co. or equivalent). Store it desiccated in a desiccator. Stable at least 1 year.
freezer. 14. Mobile phase. Dissolve 19.05 g of KH2PO4,
2. L-Epinephrine (Sigma or equivalent). Store 800 mg of 1-heptanesulfonic acid sodium salt (HPLC
it desiccated in refrigerator. grade) and 74 mg of Na2EDTA in about 1850 mL of
3. Dopamine hydrochloride (Sigma, or
deionized water. Adjust the pH to 4.0 with 85%
equivalent)
phosphoric acid. Transfer the solution to a 2000-mL
4. 3,4-Dihydroxybenzylamine hydrobromide
volumetric flask, add 80 mL acetonitrile (HPLC grade),
(DHBA)
and dilute to volume with deionized water. Stable 1
5. Phosphoric acid (0.015 mol/L). Dilute 1 mL
week when stored in a refrigerator. Immediately before
of 85% phosphoric acid to 1 L with water.
6. Phosphoric acid (0.3 mol/L). Dilute 10 mL of using, filter through a 0.45 m filter and degas by
85% phosphoric acid to 500 mL with water. stirring under vacuum for 5 to 10 min.
7. Sulfuric acid (0.7 mol/L). Dilute 20 mL of 15. Standards. The working standards are
concentrated sulfuric acid to 500 mL with water. included with each set of determinations as described in
8. Ammonium sulfate (2 mol/L). Place 132 g of the procedure.
ammonium sulfate in a 500-mL volumetric flask, and Stock catecholamine standards (100 mg/L as
dilute to volume with deionized water. Stable when free base)
stored in refrigerator. Norepinephrine. Dissolve 60.8 mg of
9. Sodium metabisulfite (5 g/L, 26.3 mmol/L). norepinephrine hydrochloride in 500 mL volumetrically
Dissolve 1 g of sodium metabisulfite in 200 mL of with 0.015 mol/L phosphoric acid.
water. Prepare fresh daily. Epinephrine. Dissolve 50.0 mg of epinephrine
10. Disodium ethylenediaminetetraacetic acid, in 500 mL volumetrically with 0.015 mol/L phosphoric
Na2EDTA (50 g/L, 149 mmol/L). Dissolve 5 g of acid.
Dopamine. Dissolve 61.9 mg of dopamine
Na2EDTA in 100 mL of water. hydrochloride in 500 mL volumetrically with 0.015
11. Tris-EDTA buffer, pH 8.6 (3 mol/L). Add mol/L phosphoric acid.
about 120 mL of water to 72 g of Trizma base (Sigma These standards are stable for at least 6
Chemical Co. or equivalent) and 10 g of Na2EDTA. Stir months when stored tightly sealed in amber glass bottles
until dissolved. The solution may require warming for in a refrigerator.
complete dissolution. When the solution is at room Combined working standard (1 mg/L
temperature, adjust the pH to 8.6 with concentrated HCl. norepinephrine, epinephrine, and dopamine). Pipet 1.0
Transfer to a 200-mL volumetric flask, and dilute to mL of each stock standard into a 100-mL volumetric
volume with deionized water. Stable for 4 to 6 months flask, and dilute to volume with 0.015 mol/L phosphoric
when stored in refrigerator. Warm to room temperature acid. Prepare fresh daily.
before using. Check pH before use. Norepinephrine working standard (1 mg/L).
12. Phosphate buffer, pH 7 (85.2 mmol/L). Dilute 1.0 mL of norepinephrine stock standard to 100
Dissolve 9.64 g of Na2HPO4, 2.36 g of KH2PO4, and mL with 0.015 mol/L phosphoric acid. Stable as a
chromatographic marker for at least 6 months when
20 g of Na2EDTA in about 950 mL of deionized water.
stored in a refrigerator.
Adjust to pH 7.0 with 2 mol/L NaOH. Transfer to a Stock internal standard (100 mg/L). Dissolve
1000-mL volumetric flask, and dilute to volume with 79.1 mg of DHBAHBr in 500 mL volumetrically with
deionized water. Store in refrigerator. Warm to room 0.015 mol/L phosphoric acid. Stable for at least 6
temperature and check pH before use. Discard after 3 months when stored tightly sealed in a refrigerator.
months, or earlier if bacterial growth appears. Working internal standard (1 mg/L DHBA).
13. Acid-washed alumina. Acid-washed alumina Dilute 1 mL of DHBA stock standard to 100 mL with
is prepared according to the procedure of Anton and 0.015 mol/L phosphoric acid. Prepare fresh daily.
Sayre [25]. Add 100 g of neutral alumina (Sigma, 16. Controls. Controls are available commercially
activity grade 1, or equivalent) to 500 mL of 2 mol/L but, in my experience, have not given satisfactory
HCl in a 1000-mL beaker. Cover with a watch glass, and chromatograms, partly because these controls contain
heat for 45 min at 90C to 100C, with continuous and very little epinephrine. Controls satisfactory for HPLC
rapid stirring. Remove from heat and allow 1.5 min for assay may be prepared from urine collections. To 100
the heavier particles of alumina to settle. Discard the mL of urine which has been acidified to a pH less than 3,
supernatant and the finer particles of alumina. Wash the add 1 mL of 5 g/L sodium metabisulfite and 1 mL of 50
precipitate twice with fresh 250-mL portions of 2 mol/L g/L Na2EDTA. If particulate matter is present, filter
HCl at 70C for 10 min, discarding the supernatant and
under vacuum through coarse filter paper. Add
the finer alumina particles each time. Wash the
catecholamine standards as required for controls in the
precipitate with 500 mL of 2 mol/L HCl at 50C for 10
abnormal range. Mix well. Place aliquots in scintillation
min. Decant the HCl. Wash the precipitate repeatedly
vials. Store in freezer at 20C. Stable for about 6
(about 20 to 25 times) with fresh 200-mL portions of
months.
deionized water, decanting the finer particles each time
17. Columns. Columns are filled with the weak
until the pH of the supernatant is 3 to 4. Transfer the
carboxylic acid ion-exchange resin Bio-Rex 70 and are
alumina to an evaporating dish; heat at 120C for 1 h and
included in Catecholamines by Column Kit (Bio-Rad
361
Laboratories, Richmond, CA 94804) or can be ordered 14. Wash the alumina twice with aliquots of about
separately (Bio-Rad). Columns can be prepared directly 2 mL of water. Agitate the alumina thoroughly
from the Bio-Rex 70 resin itself [2,14,16] (click here). by squirting the water vigorously from a wash
However, the mesh size of the bulk Bio-Rex 70 is not so bottle into the reaction vial. Allow the alumina
closely controlled as it apparently is for the prepacked particles to settle, and remove the water by
columns, and good flow characteristics through the aspiration after each wash. Remove as much
columns may be more difficult to obtain than with the water as possible after the second wash without
prepacked columns. removing any alumina.
15. Add 300 L of 0.3 mol/L phosphoric acid to the
Assay alumina to elute the catecholamines. Vortex
Equipment: (1) HPLC equipped with either a mix gently, and let stand for 5 min. Transfer the
fluorometric detector (Kratos Analytical Instruments supernatants to polypropylene centrifuge tubes.
Model FS970 or equivalent) or an electrochemical Centrifuge at 500 g in a microcentrifuge for at
detector (Bioanalytical Systems Model LC-4 or least 5 min. Transfer the supernatants to
equivalent) with glassy carbon electrode, (2) integrating separate tubes for injection into the HPLC.
recorder, (3) du Pont Zorbax ODS column, 7 m particle Stable for at least 1 week when stored tightly
size (4.6 mm inner diameter 25 cm) or equivalent, (4) sealed in a refrigerator.
rotary mixer or reciprocal shaker, (5) microcentrifuge
capable of 500 g.
HPLC Operation
Extraction of Standards and Samples 1. Set fluorometer at 200 nm excitation
1. Prepare the catecholamine isolation columns, wavelength, and insert 300 nm emission filter.
one for each sample and standard. Shake the Set range at 0.1 A full scale. For the
columns, or rotate them gently to resuspend the electrochemical detector, set the applied
resin, and then place each column upright to potential to +0.65 to +0.75 volts versus the
allow the resin to settle. Ag/AgCl electrode. The chromatographic
2. Transfer a 5-mL aliquot of each sample to a 30- conditions and detector settings are summarized
mL beaker. in Table 3.
3. For low, middle, and high standards, transfer 2. Pump at least 10 mL of deionized water
100, 200, and 400 L, respectively, of the through the column.
combined working standard to separate 30-mL 3. Equilibrate the column with at least 100 mL of
beakers. These standards correspond to 20, 40, mobile phase. Set flow rate to 2.5 mL/min.
and 80 ng of norepinephrine, epinephrine, and 4. Inject 10 L of the norepinephrine working
dopamine per milliliter. standard, and note the norepinephrine retention
4. Add 200 L of the working DHBA internal time. If necessary, add acetonitrile to the mobile
standard and 15 mL of pH 7 phosphate buffer to phase to decrease retention, and reinject the
each beaker. norepinephrine working standard until the
5. Adjust the pH to 6.50 0.02 using 2 mol/L retention time is about 2.5 min.
NaOH. 5. Calibrate the assay with the 80 ng/mL
6. Pour each sample and standard onto a separate combined standard (see step 3 of Extraction of
column. Allow to drain completely. Discard Standards and Samples), using peak areas and
eluate. the internal standard procedure (see
7. Add 10 mL water. Allow to drain completely. Calculations). For studies of small changes
Discard eluate. from the normal range, calibrate with the 40
ng/mL combined standard.
8. Add 1.5 mL of 0.7 mol/L sulfuric acid. Allow
to drain completely. Discard eluate. 6. Inject from 20 to 50 L of standards and
9. Elute the catecholamines into test tubes with 4 samples. Increase the injection volume for
mL of 2 mol/L ammonium sulfate. samples with low concentrations of
catecholamines.
10. Add 100 L of 5 g/L sodium metabisulfate to
7. At the end of the run, pump at least 20 mL of
each tube and mix.
deionized water through the column and all
11. NOTE: Complete this step for each sample other components of the HPLC system, and
individually, one at a time. Add 0.5 mL of 3 then pump at least 20 mL of acetonitrile
mol/L Tris/EDTA buffer to each tube. through the column. Typical chromatograms of
Immediately adjust the pH to 8.60 0.03 using standards and samples are given in Figures 11
2 mol/L NaOH. Pour into a 5-mL conical and 12.
reaction vial containing approximately 50 mg
of acid-washed alumina. Put the vial on a rotary
mixer or reciprocal shaker.
12. Rotate or shake each vial for at least 15 min. Calculations
13. Allow alumina to settle, and remove Measure peak areas, and calculate concentrations of
supernatant by aspiration. norepinephrine, epinephrine, and dopamine by the
internal standard procedure.
362
__Au__
Cu = _ADHBA,u_ Cs
__As__
ADHBA,s
where Au = peak area of the unknown

ADHBA,u = DHBA peak area of the unknown
As = peak area of the standard
ADHBA,s = DHBA peak area of the standard
Cu = concentration of the unknown (ng/mL)
Cs = concentration of the standard (ng/mL)
A typical standard curve is presented in Figure 13.
Many chromatographic data processors will do these

calculations automatically after the calibration step is
completed. For each catecholamine:
g of catecholamine per 24 hr = g/L 24-hr volume in

liters
Interferences
Drugs such as -methyldopa (Aldomet), isoproterenol,
and their metabolites are eluted after the catecholamines
and do not interfere. L-Dopa is metabolized to dopamine
and will cause an elevation in this analyte. -Methyl-L-
tyrosine (Demser) can be eluted close to norepinephrine
in many chromatographic systems. This drug blocks
tyrosine decarboxylase and is frequently given to control
catecholamine synthesis by persons with suspected
pheochromocytomas.
Notes
1. Many laboratories follow a procedure in which
the order of the sample cleanup procedures is
reversed; that is, the sample is first added to
alumina and then eluted with 50 mM acetic
acid, 5 mM sodium bisulfite. This material is
then added to the Bio-Rex column, where the
catecholamines are eluted with boric acid (40
g/L).
2. Sodium octyl sulfate can also be used as the ion
pair.
3. A useful alternative mobile phase that may
perform better on some manufacturers columns consists
of 28.3 g of monochloroacetic acid, 9.35 g of NaOH, 1.5
g of disodium EDTA, and 210 mg of sodium octyl
sulfate in 2 L of water, pH 3.00 to 3.05. The amounts of
octyl sulfate and organic solvent (such as acetonitrile)
may vary as the column ages so that the proper retention
times and resolution are maintained.
363
Cerebrospinal Fluid (CSF) Protein Quantitation

Danyel H. Tacker and Anthony O. Okorodudu
Name: CSF Proteins

Clinical significance: Central nervous system (CNS) disease, trauma Refer to Chapter 47,
Nervous System, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular formula: N/A See information for individual proteins
Molecular mass: N/A See information for individual proteins
Merck Index: N/A
Principles of Analysis and Current Usage i
Cerebrospinal fluid (CSF) is the fluid in the ventricles of The proteins present in CSF, or which are measured in
the brain, in the space between the arachnoid and the pia CSF, include:
mater, and surrounding the spinal cord. It is an ultrafiltrate Albumin, which passively diffuses through the
of plasma that passes through the blood-brain barrier BBB. Albumin is the predominant protein in the
(BBB) in the choroid plexus [1,2]. Selective transport and plasma, as well as the CSF, and has a primary
diffusion of proteins, electrolytes, and other molecules and function in CSF as a transporter of a variety of
bloodborne factors occurs at the level of this highly compounds [1-4].
selective form of capillary endothelium. Thus CSF has Immunoglobulins, which are usually excluded
protein constituents and an electrophoretic pattern similar from the CSF by the BBB. Immunoglobulins that
to that of plasma/serum but at vastly lower concentrations appear in CSF have two potential sources: (1)
in normal health states. In disease states, the permeability massive failure of the BBB and leakage into the
and function of the BBB can be compromised, increasing CSF from the plasma or (2) intrathecal production
the protein concentrations of CSF relative to in cases of CNS infection [1-3,5-8].
plasma/serum. Transthyretin/prealbumin, which is synthesized
The choroid plexus is also the site of in-situ production and secreted in situ. Transthyretin is a shuttle
and secretion of various proteins, peptides, and hormones protein and transporter of thyroid hormones; it
in healthy and disease states. Thus alterations in the complexes with retinol-binding protein to aid in
protein concentration of CSF relative to plasma/serum, as transport of vitamin A/retinols [1-3,5].
well as the appearance of proteins unique to the CSF fluid, 2-Transferrin, which is a desialated form of
can give an indication of the general health of the central transferrin found only in the CNS and functions
nervous system (CNS) [2]. Finally, the appearance of in iron complexation and maintenance of metal
CSF-specific proteins in non-CSF body fluids can ion redox status. 2-Transferrin can be detected
indicate CNS trauma, which can be helpful in diagnosis. long (months and sometimes years) after CNS
trauma in collected body fluids [1-3,5,9,10].
Like serum, CSF is a complex solution including ions, C-Reactive protein (CRP), which is usually found
nutrients, and proteins. The entry of ions present in CSF in very low concentrations or is altogether absent
(H+, K+, Ca2+, Mg2+, bicarbonate, etc.) is specifically from CSF. CRP is an acute-phase reactant and
regulated by ion channels in the membranes composing the appears to have a role in modulating
BBB. However, nutrients and wastes (i.e., glucose, urea, inflammatory response [1,2,5,11].
creatinine) diffuse freely through the BBB. Bilirubin is
2-Macroglobulin, which is normally excluded
typically not present in CSF in measurable quantities [1,2].
owing to its size. 2-Macroglobulin is a protease
inhibitor and an acute-phase protein. Presence of
i
CSF Proteins this protein in the CSF usually only occurs when
Previous and current authors of this method: the integrity of the BBB has been heavily
First edition: Gayle Birkbeck compromised [1-3,5].
Methods edition: Gayle Jackson -Amyloid and tau ( proteins, which are
Second edition: Not updated normally absent from CSF. -Amyloid is a
Third edition: Not updated cleavage product of amyloid precursor protein
Fourth edition: Gayle Jackson (function is not currently described), which forms
Fifth edition: Danyel H. Tacker, Anthony O. extracellular amyloid plaques when not properly
Okorodudu removed from sites of production. Protein
364
normally associates with tubulin to stabilize Preferred & Other Methods: Specific for each protein;
microtubule structures present in neurons; see below.
hyperphosphorylation of protein causes it to
form massive complexes, which result in Total Protein: [1-3,5,18-21]
pathological plaques that are neurotoxic. Both of Accurate measurement of CSF protein is difficult because
these proteins are associated with CSF proteins are present in small quantities and because
neurodegeneration [2]. qualitative differences exist between the amounts of
2-Microglobulin, which is normally present in immunoglobulin and the amounts of albumin present in
low concentrations and is largely an patient specimens. Methods used for CSF total protein
investigational marker in the CSF. 2- measurements can be sensitive but can give inaccurate
Microglobulin composes part of the major results when the albumin-to-gamma globulin ratio
histocompatibility complex I protein complex, changes.
which is responsible for non-self recognition A. Preferred Method: Trichloroacetic Acid (TCA) +
(primarily of viruses and tumor antigens). 2- Nephelometric or Turbidimetric Analysis. The TCA
Microglobulin is amyloidogenic [1-3,5,12,13]. precipitates the protein, which is then read using
Protein 14-3-3, which is normally absent from nephelometric or turbidimetric techniques. Light
CSF. Protein 14-3-3 is a general name for a scatter or absorbance is proportional to protein
highly conserved class of intracellular signaling concentration. Comments: Simple and fast, but
and regulatory proteins. Presence of 14-3-3 requires the largest sample volume. Alternate
protein in the CSF is observed in spongiform precipitating agents include benzalkonium chloride
encephalopathy (e. g., Creutzfeldt-Jakob disease) and benzethonium chloride.
[14]. B. Preferred Method: Sulfosalicylic Acid (SSA): 3%
Myelin basic protein, which is normally absent Sulfosalicylic and 7% Sodium Sulfate; Total Protein,
from CSF. MBP was originally isolated from Turbidimetric. The assay principle is based on
myelin sheaths. Its presence in the CSF is precipitation of protein, with the resultant turbidity
associated with multiple sclerosis [5,7,15,16]. measured in a spectrophotometer. Comments: CSF
Fibronectin, which is normally present in low (500 L), commonly used; sulfosalicylic acid alone
concentrations. Fibronectin complexes with results in greater turbidity with albumin than with
integrins to stabilize the extracellular matrix of all globulin. The same phenomenon is seen with
tissues. Presence of increased fibronectin in CSF sulfosalicylic acid-sodium sulfate in combination.
is associated with trauma and disruption of the C. Historical Method: Biuret Method/Colorimetric
extracellular matrix of the CNS [1,2,5,17]. Analysis. Copper ions in the Biuret solution interact
with amide groups (e.g., lysine) on amino acid
CSF has a number of functions in the CNS [1,2]. Firstly, chains. Albumin has the best binding/detection.
CSF surrounds the brain in the subarachnoid space, fills Comments: Rare; low color yield restricts usage to
the ventricles, and surrounds the spinal cord, providing a very sensitive instruments with rate-measurement
fluid layer of insulation against mechanical trauma to the capability.
fragile tissues of the CNS. Since it completely surrounds D. Historical Preferred Method: Lowry Method. Two-
these tissues, CSF also gives tissues ready access to step assay that incorporates the Biuret reaction with a
nutrients, hormones, and other factors within the CNS second, Folin-Ciocalteu reagent step (whereby
compartment. Further, the ions present in CSF allow for molybdate and tungstate bind to tryptophan and
maintenance of the ambient ionic charge needed for tyrosine residues on the protein). This dual reaction
efficient CNS function. Finally, because there is no lymph gives greater sensitivity but preferentially binds to
drainage in the CNS, the CSF provides an important route albumin. Comments: CSF (200 L), rarely used; time
for the excretion of CNS waste. consuming; influenced by endogenous phenols and
drugs.
The constant production of CSF in the CNS also aids in the E. Other Method: Coomassie Brilliant Blue Dye
maintenance of intracranial pressure and homeostasis and Binding; Total Protein, Chemical,
delays alterations of these when acute changes in Spectrophotometric. In the assay, Coomassie Brilliant
peripheral pressure and homeostasis occur. In short, CSF Blue G-250 binds to protein, with a resultant color
provides protection against abrupt pathological change change from brownish orange to an intense blue
because of its isolated location and highly regulated color, and the absorbance of the dye shifts from 465
composition. to 595 nm. Comments: CSF (25 to 100 L),
infrequently used; problem with standardization;
Reference and Preferred Methods different proteins show variability in sensitivity and
Reference Method: N/A Method used is dependent on variation in standard curves.
the protein and type of analysis.
365
Specific Proteins: [1-8,11-19,22-37] mixture for a second incubation. Excess sample

A. Immunonephelometry light scatter is proportional is washed away or removed in a flow cell, and a
to protein-antibody complex concentration. substrate for the chemiluminescent compound is
Typically, the target protein is selectively bound by a applied. After electrical stimulation of the
specific antibody, and the resulting antigen-antibody sample, the chemiluminescent compound is
complex increases the light scatter of the sample. The excited and gives off photons of light. These
scattered light is detected at an angle >180 degrees photons are counted, and the amount of signal is
from the incident light beam passing through the proportional to target protein concentration.
sample.
C. High-performance liquid chromatography
B. Immunoassay (IA) protein-specific antibodies bind (HPLC) with coulometric or other detection
target proteins in the CSF sample. Depending on the method is typically used for peptides and
type of IA, tagging of the detection antibody, and hormones present in very low concentrations.
other analytical specifications, the signal-to- With HPLC, processed samples are injected into a
concentration relationships vary. Examples of IA column under high pressure and carried by a
used in the above-referenced publications include: liquid phase. Depending on the column, the liquid
1. IA after high-performance affinity phase, and the protein chemistry, fractions of
chromatography (for ultra-low, specific IgG proteins in the sample will elute at different times.
detections) These fractions can be subsequently analyzed in a
2. Enzyme IA an example is fibronectin, but any variety of ways. Use of internal standards and in-
enzyme-linked specific antibody used for run calibrators can allow for quantitation.
detection falls in this category. Enzyme-linked
detection antibodies react with a substrate added Electrophoretic Patterns (Semiquantitative): [3,7-
after the detection antibody has complexed with 11,25,36,38-40]
the target protein and excess has been washed A concentration step is necessary for this procedure.
away. The reaction with the substrate produces a Polyacrylamide and other gel techniques are used for
signal (typically colorimetric) that is electrophoretic separation of protein bands in serum and
proportional to the target protein concentration CSF. In short, CSF is applied to an agarose gel and
and rate-limited by the presence of the enzyme separated in an electric field through a liquid buffer phase.
after the washing steps. Dye-binding techniques (utilizing bromocresol purple,
3. Radio IA an example is 2 microglobulin, but Coomassie Brilliant Blue, and other dyes) are used for
new and emerging assays often start with RIA visualization and densitometric quantitation. Densities of
and then progress to other IA techniques. RIA is bands are used to express protein fraction concentration
typically competitive in nature, whereby a based on their percentages of total protein, usually
radioactive, tagged target protein is added to the measured via nephelometric or turbidimetric detection
sample. The tagged protein and the native (above).
protein compete for antibody binding (typically
to a solid substrate). After thorough washing, the The most common application of CSF electrophoresis is
radioactivity on the substrate is measured, for the visualization of oligoclonal bands in the gamma
usually via scintillation counting. Owing to the region of the gel. Banding in the gamma region suggests
nature of competitive IA, concentration of the that in-situ expression of monoclonal antibodies is taking
target protein in the sample is typically inversely place and occurs in a variety of CNS diseases, including
proportional to the signal/scintillation detected. multiple sclerosis and malignancy.
4. Electrochemiluminescence (ECLIA) and related
immunoassay technologies are used for Specialized/Emerging/Research Techniques: [15,34,38]
measuring specific proteins. A concentration step A. Enzyme-linked immunosorbent assay (ELISA) and
is usually necessary for CSF protein quantitation. related methods are used for specialized testing of
Depending on the size of the target, the detection specific and uncommon proteins. ELISA is a
can either be proportional/indirect, or inversely common first step in research analysis and often
proportional/competitive. For most proteins, the expands to automated IA. IAs were discussed
target is large enough that the assay is indirect. previously.
Using this approach, a trapping antibody is
bound to a substrate (often a flat detection B. Specialized detections using Western blot are usually
surface or a paramagnetic bead), and the target limited to research but are possible as long as specific
protein in the sample is first incubated with the antibodies are available. In Western blotting, proteins
trapping antibody. A second antibody (the are extracted from the sample and concentrated. The
detection antibody, which is tagged with a concentrated protein extract is electrophoretically
chemiluminescent compound) is added to the separated using an agarose gel preparation then
366
incubated with specific antibodies (which are

typically linked to enzymes). Colorimetric detection Depending on the assay used for CSF analysis, a number
is then possible, as is semiquantitation using of factors could cause interference. For example,
densitometric analysis of protein bands. colorimetric techniques employing spectrophotometric
detection can be susceptible to interference from
Specimen hemolysis, icterus, uremia, drugs with spectral activity,
Preparation: [2] and high ammonium concentrations (for Biuret and Lowry
There are many side effects associated with the collection reactions). Also, other conditions or therapies (i.e.,
of CSF (typically via lumbar puncture, also called a spinal dextran, contrast media, and high concentrations of
tap) that are undesirable and even life-threatening. The immunoglobulins) that increase turbidity of the CSF can
risks associated with CSF collection must be weighed affect nephelometric and turbidimetric analysis.
against the need for diagnostic information, and volume of
collection must be kept to an absolute minimum for the CSF Proteins Reference Intervals
testing required. Patients should remain calm. Opening Reference Intervals: Please consult Table 1: Available
pressure should be between 90 and 180 mm of water (for Reference Intervals for CSF Proteins
normal adults) but is dependent on postural changes and
elevations, as well as patient physiological status and age Calculations: [1-3,20]
[2]. An opening pressure greater than 200 mm of water in
a relaxed patient suggests serious disease and limits CSF A. Albumin Index
collection volume to 2.0 mL.
Albumin Index = CSF albumin (mg/dL)__
Draw/Collection: [2,5] Serum albumin (mg/dL)
When opening pressure allows for full collection, three
tubes are typically drawn. The first tube is used for Interpretation: For adults, an albumin index < 9 is
chemistry and immunology (unless traumatic; bloody normal. An index of 9 to 14 reflects slight
specimens will affect protein measurements and impairment of the BBB. An index of 14 to 30
determinations), the second tube is for culturing, and the reflects moderate impairment of the BBB, and an
third tube is for cell counting/differentials. Tubes may be index > 30 reflects severe impairment of the
substituted or added, but volumes must be kept to a BBB.
minimum. Maximal volume drawn is 20 mL in an adult
(total CSF volume typically 90 to 150 mL), and about 3 B. IgG Ratio
mL in a neonate (total CSF volume typically 10 to 60 mL).
Collection of blood samples for comparisons and index IgG Ratio = CSF IgG (mg/dL)__
calculations is optimal; when collected 2 hours before the Serum IgG (mg/dL)
CSF collection, the best comparisons are obtained
(because of the delayed equilibrium of the CSF). Interpretation: For adults, an IgG ratio of 3.0 to
8.7 is normal. IgG ratios > 8.7 reflect BBB
Storage: [2] impairment.
CSF degradation begins within 1 hour, so all efforts should
be made to promptly store the sample tubes properly. All C. CSF IgG Index
samples should be centrifuged upon receipt. If analysis
cannot take place immediately, store CSF at 4C for < 72 CSF IgG Index =
hours. Freezing CSF at 20C keeps constituents stable for
up to 6 months. Freezing CSF at 70C keeps constituents [CSF IgG (mg/dL) / serum IgG (g/dL)]
stable indefinitely. Blood samples should be handled and [CSF Alb (mg/dL) / serum albumin (g/dL)]
stored normally.
Interpretation: For adults, a CSF IgG index < 0.8
Patient Risks Involved: [2] is normal. A CSF IgG index > 0.8 suggests
The most common untoward effect of lumbar puncture is intrathecal IgG production.
headache. Infections, meningitis, and death have occurred NOTE: With additional time-sensitive information
as a result of lumbar puncture procedures. about the samples used, IgG synthesis rates may
be calculated.
Interferences [1,2]
The location and specific functions of CSF make it Interpretation and Clinical Utility [1,2,14,22,38]
particularly susceptible to diurnal variation and hormonal Interpretation of CSF Electrophoresis:
factors, which can affect specific protein levels. Such Normally, CSF samples show an evident
influences should be considered accordingly when transthyretin/prealbumin band (typically 2% to 7% of total
analyzing CSF. sample protein content), followed by bands for albumin
367
(56% to 76%), 1 (2% to 7%), 2 (4% to 12%), and 1/2 Neoplastic Disease:
(8% to 18%) globulins. The region (3% to 12% of total Different neoplasms produce different alterations of CSF
sample protein content) should be faintly staining and protein content, the description of which is beyond the
diffuse. scope of this summary. Markers used in diagnosing
neoplasm in CSF fluid include 2-microglobulin,
If the region shows multiple distinct bands, this is called fibronectin, myelin basic protein, and others.
an oligoclonal banding pattern. Oligoclonal bands are
typically associated with demyelinating disease. However, Other CNS Disease:
if the region shows only a single band, this pattern is Terminal spongiform encephalopathies are typically
typically associated with infectious disease, such as associated with elevations in Protein 14-3-3. Imaging and
meningitis. case-relevant symptomatic information are usually used
for diagnosis.
When CSF electrophoresis is performed alongside a serum
sample from the same patient, collected at generally the Alzheimers disease diagnosis is made with imaging
same time as the CSF sample, then an assessment of BBB results, case-relevant symptomatic information, and
integrity can be made. Typically, the CSF lane on the prediction of early risk of disease by determining the ratio
electrophoresis will be faint when compared to the serum of protein to 2-amyloid protein in CSF.
lane. If the CSF lane appears to be equal in density or
darker than the serum lane, then the BBB is probably CSF Proteins Assay Performance Goals
compromised. Calculation of the CSF IgG index and/or the Total protein analysis of CSF employs techniques also
albumin index should always accompany CSF used for serum and plasma protein quantitation. This is a
electrophoretic analysis; the calculations give a more exact benefit to CSF protein quantitation in that the available
assessment of BBB integrity than the density of sample assays are largely automated and come with reliable
staining. control materials. Typically, automated versions of the
SSA and TCA techniques show coefficients of variation
Disease-Specific Interpretation: (CVs) 5% and wide measuring ranges.
Infection:
In CNS infection, the BBB is compromised, and increased Immunoassay performance will vary according to the
CSF protein levels relative to plasma protein/serum levels detection technique measured. For automated IA assays,
are observed. Other indicators, such as elevation of CRP CVs tend to be slightly wider than the serum/plasma
and immunoglobulins, aid in diagnosis. Electrophoresis of version of the assay. Depending on the availability of a
CSF can reveal oligoclonal banding patterns of reference material for a given protein, degrees of assay
intrathecally produced immunoglobulins. standardization, and thus the resulting assay performance,
will vary.
CNS Trauma:
Subarachnoid hemorrhage and other CNS trauma increases Electrophoretic assay performance is largely based on
BBB permeability. Total protein will be elevated in the satisfactory visualization of all relevant bands when
CSF specimen relative to plasma protein concentrations, compared to control samples and serum samples
and typically absent components such as bilirubin and accompanying the CSF sample on the gel. When analysis
methemoglobin are detectable. is semiquantitative and tied to band density as a percent of
total sample protein concentration, then cumulative
In cases of rhinorrhea or otorrhea, the presence of 2- statistics can be generated and performance assessed. CVs
transferrin is suggestive of CSF leakage outside the CNS. 10% for major bands and 20% for minor bands are
Transthyretin detection is only useful in this type of generally acceptable.
analysis if present in very high concentrations, since
transthyretin is also seen in plasma. HPLC technique performance, when made
semiquantitative through the use of internal and assay
Demyelinating Disease: controls, varies according to the targets being detected and
The intrathecal production of immunoglobulins in the the degree of automation used in sample preparation and
choroid plexus occurs in demyelinating diseases such as assay.
multiple sclerosis and produces oligoclonal banding
patterns on electrophoresis. The ratio of IgG in the CSF References
relative to that of the plasma will be elevated, and specific 1 Burtis CA, Ashwood ER, Bruns DE, eds. Tietz
studies for myelin basic protein in the CSF are typically Textbook of Clinical Chemistry and Molecular
positive. Diagnostics. 4th ed. Philadelphia: Saunders;
2006:577-595.
2 McPherson R, Pincus M, eds. Henrys Clinical
Diagnosis and Management by Laboratory
368
Methods. 21st ed. Philadelphia/China: Saunders; globulin) in human cerebrospinal fluid and
2007:426-434. seminal plasma. Clin Chim Acta 1981;113:207-
3 Wise BL. The quantitation and fractionation of 12.
proteins in cerebrospinal fluid. Am J Med 18 Deaton CD, Maxwell KW, Smith RS, Creveling
Technol 1982;48:821-7. RL. Use of laser nephelometry in the
4 Bonate PL. Quantification of albumin in measurement of serum proteins. Clin Chem
cerebrospinal fluid. Anal Biochem 1988;175:300- 1976;22:1465-71.
4. 19 Liu TZ, Sun CF. Laser nephelometric quantitation
5 Wu AHB, ed. Tietz Clinical Guide to Laboratory of protein in urine and cerebrospinal fluid. J
Tests. 4th ed. Philadelphia: Saunders; 2006. Formos Med Assoc 1990;89:1067-71.
6 Arseneault JJ. An enzyme immunoassay for the 20 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ.
quantitation of IgG in serum and cerebrospinal Protein measurement with the Folin phenol
fluid. Clin Chim Acta 1980;107:73-84. reagent. J Biol Chem 1951;193:265-75.
7 Gerson B, Cohen SR, Gerson IM, Guest GH. 21 Mauck JC. The quantitation of urinary and spinal
Myelin basic protein, oligoclonal bands, and IgG fluid proteins by a modified biuret method. Clin
in cerebrospinal fluid as indicators of multiple Biochem 1976;9:188-91.
sclerosis. Clin Chem 1981;27:1974-7. 22 Aksamit AJ. Cerebrospinal fluid 14-3-3 protein:
8 Matikainen MT. Solid-phase immunoassay variability of sporadic Creutzfeldt-Jakob disease,
methods for quantitation of IgG and viral laboratory standards, and quantitation. Arch
antibodies in cerebrospinal fluid and its Neurol 2003;60:803-4.
electrophoretic fractions. J Neurosci Methods 23 Coore J, Ambler J. Potential problem with
1981;4:277-86. standardization of prealbumin (transthyretin)
9 Rouah E, Rogers BB, Buffone GJ. Transferrin results in immunoturbidimetry. Clin Chem
analysis by immunofixation as an aid in the 1991;37:588-9.
diagnosis of cerebrospinal fluid otorrhea. Arch 24 Croci D, Nespolo A, Bosoni MA, Tarenghi G. A
Pathol Lab Med 1987;111:756-7. simple immunoturbidimetric method for IgG and
10 Zaret DL, Morrison N, Gulbranson R, Keren DF. albumin quantitation in cerebrospinal fluid and
Immunofixation to quantify beta 2-transferrin in serum. J Clin Chem Clin Biochem 1989;27:863-
cerebrospinal fluid to detect leakage of 8.
cerebrospinal fluid from skull injury. Clin Chem 25 Dinesen B, Nielsen S. A sensitive enzyme
1992;38:1908-12. immunoassay for the quantitation of human IgM.
11 Hansson LO, Lindquist L, Linne T, Sego E. Scand J Clin Lab Invest 1979;39:449-54.
Quantitation of C-reactive protein in 26 Hirohata S, Inoue T, Yamada A, Hirose S,
cerebrospinal fluid and serum by zone Miyamoto T. Quantitation of IgG, IgA and IgM
immunoelectrophoresis assay (ZIA). J Immunol in the cerebrospinal fluid by a solid-phase
Methods 1987;100:191-5. enzyme-immunoassay. Establishment of normal
12 Mavlight GM, Stuckey SE, Cabanillas FF, control values. J Neurol Sci 1984;63:101-10.
Keating MJ, Tourtellotte WW, Schold SC, 27 Javors MA, Bowden CL, Maas JW. 3-methoxy-4-
Freireich EJ. Diagnosis of leukemia or lymphoma hydroxyphenylglycol, 5-hydroxyindoleacetic
in the central nervous system by beta 2- acid, and homovanillic acid in human
microglobulin determination. N Engl J Med cerebrospinal fluid. Storage and measurement by
1980;303:718-22. reversed-phase high-performance liquid
13 Lutz CT, Cornell SH, Goeken JA. Establishment chromatography and coulometric detection using
of a reference interval for beta 2-microglobulin in 3-methoxy-4-hydroxyphenyllactic acid as an
cerebrospinal fluid with use of two commercial internal standard. J Chromatogr 1984;336:259-69.
assays. Clin Chem 1991;37:104-7. 28 Meucci G, Rossi G, Bettini R, Montanaro D,
14 Conly JM, Ronald AR. Cerebrospinal fluid as a Gironelli L, Voci L, Bianchi F. Laser
diagnostic body fluid. Am J Med 1983;75:102-8. nephelometric evaluation of albumin, IgG and
15 Chamczuk AJ, Ursell M, OConnor P, Jackowski alpha 2-macroglobulin: applications to the study
G, Moscarello MA. A rapid ELISA-based serum of alterations of the blood-brain barrier. J Neurol
assay for myelin basic protein in multiple Sci 1993;118:73-8.
sclerosis. J Immunol Methods 2002;262:21-7. 29 Mingioli ES, Strober W, Tourtellotte WW,
16 Cohen SR, Herndon RM, McKhann GM. Whitaker JN, McFarlin DE. Quantitation of IgG,
Radioimmunoassay of myelin basic protein in IgA and IgM in the CSF by radioimmunoassay.
spinal fluid. An index of active demyelination. N Neurology 1978;28:991-5.
Engl J Med 1976;295:1455-7. 30 Nerenberg ST, Prasad R. Radioimmunoassays for
17 Gressner AM, Wallraff P. Laser nephelometric Ig classes G, A, M, D, and E in spinal fluids:
quantitation of fibronectin (cold-insoluble
369
normal values of different age groups. J Lab Clin simultaneous determination of cerebrospinal fluid
Med 1975;86:887-98. and serum immunoglobulin-G and albumin. Trans
31 Phillips TM, More NS, Queen WD, Holohan TV, Am Neurol Assoc 1970;95:323-5.
Kramer NC, Thompson AM. High-performance 37 Vatassery GT, Krezowski AM, Sheridan MA.
affinity chromatography: a rapid technique for the Comparison of manual methods of determination
isolation and quantitation of IgG from cerebral of albumin in human cerebrospinal fluid by the
spinal fluid. J Chromatogr 1984;317:173-9. bromcresol green and immuno-precipitation
32 Poulik MD, Perry DJ, Vokac E, Sekine T. Post- methods. Clin Biochem 1980;13:78-80.
gamma globulin. II. Radioimmunoassay 38 Mashige F, Shimizu T, Iijima S, Ohkubo A.
determination of levels of post-gamma globulin Analysis for cerebrospinal fluid proteins by
and beta 2-microglobulin. Clin Chim Acta sodium dodecyl sulfate-polyacrylamide gel
1983;128:249-60. electrophoresis. Clin Chem 1992;38:2008-12.
33 Pressac M, Vignoli L, Aymard P. 39 Sorjonen DC, Cox NR, Swango LJ.
Immunoturbidimetric assay of transthyretin by Electrophoretic determination of albumin and
random-access analysis. Clin Chem 1988;34:176. gamma globulin concentrations in the
34 Schwerer B, Bach M, Bernheimer H. ELISA for cerebrospinal fluid of dogs with
determination of albumin in the nanogram range: encephalomyelitis attributable to canine distemper
assay in cerebrospinal fluid and comparison with virus infection: 13 cases (1980-1987). J Am Vet
radial immunodiffusion. Clin Chim Acta Med Assoc 1989;195:977-80.
1987;163:237-44. 40 Bichsel P, Vandevelde M, Vandevelde E, Affolter
35 Tourtellotte WW, Tavolato B, Parker JA, Comiso U, Pfister H. Immunoelectrophoretic
P. Cerebrospinal fluid electroimmunodiffusion. determination of albumin and IgG in serum and
An easy, rapid, sensitive, reliable, and valid cerebrospinal fluid in dogs with neurological
method for the simultaneous determination of diseases. Res Vet Sci 1984;37:101-7.
immunoglobulin-G and albumin. Arch Neurol 41 Soldin SJ, Brugnara C, Wong EC, eds. Pediatric
1971;25:345-50. Reference Intervals. 7th ed. Washington DC:
36 Tourtellotte WW, Tavolato BA, Parker JA, AACC Press; 2007.
Walker JE, Comiso PC. Electroimmunodiffusion:
a simple, rapid, reliable, and valid method for the
370
Tables and Figures

Table 1: Available Reference Intervals for CSF Proteins
Analyte Method Range (mg/dL)

1-Acid glycoprotein Radial immunodiffusion (RID) 0.28-0.54
Albumin [4,18,24,28,34] IA techniques, turbidimetric, Biuret/colorimetric 0-45 (3 mo 4 y)

10-30 (>4 y)
2-HS-glycoprotein Enzyme immunodiffusion (EID) 0.17
IgA [30] Radioimmunoassay (RIA) (in mg/dL by age):

15-20 y: 0.07 0.04
21-40 y: 0.07 0.03
41-60 y: 0.10 0.03
61-87 y: 0.11 0.06
IgD RIA (in U/mL by age):
15-20 y: 3.56 2.0
21-40 y: 3.02 1.3
41-60 y: 2.96 0.88
61-87 y: 3.20 0.9
IgG RIA (in mg/dL by age):
15-20 y: 3.5 2.0
21-40 y: 4.2 1.4
41-60 y: 4.7 1.0
61-87 y: 5.8 1.6
IgM RIA (in mg/dL by age):
15-20 y: 0.020 0.009
21-40 y: 0.016 0.003
41-60 y: 0.017 0.004
61-87 y: 0.017 0.005
2-Macro-globulin Laser nephelometry 0.11-0.45 mg/dL
2-Microglobulin [12,13] Nephelometry 1.5 0.2 mg/L
2-Transferrin [9,10] Electrophoresis, immunofixation electrophoresis, CSF, positive

Western blot
Myelin basic protein [7,15,16] RIA, ELISA <2.5 ng/mL
Total protein [1,2,41] Turbidimetry, nephelometry, Biuret/colorimetric (in mg/dL)

Adult 18-32
Pediatric:
Premature: 15-130
Full-term at birth: 40-120
<1 month: 20-80
>1 month: 15-40
Transthyretin/prealbumin Nephelometry, turbidimetry, densitometry ~2% of total protein
[23,33] concentration
NOTE: Reference intervals will vary by the method used and should be established in-house. In newborns, CSF proteins are
higher, owing to high permeability of the newborns BBB. Reference intervals for specific proteins have been published and
should be established or verified accordingly.
371
Procedure: Sulfosalicylic Acid Turbidimetric Assay 2. To prepare the standard curve, dilute a control
serum of 70 g/L of total protein with saline so that
Principle: Protein is precipitated as a fine, white 5 dilutions with protein concentration between
precipitate by the addition of sulfosalicylic acid. The 100 and 1500 mg/L are prepared as follows:
resulting turbidity is determined spectrophotometrically at
430 nm. Stock Volume Total Volume
Reagents Concentration (mg/L) (mL)
1. Sulfosalicylic acid, 30 g/L (118 mmol/L). 2.0 100
Dissolve 30 g of sulfosalicylic acid in 800 mL of 1.5 100
distilled water, and bring to 1 L. Store in a brown 1.0 100
bottle. Stable for 1 year at room temperature. 0.5 100
2. Control. Ortho Diagnostics CSF control (Ortho 0.2 100
Diagnostics, Raritan, NJ). 0 100
Assay
Equipment: Spectrophotometer capable of reading Do not use a serum control with elevated lipid or
at 430 nm. bilirubin content.
1. Add 1.0 mL of 3% sulfosalicylic acid to three 3. Standard curve. Construct the standard curve,
test tubes; one for an instrument blank, one for plotting absorbance at 430 nm versus protein
the Ortho standard, and one for the patient. concentration. This should be a straight line that
2. Add 0.2 mL of Ortho Diagnostics CSF control to passes through the origin.
the standard tube (click here). 4. The standard curve needs to be run only when
3. Add 0.2 mL of CSF to the patients tube. Add 0.2 there is a change in the spectrophotometer (i.e.,
mL of saline solution to blank tube. bulb change) or that lot of quality control material
4. If the spinal fluid is strongly colored, prepare a or the quality control sample does not give the
patient blank using 0.2 mL of CSF and 1.0 mL of expected value.
saline. Notes
5. Cover all test tubes with plastic cups or Parafilm, 1. The first drop of spinal fluid should be added
and gently invert several times. slowly to the sulfosalicylic acid reagent. If a
6. Let stand 10 min at room temperature. dense cloud forms immediately, the protein
7. Set absorbance at zero on the spectrophotometer concentration is very high, and dilution will be
at 430 nm, using the instrument blank of necessary. In this case, the rest of the 0.2-mL
sulfosalicylic acid plus saline. sample should not be added to the acid but instead
8. Invert test tube gently to mix, introduce sample returned to the original tube. The spinal fluid
into cuvette, and read absorbance (A) of all should be diluted with saline.
standards and patient samples. 2. If red blood cells are present in the spinal fluid,
9. Dilute samples 1:2 with saline, and rerun if the fluid must be centrifuged before protein
concentration is greater than 2000 mg/L. analysis.
Calculations
1. Correct absorbance of samples with blank:
Apatient Apatient blank = Acorrected patient
2. Read the corrected absorbance from the standard

curve (see below) to obtain the protein
concentration.
Controls and Standards
1. Since sulfosalicylic acid produces different
degrees of turbidity with equal concentrations of
different types of CSF proteins, it is necessary to
construct a curve using a standard that has the
same albumin-globulin ratio as the spinal fluid
being tested. Usually spinal fluid and serum have
a very similar albumin-globulin ratio, so a normal
serum control with the proper dilutions can be
used to construct the curve. The albumin-globulin
ratio of the standard should be between 1.0 and
1.5.
372
Ceruloplasmin
Ceruloplasmin
Ahmed Zakaria
Name: Ceruloplasmin: iron(II)-oxygen oxidoreductase (ferroxidase)

Merck Index: 14: 2006
Principles of Analysis and Current Usage i molecular oxygen, while ferric iron thereby becomes
Ceruloplasmin is a single polypeptide synthesized in the available for binding to and transport by transferrin.
liver [1]. It contains six to eight atoms of copper in its
structure [2], carries up to 90% of copper in plasma, and
performs ferroxidase (EC 1.16.3.1.) or iron (II):oxygen
oxidoreductase activity essential for normal iron
homeostasis. Ceruloplasmin catalyzes the oxidation of
ferrous iron Fe (II) to ferric iron Fe (III), which is carried
in the plasma by transferrin [3].
Holmberg and Laurel isolated and purified ceruloplasmin

from plasma and established its molecular mass, copper
content, and oxidase activity [4]. The purified protein has
an intense blue color and imparts a green color to serum
when the albumin is abnormally low. Human In 1967, Curzon and Speyer developed a colorimetric
ceruloplasmin is a 132-kDa protein, and the subunit-like method utilizing p-phenylenediamine as a substrate [7]
fragments that are found during its preparation are formed (Table 2, Method 1), whereas Schosinsky et al. used an o-
because of autolytic degradation [2]. Human dianisidine dihydrochloride as a substrate (Table 2,
ceruloplasmin is a single polypeptide chain and contains Method 2) [8]. Winkles et al. developed an automated
1046 amino-acid residues and four glucosamine assay for ferroxidase using a centrifugal analyzer [9]. They
oligosaccharides [5,6]. showed that the oxidase activity of ceruloplasmin towards
Fe (II) has physiological significance both in iron
Numerous assays were developed based on the activity of metabolism and as a serum antioxidant. Furthermore, the
the ceruloplasmin oxidase, using polyamine or polyphenol ferroxidase activities correlated significantly with
substrates such as p-phenylenediamine, N,N- ceruloplasmin and copper concentrations.
dimethylphenylenediamine, in addition to ferrous iron,
benzidine, and others. However, the measurements by Erel described an automated measurement of serum
these assays are relatively nonspecific. The principle of ceruloplasmin ferroxidase activity in which Fe (II) ions
ceruloplasmin oxidase assays is based on the conversion of were used as the substrate in acetate buffer pH 5.8 [10].
substrate to the oxidized colored product that is measured The ceruloplasmin catalytic activity was determined by
photometrically. Ceruloplasmin utilizes the electron oxidation of ferrous ions to ferric ions and a chromogen, 3-
chemistry of bound copper ions to couple iron oxidation (2-pyridyl)-5,6-bis(2-[5-furylsulfonic acid])-1,2,4-triazine
with the four-electron reduction of dioxygen (Scheme 1). was added at the end of the incubation period. This
The reduced form of ceruloplasmin slowly reduces chromogen forms a colored complex with the remaining
nonoxidized ferrous ions but not with ferric ions. The
i colored Fe (II) complex was measured photometrically at
Ceruloplasmin 590 to 610 nm. The difference in the ferrous ion
Previous and current authors of this method: concentration before and after the enzymatic reaction
First edition: Not done indicated the amount of oxidized ferrous ion [10]. The
Methods edition: Gayle Jackson range for serum ceruloplasmin ferroxidase activity in
Second edition: Not updated healthy persons was 198 to 1107 U/L.
Fifth edition: Ahmed Zakaria
373
Ceruloplasmin
The enzymatic methods measure the oxidase activity of The antisera used in most of the immunochemical methods
ceruloplasmin by using either p-phenylenediamine or o- were significantly different in their antibody activity [13].
dianisidine dihydrochloride as the substrate. The p- The reason is in part due to ceruloplasmin degradation or
phenylenediamine method is relatively simple and has loss of copper from its structure, which subsequently
been utilized in clinical laboratories since the early 1960s. produces variable antigenicity if used as an immunogen.
Nevertheless, the method is not specific for ceruloplasmin
for several reasons. Catalytic oxidation of the substrate by The variation in antiserum specificity makes it difficult to
free metals such as cupric and ferrous ions occurs, causing standardize ceruloplasmin measurement between
falsely elevated levels. The reaction is inhibited by high laboratories.
concentrations of blood urea nitrogen (>1 g/L), uric acid
(>200 mg/L), or bilirubin (163 mg/L) [8]. The oxidation of Specimen
p-phenylenediamine is decreased by ascorbic acid and Serum is the preferred specimen. EDTA and citrate are not
increased by exposure to visible light. The enzymatic acceptable if enzymatic methods are used. Assays should
assays are also subject to negative interference by certain be performed on fresh specimens. If this is not possible,
anions and organic compounds [7]. Since the o-dianisidine serum should be frozen until the assays are performed. No
is more stable and forms a more stable product than p- fasting or other preparation is usually needed [14].
phenylenediamine, the use of o-dianisidine as a substrate is Ceruloplasmin is stable for 3 days at 4C and 4 weeks at
the preferred method, based on the oxidase activity 20C. Ceruloplasmin is susceptible to degradation or
measurements. The direct enzymatic methods were oxidative modification, which alters its reactivity with
preferred until the development of immunochemical antibodies.
methods, which were considered specific and easy to
perform in hospital laboratories (Table 2, Methods 3 and Interferences
4). The immunochemistry methods are subject to interfering
substances, including hemoglobin (500 mg/dL), lipids
Reference and Preferred Methods (1000 mg/dL), and bilirubin (30mg/dL). Increasing
There is no specific reference method for ceruloplasmin. triglyceride may contribute to the depression of
Immunoturbidimetry or immunonephelometry are the ceruloplasmin concentration up to 50% [15,16].
preferred methods for quantitative determination of human
ceruloplasmin concentration on automated instruments Ceruloplasmin Reference Interval
[11]. Methods based on nephelometric and turbidimetric The reference intervals for ceruloplasmin are shown below
assays demonstrate acceptable imprecision compared to [14]:
the earliest immunochemical assays, radial
immunodiffusion, and rocket laurel Table 1: Ceruloplasmin Reference Intervals
immunoelectrophoresis [12]. In the immunochemical Age Concentration (mg/L)
methods, a specific antibody is used against ceruloplasmin. Cord blood (term) 50330
<5 months 150560
Recently, more specific anti-human ceruloplasmin 56 months 260830
antibodies have become available, and a variety of manual, 736 months 310900
automated immunoturbidimetric and >312 years 250450
immunonephelometric assays are commercially available
on several manufacturers platforms. >1219 years Male 150370
Female 220500
Pesce and Bodourian described a kinetic nephelometric Adults, Male 220400
assay for the measurement of ceruloplasmin in serum [12].
Adults, Female (No 250600
The reaction rate was not affected by temperature changes
contraceptive)
between 30C and 37C or pH between 6.0 and 8.4.
(Contraceptive estrogen) 270660
Ceruloplasmin concentrations measured by the centrifugal
(Pregnant ) 3001200
analyzer were compared to results obtained by
nephelometric, radial immunodiffusion, and an enzymatic
procedure using p-phenylenediamine as the substrate [12].
The correlation coefficients were 0.93, 0.88, and 0.86,
respectively. They attributed the bias between the Normal adult plasma concentration is reached 3 to 6
nephelometric and the RID techniques to the different months after birth. Females taking oral contraceptives
antisera specificities, since the antisera were from different show higher concentrations of ceruloplasmin when
sources. Unlike the enzymatic procedures, measurement of compared with age-matched females [17]. Pregnant
ceruloplasmin with immunochemical methods does not females have ceruloplasmin concentrations that are
require rigorously controlled temperature or pH. approximately twice those of nonpregnant women. Aging
374
Ceruloplasmin
does not appear to result in any significant changes in References

ceruloplasmin concentrations when compared with healthy 1 Ryden L. Single-chain structure of human
young adults [18]. ceruloplasmin. Eur J Biochem 1972;26:380-386.
2 Kingston IB, Kingston BL, Putnam FW.
Interpretation Chemical evidence that proteolytic cleavage
Ceruloplasmin is synthesized in the liver. Its primary causes the heterogeneity present in human
function is to transport copper, and in addition, it plays an ceruloplasmin preparations. PANS, USA
important role in iron transport, metabolism, and 1977;74:5377-5381.
homeostasis. Its major function may be to facilitate the net 3 Osaki S, Johnson DA, Frieden E. The possible
flux of iron between cells and tissues significance of the ferrous oxidase activity of
[19,20]http://www.jbc.org/cgi/ijlink?linkType=ABST&jou ceruloplasmin in normal human serum. J Biol
rnalCode=sci&resid=279/5351/714. As a result of Chem 1966;241:2746-2751.
defective copper transport, ceruloplasmin concentrations 4 Holmberg CG, Laurell CB. Investigations in
are usually but not always reduced in Wilsons disease serum copper III. Ceruloplasmin as an enzyme.
[21]. However, in Wilsons disease, copper is accumulated Acta Chem Scand. 1951;5:476-480.
in tissues including liver (hepatolenticular degeneration), 5 Dwulet FE, Putnam FW. Complete amino acid
brain, and cornea [21]. Decreased ceruloplasmin levels sequence of a 50,000-dalton fragment of human
have been observed with nephrosis, tropical and ceruloplasmin. PNAS, USA 1981;78:790-794.
nontropical sprue, Menkes syndrome (the defect is 6 Takahashi N, Ortel TL, Putnam FW. Single-chain
secondary to poor absorption and utilization of dietary structure of human ceruloplasmin: the complete
copper), scleroderma of the small bowel, primary biliary amino acid sequence of the whole molecule.
cirrhosis, and primary biliary atresia [22]. PANS USA 1984;81:390-394.
7 Curzon G, Speyer BE. Inhibitors of
Ceruloplasmin is an acute-phase reactant, and blood levels ceruloplasmin. Biochem J 1967;105:243-250.
may increase in infections or inflammatory disease [23]. 8 Schosinsky KH, Chavarria M, Lehmann HP.
High levels occur in pregnancy, with estrogens, and with Measurement of ceruloplasmin from its oxidase
oral contraceptive use when the agent contains estrogen as activity in serum by use of o-dianisidine
well as progesterone [24]. dihydrochloride. Clin Chem 1974;20:1556-1563.
9 Winkles J, Joens AF, Winyard P, Blacke DR,
Excessive therapeutic zinc may block intestinal absorption Lunec J. An automated method for the kinetic
of copper and lead to a copper deficiency syndrome measurement of ferroxidase activity. Ann Clin
characterized by hypochromic microcytic anemia with Biochem 1988;25:250-254.
leukopenia/neutropenia and very low concentrations of 10 Erel O. Automated measurement of serum
ceruloplasmin [25]. ferroxidase activity. Clin Chem 1998;44:2313-
2319.
Ceruloplasmin is also associated with diabetes and 11 Blirup-Jensen S. Protein standardization III:
cardiovascular disease. It is involved in promoting LDL method optimization. Basic principles for
oxidization and inhibits nitric oxide (NO) production [26]. quantitative determination of human serum
proteins on automated instruments based on
Ceruloplasmin Performance Goals turbidimetry or nephelometry. Clin Chem Lab
Both the Clinical Laboratory Improvement Amendments Med 2001;39:1098-1109
of 1988 (CLIA 88) and biological variability may be used 12 Pesce MA, Bodourian SH. Nephelometric
to determine the performance goal of an analytical assay. measurement of ceruloplasmin with a centrifugal
The intraindividual and interindividual biological variation analyzer. Clin Chem 1982;28:516-519.
of ceruloplasmin concentration is 5.8% and 11.1%, 13 Buffone GJ, Brett EM, Lewis SA, Iosefsohn M,
respectively. Therefore, the desired performance for Hicks JM. Limitations of immunochemical
ceruloplasmin assays is defined by the biological variation measurement of ceruloplasmin. Clin Chem
imprecision estimate as 2.9% (50% of intraindividual 1979;25:749-751.
variation, which is 5.8%). Using these criteria, the total 14 Roberts WL, McMillin GW, Burtis CA, Burns
error allowable in ceruloplasmin measurement is estimated DE. Tietz Textbook of Clinical Chemistry. 4th ed.
to be 7.8% [27]. The overall ceruloplasmin performance Philadelphia: Saunders; 2006.
ranges are considered to be well within the capability of 15 Bornhorst JA, Roberts RF, Roberts WL. Assay-
most automated immunoassays currently available in specific differences in lipemic interference in
clinical laboratories. The analytical performance of the native and Intralipid-supplemented samples. Clin
immunonephelometric methods is summarized in Table 3. Chem 2004;50:2197-2201
The data is from the 2007 College of American 16 IMMAGE Immunochemistry Systems:
Pathologists surveys. Beckman Chemistry Information Package Insert;
2005.
375
Ceruloplasmin
17 Milne DB, Johnson PE. Assessment of copper ed. Boston/Toronto: Little, Brown & Company;
status: affect of age and gender on reference 1986.
ranges in healthy adults. Clin Chem 1993;39:883- 23 Denko CU. Protective role of ceruloplasmin in
887. inflammation. Agents Actions 1979;9:333-343.
18 Gaggiotti G, Orlandoni P, Ambrosi S, OnoratoG, 24 Carruthers ME, Hobbs CB, Warren RL. Raised
Piloni S, Amadio L et al. The influence of age and serum copper and ceruloplasmin levels in subjects
sex on nutritional parameters in subjects aged 60 taking oral contraceptives. J Clin Pathol
years and over. Arch Gerontol Geriat 1966;19:498-500.
1995;12:117-128. 25 Irving JA, Mattman A, Lockitch G, Farrell K,
19 Mukhopadhyay CK, Attieh ZK, Fox PL. Role of Wadsworth LD. Element of caution: a case of
ceruloplasmin in cellular iron uptake. Science reversible cytopenias associated with excessive
1998;279:714-717. zinc supplementation. Can Med Assoc J
20 Attieh ZK, Mukhopadhyay CK, Seshadri V, 2003;169:129-131.
Tripoulas NA, Fox PL. Ceruloplasmin ferroxidase 26 Mukhopadhyay C, Ehrenwald E, Fox P.
activity stimulates cellular iron uptake by a Ceruloplasmin enhances smooth muscle cell and
trivalent cation-specific transport mechanism. J endothelial cellmediated low density lipoprotein
Biol Chem 1999;274:1116-1123. oxidation by a superoxide-dependent mechanism.
21 Percival SS, Harris ED. Copper transport from J Biol Chem 1996;271:14773-14778.
ceruloplasmin: characterization of the cellular 27 Fraser GC. Biological Variation: From Principle
uptake mechanism. Am J Physiol Cell Physiol to Practice. Washington, DC: AACC Press; 2001.
1990;258:C140-C146.
22 Wallach J. Interpretation of Diagnostic Tests: A
Handbook Synopsis of Laboratory Medicine. 4th
Table 2: Methods Summary

Method Principle of Analysis: Comments:
1 Colorimetric. Ceruloplasmin oxidizes p- Historical; substrate requires special

phenylenediamine in the presence of oxygen to form a precautions and is unstable and
purple product with an absorption maximum at 530 nm nonspecific
2 Colorimetric. o-Dianisidine dihydrochloride is oxidized Serum; preferred chemical
by ceruloplasmin at pH 5 in the presence of oxygen; the oxidation
product formed is solubilized in sulfuric acid, 9 mol/L,
and measured at 540 nm
3 Radial immunodiffusion. Specific antibody to Serum; simple immunochemical
ceruloplasmin forms precipitin ring in agarose medium method; less frequently used
4 Immunonephelometric. Specific antibody to Serum; requires specific
ceruloplasmin forms complex that scatters light instrumentation; currently used
376
Ceruloplasmin
Table 3: Analytical Performance of Ceruloplasmin Assays

METHOD SURVEY College of American Pathologists Survey 2007
NEPHELOMETRY No. of Mean SD CV Median Low High

Labs mg/dL Value Value
Beckman Array S2B 11 33.8 1.3 3.7 34.0 32.0 36.0

S2C 12 3.6 0.5 14.4 4.0 3.0 4.0
Beckman Immage S2B 99 34.4 1.8 5.1 35.0 31.0 38.0

S2C 105 3.4 0.5 14.5 3.0 3.0 4.0
Dade Behring S2B 74 38.2 2.3 6.1 38.0 31.0 44.0

Nephelometer systems S2C 75 3.9 0.4 10.2 4.0 3.0 5.0
TURBIDIMETRIC
Roche COBAS Integra S2B 21 43.0 2.8 6.5 43.0 38.0 50.0
S2C 12 4.7 0.6 14.0 0.0 4.0 6.0
Roche Hitachi/Cobas c S2B 18 40.8 1.1 2.7 41.0 39.0 43

S2C 5 - - - 5.0 4.0 8.0
Olympus AU system S2B 5 - - - 40.0 36.0 42

377
Chloride
Chloride
William J. Korzun
Name: Chloride
Clinical significance: Refer to Chapter 28, Physiology and Pathophysiology of Body Water and
Electrolytes, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Atomic symbol: Cl
Merck Index: 2062
Chemical class: Inorganic anion
NRSCL Definitive and
Reference Methods: NCCLS RS10-9
i principle, one can determine the absolute amount of

silver ions generated from the number of coulombs
Chloride is the major anion in extracellular body fluids (current time) produced and Faradays constant
and is measured in serum, plasma, urine, sweat, and (96,487 coulombs per equivalent). In practice, the time
occasionally other body fluids. The most common required to titrate a chloride standard solution or
method in current use is ion-selective electrode (ISE), unknown sample with a constant current is measured.
representing greater than 99% of College of American The unknown concentration is then calculated according
Pathologists (CAP) Comprehensive Chemistry survey to the following formula:
participants in 2007. Amperometric/coulometric titration
and mercuric nitrate/thiocyanate methods accounted for Chloride conc. (standard) = Chloride conc. (unknown)
the non-ISE-based methods in use. Titration time (standard) Titration time (unknown)
Ion-selective electrode methods are used in most Methodological precision in coulometric titration is
automated chemistry analyzers, ranging from large, limited by the rate of mass transfer of silver ions from
multichannel instruments to compact, physicians-office the generating electrode to the bulk solution. The time
lab equipment. The measurement technology (Table 1,
needed to detect the end-point once all the Cl is
Method 1) and limitations are essentially the same as consumed is called the blank time and depends on the
described for sodium and potassium. The ion-selective
rate of mixing in the vessel and the rate of Ag+
sensing element is typically silversilver chloride or
formation. For maximum precision, the blank time
silver sulfide. ISE methods using direct, undiluted
should be a small fraction of the total titration time. At
specimens and using indirect, diluted specimens are both
very high chloride concentrations (400 mmol/L under
used for chloride measurement.
typical test conditions), the physical bulk of precipitate
interferes with mass transfer and obscures detection of
The coulometric titration method is used in some
the end-point. Dirty electrodes will cause poor
analyzers and is also the basis for the reference method
reproducibility of results, and some workers recommend
for chloride. Silver ions, generated at a constant voltage
cleaning with ammonium hydroxide and nitric acid
from a silver electrode, react with chloride ions to form
solutions between each titration [2]. Gelatin in the
insoluble silver chloride (Table 1, Method 2) [1,2]. The
original procedure and polyvinyl alcohol, which is more
end-point is detected amperometrically by a second pair
stable, in current procedures prevent reduction of silver
of electrodes that specifically measure the free silver
chloride at the indicating electrodes and promote
ions that result when all chloride ions are consumed. In
uniform deposition of excess silver ions on the indicator
cathode. This results in a smooth amperometric current
i and a reproducible detection of the end-point. It is
Chloride
Previous and current authors of this method: necessary for the titration mixture to have an acid pH to
First edition: W. Gregory Miller prevent the formation of poorly soluble basic silver salts.
Methods edition: F. Phillip Anderson, W. Gregory Nitric acid and acetic acid in the titration reagent ensure
Miller an acid pH. Nitric acid also provides ionic conductivity,
Second edition: F. Phillip Anderson, W. Gregory and the acetic acid sharpens the end-point by reducing
Miller the slight solubility of silver chloride [1].
Third edition: F. Phillip Anderson, W. Gregory
Miller The mercuric thiocyanate method employs the
Fourth edition: F. Phillip Anderson, W. Gregory quantitative displacement of thiocyanate by chloride
Miller from mercuric thiocyanate and subsequent formation of
Fifth edition: William J. Korzun a red ferric thiocyanate complex, which is measured
378
Chloride
colorimetrically at 525 nm (Table 1, Method 3) [3]. The produce unreliable results because chloride
sensitivity and linear range of the reaction are adjusted concentration varies with sweating rate and because the
by addition of an excess of mercuric ions as mercuric quantity of chloride may be insufficient for accurate
nitrate. The chloride will first combine with the free analytical measurement. Children under 2 to 3 weeks of
mercury ions (colorless) and then displace any age may have unusually high sweat electrolyte
thiocyanate from mercuric thiocyanate to produce a concentrations. It is therefore advisable to delay the
colored product measured at 525 nm (Figure 1). This sweat test until after this time [7]. Sweat specimens
approach limits the minimum detectable chloride to 80 should always be collected in duplicate. This procedure
mmol/L but greatly improves the sensitivity (absorbance is important in the diagnosis of cystic fibrosis.
change per millimole) in the clinically important serum
range of 80 to 125 mmol/L. The ferric thiocyanate Interferences
reaction is very temperature sensitive, and a constant All the methods for chloride will show positive
temperature must be maintained to obtain accurate interference from other halides. The only clinically
results. Failure to do so can result in a significant drift in important interfering halide is bromide, which is
some automated methods. administered in some drug preparations. Bromide and
chloride do not react equivalently in all analytical
A less common spectrophotometric method that has been systems. Bromide ion showed a reactivity equivalent to
used by some automated analyzers is based on the 2.3 chloride ions with one ion-selective electrode
reaction between ferric nitrate and chloride ions to form instrument, an equivalency to 1.6 chloride ions with the
a ferric chloride complex which absorbs light at 340 nm continuous-flow mercuric thiocyanate method, but an
(Table 1, Method 4). The definitive method for chloride equivalency to 1 chloride ion with the coulometric
analysis is isotope dilutionmass spectrometry using titration method [8]. Since one would like to be able to
samples spiked with 37Cl (Table 1, Method 5) [4]. detect bromism, a system that is highly responsive to
bromide may be desirable.
Coulometric titration is both an American Association Ion-selective, electrode-based methods are generally not
for Clinical Chemistry (AACC) selected method [2] and affected by bilirubin or triglycerides. Gross hemolysis
the National Institute of Standards and Technology may, however, result in a negative bias as a result of
(NIST) reference method for serum chloride [4] because dilution. The ferric thiocyanate procedure is subject to
of its precision and its freedom from interferences (see interference from bilirubin, hemoglobin, and lipemia if
Procedure: Serum Chloride by Coulometry). Using the the assay does not include a protein-separation step, such
NIST procedure, 14 evaluating laboratories achieved an as dialysis. However, procedures using small sample
average standard deviation of approximately 0.5 volumes (0.01 mL of serum to 1 mL of reagent) show no
mmol/L, with a maximum bias of 0.5 mmol/L, analyzing interference from bilirubin up to 220 mg/L or from
seven serum pools over a chloride range of 79.2 to 116.8 triglycerides up to 6 g/L [3].
mmol/L. Coulometric titration, ion-selective electrode,
and mercuric-ferric thiocyanate methods all yielded Chloride Reference Interval
clinically identical results [5]. Serum or Plasma, 101 to 111 mmol/L.
These values were determined as the central 95% range
Specimen obtained from analysis of 1205 (596 males, 609 females)
Serum, heparinized plasma, urine, and other body fluids healthy, ambulatory persons 1 to 78 years old. No
are acceptable. Serum, plasma, and other fluids should variation with sex or age was observed. An automated
be promptly separated from cells to avoid shifts in the mercuricferric thiocyanate method, which was
ionic equilibrium with metabolism and pH changes. calibrated against the coulometric titration method using
Chloride in serum, plasma, urine, and other fluids is primary aqueous standards, was used. Reference
stable for at least 1 week at room, refrigerator, or freezer intervals obtained with use of ISE-based methods are
temperatures. Some ion-selective analyzers can measure generally lower by 2 to 3 mmol/L. A slight diurnal
chloride activity in whole blood, and such measurements variation in chloride has been described; maximum
should be made within 2 hours to avoid ionic shifts in values occur from noon to 2:00 pm, and lowest values
electrolytes. Samples collected in gel-barrier tubes do are seen at night. Chloride concentrations may decrease
not show any difference in chloride when compared with slightly following meals because of the role of chloride
samples collected in tubes not containing gel. in gastric acid production.
Sweat for chloride measurement is collected after Urine, 110 to 250 mmol/24 h.
iontophoretic delivery of pilocarpine to skin sweat Urine chloride levels are strongly influenced by dietary
glands to stimulate sweating. The area of skin that must intake.
be stimulated depends on the volume of sweat needed
for analysis. Ion-selective electrode methods require less Sweat [6,7,9,10]
volume than coulometric titration methods. In any event,
Normal 0 to 40 mmol/L
the rate of sweating must be greater than 1 g/m2/min to
Indeterminate for cystic fibrosis 40 to 60 mmol/L
obtain reliable results [6]. Lower sweating rates will
Suggestive of cystic fibrosis >60 mmol/L
379
Chloride
2 Dietz AA, Bond EE. Chloride, coulometric-

Reliable sweat chloride measurements depend on an amperometric methods. In: Faulkner, WR,
adequate rate of sweating being achieved, as discussed Meites S, eds. Selected Methods of Clinical
earlier. A repeat determination of sweat chloride (new Chemistry. Vol 9. Washington, DC: American
collection and analysis) should always be made to Association for Clinical Chemistry; 1982:149-
confirm a positive finding. Intermediate values do not 152.
indicate heterozygote status. Adults have a more variable 3 Levinson SS. Chloride, colorimetric method.
sweat composition than children and can have sweat In: Faulkner, WR, Meites S, eds. Selected
chloride values up to 70 mmol/L. Methods of Clinical Chemistry. Vol 9.
Washington, DC: American Association for
Interpretation Clinical Chemistry; 1982:143-148.
Blood chloride concentration changes in a variety of 4 Velapoldi RA, Paule RC, Schaffer R, Mandel
electrolyte disturbances, often paralleling changes in TJ, Gramlich JW. Standard reference materials:
sodium concentrations. Thus chloride levels increase in a reference method for the determination of
dehydration and are decreased in overhydration, chloride in serum. NBS Special Pub No. 260-
congestive heart failure, Addisons disease, and other 67. Washington, DC: U.S. Government Printing
disturbances of sodium-water balance. Primary Office; 1979.
decreases in chloride are seen after gastric suction, 5 Geisinger KR, Geisinger KF, Wakely PE Jr,
vomiting, and diarrhea. Batsakis JG. Serum chloride: a CAP survey.
Am J Clin Pathol 1980;74:546-551.
Chloride Performance Goals 6 Clinical Laboratory Standards Institute. Sweat
The current targets for acceptable performance for testing: specimen collection and quantitative
chloride measurements are reported values that are 5% analysis. Approved Guideline CLSI C34-A2.
of the peer-group mean. Coulometric titration at normal Villanova, PA: CLSI; 2000.
serum concentrations of chloride gives within- 7 Hammond KB, Johnston EJ. Sweat test for
laboratory, between-day coefficients of variation (CV) of cystic fibrosis. In: Faulkner WR, Meites S, eds.
1% to 1.5% for automated methods and 2% to 2.5% for Selected Methods of Clinical Chemistry. Vol 9.
manual methods. Analysis with automated ion-selective Washington, DC: American Association for
electrode or mercuric thiocyanate methods gives a CV of Clinical Chemistry; 1982:347-351.
1%. The manual coulometric titration method and any of 8 Elin, RJ, Robertson, EA, Johnson, E. Bromide
the automated methods are recommended for routine interferes with determination of chloride by
analysis of serum. Because of the wide range of chloride each of four methods. Clin Chem 1981;27:778-
concentrations possible in urine and sweat, chloride 779.
analyses are best done with a coulometric method. Ion- 9 Gibson LE, Cooke RE. A test for concentration
selective electrode methods are also satisfactory within of electrolytes in sweat in cystic fibrosis of the
the manufacturers documented reportable range. pancreas, utilizing pilocarpine by iontophoresis.
Mercuric thiocyanate methods are generally not suitable J Pediatr 1959;55:545-549.
for urine or sweat because of the limited range of 10 Warwick WJ, Hansen L. Measurement of
concentration sensitivity. chloride in sweat with the chloride-selective
electrode. Clin Chem 1978;24:050-2053.
Chloride concentrations are relatively stable in healthy 11 Gonzalez-Revalderia J, Garcia-Bermejo S,
individuals. Intraindividual variability was 1.2% in Menchen-Herreros A, Fernandez-Rodriguez E.
healthy individuals over a 20-week period [11]. Towards narrower analytical goals in routine
laboratory work. Clin Chem 1991;37:596.
References
1 Cotlove E. Chloride: Standard Methods of
Clinical Chemistry. Vol 3. New York:
Academic Press; 1961:81-92;
380
Chloride
Table 1: Methods of Chloride Measurement

Method 1: Ion-selective electrode; quantitativepotentiometric end-point, or kinetic
Principle of analysis: Ag/AgCl(s); test solution | reference electrode
Comments: Most frequently used, good accuracy and precision; manual or automated
Method 2: Coulometric titration; quantitativetitration, end-point
Principle of Analysis: Ag+ + Cl AgCl()
Comments: Reference method, highly accurate; manual or automated
Method 3: Mercuric/ferric thiocyanate; quantitativeend point
Principle of analysis: 2 Cl + Hg(SCN)2 HgCl2 + 2(SCN)
3 (SCN) + Fe3+ Fe(SCN)3 (red); Amax, 525 nm
Comments: Less frequently used, good accuracy and precision; manual or automated
Method 4: Ferric chloride; quantitativeend point
Principle of analysis: Fe3+ + Cl (FeCl)2+
Comments: Uncommon
Method 5: Isotope dilution; mass spectrometric
Principle of analysis: Dilution of common isotopes with 37Cl
Comments: Definitive method; used in research or reference laboratories
Figure 1
Difference spectra of mercuric ferric thiocyanate complexed with chloride. Reagent versus water, long dashes; chloride
(100 mmol/L) plus reagent versus water, short dashes; chloride (100 mmol/L) plus reagent versus reagent, solid curve.
381
Chloride
Procedure: Serum Chloride by Coulometry acid/PVA solution. Raise each vial, in turn, into
position. Record each of the blank times,
Principle average the last four values, and enter that
This coulometricamperometric titration is based on the number on the BLANK thumbwheel switch.
method of Cotlove [1]. The silver ions generated by the 4. Titration. Raise a vial containing 5.5 mL of
instrument combine with the chloride in the test sample. acid/PVA solution and 0.1 mL of chloride
The end-point is detected by a second pair of electrodes standard into position, and press the
which measure the free silver ions present when all the TITRATION switch. The reagent blank time
chloride ions have reacted. will automatically be subtracted, and the value
displayed will be in meq/L. If replicate
Reagents determinations of the standard are not within +1
1. Nitric acetic acid reagent (0.1 mol/L HNO3 + meq/L of the assigned value, use the
1.76 mol/L acetic acid). To 800 mL of distilled water, COMPENSATE switch and knob on the back
add 6.4 mL of concentrated nitric acid and 100 mL of of the instrument to adjust the calibrator. If the
glacial acetic acid. Mix. Stable for 6 months at room calibration of the instrument is adjusted with
temperature. the COMPENSATE switch, all specimens in
2. Polyvinyl pyrrolidone (PVA) solution, 18 the run must be analyzed with the switch in that
g/L. 1.8 g of powdered PVA is placed in 100 mL of position.
cool distilled water. Heat just to boiling with stirring to To analyze controls and patient specimens,
dissolve PVA. Cool. Use immediately to prepare raise a vial containing 5.5 mL of acid/PVA
acid/PVA solution. solution and 0.1 mL of specimen into position.
3. Acid/PVA solution. Add the 100 mL of PVA When the titration is complete, the displayed
solution to the 900 mL of nitric acetic acid reagent. Mix. reading will be the chloride concentration in the
This is stable at room temperature for 6 months. specimen.
4. Chloride standard, 100 mEq/L of Cl. Place Calculations
5.845 g of dry NaCl in a 1-L volumetric flask, and add For the chloridometer utilized in the above
approximately 900 mL of distilled water to dissolve the procedure, no calculation is required. Rinse electrodes
salt. Bring volume to 1 L with distilled water and mix after each sample.
well. Stable for 6 months at room temperature. For titrators without direct mEq/L readout, set
test switch to high. Run standard, control, and patient
Specimen samples, recording time in seconds for each. Calculate
the mEq/L as follows:
Time unknown time blank mEq/L Std = mEq/L of unknown
Time standard time blank
Assay
Equipment: The assay is performed in our
laboratory with a Labconco Digital Chloridometer Example:
Model 442-5000. Different instruments will require Titration time of standards: 42.3, 42.7 sec
slightly different protocols, which will be found in the Titration time of blanks: 1.7, 2.1 sec
individual instrument operation manuals. Titration time of unknown: 41.2 sec
1. Preparation of sample. Dispense 4.0 mL of
acid/PVA solution into vial. Pipet 0.1 mL of 41.2 1.9 100 = 96.8 = 97 mEq/L
standard, sample, or control into appropriate 42.5 1.9
vials, and rinse pipet in acid/PVA solution.
Mix. For blanks, use 4.0 mL of acid/PVA
solution. NOTE: Rinsing the pipet is absolutely
necessary to avoid viscosity errors between
aqueous standards and serum.
2. Daily setup. With the instrument off, verify
that the generator electrode is the same length
and thickness as the other electrodes. If not,
feed additional silver wire from the spool, clip
off the thinned-out tip so that the fresh wire is
at the same length as the other electrodes.
Tighten the binding post/electrical connection,
and polish all of the electrodes. Rinse
thoroughly with deionized water and blot dry.
Make sure that no liquid is bridging any of the
electrodes at their common mounting post.
3. Reagent blank time delay adjustment. Turn the
instrument on. Set the TITRATION switch to
AUTO. Fill each of six vials with 5.5 mL of
382
Cholesterol
Cholesterol
John R. Burnett and Ken Robertson
Name: Cholesterol
Clinical significance: Refer to Chapter 37, Coronary Artery Disease: Lipid Metabolism, in the 5th
Molecular formula: C27H46O

Merck Index: 2181
Chemical class: Sterol, lipid
H3C
CH3
CH3
CH3
CH3
Structure: HO
i rest in the free form. This has some analytical implications,
since in some chemical reactions, the color development
Cholesterol is a sterol compound that is found in all with the ester cholesterol is greater in intensity than that with
animal tissues and serves many important physiological free cholesterol. This in turn can lead to a large positive bias.
functions, including being a substrate for the synthesis of In some enzymatic reactions, hydrolysis of longer chain
bile acids and steroid hormones; it is an important cholesteryl esters, such as cholesteryl arachidonate, is not
component in cell membranes. Because cholesterol complete, resulting in a negative bias. For this reason,
appears to be involved in atherosclerosis, cholesterol understanding the chemistry of the various cholesterol
measurement is one of the most common laboratory tests methods and recognizing their limitations is of utmost
used today [1]. importance in the selection of the method to be used in the
laboratory. Most often, compromises must be made amongst
The history of this famous organic compound goes back to simplicity, speed, convenience, accuracy, and precision. The
the 19th century. Liebermann [2], in 1885, first described the result of these compromises up to now has been an assay that
color reaction of sulfuric acid with a solution of cholesterol still poses many problems for most laboratories and produces
in acetic anhydride. Four years later, Burchard [3] reported results of questionable reliability [17].
that a more intense blue-green color is produced when acetic
anhydride and sulfuric acid are added to a solution of Classification of Cholesterol Methods
cholesterol in chloroform (Table 1, Method 1). Since that Single-Step Methods
time, this Liebermann-Burchard reaction has been widely In these direct assays, there is no sample preparationthat
used as a colorimetric reaction for the estimation of is, no isolation and purification of the steroid or steroids.
cholesterol in biological fluids. Literally hundreds of Thus direct procedures are those carried out on serum or
cholesterol methods have been reported since then. For a plasma samples without any prior solvent extraction steps.
historical rsum and authoritative critique of cholesterol These methods are simple and rapid and require little
methods, consult review articles by Zak and co-workers [4- manipulation of the sample and therefore are suitable for
9], Tonks [10], Naito and David [11], Naito [12-14], and automation. However, depending upon the chemical
Martnek [15]. The College of American Pathologists (CAP) reaction, these procedures are likely to exhibit both positive
Check Sample exercise (Special Topics No. ST-104) written and negative errors resulting from the presence of proteins,
by Werner et al. [16] is an excellent technical review. bilirubin, hemoglobin, vitamins A, C, and D, steroid
hormones, uric acid, turbidity, and differences in the
Total cholesterol measurement includes both the esterified chromogenicity of free cholesterol and ester cholesterol.
and free forms of the sterol. In serum or plasma, two thirds Many automated procedures are based on direct methods.
of the total cholesterol exists in the esterified form, with the They are usually based on such acceptable methods as
Zlatkis et al. [18], Huang et al. [19], Pearson et al. [20], and
enzymatic procedures [21-42].
i
Cholesterol
Previous and current authors of this method: Two-Step Methods
First edition: Herbert K. Naito In two-step assays, an organic-phase extraction step is
Methods edition: Herbert K. Naito introduced before measurement of cholesterol and other
Second edition: Herbert K. Naito chemically related steroids. This pretreatment step removes
Third edition: Herbert K. Naito many nonspecific chromogens that might interfere with the
Fourth edition: Herbert K. Naito assay, with the relative chromogenic response of each
Fifth edition: John R. Burnett, Ken Robertson interfering substance being dependent upon the chemistry
383
Cholesterol
involved. Since no saponification step is involved in these B reaction generally is carried out in a strong acid medium
methods, the deleterious effect of differential color response sulfuric acid, acetic acid, and acetic anhydride. In the
between free and ester cholesterol still remains, especially in chemical reaction, cholesterol goes through a stepwise
Liebermann-Burchard reactions. However, for most routine oxidation, with each step yielding a cholestapolyene
clinical work, this two-step procedure has been well molecule, which has one more double bond than the
accepted, with the values obtained corresponding closely compound from which it was derived. The initial L-B step
with that of the Abell et al. method [43]. The correlation is involves protonation of the OH group in cholesterol and
even closer when a calibration factor is added to correct for subsequent loss of water to give the carbonium ion 3,5-
free and ester cholesterol differential color development. The cholestadiene, which is the first step in the color reaction.
methods of Zak and Ressler [9], Carr and Drekter [44], Bloor Sequential oxidation of this allylic carbonium ion by SO
[45], and Chiamori and Henry [46] are examples of two-step yields a cholesta-hexenesulfonic acid chromophoric
procedures. compound with absorbance maxima (Amax) of 610 and 410
nm. This chemical reaction for cholesterol determination has
Three-Step Methods undergone many modifications through the years and has
Three-step procedures involve, in addition to extraction of included the measurement of either free or esterified
cholesterol, a saponification step that hydrolyzes the fatty- cholesterol or both. Digitonin, a saponin, reacts with the free
acid moiety from the cholesterol ester. Consequently, one OH (at the C-3 position) group of the A ring of the sterol to
measures only free cholesterol. The CDC-proposed reference form an insoluble complex, cholesterol digitonide. The
method [47] for cholesterol, a modification of the method of measurement of the sterol content in the supernatant solution
Abell et al. [43], belongs in this classification. provides an estimate of esterified cholesterol. To obtain free
cholesterol, one subtracts the esterified value from the total
Four-Step Methods (free plus ester) cholesterol value.
Four-step methods go a step further than the three-step
method of extraction, saponification, and color development. Iron-Salt-Acid Reaction
The total extractable steroids are purified for cholesterol In 1953, Zlatkis and Boyle [18] proposed a new colorimetric
determination by the addition of a saponin, digitonin. The procedure for cholesterol, dependent on the magenta color
reactive site on the C-3 position on the produced when a solution of ferric chloride in concentrated
cyclopentanoperhydrophenanthrene ring in cholesterol is a sulfuric acid is added to a solution of cholesterol in glacial
hydroxyl group that is esterified by the digitonin. This causes acetic acid. The color developed in this reaction is more
the complex to be precipitated. The addition of the digitonin intense and more stable than that developed in the L-B
step also eliminates the effect of interfering nonspecific reaction. This reaction (Table 1, Method 3) involves acetic
chromogenic constituents. Empirically, four-step procedures acidsulfuric acid in the absence of acetic anhydride. In this
should be more accurate and precise than any other
reaction, however, Fe3+ must be added to obtain the desired
procedure, with the possible exception of enzymatic
chromogen. As in the L-B procedure, the initial step is the
cholesterol determinations. However, unless extreme
protonation of the OH group in the cholesterol molecule and
precautions are taken, multiple steps may also mean multiple
subsequent loss of water to form the carbonium ion (3,5-
errors. For instance, whereas digitonin precipitation
cholestadiene). Serial oxidation of this allylic carbonium ion
enhances the accuracy of the method, the digitonin must be
completely decomposed and removed or it will cause by Fe3+ yields a tetraenylic cation with an absorbance
additional color development and thus positive error. maximum of 563 nm. The iron-salt-acid procedures are
about sevenfold more sensitive than the L-B methods. This
The Schoenheimer and Sperry [48] and Sperry and Webb increased sensitivity may be attributed largely to the
[49] methods are four-step procedures for cholesterol stabilizing effect on enylic cation formation at higher H2SO4
determinations. These two methods also employ concentrations. In general, increasing the H2SO4
Liebermann-Burchard reagents. Although the Abell et al. concentration would be expected to improve the stability of
[43] method may be more tailored to serum or plasma each of the carbonium ions formed in the stepwise oxidation
cholesterol analysis, four-step methods may be more suitable of the sterol, thereby making it much more likely for one to
for tissue extracts [11], since certain tissues may have observe carbonium-ion formation in the iron-salt-acid
appreciable noncholesterol and cholesterol esters that may reaction than in the L-B reaction.
contribute to the inaccuracy of the final development of
color. Measurements of serum cholesterol by the FeCl3-H2SO4
reaction were automated by Levine and Zak [50] and by
Methods of Analysis
Literally hundreds of cholesterol methods have been Block and co-workers [51].
published, usually as modifications of the following
In 1969, a micromethod for serum cholesterol, which
reactions: (1) Liebermann-Burchard, (2) iron-salt-acid, (3) p-
toluenesulfonic acid, or (4) enzymatic end-point. requires only 20 L of serum, was described by Jordan and
Knoblock [52]. In this technique, cholesterol is precipitated
Liebermann-Burchard Reaction as a dextran sulfate complex to avoid interference from
Among the nonenzymatic colorimetric reactions for bilirubin. The precipitate is dissolved in glacial acetic acid,
cholesterol, the Liebermann-Burchard (L-B) procedure and ferric chloride reagent is then added. Even greater
(Table 1, Method 1) is perhaps the most widely used. The L- sensitivity can be achieved by fluorometric detection of the
384
Cholesterol
FeCl3-H2SO4 chromogen [53]. An automated fluorometric In this reaction, cholesterol concentration can be determined
technique for serum cholesterol was described by Robertson by amperometric measurement of the rate of oxygen
and Cramp [54]. depletion.
p-Toluenesulfonic Acid Reactions Other assays make use of the ability of H2O2 to oxidize
These methods are based on reactions of cholesterol with p- compounds to produce colored species that can be measured
toluenesulfonic acid (p-TSA), acetic anhydride, glacial acetic spectrophotometrically:
acid, and H2SO4 (Table 1, Method 4).
_
2H2O2 + phenol + 4-aminophenazone peroxidase
Enzymatic End-Point Reactions quinone imine dye + 4H2O 3a
Developments over the past 30 years have been toward more
rapid, direct analyses of cholesterol in serum or whole blood.
_
Pretreatment of samples increases the possibility for error Acetaldehyde + NADP+ aldehyde dehydrogenase
and increases turnaround time. These demands have led to acetate + NADPH + H+ 3b
the most important development in cholesterol or
measurements, which is the introduction of enzymatic _
techniques. H2O2 + methanol catalase formaldehyde + H2O 4
Enzymatic techniques (Table 1, Method 5) for determining Formaldehyde + acetylacetone 3,5-diacetyl-1,4-dihydrolutidine 5

cholesterol have emerged to compete with the classical L-B or
reaction and have become the most popular method for _
H2O2 + ethanol catalase acetaldehyde + H2O 6
cholesterol analysis. The original work used preliminary
alkaline saponification of the sample to produce only free
peroxidase
cholesterol [9,38,44]. In the next step, cholesterol oxidase, H2O2 + homovanillic acid
an enzyme almost entirely specific for cholesterol, was 2,2-dihydroxy-3,3-dimethyloxy-biphenyl-5,5-diacetic acid 7
added. This caused the breakdown of cholesterol to cholest-
4-en-3-one and hydrogen peroxide, after which several or
different reaction systems have been used to produce a final _peroxidase
H2O2 + o-dianisidine chromophore + 2H2O 8
chromogen.
Subsequent developments led to the technique of enzymatic Reaction 3a (Trinders reaction), which forms the quinone
hydrolysis of cholesteryl esters by employing cholesterol imine dye (absorbance maximum 500 to 525 nm), is the
esterase. The commercial kits on the market now offer a total basis of the majority of the methods currently on the market
enzymatic procedure, utilizing the enzyme cholesterol [21,24,27,37,39,55]. Other indicator systems that have been
esterase to replace the chemical saponification. Cholesterol used in equation 3 include 4-aminophenazone plus 2-
esterase is specific for cholesteryl esters, splitting the esters hydroxy-3,5-dichlorobenzenesulfonic acid and 3,3,5,5-
into free cholesterol and free fatty acids. The cholesterol tetramethylbenzidine.
oxidase reaction follows, with the same enzyme reaction as
described above. The amount of color produced is directly The action of catalase in conjunction with methanol and
proportional to the amount of serum cholesterol. Because of acetylacetone to produce dihydrolutidine (reaction 4 and
the increased use of enzymatic methods, a more extensive reaction 5) is measured at 405 nm. Reaction 5, monitored at
discussion is provided here. 405 nm [22,30,35], has the advantage of not being subject to
bilirubin interference and has been developed as a rate
The first chemical step in the enzymatic methods for procedure to improve on the original time-consuming end-
cholesterol uses the enzyme cholesterol esterase to hydrolyze point assay.
the cholesteryl esters present in the serum to free cholesterol
and free fatty acids: Reaction 3b is monitored at 340 nm as an end-point reaction
[56].
_
Cholesteryl esters + H2O cholesterol esterase Additional methods that measure cholesterol by analysis of
cholesterol + free fatty acids 1 the H2O2 produced by
cholesterol oxidase include:
As discussed previously, the second step uses the enzyme (a) Polarography [38,57] employing electrodes specific
cholesterol oxidase in the presence of oxygen to oxidize the for hydrogen peroxide
cholesterol (both the free cholesterol found in the serum and (b) Action of peroxidase to oxidize homovanillic acid
the free cholesterol generated in step 1) to cholest-4-en-3- to a fluorescent compound measured at 470 nm [33]
one and H2O2: (reaction 7)
_
Cholesterol + O2 cholesterol oxidase (c) the condensation with o-dianisidine in the presence
of peroxidase to produce a colored compound
cholest-4-en-3-one + H2O2 2 which is measured at 450 nm [28,58] (reaction 8)
385
Cholesterol
According to a 2007 CAP Chemistry Participant Summary, of the cholesteryl esters, extraction with petroleum ether, and
99% of the participants in the Chemistry Proficiency Survey color development with acetic acidacetic anhydride
used an enzymatic method to measure total cholesterol [59]. H2SO4 reagent. With proper equipment and good technical
skills, this procedure can be used to produce accurate,
High-Performance Liquid Chromatography precise results [68].
For many years, high-performance liquid chromatography
(HPLC) has been studied as a technique for the measurement In practice, however, although the CDC method could be
of cholesterol in serum, particularly in the research field. used in most laboratories for calibrating their methods, it is
One hindrance, however, has been that cholesterol with only time consuming and requires particular expertise. Similarly,
a single double bond is a poor absorber of ultraviolet light, ID/GC/MS techniques are not practical for routine
making detection difficult. Rather than trying to perform laboratories. In reality, only the enzymatic techniques
these measurements at < 210 nm, one suggestion to capable of being automated are suitable for use in most
overcome this problem is to produce benzoyl derivatives of working clinical chemistry laboratories.
the cholesterol, allowing detection at 228 nm possible
[60,61]. Liebermann-Burchard Methods
The L-B reaction has been one of the most used cholesterol
It has also been reported that the 4-3-cholestanone produced chemistries. However, its wide usage does not automatically
after treatment with cholesterol oxidase is detectable at 241 ensure reliable cholesterol analysis. Much depends on
nm [62]. whether cholesterol (more specifically cholesteryl ester) is
hydrolyzed and purified before analysis. For example, 7-
Other work has looked at the derivatization of cholesterol dehydrocholesterol and 7--cholesterol give about two and
with 2-[2-(isocyanate)ethyl]-3-methyl-1,4-naphthoquinone, four times more color, respectively, with the L-B reaction
or at the use of fluorescent detection following HPLC. Such than cholesterol itself does. Also, the color intensity
techniques do, however, take additional time, and each produced by cholesteryl esters is greater than that produced
derivatization step may introduce further error into the by free cholesterol. Thus the total cholesterol is
results. At this juncture, these sensitive methods, although of overestimated unless saponification is carried out first, as in
use for research purposes, are not suitable for the routine the methods of Schoenheimer and Sperry [48], Sperry and
clinical laboratory, where automation is required [63]. Webb [49], and Abell et al. [43]. Because of these biases,
any procedure that uses an L-B type of reagent and does not
Reference and Preferred Methods include saponification and purification (for example, with
Definitive Methods digitonin) steps will give falsely elevated concentrations for
Definitive techniques permit measurement of the total cholesterol when pure cholesterol is used as a standard.
concentration of a substance in a biological sample and The Bloor method [45] is an example of a method yielding
give results accepted as the nearest attainable to true results with a large positive bias. Microgram quantities of
values. The Joint Committee for Traceability in vitamin A (retinol) give large positive errors by this method.
Laboratory Medicine (JCTLM) [64] recognizes two
isotopic dilution/gas chromatographymass spectrometry Enzymatic End-Point Methods
(ID/GC-MS) techniques as definitive methods. They are The enzymatic procedures are direct methods amenable to
the methods of (a) Deutsche Gesellschaft fur Klinische automating. The primary advantages of the enzymatic
Chemie (DGKC) and (b) National Institute of Standards methods are that one avoids the use of the harsh reagents
and Technology (NIST)previously NBS. Such required by the chemical determinations, and the
techniques have coefficients of variation (CV) no greater interferences of other blood constituents are greatly reduced
than 0.5% and a total uncertainty of no greater than 1% because of the specificity of the enzymes. A few cholesterol
(Cohen et al. [65]). Schaffer et al. [66] compared the analogs may interfere, but the magnitude of the interference
Karolinska Institute and NIST ID-MS methods for seen in the enzymatic assays is much less than the
cholesterol and found a 0.2% mean difference between interference seen with chemical measurements or is greatly
the two methods. The Karolinska Institute standard was reduced. The option to determine the free and esterified
found to contain lathosterol (5--cholest-7-en-3-ol-3[]). cholesterol is available with these methods.
The NBS method appeared to be more precise, probably
because of the more complex and time-consuming A few precautions are necessary when one is using the
protocol for sample preparation and MS. For more enzymatic kits that have quinone imine dye as the
information on definitive cholesterol methods, consult chromogen. Blanking can cause problems with turbid
Cohen et al. [65] and Wolthers et al. [67] samples for manual procedures. For automated procedures,
all-glass tubing must be used after the point of the formation
The only method currently classified by JCTLM as a of the quinone imine dye complex, since Tygon tubing will
Reference Method is the University of Ghents ID/GC/MS absorb the dye, slowly releasing it in subsequent samples and
method [64]. causing carryover problems.
The Centers for Disease Control and Prevention (CDC, Some variation in results of different methods may be caused
Atlanta, GA) has proposed a modification of the procedure by measurement at different wavelengths. For example, a
of Abell et al. [24,27,29,30,34,39,43] for use as a reference range of 460 to 560 nm has been proposed for the Trinder-
method for cholesterol [47]. This method includes hydrolysis based reaction (reaction 3), with 500 nm being the maximum
386
Cholesterol
absorbance of the dye. Commonly suggested wavelengths to 180 mg/dL (1.6 to 2.0 mmol/L)not enough to cause
for taking measurements are 505, 520, and 525 nm. interference problems with HDL cholesterol measurements
(i.e., incomplete precipitation of LDL, VLDL, and
The problem of incomplete ester hydrolysis and thus of chylomicrons). However, with markedly
standardization is the primary outstanding problem with hypertriglyceridemic samples, that is, greater than 1,000
enzymatic methods for measuring cholesterol. This mg/dL (11.3 mmol/L), interference can occur with
incompleteness of hydrolysis may be generally true of all colorimetric assays. Ethanol consumption causes acute but
cholesteryl esters, but one study indicates that it may be the transient elevations in serum triglyceride concentrations,
result of the specificity of the esterase [62]. For example, of which are evident in carbohydrate-sensitive
the various commercially available enzymatic cholesterol hypertriglyceridemic persons. Therefore, it would be
reagents with aqueous preparations of cholesteryl esters, advisable to request that the patient refrain from drinking
those that have arachidonic acid or acetate as esters appear to alcohol for 72 hours before the day of blood drawing.
demonstrate limited reactivity. The species source of the
cholesterol esterase may also affect the specificity of the Blood-Drawing Techniques
reaction. Siedel et al. [31,32] and Deeg and Ziegenhorn [69] It is well recognized that plasma volume increases and the
reported a new commercial enzymatic preparation for concentrations of nondiffusible plasma components decrease
cholesterol with an esterase from a microbial source. It when a standing subject assumes a recumbent position, a
reportedly had improved hydrolytic activity, virtually result of redistribution of water between the vascular and
completely hydrolyzing the cholesteryl esters (99.5%), as extravascular compartments. A significant reduction in total
verified by thin-layer chromatography. Siedel et al. [31,32] plasma cholesterol has been measured after 5 min, and
found that the new reagent (Monotest Cholesterol, High decreases of as much as 10% to 15% have been recorded 20
Performance, BMD, Inc., Houston, TX), with improved min after a recumbent position is assumed. The effect on
lipolytic efficiency, yielded results that were identical to cholesterol concentration when the subject changes from a
those measured with a candidate reference procedure standing position to a sitting position is also significant
involving alkaline cholesterol ester saponification. though somewhat smallerabout 6% after 10 to 20 min.
The esterases may be the source of much of the difference If a tourniquet is used, it should be removed before blood
seen between results obtained by various enzymatic methods sampling, since prolonged use of a tourniquet has been
and results obtained by reference procedures, with the reported to increase lipid values. Serum cholesterol
magnitude of the bias ranging from 4% to 6% of the total concentrations were found to increase an average of 10% to
cholesterol content [60-63,70,71]. To help circumvent this 15% after 5 min of occlusion. From a practical standpoint,
analytical problem, secondary serum calibrators are required errors of this magnitude would not normally be encountered,
for the assays to compensate for the incomplete cholesterol since the tourniquet is usually removed within 30 to 60 sec,
esterase reaction. and the changes occurring during this period are
insignificant. Increases of 2% to 5% have been observed
A properly calibrated enzymatic method is certainly the after about 2 min; these increases may be seen if difficulties
preferred method for routine laboratory analysis. Enzymatic arise during sampling or if blood is taken by an
methods can be as accurate and precise as the Abell-Kendall inexperienced technician.
reference method. Since these methods are readily available
as kits, this technique is not presented in detail. A reference Choice of Plasma or Serum
L-B method is presented, since this may be used for the If phlebotomy is properly performed, either plasma or serum
initial calibration of an enzymatic method. is usually suitable for total cholesterol, triglyceride, or
phospholipid determinations. Because the plasma and serum
Specimen cholesterol values differ by 3% to 5%, the NCEP Laboratory
Patient Preparation Standardization Panel recommended that if plasma is used,
Determinations of lipid constituents in plasma or serum are the cholesterol values should be multiplied by 1.03 to
usually performed on blood drawn from patients fasting for convert the values equivalent to serum values [72]. Plasma is
12 to 16 hours. The fasting sample is essential for usually preferred when lipids and lipoproteins are being
triglyceride analysis, since triglyceride concentrations chemically analyzed. If plasma is chosen, the suggested
increase as soon as 2 hours postprandially and reach a anticoagulant is solid EDTA (1 mg/mL of blood), and the
maximum at 4 to 6 hours. Nonfasting samples are not blood cells should be separated as soon as possible (within 2
suitable for analysis, since elevated results caused by normal hours).
assimilation of food cannot be distinguished from elevated
results resulting from abnormal lipid metabolism or inborn Certain anticoagulants, such as fluoride, citrate, and oxalate,
errors of metabolism. Like total cholesterol, HDL cholesterol cause rather large shifts of water from the red blood cells to
concentrations change very little between fasting and the plasma, which result in the dilution of plasma
nonfasting conditions. In subjects in whom components. Heparin, on the other hand, causes no
hypertriglyceridemia is induced by nonfasting conditions, detectable change in red-blood-cell volume and decreases
interference with some HDL cholesterol assays becomes an total cholesterol concentration by only 1% or less. Although
issue, causing a positive bias. In most healthy individuals, EDTA causes slightly greater changes than heparin does in
the 12-hour fasting serum triglyceride concentration is about cholesterol concentration, it is generally preferred for
80 to 100 mg/dL (0.9 to 1.1 mmol/L) and nonfasting is 140 lipoprotein analysis for several reasons. EDTA retards the
387
Cholesterol
auto-oxidation of unsaturated fatty acids and cholesterol by triglycerides to free glycerol and fatty acids takes place and
chelating heavy metal ions such as Cu2+. Oxidation leads to effectively reduces the true triglyceride concentration. In
alterations in the physical properties of the lipoproteins, addition, the sharpness of the lipoprotein bands decreases
gross denaturation, and apparent degradation of lipoproteins. with time when electrophoresis is performed. Therefore, if
In addition, EDTA can reduce changes that can occur as a more than short-term storage is anticipated, serum and
result of contamination by phospholipase Cproducing plasma samples (but not whole blood) should be frozen to
bacteria. Such contamination is associated with spurious prevent the spontaneous hydrolysis and oxidation of the
increases in total glyceride concentration that are caused by unsaturated lipids, which occurs during storage, especially at
the production of partial glycerides from some warmer temperatures.
phospholipids. It is also associated with the artifactual
appearance of chylomicron-like particles and decreases in - Stored serum or plasma samples must be adequately mixed
, -, and pre-lipoproteins. These changes are minimized by before use. Lipids will layer by density within the sample
the addition of EDTA, which is an antioxidant and an when stored and give different results, depending on the
inhibitor of phospholipase C activity. layer from which an aliquot is obtained. Simple inversion of
the tube that contains the serum or plasma is not sufficient,
It is good practice to perform the lipid analysis as soon as particularly if the sample was frozen. Instead, the sample
possible. If analysis is substantially delayed after the sample should be carefully mixed on a vortex mixer for 5 to 10 sec.
is drawn, alterations in lipoprotein composition can result Particular attention should be given to those samples that
from the exchange of cholesterol esters and triglycerides have appreciable amounts of standing chylomicrons. One
between HDL and the other lipoproteins. Several of these can obtain accurate lipid results only when the chylomicrons
processes can occur simultaneously and influence results. are dispersed evenly through the sample before taking an
Their cumulative effects can either be additive (they may aliquot for lipid analysis. If pronounced chylomicronemia
increase the error of the analysis) or compensatory (the exists, the sample should be diluted with saline solution so
processes may have opposite effects on the results that positive-displacement errors do not occur with the
minimizing the error), depending on the exact procedure. excessive macromolecules.
Storage of Samples If an organic solvent extraction method is used, the extract

Two of the common variables of sample handling that can be stored at 4C for short periods (a few days) if one is
influence measured lipid and lipoprotein values are the concerned only with total cholesterol, triglyceride, or
length of time and the conditions under which samples are phospholipid analyses. On the other hand, if one is interested
stored before analysis. It is generally recommended that in fatty acid profiles, the extract should be frozen
plasma be stored in the liquid state when it is to be used for immediately to inhibit auto-oxidation. Care must be taken
lipid and, particularly, lipoprotein analysis or for lipoprotein when one is storing the purified lipid extracts. Cork or
electrophoresis studies. In addition, the determinations rubber stoppers or caps with paper, cork, or rubber liners
should be performed promptly. In practice, however, delays should be avoided when one is storing lipid extracts in
can occur for a variety of reasons, and it may even be organic solvents. It is recommended that tubes specifically
necessary to analyze frozen samples. Several groups of designed for low-temperature storage should be used. Screw-
workers have measured lipoprotein levels in samples that capped tubes with Teflon-lined caps are good storage vials if
have been stored at different temperatures for varying a good seal is formed. Evaporation of the solvent can be a
periods of time, frozen and unfrozen [15,73]. The problem if care is not taken to ensure the selection of a good
measurements were made with the analytical ultracentrifuge, storage container. A well-sealed container or vial also helps
and the results generally indicated that the levels of all to keep oxygen out of the sample and prevent oxidation of
lipoproteins may decrease with storage. No definitive studies the unsaturated lipids. If the samples are to be stored in this
have been made determining the optimum freezing form for a long period, it is best to overlay the sample extract
temperature and the length of storage permissible at this with nitrogen before the vial is sealed.
temperature. Cooper [74] reported that when specimens are
stored in the liquid state at 4C for longer than 1 week, high- Interferences
density lipoprotein (HDL) cholesterol values, reproducibility Because enzymatic techniques are the only ones in routine
of the test, and resolution of lipoprotein patterns decrease. use, only interferences relating to these will be discussed.
The HDL cholesterol values were found to be more stable
when the specimens were stored at 60C than when the Only a few interferences have been reported for the
specimens were stored at 20C or 5C over 10 months. At enzymatic method. Haeckel and Perlick [75] reported that
present, the general consensus is that freezing at 60C the coupled reactions of cholesterol esterase, cholesterol
provides the longest stable storage and may allow for oxidase, catalase, and aldehyde dehydrogenase (reaction 6
reproducible results even after a year or more. Freezing at and reaction 3b) are not inhibited by bilirubin. In contrast,
20C will keep samples stable for a few months if self- enzymatic analyses of serum cholesterol, by action of
defrosting freezer units are avoided, since samples stored in peroxidase with 4-aminoantipyrine and phenol, are inhibited
these units periodically undergo thawing and freezing, by bilirubin [39,40]. Miner-Williams [76] has reported that
causing a breakdown of the lipid and lipoprotein certain surfactants can cause inhibition of cholesterol
components. If the samples are to be used in a few days, oxidase activity.
refrigeration at 4C is sufficient. However, even at
refrigerated temperatures, spontaneous hydrolysis of the
388
Cholesterol
Although ascorbic acid (vitamin C) is known to interfere CAD risk factors and the concentrations of both LDL
with the Trinder reaction, in practice its effect in cholesterol cholesterol and HDL cholesterol. Current recommendations
assays is not great. Martinello et al. [77] have shown that for treatment approaches based on risk of CAD and LDL
oral doses as high as 4 g do not interfere significantly, owing cholesterol levels are shown in Table 4 [80].
to the serum concentrations not reaching high enough levels
and also to the fact that by the time the sample is normally Cholesterol Performance Goals
assayed, the ascorbic acid has been oxidized. Many The current Clinical Laboratory Improvement Amendments
cholesterol assays are performed on fasting samples, making (CLIA) target for acceptable performance for cholesterol
it less likely that vitamin C has been consumed shortly measurements is 10% of the peer-group mean. Precision
before the sample is collected and assayed. However, mega- goals based on the work of Fraser et al. [81] indicate that a
doses (e.g., 15 g) or large intravenous dosages (e.g., 2 g IV) precision goal of a maximum standard deviation of 5.4
may interfere, especially if the assay is performed soon after mg/dL at a cholesterol concentration of 200 mg/dL is
collection. This interference could be a potential problem appropriate (a CV of 2.7%). The intraindividual variation in
with point-of-care assays. cholesterol in healthy adults has been found to range from
approximately 4% to 8% over a 5-month period [82].
Cholesterol Reference Interval
The 95th-percentile limit for normal total cholesterol values, The NCEP Laboratory Standardization Panel recommended
once advocated by Fredrickson et al. [78] for lipoprotein laboratory standards for precision and accuracy of total
phenotyping, is shown in Table 2. cholesterol measurements in the laboratory [72]. For the
interim goal for precision and accuracy, they recommended
Although it is customary to use normal limits or reference 5.0% CV or less and an average overall bias not to exceed
intervals to determine abnormal values for most blood 5.0% from the reference method values (which is
constituents, the determination of abnormal cholesterol equivalent to 9.5% on a single measurement). They also
levels is an exception to the rule. Prior to 1988, most clinical recommended that precision and accuracy should decrease to
laboratories and practicing physicians used the 95th- 3.0% CV or less and 3.0% or less, respectively. The 2007
percentile reference intervals when evaluating whether a CAP Chemistry Participant Summary [59] precision data
sample had normal or abnormal cholesterol values. These indicates that participants perform well against these criteria,
nomograms are based on sampling apparently normal with all method CVs across the CAP range of values (106
persons in the U.S. population and arbitrarily defining to 268 mg/dL) being 2.8% to 3.1%. Some users of enzymatic
hyperlipidemia as being present when the plasma cholesterol techniques are achieving CVs < 2% across that range.
or triglycerides, or both, are above the 95th-percentile value
for the population to which the persons belong. For example, References
a 40-year-old man would have been considered normal 1 Naito HK, Hoff HF. Nutrition and pathogenesis of
with a total cholesterol of 260 mg/dL (6.20 mmol/L). the blood vessel. In: Blend J, ed. Medical
Unfortunately, because of the way we have defined normal Applications of Clinical Nutrition. New Canaan,
in the past, many test results do not correlate very well with CT: Keats Publishing; 1983:178-221.
disease states or health-risk conditions on an individual 2 Liebermann C. Ueber das oxychinoterpen. Dtsch
basis. Thus, the normal range may not be a healthy level. Chem Geselsch 1885;18:1803-1809.
3 Burchard H. Beitrge zur kenntnis des cholesterins.
Interpretation Chem Zentralbl 1890;61:25-27.
There is a well-established curvilinear relationship between 4 Zak B. Cholesterol methodologies: a review. Clin
cholesterol and risk of atherosclerotic CAD. CAD risk is Chem 1977;23:1201-1214.
approximately 2% to 3% higher for each 1% increase in total 5 Zak B, Epstein E. New cholesterol reagent. Clin
cholesterol concentration. However, most patients with CAD Chem 1961;7:268-270.
have total cholesterol concentrations that are average for 6 Zak B, Weinger LM, Walsh B. Spectrophotometric
their society. The measurement of total cholesterol includes study of bilirubin interference in the Huang reaction
both atherogenic particles (mainly LDL) and for cholesterol. Clin Chim Acta 1970;30:697-705.
cardioprotective particles (HDL) which are independently 7 Zak B, Dickenman RC, White EG, Burnett H,
related to CAD risk. The association of LDL cholesterol is Cherney PJ. Rapid estimation of free and total
therefore stronger than that for total cholesterol and is further cholesterol. Am J Clin Pathol 1954;24:1307-1315.
marked when accompanied by triglyceride elevation. 8 Zak B, Epstein E, Baginski ES. Review and critique
Furthermore, a ratio of total cholesterol to HDL cholesterol of cholesterol methodology. Ann Clin Lab Sci
is considered a better indicator of CHD risk than total 1972;2:101-125.
cholesterol alone. In epidemiological studies, the risk has 9 Zak B, Ressler N. Methodology in determination of
been shown to increase exponentially when the ratio is > 4.5. cholesterol. Am J Clin Pathol 1955;25:433-446.
10 Tonks DB. The estimation of cholesterol in serum: a
Screening for dyslipidemia is indicated for all adults up to 75 classification and critical review of methods. Clin
years of age, regardless of CAD risk status, and for adults Biochem 1967;1:12-29.
older than 75 years of age who have multiple CAD risk 11 Naito HK, David JA. Laboratory considerations:
factors [79]. The current guidelines for classification of determination of cholesterol, triglyceride,
serum lipids are shown in Table 3. Treatment approaches for phospholipid, and other lipids in blood and tissues.
patients with dyslipidemias are based upon the number of
389
Cholesterol
In: Story JA, ed. Lipid Research Methodology. New 30 Ziegenhorn J. Enzymatische bestimmung des
York: Alan R. Liss; 1984:1-76. gesamt-cholesterins im serum mit
12 Naito HK. In: Kaplan LA, Pesce AJ, eds. Clinical Analysenautomaten. Z Klin Chem Klin Biochem
Chemistry: Theory, Analysis and Correlation. St. 1975;13:109-115.
Louis: Mosby;1984: 550-593, 918-926. 31 Siedel J, Schlumberger H, Klose S, Ziegenhorn J,
13 Naito HK. Cholesterol: review of methods. Check Wahlefeld AW. Improved reagent for the enzymatic
Sample PTS 85-1. Chicago: American Society of determination of serum cholesterol. J Clin Chem
Clinical Pathology; 1985:1-17. Clin Biochem 1981;19:838-839.
14 Kritchevsky D. Cholesterol. New York: John Wiley 32 Siedel J, Hgele EO, Ziegenhorn J, Wahlefeld AW.
& Sons;1985:232-55. Reagent for the enzymatic determination of serum
15 Martnek RG. Review of methods for determining cholesterol with improved lipolytic efficiency. Clin
cholesterol and cholesterol esters in serum. J Am Chem 1983;29:1075-1080.
Med Technol 1970;32:64-86. 33 Huang HS, Kuan JC, Guilbault GG. Fluorometric
16 Werner M, Williams D, Skendzel LP. Operations enzymatic determination of total cholesterol in
characteristics of analytical methods: comments and serum. Clin Chem 1975;21:1605-1608.
application to cholesterol assay. Check Sample ST- 34 Huang HS, Kuan SS, Guilbault GG. Amperometric
104. Chicago: American Society of Clinical determination of total cholesterol in serum with use
Pathologists; 1978. of immobilized cholesterol ester hydrolase and
17 Naito HK. Reliability of lipid and lipoprotein cholesterol oxidase. Clin Chem 1977;23:671-676.
testing. Am J Cardiol 1985;56(suppl.):65-95. 35 Knob M, Rosenmund H. Enzymatische bestimmung
18 Zlatkis A, Zak B, Boyle AJ. A new method for the des gesamtcholesterins im serum mit Zen-
direct determination of serum cholesterol. J Lab trifugalanalyzern. Z Klin Chem 1975;13:493-498.
Clin Med 1953;41:486-492. 36 Lie RF, Schmitz JM, Pierre KJ. Cholesterol
19 Huang TC, Wefler V, Raferty A. A simplified oxidase-based determination, by continuous-flow
spectrophotometric method for determination of analysis, of total and free cholesterol in serum. Clin
total and esterified cholesterol with tomatine. Anal Chem 1976;22:1627-1630.
Chem 1963;35:1757-1758. 37 Witte DL, Barrett DA, Wycoff DA. Evaluation of
20 Pearson S, Stern S, McGavack TH. A rapid method an enzymatic procedure for determination of serum
for the determination of total cholesterol in serum. cholesterol with the Abbott ABA-100. Clin Chem
Anal Chem 1953;25:813-814. 1974;20:1282-1286.
21 Allain CC, Poon LS, Chan CS, Richmond W, Fu 38 Noma A, Nakayama K. Polarographic method for
PC. Enzymatic determination of total serum rapid micro-determination of cholesterol with
cholesterol. Clin Chem 1974;20:470-475. cholesterol esterase and cholesterol oxidase. Clin
22 Pesce MA, Bodourian SH. Enzymatic rate method Chem 1976;22:336-340.
for measuring cholesterol in serum. Clin Chem 39 Pesce MA, Bodourian SH. Enzymatic measurement
1976;22:2042-2045. of cholesterol in serum with the CentrifiChem
23 Kumar A, Christian GD. Enzymatic assay of total centrifugal analyzer. Clin Chem 1977;23:280-282.
cholesterol in serum or plasma by amperometric 40 Pesce MA, Bodourian SH. Interference with the
measurement of rate of oxygen depletion following enzymic measurement of cholesterol in serum by
saponification. Clin Chim Acta 1977;74:101-108. use of five reagent kits. Clin Chem 1977;23:757-
24 Borner K, Klose S. Enzymatische bestimmung des 760.
gesamtcholesterins mit dem Greiner Selective 41 Richmond W. Preparation and properties of a
Analyzer (GSA-11). J Clin Chem Clin Biochem cholesterol oxidase from Nocardia sp. and its
1977;15:121-130. application to the enzymatic assay of total
25 Flegg HM. An investigation of the determination of cholesterol in serum. Clin Chem 1973;19:1350-
serum cholesterol by an enzymatic method. Ann 1356.
Clin Biochem 1973;10:79-84. 42 Richmond W. Use of cholesterol oxidase for assay
26 Fruchart JC, Tsumbu BB, Dewailly P, Sezille G, of total and free cholesterol in serum by continuous-
Jaillard J. Automation of fluorometric enzymatic flow analysis. Clin Chem 1976;22:1579-1588.
determination of free and total cholesterol: 43 Abell LL, Levy BB, Brodie BB, Kendall FE.
application to the determination of the esterification Simplified methods for the estimation of the total
rate. Ann Biol Clin (Paris) 1975;33:453-457. cholesterol in serum and demonstration of its
27 Heuck CC, Wirth A. Simultane bestimmung von specificity. J Biol Chem 1952;195:357-366.
freiem cholesterin und gesamtcholesterin auf dem 44 Carr JJ, Drekter IJ. Simplified rapid technic for the
Technicon AA1. J Clin Chem Clin Biochem extraction and determination of serum cholesterol
1977;15:213-216. without saponification. Clin Chem 1956;2:353-368.
28 Tarbutton PN, Gunter CR. Enzymatic determination 45 Bloor WR. The determination of small amounts of
of total cholesterol in serum. Clin Chem lipid in blood plasma. J Biol Chem 1928;77:53-73.
1974;20:724-725. 46 Chiamori N, Henry RJ. Study of the ferric chloride
29 Trocha PJ. Improved continuous flow enzymatic method for the determination of total cholesterol
determination of total serum cholesterol and and cholesterol esters. Am J Clin Pathol
triglycerides. Clin Chem 1977;23:146-147. 1959;31:305-309.
390
Cholesterol
47 Cooper GR, Gill JB, Biegeleisen JJ, Miller DT, detection of oxidised cholesterols in HPLC
Bayse DD. CDC point of reference for total separations. Anal Chim Acta 1980;117:71-79.
cholesterol measurements [abstract]. Clin Chem 63 Nakajima M, Yamato S, Wakabayashi H, Shimada
1980;26:966-967. K. High-performance liquid chromatographic
48 Schoenheimer R, Sperry WM. A micromethod for determination of cholesterol and cholestanol in
the determination of free and combined cholesterol. human serum by precolumn derivatization with 2-
J Biol Chem 1934;106:745-760. [2-(isocyanate)ethyl]-3-methyl-1,4-naphthoquinone
49 Sperry WM, Webb M. A revision of the combined with platinum catalyst reduction and
Schoenheimer and Sperry method for cholesterol electrochemical detection. Biol Pharm Bull
determination. J Biol Chem 1950;187:97-106. 1995;18:1762-1764.
50 Levine JB, Zak B. Automated determination of 64 Joint Committee for Traceability in Laboratory
serum total cholesterol. Clin Chim Acta Medicine-DB website. Available from:
1964;10:381-384. http://www.bipm.org/jctlm/
51 Block WD, Jarett KJ Jr., Levine JB. An improved 65 Cohen L, Jones RL, Batra KV. Determination of
automated determination of serum total cholesterol total ester and free cholesterol in serum and serum
with a single color reagent. Clin Chem lipoproteins. Clin Chim Acta 1961;6:613-619.
1966;12:681-689. 66 Schaffer R, Sniegoski LT, Welch MJ, White VE,
52 Jordan WJ Jr., Knoblock EC. A micromethod for Cohen A, Hertz HS et al. Comparison of two
total cholesterol eliminating the effect of high isotope dilution/mass spectrometric methods for
bilirubin levels. Clin Chem 1969;15:807-808. determination of total serum cholesterol. Clin Chem
53 Solow EB, Freeman LW. A fluorometric ferric 1982;28:5-8.
chloride method for determining cholesterol in 67 Wolthers BG, Heindriks FR, Muskiet HJ, Groen A.
cerebrospinal fluid and serum. Clin Chem Mass fragmentographic analysis of total cholesterol
1970;16:472-476. in serum using heptadeuterated internal standard.
54 Robertson G, Cramp DG. An evaluation of Clin Chim Acta 1980;103:305-315.
cholesterol determination in serum and serum 68 Duncan IW, Mather A, Cooper GR. The Procedure
lipoprotein fractions by a semiautomated for the Proposed Cholesterol Reference Method.
fluorimetric method. J Clin Pathol 1970;23:243- Atlanta, GA: Centers for Disease Control; 1982.
245. 69 Deeg JR, Ziegenhorn J. Kinetic enzymic method for
55 Nobbs BT, Smith JM, Wallar AW. Enzymic automated determination of total cholesterol in
determination of plasma cholesterol on discrete serum. Clin Chem 1983;29:1798-1802.
automatic analyzers. Clin Chim Acta 1977;79:391- 70 Cooper GR, Ullman MD, Hazelhurst J, Miller DT,
398. Bayse DD. Chemical characteristics and sources of
56 Carlson SE, Goldfarb S. A sensitive enzymatic error of an enzymatic cholesterol method. Clin
method for the determination of free and esterified Chem 1979;24:1074.
tissue cholesterol. Clin Chim Acta 1977;79:575- 71 Wiebe DA, Bernert T. Influence of incomplete
582. cholesterol ester hydrolysis on enzymic
57 Yao T, Sato M, Kobayashi Y, Wasa T. measurements of cholesterol. Clin Chem
Amperometric assays of total and free cholesterols 1984;30:352-356.
in serum by the combined use of immobilized 72 U.S. Department of Health & Human Services,
cholesterol esterase and cholesterol oxidase reactors Public Health Service, National Institutes of Health.
and peroxidase electrode in a flow injection system. Recommendations for Improving Cholesterol
Anal Biochem 1985;149:387-391. Measurement: A Report from the Laboratory
58 Sale FO, Marchesini S, Fishman PH, Berra B. A Standardization Panel of the National Cholesterol
sensitive enzymatic assay for determination of Education Program. NIH Publication No. 90-2964.
cholesterol in lipid extracts. Anal Biochem Bethesda, MD: U.S. DHHS, PHS, NIH; 1990.
1984;142:347-350. 73 Manual of Laboratory Operations, Lipid Research
59 C-A Chemistry/Therapeutic Drug Monitoring Clinics Program. Vol. 1, Lipid and lipoprotein
Participant Summary, Surveys 2007. Washington, analysis. DHEW Publ. No. NIH 75-628,
DC: College of American Pathologists; 2007. Washington, DC: U.S. Government Printing Office;
60 Hojo K, Hakamata H, Ito A, Kotani A, Furukawa C, 1974 May.
Hosokawa YY et al. Determination of total 74 Cooper GR. High density lipoprotein reference
cholesterol in serum by high-performance liquid materials. In: Lipel K, ed. Report of the High
chromatography with electrochemical detection. J Density Lipoprotein Methodology Workshop.
Chromatogr A 2007;1166:135-141. DHEW Publ. No. NIH 79-1661. Washington, DC:
61 Kasama T, Byun DS, Seyama J. Quantitative U.S. Government Printing Office; 1979:178.
analysis of sterols in serum by high-performance 75 Haeckel R, Perlick, M. A new enzymatic
liquid chromatography. Application to the determination of cholesterol: the use of aldehyde
biochemical diagnosis of cerebrotendinous dehydrogenase to measure H2O2 producing
xanthomatosis. J Chromatogr 1987;400, 241-246. reactions, II. J Clin Chem Clin Biochem 1976; 14:
62 Ogren L, Csiky I, Risinger L, Nilsson LG, 411-414.
Johansson G. A post-column enzyme reactor for
391
Cholesterol
76 Miner-Williams W. Surfactant inhibition of prevention of atherogenesis. Endocr Pract;

cholesterol oxidase. Clin Chim Acta 1980;101:77- 2000;6:162-213.
84. 80 Summary of the second report of the National
77 Martinello F, da Silva EL. Ascorbic acid Cholesterol Education Program (NCEP) expert
interference in the measurement of serum panel on detection, evaluation, and treatment of
biochemical parameters: In vivo and in vitro high blood cholesterol in adults. JAMA
studies. Clin Biochem 2006;39:396-403. 1993;269:3015-3023.
78 Fredrickson DS, Levy RI, Lees RS. Fat transport in 81 Fraser CG, Peterson PH, Ricos C, Heackel R. In:
lipoproteinsan integrated approach to Heackel R, ed. Evaluation of Methods in Laboratory
mechanisms and disorders. N Engl J Med Medicine. New York: VCH Publishers; 1993:87-99.
1967;276:148-156. 82 Fraser CG. Biological variation in clinical
79 Anderson RJ, Bergman DA, Davidson J, Palumbo chemistry. An update: collated data, 1988-1991.
PJ, Rettinger HI, Rubenstein HA et al. The Arch Pathol Lab Med 1992;116:916-923.
American Association of Clinical Endocrinologists
medical guidelines for clinical practice for the
diagnosis and treatment of dyslipidemia and
Table 1: Methods for Serum Cholesterol Analysis

Method 1: Liebermann-Burchard (L-B); one-, two-, three-, or four-step method
Principle of analysis: Cholesterol extracted and reacted with strong acid (sulfuric acid) and acetic anhydride to form
colored cholestahexaenesulfonic acid molecule (Amax, 410 nm); nonesterified cholesterol precipitated by digitoxin,
and remaining cholesterol measured and free cholesterol calculated; Total Esterified = Free
Comments: Very common method; total cholesterol reaction overestimates concentration of esterified cholesterol;
unstable color
Method 2: Abell et al. [7]; three-step method
Principle of analysis: Cholesterol extracted with zeolite, esters chemically hydrolyzed (saponification), and total
cholesterol measured by L-B reaction
Comments: Considered current reference method; laborious
Method 3: Iron-salt-acid; two-step method
Principle of analysis: Similar to reaction conditions of Method 2, except Fe3+ ions are added to yield tetraenylic
cation (Amax, 563 nm)
Comments: Not frequently used; sevenfold more sensitive than L-B method; free and esterified cholesterol give
same color; no need to hydrolyze esters
Method 4: p-Toluenesulfonic acid (p-TSA); three-step method
Principle of analysis: Similar to method 3; p-TSA reacts with cholesterol derivative to form chromophore (Amax,
550 nm)
Comments: Rarely used; free and esterified cholesterol give same color
Method 5: Enzymatic end-point; one-step method
_
a. Cholesterylesters cholesterol esterase cholesterol + fatty acids
_
b.* Cholesterol + O2 cholesterol oxidase cholest-4-en-3-one + H2O2
_
c. H2O2 + 4-aminophenazone (or other dye) peroxidase oxidized dye
(Amax, 500 nm) + H2O2
Comments: Most common method; accurate and easily automated; future reference method
*Can monitor reaction by following O2 consumption with oxygen electrode.
Table 2: Historical Reference Intervals for Total Serum Cholesterol [79]

Age Men* Women* Suggested Normal Limits
(years) mg/dL mg/dL mg/dL mmol/L
0-19 172 34 179 33 120230 (3.105.95)
20-29 183 37 179 35 120240 (3.106.21)
30-39 210 33 295 37 140270 (3.626.98)
40-49 230 55 217 35 150310 (3.888.02)
50-59 240 48 251 49 160330 (4.148.53)
*Total cholesterol in mg/dL in mean standard deviation (values in parentheses are in mmol/L).
Conversion Factors: mg/dL 0.0259 = mmol/L mmol/L 38.6 = mg/dL
392
Cholesterol
Table 3: Classification of Adults Based on Total and HDL Cholesterol

Total Cholesterol
Desirable <200 mg/dL
Borderline high 200-239 mg/dL
High >240 mg/dL
HDL Cholesterol
Low <40 mg/dL
High >60 mg/dL
Table 4: Treatment Approach Based on CAD Risk Factors and LDL Cholesterol Level (mg/dL)
Nutrition Therapy, Drug
Setting Physical Activity Therapy Goal
CAD risk factors

<2 >160 >190 <160
>2 >130 >160 <130
With atherosclerotic disease >100 >130 <100
With type II diabetes mellitus >100 >130 <100
Note: Subtract one risk factor when HDL cholesterol is 60 mg/dL.
393
Cholinesterase
Cholinesterase
Danyal B. Syed
Name: Cholinesterase, acylcholine acylhydrolase, cholinesterase 2 (ChE 2),
RBC cholinesterase, AChE; Pseudocholinesterase, butyrylcholinesterase,
cholinesterase I (ChE 1), plasma cholinesterase, CHS
Clinical significance: Refer to Chapter 52, Diseases of Genetic Origin, in the 5th edition of Clinical
Enzyme number: EC 3.1.1.7 (Acetylcholinesterase, RBC cholinesterase, acetylcholine
acetylhydrolase)
EC 3.1.1.8 (Pseudocholinesterase, butyrylcholinesterase, acylcholine
acylhydrolase, plasma cholinesterase)
Number of units: Four identical subunits
Known enzyme forms: RBC, brain, liver
Chromosomal location [1,2]: Acetylcholinesterase (3.1.1.7) = 7q22;
Butyrylcholinesterase ( 3.1.1.8) = 3q26.1-26.2

i less specific for this substrate than is the red-blood-cell
enzyme acetylcholinesterase, and so its function may be
The polymorphic serine esterases that catalyze the to hydrolyze other choline esters. Cholinesterase activity
hydrolysis of the acyl-esters to respective alcohols and is usually measured for one of two reasons. The most
carboxylates are called cholinesterases. They differ in common reason is the determination of a decrease in
their substrate specificity, sensitivity to certain enzymatic activity as an indicator of exposure to
inhibitors, and their localizations at both cellular and organophosphorus compounds, including many
subcellular levels. There are two types: (1) pesticides. The second reason is to identify the presence
acetylcholinesterase (EC 3.1.1.7), also called RBC of inherited abnormal variants of cholinesterase.
cholinesterase, is a true cholinesterase found in the Variants are identified by assaying both total activity and
blood and neural synapses; and (2) pseudocholinesterase the extent of inhibition by either dibucaine or fluoride;
(EC 3.1.1.8), also referred to as plasma or serum some of these variants lead to prolonged apnea in
cholinesterase (and CHS, cholinesterase I, and patients receiving the anesthetic succinylcholine or
butyrylcholinesterase), which is synthesized by the liver mivacurium. Measurement of acetylcholinesterase
and is present in plasma [3-5]. Its true physiological activity can be performed during amniotic fluid analyses
function is unknown. The two types differ in their for neural tube defects to confirm an elevated amniotic-
respective substrate preferences. Although fluid alpha-fetoprotein (AFP) level.
pseudocholinesterase can hydrolyze acetylcholine, it is
Since pesticide poisoning can present acutely, an assay
for cholinesterase must be simple enough to be
i
Cholinesterase performed stat but accurate enough for determination
Previous and current authors of this method: of cholinesterase phenotyping. The common test
First edition: Mary Ellen King principle for measuring cholinesterase activity is based
Methods book: Mary Ellen King on the use of thiocholine esters. These substrates are
Second edition: Mary Ellen King known to be unstable in aqueous solutions. Serum
Third edition: Steven C. Kazmierczak acylcholinesterase, although active against most of the
Fourth edition: Steven C. Kazmierczak thioesters of choline, is best determined by using
Fifth edition: Danyal B. Syed butyrylthiocholine as a substrate, which is stable in
394
Cholinesterase
aqueous solution for 30 days [6]. Cholinesterase activity countries employs the use of butyrylthiocholine as
is usually measured by one of the three methods substrate and DTNB as the chromogenic agent. More
described below: recently it has been proposed [21] that the substrate
succinyldithiocholine/DTNB system be considered as a
1. Spectrophotometric. This method uses basis for a candidate reference procedure for
propionylthiocholine or acetylthiocholine as substrate standardization of the cholinesterase assays within the
(Table 1, Method 1). 5,5-dithiobis(2-nitrobenzoic acid) work of the IFCC Committee on Reference Systems on
(DTNB) is added to react with the released thiocholine Enzymes [22].
to form the yellow compound, 5-thio-2-nitrobenzoic acid
(absorption maximum, 410 mm) [7,8]. The reactions These proposals were based upon the evaluation of six
may be followed as a rate or end-point procedure. different methods for the purpose of providing
2. Titrimetric [9-13]. These methods follow the release comparison between methods for their ability to identify
of hydrogen ions from a choline ester. One titrimetric succinyldicholine-sensitive patients. Cholinesterase
method (Michel method) directly measures hydrogen- activity was measured using a total of 131 patients sub-
ion release by monitoring pH changes over a 1- to 1- grouped according to cholinesterase phenotypes with
hour period (Table 1, Method 2). A second titrimetric different sensitivity to succinyldicholine. By determining
colorimetric method quantifies the amount of hydrogen the phenotypes based on dibucaine and fluoride
ion released by the hydrolysis of acetylcholine by numbers, and by performing the DNA analysis,
following the color change of a phenol red indicator over correlation between the phenotype classification and
a 30-min incubation time (Table 1, Method 3). The color genotype was confirmed. The tested methods
change is related to the moles of substrate converted. significantly differed in their ability to discriminate
3. Electrometric [14-16]. This is a special type of between the subjects with and without
titrimetric method usually referred to as an electrometric succinyldicholine-sensitive phenotypes, as exhibited by
procedure. It uses highly automated titration the area under the receiver operating curves (ROC) as
instrumentation to continuously add base to neutralize follows (21):
the hydrogen ions released from acetylcholine by
cholinesterase (Table 1, Method 4). This instrument Methods Substrate Area under ROC
maintains the enzyme reaction at constant pH, and the 1. Succinyldithiocholine/
amount of base added is directly related to the number of 5,5-dithio-bis-nitrobenzoate (DTNB) 0.97
moles of acetylcholine hydrolyzed on a mole-per-mole 2. Propionylthiocholine / DTNB 0.94
basis. 3. Butyrylthiocholine (BTC) 0.90
4. Benzoylcholine 0.90
The presence of true acetylcholinesterase activity in
amniotic fluid can be determined by a quantitative
[17,18] or qualitative assay [19,20]. In either case, an It was concluded that the succinyldithiocholine/DTNB
inhibitor of acetylcholinesterase activity is used. The assay is the method of choice for selecting subjects
most frequently used inhibitor is 1,5-bis(4- sensitive to succinyldicholine treatment and thereby
allyldimethylammoniumphenyl)pentan-3-one dibromide developing apnea. At present, most automated
(BW 284C51). For a quantitative assay, the enzyme instruments prefer the use of butyrylthiocholine as
activity can be measured by Method 1 in Table 1. The substrate for determining the cholinesterase activity.
qualitative assay for acetylcholinesterase employs With the Vitros analyzer [23], butyrylthiocholine is also
polyacrylamide gel electrophoresis to separate used as substrate, but the hydrolysis product,
isoenzymes of cholinesterase [19,20]. After thiocholine, reduces the chromogen, potassium
electrophoresis, the gels are incubated with hexacyanoferrate III (potassium ferricyanide), to
acetylthiocholine in the presence and absence of the hexacyanoferrate II [24]. The rate of the loss of color, as
acetylcholinesterase inhibitor BW 284C51. The measured by reflectance spectrophotometry, is
thiocholine formed by the reaction reacts with copper proportional to the cholinesterase activity in the sample.
sulfate, also present in the reaction mixture, to form a The thio-substrates are normally unstable and must be
white cupric thiocholine precipitate. The zone of prepared fresh prior to use. The use of polar solvents to
acetylcholinesterase activity is identified by its enhance the stability of BTC has recently been patented
disappearance when the acetylcholinesterase inhibitor is [25]. In an earlier patented invention [26], p-
present. hydroxybenzoyl choline is used as substrate in the
presence of p-hydroxybenzoic acid hydroxylase enzyme.
Reference and Preferred Methods The cholinesterase hydrolysis product p-hydroxybenzoic
Currently there is no formal reference method for the acid is acted upon by p-hydroxybenzoic acid
determination of cholinesterase activity. Some 3 decades hydroxylase enzyme in the presence of NADPH, which
ago, propionylthiocholine/DTNB method was selected to is oxidized to NADP+ and measured by the decrease in
be the method for measuring cholinesterase activity by extinction coefficient at 340 nm. The reaction can also
the American Association of Clinical Chemistry be monitored by measuring the loss of oxygen using an
(AACC) [7]. In 1992, a national German procedure oxygen detector system and has been fully automated.
utilized butyrylthiocholine as substrate [6]. The most As is evident, this enzyme-coupled reaction to measure
commonly used procedure in Italy and many other the cholinesterase enzymatic activity avoids the use of
395
Cholinesterase
thioester compounds altogether as substrates. Qualitative gel electrophoresis is the method of choice
Colorimetric assays using substrates as described above for detection of amniotic fluid acetylcholinesterase [32].
are also available for several automated chemistry Since this test is used to confirm the suspicion of a
systems. Dibucaine inhibition assays are also available, neural tube defect after an elevated amniotic fluid AFP
or the automated cholinesterase assay can be modified in level is found, results with high clinical specificity are
the laboratory to measure the inhibition [27-30]. required. The British collaborative report showed that
the mere presence of acetylcholinesterase activity was
The temperature optimum is 37C to 40C for both sufficient evidence to confirm an elevated AFP result.
cholinesterase and acetylcholinesterase activity. The
Q10 (temperature activity coefficient) is 1.3 for Specimen
acetylcholinesterase and 1.5 for cholinesterase. A No special patient preparation is needed. Either serum or
standardized temperature of 37C has been heparinized plasma may be used for cholinesterase
recommended for assessing cholinesterase sensitivity to assay. Cholinesterase is stable for up to 80 days at room
dibucaine and fluoride. However, some of the variant temperature and for 3 years if frozen at 20C [3].
forms of cholinesterase have temperature maxima as low Samples should be stored and shipped cold to avoid wide
as 32C or as high as 51C, and both fluoride and extremes of temperature. Samples can be drawn into
dibucaine inhibition numbers decline as the either a plain red-top tube or a speckled-red-top tube
measurement temperature is raised above 25C [3]. with gel as serum separator, or for plasma into a green-
top tube with heparin as anticoagulant, employing
Described below is a simple colorimetric method based normal phlebotomy procedure. Samples should be
on the procedure of Ellman, requiring only a centrifuged and serum or plasma separated quickly from
spectrophotometer and 10 min of analysis time. This the clot or cells, avoiding hemolysis. If samples are
method has been published by the American Association frozen, repeated freeze/thaw cycles should be avoided.
for Clinical Chemistry [7]. The substrate
propionylthiocholine iodide is used, since it gives lower Samples can also be collected onto filter paper using the
blank readings and 27% greater activity at 37C than dried-blood-spot method. This sample collecting method
butyrylthiocholine iodide. Easier discrimination among is especially useful in field studies to evaluate
some common cholinesterase phenotypes is observed cholinesterase activity in young children to detect and
with use of this substrate. Between-run coefficients of monitor organophosphate pesticide exposure in children
variation (CVs) of 3% are reported [7,8]. The reaction living in agricultural communities. The amount of
conditions are summarized in Table 2. activity in dried blood has been determined to be less
than the activity in plasma or serum but exhibits
The Michel method [9] has served as a standard method significant correlation: Spearman r = 0.6 (p = 0.01) [33].
during the collection of most of the clinical data of
pesticide exposure and is still widely used. The Samples believed to contain a reversible inhibitor should
procedure is slow (10 min of equilibration plus 1 hour of be collected on ice, kept cold, and assayed as soon as
incubation), and good accuracy and precision depend possible to limit in-vitro destruction of the inhibitor.
heavily on excellent temperature control and careful pH Sample-to-reagent dilution and incubation times should
measurement [4]. A 1C change produces a 5.5% change be minimized in the assay to limit reactivation of the
in cholinesterase activity and a 3% change in enzyme during assay [4].
acetylcholinesterase activity. Newman and Que Hee [31]
investigated cholinesterase results using the Michel Interferences
method and recommended use of a reagent blank rather Moderate hemolysis will not interfere if the serum is
than correction tables for improved accuracy. They also well centrifuged to remove red-blood-cell ghosts [7].
found better inter-run precision with use of a reagent Lipemia or icterus (bilirubin up to 20 mg/dL) reportedly
blank (CV = 6%) than with use of the correction table does not interfere. The presence of oxalate or fluoride
(CV = 10%). has been reported to significantly decrease measured
cholinesterase activities [34]. Similarly citrate
The electrometric procedure, in which the neutralization anticoagulant, heavy metals, borate, some serum
of the reaction is monitored by an automatic titrator, separators, insecticides, organophosphates, carbamates,
requires a specialized instrument dedicated to the assay neuromuscular relaxants, phenothiazines, both anabolic
but offers the advantage of being a rapid, rate method and glucocorticosteroids, oral contraceptives, estrogens,
[14,15]. CVs of 4% to 5% are reported [16]. The radiographic agents such as iopanoic acid, streptokinase,
instrument can also be used to measure testosterone, and many other such compounds have been
acetylcholinesterase. observed to decrease cholinesterase activity. Plasma
cholinesterase activity has been noted to decrease after
Because of its ease of adaptation to most laboratory plasmapheresis and increase after transfusion in patients
spectrophotometers, the Ellman method is recommended who have abnormally low cholinesterase activity [35].
if cholinesterase is not available on automated
equipment.
396
Cholinesterase
Cholinesterase Reference Intervals worker exposure to certain toxic pesticides, principally

Each laboratory should establish its own reference range organophosphates and carbamates. These compounds
based upon population, instrument and methodology have sufficient structural similarity to acetylcholine to
used. Generally it is between 4 and 19 U/mL or 4,000 bond to both cholinesterase enzymes; organophosphates
and 19,000 U/L. Cholinesterase values in neonates and phosphorylate and the carbamates acetylate a serine
infants younger than 6 months are 40% to 50% of adult residue of the enzymes active sites. The stability of
levels [36]. Young adult females, 18 to 35 years of age, these bonds depends on their sensitivity to hydrolysis;
have cholinesterase levels 64% to 74% of adult male half-lives of the various inhibitor-enzyme complexes
values. Older females and males (70 to 80 years of age) range from minutes to weeks. The carbamates, which
have values not significantly different from those of form bonds that hydrolyze with relative ease in vivo and
young adult females [37]. Cholinesterase activity in vitro, are referred to as reversible inhibitors. Other
decreases during pregnancy [38]. compounds containing quaternary or tertiary amines or
other positively charged atoms may also reversibly
Acetylcholinesterase levels in neonates are 63% of adult inhibit cholinesterases; these include physostigmine and
levels [39] and rise gradually to adult levels by 1 year of sulfonium and arsonium compounds [4].
age [40]. There are no male/female differences in
acetylcholinesterase values in adults [37]. The laboratory picture resulting from pesticide poisoning
is dependent on the pesticide involved and the level and
The most common phenotype (U, usual) is not time frame of exposure. Some compounds preferentially
associated with a prolonged response to succinylcholine inhibit cholinesterase; others inhibit acetylcholinesterase
and is inhibited approximately 84% by dibucaine and more strongly. Pesticides that preferentially inhibit
80% by fluoride (see Table 3) [41]. Other phenotypes cholinesterase after low-level exposure will often inhibit
that can occur include dibucaine-resistant A (atypical) acetylcholinesterase as well if a longer exposure occurs
and fluoride-resistant F (fluoride). The so-called silent [45]. For example, a person exposed to slight to
phenotypes, S1 and S2, display little or no cholinesterase moderate levels of a cholinesterase inhibitor will first
activity in the homozygous state. The homozygous have a drop in cholinesterase activity as the enzyme is
variant forms are usually not inhibited more than 20% to inactivated in vivo. If the low-level exposure continues,
50% by the relevant inhibitor. red cell acetylcholinesterase may also decline. This same
pattern of low cholinesterase and low
Acetylcholinesterase is present in fetal serum but is not acetylcholinesterase activities could, however, be found
normally detected in amniotic fluid. The intraindividual after a single, recent, moderate to severe exposure to the
variation for cholinesterase activity in healthy adults has same pesticide. When the exposure to the inhibitor stops,
been reported to be less than 7% [3,4,5,7]. cholinesterase activity returns to normal before
acetylcholinesterase activity, because regeneration of
The cholinesterase activity level in cerebrospinal fluid cholinesterase through liver synthesis of the enzyme
(CSF) has been observed to be 1/20 to 1/100 of the proceeds faster than bone-marrow replacement of the red
blood cells, which contain acetylcholinesterase. A
corresponding serum/plasma activity. In clinical
person recovering from pesticide exposure that resulted
conditions where bleeding into CSF takes place, the CSF
in depressed cholinesterase and acetylcholinesterase
cholinesterase enzymatic activity has been demonstrated levels will have normal levels of cholinesterase with low
to increase to 1/4 to 1/2 of corresponding plasma activity levels of acetylcholinesterase for a time, the same picture
level [42]. that another person would have after a mild exposure to
a pesticide preferentially inhibiting acetylcholinesterase.
Interpretation
Plasma cholinesterase (CHS) is found in plasma, liver, A considerable decline in the activities of either or both
brain, kidney, intestine, and pancreas. In contrast, RBC enzymes can occur before clinical symptoms appear. For
cholinesterase (acetylcholinesterase) is found in red example, values as low as 10% of original levels may be
blood cells, nerve tissue, skeletal muscle, and placenta. seen in a person exposed to low levels of pesticide over a
The two enzymes have different Kms for acetylcholine long period of time [45]. However, mild illness can
and succinylcholine, as tabulated in Table 4 [43,44]. occur while either or both enzyme levels are within
population-reference intervals, particularly after short-
Some cholinesterase variants are unable to break down term moderate exposure to pesticide. In nearly all severe
succinylcholine, a muscle relaxant used in anesthesia, poisonings, both acetylcholinesterase and cholinesterase
causing prolonged apnea and paralysis. It is important to activities drop sharply.
determine if a cholinesterase variant is the cause of an
unusual response to succinylcholine so that the drug can Because of the relatively wide population reference
be avoided or its dosing modified if the patient needs intervals for cholinesterase and acetylcholinesterase, it is
anesthesia in the future. The frequency of response is best to interpret a persons enzyme activity levels in
cited in Table 5 [5]. relation to his or her personal reference values. This is
particularly important in the evaluation of possible low-
Measurements of RBC acetylcholinesterase (AChE) and level pesticide exposure, where only a small decline in
plasma cholinesterase (CHS) are performed to monitor one or both enzymes may occur. A worker should have
397
Cholinesterase
at least two baseline measurements made of neurodegeneration than those of the Abeta peptide alone
cholinesterase and acetylcholinesterase before beginning both in vitro and in vivo.
work with pesticides. Intraindividual CVs average 7% to
11% for each enzyme, with maximum fluctuations of The fact that AChE is able to enhance amyloid
20% to 25% in healthy persons [45]. If either enzyme formation, the suggestion that the AChE activity
activity drops 20% to 30% or more on a subsequent associated with amyloid formation is exhibited by a 3.5
analysis, pesticide exposure should be considered. Drops kDa peptide (H peptide) containing a tryptophan of the
of 50% to 75% are clear danger signals that poisoning enzyme peripheral binding site [56-58], and that such an
has occurred, and hospitalization may be required [46]. effect is sensitive to drugs that block this AChE activity
of the H peptide, has resulted in the use of AChE
Other conditions associated with reduced cholinesterase inhibitors such as donepezil, rivastigmine, and
activity include liver disease, malnutrition, acute galantamine, which are now commonly used
infection, and some anemias. The presence of these therapeutically in the treatment of mild to moderate
conditions must be considered in the evaluation of Alzheimers disease, possibly to reduce the
enzyme changes. Acetylcholinesterase levels reflect the neurodegeneration and neurotoxicity [59]. However, the
number and age of the red blood cell population, with level of efficacy of these drugs is still being debated.
the highest levels in the youngest erythrocytes [4].
Because the circulating levels of cholinesterase and Cholinesterase Performance Goals
acetylcholinesterase are controlled by different Reportable Range (Dynamic Range, Analytical
physiological mechanisms, a drop in both enzyme levels Range): varies from instrument to instrument, based on
points very strongly to exposure to an inhibitor such as a assay method, temperature and substrate. Based on the
pesticide. butyrylthiocholine/ferricyanide method, as
recommended by the German Society for Clinical
The presence of both true acetylcholinesterase activity Chemistry, the lower limit of reportable range
and elevated AFP levels in amniotic fluid is considered (sensitivity) is generally between 10 and 200 U/L (0.01
presumptive evidence of an open neural tube defect in a to 0.2 U/mL), and the upper limit is 11,000 to 25,000
fetus. The British Collaborative Acetylcholinesterase U/L (11.0 to 25.0 U/mL). Samples with cholinesterase
Study [32] showed that 99.5% of all pregnancies activity greater than the upper limits of the dynamic
associated with an anencephaly and open spina bifida range must be diluted and reanalyzed according to
had positive acetylcholinesterase results by gel instrument manufacturers recommendations or a
electrophoresis. Pregnancies associated with exomphalos laboratorys own protocol.
(75%), miscarriage (47%), and other serious congenital
malformations (50%) also had positive Precision: The within-run precision varies from
acetylcholinesterase results. Only 6.4% of pregnancies instrument to instrument and method to method, and
not associated with serious fetal malformations or generally ranges from 0.5% to 2.0 %, whereas between-
miscarriage had positive acetylcholinesterase results. day CV within a laboratory ranges from 1.4% to 5%.
However, all but one of the false-positive cases were The cholinesterase (pseudocholinesterase) activity is not
associated with significant contamination of blood, a regulated analyte according to the Clinical Laboratory
either of fetal or unspecified origin. Improvement Amendments of 1988 (CLIA 88), and the
CAP criteria for acceptable performance in external
The involvement of cholinesterase in triglyceride proficiency testing is peer group 30%.
metabolism and correlation of its activity with plasma
LDL-cholesterol [47], HDL-cholesterol [47-51], and References
triglyceride concentration has been reported. It has also 1 Getman DK, Eubanks JH, Camp S, Evans GA,
been shown to exhibit significant association with the Taylor P. The human gene
metabolic syndrome, as well as diabetes. This encoding acetylcholinesterase is located on the long
connection between the cardiovascular risk factors, arm of chromosome 7.
metabolic syndrome, and cholinesterase activity has Am J. Hum Genet 1992;51:170-177.
recently been confirmed [52]. These studies established 2 Soreq H. The Multileveled regulation of the human
heritability for plasma cholinesterase activity in the cholinesterase genes
general population and identified two chromosomal and their protein products. Biological Sciences,
locations that contribute to this heritability between the Biochemistry, Report
cardiovascular risk factors, metabolic syndrome, and A372772. Jerusalem, Israel: Hebrew University;
cholinesterase activity. September 30, 1993.
3 Brown SS, Kalow W, Pilz W, Whittaker M. The
It has been suggested that the low-activity K variant has plasma cholinesterases: a
a causative role in Alzheimers disease [53-55]. new perspective. Adv Clin Chem 1981;22:1-123.
Biochemical studies have also indicated that 4 Willis JH. Blood cholinesterase: assay methods
acetylcholinesterase (AChE) induces amyloid fibril and considerations. Lab Management
formation, resulting in the formation of highly toxic 1982;20:53-64.
AChE-amyloid Abeta complexes, triggering more 5 Lehmann H, Liddell J. The cholinesterase
variants. In: Stanbury JB, Wyngaarden JB, and
398
Cholinesterase
Fredrickson DS, eds. The Metabolic Basis of for neural-tube defects in early pregnancy.
Inherited Disease. 3rd ed. New York: McGraw- Lancet 1979;1:685-688.
Hill; 1972:1730-1736. 20 Barlow RD, Cuckle HS, Wald NJ. A simple
6 Working Group on Enzymes. Proposal of method for amniotic fluid gel-
standard methods for the determination of acetylcholinesterase determination, suitable for
enzyme catalytic concentrations in serum and routine use in the antenatal diagnosis of open
plasma at 37 C. Eur J Clin Chem Clin Biochem neural tube defects. Clin Chim Acta
1992;30:163-170. 1982;119:137-142.
7 Dietz AA, Rubenstein, HM, Lubrano T. 21 Andrea Mosca A, Bonora R, Ceriotti F,
Colorimetric determination of serum Franzini C, Lando G, Maria Cristina Patrosso
cholinesterase and its genetic variants by the MC et al. Italian Society of Clinical
propionylthiocholine-dithiobis (nitrobenzoic Biochemistry and Clinical Molecular Biology
acid) procedure. In: Cooper GR, King JS, eds. (SIBioC) Working Group on Enzymes. Assay
Selected Methods for the Small Clinical using succinyldithiocholine as substrate: the
Chemistry Laboratory. Vol. 8. Washington, DC: method of choice for the measurement of
American Association for Clinical Chemistry; cholinesterase catalytic activity in serum to
1977:41-46. diagnose succinyldicholine sensitivity. Clin
8 Price EM, Brown SS. Scope and limitations of Chem Lab Med 2003;41:317-322.
propionylthiocholinesterase in the 22 Panteghini M, Ceriotti F, Schumann G,
characterization of cholinesterase variants. Clin Siekmann L. Establishing a reference system in
Biochem 1975;8:384-390. clinical enzymology. Clin Chem Lab Med
9 Michel HO. An electrometric method for the 2001;39:795-800.
determination of red blood cell and plasma 23 Ortho-Clinical Diagnostics. Test methodology:
cholinesterase activity. J Lab Clin Med CHE, Cholinesterase. Part No. MP2-91, CAT
1949;34:1564-1568. No. 126 2096. Rochester, NY: Ortho-Clinical
10 Witter RF, Grubbs LM, Farrior WL. A Diagnostics, Inc; 1996.
simplified version of the Michel method for 24 Kenedel Bttger. Klin Wschr. 1967;45:325-327.
plasma or red cell cholinesterase. Clin Chim 25 Weber F, Gottschalk N, Kytzia H-J, Muecke M,
Acta 1966;13:76-78. Nagel R. Stabilized cholinesterase substrate
11 Caraway WT. Photometric determination of solution. Application Number 11/566788.
serum cholinesterase activity. Am J Clin Pathol Filing date 12/05/06. Roche Diagnostics
1956;26:945. Operations, Inc. US patent 20070178546.
12 Crane CR, Sanders DC, Abbot JN. Available at
Cholinesterase use and interpretation of <http://www.freepatentsonline.com/70178546.h
cholinesterase measurements. In: Sunshine I, tml>
ed. Methodology for Analytical Toxicology. 26 Motonaga H Naito, Shino M. Test Lab Co Ltd
Cleveland, OH: CRC Press; 1975:86-104. (JP). Method for determination of
13 Ellin RI, Vicario P. A pH method for measuring cholinesterase activity. Patent Number 4565780
blood cholinesterase. Arch Environ Health pub 1986. Available at
1975;30:263-265. <http://www.freepatentsonline.com/4565780.ht
14 Nabb DP, Whitfield F. Determination of ml>
cholinesterase by an automated pH stat method. 27 Holownia P, Newman DJ, Bruno C. Automated
Arch Environ Health 1967;15:147-152. dibucaine number measurement with DuPont
15 Aldrich FD, Walker GF, Patnow CA. A Dimension ES and AR analyzers. Clin Chem
micromodification of the pH-stat assay for 1995;41:664-667.
human blood cholinesterase. Arch Environ 28 Naccarato WF. DuPont Imaging and Medical
Health 1969;19:617-620. Products, Diagnostics Division. New
16 Serat WF, Van Loon AJ, Lee MK. application of the pseudocholinesterase method:
Microsampling techniques for human blood determination of dibucaine and fluoride
cholinesterase analysis with the pH stat. Bull number. DuPont technical bulletin IPDP-2156.
Environ Contam Toxicol 1977;17:542-550. Wilmington, DE: DuPont Imaging; 1980:1-4.
17 Hay DL, Ibrahim GF, Horacek I. Rapid 29 Denblaauwen DH, Poppee WA, Tritscher W.
acetylcholinesterase screening test for neural Cholinesterase (EC 3.1.1.8) with butyryl iodide
tube defect. Clin Chem 1983;29:1065-1069. as substrateage-dependent and sex-dependent
18 Hullin DA, Laurence KM, Elder GH, Roberts reference values with special reference to
A, Newcombe RG. Amniotic fluid hormonal effects and pregnancy. J Clin Chem
cholinesterase measurement as a rapid method Biochem 1983;21:381-386.
for the exclusion of fetal neural-tube defects. 30 Cattozzo G, Franzini C, Rettodini M. Dibucaine
Lancet 1981;1:325-326. number measured with the Ektachem system
19 Smith AD, Wald NJ, Cuckle HS, Stirrat GM, [tech brief]. Clin Chem 1993;39:1545-1546.
Borrow M, Lagercrantz H. Amniotic fluid 31 Newman MA, Que Hee SS. Interconversion and
acetylcholinesterase as a possible diagnostic test comparison of the results of three methods for
399
Cholinesterase
acetyl cholinesterase in serum. Clin Chem Butyrylcholinesterase activity and metabolic

1984;32:308-310. syndrome in obese patients. Clin Chem Lab
32 Wald NJ, Cuckle HS. Report of the Med 2005;43:285-288.
Collaborative Acetylcholinesterase Study: 48 Abbott CA, Mackness MI, Kumar S, Olukoga
amniotic fluid acetylcholinesterase AO, Gordon C, Arrol S et al. Relationship
electrophoresis as a secondary test in the between serum butyrylcholinesterase activity,
diagnosis of anencephaly and open spina bifida hypertriglyceridaemia and insulin sensitivity in
in early pregnancy. Lancet 1981;2:321-327. diabetes mellitus. Clin Sci (Lond) 1993;85:77-
33 Hilborn ED, Padilla S. A dried blood spot 81.
method to evaluate cholinesterase activity in 49 Annapuma V, Senciall I, Davis AJ, Kutty KM.
young children. Arch Environ Health Relationship between serum
2004;59(9):467-70. pseudocholinesterase and triglycerides in
34 Kramer S, Watson LL, Lodato DT, Widsor- experimentally induced diabetes mellitus in
Long K, Gilmore JM, Augustyn JM. Assay of rats. Diabetologia 1991;34:320-324.
pseudocholinesterase on the dimension clinical 50 Kutty KM, Payne RH. Serum
chemistry system. Clin Chem 1991;37:913. pseudocholinesterase and very low density
35 Evans RT, MacDonald R, Robinson A. lipoprotein metabolism. J Clin Lab Anal
Suxamethonium apnoea associated with 1994;8:247-250
plasmapheresis. Anaesthesia 1980;35:198-201. 51 Rustemeijer C, Schouten JA, Voerman HJ,
36 Zsigmond EK, Downs JR. Plasma Beynen AC, Donker AJ, Heine RJ. Is
cholinesterase activity in newborns and infants. pseudocholinesterase activity related to markers
Can Anaesth Soc 1971;18:278-283. of triglycerol synthesis in type II diabetes
37 Shanor S, Van Hess GR, Baart N, Erdos EG, mellitus? Clin Sci (Lond) 2001;101:29-35.
Folders MD. The influence of age and sex on 52 Valle A, OConnor DT, Taylor P, Zhu G,
human plasma and red cell cholinesterase. Am J Montgomery GW, Slagboom PE, Martin NG,
Med Sci 1961;242:357-361. Whitfield JB. Butyrylcholinesterase: association
38 Howard J, East N, Chaney J. Plasma with the metabolic syndrome and identification
cholinesterase activity in early pregnancy. Arch of two gene loci affecting activity. Clin Chem
Environ Health 1978;33:277-279. 2006;52:1014-1020.
39 Kaplan E, Herz F, Hsu KS. Erythrocyte 53 Lehmann DJ, Johnston C, Smith AD. Synergy
acetylcholinesterase activity in ABO hemolytic between the genes for butyrylcholinesterase K
disease of the newborn. Pediatrics 1964;33:205- variant and apolipoprotein E4 in late-onset
211. confirmed Alzheimers disease. Hum Mol
40 Kaplan E, Tildon JT. Changes in red cell Genet 1997;6:1933-1936.
enzyme activity in relation to red cell survival 54 Panegyres PK, Mamotte CD, Vasikaran SD,
in infancy. Pediatrics 1963;32:371-375. Wilton S, Fabian V, Kakulas BA.
41 Brock A. Immunoreactive plasma Butyrylcholinesterase K variant and
cholinesterase (E.C. 3.1.1.8) substance Alzheimers disease. J Neurol 1999;246:360-
concentration, compared with cholinesterase 70.
activity concentration and albumin: inter- and 55 Ghebremedhin E, Thal DR, Schultz C, Braak H.
intraindividual variations in a healthy Age-dependent association between
population group. J Clin Chem Clin Biochem butyrylcholinesterase K-variant and Alzheimer
1990;28:851-856. disease-related neuropathology in human
42 Kambam JR, Horton B, Parris, WC, Hyman brains. Neurosci Lett 2002;320:25-28.
SA, Berman, LM, Sastry BVR. 56 De Ferrari GV, Canales MA, Shin I, Weiner
Pseudocholinesterase activity in human LM, Silman I, Inestrosa NC. A structural motif
cerebrospinal fluid. Anaesth Analg of acetylcholinesterase that promotes amyloid
1989;68:486-488. -peptide fibril formation. Biochemistry
43 Shukuya R. On the kinetics of the human blood 2001;40:10447-10457.
cholinesterase. J Biochem 1953;40:135-140. 57 Inestrosa NC, Sagal JP, Colombres M.
44 Foldes FF, van Hees GR, Shanor SP, Baart N. Acetylcholinesterase interaction with Alzheimer
Interrelationship of suxamethonium, amyloid beta. Subcell Biochem 2005;38:299-
suxethonium and succinyl monocholine and 317.
human cholinesterase. Fed Proc 1966;15:422. 58 Inestrosa NC, Dinmarca, MC, Alvarez A.
45 Hayes WJ. Pesticides Studied in Man. Amyloid-cholinesterase interactions.
Baltimore: Williams & Wilkins; 1982. Implications for Alzheimer disease. FEBS
46 Lotti M. Cholinesterase inhibition: complexities Journal 2008;275:625-632.
in interpretation. Clin Chem 1995;41:1814- 59 Seow D, Gauthier S. Pharmacotherapy of
1818. Alzheimer disease. Can J Psych 2007;52:620-
47 Alcantara VM, Oliveira LC, Rea RR, Suplicy 630.
HL, Chautard-Freire-Maia EA.
400
Cholinesterase
Tables
Table 1: Cholinesterase Methods Summary
Method 1: Colorimetric; spectrophotometric, kinetic or end-point
Comment: Serum or plasma; recommended method

Method 2: pH change; pH measurement
_
Principle of analysis: Acetylcholine CHS choline + acetate + H+
Comments: Serum or plasma; wide interlaboratory variation of results
Method 3: Colorimetric; pH dye color change, end-point
Comments: Serum or plasma; wide interlaboratory variation of results

Method 4: Electrometric; titrate acid released, kinetic or end-point
_
Acetylcholine CHS choline + acetate + H+
H+ + added base H2O and constant pH
Comments: Serum or plasma; requires automated pH titration
DTNB, 5,5-Dithiobis(2-nitrobenzoic acid).
Table 2: Conditions for Cholinesterase Analysis

Temperature 37C
Wavelength 410 nm
pH 7.6
Final concentration of reagent PCTI: 2 mmol/L
DTNB: 0.253 mmol/L
Phosphate: 25 mmol/L
Dibucaine: 0.03 mmol/L
Fluoride: 4 mmol/L
Sample volume 10 L
Fraction of sample volume 0.002 (dilution corrected)
Reaction time 10 min
Coefficient of variation 3%4
Linearity 0 through normal range
PCTI, Propionylthiocholine; DTNB, 5,5-dithiobis(2-nitrobenzoic acid).

401
Cholinesterase
Table 3: Reference Values for Genetic Variations of Cholinesterase*

Phenotype Activity at 37C, Percent inhibition by
mol/min/mL Dibucaine Fluoride
U 8.44 1.78 83.6 1.3 79.7 1.2
A 1.90 0.61 19.9 2.7 84.0 1.8
AD 1.30 0.37 20.7 4.1 82.3 3.4
S1 0.03 0.01 5.3 4.3 35.7 6.1
S2 0.18 0.05 67.6 4.3 67.6 1.7
F 3.57 71.8 53.6
AF 3.65 0.47 60.2 3.1 68.3 1.6
FS 3.47 76.7 64.9
UA 5.84 1.76 72.7 3.1 80.0 1.6
UF 5.99 1.26 79.8 1.2 73.0 1.7
US 4.61 0.57 84.4 0.8 79.3 1.4
*Derived from Dietz AA, Rubenstein HM, Lubrano T. Colorimetric determination of serum cholinesterase and its genetic
variants by the propionylthiocholine-dithiobis (nitrobenzoic acid) procedure. In: Cooper GR, King JS, eds. Selected
Methods for the Small Clinical Chemistry Laboratory. Vol 8 Washington, DC: American Association for Clinical
Chemistry; 1977.
Values are mean standard deviation.
Phenotypes associated with prolonged apnea in response to succinylcholine administration.
Table 4: Michaelis-Menten Constants for Cholinesterases

Substrate CHS (Cholinesterase) AChE (Acetylcholinesterase)
Acetylcholine 1.2 to 1.8 10-3 M[31,35] 3.7 10-4 M[31]
Succinylcholine 1. 10-3 M[35] Does not hydrolyze succinylcholine
Table 5: Prolonged Response to Succinylcholine

Phenotypes Estimated Frequency Sensitivity to Succinylcholine
Homozygote
UU Normal population 1/3200 moderately sensitive
AA 1/2800 All very sensitive
FF 1/300,000 All very sensitive
SS 1/140,000 All very sensitive
Heterozygote
UA 1/26 1/500 moderately sensitive
UF 1/280 1/200 moderately sensitive
US 1/190 Not known
AF 1/29,000 All very sensitive
AS 1/20,000 All very sensitive
FS 1/200,000 All very sensitive
402
Cholinesterase
Figures
Cholinesterase Figure 1: Polyacrylamide gel electrophoreses assay for acetylcholinesterase.
Polyacrylamide gel electrophoresis assay for acetylcholinesterase performed as described in the text. A, Amniotic fluid B
incubated in presence of substrate plus inhibitor. Notice absence of second band at arrow position. B, Amniotic fluid
incubated in presence of substrate. Notice second band at arrow position. Presence of a cholinesterase activity at second
position (arrow) is inhibited by BW 284C51, evidence that this is true acetylcholinesterase activity. An elevated amniotic
alpha-fetoprotein (>3 SD from mean) plus the presence of true acetylcholinesterase activity is strong evidence for the fetus
having a neural tube defect. C, Fetal calf serum that has a cholinesterase activity (arrow) that migrates slightly slower than
true acetylcholinesterase activity found in amniotic fluid. It is not inhibited by BW 284C51.
Procedure: Colorimetric Assay for Cholinesterase Stable for 3 months at 4C to 8C. Check weekly for
bacterial growth and pH and stability.
Principle 2. Propionylthiocholine iodide solution (PCTI),
Propionylthiocholine is used as a substrate for 20 mmol/ L. Place 606 mg of PCTI in a 100-mL
cholinesterase in the following coupled reactions: volumetric flask, and dissolve with approximately 80
_
Propionylthiocholine + H2O EC 3.1.1.8 propionic mL of distilled water. Bring volume to 100 mL with
distilled water. Stable for 1 day at ambient temperature.
acid + thiocholine
3. 5,5-Dithiobis(2-nitrobenzoic acid) (DTNB),
167 mg/L (0.421 mmol/L). Weigh out 167 mg of
Thiocholine + DTNB oxidized thiocholine + 5-thio-2- DTNB, dissolve, and dilute to a final volume of 1 L with
nitrobenzoic acid 0.036 M phosphate buffer. Stable for 12 months when
stored in a brown glass bottle at 4C.
Phosphate buffer pH 7.6 is used, and the rate of reaction 4. Dibucaine HCl (Nupercaine Hydrochloride,
is followed at 410 nm, an absorption maximum of 5- Ciba), 0.3 mmol/L. Place 57 mg into a 500-mL
thio-2-nitrobenzoic acid. In addition to total activity, volumetric flask, dissolve in distilled water, and dilute to
dibucaine and fluoride inhibitions (measured separately) a final volume of 500 mL with distilled water. Stable for
are determined when cholinesterase phenotyping is 1 year at room temperature.
requested. Only glass containers may be used, since 5. Sodium fluoride, 40 mmol/L. Weigh out 84
inhibitors may be extracted from some plastics. mg of NaF, dissolve in approximately 40 mL of distilled
Reagents water, and dilute to a final volume of 50 mL. Stable for 1
1. Phosphate buffer, 0.036 mol/L, pH 7.6. day at room temperature.
Prepare a solution of 0.036 M Na2HPO4 by dissolving 6. Quinidine sulfate, 14 mmol/L (5 g/L). Weigh
out 500 mg, dissolve in approximately 90 mL of distilled
4.73 g of the salt in approximately 800 mL of distilled water, and dilute to a final volume of 100 mL. Stable for
H2O and diluting to 1 L. Prepare a 0.1 M solution of 1 year at room temperature.
KH2PO4 by dissolving 13.6 g of the salt in a final 7. Working substrate reagents. Prepare 3 mL
volume of 1 L of distilled water. Mix 50 mL of KH2PO4 of each solution per test sample.
with 950 mL of Na2HPO4 to achieve a final pH of 7.6. 1. Dilute PCTI with an equal volume of distilled
water.
403
Cholinesterase
2. Dilute PCTI with an equal volume of dibucaine Procedure: Qualitative Analysis for Amniotic Fluid
solution. Acetylcholinesterase
3. If NaF is used, dilute PCTI with an equal
volume of sodium fluoride solution. Principle
Elevated levels of true acetylcholinesterase activity in
Assay amniotic fluid have been associated with a high
Equipment: Spectrophotometer or photometer set at 410 probability of fetal open neural tube defects. Positive
nm, with bandpass 10 nm, and 37C water bath. identification of the activity can be determined by
This procedure measures the activity of polyacrylamide gel electrophoresis. Different bands of
cholinesterase and the effect of two inhibitors, dibucaine cholinesterase activity are separated after electrophoresis
and fluoride. at pH 8.1. The activities are identified by incubation of
the gels with substrate (acetylthiocholine) and an
Assay for Total Activity indicator (copper sulfate). Cholinesterases will cleave
1. Prepare two test tubes for each sample or the substrate to acetic acid and thiocholine. The
control, one labeled test and the other blank. thiocholine reacts with Cu2+ to form complexes that
2. Add 3 mL of DTNB buffer to each test tube, form a white precipitate at the positions in the gel to
and allow to equilibrate 5 minutes in a 37C water bath. which the enzyme activities have migrated. The true
3. Add 1 mL of working PCTI substrate diluted serum acetylcholinesterase activity is defined by its
with water to each tube labeled test. ability to be inhibited by the specific inhibitor, BW
4. Add 1 mL of a 1:100 dilution of serum. 284C51that is, 1,5-bis(4-
(Samples with low activity should be reanalyzed with a allyldimethylammoniumphenyl)pentan-3-one dibromide.
smaller dilution, such as 10-fold or 50-fold.) Dilute the
serum with distilled water shortly before the assay, since Reagents
the cholinesterase may slowly lose activity after dilution. 1. Polyacrylamide 7.7% in buffer preservative.
5. After exactly 3 min, add 1 mL of quinidine Gels purchased as part of AlkPhor System from
reagent to stop the reaction. TechAmerica Diagnostics, San Marcos, CA (Also
6. Prepare blank readings by reversing steps 2 and available from Quantimetrix Corp., Redondo Beach CA,
5; that is, first add quinidine reagent to the blank tube, USA).
and then serum, and then add the substrate solution. 2. Tris-glycine buffer, pH 8.1.
7. Read absorbance of each test reaction against 3. Stock. To a 1-L volumetric flask, add 29 g of
the corresponding blank within 5 min. glycine and 6 g of tris(hydroxymethyl)aminomethane.
Dissolve salts in 800 mL of distilled water, and adjust
Assay for Dibucaine or Fluoride Inhibition pH to 8.1 with 3 M HCl. Adjust volume to 1 L. Store at
1. Same as for step 1 above. 4C to 8C. Stable for 1 year.
2. Add 3 mL of DTNB buffer to each test tube, Working buffer. Prepare a 1:10 dilution by
and allow to incubate 5 min in a 37C water bath. using 100 mL of stock Tris-glycine buffer and 900 mL
3. Add 1 mL of working substrateinhibitor of distilled water. Check that pH is 8.1 (+0.1); adjust if
solution, that is, PCTI diluted with either dibucaine or necessary. Discard after use.
NaF. 4. Preincubation indicator solution, pH 6.5
4. Follow steps 4 to 7 of the assay for total (final concentration of components in parentheses).
activity. To a 500-mL volumetric flask, add:
Calculations
Na2SO4 (1.06 mol/L) 75.3 g
The cholinesterase activity is expressed in international
units per milliliter (IU/mL = mol/min/mL) at 37C and Sodium maleate (0.07 mol/L) 4.834 g
calculated from the following equation: CuSO4 (0.004 mol/L) 0.500 g
Glycine (0.02 mol/L) 0.750 g
U/mL = x Aunknown = 14.71 Aunknown y 13.6 z MgCl2 (0.003 mol/L) 0.305 g
where x is the total reaction volume (in milliliters), y is Dissolve salts in approximately 450 mL of
the volume of undiluted sample used, Aunknown is the distilled water. Adjust pH to 6.5 with 3 M
reaction absorbance corrected for the blank, 13.6 is the NaOH, and adjust volume to 500 mL with
millimolar absorptivity of the 5-thio-2-nitrobenzoate for additional distilled water. Stored at room
a 1-cm cuvette, and z is the number of minutes of temperature. This reagent is stable for 4 weeks
incubation. Inhibition by dibucaine or fluoride is if no bacterial growth is evident. Check pH
calculated as follows: before use and adjust to 6.5 if necessary.
Percent inhibition = 5. Indicator with acetylthiocholine. To a 50-mL
100 % U/mLwith inhibitor 100% volumetric flask, add 0.1 g of acetylthiocholine iodide.
U/mLwithout inhibitor Adjust volume to 50 mL with preincubation indicator
solution. Prepare fresh.
6. Stock acetylcholinesterase inhibitor (BW

284C51) solution (10-3 M). To prepare this solution,
404
Cholinesterase
wear a surgical gown and mask and disposable plastic 10. Remove each tube. Gently insert the needle into
gloves. Work in a hood. Take a 25-mg vial of BW the tube, moving it alongside the walls while expelling
284C51 (purchased from Sigma Chemical Co., St. Louis, water. Be careful not to gouge the gels. Place each gel in
MO, A9013) and dissolve in 44 mL of distilled water. a 25 75 mm test tube. Label each tube appropriately.
Store in refrigerator for 3 months. 11. To one of each patient and positive control
7. Indicator with acetylthiocholinesterase tubes, add 5 mL of the preincubation indicator solution.
inhibitor. To a 50-mL volumetric flask, add 500 L of Also incubate the negative control in this solution. Let
acetylcholinesterase inhibitor 10-3 mol/L. Bring to stand for 30 min.
volume with preincubation indicator solution. Prepare 12. To the other set of patient and positive control
fresh. tubes, add 5 mL of the indicator + acetylcholinesterase
8. Indicator with acetylthiocholinesterase inhibitor solution. Cap and let stand for 30 min.
inhibitor and acetylthiocholine. To a 50-mL 13. Pour substrate out of all tubes. Pipet any
volumetric flask, add 500 L of stock remaining substrate (with a Pasteur pipet) from bottom
acetylcholinesterase inhibitor plus 0.1 g of of tube, being careful not to gouge the gels. Wear
acetylthiocholine. Bring to 50 mL final volume with disposable gloves, taking care not to let the inhibitor
preincubation indicator solution. Prepare fresh. solution come into contact with your skin. Flush contents
WARNING: Acetylcholinesterase inhibitor is extremely down sink.
toxic, and great care should be used when handling. 14. To the set of patients and positive and negative
Never pipet by mouth, and wash hands thoroughly after controls that had the preincubation indicator solution,
use. add 5 mL of indicator and acetylthiocholine solution. Let
stand overnight.
Assay 15. To the set of patients and positive control that
Equipment: Any commercially available system for were preincubated with indicator + acetylcholinesterase
column gel electrophoresis. inhibitor, add 5 mL of indicator + acetylcholinesterase
1. Remove gels from storage container, shaking inhibitor + acetylthiocholine solution. Let stand
off excess buffer. Two tubes will be needed for each overnight.
patient and a positive control (aliquots of a positive
amniotic fluid should be stored in 70C freezer). One Evaluation (Figure 1)
tube will be needed for a negative control (normal Zones of positive acetylcholinesterase activity will be
amniotic fluid). (Each patient and the positive control seen in the gels incubated with acetylthiocholine as
will have separate tubes for the inhibitor.) distinct bands, with the distance of migration similar to
2. Insert gel tubes, unfilled end up, in the upper that of the positive control. The gels incubated with the
bath of the electrophoretic chamber. The top of the gel acetylcholinesterase inhibitor will show no bands in the
tubes should be flush (level) with the top of the silicone same area.
tube adaptor. Be careful not to touch the end of the gel. Tubes negative for acetylcholinesterase activity
Plug all unused tube holes with plastic stoppers. will show no banding pattern in either tube at this
3. Place 800 mL of the working buffer in the position.
lower bath of the electrophoretic chamber and 200 mL of
Notes
buffer in the top portion. Be certain that no bubbles have
1. Amniotic fluid that is highly positive for AFP
become trapped at the bottom of the gel tubes. Squirt
should be stored in small aliquots at 70C. Amniotic
some buffer (using a Pasteur pipet) into the top of the
fluid with normal levels of AFP should also be frozen.
tubes to free any bubbles.
These will serve as positive and negative controls,
4. Place upper bath over lower bath. Place cover
respectively, for acetylcholinesterase activity.
over upper bath, and plug in the two white plugs over the
2. If an amniotic fluid positive for
metal prongs.
acetylcholinesterase cannot be obtained, fetal calf serum
5. Turn the power on and pre-run the gels at a
can be used instead. Fetal calf serum contains a
constant voltage of 200 volts. Electrophorese the gels for
cholinesterase activity that migrates just slightly slower
2 hours.
than the true acetylcholinesterase found in amniotic
6. At the end of 2 hours, turn the power off.
fluid. The fetal calf serum activity is not inhibited by
Remove the lid from the upper portion of the bath.
BW 284C51.
Apply 75 L of amniotic fluid to each tube. Introduce 3. Amniotic fluid that is contaminated with fetal
the tip slowly through the buffer until it touches the gel blood may give false-positive results for
surface. Carefully layer the sample onto the gel. acetylcholinesterase activity. Fetal blood has this
7. Replace the cover lid, and plug in the two white activity. If the fluid is obviously bloody, a cell count and
plugs. a stain should be performed to determine the number of
8. Turn on power and set power at a constant fetal cells present. If there are over 60 million fetal cells
voltage of 200 volts. Electrophorese at 200 volts for 2 per milliliter, positive results should be interpreted with
hours. caution. A repeat amniotic fluid specimen might be
9. When electrophoresis is complete, remove the appropriate.
gels from the tubes using a large syringe with a long,
thin needle. Fill the syringe with distilled water.
405
Cytomegalovirus (CMV)
Terry Pry and Greg Maine
Name: Cytomegalovirus (CMV)

Clinical significance: Refer to Chapter 7, Immunological Reactions, and Chapter 8, Immunochemical
Techniques, in the 5th edition of Clinical Chemistry: Theory, Analysis, and Correlation.
Principles of Analysis and Current Usage immune response is by definition abnormal. Although
Human cytomegalovirus (CMV) is a member of the quantitative culture methods have been used to predict
Herpesviridae family. CMV infections are very disease or follow a patients response to therapy, they
common, with 40% to 100% of the general population are relatively insensitive. Commercially available
showing prior exposure by serology [1]. The majority of molecular methods have been introduced to complement
persons with acute CMV will be asymptomatic. CMV is nonmolecular methods and include diagnostic CMV
capable of infecting and damaging many different cell methods such as real-time PCR [5]. Emphasis is on the
types, such as salivary glands, pancreas, adrenals, lung, search for improved diagnostic testing for early detection
liver, eye, ear, placenta, gastrointestinal tract, heart, of the increased viral replicative activity that presumably
ovaries, skin, and various components of the central precedes the onset of CMV disease in infected
nervous system. The ability of CMV to infect white individuals.
blood cells facilitates dissemination of virus within the
host. As with other herpes viruses, CMV shares a CMV-IgM Test
characteristic ability to remain dormant within the body CMV-IgM assays are currently based on ELISA
over a long period, reactivating and shedding when the technology. A specific IgM response may indicate a
hosts immune system is suppressed or compromised. recent primary infection, or it may be associated with a
nonprimary CMV infection in the following cases: (1)
Complications of primary CMV infection in IgM levels may occasionally persist for up to several
immunocompetent persons are rare except in newborns. years following primary infection, and (2) detection of
Intrauterine transmission of primary CMV infection false-positive IgM antibodies may occur in patients with
(approximately 40%), especially during the first autoimmune diseases such as systemic lupus
trimester, has the potential to cause significant fetal erythematosus (SLE) due to presence of rheumatoid
damage and subsequent severe pathology in newborns. factor (RF), since CMV can establish latency in the host
CMV is the most common congenital infection, [6]. Pregnant women with CMV IgG- and IgM-positive
occurring in approximately 1% of all live births [2]. results require further testing using a CMV IgG avidity
CMV is also associated with significant disease and may assay as described below. In the absence of a
be life threatening in immunocompromised individuals. commercially available gold standard CMV IgM assay,
Among the herpesviruses, CMV is responsible for most comparisons between various commercially available
morbidity and mortality in immunocompromised hosts assays yield variable results, making it difficult to assess
[3]. CMV is also a serious pathogen in patients who have differences in sensitivity and specificity observed
received an organ or bone marrow transplant. Estimates between different commercial assays [7]. The growth
indicate that one third of transplant recipients experience and purification of the virus, as well as the source of the
CMV disease, and up to half will have lung infections, viral antigen utilized in the CMV IgM assay, affects
resulting in a patient mortality rate of up to 50% [4]. assay sensitivity and specificity with state-of-the-art
assays utilizing recombinant antigens [8].
Laboratory methods for the diagnosis of CMV are
evolving, and considerable progress has been made over CMV-IgG Test
the past 10 years in the design of improved diagnostic Semi-quantitative CMV-IgG assays are also based on
techniques. Their use can clearly inform patient ELISA. This assay documents a past infection, as well as
management. Traditional methods for laboratory indicating the potential for an active infection, since
diagnosis of CMV infection include serology, virus reactivation of a latent infection or reinfection with
culture, modified culture (shell vial assay), and antigen another strain of the virus may occur. An IgG assay may
detection (antigenemia and NAT). Since CMV can be also identify an active infection by demonstration of IgG
transmitted from mother to fetus or from organ donor to seroconversion or a significant IgG rise between paired
recipient, automated serological assays have proven sera taken 2 to 3 weeks apart [9]. Because no
useful in the prenatal diagnosis of congenital CMV international standard exists for CMV IgG, definition of
infection and the determination of donor/recipient either a protective antibody level or a CMV-IgG
serostatus for CMV. CMV IgG avidity testing in international unit (IU) has not yet been defined.
pregnant women, especially during the first trimester, Comparisons between commercial IgG assays utilize
can be used to distinguish between a primary and arbitrary units and correlate well with each other.
nonprimary CMV infection, thereby stratifying the risk
of transmission of maternal CMV infection to the fetus. CMV IgG Avidity Test
Serological assays have limited utility in The IgG avidity assay measures the functional binding
immunocompromised patients whose ability to mount an affinity (Avidity Index) of the IgG class of antibody in
406
response to infection. Avidity is a measure of the overall of CMV [17,18]. Quantitation of CMV DNA is
strength of binding of an antigen with many antigenic performed by dedicated instruments and involves DNA
determinants and multivalent antibodies, whereas affinity extraction and analysis. Primers used in different assays
refers to the strength of binding between a single may be derived from different viral genomic sites, and
antigenic determinant and an individual antibody hence there may be variation between manufacturers.
combining site. The IgG avidity assay has been Quantitative PCR is available as commercial kits using
implemented in a diagnostic algorithm for pregnant various technologies. Most assays utilize the highly
women [10]. This diagnostic algorithm is based on conserved early or immediate-early genes. Real-time
screening pregnant women with a CMV IgG assay and a PCR CMV assays provide sensitivity and specificity
sensitive IgM assay, followed by reflex testing of CMV equivalent to that of conventional PCR combined with
IgG and IgM-positive specimens with a CMV IgG Southern blot analysis, and since amplification and
avidity assay. Interpretation of a positive IgM assay in detection steps are performed in the same closed vessel,
pregnant women is assisted by determination of the these assays do not suffer from false-positive results due
avidity index of CMV-specific IgG, since low-avidity to contamination. The presence of CMV DNA in plasma
IgG is produced during primary CMV infection. As the may be associated with active viral replication and
infection progresses, the avidity of the IgG increases, disease development [19]. Uniform reporting of results
and therefore high-avidity IgG is a marker of between laboratories is currently lacking and will benefit
nonprimary CMV infection. Detection of low-avidity from assay standardization and uniform guidelines.
CMV IgG in specimens from pregnant women indicates
that primary CMV infection has occurred within the past The detection of CMV viral transcript mRNA is now
18 to 20 weeks, whereas detection of high-avidity CMV performed using nucleic acid sequencebased
IgG excludes primary infection [11]. amplification that allows specific amplification of
unspliced viral mRNAs in a DNA background. This is a
CMV Neutralization Test promising technique for monitoring CMV infection in
The presence of neutralizing antibodies can be utilized to transplant recipients [19].
distinguish primary from nonprimary infection. These
antibodies appear shortly after primary infection (13 to Reference and Preferred Methods
15 weeks), and therefore the presence of high titer Since internationally recognized reference methods or
neutralizing antibody during a CMV infection may be standards do not yet exist for various CMV diagnostic
used to rule out a primary infection [12]. Neutralization methods, it is important for users to identify the specific
assays have been applied to an ELISA format but have method utilized with each test result.
not yet been commercialized because they are labor
intensive. Virological and molecular detection of CMV and
serological demonstration of a specific immune response
Virus Isolation in Tissue Culture are used for diagnosis. An acute infection is usually
Fetal infection can be assessed by detection of CMV diagnosed through detection of the virus in body fluids
virus in cultured amniotic fluid (AF). Virus isolation or deduced by the presence of a specific IgM antibody
from AF is considered the gold standard method for immune response along with the seroconversion from
detection of fetal infection, owing to the high sensitivity the IgG antibody negative to IgG antibody positive.
and 100% specificity of the method. Inoculated cultures Serological tests are highly specific and sensitive, and
are prepared using human diploid fibroblast cells [13]. although antibody can decline with age and severe
The shell vial assay is the preferred method because of immunosuppression, IgG seropositivity is usually
its shortened incubation, providing results within 16 to lifelong. Techniques for detecting the virus itself are
72 hours. The method utilizes immunofluorescence to improving rapidly, and nucleic acid tests such as
detect binding of monoclonal antibodies to viral proteins polymerase chain reaction are now available at large
synthesized shortly after infection. The method has been centers. The accepted gold standard for the diagnosis of
described in numerous publications and has been studied congenital CMV infection is the detection of virus in
either exclusively or in combination with PCR methods urine or saliva within 2 weeks of birth.
[14-16].
Specimen
PCR Test CMV is detected and quantified most commonly using
Real-time PCR combines PCR chemistry with plasma, but other specimens include amniotic fluid,
fluorescent probe detection of amplified product in the cerebrospinal fluid, urine, throat washing, semen, and
same reaction vessel. In general, both PCR and tissue samples, particularly with the use of PCR. In
amplified product detection are completed in an hour or detecting congenital CMV, isolation of CMV utilizes
less, which is considerably faster than conventional PCR saliva or urine within 2 weeks of birth. Neonatal blood
detection methods. dried on paper has also been utilized for viral DNA
testing [20]. CMV virus is very labile, and tissue culture
One of the earliest applications of real-time PCR for samples should be kept at 4C to 8C and tested within
testing in the clinical microbiology laboratory was for 48 hours. Freezing of amniotic fluid specimens destroys
detection of CMV, and an extensive literature exists CMV infectivity [13].
describing considerations and implementation of
applying real-time PCR to the routine laboratory testing Interferences
407
CMV-IgM Assay Philadelphia, USA, 1995, W.B.Saunders,

Detection of false-positive IgM antibodies may occur in pp.312-353.
patients with autoimmune diseases such as systemic 3 Rubin RH, Cosimi A, Tulkoff-Rubin NE,
lupus erythematosus (SLE) and due to the presence of Russell PS, Hirsch MS. Infectious disease
rheumatoid factor (RF). Cross-reactivity with Epstein- syndromes attributable to cytomegalovirus
Barr virus (EBV) and polyclonal stimulation by EBV recipients and their significance among renal
can also result in false-positive results. Conventional transplant recipients, Transplantation 1977; 24:
PCR may exhibit false positives due to possible 458-464.
contamination by amplification products. 4 Schmidt GM, Horak DA, Niland JC, Duncan
SR, Forman SJ, Zaia JA. A randomized,
Interpretation controlled trial of prophylactic ganciclovir for
Diagnosis of primary infection may be achieved in the cytomegalovirus pulmonary infection in
majority of cases through concurrent analysis of serum recipients of allogenic bone marrow transplants,
antibodies, virus detection, and clinical signs. N Engl J Med 1991; 324: 1005-1011.
5 Szczepura A, Westmoreland D, Vinogradova Y,
Assessment of CMV infection in pregnancy (Figures 1a Fox J, Clark M. Evaluation of molecular
and 1b) begins with maternal serology, which should techniques in prediction and diagnosis of
establish recent primary, latent, or nonprimary infection. cytomegalovirus disease in
Primary CMV infection occurs when the CMV virus immunocompromised patients, Health
infects a previously uninfected, seronegative individual; Technology Assessment 2006; 10: 1-176.
primary infection is related to worse fetal outcomes. 6 Krast YJ, Stals FS, Christiaans MH, Lazzarotto
Seroconversion can be demonstrated by the production T, Landini MP, Bruggeman, CA. IgM antibody
of CMV IgM and IgG antibodies (low avidity) in a detection of ppUL80a and ppUL32 by
seronegative individual. Documentation of CMV immunoblotting: an early parameter for
seroconversion is rare in pregnant women, since they are recurrent cytomegalovirus infection in renal
not routinely screened for CMV antibodies prior to transplant recipients. J Med Virol 1996; 48:
gestation. In these cases, the detection of CMV IgM has 289-294.
been used as a marker of active or recent CMV infection 7 Lazzarotto T, Brojanac S, Maine GT, Landini
[21]. Since CMV-specific IgM can be produced during MP. Search for cytomegalovirus-specific
nonprimary CMV infections, a CMV IgM-positive result immunoglobulin M: Comparison between a
by itself does not permit an accurate assessment of the new Western blot, conventional Western blot
risk of CMV infection to the fetus. Nonprimary infection and nine commercially available assays. Clin
occurs when a previously infected, seropositive Diagn Lab Immunol 1997; 4: 483-486.
individual experiences (1) reactivation of a previous 8 Maine GT, Lazzarotto T, Landini M. New
CMV infection from latency or (2) reinfection with a Developments in the diagnosis of maternal and
new strain of CMV. Nonprimary CMV infection is congenital CMV infection, Expert Rev Diagn
usually characterized by the presence of high avidity 2001; 1: 19-29.
CMV IgG antibodies. The IgG avidity assay is a 9 Rawlinson WD. Broadsheet Number 50:
powerful tool, but it should be used and interpreted Diagnosis of human cytomegalovirus infection
properly. Interpretable results can be achieved mainly and disease, Pathology 1999; 31: 109-115.
for sera obtained within the first trimester of pregnancy. 10 Munro SC, Hall B, Whybin LR, Leader L,
Robertson P, Maine G, Rawlinson WD.
CMV Performance Goals Diagnosis and Screening for Cytomegalovirus
The 2007 College of American Pathologists (CAP) in Pregnant Women. J Clin Microbiol 2005; 43:
Quality Assurance Program (QAP) listed the frequency 4713-4718.
of the following anti-CMV (total) methodologies among 11 Lazzarotto T, Spezzacatena P, Pradelli P, Abate
171 participants: enzyme immunoassay (43%), D, Varani S, Landini MP. Avidity of
fluorescent EIA (9%), particle agglutination (18%), and immunoglobulin G directed against human
solid phase red cell ADH (25%). Satisfactory qualitative cytomegalovirus during primary and secondary
evaluation criteria (90% participant consensus of infection in immunocompetent and
results against target) was achieved within each method immunocompromised subjects, Clin Diag Lab
group for all samples, save for particle agglutination, Immunol 1997; 4: 469-473.
which only achieved 36% consensus. 12 Eggers M, Metzger C, Enders G. Differentiation
between acute primary and recurrent human
References cytomegalovirus infection in pregnancy, using a
1 de Jong, MD, Galasso GJ, Gazzard B, Griffiths microneutralization assay, J Med Virol 1998;
PD, Jabs DA, Kern ER et al. Summary of the II 56: 351-358.
International Symposium on Cytomegalovirus, 13 Hodinka RL, Human cytomegalovirus. In:
Antiviral Res 1998; 39:141-162. Murphy PR, Baron EJ, Pfaller MA, Tenover
2 Stagno S. Cytomegalovirus. In: Infectious FC, Yolken RH, editors. Manual of Clinical
Diseases of the Fetus and Newborn Infant (4th microbiology. 1999;p.889-991.
Edition). Remington JS, Klein JO, (Editors). 14 Mendelson E, Aboudy Y, Smetana Z,
Tepperberg M, Grossman Z. Laboratory
408
assessment and diagnosis of congenital viral Microbiology: Applications for Routine

infections: Rubella, cytomegalovirus (CMV), Laboratory Testing, Clin Micro Rev 2006; 19:
varicella-zoster virus (VZV), herpes simplex 165-256.
virus (HSV), parovirus B19 and human 19 Gerna G, Baldanti F, Lilleri D, Parea M,
immunodeficiency virus (HIV), Reproductive Alessandrino E, Pagani A et al. Human
Toxicology 2006; 21: 350-382. cytomegalovirus immediate-early mRNA
15 Weiner CP, Grose C. Prenatal diagnosis of detection by nucleic acid sequence based
congenital cytomegalovirus infection by virus amplification as a new parameter for pre-
isolation from amniotic fluid. Am J Obstet emptive therapy in bone marrow transplant
Gynecol 1990; 163: 1253-1255. recipients, J Clin Microbiol 2000; 38: 1845-
16 Nicolini U, Kustermann A, Tassis B, Fogliani 1853.
R, Galimberti A, Percivalle E, et al. Prenatal 20 Barbi M, Binda S, Caroppo S. Diagnosis of
diagnosis of congenital cytomegalovirus congenital CMV infection via dried blood spots,
infection. Prenat Diagn 1994; 14: 903-906. Rev Med Virol 2006; 16: 385-392.
17 Boeckh M, Huang M, Ferrenberg J, Stevens- 21 Nielsen SL, Sorensen I, Andersen HK. Kinetics
Ayers T, Stensland L, Nichols WG, et al. of specific immunoglobulins M, E, A and G in
Optimization of quantitative detection of congenital, primary and secondary
cytomegalovirus DNA in plasma by real-time cytomegalovirus infection studied by antibody-
PCR. J Clin Microbiol 2004; 42: 1142-1148. capture enzyme-linked immunosorbent assay, J
18 Espy M, Uhl J, Sloan L, Buckwalter S, Jones Clin Microbiol 1988; 26: 654-661.
M, Vetter E et al. Real-Time PCR in Clinical
Table 1: Summary of CMV Methods
Method 1. CMV IgG

Principle: ELISA methodology
Usage: Marker of past CMV infection
Comments: Limitation of result: Not conclusive of latent or active infection. Comparisons between
commercial IgG assays correlate well with each other, but no international standard yet exists.
Method 2. CMV IgM
Principle: ELISA methodology
Usage: Marker of active or recent CMV infection
Comments: Improved CMV IgM assays employing recombinant antigens or peptides are a significant
advance over conventional viral lysate-based immunoassays. IgM antibodies may persist for months or years.
Method 3. CMV IgG Avidity
Principle: Two CMV IgG ELISA tests are performed on the same sample. After primary antibody incubation,
the solid phase of each assay is washed with either buffer (control) or a chaotropic reagent (urea, diethylamine) to
remove low-avidity antibodies. After another wash, the conjugate is added and the signal measured. The avidity
index is the ratio of (CMV IgG titer with chaotropic wash step/CMV IgG titer with buffer wash step) 100, that
is, the ratio of CMV high-avidity IgG over the total CMV IgG titer.
Usage: Differentiate between primary infection (low avidity IgG) and recurrent or persistent infection
(high avidity IgG).
Comments: Avidity testing should be performed early in gestation (within the first trimester). Fully
automated high throughput analytical systems available with integrated avidity assays.
Method 4. Fetus Culture (Shell Vial Assay)
Principle: Gold standard for detection of fetal infection, owing to 100% specificity. Immunofluorescence
can confirm presence of CMV virus.
Usage: Virus isolation from amnionic fluid remains key method for demonstration of fetal infection and
is used either exclusively or in conjunction with molecular methods.
Comments: Strict handling of sample due to labile virus. CMV identified after 16 to 72 hours in culture
Method 5. Real-Time PCR
Principle: DNA extraction, amplification of DNA, detection using fluorescent probes.
Usage: Detection of viral DNA in clinical samples
Comments: General agreement that PCR is more sensitive than culture. Samples can be frozen for
subsequent analysis by other labs.
Figure 1
409
Figure 2. Proposed diagnostic algorithm for CMV serology screening in pregnant women.
410
Copper
Copper
K. Raja, R. Swaminathan
Name: Copper
Principles of Analysisi (BCA) [7] and 3-[2-[2-(2-hydroxyimino-1-methyl-

propylideneamino)-(ethylamino]-ethyl-imino]butan-2-
Copper, an essential yet potentially toxic element, is a one oxime [8]. Some of these methods have been adapted
transition group IB metal in period 4 of the periodic table. for use in clinical laboratories with commercial reagent
Copper forms two series of inorganic salts: cuprous and kits being available for the determination of serum copper.
cupric salts. The cuprous salts are unstable in solution, In the method using 3,5,Di-Br-PAESA, copper bound to
converting readily into the corresponding cupric salts. ceruloplasmin is released by a reducing agent at pH 4.7;
the released copper forms a stable complex with the
Analytical techniques applied to the measurement of chromogen [1].
copper in biological fluids and tissues include
spectrophotometry (Table 1, Method 1) [1-8], flame (Table In flame atomic absorption spectrophotometry (Table 1,
1, Method 2) [9-12], and flameless (Table 1, Method 3) Method 2), monochromatic light from a hollow-cathode
[13-17] atomic absorption spectrometry, inductively copper lamp is absorbed by ground-state atoms in a lean
coupled plasma mass spectrometry (ICP-MS)(Table 1, air-acetylene flame. The amount of light absorbed at 324.7
Method 4) [18], and inductively coupled plasma optical nm is directly proportional to the concentration of atoms in
emission spectrometry (ICP-OES) (Table 1, Method 5) the gaseous state in the light path and hence to the
[19]. Other methods using emission spectrography [20] concentration of copper in the solution. Sensitivity will
neutron activation analysis [21], anodic stripping vary depending on the kind of burner used and on burner
voltammetry [22] and total reflection fluorescence analysis slit length.
[23] have also been described in the literature. However,
they are not always practical to be used in busy clinical A number of publications have described methods for the
laboratories. Flame atomic absorption spectrophotometry measurement of serum and urine copper by flameless
remains by far the most widely used method, followed by atomic absorption spectrophotometry (Table 1, Method 3)
ICP-OES and ICP-MS. A few laboratories, however, still [13,14,17]. The sensitivity of this method is sufficient for
use colorimetric methods. measurements to be made directly on urine or on a 10- to
50-fold dilution of serum. Although acidification of urine
Many spectrophotometric methods that utilize the has been recommended by some authors, its necessity
formation of colored copper complexes have been remains questionable.
published. Chromogens used in the analysis of copper
include the following: 4-(3,5-dibromo-2-pyridylazo)-N- Inductively coupled plasma optical emission/mass
ethyl-N-sulfopropylaniline (3,5,Di-Br-PAESA) [1], 2- spectrometry (ICP-OES/MS) are relatively new techniques
(5bromo-2-pyridylazo(-5-(n-propyl-N-sulfopropylamino) in clinical laboratories and have the potential to measure
aniline [2], multiple elements within a single sample [18,19,24]. This
2-(5-nitro-2-pyridylazo)-5-(N-propyl-N-sulfopropylamino) capability saves time and volume of sample required for
phenol (nitro-PAPS) [3], ,,,-tetrakis(4-N- analysis. The instruments are relatively expensive to
trimehylaminophenyl)porphine [4], 2,5-dimercato-1,3,4- purchase and operate. The methods can, however, detect
thiadiazole(DMTD) [5], di-2-pyridyl ketone many elements besides copper at very low concentrations
benzoylhydrazone (dPKBH) [6], 2,2-bicinchoninic acid in various sample types and over a wide dynamic range. In
these techniques, the sample is introduced into the
inductively coupled plasma (ICP) as an aerosol. The larger
i
Copper droplets are excluded from entry via the spray chamber,
Previous and current authors of this method: but the smaller droplets get swept into the ICP, which is a
First edition: Not done hard but rapid atomization/ionization technique. The
Methods edition: Nancy W. Alcock very high temperature of the plasma leads to the aerosol
Second edition: Not updated droplets being desolvated, vaporized, atomized, and/or
Third edition: Not updated ionized. The ionization efficiency is a function of the
Fourth edition: Not updated elemental ionization potential, as well as plasma
Fifth edition: K. Raja, R. Swaminathan temperature and sample matrix. In ICP-OES, the
characteristic radiation emitted by the excited atoms/ions is
411
Copper
detected and electronically converted to elemental extreme care is required to avoid contamination, especially
concentrations. In ICP-MS, the ions are extracted through at the low concentrations when using EAAS.
a series of cones prior to reaching the analyzer vacuum
compartment. The mass spectrometer separates the ions When atomic absorption/ICP spectrometry is not available,
according to their mass-to-charge (m/z) ratio. As the ions a colorimetric method based on 3,5-Di-Br-PAESA can be
produced within the plasma are virtually singly-charged used either manually or on an automated instrument.
ions, the m/z ratio equates to the mass of the ion. The Reagents for this method can be obtained form commercial
signal measured by the detector (usually an electron sources in the form of a kit (e.g., Randox copper assay). In
multiplier) is thus proportional to the concentration of the this method, the sample or standard is mixed with reagent
element(s) in the sample. The most common type of mass containing ascorbic acid and allowed to stand for 60
analyzer used in ICP-MS instruments is a quadrupole. seconds before taking an absorbance reading. The
However, time of flight and high-resolution magnetic chromogen is added, and the final absorbance is taken 5
sector analyzers are also used in some instruments, where minutes later.
very accurate mass resolution of complex mixtures is vital.
The process of ICPMS can also be applied to ascertain the Interferences and Contaminants
analyte concentration in a sample via isotope dilution As methods have progressively become more sensitive,
analysis, provided the isotopic composition is known. With thereby allowing measurements to be made on small
this method, a known quantity of the analyte enriched in volumes and down to trace (ppm/ppb)/ultratrace (ppb/ppt)
one isotope is added to the sample, and the resulting levels, the significance of contamination as a serious
change in isotope ratio is used to calculate the source of error during the analytical process has become
concentration of the element in the sample. even more apparent. An assessment of the potential source
of contamination and the need to take steps to prevent it
The process of ICP-MS does, however, suffer from are therefore important when contemplating analytical
spectral, matrix, and physical interferences, with spectral work from sample collection to the final analysis.
interferences being probably the most serious. For
example, two or more atomic ions arising from the Contamination can arise from the environment, the
matrix/solvent/plasma can combine to form a molecular reagents, apparatus, or even the operator. Environmental
(polyatomic) species of equivalent mass to the contamination is the presence of particulate/gaseous matter
element/isotope of interest. The analysis of 63Cu can be in the air, which can arise from laboratory vents and/or the
affected by the interaction between 40Ar and 23Na [24]. air-conditioning unit(s) (especially when associated with
Interferences can be reduced by the use of a reactant gas in inefficient filtration), the slow deterioration of solid
a collision or reaction cell placed between the extraction objects, paints, cements or other construction materials,
optics and mass analyzer. The interference is diminished plastics, or from chemical processes upon materials. Any
by the reactant gas (e.g., ammonia), either converting the ongoing building works or high volume of traffic in the
interfering polyatomic to a different species or reducing its vicinity of the laboratory can also contribute markedly to
kinetic energy. the degree of contamination. It has been estimated that air
within an analytical lab can contain as much as 200 g/m3
Currently, the most widely used method for the of particulate matter, including copper [25]. Whereas the
measurement of copper in body fluids or tissues is either heavier particles tend to settle, the lighter particles can stay
flame or flameless atomic absorption spectroscopy. Only airborne and can, depending on the ventilation system and
6% of laboratories participating in the UK National air currents within a laboratory, be carried appreciable
External Quality Assurance Service use a colorimetric distances. There are a number of ways of avoiding or
method for copper estimation. minimizing particulate/environmental contamination. The
more sophisticated and expensive approaches include the
Reference and Preferred Methods construction of a clean room. The air entering the room
A reference method for copper has not been published. In gets filtered using HEPA (high-efficiency particulate air)
the absence of a reference method, flame atomic filtrationwhich has the capacity to remove the vast
absorption spectrophotometry (FAAS) is the method of majority of particulate material (>99.9% of particles that
choice when the concentration is 1.5 mol/L (96 g/L) or are 0.3 m or larger in size). In addition, the room could
greater. Flameless or electrothermal atomic absorption also be maintained at positive pressure so that the flow of
spectrophotometry (EAAS) is more sensitive than FAAS; the air is in one direction only (i.e., from laboratory to
the detection limit is around 30 nmol/L (2 g/L). However, outside). Very specialized centers may have, in addition,
EAAS is also more sensitive to low levels of an antechamber area that separates the main laboratory
contamination. The EAAS method may be preferable for from the clean room and provides an area where the
untreated urine. The assay time is, however, longer for analyst(s) can put on clean apparel.
EAAS typically minutes as compared to seconds per
specimen when using the flame instrument. This is the More practical approaches, however, include use of
main disadvantage of the flameless technique. In addition, laminar flow hoods. These self-contained units provide
412
Copper
HEPA-filtered areas for the safe handling of samples, The most popular plastics for trace analysis are
reagents, standards, and so forth. To avoid the possibility Perfluoroalkoxy (PFA) and high-/low-density polyethylene
of any hazardous materials being blown at the analyst, (HDPE/LDPE). PFA is not only resistant to inorganic
exhausted laminar flow systems are recommended. acids (HF, HNO3, HCl)/bases and oxidizing agents, but
However, correct operation of the unit (e.g., the can also be microwaved, thus providing an ideal material
appropriate positioning of front shielding panel) is for microwave digestion vessels. However, the cost for this
necessary for optimal performance. type of plastic is high. LDPE or HDPE bottles are also
resistant to inorganic acids/bases and are typically used for
In laboratories, where the likelihood of a clean- storing diluted samples/sample digests or working
room/laminar-flow hood is not possible, owing to the cost standards. These bottles can withstand solutions of 10%
and/or the availability of space, the contamination risk can v/v HNO3 over extended periods of time.
be minimized by ensuring good laboratory working
practices. Usage of an ordinary enclosed room combined Contamination from glassware can be reduced by soaking
with decent filtration of air vents/air-conditioning unit and it overnight in a dilute detergent solution (e.g., Decon at
regular cleaning of walls, equipment, and floors with 5%), rinsing with deionized water, followed by soaking in
deionized water/detergent (and lint-free towels or dust- 10% HNO3 for at least 24 hours and rinsing thoroughly
adsorbing cleaning tools) may be adequate to ensure the (3) with good quality deionized water. Alternatively, high
production of reliable results within an analytical setting. quality quartz (type III) may be used, as this form of glass
Sensible approaches to the working practices, such as tends to have a lower level of elemental contamination.
avoiding unnecessary shelving/cluttering/storage (which Some workers also recommend soaking or rinsing of
can accumulate dust particles), working on easily cleaned plastic ware in dilute acid prior to rinsing with deionized
plastic sheets/trays, minimizing the number of water. All lab ware should be air or oven dried before use.
transfers/preparative steps and storage of
reagents/consumables in clean, closed cupboards/snap-top It is important to note that often it is the contaminants from
plastic boxes helps to further reduce contamination. the atmosphere and/or lab ware that is the cause of
interference rather than the impurities from the water.
The analysis of copper involves some initial preparative
step(s) of the sample, so that the resulting solution is as The analyst should also take every precaution in
homogenous/particle free as possible. Reagents, if not of preventing contamination. Careless manipulation of
suitable high quality, can thus introduce contaminants equipment or lab ware (e.g., by using bare hands) can
within the assay, especially when the analyst is dealing cause contamination arising from dried or dead skin,
with low concentrations. This factor becomes even more sweat, and so forth. Use of hand lotions or creams can also
important when considering the fact that the volume of be a potential contaminant. Touching a dirty or metallic
reagents used is often in excess of the volume of sample. surface, with or without gloves, before proceeding with a
High-quality reagents should also be dedicated solely for protocol can also be a further source of contamination.
trace-metal work and must not be used for general Wearing of jewelry should also be avoided in the
laboratory work. Moreover, pipetting should never be laboratory, since it can contribute to contamination if it
carried out from the original reagent stock bottle. Always comes into contact with lab ware or a sample. The analyst
transfer a small volume of the reagent to a suitable should also wear a fastened, lint-free lab coat and powder-
container prior to use. Most preparations/dilutions within free gloves when carrying out all analytical procedures.
the analytical laboratory involved in elemental analysis
also require the use of deionized/double distilled water as Specimen
the primary reagent. A large proportion of laboratories use Blood
water that is produced by commercially available reverse Venous blood should be collected into a suitable
osmosis (i.e., passage of water through a membrane disposable vacutainer, using either a venipuncture vacuum
against osmotic pressure) and/or ion-exchange (i.e., system or an indwelling butterfly cannula (attached to a
passage through a resin) purification systems. Many plastic syringe). If possible, the first 5 mL of blood should
commercial companies (e.g., Elga) make use of both be discarded or used for tests other than elemental analysis
technologies within the same unit. (e.g., for general biochemistry). Subsequent collections
should then be made in suitable trace metalfree
Another potential source of contamination is the apparatus vacutainers and used for determination of copper and other
(e.g., containers, sample vials) that comes into contact with elements. There are currently three suppliers of suitable
the sample. Glass, a material that is extensively used blood collection tubes: Beckton Dickinson, Greiner, and
within a laboratory, is a potential source of elemental Sarstedt. Although plasma can be used for analysis,
contamination and should, wherever possible, be replaced heparin, which is used as anticoagulant, may be
with plastic. The analyst should in fact attempt to use contaminated with copper. Trace metalfree tubes for
plastics for every step of the protocol, namely sample pediatric samples are currently unavailable. To avoid the
collection, preparation, and storage of reagents/standards. possibility of contamination, pediatric samples should be
413
Copper
collected in anticoagulant-free tubes and centrifuged fairly post/van/courier. If there is to be a delay in transportation,
soon after clotting to obtain the serum fraction. the samples can be kept refrigerated or in a freezer. To
The transfer of serum or plasma to secondary tubes standardize calculations and interpretation of results, and
constitutes a further possible source of contamination. to correct for differing fluid content, tissues need to be
Trace metalfree tubes (e.g., Teklab) or polystyrene tubes completely dried prior to analysis.
(Sarstedt) or Hitachi cups (Clinicon) may be suitable. It is
preferable not to use colored caps or caps with o-rings to Hair
avoid the possibility of contamination. It is advisable to Though easy to obtain, this sample may suffer from
check each new batch of primary/secondary collection contamination from sweat, the environment (dust), water,
tubes for contamination by using water/dilute and previous beauty formulas (e.g., shampoos, hair dyes,
acid/serum/plasma as a matrix [26]. and other treatments) [28]. Diet has also been reported to
have an influence [29]. Moreover, because concentrations
Capillary blood sampling is more prone to contamination are affected by distance from the scalp, age, and rate of
and should be best avoided. hair growth, the use of hair copper concentration reflecting
nutritional status remains questionable [30].
Urine
A timed collection (24-hour) is the preferred specimen Reference Intervals
when assessing copper output, especially in the diagnosis Table 2 gives the reference ranges for serum and urine
of Wilsons disease [27]. It is necessary for collections to copper. Serum copper concentrations in females are
be made into acid-washed or copper-free plastic ware slightly higher than in males. Concentrations in children
containers. Metal containers should not be used. Subjects younger than 1 year are lower than in adults.
should also refrain from collecting urine in another
container and then transferring it to the allocated container. Table 2. Reference Values for Serum and Urine
Acid washing is accomplished by swirling a 10% to 20% Copper
HNO3 solution within a clean/new plastic container. After Serum/Plasma Age mol/L mg/L
disposing of the acid, the container (including the lid) must 0-4 months 1.4-7.2 0.09-
be rinsed out three times with deionized or double distilled 0.46
water then dried at room temperature. Containers need to 4-6 months 3.9-17.3 0.25-1.1
be kept away from possible sources of contamination 7-12 months 7.9-20.5 0.5-1.3
during these steps. It is not necessary to acidify urine after Over 1 year 12-25 0.76-1.6
collection for the measurement of copper. After Adult males 11.0-22.0 0.7-1.4
ascertaining the weight/volume of the collected urine, an Adult females 12.6-24.3 0.8-1.55
aliquot could be prepared (i.e., transferred to a sterile Pregnancy: 16 27-40 1.7-2.5
plastic universal container) and kept refrigerated if there is weeks to term
to be a delay in the analysis. Wilsons <4.0 <0.25
disease
If refrigerated, samples should be left to attain room
Urine mol/24hrs g/24hrs
temperature before analysis. Furthermore, it is better to
Adults <0.95 <60
mix urine samples 25 to 30 minutes before analysis and
Wilsons >1.57 >100
then to leave them standing. This period allows the
disease
particulate matter (e.g., urate and phosphates) to settle.
Thus the sample that is subsequently aspirated is usually Post <12 <760
free from particulate interferences. Pilot studies using penicillamine
urine samples and a known copper standard solution have challenge
revealed that the risk of copper co-precipitating or being normal adults
adsorbed onto the precipitates is negligible. Post >25 >1590
penicillamine
Tissue, Including Liver Biopsies challenge
Samples should be obtained with care. Use clean and Wilsons
metal-free instruments, if possible, to obtain and disease
manipulate soft-tissue specimens. Moreover, use clean
tools for each specimen to minimize carry-over Interpretation
contamination. Once obtained, the samples should be Copper is an essential, yet potentially toxic element. It is
stored and transported in suitable containers. Freshly involved in various cellular processes, primarily as a
obtained liver biopsies should be placed on a clean piece cofactor of enzyme/protein systems involved in
of filter paper moistened with deionized/double distilled catecholamine metabolism, oxidative metabolism,
water, and then placed into a plastic sterile universal neurotransmitter synthesis, iron absorption, free radical
container. The sample could then be sent by scavenging (SOD) and collagen cross-linking (lysyl
oxidase) [31-33]. Body concentrations of copper are
414
Copper
usually low (100 to 150 mg), with the highest skeletal abnormalities, and the strikingly peculiar steely
concentrations being in the liver (10 to 20 mg) and brain. hair (possibly due to defective formation of disulphide
Copper homeostasis is controlled mainly at the level of bonds in keratin). Affected infants tend to be identified
excretion [34,35]. The average safe daily intake of dietary soon after birth, and most infants with the untreated, severe
copper in adults is 1.5 to 3.0 mg (24 to 47 mol) [36]. (classic) form of Menkes die before their fourth birthday.
Useful dietary sources include beans, peas, whole grain Diagnosis can be confirmed by low serum
products, liver, seafood (oysters, crab, lobster), meats, and copper/ceruloplasmin levels, abnormal catecholamine
nuts (almonds, walnuts). The intestine normally absorbs concentrations in blood and CSF, microscopic examination
about 30% to 50% of ingested copper via both passive and of hair, accumulation of copper in cultured fibroblasts
active (energy-dependent) mechanisms [37,38]. It is then (skin biopsy) or placenta, or prenatally by measuring the
transported to the liver, bound mainly to albumin, and to a copper content of chorionic villi in the first trimester.
lesser extent amino acids. The liver plays a key role in
maintaining a physiological concentration of plasma Toxicity of copper may arise from either a genetic disorder
copper and in controlling copper homeostasis. It takes up or from acute or chronic copper ingestion. Acute ingestion
the copper from the circulation via a carrier-mediated of copper salts produces nausea, vomiting, diarrhea,
process and incorporates it into ceruloplasmin (the blue circulatory collapse, intravascular hemolysis and in some
glycoprotein), metallothionein, or glutathione. cases death. Acute toxicity has been reported in subjects
ingesting copper-containing solutions (either deliberately
In plasma, 90% of copper is bound to ceruloplasmin, with or by accident), in patients given excessive
about 10% loosely bound or associated with albumin. A supplementation or low levels of other essential nutrients,
small fraction is complexed with low molecular weight and in those on maintenance hemodialysis (leaching of the
compounds such as amino acids. element from copper-containing dialysis membranes
[Cuprophan]). In situations of acute toxicity, serum copper
Frank clinical copper deficiency is unusual. It has concentration will be high, but the ceruloplasmin
however, been reported in malnourished children, in concentration will be normal.
infants on low-protein-diet, cows-milk formulas, after
long-term parenteral or enteral [39,40] nutrition, after Chronic poisoning with copper leads to gross hepatic
severe and persistent diarrheal illness, or when intake of copper overload with severe liver disease (cirrhosis) in
zinc has been excessive. Copper deficiency presents as a young children. Indian childhood cirrhosis has been
microcytic, hypochromic anemia with marked neutropenia ascribed to boiling or storage of milk in corroded copper or
and is resistant to iron therapy. Children and neonates may brass vessels. There have also been reports of poisoning in
also have defective collagen and elastin synthesis and may young children as a result of high copper content in well
develop bone disease (osteoporosis). water. Liver copper content in such cases can exceed that
found in overload due to Wilsons disease.
Subclinical copper deficiency may be more widespread
then previously thought and has been suggested as a risk Wilsons disease (hepatolenticular degeneration) is an
factor for cardiovascular disease [41]. autosomal recessive disorder caused by a defect in the
gene encoding for the P type ATP7B transporter, mapping
The rare X-linked recessive neurodegenerative disorder to chromosome 13 [43]. Well over 100 mutations have
Menkes disease (prevalence 1 in 250,000 births), is been identified so far, thus making routine DNA screening
characterized by a failure of copper transport (efflux) an unlikely proposition. The condition is associated with
across the intestinal mucosa due to a defect in the MNK impaired incorporation of copper into ceruloplasmin, and
gene located on long arm of chromosome 13, which codes its subsequent excretion via the bile. Copper consequently
for a copper transporter [42,43]. Although copper accumulates within the liver, where it can induce free-
concentrations are low in plasma, brain, and liver, copper radical formation. With progressive hepatic loading,
paradoxically accumulates in many other tissues of the copper diffuses into the blood and accumulates in other
body (i.e., kidney, spleen, and skeletal muscle) due to tissues, with damaging effects particularly on the brain.
abnormal copper transport. The intake into cells appears to Consequently, the subject presents with progressive
be normal, but there is defective utilization intracellularly. neurological symptoms (characterized by tremor,
The condition is therefore one of functional copper dysphagia, dystonia, incoordination, and unusual
deficiency as a result of a deficiency of copper-dependent behavior), low serum concentrations of copper and
enzymes. As there are many variations of Menkes disease, ceruloplasmin, raised urinary copper excretion, and
symptoms range from mild to severe, with onset occurring characteristic copper deposits in the Descemets membrane
in infancy (classical form) or childhood/adulthood (milder in the limbus of the cornea (Kayser-Fleischer rings) [44].
variants such as occipital horn syndrome). Patients may
thus present with progressive cerebral degeneration, Copper Performance Goals
mental retardation, poor immune function, poor thyroid The current target for acceptable performance criteria is
function, depigmentation, severe anemia, hypotonia, 7.7% [45]. It is evident from the 2007 College of
415
Copper
American Pathologists (CAP) survey that many methods 13 Matousek JP, Stevens BJ. Biological applications
do not achieve this target. Atomic absorption with of the carbon red atomizer in atomic absorption
deuterium background correction is the best of the current spectroscopy: preliminary studies on iron, copper,
methods used in the United States. lead and zinc in blood and plasma. Clin
Chem1971; 17: 363-368.
References 14 Sunderman FW-Jr. Electrothermal atomic
1 Abe A, Yamashita S, Noma A. Sensitive, direct absorption spectrophotometry of trace elements in
colorimetric assay for copper in serum. Clin biological fluids. Anal Clin Lab Sci 1975; 5: 421-
Chem 1989; 35: 552-554. 433.
2 Makino T. Sensitive, direct colorimetric assay of 15 Alcock NW. Matrix effects in flameless atomic
serum copper using 5-Br-PSAA. Clin Chim Acta absorption of copper. Clin Chem 1984; 30: 973.
1989; 185: 7-16. 16 Halls DJ, Fell GS, Dunbar P.M. Determination of
3 Yamashita S, Abe A, Noma A. Sensitive, direct copper in mine by graphite furnace atomic
procedures for simultaneous determinations of absorption spectrometry. Clin Chim Acta 1981;
iron and copper in serum, with use of 2-(5-nitro- 114: 21-27.
2-pyridylazo)-5-(N-propyl-N- 17 Polyakova E, Shuvaeva O. Determination of
sulfopropylamino)phenol (nitro-PAPS) as ligand. calcium, magnesium, iron, copper, zinc and
Clin Chem 1992; 38: 1373-1375. phosphorus in blood serum by arc atomic
4 Makino T, Itoh JI. A highly sensitive colorimetric absorption spectrometry. J Anal Chem 2005; 60:
determination of serum copper using alpha, beta, 937-941.
gamma, delta-tetrakis(4-N- 18 Townsend AT, Miller KA, McLean S, Aldous S.
trimethylaminophenyl)porphine. Clin Chim Acta The determination of copper, zinc, cadmium and
1981; 111: 1-8. lead in urine by high resolution ICP-MS. J Anal.
5 Ahmed MJ, Jahan I, Banoo S. A simple Atomic Spectrum 1998; 13: 1213-1219.
spectrophotometric method for the determination 19 Rahil-Khazen R, Henriksen H, Bolann BJ, Ulvik
of copper in industrial, environmental, biological RJ. Validation of inductively coupled plasma
and soil samples using 2,5-dimercapto-1,3,4- atomic emission spectrometry technique (ICE-
thiadiazole. Anal Sci 2002; 18: 805-810. AES) for multi-element analysis of trace elements
6 Pinto JJ, Moreno C, Garcia-Vargas M. A simple in human serum. Scand J Clin Lab Invest 2000;
and very sensitive spectrophotometric method for 60: 677-686.
the direct determination of copper ions. Anal 20 Niedemeier W, Creitz EE, Hollay HL. Trace
Bioanal Chem 2002; 373: 844-848. metal composition of synovial fluid from patients
7 Brenner AJ, Harris ED. A quantitative test for with rheumatoid arthritis. Arthritis & Rheumatism
copper using bicinchoninic acid. Anal Biochem 2005; 5: 439-444.
2002; 226: 80-84. 21 Fritze K, Aspin N, Holmes TH. Routine
8 Dalman O, Tufekci M, Nohut S, Guner S, determination of copper in biological materials by
Karabocek S. Spectrophotometric determination neutron activation. Radiochem Acta 1964; 3: 204-
of copper in pharmaceutical and biological 206.
samples with 3-[2-[2-(2-hydroxyimino-1-methyl- 22 Gozzo ML, Colacicco L, Calla C, Barbaresi G,
propylideneamino)-(ethylamino]-ethyl-imino]- Parroni R, Giardina B, Lippa S. Determination of
butan-2-one oxime). J Pharm Biomed Anal 2002; copper, zinc, and selenium in human plasma and
27: 183-189. urine samples by potentiometric stripping analysis
9 Dawson JB, Ellis DJ, Newton-John H. Direct and constant current stripping analysis. Clin Chim
estimation of copper in serum and urine by atomic Acta 1999; 285: 53-68.
absorption spectrophotometry. Clin Chim Acta 23 Yap CT. X-ray total Reflection fluorescence
1968; 21: 33-42. analysis of iron, copper, Zinc and Bromine in
10 Spector H, Glusman S, Jatlow P, Seligson D. human serum. Spectroscopy 1988; 42: 1250-1253.
Direct determination of copper in urine by atomic 24 Sampson B. Clinical application of inductively
absorption spectrophotometry. Clin Chim Acta . coupled plasma mass spectrometry. CPD Clin
1971; 31: 5-11. Biochem 2004; 6: 24-31.
11 Weinstock N, Uhlemann M. Automated 25 Murphy TJ. Accuracy in trace analysis: sampling,
determination of copper in undiluted serum by sample handling and analysis, National Bureau of
atomic absorption spectrophotometry. Clin Chem Standards Special Pub No. 422, Proceedings of
1981; 27: 1438-1440. the 7th Materials Research Symposium,
12 Ji X, Ren J. Determination of copper and zinc in Washington, DC, 1976, U.S. Government
serum by derivative atomic absorption Printing Office.
spectrometry using the micro sampling technique.
Analyst 2002; 127: 416-419.
416
Copper
26 Hall M, Loscombe S, Taylor A. Trace Element 37 Wapnir RA, Steil L. Intestinal absorption of
contamination from blood specimen containers. copper: effect of sodium. Proc Soc Exp Biol Med.
Trace Elements in Medicine. 1988; 5: 126-129. 1987; 185: 277-282.
27 Martins da Costa CM, Baldwin D, Portmann B, 38 Zerounian NR, Redekosky C, Malpe R, Linder
Lolin Y, Mowat AP, Mieli-Vergani G. Value of MC. Regulation of copper absorption by copper
urinary copper excretion after penicillamine availability in the CaCO-2 cell intestinal model.
challenge in the diagnosis of Wilsons disease. Am J Physiol 2003; 284: G739-G747.
Hepatol 1992; 15: 609-615. 39 Higuchi S, Higashi A, Nakamura T, Matsuala I.
28 McKenzie JM. Alteration of the zinc and copper Nutritional copper deficiency in severely
concentration of hair. Am J Clin Nutr 1978; 31: handicapped patients on a low copper enteral diet
470-476. for a prolonged period: estimation of the required
29 Popko J, Olszewski S, Hukalowicz K, dose of dietary copper. J Paed Gastroenterol Nutr
Markiewicz R, Borawska MH, Szeparowicz P. 1988; 7: 583-587.
Lead, Cadmium, Copper and Zinc Concentrations 40 Oliver A, Allen KR, Taylor J. Trace element
in blood and hair of mothers of children with concentrations in patients on home enteral
locomotor system malformations. Pol J Environ feeding: two cases of severe copper deficiency.
Stud 2003; 12: 375-379. Ann Clin Biochem 2005; 42: 136-140.
30 Rivlin R. Misuse of hair analysis for nutritional 41 Klevay LM. Cardiovascular disease form copper
assessment. Am J Med 1983; 75: 489-493. deficiency- A history. J Nutr 2000;130:489S-
31 Vulpe CD, Packman S. Cellular copper transport. 492S.
Ann Rev. Nutr 1995; 15: 293-322. 42 Andrews NC. Mining copper transport genes.
32 Taylor A. Detection and monitoring for disorders Proc Natl. Acad Sci 2001; 98: 6543-6545.
of essential trace elements. Ann Clin Biochem 43 Frydman M, Bonne-Tamir B,Farrer LA,
1996; 33: 486-510. Conneally PM, Magazanik A, Ashbel S et al.
33 Uauy R, Gonzalez M, Olivares M. Essentiality of Assignment of the gene for Wilson disease to
copper in humans. Am J Clin Nutr 1998; 67: chromosome 13. Linkage to esterase D locus.
952S-959S. Proc. Natl Acad Sci 1985; 82: 1819-1821.
34 Linder MC. Biochemistry and molecular biology 44 De Bie P, Muller P, Wijmenga C, Klomp LWJ.
of copper in mammals. In: Handbook of copper- Molecular pathogenesis of
pharmacology and toxicology, Totowa, NJ, . Wilsons and Menkes disease: Correlation of
Massaro EJ, Totowa NJ, eds. Humana. 2002;3-32 mutations with moleculare defects
35 Linder MC, Hazegh-Azam M. Copper and disease phenotypes. J Med Genet
biochemistry and molecular biology. Am J Clin 2007;44:673-678.
Nutr 1996; 63: 797S-811S. 45 Fraser CG. Biological variation: From principles
36 National Research Council: Havel RJ, Calloway to practice. Washington AACC
DH, Gussow JD, Mertz W, Nesheim MC eds. press 2001.
Recommended Dietary allowances. 10th edn.
Washington DC. Natl Acad Press. 224-30.
417
Copper
Table 1. Methods for Copper Analysis in Biological Materials
Method Principle Usage Comments
1. Spectrophotometric Copper complexes with S

chelator to form a strongly
chromogenic complex
2. Flame atomic absorption Amount of light from copper S, U, tissue 1, 2
spectrometry hollow cathode lamp
absorbed by ground state
atoms
3. Flameless atomic As for Method 2, but electro S, U, tissue 1, 2
absorption spectrometry thermal heating used to
produce atoms
4. Inductively coupled plasma Atomization followed by S, U, tissue 1, 2, 3
mass spectrometry (ICP- ionization. Detection by
MS) mass/charge ratio
5. Inductively coupled Atomization followed by S, U, tissue 1, 2, 3
plasma optical emission ionization. Emitted light by
spectrometry(ICP-OES) excited atoms/ions is
detected
1 Digestion required for tissues.

2 Deproteinization or dilution required for serum.
3 Multi-element analysis.
S, Serum; U, urine.
Analysis of Copper centrifuge (3000 g for 10 to 15 min). The presence of an

acidic milieu not only induces the proteins to precipitate
Less than 10% of the analytical laboratories still use but also liberates the copper bound to the proteins.
colorimetric methods for the analysis of copper. Flame Analyze the clear supernatant, ensuring that the aspiration
atomic absorption spectrometry remains by far the most tube is directed away from any deposits precipitated on the
common analytical method used in the analysis of copper side of the tube. By comparing the signal to a calibration
in biological specimens. It is thus imperative that the curve that has been established by similarly preparing and
presenting solution is as homogenous and particle-free as running a set of working aqueous standards (0 to 50
possible to avoid blockage of the aspiration tube and/or mol/L), the concentration of copper can be determined.
interference of the absorption signal. To aid in correcting Note that blanking is carried out using a solution
for background contamination, it is advisable to include a containing water (instead of serum) and 10%TCA at the
blank which has been prepared in a similar fashion to the same volumes as for the standards/serum samples. The
samples and using the same reagents. working standards are normally prepared from a
commercially available, high-quality stock copper
Analysis of Serum Copper solution. To avoid pipetting errors, it is essential that the
The preparative step(s) for producing a same dispenser/pipette(s) be used for the preparation of
clear/homogenous/particle-free solution varies from one both the standards and the samples.
laboratory to another, but the principle of the actual There are many models of flame atomic absorption
analytical process thereafter is very similar. The method currently in use todayof which Perkin Elmer Model
described below is one example of a method used in the 3100 is an example. Each atomic absorption instrument,
analysis of Cu in serum/plasma samples. however, needs to be optimally set up prior to analysis.
The hollow cathode lamp must be run at a suitable current
Prepare a 20%w/vTCA solution (100 g in 500 mL of (usually 10 to 15 mA), and the wavelength must be set at
deionized water), and dilute this 1:1 to get a 10% solution. 324.7 nm. In instruments where the hollow cathode lamp is
To a plastic (polystyrene/polypropylene) tube containing not automatically adjusted, manual alterations are required.
1.5 mL of 10% TCA, add a small volume of serum/plasma Adjust the lamp gently (in/out and
(100 to 250 l), mix well (manual/automated vortexer), and clockwise/anticlockwise), and adjust the screws on the
418
Copper
lamp housing to get the best possible intensity reading. within samples. If the concentration within a sample
Because only a small proportion of sample that is aspirated exceeds the concentration of the top-working standard,
actually reaches the flame, and the sample is within the dilute the sample with water and re-run, making sure to
light path very transiently, the burner-nebulization system correct for the dilution.
is very inefficient (only around 5% efficient). The
sensitivity of the analysis can be improved by using a Analysis of Tissue Copper (e.g., Liver Biopsy)
burner head having a longer slit (i.e., 10 cm). Moreover, The measurement of copper in a liver biopsy sample is
pre-mix burners are more popular, since they experience often the cornerstone in establishing the diagnosis of
less interference and have better signal/noise ratio than Wilsons disease. As stated above, the fluid content of
total consumption burners. The burner head, however, tissue may vary from sample to sample, depending on
must be correctly aligned so that the maximal signal can be storage conditions and other factors. It is thus necessary to
detected. For checking the burner alignment, two steps are totally dry the sample prior to digestion with concentrated
necessary: nitric acid. Transfer the tissue sample from the filter paper
1. Rest a piece of fairly stiff paper/card in the middle of to a clean, metal-free, plastic tube. If larger pieces are
the burner head and perpendicular to the slit. Turn available, they should be sliced into smaller pieces using a
the horizontal/vertical adjustment screws/knobs in clean scalpel. Dry the samples overnight with the lid of the
turn, so that the burner head slot is centered tube open but covered. In cases where larger pieces have
approximately below the light beam. Lower the been provided, heating the samples to temperatures up to
height of the burner head so that it is well below the 100C on a hot plate or in an oven for a further 24 hours
beam. Whilst monitoring the absorption signal on helps with the drying process. The sample(s) are deemed
continuous mode, auto-zero the reading and raise the to be completely dry when there is no further change in
burner head until a small change in absorbance mass with time.
becomes apparent (i.e. burner head interfering with
beam). Lower the burner head gently until the After drying, samples can be digested in a small volume of
absorbance reads zero again. A further marginal concentrated nitric acid. Blank tubes (no tissue) and QC
lowering of the burner head ( to turn of the samples (lyophilized liver sample of known concentration)
screw) will suffice for not only copper but most are also prepared at the same time to correct for
other analyses. background noise (e.g., contamination) and to check for
2. Once the flame is lit and warmed up, aspirate a dilute accuracy and precision, respectively. Overnight exposure
standard solution of copper while continuously to nitric acid is usually sufficient for complete digestion of
monitoring the absorbance, make fine adjustments to samples. In rare cases, heating at 70C to 80C for 2 hours
the burner by gently tweaking the horizontal and (hot plate) may be required to ensure complete digestion.
vertical screws in turn until the maximum If available, the technique of high-pressure closed vessel
absorbance is reached. microwave digestion can be used to efficiently convert the
samples to homogenous solutions. The digested sample is
To obtain acceptable sensitivity, the oxidant/fuel then diluted with deionized water. In cases where fat-
(air/acetylene) mixture for the flame should also be containing samples have been provided, a white insoluble
optimized. Copper analysis usually performs better with a precipitate may be evident. In such cases, briefly
fairly hot flame. Thus the analyst may need to make centrifuge the digest (10,000 rpm for 2 min) prior to
adjustments to the fuel/oxidant knobs so that a lean, bluish analysis.
flame is evident. A 2:1 oxidant/fuel ratio usually suffices
in the case of copper. Optimization of the ratio can be Samples can then be analyzed as for the urine samples and
established by running a dilute copper standard solution using the same set of aqueous standards. Where tissue
through the flame and under continuous reading mode, levels exceed the top aqueous standard concentration,
making fine adjustments to the fuel/oxidant knobs to dilute the digest and re-run, and correct for dilution.
acquire the maximal absorbance reading.
The Cu content (g/g) in the sample/digest, can be
Analysis of Urine Copper calculated from the following equation:
The instrument must be optimized as for the serum [Cu]M total volume of digested sample (in mL)
measurements. In most cases, pretreatment of urine is not 63.54(atomic wt of copper)
necessary for copper measurement. Following blanking
with water, establish a calibration curve by aspirating a Divided by 1000x dry wt (g)
number of aqueous copper standards over the
concentration range of interest. Ensure that the response is The above-mentioned procedures and processes
linear over the range by comparing the absorbance at each mainly refer to the technique of flame AAS.
concentration. Run the urine samples against the Electrothermal/Graphite furnace AA is equally
established calibration to ascertain the concentration suitable for the analysis of copper in biological
419
Copper
samples. However, despite the increased sensitivity the last decade or two, the use of ICP (ICP-OES and
(100 to 1000 times) and the ability to utilize small ICP-MS) has grown in popularity as an analytical tool.
volumes (e.g., pediatric samples), the instrument takes The key advantage of these instruments is the ability
much longer to analyze each sample (minutes as to perform multi-element analysis on a single sample.
opposed to seconds). Moreover, the set-up is more Moreover, as a result of enhanced
complex than the flame system. The furnace- sensitivity/specificity, a sample can be analyzed by
programming steps (for drying, charring and simply diluting it many-fold. This procedure
atomization) varies from matrix to matrix, thereby minimizes not only the preparative steps required but
requiring a greater degree of analyst expertise. Over also the matrix effects. Major drawbacks for the use of
ICP-MS are the high running costs and the
requirement for a greater degree of expertise.
420
Cortisol
Cortisol
R. Swaminathan
Name: Cortisol, hydrocortisone, compound F, glucocorticoid
Clinical significance: Refer to Chapter 51, Adrenal Hormones and Hypertension, in the 5th
Molecular formula: C21H30O5
Merck Index: 4689
Structure:
soluble metabolites and glucuronide conjugates. The

Principles of Analysis and Current Usage i excretory products of cortisol that contain the
dihydroxyacetone group are known as 17-
Cortisol, also known as compound F and hydroxycorticosteroids (17-OH-CS). The
hydrocortisone, is the most important glucocorticoid measurement of urinary 17-OH-CS has been used in
secreted by the zona fasciculata and zona reticularis the past as an indirect measure of cortisol secretion
of the adrenal gland. Under basal conditions, only rate. Cortisol in urine is a reflection of free,
about 5% to 10% of circulating cortisol is free. The circulating cortisol in blood.
rest is bound to proteins: 85% to transcortin (cortisol-
binding globulin) and 5% to 10% to albumin. It is the Many different types of methods have been used for
free fraction which is physiologically active, although the determination of cortisol in biological fluids
the albumin-bound fraction is also available because (Table 1). Before the availability of immunoassay
the binding affinity of albumin for cortisol is low. methods for serum and urine cortisol, urinary 17-OH-
CS estimations by colorimetric or fluorometric
Unbound cortisol is filtered at the glomerulus and methods were used for the assessment of adrenal
excreted. This fraction constitutes less than 1% of the function [1]. With the development of
total cortisol synthesized daily; the rest is excreted as immunoassays, these methods have become obsolete.
A fluorometric method developed by Mattingley,
i where cortisol is extracted and then converted by
Cortisol
ethanolic sulfuric acid to a fluorogenic chromogen,
Previous authors of this method:
has been used to measure plasma cortisol (Table 1,
First edition: Suman Patel, John A. Lott
Method 1) [2]. This method is also obsolete. These
Methods edition: Suman Patel, John A. Lott
methods lack specificity, require a large volume of
sample, and are labor intensive. Several competitive
Third edition: Suman Patel, John A. Lott
protein binding assays were then used (Table 1,
Fourth edition: Suman Patel, John A. Lott
Method 2) [3]. With the development of more
Fifth edition: R. Swaminathan
specific assays these methods are also no longer used.
421
Cortisol
gas chromatography [12]. In this method [12],

Immunoassays are the main methods for estimation labeled cortisol as internal standard is added to
of cortisol in clinical laboratories. A large number of plasma and the plasma extracted with
isotopic (Table 1, Method 3[i]) and non-isotopic dichloromethane, purified using Sephadex LH-20,
(Table 1, Method 3[ii]) immunoassays for the followed by derivatization and detection by gas-
measurement of cortisol have been described [4]. liquid chromatography.
These immunoassays are competitive immunoassays,
where cortisol in the sample competes with labeled The reference method for plasma cortisol utilizes
cortisol for a limited number of anti-cortisol liquid chromatographymass spectrometry (LC-MS)
antibodies. Almost all immunoassays used in clinical [13]. In this method, an internal standard is added to
laboratories do not require an extraction step [5]. the sample, followed by extraction with ethyl acetate.
Initially, labels used in these assays were radioactive. Liquid chromatographymass spectrometry (LC-MS)
Later, others were used, including enzymes, interfaced with electrospray ionization is used, and
luminescent, chemiluminescent, and fluorescent selective monitoring of the [M+H]+ ions of cortisol
labels. Many commercial kits are available for the and isotopically labeled cortisol are used for
determination of plasma and urinary cortisol. Many quantitation. A similar method but using
of the immunoassays are now automated and are dichloromethane to extract cortisol has also been
available on semi-automated or automated analyzer described [14]. Guo et al. have described a much
platforms. These immunoassays can be simpler sample preparation. In this method, the
heterogeneous assays, where a separation step is sample after addition of internal standard is mixed
required to partition the bound from unbound with acetonitrile, centrifuged, and the supernatant is
fractions, or homogenous assays, where no separation injected into the LC-MS [15]. A recent study
is required. The specificity of immunoassays depends suggests that care is necessary in using labeled
largely on the properties of the antibody. internal standard, since naturally occurring isotopes
of cortisol can cause interference [16].
Chromatographic methods are an alternative to
immunoassay for the determination of cortisol, and A preferred method which gives good accuracy and
they have the advantage that there is less interference, precision is the HPLC method. However, this method
but these methods are time consuming and not is labor intensive, requires expertise, and is slow.
suitable for routine analysis. Of the chromatographic
methods, high-performance liquid chromatography Direct (without extraction) immunoassay methods are
(HPLC) is often used in clinical laboratories [6]. All the most frequently used methods in laboratories, and
HPLC methods require extraction of cortisol from the many are widely available on automated platforms
sample, by either a liquid-liquid extraction or by use [17]. More than 90% of cortisol is bound to proteins,
of solid-phase extraction columns. Both reversed- so the bound fraction needs to be quantitatively
phase and normal-phase columns are used for the displaced before analysis in these direct assays.
chromatographic analyses, and the column effluents Methods used for the displacement include use of
are monitored by either ultraviolet (254 nm) or agents such as 8-anilino-1-naphthalene sulfonic acid
fluorescence spectroscopy [6,7]. (ANS), danazol, and salicylate. Heat treatment or
lowering the pH are also effective in releasing bound
Other chromatography methods include gas cortisol. It is important to note that the efficiency of
chromatography [8] and mass spectrometry displacement by agents such as danazol may depend
interfaced to gas chromatography (GC-MS) [9] or upon the concentration of corticosteroid-binding
liquid chromatography (LCMS) [10]. Methods using globulin (CBG). In situations of high CBG
capillary electrophoresis have also been described concentration such as pregnancy, the displacement of
[11]. cortisol may not be complete. Furthermore, high
concentration of the releasing agent may affect the
In methods using gas-liquid chromatography, cortisol binding of cortisol to the antibody.
is extracted, derivatized, and detected by flame
ionization [6]. In GC-MS methods, quantitation is by Antibodies used in immunoassays are both
mass spectrometry. LC-MS methods are similar to polyclonal and monoclonal. Common derivates used
HPLC methods, except the detection is by MS [9]. in the preparation of antibodies are cortisol-21-
hemisuccinate and cortisol-3-carboxy methyloxime
Reference and Preferred Methods conjugate. Measurement principles used in the non-
The definitive method for measurement of plasma isotopic assays include photometry, fluorescence,
cortisol is isotope dilution mass spectrometry using chemiluminescence, electro-chemiluminescence, and
422
Cortisol
time-resolved fluorescence. Table 2 gives details of undergoing metyrapone testing, and patients
some of the automated immunoassays in common with 21-hydroxylase deficiency. In addition,
use. Assay formats used include both heterogeneous immunoassays may give discrepant results
(separation of bound and free hormone required) and in patients whose hypothalamic-pituitary
homogenous (no separation required). Assays used in axis is activated, such as in critical illness. In
the Roche system and Advia Centaur are examples of these patients, stimulation of the
heterogeneous assays. The most common methods hypothalamic-pituitary-adrenal axis causes
used in United States and the United Kingdom, increase in precursors of cortisol, which can
according to data from CAP and the United Kingdom lead to varying degrees of cross-reactivity.
National External Quality Assessment Service This was shown in a recent study of
(UKNEQAS), are given in Table 3. The Bayer Advia critically ill patients [21], where cortisol
Centaur cortisol assay is a competitive immunoassay, response to ACTH (Synacthen test) was
where cortisol in the sample and acridinium ester compared between 3 common
labeled cortisol competes for a polyclonal rabbit anti- immunoassays and an HPLC method. This
cortisol antibody. Salicylate is added to release the study found that the Abbott TDX method
bound cortisol. The Roche Elecys assay is a and IMMULITE 2000 method gave higher
competitive immunoassay with competition between results than HPLC by 95% and 79%,
the cortisol in the sample and ruthenium complex respectively. Limits of agreement between
labeled cortisol for a polyclonal ovine anticortisol HPLC and the three methods were: 62% to
antibody. Treatment with danazol releases the bound 480%, 53% to 590% and 54% to 770%,
cortisol. The Abbott fluorescent polarization assay is respectively, for the Centaur, TDX, and
an example of a homogenous assay. Most of the IMMULITE methods.
common automated immunoassays give satisfactory 4. Urine cortisol methods: Many
performance. Assays that give relatively poor immunoassays are used for the
performance are the Abbott AxSYM and DPC determination of urinary free cortisol, some
IMMULITE 2000 methods. after extraction of urine with
dichloromethane and others without prior
Although cortisol immunoassays are automated, easy extraction. In methods without prior
to perform, and widely used, several problems have extraction, the measured concentrations are
been identified with these assays. nearly twofold higher than those determined
by HPLC methods [22,23].
1. Commercial immunoassays for cortisol
show a positive bias when compared to Nevertheless, immunoassay techniques are the
either isotope-dilution mass spectrometry or current method of choice and most widely used
gas chromatography mass because of their overall acceptable specificity,
spectrophotometric methods [18]. Murphy et sensitivity, and precision and because results are
al. [19] reported that without available fairly rapidly.
chromatographic purification, all methods
grossly overestimated the amount of cortisol Specimen
present in urine. Both serum and heparinized plasma are suitable
2. Immunoassay methods work reasonably specimens for the measurement of cortisol in blood.
well in patients with normal serum protein Specimens can be stored at 2C to 8C for 2 days.
levels. However, in patients with low protein For longer storage, specimens must be frozen [5].
levels, such as patients in intensive care Freeze/thaw cycles have not been found to
units, these assays can give misleading significantly alter cortisol concentrations.
results. In a recent study, it was reported that
the Bayer Advia Centaur cortisol assay gave Salivary cortisol has been found to be an excellent
consistently lower results when the albumin indicator of serum free cortisol concentrations
concentration was lower than 24 g/L [20]. [24,25]. A specimen is typically collected using a
3. Specificity. Antibodies used in these special device containing a cotton swab, with or
immunoassays show varying degrees of without prior stimulation by chemicals such as citric
cross-reactivity to related steroids (see acid. Specimens are stable in the presence of citric
above), such as prednisolone, 11- acid at a concentration of 10 g/L. In a citric-acid-
deoxycortisol, and 21-deoxycortisol. This treated cotton swab, cortisol is stable for as long as 6
can be a problem in patients taking weeks at room temperature [26]. Presence of citric
prednisone/prednisolone, patients acid (or lemon juice) may cause apparent under-
423
Cortisol
recovery in extraction-based assays [27]. Several unchanged, with very little tubular reabsorption.
commercial collecting devices are available for Although less than 1% of the daily cortisol secretion
saliva, and of these the PE Salinette was found to be is excreted in the urine, urine cortisol measurements
the best [28]. reflect cortisol secretion rate. The reference range for
urinary free cortisol varies with the method used.
Free cortisol in urine can be determined by Typically, methods without extraction give much
immunoassay or HPLC methods. These procedures higher values (e.g., up to 325 g/d (900 nmol/d)).
require an extraction step using methylene chloride, When an extraction method is coupled to an HPLC
and it is customary to monitor the efficiency of the analysis, the reference range is 10 to 54 g/d (27 to
extraction technique by extraction of a urine control 150 nmol/d), and this is similar to that obtained using
with known cortisol concentration. A 24-hour urine an LC-MS/MS method [30]. Several studies have
sample is preferred. Urine collected for a 24-hour shown a difference in cortisol excretion rate between
sample should be refrigerated during the collection males and females.
period. Some laboratories use boric acid
(approximately 10 g of boric acid per liter) or acetic Problems with urine free cortisol measurements
acid (approximately 7 mL of glacial acetic acid per include incomplete urine collection and cross-
liter), placed in the collection container at the start of reactivity (especially in methods without prior
the collection, to preserve the specimen. If the sample extraction). Sensitivity and specificity of urine
cannot be analyzed within a day of collection, it cortisol measurement for the diagnosis of Cushings
should be frozen at 20C. syndrome were reported to be 45% to 71% and
100%, respectively [31]. Owing to the difficulty in
Interferences collecting complete 24-hour urine samples, some
All immunoassays are prone to interference. Presence authors recommend the use of urine
of heterophilic antibodies in patient serum can give cortisol/creatinine ratios in early morning urine as an
false results. Although specific antibodies are used, alternative. This is especially useful in children [32].
almost all immunoassays for cortisol used in clinical In addition, variation in the activity of 11-
laboratories show cross-reactivity with other related hydroxysteroid dehydrogenase type 2, which
steroids such as prednisolone. Cross-reactivity inactivates cortisol to cortisone, can significantly
against prednisolone varies from 27% for the Advia affect the amount of cortisol excreted by the kidney.
Centaur to 171% for the Roche system [29]. Cross- It has also been suggested that measurement of both
reactivity against prednisone is lower (0.3% to 6.6%). urinary cortisol and urinary cortisone can provide
However, prednisone is converted to prednisolone in more meaningful information [33].
the body, and therefore these assays are not suitable
for measurement of cortisol in patients taking Serum free cortisol can be measured by equilibrium
prednisone or prednisolone. Cross-reactivity to 11- dialysis method, steady-state gel filtration, or by
deoxycortisol varies from 4.15% in the Roche system ultra-filtration methods [34-36]. All these procedures
to 21.6% in the Beckman Access analyzer. In patients are time consuming and labor intensive, and some
undergoing metyrapone testing, 11-deoxycortisol will authors have suggested that calculation of a free
be high, and these assays will therefore give falsely cortisol index from serum total cortisol and CBG
high values for cortisol. Some of the assays also concentrations may be more useful [37]. An
show cross-reactivity against 21-deoxycortisol, which alternative approach is the measurement of cortisol in
is elevated in patients with 21-hydroxylase saliva. Salivary cortisol is a good index of free serum
deficiency. Some of the interferences (e.g., due to 11- cortisol [24]. The concentration of cortisol in saliva is
deoxycortisol) could be removed by prior extraction independent of salivary flow rate and reflects acute
of samples with carbon tetrachloride. changes in serum cortisol [25].
Interpretation Cortisol Reference Interval

Cortisol secretion is pulsatile, so isolated serum Reference values vary to some extent, depending on
cortisol values are of limited value. Stimulation and the methods used for cortisol measurement. Other
suppression tests are required to diagnose factors affecting the reference interval are age,
adrenocortical disorders. gender, and diurnal variations. Values presented in
Table 4 are compiled from the literature. Because of
Twenty-four-hour urine cortisol excretion reflects the decreased cortisol production, children have lower
overall cortisol secretion rate and is not influenced by plasma cortisol levels and urinary excretion rates than
the circadian rhythm. Only the free fraction in plasma adults do. Plasma cortisol concentrations exhibit a
is filtered at the glomerulus, and this is excreted circadian rhythm, reaching a peak between 4 and 6
424
Cortisol
am; lowest values are seen by midnight. Total and 7 Lensmeyer G, Carlson I, Wiebe D, Devos D.
free plasma cortisol and cortisol-binding globulin Determination of cortisol ( C), cortisone
levels are elevated during pregnancy. Table 4 also (CN), corticosterone (CC), prednisone (P)
shows cortisol concentrations in saliva and urine and prednisolone (PL) in serum with solid
reported in the literature. reversed-phase extraction and high
performance liquid chromatography
(HPLC). Clin Chem 1992;38:945.
Cortisol Performance Goals 8 Woitz HH, Chattaraj S. The role of gas-
The current target for acceptable performance criteria liquid chromatography in steroid hormone
as mandated by the Clinical Laboratory Improvement analysis. J Chromatogr Sc 1973;11:167-74.
Amendments 88 is 25% of the peer-group mean 9 Hsu FF, Wang L, Bier DM. Determination
[38]. Precision goals suggested by Fraser et al. are a of cortisol in human plasma by gas
maximum standard deviation of 23 g/L (63 nmol/L) chromatography/negative ion chemical
at a cortisol concentration of 300 g/L (828 nmol/L) ionization/mass spectrometry. Anal Biochem
(a coefficient of variation [CV] of 7.6%) [39]. It is 1994:216:401-5.
evident from a recent study that intralaboratory 10 Shibasaki H, Furuta T, Kasuya Y.
imprecision (%CV) of serum/plasma cortisol assays Quantification of corticosteroids in human
by different vendors varies from 5% to 12% for plasma by liquid chromatography
cortisol concentrations ranging from 100 g/L to 300 thermospray mass spectrometry using stable
g/L (275 nmol/L to 828 nmol/L)[40]. Data from the isotope dilution. J Chromatogr B Biomed
2007 CAP survey shows that the most widely used Appl 1997;892:7-14.
method in the United States, the Bayer Advia Centaur 11 Schmalzing D, Nashabeh W, Yao XW,
method, achieves this analytical goal. Cortisol assays Mhatre R, Regnier FE, Afeyan NB, Fuchs
by the Abbott AxSYM and DPC IMMULITE 1000 M. Capillary electrophoresis based
methods show precision of 12.4% and 12.6%, immunoassay for cortisol in serum. Anal
respectively. The newer DPC IMMULITE 2000 and Chem 1995;67:606-12.
2500 assays show better performance [41]. 12 Siekmann L, Breuer H. Determination of
cortisol in human plasma by isotope
References dilution-mass spectrometry. Definitive
1 Porter CC, Silber RH. A quantitative color methods in clinical chemistry. I. J Clin
reaction for cortisone and related 16,21- Chem Clin Biochem. 1982;20:883-92.
dihydroxy-20-ketosteroids. J Biol Chem 13 Tai SS, Welch MJ. Development and
1950;185:201-7. evaluation of a candidate reference method
2 Mattingly DA. Simple fluorometric method for the determination of total cortisol in
for the estimation of free 11- human serum using isotope dilution liquid
hydroxycorticoids in human plasma. J Clin chromatography/mass spectrometry and
Pathol 1962;15:374-9. liquid chromatography/tandem mass
3 Turner AK, Carrol CJ, Pinkus JL, Charles D, spectrometry. Anal Chem 2004;76:1008-14.
Chattoraj SC. Simultaneous competitive 14 Jung PG, Kim B, Park SR, So HY, Shi LH,
protein binding assay for cortisol, cortisone, Kim Y. Determination of serum cortisol
and prednisolone in plasma, and its clinical using isotope dilution-liquid
application. Clin Chem 1973;19:731-6. chromatography-mass spectrometry as a
4 Price CP, Newmann DJ. Principles and candidate reference method. Anal Bioanal
Practice of Immunoassay. New York: Chem 2004;380:782-8.
Stockton press; 1996. 15 Guo T, Taylor RL, Singh RJ, Soldin SJ.
5 Demers L. The adrenal cortex. In: Burtin Simultaneous determination of 12 steroids
CA, Ashwood ER, Bourns DE, eds. Tietz by isotope dilution liquid chromatography
Textbook of Clinical Chemistry and photospray ionization tandem mass
Molecular Diagnostics. 4th ed. Philadelphia: spectrometry. Clin Chim Acta 2006;372:76-
Saunders; 2006:2003-52. 82.
6 Hariharan M, Naga S V, Noord T, Kindt EK. 16. Duxbury K, Owen L, Gillingwater S, Keevil
Simultaneous assay of corticosterone and B. Naturally occurring isotopes of an analyte
cortisol in plasma by reverse-phase liquid can interfere with doubly deuterated internal
chromatography. Clin Chem 1992;38:346- standard measurement. Ann Clin Biochem
51. 2008;45:210-2.
425
Cortisol
17 Wild D. The Immunoassay Handbook. 2nd cortisol immunoassays. Clin Biochem

ed. London: Nature PublishingGroup; 2001. 2004;37:489-93.
18 De Brabandere VI, Thienpont LM, Stockl D, 30 McCann SJ, Gillingwater S, Keevil BG.
De Leenheer AP. Three routine methods for Measurement of urinary free cortisol using
serum cortisol evaluated by comparison with liquid chromatographytandem mass
an isotope dilution gas chromatography- spectrometry: comparison with the urine-
mass spectrometry method. Clin Chem adapted ACS:180 serum cortisol
1995;41:1781-3. chemiluminescent immunoassay and
19 Murphy BE. How much "UFC" is really development of a new reference range. Ann
cortisol? Clin Chem 2000;46:793-4. Clin Biochem 2005;42:112-8.
20 Barnes SC, Swaminathan R. Effect of 31 Findling JW, Raff H. Cushing's syndrome:
albumin concentration on serum cortisol important issues in diagnosis and
measured by the Bayer Advia Centaur assay. management. J Clin Endocrinol Metab
Ann Clin Biochem 2007;44:79-82. 2006;91:3746-53.
21 Cohen J, Ward G, Prins J, Jones M, 32 Moore RA, Smith RF, Asplin CM. Simple
Venkatesh B. Variability of cortisol assays test for nocturnal hypoglycaemia in diabetic
can confound the diagnosis of adrenal patients. Lancet 1979;1:409-10.
insufficiency in the critically ill population. 33 Lin CL, Wu TJ, Machacek DA, Jiang KS,
Intensive Care Med 2006;32:1901-5. Kao PC. Urinary free cortisol and cortisone
22 Horie H, Kidowaki T, Koyama Y, Endo T, determined by high performance liquid
Homma K, Kambegawa A, Aoki N. chromatography in the diagnosis of
Specificity assessment of immunoassay kits Cushings syndrome. J Clin Endocrinol
for determination of urinary free cortisol Metab 1997;82:151-5.
concentrations. Clin Chim Acta 34 Beim T. Free cortisol in human plasma.
2007;378:66-70. Horm Res 1980;12:643-50.
23 Ching SY, Lim EM, Beilby J, Bhagat C, 35 Robin P, Predine J, Milgrown E. Assay of
Rossi E, Walsh JP, Pullan P. Urine free unbound cortisol in plasma. J Clin
cortisol analysis by automated immunoassay Endocrinol Metab 1977;46:277-83.
and high-performance liquid 36 le Roux CW, Sivakumaran S, Alaghband-
chromatography for the investigation of Zadeh J, Dhillo W, Kong WM, Wheeler MJ.
Cushing's syndrome. Ann Clin Biochem Free cortisol index as a surrogate marker for
2006;43:402-7. serum free cortisol. Ann Clin Biochem
24 Raff H. Salivary cortisol: a useful 2002;39:406-8.
measurement in the diagnosis of Cushings 37 le Roux CW, Chapman GA, Kong WM,
syndrome and the evaluation of the Dhillo WS, Jones J, Alaghband-Zadeh J.
hypothalamic pituitary-adrenal axis. Free cortisol index is better than serum total
Endocrinologist 2000;10:9-17. cortisol in determining hypothalamic-
25 Vining RF, McGinley RA. The pituitary-adrenal status in patients
measurement of hormones in saliva: undergoing surgery. J Clin Endocrinol
possibilities and pitfalls. J Steroid Biochem Metab 2003;88:2045-8.
1987;27:81-94. 38 Clinical Laboratory Improvement
26 Aardal E, Holm AC. Cortisol in saliva: Amendments. Available at
reference ranges and relation to cortisol in <http://www.cms.hhs.gov/clia/> Accessed
serum. Eur J Clin Chem Biochem 2008.04.16.
1995;33:927-32. 39 Fraser CG, Hyltoft Petersen P, Rics C,
27 Vialard-Miguel J, Belaidi N, Lembeye L, Haeckel R. Quality specifications. In:
Corcuff JB. Lemon juice alters cortisol Haeckel R, ed. Evaluation Methods in
assays in saliva. Clin Endocrinol (Oxf) Laboratory Medicine. New York: UCH
2005;63:478-9. Publishers; 1993:87-99.
28 Grosche M, Raul M. Influence of 40 Steele BW, Wang E, Palmer-Toy DE,
commercial collection devices for saliva on Killeen AA, Elin RJ, Klee GG. Total long-
the reliability of salivary steroid analysis. term within-laboratory precision of cortisol,
Steroids 2006;71:1097-100. ferritin, thyroxine, free thyroxine, and
29 Roberts RF, Roberts LW. Performance thyroid-stimulating hormone assays based
characteristics of five automated serum on a College of American Pathologists fresh
frozen serum study: do available methods
426
Cortisol
meet medical needs for precision? Arch

Pathol Lab Med 2005;129:318-22.
41 College of American Pathologists.
Comprehensive Chemistry Survey Report C-
B, 2007.
42 Clark PM, Neylon I, Raggatt PR, Sheppard
MC, Stewart PM. Defining the normal
cortisol response to the short Synacthen test:
implications for the investigation of
hypothalamic-pituitary disorders. Clin
Endocrinol 1998;49:287-92.
43 Burtis CA, Ashwood ER, Bruns DE. Tietz
Textbook of Clinical Chemistry and
Molecular Diagnostics. 4th ed. Philadelphia:
Saunders; 2006.
44 Heckmann M, Wudy SA, Haack D, Pohlandt
F. Reference range for serum cortisol in well
preterm infants. Arch Dis Child Fetal
Neonatal Ed. 1999;81:F171-4.
45 MacMahon W, Thompson J, Bowers W,
Sgoutas D. A simplified ultrafiltration
method for determination of serum free
cortisol. Clin Chim Acta 1983;131:171-84.
427
Cortisol
Table 1: Methods of Cortisol Analysis
Method Type of Analysis Principle Usage Comments
1.Fluorometry Chemical After extraction, ethanolic S, P Historical interest

derivatization sulfuric acid converts
cortisol to fluorogenic
chromogen
2. Competitive protein Radioassay After extraction, cortisol in S Historical interest
binding sample competes with
125
I-labeled cortisol for a
cortisol-binding protein
Heterogeneous assay
3. Immunoassay
(i) Radioimmunoassay Competitive radio As above, but antibody S, U Mainly of
assay specific for cortisol is the historical
binding agent interest
(ii) Non-isotopic Labeled with a non-isotopic

immunoassay label such as enzyme S, U
Separation of bound and free

label required
(a) Heterogeneous
No separation required Commonly used
method; no
extraction
(b) Homogeneous required
Commonly used
method; no
extraction
required
4. High-performance liquid Chromatographic Cortisol separated from other S, P, U Extraction required
chromatography Separation compounds by Not suitable for
(HPLC) chromatography and large number of
detected by native routine samples
fluorescence or UV Less interference
absorption
5. Gas chromatography As above Cortisol derivatized and S, P, U As above
separated by
chromatography and
detected by flame
ionization
6. Gas chromatography-mass As above; detection by MS S, P, U As above
spectrometry
7. Liquid chromatography- As above As for GC-MS S, P, U Non-extraction
mass spectrometry methods
available
Candidate
reference
method
428
Cortisol
Table 2: Commonly Used Automated Systems for the Measurement of Cortisol
Instrument Principle End point Manufacturer

Roche Elecys /E170 Paramagnetic beads Electro-chemiluminescence Roche
modular
ACS180and Centaur Paramagnetic particles Chemiluminescence Bayer/Siemens
IMMULITE 2000/2500 Coated beads Chemiluminescence DPC/Siemens
Access/DXI Paramagnetic particles Alkaline phosphatase Beckmann
Architect Paramagnetic particles Chemiluminescence Abbott
TDX Fluorescence Fluorescence Abbott

polarization
VITROS ECi Coated beads Enhance Ortho
chemiluminescence
TOSOH AIa Magnetic microbeads Kinetic fluorescence Tosoh
DELFIA Coated microtiter plate Time-resolved fluorescence Wallac
Table 3: Cortisol assays commonly used in the USA and UK
USA UK
(%) (%)
Bayer Advia Centaur ` 29.3 27.4
Beckman Access/DXI 24.3 11.1
DPC Immulite 2000/2500 11.9 16.7
RocheElcys/E170 modular 11.1 32.1
Abbott Architect 0.7 6.0
Abbott Axsym 4.2 2.0
Others 1.7 2.8
Data from CAP and UK NEQAS

429
Cortisol
Table 4: Cortisol Concentration in Serum, Saliva and Urine
Reference Ranges
Serum Total Cortisol

Adults am 51-230 g/L (140 635 nmol/L) [42]
4 pm 31-100 g/L ( 85 275 nmol/L)
midnight <51 g/L ( <140 nmol/L)
Children am 30-210 g/L ( 83 580 nmol/L) [43]

Infants 20-110 g/L ( 55 304 nmol/L) [43]
Preterm infants 26-203 g/L ( 73 562 nmol/L) [44]
Urine Free Cortisol

Adults unextracted <326 g/d (<900 nmol/d) [23]
extracted 20-90 g/d (55 248 nmol/d) [43]
HPLC <62 g/d (<170 nmol/d) [23]
LCMS <60 g/d (<165 nmol/d) [30]
Serum Free Cortisol

Adults ultra filtrate 2.0-22 g/L ( 6 - 61 nmol/L) [45]
Saliva
Adults 8 am 1.3-9.8 g/L (3.5 27 nmol/L) [34]
10 pm <2.0 nmol/L (<6 nmol/L)
430
Creatine Kinase
Creatine Kinase
Mauro Panteghini
Name: Creatine kinase (CK, adenosine triphosphate: creatine N-phosphotransferase)
Molecular mass: Varies from 78,500 to 85,100 D; molecular mass of individual
subunits is half that of intact molecule
Known isoenzymes: CK1 (CK-BB), CK2 (CK-MB), CK3 (CK-MM), CKmt (mitochondrial)
i active site of the enzyme. Activity can be partially

restored by incubating the enzyme preparation with
Creatine kinase (CK) is a dimer composed of two sulfhydryl compounds, such as N-acetylcysteine,
subunits, each with a molecular weight of about 40,000 dithiothreitol (Cleland reagent), and glutathione. The
D. Three different pairs of subunits can exist: BB (or current agent of choice is N-acetylcysteine, which has
CK1), MB (or CK2), and MM (or CK3). All three of these the advantage of being a very soluble substance used at a
isoenzyme species are found in the cytosol of the cell. final concentration of 20 mmol/L in the assay reagent.
However, there exists a fourth isoenzyme (CKmt)
located between the inner and outer membranes of Numerous photometric, fluorometric, and coupled-
mitochondria, and it constitutes, in the heart for example, enzyme methods have been developed for the assay of
up to 15% of the total CK activity. A summary of CK activity, using either the forward (Cr CrP) or the
enzyme activity distribution is given in Table 1 [1]. reverse (Cr CrP) reaction. Currently, all commercial
assays for total CK are based on the reverse reaction that
CK catalyzes the formation of ATP, a high-energy proceeds about six times faster than the forward reaction.
substance that is required for contractile or transport The ATP produced is measured by hexokinase
systems. The enzyme also catalyzes the reversible (HK)/glucose-6-phosphate dehydrogenase (G6PD)
phosphorylation of creatine (Cr) with ATP as the donor coupled reactions that ultimately convert NADP+ to
of the phosphate group. The reaction to form creatine NADPH, which is monitored spectrophotometrically at
phosphate (CrP) (the forward reaction) is as follows: 340 nm. The rate of increase in absorbance is a measure
of CK activity present in the specimen.
Creatine + ATP pH 6.7 CK pH 9.0 2+
creatine phosphate + ADP Creatine phosphate + ADP CK, Mg , pH 6.7
creatine + ATP
At neutral pH, the formation of ATP is favored; a pH of
9.0 is optimal for the formation of CrP, another high- ATP + glucose HK glucose-6-phosphate + ADP
energy compound. Mg2+ is an obligatory activating ion
Glucose-6-phosphate + NADP+ G6PD 6-
that forms complexes with ATP and ADP. The optimal
phosphogluconate + NADPH + H+
concentration range for Mg2+ is quite narrow, and excess
Mg2+ is inhibitory. Many metal ions, such as Mn2+, Ca2+,
Zn2+, and Cu2+, inhibit enzyme activity, as do Oliver first reported this method [3] that Rosalki
iodoacetate and other sulfhydryl-binding reagents. subsequently modified and improved by adding AMP to
Activity is inhibited by excess ADP and by citrate, inhibit adenylate kinase (AK) and cysteine to activate
fluoride, nitrate, acetate, iodide, bromide, malonate, and CK [4]. Subsequently, Szasz and colleagues optimized
L-thyroxine [2]. the assay by adding N-acetylcysteine to activate CK,
EDTA to bind Ca2+ and increase the stability of the
The enzyme is relatively unstable in serum, activity reaction mixture, and adenosine pentaphosphate (Ap5A)
being lost as a result of sulfhydryl group oxidation at the in addition to AMP to inhibit AK [5]. A reference
method based on this previous experience was developed
by the International Federation of Clinical Chemistry
i
Creatine Kinase and Laboratory Medicine (IFCC) for the measurement of
Previous and current authors of this method: CK at 37C [6].
First edition: Lenox B. Abbott, John A. Lott
Methods edition: Kory M. Ward, John A. Lott Reference and Preferred Methods
Second edition: Kory M. Ward, John A. Lott The above reported Oliver-Rosalki reaction scheme is
Third edition: Kory M. Ward, John A. Lott the basis for the reference method [6]. The reaction and
Fourth edition: Kory M. Ward, John A. Lott measurement conditions of the reference method are
Fifth edition: Mauro Panteghini
431
Creatine Kinase
listed in Table 2, and the procedure is described on page The suggested upper reference limits when standardized
11. methods for CK determination are employed are:
Males up to 170 U/L
The major differences between the commercially Females up to 145 U/L
available optimized assays include the choice of buffer
and source of G6PD, the latter determining the type of Interpretation
pyridine nucleotide that must be used. Imidazole is the Serum CK is increased in nearly all patients when there
best buffer system, but Tris can also be used. The G6PD is injury, inflammation, or necrosis of skeletal or heart
is usually obtained from yeast or from the bacterium muscle.
Leuconostoc mesenteroides; the former requires NADP, Serum CK activity is greatly elevated in all types of
whereas the latter can use either NADP or the less muscular dystrophy. In progressive muscular dystrophy
expensive NAD. (particularly Duchenne sex-linked muscular dystrophy),
enzyme activity in serum is highest in infancy and
Specimen childhood (7 to 10 years of age) and may be increased
Specimens for CK analysis include serum and plasma long before the disease is clinically apparent. Serum CK
heparin. Anticoagulants, other than heparin, should not activity characteristically falls as patients get older and
be used in collection tubes because they inhibit CK as the mass of functioning muscle diminishes with the
activity. Collection of specimen in gel separator tubes progression of the disease. About 50% to 80% of the
does not appear to affect CK activity when compared asymptomatic female carriers of Duchenne dystrophy
with tubes not containing gel [7]. show threefold to sixfold increases of CK activity. Quite
high values of CK are noted in viral myositis,
CK activity in serum is relatively unstable and rapidly polymyositis, and similar muscle diseases. However, in
lost during storage. Average stabilities are less than 8 neurogenic muscle diseases, such as myasthenia gravis,
hours at room temperature, 48 hours at 4C, and 1 month multiple sclerosis, poliomyelitis, and parkinsonism,
at 20C. Therefore, the serum specimen should be serum enzyme activity is not increased [11,12].
stored at 80C if analysis is delayed for more than 30
days. In acute rhabdomyolysis due to crush injuries, with
severe muscle destruction, serum CK activities
Interferences exceeding 200 times the upper reference limit may be
A moderate degree of hemolysis (up to 0.32 g/dL of found. If the CK remains below 5000 U/L (about 30
hemoglobin) does not significantly influence the times the upper reference limit) during the first 3 days
measured CK activity because erythrocytes contain no after the insult, the probability of developing acute renal
CK activity [8]. However, severely hemolyzed failure appears to be low [13]. Very high activity is also
specimens are unsatisfactory because enzymes and encountered in malignant hyperthermia, a familial
intermediates (AK, ATP, and glucose-6-phosphate) disease characterized by high fever and brought on by
liberated from the erythrocytes may affect the lag phase administration of inhalation anesthesia (usually
and the side reactions occurring in the assay system. halothane) to the affected individual. Serum CK can also
be increased by direct trauma to muscle, including
Turbid and icteric samples can be analyzed; appropriate intramuscular injections and surgical interventions.
values are obtained if the starting absorbance is not too
high. The changes of serum CK and its MB isoenzyme
following an acute myocardial infarction have been for
CK Reference Intervals many years the mainstay for the diagnosis [14]. It is
Serum CK activity is subject to a number of now, however, more advantageous to use more cardiac-
physiological variations. Sex, age, muscle mass, physical specific nonenzymatic markers, such as cardiac troponin
activity, and race all interact to affect measured serum I or T [15]. Other cardiac conditions have been reported
activity [9]. Males generally have a larger muscle mass, to increase serum CK and CK-MB in serum (e.g.,
which results in higher serum CK activities than those in coronary artery bypass surgery, cardiac transplantation,
females. Race also has an effect on CK activities, and myocarditis, and pulmonary embolism).
the mean activity in whites is 66% of the mean activity
in blacks. In Caucasian subjects, the reference interval Serum CK activity demonstrates an inverse relationship
was found to be 46 to 171 U/L for males and 34 to 145 with thyroid activity. About 60% of hypothyroid
U/L for females when measured with an assay traceable subjects show an average elevation of CK activity
to the IFCC 37C reference procedure [10]. fivefold more than the upper reference limit. The major
isoenzyme present is CK-MM, suggesting muscular
Newborns generally have higher CK activities, resulting involvement.
from skeletal muscle trauma during birth. Serum CK in
infants decreases to the adult reference interval by 6 to Total CK activity may be significantly increased after
10 weeks. Table 3 contains details of CK activities in exercise. Strenuous, prolonged exercise will result in
apparently healthy populations. large increases of serum CK activities [16]. In untrained
persons, serum CK appears to increase proportionately
to the duration and intensity of the exercise; however,
432
Creatine Kinase
conditioned persons often show smaller changes in kinase inhibitors. Clin Chem 1977; 23: 1888-
serum CK activity. Sustained exercise, such as in well 92.
trained, long-distance runners, increases the CK-MB 6 Schumann G, Bonora R, Ceriotti F, Clerc-
content of skeletal muscle, owing to the phenomenon of Renaud P, Ferrero CA, Ferard G, et al. IFCC
fetal reversion, in which fetal patterns of protein primary reference procedures for the
synthesis reappear. Thus serum CK-MB isoenzyme may measurement of catalytic activity
increase in such circumstances. This explanation may concentrations of enzymes at 37 C. Part 2.
also account for the elevated CK-MB values sometimes Reference procedure for the measurement of
observed in chronic renal failure (uremic myopathy). catalytic concentration of creatine kinase. Clin
Chem Lab Med 2002; 40: 635-42.
During normal childbirth, there is a sixfold elevation in 7 Koumantakis G, Smith C, Wyndham L. More
maternal total serum CK activity. Surgical intervention on the PST blood collection tubes. Ann Clin
during labor further increases the activity of CK in Biochem 1991; 28: 423-4.
serum. 8 Frank JJ, Bermes EW, Bickel MJ, Watkins BF.
Effect of in vitro hemolysis on chemical values
A number of drugs can increase serum CK activities. for serum. Clin Chem 1978; 24: 1966-70.
Release of CK from skeletal muscle may occur after 9 Meltzer HY. Factors affecting serum creatine
pharmacological doses of drugs listed in Table 4. phosphokinase levels in the general population:
the role of race, activity, and age. Clin Chim
Decreased serum CK activity is observed in immobilized Acta 1971; 33: 165-72.
persons (e.g., patients with spinal cord injury) and in 10 Schumann G, Klauke R. New IFCC reference
patients with cancer cachexia. procedures for the determination of catalytic
activity concentrations of five enzymes in
CK Performance Goals serum: preliminary upper reference limits
For a single laboratory, the analytical goals for desirable obtained in hospitalized subjects. Clin Chim
performance of CK assays, derived from biological Acta 2003; 327: 69-79.
variation of the enzyme, are as follows [17]: 11 Hood D, Van Lente F, Estes M. Serum enzyme
alterations in chronic muscle disease. A biopsy-
Imprecision (CV): 11.4% based diagnostic assessment. Am J Clin Pathol
Bias: 11.5% 1991; 95: 402-7.
Total error: 30.3% 12 Panteghini M. Enzyme and muscle diseases.
Curr Opin Rheumathol 1995; 7: 469-74.
The current interlaboratory coefficient of variation (CV) 13 Beetham R. Biochemical investigation of
on proficiency testing surveys at a CK activity of suspected rhabdomyolysis. Ann Clin Biochem
approximately two times the upper reference limit ranges 2000; 37: 581-7.
from approximately 1.5% to 5% in different peer groups, 14 Dolci A, Panteghini M. The exciting story of
implying that current analytical techniques are adequate cardiac biomarkers: from retrospective
in terms of imprecision [18]. In an international study detection to gold diagnostic standard for acute
assessing the accuracy of CK results from six diagnostic myocardial infarction and more. Clin Chim
companies, all the evaluated assays showed significant Acta 2006; 369: 179-87.
bias when compared to the IFCC reference procedure 15 Panteghini M. The new definition of
[19]. Given the low assay imprecision, > 95% of myocardial infarction and the impact of
participating laboratories applying Abbott, Beckman, troponin determination on clinical practice. Int J
and Roche analytical systems were, however, expected Cardiol 2006; 106: 298-306.
to comply traceability within the biologically derived 16 Lott JA, Landesman P. The enzymology of
total error budget [19]. skeletal muscle disorders. CRC Crit Rev Clin
Lab Sci 1984; 20: 153-90.
References 17 Ricos C, Garcia-Lario JV, Alvarez V, Caval F,
1 Tsung SW. Creatine kinase isoenzyme patterns Domenech M, Hernandez A, et al. Biological
in human tissue obtained at surgery. Clin Chem variation database, and quality specifications
1976; 22: 173-5. for imprecision, bias and total error (desirable
2 Bais R, Edwards JB. Creatine kinase. CRC Crit and minimum). The 2004 update.
Rev Clin Lab Sci 1982; 16: 291-335. www.westgard.com/guest26.htm
3 Oliver IT. A spectrophotometric method for the 18 College of American Pathologists. Survey
determination of creatine phosphokinase and 2007. CARB-B Cardiac MarkersParticipant
myokinase. Biochem J 1955; 61: 115-22. summary. Rev 4/2007.
4 Rosalki SB. An improved procedure for serum 19 Jansen R, Schumann G, Baadenhuijsen H,
creatine phosphokinase determination. J Lab Franck P, Franzini C, Kruse R, et al. Trueness
Clin Med 1967; 69: 696-705. verification and traceability assessment of
5 Szasz G, Gerhardt W, Gruber W. Creatine results from commercial systems for
kinase in serum. 3. Further study of adenylate measurement of six enzyme activities in serum.
An international study in the EC4 framework of
433
Creatine Kinase
the Calibration 2000 project. Clin Chim Acta Chicago: Year-Book Medical Publishers, 1985;
2006; 368: 160-7. pp. 163-97.
20 Cherian GA, Hill JG. Age dependence of serum 23 Young DS, Pestaner LC, Gibberman V. Effects
enzymatic activities (alkaline phosphatase, of drugs on laboratory tests. Clin Chem 1975;
aspartate aminotransferase, and creatine kinase) 21: 286D.
in healthy children and adolescents. Am Soc 24 Thompson PD, Clarkson P, Karas RH. Statin-
Clin Pathol 1978; 70: 783-9. associated myopathy. JAMA 2003; 289: 1681-
21 Miller GW, Chinchilli VM, Gruemer H, Nance 90.
WE. Sampling from a skewed population 25 Guthrie RM, Lott JA. Abnormal serum CK-MB
distribution as exemplified by estimation of the following an amitriptyline overdose: a case
creatine kinase upper reference limit. Clin report and review of the literature. J Family
Chem 1984; 30: 18-23. Pract 1986; 22: 550-5.
22 Lott JA, Wolf PL. Creatine kinase. In: Clinical
enzymology: a case-oriented approach.
Table 1: CK Activities and Cytoplasmic Isoenzyme Patterns in Human Tissues*
Tissue Relative CK Activity CK-MM% CK-MB% CK-BB%

(expressed as multiple
of serum activity)
Skeletal muscle
(type I, slow twitch) 50,000 97 3 <1
Skeletal muscle
(type II, fast twitch) 50,000 99 1 <1
Myocardium 10,000 78 22 <1
Brain 5000 0 0 100
Gastrointestinal tract 5000 3 1 96
Urinary bladder 4000 2 6 92
*Adapted from Panteghini 1995 [12]
434
Creatine Kinase
Table 2: Measurement Conditions for IFCC Reference Method
Temperature 37C
Wavelength 340 nm
Buffer, concentration Imidazole, 50 mmol/L
pH (at temperature) 6.5 (37C)
Volume fraction of sample 0.0435 (1:23)
Concentration of reagents in the final complete reaction mixture:
Creatine phosphate 30 mmol/L
ADP 2 mmol/L
EDTA 2 mmol/L
Mg acetate 10 mmol/L
N-acetylcysteine 20 mmol/L
AMP 5 mmol/L
Diadenosine-5-pentaphosphate 0.01 mmol/L
D-glucose 20 mmol/L
Pyridine nucleotide (NADP) 2 mmol/L
Hexokinase (37C) 4000 U/L
G6PD (37C) 2800 U/L
Table 3: Serum CK Activities in Apparently Healthy Populations

Population: Newborns
Subpopulation: Males and females with physiological vaginal deliveries
Enzyme activity: Up to 10 times the upper reference limits in adults
Population: Infants
Subpopulation: Males and females
Enzyme activity: Up to 3 times the upper reference limits in adults [16]
Population: Children and adolescents [9,20]

Subpopulation:
a. Males
b. Females
c. Males and females
Enzyme activity:
a. CK remains within the adult range from 5 to 12 years of age; thereafter, it rises maximally to 1.5 to 2.5
times the upper reference limit in male adults
b. CK remains fairly constant from 5 to 12 years and then gradually declines
with age up to 20 years
c. Male CK activities are 2 times those for females of the same age
Population: Adults [9,21,22]

Subpopulation:
a. Males and females
b. Whites and blacks
c. Ambulatory, hospitalized
d. Hospitalized, bedridden
Enzyme activity:
a. Mean CK activity for men is 1.5-fold that of women
b. Blacks have 1.2- to 1.3-fold higher CK activities than whites of the
same sex
c. Ambulatory patients show a trend toward higher CK activities
d. CK activities can be very low during inactivity
435
Creatine Kinase
Table 4: Drugs That Can Increase Serum CK Activities
Condition Relative Increase Comments

Drug-induced changes at pharmacological doses:
Danazol Slight [16] Some patients
Halofenate Slight [23] Mild, transient effect
Pindolol Slight [16] Some patients
Stanozolol Slight [16] Some patients
Succinylcholine Slight [16] Some patients; dramatic increases
if hyperpyrexia occurs
Statins Slight/Moderate [24] Myopathy observed in some patients;
may produce rhabdomyolysis (0.02%)
Aminocaproic acid Moderate [16] Myopathy observed in some patients
Amphotericin B Moderate [16] Proximal myopathy observed
Carbenoxolone Moderate [16,23] May cause hypokalemic myopathy
Haloperidol Moderate [16] Dystonia produced in some patients
Labetalol Moderate [16]
Lidocaine Moderate [16]
D-Penicillamine Moderate [16] Produced polymyositis
Perphenazine Moderate [16] Produced dystonia
Prochlorperazine Moderate [16] Produced dystonia
Quinidine Moderate [16]
Cyclopropane Pronounced [16] Dramatic increases if hyperpyrexia occurs
Diethyl ether Pronounced [16] Dramatic increases if hyperpyrexia occurs
Halothane Pronounced [16,23] Pronounced effect if causes malignant hyperpyrexia
Drug-induced changes in case of abuse or poisoning:
Amphetamine Pronounced [16] Huge increases
Barbiturates Pronounced [16] Huge increase in comatose patients
Ethanol Pronounced [16] Usually a result of alcoholic myopathy
Heroin Pronounced [16] Huge increases; may produce rhabdomyolysis
Amitriptyline Moderate [25] In suicide cases
Procedure Calculations
Average the change of absorbance (A) per min
Principle values. If the A values are decreasing or appear
CK is measured by a coupled reaction sequence inconsistent (e.g., not linear), repeat the assay with a
described earlier (Oliver-Rosalki method). After a 5-min diluted specimen.
incubation to allow the lag phase to go to completion,
the change in absorbance at 340 nm is monitored at 37C CK activity (U/L) = A/min ___1__ total vol_
(measurement interval, 2 min). The CK activity is molar absorptivity sample vol
directly related to the increase in absorbance of NADPH,

obtained from the zero-order portion of the curve of
= A/min _1_ _2.30_ 1000
absorbance versus time. The results are expressed in
U/L. 6.30 0.10
Reagents = A/min 3651

Reagents may be purchased from several
manufacturers.
Note: The molar absorptivity of NADPH is 630 m2/mol
Assay at 339 nm. The method should be linear to 1500 U/L.
Follow instructions of the manufacturers
analytical system (i.e., instrument, reagents, and
calibrators/controls).
436
Creatine Kinase Isoenzymes
Creatine Kinase Isoenzymes

Mauro Panteghini
Name: Creatine kinase (CK, adenosine triphosphate: creatine N-phosphotransferase)

Clinical significance: Refer to Chapter 36, Cardiac and Muscle Disease, in the 5th edition of
Molecular mass: Varies from 78,500 to 85,100 D; molecular mass of individual
subunits is half that of intact molecule
Known isoenzymes: CK1 (CK-BB), CK2 (CK-MB), CK3 (CK-MM), CKmt (mitochondrial)
Biochemical reaction: (See below.)

i membrane of mitochondria. CK-MM is found
Creatine kinase (CK) is a dimeric enzyme that catalyzes predominantly in skeletal muscle; the richest source of
the reversible phosphorylation of creatine by ATP. CK CK-MB is myocardial muscle, and CK-BB is found
comprises two subunits which are designated M and mainly in brain and intestine. For details of relative CK
B; hence the protein may exist in three forms, CK- activities and isoenzyme composition of different tissues
MM, CK-MB, and CK-BB, respectively. All three of refer to Table 1. The CK activity in serum of healthy
these isoenzymes are found in the cytosol of the cell. people is due almost exclusively to MM activity (though
There is also a fourth isoenzyme (CKmt) located in the small amounts of CK-MB may be present) and is the
result of physiological turnover of muscle tissue.
i
Creatinine Kinase isoenzymes When electrophoresed, CK-MM runs closest to the
Previous and current authors of this method: cathode, CK-MB has intermediate mobility, and CK-BB
First edition: John F. Chapman, Linda L. Woodard, moves farthest from the point of application toward the
Lawrence M. Silverman anode. CKmt, which runs more cathodally than the MM
Methods edition: John F. Chapman, Linda L. Woodard, fraction, is usually associated with tissue necrosis that
Lawrence M. Silverman accompanies severe anoxic shock and severe liver
Second edition: John F. Chapman, Linda L. Woodard, disease. CK activity may also be found in a
Lawrence M. Silverman macromolecular formthe so-called macro-CK. Macro-
Third edition: Monica Payne, Peter E. Hickman CK is found, often transiently, in sera of up to 6% of
Fourth edition: Not updated hospitalized patients, but only a small proportion of
Fifth edition: Mauro Panteghini these have increased CK activities in serum. It exists in
437
2
two forms, types 1 and 2. Macro-CK type 1 is a complex MM remaining cathodic to the point of application. The
of CK, typically CK-BB, and an immunoglobulin, often discriminating power of electrophoresis also allows the
IgG. It often occurs in women older than 50. Macro-CK detection of abnormal CK bands (e.g., macro-CK). The
type 2 is oligomeric CKmt found predominantly in disadvantages of electrophoresis include that the
adults who are severely ill with malignancies or in turnaround time is relatively long, the procedure is
children who have notable tissue distress [1]. highly labor intensive and not adaptable to clinical
chemistry analyzers in emergency situations, and
Both M and B subunits have a C-terminal lysine residue, interpretative skills are required.
but only the former is hydrolyzed by the action of
carboxypeptidases present in blood. Carboxypeptidases Immunochemical methods are applicable to the direct
B (EC 3.4.17.2) or N (arginine carboxypeptidase; EC measurement of CK-MB. In the immunoinhibition
3.4.17.3) sequentially hydrolyze the lysine residues from technique, an anti-CK-M subunit antiserum is used to
CK-MM to produce two CK-MM isoformsCK-MM2 inhibit both M subunits of CK-MM and the single M
(one lysine residue removed) and CK-MM1 (both lysine subunit of CK-MB and thus allow determination of the
residues removed). The loss of the positively charged enzyme activity of the B subunit of CK-MB, the B
lysine produces a more negatively charged CK molecule subunits of CK-BB, and macro-CKs. To determine CK-
with greater anodic mobility at electrophoresis. Because MB, this technique assumes the absence of CK-BB (and
CK-MB has only one M subunit, the dimer coded by the of the other sources of interference such as macro-CKs)
M and B genes is named CK-MB2 and the lysine- from the tested serum. Because the CK-B subunit
hydrolyzed dimer is named CK-MB1. The assay of the accounts for half of the CK-MB activity, the change in
CK isoforms requires special techniques, such as high- absorbance should be doubled to obtain CK-MB activity.
voltage electrophoresis (with gel cooling), HPLC, This results in a significant decrease of the analytical
chromatofocusing, or immunoassay [2]. sensitivity of the method. If present, atypical macro-CK
may result in falsely elevated CK-MB results. Owing to
For many years, CK isoenzyme analysis has been used its low sensitivity and specificity, the immunoinhibition
to confirm or refute a diagnosis of acute myocardial technique has been largely supplanted by mass assays of
infarction. CK-MB content in myocardium is CK-MB.
approximately 20% of total myocardial CK, so damage
to myocardium is accompanied by release of CK-MB In contrast with immunoinhibition, which measures the
into the circulating blood in higher proportion. It is now, CK-MB isoenzyme by determination of its catalytic
however, more advantageous to use more cardiac- activity, mass immunoassays measure CK-MB protein
specific nonenzymatic markers, such as cardiac troponin concentrations. A number of mass assays using various
I or T [3]. labels are now commercially available and are used for
routine determination of CK-MB. Measurements use the
Historically, CK-MB has been measured by determining sandwich technique, in which one antibody
the catalytic activity after separation of the isoenzymes specifically recognizes only the MB dimer. The
by electrophoresis or ion-exchange chromatography. sandwich technique ensures that only CK-MB is
Subsequently, immunological techniques measuring estimated, because neither CK-MM nor CK-BB reacts
catalytic activity were introduced. These methods were with both antibodies. Mass assays are more sensitive
limited in their use, since interferences with macro-CK than activity-based methods, with a limit of detection for
or CKmt gave low specificity. Currently, the most CK-MB usually <1 g/L [4]. Other advantages include
common approach is to measure concentrations of the sample stability, noninterference with hemolysis, drugs,
CK-MB protein (mass) by using immunoassays with or other catalytic activity inhibitors, full automation, and
monoclonal antibodies for the MB dimer [4]. fast turnaround time.
Reference and Preferred Methods Specimen

At the present time, there is no reference method for CK Fresh serum free from hemolysis is the specimen of
isoenzyme analysis. The techniques most commonly choice for analysis of the CK isoenzyme pattern. Of the
used are electrophoresis and various immunological three commonly seen isoenzymes, CK-BB activity is the
methods. least stable. Adding a thiol such as 2-mercaptoethanol to
the serum improves its stability [5]. CK-MB activity is
Electrophoretic methods are useful for separation of all not significantly reduced when the separated serum is
of the CK isoenzymes. The isoenzyme bands are stored for up to 48 hours at 4C or 1 month at 20C.
visualized by incubating the support (e.g., agarose or
cellulose acetate) with a concentrated CK assay mixture Since mass measurement is not subject to the loss of
using the reverse reaction. The NADPH formed in this enzyme activity, CK-MB protein concentration in serum
reaction is then detected by observing the bluish-white is stable for weeks, whether the specimen is stored under
fluorescence after excitation by long-wave (360 nm) refrigeration, and for several years if stored at 20C.
ultraviolet light. NADPH may be quantified by
fluorescence densitometry, which is capable of detecting Interferences
bands of 2 to 5 U/L. The mobility of CK isoenzymes at EDTA, oxalate/fluoride, and citrated plasma must not be
pH 8.6 toward the anode is BB > MB > MM, with the used, since these inhibit CK activity. Because CK-MB
438
3
mass assays are not subject to the interferences of CK-MB Performance Goals
activity assays, hemolysis and various anticoagulants For a single laboratory, the analytical goals for desirable
generally do not interfere. performance of CK-MB mass assays, derived from
biological variation of the protein, are as follows [9,10]:
CK Isoenzymes Reference Intervals
With the CK-MB mass assay, the upper reference limit Imprecision (CV): 9.2%
for males is 5.0 g/L, with values for females being less Bias: 16.0%
than for males, although many laboratories use a single Total error: 31.3%
reference interval (male). Sustained exercise, such as in
well trained, long-distance runners, increases the CK- The current interlaboratory coefficient of variation (CV)
MB content of skeletal muscle, which may produce on proficiency testing surveys at a CK-MB mass
abnormal serum CK-MB concentrations. concentration around the upper reference limit ranges
from 3.7% to approximately 9% in different peer groups
To better separate non-myocardial infarction from [11]. At higher CM-MB concentrations (70 g/L), data
myocardial infarction patients, the use of a relative from the College of American Pathologists survey show
index (RI) is necessary. The RI relates CK-MB mass CVs between 2.5% and 6% [12].
concentration in g/L to measured total CK activity in
U/L. Results are expressed as a percentage: References
1 Panteghini M, Bais R, van Solinge WW.
Physiological: 3% Enzymes. In: Burtis CA, Ashwood ER, Bruns
Equivocal: 3 to 5% DE, eds. Tietz Textbook of Clinical Chemistry
Consistent with myocardial necrosis: >5% and Molecular Diagnostics. 4th ed.
Philadelphia, PA: Elsevier Saunders, 2006:597-
To appropriately use the RI, blood sampling between 8 643
and 36 hours from symptom onset is necessary. 2 Panteghini M. Serum isoforms of creatine
kinase isoenzymes. Clin Biochem 1988; 21:
CK-BB may be elevated in neonates, particularly in 211-8.
brain-damaged or very low birth weight newborns. The 3 Panteghini M. The new definition of
presence of CK-BB in blood, usually at low myocardial infarction and the impact of
concentrations, may however represent a physiological troponin determination on clinical practice. Int J
finding in the first days of life. Cardiol.. 2006; 106: 298-306.
4 Panteghini M. Diagnostic application of CK-
MB mass determination. Clin Chim Acta 1998;
Interpretation 272: 23-31.
Detection of CK-MB in increased amounts has been the 5 Abbott LB, Lott JA. Reactivation of serum
hallmark for the biochemical diagnosis of acute creatine kinase isoenzyme BB in patients with
myocardial infarction for the past several decades [6]. malignancies. Clin Chem 1984; 30: 1861-3.
However, the interpretation of results in the presence of 6 Adams JE III, Abendschein DR, Jaffe AS.
other illness, particularly skeletal muscle injury, is Biochemical markers of myocardial injury. Is
difficult [7]. It is important to remember that CK-MB MB creatine kinase the choice for the 1990s?
concentrations in serum may not always accurately Circulation 1993; 88: 750-63.
reflect myocardial damage. There have been numerous 7 Adams JE 3rd, Sicard GA, Allen BT, Bridwell
reports of CK-MB elevation associated with non-cardiac KH, Lenke LG, Davila-Roman VG, et al.
conditions, including polymyositis, malignancy, and Diagnosis of perioperative myocardial
hypothyroidism. Serial sampling after the onset of chest infarction with measurement of cardiac
pain, showing the typical rise and fall of CK-MB values, troponin I. N Engl J Med 1994; 330: 670-4.
significantly increases the diagnostic accuracy for 8 Jaffe AS, Babuin L, Apple FS. Biomarkers in
myocardial infarction. acute cardiac disease. The present and the
future. Circulation 2006; 48: 1-11.
CK-MB determination can be used with some success to 9 Ricos C, Garcia-Lario JV, Alvarez V, Caval F,
estimate the extent of myocardial necrosis to assist with Domenech M, Hernandez A, et al. Biological
assessment of infarct prognosis. When comparing peak variation database, and quality specifications
CK-MB with estimates of infarct size, good correlations for imprecision, bias and total error (desirable
can be obtained (Table 2). A problem with using CK- and minimum). The 2004 update.
MB for this purpose is the requirement for frequent www.westgard.com/guest26.htm.
sampling to ensure that peak CK-MB values are 10 Ross SM, Fraser CG. Biological variation of
correctly identified. It appears in some way provocative, cardiac markers: analytical and clinical
but it is quite possible that within a few years, CK-MB considerations. Ann Clin Biochem 1998; 35:
determination will no longer be offered, because cardiac 80-4.
troponin measurement is clearly superior in this clinical 11 Reference Centre for Quality Assurance
setting [8]. External Quality Assessment Scheme (Regione
439
4
Toscana, Italy). Comprehensive Cardiac 13 Panteghini M. Enzyme and muscle diseases.

Markers Survey Report 2006. Curr Opin Rheumathol 1995; 7: 469-74.
12 College of American Pathologists. Survey
2007. CARB-B Cardiac Markers Participant
Summary. Rev 4/2007.
Tables
Table 1: CK Activities and Cytoplasmic Isoenzyme Patterns in Human Tissues*
Tissue Relative CK activity CK-MM% CK-MB% CK-BB%

(expressed as multiple
of serum activity)
Skeletal muscle
(type I, slow twitch) 50,000 97 3 <1
Skeletal muscle
(type II, fast twitch) 50,000 99 1 <1
Myocardium 10,000 78 22 <1
Brain 5000 0 0 100
Gastrointestinal tract 5000 3 1 96
Urinary bladder 4000 2 6 92
*Adapted from Panteghini 1995 [13]
Table 2: Decision Limits of CK-MB Mass Peak for Infarct Size Definition
Microscopic myocardial infarction (MI) (focal necrosis) <10 g/L

Small MI (<10% left ventricle) 10-60 g/L
Medium MI (10-30% left ventricle) 60-225 g/L
Large MI (>30% left ventricle) >225 g/L
440
Creatinine
Creatinine
Edmund J. Lamb
Name: Creatinine
Clinical significance: Increased concentrations in blood indicate renal impairment Refer to Chapter
30, Renal Function, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular formula: C4H7N3O
Merck Index: 2552
Chemical class: Creatine end-product metabolite
Structure:
Reference method: GC-IDMS

i was then commonly adsorbed onto aluminum
Creatinine is a widely used test of kidney function. Most magnesium silicate clays, including Fullers earth
chemical methods for measuring creatinine are primarily (Floridin) and Lloyds reagent, theoretically isolating it
based on the reaction with alkaline picrate (Table 1, from potential interferents (Table 1, Method 2) prior to
Method 1). In this reaction, first described by Jaffe in reaction with alkaline picrate. Such techniques were in
1886 [1], creatinine reacts with picrate ion in an alkaline widespread use in the 1970s, and a variation of this
medium to yield an equimolar orange-red Janovski procedure was originally proposed as a reference method
complex. Despite considerable literature on the subject, [5]. However, it became apparent that such methods did
the reaction mechanism and the structure of the product not eliminate interference from noncreatinine
remain unclear [2]. Use of this reaction to measure chromogens. Further, they were imprecise and
creatinine in urine was first described by Folin in 1904 unsuitable for automation, and by the 1980s they were
[3]. He later extended this principle to creatinine not widely used [6]. Other approaches that have been
measurement in deproteinized blood [4]. The reaction is used in an attempt to improve the general specificity of
nonspecific for creatinine, and a variety of modifications the Jaffe reaction have included acid blanking, the use of
have been introduced in an attempt to remove cation exchange resins [7,8], solvent extraction, and
interference from noncreatinine chromogens (see oxidation of interferents with compounds such as cerium
below). Measurements of creatinine in serum and plasma sulfate; generally these modifications have not proved
are generally considered equivalent, and the term serum practical. The use of dialysis membranes in continuous
will be used throughout this chapter. flow automation in the 1960s reduced interference from
protein, although other noncreatinine chromogens
End-Point Assays remained dialyzable.
-Ketomethylene groups found in proteins react with
alkaline picrate to form highly colored complexes. Early Kinetic Assays
end-point methods for the measurement of creatinine in The Jaffe reaction is first order with respect to
serum therefore required the use of a protein-free filtrate, creatinine, picrate, and hydroxide concentrations.
generated for example using tungstic acid. Creatinine Kinetic alkaline picrate methods for the determination of
creatinine (Table 1, Method 3) take advantage of the fact
that creatinine and noncreatinine chromogens have a
i differential rate of color development, thus allowing a
Creatinine
Previous and current authors of this method: rate-dependent separation of creatinine from interfering
First edition: Robert L. Murray substances. Deproteinization and/or adsorption of
Methods edition: Robert L. Murray interferents is not necessary, since the reaction between
Second edition: Robert L. Murray alkaline picrate and noncreatinine chromogens is slow
Third edition: Steven C. Kazmierczak and does not significantly occur during the usual kinetic
Fourth edition: Steven C. Kazmierczak reaction interval. This approach significantly reduces but
Fifth edition: Edmund J. Lamb does not eliminate interferences from noncreatinine
441
Creatinine
chromogens. Such methods gained in popularity with the reaction to reach equilibrium. This shortcoming can be
advent of instruments capable of making accurate overcome by a kinetic approach, but with a further
absorbance readings at precise, highly reproducible reduction in sensitivity. The approach has not been
intervals [6,9]. Kinetic Jaffe assays have been popular, partly due to poor sensitivity, poor precision,
implemented on various automated microprocessor- and the relatively high cost of reagents [6].
controlled instruments and are currently the most widely Creatininase and creatinase. An alternative, more
used approach to creatinine measurement. Extensive popular, approach has utilized the enzyme creatinase
literature exists on the choice of reactant concentrations (EC 3.5.3.3; creatine amidinohydrolase), which yields
and reading interval, as well as on the choice of sarcosine and urea, the former being measured with
wavelength and reaction temperature [6]. Brief further enzyme-mediated steps using sarcosine oxidase
comments include: (EC 1.5.3.1; yielding glycine, formaldehyde, and
Reading interval. Some noncreatinine chromogens have hydrogen peroxide) and peroxidase (Table 1, Method 5)
a rapid reaction with alkaline picrate (e.g., acetoacetate), [12]. The hydrogen peroxide can be detected with a
whereas others show a slower reaction (e.g., protein). To variety of methods. The potential interference due to
avoid the period of maximal reaction with these ascorbic acid can be overcome by the inclusion of
interferents, the majority of kinetic methods read over ascorbate oxidase. The influence of endogenous
the approximate time interval 10 to 80 seconds after intermediate creatine and urea can be overcome by a
mixing of the reagents and sample. preincubation step, initiating the reaction with
Picrate concentration. With respect to picrate, the Jaffe creatininase. This system has been incorporated in a
reaction is pseudofirst order up to 30 mmol/L picrate, point-of-care testing device using polarographic
with the majority of methods employing a concentration detection [13]. An alternative detection system involves
between 3 and 16 mmol/L. At concentrations above 6 measurement of the reduction of NAD by formaldehyde
mmol/L, the rate of color development becomes in the presence of formaldehyde dehydrogenase [14].
nonlinear, so a two-point fixed interval rather than Creatinine deaminase. The third enzyme system used for
multiple-data-point approach is required. creatinine quantitation is creatinine deaminase (EC
Wavelength. While the absorbance maximum of the 3.5.4.21; creatinine iminohydrolase), which catalyzes the
Jaffe reaction is approximately 490 nm, improved conversion of creatinine to N-methylhydantoin and
method linearity and reduced blank values have been ammonia (Table 1, Method 6) [15]. Early methods
reported at other wavelengths, the choice varying with concentrated on the detection of ammonia using either
hydroxide concentration. Generally, the absorbance of glutamate dehydrogenase [16], an ammonia electrode
the complex is measured between 510 and 520 nm. [17], or the Berthelot reaction. An alternative approach
Picrate ion, present in excess in the reaction solution, involves the enzyme N-methylhydantoin amidohydrolase
absorbs significantly at wavelengths below 500 nm. [18]. In all of these procedures, the reaction mixture
Temperature. The rate of Jaffe complex formation and must be free of ammonia-producing and ammonia-
the absorptivity of the complex are temperature- consuming materials. In addition, a correction for
dependent, measurable differences being observed even endogenous ammonia must be made. Although the
between 25C and 37C. The absorbance maximum of concentration of physiologically produced ammonia is
the colored product is also temperature sensitive. relatively low, substantial amounts of ammonia can be
Consequently, temperature control is an important formed by protein deamination if blood is left at room
component of assay reproducibility. temperature.
Enzymatic Assays Dry Chemistry Systems

Several creatinine-degrading enzymes have been A number of multilayer dry reagent methods have been
investigated for use in creatinine analysis. This approach described for the measurement of creatinine using
was first attempted in 1937 by Miller and Dubos, using enzyme-mediated reactions. An early two-slide
crude extracts of aerobic bacteria capable of forming approach employed creatinine deaminase, with the
creatinine-decomposing enzymes [10]. However, owing ammonia diffusing through a semipermeable and
to the requirement for large quantities of pure enzyme, optically opaque layer to react with bromophenol blue to
these approaches have only recently become give an increase in absorbance at 600 nm. A second
commercially feasible. There are primarily three multilayer film lacking the creatinine deaminase enzyme
approaches, described below. was used to quantitate endogenous ammonia, enabling
Creatininase. The enzyme creatininase (EC 3.5.2.10; blank correction [19]. A later, more precise, single slide
creatinine amidohydrolase) catalyzes the conversion of method utilized the creatininase/creatinase reaction
creatinine to creatine. The creatine is then detected with sequence [20]. Lidocaine metabolites have been reported
a series of enzyme-mediated reactions involving creatine to interfere with this method [21]. The creatinine
kinase, pyruvate kinase, and lactate dehydrogenase, with deaminase system described above has also been utilized
monitoring of the decrease in absorbance at 340 nm due and adapted for use as a point-of-care testing device
to NADH disappearance (Table 1, Method 4) [11]. [22]. In all cases, the color produced in the film is
Initiating the reaction with creatininase allows for the quantitated by reflectance.
removal of endogenous creatine and pyruvate in a
preincubation reaction. The kinetics of the reaction are Other Chemical Methods
poor, and a 30-min incubation is required to allow the
442
Creatinine
Other chemical approaches to the measurement of general use. Measurement of serum and urine creatinine
creatinine have been tried. These include (1) reaction is a core test for most diagnostic laboratories.
with 1,4-naphthoquinone-2-sulphonate [23]; (2) use of o- Laboratories must often provide results in a STAT mode
nitrobenzaldehyde to convert creatinine to on a 24-hour-a-day basis, frequently on samples of small
methylguanidine and its reaction with -naphthol and volume (i.e., from pediatric patients) that may be
sodium hypochlorite under alkaline conditions [24]; and compromised by interferences (e.g., bilirubin in
(3) reaction with 3,5-dinitrobenzoic acid (DNBA) jaundiced infants or adults). These factors will influence
[25,26] or its derivatives [27,28]. The DNBA procedure a laboratorys choice of methodology. The methodology
has been adapted for use with the dry-strip technology for the measurement of creatinine is complex by virtue
[29]. In this procedure, creatinine percolates through the of the number of variants of the Jaffe reaction and the
initial layers of the strip to react with DNBA to form a innovations attempted using enzymatic procedures to
reddish chromogen. The reaction is monitored by overcome the limitations of the former. Kinetic Jaffe
reflectance spectrophotometry at 560 nm. A reagent-free approaches predominate in wet chemistry analytical
mid-infrared method for urinary creatinine measurement systems and are relatively inexpensive. Enzymatic
has been described [30]. None of these reactions are methods tend to be more expensive than Jaffe assays and
widely used in clinical laboratories. are the methods used in dry chemistry systems, including
some point-of-care testing devices. Any laboratorian
High-Performance Liquid Chromatography assessing a new creatinine method (e.g., as part of an
Many authors have reported separation and quantitation analyzer purchase) should review the data for that
of creatinine with high-performance liquid method on interference due to bilirubin, protein, glucose,
chromatography (HPLC) (Table 1, Method 7), using and ketones/ketoacids. Bilirubin will also be an
both cation-exchange [31-33] and reversed-phase important consideration in enzymatic procedures that
techniques [34,35]. Separation is followed by generate hydrogen peroxide. Despite criticism of the
quantitation using either an on-stream Jaffe reaction Jaffe methods, there is invariably a good correlation
[31], use of creatinines native absorbance at between them and enzymatic procedures, with the
approximately 230 nm [32-37] or enzymatic detection differences likely to be due as much to calibration as to
[38]. All HPLC approaches are reported to be rapid, interference. However, there is general agreement that
specific, and precise; several have been proposed as with appropriate standardization, commercial enzymatic
candidate reference methods [33,38,39]. Thienpont et al. assays can, on most systems and in most samples,
reported between-day and within-run imprecision of provide results comparable to those of HPLC or IDMS.
<1% with deviation from gas chromatography isotope Provided the increased costs can be justified, it seems
dilution-mass spectrometry (GC-IDMS) target values of possible and probably desirable that enzymatic assays
+0.1% [39]. will gradually replace the kinetic Jaffe assay currently in
widespread use.
Isotope-Dilution Mass Spectrometry
The use of mass fragmentography as a reference method Of 722 participants in the United Kingdom National
for creatinine was first proposed by Bjorkhem et al. in External Quality Assessment Scheme for serum
1977 [40]. GC-IDMS is considered the method of choice creatinine in May 2007, 73% used the kinetic Jaffe
for establishing the true concentration of creatinine in reaction, 7% used an end-point Jaffe, 4% used an
serum because of its excellent specificity and precision. OLeary modification of a kinetic Jaffe assay, 3% used
Definitive methods have been described by several an enzymatic assay, and 13% used a dry-slide enzymatic
authors [41-43], and the Joint Committee on Traceability method. Of the kinetic Jaffe participants, the major
in Laboratory Medicine (JCTLM) nominated three GC- reagent manufacturers were Abbott Diagnostics (16%),
IDMS methods as reference measurement procedures Beckman Coulter Ltd (13%), Roche Diagnostics Ltd
[44]. However, creatinine requires prior derivatization (36%, Modular and Integra platforms), and Olympus Ltd
before GC analysis and a cation-exchange clean-up step (23%). A similar pattern of usage is seen in the Unites
is also necessary because creatine is derivatized into the States. Of 2931 participants in the 2007 College of
same chemical species as creatinine [44]. This limits the American Pathologists (CAP) program for serum
throughput of analysis and hence the general creatinine, 74% used the kinetic Jaffe reaction, 4% used
applicability of the technique. Recently, direct methods an end-point Jaffe, 3% used an enzymatic assay, and
in which HPLC is coupled with IDMS have been 18% used a dry-slide enzymatic method. Of the kinetic
reported, offering the potential for quicker turnaround Jaffe participants, the major reagent manufacturers were
[45,46]. Abbott Diagnostics (6%), Beckman Coulter Ltd (42%),
Dade Behring (37%), Bayer Diagnostics (2%), Roche
Reference and Preferred Methods Diagnostics (5%), and Olympus Ltd (6%).
As described above, GC-IDMS has been proposed by the
JCTLM as the definitive reference method for serum Specimen
creatinine measurement. HPLC methods have often been The methods described are suitable for the measurement
shown to give comparable performance. Although these of creatinine in serum, plasma, or diluted urine. Such
methods are generally used by commercial specimens are stable for at least 7 days at 4C. Serum
manufacturers to establish traceability of their assays, creatinine is also stable during long-term frozen storage
they are presently unlikely to find their way into more and after repeated thawing and refreezing [47].
443
Creatinine
However, it should be noted that delayed separation such as borate and phosphate, together with surfactant,
(beyond 14 h) of serum from erythrocytes leads to a have been used to minimize the effects of this
significant increase in apparent serum creatinine interference. A popular maneuver in this context has
concentration using some kinetic Jaffe (but not been the addition of potassium ferricyanide before the
enzymatic) assays, possibly due to release of alkaline picrate (OLeary method) which oxidizes
noncreatinine chromogens from the red cells [48]. bilirubin to biliverdin, hence reducing its interference
Bacterial contamination, which can occur in samples [58,59]. Enzymatic (creatinine deaminase/creatininase)
stored for long periods of time, has been reported to assays are also not immune to negative interference by
falsely lower creatinine values measured using the Jaffe bilirubin, which reacts with the peroxidase detection
reaction, purportedly due to bacterial production of a system [57]. This problem has been approached with the
substance that retards the reaction [49]. Creatinine addition of potassium ferricyanide (with limited success)
concentration increases in blood after meals containing or bilirubin oxidase.
cooked meat: ideally blood for serum creatinine
measurement should be obtained in the fasting state [50]. Reference Interval and Interpretation
Given the foregoing discussion, reference intervals for
Interferences serum creatinine are clearly method dependent. A
The Jaffe reaction is not specific for creatinine. Many selection of the reported reference values for different
compounds have been reported to produce a Jaffe-like methods are listed in the table below. Creatinine is
chromogen, including protein, glucose, ascorbic acid, formed by the condensation of creatine into a ring
levulose, ketone bodies [51,52], pyruvate, guanidine, structure, with the concomitant loss of a water molecule.
aminohippurate, uric acid, blood-substitute products The reaction occurs spontaneously in the body. Because
[53], and cephalosporins [54]; the reader is referred to the bulk of creatine resides in muscle, the amount of
several comprehensive reviews [6,55,56]. Ascorbic acid, creatinine present in a person reflects muscle mass.
levulose, glucose, and uric acid are capable of reducing Therefore, women have lower serum creatinine
picrate to picramate, and this product, with its concentrations than men, and children and infants have
absorbance maximum of 485 nm, causes overestimation lower serum creatinine concentrations than adults.
of creatinine. Pyruvate, acetone, acetoacetic acid, amino Reference ranges in older people are similar to those in
acids, and proteinor any compound with an active younger adults, despite the decline in kidney function
methyl or methylene groupintroduce error by coupling that occurs on average with ageing. Typically, reference
with picrate by a mechanism analogous to that of intervals for serum creatinine measured by Jaffe methods
creatinine. All these compounds can produce colored are 80 to 115 mol/L (0.9 to 1.3 mg/dL) in men and 53
compounds, which then increase the absorbance in the to 97 mol/L (0.6 to 1.1 mg/dL) in women [60].
region of the creatinine-picrate complex. The effect of However, as can be seen from the table, lower reference
ketones and ketoacids is probably of the greatest ranges are generally reported with enzymatic or
significance clinically, although the effect is very compensated assays. Serum creatinine concentration
method dependent. Thus reports on acetoacetate in patients with untreated end-stage renal disease
interference vary from a negligible increase to an (ESRD) may exceed 1000 mol/L (11 mg/dL).
increase of 310 mol/L in the apparent creatinine Urinary creatinine excretion is higher in men (124 to 230
concentration at an acetoacetate concentration of 8 mol/kg/d [14 to 26 mg/kg/d]) than in women (97 to 177
mmol/L. Kinetic Jaffe approaches have significantly mol/kg/d [11 to 20 mg/kg/d]). Creatinine excretion
reduced, but not eliminated, interferences from decreases with age: typically, for a 70 kg man, creatinine
noncreatinine chromogens. The degree of interference excretion will decline from approximately 14.5 to 9.1
from these compounds is dependent on the precise mmol/d with advancing age from 30 to 80 years [61].
reaction conditions chosen and the concentration of the Measurement of urinary creatinine excretion can be a
interferent present in the patients sample. It is generally useful indication of the completeness of a timed urine
considered that noncreatinine chromogens contribute collection. Creatinine excretion is often used as a method
approximately 20% of the apparent, Jaffe-measured, of normalizing the urinary excretion of analytes. That is,
creatinine concentration in normal serum samples. the excretion of the test analyte (in mmol or grams) is
Noncreatinine chromogens do not generally contribute to divided by the total amount of creatinine (in mmol or
measured urinary creatinine concentration. grams) excreted in the same urine specimen. This
method is a rough correction for volume differences
Bilirubin or other hemoglobin degradation products are between patient specimens. Similarly, expressing the
negative interferents in the Jaffe reaction, probably by concentration of a substance as a ratio to the creatinine
their oxidation in strong base to colorless compounds concentration is a useful method of adjusting for urinary
[57]. The resulting decrease in background decreases the concentration differences in random (spot) urine
measured absorbance and is interpreted as a lower samples.
creatinine concentration. The addition of buffering ions,
444
Creatinine
Serum Creatinine Reference Intervals

Study Method Age Serum Creatinine (mol/L)
Male n Female n
Gardner and Technicon 20-29 y 60-110 321 50-110 384
Scott (1980) [62] Autoanalyzer 30-39 y 60-120 537 50-110 698
method N11b 40-49 y 50-110 476 50-110 560
50-59 y 60-140 294 50-120 279
60-69 y 70-140 116 50-120 142
>69 y 50-160 38 60-140 50
Soldin and Hicks Kodak Ektachem 0-1 week 53-97 >100 53-97 >100
(1995) [63] (enzymatic) 1 week-1 month 27-62 >100 27-62 >100
1 month-1 y
1-18 y 18-35 >100 18-35 >100
18-62 >100 18-62 >100
Mazzachi et al. Hitachi 917 (rate- adults 62-106 293 44-80 269
(2000) [64] blanked,
compensated Jaffe)
Mazzachi et al. Hitachi 917 adults 59-104 293 45-84 269
(2000) [64] (creatininase/
creatinase)
Sugita et al. Hitachi 736 0-1 y 4-29* 18
(1995) [14] (creatinine 2-5 y 4-40* 16
amidohydrolase/sar 6-9 y 18-46* 41
cosine oxidase) 10 y 19-52* 15
adult 55-96 235 40-66 355
All values have been rounded up or down to the nearest whole number.
* males and females combined
To convert plasma creatinine in mol/L to mg/dL, multiply by 0.011
Creatinine Performance Goals GFR, and vice versa. Miller et al. have calculated that
Different methods for assaying serum creatinine have allowing for a total error in eGFR of 15%, the maximal
varying degrees of imprecision and accuracy. Targets for allowable bias of a creatinine assay at a concentration of
desirable imprecision are generally based on knowledge 88 mol/L (1.00 mg/dL) would be 3 mol/L (0.034
of the biological variation of an analyte [65]. Mean intra- mg/dL), a target achieved by very few laboratories or
individual biological variation (CVi) for serum methods [70]. As a result of reaction with noncreatinine
creatinine has been reported as 4.1% to 4.3% [66,67], chromogens, end-point Jaffe methods were typically
indicating a desirable analytical performance goal (based judged to overestimate true serum creatinine
on 50% of the upper estimate of the CVi) of 2.2% and a concentration by approximately 20% at physiological
desirable total error goal of 7.6% [44]. With the advent concentrations [44], and most commercially available
of automated kinetic analysis, within-laboratory, creatinine methods continue to demonstrate significant
between-day imprecision of approximately 3.0% can be positive bias compared to HPLC or IDMS methods,
expected at pathological concentrations, with slightly particularly at concentrations within the reference range
inferior performance within the reference interval [68]. [71,72]. A recent College of American Pathologists
A recent report suggests intra-laboratory imprecision at a (CAP) survey confirmed that most routine methods,
concentration of 88.4 mol/L (1.00 mg/dL) varying including some enzymatic assays, had significant
between 2.0% and 8.4% [44]. Clearly, many laboratories positive bias compared to a result obtained using an
do not perform to desirable performance standards IDMS reference procedure [70]. For example, at a true
defined in terms of biological variation [69], although concentration of 80 mol/L (0.90 mg/dL), kinetic
reference HPLC and IDMS methods do appear capable alkaline picrate assays on the Abbott Aeroset, Bayer
of achieving these performance goals. Advia 1650, Beckman LX20, Olympus AU5200, and
Vitros 950 systems had mean positive biases of 12
Additionally, accuracy remains a major issue and mol/L (0.14 mg/dL), 18 mol/L (0.20 mg/dL), 5
impinges on achievable between-laboratory agreement. mol/L (0.06 mg/dL), 11 mol/L (0.12 mg/dL), and 9
Acceptable performance criteria (CLIA-88) for mol/L (0.10 mg/dL), respectively. Variation in bias was
measurement of creatinine required laboratories to be related more to instrument manufacturer than method
accurate to within 27 mol/L (0.3 mg/dL) or 15% of principle, suggesting a strong influence of calibration
the peer group mean, whichever is greater. Given recent differences on bias. Indeed, one manufacturer (Roche)
emphasis on the use of estimated glomerular filtration had achieved almost zero bias across a range of assay
rate (eGFR) based upon serum creatinine data, this target principles and analytical platforms. This echoes earlier
is too liberal. The more a method overestimates true CAP data demonstrating that systematic differences in
creatinine, the greater will be the underestimation of the calibration of serum creatinine assays account for
445
Creatinine
85% of the observed differences between laboratories 5 Haeckel R. Assay of creatinine in serum, with
[73]. In 2003, the International Measurement Evaluation use of fullers earth to remove interferents. Clin
Programme (IMEP)-17 survey of approximately 1000 Chem 1981;27:179-83.
laboratories in 35 countries demonstrated almost 6 Spencer K. Analytical reviews in clinical
universal overestimation of serum creatinine in a pool biochemistry: the estimation of creatinine. Ann
with IDMS certified concentration of 75 mol/L (0.83 Clin Biochem 1986;23:1-25.
mg/dL). Of the 14 method groups, 13 demonstrated 7 Weatherburn MW, Trotman RB, Jackson SH.
mean positive bias compared to the reference value, Specific method for serum creatinine
typically varying between 10% and 15%. (The exception determination based on ion exchange
was the Roche enzymatic method.) A pool at chromatography and an automated alkaline
pathological concentration (169 mol/L, [1.86 mg/dL]) picrate reaction -- a proposed reference method.
showed an approximately equivalent number of Clin Biochem 1978;11:159-66.
laboratories with positive and negative bias (see 8 Polar E, Metcoff J. True creatinine
www.imep.ws for further information). Calibration chromogen determination in serum and urine by
differences contribute not only to bias but also to poor semi-automated analysis. Clin Chem
interlaboratory agreement. Proficiency studies 1965;11:763-70.
demonstrate that whilst between laboratory CVs of <3% 9 Bowers LD, Wong ET. Kinetic serum creatinine
are achievable within method groups, overall between- assays. II. A critical evaluation and review. Clin
laboratory agreement across methods is much poorer. Chem 1980;26:555-61.
Systematic variation between laboratories of 10 Miller BF, Dubos R. Determination by a
approximately 18 mol/L (0.2 mg/dL) is common specific, enzymatic method of the creatinine
[44,70]. content of blood and urine from normal and
nephritic individuals. J Biol Chem
Clearly, worldwide standardization of creatinine 1937;121:457-67.
measurement is a crucial aim. Recently, standardized 11 Moss GA, Bondar RJ, Buzzelli DM. Kinetic
serum matrix reference materials (SRM 967) with enzymatic method for determining serum
known creatinine concentrations (71 mol/L [0.80 creatinine. Clin Chem 1975;21:1422-6.
mg/dL] and 354 mol/L [4.00 mg/dL]) have been 12 Fossati P, Prencipe L, Berti G. Enzymic
prepared by the National Institute of Standards of creatinine assay: a new colorimetric method
Technology (NIST) and submitted to the JCTLM for based on hydrogen peroxide measurement. Clin
inclusion in a list of standardized reference materials Chem 1983;29:1494-6.
[44]. It is anticipated that the use of this material, in 13 Rowe DJ, Omar H, Barratt SL, Biggs P. An
combination with GC-IDMS reference methodology, evaluation of blood creatinine measurement by
will assist reagent manufacturers in achieving better creatinase on the NOVA M7 blood gas analyzer.
consensus between methods [74]. Although undoubtedly Clin Chim Acta 2001;307:23-5.
desirable, it must be recognized that standardization is 14 Sugita O, Uchiyama K, Yamada T, Sato T,
only one arm of the problem. Standardization will not Okada M, Takeuchi K. Reference values of
solve the problem of different reactivity with serum and urine creatinine, and of creatinine
noncreatinine chromogens across different patient clearance by a new enzymatic method. Ann Clin
samples, which can only be resolved by the use of highly Biochem 1992; 29: 523-8.
specific creatinine methods. 15 Tabata M, Kido T, Totani M, Murachi T.
Acknowledgments Automated assay of creatinine in serum as
simplified by the use of immobilized enzymes,
Data on current method usage was provided by Finlay creatinine deiminase, and glutamate
Mackenzie from the United Kingdom National External dehydrogenase. Anal Biochem 1983;134:44-9.
Quality Assessment Scheme, Birmingham, UK. 16 Tanganelli E, Prencipe L, Bassi D, Cambiaghi
S, Murador E. Enzymic assay of creatinine in
References serum and urine with creatinine iminohydrolase
1 Jaffe M. Uber den Niederschlag welchen and glutamate dehydrogenase. Clin Chem
Pikrinsaure in normalen Harn erzeugt und uber 1982;28:1461-4.
eine neue Reaction des Kreatinins. Z Physiol 17 Thompson H, Rechnitz GA. Ion electrode based
Chem 1886;10:391-400. enzymatic analysis of creatinine. Anal Chem
2 Vasiliades J. Reaction of alkaline sodium picrate 1974;46:246-9.
with creatinine: I. Kinetics and mechanism of 18 Fossati P, Ponti M, Passoni G, Tarenghi G,
formation of the mono-creatinine picric acid Melzi dEril GV, Prencipe L. A step forward in
complex. Clin Chem 1976;22:1664-71. enzymatic measurement of creatinine. Clin
3 Folin O. Beitrag zur Chemie des Kreatinins und Chem 1994;40:130-7.
Kreatins in Harne. Z Physiol Chem 19 Sundberg MW, Becker RW, Esders TW,
1904;41:223-4. Figueras J, Goodhue CT. An enzymic creatinine
4 Folin O. On the determination of creatinine and assay and a direct ammonia assay in coated thin
creatine in blood, milk and tissues. J Biol Chem films. Clin Chem 1983;29:645-9.
1914;17:475-81.
446
Creatinine
20 Mauck JC, Mauck L, Novros J, Norton GE. by high-performance liquid chromatography.

Development of a single-slide KODAK Clin Chem 1978;24:747-50.
EKTACHEM thin film assay for serum and 35 Spierto FW, MacNeil ML, Culbreth P, Duncan
urine creatinine [Abstract]. Clin Chem I, Burtis CA. Development and validation of a
1986;32:1197-8. liquid-chromatographic procedure for serum
21 Sena SF, Syed D, Romeo R, Krzymowski GA, creatinine. Clin Chem 1980;26:286-290.
McComb RB. Lidocaine metabolite and 36 Ambrose RT, Ketchum DF, Smith JW.
creatinine measurements in the Ektachem 700: Creatinine determined by "high-performance"
steps to minimize its impact on patient care. liquid chromatography. Clin Chem
Clin Chem 1988;34:2144-8. 1983;29:256-9.
22 Gray MR, Phillips E, Young DM, Price CP. 37 Paroni R, Arcelloni C, Fermo I, Bonini PA.
Evaluation of a rapid specific ward-based assay Determination of creatinine in serum and urine
for creatinine in blood. Clin Nephrol by a rapid liquid-chromatographic method. Clin
1995;43:169-73. Chem 1990;36:830-836.
23 Sullivan MS, Irrevere F. A highly specific test 38 Linnet K, Bruunshuus I. HPLC with enzymatic
for creatinine. J Biol Chem 1958;223:530-33. detection as a candidate reference method for
24 Van Pilsum JF, Martin RP, Kito E, Hess J. serum creatinine. Clin Chem 1991;37:1669-
Determination of creatine, creatinine, arginine, 1675.
guanidinoacetic acid, guanidine, and 39 Thienpont LM, Van Landuyt KG, Stockl D, De
methylguanidine in biological fluids. J Biol Leenheer AP. Candidate reference method for
Chem 1956;1956:225-36. determining serum creatinine by isocratic
25 Benedict SR, Behre JA. Some applications of a HPLC: validation with isotope dilution gas
new color reaction for creatinine. J Biol Chem chromatography-mass spectrometry and
1936;114:515-532. application for accuracy assessment of routine
26 Langley WD, Evans M. The determination of test kits. Clin Chem 1995;41:995-1003.
creatinine with 3,5-dinitrobenzoate. J Biol Chem 40 Bjorkhem I, Blomstrand R, Ohman G. Mass
1936;115:333-41. fragmentography of creatinine proposed as a
27 Parekh AC, Cook S, Sims C, Jung DH. A new reference method. Clin Chem 1977;23:2114-21.
method for the determination of serum 41 Welch MJ, Cohen A, Hertz HS, Ng KJ,
creatinine based on reaction with 3.5- Schaffer R, Van der Lijn P, et al. Determination
dinitrobenzoyl chloride in an organic medium. of serum creatinine by isotope dilution mass
Clin Chim Acta 1976;73:221-31. spectrometry as a candidate definitive method.
28 Sims C, Parekh AC. Determination of serum Anal Chem 1986;58:1681-5.
creatinine by reaction with methyl-3,5- 42 Stockl D, Reinauer H. Candidate reference
dinitrobenzoate in Methyl Sulfoxide. Ann Clin methods for determining target values for
Biochem 1977;14:227-32. cholesterol, creatinine, uric acid, and glucose in
29 Stevens JF, Tsang W, Newall RG. Measurement external quality assessment and internal
of bilirubin, cholesterol and creatinine in serum accuracy control. I. Method setup. Clin Chem
and plasma, by solid-phase reflectance 1993;39:993-1000.
spectroscopy. J Clin Pathol 1983;36:598-601. 43 Thienpont LM, Leenheer AP, Stockl D,
30 Shaw RA, Low-Ying S, Leroux M, Mantsch Reinauer H. Candidate reference methods for
HH. Toward reagent-free clinical analysis: determining target values for cholesterol,
quantitation of urine urea, creatinine, and total creatinine, uric acid, and glucose in external
protein from the mid-infrared spectra of dried quality assessment and internal accuracy
urine films. Clin Chem 2000;46:1493-5. control. II. Method transfer. Clin Chem
31 Brown ND, Sing HC, Neeley WE, Koetitz SE. 1993;39:1001-6.
Determination of "true" serum creatinine by 44 Myers GL, Miller WG, Coresh J, Fleming J,
high-performance liquid chromatography Greenberg N, Greene T, et al.
combined with a continuous-flow Recommendations for improving serum
microanalyzer. Clin Chem 1977;23:1281-3. creatinine measurement: a report from the
32 Harmoinen A, Sillanaukee P, Jokela H. Laboratory Working Group of the National
Determination of creatinine in serum and urine Kidney Disease Education Program. Clin Chem
by cation-exchange high- pressure liquid 2006;52:5-18.
chromatography. Clin Chem 1991;37:563-565. 45 Stokes P, O'Connor G. Development of a liquid
33 Rosano TG, Ambrose RT, Wu AH, Swift TA, chromatography-mass spectrometry method for
Yadegari P. Candidate reference method for the high-accuracy determination of creatinine in
determining creatinine in serum: method serum. J Chromatogr B Analyt Technol Biomed
development and interlaboratory validation. Clin Life Sci 2003;794:125-36.
Chem 1990;36:1951-1955. 46 Owen LJ, Wear JE, Keevil BG. Validation of a
34 Soldin SJ, Hill JG. Micromethod for liquid chromatography tandem mass
determination of creatinine in biological fluids spectrometry assay for serum creatinine and
447
Creatinine
comparison with enzymatic and Jaffe methods. Scottish new town. J Clin Pathol 1980;33:380-
Ann Clin Biochem 2006;43:118-23. 5.
47 DiMagno EP, Corle D, O'Brien JF, Masnyk IJ, 63 Soldin SJ, Hicks JM. Pediatric Reference
Go VL, Aamodt R. Effect of long-term freezer Ranges. Washington: AACC Press, 1995.
storage, thawing, and refreezing on selected 64 Mazzachi BC, Peake MJ, Ehrhardt V.
constituents of serum. Mayo Clin Proc Reference range and method comparison
1989;64:1226-34. studies for enzymatic and Jaffe creatinine
48 Shepherd J, Warner MH, Kilpatrick ES. assays in plasma and serum and early morning
Stability of creatinine with delayed separation urine. Clin Lab 2000;46:53-5.
of whole blood and implications for eGFR. Ann 65 Fraser CG, Hyltoft Petersen P, Libeer JC, Ricos
Clin Biochem 2007;44:in press. C. Proposals for setting generally applicable
49 Dilena BA. Bacterial interference with quality goals solely based on biology. Ann Clin
measurement of creatinine in stored plasma. Biochem 1997;34:8-12.
Clin Chem 1988;34:1007-8. 66 Gowans EM, Fraser CG. Biological variation of
50 Preiss DJ, Godber IM, Lamb EJ, Dalton RN, serum and urine creatinine and creatinine
Gunn IR. The influence of a cooked-meat meal clearance: ramifications for interpretation of
on estimated glomerular filtration rate. Ann results and patient care. Ann Clin Biochem
Clin Biochem 2007;44:35-42. 1988;25:259-63.
51 Soldin SJ, Henderson L, Hill JG. The effect of 67 Fraser CG, Cummings ST, Wilkinson SP,
bilirubin and ketones on reaction rate methods Neville RG, Knox JD, Ho O, et al. Biological
for the measurement of creatinine. Clin variability of 26 clinical chemistry analytes in
Biochem 1978;11:82-6. elderly people. Clin Chem 1989;35:783-786.
52 Watkins PJ. The effect of ketone bodies on the 68 Blijenberg BG, Brouwer HJ, Kuller TJ,
determination of creatinine. Clin Chim Acta Leeneman R, van Leeuwen CJ. Improvements
1967;18:191-6. in creatinine methodology: a critical
53 Ali AC, Mihas CC, Campbell JA. Interferences assessment. Eur J Clin Chem Clin Biochem
of o-raffinose cross-linked hemoglobin in three 1994;32:529-37.
methods for serum creatinine. Clin Chem 69 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
1997;43:1738-43. Hernandez A, Jimenez CV, et al. Current
54 Swain RR, Briggs SL. Positive interference databases on biological variation: pros, cons
with the Jaffe reaction by cephalosporin and progress. Scand J Clin Lab Invest
antibiotics. Clin Chem 1977;23:1340-2. 1999;59:491-500.
55 Bowers LD. Kinetic serum creatinine assays I. 70 Miller WG, Myers GL, Ashwood ER, Killeen
The role of various factors in determining AA, Wang E, Thienpont LM, et al. Creatinine
specificity. Clin Chem 1980;26:551-4. measurement: state of the art in accuracy and
56 Weber JA, van Zanten AP. Interferences in interlaboratory harmonization. Arch Pathol Lab
current methods for measurements of Med 2005;129:297-304.
creatinine. Clin Chem 1991;37:695-700. 71 Carobene A, Ferrero C, Ceriotti F, Modenese
57 Kroll MH, Elin RJ. Interference with clinical A, Besozzi M, de Giorgi E, et al. Creatinine
laboratory analyses. Clin Chem 1994;40:1996- measurement proficiency testing: assignment of
2005. matrix-adjusted ID GC-MS target values. Clin
58 Knapp ML, Mayne PD. Development of an Chem 1997;43:1342-7.
automated kinetic Jaffe method designed to 72 Lawson N, Lang T, Broughton A, Prinsloo P,
minimise bilirubin interference in plasma Turner C, Marenah C. Creatinine assays: time
creatinine assays. Clin Chim Acta for action? Ann Clin Biochem 2002;39:599-
1987;168:239-46. 602.
59 O'Leary N, Pembroke A, Duggan PF. A 73 Ross JW, Miller WG, Myers GL, Praestgaard J.
simplified procedure for eliminating the The accuracy of laboratory measurements in
negative interference of bilirubin in the Jaffe clinical chemistry: a study of 11 routine
reaction for creatinine. Clin Chem chemistry analytes in the College of American
1992;38:1749-51. Pathologists Chemistry Survey with fresh
60 Tietz NW, Shuey DF, Wekstein DR. frozen serum, definitive methods, and reference
Laboratory values in fit, aging individuals-- methods. Arch Pathol Lab Med 1998;122:587-
sexagenarians through centenarians. Clin Chem 608.
1992;38:1167-85. 74 Panteghini M, Myers GL, Miller WG,
61 Perrone RD, Madias NE, Levey AS. Serum Greenberg N. The importance of metrological
creatinine as an index of renal function: new traceability on the validity of creatinine
insights into old concepts. Clin Chem measurement as an index of renal function. Clin
1992;38:1933-53. Chem Lab Med 2006;44:1287-92.
62 Gardner MD, Scott R. Age- and sex-related 75 Brochner-Mortensen J, Rodbro P. Selection of
reference ranges for eight plasma constituents routine method for determination of glomerular
derived from randomly selected adults in a
448
Creatinine
filtration rate in adult patients. Scand J Clin Lab filtration rate from serum creatinine. J Am Soc
Invest 1976;36:35-43. Nephrol 2000;11(suppl):155A.
76 Walser M. Assessing renal function from 80 Lamb EJ, Tomson CRV, Roderick PJ.
creatinine measurements in adults with chronic Estimating kidney function in adults using
renal failure. Am J Kidney Dis 1998;32:23-31. formulae. Ann Clin Biochem 2005;42:321-45.
77 Cockcroft DW, Gault MH. Prediction of 81 Levey AS, Coresh J, Greene T, Stevens LA,
creatinine clearance from serum creatinine. Zhang YL, Hendriksen S, et al. Using
Nephron 1976;16:31-41. standardized serum creatinine values in the
78 Levey AS, Bosch JP, Lewis JB, Greene T, modification of diet in renal disease study
Rogers N, Roth D. A more accurate method to equation for estimating glomerular filtration
estimate glomerular filtration rate from serum rate. Ann Intern Med 2006;145:247-54.
creatinine: a new prediction equation. 82 Levey AS, Coresh J, Greene T, Marsh J,
Modification of Diet in Renal Disease Study Stevens LA, Kusek JW, et al. Expressing the
Group. Ann Intern Med 1999;130:461-70. Modification of Diet in Renal Disease Study
79 Levey AS, Greene T, Kusek JW, Beck GJ. A equation for estimating glomerular filtration
simplified equation to predict glomerular rate with standardized serum creatinine values.
Clin Chem 2007;53:766-72.
449
Creatinine
Table 1: Creatinine Methods Summary Table
Method 1: Jaffe; spectrophotometric, end point, quantitative

Principle of analysis: Creatinine + alkaline picrate Janovski complex (red)
Comments: Serum, plasma, diluted urine; described by Jaffe, 1886. Significant interference from noncreatinine
chromogens.
Method 2: Jaffe/Fullers earth; same as Method 1; creatinine isolated before analysis; can be removed with buffer or
picrate reagent added directly to creatinine adsorbent suspension
Comments: Serum, plasma, diluted urine; alternatively cation exchangers used as adsorbent. Interference from
noncreatinine chromogens reduced. No longer in routine use.
Method 3: Jaffe, kinetic; spectrophotometric, quantitative, kinetic analysis during early color formation
Comments: Serum, plasma, diluted urine; requires automated equipment for accurately timed and precise
absorbance measurements. Precise conditions vary between instruments. Interference from noncreatinine
chromogens reduced. This is currently the most popular method in diagnostic laboratories.
Method 4: Creatininase (creatinine amidohydrolase); enzymatic hydrolysis to creatine, which reacts in indicator reactions
monitored spectrophotometrically at 340 nm
_
Creatinine + H2O creatinine amidohydrolase creatine
_
Creatine + ATP CK creatine phosphate + ADP
_
ADP + phosphoenolpyruvate pyruvate kinase ATP + pyruvate
_
Pyruvate + NADH + H+ LD NAD+ + lactate
(CK, creatine kinase; LD, lactate dehydrogenase)
Comments: Poor sensitivity and precision, not widely used.
Method 5: Creatininase (creatinine amidohydrolase) is coupled with creatinase, generating creatine and sarcosine
sequentially. A further reaction with sarcosine oxidase leads to the production of hydrogen peroxide, which can be
detected by several indicator reactions.
_
Creatinine + H2O creatinine amidohydrolase creatine
_
Creatine + H2O creatinase sarcosine + urea
_
Sarcosine + O2 + H2O sarcosine oxidase formaldehyde + glycine + H2O2
_
Indicator (reduced) + H2O2 peroxidase indicator (oxidized) + 2H2O
Comments: Widely used specific method. Bilirubin interference remains a problem.
Method 6: Creatinine iminohydrolase; enzymatic hydrolysis of creatinine with formation of ammonia, which can be
quantitated spectrophotometrically or electrometrically
_
Creatinine creatinine iminohydrolase N-methylhydantoin + NH3
NH3 measured by: GLDH reaction, ammonia electrode, or colorimetrically
Comments: Requires high enzyme purity, used in some dry-slide techniques.
Method 7: High-performance liquid chromatography (HPLC); cation-exchange or reversed-phase chromatographic
separation of creatinine from other compounds
Principle of analysis: Creatinine quantitated by Method 1 or absorption at 230 nm or by enzymatic assay
Comments: Highly specific and precise, not used in routine practice, possible reference method.
Method 8: Isotope-dilution mass spectrometry (IDMS)
Principle of analysis: Measurement by IDMS following either initial GC or HPLC separation. GC approaches
require prior clean-up step
Comments: Highly specific and precise measurement. Reference method proposed by JCTLM. HPLC-IDMS
approaches may be applicable to routine use in the future.
450
Creatinine
Appendix: Glomerular Filtration Rate, Creatinine collection period. The test is initiated by having patients
Clearance, and Estimated GFR empty the bladder at the beginning of the timed period.
Urine is collected throughout the period, the bladder is
The GFR is widely accepted as the best overall measure again emptied at the end of the time period, and a blood
of kidney function, enabling a statement of the complex sample is obtained. Creatinine determinations are
functions of the kidney in a single numerical expression. performed on both samples, the urine volume measured
A decrease in GFR precedes kidney failure in all forms and the clearance calculated using the above formula.
of progressive disease. A number of methods are used to
measure the GFR; most involve the kidneys ability to In adults, the intra-individual day-to-day coefficient of
clear either an exogenous or endogenous marker. The variation for repeated measures of creatinine clearance
renal clearance of a substance is defined as the volume exceeds 25% [75]. Although tubular secretion
of plasma from which the substance is completely undermines the theoretical value of creatinine as a
cleared by the kidneys per unit time. marker of GFR, in the context of creatinine clearance,
this has previously been offset to some extent by the use
Provided a substance S is in stable concentration in the of nonspecific Jaffe methods to measure serum
plasma, physiologically inert, freely filtered at the creatinine, which leads to an overestimation of serum
glomerulus, and neither secreted, reabsorbed, concentration (see above). Nevertheless, creatinine
synthesized, nor metabolized by the kidney, then the clearance usually equals or exceeds inulin GFR in adults
amount of that substance filtered at the glomerulus is by a factor of 10% to 40% at clearances above 80
equal to the amount excreted in the urine (i.e., the mL/min. However, as GFR falls, tubular secretion of
amount of S entering the kidney must exactly equal the creatinine rises proportionately, and the creatinine
amount leaving it). The amount of S filtered at the clearance can reach nearly twice that of inulin [76].
glomerulus = GFR multiplied by plasma S concentration: Hence at best, creatinine clearance can only provide a
GFR PS. The amount of S excreted equals the urine S crude index of GFR.
concentration (US) multiplied by the urinary flow rate
(V, volume excreted per unit time). Estimated GFR
The mathematical relationship between serum creatinine
Since filtered S = excreted S and GFR can be improved by correcting for confounding
variables. Many equations have been derived which
GFR PS = US V estimate GFR using serum creatinine corrected for some
or all of gender, body size, race, and age. These may
GFR = (US V)/PS produce a better estimate of GFR than serum creatinine
alone. The Cockcroft and Gault and Modification of Diet
where: GFR = clearance in units of milliliters of plasma in Renal Disease (MDRD) Study equations have been
cleared of a substance per minute widely used in adults. Two further equations, Schwartz
US = urinary concentration of the substance and Counahan-Barratt, are recommended for use in
V = volumetric flow rate of urine in milliliters children.
per minute
PS = plasma concentration of the substance The Cockcroft and Gault equation, published in 1976, is
one of the earliest and most widely used of these
The term (US V)/PS is defined as the clearance of equations [77].
substance S and is an accurate estimate of GFR,
providing the aforementioned criteria are satisfied. Inulin Creatinine clearance (mL/min) = [140 age (years)]
satisfies these criteria and has long been regarded as the weight (kg)/(0.814 serum creatinine (mol/L)
most accurate (gold standard) estimate of GFR. A (0.85 if female)
variety of other exogenous (radioisotopic and non-
radioisotopic) and endogenous markers have been used The observed correlation coefficient between their
to estimate clearance. Commonly, clearance estimations predicted creatinine clearance and measured creatinine
are adjusted to an average body surface area (BSA) of clearance was 0.83 (R2 0.69), with predicted and mean
1.73 m2. measured clearance values differing by 20% or less in
Creatinine Clearance 67% of patients.
Because creatinine is endogenously produced, released
into body fluids at a constant rate, and freely filtered at The 6-variable (6-v) MDRD formula was published in
the glomerulus, its clearance can be measured as an 1999 following a retrospective re-analysis of the MDRD
indicator of GFR. Creatinine clearance has in the past Study [78]. GFR was determined in 1628 predominantly
been seen as more sensitive for detection of renal middle-aged patients with known kidney disease, using
125
dysfunction than serum creatinine measurement. I-iothalamate and expressed per 1.73 m2 BSA.
However, it requires a timed urine collection, which Creatinine was measured in a central laboratory using a
introduces its own inaccuracies, and is inconvenient and kinetic Jaffe assay on a Beckman Astra CX3 analyzer.
unpleasant. To do the test, one needs a precisely timed Stepwise multiple regression modeling on log-
urine collection and a blood sample taken during the transformed data was used to determine the set of
451
Creatinine
variables that best predicted GFR. The inclusion of MDRD equation, the constant factor 186 is replaced with
serum urea and albumin concentrations was found to the constant 175.
improve the predictive strength of the model (R2 0.903
against 125I-iothalamate GFR). There are several clinical situations in which precise
knowledge of GFR is important, and where reliance on
GFR (mL/min/1.73 m2) = 170 [serum creatinine equation-based estimates of GFR should be avoided.
(mol/L) 0.011312]-0.999 [age]-0.176 [0.762 if These include cancer chemotherapy or the use of any
patient is female] [1.180 if patient is black] other drug that is renally excreted and has a narrow
[serum urea (mmol/L) 2.801]-0.170 [serum albumin therapeutic margin; the assessment of potential living,
(g/L) 0.1]+0.318 related kidney donors; and the assessment of GFR in
patients with muscle-wasting disorders, including spina
The lack of requirement of knowledge of body weight bifida and paraplegia. When accurate GFR information
was a practical advantage compared to the Cockcroft and is required, reference methods should be used.
Gault equation, which also gave poorer accuracy in their
hands. The requirement for measurement of albumin and Serum creatinine is an imperfect marker of GFR, and
urea in addition to creatinine was clearly a limitation of therefore it is not altogether surprising that formulae
the 6-v MDRD equation, increasing both analytical costs based predominantly upon it are imperfect. Their use
and the contribution of analytical variation. Recognizing cannot circumvent the very significant spectral
this, these workers subsequently published (in 2000) a interferences affecting serum creatinine measurement,
simplified 4-variable (4-v) version of the formula, which and the formulae are critically susceptible to variations
does not require these measurements and did not result in creatinine assay calibration and specificity. However,
in appreciable loss of accuracy (R2 0.892 against 125I- despite these major limitations, the use of estimates of
iothalamate GFR), with 90% of subjects having a GFR GFR has been shown to improve the recognition of
estimate within 30% of the measured value [79]. patients with chronic kidney disease, compared to use of
serum creatinine alone. Internationally,
GFR (mL/min/1.73 m2) = 186 [serum creatinine recommendations now generally favor use of the 4-v
(mol/L) 0.011312]-1.154 [age]-0.203 [1.212 if black] MDRD formula for estimating GFR in adults.
[0.742 if female]
The 4-v MDRD equation has several advantages

compared to the Cockcroft and Gault equation: It was
developed and validated in a large population; there is no
requirement for patient weight; the validation was
against an iothalamate clearance estimate of GFR, and
reported GFR is corrected for BSA. Unlike the
Cockcroft and Gault equation, the model also takes into
account the fact that the relationship between serum
creatinine and GFR differs between Caucasians and
African Americans. Since its publication, a series of
studies have compared the applicability of the Cockcroft
and Gault and 4-v MDRD equations in a variety of
clinical settings [80]. From these, a general conclusion
can be drawn: For the detection of patients with GFR
<60 mL/min/.73 m2, the MDRD equation provides a
more accurate and clinically acceptable assessment of
GFR compared to the Cockcroft and Gault equation.
Both equations demonstrate deteriorating performance as
GFR increases, and neither adequately estimates GFR in
patients with physiological-range GFRs.
Interlaboratory differences in creatinine measurement

have been discussed previously. These differences,
compounded by poor precision around the upper limit of
the reference interval, can have a significant impact on
the estimation of GFR using these equation-based
approaches. Further, the MDRD equation was anchored
to a nonspecific kinetic rate assay for creatinine on an
analytical platform that will not be available in
perpetuity. Recognizing this, the authors of the MDRD
study have published a re-analysis of their data, enabling
the MDRD formula to be anchored to an ID-MS
creatinine assay [81,82]. In this version of the 4-v
452
CICLOSPORIN (CYCLOSPORINE A)
David W. Holt, Denise A. McKeown and Atholl Johnston
Name: rINN Ciclosporin, USAN Cyclosporine

Also known as: cyclosporin A, Neoral and Sandimmune
Clinical significance: Refer to Chapter 54, Laboratory Evaluation of the Transplant Recipient, in the
5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular formula: C62H111N11O12
Merck Index: 2748
Chemical class: Cyclic peptide
CH3
|
CH
||
CH
|
CH3 CH2
| |
CH --- CH3 CH3 CH --- CH3 CH3
| | | |
CH2 CH3 CH --- CH3 CH3 CH --- HO CH2 CH3
| | | | | | |
CH3 --- N --- CH --- CO --- N ---------- CH --- C --- N --- CH --- CO --- N --- CH --- C --- N --- CH2
| L L || L | L || |
CH3 CO O H O CO
| | |
CH3 --- CH --- CH2 --- CH N --- CH3
|
CH3 --- N H O H
| D L | L || L |
OC --- CH ---------- N --- CO --- CH --- N --- CO --- CH --- N --- C --- CH --- N --- CO --- CH L
| | | | | | |
CH2 H CH3 CH3 --- CH2 CH3 CH --- CH3 CH2
| | | |
O CH3 CH3 CH --- CH3
| |
CH2 CH3
|
CH2 --- OH
Structure:
Principles of Analysis and Current Usage prescribed, including psoriasis, atopic dermatitis,
rheumatoid arthritis, nephrotic syndrome and Behets
Ciclosporin is a unique immunosuppressive agent disease.
isolated from Tolypocladium inflatum Gams (NRRL
8044). It is approved for the prevention of graft rejection Ciclosporin is metabolized extensively to more than 30
following kidney, liver, heart, combined heart-lung, metabolites [1,2], and the ratio of metabolite to parent
lung, or pancreas transplantation. It can also be used for drug in the circulation is often greater than 3:1. There is
the treatment of transplant rejection in patients who have little evidence that the metabolites possess any
previously received other immunosuppressive agents. significant pharmacological activity, so assay
Following bone marrow transplantation, it can be used to development has focused on the measurement of the
prevent graft rejection and for the prophylaxis of graft- parent compound. The choice of sample matrix has been
versus-host disease (GVHD). There are several determined by the partitioning of ciclosporin between
autoimmune indications for which ciclosporin can be plasma and red cells. The drug distributes extensively
into red cells such that plasma/serum concentrations are
only about one third of those in whole blood [3]. In
addition, the distribution into the red cell is temperature
i
Ciclosporin (Cyclosporine A) dependent [4]. Thus the combination of much lower drug
Previous and current authors of this method: concentrations in plasma/serum, and the need to separate
First edition: Not done samples at a fixed temperature to maintain consistency,
Methods edition: Larry D. Bowers has resulted in the use of whole blood as the matrix for
Second edition: Not updated therapeutic drug monitoring (TDM) of the drug [5].
Third edition: Leslie M. Shaw
Fourth edition: Leslie M. Shaw, Larry D. Bowers Essentially, two analytical techniques have been used to
Fifth edition: David W. Holt, Denise A. McKeown, measure the drug: immunoassays and high-performance
Atholl Johnston liquid chromatography (HPLC). During the early clinical
use of ciclosporin, TDM was performed using a
453
polyclonal antibody-based radioimmunoassay kit immunoassay has an average bias of, say, 10%
produced by the manufacturer of the drug (Sandoz compared with HPLC, no universal correction factor can
Pharma, Basel, Switzerland, now Novartis Pharma)[6]. be applied to an immunoassay result to take into account
This kit was withdrawn some time ago, as were its interference due to ciclosporin metabolites. For clinical
successor monoclonal antibodybased samples from kidney transplant patients, immunoassay
radioimmunoassays, one relatively specific for the parent results can have a bias of approximately 10% to 40%,
compound, the other with a broad range of cross- compared with a chromatographic assay. This bias can
reactivity for the parent compound and its metabolites be substantially higher for samples from patients with
[7]. A number of commercial immunoassays have been hepatic dysfunction, since cross-reacting metabolites
developed over the intervening 2 decades [8,9]. tend to accumulate in the blood [7]. However, it must be
Immunoassays are very widely used to measure the drug, remembered that both immunoassay and
partly because of their relative ease of use [10]. Over the chromatographic techniques are prone to bias from
years, there has been some interest in the use of calibrator inaccuracy, whether the calibrators are
immunoassays that cross-react with both ciclosporin and prepared in-house or supplied with commercial kits
its major metabolites. This is because some centers [44,45]. Between-center differences might be minimized
wanted a measure of the high concentrations of if common calibrators or control material were used
ciclosporin metabolites that can accumulate in some more widely [46]. Calibrator and control material
clinical settings [11]. However, interest in nonspecific carrying a CE mark (the symbol indicating that the
measurements has declined substantially. product conforms to a number of European Community
Directives on manufacturing standards) and U.S. Food
A popular assay for such nonspecific measurements has and Drug Administration (FDA) approval for tacrolimus
been the polyclonal antibody fluorescence polarization is already available, and similar material for ciclosporin
immunoassay produced by Abbott Diagnostics. should be available in the near future [47]. A certified
Immunoassays in current use are listed in Table 1. They reference material would be helpful, since it would allow
all have a relatively high specificity for the parent both those preparing in-house calibrators and diagnostics
compound, but their lack of total specificity for the companies preparing kit calibrators to have access to a
parent drug is well documented [12,13]. common material with a defined value.
Until comparatively recently, the only chromatographic Despite the lack of a selective chromophore, HPLC-UV
technique available for the measurement of ciclosporin has the required sensitivity for the analysis of ciclosporin
was HPLC with ultraviolet (UV) detection. The [48,49]. However, HPLC-UV is susceptible to
molecule contains no selective chromophores, but low interferences from the sample matrix, co-elution from
wavelengths can be used for detection. Although a wide ciclosporin metabolites, and co-administered drugs and
variety of reversed-phase HPLC-UV methods have been their metabolites, resulting in laborious extraction
developed [14-18], they have been largely superseded by procedures and lengthy chromatographic run times [48-
the introduction of assays based on mass-spectrometric 50].
detection. This has brought both selectivity and
sensitivity to the measurement of the drug, and for the As a result there has been an increasing trend towards
first time, the possibility of attaining a gold standard the use of HPLC-MS and HPLC-MS/MS for the
reference technique. These single (HPLC-MS) [19-26] measurement of ciclosporin and other
and tandem (HPLC-MS/MS) [27-38] mass-spectrometric immunosuppressive drugs, largely because of the
assays have been developed for the analysis of availability of bench-top apparatus at competitive prices.
ciclosporin alone, for the simultaneous, multi-analyte The increased specificity of these techniques has allowed
analysis of the drug and its metabolites, and for the use of less tedious extraction procedures and shorter
ciclosporin together with other immunosuppressive chromatographic run times [48,49]. There have also been
drugs. Typical chromatographic assays currently in use advances towards the automation of these techniques.
are summarized in Table 1. The trend away from the use of HPLC-UV towards the
use of mass-spectrometric techniques is reflected by
Reference and Preferred Methods recent returns to the International Proficiency Testing
There is no defined reference method for ciclosporin. Scheme (www.bioanalytics.co.uk, accessed February 15,
Consensus documents on ciclosporin monitoring have 2008), with only 4 centers reporting use of HPLC-UV,
advocated that a well-validated HPLC method should be but 66 centers reporting use of HPLC-MS(/MS).
used for the selective measurement of ciclosporin ]5,39- However, immunoassay is still the predominant assay
43]. The immunoassays in current use have all been technique, with 470 centers (87%) reporting results with
shown to correlate with HPLC results but almost this methodology. Immunoassays require less technical
invariably produce a higher result. The bias associated input to operate, can be automated, and provide fast
with immunoassays is very dependent on the specificity turnaround times. However, the advantages of
of the antibody for the parent molecule. Within-subject automation and rapid analysis times are now being seen
and between-subject, the proportion of cross-reacting for HPLC-MS(/MS). There are recent reviews on the use
metabolites can vary, depending on such factors as of HPLC-MS(/MS) for the analysis of
hepatic function and concomitant drug therapy. immunosuppressive drugs by McKeown et al. [50],
Although it is possible to say that a particular Taylor et al. [49], and Deters et al.[48].
454
Specimen shorter time frame than traditional HPLC.) Vogeser et al.

The debate over which sample matrix to use for the [57] highlighted the interference that isobaric ions of
TDM of ciclosporin continued for some years after the ciclosporin metabolites can have on the HPLC-MS/MS
introduction of the drug. For the practical reasons analysis of the drug, utilizing the analogue ciclosporin D
mentioned above, whole blood was adopted as the as the internal standard. Analytical data from one patient
matrix of choice. The introduction of the more sensitive sample investigated in detail gave results of 277, 254,
and selective mass-spectrometric assays has enabled and 65 g/L for analyses made by immunoassay and
researchers to investigate alternative samples for HPLC-MS/MS with extended or limited
ciclosporin measurement. Assays validated for one chromatography, respectively. The inaccurate result
matrix should not be used for other matrices without at obtained using a short chromatographic run time was
least partial revalidation. probably due to the co-elution of a ciclosporin
metabolite in the internal standard channel.
Keevil et al. [32] and Yonan et al. [51] have investigated
the use of fingerprick (capillary blood spot) samples Co-eluting sample matrix components can affect the
using sample volumes as low as 10 L. Mendonza et al. efficiency of ionization of the analyte of interest, and
[33] have investigated the use of saliva as the sample this can result in ion suppression or ion enhancement of
matrix as a measure of unbound drug in the circulation. the signal [55,56,58]. These interferences can arise from
The separation of bound and unbound fractions of the sample of interest or a previous sample injection, or
ciclosporin in plasma is complex. However, saliva from buildup of interfering compounds on the analytical
analysis avoids the need for this step; it is only the column. Ion suppression can also occur as a result of
unbound portion that diffuses into the salivary glands. additives in the mobile phase. Electrospray ionization
These alternative matrices are less invasive to collect (ESI) is more prone to ion suppression than atmospheric
than venous blood and lend themselves to allowing pressure chemical ionization (APCI).
patients to collect their own samples outside the hospital
environment. The enhanced sensitivity associated with Although the majority of matrix effects are observed in
mass-spectrometric techniques also allows the the void volume of the chromatographic run, explaining
measurement of the drug in individual subsets of cells why problems can arise using the dilute-and-shoot
[52-54] to study the pharmacokinetic/pharmacodynamic approach, they can also occur later in the
relationships of the drug at its target level. chromatographic run [56]. Matrix effects can be
minimized or eliminated by improving extraction
Ciclosporin is relatively stable in whole blood, allowing techniques and/or chromatographic conditions [55].
the shipment of samples at ambient temperatures. If Taylor et al. [58] and McKeown et al. [50] have
samples are to be analyzed within a week of receipt, they demonstrated how ion suppression can be minimized if
should be kept refrigerated at approximately 4C. If they drug-free whole blood is extracted using solid-phase
are to be stored for longer periods before extraction (SPE) rather than by a simple protein
analysis/reanalysis, they should be deep frozen at precipitation (PP) method. Matuszewski et al. [55]
approximately 20C. Deep frozen, the drug is stable for provide a detailed account on how to assess for the
at least 1 year and through 3 freeze/thaw cycles. presence of matrix effects.
Interferences The use of internal standards that are analogs of the

A challenging problem associated with ciclosporin analyte of interest labeled with stable isotopes can aid in
measurement is interference from metabolites of the eliminating the problems of matrix effects, since both
drug, either the variable cross-reactivity with antibodies compounds should be affected to the same extent [56].
used in immunoassays or interference in Until recently, a deuterated analogue of ciclosporin was
chromatographic assays. not readily available. Salm et al. [36] described the
HPLC-MS/MS analysis of ciclosporin using ciclosporin-
Some users tend to consider HPLC-MS(/MS) to have D12 as the internal standard. As alternatives, authors
absolute specificity for the analyte, which has given rise have used structural analogs of ciclosporin (e.g., D, B, or
to a dilute (protein precipitation) and shoot (limited G) and nonstructurally related compounds (e.g.,
chromatography) approach to the analysis. However, ascomycin) as internal standards. Recently, Taylor et al.
HPLC-MS(/MS) is prone to interferences from co- [59] compared the use of the deuterated and D analogues
eluting compounds, the sample matrix, and detector and ascomycin as internal standards for the analysis of
cross-talk [55,56]. (With tandem mass spectrometers, ciclosporin. The results were as expected; the deuterated
cross-talk occurs as a result of ions from the previous analog gave superior resultsas judged by the
scan events not being cleared from the collision cell comparison of linearity, between-day precision, and the
before the next scan event occurs. This is more of a results from 104 whole-blood samples from solid organ
problem in multi-analyte methods in which ions of transplant recipients receiving ciclosporin therapy
similar masses are formed. Many modern mass followed by the D analogue and then ascomycin. Taylor
spectrometers are designed to reduce cross-talk, but it et al. also hypothesized that the negative bias reported
may present problems in the use of the emerging ultra- for the patient sample results, when using the D analog
performance liquid chromatography [UPLC] systems, as the internal standard, may have been due to
which enable the analysis of many compounds in a
455
interfering ciclosporin metabolite ions in the ciclosporin the order 500 to 1800 g/L, outside the calibration range
D mass channel. for most immunoassays, although two kit manufacturers
have produced an extended calibration range for their
In an effort to promote consistency in the performance of assays. It is also worth noting that the relative
HPLC-MS(/MS) assays for the measurement of differences between several of the immunoassays are
ciclosporin, the manufacturer of the drug, Novartis smaller when C2 sample results are compared with C0
Pharma, has donated stocks of the deuterated analogue to results, probably because the proportion of metabolites
the International Association of Therapeutic Drug in the 2-hour sample is smaller [72].
Monitoring and Clinical Toxicology for distribution
amongst its members. Details on how to obtain the There are few reasons to measure ciclosporin in patients
standard can be found on the Associations website, receiving the drug for autoimmune indications. Doses
www.iatdmct.org. are generally lower than those used for transplant
indications, and there is often an easily observable
Although HPLC-MS(/MS) methods are very sensitive pharmacodynamic marker of efficacy [73-75]. In some
and have the potential to be highly selective, they still instances, it is necessary to ensure that the drug has been
need to be well validated and any pitfalls addressed absorbed and to rule out noncompliance if there is
before they can be used reliably for patient management. therapeutic failure.
A useful scheme to follow for assay validation is that
described in the FDA bioanalytical methods validation For compliance testing, it is important to have
document established the lowest concentration that can be
(http://www.fda.gov/cder/guidance/4252fnl.pdf, measured reliably by an assay. The lower limit of
accessed 2008-02-15). Viswanathan et al. [60], Peters et quantification (LLOQ) is the concentration of the lowest
al. [61], and Shah et al. [62,63] also provide good calibrator, and concentrations measured below that
guidance on the validation of bioanalytical methods. should be reported as less than the LLOQ value. Ideally,
a concentration of about 25 g/L should be measurable,
Ciclosporin Therapeutic Range but this is not always possible with immunoassays or
Therapeutic drug monitoring is recommended for HPLC-UV assays. It is prudent to assay a ciclosporin-
ciclosporin because the drug has erratic free blood sample periodically as a check on assay
pharmacokinetics, it is difficult to distinguish its sensitivity.
therapeutic efficacy from symptoms and signs of
toxicity, there is the potential for pharmacokinetic drug Most centers report results in g/L or ng/mL, since the
interactions with other prescription or nonprescription literature uses these units almost exclusively. For those
drugs, and long-term, there are problems related to who wish to report in mol/L, then 1 g/L is equivalent
patient compliance with the prescribed dose. to 0.8315 mol/L.
Most centers monitor the drug using a pre-dose, or The inherent differences between the results obtained by
trough (C0), blood sample. For this approach to the immunoassays (related to their cross-reactivity with
monitoring, ciclosporin concentrations are of the order ciclosporin metabolites) and the generally lower values
100 to 400 g/L, depending on the indication for which obtained by chromatographic assays impacts on the
the drug is being used. However, it has been known for target concentration ranges for this drug. These
some time that ciclosporin efficacy is related to exposure methodological differences have given rise to method-
during a dosing interval, based on the area under the related target ranges. Such differences must be kept in
time-concentration curve (AUC), and that the trough mind if a center changes methodology or if clinicians
sample is not a good reflection of exposure [64]. For move between centers at which different methods are in
logistical reasons, measurement of AUC is difficult, use.
even if based on a limited sampling approach rather than
on several samples collected across a dose interval [65]. Interpretation
For this reason, the measurement of AUC has seen The object of TDM is to ensure that there is sufficient
negligible uptake. Following the introduction of the exposure to the drug to achieve therapeutic efficacy
Neoral oral formulation, designed to reduce some of the while avoiding excessive exposure that might lead to
pharmacokinetic variability associated with absorption acute and chronic nephrotoxicity or other unwanted
[66], there was considerable interest in monitoring, effects. There are many publications in which
based on a sample collected 2 hours post-dose (C2)[67]. therapeutic concentration ranges have been suggested,
The reason for using this time point is that it is a good but by and large, the approach to establishing them has
approximation for AUC0-4, when most of the variability been empirical rather than systematic [5,40]. There are
in the 12-hour AUC occurs [68]. Although the uptake of no randomized trials in which TDM is compared with
this monitoring strategy has not been widespread, there empirical dosing. The concept of a generally applicable
is a substantial body of literature, including a meta- therapeutic range for this drug cannot be justified. There
analysis of results and consensus guidelines [69,70]. are several factors that affect the interpretation of the
measured ciclosporin concentration, including the
Measurement of ciclosporin in C2 samples has transplant type, the time since transplantation, other drug
methodological implications [71]. Concentrations are of
456
therapy, the time of blood sampling relative to the last samples from transplant patients because of antibody
dose, and the assay method employed [76,77]. cross-reactivity with ciclosporin metabolites. Added to
this is the variability in calibration accuracy [45].
Applying data on therapeutic concentration ranges that Bioanalysts should be aware of these assay differences
have been generated over the almost 30-year history of and strive to minimize them, but it is fair to say that
ciclosporin TDM is complicated by a number of changes clinicians are often more concerned about the
that have taken place over that period: concomitant drug consistency of the result rather than the absolute number.
therapy, changes in the selection of patients and This is because they may be looking at trends in
transplanted organs, and advances in surgical techniques. concentration within fairly broad limits and are anxious
To avoid long-term toxicity, principally chronic allograft to know that an assay is performing in the same manner
nephropathy, there is a trend towards the use of lower over a period of time. Between-method differences may
doses of the calcineurin inhibitors, both ciclosporin and assume more importance if the assay method is changed,
tacrolimus, by combining them with mycophenolic acid if a patient visits more than one center at which different
and/or an mTOR inhibitorsirolimus or everolimus. assays are performed, or if different assays are used in
The introduction of induction therapy using monoclonal multicenter clinical studies.
anti-IL2 antibodies has also had an impact on ciclosporin
dosing and its target concentrations [78,79]. As a result, Consistency of assay performance and comparability
lower target concentrations of ciclosporin have been with peers using the same method is best addressed by
proposed, based either on empirical observation or by participation in an external proficiency testing scheme
analysis of data from concentration-controlled clinical [81]. Data from these schemes can also be used to assess
studies. assay performance, especially if pooled samples from
patients receiving ciclosporin are circulated. Choosing
Given the caveats that can be placed on the interpretation which assay to use in a clinical setting will be heavily
of any given measured concentration, it is not surprising influenced by the apparatus available, the technical
that centers often adopt an in-house set of ranges that are expertise available, and the sample throughput. When
based on their own experience, patient demographics, reviewing published data on assay kits, remember that
combination of drug therapy, and assay methodology. As changes in kit configuration may have occurred since
was noted in a previous review [77], the last set of publication of the data. When comparing absolute
guidelines, based on data from 21 centers monitoring the differences between assays, it is useful to check whether
drug in kidney transplant patients [40], there was a common control material has been analyzed by each
sixfold range in the concentrations considered effective assay. In addition, the analytical platform on which an
and a threefold range in the concentrations considered immunoassay was tested may differ from that in current
toxic. In general, higher concentrations are aimed for use, and the names of many diagnostics companies have
early after transplantation, when the immunological risk changed owing to mergers and acquisitions. In some
is highest. After a period of 3 to 6 months, there is locations, diagnostic kits may only be available through
usually a gradual tapering of the dose, together with a local agent and may be rebadged under another name.
more reliance on adjunctive therapy. It is also common
for method-specific concentration ranges to be used, The Tables below show the immunoassays and
with allowance being made for the higher results chromatographic assays currently in use to measure
produced by methods with least specificity for the parent ciclosporin. Typical data for the measurement of
compound. ciclosporin in a blood sample spiked with the drug and
in a pooled sample from patients receiving the drug are
For those centers new to this field, it is always worth shown in Figure 1, using data available online from the
consulting recent literature on the interpretation of International Proficiency Testing Scheme
ciclosporin concentrations, but one must always take (www.bioanalytics.co.uk). It is clear that the between-
into consideration any trends in prescription or clinical method differences are influenced by both calibration
practice subsequent to these publications. Consulting and antibody specificity and that HPLC assays are
those already in the field, who are engaged in a similar subject to some wide variations in performance. The
transplant setting, is probably of more value. It must be package inserts for each kit give details of assay
remembered that the measured concentration of performance and cross-reactivity with ciclosporin
ciclosporin is only one part of the dosing algorithm, metabolites. However, bear in mind that in vitro tests of
which must also take into account clinical and other cross-reactivity do not set out to investigate the various
laboratory data. combinations of ciclosporin metabolites that might be in
any sample and cannot reproduce every clinical situation
Ciclosporin Performance Goals that could arise. All immunoassays have the potential to
Guidelines that have been developed for the monitoring produce higher-than-true values, especially in clinical
of therapeutic drugs indicate that inaccuracy and situations in which ciclosporin metabolites may be
imprecision of the analytical method used should not relatively high, such as following liver and heart
exceed 10% [80]. For the assays currently available transplantation, during hepatic dysfunction, and when
(see below), this is not likely to be the case. For the kinetics of ciclosporin have been altered by
immunoassays, it is not possible to achieve so close an pharmacokinetic drug interactions.
agreement with a chromatographic method for many
457
There are clear differences between the various methods Abbott TDx monoclonal and polyclonal
related to calibration and selectivity which must be taken immunoassays. Ann Clin Biochem
into consideration when a result is interpreted. In 1995;32:208-9.
general, the most selective assays are to be preferred, but 12 Steimer W. Performance and specificity of
the final choice is likely to be governed by operational monoclonal immunoassays for cyclosporine
constraints such as access to appropriate apparatus. All monitoring: how specific is specific? Clin
the methods can be used to investigate the Chem 1999;45:371-81.
pharmacokinetics of ciclosporin following chronic 13 Soldin SJ, Steele BW, Witte DL, Wang E, Elin
doses, but preference should be given to the most RJ. Lack of specificity of cyclosporine
selective assays. For single-dose studies and for the immunoassays. Arch Pathol Lab Med
investigation of pharmacokinetic drug interactions, a 2003;127:19-22.
chromatographic assay with mass-spectrometric 14 Khoschsorur G, Semmelrock HJ, Rodl S, Auer
detection is the method of choice. T, Petek W, Iberer F et al. Rapid, sensitive,
high-performance liquid chromatographic
References method for the determination of cyclosporin A
1 Wenger RM. Structures of cyclosporine and its and its metabolites M1, M17 and M21. J
metabolites. Transplant Proc 1990;22:1104-8. Chromatogr B Biomed Sci Appl 1997;690:367-
2 Christians U, Sewing KF. Cyclosporin 72.
metabolism in transplant patients. Pharmacol 15 Brozmanova H, Grundmann M, Safarcik K,
Ther 1993;57:291-345. Jegorov A. High-performance liquid
3 van den Berg JW, Verhoef ML, de Boer AJ, chromatographic method for therapeutic drug
Schalm SW. Cyclosporin A assay: conditions monitoring of cyclosporine A and its two
for sampling and processing of blood. Clin metabolites in renal transplant patients. J
Chim Acta 1985;147:291-7. Chromatogr B Biomed Sci Appl 2000;749:93-
4 Hows JM, Smith JM. In vitro stability of 100.
cyclosporin A. J Clin Pathol 1983;36:720-1. 16 Chimalakonda AP, Shah RB, Mehvar R. High-
5 Shaw LM, Bowers LD, Demer L, Freeman DJ, performance liquid chromatographic analysis of
Moyer T, Sanghvi A. Critical issues in cyclosporin A in rat blood and liver using a
cyclosporine monitoring: report of the Task commercially available internal standard. J
Force on Cyclosporine Monitoring. Clin Chem Chromatogr B Analyt Technol Biomed Life Sci
1987;33:1269-88. 2002;772:107-14.
6 Jones SR, Gunson B, McLean S, McMaster P, 17 Ouyang J, Baeyens WR, Duan J, Delanghe J.
Ratcliffe JG. Cyclosporine levels in whole Improvement of cyclosporin A determination in
blood. Comparison of the polyclonal Sandoz whole blood by reversed-phase high-
Ciclosporin Radioimmunoassay Kit with the performance liquid chromatography. Biomed
specific antibody in the new Sandimmun Kit. Chromatogr 2003;17:404-10.
Transplantation 1988;45:991-2. 18 Baldelli S, Zenoni S, Merlini S, Perico N,
7 Holt DW, Johnston A, Marsden JT, Vernillet L, Cattaneo D. Simultaneous determination of
Keown PA, Rosano TG et al. Monoclonal everolimus and cyclosporine concentrations by
antibodies for radioimmunoassay of HPLC with ultraviolet detection. Clin Chim
cyclosporine: a multicenter comparison of their Acta 2006;364:354-8.
performance with the Sandoz polyclonal 19 Christians U, Jacobsen W, Serkova N, Benet
radioimmunoassay kit. Clin Chem LZ, Vidal C, Sewing KF et al. Automated, fast
1988;34:1091-6. and sensitive quantification of drugs in blood
8 Schutz E, Svinarov D, Shipkova M, Niedmann by liquid chromatography-mass spectrometry
PD, Armstrong VW, Wieland E et al. with on-line extraction: immunosuppressants. J
Cyclosporin whole blood immunoassays Chromatogr B Biomed Sci Appl 2000;748:41-
(AxSYM, CEDIA, and Emit): a critical 53.
overview of performance characteristics and 20 McMahon LM, Luo S, Hayes M, Tse FL. High-
comparison with HPLC. Clin Chem throughput analysis of everolimus (RAD001)
1998;44:2158-64. and cyclosporin A (CsA) in whole blood by
9 Cattaneo D, Zenoni S, Murgia S, Merlini S, liquid chromatography/mass spectrometry using
Baldelli S, Perico N et al. Comparison of a semi-automated 96-well solid-phase
different cyclosporine immunoassays to extraction system. Rapid Commun Mass
monitor C0 and C2 blood levels from kidney Spectrom 2000;14:1965-71.
transplant recipients: not simply overestimation. 21 Zhou L, Tan D, Theng J, Lim L, Liu YP, Lam
Clin Chim Acta 2005;355:153-64. KW. Optimized analytical method for
10 Morris RG, Lam AK. Cyclosporin monitoring cyclosporin A by high-performance liquid
in Australasia: survey of laboratory practices in chromatography-electrospray ionization mass
2000. Ther Drug Monit 2002;24:471-8. spectrometry. J Chromatogr B Biomed Sci Appl
11 Roberts NB, Dutton J, Tewari A, Bone MJ, 2001;754:201-7.
Sells RA, Ward RG et al. A method of relating
whole blood cyclosporin A results by HPLC to
458
22 Deters M, Kirchner G, Resch K, Kaever V. chromatography tandem mass spectrometry and

Simultaneous quantification of sirolimus, its application in C2 monitoring. Ther Drug
everolimus, tacrolimus and cyclosporine by Monit 2002;24:757-67.
liquid chromatography-mass spectrometry (LC- 33 Mendonza A, Gohh R, Akhlaghi F.
MS). Clin Chem Lab Med 2002;40:285-92. Determination of cyclosporine in saliva using
23 Poquette MA, Lensmeyer GL, Doran TC. liquid chromatography-tandem mass spectro-
Effective use of liquid chromatography-mass metry. Ther Drug Monit 2004;26:569-75.
spectrometry (LC/MS) in the routine clinical 34 Ceglarek U, Lembcke J, Fiedler GM, Werner
laboratory for monitoring sirolimus, tacrolimus, M, Witzigmann H, Hauss JP et al. Rapid
and cyclosporine. Ther Drug Monit simultaneous quantification of
2005;27:144-50. immunosuppressants in transplant patients by
24 Zaater MF, Tahboub YR, Najib NM. Liquid turbulent flow chromatography combined with
chromatographic-electrospray mass tandem mass spectrometry. Clin Chim Acta
spectrometric determination of cyclosporin A in 2004;346:181-90.
human plasma. Anal Bioanal Chem 35 Koal T, Deters M, Casetta B, Kaever V.
2005;382:223-30. Simultaneous determination of four
25 Vollenbroeker B, Koch JH, Fobker M, immunosuppressants by means of high speed
Suwelack B, Hohage H, Muller U. and robust on-line solid phase extraction-high
Determination of cyclosporine and its performance liquid chromatography-tandem
metabolites in blood via HPLC-MS and mass spectrometry. J Chromatogr B Analyt
correlation to clinically important parameters. Technol Biomed Life Sci 2004;805:215-22.
Transplant Proc 2005;37:1741-4. 36 Salm P, Taylor PJ, Lynch SV, Warnholtz CR,
26 Koseki N, Nakashima A, Nagae Y, Masuda N. Pillans PI. A rapid HPLC-mass spectrometry
Simultaneous quantitative determination of cyclosporin method suitable for current
cyclosporine A and its three main metabolites monitoring practices. Clin Biochem
(AM1, AM4N and AM9) in human blood by 2005;38:667-73.
liquid chromatography/mass spectrometry using 37 Ceglarek U, Casetta B, Lembcke J, Baumann S,
a rapid sample processing method. Rapid Fiedler GM, Thiery J. Inclusion of MPA and in
Commun Mass Spectrom 2006;20:733-40. a rapid multi-drug LC-tandem mass
27 Taylor PJ, Jones CE, Martin PT, Lynch SV, spectrometric method for simultaneous
Johnson AG, Pond SM. Microscale high- determination of immunosuppressants. Clin
performance liquid chromatography- Chim Acta 2006;373:168-71.
electrospray tandem mass spectrometry assay 38 Bogusz MJ, Enazi EA, Hassan H, bdel-Jawaad
for cyclosporin A in blood. J Chromatogr B J, Ruwaily JA, Tufail MA. Simultaneous LC-
Biomed Sci Appl 1998;705:289-94. MS-MS determination of cyclosporine A,
28 Simpson J, Zhang Q, Ozaeta P, Aboleneen H. A tacrolimus, and sirolimus in whole blood as
specific method for the measurement of well as mycophenolic acid in plasma using
cyclosporin A in human whole blood by liquid common pretreatment procedure. J Chromatogr
chromatography-tandem mass spectrometry. B Analyt Technol Biomed Life Sci
Ther Drug Monit 1998;20:294-300. 2007;850:471-80.
29 Brignol N, McMahon LM, Luo S, Tse FL. 39 Kahan BD, Shaw LM, Holt D, Grevel J,
High-throughput semi-automated 96-well Johnston A. Consensus document: Hawk's Cay
liquid/liquid extraction and liquid meeting on therapeutic drug monitoring of
chromatography/mass spectrometric analysis of cyclosporine. Clin Chem 1990;36:1510-6.
everolimus (RAD 001) and cyclosporin a (CsA) 40 Oellerich M, Armstrong VW, Kahan B, Shaw
in whole blood. Rapid Commun Mass Spectrom L, Holt DW, Yatscoff R et al. Lake Louise
2001;15:898-907. Consensus Conference on cyclosporin
30 Streit F, Armstrong VW, Oellerich M. Rapid monitoring in organ transplantation: report of
liquid chromatography-tandem mass the consensus panel. Ther Drug Monit
spectrometry routine method for simultaneous 1995;17:642-54.
determination of sirolimus, everolimus, 41 Oellerich M, Armstrong VW, Schutz E, Shaw
tacrolimus, and cyclosporin A in whole blood. LM. Therapeutic drug monitoring of
Clin Chem 2002;48:955-8. cyclosporine and tacrolimus. Update on Lake
31 Keevil BG, Tierney DP, Cooper DP, Morris Louise Consensus Conference on cyclosporin
MR. Rapid liquid chromatography-tandem and tacrolimus. Clin Biochem 1998;31:309-16.
mass spectrometry method for routine analysis 42 Holt DW, Armstrong VW, Griesmacher A,
of cyclosporin A over an extended Morris RG, Napoli KL, Shaw LM. International
concentration range. Clin Chem 2002;48:69-76. Federation of Clinical Chemistry/International
32 Keevil BG, Tierney DP, Cooper DP, Morris Association of Therapeutic Drug Monitoring
MR, Machaal A, Yonan N. Simultaneous and and Clinical Toxicology working group on
rapid analysis of cyclosporin A and creatinine immunosuppressive drug monitoring. Ther
in finger prick blood samples using liquid Drug Monit 2002;24:59-67.
459
43 Morris RG, Holt DW, Armstrong VW, switching approach. J Chromatogr B Analyt
Griesmacher A, Napoli KL, Shaw LM. Analytic Technol Biomed Life Sci 2007;857:92-9.
aspects of cyclosporine monitoring, on behalf 55 Matuszewski BK, Constanzer ML, Chavez-Eng
of the IFCC/IATDMCT Joint Working Group. CM. Strategies for the assessment of matrix
Ther Drug Monit 2004;26:227-30. effect in quantitative bioanalytical methods
44 Holt DW, Johnston A. Monitoring new based on HPLC-MS/MS. Anal Chem
immunosuppressive agents. Are the methods 2003;75:3019-30.
adequate? Drug Metabol Drug Interact 56 Annesley TM. Ion suppression in mass
1997;14:5-15. spectrometry. Clin Chem 2003;49:1041-4.
45 Holt DW, Denny K, Lee T, Johnston A. 57 Vogeser M, Spohrer U. Pitfall in the high-
Cyclosporin MeasurementWho cares about throughput quantification of whole blood
inaccuracy? Accred Qual Assur 2001;6:405-8. cyclosporin A using liquid chromatography-
46 Annesley TM. Application of commercial tandem mass spectrometry. Clin Chem Lab
calibrators for the analysis of Med 2005;43:400-2.
immunosuppressant drugs in whole blood. Clin 58 Taylor PJ. Matrix effects: the Achilles heel of
Chem 2005;51:457-60. quantitative high-performance liquid
47 Cooper DP, Napoli KL, Taylor PJ, Hammett- chromatography-electrospray-tandem mass
Stabler C, Graham KS, Hammon GD et al. spectrometry. Clin Biochem 2005;38:328-34.
Multicentre validation of the MassTrak reagent 59 Taylor PJ, Brown SR, Cooper DP, Salm P,
kit for the quantification of tacrolimus in whole Morris MR, Pillans PI et al. Evaluation of 3
blood using HPLC-MS/MS. Clin Chem internal standards for the measurement of
53[Suppl], A103. 2007. cyclosporin by HPLC-mass spectrometry. Clin
48 Deters M, Kaever V, Kirchner GI. Liquid Chem 2005;51:1890-3.
chromatography/mass spectrometry for 60 Viswanathan CT, Bansal S, Booth B,
therapeutic drug monitoring of DeStefano AJ, Rose MJ, Sailstad J et al.
immunosuppressants. Analytica Chimica Acta Quantitative bioanalytical methods validation
2003;492:133-45. and implementation: best practices for
49 Taylor PJ. Therapeutic drug monitoring of chromatographic and ligand binding assays.
immunosuppressant drugs by high-performance Pharm Res 2007;24:1962-73.
liquid chromatography-mass spectrometry. 61 Peters FT, Drummer OH, Musshoff F.
Ther Drug Monit 2004;26:215-9. Validation of new methods. Forensic Sci Int
50 McKeown DA, Lee TD, Holt DW. Therapeutic 2007;165:216-24.
drug monitoring: measurement of 62 Shah VP, Midha KK, Dighe S, McGilveray IJ,
immunosuppressants using HPLC with mass Skelly JP, Yacobi A et al. Analytical methods
spectrometric detection. In: Niessen MA, ed. validation: bioavailability, bioequivalence and
The Encyclopedia of Mass Spectrometry. Vol pharmacokinetic studies. Conference report.
8. Hyphenated Methods. Amsterdam: Elsevier; Eur J Drug Metab Pharmacokinet 1991;16:249-
2006:813-23. 55.
51 Yonan N, Martyszczuk R, Machaal A, Baynes 63 Shah VP, Midha KK, Findlay JW, Hill HM,
A, Keevil BG. Monitoring of cyclosporine Hulse JD, McGilveray IJ et al. Bioanalytical
levels in transplant recipients using self- method validation--a revisit with a decade of
administered fingerprick sampling. Clin progress. Pharm Res 2000;17:1551-7.
Transplant 2006;20:221-5. 64 Mahalati K, Kahan BD. Pharmacological
52 Falck P, Guldseth H, Asberg A, Midtvedt K, surrogates of allograft outcome. Ann Transplant
Reubsaet JL. Determination of ciclosporin A 2000;5:14-23.
and its six main metabolites in isolated T- 65 David OJ, Johnston A. Limited sampling
lymphocytes and whole blood using liquid strategies for estimating cyclosporin area under
chromatography-tandem mass spectrometry. J the concentration-time curve: review of current
Chromatogr B Analyt Technol Biomed Life Sci algorithms. Ther Drug Monit 2001;23:100-14.
2007;852:345-52. 66 Holt DW, Johnston A. Cyclosporin
53 Falck P, Asberg A, Guldseth H, Bremer S, microemulsion: a guide to usage and
Akhlaghi F, Reubsaet JL et al. Declining monitoring. BioDrugs 1997;7:175-97.
intracellular T-lymphocyte concentration of 67 Belitsky P, Dunn S, Johnston A, Levy G.
cyclosporine a precedes acute rejection in Impact of absorption profiling on efficacy and
kidney transplant recipients. Transplantation safety of cyclosporin therapy in transplant
2008;85:179-84. recipients. Clin Pharmacokinet 2000;39:117-25.
54 Ansermot N, Fathi M, Veuthey JL, Desmeules 68 Cantarovich M, Besner JG, Barkun JS, Elstein
J, Hochstrasser D, Rudaz S. Quantification of E, Loertscher R. Two-hour cyclosporine level
cyclosporine A in peripheral blood determination is the appropriate tool to monitor
mononuclear cells by liquid chromatography- Neoral therapy. Clin Transplant 1998;12:243-9.
electrospray mass spectrometry using a column-
460
69 Knight SR, Morris PJ. The clinical benefits of immunosuppressive drugs. Clin Ther
cyclosporine C2-level monitoring: a systematic 1999;21:1632-52.
review. Transplantation 2007;83:1525-35. 77 Johnston A, Holt DW. Immunosuppressant
70 Levy G, Thervet E, Lake J, Uchida K. Patient drugs - the role of therapeutic drug monitoring.
management by Neoral C(2) monitoring: an Br J Clin Pharmacol 2001;52 Suppl 1:61-73
international consensus statement. 78 Loichot C, tue-Ferrer D, Bernard N, Bonardet
Transplantation 2002;73(Suppl):S12-S18. A, Boulieu R, Kergueris MF et al. Cyclosporine
71 Holt DW, Johnston A, Kahan BD, Morris RG, monitoring in renal transplant recipients with
Oellerich M, Shaw LM. New approaches to induction therapy: C2 levels in patients
cyclosporine monitoring raise further concerns monitored on C0. Fundam Clin Pharmacol
about analytical techniques. Clin Chem 2006;20:91-6.
2000;46:872-4. 79 Schiff J, Cole E, Cantarovich M. Therapeutic
72 Johnston A, Chusney G, Schutz E, Oellerich M, monitoring of calcineurin inhibitors for the
Lee TD, Holt DW. Monitoring cyclosporin in nephrologist. Clin J Am Soc Nephrol
blood: between-assay differences at trough and 2007;2:374-84.
2 hours post-dose (C2). Ther Drug Monit 80 Shaw LM, Annesley TM, Kaplan B, Brayman
2003;25:167-73. KL. Analytic requirements for
73 Holt DW, Johnston A. Cyclosporin monitoring: immunosuppressive drugs in clinical trials.
its role in autoimmune indications. J Ther Drug Monit 1995;17:577-83.
Autoimmun 1992;5 Suppl A:177-82. 81 Holt DW, Johnston A. Monitoring
74 Griffiths CE, Dubertret L, Ellis CN, Finlay AY, immunosuppressive drugs: the rationale for
Finzi AF, Ho VC et al. Ciclosporin in psoriasis proficiency testing. Ligand Assay 1998;3:2-6.
clinical practice: an international consensus 82 Bowers LD, Mathews SE. Investigation of the
statement. Br J Dermatol 2004;150 Suppl mechanism of peak broadening observed in the
67:11-23.:11-23. high-performance liquid chromatographic
75 Griffiths CE, Katsambas A, Dijkmans BA, analysis of cyclosporine. J Chromatogr
Finlay AY, Ho VC, Johnston A et al. Update on 1985;333:231-8.
the use of ciclosporin in immune-mediated 83 Annesley TM, Clayton L. Simple extraction
dermatoses. Br J Dermatol 2006;155 Suppl 2:1- protocol for analysis of immunosuppressant
16.:1-16. drugs in whole blood. Clin Chem
76 Shaw LM, Holt DW, Keown P, 2004;50:1845-8.
Venkataramanan R, Yatscoff RW. Current
opinions on therapeutic drug monitoring of
461
Table 1: Methods for Ciclosporin (CSP) Measurement

IMMUNOASSAYS
1. Radioimmunoassay
CYCLO-Trac SP-Whole Blood - DiaSorin Inc.
Competitive binding of 125I-CSP and nonlabeled CSP to a limited amount of monoclonal specific antibody, using
a methanolic extract of a whole blood sample. Double antibody phase separation technique. Sample size: 100 L
of whole blood collected into EDTA anticoagulant. Calibration range 0 to 1200 g/L.
Comments:
Problems related to the use and disposal of radioactive materials; some cross-reactivity with CSP metabolites.
2. Fluorescence Polarization Immunoassay (FPIA)
TDx Cyclosporine Monoclonal Whole Blood Abbott Laboratories
A competitive immunoassay performed on the TDx platform, involving competitive binding of fluorescein-
labeled CSP and nonlabeled CSP to a limited amount of monoclonal specific antibody, using a whole blood
sample pretreated with a solubilization reagent and precipitation reagent. The polarization of fluorescent light is
inversely proportional to the concentration of CSP in the sample. Sample size: 150 L of whole blood
(anticoagulant not specified). Calibration range 0 to 1500 g/L
Comments:
Relatively high cross-reactivity with CSP metabolites.
3. Fluorescence Polarization Immunoassay (FPIA)
Cyclosporine AXSYM Abbott Laboratories
A competitive immunoassay performed on the AXSYM platform, involving competitive binding of fluorescein-
labeled CSP and nonlabeled CSP to a limited amount of monoclonal specific antibody, using a whole blood
sample pre-treated with a solubilization reagent and precipitation reagent. The polarization of fluorescent light is
inversely proportional to the concentration of CSP in the sample. Sample size: 150 L of whole blood collected
into EDTA or heparin anticoagulant. Calibration range 0 to 800 g/L
Comments:
Some cross-reactivity with CSP metabolites.
4. Chemiluminescent Microparticle Immunoassay (CMIA)
ARCHITECT Cyclosporine Abbott Laboratories
(Available in Europe from March 2008)
A two-step immunoassay performed on the Architect platform. Samples are solubilized and extracted with a
precipitation reagent. The sample and monoclonal antibodycoated paramagnetic microparticles are combined.
CSP binds to the antibody on the beads, which are washed, and a CSP acridinium conjugate is added to occupy
vacant CSP antibody sites. A trigger solution then initiates chemiluminescence. Sample size: 200 L of whole
blood collected into EDTA anticoagulant. Calibration range 0 to 1500 g/L
Comments:
Initial reports suggest relatively little cross-reactivity with CSP metabolites.
5. Enzyme Multiplied Immunoassay Technique (EMIT)
Emit 2000 Cyclosporine Specific Assay Dade Behring
A homogeneous competitive immunoassay in which the drug competes with enzyme-labeled CSP for antibody
sites. Unbound enzyme reacts with nicotinamide adenine dinucleotide (NAD), converting it to NADH, which can
be measured spectrophotometrically. The sample requires pretreatment to solubilize the drug and precipitate
protein components before analysis. Sample size: 100 L of whole blood collected into EDTA anticoagulant. Two
calibration ranges available 0 to 500 g/L and 0 to 2000 g/L.
Comments:
Some cross-reactivity with metabolites. Several analytical platforms can be used.
6. Affinity Chrome-Mediated Immunoassay (ACMIA)
Cyclosporine Felx Reagent Cartridge CSA and CSAE
Dimension Clinical Chemistry System Dade Behring
An automated immunoassay technique with no pretreatment step, performed with a method specific reagent
cartridge on the Dimension platform. The sample is mixed and lysed on the analytical platform then mixed with a
-galactosidase-CSP conjugate. Magnetic particles coated with CSP then bind unbound antibody-enzyme
conjugate, which are separated magnetically. -galactosidase is then used to catalyze the hydrolysis of a
chromophore, and the change in absorption measured. Samples must remain on the analytical platform for 30
minutes prior to analysis. Sample size: 200 L of whole blood collected into EDTA anticoagulant. Calibration
range 25 to 500 g/L (CSA), 350 to 2000 g/L (CSAE).
462
Comments:
Some cross-reactivity with metabolites. Performed on a random-access analyzer.
7. Cloned Enzyme Donor Immunoassay (CEDIA)
CEDIA Cyclosporine Plus Microgenics
A homogeneous enzyme immunoassay which uses genetically modified -galactosidase as two inactive
fragments. Samples are lysed off the analytical platform, but no other pretreatment is required. CSP in the sample
competes for antibody with CSP conjugated to one enzyme fragment. Higher concentrations of CSP allow the
enzyme fragments to reassociate and form active enzyme, which cleaves a colored substrate. Sample size: 100 L
of whole blood collected into EDTA anticoagulant. Calibration range 25 to 450 g/L (CSA), 450 to 2000 g/L
(CSAE).
Comments:
Some cross-reactivity with metabolites. Several analytical platforms can be used.
8. Direct Chemiluminescence
Cyclosporine (CsA) Siemens Healthcare Diagnostics
A competitive immunoassay using direct chemiluminescent technology, performed on the ADVIA Centaur and
ADVIA Centaur XP systems. Samples must be manually pretreated to lyse the cells. The labeled antibody binds
to streptavidin coupled to paramagnetic particles on a solid phase. Sample size: 100 L of whole blood collected
into EDTA anticoagulant. Calibration range 0 to 2000 g/L.
Comments:
Preliminary data suggest relatively low cross-reactivity with metabolites. Performed on a random access analyzer.
CHROMATOGRAPHIC ASSAYS
High-Performance Liquid Chromatography Techniques
a. Ultraviolet (HPLC-UV) detection at low wavelengths, 205 to 227nm.
b. Mass spectrometric (HPLC-MS) detection, typically by monitoring the stable sodium adduct [M+Na]+.
c. Tandem mass-spectrometric (HPLC-MS/MS) detection, typically by multiple reaction monitoring of the less
stable ammonium adduct [M+NH4]+ precursor ion to the [M+H]+ product ion.
Principle of analysis: Extraction of CSP from whole blood, using protein precipitation followed by a variety of
off-line and semi-automated on-line clean up techniques; chromatography typically on reversed-phase HPLC
systems; then detection by one of the above methods (a-c)
Comments: Well validated HPLC based techniques provide a selective measurement of CSP and its metabolites.
463
Figures
Figure 1A
Figure 1B
Figure 1: Blood ciclosporin concentrations in A, a ciclosporin-free whole blood sample spiked to a nominal concentration
of 150 g/L and B, in a pooled whole blood sample from transplant patients receiving ciclosporin following bone marrow
transplantation. The data are presented as box and whisker plots, by analytical technique.
ACMIA, Affinity Chrome-Mediated Immunoassay; CEDIA, cloned-enzyme donor immunoassay; EMIT, enzyme-
multiplied immunoassay technique; FPIA TDx, fluorescence polarization immunoassay performed on the TDx platform;
FPIA AxSYM, fluorescence polarization immunoassay performed on the AxSYM platform; RIA CYCLO-Trac, radio-
immunoassay using CYCLO-Trac SP kit; HPLC/MS, high-performance liquid chromatography with MS or MS/MS
detection; HPLC/UV, high-performance liquid chromatography with UV detection. (Data from International Ciclosporin
Proficiency Testing Scheme [unpublished].)
464
Figure 2: Typical HPLC-UV chromatograms of CSP. A, 250 g/L CSP standard (4.3 min) and 750 g/L CSP-D (5.6 min).
B & C, Chromatograms of whole blood extracts from transplant patients receiving the drug; whole blood CSP
concentrations were 79 g/L (B) and 298 g/L (C).
465
Good chromatographic conditions

e.g. extended chromatography or rapid
No Further Extraction chromatography utilising ballistic gradients and
divertor valves
Manual LLE
e.g. in house, utilising 0.1M sodium hydroxide and
Liquid-Liquid methyl-tert-butyl ether or chlorobutane
Extraction
(LLE) Semi-automated 96-well plate LLE
e.g. utilising the Tomtec Quadra 96 workstation
Manual SPE
Protein
Precipitation
Semi-automated off-line SPE
e.g. utlising the Gilson ASPEC XL4
Solid Phase Extraction Semi-auotmated 96-well plate SPE

(SPE) e.g. utilising the Tomtec Quadra 96 workstation
On-line SPE (or LC/LC) utilising an additional

extraction column, additional HPLC pumps, and a
6-port column switching valve, prior to the analytical
column
Turbo Flow On-line TFC utilising a specially designed TFC

extraction column, additional HPLC pumps, and two
Chromatography 6-port column switching valve, prior to the analytical
(TFC) column
Figure 3: Schematic diagram of several sample preparation methods that have been used to extract ciclosporin from whole
blood.
466
Figure 4: HPLC-MS/MS chromatograms of ciclosporin extracted from a whole-blood sample spiked at a concentration of
25 g/L (left-hand panels). The MRM transitions monitored were: A, m/z: 1203/100 for the [M+H]+ precursor
ion; B, m/z: 1201/1089 for the [M-H]- precursor ion, and C, m/z: 1220/1203 for the [M+NH4]+ precursor ion. Note
the increase in sensitivity when monitoring the [M+NH4]+ precursor ion compared with the [M+H]+ in positive
ionization mode. The comparable transitions for the ciclosporin-D12 internal standard are shown in the right-hand
panels.
467
Figure 5: Typical calibration curve used for the analysis of pre-dose (trough, C0) samples from patients receiving
ciclosporin therapy. Whole-blood calibrators over the concentration range 25 to 500 g/L were extracted, and the
concentration was plotted against the ratio of peak area for the drug and the internal standard (ciclosporin-D12).
Procedure: High-Performance Liquid Internal Standards: Methods have used structural

Chromatography analogues such as ciclosporin D and various structurally
HPLC-UV unrelated compounds [15,18].
Sample Preparation UV Detector Wavelength: Low wavelengths ranging
Whole blood samples are typically protein precipitated from 205 nm to 227 nm [15,17].
using reagent mixtures such as aqueous 10% zinc Figure 2 shows an HPLC-UV chromatogram for
sulphate solution (ZnSO4):acetonitrile:methanol ciclosporin, using the D analogue as the internal
(50:20:30, v/v/v),[15] or 5% ZnSO4 and acetone [18]. standard.
The supernatant is then liquid-liquid extracted (e.g.,
diethyl ether then partitioned twice with hexane [15] or HPLC-MS(/MS)
undergoes SPE (e.g., Bond-Elut cartridge, C18, 200 mg, The following section summarizes HPLC-MS and
3 mL). Sample volumes up to 1 mL can be required to HPLC-MS/MS method conditions that can be used for
achieve lower limits of quantitation (LLOQ) of 20 g/L the analysis of ciclosporin and for the simultaneous
[18]. multi-analyte analysis of ciclosporin with other
Chromatography and Detector Conditions immunosuppressive drugssirolimus (SIR), everolimus
Mobile Phase: Typically, mixtures of acetonitrile- (EVR), tacrolimus (TAC) and mycophenolic acid
deionized water, acetonitrile-methanol-deionized water, (MPA).
or acetonitrile-methanol-phosphate buffer [15,16,18]. Sample Preparation
Analytical Column: Beckman Ultrasphere C8 (75 Whole blood can be protein precipitated using the
4.6mm, 3 m) or a Hypersil ODS C18 (100 4.6mm, 5 successive addition of reagents or the addition of
m) are examples of the types of reverse-phase columns premixed reagents (83). Methanol:0.4M ZnSO4 (4:1,
that have been utilized for ciclosporin analysis [17,18]. v/v) [19] and acetonitrile:0.1M ZnSO4 (70:30, v/v) [36]
Column Oven: Elevated temperatures of 60C [18] or are examples of the premixed reagents utilized.
70C [15] tend to be used to obtain acceptable peak Annesley et al. [83] reported cleaner extracts with the
shapes. Investigations have shown that the peak successive addition of deionized water (water
broadening observed during the chromatographic hemolysis), followed by 0.1 M ZnSO4, then methanol, in
analysis of ciclosporin is attributable to the presence of comparison to premixed reagents.
rapidly interconverting conformers of the ciclosporin Following protein precipitation, the extracts are either
molecule [82]. analyzed by HPLC-MS(/MS), or further off-line or on-
Run Times: Chromatographic analysis can take up to 18 line sample cleanup procedures can be utilized prior to
min for ciclosporin [16], and 36 min [18] and 40 min HPLC-MS(/MS) analysis (see Figure 3).
[15] for the multi-analyte analysis of ciclosporin and Sample volumes as low as 100 L have been used to
everolimus, or ciclosporin and two of its metabolites, achieve a LLOQ of 7.5 g/L for HPLC-MS [19]. Keevil
respectively. et al. [31] reported a HPLC-MS/MS method for
468
ciclosporin that was linear over a wide analytical range m/z: 1201/1089, ciclosporin D m/z: 1215/907, and
(1 to 5000 g/L), using only 100 L of sample. ciclosporin-D12 m/z: 1213/1101. Typical
Chromatography Conditions chromatograms for the analysis of an extracted
Mobile Phase: ciclosporin calibrator together with ciclosporin-D12 are
HPLC-MS: Typically methanol-deionized water [21] or shown in Figure 4.
acetonitrile-deionized water [25] systems, sometimes For reasons related to assay sensitivity (see Figure 4),
with the addition of formic acid [19] or acetic acid [24]. the less stable ammonium adduct precursor ion
Both isocratic and gradient elution systems are utilized. [M+NH4]+, which fragments back to the [M+H]+ ion, is
HPLC-MS/MS: Typically methanol-deionized water or typically monitored for ciclosporin (m/z: 1220/1203),
acetonitrile-deionized water and may be supplemented ciclosporin D (1234/1217), and ciclosporin-D12
with ammonium acetate [29,36] or ammonium formate (1232/1215) [48]. The mobile phase must be
[38] (important for tandem mass spectrometric analysis, supplemented with ammonium ions to form this adduct.
see below). Both isocratic and gradient elution systems Analyzers: Typically, single quadrupole and triple
are utilized. quadrupole analyzers are used for the measurement of
Analytical Column: Various reverse-phase HPLC ciclosporin in clinical samples. However, any suitable
columns have been used, including C8, C18, phenyl- analyzer can be used.
hexyl [19,34,36]. Ciclosporin Calibrator and Quality-Control Sample
Column Oven: Elevated temperatures of 65C [38] or Preparation
70C [36] are often still used. However, the specificity In order to determine the concentration of whole blood
of mass spectrometry allows rapid isocratic and gradient samples from patients receiving ciclosporin therapy, it is
elutions compared to HPLC-UV, which aids in necessary to construct a calibration curve. A sufficient
improving the broad ciclosporin peak shape. number of whole blood calibrators (Cals) (n 6),
Run Times: Simultaneous HPLC-MS and HPLC-MS/MS analyzed in duplicate, must be used to cover the
analysis of ciclosporin, SIR, EVR and TAC can be expected concentration range. For the TDM of
achieved in less than 10 min [19] and 3 min [35], ciclosporin C0 and C2 samples, a range of 25 to 2000
respectively. Recently, Salm et al. [36] reported a total g/L should be suitable. Two ciclosporin-free samples
analysis time of 2 min for the HPLC-MS/MS analysis of should be analyzed, one with internal standard (zero
ciclosporin. sample) and one without internal standard (blank
Internal Standards: The stable isotope ciclosporin-D12 sample); neither should be included when fitting the
[20,36], structural analogues such as ciclosporin D calibration curve.
[19,35], and various structurally unrelated compounds, (http://www.fda.gov/cder/guidance/4252fnl.pdf,
such as ascomycin. accessed February 15, 2008)
Mass Spectrometry Quality-control (QC) samples (n 3) that cover the
Ion Source: Electrospray (ESI) or atmospheric pressure entire calibration range should be run in duplicate, with a
chemical ionization (APCI), operated in positive or specified frequency, depending on the size of the
negative ionization mode. analytical run, to determine accuracy and precision. The
Principle of HPLC-MS: Following chromatography, the low QC should be 3 times the concentration of the
molecules enter the mass spectrometer, where they are LLOQ, the mid QC approximately midway between the
ionized and detected according to their mass-to-charge low and high QC, and the high-range QC near the upper
ratio (m/z). Various ions are formed; [M-H]- m/z: 1201, end of the concentration range.
[M+H]+ m/z: 1203, [M+Na]+ m/z: 1225 and [M+K]+ m/z: (http://www.fda.gov/cder/guidance/4252fnl.pdf,
1241. However, ciclosporin (m/z: 1225) and the D accessed February 15, 2008)
analogue (m/z: 1239) are typically detected by the single The purity, storage, and expiry/retest dates of the
ion monitoring (SIM) of their stable sodium adducts ciclosporin reference material used to prepare the Cals
[M+Na]+ [48]. and QCs should be known. If possible, all critical
Principle of HPLC-MS/MS: Following chromatography, analytical measurements should be made from
the molecules enter the mass spectrometer, where they equipment which is traceable to a national/international
are ionized (precursor ion), filtered according to their standard (e.g., glassware, calibration check weights for
m/z, and undergo fragmentation using a collision gas balances, etc.).
(e.g., Ar, N2 or He) to produce product ions, which are Preparation of Ciclosporin Calibrator and Control
again filtered according to their m/z. Analytes are Stock Solutions:
detected by monitoring the precursor-to-product-ion Prepare two separate stock solutions for the Cals and
transitions in multiple reaction monitoring (MRM) mode QCs by dissolving 10 mg ciclosporin in 10 mL of
[48]. methanol. Remember to correct for purity.
Ciclosporin can gain [M+H]+ or lose [M-H]- a proton Preparation of Ciclosporin Whole-Blood Calibrators:
when analyzed in positive or negative ionization modes, Pipette 100 L of the Cal stock solution into a 10-mL
respectively. MRM transitions monitored in positive volumetric flask, and make up to the mark with
ionization mode are m/z: 1203/425,100 (ciclosporin), ciclosporin-free human blood to produce a 10 g/mL Cal
m/z: 1217/425 (ciclosporin D) and m/z: 1215/437,100 sub-stock (CSS). The percentage solvent content should
(ciclosporin-D12), for ciclosporin and its internal always be 5% when spiking biological matrices.
standards. In negative ionization mode, the following Prepare working calibrators by diluting the CSS with
MRM transitions are typically monitored for ciclosporin human blood as follows:
469
Cal Concentration
Calculations
Vol of CSS (mL) Total Vol Blood (mL) (g/L) Calibrators:
0.000 10 0 The peak area ratio (PAR) for ciclosporin is
0.025 10 25 calculated as follows:
0.050 10 50 Peak area of calibrator
0.100 10 100 Peak area of internal standard
A calibration curve is generated by using the simplest
0.250 10 250
regression model that adequately describes the
0.500 10 500 concentration-response (PAR) relationship.
1.000 10 1000 (http://www.fda.gov/cder/guidance/4252fnl.pdf,
1.500 10 1500 accessed February 15, 2008) For a calibration curve to
2.000 10 2000 be considered acceptable, the back-calculated
concentrations should be within 15% of the expected
Preparation of Ciclosporin Whole-Blood Quality value for all non-zero calibrators, except the LLOQ
Controls: (20%). At least 75% of the calibrators must fall within
Pipette 100 L of the QC stock solution into a 10-mL these criteria, including the LLOQ and the upper limit of
volumetric flask, and make up to the mark with quantification (ULOQ). Figure 5 shows a typical
ciclosporin-free human blood to produce a 10 g/mL QC calibration curve for ciclosporin.
sub-stock (QCSS). The percentage solvent content Quality Controls and Patient Samples:
should always be 5% when spiking biological The concentrations for the QCs and patient samples are
matrices. calculated using the calibration regression model. For an
Prepare working control samples by diluting the QCSS analytical run to be considered acceptable, at least 67%
with human blood as follows: of the QCs must be within 15% of their expected
concentrations
Vol of QCSS (mL) Total Vol Blood (mL) QC Concentration (g/L)
0.075 10 75 (http://www.fda.gov/cder/guidance/4252fnl.pdf,
0.750 10 750 accessed February 15, 2008).
1.750 10 1750
470
D-Xylose
D-Xylose
Michael D.D. McNeely
Name: D-Xylose
Clinical significance: Refer to Chapter 34, The Pancreas: Function and Chemical Pathology, in the
Molecular mass: 150.13
Merck Index: 9897
Chemical class: Carbohydrate
Structure: click here
Other methods include the measurement of serum-

Principles of Analysis and Current Usage reducing substances with alkaline ferricyanide,[6] breath
Three general analytical approaches have been used for
testing with 14C-labeled xylose,[7] an enzyme method
the measurement of d-xylose: reaction with ortho-
using d-xylose isomerase and d-glucitol
toluidine,[1] para-bromoaniline,[2] and
dehydrogenase,[8] and various procedures that use acids
phloroglucinol.[3]
to dehydrate pentoses to furfurals, which are determined
by reaction with reagents such as aniline,[9] orcinol
In hot (100 C) glacial acetic acid, o-toluidine reacts
ferric chloride (Bials reagents),[10] benzidine,[11] or
with xylose and other 5- and 6-carbon sugars to form
xylidine.[12] These approaches have no advantage over
colored aromatic complexes with different maximal
the three described above, and they are rarely used.
absorbances (method 1, D-Xylose Methods Summary
Table).
The o-toluidine method was very convenient when
Using a two-wavelength technique, one can subtract the
ortho-toluidine was a common manual reagent for
absorbance resulting from the reaction of other sugars to
glucose. On the other hand, o-toluidine reagent is a
assess the amount of color produced by xylose.
noxious, possibly carcinogenic, and corrosive material
Measurements are made at 480 and 630 nm (D-Xylose
that is difficult to pipet accurately. When plasma glucose
Figure 1). Borate is added to form a complex with
concentrations are greater than 3 g/L, the absorbance of
hydroxyl groups and allow greater reactivity and
the reaction solution becomes impossible to read
absorbance. The method of Goodwin is the preferred
accurately. The method is susceptible to interference
version.[4] The advantage of this approach is that it can
from other pentoses. Hexoses other than glucose may
be performed directly on serum, plasma, and urine. The
inaccurately elevate or depress the results by altering the
reaction time is only 10 min.
A480/A630 ratio. Fortunately, the plasma and urine
The phloroglucinol method, as described by Eberts,[3] is concentrations of sugars other than glucose is very low
based upon Tollens test[5] for pentoses. It relies on the and quite stable under conditions of the clinical test
reaction of furfural, which is produced by the acid- protocol.
catalyzed dehydration of xylose, with the phloroglucinol
to produce a colored complex (method 2, D-Xylose The p-bromoaniline analytical approach is
Methods Summary Table). The reaction is monitored at recommended. In our experience, the p-bromoaniline
554 nm. This method is less susceptible to glucose method is the most consistent assay, particularly if
interference than the other two, requires a single reagent performed only occasionally. Glucose and other sugars
without protein precipitation, requires a 4-min do not interfere, except at very high levels. It is
incubation step, and does not require dual-wavelength necessary to perform a protein precipitation on serum
measurements. and plasma. Nevertheless, d-xylose analyses do not
require a rapid turnaround time, and the relatively long
The most commonly used method to measure d-xylose is incubation time (70 C and room temperature, for a total
the p-bromoaniline procedure (method 3, D-Xylose of 80 min) is seldom a constraining feature. A definitely
Methods Summary Table). The basis of the method is negative feature of this method is that it uses a reagent
dehydration of d-xylose (a pentose) to furfural, which (thiourea) that is considered to be a carcinogen in
then combines with p-bromoaniline to produce a pink animals.[13]
chromogen in the presence of a thiourea antioxidant.[2]
The phloroglucinol method should also be considered as
The reaction conditions are mild in order to prevent the a recommended alternative method. Its short reaction
formation of furfurals from other carbohydrates, time and use of nontoxic reagents are attractive features.
glucuronic acid, and ascorbic acid. The reaction is A comparison of this procedure to a highly specific gas-
monitored at 520 nm. chromatographic (GC) procedure[14] showed an
471
D-Xylose
excellent correlation for both serum (r = 0.98) and urine passive diffusion must play a part. A weak carrier
(r = 0.99). The slopes of the regression lines comparing mechanism has been postulated.
the results of the phloroglucinol assay to the results of
the GC procedure were both 1. The phloroglucinol Of an ingested dose, some 30% to 40% is excreted in the
method did have a slight positive bias compared to the urine. After an intravenous dose of d-xylose,
GC method at lower concentrations of d-xylose. A approximately 40% appears in the urine. Up to 20%
comparison of the reactions listed in d-Xylose Methods undergoes cellular metabolism, since radioactive
Summary Table is found in d-Xylose Table: Comparison remnants can be detected in the expired breath as CO2.
of reaction conditions. This leaves about 40%, which either is not absorbed or is
metabolized by some unknown pathway. Studies have
Specimen and Patient Protocol demonstrated that it is not stored as glycogen.
The patient is instructed to fast overnight but is Presumably, the remaining amount is metabolized
encouraged to drink an ample amount of water. At the (possibly in the liver).[17]
start of the test, the patients bladder is emptied, and then
a blood specimen is drawn into a fluoridated or Whatever its fate, the essential fact is that d-xylose is
iodoacetate phlebotomy tube. Twenty-five grams of d- primarily absorbed by a passive mechanism that is
xylose, dissolved in 150 to 200 mL of hot tap water and largely dependent on the general integrity of the
cooled with 300 to 350 mL cold tap water, is ingested by gastrointestinal mucosa. Once absorbed, a large fraction
the patient (0.5 g per kilogram of body weight to a of it is excreted rapidly into the urine.
maximum of 25 g is given to smaller subjects). All urine
passed over the next 5 h is collected in a single This phenomenon is the basis of the d-xylose absorption
container. Urine specimens are stable for 48 h at 4 C test in which the urinary excretion of d-xylose after an
and for many weeks if frozen. Blood specimens may be oral load is directly correlated with the passive
drawn at 1 and 2 h. absorptive capacity of the small intestine.
Fluoridated plasma samples are preferred over serum,
but both can be used. The samples, if separated within 2 Variations in the clinical test have centered on the dose
h, are stable at least 2 weeks in the refrigerator and for of d-xylose, the timing of specimen collection, the
many months when frozen. Heparin anticoagulant and choice of blood or urine, and the analytical technique for
iodoacetate preservative are also acceptable. d-xylose quantitation.
d-Xylose Reference Interval Dose
Adults given the full 25 g dose will excrete at least 25% The standard dose of d-xylose has been 25 g.
over the next 5 h if renal function is normal. One-hour This amount is considered too large by some workers
and 2-h plasma levels usually rise above 300 mg/L. who believe that it causes an osmotic assault on the
small intestinal lumen, which then reduces sugar
Children should excrete 25% of the dose into the urine, absorption. To counter this criticism, a 5 g dose has been
but the plasma levels may be slightly lower (80 to 280 recommended. However, the smaller doses have also
mg/L). been criticized because of the observation that up to 10 g
of d-xylose can be metabolized by intestinal bacteria.
With a 5 g dose, the 1-h plasma concentration should be This claim has been countered by Kendall, who
above 200 mg/L in young children. In adults and demonstrated that there is little effect on D-xylose from
cooperative children, a 5-h urine collection is best. An Klebsiella, Enterobacter aerogenes, Proteus mirabilis,
excretion of less than 30% of the oral dose is considered and Streptococcus fecalis.[18]
abnormal. d-Xylose is relatively expensive ($0.40/g), and
therefore the use of a 5 g rather than a 25 g dose will
Interpretation result in a saving of $8.00 per test.
The assessment of d-xylose absorption is used as a Prolonged administration of d-xylose to rats has
general test of the small intestines ability to absorb been shown to be carcinogenic. However, specially
nutrients. d-Xylose is a dextrorotatory aldopentose, formulated d-xylose has recently been approved for use
which is known to be absorbed at a much slower rate in this test (Xylo-Pfan, Adria Laboratories, Inc., P.O.
than glucose.[15] Helmer and Fouts in 1937[16] Box 16529, Columbus, OH 43215).
observed that two patients with gastrointestinal disease
absorbed xylose at a slower than expected rate. This Timing
started the investigations that ultimately led to the After its absorption, the blood concentration of
verification of the test as an indicator of gastrointestinal d-xylose peaks at 2 h in adults and 1 h in children.[19]
disease. However, this is a highly variable phenomenon, and
therefore the collection of a 5-h, postingestion urine is
We now know that d-xylose is absorbed most efficiently advocated. The administration of a 5 g dose permits the
in the jejunum. Since there is no active renal tubular use of a 2-h urine collection.
reabsorption, the sugar is excreted rapidly into the urine. A test relying solely on a 1-h blood sample was
It has been demonstrated that d- and l-xylose are first reported by Rolles in 1973 and confirmed by
absorbed at different rates. Thus some factor other than others.[20,21]
472
D-Xylose
The 1-h approach has been criticized by of d-xylose isomerase with d-glucitol
Lamabadusuriya,[22] Christie,[23] and Liebman,[24] dehydrogenase. Anal Biochem 1977;80:41-50.
who cited a sensitivity of 39% to 95% and a specificity 9 Goodwin JF. Micromethod for measuring
of 24% to 78%. Kraut and Lloyd-Still[25] used a 5 g pentoses by use of an aniline reagent. Clin
dose and a 60-min collection and compared their results Chem 1971;17:397-9.
with those of intestinal biopsy. They obtained a 10 Rozental M, Tomaszewski L. A new simple
sensitivity of 68% to 96%, depending on the category of ultramicromethod for the determination of d-
the disease, and a specificity of 100%. xylose in blood and urine. Clin Chem Acta
Santini and Sheehy[26] used a 5 g dose with a 1974;50:311-7.
5-h urine collection in a study of 125 normal subjects. 11 Tauber H. Color test for pentoses. Proc Soc Exp
All the subjects excreted more than 30% of the ingested Med Biol 1937;37:600-1.
dose, whereas 60 patients with untreated tropical sprue 12 Suminokura K. Colorimetric method for
excreted 5% to 15%, and 36 patients with tropical sprue determination of pentoses. J Biochem
excreted 14% to 30% of the ingested dose. 1931;14:343-59.
Hill and Cherian showed that the specificity of 13 Department of Health and Human Services,
the test was 95% and the sensitivity was 70% for celiac USPHS, Centers for Disease Control, National
disease and 25% for nonceliac gastrointestinal disorders. Institute for Occupational Safety and Health.
All factors considered, the use of a 25 g dose in Registry of toxic effects of chemical substances.
combination with a 5-h urine collection is recommended. Vol. 11 981-1982.
For routine use, a blood sample is not necessary. 14 Johnson SL, Bliss M, Mayersohn M, Conrad
Low serum levels of d-xylose or low excretion KA. Phloroglucinol-based colorimetry of xylose
rates are seen in a wide variety of conditions affecting in plasma and urine compared with a specific
the jejunum. Impaired absorption indicates decreased gas-chromatographic procedure. Clin Chem
mucosal absorption ability. In addition, low levels are 1984;30:1571-4.
associated with ascites (where much of the dose may be 15 Cori CF. Fate of sugar in animal body; rate of
retained in the abdominal cavity), vomiting, delayed absorption of hexoses and pentoses from
gastric emptying, improper urine collection, high-dose intestinal tract. J Biol Chem 1925;66:691-715.
aspirin therapy, neomycin, colchicine, indomethacin, and 16 Helmer OM., Fouts PJ. Gastrointestinal studies:
atropine. Impaired renal function will cause a dramatic excretion of xylose in pernicious anemia. J Clin
reduction in xylose excretion. Invest 1937;16:343-9.
17 Hiszczynskyj R, Keitges PW. D-Xylose
References absorption test. Advanced Clinical Chemistry
1 Goodwin JF. Method for simultaneous direct No. ACC-3 (1972), American Society of
estimation of glucose and xylose in serum. Clin Clinical Pathologists, Commission on
Chem 1970;16:85-91. Continuing Education, Check sample program,
2 Roe JH, Rice EW Photometric method for 1973.
determination of free pentoses in animal tissues. 18 Kendall MJ, Nutter S, Hawkins CF. Bacteria
J Biol Chem 1948;173:507-12. and the xylose test. Lancet 1972;1:1017-8.
3 Eberts TJ, Sample RHB, Glick MR, Ellis GH. 19 Rolles CJ, Nutter S, Kendall MJ, Anderson CM.
A simplified colorimetric micromethod for One-hour blood-xylose screening-test for
xylose in serum or urine with phloroglucinol. coeliac disease in infants and young children.
Clin Chem 1979;25:1440-3. Lancet 1973;2:1043-5.
4 Goodwin JF. Method for simultaneous direct 20 Schaad U, Gaze H, Hadorn B. Value of 1-hour
estimation of glucose and xylose in serum. Clin blood-xylose test in diagnosis of childhood
Chem 1970;16:85-91. coeliac disease. Arch Dis Child 1978;53:420-2.
5 Oshima K, Tollens B. Ueber Spectral- 21 Buts JP, Morin CL, Roy CC, Weber A, Bonin
reactionen des Methylfurfurols. Ber Dtsch A. One-hour blood xylose test: a reliable index
Chem Ges 1901;34:1425. of small bowel function. J Pediatr 1978;92:729-
6 Haeney MR, Culank LS, Montgomery RD, 33.
Sammons HG. Evaluation of xylose absorption 22 Lamabadusuriya SP, Packer S, Harries JT.
as measured in blood and urine: a one-hour Limitations of xylose tolerance test as a
blood xylose screening test in malabsorption. screening procedure in childhood coeliac
Gastroenterology 1978;75: 393-400. disease. Arch Dis Child 1975;50:34-9.
7 Roberts RK, Campbell CB, Bryant SJ, Adames 23 Christie DL. Use of the one-hour blood xylose
L. Xylose-1-14C absorption test: the use of test as an indicator of small bowel mucosal
urine, serum and breath analysis, and disease. J Pediatr 1978;92:725-8.
comparison with a colorimetric assay. Aust NZ 24 Liebman WM. Xylose test in malabsorption. J
J Med 1976;6:532-6. Pediatr 1979;94:508-9
8 Kersters-Hilderson H, Van Doorslaer E, De 25 Kraut JR, Lloyd-Still JD. The 1-hour xylose test
Bruyne CK, Yamanaka K. Quantitative in the evaluation of malabsorption in infants
determination of d-xylose by a coupled reaction and children. Am J Clin Nutr 1980;33:2328-33.
473
D-Xylose
26 Santini R, Sheehy TW, Martnez-de Jess J. Gastroenterology 1961;40:772-4.

The xylose tolerance test with a five-gram dose.
474
D-Xylose
Tables
D-Xylose Methods Summary Table

Method 1: o-Toluidine
Principle of analysis: o-Toluidine reacts with pentoses to form products with different absorbance maxima;
differential spectrophotometry is carried out
Comments: Serum, plasma, urine; 10-min incubation at 100 C, no precipitation step; glucose may interfere at
high concentration
Method 2: Phloroglucinol
Principle of analysis: Acetic and HCl acid produce furfural, which reacts with phloroglucinol to produce a
colored compound with absorbance at 554 nm
Comments: Serum, plasma, urine; 10-min incubation; single reagent; no protein precipitation step
Method 3: p-Bromoaniline
Principle of analysis: p-Bromoaniline acetate and thiourea antioxidant react with furfural to develop a pink
color, monitored at 520 nm
Comments: Serum, plasma, urine; standard method; protein precipitation step and 70-min incubation
Comparison of Reaction Conditions for d-Xylose

Condition p-Bromoaniline o-Toluidine Phloroglucinol
Temperature 70 C; 100 C 100 C
ambient temperature
Sample volume Plasma, 1 mL Serum, 0.1 mL Serum, 50 L
Urine, 1 mL Urine, 5 L
Sample pretreatment Deproteinization None None
of serum
Fraction of sample 0.167 0.0196 Serum, 0.0099
volume Urine, 0.001
Reagent p-Bromoaniline o-Toluidine Phloroglucinol
Reaction time 80 min 13 min 4 min
Linearity 13.32 mmol/L 13.2 mmol/L
Precision 4.4% urine Coefficient of variation
7.4% blood 3.0% to 5.2%
Interferences Substances other Bilirubin Glucose
than xylose that Hemolyzed serum Ribose
produce furfurals Glucose Arabinose
Methylxanthines
475
D-Xylose
Figures
D-Xylose Figure 1
Absorbance spectra of ortho-toluidine reaction with D-xylose, dashed curve, and D-glucose, solid curve (each at 1 g/L).
Absorbance versus reagent blank.
D-Xylose Figure 2
Absorbance spectra of p-bromoaniline reaction with D-xylose (100 mg/L). Reagent blank versus water, dashed curve; D-
xylose versus reagent blank, solid curve.
476
D-Xylose
D-Xylose Figure 3
Absorbance spectra of phloroglucinol reaction with D-xylose (5 mg/L). Reagent blank versus water, dashes; D-xylose
versus water, dashes with dots; D-xylose versus reagent blank, solid curve.
undissolved excess solid. Stable for 2 months at room

Procedure: p-Bromoaniline Reaction for d-Xylose temperature.
4. Saturated benzoic acid. Add 15 g of benzoic
Principle acid to 500 mL of distilled water. Allow to mix for 48 h
Urine or protein-free supernatant of plasma is
at room temperature. Use if solid benzoic acid is still
added to a reagent consisting of p-bromoaniline in
visible.
glacial acetic acid to produce a pink chromogen. The
5. Stock Place 200 mg of d(+)-xylose in a 100
absorbance spectrum for the chromogen is shown in D-
mL volumetric flask and dissolve in and bring to mark
Xylose Figure 2.
with saturated benzoic acid. Stable for 12 months at
Reagents refrigerated temperature.
1. Zinc sulfate, ZnSO4 7H2O, 50 g/L (0.174 6. Working Dilute 5 mL of stock standard to 100
mL with saturated benzoic acid. Stable for 3 months at
mol/L). Place 50 g of ZnSO4 7H2O in a 1 L flask. refrigerated temperature.
Dissolve in approximately 900 mL of distilled water. 7. Working standard, 200 mg/L (1.33 mmol/L).
Bring to volume, and mix. Store at room temperature for Dilute 10 mL of stock standard to 100 mL with saturated
up to 1 year. benzoic acid. Stable for 3 months at refrigerated
2. Barium hydroxide, Ba(OH)2 8H2O (0.158 temperature.
mol/L). Dissolve 25 g in 500 mL of distilled water; then Assay
boil, cool, and filter. Stable at room temperature for up Equipment: Heating bath set at 70 2 C;
to 1 year.[27] spectrophotometer with a band pass less than 10 nm at
3. p-Bromoaniline reagent (116 mmol/L) in 520 nm.
saturated thiourea. To prepare saturated thiourea, add 1. Prepare protein-free supernatant of plasma by
7.5 g of thiourea to 150 mL of glacial acetic acid, and mixing 1.0 mL of plasma, 5.0 mL of distilled
allow to mix overnight. (Caution: Believed to be an water, and 2.0 mL of 0.158 M barium
animal carcinogen.) Place 2.0 g of p-bromoaniline in a hydroxide. Then add 2.0 mL of 0.174 M zinc
100 mL volumetric flask, and bring to mark with the sulfate, mix, and centrifuge at 500 g for 5 min.
saturated solution of thiourea, which is decanted off the
477
D-Xylose
2. Prepare 1:50 and 1:100 aqueous dilutions of 6. Working xylose Dilute 5 mL of stock standard
urine (stable 48 h in refrigerator, longer if to 100 mL with saturated benzoic acid. Stable for 3
frozen). months at refrigerated temperature.
3. Pipet 1.00 mL of distilled water, standards, 7. Working standard, 200 mg/L (1.33 mmol/L).
plasma protein-free supernatant, and urine Dilute 10 mL of stock standard to 100 mL with saturated
dilutions into a series of labeled, duplicate glass benzoic acid. Stable for 3 months at refrigerated
tubes. temperature.
4. Add 5.00 mL of p-bromoaniline reagent to all
tubes. Assay
Equipment: Heating block (105 C);
5. Incubate one set of tubes in a heating bath at 70
spectrophotometer (band pass <10 nm) capable of
2 C for 10 min. Cool under running water.
reading at 554 nm.
The second set will serve as blanks.
1. Set heating block at 105 C (under fume hood).
6. Place all tubes in the dark at room temperature
for 70 min. 2. Run standards and patient samples in duplicate.
Label 13 100 mm disposable tubes as follows:
7. Read absorbance of each tube within 30 min at 1 for each serum blank, 1 for each standard
520 nm. Unheated tube serves as blank. If blank, 2 for each standard, 2 for each patient.
absorbance is too high, dilute both tubes with p-
3. Pipet 5 mL of working reagent into each tube.
bromoaniline reagent, reread, and make
appropriate correction. 4. Add 50 L of distilled H2O to standard blank.
Calculation Add 50 L of any normal (xylose-free) serum to
Plasma: serum blank. Add 50 L of plasma, serum, or
standard to respective tubes or 5 L of urine (if
analyzing a urine). Vortex mix well.
5. Simultaneously transfer all tubes to heating
block, and set clock for exactly 4 minutes.
Urine: NOTE: Do not add tubes at intervals because
temperature will drop during incubation, and different
tubes will be affected differently.
6. After incubation, immediately cool tubes in an
ice-water bath. Mix again.
7. Read absorbance of each tube at 554 nm, using
the standard blank to set the spectrophotometer
to 100%T (zero absorbance).
D-Xylose Reference Interval
click here
Calculation
Procedure: Phloroglucinol Reaction for d-Xylose Serum:
Principle A554 patient A554 serum blank mmol/L standard
Simplified colorimetric procedure using A554 standard
phloroglucinol (Trinder and Tollen application) wherein = mmol/L
d-xylose is heated in strong acid to form furfural. Use the standard whose absorbance is closest
Furfural reacts with phloroglucinol to produce a pink- to that of the sample.
colored compound, with a high molar absorptivity at 554
nm (d-Xylose Figure 3).
Urine:
Reagents 1. Calculate as for serum, but multiply urine
1. Glacial acetic acid results by 10 (5 L of urine and 50 L of
2. Concentrated HCl standard were used) to find mmol/L.
3. Working phloroglucinol reagent (36 2. Convert mmol/L to mg/L: mmol/L 150
mmol/L). Place 0.5 g of phloroglucinol (Sigma mg/mmol = mg/L.
Chemical Co., St. Louis, P-3502) in an Erlenmeyer flask, 3. Calculate for 5 h = mg/L Total volume (in
and dissolve in 100 mL of glacial acetic acid and 10 mL liters).
of concentrated HCl. Keep in a dark bottle. Stable only
4. Divide by 1000 mg/g, and report as g/5-h urine.
for 4 days at room temperature. Use gloves in handling
reagent.
4. Stock xylose Place 200 mg of d(+)-xylose in a
Notes
100 mL volumetric flask and dissolve in and bring to
mark with saturated benzoic acid. Stable for 12 months 1. Perform a glucose analysis on the patients
at refrigerated temperature. serum specimen. Record on slip. Slight
interference is seen in samples with high level
of glucose (>5000 mg/L).
478
D-Xylose
2. Color reaction is linear to 10.0 mmol/L. on the requisition. The total volume of the urine must be
known to calculate results and interpret the test. If this
information is unavailable, send out measured value
D-Xylose Reference Interval (mm/L) and free text comment that no reference
click here intervals are available for uncalculated results.
3. Both urine and serum samples are stable at 28
d-Xylose in Serum/Plasma and Urine C for 1 week.
I. Principle of Test and Clinical Significance
III. Reagents
A. Principle: Simplified colorimetric procedure
using phloroglucinol (Trinder and Tollen application). d- A. Phloroglucinol (Sigma P-3502)
Xylose is heated in strong acid to form furfural. Furfural B. Glacial acetic acid (Fischer A38-212)
reacts with phloroglucinol to produce a pink colored C. Hydrochloric acid (Fischer A144st-212)
compound with high molar absorptivity at 550 nm. This D. d-Xylose (Sigma X-1500)
test can be used to differentiate between intestinal E. Saturated benzoic acidadd benzoic acid to
malabsorption and pancreatic insufficiency. 500 mL distilled water until it settles out.
F. Working reagentWeigh out 0.5 gm
B. Clinical Significance: Xylose is not normally phloroglucinol and place in a 200 mL dark brown bottle.
present in significant amounts in the blood. When given Carefully add 100 mL glacial acetic acid and 10 mL
orally, it is passively absorbed in the proximal small concentrated HCl to bottle. Mix together under fume
(duodenojejunal) intestine (to about 60%) and most is hood. Stable for 4 days at room temperature. This
excreted by the kidneys. The amount of xylose reagent is very caustic. Wear gloves and eye shields
recovered in the urine or blood in a specified time when handling and preparing reagent and perform test
interval after administration is used to evaluate intestinal under the hood.
absorption. Low absorption of xylose is observed in
intestinal malabsorption. In malabsorption due to IV. Equipment
pancreatic insufficiency, the absorption of xylose will be
essentially normal. Therefore, this test is of some help A. Heating block (Thermolyne Dri-Bath Model
to distinguish between these two types of malabsorption. #DB-17615)
B. Thermometer that reads up to 105 C
NOTE: The accuracy of the procedure depends C. Spectrophotometer (PerkinElmer Jr. model 35)
not only on the rate of absorption but also on the rate of D. 13 x100 disposable test tubes
excretion of xylose by the kidneys. In order to eliminate E. 5 mL serological pipet
misinterpretations as a result of renal retention, a blood F. 50 L pipettor (MLA)
determination of xylose is carried out along with the G. MLA adjustable volume pipettor (adjusted to
determination of xylose in urine. A normal blood xylose 450 l)
level in the presence of decreased urine xylose would H. 12 x 75 mm glass cuvettes
suggest renal retention.
V. Calibration
Low values for d-xylose absorption are also
observed in a number of other conditions, including A. Stock d-xylose standard (50 mmol/L). Weigh
celiac disease, tropical sprue, Crohns disease, out 0.3755 g d-xylose. Place in 50 mL volumetric flask
immunoglobulin deficiency, pellagra, ascariasis, and QS to line with saturated benzoic acid.
vomiting, delayed gastric emptying, inadequate
hydration, decreased circulation, intrinsic renal disease, B. Working Standards
thyroid disease, sequestration in body fluids as occurs in
pregnancy and ascites, and incomplete urine collection. 1. 5.0 mmol/Lusing volumetric pipet add 1 mL
of the 50 mmole/L stock d-xylose standard and 9 mL of
II. Sample (Specimen) saturated benzoic acid to a 16 x 100 mm test tube. Mix
thoroughly.
A. Patient Preparation Instructions: A 25 g oral 2. 1.0 mmol/Ladd volumetrically 1 mL of the 5
dose of xylose in water is given to a fasting patient. mmol/L standard and 4 mL of saturated benzoic acid in a
Blood should be drawn at 1 and 2 h intervals after the d- 16 x 100 mm test tube. Mix thoroughly.
xylose is administered. All urine should be collected for
5 h following the dose. All the above standards are stable for 3 months at room
temperature.
B. Sample
Routinely run the 5 and 1 mmol/L standards every time
1. A minimum of 150 l serum or plasma. the assay is run and select the one with O.D. closest to
2. A minimum of 50 l urine is required. Urine the O.D. of patient sample for calculations. Standard
should be mixed and measured. Send a 10 ml aliquot of duplicates should have <5% variation between tubes.
the 5 h collection for analysis. Record the total volume
479
D-Xylose
VI. Quality Control F. After incubation, immediately cool tubes (use

ice water). Mix again.
A. Run any patient with a normal glucose as a
negative OC (serum) with each run. G. Read on PerkinxElmer Jr.:
B. Run an aliquot of the spiked pool as positive
QC (found in -70 C freezer) with each run. 1. Pour solutions to be read into clean 12 x 75
C. The QC should be within + 2SD of the mean spectrophotometer cuvettes.
from several interassay determinations (see overlap 2. Set to 550 filter.
policy). See range in QC notebook. Record QC on run 3. Set to absorbance.
sheet and in QC notebook. 4. Zero against proper blank (standard or serum).
D. If QC is out of range, repeat the assay with 5. Read absorbance of patients and QC.
fresh standards and reagent. 6. Write down absorbance values of QC # and
E. If QC is still out of range, consult supervisor. patients on worksheet.
To make spiked pool: Add 1 mL of 50 mm/L standard 7. Calculate results.
to 24 mL of normal serum, not normal control (there is
color interference). (Glucose should be normal.) The H. After incubation, solutions to be read are stable
value will be approximately 2 mm/L. Mix well, aliquot, for 1 h. Keep on ice until they are read.
and store in -70C freezer. Run several times to
accumulate data and calculate range. I. Solutions may be discarded down drain under
hood but must be followed by running water for 12
VII. Procedure min.
WEAR GLOVES AND PROTECTIVE J. Clean 12 x 75 cuvettes by washing in solution

EYEWEAR DURING COMPLETE PROCEDURE. of Acationox (C6330 American Scientific), 1 capful to 2
L water. Rinse well with distilled water. Place upside-
A. Set Dow heat block to 105 C (under hood). down in rack and dry in oven.
Acceptable temperature 105 C + 3 C. Write down
temperature on run sheet at beginning of incubation. VIII. Calculation
B. Run standards and patients in duplicate. Label
13 x 100 disposable tubes: O.D. patient x mmol/L std. = mmol/L
O.D. standard
1 serum blank
1 standard blank Urine:
2 for each standard
2 for each patient 1. Multiply urine results by 10 (dilution factor) =
2 for each QC mmol/L.
2. Convert mmol/L to mg/dL: mmol/L x 15 =
C. Pipet 5 ml of working reagent into each tube mg/dL.
(volumetrically). 3. Calculate for 5 h = mg/dL x TV/100.
4. Divide by 1000 and report as grams/5 h urine.
D. Pipet QC, standards and patients into
appropriate tube as follows (using calibrated pipets): Expected Results:
1. Standard blank: 50 L H2O to standard blank Blood:

(zero standards against this) Adult (after 25 g dose) > 25 mg/dL(> 1.66 mm/l)
2. Serum blank: 50 L of any normal (xylose- Child (after 25 g dose) > 30 mg/dL (> 2.0 mm/l)
free) serum-to-serum blank (zero all patients against Urine:
this). Adult (after 25 g dose) > 4 g/5 h urine
3. 50 l of plasma, serum or standard to respective Child (1633% of given dose)
tubes.
4. 50 l urine (make 10 dilution of urine before X. Procedural Notes
pipetting into tube with working reagent). Mix well
(vortex) and cap with Multifit stoppers (Kew Scientific). A. This procedure is sensitive to pipetting
technique. Be sure to use a volumetric pipet when
E. Simultaneously add all tubes to heat block and dispensing the working regent and to use good pipetting
set clock for 4 minutes. NOTE: Do not add tubes at technique when adding the 50 L of sample to the
intervals because temperature will drop during reagent.
incubation and different tubes will be affected B. This procedure may be run in triplicate instead
differently. of duplicate to ensure that absorbances agree within 5%.
When running in triplicate, the best 2 out of 3
absorbances may be used in calculating the average.
480
D-Xylose
C. When cooling the tubes after inoculation, make

sure to immerse the tubes at least halfway in the ice-
water bath and to leave them in a sufficient amount of
time to allow complete cooling (about 5 min).
XI. Procedure Performance
A. Color reaction is linear to 10.0 mmol/L.

Samples exceeding this amount should be diluted 10x
with distilled water and repeated.
B. Patients with high glucose (>500 mg/dL) may
have false positive results. Run and record on the
worksheet a glucose on each patients specimen to check
that it is normal. If it is >500 mg/dL, attach a comment
to the patients result, noting that there will be a slight
positive interference. False negative results may occur if
patient is on high-dose aspirin therapy or taking
neomycin, calcine, indomethacin, or atropine.
C. Day-to-day precision at a level of 1.9, SD =
0.12, CV = 6.3%
XII. References
Eberts, T.J., et al.: A simplified colorimetric
micromethod for xylose in serum or urine, with
phloroglucinol, Clin. Chem. 25(8):1440-1443, 1979.
481
Dehydroepiandrosterone and Its Sulfate (DHEA and DHEA-S)

Gus Koerbin
Name: Dehydroepiandrosterone, DHEA, Prasterone,
3-hydroxyandrost-5-en-17-one
Clinical Significance: Refer to Chapter 48, General Endocrinology, in the 5th edition of Clinical
Molecular formula: DHEA, C19H28O2; DHEA-S, C19H28SO5
Molecular mass: DHEA, 288.41; DHEA-S, 368.51
Merck Index: DHEA, 7606 (Prasterone)
Structure:
Principles of Analysis and Current Usage closely related steroids (Table 1, Method 4). Nieschlag
[5] described an RIA procedure for both steroids. For
The measurement of dehydroepiandrosterone (DHEA) DHEA, plasma was extracted with ether and then
and dehydroepiandrosterone sulfate (DHEA-S) is purified by TLC before RIA. DHEA-S was extracted
important for the assessment of adrenal androgen from plasma into ethyl acetate, hydrolyzed in acetic acid
production [1]. Gas-liquid chromatography (GLC) was at 75C to form free DHEA, and then purified by TLC.
first used in a number of procedures (Table 1, Method 1) Separation of bound from free labeled DHEA was
[2]. In each procedure, direct measurement was not achieved with globulin-dextran-coated charcoal.
possible, since DHEA and its sulfate are thermo-labile
and required derivatization. Direct RIA measurement of DHEA-S in plasma was
described by Buster and Abraham [6] and Cattaneo [7].
An exhaustive double-isotope derivative method was Because of the relatively high concentration of DHEA-S
suggested by Gandy (Table 1, Method 2) [3]. This in plasma, simple dilution of plasma reduced
procedure required five chromatographic steps and interferences, making extraction and chromatography
quantitation by liquid scintillation. unnecessary. Haning [8] improved the direct assay for
DHEA-S by using an antiserum that reduced previously
A competitive protein-binding assay [4] for total DHEA observed cross-reactivity with other steroid sulfates. An
and DHEA-S in plasma actually measured the derivative RIA from blood spotted on filter paper was developed by
5-androstenediol (5-A-diol) (Table 1, Method 3). Grueters [9]. Determination of urinary DHEA-S was
reported by Causon, using high-performance liquid
Early radioligand assays for DHEA and its sulfate chromatography and RIA [10]. It was generally found
generally required chromatographic separation before that analysis of DHEA by RIA required extensive
RIA because of cross-reactivity of the antisera with these treatment of plasma before RIA. This was necessary
because of the relatively low concentration of DHEA
i
compared with DHEA-S and other interfering steroids.
Dehydroepiandrosterone and Its Sulfate (DHEA and
DHEA-S) Assay of DHEA-S using solid-phase enzyme-linked
Previous and current authors of this method: immunosorbent assay (ELISA) based on the principle of
First edition: Not done competitive binding is now the most commonly
Methods edition: Karen L. Nickel performed assay (Table 1, Method 5). Microtiter wells or
Second edition: Not updated micro-particles (beads) are coated with a polyclonal
Third edition: Not updated antibody directed towards a unique antigenic site of the
Fourth edition: Karen L. Nickel DHEA-S molecule. Endogenous DHEA-S of a patient
Fifth edition: Gus Koerbin sample competes with a DHEA-S horseradish
482
peroxidase conjugate for binding to the coated antibody. Interpretation

After incubation, the unbound conjugate is removed. The DHEA is produced in substantial quantity in the adrenal
amount of bound peroxidase conjugate is inversely gland. It is converted to its sulfated form in either the
proportional to the concentration of DHEA-S in the adrenal gland or the liver and can be desulfated in
sample. peripheral tissues. The circulating concentration of
DHEA-S is many times higher than DHEA.
There is no reference method for either DHEA or DHEA has weak androgenic actions and may be
DHEA-S. Determination of DHEA and DHEA-S in considered a prohormone, with conversion to
serum or plasma has generally replaced urinary analysis. testosterone and/or estrogen occurring in peripheral and
However, total 17-ketosteroid assay in urine is used by genital tissue [14]. Adrenal production of DHEA is an
some clinicians as a screen for hypersecretion of adrenal important contributor to androgen production and effects
androgens. in women but not in men.
Analysis of DHEA-S in plasma is routinely performed The steady decrease in serum levels of these steroids
by solid-phase competitive chemiluminescent enzyme from early adulthood onwards suggests that they may be
immunoassay. markers of aging and that high levels may protect against
With direct RIA, neither extraction nor chromatography the pathological consequences of becoming old. Studies
is required in most methods. These procedures use a on humans demonstrate associations of DHEA and
tritiated isotope. Commercial availability of 125I-(3)- DHEA-S with survival and various measures of health
radioiodinated DHEA-S and antiserum raised to DHEA- status, including diabetes, cardiovascular disease,
S-3-betamonohemisuccinate-human serum albumin obesity, cognitive impairment, physical limitations, and
allow the use of methods with reduced counting time and depressive symptoms [15,16].
greater ease of handling.
Despite extensive research on these steroids, little is
Specimen known about their specific functions or the mechanisms
Serum, heparinized plasma, or EDTA plasma may be by which they affect health [17]. Research suggests that
used for analysis of DHEA or DHEA-S. Patients should DHEA may counterbalance the immunosuppressive
not be given any steroids or gonadotropin medications effects of glucocorticoids, although the receptor for
for at least 48 hours prior to sample collection [11]. The DHEA has not been fully characterized [18]. DHEA has
specimen is stable for 2 days at 4C and for many been shown to have biological actions on hemostasis,
months if frozen. cell proliferation, lipid metabolism, stress response, and
immune function [15]. Knowledge about causal effects
Interferences of DHEA-S for health in humans is limited and often
Heterophilic antibodies are the most common cause of inconclusive because many of the experimental studies
interference when using ELISA techniques. The use of have been performed on nonprimates, which have far
separator gel tubes (SST) may affect results. No lower concentrations of these steroids than humans [19].
significant differences have been noted between
concentrations in specimens collected in plastic and DHEA-S is important in the investigation of hirsutism
glass SST tubes [12]. If using RIA, impure absolute and alopecia in women. It is also useful in the
ethanol may leave a residue on drying that interferes assessment of delayed puberty and adrenal disorders.
with binding. Care must be taken to avoid contamination Plasma levels increase steadily from about 7 years of
of the alcohol. age, declining after about 30 years of age. Diurnal
variation is not observed.
DHEA and DHEA-S Reference Interval
Normal values should be established by each individual DHEA-S is frequently analyzed in conjunction with free
laboratory for each particular assay. Normal adult males testosterone in the investigations of hirsutism and
have typical values of 80 to 550 g/dL, (2.2 to 15.0 hyperandrogenism. One of these two hormones is likely
mol/L). A typical adult female reference interval is 37 to be elevated in these conditions. High levels are also
to 430 g/dL (1.0 to 11.6 mol/L) and for seen in polycystic ovary syndrome. Extremely high
postmenopausal women < 190 g/dL (<5.1mol/L). levels (>700 g/dL) in women are suggestive of
DHEA-S levels in women also vary according to the hormone-secreting adrenal tumors.
menstrual cycle, with the follicular and luteal phases
showing lower levels than midcycle. DHEA-S DHEA and DHEA-S Performance Goals
concentrations in infants decrease between 1 and 5 Survey data from the 2007 College of American
months after birth but are generally less than 150 g/dL Pathologists (CAP) Participant Summary Report show
(4 mol/L). Between 6 months and 7 years of age, imprecision values (% coefficient of variation) for
DHEA-S is normally less than 18 g/dL (0.05 mol/L) measurement of DHEA-S range from 4.2% to 14.3% at
in boys and less than 37 g/dL (1.0 mol/L) in girls concentrations of approximately 40 g/dL (1.0 mol/L)
[13]. In teenage children, indicatively, values are 3.7 to and 4.5% to 12.6% at 140 g/dL (3.8 mol/L) for assays
170 g/dL (0.1 to 4.6 mol/L). with > 10 participants. Less than 5% of all assays are
performed using RIA, and these methods have higher
483
imprecision, 5.4% to 14.3%, when compared with 9 Grueters A, Korth-Schutz S. Longitudinal study
ELISA, 4.2% to 11.8%. of plasma dehydroepiandrosterone sulfate in
preterm and full-term infants. J Clin Endocrinol
DHEA-S is not regulated under the Clinical Laboratory Metab 1982;55:314-320.
Improvement Amendments (CLIA 88) for proficiency 10 Causon RC, Collins SL, Fry DE. Determination
testing. Analytical performance goals set by the Royal of urinary dehydroepiandrosterone sulphate by
College of Pathologists of Australasia (RCPA) are 37 combined high-performance liquid
g/dL, up to 370 g/dL and 10% > 370 g/dL (1.0 chromatography and radioimmunoassay. J
mol/L up to 10 mol/L and 10% > 10 mol/L. Chromatogr 1982;227:485-491.
11 Inter Science Institute. Current Unique and
The 2007 CAP data show that there is a lack of Rare Endocrine Assays. Inglewood, CA:
harmonization between commonly used DHEA-S InterScience Institute; 1997.
ELISA assays. There are differences in concentration 12 Reinartz JJ, Ramey ML, Fowler MC et al.
between the Roche assays (16% of all assays) and other Plastic vs glass SST evacuated serum-separator
assays, with the Roche assays showing values up to 60% blood-drawing tubes for endocrinologic
higher at 100 g/dL (2.7 mol/L). The RCPA 2007 data analytes. Clin Chem 1993;39:2535-2536.
also show a lack of harmonization with the Roche 13 Babalola AA, Ellis G. Serum
assays. In this program, differences in concentrations are dehydroepiandrosterone sulfate in a normal
up to 30%. Within-individual biological variation has pediatric population. Clin Biochem
been found to be about 4% and between-individual 1985;18:184-189.
variation approximately 29% [20]. 14 Labrie F, Luu-The V, Belanger A et al. Is
dehydroepiandrosterone a hormone? J
References Endocrinol 2005;187:169-96.
1 Nelson DH. Adrenal androgens. In: The 15 Feldman HA, Johannes CB, Araujo AB, Mohr
Adrenal Cortex: Physiological Function and BA, Longcope C, McKinlay JB. Low
Disease. Vol 18. Major Problems in Internal dehydroepiandrosterone and ischemic heart
Medicine. Philadelphia: Saunders; 1980. disease in middle-aged men: prospective results
2 Kumari GL, Collins WP, Sommerville IF. from the Massachusetts Male Aging Study. Am
Further studies on the gas-liquid J Epidemiol 2001;153:79-89.
chromatographic determination of C19-steroids 16 Trivedi DP, Khaw KT.
in human plasma using nickel-63 electron Dehydroepiandrosterone sulfate and mortality
capture detection. J Chromatogr 1969;41:22-36. in elderly men and women. J Clin Endocrinol
3 Gandy HM, Peterson RE. Measurement of Metab 2001;86:4171-7.
testosterone and 17-ketosteroids in plasma by 17 Celec P, Starka L. Dehydroepiandrosterone: is
the double isotope dilution derivative the fountain of youth drying out? Physiol Res
technique. J Clin Endocrinol 1975;4:505-512. 2003;52:397-407.
4 Rosenfield RL. A competitive protein binding 18 Butcher SK, Lord JM. Stress responses and
method for the measurement of unconjugated innate immunity: aging as a contributory factor.
and sulfate-conjugated dehydroepiandrosterone Aging Cell 2004;3:151-60.
in peripheral plasma. Steroids 1971;17:689- 19 Dhatariya K, Bigelow ML, Nair KS. Effect of
696. dehydroepiandrosterone replacement on insulin
5 Nieschlag E, Loviaux DL, Lipsett MB. sensitivity and lipids in hypoadrenal women.
Radioligand assay for delta-5-3-beta- Diabetes 2005;54:765-9.
hydroxysteroids. I. 3-Beta-hydroxy-5- 20 Westgard QC. Desirable Specifications for
androstene-17-one and its sulphate. Steroids Total Error, Imprecision, and Bias, Derived
1972;19:669-679. from Biologic Variation. Available at
6 Buster JE, Abraham GE. Radioimmunoassay of <http://www.westgard.com/biodatabase1.htm>
plasma dehydroepiandrosterone sulphate. Anal Accessed 2008.10.18.
Lett 1972;5:543-551. 21 Anderson DC, Hopper BR, Lesley BL, Yen SC.
7 Cattaneo S, Forti G, Fiorelli G, Barbieri U, A simple method for the assay of eight steroids
Serio M. A rapid radioimmunoassay for in small volumes of plasma. Steroids
determination of dehydroepiandrosterone 1976;28:179-196.
sulphate in human plasma. Clin Endocrinol 22 Smith MR, Rudd BT, Shirley A, Rayner PH,
1972;4:505-512. Williams JW, Dugnan ND, Bertrand PV. A
8 Haning RV, Carlson IH, Shapiro SS, Nolten radioimmunoassay for the estimation of serum
WE. Testosterone free index correlates best dehydroepiandrosterone sulphate in normal and
with dehydroepiandrosterone sulphate. Fertil pathological sera. Clin Chim Acta 1975;65:5-
Steril 1981;36:757-765. 13.
484
Table 1: Methods of DHEA and DHEA-S Analysis

Method 1: Gas-liquid chromatography (GLC)
Principle of analysis: Extraction; analyte partially purified by chromatography, derivatized, analyzed by GLC,
detected to electron capture
Comments: Plasma; rarely used
Method 2: Double-isotope derivatization
Principle of analysis: Radioactive derivatives are formed, purified by chromatography, and quantitated by
radioisotopic counting
Comments: Plasma; very rarely used
Method 3: Competitive protein binding (CPB)
Principle of analysis: Extraction, purification by thin-layer chromatography; enzymatic conversion to 5-
androstenediol; CPB using testosterone-binding globulin
Comments: Plasma; very rarely used
Method 4: Radioimmunoassay (RIA)
a. DHEA-S: Dilution of plasma and direct RIA using iodinated or tritiated tracer and antiserum
b. DHEA: Extraction, chromatography, RIA using tritiated tracer and antiserum
Comments:
a. Plasma; in common use
b. Plasma; in common use
Method 5: Enzyme-linked immunosorbent assay (ELISA)
Principle of analysis: Solid-phase ELISA based on the principle of competitive binding.
Comments: Serum, plasma; most widely used assay.
485
Table 2: Comparison of Some Radioimmunoassay Methods for DHEA-S

Buster [6] Cattaneo [7] Smith [22] Haning [8]
1972 1972 1975 1981
Extraction No No Yes No
Recovery correction No No Yes No
Antiserum DHEA-3- DHEA-3-BSA DHEA- DHEA-BSA
hemisuccinate hemisuccinate
Isotope 3H-DHEA 3H-DHEA-S 3H-DHEA-S 3H-DHEA-S
(not S)
Incubation Overnight, 4C 2 hr, 4C 2 hr, room Overnight, 4C
4C temperature
Separation Charcoal-dextran Charcoal-dextran Ammonium Charcoal-dextran
sulfate
Cross-reactivity DHEA 100% A-S 50%; DHEA 129% EpiS 33%
A-dione 13% DHEA 25%; A-dione 30% A-S 20%
Andro 11%; A-diol 3% Epi 8%
EtioS 10%
Recovery Not stated Satisfactory 87% 95%
Interassay precision 18.8% 7.0% 12.3% 11.2%
Sensitivity 10 pg 50 pg 800 pg 35 pg
Normal range (g/dL): M, 199334 M, 169315 M and F, M, 230470
F, 82338 F, 150328 98456 F, 130350
A-diol, 5-Androstenediol; A-dione, 5-androstenedione; A-S, conjugated androstene; BSA, bovine serum albumin; DHEA,
dehydroepiandrosterone; Epi, epiandrosterone; EpiS, epiandrosterone sulfate; EtioS, etiocholanolone sulfate; F, female; M,
male; S, sulfate (conjugated).
Standard
Volume (mL) reaction
Procedure: DHEA-S by Radioimmunoassay
Standard of inter-of absolute tube
Principle (pg/mL) mediate B ethanol (pg/tube) (pmol/tube)
The plasma sample is diluted 300- to 2000-fold in 10,000 10.0 0 1000 2.71
buffer. For routine analysis, a dilution of 1 to 1000 is 5,000 5.0 5.0 500 1.36
2,500 2.5 7.5 250 0.68
convenient. The radioimmunoassay is then performed by 1,000 1.0 9.0 100 0.27
use of an antiserum to DHEA-3-hemisuccinate-BSA and 500 0.5 9.5 50 0.14
tritiated DHEA-S. Incubation is conducted at 4C for 2
hr. Separation of bound and free is achieved using a Pipet intermediate B according to the scheme above. For
charcoal-dextran mixture. One obtains a concentration of convenience, use 10.0 mL volumetric flasks. DHEA-S
DHEA-S in the specimen by comparing its observed standard solutions should be stored at 2C to 8C.
binding with that of standard DHEA-S. This procedure is Always allow ethanolic solutions to come to ambient
based on the method of Cattaneo.[7] temperature, and mix well before opening and use. Care
Reagents must be taken to avoid evaporation (and concentration)
1. Tris buffer, 0.1 mol/L, pH 7.4. Dissolve 12.1 of standard solutions. Properly stored, standard solutions
g of Tris (tris[hydroxymethyl]aminomethane) in 950 mL are stable for 6 months.
of distilled deionized water. Adjust pH to 7.4 with 6 M 5. DHEA-S antiserum. An antiserum is raised
HCl or 6 M NaOH, as needed. Adjust volume to 1000 using either DHEA-3-hemisuccinate or DHEA-3-BSA.
mL and mix. Stable for 2 months at 4C to 8C. This material is also commercially available from
2. DHEA-S standards. Weigh 20 mg of DHEA- Radioassay Systems Laboratory, Carson, CA. The
S (Sigma Chemical, St. Louis, or equivalent) and antiserum is diluted appropriately to achieve proper
dissolve in 150 mL of absolute ethanol. Swirl to binding characteristics so that maximum binding is 40%
dissolve, and bring to 200 mL. Label Stock DHEA-S to 50%, and the 50% intercept is about 300 pg per tube.
Standard. This solution contains 100 g/mL (0.27 This usually requires about a 1:1000 to 1:2000 dilution.
mmol/L). Aliquot and store at 20C.
3. Intermediate A. Dilute 1.0 mL of stock 6. Tritiated DHEA-S. Stock isotope can be
standard to 100 mL in absolute ethanol. This solution obtained from New England Nuclear (0.25 mCi, 0.0012
contains 1 g/mL (2.7 mol/L). mg in 0.25 mL of benzene).
4. Intermediate B. Dilute 1.0 mL of intermediate 7. Intermediate trace. Dilute stock isotope 1:10
A to 100 mL in absolute ethanol. This solution contains in absolute ethanol. This solution contains 96 ng/mL
10 ng/mL (or 10,000 pg/mL, or 27 nmol/L). Use this (0.26 nmol/L).
solution for preparing working standards. 8. Working trace. Dilute intermediate trace 1:50
Working standards in absolute ethanol. This solution contains 1.9 ng/mL (52
486
pmol/mL) and is equivalent to 96 pg/tube (0.26 following formula:

pmol/tube). Store all isotope solutions at 2C to 8C.
9. Charcoal-dextran solution. Weigh 250 mg of % B/B0 = Counts (standard, unknown) NSB 100%
washed Norite A (Amend Drug Co., Philadelphia, PA) Counts (zero standard) NSB
and 25 mg of Dextran T-70 (Pharmacia Fine Chemicals,
Piscataway, NJ). Add 100 mL of Tris buffer. Mix and 3. Plot a smooth curve through the standard points
store at 2C to 8C. Suspend well before use. using logit-log graph paper, placing %B/B0 on
10. Scintillation fluid. Mix 64 mL of Spectrofluor the ordinate and DHEA-S concentration on the
PPOPOPOP (Aldrich Chemicals, Milwaukee, WI) and abscissa.
25 mL of methanol with 1 L of toluene. 4. Calculate the final DHEA-S concentration
Assay using the following formula:
Equipment: Automatic liquid scintillation spectrometer
and vortex mixing apparatus. mg/L = [pg/tube from graph][107][1000 mL/L]__
1. Dilution of plasma samples and controls: Dilute [0.1 mL of plasma/tube][1,000,000 pg/mg]
0.1 mL of well-mixed plasma with 2.0 mL of
Tris buffer. Vortex mix. Then continue dilution The factor 1071 is the usual dilution factor. If a different
by pipetting 0.1 mL of diluted plasma and dilution protocol is used, use the appropriate factor.
adding 5.0 mL of Tris buffer. Vortex mix and 1 mg/L = 2.71 nmol/L
use this for the assay. The overall dilution is
1:1071. Procedure: DHEA by Radioimmunoassay
2. Remove working antiserum, isotope, and Principle
standards from storage, and allow to come to An aliquot of plasma is mixed with 3H-DHEA,
ambient temperature. Mix gently. incubated for 30 min, and then extracted with diethyl
3. Label duplicate 10 75 mm glass tubes for ether. After drying, the extract is dissolved in isooctane
zero, TC (total counts), NSB (nonspecific for chromatography. The DHEA is separated from other
binding), standards, and for each specimen and steroids by Celite partition chromatography. One then
control. performs radioimmunoassay using an antiserum to
4. Pipet 0.1 mL of absolute ethanol into the tubes DHEA-17-o-carboxymethyl-HSA and tritiated DHEA.
labeled zero and 0.1 mL of each working Incubation proceeds overnight at 4C. Separation of
standard into the appropriately labeled tubes. bound and free forms is achieved with charcoal-dextran.
Dry under nitrogen, taking care not to overdry. One can obtain the concentration of DHEA in the
5. Pipet 0.05 mL of working isotope into all tubes. specimen by comparing its observed binding with that of
Dry under nitrogen, taking care not to overdry. the DHEA standards. Correction is made on each
6. Pipet 0.1 mL of Tris buffer into the zero and all specimen for individual recovery losses. This procedure
standard tubes. Pipet 0.3 mL of Tris buffer into is based on the methods of Anderson [21].
TC and NSB tubes. Pipet 0.1 mL of diluted Reagents
plasma samples and controls into appropriately 1. Solvents. Spectroquality solvents (Merck,
labeled tubes. Vortex mix gently. Rahway, NJ), diethyl ether, isooctane, ethyl acetate,
7. Add 0.2 mL of diluted working antiserum to propylene glycol, and ethanol, are used without further
each tube except the NSB tubes. Vortex mix purification.
gently. 2. Chromatographic solutions:
8. Incubate all tubes at 4C for 2 hr. After 1. 10% ethyl acetate Add 10 mL
incubation, place tubes into an ice bath for 10 of ethyl acetate to 90 mL of isooctane.
min. 2. 15% ethyl acetate Add 15 mL
9. Add 0.2 mL of charcoal-dextran solution to all of ethyl acetate to 85 mL of isooctane.
tubes except TC. Maintain charcoal slurry using Solutions should be prepared fresh
an ice bath above a magnetic stirrer. Add slurry before each chromatography
rapidly, using automatic pipettes. Vortex mix procedure.
tubes immediately after addition of charcoal, 3. Celite 545, reagent grade. Obtained from
and continue incubation at 4C for 10 min. Alltech Associates, State College, PA, washed in
10. Centrifuge all tubes at 3000 g for 15 min. benzene, dried at 800C for 16 to 18 h, and kept dry at
11. Transfer 0.2 mL of each tube supernatant into 100C until use.
labeled counting vials, using an automatic 4. Phosphate-gelatin buffer (P-G buffer) (0.4
pipet. Add 10 mL of scintillation fluid, mix, and mol/L, pH 7.4, 0.1% gelatin). Weigh 1 g of gelatin
count. (Knox unflavored) and add to 500 mL of deionized or
Calculations distilled water. Heat to 60C to dissolve, and then cool.
1. Determine the average TC, NSB, and zero Add the following to the gelatin solution: 5.38 g of
standard counts; the last should be 40% to 50% NaH2PO4, 8.80 g of Na2HPO4, 9.00 g of NaCl, and
of TC.
2. Calculate %B/B0 binding for each standard, 1.00 g of NaN3 (all reagent grade). Stir to dissolve.
unknown, and control, according to the Adjust pH to 7.4 if necessary, using 1 M NaOH or 1 M
487
HCl. Adjust volume to 1 L and mix. Store at 2C to 8C. Equipment: Automatic liquid scintillation spectrometer.
Stable for 1 month after preparation. Extraction of Plasma
5. DHEA standards. Weigh 24 mg of DHEA 1. Add 0.1 mL of 3H-DHEA solution (spiking
(Sigma Chemical, St. Louis) and add to 150 mL of isotope) having about 5000 dpm to 0.5 mL of
absolute ethanol. Swirl to dissolve, and adjust volume to plasma. Vortex mix.
200 mL with ethanol. Label stock DHEA standard. 2. Incubate spiked plasma for 30 min at ambient
This solution contains 120 g/mL. Store at 2C to 8C temperature.
for 6 months. 3. Into separate liquid scintillation vials, pipet 0.1
6. Intermediate A. Dilute 10.0 mL of stock mL of spiking isotope (in duplicate). Add 0.4
standard to 100 mL in absolute ethanol. This solution mL of P-G buffer, and save for later counting.
contains 12 g/mL. Store at 2C to 8C for 6 months. These are total recovery tubes.
7. Intermediate B. Dilute 1.0 mL of intermediate 4. Add 10 mL of diethyl ether to each plasma
A to 100 mL in absolute ethanol. This solution contains sample, and vortex for 1 min.
120 ng/mL (120,000 pg/mL or 416 pmol/mL). Store at 5. Freeze extract on dry ice, and pour off ether.
2C to 8C for 6 months. 6. Dry extract, and redissolve in 1.5 mL of
isooctane.
Preparation of Working Standards
Standard in Preparation of Columns
Volume (mL) reaction
Standard P-G tube 1. Plug 5-mL siliconized pipets (Kimble) with a
(pg/mL) intermediate buffer (pg/tube) (pmol/tube) small ball of glass wool at the bottom, and
C 6000 0.5 mL of B 9.5 600 2.08 mount on a supporting rack. Use separate
D 3000 5.0 mL of C 5.0 300 1.04 columns for each specimen and control.
E 1500 5.0 mL of D 5.0 150 0.52
F 750 5.0 mL of E 5.0 75 0.26 2. Add a slurry of Celitepropylene glycol
G 375 5.0 mL of F 5.0 38 0.13 mixture equivalent to 0.75 g (about 0.4 mL) to
each pipet. Allow to drain.
For convenience in preparing working 3. Add 5 mL of isooctane to each column, and
standards, use 10-mL volumetric flasks. Prepare working allow to drain.
standards fresh before use and discard unused volumes. Sample Chromatography
Care must be taken in handling of ethanolic 1. The extract from ether extraction is layered on
standard solutions. Always allow ethanol to come to top of the Celite column and allowed to drain.
ambient temperature, and mix well before opening and 2. The column is washed with 5 mL of isooctane.
use. Care must be taken to avoid evaporation (and 3. Add 5 mL of 10% ethyl acetateisooctane to
concentration) of standard solutions. each column, allow to drain, and discard eluate.
DHEA antiserum. An antiserum is raised 4. Add 5 mL of 15% ethyl acetateisooctane to
using DHEA-17-o- each column, allow to drain, and collect and dry
carboxymethyl-HSA. This antiserum is commercially the eluate.
available from Radioassay Systems Laboratory, Carson, Radioimmunoassay
CA. The antiserum is diluted appropriately to achieve 1. Redissolve the dried chromatography eluates in
proper binding characteristics with a maximum binding 4 mL of P-G buffer, vortex, and heat at 50C
of 40% to 50% of total counts and a %B/B0 equaling for 5 min. Vortex again after letting it stand for
50% at about 200 pg/tube. The dilution is typically 10 min at ambient temperature.
1:2500. Aliquot and store at 20C. 2. Pipet duplicate 0.5-mL aliquots of each
8. Tritiated DHEA. Stock isotope can be reconstituted eluate into liquid scintillation vials
obtained from New England Nuclear. Stock solutions (1 for recovery determination. These are sample
mCi/mL) are stable at 4C for 2 months. Intermediate recovery counts tubes.
aliquots are diluted in absolute ethanol to achieve 12.5 3. Label 10 75 mm glass tubes in duplicate for
Ci/mL and are stored at 4C. Prepare working trace TC, NSB, standards, samples, and controls.
solutions just before use by diluting intermediate trace in 4. Pipet two 0.5-mL and two 0.2-mL aliquots of
buffer to a level of 5000 disintegrations per minute P-G buffer eluates in P-G buffer into labeled 10
(dpm) per 0.1 mL. Discard unused working trace after 75 mm tubes.
each preparation. 5. Pipet 0.1 mL of working standards into labeled
9. Celite/propylene glycol mixture. Add 200 g standard tubes.
of Celite to 100 mL of propylene glycol. Slurry well 6. Add 1.1 mL of P-G buffer to TC tubes, 0.6 mL
before use. to NSB tubes, 0.4 mL to standard tubes, and 0.3
10. Charcoal-dextran solution. Weigh 250 mg of mL to sample tubes that contain only 0.2 mL of
washed Norite A (Amend Drug Co. Irvington, NJ) and eluate. Vortex gently.
25 mg of Dextran T-70 (Pharmacia Fine Chemicals, 7. Add 0.1 mL of working tracer to each tube.
Piscataway, NJ). Add 100 mL of P-G buffer. Mix and 8. Add 0.1 mL of working antiserum to each tube
store at 2C to 8C. Slurry well before dispensing. except TC and NSB. Vortex tubes gently.
11. Scintillation fluid. Mix 64 mL of Spectrofluor 9. Incubate tubes at 37C for 30 min and then
PPOPOPOP and 25 mL of methanol with 1 L of toluene. overnight at 4C.
Assay
488
10. Add 0.5 mL of charcoal-dextran solution to all Procedure: DHEA-S by ELISA

tubes except TC. Maintain charcoal slurry using 1. Secure the desired number of microtiter wells in the
an ice bath above a magnetic stirrer. Add slurry holder.
rapidly using an automatic pipettor. Vortex mix 2. Dispense 25 L of each standard, controls, and
tubes immediately after addition of charcoal, samples with new disposable tips into appropriate wells.
and continue incubation for 10 minutes at 4C. 3. Dispense 200 L enzyme conjugate into each well.
11. Centrifuge all tubes at 3000 g for 15 min. 4. Thoroughly mix for 10 seconds. It is important to
12. Transfer 0.5 mL of each tube supernate into have a complete mixing in this step.
labeled counting vials using an automatic pipet. 5. Incubate for 60 minutes at room temperature without
Add 10 mL of scintillation fluid, mix, and covering the plate.
count. Also at this time, count the total recovery 6. Briskly shake out the contents of the wells.
and sample recovery vials for each sample. Rinse the wells 3 times with diluted sash Solution (400
L per well). Strike the wells sharply on absorbent paper
Calculations to remove residual droplets.
1. Determine the average TC, NSB, zero standard IMPORTANT NOTE:
counts, total recovery, and sample recovery. The sensitivity and precision of this assay is markedly
2. Calculate B/B0 binding for each standard, influenced by the correct performance of the washing
unknown, and control, according to the procedure!
following formula: 7. Add 100 L of substrate solution to each well.
8. Incubate for 15 minutes at room temperature.
% B/B0 = Counts (standard, unknown) NSB 100% 9. Stop the enzymatic reaction by adding 50 L of stop
Counts (zero standard) NSB solution to each well.
10. Read the absorbance at 450 10 nm with a
3. Plot a smooth curve through the standard points microtiter-plate reader within 10 minutes after adding
using logit-log graph paper, placing %B/B0 on the stop solution.
the ordinate and DHEA concentration on the
abscissa. Calculation of Results
4. Calculate the final DHEA concentration using 1. Calculate the average absorbance values for each set
the following formula: of standards, controls, and patient samples.
2. Construct a standard curve by plotting the mean
ng/mL = [pg/tube from graph][Dilution factor]______ absorbance obtained from each standard against its
[0.5 mL of plasma][Fract. rec.][1000 pg/ng] concentration with absorbance value on the vertical (Y)
axis and concentration on the horizontal (X) axis.
where the dilution factor is 4.0/0.5 = 8, or 4.0/0.2 = 20, 3. Using the mean absorbance value for each sample,
depending on volume of redissolved eluate used determine the corresponding concentration from the
Fract. rec. is fractional recovery, defined as standard curve.
4. Automated method: Computer programs using cubic
Average sample recovery counts spline, 4 PL (4 Parameter Logistics) or Logit-Log can
Average total recovery counts generally give a good fit.
5. The concentration of the samples can be read directly
and 0.5 mL is volume of plasma extracted: from this standard curve. Samples with concentrations
DHEA 1 ng/mL = 3.47 nmol/L higher than that of the highest standard have to be
further diluted. For the calculation of the concentrations,
Notes this dilution factor has to be taken into account.
1. Any residue left from a contaminated solvent
will potentially interfere with binding in this A typical example of a standard curve for the DHEA-S
procedure. Recoveries less than 70% indicate ELISA:
deficiencies in the extraction-chromatography Standard Optical Units (450 nm)
procedure. Standard 0 (0 g/dL)
2. Each Celitepropylene glycol system should be Standard 1 (10 g/dL)
calibrated with 15% ethyl acetateisooctane Standard 2 (50 g/dL)
and radiolabeled DHEA to ensure proper Standard 3 (100 g/dL)
elution of peak. Standard 4 (250 g/dL)
3. Care must be taken in handling ether so that no Standard 5 (500 g/dL)
peroxides are present. Standard 5 (1000 g/dL)
4. Cross-reactivity of the antiserum should be
characterized, especially toward other
conjugates of DHEA. Chromatography,
however, eliminates most of any cross-reacting
steroid present.
489
Digoxin and Digitoxin

Randal J. Schneider
Name: Digoxin and digitoxin
Molecular formula: C41H64O14 (- and -acetyldigitoxin, digoxin)
C41H64O13 (digitoxin)
Molecular mass: 780.97 D and 764.92 D, respectively
Merck Index: 3148, 3143
Chemical class: Cardiac glycoside
O
Digoxin
O Lactone Ring
OH CH3
H
CH3 H
Steroid Nucleus
H
H OH
O
CH3 H
Sugar Residues
3
O OH
Structure: H
Principles of Analysis and Current Usage used to control cardiac arrhythmias through
improvement of the strength of myocardial contraction.
The term digitalis is used to represent all steroid However, intoxication is a frequent complication of
glycoside compounds sharing common features of digitalis therapy and can be life threatening, frequently
chemical structure (a steroid nucleus with an unsaturated requiring stat analysis for digitalis levels and in severe
lactone ring at the C17 position and one or more sugar cases, administration of the digoxin-specific antibody
residues at C3) and common effects on cardiac activity. fragments (Digibind, DigiTAb). Determination of
Although digoxin and digitoxin are the two digitalis digitalis concentrations in blood has been helpful in the
preparations used in the United States, digoxin is more diagnosis of toxicity and the establishment of an optimal
frequently prescribed. dosage, because a reasonably close correlation exists
between blood and tissue concentrations of digitalis.
Digoxin and digitoxin are obtained in the crystalline
form from the leaves of Digitalis lantana and Digitalis Many techniques have been used to determine blood
purpurea, respectively. Digoxin is sparingly soluble in levels of digoxin. They include gas chromatography [1]
water, chloroform, ether, ethyl acetate, and acetone, but (Table 1, Method 1), methods based on inhibition by
soluble in dilute (50%) alcohol or pyridine, whereas digoxin of sodium-potassium-activated adenosine
digitoxin, a nonpolar molecule (structure is that of triphosphatase (ATPase) in red blood cells [2] or in
digoxin without the C12-OH group), is relatively more microsomes [3] (Table 1, Method 2), and an isotopic
soluble in organic solvents. Digoxin and digitoxin are displacement method using a sodium/potassium-
activated ATPase (Table 1, Method 3) [4]. However,
these methods are relatively tedious and have been
i
Digoxin and Digitoxin largely replaced by more practical and sensitive
Previous and current authors of this method: immunoassay techniques.
First edition: I-Wen Chen, Linda A. Heminger
Methods edition: I-Wen Chen, Linda A. Heminger Digoxin was one of the first hapten-like small molecules
Second edition: I-Wen Chen, Linda A. Heminger to be analyzed by the RIA technique (Table 1, Method 4)
Third edition: I-Wen Chen, Linda A. Heminger [5]. Digoxin antisera were successfully raised in animals
Fourth edition: I-Wen Chen, Linda A. Heminger immunized with digoxin coupled to a variety of
Fifth edition: Randal J. Schneider
490
macromolecules, including poly-l-lysine, human serum bound digoxin to be eluted from the column. The
albumin, and bovine serum albumin. Unlike thyroid column eluate is mixed with the substrate tablet
hormones, a digoxin molecule contains neither iodine (containing o-nitrophenyl--d-galactopyranoside), and
nor any functional groups that can be iodinated. Tritiated the enzyme activity is quantified by measurement of the
digoxin preparations were initially the only radioligands rate of change of absorbance at 405 nm. The -
available for digoxin RIA. Tritiated radioligands are galactosidase activity present in this fraction is
beta-particle emitters and must be counted by means of proportional to the digoxin concentration in the serum.
liquid scintillation counters. This is expensive and
inconvenient in comparison to the counting of gamma- A fluorescent probe has also been used as an alternative
ray emitters such as iodine (125I) by a crystal to radioactive labels in the determination of digoxin. In
scintillation well counter. This problem was overcome the homogeneous fluorescence polarization
by the development of techniques that attach a functional immunoassay developed for digoxin, the label is a
group that can be easily iodinated to a digoxin molecule. fluorescent dye (fluorescein) (Table 1, Method 7). When
Tyrosine, histamine, and tyramine have been coupled to the digoxin-fluorescein complex is excited by a
a digoxin or digoxigenin molecule for radioiodination. polarized beam of light, it emits polarized fluorescent
An example of a derivative formed in this way is 3-O- light. The degree of polarization depends on the extent
succinyl digoxigenin tyrosine. When this compound is of rotation of the molecule during the period between
radioiodinated, a monoiodinated and di-iodinated excitation and emission (rotational relaxation time). The
derivative can be obtained. It is important that the rotational relaxation time is proportional to the size of
radioiodinated 3-O-succinyl digoxigenin tyrosine used in the molecule; the larger the molecule, the longer the
a digoxin RIA be free of the di-iodinated derivative, rotational relaxation time and the larger the degree of
since the di-iodinated derivative is known to bind to fluorescence polarization. Therefore, in the fluorescence
thyroxine-binding globulins [6]. Nearly every separation polarization immunoassay, the quantity of antibody-
technique used in RIAs has been applied successfully to bound digoxin-fluorescein complex (a much larger
the RIA of digoxin. molecule than the unbound complex) present can be
determined, in the presence of the unbound complex, by
Other types of labels have been developed as alternatives measurement of the increase in the degree of
to radionuclides. In the homogeneous enzyme-multiplied polarization of the emitted fluorescence.
immunoassay technique (EMIT) (Table 1, Method 5),
the enzyme glucose-6-phosphate dehydrogenase is More recently, assays based on drug-coated
chemically coupled to digoxin and used as a label. When microparticles have been developed. The particle-
the enzymedigoxin conjugate is bound to the digoxin enhanced turbidimetric immunoassay method (PENTIA,
antibody, the enzyme is inactivated. In the presence of Table 1, Method 8) is based on competition between
digoxin in a sample, some of the enzymedigoxin exogenous drug in the specimen and digoxin-coated
conjugate does not bind to the digoxin antibody, and the microparticles for anti-digoxin antibody. The rate of
enzyme is active. Therefore, one can monitor the extent absorbance change is directly proportional to the rate at
of digoxin-antibody bindingand thus the digoxin which the microparticles agglutinate in the presence or
concentrationby measuring the enzyme activity absence of digoxin in the specimen. Other microparticle-
spectrophotometrically without physically separating the based assays include latex (Table 1, Method 9),
bound and free fractions. Although the EMIT was magnetic enzyme immunoassay (Table 1, Method 10),
developed originally as a manual assay system, it has and magnetic microparticles (Table 1, Method 11). In
been successfully adapted to automated chemistry these assays, the microparticles serve as the solid phase
analyzers. for capture of the antigen of interest, followed by a
detection method such as chemiluminescence or
A sensitive affinity columnmediated immunometric fluorescence.
assay (Table 1, Method 6) has also been developed [7].
In this assay, the F(ab)2 fragment of the digoxin All methods described in Table 1 are also applicable to
(and for some, developed for) the measurement of
antibody is labeled with -galactosidase. An excess
circulating digitoxin [3,8,9]. As with digoxin, digitoxin
amount of this enzymeantibody conjugate is incubated
is determined in the clinical laboratory almost
with the sample so that digoxin contained in the sample
exclusively by immunoassay. However, only a few
can be rapidly and quantitatively bound to the antibody.
commercial immunoassay kits are currently available
To accomplish this, equal volumes of serum sample and
because digitoxin analyses are infrequently requested,
reagent containing -galactosidaseantibody conjugate
owing to the lower frequency of digitoxin prescribing as
are mixed and incubated manually for 10 minutes to 2
compared to digoxin.
hours at room temperature. The sample cup is then
introduced into the instrument along with a digoxin test
A review of the laboratory methods reported in the
pack containing ouabain (digoxin analog)-bovine serum
College of American Pathologists (CAP) Proficiency
albuminSephadex G-10 affinity column positioned in
Summary Report (2007 CAP Z-B) indicates more than
the pack head and a substrate tablet. The system then
half of the methodologies include microparticle
automatically pumps the sample through the column,
technologies. The majority of the remaining methods are
allowing only the enzymeantibody conjugate with the
491
based on enzymes, fluorescence, and specimen is not to be tested within 24 hours, it should
chemiluminescence. preferably be stored frozen.
Gas chromatography, ATPase inhibition, and isotope- Time of blood collection is an important factor in
displacement methods are described briefly in Table 1 determining digoxin toxicity. Serum digoxin levels will
for historic interest or for specialized use. rise sharply during the first hour after an oral dose,
indicating post-absorption uptake. This is followed by a
Reference and Preferred Methods sharp decrease as digoxin is taken up by the myocardium
Recently a candidate reference method for digoxin and and other tissues. Usually 4 to 6 hours are needed for
digitoxin was developed and implemented by the serum and tissue stores of digoxin to reach equilibrium.
Institute for Standardization and Documentation in the Thereafter, serum digoxin levels tend to stabilize and
Medical Laboratory (INSTAND). The method employs decrease very slowly over the next 24 to 40 hours.
high-pressure liquid chromatography with isotope- Clinically, the time of peak tissue (not plasma)
dilution tandem mass spectroscopy (LC-IDMS/MS) concentrations is used to accurately correlate plasma and
[10]. The method is used to set target values for tissue concentrations. One obtains the greatest diagnostic
materials used in external quality-assessment surveys. accuracy by drawing the blood sample at a standard time
Prior to the development of the IDMS/MS method, an during the stable phase, when serum digoxin levels
interlaboratory transferability study for digoxin assays reflect the average cardiac concentrations.
sponsored by the Centers for Disease Control, the Recommendations are to collect specimens 8 hours or
National Committee on Clinical Laboratory Standards, more after the dose.
and the American Association for Clinical Chemistry
developed a candidate reference method. The RIA Digoxin concentrations cannot be accurately determined
method was based on tritium-labeled digoxin as the in a patient who is being switched from digitoxin to
radioligand and charcoal as the separation agent [11]. digoxin therapy, since both drugs may be present in the
serum. The digoxin antisera used in most commercial
Although most commercial kits are well-characterized digoxin immunoassay kits cross-react significantly with
systems, their performance characteristics should be digitoxin (2% to 6% cross-reactivity, with some as high
evaluated carefully in each laboratory before they are as 40%). The problem of cross-reaction is further
selected for routine use. Immunoassay methods for exacerbated by the fact that therapeutic levels of
digoxin have been fully automated. The sensitivity of the digitoxin are about 10 times higher than those of
EMIT assay is relatively low, primarily because of the digoxin. The digitoxin antisera also cross-react with
steric hindrance introduced into the antigen-antibody digoxin (2% to 4% cross-reactivity), but because of the
reaction by the presence of the enzyme macromolecule higher therapeutic levels, the interference by digoxin is
and also because of the need to differentiate the less significant in the digitoxin immunoassay.
antibody-enzyme complex from the unbound enzyme-
substrate complex (bound from unbound) and the Interferences
competitive mode of the assay (competition between The presence of digoxin-like immunoreactive substances
labeled and unlabeled antigens for the limited number of (DLIS) in human plasma and other body fluids can
antibody-binding sites). Some of the limitations have cross-react with anti-digoxin antibodies, causing falsely
been circumvented by the affinity columnmediated increased digoxin concentrations [13]. Increased DLIS
immunometric assay and by the homogeneous concentrations have been found in patients with
fluorescence polarization immunoassay. Non-isotopic conditions typically characterized by volume expansion,
immunoassays have been semiautomated. Pretreatment such as uremia, essential hypertension, liver disease, and
steps may be required, because endogenous fluorescent pregnancy. Exogenous DLIS have been found in serum
compounds bound to serum proteins interfere with the following ingestion of various Chinese medicines, and
assay. It has been shown that digoxin binds to the protein following therapy with spironolactone. FPIA procedures
pellet during the trichloroacetic acid precipitation step, generally exhibit the greatest degree of interference.
and thus falsely high or low values may be expected Falsely decreased digoxin results due to the presence of
with patient specimens having abnormally low or high DLIS have been described. Negative interference in
protein contents [12]. Although the non-isotopic reagents digoxin measurements have been described in the MEIA
are considerably more expensive than those of RIA, they procedures. Positive and negative interference due to the
are attractive alternatives to RIA in laboratories that presence of DLIS can be overcome by measuring free
have such units already in use. The assay precision of digoxin rather than total digoxin concentrations. The
these two systems compared favorably with that of RIA. presence of hemoglobin, lipemia, or icterus usually does
not interfere with the commonly used immunoassay
Specimen procedures unless these compounds are in extremely
In most instances, either serum or plasma may be used increased amounts [14].
for the assay, though some commercial manufacturers
specifically indicate that plasma is not to be used with Digoxin and Digitoxin Therapeutic Range
their assay kits. Digoxin in serum is quite stable, with no As with all diagnostic tests, it is advisable for the
appreciable loss of digoxin observed in serum stored at individual laboratory to accumulate and analyze its own
room temperature for up to 2 weeks. However, if the data to establish a range of toxic and nontoxic values;
492
differences in assay techniques may affect assay results. of body weight. The higher doses are necessary because
One study found that 90% of nontoxic patients had of the greater ratio of cardiac mass to total body mass in
serum digoxin levels below 2 ng/mL (2.56 nmol/L), children. These higher doses are reflected in higher
whereas 86% of the toxic patients had levels greater than serum digitalis concentrations in children.
2 ng/mL [15].
Digitalis toxicity occurs frequently and can be life
Beller et al. [16] reported a mean serum digitoxin threatening. Common clinical symptoms of a digitalis
concentration of 20 11 ng/mL (26.1 nmol/L), with a overdose are anorexia, nausea and vomiting, blurred
range of 9 to 32 ng/mL (11.8 to 41.8 nmol/L), for vision, flickering objects of yellow and green color, and
nontoxic patients on oral doses of 0.07 to 0.15 mg/day disorientation. The most common important myocardial
(0.89 to 1.96 mol/day) and a mean concentration of 34 manifestation of digitalis toxicity is cardiac arrhythmia.
18 ng/mL (44.4 nmol/L), with a range of 16 to 52
ng/mL (20.9 to 67.9 nmol/L), for toxic patients on doses Clinical diagnosis of digitalis toxicity is often difficult in
of 0.10 mg/day (1.31 mol/day). patients with serious heart disease, because most of its
manifestations could be caused by either the disease or
It is generally understood that infants are more resistant the drug. Immunoassays of digitalis were initially
to cardiac glycosides than adults are. OMalley et al. developed for diagnosis and avoidance of digitalis
found that a mean nontoxic digoxin level of 3.8 1.2 toxicity. Mean serum digitalis concentrations in patients
ng/mL (4.87 nmol/L) for eight infants younger than 1 with drug toxicity (arrhythmia) are about two to three
month was significantly higher than 1.4 1.1 ng/mL times higher than those in patients receiving
(1.79 nmol/L) for 16 adults, but the digoxin therapeutically appropriate doses. Although this
concentration in infants older than 1 month (1.4 0.5 difference is statistically significant in most studies,
ng/mL, n = 5) was no higher on average than that of the substantial overlap exists between groups with and
adult controls [17]. without toxicity. Therefore, knowledge of high serum
digitalis levels is helpful in the clinical recognition of
Interpretation digitalis toxicity, but under no circumstances can it be
Digitalis is a frequently prescribed drug in clinical used as the sole basis for determining the presence of
practice in the United States. Because of its ability to drug toxicity in place of sound clinical judgment. Serum
increase the strength of cardiac contraction, digitalis digitalis levels are especially helpful in the interpretation
preparations have been used medically for the control of of arrhythmias. In the presence of high digitalis levels,
supraventricular tachycardias and the treatment of arrhythmias by digitalis toxicity are predicted, whereas
congestive heart failure. Oral administration is the least arrhythmias in patients with low serum digitalis levels
expensive and the safest route of digitalis administration. are probably related to the underlying disease.
The intravenous route is used only when conditions such Furthermore, hypokalemia can precipitate digitalis
as coma and vomiting prevent oral administration or toxicity because of the combined effects an elevated
when rapid onset of digitalis action is imperative, as in digoxin with a low potassium will have on normal
patients with pulmonary edema from acute left- cardiac rhythm. The digitalis assay can also be used to
ventricular failure. Digoxin, the most commonly used detect noncompliant patients, especially when they are
digitalis preparation in the United States, has an average known to have had adequate digitalis levels in the past.
excretory half-life of about 35 hours. Since digoxin is
eliminated primarily by renal excretion, patients with Fab fragments of digoxin-specific antibodies have been
reduced renal function show a longer plasma half-life used in the management of advanced, life-threatening
as long as 5 days in patients who are essentially digitalis toxicity unresponsive to conventional
anephric. In contrast, digitoxin is more than 90% bound therapeutic modalities [18]. The use of Fab fragments
to serum albumin, has a half-life averaging 5 to 9 days, permits glomerular filtration and relatively rapid renal
and is relatively uninfluenced by impaired renal excretion of antibody-bound digoxin, resulting in
function. progressive removal of digoxin from receptor sites and
rapid reversal of advanced digoxin toxicity without signs
Digitalis is given until a therapeutic effect is achieved of adverse effects. Because of the presence of digoxin-
(such as relief of cardiac failure) or until the earliest specific antibodies, serum samples obtained from
toxic effects (such as anorexia or arrhythmia) appear. patients receiving antibody treatment cannot be assayed
The typical initial oral dose of digoxin in adults is 0.75 directly. They must be either deproteinized or subjected
to 1.5 mg. This is followed by the administration of 0.25 to equilibrium dialysis before the assay. Total and free
to 0.5 mg every 6 hours until adequate digitalization is digoxin concentrations can be obtained from the
achieved. The oral maintenance dose ranges from 0.0625 deproteinized and equilibrated dialysis samples,
to 0.75 mg daily. The initial oral dose of digitoxin of 1.0 respectively. It has been reported that free (active)
to 2.5 mg is administered in several equal aliquots every digoxin serum concentrations decreased to undetectable
6 hours. The dose should be administered within 36 to levels within minutes after administration of the digoxin-
40 hours. The oral maintenance dose ranges from 0.005 specific Fab fragments, whereas the total digoxin serum
to 0.2 mg daily in divided doses. Digitalis doses in concentrations decreased more gradually. Immunoassays
children are about 50% larger than those used in adults that measure total digoxin levels vary in their cross-
when calculated on the basis of milligrams per kilogram reactivities to digoxin-specific Fab fragments, and
493
laboratories should be aware of this limitation for 8. Lukas DA, Peterson RE. Double isotope
potentially overestimating digoxin levels generated after dilution derivative assay of digitoxin in plasma,
the administration of the Fab antidote. Some laboratories urine, and stool of patients maintained on the
have validated ultra-filtration systems for separating free drug. J Clin Invest 1966;45:782-95.
from Fab-bound digoxin to provide healthcare 9. Oliver GC Jr, Parker BM, Brasfield DL, Parker
professionals with a more accurate bioavailable serum CW. The measurement of digitoxin in human
digoxin level. serum by radioimmunoassay. J Clin Invest
1968;47:1035-42.
Digoxin Performance Goals 10. Kaiser P, Akerboom T, Wood WG, Reinauer H.
Survey data from the 2007 CAP Participant Summary A novel LC-IDMS/MS method for the
Report show imprecision values (% coefficient of determination of the cardiac glycosides digoxin
variation) for digoxin measurements range from 4.6% to and digitoxin using caesium adducts. Clin Lab
14.1% at a mean digoxin concentration of approximately 2006;52:37-42.
1.24 ng/mL. Acceptable performance criteria (Clinical 11. Fast DM, Hannon WH, Burtis CA, Bayse DD.
Laboratory Improvement Amendments, 1988) for An interlaboratory study of the determination of
measurement of digoxin and free digoxin require that digoxin by immunoassay. Clin Chem
laboratories be accurate to within the greater of 20% or 1980;26:480-6.
0.2 ng/mL and 3S.D. or 10% of the peer-group 12. Porter WH, Haver VM, Bush BA. Effect of
mean, respectively. Desirable specifications for protein concentration on the determination of
analytical imprecision derived from studies of biological digoxin in serum by fluorescence polarization
variation indicate an assay SD of 0.03 and 0.08 ng/mL at immunoassay. Clin Chem 1984;30:1826-9.
concentrations of 0.8 and 2.0 ng/mL [19]. 13. Dasgupta A. Endogenous and exogenous
References digoxin-like immunoreactive substances:
1. Watson E, Kalman SM. Assay of digoxin in impact on therapeutic drug monitoring of
plasma by gas chromatography. J Chromatogr digoxin. Am J Clin Pathol 2002;118:132-40.
1971;56:209-18. 14. Miller JJ, Straub RW Jr, Valdes R Jr. Analytical
2. Bertler A, Redfors A. An improved method of performance of a monoclonal digoxin assay
estimating digoxin in human plasma. Clin with increased specificity on the ACS:180. Ther
Pharmacol Ther 1970;11:665-73. Drug Monit 1996;18:65-72.
3. Burnett GH, Conklin RL. The enzymatic assay 15. Park HM, Chen IW, Manitasas GT, Lowey A,
of plasma digoxin. J Lab Clin Med Saenger EL. Clinical evaluation of
1971;78:779-84. radioimmunoassay of digoxin. J Nucl Med
4. Brooker G, Jelliffe RW. Serum cardiac 1973;14:531-3.
glycoside assay based upon displacement of 3 16. Beller GA, Smith TW, Abelmann WH, Haber
H-ouabain from Na-K ATPase. Circulation E, Hood WB Jr. Digitalis intoxication. A
1972;45:20-36. prospective clinical study with serum level
5. Butler VP Jr, Chen JP. Digoxin-specific correlations. N Engl J Med 1971;284:989-97.
antibodies. Proc Natl Acad Sci U S A 17. OMalley K, Coleman EN, Doig WB,
1967;57:71-8. Stevenson IH. Plasma digoxin levels in infants.
6. Painter K, Vader CR. Interference of iodine- Arch Dis Child 1973;48:55-7.
125 ligands in radioimmunoassay: evidence 18. Wenger TL, Butler VP Jr, Haber E, Smith TW.
implicating thyroxine-binding globulin. Clin Treatment of 63 severely digitalis-toxic patients
Chem 1979;25:797-9. with digoxin-specific antibody fragments. J Am
7. Leflar CC, Freytag JW, Powell LM, Strahan JC, Coll Cardiol 1985;5(5 Suppl A):118A-123A.
Wadsley JJ, Tyler CA et al. An automated, 19. Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
affinity-column-mediated, enzyme-linked Hernandez A, Jimenez CV et al. Current
immunometric assay for digoxin on the Du Pont databases on biological variation: pros, cons
aca discrete clinical analyzer. Clin Chem and progress. Scand J Clin Lab Invest
1984;30:1809-11. 1999;59:491-500.
494
Tables
Table 1: Methods of Digoxin Analysis

Method 1: Gas chromatography
Principle of analysis: Physicochemical separation with 3H-digoxin as internal standard
Comments: Plasma; may be applied to other biological specimen; requires prepurification by thin-layer
chromatography and derivatization; tedious and time consuming
Method 2: ATPase inhibition
Principle of analysis: Dose-dependent inhibition of Na- and K-activated ATPase
Comments: Plasma; may be applied to other biological specimens; requires extraction; tedious and time
consuming; less precise and less specific than RIA
Method 3: Isotope displacement
Principle of analysis: Displacement of 3H-ouabain from purified Na-K ATPase preparation by digoxin
Comments: Plasma; may be applied to other biological specimens; requires extraction and beta counting; tedious,
nonspecific
Method 4: Radioimmunoassay
Principle of analysis: Competitive binding of 3H- or 125I-labeled and unlabeled digoxin with digoxin antibody
Comments: Plasma or serum; requires phase separation and beta-particle or gamma-ray counting; fully
automated; most sensitive and precise
Method 5: Enzyme-multiplied immunoassay techniques
Principle of analysis: Competitive binding for antibody between endogenous digoxin and enzyme-labeled
digoxin. Antibody binding to enzyme-digoxin complex inhibits enzymatic activity.
Comments: Serum; lacks sensitivity; requires sample pretreatment
Method 6: Enzyme immunometric assay
Principle of analysis: Quantitative binding of digoxin by excess amount of enzyme-labeled antibody
Comments: Plasma or serum; requires no sample pretreatment but requires phase separation by affinity column;
system semiautomated
Method 7: Fluorescence polarization immunoassay
Principle of analysis: Competitive binding of fluorescein-labeled and unlabeled digoxin with digoxin antibody
Comments: Plasma or serum; requires sample pretreatment but requires no phase separation; quantitation
depends on degree of polarization of emitted fluorescence; system semiautomated
Method 8: Particle-enhanced turbidimetric immunoassay
Principle of analysis: Competitive binding for antibody between endogenous drug and drug coated onto
microparticles. Antibody binding of digoxin in the specimen to microparticles slows the agglutination process that
is measured photometrically.
Method 9: Latex microparticle agglutination
Principle of analysis: Competition between digoxin and digoxin-labeled latex microparticles for digoxin-
specific antibodies. The level of agglutination is measured turbidimetrically.
Method 10: Microparticle enzyme immunoassay
Principle of analysis: Digoxin in the specimen is sandwiched between bound antibodies to microparticles and
digoxin-specific, enzyme-labeled antibodies. Digoxin is proportional to the fluorescence generated from the
enzyme-substrate interaction.
Method 11: Magnetic particle immunoassay
Principle of analysis: Antibody conjugate reagent mixing with digoxin in the specimen. Bound digoxin is
separated from excess antibody conjugate by magnetic microparticles coated with the digoxin analog, ouabain.
The remaining supernatant containing the bound digoxin-antibody-enzyme complex is mixed with substrate.
Enzymatic conversion of the substrate is proportional to the amount of digoxin the specimen.
495
Drug Screens
Drug Screens
Christopher R. McCudden
Name: Drug screens

to detect the drug. Screening methods typically use

Principles of Analysis and Current Usage analytical techniques that give either a positive or
negative finding, whereas confirmation tests may be
either qualitative or quantitative. Numerous possible
Drug screens are a general term for analysis of toxic
analytical permutations exist for the screening-
and controlled substances. These tests are known by a confirmation format (see Table 1).
variety of names such as coma panel, toxic screen,
complete drug screen, comprehensive screen, drug-
Most drug screens fall into one of two classes: (1) a
abuse screen, stimulant panel, or overdose panel. specific screen for a particular known or suspected drug
Despite the fact that these test names suggest an (or class of drug), or (2) a comprehensive screen, with
exhaustive list of compounds, drug screens are analysis of as many drugs as are technically and
individually defined by each laboratory and may only economically feasible for the laboratory to measure.
contain a handful of commonly encountered drugs.
Searching for one particular compound is much easier
There are many different types of qualitative drug than screening for a large number of possible
screens, each developed with that particular laboratorys compounds. If a single known or suspected drug is of
interests, finances, and experience in mind. There are
interest, the screening test can be optimized to ensure
various types of urine screening tests used which rely on adequate specificity and sensitivity. If the screen is
spot tests, immunoassays, or chromatography. Whereas broadened to include several compounds, a series of
immunoassays are directed toward specific drugs or drug
specific analyses for each compound can be combined
groups, chromatographic systems such as TLC and GC- together to form a limited or panel type of screen. In
MS can detect a much wider range of compounds and this format, tests are performed on each compound using
metabolites.
a procedure that optimizes sensitivity. If the number of
compounds is increased further, the time required for
To define a specific drug screen, one must consider the performing specific individual assays for a large group
reasons for screening. Areas in which drug screening is
of drugs will eventually limit the number of drugs that
often indicated include (1) clinical toxicology, (2) can be included in the screen. When it is necessary to
forensic/medicolegal toxicology, (3) employee-screening expand the screen to include many substances, less-
programs, (4) drug trials, and (5) drug-abuse programs.
specific assays that enable the analyst to look for a vast
Although screening for forensic purposes is still largely array of compounds at a single glance may be employed.
the domain of the coroners laboratory, other drug- The actual number of drugs included in this expanding
screening programs have been adopted by the clinical
screening process varies from laboratory to laboratory
laboratories with special emphasis on clinical toxicology and depends heavily on the experience of the analyst
(the overdose patient) and drug-abuse screening performing the assay.
programs.
When defining the screen in a particular laboratory, it is
Defining the Test
important to consider the clinical need to detect a given
Qualitative analysis for a drug or a group of drugs can
compound. In the case of drugs of abuse, certain regions
generally be divided into three basic steps: (1) isolation
of the country are more or less likely to encounter a
or separation of the compounds of interest from the
particular drug. An example of this is shown in Figure
biological specimen, (2) analysis of the drug content of
1A, where PCP is commonly encountered in parts of
the partially purified sample by a screening method, and
Texas but is less frequently seen in the Southeast and
(3) if positive, employment of one or more confirmatory
Northwest. In contrast (Figure 1B), cocaine is prevalent
methods that are different from the method used initially
in both Texas and throughout the Southeast. Historical
experience can be a valuable tool when deciding which
i tests to employ on a limited budget.
Drug screen
Once the laboratory has decided on the test menu, it
First edition: F. Michael Hassan
must gather the appropriate analytical data on these
Methods edition: F. Michael Hassan
drugs and their metabolites. The necessity of
Second edition: F. Michael Hassan
understanding metabolic pathways of drugs of interest
Third edition: F. Michael Hassan
cannot be overemphasized. As many drugs are present in
Fourth edition: F. Michael Hassan
urine only as metabolites, it is essential to have standards
Fifth edition: Christopher R. McCudden
496
Drug Screens
of both known metabolites and the parent drugs for of a specific drug (or class of drugs, see Table 3), which
accurate identification and confirmation. is indicated by the formation of a colorimetric reaction
product [5,8-10]. Although spot tests offer inexpensive
General Analytical Procedures testing with rapid turnaround times, they are relatively
A set of general references is provided below. These nonspecific and often require subjective interpretation.
detail not only general screening procedures but also
specific individual assay techniques and technologies Ultraviolet Spectroscopy
involved in toxicological analyses [1-6]. As with the spot tests, ultraviolet analysis is best
employed to test for a specific drug or confirm the
Extraction presence of compounds detected by other techniques.
With the exception of spot tests and immunoassays, most The ultraviolet scan has relatively poor specificity unless
procedures require significant preparation of the the unknown drug is isolated by other test procedures,
specimen before actual analysis. For the drug screen, such as thin-layer chromatography, or extracted by
isolation is usually accomplished by solvent extraction procedures that minimize spectral interferences from
procedures or by column chromatography, employing body fluids and from other drugs. As evident in Figure 2,
exchange resins or solid, bonded-phase stationary compounds may have overlapping or similar ultraviolet
phases. Solvent extraction of urine samples involves a spectra, necessitating further confirmatory techniques.
two-phase partitioning system in which the drug is Sources have compiled spectra of most commonly
separated from an aqueous phase and redistributed into a encountered drugs [5]. However, whereas ultraviolet
second phase, which is usually organic. This removes spectra are readily obtained, sample preparation time can
many of the interfering compounds present in the be lengthy (Table 2).
specimen. The degree of complexity of the extraction
depends on the type of analytical method used and the Immunochemical Methods
degree of freedom from interfering compounds required Immunochemical methods rely on antibodies and
of the processed specimen. various chemical, enzyme, and small-particle labels to
identify drugs. There are numerous different formats for
Since drugs are extracted from matrices according to antibody-based drug detection that have become popular
their ionization state, partitioning of the specimen because of their rapid turnaround times, accuracy, and
constituents into acidic, neutral, and basic fractions is ease of use. Currently available formats include enzyme-
possible. Enhanced extraction efficiencies are obtained multiplied immunoassay technique (EMIT), cloned-
by preparation of extractions from specimens made enzyme donor immunoassay (CEDIA), colloid metal
acidic or alkaline to yield higher recoveries of drug. immunoassay (CMI), fluorescence immunoassay,
When shortened turnaround times are a necessity, single fluorescence polarization immunoassay (FPIA),
pH extractions can be performed. These shortened immunochromatography, lateral-flow immunoassay,
extractions are essentially a compromise, since most microparticle immunoassays (turbidimetric or
drug groups are extracted to some extent but with a nephelometric), and radioimmunoassay (RIA). Because
lower overall recovery. these assays are antibody based, is important to consider
the drug target and related compounds that may be
For extraction of drugs, the most commonly used resin is detected. Commercial assay users should be familiar
known as XAD-2 (non-ionic polystyrene exchange with the metabolism of target drugs and consult the
resin). This resin is typically immobilized in a column manufacturers package inserts for compound cross-
bed, and urine is applied. Drugs are adsorbed onto the reactivity. External proficiency tests also provide a
resin and then subsequently eluted with appropriate useful source for antibody specificity.
solvents. As with conventional solvent extractions, the
proper selection of sample pH and solvents for elution Enzyme Immunoassay
optimize the selection process. Column extraction is Commercially available EMIT kit assays for an array of
amenable to batch extraction using semiautomated individual drugs or drug classes are offered by several
procedures [7]. manufacturers (Table 2). The sample is added to a drug-
specific antibody mixture, and then an enzyme-labeled
Drug extraction can also be achieved using other drug reagent is introduced. Drug in the patient specimen
compounds such as bonded-phase sorbents. Bonded- competes with conjugated drug for antibody. A
phase sorbents are usually composed of silica gels and colorimetric reaction occurs between the enzyme and
enable convenient batch processing in commercially substrate, where the amount of drug in the specimen is
available columns. As with XAD-2 resins, column- proportional to the signal. These assays offer good
bound drugs can be selectively eluted off with various specificity and adequate sensitivities for detection of
solvents. drugs in both overdose and drug-abuse situations. The
EMIT test menu includes an array of both prescription
Spot Tests drugs and drugs of abuse (see Table 4).
Spot tests are among the simplest and quickest methods
for initial drug screening or a confirmation test (Table As with other immunoassay, is important to consider the
2). Usually an unprocessed specimen (urine or serum) is drug target and related compounds that may cross-react
added directly to test reagents to assay for the presence with the antibody. For example, the Emit II Plus opiate
497
Drug Screens
assay (Syva, Dade Behring, USA) is targeted towards relative to an internal standard (Figure 3A), using
urinary morphine, morphine-3-glucuronide, and codeine thermal conductivity detection. More compound-specific
but can cross-react with high concentrations of synthetic detectors are available for nitrogen and phosphorus, or
opiates such as oxycodone and hydromorphone. halogens. For comparison of detection methods, contrast
Similarly, the Emit II plus assay for amphetamine Figure 3A with Figure 3B, which shows a sample
detects related compounds such as detected using nitrogen detection.
methylenedioxymethamphetamine (MDMA or
Ecstasy), as well as unrelated drugs like bupropion and Alternatively, a gas chromatograph can be interfaced
propranolol. with a mass spectrometer detector (GC/MS). This
combination provides the most objective analytical data
Radioimmunoassay of existing methods for structural elucidation of an
Although infrequently used today, RIA procedures have unknown compound (Figure 4). The GC/MS system has
historically provided the analyst with commercially been utilized both as a confirmatory procedure and as a
available assays for a select number of individual drugs primary screening method for drug screens when
[4]. As with EMIT assays, RIA specificity depends improved turnaround times are necessitated (immediate
completely on the antibody. RIAs have the highest overdose screening) [19]. The drawbacks of
analytical sensitivities, enabling detection of drugs at conventional GC and GC/MS are (1) the considerable
concentrations found in the majority of clinical technical expertise required to develop and perform
specimens. However, most assays require more than 30 testing and (2) high initial equipment costs.
min of incubation time, and few laboratories can offer
immediate results for more than one or two of these High-Performance Liquid Chromatography (HPLC)
assays. Radioactive-material handling and waste HPLC is another technique that can be used to separate
disposal are also a significant drawback. As with EMIT complex mixtures of drugs extracted from biological
assays, users should be aware of antibody specificity. fluids. Although specimens usually require extraction
before analysis, HPLC has an advantage over GC in that
Thin-Layer Chromatography (TLC) the sample does not have to be volatile or derivatized.
Despite its technical demands, TLC serves as an However, HPLC has limited ability to separate many
excellent screening or confirmation technique for both different classes of drugs simultaneously using the same
large laboratories and those with more limited column and buffer conditions. Therefore, several runs
capabilities. The method requires no instrumentation and may be required, using columns and conditions with
is relatively inexpensive (Table 2). Additional specificity for a given class of drugs (e.g.,
advantages include the flexibility to perform benzodiazepines). Like other forms of chromatography,
simultaneous analyses of a larger number of drugs on a HPLC is also limited in the number of patient samples
larger number of specimens. Sensitivities of most TLC that can be processed at a time, particularly when runs
screening procedures are adequate for both overdose and are long (>60 min). It is noteworthy that application of
drug-abuse cases, and specificity is often acceptable. The these traditional methods to ultra-pressure
major limitation of this method is lengthy turnaround chromatography (UPLC) will eventually shorten analysis
time. Depending on the method utilized, the procedure times. As with GC, it is possible to connect several
may take 2 to 4 hours [11]. Modified methods have been detectors, spectrophotometers, fluorometers,
introduced that are based on the same principles of TLC electrochemical detectors, or mass spectrometers in
but with accelerated development and detection steps series to characterize the compounds partially as they are
[12,13]. eluted from the column.
Gas Chromatography (GC) and Gas Chromatography Liquid Chromatography and Mass Spectrometry or
Mass Spectrometry (GC/MS) Tandem Mass Spectrometry
GC offers one of the most powerful analytical tools for When coupled to mass spectrometry (MS) or tandem
the separation of complex mixtures in biological fluids. mass spectrometry (MS/MS), liquid chromatography
As with thin-layer chromatography, GC enables the (LC) has the potential to definitively identify a vast array
evaluation of a large number of possible constituents of compounds. Although it is not yet widely utilized,
simultaneously (Table 2). Unlike thin-layer LC-MS/MS has a few advantages over GC-MS,
chromatography, however, GC techniques cannot screen including analysis of polar, nonvolatile, and heat-labile
a large number of patient specimens simultaneously; drugs [20,21]. As with HPLC, some of the laborious
therefore individual samples must be analyzed derivatization steps may be omitted. In addition, reagent
consecutively, lowering the overall throughput. costs may be substantially reduced in comparison with
immunoassays. As with GC-MS, limitations of LC-MS
For analysis by GC, specimens must generally be drug screening are the technical expertise required to
extracted, and the drug must either be volatile or develop and validate the assays and the initial capital
derivatized to make it volatile [14-18]. For screening costs [22,23].
purposes, samples can be separated on a single column
or, to enhance analytical specificity and sensitivity, can Point-of-Care Testing
be injected into various compound-selective columns. As with all critical analytes, it is attractive to consider
Compounds are identified by their retention times screening for drugs of abuse as point-of-care (POC) tests
498
Drug Screens
(Table 2). These potentially offer rapid turnaround times, Reference and Preferred Methods
simplified work processes, and could alleviate the As is so often found with esoteric laboratory tests, drug-
relatively high cost of maintaining low-volume, 7-day, screening procedures do not have a reference method.
24-hour service in the central testing laboratory [24,25]. The optimal reference method is difficult to establish,
since it would entail (1) defining which compounds
For POC drug testing, urine is the preferred specimen, should be included in various screens, (2) establishing a
since the more common drugs of abuse can be detected preferred separation or extraction scheme, (3) describing
for longer time periods in urine compared with serum. the best screening method, and (4) suggesting the best
The amount of sample required ranges from a few drops confirmation procedures. Given that the drugs of interest
to as much as 20 to 30 mL. Other samples have also invariably change over time, these criteria cannot readily
been considered for POC drug testing, including sweat, be achieved.
breath, and saliva [25]. Of these, saliva is considered an
attractive, readily accessible and noninvasive alternative The type of screen required is dependent on both
to urine. However, because some urinary drug economic and clinical needs. If there is good clinical-
metabolites differ from those found in saliva, traditional laboratory interaction, it is possible to offer more
cutoffs for urinethose specified by the U.S. Substance efficient specific testing. For example, in the acute care
Abuse and Mental Health Services Administration setting, an agitated or hyperactive patient does not need
(SAMSHA), for examplemay not be appropriate. to be screened for morphine, tricyclic antidepressants,
and benzodiazepines, because drugs such as
Advantages of POC testing include rapid turnaround amphetamines, phencyclidine, and the cocaine agents are
time and simplified chain-of-custody issues more probable causes of this altered mental status.
(confirmation tests are still required for tests with legal Therefore, the laboratory can offer faster, more accurate
implications). The demands for proper specimen service if the drug screen is limited to a few agents.
identification, labeling, and transport can also be Similarly, most drug-abuse screening related to
simplified with the use of POC devices that are used at employee testing or drug-rehabilitation programs
the bedside [26]. require only the testing for a limited number of drugs.
Examples of (1) comprehensive, (2) stimulant panel, and
Devices for POC testing have been designed for single- (3) drug-abuse screening procedures are presented
or multiple-drug detection. Test devices include a range below.
of formats, including dipsticks, cards, and plastic
cassettes. The methods used in these devices are Specimen
typically immunochromatographic, analogous to home The most commonly used samples for drug screens are
pregnancy tests, which permit simple visual blood or urine (Table 5). However, the type of specimen
interpretation of results. One example of a POC device is used for drug screens depends on many factors. Whole
based on competitive binding of antibody to drug present blood or serum samples are useful for quantitative
in urine and drug conjugate bound to a porous analysis of circulating agents such as therapeutic drugs.
membrane. If the urine contains drug, dye-conjugated This is particularly true when drug levels correlate with
antibodies bind to an immobilized antibody strip, toxicity and there are available antidotes (e.g., N-
resulting in the formation of a colored line. Although the acetylcysteine for acetaminophen poisoning). For past
performance of POC tests for drugs is improving [27], exposure and drugs-of-abuse screening, it is generally
they are not as accurate as immunoassays or GC-MS and accepted that the optimal specimen is urine. Immediate
are more expensive [28]. In addition, it is essential that and appropriate collection is critical for all drug screens
users be appropriately trained with the testing device and (see Table 6 for drug-detection windows).
be aware of its limitations. The National Academy of
Clinical Biochemistry (NACB) has published detailed For forensic drug analysis, a wide variety of specimens
guidelines on POC drugs-of-abuse screening [25]. may be employed, including vitreous fluid, hair, urine,
blood, and body-cavity fluid. For drugs-of-abuse testing
Future Technologies in the workplace, alternate specimens such as hair and
There are several developing technologies that may be saliva are being researched. At this time, these
used for drug screening and confirmation in the future. specimens are not used, owing to lack of standardization,
These newer technologies include infrared spectroscopy lack of reference ranges, and potential ethical bias, but it
[29], nuclear magnetic resonance spectroscopy [30], and is noteworthy that the U.S. Department of Health and
capillary zone electrophoresis [31,32]. CZE has been Human Services (DHHS) is actively exploring the use of
used experimentally to screen opiates [31,32], but alternate specimens [33]. This may accelerate the
because of the very small injection volumes, it is subject development of methods for such specimens, as well as
to detection-limit problems. This requires derivatization, the use of point-of-collection testing.
which essentially nullifies the ease of use and analytical
time savings. The other methods are expensive and Interpretation
largely experimental, so they will not be discussed here. Interpretation of results is largely method dependent.
When using tests that rely on antibodies, such as EMIT
assays, one should consider drug cross-reactivity. For
example, the EMIT amphetamine assay can cross-react
499
Drug Screens
with ephedrine and phenylpropanolamine, causing false- unknowns. Where appropriate, it is important to monitor
positive results. Thus even though the theoretical all three stages of the procedure: extraction or
specificities of immunochemical assays are generally preparation, screening, and confirmation. Screening test
excellent, all positive results encountered must be performance can also be monitored by correlating data
confirmed by a secondary procedure. In fact, if potential with confirmation test results (e.g. GC-MS).
medicolegal specimens are routinely analyzed (that is,
employee, athletic, or forensic screening), a second References
confirmatory analytical procedure must be employed. 1 Wu A. Urine adulteration before testing for
drugs of abuse. In: Shaw L, Kwong T, eds. The
A discussion of drug-abuse screening would not be Clinical Toxicology Laboratory: Contemporary
complete without some comment on the proper Practice of Poisoning. Washington, DC: AACC
processing of specimens, including collection, Press; 2001.
transportation, storage, and analytical phases of the 2 Mikkelsen SL, Ash KO. Adulterants causing
process. Since the ramifications of this type of screening false negatives in illicit drug testing. Clin
are of concern, both analytical accuracy and specimen Chem. 1988;34:2333-2336.
integrity must be assured. Where chain-of-custody is 3 Warner A. Interference of common household
required, all urine collections should be witnessed, chemicals in immunoassay methods for drugs
followed by a foolproof sample-identification method. of abuse. Clin Chem. 1989;35:648-651.
Results from forensic and nonstandard specimens should 4 Thoma JJ, Bondo PB, Sunshine I. Guidelines
be interpreted with caution. for Analytical Toxicology Programs. Cleveland:
CRC Press; 1977.
Medicolegal specimens should be properly stored 5 Sunshine I, Jatlow PI, Chemical Rubber
(refrigerated at 4C) and secured until time of Company. Methodology for Analytical
transportation. To assure their integrity, they must be Toxicology. Cleveland: CRC Press, 1975.
adequately sealed; this is best done with standard legal- 6 Moffat AC, Osselton MD, Widdop B, Clarke
specimen tape. The number of people handling the EGC. Clarkes Analysis of Drugs and Poisons
specimen during its transportation should be kept to a in Pharmaceuticals, Body Fluids and
minimum, and the sample should be delivered directly to Postmortem Material. 3rd ed. London/New
laboratory personnel. The assignment of an accession York: Pharmaceutical Press; 2004:2 v(xxi, 3-
number by the laboratory must be straightforward and 564, viii, 7-1935).
easy to follow through all phases of the testing procedure 7 Stimpfl T, Jurenitsch J, Vycudilik W. General
(such as instrument logs, labeling of chromatogram data unknown screening in postmortem tissue and
sheets, and final reports). All steps must be thoroughly blood samples: a semi-automatic solid-phase
documented when testing has legal ramifications. extraction using polystyrene resins followed by
liquid-liquid extraction. J Anal Toxicol.
Drug-Screening Performance Goals 2001;25:125-129.
Quality Control 8 Berry DJ, Grove J. Emergency toxicological
As with all laboratory tests, the performance and screening for drugs commonly taken in
reliability of drug screens must be continuously overdose. J Chromatogr. 1973;80:205-220.
monitored. One can accomplish this by employing an 9 Faulkner WR, American Association for
internal quality-control program and by participating in Clinical Chemistry. Selected Methods of
external proficiency surveys, such as those conducted by Emergency Toxicology. Washington, DC:
the College of American Pathologists (CAP). AACC Press; 1986:x,102.
Proficiency surveys enable labs to monitor screening 10 Fischer DS. A method for the rapid detection of
accuracy and allow evaluation of both the sensitivity and acute iron toxicity. Clin Chem. 1967;13:6-11.
specificity of the methods utilized. CAP proficiency 11 Davidow B, Li Petri N, Quame B. A thin-layer
surveys from 2006-2007 are summarized in Table 7. The chromatographic screening procedure for
data summary includes cannabinoids (THC-COOH), detecting drug abuse. Techn Bull Reg Med
cocaine (benzoylecgonine), amphetamine (amphetamine Technol. 1968;38:298-303.
and methamphetamine), opiates (heroin, opium, codeine, 12 Blass KG, Thibert RJ, Draisey TF. A simple,
morphine, and 6-acetyl morphine), and PCP rapid thin-layer chromatographic drug
(phencyclidine). Additional information resources are screening procedure. J Chromatogr.
available on the CAP website, http://www.cap.org. 1974;95:75-79.
13 Lillsunde P, Korte T. Comprehensive drug
Labs are also required to perform internal quality control screening in urine using solid-phase extraction
(QC), where known control materials spiked with an and combined TLC and GC/MS identification.
assortment of drugs are tested daily. Control materials J Anal Toxicol. 1991;15:71-81.
can be obtained commercially or can be prepared in- 14 Bastos ML, Jukofsky D, Mule SJ. Routine
house if the laboratory possesses the proper licenses for identification of drugs of abuse in human urine.
handling controlled substances. Control samples should 3. Differential elution of the XAD-2 resin. J
be processed in the same manner as patient specimens Chromatogr. 1973;81:93-98.
and should be submitted to the same rigorous analyses as
500
Drug Screens
15 Mule SJ. Routine identification of drugs of 27 Moody DE, Fang WB, Andrenyak DM, Monti
abuse in human urine. I. Application of KM, Jones C. A comparative evaluation of the
fluorometry, thin-layer and gas-liquid instant-view 5-panel test card with OnTrak
chromatography. J Chromatogr.1971;55:255- TesTcup Pro 5: comparison with gas
266. chromatography-mass spectrometry. J Anal
16 Mule SJ, Bastos ML, Jukofsky D, Saffer E. Toxicol. 2006;30:50-56.
Routine identification of drugs of abuse in 28 George S, Braithwaite RA. Use of on-site
human urine. II. Development and application testing for drugs of abuse. Clin Chem.
of the XAD-2 resin column method. J 2002;48:1639-1646.
Chromatogr. 1971;63:289-301. 29 Petibois C, Deleris G, Cazorla G. Perspectives
17 Fujimoto JM, Wang RI. A method of in the utilisation of Fourier-transform infrared
identifying narcotic analgesics in human urine spectroscopy of serum in sports medicine:
after therapeutic doses. Toxicol Appl health monitoring of athletes and prevention of
Pharmacol. 1970;16:186-193. doping. Sports Med (Auckland, NZ).
18 Breiter J, Helger R, Lang H. Evaluation of 2000;29:387-396.
column extraction: a new procedure for the 30 Grootveld M, Algeo D, Silwood CJ, Blackburn
analysis of drugs in body fluids. Forensic Sci. JC, Clark AD. Determination of the illicit drug
1976;7:131-140. gamma-hydroxybutyrate (GHB) in human
19 Deutsch DG, Bergert RJ. Evaluation of a saliva and beverages by 1H NMR analysis.
benchtop capillary gas chromatograph-mass BioFactors (Oxford, England). 2006;27:121-
spectrometer for clinical toxicology. Clin 136.
Chem. 1985;31:741-746. 31 Alnajjar A, Idris AM, Multzenberg M, McCord
20 Maurer HH. Screening procedures for B. Development of a capillary electrophoresis
simultaneous detection of several drug classes method for the screening of human urine for
used for high throughput toxicological analyses multiple drugs of abuse. J Chromatogr B Analyt
and doping control: a review. Comb Chem High Technol Biomed Life Sci. 2007;856:62-67.
Throughput Screen. 2000;3:467-480. 32 Wei F, Zhang M, Feng YQ. Application of
21 Politi L, Groppi A, Polettini A. Applications of poly(methacrylic acid-ethylene glycol
liquid chromatography-mass spectrometry in dimethacrylate) monolith microextraction
doping control. J Anal Toxicol. 2005;29:1-14. coupled with capillary zone electrophoresis to
22 Polettini A. Systematic toxicological analysis of the determination of opiates in human urine.
drugs and poisons in biosamples by hyphenated Electrophoresis. 2006;27:1939-1948.
chromatographic and spectroscopic techniques. 33 Bush DM. The U.S. Mandatory Guidelines for
J Chromatogr B Biomed Sci Appl. 1999;733:47- Federal Workplace Drug Testing Programs:
63. current status and future considerations.
23 Svensson JO, Andersson M, Gustavsson E, Forensic Sci Int. 2008;174:111-119.
Beck O. Electrospray LC-MS method with 34 Verstraete AG. Detection times of drugs of
solid-phase extraction for accurate abuse in blood, urine, and oral fluid. Ther Drug
determination of morphine-, codeine-, and Monit. 2004;26:200-205.
ethylmorphine-glucuronides and 6- 35 Figure adapted from:
acetylmorphine in urine. J Anal Toxicol. http://www.questdiagnostics.com/employerso
2007;31:81-86. lutions/dti_11_2005/dti_12.html (Accessed
24 Harvey MA. Point-of-care laboratory testing in 2009-03-03.
critical care. Am J Crit Care. 1999;8:72-83; 36 Moeller MR, Steinmeyer S, Kraemer T.
quiz 4-5. Determination of drugs of abuse in blood. J
25 Ian D, Watson RB, Hammett-Stabler C, Chromatogr B Biomed Sci Appl. 1998;713:91-
Nicholes B, Smith B, George S, Verstraete A, 109.
Goldberger B. Drugs and alcohol. In: Nichols 37 Natelson S. Techniques in Clinical Chemistry.
JH, ed. NACB Laboratory Medicine Practice Springfield, IL: Charles C Thomas; 1971:649-
Guidelines: Evidence-Based Practice for Point- 651.
of-Care Testing. Washington, DC: National 38 Curry A. Poison Detection in Human Organs.
Academy of Clinical Biochemistry; 2007. Springfield, IL: Charles C Thomas; 1971:67.
26 Yang JM, Lewandrowski KB. Urine drugs of 39 Jain NC, Leung WJ, Budd RD, Sneath TC.
abuse testing at the point-of-care: clinical Thin-layer chromatographic screening and
interpretation and programmatic considerations confirmation of basic drugs of abuse in urine. J
with specific reference to the Syva Rapid Test Chromatogr. 1975;115:519-26.
(SRT). Clin Chim Acta. 2001;307:27-32.
501
Drug Screens
Table 1. Suggested Confirmation Procedures for Screening Tests
Screen Confirmations
Spot test UV, TLC, or GC/MS
TLC UV or GC/MS
EMIT TLC, GC/MS, or UV
RIA TLC or GC/MS
GC GC/MS
Table 2. Types of Assays Used in Screening Procedures
Assay 1: Spot test

Principle: Drug or agent in specimen reacts chemically, giving specific color Advantages: Very rapid test;
indicates possible drug group present
Disadvantages: Not specific; different spot test required for each group of drugs; limited to a few drug groups
Assay 2: Ultraviolet spectroscopy
Principle: Extracted drug identified by its specific absorbance spectrum
Advantages: Inexpensive; spectrum often gives specific identification and quantitation of compound
Disadvantages: Not all drugs have suitable absorbance spectra; limited sensitivity; some degree of ambiguity
among drugs of same class; extensive sample preparation
Assay 3: Enzyme-multiplied immunoassay technique (EMIT)
Principle: Drug in the specimen competes with a conjugated drug-enzyme complex for a specific antibody;
unbound drug-enzyme conjugate is detected colorimetrically
Advantages: Rapid for single or several assays; most convenient method for detection of certain drugs
Disadvantages: Class specific; assays must be done separately for each class of drugs or individual drug; costly
Assay 4: Thin-layer chromatography
Principle: Extracted drug is separated by TLC and detected by specific chemical staining reaction or
physicochemical properties
Advantages: Can separate and detect more compounds than any other method; inexpensive
Disadvantages: Not possible to detect all drugs with one extraction or one chromatography system; detection is
subjective; long turnaround times
Assay 5: Gas chromatography (GC) and GC/mass spectrometry (MS)
Principle: Extracted drug is separated by GC and identified by retention time; with GC/MS, exact mass is used
to identify compounds.
Advantages: Can separate and detect large variety of drugs; GC/MS provides definitive identification of drugs
Disadvantages: Not possible to detect all drugs; extraction and confirmation necessary
Assay 6: Point-of-Care Tests
Principle: Variable, but typically drug in specimen interacts with conjugated antibody and is immobilized by
capture antibody for visual detection
Advantages: Rapid and easy to use
Disadvantages: Test devices are expensive; limited performance; only applicable to a subset of drugs
502
Drug Screens
Table 3. Spot Tests

Drug: Acetaminophen
Specimen required: Urine
_
Reaction: o-Cresol + acetaminophen NH4OH blue color
Test time: 15 min
Comments: Highly sensitive and relatively specific
Reference: [8]
Drug: Ethanol
Specimen required: Urine, serum
Reaction: Microdiffusion into dichromate
2K2Cr2O7 + 10H2SO4 + 3C2H5OH 2Cr2(SO4)4 + 2K2SO4 + 3CH3COOH +
11H2O + 4H+ (green to blue color)
Test time: 15 to 30 min
Comments: Good sensitivity; nonspecific for ethanol
Reference: [5]
Drug: Salicylate
Reaction: Trinders solution: Salicylate + FeCl3 violet-colored complex
Test time: 2 min
Comments: Good specificity if serum used; good sensitivity
Reference: [37]
Drug: Carbamates (meprobamate)
Reaction: Furfurol + meprobamate + antimony trichloride black color on thin-layer chromatography plate
Test time: 5 minutes
Comments: Not specific for meprobamate; good sensitivity
Reference: [38]
Drug: Imipramine/desipramine
Reaction: Forrest reagent K2CrO3 (acidic) + imipramine green-colored complex
Test time: 2 min
Comments: Phenothiazines may interfere
Reference: [5]
Drug: Ethchlorvynol
Reaction: Diphenylamine + ethchlorvynol in acid red color
Test time: 10 min
Comments: Good sensitivity and specificity
Reference: [9]
Drug: Phenothiazines
Reaction: FPN reagent (ferric chloride/perchloric acid/nitric acid) FeCl3 (acidic oxidizing agent) red- to violet-colored
complex
Test time: 2 min
Comments: Nonspecific; poor sensitivity for some phenothiazines
Reference: [5]
Drug: Iron
Specimen required: Serum
Reaction: Bathophenanthroline color change with iron (blue)
Test time: 15 minutes
Comments: Will not react at normal serum concentrations
Reference: [10]
503
Drug Screens
Table 4. EMIT Menu of Drug Tests

Urine Serum
Amphetamine Acetaminophen
Barbiturates Anticonvulsants:
Cannabinoids Carbamazepine, ethosuximide,
Cocaine phenobarbital, primidone, valproic Acid
Ethanol Antimicrobial drugs:
LSD Amikacin, gentamicin, tobramycin,
Methadone vancomycin
Methaqualone Caffeine
Opiates Cardioactive drugs:
Phencyclidine Digoxin, disopyramide, lidocaine, N-
Propoxyphene acetylprocainamide, procainamide,
Propoxyphene quinidine
Adulterates: Methotrexate
Bleach, chromate, pH, Salicylate
specific gravity, nitrates Theophylline
Table 5. Specimens of Choice in Drug Screen

Specimen: Urine
Volume required: 30 mL
Indication: Drug abuse screening; overdose screening; employment screening
Advantages: Generally easy to obtain in high volume; most drugs found in sufficient concentration to enable
identification
Disadvantages: Contains many metabolic products that may interfere with identification; parent drug may not be
present; quantitation offers little correlation with clinical effects
Specimen: Blood, serum, and plasma (tube type may depend on analyte)
Indication: Overdose screening; therapeutic drug monitoring
Advantages: Parent drug present; quantitative level may assist with patient management (therapeutic and toxic
reference levels often known)
Disadvantages: Limitation of sample volume; concentration of select drugs often too low to enable detection
(especially in non-overdose situations)
Specimen: Gastric lavage or emesis
Indication: Overdose screening
Advantages: Parent drug present
Disadvantages: Matrix problems (interference from foodstuffs); drugs quickly absorbed may be missed by
gastric screen; drugs not orally ingested will not be detected
504
Drug Screens
Table 6. Detection Window for Commonly Encountered Drugs of Abuse [34].

Drug/Metabolite Acute/Single Use Chronic/Heavy Use
Alcohol 0-4 hrs 6-12 hrs
Amphetamines 2-7 hrs 2-4 days
Anabolic Steroids 4-6 hrs 2-3 weeks (oral dose)
1-3 months injected dose
Barbiturates 2-4 hrs 1-4 days short-acting
(secobarbital, allobarbital)
2-3 weeks long-acting
(phenobarbital, barbital)
Benzodiazepines 2-7 hrs 4-6 weeks with long-acting
Cannabinoids 6-18 hrs Up to 30 days with chronic use
Cocaine metabolite 1-4 hrs 2-4 days
(benzoylecgonine)
Lysergic acid diethylamide (LSD) 2 hrs 1-4 days
Mescaline 1-2 hrs 2-4 days
Methadone 2 hrs 2-6 days
Methamphetamines 1-3 hrs 2-4 days
Methaqualone 3-8 hrs Up to 10 days
Methylenedioxymethamphetamine 1 hr 2-3 days
(MDMA)
Nicotine 4-6 hrs 7-14 days
Opiates (heroin, morphine, codeine) 2 hrs 2-3 days
Oxycodone 1 hr 1-2 days
Phencyclidine (PCP) 5-7 hrs 6-10 days
Propoxyphene 4-6 hrs 3-6 days
Psilocybin (mushrooms) 2 hrs 1-3 days
Rohypnol 1 hr 8 hrs
Gammahydroxybutyric acid (GHB) 1 hr 8 hrs
Tricyclic antidepressants (TCA) 8-12 hrs 2-7 days
*NOTE: Detection depends on dose, usage, physiological factors, and excretion rates. Data represent average rates.
Table 7. Summary of College of American Pathologists Proficiency Testing from 2006-2007.

DRUG CLASS
Cocaine
Amphetamine Metabolite Cannabinoid Opiate PCP
Cutoff (1000 ng/mL) (300 ng/mL) (20 ng/mL) (300 ng/mL) (25 ng/mL)
METHOD Sensitivity Sensitivity Sensitivity Sensitivity Sensitivity
CEDIA 98% 100% 100% 100% 100%
CMI 97% 99% 99% 100% 99%
EIA 98% 99% 99% 99% 100%
FIA 57%** 99% 94% 99% 100%
FPIA 99% 100% 99% 100% 99%
IC 63%** 99% 98% 99% 94%
LFI 55%** 96% 92% 98% 98%
MIA 63%** 99% 96% 100% 100%
Specificity of
All Assays 99% 100% 100% 100% 100%
Combined
*Data exclude PT samples below the cutoff and methods with different cutoff values than those listed above.
** No consensus was reached for some samples in these assays.CEDIA, cloned-enzyme donor immunoassay; CMI, colloid
metal immunoassay; EIA, enzyme immunoassay; FIA, fluorescence immunoassay; FPIA, fluorescence polarization
immunoassay; IC, immunochromatography; LFI, lateral-flow immunoassay; MIA, microparticle immunoassay.
505
Drug Screens
Figures
Drug Screen: Figure 1
A.B.
Prevalence of drug positivity in employer drug screens from January to June 2005 [35].
A, Phencyclidine-positive (PCP-positive) rate. B, Cocaine-positive rate. ND, Not determined.
506
Drug Screens
Absorbance Spectra for A, Theophylline in 0.1 M NaOH. B, Sulfamethoxazole in 0.1 M NaOH.
Gas chromatograms for A, Nonselective detector (flame ionization detector). B, Selective detector (nitrogen-phosphorus
detector).
507
Drug Screens
GC-MS chromatogram of serum samples spiked with THC (2 ng/mL), OH-THC (5 ng/mL) and THC-COOH (20 ng/mL).
Section 1: m/z 313, 328, 316, 331; Section 2: m/z 313, 314, 358; Section 3: m/z 313, 357, 372, 375.
Gas chromatographic pattern of standard solutions of amphetamines. 1,d-amphetamine; 2, phentermine; 3,

methamphetamine; 4, phenylpropanolamine; 5, pseudoephedrine; 6, ephedrine; 7, phenmetrazine. Note that nicotine has
the same retention time as phenylpropanolamine.
508
Drug Screens
Gas chromatographic pattern of patient urine extract to confirm positive-EMIT cocaine. Note positive cocaine (peak 2) and
its methyl ester (peak 1).
Drug-Screening Methods and Procedures brown bottle at room temperature. This is stable for 1
year at room temperature.
Procedure: Comprehensive Drug Screen 3. o-Cresol reagent (0.096 mol/L). Add 10 mL
Principle of o-cresol to a 1 L volumetric flask, and dilute to mark
Urine or gastric specimens are extracted with organic with distilled water. Store at room temperature. This is
solvents at a selected pH followed by analysis by thin- stable for 1 year at room temperature.
layer chromatography. Selected ancillary spot tests and 4. Forrest reagent
enzyme immunoassays are simultaneously performed on Nitric acid (7.8 mol/L). Add 50 mL of
urine. Positive findings are confirmed by use of concentrated nitric acid to 50 mL of distilled water.
appropriately chosen techniques. Perchloric acid (1.85 mol/L). Add 20 mL of 70%
Indications perchloric acid to 50 mL of distilled water. Allow to
The comprehensive drug screen is ordered to establish cool, and bring volume to 100 mL.
information regarding drug usage as it may pertain to Sulfuric acid, 12 N (6 mol/L). Add 30 mL of
cases of a toxicological nature, including accidental or concentrated sulfuric acid slowly, with constant stirring,
suicidal overdose with coma or for evaluation of to 50 mL of distilled water. Allow to cool, and bring
potential drug involvement in other medical volume to 100 mL.
emergencies, such as seizures, jittery baby syndrome, or Potassium dichromate, 2 g/L (6.8 mmol/L). Add 200 mg
drug interactions. of potassium dichromate to 80 mL of distilled water.
Reagents Dissolve, and bring to a final volume of 100 mL.
All reagents should be of chromatographic Mix 100 mL of each of the above reagents together. This
grade. solution is stable for 1 year at room temperature when
1. Trinders solution: Fe(NO3)3, 100 mmol/L; stored in an opaque (amber) glass container.
HgCl2, 14.7 mmol/L; HCl, 120 mmol/L. Add 4 g of 5. FPN reagent (ferric chloride/perchloric
acid/nitric acid)
ferric nitrate and 4 g of mercuric chloride to a 100 mL Nitric acid (7.8 mol/L). See above, under Forrest
volumetric flask. Dissolve salts in 12 mL of 1 M reagent.
hydrochloric acid. Dilute to mark with distilled water. Perchloric acid (1.85 mol/L). See above, under Forrest
Store at room temperature. This is stable for 6 months. reagent.
2. Potassium dichromate (3.4 mmol/L). Add 1 g Ferric chloride (0.308 mol/L). Dissolve 5 g of ferric
of potassium dichromate to a 1 L volumetric flask. chloride in 50 mL of distilled water in a volumetric
Dissolve salt in 500 mL of distilled water. Add 0.1 g of flask. Bring to final volume of 100 mL.
silver nitrate to the flask, and then slowly, with stirring, Mix these solutions in the following proportions: 1 part
add 500 mL of concentrated sulfuric acid. Store in a ferric chloride, 9 parts perchloric acid, and 10 parts nitric
509
Drug Screens
acid. This is stable for 1 year at room temperature when 16. Standards
stored in an opaque (amber) glass container. Ethanol (789 mg/L, 17.15 mmol/L). Pipet 0.1 mL of
6. Concentrated hydrochloric acid. absolute ethanol into a 100-mL volumetric flask. Dilute
7. Ammonium hydroxide, 4 mol/L. Dilute 284 to mark with distilled water. Store at 4C. This is stable
mL of concentrated ammonium hydroxide to 1 L with for 2 months.
distilled water. This is stable for 2 years at room Salicylate standard (100 mg/L, 724 mol/L). Dissolve
temperature. 10 mg of salicylic acid (USP grade) in 0.5 mL of
8. EMIT d.a.u. (drug-of-abuse urine) opiate methanol in a 100-mL volumetric flask, and dilute to
and benzodiazepine assay kits. These kits are usually 100 mL with distilled water. Store at 4C. This is stable
stable for 1 year before opening. Follow manufacturers for 6 months.
recommendations for storage and use. Acetaminophen urine control (100 g/mL, 0.66
9. Extraction buffer. Prepare a saturated mol/mL). Add 10 mg of acetaminophen to a 100-mL
ammonium chloride solution by dissolving 600 g in 1 L volumetric flask. Dissolve in 1 mL of methanol. Dilute
of hot, distilled water. Cool, and add concentrated to mark with urine previously noted to be free of the
ammonium hydroxide to adjust the pH to 9.5. Store at drug. Refrigerate at 4C. This is stable for 6 months.
room temperature. This is stable for 6 months. Urine drug control. Select a suitable commercially
10. Extraction solvent. Prepare a mixture of available control or prepare one using the following
dichloromethane and isopropanol (90:10 by volume). procedure:
Store at room temperature. Prepare weekly. 1. Collect approximately 2 L of drug-free urine (urine
11. Methanolic 0.1 M hydrochloric acid. Add 0.9 previously noted to be negative for all drugs through
mL of concentrated hydrochloric acid to a 100 mL testing). To clarify, filter through Whatman No. 1 filter
volumetric flask. Dilute to mark with methanol. Store at paper.
room temperature. This is stable for 1 year. 2. Prepare a stock drug standard by adding 4 mg each of
12. Reconstitution solvent. Prepare a 1:1 (v/v) phenobarbital, secobarbital, morphine, codeine,
mixture of dichloromethane and methanol. Store at room meperidine, methyprylon, and oxazepam and 10 mg each
temperature. This is stable for 1 year. of d-amphetamine and methamphetamine to a 10-mL
13. Thin-layer chromatographic developing volumetric flask. Dissolve the drugs in methanol, and
solution. Mix 85 mL of ethyl acetate, 10 mL of dilute to mark with methanol.
methanol, and 5 mL of ammonium hydroxide. Prepare 3. Transfer the entire contents of the drug standard flask
fresh for each development. to a 2-L volumetric flask. Rinse the flask several times
14. Thin-layer chromatographic sprays with distilled water, and add rinses to the 2 L flask.
Ninhydrin (5.61 mmol/L). In a 100-mL volumetric flask, Dilute to the 2-L mark with the urine pool. Mix well.
dissolve 100 mg of 1,2,3-indantrione in 5 mL of acetone, 4. Divide into 20-mL aliquots, and store in plastic vials.
and dilute to mark with acetone. Prepare fresh daily. 5. Store at 20C until used. This is stable for 1 year.
Diphenylcarbazone (2 mmol/L). In a 100-mL volumetric 6. Control contains 2 g/mL of each drug except the
flask, dissolve 10 mg of diphenylcarbazone in 5 mL of amphetamines, which have concentrations of 5 g/mL.
acetone, and dilute to mark with a 1:1 acetone-water 17. Thin-layer chromatographic reference
solution. This is stable for 1 month at room temperature. standard (1 g/mL). Add 10 mg each of
Mercuric sulfate (11.5 mmol/L). Add 0.5 g of mercuric propoxyphene, methadone, chlorpromazine, meperidine,
oxide to 20 mL of concentrated sulfuric acid. Add this quinine, phenobarbital, phenylpropanolamine, and
acid solution slowly to 150 mL of distilled water, and morphine to a 10-mL volumetric flask. Dissolve in
bring to a final volume of 200 mL. This is stable for 6 methanol, and dilute to mark with methanol. Refrigerate
months at room temperature. at 4C. This is stable for 6 months.
15. Iodoplatinate
18. Individual drug standards (1 g/mL).
Stock solution (0.38 mol/L). Dissolve 10 g of platinum
Prepare in the same manner as previous standard with
chloride in 100 mL of distilled water. refrigerate at 4C.
drug of choice. This is stable for 6 months at 4C.
This is stable for 1 year.
Assay
Working solution. Add 5 mL of stock solution to 3 g of
Equipment:
potassium iodide in 100 mL of distilled water. Dilute
Porcelain spot-test plates
this mixture to 125 mL with distilled water, and then
Conway microdiffusion dish
dilute this with an equal volume of methanol (final
Teflon 125-mL volume separatory funnels (Fisher
volume 250 mL). Refrigerate at 4C. This is stable for 6
Scientific)
months.
60-mL glass centrifuge tubes
Dragendorff reagent. Dissolve 1.3 g of bismuth
Sample concentrator/evaporator (Brinkmann, Buchii
subnitrate in a solution composed of 60 mL of distilled
Rotavap, or other convenient evaporation system)
water and 15 mL of glacial acetic acid. Dissolve 12 g of
Thin-layer chromatographic apparatus
potassium iodide in 30 mL of distilled water. Combine
Spotting platen
the two solutions. Dilute this mixture with 100 mL of
20 20-cm glass, silica-gel 60 F-254 plates
distilled water and 25 mL of glacial acetic acid.
Refrigerate at 4C. This is stable for 6 months. Developing tank to accommodate 20 20-cm plates
Chromatographic spray cans
Air blower
510
Drug Screens
Short (254 nm)/long (366 nm) combination

ultraviolet lamp Thin-Layer Chromatography (TLC)
TLC spotting. Use a conditioned (preheated for 1 hr at
Procedure: Comprehensive Drug Screen 70C) TLC plate. Using a spotting platen, mark the spot
(Alternative) origin at 2 cm from the bottom of the plate. From this
mark, measure 15 cm upward, and draw the end line.
Ancillary Tests Apply approximately 1 L of reference standard in the 3,
Salicylate spot test. Add 3 drops of salicylate control or 10, and 17 positions on the plate. Use positions 9, 11,
patient specimen to a porcelain dish. Serum is the first and 12 to spot drug standards of interest (drugs indicated
specimen of choice, followed by urine. Gastric lavage or by history). Apply extracted control and patient
emesis is unacceptable without special treatment. (If specimens in open positions. Apply approximately 3 L
urine is used, boil the specimen first to remove diacetic at a time, and spot 10 times, allowing the applied spots
acid, a potentially interfering substance.) Add 3 drops of to dry between applications.
Trinders reagent. A violet color indicates the presence Plate development. Prepare the TLC development
of salicylate. solvent, and pour into tank lined with 20 20-cm filter
paper on both sides. Position the plate or plates in the
Alcohol spot test. To the center well of a development tank, and allow time for migration to
microdiffusion dish, add 0.5 mL of potassium proceed to the end line (approximately 40 min).
dichromate reagent. To the outer ring, add 0.5 mL of
alcohol control or patient specimen (urine or serum). Detection. Air-dry the plate. Observe the plate under a
Lightly grease the outer ring of the microdiffusion dish long-wave (366 nm) ultraviolet lamp. Mark any
lid with vacuum grease. Cover and allow to stand at fluorescing compound by lightly tracing the spot with a
room temperature. The dichromate solution will change pencil. Mark subsequent positive spots with a pencil.
from a yellow to a light green or blue color in the Place under short (254 nm) light, and look for any
presence of methanol, ethanol, or isopropanol. absorbing spots. Spray with Ninhydrin, irradiate with
long-wave ultraviolet radiation for 5 min, and then heat
Acetaminophen spot test. To 1 mL of patient or plate at 70C for 5 min. Amines, such as
control urine, add 1 mL of concentrated hydrochloric phenylpropanolamine and d-amphetamine, will stain
acid. Heat in a boiling water bath for 10 min. Dilute 0.1 pink. Spray with diphenylcarbazone, followed by the
mL of this sample with 0.9 mL of o-cresol reagent. Add mercuric sulfate spray. Observe the blue-violet colors of
2 mL of 4 M ammonium hydroxide. Blue color indicates barbiturates, phenytoin, glutethimide, and ethchlorvynol.
presence of acetaminophen. Re-spray with diphenylcarbazone, and heat plate at 70C
for 10 min. The pink, red, and violet colors of
EMIT II Plus d.a.u. For amphetamines, barbiturates, metabolites of phenothiazine drugs will appear. Observe
benzodiazepines, cannabinoids (THC), cocaine, opiates, under long-wave ultraviolet lamp. Observe the yellow
phencyclidine (PCP), methadone, methaqualone, and fluorescence of the benzodiazepine drugs and their
propoxyphene assays. Use as described by metabolites. Spray with iodoplatinate and Dragendorff
manufacturer. Cross-reactivity of some compounds is reagents. Most nitrogenous basic drugs will stain dark.
indicated in package inserts. Soak the finished plate in water for 30 seconds, remove,
and dry with a jet of hot air. Observe the chalk-white
Extraction of Urine colors of methyprylon and carbamates such as
1. Process the drug-screen control along with meprobamate. Rf values (distance of spot migration
patient specimen.
divided by distance of developing solvent migration),
2. To a 125-mL separatory funnel, add 20 mL of
spray reactions, and ultraviolet detections of unknowns
urine and 2 mL of ammonium chlorideammonium
should be compared to known standards.
hydroxide buffer, and mix.
3. Add 30 mL of extraction solvent, and extract
Confirmation
for 10 min.
The presence of drugs indicated as being positive by the
4. Allow layers to separate. Collect bottom solvent
thin-layer chromatographic or ancillary test screening
layer in a 50-mL glass centrifuge tube, and centrifuge for
methods must be confirmed by a second analytical
5 min at 500 g at room temperature.
procedure different from that used to screen for the
5. With a glass stirring rod, break emulsions that
compound (see Table 1, suggested confirmation
may have formed by aggregation, and re-centrifuge.
procedures for screening tests). Occasionally, when
Aspirate off top aqueous layer and discard. Filter solvent
serum is provided and when indicated, as in overdose
through filter paper (P5, Fisher Scientific, qualitative)
cases, the quantitation of the suspected drug in serum is
into evaporation flask.
the confirmatory step in the method. At other times,
6. Add 2 drops of methanolic 0.1 M hydrochloric
additional tests must be performed on the urine to
acid, and evaporate to dryness at 50C under nitrogen.
confirm the presence of the drug.
7. Using a Pasteur pipet, reconstitute with 10
drops (approximately 150 L) of methylene dichloride
methanol solution, being sure to rinse down the sides of
the evaporation flask.
511
Drug Screens
Drugs Which Can Be Identified by a Comprehensive Methyprylon (Noludar)*

ScreenListed Alphabetically Morphine
Nortriptyline (Aventyl)*
Orphenadrine
Acetaminophen (Tylenol)*
Oxycodone
Amitriptyline (Elavil)*
Pentazocine (Talwin)
Amobarbital*
Pentobarbital*
Amoxapine (Asendin)*
Phencyclidine (PCP)
Amphetamine
Pheniramine
Atropine
Phenmetrazine (Preludin)
Benzocaine
Phenobarbital*
Benzodiazepinesincludes metabolites of:
Phenothiazinesincludes metabolites of:
Chlordiazepoxide (Librium)*
Trifluoperazine (Stelazine)
Chlorazepate (Tranxene)*
Propiomazine (Largon)
Diazepam (Valium)*
Perphenazine (Trilafon)
Oxazepam (Serax)*
Prochlorperazine (Compazine)
Benzotropine (Cogentin)
Promazine (Sparine)
Benzphetamine (Didrex)
Chlorpromazine (Thorazine)
Brompheniramine
Thioridazine (Mellaril)
Butalbital*
Phentermine (Ionamin)
Caffeine
Phenylpropanolamine
Cannabinoids
Phenyltoloxamine
Carbamazepine (Tegretol)*
Phenytoin (Dilantin)*
Carbromal
Primidone (Mysoline)*
Carisoprodol (Soma)
Procainamide (Pronestyl)*
Chlorpheniramine
Procaine
Cimetidine/ranitidine
Promethazine (Phenergan)
Cocaine
Propranolol (Inderal)
Codeine
Propoxyphene (Darvon)
Cyclobenzaprine (Flexeril)
Pyrilamine
Desipramine (Norpramine)*
Quinidine/quinine*
Dimenhydrinate (Dramamine)
Quinine
Diphenhydramine (Benadryl)
Salicylate*
Disopyramide (Norpace)
Secobarbital*
Doxepine (Sinequan)*
Strychnine
Doxylamine
Theophylline (Aminophylline)
Ephedrine/pseudoephedrine
Trimethoprim
Ethanol*
Trimipramine
Ethchlorvynol (Placidyl)*
Trip-(Pyribenzamine)
Ethinamate (Valmid)
Tripelennamine
Fluoxetine (Prozac)
Triprolidine
Flurazepam (Dalmane)
Verapamil
Glutethimide (Doriden)*
Haloperidol (Haldol)
*Quantitative assays available.
Hydrocodone
Not available in stat screen.
Hydromorphone (Dilaudid)
Not routinely reported
Hydroxyzine (Vistaril)
Imipramine (Tofranil)*
Isopropanol*
Lidocaine (Xylocaine)*
Loxapine (Loxitane)
Maprotiline (Ludiomil)
Meperidine (Demerol)
Mepivacaine
Meprobamate (Miltown)*
Methadone
Methamphetamine
Methanol*
Methapyrilene
Methaqualone (Quaalude)*
Methocarbamol (Robaxin)
Methylphenidate (Ritalin)
512
Drug Screens
Drugs Which Can Be Identified by a Comprehensive NARCOTIC ANALGESICS

ScreenListed by Class Codeine
Hydrocodone
Hydromorphone (Dilaudid)
ALCOHOLS
Meperidine (Demerol)
Ethanol*
Methadone
Isopropanol*
Morphine
Methanol*
Oxycodone
Pentazocine (Talwin)
ANTICONVULSANTS
OTHER ANALGESICS
Carbamazepine (Tegretol)*
Acetaminophen (Tylenol)*
Phenobarbital*
Benzocaine
Phenytoin (Dilantin)*
Mepivacaine
Primidone (Mysoline)*
Procaine
ANTIDEPRESSANTS
Propoxyphene (Darvon)
Amitriptyline (Elavil)*
Salicylate*
Amoxapine (Asendin)*
STIMULANTS
Cyclobenzaprine (Flexeril)
Amphetamine
Desipramine (Norpramin)*
Benzphetamine (Didrex)
Doxepin (Sinequan)*
Caffeine
Fluoxetine (Prozac)
Cocaine
Imipramine (Tofranil)*
Ephedrine/pseudoephedrine
Loxapine (Loxitane)
Methamphetamine
Nortriptyline (Aventyl)*
Methylphenidate (Ritalin)
Maprotiline (Ludiomil)
Phencyclidine (PCP)
Trimipramine
Phenmetrazine (Preludin)
ANTIHISTAMINES
Phentermine (Ionamin)
Brompheniramine
Phenylpropanolamine
Chlorpheniramine
Strychnine
Dimenhydrinate (Dramamine)
TRANQUILIZERS
Diphenhydramine (Benadryl)
Benzodiazepinesincludes metabolites:
Doxylamine
Chlorazepate (Tranxene)*
Methapyrilene
Chlodiazepoxide (Librium)*
Orphenadrine
Diazepam (Valium)*
Pheniramine
Oxazepam (Serax)*
Promethazine (Phenergan)
Carisoprodol (Soma)
Phenyltoloxamine
Haloperidol (Haldol)
Pyrilamine
Hydroxyzine (Vistaril)
Triprolidine
Meprobamate (Miltown)*
Tripelennamine (Pyribenzamine)
Methocarbamol (Robaxin)
CARDIAC DEPRESSANTS AND BETA
Phenothiazinesincludes metabolites:
BLOCKERS
Chlorpromazine (Thorazine)
Disopyramide (Norpace)
Perphenazine (Trilafon)
Lidocaine (Xylocaine)*
Prochlorperazine (Compazine)
Procainamide (Pronestyl)*
Promazine (Sparine)
Propanolol (Inderal)
Propiomazine (Largon)
Quinidine/Quinine*
Thioridazine (Mellaril)
Verapamil
Trifluoperazine (Stelazine)
HYPNOTICSSEDATIVES
MISCELLANEOUS
Amobarbital*
Atropine
Butalbital*
Benztropine (Cogentin)
Pentobarbital*
Cannabinoids
Secobarbital*
Cimetidine/ranitidine
Carbromal
Quinine
Ethchlorvynol (Placidyl)*
Theophylline (Aminophylline)
Ethinamate (Valmid)
Trimethoprim
Flurazepam (Dalmane)
Glutethimide (Doriden)*
*Quantitative assays available.
Methaqualone (Quaalude)*
Not available in STAT screen.
Methyprylon (Noludar)*
Not routinely reported
513
Drug Screens
Procedure: Stimulant Panel (Urine) assay. Report in quotation marks those test samples
measuring less than the EMIT low calibrator for that
Principle assay. Proceed to the confirmatory techniques listed
Various drugs of interest are first screened using enzyme below if positive results are obtained from the EMIT
immunoassay, and presumptive-positive cases are assay.
confirmed by either gas or thin-layer chromatographic
methods. Amphetamine Confirmation by Gas Chromatography
Indications Equipment: Gas chromatograph with flame
The stimulant panel is used to rule out the following ionization detector, utilizing a 6-foot, 2-mm inner
drugs as the cause of hyperactivity, psychosis, or other diameter glass column packed with GP 10% Apiezon
stimulated states of consciousness: cocaine, L/2% KOH on 80/100 Chromosorb W AW column of 2-
phencyclidine (PCP), d-amphetamine, mm inner diameter (Supelco, Inc., Bellefonte, PA).
methamphetamine, phenmetrazine, phentermine, 1. Sample preparation by Toxi-Lab tube extraction
phenylpropanolamine, ephedrine. a. Shake Toxi-Lab tube A to mix
The schema of analysis is as follows: contents.
b. Add 5 mL of urine, and mix for 5 min.
PCP c. Centrifuge for 2 min at 1000 g.
Cocaine Amphetamine d. Transfer half of the supernatant to a
Screen EMIT EMIT EMIT Reacti-Vial, add 2 drops of 0.1 mol of HCl per
Confirmation 1 GLC (OV-17) GC (OV-17) liter of methanol, and evaporate to dryness.
GC (Apiezon)
Confirmation 2 Toxi-Lab or Toxi-Lab equivalent for e. Save the other half of the supernatant
each drug for further confirmation if necessary.
f. Reconstitute the dried residue from
EMIT is the enzyme immunoassay d.a.u. (drug of abuse step (d) in 20 L of methanol, and vortex mix
urine) kit sold by Dade-Behring (Syva), GLC is gas- the vial.
liquid chromatography utilizing 3% OV-17 and Apiezon g. Inject 1 L of the mixed standard into
column types, and Toxi-Lab is a drug detection system the gas chromatograph.
sold by the analytical systems division of Marion h. Allow the column to re-equilibrate
Laboratories, Inc., Laguna Hills, CA. after the run.
Reagents i. Inject 4 L of the reconstituted
1. EMIT II Plus (Syva, Dade Behring) reagents. specimen.
Reagents for cocaine, phencyclidine (PCP), and 2. Chromatographic parameters
amphetamine, including negative controls, and low a. Column temperature program, 130C
calibrator. to 210C, 10C per minute
2. Toxi-Lab reagents, including A tubes b. Injection port temperature, 250C
(Marion Labs). c. Detector temperature, 270C
3. Methanol. Chromatoquality (Fisher Scientific). d. Nitrogen flow, 30 mL/min
4. 0.1M HCl in methanol. Prepare by diluting e. Attenuation, 8
8.5 mL of concentrated HCl to 1 L with methanol. f. Range, 10-11 amp/mV
5. Gas chromatography amphetamine reference g. Chart speed, 2 cm/min
standards 1 mg/mL. Add 5 mg of each of the h. Sample run time, 8 min
following to a 5-mL volumetric flask: d-amphetamine, 3. Confirm by comparing relative retention times
methamphetamine, phentermine, phenmetrazine, of unknown peaks to peaks from standard run. Drug
ephedrine, phenylpropanolamine, and nicotine. Bring to Screen: Figure 5 shows a typical separation of
volume with methanol. Store at 4C. Stable for 3 amphetamine standards.
months. 4. Perform Toxi-Lab TLC plate development for
6. Gas chromatography PCP, cocaine, and stimulants as described by the manufacturer for
benzoylecgonine standards 1 mg/mL. Add 5 mg of secondary confirmation.
each of the following to a 5-mL volumetric flask:
cocaine, benzoylecgonine, and phencyclidine. Bring to Cocaine Confirmation by Gas Chromatography
volume with methanol. Stable for 1 month at 4C. Equipment: Same GC detector and column as
Note. A primary cocaine metabolite, the ecgonine above but column packing is 3% OV-17 on 100/120
methyl ester, is presently not routinely available. It is Supelcoport.
suggested that urines known to have tested positive for 1. Sample preparation by Toxi-Lab tube extraction
cocaine be used as a source of this metabolite. a. Shake Toxi-Lab tube A to mix
Assay contents.
EMIT Screening Procedure b. Add 5 mL of urine, and mix for 5 min.
Perform the EMIT PCP, cocaine, and c. Centrifuge for 2 min at 1000 g.
amphetamine screening procedures according to d. Transfer all the upper fraction to a
manufacturers directions. Phentermine, phenmetrazine, Reacti-Vial.
ephedrine, and phenylpropanolamine are detected, as are
other amphetamines with the amphetamine EMIT d.a.u.
514
Drug Screens
e. Extract the remaining bottom layer for l. Allow the column to re-equilibrate
benzoylecgonine by adding 2 mL of after the run.
dichloromethane. m. Inject 1 L of the unknown extract,
f. Mix for 2 min. and initiate run program.
g. Centrifuge for 2 min at 1000 g. 2. Chromatographic parameters
h. Remove the colored aqueous layer, a. Column temperature program, 150C
and transfer the dichloromethane to the same to 270C, 15C per minute
evaporation vial as in step (d). b. Injection port temperature, 250C
i. Evaporate to dryness at 50C under a c. Detector temperature, 250C
gentle stream of nitrogen. d. Nitrogen carrier flow, 20 mL/min
j. Reconstitute in 20 L of methanol, and e. Attentuation, 8
vortex mix. f. Chart speed, 1.2 cm/min
k. Inject 1 L of the GC cocaine standard g. Sample run time, 25 min
into the gas chromatograph, and follow run 3. Confirm by comparing retention times of
program. unknown peaks to peaks from standard run. Drug
l. Allow the column to re-equilibrate Screen: Figure 6 shows a patient urine positive for
after the run. cocaine.
m. Inject 1 L of the unknown extract, 4. Perform Toxi-Lab analysis for cocaine as
and initiate run program. described by the manufacturer for secondary
2. Chromatographic parameters confirmation.
a. Column temperature program, 150C
to 270C, 15C per min Procedure: Drug Abuse Screen
b. Injection port temperature, 250C Principle
c. Detector temperature, 250C Drugs of interest are first screened by enzyme
d. Nitrogen carrier flow, 20 mL/min immunoassay, with presumptive positive cases being
e. Attenuation, 8 confirmed by TLC.
f. Chart speed, 1.2 cm/min Indications and Comments
g. Sample run time, 25 min The screen is used to rule out the presence of select
3. Confirm by comparing relative retention times drugs of abuse as part of employee-screening programs,
of unknown peaks to peaks from standard run. Drug to monitor clients in drug-rehabilitation programs, or for
Screen: Figure 6 shows a positive urine cocaine any clinical situation requiring knowledge of drug usage.
chromatogram.
4. Perform Toxi-Lab analysis for cocaine as As with the comprehensive drug screen, the drugs
described by the manufacturer for secondary chosen to be included in this screen will be selected
confirmation. based upon the current prevalence of their abuse, along
with technical and economic considerations. The
PCP Confirmation by Gas Chromatography constituents of the screen will be continuously changing,
Equipment: as above for Cocaine just as drug-usage patterns are. When a limited approach
1. Sample preparation by Toxi-Lab extraction to drug-abuse screening is the goal, as presented here,
a. Shake Toxi-Lab tube A to mix the participation by commercial suppliers of drug assay
contents. kits is essential for a successful program.
b. Add 5 mL of urine, and mix for 5 min. Reagents and supplies
c. Centrifuge for 2 min at 1000 g. 1. EMIT d.a.u. assay kits. (Dade-Behring)
d. Transfer the entire upper fraction to a 2. Toxi-Lab A and B drug detection systems
Reacti-Vial. and cannabinoid (THC) assay kits. (Analytical
e. Extract the remaining bottom layer for Systems, Inc., Laguna Hills, CA)
benzoylecgonine by adding 2 mL of Assay
dichloromethane. 1. Perform the following EMIT d.a.u. enzyme
f. Mix for 2 min. immunoassays as described by the manufacturer:
g. Centrifuge for 2 min at 1000 g. amphetamines, barbiturates, benzodiazepines,
h. Remove the colored aqueous cannabinoids (THC), cocaine, opiates, phencyclidine
layer, and transfer the dichloromethane (PCP), methadone, methaqualone, and propoxyphene.
to the same evaporation vial as in step 2. If found to be presumptively positive by
(d). immunoassay, confirm with the appropriate Toxi-Lab
i. Evaporate to dryness at 50C under a procedure as described by the manufacturer.
gentle stream of nitrogen. a. Positive amphetamines. Confirm using
j. Reconstitute in 20 L of methanol, and the Toxi-Lab Sympathomimetic Amines
vortex mix. (Amphetamine) Differentiation special
k. Inject 1 L of the GC cocaine procedure using Toxi-Lab A system.
standard into the gas chromatograph, b. Positive barbiturates. Confirm using
and follow run program. the Toxi-Lab B system.
515
Drug Screens
c. Benzodiazepines. Confirm using the

Toxi-Lab Benzodiazepines: Hydrolysis
special procedure using Toxi-Tubes B and
Toxi-Lab A system.
d. Cocaine. Confirm using conventional
Toxi-Lab A system for basic drugs and the
Benzoylecgonine: Extraction and Detection
special procedure.
e. Phencyclidine. Confirm using the
Toxi-Lab Phencyclidine: Confirmation by
Remigration special procedure.
f. Methaqualone, methadone, and
propoxyphene. Confirm using the Toxi-Lab A
system. For methaqualone, also use the
Methaqualone: Confirmation with Sodium
Borohydride special procedure.
g. Opiates. Confirm using the
conventional Toxi-Lab A system and the
Morphine and Codeine: Confirmation using
Sodium Hydroxide special procedure. In
addition, the special procedure Morphine:
Hydrolysis of the Glucuronide Conjugate may
be required.
h. Cannabinoids. Confirm using the
Toxi-Lab Cannabinoid (THC) assay system.
516
Estradiol
Estradiol
Greg Ward
Name: Estradiol, E2 , estra-1,3,5(10)-triene-3,17-diol

Clinical significance:
Structure: C18H24O2
Chemical Class: Steroid
Refer to Chapter 44, Pregnancy, and Chapter 50, The Gonads, in the 5th edition of Clinical Chemistry: Theory, Analysis,
Correlation.
Principles of Analysis and Current Usage Unbound antibody binds to an estradiol derivative
Early methods of estradiol analysis used the competitive- coupled to paramagnetic particles (Solid Phase). The
binding assay format with tritiated estradiol as the acridinium ester remaining on the solid phase is reacted
radiolabel. Estradiol is bound to protein, and these chemically to release light. An inverse relationship exists
assays used organic extraction with diethyl ether prior to between the amount of estradiol in the patient sample
assay of total estradiol levels. Early commercial and the amount of relative light units (RLUs) detected by
immunoassay kits achieved separation of free from the ACS:180 system.
bound with dextran-coated charcoal [1]. Later
immunoassays used estradiol-6-(o- Comparison of eight automated, non-isotopic estradiol
carboxymethyl)oxime-[125I]-histamine as tracer. Organic assays with an isotope-dilution gas chromatography
extraction was replaced by chemical agents such as mass spectrometry (ID-GC-MS) reference method
danazol, dihydrotestosterone, mesterolone, or showed variable agreement, with slopes of 0.87 to 1.20
anilinonaphthalene sulfonic acid to displace estradiol observed. Further, the interassay precision at an estradiol
from its binding proteins [2]. These direct estradiol concentration of 18 pg/mL (66 pmol/L) varied from
immunoassays used either double-antibody techniques or 6.9% to 42.6% [5]. Liquid chromatographytandem
antibody bound to solid phase for separation of free mass spectrometry (LC-MS/MS) demonstrated poor
tracer from antibody-bound tracer. Further SHBG agreement with direct immunoassay and observed
interference [3] and binding of iodinated tracers to interference in immunoassays at low estradiol
SHBG [4] has been reported in direct assays, leading to concentrations [6]. The functional sensitivity of this LC-
potentially discordant results. MS/MS method was 6.3 pg/mL (23.2 pmol/L). In
general, current direct estradiol immunoassays have a
Direct radioisotopic methods have been replaced with wide measuring range, with the desired aim of a single
more sensitive, direct, automated non-isotopic assay for clinical application in menopause (low levels)
proprietary assays for total estradiol. One example is the and assisted reproduction (high levels). This analytical
Vitros immunodiagnostic method by the Ortho Clinical tradeoff has not been successful, and as a result, the
Diagnostics Division of Johnson and Johnson functional sensitivity of these immunoassays is
(Rochester, NY), which uses a competitive inadequate for use at low concentrations. Future
immunoassay technique. The estradiol in the sample estradiol immunoassay design may require two reagent
competes with estradiol labeled with horseradish formulations with different antibody concentrations,
peroxidase for biotinylated antibody to estradiol. The providing for precise measurement of estradiol at either
antibody complex is captured by streptavidin bound to low- or high-concentration ranges. It has been suggested
plastic wells. The peroxidase is used to catalyze the that extraction of samples prior to immunoassay allows
oxidation of a luminol derivative with the production of for the measurement of low estradiol levels [7,8].
light. The Siemens (Ciba Corning Bayer Corporation) Overestimation of estradiol levels at concentrations less
Advia Centaur or ACS estradiol procedure is a than 30 ng/L (110 pmol/L) has been observed for all
competitive chemiluminescent immunoassay. Estradiol types of immunoassay with indirect (extraction)
in the patient serum binds to an acridinium ester (AE)- immunoassays, correlating better with ID-GC-MS than
labeled mouse monoclonal antibody (Lite Reagent). direct immunoassays [9]. The need to measure estradiol
in clinical practice for men, children, and
postmenopausal women with either fracture risk or
i
Estradiol (E2) breast cancer risk, together with the inaccuracy and
Previous and current authors of this method: imprecision of immunoassays at low estradiol levels, has
First edition: Not done led to the development of MS methods which can
Methods book: Not done measure low levels of estradiol using low sample
Second edition: Not done volumes [10,11,12,13,14]
Third edition: Not done
Fourth edition: Unidentified author
Fifth edition: Greg Ward
517
Estradiol
It has been recommended that preanalytical factors and

Reference and Preferred Methods procedures, both technical and biological, which
There are currently three reference methods for estradiol contribute to error in the measurement of estradiol be
[15,16,17]. Of the 1255 laboratories reporting estradiol addressed and standardized prior to method
levels in the 2007 College of American Pathologists harmonization and development of reference intervals
(CAP) Survey Participant Summary Report, [22].
approximately 98.4% used direct non-isotopic
immunoassay. Two laboratories used mass spectrometry, Interferences
and 18 (1.5%) used radioimmunoassay. Specimens collected in glass gel-separator tubes have
been shown to have significantly lower results compared
The problems with estradiol assays are the same as those with specimens collected in plain evacuated tubes [23].
described for testosterone assays, with repeated calls for
standardization [18,19]. The availability of reference Direct assays have been reported to yield incorrect
materials linked to reference methods will allow for results in neonates [24]. Heterophile antibodies such as
appropriate standardization of methods by manufacturers human anti-mouse antibody can interfere in many
and laboratories. The increasing clinical requirement for immunoassays.
measurement of low estradiol levels will see further
development and utilization of MS methods in routine Samples from patients receiving estrogen replacement
laboratories [20]. therapy, including Premarin, may yield spurious
Specimen estradiol results because this replacement therapy can
The choice of serum or plasma is dependent in part on suppress estradiol release. Further cross-reactivity of
the manufacturer of the test system. For example, for the exogenous estrogens in estradiol assays is variable.
ACS 180 system, serum is the preferred specimen, but
for the Vitros system, serum, EDTA, or heparinized Unconjugated estriol has been reported to interfere in
plasma are all acceptable. The sample should be some estradiol immunoassays [25].
refrigerated if not analyzed immediately. Exposure to air
should be minimized. Samples may be stored at 20C Estradiol Reference Interval
but not frozen and thawed more than once [21]. It is Early discussions of estradiol reference intervals
important that information relating to the clinical status considered total estradiol composed of the conjugated
of the woman with respect to the luteal cycle accompany and unconjugated forms [26]. The reference interval is
the specimen. dependent on the stage of the luteal cycle of the woman.
Reference Intervals for Adults, Reported in pmol/L and pg/mL

Patient Vitros Vitros ACS 180 Univ. Virginia
pmol/L pg/mL pg/mL pg/mL
Female follicular 97-592 26-161 28-172 30-65 early
Pre-ovulary peak 685-1404 187-382
Female luteal 120-738 32-201 55-246 30-200 mid
Postmenopausal 19.7-141 5.4-38 13-93 <30

Males 19.7-242 5.4-66 21-76 30-56
Interpretation luteum [28]. During pregnancy, the placenta is a source

The adrenals, testes, ovaries, and placenta all produce of estradiol secretion. The primary function of estradiol
estradiol. In serum or plasma, 1% to 3% is nonprotein is to stimulate growth of the female sex organs and
bound, 40% is bound to sex-hormone-binding globulin development of secondary sexual characteristics.
(SHBG), and the remainder is bound to albumin [2]. The
principal and most potent estrogen secreted by the ovary Estradiol plays an essential role throughout the human
is estradiol-17. Estradiol promotes the development of menstrual cycle [28]. For interpretation, it is important to
the female secondary sexual characteristics, uterine know the clinical status of the woman with respect to the
growth, thickening of the vaginal mucosa, thinning of luteal cycle. During the early follicular phase, the
the cervical mucus, and development of the ductal estradiol level is relatively constant and low. By day 7,
system of the breast. the dominant follicle is established, and the estradiol
level rises significantly. The elevated estradiol level
Measuring the circulating level of estradiol is important suppresses the FSH level by negative feedback on the
for assessing ovarian function and monitoring follicular hypothalamus and pituitary gland and triggers a rapid
development for assisted-reproduction protocols [27]. In rise of LH. The estradiol level falls significantly as LH
normal, nonpregnant females, estradiol is secreted reaches its peak. Normally, ovulation occurs 10 to 12
mainly by the combined function of the theca and hours after the LH peak and 24 to 36 hours after the
granulosa cells of the developing follicle and the corpus estradiol peak. During the luteal phase, the estradiol
518
Estradiol
level increases again, reaching its peak luteal level about Oestradiol assays: applications and guidelines
8 days after ovulation. The elevated estradiol level is for the provision of a clinical biochemistry
involved in the regression of the corpus luteum. Unless service. Ann Clin Biochem 1988;25:466-83.
fertilization of the ovum takes place, the estradiol level 3 Masters AM, Hahnel R. Investigation of sex-
decreases, signaling the start of a new cycle. hormone binding globulin interference in direct
immunoassays for testosterone and estradiol.
Elevated estradiol levels in females may also result from Clin Chem 1989;35:979-984.
primary or secondary ovarian hyperfunction [29]. Very 4 Micallef JV, Hayes MM, Latif A, Ahsan R,
high estradiol levels are found during the induction of Sufi SB. Serum binding of steroid tracers and
ovulation [27] for assisted reproduction therapy or in its possible effect on direct steroid
pregnancy. Automated immunoassays have suitable immunoassay. Ann Clin Biochem 1995;32:566-
performance for these protocols [30,31]. Decreased 574.
estradiol levels in females may result from either the 5 Yang DT, Owen WF, Ramsay CS, Xie H,
lack of ovarian synthesis (primary ovarian hypofunction Roberts WL. Performance characteristics of
and menopause) or a lesion in the hypothalamus- eight estradiol immunoassays. Am J Clin Pathol
pituitary axis (secondary ovarian hypofunction). 2004;122:332-337.
Estradiol levels are normally low in males. Elevated 6 Nelson RE, Grebe SK, OKane DJ, Singh RJ.
estradiol levels in males may be due to increased Liquid chromatography-tandem mass
aromatization of androgens, resulting in gynecomastia spectrometry assay for simultaneous
[29,32]. measurement of estradiol and estrone in human
Estrogen replacement therapy can effect estradiol 7 Dighe AS, Sluss PM. Improved detection of
concentrations, complicating interpretation. Target serum estradiol after sample extraction
concentrations for therapeutic estradiol levels have been procedure. Clin Chem 2004;50:764-766.
published; however, use of these target values is not 8 Ankarberg-Lindgren C, Norjavaara E. A
valid, owing to some or all of assay variability, lack of purification step prior to commercial sensitive
method documentation, lack of reference to timing of immunoassay is necessary to achieve clinical
sample in relation to patch change or gel application, usefulness when quantifying serum 17beta-
small number of women in studies, failure to specify estradiol in prepubertal children. Eur J
type of patch studied, and lack of standardization of gel Endocrinol 2008;158:117-124.
area [33]. 9 Lee JS, Ettinger B, Stanczyk FZ, Vittinghoff E,
Hanes V, Cauley JA, Chandler W et al.
Estradiol Performance Goals Comparison of methods to measure low serum
Survey data from the CAP 2007 participant summary estradiol levels in postmenopausal women. J
report shows imprecision values (% coefficient of Clin Endocrinol Metab 2006;91:3791-3797.
variation [CV]) for estradiol ranging from approximately 10 Santen RJ, Demers L, Ohorodnik S, Settlage J,
3% to 18% at a mean concentration of 360 pg/mL and Langecker P, Blanchett D, Goss PE, Wang S.
from 3% to 23% at a mean concentration of 1608 pg/mL. Superiority of gas chromatography/tandem
mass spectrometry assay (GC-MS/MS) for
Acceptable Clinical Laboratory Improvement estradiol for monitoring aromatase inhibitor
Amendments performance criteria (CLIA 88) for therapy. Steroids 2007;72:666-671.
measurement of estradiol requires that laboratories be 11 Hsing AW, Stanczyk FZ, Blanger A,
accurate to within 3 SD of the peer-group mean. Schroeder P, Chang L, Falk RT, Fears TR.
Within-subject biological variation has been determined Reproducibility of serum sex steroid assays in
to be approximately 11.3% [34]. Desirable specifications men by RIA and mass spectrometry. Cancer
for analytical imprecision derived from studies of Epidemiol Biomarkers Prev 2007;16:1004-
biological variation indicate that an assay imprecision of 1008.
no greater than 5.15% is required for an individual 12 Kushnir MM, Rockwood AL, Bergquist J,
laboratory. Varshavsky M, Roberts WL, Yue B, Bunker
AM, Meikle AW. High-sensitivity tandem mass
Laboratories will meet CLIA 88 performance criteria, spectrometry assay for serum estrone and
since this only requires agreement with the same estradiol. Am J Clin Pathol 2008;129:530-539.
method. Estradiol methods currently do not meet the 13 Guo T, Gu J, Soldin Op, Singh RJ, Soldin SJ.
precision, specificity, and accuracy required for clinical Rapid measurement of estrogens and their
use [20]. Harmonization of immunoassay methods metabolites in human serum by liquid
against reference methods and the use of MS at low chromatographytandem mass spectrometry
estradiol levels is required to address these issues. without derivatization. Clin Biochem
2008;41:736-741.
References 14 Sluss PM, Hayes FJ, Adams JM, Barnes W,
1 Toro G, Ackerman PG. Practical Clinical Williams G, Frost S, Ramp J et al. Mass
Chemistry. Boston: Little Brown & Co; spectrometric and physiological validation of a
1974:571-572. sensitive, automated, direct immunoassay for
2 Ratcliffe WA, Carter GD, Dowsett M et al. estradiol using the Architect. Clin Chim Acta
519
Estradiol
2008;388:99-105. DL. Analytical performance of a new two-step

15 Siekmann L. Determination of oestradiol-17 ADVIA Centaur estradiol immunoassay during
beta in human serum by isotope dilution-mass ovarian stimulation. Clin Chem Lab Med
spectrometry. Definitive methods in clinical 2006;44:105-109.
chemistry. II. J Clin Chem Clin Biochem 31 Taieb J, Mendez Lozano DH, Benattar C,
1984;22:551-557. Messaoudi C, Pos C. Enlightenment about the
16 Thienpont LM, Verheghe PG, Van Brussel KA, new Architect-i2000 estradiol (Abbott
De Leenheer AP. Estradiol-17 beta quantified Laboratories) immunoassay during in-vitro
in serum by isotope dilution-gas fertilization. Clin Biochem 2007;40:1423-1426.
chromatography-mass spectrometry: reversed- 32 Martin RM, Lin CJ, Nishi MY, Billerbeck
phase C18 high-performance liquid AEC, Latronico AC, Russell DW, Mendonca
chromatography compared with immuno- BB. Familial hyperestrogenism in both sexes:
affinity chromatography for sample clinical, hormonal, and molecular studies of
pretreatment. Clin Chem 1988;34:2066-2069. two siblings. J Clin Endocrinol Metab
17 Tai SSC, Welch MJ. Development of a 2003;88:3027-3034.
reference measurement procedure for the 33 Armston A, Wood P. Hormone replacement
determination of estradiol-17 in human serum therapy (oestradiol-only preparations): can the
using isotope-dilution liquid chromatography laboratory recommend a concentration of
tandem mass spectrometry. Anal Chem plasma oestradiol to protect against
2005;77:6359-6363. osteoporosis? Ann Clin Biochem 2002;39:184-
18 Middle J. Standardization of steroid hormone 193.
assays. Ann Clin Biochem 1998;35:354-363. 34 Fraser CG. Biological Variation: From
19 Stanczyk FZ, Lee JS, Santen RJ. Principles to Practice. Washington, DC: AACC
Standardization of steroid hormone Press; 2001.
immunoassays: why, how, and when? Cancer
Epidemiol Biomarkers Prev 2007;16:1713-
1719.
20 Demers LM. Testosterone and estradiol assays:
current and future trends. Steroids 2008;May 15
Epub.
21 Diver MJ, Hughes JG, Hutton JL, West CR,
Hipkin LJ. The long-term stability in whole
blood of 14 commonly-requested hormone
analytes. Ann Clin Biochem 1994;31:561-5.
22 Raff H, Sluss PM. Pre-analytical issues for
testosterone and estradiol assays. Steroids
2008;May 21 Epub.
23 Reinartz JJ, Ramey ML, Fowler MC et al.
Plastic versus glass SST evacuated serum-
separator blood-drawing tubes for
endocrinologic analytes. Clin Chem
1993;39:2535-2536.
24 Diver MJ. Plasma estradiol concentrations in
neonates. Clin Chem 1987;33:1394.
25 Cao Z(T), Swift TA, West CA, Rosano TG, Rej
R. Immunoassay of estradiol: unanticipated
suppression by unconjugated estriol. Clin Chem
2004;50:160-165.
26 Tietz NW. Clinical Guide to Laboratory Tests.
Philadelphia: Saunders; 1983:180.
27 Muse K, Wilson EA. Monitoring ovulation
induction: use of biochemical and biophysical
parameters. Semin Reprod Endocrinol
1986;4:301-09.
28 Speroff L, Glass RH, Kase NG, eds. Clinical
Gynecologic Endocrinology and Infertility. 4th
ed. Baltimore: Williams & Wilkins; 1989:91-
119.
29 Kicklighter EJ, Norman RJ. The gonads. In:
Kaplan LA, Pesce AJ. Clinical Chemistry:
Theory, Analysis, Correlation. 2nd ed. St Louis:
Mosby; 1989:650-663.
30 Massart C, Gibassier J, Laurent MC, Le Lannou
520
Estriol
Estriol
Lawrence A. Kaplan
Name: Estriol, 1,3,5,(10)-estratriene-3,16, 17-triol, E3

Clinical significance: Refer to Chapter 44, Pregnancy, and Chapter 50, The Gonads, in the 5th
edition of Clinical Chemistry: Theory, Analysis, Correlation
Merck Index: 3654
Structure:
Principles of Analysis and Current Usage already significant biological and analytical errors. The
Estriol is the major estrogen steroid produced during introduction of reporting estrogen output normalized to
pregnancy. The concentration of plasma and urinary creatinine output, the estrogen/creatinine (E/C) ratio[1,2]
estriol rises a thousandfold during pregnancy, and by the improved the assay by eliminating the need for an
third trimester estriol represents 85% to 90% of the C18 accurate, timed collection. The premise of the E/C ratio
estrogenic steroids present in plasma and urine. The is that the kidneys excrete both the conjugated estriol
serum and urine levels of the estriol derivatives are and creatinine by similar mechanisms and that a random
compared in Estriol Table: Estriol metabolites present in urine measured for an E/C ratio yields results as valid as
plasma and urine. Urinary estriol is present only in the results obtained from the longer collection.[3]
conjugated forms, since the protein-bound, unconjugated
estriol is not filtered at the glomerulus. Urinary estriol output is dependent not only on fetal and
placental biochemical pathways that result in the
Estriol is most frequently measured as part of the synthesis of estriol, but on maternal hepatic conjugation
biochemical monitoring of high-risk pregnancies, and and renal clearance of the estriol as well. It has therefore
thus the sample analysis time required before a physician been suggested that monitoring urinary estrogen is less
can obtain a result is an important consideration when valid than measurement of serum estriol for the
one is choosing a method. The type of method used will assessment of the status of the fetoplacental unit (fetus
depend most importantly on the decision to monitor and placenta) because of the dependence of the urinary
estriol in either serum or urine. In either case, the measurement on both maternal and placental
monitoring is usually initiated in the beginning of the function.[4,5] The measurement of plasma estriol has
third trimester, when estriol levels rise to measurable been recommended, since it monitors only the
amounts. The absolute measurement at any one time is fetoplacental unit. Unconjugated estriol, with a half-life
less important than changes over time. of only 20 min, reflects most specifically the output of
the fetoplacental unit at the time of phlebotomy.[4,6]
The urinary procedures were traditionally performed on
a 24-h sample. This is a difficult specimen to collect Few studies have demonstrated an overwhelming utility
accurately and repeatedly. When reported as milligrams of one measurement (urine versus plasma) over the
of estrogen per 24 h, the collection error is added to the other. Many studies have demonstrated a good
correlation between plasma estriol and urinary estriol
measurements; included are studies both on entire
i
Estriol populations and on individual, high-risk
Previous and current authors of this method: pregnancies.[7,8] Estriol output by the fetoplacental unit
First edition: Lawrence A. Kaplan is variable, and circadian rhythms and hour-to-hour
Methods edition: Lawrence A. Kaplan fluctuations have both been noted.[9,10] These
Second edition: Lawrence A. Kaplan variations appear to affect the concentration of plasma
Third edition: Not updated estriol more than urinary estriol. Urinary estrogens, the
Fourth edition: Lawrence A. Kaplan E/C ratio, and total serum estriol seem to have maximal
Fifth edition: Not updated values at around midday with minimum concentrations
noted about midnight; the daytime concentrations for
521
Estriol
serum free estriol are greater than the nighttime chromophore is excited at 530 nm, an intense, yellow-
values.[9] green fluorescence, which can be monitored at 550
nm,[18,20] is produced. The fluorescent reaction has
Urine been adapted to continuous-flow, automated
The estrogen composition of urine of a pregnant woman analysis.[16,21,22] Ittrich also found that hydroquinone
in the third trimester is approximately 80% to 95% can enhance the color formed in the Kober reaction.[23]
estriol conjugates (sulfate and glucuronate) with the Trichloroacetic acid has been used in place of p-
remainder consisting of conjugates of estrone and nitrophenol.[24]
estridiol.[11,12] The oldest, and still most widely used,
procedure for the analysis of estrogen in urine is the Brown and his co-workers employed several extraction
colorimetric reaction developed by Kober in 1931.[13] steps to separate estriol from estrone and
This method is based on the reaction of all estrogens estradiol.[25,26] The estrogens were further purified
with sulfuric acid to form a pinkish red Kober and separated on alumina columns, where the fractions
chromogen of a yet unknown structure[14] (method 1, were monitored by the Kober reaction. This complex
Estriol Methods Summary Table). Oliver et al. suggested procedure allows for quantitation of each individual
that at least three chromogenic reaction products that estrogen (estriol, estrone, and estradiol).
form a colloidal suspension of differing particle size are
produced by the Kober reaction.[15] Ferrous sulfate is Urinary estriol can be specifically quantitated by a
often added to the color reaction to serve as a catalyst variety of chromatographic techniques, all of which
and to minimize certain interferences.[14-16] Since most require a hydrolysis step to convert estriol to the
of the urinary estrogen in the third trimester is estriol, unconjugated form. In addition, all chromatographic
this assay is effective in monitoring changes in excretion assays require an extraction and partial purification of
of estriol during pregnancy though it is not specific for the estrogens before analysis. Liquidliquid partitioning,
this compound. column extractions, and carbon black adsorption[27]
have been used successfully. Gasliquid
The Kober reaction is usually performed after a chromatographic techniques usually require an
hydrolysis step (acid or enzymatic) to convert extraction of the hydrolyzed estrogens before volatile
conjugated estriol into the free form. Performing the derivatives, such as acetate, methyl, and tetramethylsilyl
hydrolysis step before extraction has been shown to conjugates, are formed. Most separations have been
increase the final estrogen yield by about 20%, but for a performed on 3% OV-1 columns, with nitrogen gas as
screening procedure this may.[17] Hydrolysis of the the mobile phase and flame ionization as the detection
estriols before extraction can also lead to interferences. method.[28,29] Some methods utilize temperature
The Kober reaction can be performed directly on the programming (method 3, Estriol Methods Summary
hydrolyzed urine or on unhydrolyzed or hydrolyzed Table). High-performance liquid chromatography
estrogens that have been first extracted into an organic (HPLC) analyses of urinary estriol using reversed-phase
phase (such as ethyl acetate), which is removed by columns have been reported. These methods employ
evaporation. In the latter case, the Kober reaction is then several different types of detection methods (method 4,
performed on the residue. A pinkish red color is formed Estriol Methods Summary Table) including
with estrogens, and a brownish color results from measurement of absorbance at 280 nm (see Estriol:
nonestrogenic compounds that are also extracted. Since Figure 1 for absorbance spectrum),[30] measurement of
the interfering compounds produce a linear background native fluorescence at 308 nm after excitation at 220 nm
absorbance over the spectral region about the absorbance (see Estriol: Figure 2 for the fluorescence spectrum),[31]
maximum of the Kober chromogen,[17] the Allen and derivatization with dansyl chloride and measurement
correction has been used to correct the absorbance for of the fluorescent derivative.[32]
background interference. The Kober chromogen in
aqueous solutions has an absorbance maximum at 514 Immunoassays developed for measurement of urine
nm, and the Allen correction measurements are made at estriol include radioimmunoassay (RIA) and enzyme
472 and 556 nm.[17] immunoassays (EIA) (methods 5 and 6, Estriol Methods
Summary Table). These assays require a hydrolysis step
An important modification of the Kober reaction was the to convert estriol conjugates to the free form. Several
extraction of the Kober chromophore into chloroform
RIAs have been reported using tritium (3H) or
(containing 2% p-nitrophenol plus 1% ethanol), the so-
called Ittrich extraction.[18] This extraction step was radioactive iodine (125I) labeling and employing a
designed to remove the estrogen reaction product from variety of techniques for separating bound from free
other, more water-soluble, interfering chromogens. The label, including dextran-coated charcoal,[33] double-
Ittrich chromogen can be measured colorimetrically or antibody precipitation,[34] and ammonium sulfate
fluorometrically. In chloroform, the chromophore has an precipitation.[35] The EIA is a heterogeneous assay
absorbance maximum between 530 and 540 nm[17,18] requiring polyethylene glycol to separate the antibody-
(method 2, Estriol Methods Summary Table). The Allen bound estriolenzyme conjugate from the free label.[36]
correction has also been applied to the chloroform A unique luminescence immunoassay has been
extract, with the secondary absorbance measurements described for the measurement of total urinary
made at 505 and 565 nm.[19] When the extracted estrogens.[37] After enzymatic treatment of the urine to
522
Estriol
hydrolyze the conjugates, the estrogens are allowed to reversed-phase chromatography (method 9, Estriol
compete for antibody binding sites with 17-estradiol Methods Summary Table). Detection and quantitation of
conjugated to aminobutylated isoluminol. The amount of the estriol peak is either by electrochemical analysis at
light emitted in the subsequent luminescence assay is +0.75 volts using a glassy carbon electrode[51] or by
inversely related to the concentration of total estrogens fluorescence at 308 nm after excitation at 280
in the sample. nm.[52,53]
Plasma Estriol A fluorescence polarization assay (method 10, Estriol

The thousandfold lower estriol concentration in serum as Methods Summary Table) for the measurement of total
compared to urine has, for the most part, precluded the estriol in serum, plasma, or urine is also available.[54]
use of colorimetric or fluorometric assays[33] for This assay, which is partially automated, requires no
analysis of blood estriol. Most of the assays for serum extraction of the sample. Hydrolysis of total (conjugate
estriol, whether one is measuring total or only the and free) estriol is achieved by the use of Escherichia
unconjugated fraction, have been RIAs (method 7, coli glucuronidase to hydrolyze the glucuronide
Estriol Methods Summary Table). The RIAs use either conjugate.
3H- or 125I-estriol as the tracer and employ a variety of
procedures to separate bound from free label,[38] Saliva
including polyethylene glycol,[39] ammonium Saliva has been investigated as a possible sample for the
sulfate,[40] a second precipitating antibody,[10,41] and analysis of estriol. Salivary unconjugated estriol has
solid-phase bound systems.[8,42] The use of the last been measured by radioimmunoassay (RIA), either as a
technique has led to the automation of the assay.[43] manual assay, employing 3H-estriol and a charcoal
Measurement of total serum estriol requires the adsorption step for separating bound from free
hydrolysis of estriol conjugates to the free form before tracer,[55] or as an automated, continuous-flow RIA
the analysis; either acid or enzyme hydrolysis is used. using 125I-estriol as the tracer.[40]
Since there can be significant antibody cross-reactivity Other Measurements
between unconjugated estriol and its conjugated forms Although most analyses for estriol measure either the
(present in five to 10 times higher concentrations), the unconjugated fraction or total estriol after a hydrolysis
analysis for free estriol often requires separation of the step, assays have been developed that allow
free estriol from its conjugates. This is most frequently measurement of the individual conjugates of estriol. This
accomplished by a liquidliquid extraction step, in includes a radioimmunoassay that specifically measures
which the less water-soluble, unconjugated estriol is estriol 3-sulfate 16-glucuronide[56] and an HPLC assay
extracted into the organic phase with high efficiency. that measures three other conjugates of estriol in
Alternatively, gel exclusion chromatography (such as addition to the diconjugate.[57] There does not appear
Sephadex) has been used for this purpose.[44] RIAs that to be widespread clinical use of this type of analysis at
directly measure serum unconjugated estriol without a this time.
prior extraction step are also available.[38,45,46] These
methods employ solid-phase procedures[38,46] and Reference and Preferred Methods
double-antibody techniques to separate bound from free There is no reference method for the measurement of
label and can be automated as well.[46] either urine or serum estriol. The lack of a reference
method may be partly ascribed to the lack of a consensus
The measurement of total serum estrogens by EIA has as to how to best monitor estrogens during pregnancy.
been reported by several groups.[47,48] These assays
are not specific for estriol but measure total serum Urine
estrogens, including conjugates of estriol. These assays The original colorimetric method for the measurement of
utilize a horseradish peroxidase antibody label, and the urinary estriol, the Kober reaction, has sufficient
reaction is monitored at 463 nm (method 8, Estriol sensitivity (0.2 g) to detect only total pregnancy
Methods Summary Table). An EIA that allows direct estrogens in urine. When the product of the Kober
measurement of serum unconjugated (as well as reaction is measured by fluorometry, the sensitivity can
conjugated) estriol is also available.[49] This assay be increased to the low-nanogram range needed to
employs a double-antibody precipitation technique to measure serum estriol.[33] However, the poor
separate bound from free tracer. A competitive specificity of the method and the variability of
heterogeneous EIA has been developed for endogenous fluorescence reduce the usefulness of this
determination of unconjugated estriol in serum.[50] assay unless the chromogen is further purified. Both the
This assay consists of a fluoresceinated capture colorimetric and fluorometric procedures have been
antibody, an estriol labeled alkaline phosphatase adapted to automated, continuous-flow
conjugate, and a magnetic particle solid phase. The assay analysis[16,22,24,58] to increase the throughput and
produces P-nitrophenoxide, which is measured at 405 precision. Manual assays have reported a between-day
nm. HPLC assays for unconjugated serum estriol have percent coefficient of variation (%CV) of about
also been reported.[51,52,53] The estriol is extracted by 7.5%,[17] whereas the automated assays have reported
liquidliquid partitioning or adsorption chromatography CVs of 4% to 5% (X, 12 to 29 mg/L).[22]
and is separated from interfering compounds by
523
Estriol
The HPLC analysis of urinary estriol with fluorescence because of the cost of reagents, most laboratories batch
detection may have the advantage of increased their estriol analyses and cannot provide clinicians with
sensitivity and fewer possible interfering estriol measurements throughout the day. Low-cost,
chromatographic peaks. Gas chromatographic analysis is rapid, automated HPLC analysis also can easily provide
probably not suited for clinical use because of the labor an alternative if results are to be available 24 hours a
and time necessary for such assays. Measurement of day. Estriol Table: Comparison of representative assays
urinary estriol by RIA has good precision and compares the assay characteristics of several methods of
specificity. estriol analysis. Specimen
Either 24-h collections or single voidings are appropriate
Plasma urine specimens. Serum or plasma specimens are both
The antibody used to measure total estriol after a valid for blood analysis. When one is monitoring
hydrolysis step usually has a cross-reactivity of less than patients with consecutive sample analyses, the random
1% to 2% with estradiol or estrone. Similarly, when one urine or serum specimens should be obtained at
is measuring free serum estriol, the antibody should have approximately the same time of the day to minimize any
low cross-reactivity with the conjugated forms of estriol, error caused by biological variation. It has been shown
unless an extraction step is used to separate the that bacterial contamination of urine left unpreserved
conjugated from the unconjugated forms. leads to an apparent increase in estrogens measured by
an automated fluorometric assay.[59] Preservation of
125I seems to be the label of choice in most RIAs. The urine during collection with thimerosal (about 25 mg/L)
use of a second, precipitating antibody to separate bound or storage at 4 C effectively prevents changes in
from free radioactivity may have a small advantage over measured estrogen.
other techniques (polyethylene glycol, ammonium
sulfate) in terms of reproducibility of technique.[10] An Certain antibiotics, most notably ampicillin, have been
important advantage of solid-phase RIAs is that they shown to cause a decrease in the urinary excretion of
can be readily adapted to automated analysis with a estriol conjugates by reducing the intestinal bacteria that
concomitant increase in precision.[40,46] Automated, hydrolyze estriol conjugates.3 This reduces the
solid-phase free estriol assays are commercially recirculation of estriol into serum and its eventual
available.[46] urinary excretion.
In addition to having the capability of being Interferences

semiautomated, the HPLC assays are more flexible in The direct Kober assay, performed on prehydrolyzed,
terms of performing one or many analyses per analytical unextracted urine, is subject to interferences from a wide
run at a reasonable cost. The very quick start-up times number of compounds, especially those that form
allow HPLC to be used to provide rapid stat analysis. acetaldehyde during the acid hydrolysis step. The
acetaldehyde reacts with the hydroquinone to give false-
The EIAs are quite sensitive and are able to detect negative results. One of the most important of these
approximately 0.2 nmol of estriol. However, many of interfering compounds is glucose, which, in the
these assays suffer from a lack of specificity; they have monitoring of the high-risk pregnancies of diabetic
significant cross-reactivity with several estrogenic women, can cause significant negative interferences.
compounds and are primarily designed to measure total Glucose, at concentrations as low as 2.5 g/L, and
plasma estrogens. methenamine mandelate, which interfere with the acid
hydrolysis step, cause an overall low recovery of estriol
The method of choice for the monitoring of pregnancy and false-negative results. In addition to glucose,
estrogens will depend on factors to be evaluated by each galactose, lactose, and mannose have also been found to
laboratory: cost, ease and reproducibility of analysis, and interfere.[60] Glucose can also reduce the fluorescence
the clinical value of the information obtained. The lower produced by estrogens in fluorescence assays.[61] One
biological day-to-day variability of the random urine can reduce interference by glucose by performing the
estrogen/creatinine (E/C) ratio and the low cost and acid hydrolysis on diluted urine or by extracting estriol
rapid turnaround time for results from the colorimetric before performing the Kober reaction.[17] If an initial
assay still make this test an important method for chemical reduction step is performed with sodium
consideration. The analysis time for a random urine E/C borohydride (NaBH4) to eliminate glucose interference
analysis is approximately the same as for free serum during a direct Kober reaction or automated fluorescence
estriol by RIA. A rapid manual method for urinary assay, accurate results can be obtained without the
estrogens is presented below. For those laboratories with necessity of making multiple readings for an Allen
large numbers of samples, an automated Kober method correction.[21,61] The use of enzymatic hydrolysis can
on continuous-flow analysis would be appropriate. certainly prevent the interference from glucose but
requires a much longer time for the incubation step. A
For those laboratories asked to measure plasma estriol, general correction for nonspecific color can still be made
the free estriol analysis provides physicians with a using the Allen correction.
slightly more rapid turnaround time than total serum
estriol assays (because there is no hydrolysis step) and
gives results with less daily fluctuation. However,
524
Estriol
Estriol Reference Interval will be diminished, decreasing the levels of conjugated

The reference interval for both urinary and serum and estriol in plasma. It is important that the status of
plasma estriol varies with gestational age. The biological maternal organ function be considered when one is
variability of the fetoplacental unit (as the 95th interpreting the results of urine or serum estriol (total or
percentile range) is very broad during the last month of free) measurements, since they can cause changes in the
pregnancy. This is illustrated in Estriol: Figure 4 for results that do not reflect the status of the fetoplacental
plasma estriol. The following reference intervals can unit.
serve to illustrate expected estriol values in the last
trimester of pregnancy. Concentrations are also higher in Given a healthy mother, serum or urine estriol levels can
mothers with twin pregnancy than in those with be very helpful in the monitoring of a high-risk
singleton pregnancies with concentrations just before pregnancy (as in diabetics, women over 35 years of age,
delivery significantly correlated with total birth and drug addicts). The very broad range of serum
weight.[62] Each laboratory must determine its own (Estriol: Figure 4), urine, and salivary estriol levels
reference interval. The reference intervals are shown in indicates that the population variation is much greater
Estriol Table: Reported reference intervals. than the intraindividual variation. Thus, even with estriol
concentrations below the 95th percentile, good outcomes
The levels of salivary estriol (all unconjugated) are can be obtained, though very low values in the last
approximately one-tenth the levels of serum month of pregnancy, less than 4 g/L (13.9 nmol/L) of
unconjugated estriol, with the plasma to salivary estriol free estriol, are a cause for concern. Most often, rather
concentrations ranging from 6.1 to 14.0 ng/mL.[40,63] than be concerned with the absolute concentrations
The daily fluctuations in total serum estriol appear within the reference interval, physicians follow
greater than those of free serum estriol, though the consecutive measurements. A decrease in serum free
percentage of fluctuation for both appears to be estriol of 40% to 45% from the mean of the three
similar.[8] The individual diurnal variations of plasma previous measurements has been considered to be an
estriol can vary from 15% to 45%, with total urinary early indicator of fetoplacental distress.[8] Similar
estrogens varying from 30% to 40% during the day.[10] changes in urine are used to monitor pregnancies.
However, when the estrogen output is reported as an E/C Salivary (unconjugated) estriol levels tend to parallel
ratio, the variation is reduced to 12% to 15%. plasma levels of this compound[63] and can also be used
to monitor pregnancies.[63-65] Although a saliva
Interpretation specimen has the advantages of being readily available
The rise of estriol levels in the blood and urine of a and not requiring venipuncture, relatively little work has
pregnant woman is the result of an interplay of several been performed with it.
physiological and biochemical systems. The fetal adrenal
glands produce the biochemical precursor of estriol, In addition to fetal death, there are several other
dehydroepiandrosterone sulfate (DHEAS) in increasing situations that can decrease maternal estriol levels. These
amounts during development. The DHEAS is are suppression of fetal ACTH because of corticosteroids
hydroxylated to 16-hydroxydehydroepiandrosterone administered to the pregnant woman, suppression of
sulfate (16-OH-DHEAS), which is desulfated to 16-OH- DEAS synthesis if the fetus has pituitary aplasia or the
DHEA by sulfatases in the placenta. Other placental fetus is anencephalic, and deficiency in placental
enzymes metabolize 16-OH-DHEA to estriol, which
sulfatase or 16--hydrolase activity.
crosses into maternal blood. The unconjugated maternal
estriol is rapidly conjugated to glucuronide and sulfate
Although there are reports that the measurement of the
by the maternal liver. Most of the estriol conjugates are
individual conjugates of estriol in urine or serum can
excreted by the maternal kidneys in urine. Some of the
provide clinically useful information, there is no
conjugated estriol is excreted by the liver into the
evidence that such measurements would provide
gastrointestinal tract, where bacterial enzymes hydrolyze
information not already obtained by measurement of free
the conjugates. Most of this estriol is returned to the
or total estriol. With the recent advances in techniques
maternal plasma pool of unconjugated estriol by the
for measuring the conjugates[56,57] perhaps such
enterohepatic circulation.
evidence will be forthcoming.
The levels of unconjugated estriol in maternal serum
Estriol Performance Goals
therefore reflect both a growing fetus and an intact
The interlaboratory precision for urinary RIA, as
fetoplacental unit. However, these levels are also
reported as 11% to 20% at concentrations of 4 to 6
dependent on maternal hepatic and renal function,
mg/L, respectively, which is very similar to the %CV of
intestinal flora, and the enterohepatic circulation.
the spectrophotometric assay.
Decreased hepatic conjugation because of liver disease
might increase the plasma free-estriol pool and decrease
The chromatographic procedures used for specific
urine levels of estriol. Renal disease will decrease
urinary estriol analysis appear to be both accurate and
excretion of estriol conjugates in urine but increase total
precise. Between-day coefficients of variation of 3.7%
estrogen levels in maternal serum. If the mother is being
treated with ampicillin or other antibiotics that can (mean, X, 15.9 mg/L)21 and 8.5% (X, 16.9 mg/L)[31]
inhibit enteric flora, the enterohepatic return of estriol have been reported for HPLC assays. The interlaboratory
525
Estriol
precision for serum estriol analysis has been found to be 13 Kober S. Ein kolorimetrische Bestimmung des
approximately 23% to 29% CV (X, 1 to 2 ng/mL) and Brunsthormons (Menformon). Biochem Z 1931;
6% to 9% CV (X, 25 to 28 ng/mL) for free serum estriol, 239: 209-212.
and 12% to 20% (X, 18 to 20 ng/mL) and 9% to 11% (X, 14 Brown JB. Some observations on the Kober
148 to 152 ng/mL) for total serum estriol.[66] color and fluorescence reactions of the natural
oestrogens. J Endocrinol 1952; 8: 196-210.
References 15 Oliver RWA, Jensen SB, Crossley G.
1 Dickey RP, Besch PK, Borys N, Ullery JC. Am Chemistry of the Kober reaction. Clin Chem
J Obstet Gynecol 1966;94:591-594,. 1976; 22: 679-680.
2 Osofsky HJ, Long RE, OConnell EJ, Jr, 16 Lever M, Powell JC, Peace SM. Improved
Marshall LD. Estrogen excretion during estriol determinations in a continuous flow
pregnancy in a high-risk population. Am J system. Biochem Med 1973; 8: 188-198.
Obstet Gynecol 1971; 109: 1-7. 17 Grannis GF, Dickey RP. Simplified procedure
3 Dickey RP, Grannis GF, Hanson FW. Use of for determination of estrogen in pregnancy
the estrogen/creatinine ratio and the estrogen urine. Clin Chem 1970; 16: 97-102.
index for screening of normal and high-risk 18 Ittrich G. Eine neue Methode zur chemischen
pregnancy. Am J Obstet Gynecol 1972;113: Bestimmung der oestrogenen Hormone in Harn.
880-886. Z Physiol Chem 1958; 312: 1-14.
4 Klopper A. Criteria for the selection of steroid 19 Epstein E, Zak B, Powsner ER, Feldkamp CS.
assays in the assessment of fetoplacental In Sonnenwirth, A.C., and Jarett, L., editors:
function. In Klopper, A., editor: Plasma Gradwohls clinical laboratory methods and
hormone assays in the evaluation of fetal well- diagnosis, St. Louis, 1980, The C.V. Mosby
being, New York, 1976, Churchill Division of Co., pp. 560-561.
Longmens Press, pp. 20-35. 20 Adessi G, Jale MF. Rapid fluorometric
5 Trolle D, Bock JE, Gaede P. The prognosis and estimation of estrogens in urine after twenty
diagnostic value of total estriol in urine and weeks of pregnancy. Ann Biol Clin 1972; 30:
serum and of human placental lactogen 127.
hormone in the last part of pregnancy. Am J 21 Little AJ, Aulton K, Payne RB. Comparison of
Obstet Gynecol 1976;126: 834-842. an all-aqueous automated urine pregnancy
6 Distler W, Gabbe SG, Freeman RK et al. Estriol oestrogen method with one ring organic solvent
in pregnancy. V. Unconjugated and total plasma extraction. Clin Chim Acta 1975; 65: 167-173.
estriol in the management of pregnant diabetic 22 Hammond JE, Phillips JC, Savory J.
patients. Am J Obstet Gynecol 1978; 130: 424- Evaluation of an aqueous fluorometric
431. continuous flow method for measurement of
7 Dubin NH, Crystle CD, Grannis GF, Townsley total urinary estrogens. Clin Chem 1978; 24:
JD. Comparison of maternal serum estriol and 631-634.
urinary estriol determinations as indices of fetal 23 Ittrich G. Untersuchungen ber die Extraktion
health. Am J Obstet Gynecol 1973; 115: 835- des roten Kober-Faberstoffs durch organische
841. Losungsmittel zur strogenbestimmung im
8 Kirkish LS, Barclay ML, Parra JB et al. Plasma Horn. Acta Endocrinol 1960; 35: 34-48.
estriol vs. estrogen assays in 24-hour urines as 24 Simkins A, Worth HGJ. An investigation into
an index to fetal status. Clin Chem 1978; 24: the automated measurement of urinary
1830-1832. oestrogens in pregnancy. Ann Clin Biochem
9 Townsley JD, Dubin NH, Grannis GF et al. 1975; 12: 233.
Circadian rhythms of serum and urinary 25 Brown JB. The determination and significance
estrogens in pregnancy. J Clin Endocrinol of the natural estrogens. In Sobotka, H., and
Metab 1973; 36: 289-295. Stewart, C.P., editors: Advances in clinical
10 Katagiri H, Distler W, Freeman RK et al. chemistry, New York, 1960, Academic Press,
Estriol in pregnancy. IV. Normal Inc., vol. 3.
concentrations, diurnal and/or episodic 26 Beisher NA, Brown JB, Wood C. The value of
variations, and day-to-day changes of urinary estriol measurements during pregnancy.
unconjugated and total estriol in late pregnancy Obstet Gynecol Dig 1968; 10: 37.
plasma. Am J Obstet Gynecol 1976; 124: 272- 27 Andreslini F, DiCorcia A, Lagana A et al.
280. Preliminary isolation of urinary placental estriol
11 Goebelsmann U. The uses of estriol as a before gas or liquid chromatography. Clin
monitoring tool. Clin Obstet Gynecol 1979; 6: Chem 1983; 29: 2076-2078.
223. 28 Pinelli A, Formento ML. A precise and
12 Hobkirk R, Anuman-Rajadhon Y, Nilsen M. sensitive method for the analysis of steroids in
Contribution of estriol to total urinary estrogens small urine samples by thin-layer
during pregnancy. Clin Chem 1970; 16: 235- chromatography and gas-liquid
238. chromatography. J Chromatogr 1972; 68: 67-
75.
526
Estriol
29 Gotelli GR, Kabra PM, Marton LJ. radioimmunoassay kit for the measurement of
Determination of placental estriol in urine by serum free estriol. Clin Chem 1984; 30:
gas-liquid chromatography with equilenin as 1043(A).
internal standard. Clin Chem 1977; 23: 165- 46 Jones TB, Hoffman KL. An automated solid
168. phase RIA for the direct measurement of
30 Gotelli GR, Wall JH, Kabra PM, Marton LJ. unconjugated estriol in serum. Clin Chem 1983;
Improved liquid chromatographic determination 29: 1268.
of placental estriol in urine. Clin Chem 1978; 47 Busch AM, Dijkhuizen DM, Schuwrs AH, ven
24: 2132-2134. Weemen BK. Enzyme immunoassay for total
31 Taylor JT, Krotts JG, Schmidt GJ. estrogens in pregnancy plasma or serum. Clin
Determination of urinary placental estriol by Chim Acta 1978; 89: 59-70.
reversed-phase liquid chromatography with 48 Osterman TM, Juntunen KO, Gothoni GD.
fluorescence detection. Clin Chem 1980; 26: Enzyme immunoassay of estrogen-like
130-132. substances in plasma with polyethylene glycol
32 Schmidt GJ, Vandemark FL, Slavin W. as precipitant. Clin Chem 1979; 25: 716-718.
Estrogen determination using liquid 49 Carter JH, DeBernardi M. Double antibody
chromatography with precolumn fluorescent enzyme immunoassays for unconjugated and
labelling. Anal Biochem 1978; 91: 636-645. total estriol in unextracted serum. Clin Chem
33 Ertel NH, Moskovitz M, Schiffer MA. A 1984; 30: 1043(A).
modification of the rapid method for the assay 50 Readio J, Freeman J, Lee C et al. An automated
of plasma estriol in pregnancy: use of enzyme immunoassay for the direct quantitation
unconjugated 3H to correct for losses. J Clin of unconjugated estriol on the technicon
Endocrinol 1969; 29: 1266-1268. Immuno I system. Clin Chem 1997; 43: S190.
34 Jawad JJ, Wilson EA, Kincaid HL. Improved 51 Kaplan LA, Hohnadel DC. Measurement of
radioimmunoassay for total urinary estriol. Clin unconjugated estriol in serum by liquid
Chem 1979; 25: 99-102. chromatography with electrochemical detection
35 Anderson DW, Goebelsmann U. Rapid compared with radioimmunoassay. Clin Chem
radioimmunoassay for total urinary estriol. Clin 1982; 29: 1463-1466.
Chem 1976; 22: 611-615. 52 Andreolini F, Borra C, DiCorcia A et al.
36 Korhonen MK, Juntunen KO, Stenman UH. Improved assay of unconjugated estriol in
Enzyme immunoassay of estriol in pregnancy maternal serum or plasma by adsorption and
urine. Clin Chem 1980; 26: 1829-1831. liquid chromatography with fluorometric
37 Messeri G, Caldini AL, Bolelli GF et al. detection. Clin Chem 1984; 30: 742-744.
Homogeneous luminescence immunoassay for 53 Kabra PM, Tsai FH, Marton LJ. Liquid
total estrogens in urine. Clin Chem 1984; 30: chromatography with fluorometric detection of
653-657. unconjugated estriol in serum of pregnant
38 Product guide for radioassays and non-isotopic women. Clin Chim Acta 1983; 128: 7-9.
ligand assays, ed. 3, Clin. Chem. 29:891-896, 54 Vanderbilt A, Spring T, Fino J, Shipchandler
1983. M. A fluorescence polarization immunoassay
39 Buster J, Abraham GE. Radioimmunoassay of quantitating total estriol in serum and urine.
plasma dehydroepiandrosterone sulphate. Anal Clin Chem 1984; 30: 1043(A).
Lett 1972; 5: 543. 55 Nguyen HN, Huang NH, Besch NF, Besch PK.
40 Truran PL, Read GF. Automated continuous- Development and evaluation of four
flow radioimmunoassay for salivary estriol. radioimmunoassays for estriol in saliva. Clin
Clin Chem 1984; 30: 1678-1682. Chem 1983; 29: 1254(A).
41 Goebelsmann U et al. Estriol assays in 56 Nambara T, Niwa T, Shimada K. Direct
obstetrics. J Steroid Biochem 1975; 6: 703-709. radioimmunoassay for estriol 3-sulfate 16-
42 France JT, Knox BS, Fisher PR. Evaluation of glucuronide using specific antiserum without
a new commercial solid-phase direct deconjugation. Clin Chim Acta 1985; 149: 275-
radioimmunoassay for unconjugated estriol in 280.
pregnancy plasma. Clin Chem 1982; 28: 2103- 57 Andreolini F, Borra C, Caccamo F, et al.
2105. Estriol and its conjugates in late pregnancy
43 Astill ME, Larriva ML, Greno JV et al. determined by extraction with Carbopack B and
Automated cortisol and free estriol methods liquid chromatography with fluorometric
using 125I and the ARIA system. Clin Chem detection. Clin Chem 1985; 31: 1698-1702.
1981; 27: 1055 (abstract). 58 Strickler HS, Stanchak PJ. Automated
44 Haigh WG, Hoffman LF, Barron EJ. Two-hour fluorometry as applied to the assay of individual
Sephadex column method for assay of estrogens in purified urinary extracts. Clin
unconjugated estriol in serum. Clin Chem 1980; Chem 1969; 15: 137-153.
26: 309-312. 59 Simkins A, Crawley M. Some chemical and
45 Gill-Pazaris L, OMalley M, Hubert D, bacterial contributions to analytical variations in
Mozzicato S. Development of a direct
527
Estriol
urinary oestrogen quantitations during 64 Truran PL, Read GF, Pearson JF. Salivary
pregnancy. Med Lab Sci 1978; 35: 325-334. oestriol in normal and abnormal pregnancies. Br
60 Schindler AE, Ratanasopa V, Herrmann WL. J Obstet Gynaecol 1984; 91: 1210-1215.
Acid hydrolysis of urinary estriol. Clin Chem 65 Kirkish LS, Compton AA, McCann DS.:
1967; 13: 186. Salivary estriol as an index to fetal well-being.
61 Worth HGJ. The effect of sugars on automated Clin Chem 1986; 32: 71-75.
urinary estrogen estimations in pregnancy. Clin 66 All conclusions and interpretations in the
Chim Acta 1973; 49: 53-57. Infobase with respect to the College of
62 Westergaard JG, Teisner B, Hau J et al. American Pathologists data base are those of
Placental protein and hormone measurements in the author of this section or the editors of the
twin pregnancy. Br J Obstet Gynaecol 1982; 92: Infobase, and not those of the College
72-76. 67 Shearman RP, Jools ND, Smith ID. Maternal
63 Evan JJ, Wilkinson AR, Aickin DR. Salivary and fetal venous plasma steroids in relation to
estriol concentrations during normal parturition. J Obstet Gynaecol Br Commonw
pregnancies and a comparison with plasma 1972; 79: 212-215.
estriol. Clin Chem 1984; 30: 120-121.
Tables
Estriol metabolites present in plasma and urine

Estrogen Urine (%) Plasma (%)
Unconjugated estriol 0 9
Estriol 3-sulfate 4 15
Estriol 3-glucuronide 13 15
Estriol 16-glucuronide 73 21
Estriol 3-sulfate 16-glucuronide 10 40
528
Estriol
Methods of estriol analysis

Method 1 (Urine): Kober reaction-direct or antibody, (NH4)2SO4 or dextrancharcoal
indirect (prior extraction) Usage: Frequent
Type of analysis: Spectrophotometric Comment: Requires hydrolysis to free
Principle of analysis: Estriol heated in the estriol; Specific
presence of sulfuric acid and hydroquinone Method 6 (Urine): Enzyme immunoassay
forms a pinkish red chromogen (absorbance (EIA)
maximum, 514 nm). Reaction can be Type of analysis: Spectrophotometric, end
performed with (indirect) or without (direct) point
prior extraction of estrogen. Principle of analysis: Competitive binding
Usage: Historically most popular method; assay; estriolenzyme conjugate separated from
frequently used today free enzyme and enzyme reaction (peroxidase)
Comment: Measures total estrogen; direct monitored at 463 nm
assay can have negative interference by glucose Usage: Rare
Method 2 (Urine): Ittrich Comment: Measures total estriol after
Type of analysis: Fluorometric hydrolysis
Principle of analysis: Chromophore Method 7 (Plasma): RIA free or total estriol
extracted and measured fluorometrically Type of analysis: see method 5
(excitation, 530 nm; emission, 550 nm) or Principle of analysis: See method 5;
colorimetrically extraction required for free estriol, hydrolysis
Usage: Rare for total
Comment: Measures total estrogen; Usage: Most frequent
automated on continuous-flow analyzers Comment: Very specific; automated
Method 3 (Urine): Gas-liquid chromatography Method 8 (Plasma): EIA
Type of analysis: Flame ionization Type of analysis: See method 6
detector Principle of analysis: See method 6
Principle of analysis: Volatile derivatives Usage: Rare
chromatographed on OV-1 with N2 mobile Comment: See method 6
phase and flame ionization detector Method 9 (Plasma): HPLC
Usage: Rare Type of analysis: Electrochemical or
Comment: Requires hydrolysis to free fluorescence
estriol; very specific Principle of analysis: Extracted
Method 4 (Urine): High-performance liquid unconjugated estriol is chromatographed by
chromatography (HPLC) reversed-phase chromatography and detected by
Type of analysis: Ultraviolet or electrochemical or fluorescence detection
fluorescence detection Usage: Rare
Principle of analysis: Estriol Comment: Very specific
chromatographed by reversed-phase (C18) Method 10 (Plasma): Fluorescence
chromatography, detected at 280 nm polarization immunoassay
(ultraviolet); fluorescence (excitation, 220 nm; Type of analysis: Fluorescence
emission, 608 nm); dansyl derivatives detected Principle of analysis: Fluorescent-labeled
by fluorescence estriol competes with sample estriol for binding
Usage: Rare sites; free label results in decreased polarized
Comment: Requires hydrolysis to fluorescence
unconjugated estriol; very specific Usage: Rare
Method 5 (Urine): Radioimmunoassay (RIA) Comment: Measures total estriol in plasma
Type of analysis: Liquid scintillation or and urine
gamma-ray counting
* 3H, tritium; 125I, radioactive iodine
Principle of analysis: Competitive binding
assay, employing 3H- or 125I-labeled estriol*;
free from bound label separated by second
529
Estriol
Comparison of representative assays for urine and serum estriol

Method 1 (Urine): Kober Linearity: 80 g/L
Reaction temperature: 100 C Assay time
Sample volume: 2 mL Incubation/extraction: 75 min
Label or detection system: Spectrophotometric; 514 nm Operational: 30 min
Linearity: 100 mg/L Sensitivity: 1000 ng/L
Assay time Final concentration of reagents: 5070 103 dpm/tube
Incubation/extraction: 20 min Reported precision (between run): X, %CV: 150 g/L,
Operational: 40 min 8.4%
Sensitivity: Interferences: Ampicillin
Final concentration of reagents: Hydroquinone: 180 Method 5 (Serum): RIA (free)
mmol/L; H2SO4; 24.5 mol/L Reaction temperature: Ambient
Reported precision (between run): X, %CV: Sample volume: 250 L
Interferences: Ampicillin Label or detection system: 125I
Method 2 (Urine): HPLC Linearity: 800 g/L
Reaction temperature: 60 C (hydrolysis) Assay time
Sample volume: 5 mL Incubation/extraction: 2.5 h/batch
Label or detection system: Spectrophotometric; 280 nm Operational:
Linearity: 100 mg/L Sensitivity: 100 ng/L
Assay time Final concentration of reagents: 76 103 dpm/tube
Incubation/extraction: 60 min Reported precision (between run): X, %CV: 3900 ng/L,
Operational: 15 min 8.8%
Sensitivity: 23 mg/L Interferences: Estriol 3-sulfate, ampicillin
Final concentration of reagents: Method 6 (Serum): HPLC
Reported precision (between run): X, %CV: 15.9 mg/L, Reaction temperature: Ambient
3.7% Sample volume: 2 mL
Interferences: Ampicillin Label or detection system: Chromatographic,
Method 3 (Urine): RIA* electrochemical detection
Reaction temperature: 27 C (hydrolysis) Linearity: 0.4-8 ng
Sample volume: 20 L Assay time
Label or detection system: 125I Incubation/extraction: 25 min
Linearity: 400 mg/L Operational: 25 min
Assay time Sensitivity: 1000 ng/L
Incubation/extraction: 15 min Final concentration of reagents:
Operational: 90 min Reported precision (between run): X, %CV: 14.9 g/L,
Sensitivity: 3 mg/L 9.9%
Final concentration of reagents: 90 103 dpm/tube Interferences: Ampicillin
Reported precision (between run): X, %CV: 100 mg/L, * Amersham Corp. Arlington Heights, IL.
4.3% Nuclear Medical Systems, Inc., Newport Beach, CA.
Interferences: Ampicillin Becton-Dickinson, Automated RIA, Rutherford, NJ.
Method 4 (Serum): RIA (total) dpm: Disintegrations per minute.
Reaction temperature: 37 C (hydrolysis)
Sample volume: 25 L
Label or detection system: 125I
530
Estriol
Reported reference intervals for total estriol[41,67]

Gestational week: 2830
Plasma, serum total estriol, ng/mL (nmol/L): 38140 (132485)
Urinary total estriol, mg/24 hours (mol/24 h): 518 (1762)
Figures
Estriol: Figure 1
Ultraviolet absorbance
spectrum of estriol in
methanol.
531
Estriol
Estriol: Figure 2
Fluorescence spectrum of estriol in

methanol with excitation wavelength set
at 280 nm.
Estriol: Figure 3
Chromatograms of extracted sera.

nA, Nanoamperes. A, Serum
from nonpregnant woman spiked
with estriol (E3, 10 ng/mL) and
internal standard (I.S., 500
ng/mL). B, Serum from pregnant
woman (50 L injection). C,
Serum from pregnant woman
with spiked estriol (30 L
injection). (Modified from
Kaplan, L.A., and Hohnadel,
D.C.: Clin. Chem. 29:1463-1466,
1982.)
532
Estriol
Estriol: Figure 4
Mean (solid line) and

estimated 5th and 95th
percentiles (shaded area) for
plasma unconjugated estriol
during normal pregnancy.
Estriol patterns form three
actual pregnancy conditions
are shown.
8. Stock standard solution, 500 mg/L (1.73

Procedure: Total Urine Estrogen Determination [17] mmol/L). Dissolve 50 mg of estriol in 100 mL of
absolute ethanol. Stable for 6 months when capped
Principle tightly at 6 to 8 C.
Urine is acidified and saturated with sodium chloride to
9. Working standard, 20 mg/L (6.94 mol/L).
reduce the water solubility of estriol. Estrogens are then
Dilute 2 mL of stock solution to 50 mL with absolute
extracted into ethyl acetate. An aliquot of this extract is
ethanol. This is stable for 2 to 3 months when capped
evaporated, treated with hydroquinone in 64% sulfuric
tightly at 6 to 8 C.
acid, and heated in a boiling water bath to develop the
color. Absorbance readings are taken at 472, 514, and Assay
556 nm, and the Allen correction is used in the Equipment: A spectrophotometer with a band pass < 10
calculations. This method has no interference by glucose nm; ground-glass stoppered, 50 mL extraction tubes;
because of the initial extraction. centrifuge capable of 200 g.
1. Prepare the urine sample by
Reagents thoroughly mixing it and then centrifuging it to
1. 5 M HCl (5 mol/L). Dilute 486 mL of remove solid matter. The specific gravity of
concentrated HCl to 1 L with distilled water. This is each single-voided specimen should be
stable for 1 year at room temperature. measured with a refractometer and adjusted to
2. Crystalline NaCl, reagent grade. 1.015 or less. Prepare a 1:20 dilution of urine
3. Ethyl acetate, reagent grade. for creatinine determination.
4. Nitrogen gas. 2. To each extraction tube, add, in the
5. Hydroquinone in ethanol, 20 g/L (180 following order:
mmol/L). Dissolve 0.20 g of hydroquinone (Fisher, a. 2 mL of urine
purified) in 10 mL of absolute ethanol. Prepare fresh b. 0.4 mL of 5 M HCl
daily. c. 2 g of NaCl
6. Hydroquinone, 20 g/L (180 mmol/L) in 64% d. 2 mL distilled water
H2SO4 (24.5 mol/L). Add 664 mL of concentrated
e. 3 mL ethyl acetate
H2SO4 (98%) very slowly to 336 mL of distilled water
Cap the tube securely, and shake vigorously for
in an ice bath. Mix after each addition, and allow to cool 30 sec. Centrifuge for 5 min at 200 g at room
slightly before adding more acid. Add 20 g of temperature. The top layer is used for
hydroquinone to final solution, and heat to 50 C to subsequent analysis.
dissolve. Store in refrigerator. This solution is stable for 3. To each color development tube, add as shown
2 months. below:
7. Estriol standard (1,3,5,10-estratriene- Ethyl a
3,16,17-triol) (Mann Research Laboratories, New York,
NY).
533
Estriol
column, an ultraviolet detector, and an electrochemical

Ethyl acetate
extract (top
detector. The amount of serum estriol is calculated on
layer) the basis of peak height ratio.
2% Hydroquinone
Tube in ethanol Working standard Reagents
Blank 0.2 mL 1. Diethyl ether. Peroxide-free, reagent grade.
5 g of standard 0.2 mL 0.25 mL 2. Stock phosphateEDTA buffer (1.8 and
10 g of standard 0.2 mL 0.50 mL 0.085 mol/L, respectively), pH 6.5. Place 126.48 g of
Unknown 0.2 mL 0.5 mL
KH2PO4, 56.8 g of Na2HPO4, and 335 mg of
4. Place all tubes in a heating block set for no tetrasodium EDTA in a 1 L volumetric flask. Dissolve
more than 50 C, and evaporate all solvent from salts in approximately 800 mL of distilled water. Adjust
the tubes under a stream of nitrogen. the pH to 6.5, and bring to volume with additional
distilled water. Stable for 6 months at room temperature.
5. Add 2 mL of hydroquinone in 64% H2SO4, and
3. Working phosphateEDTA buffer (180 and
place in a boiling water bath. Swirl the tubes 8.5 mol/L, respectively), pH 6.5. Dilute 100 mL of
immediately to dissolve the dried extract. stock phosphateEDTA buffer to 1 L volumetrically
Stopper the tubes, and heat for 40 min. Cool the with distilled water. Prepare fresh daily.
tubes in tap water. 4. Mobile phase. Mix 370 mL of HPLC-grade
6. Add 1.7 mL of distilled water to each tube, mix acetonitrile with 1000 mL of working phosphate
thoroughly, and leave in the cold tap water until EDTA buffer, and filter solution through a 0.5 m FH
cool. Millipore filter. Degas under vacuum for 15 min with
7. Read the absorbance of each standard and occasional swirling.
unknown tube against the reagent blank at 472, 5. Estriol standards. Prepare a stock standard by
514, and 556 nm on the spectrophotometer. dissolving 100 mg of estriol (Sigma Chemical Co., St.
Louis, MO) in 1000 mL of methanol to give a
concentration of 100 mg/L (377 mol/L). This solution
Calculations is stable for at least a year at 4 to 8 C.
1. A corrected absorbance at 514 nm, A514 (corr),
should be calculated for each sample using the Prepare an intermediate standard by diluting the stock
Allen equation: standard 100-fold with methanol. The concentration of
this standard is 1000 ng/mL (3.47 nmol/mL). The stock
A514 (corr) = A514 _
A556 + A472 standard is diluted 1000-fold with stock internal standard
solution (see propiophenone below) to yield the working
2
chromatographic standard of 100 ng of estriol/mL (0.35
where A472, A514, and A556 are the absorbances
nmol/mL). This solution is stable for 1 month at 4 to 8
observed at designated wavelengths. C if kept in a tightly closed glass-stoppered container.
2. The concentration of estrogen (as estriol) in 6. Propiophenone, 500 g/L (3.73 mol/L).
urine is calculated by the equation: Dissolve 50 mg in 100 mL of methanol. This solution is
stable for up to a year at 4 to 8 C. This is the stock
g of estrogen/mL of urine = A514(corr) 3.6F internal standard solution.
where F = g of standard/A514 (corr) of standard. Assay

Equipment: This assay was established using a Waters
3. Milligrams of estrogen excreted per 24 h: (Waters, Inc., Division of Millipore, Milford, MA)
Radial-PAK cartridge (10 m particle size, 8 mm inner
mg of estrogen/24 h = diameter by 10 cm) containing C8 stationary phase and
g of estrogen/mL of urine Urine volume mL
an LC-4A electrochemical detector (ECD) from
4. Estrogen/creatinine ratio (E/C): Bioanalytical Systems, Inc. (West Lafayette, IN). A
Waters Bondapak C18 column worked equally well
E/C = g of estrogen/mL of urine
mg of creatinine/mL of urine with minor modifications of the experimental
system.[51] In addition, an injector, pump, dual pen
Procedure: High-Performance Liquid recorder, and ultraviolet spectrophotometer (254 nm) are
Chromatography Assay for Serum Unconjugated required to complete the HPLC system. The ECD was
Estriol [51] placed first in line to monitor the column eluate, which
then passed into the spectrophotometer. The signal
Principle output from each detector is passed to one of the pens of
Unconjugated estriol is extracted from serum with the dual pen recorder. A centrifuge capable of generating
diethyl ether. After evaporation of the solvent, the 800 g is also needed.
residue is redissolved with mobile phase containing an 1. For each sample or control to be analyzed, label
internal standard and analyzed by an HPLC system, two 50 mL glass-stoppered centrifuge tubes: T,
which employs a reversed-phase chromatographic test, and S, spike.
534
Estriol
2. Pipet 2 mL of specimen to be analyzed into IVunk, volume of sample injected into HPLC system
each centrifuge tube. Pipet a known amount of (mL)
the intermediate standard, for example, 10 to 50 0.8, factor to correct for only 80% of ether extract being
L (equal to 10 to 50 ng of spike) of the used in the analysis
intermediate standard to the tube labeled S. 2 mL, volume of serum extracted
3. Add 10 mL of diethyl ether to each tube. 4. Determine the percent recovery of the estriol
Thoroughly mix the contents of the tube by added (spike) to the second aliquot by using the
repeated inversion or vortex mixing for 10 min following calculation:
at room temperature. Centrifuge tubes for 5 min
at 800 g. Percent recovery =
4. Remove exactly 8 mL of the supernatant, and
(Cspike Cunspiked) 2 mL 100%
evaporate ether at 50 C under stream of
nitrogen gas. Cis Vs
5. Reconstitute residue in 200 L of internal where Cspike and Cunspiked are the estriol
standard solution (propiophenone). concentrations (ng/mL) calculated in step 3 for spiked
6. Analyze solution by injecting 10 to 50 L of the and unspiked samples; Cis is concentration of estriol in
reconstituted sample into the HPLC system. the intermediate estriol standard; and Vs is volume of
Also inject 10 to 50 L of the working estriol intermediate estriol standard added to spike tube (step 2
standard containing propiophenone. of assay directions).
7. Chromatographic system: 5. Correct the estriol concentration for unspiked
a. Chart speed, 0.5 cm/min. sample (step 3) for the percent recovery (step 4)
b. Electrochemical detector, 2 nA/V full- to yield the final estriol concentration.
scale sensitivity with the glassy carbon
electrode set at +0.75 V. Estriol (ng/mL) = Uncorrected estriol (ng/mL)
c. Ultraviolet detector, 0.005 AUFS Percent recovery
(absorbance units at full scale) at 254 Notes
nm. 1. The propiophenone is added to correct for
volume loss of the reconstituted sample because
d. Flow rate, 3 mL/min. The mobile
of solvent evaporation.
phase should be heated to
approximately 50 C and should be 2. The average recovery of added estriol was
constantly stirred. The column is kept 99.6% over the analytical range of 2 to 24
at room temperature. ng/mL. One could assume an average recovery
of 99.6% (or 100%) during the extraction
e. Analysis time, 10 min. See Estriol:
procedure and eliminate the need to determine
Figure 3 for example of
the recovery from each sample.
chromatograms.
3. The between-day precision of this assay (as
percent coefficient of variation) was 18.8% at
Calculations 5.1 ng/mL, and 9.9% at both 14.9 and 26.0
1. Measure the peak heights (in millimeters) of the ng/mL of estriol.
estriol peak (retention time approximately 5 4. Chromatograms essentially identical to those
minutes) and the internal standard peak for the shown in Estriol: Figure 3 were obtained using
working standard and each patient sample. the Waters Bondapak C18 columns.
2. Calculate the peak height ratio (PHR) for the 5. The HPLC system showed a linear response for
standard (std) and each unknown (unk): peak height (mm) versus nanograms of estriol
injected up to 8 ng at a 50 L injection volume,
PHR = Peak height of estriol (mm)____
and the sensitivity of the system was
Peak height of internal standard (mm)
approximately 0.4 ng of estriol (equivalent to
3. The amounts of estriol in the patients spiked
an estriol concentration of 1 ng/mL).
and unspiked samples are calculated. The example for
the unspiked is as follows:
std std
Estriol (ng/mL) = PHRunk (C ) IV (0.20 mL)
PHR std (IV unk ) (0.8) 2 mL
where Cstd, concentration of estriol standard (ng/mL)

IVstd, volume of working standard injected into HPLC
system (mL)
0.20 mL, volume of methanol used to reconstitute
extract
535
Ethylene Glycol
Ethylene Glycol
Ping Wang
Name: Ethylene glycol


Merck Index: 3744
Chemical class: Alcohol
Structure:
Principles of Analysis and Current Usage colorimetry, spectrophotometry, or fluorimetry because

Ethylene glycol analysis is requested when a history of the molecule does not absorb a sufficient amount of
ingestion of this compound is likely, or when a patient is ultraviolet light to allow detection by these methods. An
seen in a comatose or intoxicated state with an osmolar early but still often used technique employs the
gap, crystalluria, and a high anion gap indicative of oxidation of ethylene glycol by periodic acid to
metabolic acidosis. The clinical toxicologist will formaldehyde [4-6]. The formaldehyde (HCHO) is
normally be called upon to either screen for the presence subsequently derivatized to a colorimetrically or
of or determine the quantity of ethylene glycol in either fluorimetrically measurable substance (Table 1, Methods
blood or urine. Depending on the circumstances, rapid 1 to 3).
identification of this compound can prevent death many
hours after ingestion. Since ethylene glycol intoxication HIO42 HCHO
OHCH2CH2OH
is usually seen as an emergency situation, laboratories
should have the capability of identifying this compound
in a short time. 1. Harger and Forney colorimetric technique:
HCHO + Fuchsine reagent violet color
The subject of crystalluria associated with ethylene (max , 555 nm)
glycol ingestion deserves special consideration as part of
the laboratory investigation for this poison. In 1946, 2. Rajagopal colorimetric technique:
Milles referred to a 1930 report of oxalate crystals found
in the kidneys of a German soldier during an autopsy 3HCHO + 2K2Cr2O7 + 10H2SO4 + 2H+
performed after death from ethylene glycol poisoning
[1,2]. Later that year, Pons et al. reported on the deaths (yellow orange)
of 10 soldiers, all of whom had oxalate crystals in their
urine [3]. Early reports such as these helped to establish 2Cr2(SO4)4 + 2K2SO4 + 11H2O + 3HCOOH
a link between crystalluria and the diagnosis of ethylene (blue green)
glycol toxicity. This observation, coupled with the need (max , 550 nm)
for rapid diagnosis of ethylene glycol ingestion, has
made the detection of crystalluria an important factor in 3. Hantzsch reaction [6]:
the diagnosis of ethylene glycol poisoning.
Acetylacetone + HCHO NH4+Diacetyllutidine
Techniques for the analysis of ethylene glycol fall into derivative
two major groups: (1) oxidation with subsequent (ex , 400 nm; em , 500 nm)
measurement by colorimetry, spectrophotometry, or
fluorimetry and (2) various chromatographic methods.
Ethylene glycol cannot be measured directly by An enzymatic method that may be adapted to an
automatic platform uses glycerol dehydrogenase to
oxidize ethylene glycol to glycoaldehyde [7,8]. The
i
Ethylene glycol absorbance change at 340 nm associated with NADH
Previous and current authors of this method: generation is proportional to the ethylene glycol
First edition: Not done concentration (Table 1, Method 4).
Methods edition: Dan Field
Second edition: Not updated glyceroldehydrogenase
Ethylene glycol +NAD+
Third edition: Not updated glycoaldehyde + NADH + H+
Fourth edition: Dan Field
Fifth edition: Ping Wang
536
Ethylene Glycol
Glycerol dehydrogenase enzymes isolated from different crystalluria in the diagnosis. A review of the literature
organisms vary in their activity in this assay. The since the turn of the century reveals that crystalluria was
enzyme from Cellulomonas species demonstrated high reported in less than 50% of the cases. On the other
activity. This method may detect ethylene glycol as low hand, the test has a high specificity because oxalate
as 10 mg/dL, and no interference has been observed crystalluria is relatively rare. Taken together with the
from ethanol, acetone, methanol, and isopropanol. usual clinical presentation of ethylene glycol toxicity,
However, this method is subject to both false-positive the presence of oxalate crystals in the urine strongly
and false-negative interferences by a variety of supports a diagnosis of ethylene glycol toxicity.
compounds listed in the section on Interferences below. However, oxalate crystals may fail to take the classic
octahedral shape of the more commonly seen dihydrate
An alternative method using oxidation of ethylene glycol molecule and instead, in monohydrate form, may mimic
by alcohol dehydrogenase enzyme [9] could be useful hippuric acid crystals (Figure 1). Such a
(Table 1, Method 5). misinterpretation has led to misdiagnosis and death
[13,15]. No evidence exists in the literature to support
alcohol dehydrogenase the often stated contention [16] that ethylene glycol
HOCH2CH2OH + NAD+
poisoning leads to hippuric acid crystalluria in humans,
HOCH2CHO + NADH + H + though there is ample research in animal studies. The
formation of hippuric crystals has been documented in
Ethylene glycol can also be estimated from the osmolal studies of dogs and cats [10,17]. Thus analysis of
gap obtained by subtraction of a calculated osmolality crystalluria remains a useful but limited technique.
from the osmolality measured by freezing-point
depression (Table 1, Method 6). With direct-injection gas chromatographic methods, the
sample needs little preparation; with 1 to 2 L of serum
(Measured osmolality Osmocalc) 62.1 g/mol 0.92 = or urine, the detection limits of these methods range
milligrams of ethylene glycol per liter of serum, where between 16 and 50 mg/L, but gas chromatographic
62.1 g/mol is the molecular weight of ethylene glycol, analyses employing these methods are said to suffer
Osmocalc is the calculated serum osmolality, and 0.92 is from low sensitivity and poor reproducibility [14]. One
the correction for volume of water in serum. This factor lowering the sensitivity of the direct injection
technique is only of value if one can be certain that no method is a characteristic of the flame ionization
other compounds have been ingested that might be detector itself. The sensitivity of the method is
contributing to the osmolal gap. proportional to the number of carbon atoms in the
compound being measured. In the boronate derivative
Various types of chromatography compose the second method, by derivatizing the ethylene glycol directly in
major analytical approach to ethylene glycol analysis. the specimen, extraction is avoided, and the method
The thin-layer technique [10] (Table 1, Method 7) and gains some of the simplicity of direct injection, as well
gas chromatography with a flame ionization detector are as better precision and sensitivity. The HPLC benzoate
most frequently used. Because of difficulties associated method has a lower limit of detection of 10 mg/L, and no
with extraction and concentration of the compound, most interference from other alcohols is reported [18].
gas chromatographic techniques use the direct injection
method (Table 1, Method 8). Alternative approaches to The HPLC-mass spectrometric technique method uses
direct injection utilize dibenzoate [11] or boronate 20 mL of urine; however this procedure is limited to
derivatives [12] of ethylene glycol (Table 1, Method 9). specialized laboratories with the necessary equipment.
With the enzymatic oxidation methods available, the
A high-performance liquid chromatographic method chemical oxidation assays are largely outdated. The
utilizing ethylene glycol as a benzoyl ester derivative has advantages of enzymatic assays are the simplicity of
been reported (Table 1, Method 10) [13]. Another reaction and the availability of required chemicals and
method combines high-performance liquid equipment.
chromatography with mass spectroscopy for detection of
glycolic acid, a major metabolite of ethylene glycol Currently, no single method of ethylene glycol analysis
(Table 1, Method 11) [14]. Since rapid metabolism is preferred, but gas chromatography is, in one form or
clears the body of the parent ethylene glycol, this is a another, most favored. The limits of detection of the
potentially useful technique for providing a suitably methods outlined above range between 10 and 100
sensitive assay. mg/L, but since most toxic ingestions involve
concentrations at least a magnitude greater, the chosen
Reference and Preferred Methods technique depends more on the available equipment and
There is no reference method for ethylene glycol. The the training level of the laboratory personnel. The direct-
choice of method depends on clinical needs and injection gas-chromatography method presented here is
availability of expertise and instrumentation. chosen for the simplicity of the procedure and the ready
availability of equipment. The method is rapid but not
Because some laboratories use the presence of suited to the analysis of large numbers of samples. In the
crystalluria as a screening test for ethylene glycol absence of gas chromatographic instrumentation, the
toxicity, it is important to consider the role of enzymatic procedures may be adapted to automatic
537
Ethylene Glycol
platforms for stat analysis in emergency cases. The dehydrogenase causes the osmolality to fall and initiates
fluorimetric method also seems to have good specificity a high anion gap. Bicarbonate-resistant metabolic
[6]. acidosis is initially due to the production of glycolic acid
from ethylene glycol but is subsequently caused by
Specimen lactate and other organic acids formed as by-products of
Serum, plasma, or urine are acceptable for analysis; it is an increased NADH/NAD ratio caused by the action of
important to note that the appearance of ethylene glycol alcohol dehydrogenase (see Figure 4). Oxalate, the end
in the urine may be delayed for many hours, depending product of ethylene glycol metabolism, precipitates
on the circumstances of ingestion. Samples should be calcium to form crystals in the renal tubules, lungs, and
processed promptly to avoid time-dependent loss of brain, which may cause significant damage to these
glycol. One report states that the concentration of tissues.
ethylene glycol in a sample stored in a refrigerator for 12
days decreased from 2850 to 800 g/L [4]. The metabolic acidosis together with the precipitated
calcium crystals are part of a well-characterized
Interferences progression of signs and symptoms [23]. Beginning with
A common interference in colorimetry and fluorometry the central nervous system, the patient will appear
is interference from antibiotics and antitussives intoxicated or comatose, frequently without the smell of
containing 1,2-hydroxyl groups in sufficient quantity to alcohol but often with nausea, vomiting, ocular
yield falsely positive results [19]. However, in one abnormalities, and convulsions. After approximately 12
method based on the Hantzsch condensation reaction, hours, the cardiopulmonary stage predominates and is
when 36 commonly used drugs were examined, only characterized by failure of these two systems. Finally,
nystatin (Mycostatin) caused interference [6]. The approximately 36 hours after the initial ingestion, the
alcohol dehydrogenase reaction is not specific for renal phase is chiefly characterized by oliguria. Death
ethylene glycol. Other alcohols such as ethanol, can occur during any phase, but with appropriate
methanol, or isopropanol can also interfere. The glycerol therapy, patients can survive massive ingestions [24].
dehydrogenase method has the advantage of being free
of interference from some alcohols such as ethanol, The commonly quoted minimum lethal dose is stated to
acetone, methanol, and isopropanol. However, high be 100 mL (1.46 moles), but in fact as little as 30 mL
concentrations of 2,3 butanediol, glycerol, and propylene have been held responsible for death in adults [25]. The
glycol may interfere. Although glycerol and propylene actual number of deaths attributable to ethylene glycol is
glycol are widely used as solvents in IV medications, unknown. However, in 1983, the National Data
they are usually not present in the patient specimen at Collection System [26] issued a report on deaths from
levels high enough to cause significant interference. ethylene glycol taken from data collected on 11% of the
Samples from patients with high levels of lactate and U.S. population. Taking into account extensive
lactate dehydrogenase [20] and those with hepatotoxicity underreporting by the participating centers (a fact
caused by acetaminophen may also give elevated results. acknowledged by the authors) and the distribution of the
In addition, lipemia may cause false-positive results. population, the extrapolated death figure from ethylene
Glycerol, propylene glycol, and hemolysis may generate glycol is less than 40 per year. However, there are no
negative interference. reliable estimates for the nationwide deaths from
ethylene glycol.
Ethylene Glycol Reference Interval Initial diagnosis and treatment should rest solely on the
Any amount of ethylene glycol detected is abnormal. criteria of history of ingestion, apparent intoxication or
coma without the stigma of alcohol (or inconsistent with
Interpretation ethanol levels obtained and clinical course), an elevated
The study of ethylene glycol toxicity has been osmolal gap, and a metabolic acidosis with an anion gap.
characterized by confusion and misinformation since the As previously noted, crystalluria of any type is evidence
compound was first developed as a nontoxic replacement but if absent does not rule out ethylene glycol poisoning.
for glycerin at the turn of the century. Early investigators
went as far as drinking various quantities as proof of Induced emesis or gastric lavage is an important step
nontoxicity, but by 1930, reports began to surface after recent ingestion of ethylene glycol. 4-MP
questioning the benign nature of ethylene glycol [21]. (fomepizole), a specific inhibitor of alcohol
Finally, the deadliness of the glycols became clear in the dehydrogenase, can prevent further metabolism of
Massengil Disaster, where greater than 100 people ethylene glycol into toxic metabolites and therefore
eventually died after ingesting an elixir of sulfanilamide should be given as an antidote. In hospitals and clinics
in which a glycol served as the medicinal vehicle [22]. where 4-MP is not available, intravenous ethanol should
Ethylene glycol causes symptoms and organ damage be administered upon suspicion of ethylene glycol
similar to that seen with ethanol toxicity. However, most ingestion. Ethanol has a tenfold greater affinity for
of the toxicities associated with ethylene glycol ingestion alcohol dehydrogenase than ethylene glycol has and
are due to organ damage and acidosis caused by the therefore causes a virtual shutdown of ethylene glycol
metabolites, especially oxalic acid and glycolic acid. metabolism and can be considered a true antidote. This
Initially, the measured serum osmolal gap becomes administration of antidote can be coincident with both
significantly elevated, but rapid metabolism by alcohol emesis and subsequent charcoal treatment. Once
ethylene glycol is metabolized, hemodialysis becomes
538
Ethylene Glycol
crucial and should be instituted as early on in the intoxicated dogs and cats. J Am Vet Med 1984;
treatment as possible; hemodialysis removes all the toxic 184: 584.
agents while allowing for simultaneous correction of 11 Peterson RL. Gas chromatographic
acid-base and electrolyte abnormalities. determination of ethylene glycol in serum, Clin
Chem 1982; 28: 75-78.
Ethylene Glycol Performance Goals 12 Robinson S. A gas chromatographic procedure
Ethylene glycol is an infrequent analyte included in the for quantitation of ethylene glycol in
Toxicology T survey of the College of American postmortem blood. J Anal Toxicol 1981; 5: 69-
Pathologists (CAP). Labs that carry out routine ethylene 72.
glycol analysis may want to exchange samples with 13 Gondolphin W. Unusual calcium oxalate
other labs for proficiency testing. Because most assays crystals in ethylene glycol poisoning. Clin
are noncommercial, home-developed tests, there is Toxicol 1980; 16: 479-486.
generally poor standardization between laboratories. In a 14 Hewlett T, Ray AC, Reagor JC. Diagnosis of
CAP survey containing ethylene glycol, only 26.1% labs ethylene glycol (antifreeze) intoxication in dogs
correctly identified ethylene glycol in the specimen. Gas by determination of glycolic acid in serum and
chromatography was the method most commonly used urine with high pressure liquid chromatography
by the labs that correctly reported the presence of and gas chromatography-mass spectrometry. J
ethylene glycol in the specimen. False-positive results of Assoc Off Anal Chem 1983; 66: 276-283.
ethanol, ibuprofen, and salicylates were reported by 15 Berger J. Neurological complications of
individual labs. As many as 70.6% labs reported that no ethylene glycol intoxication. Arch Neurol 1981;
drugs were detected. The interlaboratory CV by gas 38: 724-726.
chromatography was 13.4%. A Swedish study looking at 16 Field DL. Acute ethylene glycol poisoning. Crit
data from an external proficiency test found an Care Med 1985; 13: 872-873.
intralaboratory CV of 4.5% and an interlaboratory CV of 17 Riley J. Urine and tissue oxalate and hippurate
11.4% [27]. Results obtained using gas chromatography levels in ethylene glycol intoxication in the dog.
and enzymatic methods were generally in good Vet Hum Toxicol 1982; 24: 331-334.
agreement [27]. A U.K. study reported a between- 18 Gupta R. Liquid-chromatographic
laboratory CV of 24% in 11 participants who measured determination of ethylene glycol in plasma. Clin
ethylene glycol in whole blood [28]. No Clinical Chem 1982; 28: 32-33.
Laboratory Improvement Amendments required 19 Doedens D. Methods for determination of
performance goals have been set for ethylene glycol ethylene glycol. Vet Hum Toxicol 1983; 25: 96-
proficiency testing. 101.
20 Eder AF, Dowdy YG, Gardiner JAM, Wolf
References BA, Shaw LM. Serum lactate and lactate
1 Milles G. Ethylene glycol poisoning. Arch dehydrogenase in high concentrations interfere
Pathol 1946; 41: 631-637. in enzymatic assay of ethylene glycol. Clin
2 Boemke F. Beitrag zur Toxikologie und Chem 1996; 42: 1489-1491.
Pathologie des thylenglykols (Glysantin). 21 Anon. Possible death from drinking ethylene
Virchows Arch Pathol Anat 1943; 310: 106. glycol (Prestone). JAMA 1930; 94: 1940.
3 Pons C. Acute ethylene glycol poisoning: a 22 Geiling EMK, Cannon PR. Pathologic effects
clinicopathologic report of 18 fatal cases. Am J of elixir of sulfanilamide (diethylene glycol)
Med Sci 1946; 211: 544-552. poisoning. JAMA 1938; 111: 919-926.
4 Harger RN, Forney RB. A simple method for 23 Moriarity R. The spectrum of ethylene glycol
detecting and estimating ethylene glycol in poisoning. Clin. Toxicol. 1974; 7:583-596.
body materials: analytical results in six fatal 24 Stokes J. Prevention of organ damage in
cases. J Forensic Sci 1959; 4: 136-143. massive ethylene glycol ingestion. JAMA 1985;
5 Rajagopal G, Ramakrishnan S. A new method 243: 2065-2066.
for estimation of ethylene glycol in biological 25 Widmen C. A few cases of ethylene glycol
material. Anal Biochem 1975; 65: 132-136. intoxication. Acta Med Scand 1946; 126: 295-
6 Meola J. Fluorometry of ethylene glycol in 306.
serum. Clin Chem 1980; 26: 1709. 26 Veltri J. 1983 Annual report of the American
7 Standefer J, Blackwell W. Enzymatic method Association of Poison Control Centers National
for measuring ethylene glycol with a centrifugal Data Collection System. Am J Emerg Med
analyzer. Clin Chem 1991; 37: 1734-1736. 1984; 2: 420-433.
8 Hansson P, Masson P. Simple enzymatic 27 Jones AW, Hrd L. How good are clinical
screening assay for ethylene glycol ( ethane- chemistry laboratories at analyzing ethylene
1,2-diol) in serum. Clin Chim Acta 1989; 182: glycol? Scand. J. Clin. Lab. Invest t. 2004; 64:
95-102. 629634.
9 Eckfeldt J. Kinetic ethylene glycol assay with 28 Wilson JF, Toseland PA, Capps NE, Sandle
use of yeast alcohol dehydrogenase. Clin Chem LN, Smith BL, Sweeney G et al. External
1980; 26: 1278-1280. quality assessment of laboratory performance in
10 Kramer J. Identification of hippuric acid analysis of toxicological cases. Forens. Sci.
crystals in the urine of ethylene glycol Intern. 2001; 121: 27-32.
539
Ethylene Glycol
Tables
Table 1: Methods of Ethylene Glycol Analysis
Method 1: Oxidation-periodic acid
Type of analysis: Colorimetric
Principle: Ethylene glycol is oxidized by acid to formaldehyde and measured as a colored derivative
Usage: Historical
Comments: Limited sensitivity
Method 2: Oxidation-periodic acid
Type of analysis: Colorimetric
Principle: As above; then chromotropic acid is used to form colored derivative
Usage: Limited
Comments: Required precipitation of many interfering substances with copper sulfatecalcium hydroxide
preparation
Method 3: Oxidationperiodic acid
Type of analysis: Fluorometric
Principle: Ethylene glycol is oxidized by acid to formaldehyde and then converted to a dihydrolutidine derivative
Usage: Alternative
Comments: As in method 2. Excessive interference by elevated triglycerides, limited range of linearity
Method 4: Oxidation-enzymatic (glycerol dehydrogenase)
Type of analysis: Spectrophotometric
Principle: Ethylene glycol is oxidized by glycerol dehydrogenase to glycoaldehyde, with concurrent reduction of
NAD+ to NADH; reaction monitored at 340 nm
Usage: Limited, good for stat analysis
Comments: Interference from 2,3 butanediol, glycerol, propylene glycol, lipemia, and hemolysis. Patients with
high levels of lactate and lactate dehydrogenase and patients with acetaminophen-induced hepatotoxicity may
have falsely elevated results.
Method 5: Oxidation-enzymatic (alcohol dehydrogenase)
Type of analysis: Spectrophotometric
Principle: Alcohol dehydrogenase oxidizes ethylene glycol to its aldehyde, with concurrent reduction of NAD+
to NADH; reaction monitored at 340 nm
Usage: Limited
Comments: Interference from ethanol, methanol, and isopropanol
Method 6: Osmolal measurement
Type of analysis: Freezing-point depression
Principle: Calculated osmolality is subtracted from measured osmolality; ethylene glycol estimated: (Osmolal
gap) 62.1 0.92
Usage: Possible screen
Comments: A quick method for estimating the quantity of ethylene glycol in a known or suspected ingestion;
mol. wt. of ethylene glycol, 62.1
Method 7: Thin-layer chromatography
Type of analysis: Chromatographic
Principle: Ether extraction; developed in one of three solutions, visualized with vanillin in H2SO4
Usage: Possible screen
Comments: Sensitivity, limit of detection not reported, use limited to few labs
Method 8: Gas chromatography (GC) (various researchers)
Principle: Flame ionization, direct injection
Usage: Common
Comments: Moderate sensitivity, widespread equipment, minimal preparation
Method 9: Gas-chromatography derivative
Principle: Extraction and derivatizationbenzoyl ester, N-butylboronate, phenylboronate
Usage: Alternative
Comments: Require extraction but offer increased sensitivity and precision
Method 10: High-performance liquid chromatography (HPLC) derivative (Gupta)
Principle: Benzoyl derivative prepared directly in plasma by Schotten-Baumann reaction to allow detection of
ultraviolet absorption spectra
Usage: Experimental
Comments: No extraction, good sensitivity, low interference potential
540
Ethylene Glycol
Method 11: HPLC; gas chromatography-mass spectrometry (GC/MS) (Hewlett)

Principle: Measures glycolic acid, the major metabolite; extraction with ketone and derivatization
Usage: Experimental
Comments: Advantage of diagnosis after ethylene glycol has been metabolized; GC/MS used to confirm glycolic
derivative
Figures
Figure 1: Ethylene glycol
Drawings of oxalate crystals. A, Calcium dihydrate

oxalate crystals. B, Calcium monohydrate oxalate
crystals. C, Hippurate crystals.
Tracing of gas chromatographic analysis of ethylene glycol in

patient serum. A, Serum of healthy person. Notice endogenous
peak at 5.60 minutes. B, Serum spiked with 3000 g/mL of
ethylene glycol.
541
Ethylene Glycol
Standard curve of ethylene glycol concentration in serum versus peak height of standard.
Figure 4: Ethylene glycol Metabolism
Schema of pathway of ethylene glycol

metabolism. Boxes, Enzymatic reactions;
LDH, lactate dehydrogenase.
542
Ethylene Glycol
Procedure: Gas Chromatographic Analysis months.
Principle Use a fresh control with each run. Bring an aliquot to

A 1:1 ratio of volatile ethylene glycol (EG) and internal room temperature prior to use, taking care to make sure
standard is injected into a carbowax column within a gas that the sample is homogeneous prior to dilution with
chromatograph, where it is separated from other internal standard and injection.
volatiles. Using a flame ionization detector, the ethylene
glycol may then be quantitated by measurement of the D. 12 75 glass culture tubes
peak height. The amount of EG is quantitated by a E. 100 L Eppendorf pipette
comparison of peak height ratio to that obtained with a F. Hamilton 10 L fixed syringe for
known 1:1 ratio of an EG standard to Internal Standard. injections
Specimens of Choice Assay

Serum (1 mL) collected in a red-top tube with or without Equipment: Gas chromatographic system with
serum separator. Samples should be collected during the flame ionization detector; 6-foot column containing 5%
acute phase of toxicity. Urine (1 ml) may be used for Carbowax 20 M.
qualitative testing only. Samples should be collected in 1. Chromatograph parameters:
the acute phase of toxicity. Specimens may be Chart speed, 0.5 cm/min
transported at room temperature. Serum should be Settings for the Shimadzu gas chromatograph:
separated from the clot upon arrival. If testing will not be A. Column: 6 ft by 2 mm glass column
performed within 3 hours (non-stat), store specimens packed with 5% carbowax 20 M
frozen (<20C); do not heat upon thawing. B. Column temp 150C
C. Injector temp 240C
Indications D. Detector temp 240C
Ethylene glycol is a major constituent of E. Nitrogen flow tank 40 psi (approx 30
antifreeze. The test is often requested based on a positive mL/min)
ingestion history, acidosis of unknown etiology, or the F. Air flow tank 20 psi (approx 300
presence of oxalate crystals seen during urinalysis. The mL/min)
test is available and usually performed on a stat basis. G. Hydrogen flow tank 20 psi (approx 30
mL/min)
Reagents and Materials H Attenuation 32
A. Internal Standard: 2-Ethoxyethanol. I. Range 10
Using a 50 L Hamilton spiking syringe, deliver 20 L J. Mode single
of liquid internal standard into a 100 mL volumetric
flask, and dilute to the mark with deionized water. Ethylene glycol peak, approximately 3.0 cm (or
Prepare fresh daily as needed. Store at room 6 minutes)
temperature. Discard after 24 hours. 2. Set instrument conditions. Allow column to
B. Working Ethylene Glycol Standard bake out for at least 30 minutes, and establish a flat
(111 mg/dL). Using a glass 0.2 mL serological pipette, baseline.
deliver 0.1 mL of EG into a 100 mL volumetric flask, 3. Inject 1 L of ethylene glycol stock standard
and dilute with deionized water to bring to volume. into the instrument. Record retention time of glycol.
Prepare fresh every 24 hours. Store at room temperature. Peak height of standard should be at least 50%
Linearity Check Standards. Procedure linearity should be of full scale.
established monthly, or whenever a new column is put 4. Inject (in duplicate) 1 L of sample, standard,
into service, by analyzing the following curve prepared or control. Record retention time of all peaks.
by adding stock EG in the amounts indicated to 100 mL Figure 2 shows a representative chromatogram
volumetric flasks and diluting to the mark with for a standard and a serum sample.
deionized water.
Procedure
Concentration Stock EG Analytical runs will consist of the following:
(mg/dL) (L) 111 mg/dL working EG Standard
223 200 In-house Control
111 100 Patient sample
56 50 If patient sample is positive; a deionized water
28 25 blank
Reinjection of positive patient
C. In-House Frozen Ethylene Glycol
Control. (56 mg/dL) To a 100 mL volumetric flask, add Note: Rinse syringe with methanol between
50 l of EG stock solution. Dilute to volume with pooled each injection.
negative serum. Mix well, then aliquot 0.2 mL portions A. With an Eppendorf syringe, add 100 l of
into 1275 glass tubes. Store at 20C. Stable for 6 internal standard solution to a glass culture tube.
543
Ethylene Glycol
B. To the same tube, add 100 l of standard, Interpretation

control, or patient sample, and mix the solution. A. Any detectable amount of EG should be
C. Inject 1 l of the mixture into the gas considered abnormal. Therefore, critical values are noted
chromatograph. to be >5 mg/dL. Levels are generally between 50 and
D. Immediately press [start] on the HP integrator 300 mg/dL in serum or urine during acute toxicity,
to initiate the run. Repeat this process for mixtures of the although the compound dissipates rapidly. Generally,
standard, control, and all patient or unknowns. The EG will not be detectable in serum beyond 12 hours post
internal standard should have a peak height of ingestion, but this is concentration dependent.
approximately one half the width of the recorder paper.
Adjust injection amount or attenuation of the recorder as B. Specificity: There are no known interfering
needed. substances, although unknown metabolic endogenous
E. Note that occasionally it may be necessary to substances have been noted to present false-positive
inject the internal standard and ethylene glycol results in other laboratories (see D below). Ethanol,
individually to determine the retention time of each methanol, isopropanol, and acetone do not interfere (they
substance. appear before EG). Occasionally, small peaks with the
F. Proceed to the calculation section. RT of EG will be noted (see Sensitivity below).
C. Sensitivity: 5 mg/dL. DO NOT REPORT
Calculations and Quality Control CONCENTRATIONS BELOW 5. Report as <5 mg/dL.
A. Compare the retention time of EG D. Results may be reported up to 446 mg/dL.
standard with patients retention times. The unknown is Report results greater than this as >446 mg/dL.
positive for EG if it possesses a peak with a relative
retention time 2% of the standard, with a peak height at Notes
least 2 times greater than peaks in the H2O blank. 1. Ethanol, methanol, isopropanol, and acetone
See Figure 3 as an example of a typical standard curve. will not interfere.
NOTE: A positive test result is indicated by a 2. The method is capable of detecting as little as
concentration of 5 mg/dL or greater. For urine 50 mg/L.
specimens, further quantitative calculations are not 3. In view of the rapid conversion of ethylene
undertaken. Report as positive. glycol to unmeasured metabolites, any detectable
amount should be considered clinically significant.
B. If the integrator was programmed for
calculating the amounts of EG in the sample, enter the
control value into the QC computer to determine if the
assay is within QC specification. If within acceptable
limits, proceed with reporting of the patient serum
quantitative results unless greater than upper limit of
calibration.
C. If the integrator was not programmed to

calculate the results, proceed with manual calculations of
control and patient as follows:
1. Using a millimeter ruler,
determine the peak heights of the ethylene glycol and
internal standard peaks. Form a ratio of these values by
dividing the EG peak height by the I.S. peak height.
2. Use the same procedure for
the rest of the samples. Then divide the ratio of the
control or patient samples by the ratio of the EG
standard, and multiply this number by the concentration
of EG in the standard (111 mg/dL). The result will be in
mg/dL units.
Concentration mg/dL =
peak height ratio of patient or control 111 mg/dL
peak height ratio of EG working standard
D. For integrated or manual calculated results

>223 mg/dL, perform a 1:1 dilution of patient serum
with deionized water, then repeat analysis, multiply
resulting concentration 2. There is no need to perform
additional dilutions
544
Fecal Electrolytes and Osmolality

Felix O. Omoruyi, Anthony O. Okorodudu
Names: Electrolytes and osmolality

Clinical significance: Loss of fluids and electrolytes from the gastrointestinal system lead to
diarrhea. Excessive lost electrolytes interrupt acid-base balance.
Refer to Chapter 34, The Pancreas: Function and Chemical Pathology, and Chapter 35
Gastrointestinal Function and Digestive Disease in the 5th edition of Clinical Chemistry: Theory,
Principles of Analysis and Current Usage potassium are the predominant cations in stool water.
The measurement of fecal electrolytes is most often When electrolytes constitute most of luminal osmolality,
performed for the assessment of intestinal salt losses in the calculated fecal osmolar gap will be low (<50
cases of diarrhea. Thus most fecal samples for which mOsm/kg). When poorly absorbable substances are
electrolyte measurements are requested are usually fairly present, the fecal osmolar gap will be large (>100
liquid, with undigested fat and solid-food material mOsm/kg).
suspended in the water portion. For these most
frequently encountered stool samples, a measured Reference and Preferred Methods
aliquot of the liquid portion is removed and centrifuged There is no reference method for the analysis of stool
at 500 to 1000 g to remove particulate matter. electrolytes. The preferred methods are ISE & freezing-
point osmometry.
Sodium and potassium can be easily measured by
dilutional or non-dilutional ion-selective electrode Specimen
analysis. Chloride can be measured by the ferric The preferred sample is a diarrheic stool specimen with a
thiocyanate method, by an ion-selective electrode distinct water phase. A minimum of 1 mL can be
method, or by coulometry (Chloridometer). Osmolality assayed. The physician should be advised if the
can be measured with freezing-point or vapor-pressure laboratory receives a well-formed, solid stool specimen;
osmometers. such a sample is inappropriate for analysis.
More rarely, electrolyte analysis may be requested on a The stool specimen should be collected for a 72-h period
stool specimen that is well formed, without a separate in a pre-weighed can (i.e., paint can). On completion of
liquid phase. It is not recommended that analysis be the collection, weigh the can with the specimen, and
performed on such a sample, because the result will have record the weight. The weight of the feces is obtained by
little meaning. subtracting the weight of the can (recorded on the can)
from the total weight. The sample should be assayed
To distinguish between osmotic and secretory diarrhea without delay, because bacterial contamination will be
(see Interpretation below), it is necessary to calculate present, and metabolism will occur.
and measure fecal water osmolality, looking for an
osmolar gap. Interferences
Substances that might interfere with the analysis of
Fecal Osmolar Gap electrolytes can be found in the method description for
If poorly absorbed material is retained within the gut, serum electrolytes and osmolality. The laboratory should
water moves into the gut and an osmotic diarrhea results. ensure that the final specimen is free of particulate
This provides the rationale for the calculation of fecal matter and is not too viscous so as not to adversely affect
osmolar gap [1,2]. Osmolar gap is estimated by the the instruments used to make the analyses.
following calculation: Osmolar gap = measured
osmolality 2 (Na + K). Note that sodium and Fecal Electrolytes and Osmolality Reference Interval
Fecal electrolyte composition can be very variable. It is
i the measured and calculated osmolality which is
Fecal electrolytes and osmolality
important.
Interpretation
Methods edition: Lawrence A. Kaplan
The value in measuring electrolytes in a stool sample
from a person with diarrhea is to distinguish secretory
from osmotic diarrhea. Secretory diarrhea occurs when
Fourth edition: Lawrence A. Kaplan
there is excess secretion of solute within the gut.
Fifth edition: Felix O. Omoruyi, Anthony O.
Osmotic diarrhea occurs when there is an accumulation
Okorodudu
within the gut of poorly absorbed non-ionic particles, as
545
may be seen in conditions such as lactase deficiency.

The two conditions can be differentiated by measuring
osmolality and calculating the osmolar gap. It will be
large in osmotic diarrhea, since the non-ionic substances
will not contribute to the calculated osmolality [3].
References
1 Castro-Rodriquez JA, Salazar-Lindo E, Leon-
Bama R. Differentiation of osmotic and
secretory diarrhea by stool carbohydrate and
osmolar gap measurements. Arch Dis Child
1997; 77: 201-205.
2 Ladefoged K, Schaffalitzky de, Muckadell OB,
Jarnum S. Fecal osmolality and electrolyte
concentrations in chronic diarrhea: do they
provide diagnostic clues. Scand J Gastroenterol
1987; 22: 813-820.
3 Walmesley RN, White GH. A guide to
diagnostic clinical chemistry. Blackwell
Scientific, 1994, p387.
546
Methods of Stool Electrolyte and Osmolality

Analyses
Principle
The liquid portion is filtered or centrifuged to remove
particulate matter. The clarified specimen is then
analyzed for electrolytes by standard procedures.
Reagents
1. Distilled water. Class I, electrolyte free.
2. Filter paper. Whatman No. 1, or equivalent.
Assay
Equipment: Centrifuge capable of generating 1000 to
1500 g, filtering funnels, commercial paint can shaker.
1. Weigh the preweighed paint can containing the
collected stool sample.
2. If the sample is already quite liquid, mix it
thoroughly to achieve a homogeneous sample. A
few minutes of mixing in a paint shaker will be
sufficient.
3. Centrifuge at 1000 g for 15 min at room
temperature.
4. If the supernatant is not particle-free and clear,
filter through Whatman No. 1 filter paper, allowing the
filtrate to drip into a clean test tube.
5. Analyze the sample using the usual laboratory
method available for measuring bicarbonate,
osmolality, sodium, potassium, and chloride.
6. Calculate and measure the osmolality of the
stool water. An osmolar gap < 50 mOsm/kg suggests
that the patient has a secretory diarrhea, and an osmolar
gap > 100 mOsm/kg suggests the patient has an
osmotic diarrhea.
547
Fecal Fat and Fat Absorption

Lawrence A. Kaplan
Name: Fat absorption

Clinical significance: Refer to Chapter 34, The Pancreas: Function and Chemical Pathology, in
the 5th Edition of Clinical Chemistry: Theory, Analysis, Correlation.
Chemical class: Triglycerides
Principles of Analysis and Current Usage Isotope Tests

An assessment of fat absorption is important in the The first isotope-based test for the assessment of fat
assessment of malabsorption [1-5]. The most specific absorption was the 131I-triolein test [7]. In this test,
assessment of this function is the accurate chemical
the patient ingested a triglyceride labeled with 131I.
quantitation of fat in a 72-hour stool collection.
At prescribed time intervals after ingestion, blood
was collected for quantitation of radioactivity. In
Fecal Fat Screening Test
general, the more efficient the fat absorption, the
A successful fat screening method is the microscopic
greater the radioisotope count.
fat evaluation of a single sample. This has been done
by application of fat stains to a smear of a
representative stool sample on a microscope slide, A 14C isotope breath test has proved superior to
followed by visual semiquantitation. To perform the iodine-labeled methods. It was first described by Abt
method, an aliquot of stool on a slide is mixed with in 1966, who measured the recovery of 14C-labeled
two drops of 95% ethanol and two drops of a lipids [8]. One drawback to the use of this test was
saturated ethanolic solution of Sudan III. Under a the increased number of false-positive results
microscope, neutral fats appear as large orange or red observed in obese patients. A popular method based
drops. In some methods, fatty acids appear as lightly upon this approach was developed by Kaihara, who
staining flakes or needle-like crystals. An abnormal used 14C-tripalmitin [9,10].
number of fat droplets per high-power field (the
number varies, depending on the method) indicates
malabsorption. A 13C stable-isotope breath test has been developed
as well [11-14]. The advantage of this test over the
While this procedure is usually used as a qualitative 14C isotope breath test is the use of non-radioactive
screening test, its use as a quantitative procedure in test material. However, this method requires
place of the 72-hour fecal fat quantitation has been measurement of 13C by a mass spectrometer or by
proposed [6]. In this procedure, fecal fat in a stool NMR (see below).
sample is stained with Sudan III and assessed
microscopically by counting the number of drops and Fecal Fat Quantitation
measuring the size of the fat globules to produce a The oldest available quantitative method is a
quantitative result. gravimetric technique (Table 1, Method 1). In this
method, the fat from stool is extracted into an organic
solvent, which is then evaporated, and the residue is
weighed to determine the fat content. It has been
i suggested that the electrical capacitance of a solvent
i extract of stool be measured (Table 1, Method 2)
[15]. The principle of this test is that the presence of
fatty acids in the extract will decrease electrical
capacitance; the greater the change in capacitance as
Methods edition: Michael D.D. McNeely
measured by change in frequency of a radio-oscillator
circuit, the more fat was extracted from the stool.
Fourth edition: Michael D.D. McNeely
Bangs acid dichromate reagent [16] has been used in
Fifth edition: Lawrence A. Kaplan
a spectrophotometric analysis (Table 1, Method 3).
The basis of this reaction is that fatty acids, released
548
from extracted triglycerides by saponification, reduce Most methods employ pure tripalmitin as the
dichromate to form blue-green chromous ions by the standard.
following reaction:
Near-infrared reflectance spectrophotometry
K2Cr2O7 + 5H2SO4 + 3H+ + R-COOH (NIRRS; Table 1, Method 8) has been used to
quantitate fecal fat concentration [25]. In this
(yellow-orange)
technique, radiation in the near-infrared spectrum that
is scattered by the surface of a homogenized fecal
Cr2(SO4)4 + K2SO4 + CO2 + 7H2O specimen is measured. Measurement of the scattered
(blue-green) radiation at different wavelengths enables the
determination of total nitrogen content, total fat, and
The absorbance of the chromous ions measured at hydrolyzed fat. With the availability of the NIRRS
600 nm is proportional to the amount of fat in the equipment, this technique is both rapid and relatively
stool. simple to perform.
Thin-layer chromatography and gas-liquid Another method for the determination of total fecal
chromatography have been applied to the quantitation fatty acids is the use of nuclear magnetic resonance
of fatty acids in organic extracts of stool (Table 1, (NMR; Table 1, Method 9). The common unsaturated
Methods 4a and 4b) [17,18]. Although often used for fatty acids (oleic, linoleic, linolenic) can be
research, these methods are very rarely used for quantified using 1H-NMR by utilizing the area-per-
routine analysis, because they are technically proton (determined by integration) technique
demanding. (http://www.lipidlibrary.co.uk/nmr/nmr.html) [26].
A convenient and rapid spectrophotometric method Equations can then be used to estimate the amounts
using the sulfovanillin reaction [19,20] has been of the unsaturated fatty acids. NMR can also be used
shown to be useful with unsaturated fatty acids but to detect 13C-labeled fat [27].
not C8 or C10 saturated fatty acids. The reaction
(Table 1, Method 5), as suggested by Knight et al. The gold standard, and thus still the most commonly
[21], involves the formation of a carbonium ion at the used procedure for the determination of fecal fat for
double bond by the action of H2SO4 and the reaction the diagnosis of malabsorption syndromes, remains
of an aromatic phosphovanillin with the carbonium the 72-hour quantitative fat estimation, using
ion to form a stable carbonium complex absorbing at methods based on van de Kamer et al. [24].
approximately 525 nm.
An interesting approach is the steatocrit, in which There are no reference methods for fecal fat analysis
homogenized stool is drawn into a capillary tube (http://www.bipm.org/jctlm/), although literature on
(Table 1, Method 6). The tube is sealed at one end by this topic almost always refers to 72-hour fecal
a flame and centrifuged for 15 minutes. The top quantitative fatty acid titratimetric analysis (Table 1,
layer, representing lipid, is measured with calipers Method 7) as the gold standard. This method is also
and expressed as a percentage of the total volume the easiest to establish and requires little specialized
(steatocrit) [22,23]. equipment or specialized laboratory skills. For this
reason, it remains the most frequently performed and
The most commonly used method to measure fecal recommended method.
fat is based on the procedure of van de Kamer et al. For those laboratories that want to screen feces
[24]. This method hydrolyzes fats by reflux of a before performing the quantitative analysis, the
weighed fecal sample in alcoholic KOH. The sample microscopic technique for fat malabsorption is
is then acidified and the fatty acids extracted into recommended. It is simple to perform and reasonably
petroleum ether. The solvent is then evaporated, the sensitive. Drummey [27] reported that the technique
residue dissolved in ethanol, and the fatty acids gave no false-negative results for anyone ingesting
determined by titration against standardized NaOH more than 15% of their diet as fat. Simko [28]
using thymol blue as an indicator (Table 1, Method evaluated this approach and reported that problems
7). Many modifications of the van de Kamer cited in the past may have been attributable to poor
technique have been employed. These include the technique. The method was shown to be less reliable
conditions of saponification (i.e., alcohol when there were increased levels of soaps and when
concentration), choice of extraction solvent, indicator free fatty acids were present, because these do not
(i.e., the use of methyl red), and choice of standard. stain.
549
considered it better than either trioleic or tripalmitic

The quantitative approach to the microscopic acids.
screening technique reported by Fine and Ogunji [6]
was reported to compare well with the 72-hour fat Butler and Gehling [33] demonstrated success with a
quantitation method. However, there has been no 4-hour, 14C breath test. Tomkin examined the
other study reported using this method, so its clinical
possibility of measuring blood 14C but found this
efficacy remains uncertain and can not be
more variable than breath analysis [34]. A number of
recommended.
workers have recognized the problem of low levels of
radioactivity (false-positive results) in obese patients.
The NMR method with 1H proton analysis is used by
Although this has been said to be caused by delayed
large reference laboratories in the United States. With
gastric emptying, it is probably caused by increased
no extraction required and rapid analysis time, the
insulin. To correct for some of the inconsistencies,
capital expense for the NMR equipment seems
Strange et al. [35] added factors for age, body weight,
worthwhile if large numbers of samples are presented
metabolic rate, and respiratory quotient but did not
to these laboratories. These costs will most likely be
believe these improved the diagnostic power of the
prohibitive for small laboratories with few samples.
test.
The remaining methods seem not to be in routine use
and may be better suited for research purposes. A number of reports have demonstrated that the 13C
breath test compared well with the standard 72-hour
The isotopic methods for estimation of fat absorption fecal fat determination. The advantage of using a
have the disadvantage of the need to deal with non-radioactive tracer is certainly attractive.
radioisotopes. Also, the iodine-labeling process However, most of the recent reports that suggest high
renders the triglyceride chemically unstable, and the clinical utility of this non-isotopic breath test have
radioactive iodine can be severed from the lipid by been performed outside the United States [11-14].
the gut mucosa and absorbed independently of lipid. Thus far, the only non-isotopic breath test to be
The labeled triglyceride is often contaminated with approved by the U.S. Food and Drug Administration
radioactive free fatty acids, fatty acid esters, and is the 13C-urea breath test [36].
other fragments that have variable rates of
The steatocrit test has the advantage of rapidity and
absorption. To avoid these problems, 14C label has
simplicity when estimating the amount of fat in stool.
been increasingly used instead of the 125I label. West However, this tests clinical efficacy has been
[29] compared 14C-tripalmitin test results to changes questioned [37], and the steatocrit remains a rarely
in serum turbidity and serum triglyceride after a fat performed procedure.
meal. Only four false-positive results out of 28
healthy persons were observed, and 32 out of 32 The near-infrared reflectance spectroscopy (NIRRS)
abnormals were identified. procedure seems to be an accurate and reproducible
method [38,39]. Although NIRRS has the advantage
Meeker [30] devised a variation of the test in which 5 of not requiring stool extraction, this technique
Ci of tripalmitin-carboxyl-14C was mixed with 35 requires specialized equipment and knowledge and
to 40 mL of an intravenous fat emulsion and thus remains an interesting but not widely used
administered orally to the 124 patients first with, and procedure.
then several days later without, Viokase (650 U of
lipase). Compared to results obtained by fecal fat Major drawbacks of the gravimetric method are the
measurement, this method showed 15% false need to work with dangerous solvents (e.g.,
positives and 7% false negatives. The Viokase petroleum ether, hexane, xylene, and carbon
modification was judged useful to separate pancreatic tetrachloride) and the overestimation of lipid content
from nonpancreatic disease and was helpful in by extraction of nonpolar, nonlipid matter. There may
determining the best course of therapy. Other authors be underestimation of lipid when medium-chain
have stated that the test does not discriminate triglycerides are lost during the evaporation step [40].
adequately [31]. However, Newcomer [10] found that The electrical capacitance approach has been
distinguishing patients with and without pancreatic criticized [41] because petroleum ether extracts of
disease was improved when triolein was used rather medium-chain triglyceride (or medium-chain fatty
than tripalmitin. Ghoos [32] employed a mixed acid) and long-chain triglyceride (or long-chain fatty
acid) have distinctly different electrical capacitances.
triglyceride-1,2-dioleyl-2-14C-decanoyl glycerol and
In fact, if triolein is used as a standard, any medium-
550
chain triglyceride present is underestimated by 27%, Stool Collection

whereas medium-chain fatty acid is overestimated by Whatever technique is selected for analysis, the most
35%. These methods are rarely used in routine important steps in any fecal analysis are the
laboratories. collection and homogenization procedures.
The sulfovanillin method produces decreasing An elaborate but effective method has been described
amounts of color as the number of double bonds in in which the feces are kept deep frozen using a
the fatty acids increases, thus yielding varying results portable toilet [43]. A simpler approach is to use a
as the fatty acid composition changes [20]. In portable campers toilet, to which can be fixed a
addition, many alcohols can be dehydrated by the plastic bag. Each collection is then contained in a
reaction conditions and subsequently enter into the separate bag and frozen.
reaction. Nevertheless, the method demonstrated
good recovery of added lipid (approximately 99.6%) Preweighed 5-gallon paint cans are the most
and good precision. commonly used collection containers, since they are
relatively inexpensive and easy to use and can also be
Several workers, including van de Kamer himself used as the extracting vessels (see below).
[42], have noted that the fatty acid titration method
recovers only about 60% of the medium-chain For many years, dry sampling techniques were used
triglyceride. Recently this has become particularly in fecal fat analysis. These involved thorough mixing
important when the success of medium-chain of the entire stool collection into a homogeneous
triglyceride administration is being monitored for the paste, removal of an aliquot, and drying of the aliquot
treatment of fat malabsorption and in an oven or under a heat lamp. In addition to being
abetalipoproteinemia; it is of questionable importance a rather repugnant procedure, there is very little
in routine work. clinical significance to a result expressed as units of
fat per unit of dry feces. Furthermore, the application
Specimen of intense heat may result in some fatty acids being
Diet volatilized [44].
For all direct methods of fat measurement in stool, it
is important that the patient ingest a normal daily Wet techniques, which involve the addition of a
diet containing a minimum of 100 g of triglyceride known amount of liquid to the specimen, mixing, and
[42]. Rather than prescribing a complex diet, it is removal of an aliquot, are most commonly used.
advisable to insist that the patient ingest 1 ounce of Manual mixing or the use of a blender have been
corn oil and a glass of whole milk with each meal for suggested and are used, but these techniques are not
3 days preceding the stool collection and for the 3 only very messy but are probably also less efficient
days of the collection. than those of Newell [45] and Jover and Gordon [46],
in which the entire collection is placed in a
A 72-hour collection is considered essential. preweighed 5-gallon paint can. This can be done
Defecation is highly variable from day to day (up to directly (as above), or the frozen aliquots may be
several hundred percent), and only this collection added separately and allowed to thaw. The weight of
time will average out the usual variations. Workers the container and complete collection can then be
have reported using markers such as charcoal, Congo determined, and the weight of the collection can be
red, chromic oxide, barium sulfate, and cuprous determined by the difference. The contents of the can
thiocyanate to keep accurate track of the transit time. may then be diluted with a homogenizing solution
These markers are not generally necessary. (such as isopropanol) and the entire mixture
homogenized by vigorous agitation on a paint shaker.
It is difficult to establish criteria by which a stool can
be rejected for analysis. Some laboratories do not Interferences
wish to analyze specimens that are formed. However, Oil-containing cathartics can cause false-positive
although fatty stools are usually fluid, semifluid, or results. In addition, ingestion of Olestra, the artificial
soft, bulky, foul, or foamy borderline conditions may dietary fat, may also cause a false-positive result
produce relatively normal samples. [47]. The most important cause of false-negative
results is insufficient fat ingestion during the test
period. Contamination of the fecal collection with
urine should be avoided.
551
Available at
Fecal Fat Reference Interval <http://www.touchbriefings.co.uk/pdf/953/s
For adults, less than 5 g of fat are normally excreted erreno.pdf>
in 24 hours after ingestion of 50 g of fat in the diet. 5 Holt PR. Diarrhea and malabsorption in the
Steatorrhea is generally defined as the excretion of elderly. Gastroenterol Clin North Am
greater than 7 g of fat per day. Using the near- 2001;30:427-444.
infrared reflectance spectroscopy technique, one 6 Fine KD, Ogunji F. A new method of
group has published reference values for stool fat for quantitative fecal fat microscopy and its
ages < 0.5, > 0.5 to 1.5, > 1.5 to 4, and > 4 to 14 correlation with chemically measured fecal
years [48]. fat output. Am J Clin Pathol 2000;113:528-
34.
For more detailed information, see the detailed 7 Duffy BJ Jr, Turner DA. The differential
procedures at the end of this chapter. diagnosis of intestinal malabsorption with
I131-fat and fatty acid. Ann Intern Med
Interpretation 1958;48:1-7.
Both the total weight and fatty acid excretion per 24 8 Abt AF, von Schuching SL. Fat utilization
hours should be reported. If the daily fat excretion is test in disorders of fat metabolism: a new
greater than 17 mmol, fat malabsorption is occurring. diagnostic method applied to patients
If the daily fat excretion is between 10 and 17 mmol, suffering from malabsorption syndrome,
the patient is not normal and may be suffering from chronic pancreatitis and arteriosclerotic
mild fat malabsorption. The fecal weight is cardiovascular disease. Bull Johns Hopkins
important, since it indicates to the clinician the Hosp 1966;119:316-330.
seriousness of the diarrhea, and very heavy 9 Kaihara S, Wagner HN Jr. Measurement of
specimens with a low fat content may indicate watery intestinal fat absorption with carbon-14
diarrhea. tracers. J Lab Clin Med 1968;71:400-411.
10 Newcomer AD, Hofmann AE, DiMagno EP,
Fecal fat analysis is a primary test to diagnose fat Thomas PJ, Carlson GL. Triolein breath test:
malabsorption in a wide variety of disorders in which a sensitive and specific test for fat
pancreatic insufficiency is a major part of the disease malabsorption. Gastroenterology 1979;76:6-
state [49-53]. 13.
11 Ventrucci M, Cipolla A, Ubalducci GM,
Fecal Fat Performance Goals Roda A, Roda E. 13C-labelled cholesteryl
There are no established performance goals for fat octanoate breath test for assessing pancreatic
absorption analysis, either from proficiency programs exocrine insufficiency. Gut 1998;42:81-87.
or from clinical professional groups. A laboratory 12 Weaver LT, Amarri S, Swart GR. 13C mixed
performing quantitative fecal fat analysis needs to triglyceride breath test. Gut 1998;43(Suppl
base acceptable recovery of extracted fat and 3):S13-S19.
acceptable precision on within-lab data, as well as 13 Ashraf H, Hildebrand P, Meier R, Beglinger
guidance form their user-physicians. C, Gyr N. Induction of artificial fat
maldigestion by tetrahydrolipstatin Assessed
References by the 13C-Hiolein Breath Test in Healthy
1 Guandalini S, Frye RE. Malabsorption Volunteers Double-Blind Controlled Pilot
Syndromes. Available at Study. Digestion 2000;62:159-163.
<http://www.emedicine.com/PED/topic1356 14 Ritz MA, Fraser RJ, Di Matteo AC, Greville
.htm> Accessed 03.10.2006. H, Butler R, Cmielewski P, Davidson G.
2 Klapproth J-M, Yang VW. Malabsorption. Evaluation of the 13C-triolein breath test for
Available at fat malabsorption in adult patients with
<http://www.emedicine.com/MED/topic138 cystic fibrosis. J Gastroenterol Hepatol
4.htm> Accessed 01.02.2007. 2004;19:448-453.
3 Bai JC. Malabsorption syndromes. Digestion 15 Wolochow DA, Broitman SA, Williams LA,
1998;59:530-46. Zamcheck N. A rapid method for
4 Serrano PL, Navarro JLL, Fernndez- quantitative determination of fecal fat based
Rodrguez CM. Laboratory test and on the principle of electrical capacitance. J
equipment for diagnostic workup for Lab Clin Med 1965;65:334-341.
malabsorption syndrome. LabTech 2004. 16 Fales FW. Evaluation of spectrophotometric
552
method for determination of total fecal lipid. steatorrhea. N Engl J Med 1961;264:85-87.
Clin Chem 1971;17:1103-1108. 28 Simko V. Fecal fat microscopy: acceptable
17 James AT, Webb JPW, Kellock TD. The predictive value in screening for steatorrhea.
occurrence of unusual fatty acids in faecal Am J Gastroenterol 1981;75:204-208.
lipids from human beings with normal and 29 West PS, Levin GE, Griffin GE, Maxwell
abnormal fat absorption. Biochem J JD. Comparison of simple screening tests
1961;78:333-339. for fat malabsorption. Br Med J
18 Thompson JB, Langley RL, Hess DR, 1981;282:1501-1504.
Welsh JD. Fecal triglycerides. I. Methods. J 30 Meeker HE, Chen I, Connell AM, Saenger
Lab Clin Med 1969;73:512-520. EL. Clinical experiences in 14C-tripalmitin
19 Tomaszewski L. A new, convenient, rapid breath test for fat malabsorption. Am J
and accurate method for determination of Gastroenterol 1980;73:227-231.
total lipids in feces. Clin Chim Acta 31 Chen IW, Azmudeh K, Connell AM,
1975;61:113-120. Saenger EL. 14C-tripalmitin breath test as a
20 Frings CS, Dunn, RT. A colorimetric diagnostic aid for fat malabsorption due to
method for determination of total serum pancreatic insufficiency. J Nucl Med
lipids based on the sulfo-vanillin reaction. 1974;15:1125-1129.
Am J Clin Pathol 1970;53:89-91. 32 Ghoos YF, Vantrappen GR, Rutgeerts PJ,
21 Knight JA, Anderson S, Rawle JM. Schurmans PC. A mixed-triglyceride breath
Chemical basis of the sulfo-phospho-vanillin test for intraluminal fat digestive activity.
reaction for estimating total serum lipids. Digestion 1981;22:239-247.
Clin Chem 1972;18:199-202. 33 Butler RN, Gehling NJ. Modification of the
14
22 Phuapradit P, Narang A, Mendonca P, C triolein breath test. Clin Chim Acta
Harris DA, Baum JD. The steatocrit: a 1981;112:371-374.
simple method for estimating stool fat 34 Tomkin GH, Bell TK, Hadden DR.
content in newborn infants. Arch Dis Child Evaluation of malabsorption test using 14C
1981;56:725-727. triglyceride. Irish J Med Sci 1971;140:449-
23 Colombo C, Maiavacca R, Ronchi M, 454.
Consalvo E, Amoretti M, Giunta A. The 35 Strange RC, Reid J, Holton D, Jewell NP,
steatocrit: a simple method for monitoring Percy-Robb IW. The glyceryl (14C)
fat malabsorption in patients with cystic tripalmitate breath test: a reassessment. Clin
fibrosis. J Pediatr Gastroenterol Nutr Chim Acta 1980;103:317-323.
1987;6:926-390. 36 Klein PD. 13C breath tests: visions and
24 van de Kamer JH, Ten Bokkel Huinink H, realities. J Nutr 2001;131:1637S-1642S.
Weyers HA. Rapid method for the 37 Wagner MH, Bowser EK, Sherman JM,
determination of fat in feces. J Biol Chem Francisco MP, Theriaque D, Novak DA.
1949;177:347-355. Comparison of steatocrit and fat absorption
25 Stein J, Purschian B, Bieniek U, Caspary in persons with cystic fibrosis. J Ped
WF, Lembcke B. Near-infrared reflectance Gastroenterol Nut 2002;35:202-205.
analysis: a new dimension in the 38 Neucker AV, Bijleveld CM, Wolthers BG,
investigation of malabsorption syndromes. Swaaneburg JC, Kester AD, Kreel B, Forget
Eur J Gastroenterol Hepatol 1994;6:889- PP. Comparison of near-infrared reflectance
894. analysis of fecal fat, nitrogen, and water
26 Schneider MU, Demling L, Jones SA, with conventional methods, and fecal energy
Barker PJ, Domschke S, Heptner G, content. Clin Biochem 2002;35:29-33.
Domschke W. A new method for total stool 39 Volmer M, Kingma AW, Borsboom PCF,
fat quantification in chronic pancreatitis. Dig Wolthers BG, Kema IP. Investigation of
Dis Sci 1987, 32:494-499. applicability of a mid-infrared spectroscopic
27. P. Kunz P, Knnecke B, Kunz I, Lengsfeld method using an attenuated total reflection
H ,von Kienlin M. Natural abundance 13C- accessory and a new near-infrared
NMR spectroscopy for the quantitative transmission method for determination of
determination of fecal fat. Clin Biochem faecal fat. Ann Clin Biochem 2001;38:256-
2003;36:505-510. 263.
28 Drummey GD, Benson JA Jr, Jones CM. 40 Henry RJ. Clinical Chemistry: Principles
Microscopical examination of the stool for and Technics. New York: Harper & Row;
553
1964:833. 48 Rivero-Marcotegui A, Olivera-Olmedo JE,

41 Hallaway BE, Sandberg R. Determination of Valverde-Visus FS, Palacios-Sarrasqueta M,
fecal triglycerides and fatty acids of Grijalba-Uche A, Garca-Merlo S. Water,
medium-chain length by electrical fat, nitrogen, and sugar content in feces:
capacitance and by petroleum ether reference intervals in children. Clin Chem
extraction. Clin Chem 1970;16:408-411. 1988;44:1540-1544.
42 van de Kamer JH. Quantitative 49 Jeppesen PB, Christensen MS, Hoy CE,
determination of saturated and unsaturated Mortensen PB. Essential fatty acid
higher fatty acids in fecal fat. Scand J Clin deficiency in patients with severe fat
Lab Invest 1953;5:30-36. malabsorption. Am J Clin Nutr
43 Ghoos Y, Vantrappen G. Clean collection 1997;65:837-843.
and manipulation of stools. Lancet 50 Carroccio A, Fontana M, Spagnuolo MI,
1977;1:884-885. Zuin G, Montalto G, Canani RB et al.
44 van de Kamer JH. Total fatty acids in stool. Pancreatic dysfunction and its association
In: Seligson D, ed. Standard Methods of with fat malabsorption in HIV infected
Clinical Chemistry. Vol. 2. New York: children. Gut 1998;43:558-563.
Academic Press; 1958:34-39. 51 Kalivianakis M, Minich DM, Bijleveld
45 Newell JE. Quantitative chemical assay of CMA, van Aalderen WMC, Stellaard F,
feces, collecting, mixing, sampling. Am J Laseur M et al. Fat malabsorption in cystic
Clin Pathol 1960;34:189-190. fibrosis patients receiving enzyme
46 Jover A, Gordon RS Jr. Procedure for replacement therapy is due to impaired
quantitative analysis of feces with special intestinal uptake of long-chain fatty acids.
reference to fecal fatty acids. J Lab Clin Am J Clin Nutr 1999;69:127-134.
Med 1962;59:878-884. 52 Steer ML, Waxman I, Freedman S. Chronic
47 Balasekaran R, Porter JL, Santa Ana CA, Pancreatitis. N Eng J Med 1995;332:1482-
Fordtran JS. Positive results on tests for 1490.
steatorrhea in persons consuming olestra 53 Holt PR. Diarrhea and malabsorption in the
potato chips. Ann Intern Med 2000;132:279- elderly. Gastroenterol Clin North Am
282. 2001;30:427-44.
554
Tables
Table 1: Methods of Quantitative Fecal Fat Measurement
Method 1: Gravimetric
Principle of analysis: Fat-soluble material is extracted, and after solvent removal, is weighed.
Comments: Historical; nonspecific; requires handling of dangerous solvents
Method 2: Capacitance
Principle of analysis: Fatty acids present in a solvent extract of stool decrease electrical capacitance of the
solvent.
Comments: Historical; requires special equipment; difficult to standardize properly
Method 3: Dichromate reduction
Principle of analysis: Potassium dichromate is reduced by saponified, extracted fatty acids to form blue-
green chromous ions and measured spectrophotometrically at 600 nm.
Comments: Rare; nonspecific; different color yield with different fatty acids
Method 4: Chromatographic
a. Thin-layer chromatography
b. Gas-liquid chromatography
Principle of analysis: Saponified, extracted fatty acids are separated and quantitated.
Comments: Research only; requires special equipment; results have limited clinical significance
Method 5: Sulfovanillin reaction
Principle of analysis: Saponified, extracted, unsaturated fatty acids react with vanillin to form a colored
carbonium ion complex whose absorbance can be measured at 525 nm.
Comments: Rare; color yield dependent on number of double bonds in fatty acids; good precision
Method 6: Steatocrit
Principle of analysis: Homogenized stool specimen is drawn into a capillary tube and centrifuged.
Volume of fatty layer is described as percentage of total.
Comments: Rare; nonspecific
Method 7: Titration
Principle of analysis: Extracted, saponified fatty acids are titrated using thymol blue as an indicator.
Comments: Most common; can underestimate amount of medium-chain fatty acids
Method 8: Near-infrared reflectance assay
Principle of analysis: Radiation in the near-infrared spectrum that is scatted by the surface of a fecal
sample is measured.
Comments: Simple and reliable method; allows determination of dry weight, total fat, and hydrolyzed fat
Method 9: Nuclear magnetic resonance (NMR)
Principle of analysis: The nuclear magnetic resonance spectra of 1H protons of individual fatty acids
measured and the total amounts of fatty acids calculated.
Comments: Does not require hydrolysis or extraction of the sample.
555
Figures
Figure 1: Fecal Fat Analysis
Standard curve for fecal fat analysis: absorbance at 502 nm versus amount of tripalmitin standard.
Procedure: Microscopic Screen (Simko 1981 5. Mix thoroughly with a glass rod.
Modified) [27] 6. Fill a hemocytometer chamber with the
mixture.
Principle 7. Warm for 2 min on a 37C heating block to
Stool sample is mixed with a fat-staining dye. The hydrolyze the fecal soaps into fatty acids.
colored droplets are counted by microscopic 8. While it is still warm, count the pink fat
examination. If the number of droplets exceeds droplets in 10 high-power (400) fields, and express
normal limits, malabsorption is strongly indicated. the results using the table below:
Reagents
1. Sudan III dye reagent, 5 g/L. Dissolve 1 g Calculation and Interpretation
of Sudan III dye reagent in 100 mL of 95% ethanol Average number Approximate
and 100 mL of glacial acetic acid to form a saturated of fat droplets fat content
solution. Stable for 6 months at room temperature. 10 5%
20 10%
30 15%
Assay 35 20%
Equipment: hemocytometer, heating block, >40 >30%
and a microscope.
1. Collect stool as a single sample or as a
prolonged collection. Fat Absorption Reference Interval
2. Mix thoroughly. Normal: Less than 5% fat content.
3. Place a 500-mg aliquot of the sample in a
100-mL beaker.
4. Add 0.5 mL of a Sudan dye reagent.
556
Procedure: Chemical Quantitation of Fecal Fat topped culture tubes and drying with a stream of
Output [52] warm air. These standards are equivalent to 5, 10, 20,
30, and 40 mmol of fatty acid per liter of homogenate
Principle or 10, 20, 40, 60, and 80 mmol of fatty acid per total
A well-mixed stool sample is saponified by treatment specimen. It is convenient to prepare several sets of
with alcoholic KOH at 80C. The free fatty acids are standards at one time and store them, tightly capped,
extracted and quantitated by spectrophotometric at room temperature. They are stable for several
observation of their reaction with methyl red. The months.
amount of fat present in the sample is calculated by 9. Carboxymethyl cellulose. Add 4 g of
comparison of the unknown with a tripalmitin carboxymethyl cellulose (sodium salt) to 1 L of
standard. distilled water. Mix to dissolve, and store at 4C to
8C for up to 6 months.
Reagents
1. Alcoholic potassium hydroxide, 0.357 Assay
mol/L of ethanol. Place 2.0 g of KOH in a 100-mL Equipment: Preweighed paint cans, paint shaker,
volumetric flask, dissolve, and dilute to volume with water bath (50C to 60C), and spectrophotometer
absolute ethanol. Prepare fresh each day. (band pass 10 nm) capable of reading at 502 nm.
2. Sodium hydroxide, 1.0 mol/L. Place 4.0 g Sample preparation:
of NaOH pellets in a 100-mL volumetric flask. Add 1. The patient ingests a high-fat diet as
approximately 80 mL of distilled water to dissolve previously described.
pellets. Allow solution to cool, add water to mark, 2. Feces collected into plastic bags are placed
and mix well. Store in plastic container at room in preweighed paint cans. This is easily done when
temperature for up to 6 months. the specimens are frozen.
3. Hydrochloric acid, 0.6 mol/L. Place 50 3. When the analysis is to be performed, the
mL of concentrated HCl in a liter volumetric flask can and its contents are weighed and the weight of
containing 800 mL of distilled water, and add the fecal collection is determined by subtraction.
distilled water to mark. Mix well, and store at room 4. The cans contents are diluted to 970 g with
temperature for up to 6 months. water, and 1 L of isopropanol is added.
4. Petroleum ether (boiling point 39C to 5. Next, 2 g of NaOH pellets and 100 mL of 4
53C), redistilled. g/L carboxymethyl cellulose solution are added. A
5. Methyl red (Fisher Scientific Co., handful of washed pebbles or metal washers is also
Pittsburgh), 2 g/L. Dissolve 200 mg of powder in included.
100 mL of 95% ethanol. Solution proceeds slowly. 6. The can is placed on a commercial paint
Store at 4C to 8C for up to 1 year. shaker and agitated vigorously for 15 min.
6. Sodium acetate, 1 mol/L. Place 13.6 g of 7. Without delay, the can is opened, and an
sodium acetate trihydrate into a 100-mL volumetric aliquot of the newly formed emulsion is removed by
flask. Add approximately 80 mL of distilled water, use of a pipet with the tip removed or a plastic
and dissolve salt. Add water to mark, and mix well. drinking straw. The aliquot may be stored
Store at room temperature for up to 2 months. indefinitely in the frozen state.
7. Buffered methyl red reagent. Prepare this 8. If the sample weighs more than 970 g, or a
reagent by photometric measurement to avoid very high fat content is suspected, an additional
variations in dye lots. Add 10 mL of 1 mol/L NaOH dilution should be made by addition of water to a
to 1000 mL of 95% ethanol. Add enough indicator weight of 1825 g, 4 g of NaOH, and 200 mL of
(10 to 13 mL) to bring absorbance to a range of 0.095 carboxymethyl cellulose solution, followed by 1890
to 0.100 at 502 nm. Add 2.0 mL of 1 M sodium mL of isopropyl alcohol. Shake for 30 min, and
acetate solution. Using a magnetic stirring device, multiply the final result by 1.9.
add 1 mol/L HCl by drops until a faint orange-red Analysis:
tinge persists. Carefully add more acid until the 1. Remix aliquot of fecal emulsion vigorously.
solution has an absorbance of 0.195 0.005. If this 2. Pipet 0.20 mL into a 15 150-mm screw-
point is passed, titrate back with dilute NaOH, being topped culture tube.
very careful not to dilute the reagent excessively. 3. Add 2.0 mL of alcoholic KOH, and cap
8. Tripalmitin standards. Dissolve 2.5 mg of tightly with a Teflon-lined cap.
tripalmitin in 200 mL of heptane. A standard curve 4. Incubate for 60 min at 80C.
may be prepared by addition of 0.25, 0.50, 1.00, 1.50, 5. Add 3.6 mL of 0.6 M HCl, and cool to room
and 2.00 mL of this solution to 15 150-mm screw- temperature.
557
6. Add 10 mL of petroleum ether, and shake 5. Count for 10 min using a 1:10 dilution of the
vigorously for 2 full min. administered dose as a standard.
7. Centrifuge at 790 g for 5 min. 6. Express the result as a percentage of the
8. Transfer 5.0 mL of petroleum ether (upper original dose per mmol of CO2 per kilogram of body
phase) to a 15 125 mm screw-capped culture tube, weight.
carefully avoiding lower phase.
9. Place tubes in water bath (50C to 60C), Calculation
and evaporate ether with a gentle stream of air. 2- to 4-h rate = 4-h cumulative 2-h cumulative
10. Add 3.0 mL of buffered methyl red reagent 2
to dry residue, stopper tightly, warm for 2 min in Fat Absorption Reference Interval
water bath (50 to 60C), and shake vigorously to Normal: (2- to 4-h rate) > 2%/h.
ensure solution.
11. Read absorbance of colored solution against Fat Stain (for Examining Fat Droplets in Stool)
isopropanol at 502 nm in a spectrophotometer. I. Principle of Test and Clinical Significance
Calculation The microscopic examination of feces for fat

Plot absorbance of tripalmitin standards against is performed as a screen to assess digestive function.
concentration on rectilinear graph paper to obtain Increased fat in the stool is an indication of
standard curve, and read concentration of unknowns steatorrhea or malabsorption of fat. Neutral fat
directly from the curve. An example of a standard droplets are stained with alcoholic Sudan III,
curve is shown in Fat Absorption Figure 1. Divide appearing as yellow-orange or red droplets.
result by 3 to obtain total fatty acid excretion per 24
II. Sample (Specimen)
h.
A stool specimen is collected in a clean, fat-
Fat Absorption Reference Interval free container (glass, Pyrex, or polyethylene). Waxed
Less than 5 g of fat are normally excreted in 24 h
containers are not suitable.
when a person has ingested 50 g of fat in a diet.
III. Reagents
Procedure: Breath Test (Butler) [32]
Principle A. Sudan III dye. Fisher certified
14C-Triolein is ingested with a fat load, and one biological stain, catalog #5669. Store at room
temperature. Stable to expiration date on vial.
measures the amount of 14CO2 produced by trapping B. 95% ethanol (190 proof). Obtain
expired air and counting dissolved CO2 in a liquid from UH Pharmacy. Store stock container in
scintillation counter. flammable cabinet at room temperature. Stable to
expiration date on bottle.
Reagents C. Sudan III stain, a saturated solution
14C-Triolein. Obtained from Amersham in 95% ethanol. Add Sudan III dye to 100 mL 95%
1.
ethanol. Swirl to mix. Solution is saturated when dye
(Arlington Heights, IL) in a capsule form.
will no longer dissolve. Stable 1 year at room
2. Trapping reagent. click here
temperature.
3. Scintillation cocktail. click here
4. Collection device. click here
IV. Equipment
Assay
A. Microscope
1. Administer 5 Ci of 14C-triolein
B. Microscope slides and coverslip
(Amersham) in a gelatin capsule followed by 40 mL
C. Wooden applicator sticks
of fat (corn oil emulsion; 30% oleic acid).
2. Exhale through a 2 mL solution of hyamine
V. Calibration: N/A
hydroxide (2 mmol/L). (See the 14C-CO2 breath test
by McNeely.)
VI. Quality Control
3. Take duplicate samples for the baseline
value and each hour after the isotope has been
administered. VII. Procedure and Methodology
4. Add 10 mL (6 g/L PPO in toluene) to each
vial. A. Place a small aliquot of stool
558
suspension on a slide with a wooden applicator stick.

Add 2 drops of 95% ethanol and 2 drops of saturated
Sudan III, and mix with the applicator stick. IX. Expected Result: Negative
A. Reference range 311 mIU/mL
B. Place a coverslip over the B. No critical values.
specimen, and examine under high power (400).
Neutral fats appear as large, yellow-orange or red
X. Procedure Notes: None
droplets.
VIII. Calculations (Derivation of Results)
XI Procedure Performance: N/A
The presence of 60 or more stained droplets
of neutral fat per high-power field is considered
positive.
559
Fecal Occult Blood
Fecal Occult Blood

R. Swaminathan
Name: Fecal occult blood

Clinical significance: Refer to Chapter 35, Gastrointestinal function, in the 5th edition of Clinical
o-tolidine [5]. By employing a stabilizer, the sensitivity of
Principles of Analysis and Current Usage the method has been improved.
Fecal occult blood testing is performed to detect bleeding These tests consist of a card containing a high-quality filter
in the gastrointestinal tract. As the name implies, it refers paper impregnated with guaiac. This is stable for long
to bleeding that is not apparent to the patient. periods because the guaiac is not in solution. The
developing solution is stabilized hydrogen peroxide in an
When blood enters the upper gastrointestinal tract, the aqueous alcoholic solution. Hemoglobin and its iron-
globin part of the hemoglobin molecule is completely containing degradation products, due to the
digested by the proteolytic enzymes, and the heme is pseudoperoxidase activity, release oxygen from hydrogen
converted by bacterial action to porphyrins. Hemoglobin peroxide. The oxygen then oxidizes alpha-guaiaconic acid,
entering the lower part of the large intestine is largely a phenolic compound present in guaiac. A quinine
undigested. In normal subjects, the amount of blood lost structure is formed, and this rearranges to a blue dye by
from the gastrointestinal tract is 0.5 to 1.5 mL per day internal electron transfer [9].
[1,2,3]. This amount of blood is not usually detected by the
fecal occult blood test. Immunological Methods
These methods are more specific and use antibodies
Common methods for detection of occult blood are based against one of the components in blood, most commonly
on the detection of hemoglobin or its breakdown products. against the globin chain of the hemoglobin. Hemoglobin
Blood in feces can also be detected by macroscopic forms a complex with a conjugate of an antibody to
examination of feces for blood cells or hematin crystals or hemoglobin. The conjugate consists of a monoclonal or
spectroscopic identification of hemoglobin and its polyclonal antibody attached to a dye or enzyme which
derivatives [4,5]. will produce a colored product from the substrate present
in the system.
Tests commonly used in clinical laboratories are based on
the detection of hemoglobin, heme, or heme-derived A variety of immunochemical detection systems have been
porphyrins. Hemoglobin is detected by immunological described. These include enzyme immunoassays (EIA)
methods, heme by guaiac-based methods utilizing the [10], hemagglutination [11], latex agglutination [12], and
pseudoperoxidase activity of heme, and porphyrins by colloidal gold agglutination assay [13].
fluorimetry.
Many commercial kits have been developed for the
Guaiac-Based Methods detection of blood by immunochemical method, and some
The most common method for detection of fecal occult of these are automated. Immunochemical methods can be
blood is that based on the detection of heme. In these performed as a point-of-care method at the bedside by a
methods, pseudoperoxidase activity of heme liberates health care professional, by the patient (methods such as
nascent oxygen from hydrogen peroxide. The liberated Hemoccult ICT), or performed in the laboratory.
oxygen oxidizes a chromogen. Historically benzedrine and
o-tolidine were used as the chromogen [6]. These methods Immunochemical methods do no require dietary restriction
are sensitive, but benzedrine and o-tolidine are and therefore are better accepted by the public. Some
carcinogenic and are no longer used in clinical laboratories immunochemical methods can be automated and therefore
[7]. Other chromogens used in this type of test are are more reproducible. Immunochemical methods are also
imipramine hydrochloride and desipramine hydrochloride more sensitive, that is, the detection limits are lower than
[8]. The most common chromogen used now is guaiac, a the guaiac-based methods. With automated systems, the
natural resin extracted from Guaiacum officinale. The reproducibility is increased further as the subjective nature
sensitivity of the guaiac method is less than that based on of visual reading of a result is removed. These methods are
capable of detecting as little as 0.3 mL of blood added to
i stool [14]. These methods are also more specific for blood
Fecal Occult Blood
from the lower gastrointestinal tract, especially the colon,
New method
Fifth edition: R. Swaminathan
560
Fecal Occult Blood
because the globin released from hemoglobin in the upper peroxidases or dietary red meat. The application of this
gastrointestinal tract is hydrolyzed by proteolytic enzymes. method for routine use or screening has not been shown.
This instrument is expensive; despite its many advantages,
In one device, the MonoHaem, a combination of it may not be practical to use it in clinical laboratories at
immunological and guaiac-based tests is used. Hemoglobin present. The sensitivity (lower detection limit) of this
is immobilized by a monoclonal antibody, and the method is 0.01 mg/g feces compared to 3 to 10 mg
hemoglobin is then visualized by the guaiac-based hemoglobin/g feces for the Hemoccult method [15].
reaction. Non-hemoglobin peroxidases will give a
background blue color which is discounted. An ELISA method to detect transferrin and globin in feces
Heme-Porphyrin Test has also been described [23].
This test is based on the fact that the heme in the
hemoglobin entering the gastrointestinal tract is converted Reference and Preferred Methods
to porphyrins, probably by gut bacteria. In this method, There is no reference method for fecal occult blood
porphyrins in the feces are extracted and then quantitated determination.
by spectrofluorimetry. This allows exact quantitation of
the hemoglobin entering the gastrointestinal tract. This Many commercial methods based on guaiac are available.
method is better at detecting bleeding in the upper These include Hemoccult (Beckman-Coulter), Hema
gastrointestinal tract. Guaiac and immunological methods Screen (Immunostics, Inc), Hemoccult SENSA (Beckman-
are unreliable in detecting upper gastrointestinal tract Coulter), ColoScreen (Helena lab), and HEMDETECT
bleeding, because the hemoglobin may be digested by (DIPROmed, Handels GmbH, Weigelsdorf, Austria). All
proteolytic enzymes in the gut [15]. guaiac methods are qualitative, and the sensitivity of these
methods is not as good as the immunochemical methods.
In addition to hemoglobin, other heme-containing proteins In a recent evaluation of some of these kits, the lower
will also contribute to the amount of porphyrin detected by detection limit was found to be 0.65, 0.9, 0.6, and 0.3 mg
this method. Furthermore, the conversion of heme and Hb/g feces for ColoScreen, Hema Screen, HEMDETECT,
other heme proteins to porphyrins is an incomplete process and Hemoccult, respectively [24]. Hemoccult SENSA has
and depends on colonic transit time, site of bleeding, and a slightly lower detection limit than Hemoccult and is also
amount of luminal heme [16]. Quantification of fecal reported to be more reliable [25]. To improve the
hemoglobin is therefore likely to underestimate the amount sensitivity of the method, it has been shown that fecal
of hemoglobin from the bleeding. Comparison of fecal rehydration improves the sensitivity but with loss of
heme-porphyrin (HemoQuant) with estimation of blood specificity. Rehydration is done by adding about 25 L of
loss by the 51Cr-labelled red cell method in patients with water to each fecal smear 1 minute before adding the
osteoarthritis on aspirin showed that the HemoQuant developer. It has been shown that fecal hemoglobin
underestimated the blood loss [17]. Recovery of ingested concentrations must exceed 10 mg/g feces, equivalent to
blood by healthy volunteers was only 63% by HemoQuant, 10 mL of daily blood loss, before the Hemoccult II test
showing that absorption of heme may be a problem [18]. will be positive 50% of the time [26]. However, stools
HemoQuant measures heme and dicarboxylic porphyrins containing as little as 1 mg of Hb/g of feces can give
by fluorimetry. However, iron in the heme molecule can positive results [27].
cause quenching, and a reduction step with oxalic acid is
usually introduced to remove the iron [19]. In patients with Many commercial kits based on immunological methods
the rare inherited disorders of porphyrin metabolism, this are also available. These include Hemoccult (Beckman
test will give false-positive results. Coulter Inc, USA), HemeSelect (Beckman-Coulter Inc,
USA), MonoHaem ( Nihon pharmaceutical, Japan),
Other Methods FlexSure OBT (Beckman-Coulter Inc, USA). Of these kits,
Methods based on the detection of haptoglobin [20] and some such as HemeSelect and FlexSure OBT are no longer
calprotectin [21] have been described. These methods have available. Hemoccult ICT is a colloid agglutination
not been fully evaluated. Hoff et al. found that the method and HemeSelect is a hemagglutination assay.
calprotectin method was not as good as the FlexSure occult Many of these methods, like guaiac-based methods, are
blood test [21]. Recently a method using matrix-assisted qualitative or semi-qualitative. Quantitation is possible
laser desorption ionization/time of flight mass with immunological methods, and automated systems are
spectrometry (MALDI-TOF-MS) has been described [22]. also available. The Magstream 1000/Hem SP and the OC-
In this method, fecal sample was mixed with water, Auto Micro 80 are automated systems utilizing
ultrasonicated, and after centrifugation, the supernatant immunochemical methods to detect occult blood [28,29].
was mixed with a matrix solution and used for MALDI- Heme-porphyrin testing is a quantitative method requiring
TOF-MS. Blood was detected by identifying the water- expensive instrumentation. A comparison of methods is
soluble and chains of hemoglobin. The sensitivity of given Table 1.
the method was 10 to 100 times higher than conventional
methods, and there was no interference from plant
561
Fecal Occult Blood
The selection of a method depends on the purpose of the the collection. For guaiac-based tests, raw fruits and
assay, as well as cost and convenience. Occult blood tests vegetables which contain high levels of peroxidase (e.g.,
can be used diagnostically to detect bleeding in the turnip, broccoli, horseradish, cauliflower, parsnip, and
gastrointestinal tract or as a screening test to detect radish) should be avoided.
colorectal neoplasms. As shown in Table 2, heme-
porphyrin testing is more useful for detecting gastric For 7 days before and during the collection period, it is
bleeding, whereas guaiac or immunological methods are best to avoid nonsteroidal antiinflammatory drugs
better for the detection of bleeding from the distal bowel (NSAIDs) such as aspirin or ibuprofen. Vitamin C in
(colon and rectum). excess of 250 mg per day should be avoided for 3 days
before and during collection.
The selection of a method for screening for colorectal
neoplasms depends on the sensitivity and specificity of the Interferences
method, as well as cost and acceptability of the method to Guaiac-based methods are prone to interferences from
the population to be screened. A systematic review of the many sources. One of these is plant peroxidases. Raw
literature by the Centre for Reviews and Dissemination fruits and vegetables such as turnips, broccoli, horseradish,
(CRD) on the accuracy of the immunological and guaiac cauliflower, cantaloupe, parsnip, and red radish contain
methods came to the following conclusions: high concentrations of peroxidases [34]. These peroxidases
There is not enough evidence to suggest that one are heme proteins and have the prosthetic group ferri-
method has better diagnostic accuracy. protoporphyrin IX (hemin) and cause false-positive results
There is limited evidence suggesting that in guaiac-based tests. However, it has been shown that
immunochemical methods may have better cooking vegetables at 100C for 20 min inactivates the
overall performance, but more evidence is plant peroxidase activity [34]. Unlike hemoglobin, because
required. these peroxidases are within the cellulose cell wall, they
There is very little choice between different escape digestion by proteolytic enzymes during the
guaiac-based methods in detecting colorectal passage down the gastrointestinal tract. Inclusion of
neoplasia [30]. moderate amounts of raw vegetables had no effect on the
However, some guaiac-based methods perform poorly in Hemoccult test [35]. Furthermore, gastric acid denatures
external quality-assurance programs (see below). peroxidase activity, so in patients with normal gastric-acid
secretion, ingestion of raw vegetables should not be a
Specimen problem [36]. Ingestion of 750 g of raw, peroxidase-rich
Feces should be sampled before they come into contact fruit and vegetables daily can cause false-positive results,
with the toilet water, since hemoglobin will leach out of but this amount of vegetables is unusually large to be eaten
the sample, and toilet sanitizers may affect the results. daily.
Because blood may not be uniformly distributed within the
stools, and bleeding may be intermittent, it is important to Delaying the development of the slide by 48 hours will
collect samples from more than one area, and the reduce interferences from plant peroxidases [37].
collection should be repeated on 3 different days [31]. It is Nevertheless, many manufacturers of guaiac-based tests
best not to expose the sample to extreme heat or humidity; recommend exclusion of high-peroxidase-containing fruits
dried samples can be stored at room temperature for 14 and vegetables prior to and during collection of samples.
days [32]. If there is a delay between sample collection and
analysis, false-negative results may be seen because of the Ingestion of red meat can cause false-positive results due
degradation of pseudoperoxidase activity of heme in moist to peroxidase activity of heme in the meat [38]. The false-
feces. If samples are collected directly onto the filter paper positive rate was higher when the rehydration method was
in the test kit and allowed to dry, this problem can be used before analysis [35]. Even after cooking the meat,
prevented [33]. In some immunological methods, some peroxidase activity can be detected [34]. Studies in
collection devices with liquid preservatives are used, and healthy volunteers have shown that it takes about 3 days
samples from patients can be collected by smearing the for the false positive due to ingested meat to disappear
stools onto the card provided. For most methods, some [39]. Hence the recommendation is made that red meat
patient preparation is required to give optimum results. should be avoided for at least 3 days prior to testing.
However, the length of time it takes for the false-positive
Patient Preparation and Guidelines results to disappear after stopping red meat may vary in
Samples should not be collected if blood is visible in the different clinical situations and with people with altered
stools or urine (e.g., menstruation, active hemorrhoids, or bowel movements. However, others have not found any
urinary tract infection). Each stool sample should be significant effect of red meat on occult blood tests [40]. A
collected before the stools come into contact with toilet meta-analysis of available data until 1999 concluded that
water. For 3 days before and during the stool collection, dietary restriction may not be necessary for guaiac-based
avoid red meat (beef, lamb, and liver) (see below). A high- fecal occult blood tests [41].
residue diet is recommended for 2 days before and during
562
Fecal Occult Blood
Ascorbic acid (Vitamin C) can also cause a negative effect Interpretation

on the oxidation of alpha-guaiaconic acid, because vitamin Normal subjects lose about 1.5 mL of blood per day, based
C is a reducing agent. Ingestion of 1 to 2 g of vitamin C on radio chromium (51Cr-labeled red cells) methods [52].
daily can cause a false-negative result [42]. In-vitro studies In patients with carcinoma of the large bowel or adenomas,
suggest that a normal intake of vitamin C is unlikely to bleeding is often microscopic but even so can be in excess
cause false-positive results [43]. In a recent study, it was of 1.5 mL per day [35,53,54]. However, bleeding can be
found that consumption of 60 mg of vitamin and 500 mL intermittent and not always present. Thus it is not always
of orange juice (350 mg of vitamin C) produced variable possible to distinguish normal from pathological bleeding
results. In subjects taking high-dose supplements, false- by measuring fecal blood loss [35]. Furthermore, the
negative results were seen [24]. It is therefore amount of blood loss from tumors can vary from day to
recommended by the manufacturers of all occult blood day [55,56], and there is non-uniform distribution of
testing kits that vitamin C intake should not be more than hemoglobin and its products in the feces [57]. Peptic ulcer,
250 mg/day (for Hema Screen) or 500 mg/day (for hemorrhoids, and diverticula are also causes of fecal occult
Hemoccult SENSA). blood loss; bleeding from these conditions is highly
variable. Of all the methods available, heme-porphyrin
Ingestion of iron supplements may cause false-positive testing is more likely to detect small amounts of blood loss
results [44]. However, others have not confirmed these from the upper gastrointestinal tract, such as that seen due
findings [45]. Some test manufacturers caution against the to aspirin [17].
use of iron supplements (Hema Screen), but others do not
(Hemoccult SENSA). Fecal occult blood testing is performed to detect
gastrointestinal bleeding and is most often used as a
The use of povidone-iodine antiseptic solution was screening test to detect colorectal cancer, which is one of
associated with false-positive results with guaiac-based the most common cancers in many parts of the world. This
assays [46]. In-vitro studies showed that that as little as is also an important investigation in patients with iron-
0.005 mL of 1:1000 dilution of this solution will give a deficiency anemia to rule out gastrointestinal bleeding as
false-positive result. This interference is due to the iodine the cause of anemia. Since bleeding from any part of the
in the antiseptic causing oxidation of the alpha-guaiaconic gastrointestinal tract can cause iron deficiency, it is
acid [47]. The use of such antiseptic solutions on the important to select the appropriate test. Testing based on
perianal area or during urinary catheterization should be guaiac or immunological methods is less likely to detect
avoided prior to fecal occult blood testing using guaiac upper gastrointestinal tract bleeding; heme-porphyrin
methods. testing is better at detecting upper gastrointestinal tract
bleeding [58] and was able to detect 90% of upper
Drugs such as aspirin, other antiinflammatory drugs (such gastrointestinal tract bleeding. As little as 5 mL of blood
as ibuprofen, naproxen, corticosteroids, phenylbutazone), loss per day can be detected.
cancer chemotherapeutic agents, and alcohol in excess can
all cause a positive reaction due to loss of blood from There have been many studies on the use of fecal occult
gastric irritation. Recent studies, however, suggest that blood testing for early detection of colorectal cancer.
low-dose aspirin has no effect on the test [48]. In a Colorectal cancer is the second leading cause of cancer
comparison of subjects taking aspirin or other NSAIDs death in United States. Studies have shown that screening
with those not on these drugs, no difference in the rate of for colorectal cancer by fecal occult blood testing can
positive fecal occult blood test results was found, and it reduce cancer incidence and mortality [59]. Although
was therefore suggested that avoidance of these drugs immunological methods have been suggested to have
before fecal occult blood testing may not be necessary better sensitivity, review of the available evidence does not
[49]. Toilet sanitizers may cause false-positive results. favor one type of test over the other [60]. Combination of
Chlorine-generating sanitizers will give false-positive the two tests did not do any better than the immunological
results with guaiac methods [50]. Nonchlorine-generating method [61]. A working party from the World Health
sanitizers reduce the immunological detection of Organization and the World Organization for Digestive
hemoglobin [51]. Endoscopy suggested that the immunochemical method
may be more amenable to standardization and quality
The heme-porphyrin method cannot distinguish porphyrins control and more acceptable to patients, because dietary
of hemoglobin origin from other heme-containing proteins restriction is not required [62].
of non-human origin. Ingestion of red meat was shown to
cause a 375% increase in fecal porphyrin measured by Reference Interval
HemoQuant [19]. However, plant-derived peroxidases do The amount of blood lost from the normal gastrointestinal
not interfere with this assay. tract is approximately 0.5 to 1.5 mL per day [1,2]. When
using a quantitative test, a positive result may be seen in
2% to 16% of the population, depending on the method
used and the population studied [63,64].
563
Fecal Occult Blood
Performance Goals hemoglobin monoclonal antibody. J Immunol

Evidence from the UK External Quality Assurance Methods 1992;153:185-92.
Programme suggests that some of the guaiac-based 14 Saito H. Screening for colorectal cancer by
methods show poor analytical performance [65]. Guaiac- immunochemical fecal occult blood testing. J
based tests are qualitative, and errors can arise in Cancer Res 1996;87:1011-24.
interpreting the color changes [66]. Because of the poor 15 Schwartz S, Dahl J, Ellefson M, Ahlquist D. The
performance of these test kits and the possible effects of "HemoQuant" test: a specific and quantitative
dietary factors, it has been recently suggested that guaiac- determination of heme (hemoglobin) in feces and
based fecal occult blood tests should not be used in the other materials. Clin Chem 1983;29:2061-7.
clinical setting and only used in screening for colorectal 16 Young GP, St John DJ, Rose IS, Blake D. Haem
cancer in asymptomatic individuals [67]. Furthermore, in the gut. Part II. Faecal excretion of haem and
several national guidelines state that for the investigation haem-derived porphyrins and their detection. J
of symptomatic patients, fecal occult blood testing has no Gastroenterol Hepatol 1990;5:194-203.
value [68]. 17 Lynch NM, McHutchison JG, Young GP, Deacon
M, St John DJ, Barraclough D. Gastrointestinal
References blood loss from a new buffered aspirin
1 Ahlquist DA, McGill DB, Schwartz S, Taylor (Ostoprin): measurement by radiochromium and
WF, Owen RA. Fecal blood levels in health and Hemoquant techniques. Aust N Z J Med. 1989;19
disease. A study using HemoQuant. N Engl J :89-96.
Med. 1985;312:1422-8. 18 Young GP, St John DJ, Lynch NM, McHutchison
2 Dybdahl JH, Daae LN, Larsen S. Occult faecal JG, Barraclough D. Measurement of occult upper
blood loss determined by chemical tests and a 51 gastrointestinal tract blood loss: a direct
Cr method. Scand J Gastroenterol. 1981;16:245- comparison of radiochromium and haem-
52. porphyrin assay techniques. J Gastroenterol
3 St John DJ, Young GP. Evaluation of Hepatol 1993;8:328-33.
radiochromium blood loss studies in unexplained 19 Rose IS, Young GP, St John DJ, Deacon MC,
iron-deficiency anaemia. Aust N Z J Med Blake D, Henderson RW. Effect of ingestion of
1978;8:121-6. hemoproteins on fecal excretion of hemes and
4 Welch CL, Young DS. Spectrophotometry of porphyrins. Clin Chem 1989;35:2290-6.
occult blood in feces. Clin Chem 1983;29:2022-5. 20 Xing PX, Young GP, Ho D, Sinatra MA, Hoj PB,
5 Syed AA, Silwadi MF, Khatoon BA. Detection McKenzie IF. A new approach to fecal occult
and diagnosis of blood in feces and urine: an blood testing based on the detection of
overview. Clin Chim Acta 2002;318:1-17. haptoglobin. Cancer 1996;78:48-56.
6 Gregerson JP. Studies on occult blood tests. Arch 21 Hoff G, Grotmol T, Thiis-Evensen E, Bretthauer
Vaerauungs Krankh 1919;25:169-75. M, Gondal G, Vatn MH. Testing for faecal
7 Searle CE. Chemical carcinogens and their calprotectin (PhiCal) in the Norwegian Colorectal
significance for chemists. Chem Br 1970;6:5-10. Cancer Prevention trial on flexible sigmoidoscopy
8 Syed AA Khatoon BA, Silwadi MF. New screening: comparison with an immunochemical
reagents for detection of fecal occult blood. J test for occult blood (FlexSure OBT). Gut
Pharm Biomed Anal 2001;24:581-6. 2004;53:1329-33.
9 Kratochvil JF, Burns RH, Soikel MK, Haskin JM. 22 Wu CI, Tsai CC, Lu CC, Wu PC, Wu DC, Lin
Isolation and characterization of alpha guaiaconic SY, Shiea J. Diagnosis of occult blood in human
acid and the nature of guaiacum blue. Phytochem feces using matrix-assisted laser desorption
1971;10:2529-31. ionization/time-of-flight mass spectrometry. Clin
10 Cervantes C, Vega H, Prez L, Feli R. Enzyme Chim Acta 2007;384:86-92.
immunoassay for occult faecal blood. Z Med Lab 23 Uchida K, Matsuse R, Miyachi N, Okuda S,
Diagn 1989;30:431-6. Tomita S, Miyoshi H et al. Immunochemical
11 Turunen MJ, Liewendhal K, Paratanen P, detection of human blood in feces. Clin Chim
Aslercreutez H. Immunological detection of fecal Acta 1990;189:267-74.
occult blood in colorectal cancer. Br J Cancer 24http://www.surrey.ac.uk/GMEC/Pages/GuildfordEvaluat
1984;49:141-146. ions/PreviousReports/ReportPDFs/FOBT%20kit
12 Vaananen P, Tenhunen R. Rapid %20survey.pdf Accessed 10-3-2008.
immunochemical detection of fecal occult blood 25 St John DJ, Young GP, Alexeyeff MA, Deacon
by use of a latex-agglutination test. Clin Chem MC, Cuthbertson AM, Macrae FA, Penfold JC.
1988;34:1763-6. Evaluation of new occult blood tests for detection
13 Nagata M, Tanaka T. Detection of fecal blood by of colorectal neoplasia. Gastroenterology
colloidal gold agglutination using an anti-human 1993;104:1661-8.
564
Fecal Occult Blood
26 Stroehlein JR, Fairbanks VF, McGill DB, Go VL. 41 Pignone M et al. Meta-analysis of dietary
Hemoccult detection of fecal occult blood restriction during fecal occult blood testing.
quantitated by radioassay. Am J Dig Dis Effect Clin Pract 2001;4:150-6.
1976;21:841-4. 42 Jaffe RM, Kasten B, Young DS, MacLowry JD.
27 Rockey DC. Occult Gastrointestinal Bleeding. N False-negative stool occult blood tests caused by
Eng J Med 1999;341:3 8-46. ingestion of ascorbic acid (vitamin C). Ann Intern
28 Greenwald B. From guaiac to immune fecal Med 1975;83:824-6.
occult blood tests: the emergence of technology in 43 Garrick DP, Close JR, McMurray W. Detection
colorectal cancer screening. Gastroenterol Nurs of occult blood in faeces. Lancet 1977;2:820-1.
2005;28:90-6. 44 Lifton LJ, Kreiser J. False-positive stool occult
29 Khoury RH, Gandhi AH, Salmon BP, Gudaitis P blood tests caused by iron preparations. A
Patel SN, Gudaitis D. Performance of the OC- controlled study and review of literature.
Auto micro80 for the determination of fecal Gastroenterology 1982;83:860-3.
Occult blood. Clin Chem 2007;53 Suppl A107. 45 Anderson GD, Yuellig TR, Krone RE Jr. An
30 Burch JA, Soares-Weiser K, St John DJ, Duffy S, investigation into the effects of oral iron
Smith S, Kleijnen J, Westwood M. Diagnostic supplementation on in vivo Hemoccult stool
accuracy of faecal occult blood tests used in testing. Am J Gastroenterol 1990;85:558-61.
screening for colorectal cancer: a systematic 46 Orchard JL, Lawson R. False-positive Hemoccult
review. J Med Screen 2007;14:132-7. reaction caused by Betadine. N Engl J Med
31 Thomas WM, Pye G, Hardcastle JD, Mangham 1984;311:199.
CM. Faecal occult blood screening for colorectal 47 Blebea J, McPherson RA. False-positive guaiac
neoplasia: a randomized trial of three days or six testing with iodine. Arch Pathol Lab Med
days of tests. Br J Surg 1990;77:277-9. 1985;109:437-440.
32 Young SP, St John DJB. Faecal occult blood 48 Greenberg PD, Cello JP, Rockey DC.
tests: choice, usage and clinical applications. Clin Relationship of low-dose aspirin to GI injury and
Biochem Revs 1992;13:161-167. occult bleeding: a pilot study. Gastrointest Endosc
33 Young GP, Sinatra MA, St. John DJ. Influence of 1999;50:618-622.
delay in stool sampling on fecal occult blood test 49 Kahi CJ, Imperiale TF. Do aspirin and
sensitivity Clin Chem 1996;42:1107 - 1108. nonsteroidal anti-inflammatory drugs cause false-
34 Caligiore P, Macrae FA, St John DJ, Rayner LJ, positive fecal occult blood test results? A
Legge JW. Peroxidase levels in food: relevance to prospective study in a cohort of veterans. Am J
colorectal cancer screening. Am J Clin Nutr Med 2004;117:837-841.
1982;35:1487-9. 50 Ahlquist DA, Schwartz S, Isaacson J, Ellefson M.
35 Macrae FA, St John DJ, Caligiore P, Taylor LS, A stool collection device: the first step in occult
Legge JW. Optimal dietary conditions for blood testing. Ann Intern Med 1988;108:609-12.
hemoccult testing. Gastroenterology 51 Imafuku Y, Nagai T, Yoshida H. The effect of
1982;82:899-903. toilet sanitizers and detergents on immunological
36 Meyer GW, Komadina K, Perucca P. Vegetable occult blood tests. Clin Chim Acta 1996;253:51-
peroxidase is denatured by gastric acid: fresh 9.
vegetables do not cause false-positive stool 52 Pierson RN Jr, Holt PR, Watson RM, Keating RP.
Hemoccults in normal subjects. Gastroenterology Aspirin and gastrointestinal bleeding. Chromate
1991;101:871. blood loss studies. Am J Med 1961;31:259-65.
37 Sinatra MA, St John DJ, Young GP. Interference 53 Herzog P, Holtermller KH, Preiss J, Fischer J,
of plant peroxidases with guaiac-based fecal Ewe K, Schreiber HJ, Berres M. Fecal blood loss
occult blood tests is avoidable. Clin Chem in patients with colonic polyps: a comparison of
51
1999;45:123-6. measurements with chromium-labeled
38 Rose S, Young GP, St John DJ, Deacon MC, erythrocytes and with the Hemoccult test.
Blake D, Henderson RW. Effect of ingestion of Gastroenterology 1982;83:957-62.
hemoproteins on fecal excretion of hemes and 54 Dybdahl JH, Daae LN, Larsen S, Myren J. Occult
porphyrins. Clin Chem 1989;35:2290-6. faecal blood loss determined by a 51Cr method
39 Feinberg EJ, Steinberg WM, Banks BL, Henry and chemical tests in patients referred for
JP. How long to abstain from eating red meat colonoscopy. Scand J Gastroenterol 1984;19:245-
before fecal occult blood tests. Ann Intern Med 54.
1990;113:403-4. 55 St John DJ, Young GP, McHutchison JG, Deacon
40 Norfleet RG. Effect of diet on fecal occult blood MC, Alexeyeff MA. Comparison of the
testing in patients with colorectal polyps. Dig Dis specificity and sensitivity of Hemoccult and
Sci 1986;31:498-501. HemoQuant in screening for colorectal neoplasia.
Ann Intern Med 1992;117:376-82.
565
Fecal Occult Blood
56 Farrands PA, Hardcastle JD. Accuracy of occult cancer screening: recommendations based on
blood tests over a six-day period. Clin Oncol performance characteristics in population studies:
1983;9:217-25. a WHO (World Health Organization) and OMED
57 Rosenfield RE, Kochwa S, Kaczera Z, Maimon J. (World Organization for Digestive Endoscopy)
Nonuniform distribution of occult blood in feces. report. Am J Gastroenterol 2002;97:2499-2507.
Am J Clin Pathol 1979;71:204-9. 63 Levin B, Hess K, Johnson C. Screening for
58 Harewood GC, McConnell JP, Harrington JJ, colorectal cancer. A comparison of 3 fecal occult
Mahoney DW, Ahlquist DA. Detection of occult blood tests. Arch Intern Med 1997;157:970-6.
upper gastrointestinal tract bleeding: performance 64 Mandel JS, Bond JH, Church TR, Snover DC,
differences in fecal occult blood tests. Mayo Clin Bradley GM, Schuman LM, Ederer F. Reducing
Proc 2002;77:23-8. mortality from colorectal cancer by screening for
59 Mandel JS, Church TR, Bond JH, Ederer F, fecal occult blood. Minnesota Colon Cancer
Geisser MS, Mongin SJ et al. The effect of fecal Control Study. N Engl J Med 1993;328:1365-71.
occult-blood screening on the incidence of 65 Duncan A, Hill PG. A review of the quality of
colorectal cancer. N Engl J Med 2000;343:1603- gastrointestinal investigations performed in UK
7. laboratories. Ann Clin Biochem 2007;44:145-58.
60 Burch JA, Soares-Weiser K, St John DJ, Duffy S, 66 Selinger RR, Norman S, Dominitz JA. Failure of
Smith S, Kleijnen J, Westwood M. Diagnostic health care professionals to interpret fecal occult
accuracy of faecal occult blood tests used in blood tests accurately. Am J Med 2003;114:64-7.
screening for colorectal cancer: a systematic 67 Fraser CG. Faecal occult blood testseliminate,
review. J Med Screen 2007;14:132-7. enhance or update? Ann Clin Biochem
61 Allison JE, Sakoda LC, Levin TR, Tucker JP, 2008;45:117-21.
Tekawa IS, Cuff T et al. Screening for colorectal 68 NICE. Clinical Guidance 27. Referral for
neoplasms with new fecal occult blood tests: Suspected Cancer 2005. Available at
update on performance characteristics. J Natl www.nice.org.uk/nicemedia/pdf/cg027niceguideli
Cancer Inst 2007;99:1462-70. ne.pdf.
62 Young GP, St John DJ, Winawer SJ, Rozen P.
Choice of fecal occult blood tests for colorectal
Table 1: Comparison of Features of Fecal Occult Blood Tests

Guaiac-Based Immunochemical Heme-porphyrins
Biochemical basis Peroxidase activity Immunoreactive Hb Fluorescent porphyrins
Compounds detected Hb Globin (human) Porphyrins
Myoglobin Myoglobin
All heme All heme
Non-heme porphyrins porphyrins form haem
Method Chromogen indicator Heme agglutination, ELISA, Solvent extraction,

latex fluorimetry
Quantitation No Possibly Yes
Drugs affecting Vitamin C No No
Interference/interpretatio Ingestion of meat No Ingestion of meat
n Plant peroxidases
Sample required Fecal smear on filter paper Smear of a scoop of feces Small scoop of feces
Equipment None Depends on method Fluorimeter
Time for sample Few minutes 15 min-2 hrs, depending on 4-8 hrs
processing method
Modified from Young and St. John (Clin Biochem Rev 1992;13:161-167)
566
Fecal Occult Blood
Table 2: Hb and Hb Products in Feces According to the Site of Bleeding

Gastric Proximal large bowel Distal bowel
Intact hemoglobin - + ++
Intact heme + + ++
Heme-derived porphyrins +++ ++ +
567
Ferritin
Ferritin
Hassan M.E. Azzazy
Name: Ferritin
Clinical significance: Measure of body iron stores Refer to Chapter 39, Iron, Porphyrin
and Bilirubin Metabolism, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular mass (apoferritin): 445,000 D
Merck Index: 3957
Chemical class: Protein, ferroprotein
Principles of Analysis and Current Usage immunoassay (EIA) (Table 1, Method 4). The solid phase,
Until 1972, no method with sufficient sensitivity was such as beads or plastic tubes, is coated with antiferritin
available to measure ferritin in normal serum. It was antibodies and then incubated simultaneously with serum
therefore assumed that ferritin, known to be a storage form and enzyme conjugated to antiferritin antibodies. Enzymes
of iron found in many tissues, did not circulate. The advent employed as markers include horseradish peroxidase
of assays using radiolabeled antibodies demonstrated not (HRP) and alkaline phosphatase (AP). The ferritin in the
only the presence of ferritin in normal serum but also its serum binds to the antibody on the bead, and the
usefulness in the diagnosis of disorders resulting in either conjugated antibody then binds to the immobilized ferritin.
low or high amounts of storage iron. After washing to remove unbound materials, a substrate
for horseradish peroxidase (such as a complex
The first assay for measurement of ferritin in serum [1] azosulfonate dye [4] or o-phenylenediamine) or alkaline
was an immunoradiometric assay (IRMA). This phosphatase (p-nitrophenyl phosphate, 4-
methodology is similar to radioimmunoassay (RIA) but methylumbelliferyl phosphate, or adamantyl dioxetane
uses a labeled antibody rather than a labeled antigen. In the phosphate) is added and allowed to react with the bound
IRMA method (Table 1, Method 1), serum is incubated enzyme. The amount of reaction product present is directly
with an excess of 125I-labeled antibody. Serum ferritin and proportional to the amount of ferritin in the serum
antibody form a soluble complex that is separated from specimen.
unreacted antibody by means of an immunoabsorbent
consisting of ferritin coupled to cellulose particles. The Another immunoassay procedure employs a radial
ferritin-cellulose absorbs unreacted labeled antibody and partition sandwich enzyme immunoassay [5]. The assay is
removes it from solution after a centrifugation step. The similar to the EIA previously described. Antibodies to
radioactivity remaining in solution represents antibody ferritin are immobilized, in this case, on glass-fiber filter
bound to serum ferritin and is directly related to serum paper. The radial-partition technique employs diffusion to
ferritin concentration. A modification of the IRMA method wash away any material (in serum or in the reagents) that
called the two-site IRMA was introduced for the does not bind to the filter paper during the formation of the
measurement of serum ferritin in 1974 by Miles et al. sandwich. The amount of enzyme (alkaline
(Table 1, Method 2) [2]. A third type of assay for serum phosphatase)-antiferritin antibody in the sandwich can be
ferritin used a competitive radioimmunoassay technique determined by adding 4-methylumbelliferyl phosphate and
(Table 1, Method 3) [3]. Neither the two-site IRMA nor measuring the rate of release of the product, 4-
the competitive radioimmunoassay procedures are in methylumbelliferone, fluorometrically.
common use today.
A latex particle-agglutination immunoassay employing
Another method more frequently employed to quantitate continuous flow automation and a particle counter has
serum ferritin is a sandwich, solid-phase enzyme been described [6]. A similar automated latex-particle
agglutination immunoassay for serum ferritin has been
i described for a centrifugal analyzer [7]. This turbidimetric
Ferritin immunoassay, also known as a microparticle-enhanced
Previous and current authors of this method: turbidimetric immunoassay, has enhanced sensitivity by
First edition: Robert S. Franco attaching the antibody to microscopic latex particles.
Methods edition: Robert S. Franco
Second edition: Not updated Measurement of ferritin by a two-site, solid-phase
Third edition: Not updated (sandwich) chemiluminometric immunoassay (Table 1,
Fourth edition: Harini Patel Method 5) has also been described [8]. This assay uses a
Fifth edition: Hassan Azzazy constant amount of two antiferritin antibodies. The labeled
568
Ferritin
antibody is coupled to a chemiluminescent molecule, Specimen

acridinium ester. The capture antiferritin antibody is Serum is the recommended sample type for this assay. No
covalently bound to paramagnetic particles, which serve as patient preparation is required. Blood should be drawn in
the solid phase. The complex is separated from unbound plain red-top tubes or serum gel tubes. Specimens are
materials by placing the tubes in a magnetic rack where the stable for at least 1 week when kept refrigerated at 4C
bound fraction aggregates in the area of the magnets, and [17] and for 6 months at 20C. Frozen specimens should
the free fraction is removed. Oxidization of the acridinium not be thawed at 37C, and repeated freezing and thawing
ester occurs rapidly, coupled with the emission of light, should be avoided, as well as violent mixing, which may
which is measured by a luminometer. The ferritin denature ferritin.
concentration is directly proportional to the relative light
unit (RLU).
Interferences
A cloned-enzyme donor immunoassay (CEDIA) has been A disadvantage of the two-site IRMA for ferritin is the
introduced [9] for ferritin measurement (Table 1, Method high-dose hook effect [18]. This is a paradoxical
6) which can be run on many open, random-access decrease in radioactive counts at high concentration of
analyzers. This is a homogeneous assay for ferritin which antigen, resulting in a maximum in the curve of
utilizes recombinant DNA technology to split the enzyme radioactivity versus concentration. Since the hook effect
-galactosidase into two inactive fragments, enzyme takes place at very high serum ferritin concentration, this
acceptor (EA) and enzyme donor (ED). The ferritin problem does not affect samples with levels of ferritin in
antibody is covalently attached to the ED fragment in such the normal range. However, some patients can have
a way that it allows the recombination of the two extremely elevated ferritin levels; therefore, it is good
fragments to form the active enzyme. The reassociation of practice to run two dilutions of each sample and compare
the EA and ED molecules is inhibited by the binding of the the results. If the calculated ferritin concentration of the
ferritin antibodyED conjugate to ferritin in the sample. more dilute sample is significantly higher than that of the
Hence less ED-conjugate is available to combine with EA less dilute sample, it could be because of the hook effect.
to form active -galactosidase. The enzyme activity is EIAs, competitive RIA, and some commercially available
inversely proportional to the concentration of ferritin in the IRMAs can perform satisfactorily at high ferritin
sample. concentrations [18]; however, each method should be
verified for the absence of the hook effect. Most
Reference and Preferred Method manufacturers give hook-effect limits in their package
There is no reference method for serum ferritin. According inserts.
to a 2007 College of American Pathologists (CAP)
proficiency testing program, 100% of participating Few interferences have been described for the ferritin
laboratories reported using non-isotopic ferritin assays. assay. Immunonephelometric methods for ferritin have
Fully automated ferritin immunoassays are now available been shown to be adversely affected by the presence of
and may be preferred over manual ones because they have increased turbidity or lipemia [19]. The presence of
several advantages, including better precision, less heterophilic antibodies can interfere with immunoassay
operator dependence, and faster throughput analysis. procedures. Patient samples that give results not consistent
with the clinical status may require further investigation to
Proper standardization of the ferritin assay is a significant rule out interference due to heterophilic antibodies. In-vivo
problem. Ferritin is present as many isoforms [10]. The interferences such as ethanol, iron salts, and oral
most acidic form (pI 4.6) contains 24 H-subunits, whereas contraceptives may cause elevated values. Ferritin level is
the most basic isoferritin molecule (pI 5.8) consists of 24 decreased by erythropoietin [20].
L-subunits [11]. Diverse ferritin forms (isoferritins) react
differently in different immunoassays [12]. Variability in Ferritin Reference Interval
commercial assays has been documented over the years, In general, the reference interval is approximately 20 to
and no reference method is generally accepted [13]. Much 200 ng/mL (about 0.05 to 0.5 nmol/L) for adult males.
of the lack of consensus about ferritin levels can probably However, normal values are both age- and sex-related,
be attributed to the absence of an accepted, well-defined with premenopausal females and children under 16
standard [14]. The introduction of a third National Institute exhibiting lower values as a result of physiologically lower
for Biological Standards and Control/World Health iron stores. Postmenopausal women have higher values
Organization (NIBSC/WHO) Recombinant Ferritin when compared with females who have not reached
International Standard 94/572 appears to have reduced but menopause [21]. Release of ferritin from cultured alveolar
not eliminated the variability in commercial assays macrophages, recovered from cigarette-smoker lungs by
[15,16]. bronchoalveolar lavage, was increased compared with
nonsmokers [22]. Ferritin concentrations in capillary
(finger-prick) blood samples have been reported to be
significantly lower than ferritin concentrations in venous
569
Ferritin
blood [23]. Serum ferritin concentration may be increased Serum ferritin is especially useful in distinguishing iron
(as much as two- to threefold) after several days of fasting deficiency from the anemia of chronic disorders, because
[24]. in the latter, ferritin levels are increased [24,34]. Serum
ferritin is also increased in other anemias [35], including
In a study of infants aged 9 months [25], the reference aplastic anemia, sideroblastic anemia, and chronic
interval for ferritin in capillary samples was reported to be hemolytic anemias. In idiopathic hemochromatosis and in
between 7.1 and 224 ng/mL. The ferritin levels were multiply transfused patients, the serum ferritin may be
positively related to birth weight, and females had higher extremely high. Ferritin has also been suggested as a
concentrations than males. In other studies [26,27], nonspecific marker for many neoplastic malignancies.
pediatric reference intervals from newborns to 19-year-
olds have been reported by using different methods. There Serum ferritin concentrations in adult Stills disease
are decreased serum ferritin levels in prepubertal and early (ASD), an acute, systemic inflammatory disorder, are
pubertal boys due to extra iron being mobilized through higher in active ASD than in inactive ASD and other
redistribution to red blood cells [28]. Population norms for systemic diseases [36]. In addition, elevation in serum
serum ferritin are discussed in recent literature [29]. ferritin (SF) and red-cell ferritin (RCF) have been reported
Normal ranges need to be determined per the used assay. in HIV-infected patients [37]. SF levels are higher in
The conventional serum ferritin reference intervals are patients with more clinical symptoms, and RCF is
presented in the table below. significantly higher in asymptomatic AIDS patients and
those treated with zidovudine (AZT).
Table 2. Serum Ferritin Reference Intervals [30,31]
Age Ferritin Value (ng/mL) It has been suggested [38] that in hematological
Newborn 25-200 malignancies, tissue-specific isoferritins may cross-react,
1 month 200-600 resulting in poor correlations when different immunoassay
2-5 months 50-200 methods are used. Hence monitoring of ferritin should be
6 months-15 years 7-140 performed using the same method.
Adult male 20-250
Adult female 10-120 A study of pregnant women [39] reported that plasma
Iron overload (adult male) >400 ferritin concentration is significantly greater in smokers
Iron overload (adult female) >200 than in nonsmokers, suggesting that plasma/serum ferritin
values should be carefully interpreted when assessing iron
status in pregnant women who smoke. Falsely high values
Interpretation
in these women may not detect patients with iron
Ferritin consists of a protein shell (apoferritin) containing a
deficiency, who require prenatal iron supplementation. It
variable amount of iron in the hollow core. The iron
may be desirable to establish a revised cutoff point for
content, in the form of colloidal hydrous ferric oxide
normal ferritin values for smokers at different stages of
phosphate micelles, varies from almost none to 30%.
gestation.
Serum ferritin is a sensitive indicator of body iron stores; it
has been shown to correlate with stainable bone-marrow
Ferritin Performance Goals
iron [23,32]. Its concentration reflects iron stores in normal
The current CLIA performance goal for measurement of
individuals and those with iron deficiency where serum
ferritin is for laboratories to be within 3 standard
ferritin concentration of 1 ng/mL is equivalent to 8 mg of
deviations of the peer-group mean. According to the 2007
stored iron [20]. It should be noted, however, that in the
CAP Survey, % coefficient of variation (CV) for all
presence of significant tissue destruction or rapid cellular
turnover, the serum ferritin concentration is ferritin methods at a mean value of 192.4 g/L (SD 22.2)
disproportionately high when compared to bone-marrow was 11.6%.
iron. Therefore, the serum ferritin level cannot be used to The normal day-to-day variation in ferritin has been found
help in the diagnosis of iron deficiency in the presence of to be approximately 14 ng/mL in men and 26 ng/mL in
inflammation, liver disease, or malignancies such as acute menstruating women [40]. Long-term intraindividual
leukemia or Hodgkins disease. Ferritin concentration is variation in healthy adults measured over a 5-month period
significantly increased in cases of iron-overload conditions has been found to be less than 20% [41].
such as hemochromatosis [33].
Steele et al. [42] determined the long-term within-
A serum ferritin of less than 10 ng/mL almost always laboratory and between-laboratory variation of ferritin
indicates iron deficiency. Although there is some overlap assays in a study that was based on a CAP fresh frozen
between the normal and iron-deficient population, this is serum study. Five of seven of ferritin methods met within-
not usually a diagnostic problem, because serum ferritin is laboratory imprecision goals based on biological criteria.
usually increased in anemia resulting from other causes. The largest contribution to the total imprecision was the
total within-laboratory component and not the between-
laboratory component of variation. The desirable analytical
570
Ferritin
quality specifications for imprecision, bias, and total error improve the accuracy and precision of seven kits
as derived from biological variation are 7.1%, 5.2%, and for determination of ferritin in serum. Clin Chem
16.9%, respectively [43]. 1984;30:298-301.
15 Thorpe SJ, Walker D, Arosio P, Heath A, Cook
References JD, Worwood M. International collaborative
1 Addison GM, Beamish MR, Hales CN, Hodgkins study to evaluate a recombinant L ferritin
M, Jacobs A, Llewellin P. An immunoradiometric preparation as an International Standard. Clin
assay for ferritin in the serum of normal subjects Chem 1997;43:1582-1587.
and patients with iron deficiency and iron 16 Lotz J, Hafner G, Prellwitz W. Reference values
overload. J Clin Pathol 1972;25:326-329. for a homogeneous ferritin assay and traceability
2 Miles LE, Lipschitz DA, Bieber CP, Cook JD. to the 3rd International Recombinant Standard for
Measurement of serum ferritin by a 2-site Ferritin (NIBSC code 94/572). Clin Chem Lab
immunoradiometric assay. Anal Biochem Med 1999;37:821-825.
1974;61:209-224. 17 Dale JC, Pruett SK. Phlebotomy: a minimalist
3 Marcus DM, Zinberg N. Measurement of serum approach. Mayo Clin Proc 1993;68:249-255.
ferritin by radioimmunoassay: results in normal 18 Ng RH, Brown BA, Valdes R. Three commercial
individuals and patients with breast cancer. J Natl methods for serum ferritin compared and the
Cancer Inst 1975;55:791-795. high-dose hook effect eliminated. Clin Chem
4 Revenant MC. Sandwich enzyme immunoassay 1983;29:1109-1113.
for serum ferritin with polypropylene test tubes as 19 Borque L, Rus A, del Cura J, Maside C, Escanero
the solid phase. Clin Chem 1983;29:681-683. J. Automated quantitative nephelometric latex
5 Timmons R, Soto AA. Radial partition sandwich immunoassay for determining ferritin in human
enzyme immunoassay for ferritin [abstract]. Clin serum. J Clin Lab Anal. 1992;6:239-44.
Chem 1984;30:984. 20 Wu AHB, ed. Tietz Clinical Guide to Laboratory
6 Bernard AM, Lauwerys RR. Continuous-flow Tests. 4th ed. Philadelphia: Saunders; 2006.
system for automation of latex immunoassay by 21 Rubin C, Wood PJ, Archer T, Rowe DJ. Changes
particle counting. Clin Chem 1983;29:1007-1011. in serum ferritin and other acute phase proteins
7 Simo JM, Joven J, Cliville X, Sans T. Automated following major surgery. Ann Clin Biochem
latex agglutination immunoassay of serum ferritin 1984;21:290-294.
with a centrifugal analyzer. Clin Chem 22 Wesselius LJ, Nelson ME, Skikne BS. Increased
1994;40:625-629. release of ferritin and iron by iron-loaded alveolar
8 Stacy DL, Han P. Serum ferritin measurement and macrophages in cigarette smokers. Am J Respir
the degree of agreement using four techniques. Crit Care Med. 1994;150:690-5.
Am J Clin Pathol 1992;98:511-515. 23 Harju E, Pakarmen A, Larmi TA. Comparison
9 Engel WD, Khanna PL. CEDIA in-vitro between serum ferritin concentration and the
diagnostics with a novel homogeneous amount of bone marrow stainable iron. Scand J
immunoassay technique. Current status and future Clin Lab Invest 1984;44:555-556.
prospects. J Immunol Methods 1992;150:99-102. 24 Lipschitz DA, Cook JD, Finch CA. A clinical
10 Drysdale JW, Kohgo Y, Watanabe N. Ferritin evaluation of serum ferritin as an index of iron
phenotypes. In: Albertini A, ed. stores. N Engl J Med 1974;290:1213-1216.
Radioimmunoassays of Hormones, Proteins, and 25 Edmond AM, Hawkins M, Pennock C, Golding J.
Enzymes. Amsterdam: Excerpta Medica; Hemoglobin and ferritin concentrations in infants
1980:213. at 8 months of age. Arch Dis Child 1996;74:36-
11 Forman DT, Parke SL. The measurement and 39.
interpretation of serum ferritin. Ann Clin Lab Sci 26 Murthy JN, Hicks JM, Soldin SJ. Evaluation of
1980;10:345-349. the Technicon Immuno 1 random-access
12 van Suijlen JD, van Noord PC, Leijnse B. immunoassay analyzer and calculation of
Accuracy of serum ferritin determinations in pediatric reference ranges for endocrine tests, T-
tissue preparations and human serum. J Clin uptake and ferritin. Clin Biochem 1995;28:181-
Chem Clin Biochem 1990;28:43-48. 185.
13 Bock JL, Endres DB, Elin RJ, Wang E, 27 Soldin SJ, Morales A, Albalos F, Lenherr S, Rifai
Rosenzweig B, Klee GG. Comparison of fresh N. Pediatric reference ranges on the IMx for FSH,
frozen serum to traditional proficiency testing LH, prolactin, TSH, T4, T3, free T4, Free T3, T-
material in a College of American Pathologists uptake, Ig E and ferritin. Clin Biochem
survey for ferritin, folate, and vitamin B12. Arch 1995;28:603-606.
Pathol Lab Med 2005;129:323-327. 28 Anttila R, Siimes MA. Serum transferrin and
14 Iacobello C, Ghielmi S, Belloli S, Arosio P, ferritin in pubertal boys: relations to body growth,
Albertini A. Use of a reference standard to
571
Ferritin
pubertal stage, erythropoiesis, and iron ferritin levels in HIV-infected patients.

deficiency. Am J Clin Nutr 1996;63:179-183. Haematologica 1994;79:165-167.
29 Custer EM, Finch CA, Sobel RE, Zettner A. 38 Vernet M, Renversez JC, Lasne Y, Revenant MC,
Population norms for serum ferritin. J Lab Clin Charlier de Bressing C, Guillemin C et al.
Med 1995;126:88-94. Comparison of six serum ferritin immunoassays
30 Seamonds B, Anderson K, Whitaker B. Reference and isoferritin spectrotypes in malignancies. Ann
intervals for ferritin: age dependence. Clin Chem Biol Clin (Paris) 1995;53:419-427.
1980;26:1515-1516. 39 Tamura T, Goldenberg RL, Johnson KE, DuBard
31 Williams WJ, Beutler E, Erslev AJ et al, eds. MB. Effects of smoking on plasma ferritin
Hematology 3rd ed. New York: McGraw-Hill; concentrations in pregnant women. Clin Chem
1983. 1995;41:1190-1191.
32 Forman DT, Vye MV. Immunoradiometric serum 40 Borel MJ, Smith SM, Derr J, Beard JL. Day-to-
ferritin concentration compared with stainable day variation in iron-status indices in healthy men
bone marrow iron as indices to iron stores. Clin and women. Am J Clin Nutr 1991;54:729-735.
Chem 1980;26:145-147. 41 Gallagher SK, Johnson LK, Milne DB. Short-term
33 Halliday JW, Russo AM, Cowlishaw JL, Powell and long-term variation of indices related to
LW. Serum-ferritin in diagnosis of nutritional status. I. Ca, Cu, Fe, Mg, and Zn. Clin
haemochromatosis. A study of 43 families. Lancet Chem 1989;35:369-373.
1977;2:621-624. 42 Steele BW, Wang E, Palmer-Toy DE, Killeen
34 Bentley DP, Williams P. Serum ferritin AA, Elin RJ, Klee GG. Total long-term within-
concentration as an index of storage iron in laboratory precision of cortisol, ferritin,
rheumatoid arthritis. J Clin Pathol 1974;27:786- thyroxine, free thyroxine, and thyroid-stimulating
788. hormone assays based on a College of American
35 Alfrey CP. Serum ferritin assay. CRC Crit Rev Pathologists fresh frozen serum study: do
Clin Lab Sci 1978(9)3:179-208. available methods meet medical needs for
36 Van Reeth C, Le Moel G, Lasne Y, Revenant precision? Arch Pathol Lab Med 2005;129:318-
MC, Agneray J, Kahn MF, Bourgeois P. Serum 322.
ferritins and isoferrtins are tools for diagnosis of 43 Westgard QC. Desirable specifications for total
active adult Stills disease. J Rheumatol error, imprecision, and bias, derived from biologic
1994;21:890-5. variation. Available at
37 Riera A, Gimferrer E, Cadafalch J, Remacha A, <http://www.westgard.com/biodatabase1.htm>
Martin S. Prevalence of high serum and red cell
572
Ferritin
Table 1: Summary of Ferritin Methods
Method 1: Immunoradiometric assay (IRMA) (one-site)

Principle of analysis: Antibody is labeled, serum ferritinantibody complex is formed, and unreacted antibody is removed;
remaining radioactivity is directly related to serum ferritin concentration.
Comments: Not commercially available for clinical use; first method with enough sensitivity to measure normal levels,
historical usage
Method 2: IRMA (two-site)
Principle of analysis: Serum ferritin is bound to solid-phase antibody, and labeled antibody is then bound to immobilized
ferritin; after washing, the radioactivity of the bound labeled antibody is directly related to serum ferritin
concentration.
Comments: Commercially available for clinical use; hook effect at high concentration (see text)
Principle of analysis: Serum ferritin and labeled ferritin compete for antibody; antigen-antibody complexes are precipitated
and counted; radioactivity is inversely related to serum ferritin concentration.
Comments: Commercially available for clinical use; requires larger sample than the two-site IRMA does; not used for routine
measurement
Method 4: Enzyme immunoassay (EIA)
Principle of analysis: Antiferritin antibody coated on solid phase is incubated with serum and antiferritin antibody conjugated
to alkaline phosphatase or horseradish peroxidase. Serum ferritin binds to the immobilized antibody and is in turn
bound by the antibodyenzyme conjugate. Appropriate enzyme substrate is added, and products of the reaction could
be measured colorimetrically or fluorometrically.
Comments: Commercially available for clinical use; microprocessor-controlled reader available from manufacturer
Method 5: Chemiluminometric assay
Principle of analysis: In this two-site chemiluminometric immunoassay, the first ferritin antibody is labeled with acridinium
ester, and the second antibody is covalently bound to a paramagnetic particle (solid phase). Binding of the two
antibodies to ferritin forms a complex that is separated and measured with a luminometer. The light signal is directly
proportional to ferritin in the sample.
Comments: Very sensitive; very frequently used
Method 6: Cloned-enzyme donor immunoassay (CEDIA)
Principle of analysis: The enzyme -galactosidase is split into two inactive fragments, enzyme acceptor (EA) and enzyme
donor (ED), using recombinant DNA technology. Ferritin antibody is covalently attached to the ED fraction that is
free to combine with EA fraction, when not bound to ferritin, to form the active enzyme. When ED-antibody reacts
with ferritin, the association with EA fragment is inhibited, and the active enzyme is not formed. The enzyme activity
is inversely proportional to the concentration of ferritin.
Comments: Homogeneous assay; frequently used
573
Folic Acid and Folate

Sheila Dawling
Name: Folic acid, pteroylglutamic acid, PGA

Clinical significance: Refer to Chapter 41, Human Nutrition, in the 5th edition of Clinical Chemistry:
Molecular mass: 441.40 D convert molar to mass units ng/mL or g/L = nmol/L 0.4414
Merck Index: 4221
O H
O C
O
N C
H
OH
OH HN
N
N
Folic Acid
H2N N N
Name: Tetrahydrofolic acid (THF)

Merck Index:
O H
O C
O
N C
H
OH
OH HN
H
N
N
Tetrahydrofolic Acid
H2N N N
H
Name: N-5-Methyltetrahydrofolic acid (MTHF)

Clinical significance: Refer to Chapter 41, Human Nutrition, in the 5th edition of Clinical
Merck Index:
O H
O C
O
N C
H
OH
OH CH3 HN
N
N
N-5-Methyltetrahydrofolic Acid
H2N N N
H
Name: N-5-Formyltetrahydrofolic acid (Folinic acid, FTHF)

574
Molecular mass: 473.44 D convert molar to mass units ng/mL or mcg/L = nmol/L
0.4734
Merck Index: 4222
O H
O C
O
N C
H O H
OH
OH C HN
N
N
N-5-Formyltetrahydrofolic Acid
H2N N N
H
Principles of Analysis and Current Usage of the serum. The absorbance is compared to the growth
Folate species present in physiological fluids are a group permitted by standard solutions of
of chemically and biologically related B vitamer methyltetrahydrofolate (MTHF).
compounds [1]. They function as coenzymes in the
metabolism of one-carbon compounds, in purine and FBP assays utilize high-affinity folate-specific binding
pyrimidine synthesis, and in the degradation of histidine proteins (Table 1, Method 2). Early assays used radio-
to glutamic acid (Figures 1 and 2). The measurement of labeled folate (125I) and a limited amount of naturally
folate in serum and whole blood is important in the produced FBP to quantitate the concentration of
differential diagnosis of megaloblastic anemia. However, unlabeled folate in the patient sera in a standard
much of the current understanding of folate metabolism competitive binding assay format. Phase separation was
stems from research into the relationship of folate status typically with dextran-coated activated charcoal, but if
to the incidence of fetal neural tube defects [2]. There the FBP can be immobilized in some way, then the
are several different approaches to folate measurement, phases can be separated by a simple washing step.
including microbiological assays, competitive folate Incubation times could be reduced to as little as 1 hour,
binding protein (FBP) assays, and chromatographic and assays were also arranged for simultaneous
techniques. measurement of vitamin B12 and folate [5,6]. Today,
FBP assays enjoy widespread use because they are rapid,
Microbiological assays were developed first, but FBP precise, and easily automated and are available in a
methods are now standard in most laboratories. All these variety of formats suitable for modern immunoassay
methods aim to measure total folates. As mass analyzers. Their properties are compared in Table 2 and
spectrometry (MS) becomes more widely available, Figures 3 and 4.
high-performance liquid chromatography (HPLC)-
tandem MS will likely be adopted for reference methods, As a first step in all FBP assays, the natural binders
which will enable differentiation of the discrete folate present in the sample must be destroyed and the folate
species present in serum or RBCs. species released, which are subsequently all reduced to
pteroylmonoglutamic acid (PGA). Initially these steps
Microbiological assays (Table 1, Method 1) utilize the were performed off-line by boiling at alkaline pH, but
folate-dependent bacterium Lactobacillus casei [3,4]. they are now performed chemically and incorporated
Microtiter plates containing folate-depleted growth into the automated process. Popular reagents are NaOH
medium inoculated with L. casei are incubated for 18 to or KOH, then HCl, with either 2-mercaptoethanol,
40 hours at 37C with patient serum. The amount of L. monothioglycerol, dithiothreitol and KCN.
casei growth permitted is determined by measurement of
turbidity at 600 to 700 nm and reflects the folate content Once the de-binding and reduction are completed, the
folate-protein binding step must be carried out at pH 9.3
to ensure equivalent binding of PGA and MTHF, either
i
Folate as a single- or two-stage process. The one-stage assay is
Previous and current authors of this method: a conventional competitive binding assay in which
First edition: Michael D.D. McNeely sample, limited quantity of FBP, and tracer are
Methods edition: Michael D.D. McNeely combined simultaneously. Radio-labeled assays are
Second edition: Not updated seldom used nowadays, as titrated tracers have low
Third edition: Steven C. Kazmierczak specific activity and require liquid scintillation for beta-
Fourth edition: Michael D.D. McNeely particle emission and 125I tracers, while offering much
Fifth edition: Sheila Dawling greater counting activity and the added ease of gamma-
ray counting are not readily available. The two-stage
575
assay is noncompetitive and involves sequential react equally with the FBP. The purity of MTHF can be
incubations. Excess FBP is first incubated with sample, assessed spectrophotometrically at 290 and 245 nm [13].
and subsequently tracer-bound folate is added to occupy PGA has been advocated as a standard because it is
the remaining available binding sites. This approach is stable, well characterized spectrophotometrically,
more sensitive. As shown in Table 2 and Figures 3 and inexpensive, and easy to purify [13]. For these reasons,
4, the tracer may be an enzyme (alkaline phosphatase or most commercial assays use PGA as the calibration
horseradish peroxidase), a luminescent, or a material.
chemiluminescent label. Later developments in
heterogeneous assay formats utilize streptavidin and Standardization of immunoassays in the absence of an
biotin, and one assay uses a specific folate-binding accepted primary reference method and standardized
protein antibody and allows for the tracer to be attached reference materials has been discussed recently [19].
to not just the folate but also to the folate-binding protein One approach is that of harmonization or desired
or an FBP antibody. outcome standardization, in which assays are calibrated
The traditional binding agent was obtained from to read the same, ideally against a best available
unpurified milk powder or partially purified - candidate reference method. While this is not ideal,
lactoglobulin [7]. Modern techniques in protein given that absolute standardization is so far away, it may
purification have improved assay reproducibility. at least provide comparability between patient values
However, it is important to recognize that milk binders and allow collection of objective data for clinical studies
have different binding affinities for the different [20].
chemical forms of folate. Furthermore, binding is pH-
dependent, affecting the various folate species Development of a new standard reference material for
differentially [8]. The pH effect is so sensitive that the folate in serum (SRM 1955) [21] will enable reagent
amount of CO2 absorbed from the atmosphere by open manufacturers to address some of these problems. This
reagents may be sufficient to shift the binding curve standard will be available at three concentrations, with
significantly [9,10]. To complicate matters further, FBP assigned values for PGA, MTHF, 5-
exhibits marked aggregation tendencies, which also formyltetrahydrofolic acid (FTHF), and total folate
affect its folate binding characteristics [11,12]. Assays (together with vitamin B12 and total homocysteine).
using FBP as the carrier for the tracer molecule (e.g.,
Table 2, Method 2f and Figure 4) should determine the Chromatographic techniques, by contrast, aim to
effect of the addition of the tracer on the binding differentiate the folate species present in physiological
characteristics. fluids, targeting either solely the active MTHF form or
PGA and FTHF in addition to MTHF. Gas
Intracellular folate (e.g., RBC), by contrast, is chromatography-mass spectrometry achieved sufficient
predominantly polyglutamated species and is present in sensitivity for RBC folate but required complex sample
much higher concentrations than in serum but can be preparation and derivatization; high temperatures
measured by the same analytical techniques. The folate introduce the possibility of folate degradation [22].
in red blood cells more closely reflects tissue folate High-performance liquid chromatography (HPLC)
stores and is less variable as a consequence of recent methods with electrochemical oxidation detection gave
folate ingestion [13,14]. To liberate the intracellular improved specificity over microbiological and
folate, freezing and thawing are considered more immunoassays, but until recently they lacked sensitivity
complete than incubation with ascorbic acid [15], and had no internal standard and so could not be
although the latter is much more amenable to automation considered high-order reference methods [23-25].
and so is more frequently used. Incubation times are Recent developments in food science resulting from
typically 90 minutes (Table 2). The hematocrit should be grain-supplementation initiatives introduced the first
determined on the sample so the results can be expressed LC/MS methods for both supplemented (PGA) and
as a function of red blood cell mass. naturally occurring folates [26]. With the availability of
Standardization of FBP folate assays is problematic and suitable deuterated internal standards and price
is often cited as a major source of variation among competitiveness of triple quadrupole instruments
analytical methods [16]. In theory, standardization of LC/MS/MS (tandem LC/MS), applications for
folate assays where compounds are commercially physiological fluids are appearing in the literature [27-
available in pure and stable forms should be easier than 29]. With detection limits around 0.5 ng/mL from a 0.25
for a peptide analyte where no traceability to SI units is to 0.5 mL specimen, these are competitive with
available. MTHF is the most rational compound to use, immunoassays and are currently being considered as
since this is the predominant serum form, but it is prone reference methods. Sample preparation is, however,
to oxidation. The molecule has two asymmetric carbons, labor intensive in that folates must first be removed from
and the physiologically produced and active their serum binding proteins as described earlier.
diastereomer is L-MTHF [6S,S]. Until recently, only Subsequently, protonated and stabilized folates in acidic
racemic MTHF was commercially available, and the solutions are then adsorbed onto C18 solid phase
different forms do not react equally with FBP [17]. It has extraction columns. After washing, folate species are
long been recognized that the chemical nature of the eluted from the columns and injected onto the HPLC
calibration material can influence the results obtained analytical column [25,27,28]. An alternative isolation
[18]. Given carefully controlled conditions at pH 9.3, the approach is affinity chromatography. Here, folate
reaction can be manipulated so that PGA and MTHF species from the plasma are trapped in a column
576
containing FBP on a solid support, impurities are washed RBC folate content is not dependent on recent food
off, and then the purified folates are eluted onto the intake and is advocated as a better measure of body
analytical column using an automated column-switching folate load; consequently, samples can be collected at
technique [23]. Sample throughput (30 to 100 samples any time of day. Samples should be anticoagulated with
per day maximum) is still inferior to automated FBP powdered EDTA, heparin, or oxalate. Liquid
assays, and the methods are unlikely to translate into anticoagulants are not suitable, because they dilute out
mainstream analysis in the foreseeable future. Similar the RBCs. Some automated assays list only one or two
methods for RBCs are also reported [30]. permitted anticoagulants, and reference should be made
to the package insert for the specific assay. Unspun
Reference and Preferred Methods whole blood must be carefully maintained at 4C; it is
At present there is no accepted reference method for stable at this temperature for 3 days. It is recommended
folate. HPLC/tandem MS is the candidate reference that samples be protected from light, because there is no
method [27], but it is unlikely to translate into evidence that the presence of RBCs in the sample affords
mainstream analysis in the foreseeable future due to lack protection from the degradative effects seen in separated
of speed and high cost. The microbiological assay is not serum described above.
suitable for routine work but is still considered a gold-
standard procedure. However, the technique is difficult Calculation of RBC Folate from Lysate Folate
to perform, requires specialized microbiological Concentration
equipment, and may take up to 3 days to complete. Since whole blood is processed for the RBC assay, the
hematocrit (packed cell volume) must be determined by
Clinical laboratories almost exclusively use FBP assays an approved method. To obtain red blood cell folate
(Table 1, Method 2), which are available in a variety of expressed as ng folate /mL packed RBCs, a calculation
formats suitable for modern immunoassay analyzers. is required. For platforms that have individual tests for
These assays are rapid, precise, and all steps, including RBC and serum folate, the RBC lysate value is usually
the denaturing of endogenous binders and de- corrected automatically for the dilution made during the
glutamination, are easily automated. Assay formats are hemolysis step. If this is not the case, then the instrument
summarized and compared in Table 2, with examples value must be multiplied by the dilution factor used.
shown in Figures 3 and 4. Standardization of FBP folate Subsequently, all values must be corrected for the
assays is problematic, but the development of a new hematocrit, since whole blood (rather than packed red
standard reference material for homocysteine and folate cells) has been used for the analysis. Strictly speaking,
in serum (SRM 1955) will enable reagent manufacturers the correct formula to apply is:
to address some of these problems. Possible interference
with these assays is addressed in the section below. RCF = WBF [SF (100 H)] 100
100 H
RBC folate is present in much higher concentrations where RCF = red cell folate
than in serum but can be measured by the same WBF = whole blood folate
analytical techniques. Incubation with ascorbic acid for SF = serum folate
90 minutes liberates intracellular folate. The hematocrit H = % hematocrit
should be determined on the sample so the results can be
expressed as a function of red blood cell mass. In practice, serum folate is often not measured
simultaneously, and the part of the formula in brackets is
Specimen in most cases is so small as to be insignificant, so the
Samples for serum (or plasma) folate should be collected calculation is simplified to:
in the fasting state, since folate concentrations reflect RCF = WBF 100
recent food content. In the case of ingestion of folate- H
fortified foods or vitamin supplements, some of the where RCF = red cell folate
plasma content may be inactive PGA if > 300 g folic WBF = whole blood folate
acid has been ingested rather than active MTHF [31]. H = % hematocrit
Stability is maintained for 24 hours at 4C. If kept longer
before assay, the serum should be kept frozen at 20C. Interferences
Serum specimens should be protected from light, In general, hyperbilirubinemia (below 20 mg/dL) or
because there is an approximate 10% and 20% loss of lipemia (below 1000 mg/dL triglycerides) do not
folate after 8 and 24 hours of light exposure, respectively interfere in any of the reported FBP methods. The
[32]. presence of hemoglobin (below 0.5 g/dL) does not per se
interfere in the assay but will artifactually elevate the
Some studies have reported decreased folate in plasma serum folate because of the much higher concentration
when compared with serum. Since 95% of whole-blood of folate in RBCs. The hemolysis index as measured by
folate is found in red blood cells, hemolyzed samples many mainframe chemistry analyzers could be used to
should not be accepted for serum analysis, and these can reject such samples. Although FBP assays share many
be detected and rejected by mainframe chemistry features of immunoassays, the small range of specimen
analyzers that determine hemolysis indices. concentrations encountered (as opposed to a tumor
marker assay, for example) exclude the requirement to
monitor for a Hook effect caused by excess antigen in
577
the patient sample saturating the binding sites on the Folic Acid Reference Interval
capture protein. Likewise, interference by heterophile Reference intervals (normal ranges) vary somewhat by
antibodies might be expected in those FBP assays that method and are rather higher in folate-supplemented
also contain animal antisera (Table 2, Methods 2a, 2b, populations. There is, however, consensus over what
2d, 2e,) and unlikely in those assays that do not (Table 2, constitutes folate deficiency.
Methods 2c, 2f, and 2g ). Serum folate:
Reference interval 3 to 18 ng/mL (6.8 to 40.8 nmol/L)
Heterophile antibodies should be suspected when serial Deficiency <3 ng/mL (<6.8 nmol/L)
specimens from the same patient or repeat analysis of the
same sample produce markedly different results, or Red blood cell folate:
when recovery on dilution is not linear. Analysis by an Reference interval 200 to 1000 ng/mL packed cells
alternative method is usually sufficient to align the (453.1 to 2266 nmol/L)
measured result with the clinical picture, but in some Deficiency <150 ng/mL packed cells (<340 nmol/L)
cases, it may be necessary to prove that these antibodies
exist. In this case, a broad spectrum passive precipitant Interpretation
such as PEG can be used, but it is preferable to use an Folic acid (pteroylglutamic acid, PGA) is stable and is
active blocking agent such as The Scantibodies therefore the form used in dietary supplements. Once
Heterophilic Blocking Reagent Tubes (HBRT). absorbed, it must be reduced to the naturally occurring
Although FDA-approved for this purpose, the value bioactive form tetrahydrofolic acid (THF). Addition of a
obtained from a sample treated with HBR should not be methyl group and subsequent reduction produces N-5-
used as a reportable result. methyltetrahydrofolate glutamate (MTHF). This is taken
up by the bloodstream and constitutes 80% to 95% of
For the assays (Table 2, Methods 2f and 2g) that have circulating folate. MTHF is transported into cells, where
biotin-streptavidin components, specimens from patients it is polyglutamated, preventing its efflux, and it thus
receiving high-dose biotin supplements (>5mg/day) may accumulates within cells (including RBCs). Here it
be problematic. It is recommended that a minimum of 8 functions as both a methyl donor and acceptor and
hours elapse between biotin ingestion and specimen consequently exists in a variety of oxidized and reduced
collection for prolactin analysis. Isolated cases of high- states (Figure 1). Folate in the blood is 90% unbound.
titer antibodies against streptavidin are also mentioned in Ten percent is bound to albumin, 2-macroglobulin, and
the literature, but supporting data are difficult to locate. to a specific folic-acid-binding protein, FBP. FBP
The presence of antibodies to the cross-linker used for increases as the body store of folate declines, probably
attachment of the ruthenium ion in the Roche assay by regulating gene expression in a mechanism similar to
(Table 2, Method 2f) has been documented in a few many other serum binding proteins.
patients with the thyroid assays but is equally applicable
to all assays with this structure [33]. FBP assays are also The primary indication for ordering a serum or RBC
prone to cross-reactivity with other species that can also folate measurement is the differential diagnosis of a
be bound by the folate-binding protein. Methotrexate megaloblastic anemia. Iron deficiency may also be
(amethopterin), aminopterin (4-aminopteroic acid), and present, which may tend to normalize RBC size and thus
folinic acid (5-formylTHF, leucovorin, citrovorum) are confound the picture. Folate is also measured during
chemotherapeutic agents whose structures are similar to pregnancy, because low folate concentrations increase
that of folate. The mechanism of action of methotrexate the risk of neural tube defects in the fetus [2]. However,
is to inhibit cellular folate uptake and its subsequent many pregnant women are folate-supplemented
reduction to active forms. Cross-reactivity data from the automatically without the expense of a folate assay. A
package inserts are summarized in Table 3. diet deficient in folate can lead to folate deficiency after
Superficially, cross-reactivity appears marginal (<1%), some 2 to 4 months of zero dietary intake. Folate
but it should be remembered that while a serum folate deficiencies are often associated with chronic alcoholism
concentration is of the order of 10 ng/mL, a due to concomitant poor nutritive state. Other conditions
methotrexate concentration after initial administration is that place patients at risk for subnormal folate levels are
may exceed 50,000 ng/mL and about 2000 ng/mL at 48 pregnancy and lactation, hypermetabolic states such as
hours [34]. This translates to an apparent serum folate of sepsis or burns, and malignancy (and its treatment).
500 and 20 respectively at 1% cross-reactivity. RBC Many intestinal disorders will result in decreased folate
methotrexate concentrations are only 3% of these in absorption, and in extreme cases, folate injections may
plasma with acute dosing, but on long-term, low-dose be necessary. Oral contraceptives, phenytoin, and other
oral therapy, methotrexate gradually replaces anticonvulsants also decrease folate absorption from the
intracellular folate [35]. Even so, RBC folate intestines.
measurement is unlikely to be adversely affected by
methotrexate therapy [36]. Leucovorin is metabolized to Folic Acid Performance Goals
MTHF intracellularly and therefore repletes intracellular With the considerations outlined above, such as the
active folate. nature of the calibration materials, removal of binders,
and the chemical-reduction processes, it is not surprising
Chromatographic assays are not susceptible to these that there is poor agreement between different assays
types of interferences. and different laboratories. Recent attempts at
standardization with the provision of SRM 1955 at three
578
different concentrations and harmonization of assays are Desirable specifications for analytical imprecision of
making some headway into improving performance [21]. serum folate, derived from studies of biological
variation, indicate an assay imprecision of no greater
Worldwide specimen exchange events have shown poor than 12% and a total error of no greater than 39%. For
performance, with coefficient of variations (CVs) of RBC folate, an assay imprecision of no greater than 6%
18% to 41% and two- to ninefold differences, with red and a total error of no greater than approximately 27% is
cell assays significantly underperforming their serum desirable. The average intra-individual variation of both
counterparts [37,38]. Increased awareness of the serum and RBC folate is approximately 24% [40].
importance of folate has galvanized action to improve
individual assay performance and also to standardize References
between assays. Acceptable performance criteria (CLIA- 1 Uddin E. Blood folic acid studies. VI.
88) for measurement of serum and RBC folate requires Chromatographic resolution of folic acid-active
that laboratories be accurate to within 3 SD of the substances obtained from blood. J Biol Chem
peer-group mean. Although this is for the most part 1959; 234: 2373-6.
achieved, wide bias is seen between the specific assay 2 Blom HJ, Shaw GM, den Heijer M, Finnell RH.
formats, as shown in Table 4. Precision data from the Neural tube defects and folate: case far from
College of American Pathologists 2006-7 Participant closed. Nature Reviews Neuroscience 2006; 7:
Summary Reports for serum folate are shown in Table 4 724-731.
for 1725 laboratories returning results for FBP assays to 3 Molloy AM, Scott JM. Microbiological assay
CAP in 2006-7 [39,40]. Only 40% of these laboratories for serum, plasma, and red cell folate using
participated in the serum cryopreserved, microtiter plate method.
calibration/linearity/verification surveys, and the spread Methods Enzymol 1997; 281: 43-53.
of these data shows the bias between methods. Abbott 4 OBroin S, Kelleher B. Microbiological assay
AxSYM (Table 2, Method 2a) gave the highest values in on microtiter plates of folate in serum and red
all surveys, with a bias of about +26%, while Bayer cells. J Clin Pathol 1992; 45: 344-7.
Centaur (Table 2, method 2c) gave the lowest values, 5 Gutcho S, Mansbach L. Simultaneous
with a bias of about 13%. Comparative data are shown radioassay of serum vitamin B12 and folic acid.
for the 255 laboratories participating in the 2006 United Clin Chem 1977; 23: 1606-14.
Kingdom National External Quality Assurance Scheme 6 Bio-Rad Laboratories. Quantaphase assay
(UKNEQAS), showing imprecision and bias. In this instruction manual. Hercules, CA. 1993.
survey, Beckman-Coulter and Abbott Architect (Table 2, 7 Chitis J. The folate binding in milk. Am J Clin
Methods 2d and 2b and Figure 3) showed the most Nutr 1967; 20: 1-4.
negative bias, with Roche (Table 2, Method 2f and 8 Givas JK, Gutcho S. pH dependence of the
Figure 4) and DPC (Table 2, Method 2e) showing binding of folates to milk binder in radioassay
positive bias. Bias varied with folate concentration, with of folates. Clin Chem 1975; 21: 427-8.
largest difference at the clinical decision cut-off point. 9 Billen J, Zaman Z, Claeys G, Blanckaert N.
Imprecision around the deficiency cut-off (3 ng/mL) is Limited dynamic range of a new assay for
still poor, with most methods returning a CV of 13% to serum folate. Clin Chem 1999; 45: 581-2.
25%, while most methods have a CV below 10% to 12% 10 Wilson DH, Williams G, Herrmann R, Wiesner
at 6 ng/mL. The difference between the two Abbott D, Brookhart P. Issues in immunoassay
methods ranging from + to bias is due to the Architect standardization: The ARCHITECT folate model
being calibrated by harmonization, compared to the for intermethod harmonization. Clin Chem
traditional PGA calibration of the AxSYM method [20]. 2005; 51: 684-7.
11 Hansen SI, Holm J, Lyngbye J, Pedersen TG,
Relatively few U.S. laboratories are participating in RBC Svendsen IB. Dependence of aggregation and
folate schemes compared to serum folate (14%), ligand affinity on the concentration of the
indicating either a smaller number of laboratories folate-binding protein from cows milk. Arch
offering this test or a lack of satisfaction with Biochem Biophys 1983; 226: 636-42.
performance. Calibration/linearity/verification surveys 12 Salter DN, Scott KJ, Slade H, Andrews P. The
are not offered for RBC folate. Results are summarized preparation and properties of folate-binding
in Table 4. In the United Kingdom National External protein from cows milk. Biochem J 1981; 193:
Quality Assessment Scheme (UKNEQAS), RBC folate 469-76.
participation shows a continued decline over recent 13 Mortensen E. Determination of erythrocyte
years, with participation currently at 53% of that of folate by competitive protein binding assay
serum folate. Performance is poor in both schemes, with preceded by extraction. Clin Chem 1976: 22:
an overall CV of almost 40% and individual method 982-92.
CVs of 15% to 25%. In contrast with the serum data, 14 Mortensen E. Effect of storage on the apparent
RBC folate bias does not trend in a predictable manner concentration of folate in erythrocytes, as
with concentration but is variable from one sample to measured by competitive protein binding
another. In general, the Architect shows a negative bias, radioassay, Clin Chem 1978; 24: 663-668.
with Bayer and Beckmann a less pronounced positive 15 Netteland B, Bakke OM. Inadequate sample-
bias [39]. preparation technique as a source of error in
determination of erythrocyte folate by
579
competitive binding radioassay. Clin Chem mass spectrometry. Anal Chem 2005; 77:
1977; 23: 1505-6. 3586-93.
16 Owen WE, Roberts WL. Comparison of five 29 Garbis SD, Melse-Boonstra A, West CE, van
automated serum and whole blood folate assays. Breemen RB. Determination of folates in
Am J Clin Pathol 2003; 120:121-6. human plasma using hydrophilic interaction
17 Shane B, Tamura T, Stokstad R. Folate assay: a chromatography-tandem mass spectrometry.
comparison of radioassay and microbiological Anal Chem 2001; 73: 5358-64.
methods. Clin Chim Acta 1980; 100: 13-9. 30 Fazili Z, Pfeiffer CM, Zhang M, Jain R.
18 Mitchell GA, Pochron SP, Smutny PV, Guity R. Erythrocyte folate extraction and quantitative
Decreased radioassay values for folate after determination by LC/MS/MS: comparison of
serum extraction when pteroylglutamic acid results with microbiologic assay. Clin Chem
standards are used, Clin Chem 1976; 22: 647-9. 2005; 51: 2318-25.
19 Stenman UH. Immunoassay standardization: is 31 Kelly P, McPartin J, Goggins M, Weir D, Scott
it possible, who is responsible, who is capable? JM. Unmetabolized folic acid in serum: acute
Clin Chem 2001; 47: 815-20. studies inn subjects consuming fortified food
20 Lequin RM. Standardization: comparability and supplements. Am J Clin Nutr 1966; 65:
and traceability of laboratory results. Clin 1790-5.
Chem 2002; 48: 391-3. 32 Mastropaolo W, Wilson MA. Effect of light on
21 Satterfield MB, Sniegoski LT, Sharpless KE, serum B12 and folate stability. Clin Chem
Welch MJ, Hornikova A, Zhang N-F et al. 1993; 39: 913.
Development of a new standard reference 33 Ando T, Yasui J, Inokuchi N, Usa T, Ashizawa
material: SRM 1955 (homocysteine and folate K, Kamihara S, Eguchi K. Non-specific
in human serum). Anal Bioanal Chem 2006; activities against ruthenium crosslinker as a
385: 612-22. new cause of assay interference in an
22 Cheruppolil R, Santhosh-Kumar CR, Kolhouse electrochemiluminescent immunoassay. Jap J
JF. Molar quantitation of folate by gas Intern Med 2007; 46: 1225-9.
chromatography-mass spectrometry. Methods 34 Treon SP, Chabner BA. Concepts in use of
Enzymol 1997; 261: 26-34. high-dose methotrexate therapy. Clin Chem
23 Bagley PJ, Selhub J. Analysis of folate form 1996; 42:1322-1329
distribution by affinity followed by reversed- 35 Kremer JM, Galivan J, Streckfuss A, Kamen B.
phase chromatography with electrochemical Methotrexate metabolism analysis in blood and
detection. Clin Chem 2000; 46: 404-11. liver of rheumatoid arthritis patients:
24 Belz S, Frickel C, Wolfrom C, Nau H, Henze association with hepatic folate deficiency and
G. High-performance liquid chromatographic formation of polyglutamates. Arthritis Rheum
determination of methotrexate, 5- 1986; 29: 832-5.
methyltetrahydrofolic acid and folinic acid in 36 Grim J, Chladek J, Martinkova J.
serum and cerebrospinal fluid. J Chromatogr B Pharmacokinetics and pharmacodynamics of
1994; 661: 109-18. methotrexate in non-neoplastic diseases. Clin
25 Lucock MD, Hartley R, Smithells RW. A rapid Pharmacokinet 2003; 42:139-51.
and specific HPLC-electrochemical method for 37 Van den Berg H, Finglas PM, Bates C. FLAIR
the determination of endogenous 5- intercomparisons on serum and red cell folate.
methyltetrahydrofolic acid in plasma using solid Int J Vitam Nutr Res 1994; 64: 288-93.
phase sample preparation with internal 38 Gunter EW, Bowman BA, Caudill SP, Twite
standardization. Biomedical Chromatography DB, Adams MJ, Sampson EJ. Results of an
1989; 3: 58-63. international round robin for serum and whole
26 Pawlosky RJ, Flanagan VP. A quantitative blood folate. Clin Chem 1996; 42: 1689-94.
stable-isotope LC-MS method for the 39 Blackmore S, Hamilton MS. UKNEQAS
determination of folic acid in fortified foods. J Annual Report 2006 for haematinic assays and
Agric Food Chem 2001; 49:1282-6. intrinsic factor antibodies. February 2007
27 Pfeiffer CM, Fazili Z, McCoy L, Zhang M, (Personal Communication).
Gunter EW. Determination of folate vitamers 40 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
in human serum by stable-isotope-dilution Hernandez A, Jimenez CV, et al. Current
tandem mass spectrometry and comparison with databases on biologic variation: pros, cons and
radioassay and microbiologic assay. Clin Chem progress. Scand J Clin Lab Invest 1999 ; 59:
2004; 50: 423-32. 491-500.
28 Nelson BC, Satterfield MB, Sniegoski LT, 41 Mincey EW, Wilcox E, Morrison RT.
Welch MJ. Simultaneous quantification of Estimation of serum and red cell folate by a
homocysteine and folate in human serum or simple radiometric technique. Clin Biochem
plasma using liquid chromatography/tandem 1973; 6: 274-84.
Table 1: Folate Methods Summary

Method 1: Microbiological
Type of analysis: Bioassay
580
Principle: Growth of folate-dependent bacteria related to specimen folate concentration; growth monitored by
turbidimetry
Usage: Rare, historical reference method
Comments: False-negative because of presence of antibiotics or antibodies to target organism; false-positive
results because of turbidity
Method 2: Folate Binding Protein (FBP) Assays
a. Type of analysis: Competitive protein binding assay
Principle: Folates in sample compete with labeled ligand for binding to milk proteins (such as lactoglobulin);
separation of bound and free labeled-ligand by dextran-coated charcoal, pre-binding to magnetic particles or
vessel walls
b. Type of analysis: Immunometric protein binding assay (see Table 2)
Principle: Pre-incubation of specimen and FBP, with sequential addition of tracer. Tracer-labeled ligand occupies
remainder of FBP sites not occupied by serum folate. Separation of bound and free labeled-ligand by pre-binding
to magnetic particles or vessel walls. Signal generated may be radioactivity, chemiluminescence, fluorescence, or
an enzyme-generated product.
Comments: Easily automated, must destroy endogenous folate binders
Method 3: Liquid Chromatography (HPLC)
Type of analysis: Chromatographic separation after isolation, using solid-
phase or affinity extraction
Principle: Signal generated may be electrochemical or Mass Spectrometric
Usage: Specialist laboratories only
Comments: Proposed reference method, low throughput
Table 2: Folate Binding Protein Methods Summary

ULOQ ng/mL
Pre-treatment Detection Label / Separation LLOQ ULOQ
Method Solid phase / conjugate Assay Mechanism Detection System Extended by
System Bound to: method ng/mL ng/mL
Automatic Dilution
2 step
anionic microparticles /
DTT then 0.8 M ALP-folate binds to ALP / ionic attraction by 4-Methylumbelliferyl-phosphate
2a mc mouse anti-FBP 0.9 20 no
KOH excess conjugated folate fiber filter Fluorescent product
coupled to FBP
FBP
2 step Acridinium (N-sulfonyl)
paramagnetic particles
Acridinium-folate binds Acridinium ester / carboxamide yes for serum
2b DTT then KOH / mc mouse anti-FBP magnet 0.8 20
to excess conjugated folate Chemiluminescence with x2
coupled to FBP
FBP Peroxide and NaOH trigger
DTT 95 mg/L competitive binding for Acridinium ester
paramagnetic particles Acridinium ester / yes for serum
2c then 1.1 M limiting biotin-labeled magnet Chemiluminescence with 0.35 24
/ avidin folate x2, x5
NaOH FBP Acid and NaOH trigger
0.1 M HCl paramagnetic particles

competitive binding for ALP / Chemiluminescent product with
2d neutralized with / goat anti-mouse / mc magnet 0.5 20 no
limiting FBP folate Lumi-Phos 530 substrate
0.5 M NaOH mouse anti FBP
competitive binding for Luminescent product with

DTT then polystyrene bead / ALP / yes for serum
2e limiting FBP between centrifugal wash Adamantyl-1,2-dioxetane 0.8 24
NaOH/KCN mc mouse anti-FBP anti-ligand x2, x5
folate and folate-ligand phosphate substrate
monothioglycerol 2 step immunometric Electrochemiluminescence with Ru

paramagnetic particles Ruthenium complex
2f 53.3 g/L / NaOH folate-biotin binds to magnet Electron amplification with 0.6 20 no
/ streptavidin / FBP
37 g/L excess FBP Tripropylamine
Luminol derivative substrate
competitive binding for
well wall / HRP / Peracid salt peroxide generator
2g 1.2 M NaOH limiting biotin-labeled wash 0.2 20 no
streptavidin folate Electron amplification with 3-
FBP
Chloro,4-hydroxy acetanilide
a Abbott Axsym ALP Alkaline phosphatase

b Abbott Architect HRP Horseradish Peroxidase
c Bayer Centaur * mc = monoclonal; pc = polyclonal
d Beckman-Coulter Access / DxI FBP Folate Binding protein
e Diagnostic Products Immulite Series DTT Dithiothreitol
f Roche Elecsys Systems
g Vitros ECi
Table 3: Cross Reactivity Data for Folate Binding Protein Assays

581
Manufacturer and Instrument ng/mL Added % Cross Reactivity

Methotrexate Leucovorin Aminopterin
Abbott Axsym 100 <2 <3 <1
Abbott Architect 100 1.1 -0.3 1.2
Bayer Centaur not stated --- --- ---
Beckman-Coulter Access / DxI 57 <1 <1 0.2
Diagnostic Products Immulite Series 100 0.9 --- ---
Roche Elecsys Systems not stated 2.3 2.3 2.7
Vitros ECi 500 <0.04 <0.04 <0.04
Data extracted from manufacturer package inserts, experiments performed according to NCCLS Protocol EP7
Table 4: Performance of Folate Binding Assays
Survey CAP 2006 - 7 UKNEQAS 2006

Average of 6 Spot Summary n = 135 Labs
n=242 Labs
Challenges
Method % of Labs mean ng/mL cv % % of labs % cv
2a 2.5 --- --- --- ---

2b --- --- --- 20.7 ---
2c 47.9 316.3 14.1 20.0 ---
2d 22.3 254.6 12.0 39.3 ---
2e 8.3 210.1 28.8 8.1 ---
2f 11.6 405.9 14.0 6.0 ---
2g 5.0 148.4 22.4 --- ---
Others 2.4 6 ---
All Results 100.0 299.9 39.3 100.0 35 - 40
a = Abbott AxSYM; b = Abbott Architect; c = Bayer Centaur; d = Beckman-Coulter Access/DxI; e = Diagnostic Products
Immulite Series; f = Roche Elecsys Systems; g = Vitros ECi.
582
Figure 1. Structures and metabolic pathways for folates.
Folic acid
Pteroic acid
L-Glutamic acid
p-Aminobenzoic
acid O H
O C
2-Amino,
O
4-hydroxy-
pteridine N C
H
10 R
OH HN R Name
4a N 7 OH Monoglutamyl folate
N3 4 5 6
2 1 -glutamyl}n Polyglutamyl folate
9
H2N N 8a N
Reduction
R'
5 10
One Carbon Entity
OH HN N N
N Oxidized H CHO N-10-Formyl-

N
CHO H N-5-Formyl-
H2N N N H CHNH H N-5-Formimino-
H
CH N-5,10-Methenyl-
7,8-Dihydrofolate
CH2 N-5,10-Methylene-
Reduction
Reduced CH3 H N-5-Methyl-
OH HN 10
H OH R' HN
N One-Carbon N
N
H Addition N 5
H2N N N H
H2N N N
H H
5,6,7,8-Tetrahydrofolate
583
Figure 2. Diagram shows inter-relationship of transmethylation (red) and remethylation (black)

pathways and the dependence on Vitamins B12 and B6.
CS, Cystathionine--synthase; DHF, dihydrofolate; MS, methionine synthase; MTHF,

methyltetrahydrofolate; SAM, S-adenosylmethionine; SHM, serine hydroxymethyl transferase; THF,
tetrahydrofolate; TS, thymidylate synthase
584
Figure 3. Schematic for an immunometric folate binding protein assay (Table 2, Method 2d).
Reproduced with permission from http://www.beckman.com/products/testmenu/access.asp
In this assay, folic acid conjugated to alkaline phosphatase competes with folic acid released from the patient sample for
limited amount of milk FBP. After incubation with paramagnetic particles coated with antibodies to FBP, unbound entities
are removed by washing. The amount of enzyme bound to the particles is inversely proportional to the patient folate
concentration. Enzyme substrate, dioxetane-P (4-methoxy-4-(3-phosphatephenyl)spiro [1,2-dioxetane-3,2-adamantane] is
converted to a meta-stable anion, which fragments further to form an excited-state anion that emits a steady glow of light at
477 nm as it reverts to the ground state. Lumi-Phos* 530 is a micellar formulation containing dioxetane phosphate,
magnesium chloride, cetyltrimethylammonium bromide (CTAB) in 2-amino-2-methyl-1-propanol buffer (pH 9.6), together
with a fluorescent surfactant as an additional signal enhancer. This co-surfactant system promotes energy transfer to the
fluorescent head groups, causing the generation of luminescence with a broad maximum at 530 nm. It provides
ultrasensitive detection and is 10,000 times more sensitive than colorimetric substrates in solution.
H2O +
+ + H2 O
+
+
O O
+ +
C C
+ H2O +
+ +
+
H2 O
Fluorescein head group
+ CTAB
585
Figure 4. Two-step electrochemiluminescent immunometric assay. (Roche Diagnostics Table 2,

Method 2f.)
1st Incubation
Folate-Biotin
Folate in
Excess Ru-FBP conjugate
patient sample
Wash
off
Streptavidin Magnetic
coated magnetic Capture
micro-particles
Ru(tpy)2 2+
Ruthenium bis(2,2:6,2-
terpyridine)
Photons
e- TPA
H+
Ru2+ 3+
RuTPA TPA
base state
TPA+
Electron donor
e-
(Tripropylamine, TPA)
+ Platinum Electrode + ensures that the reaction
Magnet continues to cycle,
multiplying the signal
In the first step, patient sample is incubated with biotinylated folate and an excess of ruthenium-labeled folate binding
protein (FBP). Next, streptavidin-coated paramagnetic microparticles are added and incubated. A sandwich complex is
formed, which becomes bound to the magnetic particles by the interaction of streptavidin and biotin. After washing to
remove unbound components, the magnetic particles are attracted out of solution by a magnet onto the electrode.
Application of a voltage to the electrode then induces chemiluminescent emission, which is measured by the
photomultiplier. The addition of an electron donor (tripropylamine) ensures that the reaction continues to cycle, thus
improving the sensitivity.
586
Follicle-Stimulating Hormone
Angela Ferguson
Name: Follicle-stimulating hormone, FSH, or follitropin

Clinical significance: Refer to Chapter 50, The Gonads, in the 5th edition of Clinical Chemistry: Theory,
Molecular weight: 30 kDa
Reference method: None
Principles of Analysis and Current Usage as low as 0.05 mIU/mL for some instruments (e.g. the
Follicle-stimulating hormone (FSH) is a 30-kDa Abbott Architect method). Radioimmunoassays (RIA) are
glycoprotein produced by the pituitary. FSH, also called available (Siemans offers the Coat-A-Count IRMA), but
follitrop1n, contains two subunits to which are covalently they are no longer widely used.
linked carbohydrate groups, including fructose, mannose,
galactose, glucosamine, galactosamine, and sialic acid. Historically, serum concentrations and therapeutic
The -subunit is chemically similar to the -subunits of preparations of FSH were measured by in-vivo bioassays
luteinizing hormone (LH), thyroid-stimulating hormone such as the Steelman-Pohley method, which relies upon
(TSH), and human chorionic gonadotropin (hCG). The - the linear relationship between the amount of FSH
subunits of these hormones are biochemically distinct and administered and ovarian weight in female rats [2,3].
impart their hormonal and immunological specificity. Bioassays have poor precision, sensitivity, and specificity,
Whereas isolated -subunits are biologically inactive, free and they are time consuming and expensive. High-
-subunits retain some activity, but both subunits must be performance liquid chromatography is being used more
present and recombined for full activity. frequently now to quantify different recombinant
preparations of FSH and decrease the variability between
In women, FSH is released from the pituitary gland in a batches [3,4].
pulsatile fashion throughout the menstrual cycle. There is a
small peak of FSH at the end of the cycle (starting on Because of the great heterogeneity of FSH created by the
about day 26) that begins the follicular maturation of the different combinations of carbohydrate modifications, it is
next cycle. Serum FSH concentrations fall and remain low difficult to classify it as a single molecule; it is more
during the follicular phase but peak again at midcycle. The correct to consider FSH as a mixture of closely related
importance of the mid-cycle peak is not entirely clear. In compounds. This presents a problem in comparing
males, FSH is also secreted in a pulsatile fashion and acts methods, because assay-specific antibodies recognize the
on Sertoli cells to stimulate spermatogenesis. various FSH isoforms differently [5]. The earliest
reference material used to standardize FSH assays was
Reference and Preferred Methods prepared from the urine of postmenopausal women. This
Immunoassays have become the method of choice for calibrator was not appropriate for serum measurements and
measuring serum FSH, with all 1650 labs participating in was later replaced with purified pituitary extracts.
the 2007 College of American Pathologists (CAP) survey, Recombinant gonadotropin calibrators are now available.
using two-site sandwich immunoassays with direct Most assays currently use the World Health Organization
chemiluminometric technology [1]. Typically these two- (WHO) 2nd IRP 78/549 (interim replacement code
site sandwich assays consist of one antibody against the 94/632).
alpha chain and one antibody against the beta chain. These
assays are very specific, and cross-reactivity with other Specimen
pituitary glycoproteins such as LH, TSH, and hCG is Serum is the preferred specimen for measuring
effectively eliminated. Claimed analytical sensitivities are gonadotropins. FSH in serum is stable at room temperature
for 8 days and for 14 days at 4C [6]. Serum should be
1
Follicle-Stimulating Hormone kept frozen at or below 20C for long-term storage.
Previous and current authors of this method: Plasma FSH is stable for at least 8 days at 4C and at room
First edition: Not done temperature and is not affected by up to 5 freeze/thaw
Methods edition: Not done cycles [7]. Urine FSH is stable for up to 32 weeks at both
Second edition: Not done 4C and 20C, but the stability declines after that [8].
Third edition: Not done Because of the episodic and cyclic secretion of FSH,
Fourth edition: Amadeo J. Pesce pooled or serial blood specimens or timed urine collection
Fifth edition: Angela Ferguson can sometimes be useful [9]. Urinary gonadotropin
measurements are used to diagnose pubertal and pituitary
587
disorders in children, owing to the low levels of Monitoring FSH is important both in the diagnosis and
gonadotropins found in serum and the less invasive nature treatment of infertility in women. Increased
of urine collection [10]. concentrations of serum FSH are an indication of
diminished ovarian reserve, and depending on the age of
Interferences the woman, it can also suggest how she will respond to
As with all sandwich immunoassays, heterophile ovarian stimulation protocols [16]. Day 3 FSH
antibodies are a possible interferent in FSH assays [11]. concentrations > 15 mIU/mL are considered elevated and
Plasma samples containing EDTA can give a falsely associated with poor reproductive outcome, as are basal
elevated result when using the Biomerieux Vidas, and estradiol concentrations > 75 pg/mL [17,18]. The diagnosis
sodium citrate in plasma can cause results to be falsely of polycystic ovarian syndrome (PCOS) can be aided by
decreased by 20% when using the Roche Elecsys 1010 and measurement of serum testosterone, DHEA-S, LH, and
2010 and MODULAR ANALYTICS analyzers [12]. FSH. Concentrations of LH are often elevated, with low to
low-normal concentrations of FSH. Some utilize an LH:
Reference Intervals FSH ratio of > 2 to indicate PCOS [19].
Reference intervals vary with assay manufacturer. Given
here are serum FSH reference intervals for three of the five During menopause, the decrease in estradiol concentration
most widely used methods according to CAP. results in increased serum LH and FSH concentrations (see
Table in Reference Intervals) [20]. Urine-based home FSH
Table 1. kits are now available which are intended to help
ADVIA Abbott DPC determine menopausal status. These tests are not
Centaur AxSym Immulite definitive, owing to the fluctuations in FSH concentrations,
(mIU/mL) (mIU/mL) 2000 and can only suggest a trend; there have been no published
(mIU/mL) peer-reviewed studies examining their utility.
Women
Follicular Phase 2.5-10.2 4-13 2.8-11.3
Children
Mid-Cycle Peak 3.4-33.4 5-22 5.8-21
In children, FSH measurement can aid in the diagnosis of
Luteal Phase 1.5-9.1 2-13 1.2-9.0
Postmenopausal 23.0- 20-138 21.7-153
delayed or precocious puberty. Most patients present with
116.3 failed or arrested puberty due to low plasma concentrations
Men 1.4-18.1 1-8 0.7-11.1 of LH, FSH, and testosterone. The gold standard for the
diagnosis of central precocious puberty is the GnRH
Interpretation and Performance Goals stimulation test. A peak serum concentration of LH > 8
Measurements of serum FSH are used in men, women, and IU/L after administration of GnRH is considered
children to evaluate pituitary function and differentiate diagnostic for precocious puberty. It is also useful to
between hypergonadotropic and hypogonadotropic compare the ratio of FSH concentration to LH
hypogonadism. concentration after stimulation; a predominant LH
response is indicative of a pubertal response [21]. In girls
Males with peripheral precocious puberty, the secretion of sex
In men, concentrations of FSH greater than 120 mIU/mL steroids is independent of pituitary gonadotropic release,
along with decreased testosterone are indicative of primary thus the response to the GnRH stimulation test is
testicular failure. Klinefelters syndrome is associated with repressed, and concentrations of LH and FSH are low. If
dysgenesis of the seminiferous tubules, a lack of sperm there are high concentrations of FSH and LH in
production, and marked elevation of serum FSH and LH conjunction with low Tanner stages, this can be a sign of
concentrations [13]. Decreased concentrations of FSH and delayed puberty and hypergonadotropic hypogonadism.
testosterone suggest hypogonadotropic hypogonadism.
References
Women 1 College of American Pathologists. Participant
Hypergonadotropic hypogonadism arises from primary summary. Y-A Ligands (Special). 2007:6-7.).
gonadal failure. The most common cause for 2 Rose MP, Gaines Das RE, Balen AH. Definition
hypergonadotropic hypogonadism is Turners Syndrome, and measurement of follicle stimulating hormone.
which is defined as a 45 X0 karyotype and female Endocrine Reviews 2000; 21: 5-22.
phenotype [14,15]. Moderately high concentrations of FSH 3 Driebergen R, Baer G. Quantification of follicle
and LH are also found in cases of androgen insensitivity, stimulating hormone (follitropin alfa): is in vivo
where the patient has a 46 XY karyotype but is bioassay still relevant in the recombinant age?
phenotypically female. In patients without signs of Curr Med Res Opin 2003; 19: 41-46.
hyperandrogenism, elevated concentrations of FSH and 4 Loureiro RF, de Oliveira JE, Torjesen PA,
LH could indicate ovarian failure. Bartolini P, Ribela, MTCP. Analysis of intact
human follicle-stimulating hormone preparations
by reversed-phase high-performance liquid
588
chromatography. J Chromatogr 2006; 1136: 10- 13 Bojesen A, Gravholt CH. Klinefelter syndrome in
18. clinical practice. Nat Clin Pract Urol. 2007; 4:
5 Sturgeon CM, Ellis AR. Standardization of FSH, 192-204.
LH and hCG-Current position and future 14 Slap GB. Menstrual disorders in adolescence.
prospects. Mol Cell Endocrinol 2007; 260-262: Best Pract Res Clin Obstet Gynaecol 2003; 17:
301-309. 75-92.
6 Kubaski NP, Ricotta M, Hunter T, Sine, HE. 15 The Practice Committee of the American Society
Effect of duration and temperature of storage on for Reproductive Medicine. Current evaluation of
serum analyte stability: examination of 14 amenorrhea. Fertil Steril 2006; 86(Suppl 4):
selected radioimmunoassay procedures. Clin S148 155.
Chem 1982; 28: 164-165. 16 Barad DH, Weghofer A, Gleicher N. Age-specific
7 Livesey JH, Hodgkinson SC, Roud HR, Donald levels for basal follicle-stimulating hormone
RA. Effect of time, temperature and freezing on assessment of ovarian function. Obstet Gynecol
the stability of immunoreactive LH, FSH, TSH, 2007; 109: 1404-1410.
growth hormone, prolactin, and insulin in plasma. 17 Macklon NS, Fauser BC. Ovarian reserve. Semin
Clin Biochem 1980; 13: 151-155. Reprod Med. 2005; 23: 248-256.
8 Brindle E, Miller RC, Shofer JB, Klein NA, 18 Licciardi FL, Liu HC, Rosenwaks Z. Day 3
Soules MR, OConnor KA. Urinary beta- estradiol serum concentrations as prognosticators
luteinizing hormone and beta-follicle stimulating of ovarian stimulation response and pregnancy
hormone immunoenzymometric assays for outcome in patients undergoing in vitro
population research. Clin Biochem 2006; 39: fertilization. Fertil Steril. 1995; 64: 991-994.
1071-1079. 19 Carr BR. Disorders of the ovary and female
9 Demers LM. General Endocrinology. In:Kaplan reproductive tract. In: Williams RH, Foster DW,
LA, Pesce AJ, Kazmierczak SC, editors. Clinical Kronenberg HM, Larsen, PR, Wilson JD, eds.
Chemistry: Theory, Analysis, Correlation. St. Williams textbook of endocrinology. 9th ed.
Louis: Mosby; 2003.p826. Philadelphia: WB Saunders, 1998:751-817.
10 Demir A, Dunkel L, Stenman U-H, Voutilainen. 20 Burger HG, Dudley EC, Hopper JL, Groome N,
Age-related course of urinary gonadotropins in Guthrie JR, Green A et al. Prospectively
children. J Clin Endocrinol Metab 1995; 80: measured levels of serum follicle-stimulating
1457-1460. hormone, estradiol, and the dimeric inhibins
11 Boscato LM, Stuart MC. Heterophilic antibodies: during the menopausal transition in a population-
a problem for all immunoassays. Clin Chem based cohort of women. J Clin Endocrinol Metab
1988; 34: 27-33. 1999; 84: 4025-4030.
12 Youngs Effects Online [database on the internet]. 21 Lee PA. Central precocious puberty. An overview
Washington DC: AACC. c2005 [updated 2007 of diagnosis, treatment, and outcome. Endocrinol
Jun 13; cited 2007 Jun 27]. Available from: Metab Clin North Am. 1999; 28: 901-918.
http://www.fxol.org/
589
Free Thyroxine and Free Triiodothyronine

Greg Ward
Name: Free T4 (fT4) and free T3 (fT3)

Clinical significance: Refer to Chapter 49, Thyroid, in the 5th edition of Clinical
Merck Index: 9262 (thyroxine), 5337 (3,5,3-triiodothyronine)
Chemical class: Amino acid
Structure:
Thyroxine
3,5,3-Triiodothyronine
i amount of protein-bound hormone would only
The measurement of free thyroid hormones serve as a passive reservoir to maintain the free
presents two major problems. One is analytical, the hormone concentration within set limits. However,
other is clinical. The analytical problem is that there is evidence that protein-bound hormones may
serum concentration of the free (non-protein- also be available for transport to target tissues.
bound) fractions of both physiologically active
thyroid hormones, thyroxine (T4) and Physiological factors that strongly affect the
analysis of free thyroid hormones are their plasma
triiodothyronine (T3), is extremely low (both in the
concentration and the mode of transport of the
pmol/L range). The clinical problem is related to entire T4 and T3 pool. In man, approximately
the continuing controversy regarding the
0.03% of 7.4 g/dL (or 96 nmol/L) of plasma T4 is
physiological role of nonprotein/nonpeptidetype
free hormones in general. The clinical use of free found free (about 2.2 ng/dL or 29 pmol/L), whereas
hormone measurements assumes the validity of the 0.4% of the total 100 ng/dL (or 1.54 nmol/L)
free hormone hypothesis, according to which only plasma T3 is in the free form (about 400 pg/dL or
the free (i.e., non-protein-bound) fraction of the 6.2 pmol/L) under standard thermodynamic
hormone is able to enter the target cell and to exert equilibrium conditions [5].
physiological activity [1-4]. Thus the much larger
Laboratory assessment of the serum concentration
i of free thyroid hormone began in the 1960s.
Free Thyroxine and Free Triiodothyronine Similar to other free (non-protein-bound)
Previous and current authors of this method: hormones, the free thyroid hormones initially were
First edition: Not done measured by indirect equilibrium dialysis (tracer
Methods edition: Andre van der Walt dialysis) or indirect ultrafiltration [6-16].
Third edition: Not updated Although several radionuclides (e.g., 3H) could be
Fourth edition: Gyorgy Csako considered theoretically for radiolabeling of thyroid
Fifth edition: Greg Ward hormones, for obvious reasons the isotopic
590
methods for free thyroid hormones involved and Physical separation methods which
continue to involve the use of radio-iodinated separate free from bound thyroid hormone
(131I- or 125I-labeled) thyroid hormones. Since the prior to direct assay of the free fraction.
1970s, a wide variety of relatively inexpensive, Examples include direct equilibrium
technically simple, often automated non-isotopic dialysis (ED) and ultrafiltration (UF). In
free T4 (fT4) and free T3 (fT3) assays have become ED, the free-thyroid-hormone
concentration is determined at the time
available. The basic concept, performance, and
when thermodynamic equilibrium is
evolution of assays for the measurement of free
attained across a membrane or partition
thyroid hormones have been reviewed repeatedly
phase in the assay system. The final
[1,2,16-33]. Virtually all of these assays involve
measurement depends on the distribution
the use of an antibody or antibodies at some stage
of hormone between the free and bound
of the analysis; that is, they are all at least in part
state or the absolute amount of hormone in
immunoassays.
the free moiety.
Before single-test, free-thyroid-hormone ligand
According to the type of calibration involved, the
assays became widely available in routine
assay can be:
laboratories, the free-thyroid-hormone status had
1. Absolute. Calibrated by gravimetrically
been routinely evaluated by the T3 uptake (T3U)
established quantities of (pure) hormone
test. The T3U test is now named the thyroid dissolved in buffer.
hormonebinding ratio (THBR). Originally 2. Comparative. Calibrated by serum standards
described in the late 1950s [29,34], this test that in turn are traceable in calibration to absolute
measures the number of unoccupied thyroid methods.
hormonebinding sites on serum proteins,
primarily on TBG. If the proportion of thyroid Techniques for Free-Thyroid-Hormone Estimates
hormone carried by TBG is at least 60%, the T3U
correlates with the fT4 and fT3 fractions in the Index Methods Using Free Hormone Fraction
Determination
serum, and determination of the T3U, in
As mentioned earlier, indirect (tracer) equilibrium
combination with total thyroid hormone dialysis (Table 1, Method 1) was the first method
determinations, allows indirect estimation of the capable of measuring the absolute fT4
fT4 and fT3 concentrations in terms of indices (free
concentration [6]. For assay, first a minute amount
T4 index, fT4I [or less correctly, fTI], and free T3
of radiolabeled (usually 131I or 125I) thyroid
index, fT3I) [35,36]. The total thyroid hormone hormone (T3 or T4) was preincubated with the
concentrations correct for variations in binding serum sample to provide uniform labeling of the
protein such as TBG concentrations that are total thyroid hormone. The sample typically was
relatively often encountered in everyday clinical used diluted. The free moiety of hormone was
situations. separated from the protein-bound fraction by
dialysis through a semipermeable membrane, and
Free-Thyroid-Hormone Techniques the amount of radiolabeled T4or T3 in the dialysate
Free-thyroid-hormone assays are now classified as was measured. At equilibrium, the product of the
follows: dialyzable (free) fraction of total radiolabeled T4 or
Index Methods which require two T3 and the total T4 or T3 concentration of the
unrelated measurements from which the
concentration of the free hormone is sample was the mass of fT4 or fT3 in the test
derived. One is the unsaturated binding system. Correction for sample dilution yielded the
protein capacity, the total TBG free hormone concentration in the patients serum.
concentration, or the free hormone In euthyroid individuals, the %fT4 or %fT3
fraction; the other is the total hormone determined by equilibrium dialysis is inversely
concentration, usually measured by an proportional to the respective thyroid hormone
immunoassay. Examples include classic binding capacity, but the mass of fT4 or fT3 in the
fT4I, T4/TBG ratio, and indirect (tracer)
system remains essentially the same. The
equilibrium dialysis. separation of radioiodinated thyroid hormones from
Single-test ligand assays. A single contaminating iodines in the dialysate was
measurement for free hormone achieved by a variety of methods. However,
concentration is performed and requires because of the many elution or wash steps
no subdivision of the test sample (e.g., involved, these methods never became suitable for
one-step [analog] and two-step use in routine clinical laboratories. Even the use of
immunoassays). the most practicable and popular separation
591
technique based on MgCl2 precipitation of thyroid antibody. The solid phase may be a glass fiber filter
hormones [12,37] remained restricted to reference with an immobilized second antibody,
and research laboratories. magnetizable (paramagnetic) particles with a
second antibody, or a multilayer thin film. For
Index Methods Using TBG Measurement signal detection, the tracers may be labeled with an
(Calculated fT4 and fT3 ) (Table 1, Method 2) enzyme (e.g., alkaline phosphatase, glucose-6-
phosphate dehydrogenase) or with a more directly
Since the free-thyroid-hormone concentration is
measurable compound (e.g., fluorescein, an
governed by the law of mass action, knowledge of
undisclosed type of fluorescent compound, or
the total hormone concentrations, binding
acridinium ester). The signal detection may be
constants, and binding-site concentrations (i.e.,
based on colorimetry (e.g., p-nitrophenol or
binding-protein concentrations and the number of
chromogen from a patented substrate), ultraviolet
binding sites per binding-protein molecule) allows
spectrophotometry (e.g., reduced nicotine adenine
the calculation of both the free hormone
dinucleotide, NADH), fluorescence (e.g., 4-
concentration and the distribution of hormone
methylumbelliferone, or fluorescent label),
among various binding proteins. A number of
fluorescent polarization (e.g., fluorescein), and
different models based on the mass action law have
chemiluminescence (e.g., acridinium, unstable
been developed for computer-assisted calculations
intermediates from the phosphate ester of
of fT4 and/or fT3 concentrations [38-42].
adamantyl dioxetane, or Lumi-Phos 530).
Techniques for Measuring Thyroid Hormone Index Methods Using Free-Thyroid-Hormone

Binding Ratio(THBR) (Formerly Known as T3 or Indices
T4 Uptake [T3U or T4U])
Classic Free T4 and Free T3 Indices (fT4I and
Classic Isotopic THBR or T3U (and T4U) Tests fT3I) (Table 1, Method 3)
The T3U test was developed originally by In 1966, Clark and Horn [35] proposed first that
Hamolsky et al. [34,43] in the late 1950s as an because of its good correlation with the serum fT4
isotopic in-vitro diagnostic test of thyroid function. concentration, the product of T3U (obtained by
In the original T3U tests, a fixed amount of their anion-exchange resin uptake method) and
radioactive T3 was incubated with whole blood, serum protein-bound iodine (PBI, as a measure of
and the binding of radioactive T3 by red blood cells the total thyroid hormone concentration) should be
called the free thyroxine index (fTI). It is now
(RBC) was measured by counting the radioactivity generally agreed that the product of THBR and the
in the separated erythrocytes obtained after total T4 or T3that is, the fT4I or fT3Iis a
centrifugation and washing:
reliable indirect estimate of the fT4 or fT3
counts in RBC(i.e.,matrixboundcounts concentration in many patients [25,36,50-56]. Thus
T3U(%) if the THBR is to be determined, then the total
totalcounts thyroid hormone (T4 or T3) concentration should
also be measured and the fT4I or fT3I calculated
Under these circumstances, the radioactivity taken [57]. Since these indices are products of the total
up by red blood cells (the T3 uptake) was T3 or T4 concentration and a ratio, they
inversely proportional to the radioactivity bound to theoretically have concentration units. However,
T4-binding serum proteins, especially TBGthat the Committee on Nomenclature of the ATA [57]
is, a high T3 uptake meant high TBG saturation recommended that in order to avoid confusion with
and reduced available T4-binding sites. serum T4 or T3 results, the free-thyroid-hormone
index units be either omitted or simply termed
Nonisotopic THBR or T3U and T4U (T index units.
Uptake) Tests
Single Test Ligand Assays for Estimation of fT4
Since the late 1970s, non-isotopically labeled T3
or fT3
and T4 have been increasingly used as tracers in a Ligand assays for the fT4 (or FT3) fraction of serum
variety of fully automated uptake tests [44-49]. are designed not to disrupt the equilibrium between
Whereas in heterogeneous immunoassay free and protein-bound thyroxine. In practice,
formulations, the antibodies continue to require thyroxine is dissociated from the bound fraction
binding to a solid phase and thus form a classic [31]. However the more important factors for free
secondary binder, in the homogeneous T-uptake thyroxine estimates are (1) whether the dissociation
immunoassays, all reactions take place in solution, of hormone occurs identically in samples and
that is, the secondary binder is a liquid-phase standards and (2) whether samples and standards
592
show identical protein binding of the assay tracer or T3) antibody. The free thyroid hormone and the
[30]. immobilized hormone (macro) analog molecules
[2] compete for binding sites on the labeled
Two-Step Labeled Hormone (Sequential/Back- antibody. The Ortho Vitros ECi thyroid hormone
Titration) Immunoassay (Table 1, Method 4) assays and the Siemens ADVIA Centaur FT3
Early coated-tube RIAs for free thyroid hormones assays are examples of this immunoassay format.
employed a sequential, two-stage method
[21,22,58-60]. A commercial version of this Equilibrium Dialysis (Table 1, Method 7)
methodology was introduced which entailed the In direct equilibrium dialysis, the free thyroid
binding of free hormone to polyclonal antibodies hormones are first separated from the protein-
coated to the walls of assay tubes bound moieties. After dialysis, the (free) T4 and/or
(immunoextraction). Since then, other solid-
T3 in the dialysate is directly measured by a
phase materials such as micro- and macrobeads
have also been used for immobilizing the sensitive T4 or T3 immunoassay [20,79], by gas
antithyroid-hormone antibodies. In a basic two-step chromatography (GC) with electron capture
assay, after decanting and washing of serum, the detection [80,81], or by mass spectrometry.
unoccupied antibody-binding sites are Although the latter technique has the advantage of
(sequentially) back-titrated with the respective providing concomitant, absolute analysis for free
labeled thyroid hormone. As a consequence, (1) the (dialyzable) T4 and T3 in the sample, it is both
sample and the labeled hormone are never in cumbersome and expensive, with little potential for
contact with each other in a classic two-step assay, routine diagnostic use.
and (2) the more binding of labeled hormone
during the second step, the less free hormone was In indirect symmetric dialysis, the unknown serum
present, that is, the signal and the free-thyroid- containing a minute quantity of radiolabeled
hormone concentration are inversely related. In this thyroid hormone is dialyzed against the same
assay, the reaction between serum proteins and unspiked serum. The initial rate of entry of the
labeled thyroid hormone is avoided. The Abbott radiolabeled thyroid hormone into the second
AxSYM, Abbott ARCHITECT, Beckman DxI and compartment is proportional to the free hormone
Wallac Delfia free-thyroid-hormone assays are all concentration in serum. With proper calibration,
two-step nonisotopic immunoassays. the results can be expressed in concentration of the
free hormones. In the original applications, the
Labeled Hormone-Analog ( One-Step ) serum was used undiluted [82,83], but now serum
Immunoassays (Table 1, Method 5) samples are usually diluted for analysis [83-86].
Since their conception in the late 1970s [1,61-64],
the isotopic and recently the automated non- A method has recently been described which uses a
isotopic, one-step (analog) immunoassays became 96-sample well dialysis plate followed by liquid
the most common methods for free-thyroid- chromatographytandem mass spectrometry (LC-
hormone measurements. Despite their controversial MS/MS) [85]. This assay is used for routine
performance [1,2,31,65-78], they allowed the analysis of serum free thyroid hormones and has
widespread measurement of free thyroid hormones been described as a major step forward in the
in routine diagnostic laboratories. In their initial analysis of free thyroid hormones [86].
isotopic form, these competitive assays for free
thyroid hormone invariably required only a single Ultrafiltration (Table 1, Method 8)
incubation step, hence the name one-step. The FT4 and fT3 concentrations also can be measured
initial labeled analogs were small molecules, but
these compounds bound strongly to albumin. To by ultrafiltration of diluted or undiluted serum
avoid protein binding, the Siemens ADVIA samples, either in dialysis bags or special
ultrafiltration devices. As with equilibrium dialysis,
Centaur free thyroxine method uses a high
molecular weight IgG-T4 compound labeled with ultrafiltration may be direct or indirect. In direct
acridinium ester analog [78]. The Roche Elecsys or ultrafiltration, the free-thyroid-hormone
E170 assays are one-step assays which have a two- concentration is measured directly, usually by a
step incubation [30]. highly sensitive immunoassay, in the ultrafiltrate
[87-89]. In indirect ultrafiltration [8,12,13,90], the
test sample is enriched with trace amounts of
Labeled Antibody Method (Table 1, Method 6)
This method is also known as the solid-phase radiolabeled thyroid hormone prior to
antigen-linked technique (SPALT). In its original centrifugation, then the radioactivity is determined
isotopic version, the method is one step, with the in the resultant ultrafiltrate. Using 125I and 131I
sample to be tested simultaneously incubated with for the labeling of T4 and T3, respectively,
solid phasebound thyroid hormone analog and a simultaneous determination of fT4 and fT3 is also
limited amount of labeled antithyroid-hormone (T4 possible. Compared to equilibrium dialysis
593
methods, ultrafiltration methods are faster to consequences of testing at room temperature

perform. They are, however, also cumbersome and instead of the more appropriate 37C have been
difficult to run in large batches. Further, because of well documented recently for a two-step
the very high ratio between protein-bound and free immunoassay for fT4 [102,103].
thyroid hormones, even small protein leakage
through the ultrafiltration membrane [87,88,91,92] Incubation Time
readily invalidates the free hormone measurement. Obviously, all equilibrium assays require that the
incubation time be sufficiently long to reach
Centrifugal ultrafiltration followed by isotope equilibrium in the assay system. Nevertheless,
dilution MS/MS was first described by Soldin et al. historically there were several instances when the
[91]. This procedure was used for fT4 but lacked incubation times were found to be inadequate for
the sensitivity for fT3 determination. The technique fT4 immunoassays, and the manufacturers were
was modified [92] to improve sensitivity, allowing
eventually forced to correct this design deficiency.
simultaneous measurement of fT4 and fT3.
Witherspoon et al. [104] showed that similar
problems exist for isotopic T3U assays as well.
Gel Filtration/Gel Bead Dialysis (Table 1,
Method 9)
Polyacrylamide [93] and Sephadex [94-96] gel Incubation Milieu
filtration and gel-bead dialysis [97] have been also Different ions have been shown to have different
employed for the measurements of free thyroid effects on the binding of thyroid hormones to their
hormones. Although these methods are faster than specific proteins. For instance, both chloride and
equilibrium dialysis, they are also tedious and phosphate ions inhibit the binding of T4 to serum
cumbersome for large numbers of samples, and in proteins [105]. Barbital ions preferentially inhibit
their usual indirect version, they are subject to error the binding of T3 to TTR (transthyretin,
arising from radioactive impurities of the tracer. prealbumin) [106]. In addition, both endogenous
Besides, these methods may not always yield a true substances (e.g.,, various phenol compounds in
free-thyroid-hormone fraction but rather a fraction uremia, FFAs in diabetes) and drugs (e.g.,
that is only proportional to the free-thyroid- salicylates, furosemide) are known to specifically
hormone concentration [95]. inhibit thyroid-hormone binding to serum proteins
and to raise the apparent fT4 and fT3
Column Adsorption Chromatography (Table 1,
concentrations [106].
Method 10)
Although this method was once considered by
some investigators as a version of a resin uptake Hormone Sequestration
test, it is better described as a direct equilibrium The loss of thyroid hormone from the serum
method for the real measurement of free thyroid compartment (hormone sequestration) into
hormones [1,20]. In this method, Sephadex LH-20 another compartment is a general feature of all
chromatography is used to separate free from assays for free thyroid hormone [1,2], including
protein-bound hormone after equilibrium has been direct equilibrium dialysis with undiluted sera
reached on the column [84,98,99]. The protein- [1,50,107]. This hormone loss may be part of the
bound thyroid hormones (T4 or T3) emerge from assay design (e.g., sequestration in a dialysis cell or
by an antithyroid-hormone antibody) or may be
the column first and are discarded before elution of incidental to the assay (sequestration to extraneous
the free-thyroid-hormone (T4 or T3) fraction, sites such as adsorption of hormone to glass)
which is then measured. [1,101].
Critical Analytical Issues Serum Dilution

Issues that are particularly critical for valid free- Except for ultrafiltration of undiluted sera, all
thyroid-hormone measurements are discussed methods of free hormone measurements involve
below. For further details, see reports by Ekins sample dilution with buffer [1,2]. In accordance
[1,2,66,100], Nelson and Wilcox [101], Midgley with the mass action law, dilution of serum
[31], Stockigt [30] and Baloch [32]. containing protein-bound and free thyroid hormone
causes the free hormone concentration to fall, but
Incubation Temperature the large protein-bound pool of thyroid hormones
Because the temperature coefficients governing has a large reservoir capacity to sustain the free
thyroid-hormone binding to various transport hormone level in the diluent. For instance, a 100-
proteins differ (TBG being the most sensitive), the fold dilution causes only an approximately 2%
relative change in free hormone concentrations at depletion of the protein-bound T4 hormone pool
37C and at room temperature will vary between and a similar depression of the ambient fT4
standards and samples that contain different
concentration in the diluted sample [1]. Because
relative amounts of transport proteins. The
594
the fT3 concentration is higher than the fT4 changes in the fT4, fT3 or T-uptake values, but
concentration relative to the protein-bound, the storage at 4C is preferred if the test is not
dilution causes a somewhat more conspicuous drop performed within 24 hours. FT4 and FT3 in serum
in the fT3 concentration (5 to 20% decrease at a appear to be stable for 24 hours in whole blood
1:100 serum dilution) [1,2,108-113]. stored at temperatures between 15C and 35C
[112]. No change in estimated FT4 concentration
Reference and Preferred Method has been observed in samples stored at 4C to 8C
There are no reference methods for either fT4 or for either 5 days as whole blood or for 13 days as
fT3. However, both equilibrium dialysis (ED) and stored serum [113]. If the time between sample
ultrafiltration (UF) have been proposed as collection and analysis is longer than 6 days, the
separation techniques for a reference measurement sample should be stored frozen at 20C or below
procedure. Problems with UF include non- in non-cycling freezers. Frozen samples should be
equilibrium, absorption, protein leakage, and mixed well by inversion after thawing and before
temperature control. Problems with ED are dilution assay. Repeated freezing and thawing and mixing
and dialysis-buffer composition [108]. Free fatty by vortexing or vigorous agitation should be
acids and drugs which can displace thyroid avoided, since this may result in denaturation of
hormones from the binding proteins are potential proteins. Ideally, blood should be collected in the
confounders for the separation of free from bound total absence of heparin [114], and clinicians
thyroid hormone prior to definitive analysis [108]. should be made aware that thyroid function tests
are contraindicated in heparin-treated patients. The
A reference measurement system for free thyroxine popularity of non-isotopic assays for free-thyroid-
has been proposed which uses ED followed by a hormone measurement presents a special problem
trueness-based measurement procedure [109,110]. with respect to specimen acceptability.
An ED LC-MS/MS procedure has recently been Consequently, particular attention should be paid to
proposed which could be considered a candidate the manufacturers specific recommendations for
reference method for FT4 [111]. There is currently specimen collection when non-isotopic methods are
no proposed reference method for fT3, however the involved in testing. Finally, because the law of
new LC-MS/MS methods should facilitate mass action governs the free-thyroid-hormone
development. concentration, it is of great importance that
specimens with results above the assay range
According to the 2007 General Ligand Survey by should not be diluted for re-assay for any of the
the College of American Pathologists (CAP), all free-thyroid-hormone or hormone-binding tests but
participating laboratories used non-isotopic should be reported as such.
methods for fT4 and fT3 assays. Greater than 99%
of these assays have imprecision of less than 10% Interferences
and less than 8%, respectively, for fT4 and fT3, The use of hemolyzed or lipemic samples is not
which is satisfactory imprecision for routine recommended. Siliconized collection tubes have
clinical use. been claimed to produce artifactual results in some
solid-phase fT4 assays, but these observations could
There has been considerable discussion over many not be confirmed in later studies [24,115-127].
years on the validity of immunoassay estimates of Apart from heparin, most blood-collection-tube
free-thyroid-hormone concentration. A recent study additives have little if any effect on the
[85] demonstrated that two non-isotopic performance of isotopic assays. However, some of
immunoassays for FT4 correlated well with ED LC- the non-isotopic assays are profoundly affected by
MS/MS but had quite different results which could certain tube additives. For instance, some fT4
be attributed to standardization issues. Other methods are incompatible with the presence of
studies using LC-MS/MS have shown poor EDTA, citrate, or fluoride.
correlation with modern automated non-isotopic
immunoassay [91,92]. The development of Patients on heparin therapy may have increased
reference methods for free thyroid hormones will free fatty acids in stored serum samples as a result
allow for better standardization of these analytes of heparin-induced lipoprotein lipase activity.
with improved analytical accuracy. These free fatty acids displace T4 from TBG and
may cause an artifactual increase in the FT4
concentration [30,32].
Specimen
The preferred specimen is a serum sample that has Drugs such as furosemide, disalicylic acid,
been removed from the clot as rapidly as possible. phenytoin, and carbamazepine, which displace T4
Fresh serum samples may be stored at room from TBG, cause dilution artifacts in FT4 assays
temperature for up to 1 week without appreciable which dilute samples as part of the analytical
procedure. Underestimation of FT4 is observed, and
595
this is greatest in methods which have the least limits of the population-based reference interval
sample dilution [30,115]. [113,121].
Abnormal binding proteins can cause artifactual FT4 Concentration

results. Examples include autoantibodies, Reference Interval for (Nonpregnant) Adults:
rheumatoid factor, and familial dysalbuminemic 0.72.0 ng/dL (926 pmol/L)
hyperthyroxinemia [32]. Autoantibodies can cause (two-step and one-step immunoassays)
either false-positive or false-negative results, 0.82.3 ng/dL (1030 pmol/L)
depending on the assay design [116]. Autoantibody (indirect [tracer] equilibrium dialysis)
interference can be assessed by incubating patient 0.82.7 ng/dL (1035 pmol/L)
serum with radiolabeled thyroid hormone, followed (direct equilibrium dialysis)
by polyethylene glycol (PEG) precipitation of A trend for defining the normal range between 0.8
autoantibody-bound radiolabel [116-118]. It has and 1.9 ng/dL (10 to 25 pmol/L) is evident, but it is
recently been demonstrated [118] that samples also apparent that the reference intervals vary not
could be re-assayed for FT4 after PEG treatment in only with the type of methodology (see earlier
a patient with autoantibody interference. In general, discussion), but also among kits based on the same
analog ligand assays are the most susceptible to analytical concept. This reinforces the need to
thyroid hormone autoantibodies. It is generally establish, or at least verify, the reference interval
assumed that two-step assays are not as susceptible for the local population. Reference intervals for
because protein is removed prior to incubation with thyroid hormones have been determined for a
tracer. Recently the results of seven different number of non-isotopic immunoassay platforms
automated immunoassay systems were compared and MS/MS methods, including DPC IMMULITE
with ED/RIA in a patient with thyroid hormone [122]; ADVIA Centaur [123]; Abbott
autoantibodies [117]. The one-step assays ( DPC ARCHITECT [124,125]; Beckman Access [126];
IMMULITE 2500, Siemens ADVIA Centaur, Roche E170 [127]; UF/LC-MS/MS [92]; and
Tosoh AIA 1800, Roche Elecsys E170) all had ED/LC-MS/MS [85].
artifactually elevated results. The Wallac Delfia
two-step assay results were consistent with FT3 Concentration
ED/RIA, while the Abbott ARCHITECT and
Reference Range for (Nonpregnant) Adults:
Beckman DXi 800 had comparative results that
143-468 pg/dL ( 2.2-7.2 pmol/L)
were considered borderline low. One suggestion for
(one-step immunoassay)
the latter observation was incomplete washing
250-550 pg/dL ( 3.8-8.5 pmol/L)
away of patient sample prior to the addition of
(one-step immunoassay)
tracer in the two automated immunoassay systems.
260-480 pg/dL ( 4.0-7.4 pmol/L)
(indirect [tracer] equilibrium dialysis)
Nonspecific interference by ruthenium cross-
linking causing artifactual elevation of results has
Like fT4, the reference range for fT3 varies with
been observed in the Roche E170
electrochemiluminescent one-step assays for FT4 both the methodology and the actual assay kit.
and FT3 [119,120]. Therefore, verification or establishment of a
reference interval based on the local population is
Free Thyroxine and Free Triiodothyronine necessary.
Reference Intervals
As with most clinical diagnostic tests, commonly In a large study with analog fT3 RIA, results from
cited reference intervals for free thyroid hormones 41,655 successive patients were analyzed to obtain
and thyroid-hormone binding are based on the the distribution curve for the population [18].
study of healthy adult males and nonpregnant Using fT3 alone as a discriminator, 14% of the
females. For correct interpretation, it is important patients were classified as hyperthyroid when
to note, however, that a number of factors (e.g., applying a reference interval of 3 to 8 pmol/L.
age, gender, pregnancy) can alter the concentration When the borderline hyperthyroid range was
of free thyroid hormones under physiological defined as 7 to 9 pmol/L, 16.8% of the cases were
conditions. Therefore, a brief review of these included in this category, whereas when the
factors is also necessary [113]. It is important to borderline cases were defined as those results
recognize that FT4 and total T4 population-based falling between 8 and 9 pmol/L, 7.4% of the
reference intervals are wide when compared to the patients were included in this category.
narrow reference interval for individuals. Thus a
measurement of FT4 might be outside an T3 and T4 Uptake, Thyroid HormoneBinding
individuals own set point, indicative of an
abnormality, but the result could still be within the Ratio (THBR), and fT4 and fT3 Indices
596
Reference Range for (Nonpregnant) Adults: decreases to adult levels by the end of puberty), the
THBR: 0.831.17 (unitless) fT4I will also change with age. In general, higher
fT4I: 4.213.0 (unitless) values occur in children of all ages than in adults.
The Committee on Nomenclature of the American Gender

Thyroid Association (ATA) [51] strongly Whereas no significant gender-related differences
recommends expression of the T3U as a ratio were found in children [133], the differences
(THBR) and considers it mandatory when one is between males and females were statistically
calculating the fT4I or (possibly) fT3I. The significant between 16 and 49 years [129].
Although these differences were relatively small,
Committee also recommended that if the THBR is
the fT4 and fT3 concentrations and the T4/TBG
reported directly, it should always be reported with
a serum total T4 concentration so that the fT4I can ratios were lower in women during that period than
in men. The differences are apparently related to
be calculated. In other words, the T3U or T4U test
hormonal differences between males and women,
or, more appropriately, the THBR is useful only as with an estrogen effect in women during the years
a means of calculating the fT4I (or, by using total of fertility.
T3, the fT3I):
Race, Posture, Immobilization, Dietary
fT4I = THBR total T4 concentration Influences, Nutrition
At the present time, there is no evidence of an
appreciable effect of these variables on free thyroid
Therefore, if the reference range of the THBR is hormones [113,134].
0.83 to 1.17 and that of total T4 concentration is 45
to 115 g/L (or 58 to 148 nmol/L), the reference Pregnancy
range of the fT4I would be 37.4 to 134 g/L (or 48 Although the magnitude of changes varied from
to 173 nmol/L). The ATA Committee [57] 0% to 65% with widely different techniques, a
recommended omission of the units from the fT4I consistent decrease was observed with most
(or fT3I) to avoid confusion with measurements of methods in fT4 and fT3 values toward the final
T4 (or T3) itself. trimester of pregnancy [17,38,128,134-145]. The
decreased fT4 concentrations probably represent
Physiological Variations in Free Thyroid physiological adaptation of the mother to increased
Hormones requirements of the fetus for thyroid hormones
Age [143,145]. Recently, UF/LC-MS/MS has been used
There are considerable differences in fT4 and fT3 to determine reference intervals for pregnant
subjects [145]. In this study, the free thyroxine
concentrations among various age groups levels similarly decreased through the pregnancy.
[17,38,128-133]. The most dramatic changes occur
at birth. Within 24 to 28 hours after birth, the fT4 Lactation
concentration more than doubles, and the fT3 Up to 16% decreases have been reported in the
concentration increases threefold. Full-term infants serum fT4 concentrations of healthy lactating
have higher concentrations of both fT4 and fT3 women [146].
than preterm neonates [130]. Under physiological
conditions, the serum fT4 concentration is highest Menstrual Cycle
Although both serum TBG and total thyroid
during the immediate postnatal period then falls
hormone concentrations may fluctuate slightly with
rapidly during the first few weeks of life in full-
the estrogen effect during the normal menstrual
term infants. The adult concentration of fT4 is
cycle, no definitive data exist for a possible effect
established only at the beginning of the third on concentrations of free thyroid hormone [147].
decade of life. The age-related changes are
comparatively small in the THBR. The THBR is Diurnal Variations
lower in newborns (mean = 0.90) than adults (mean It is well established that serum thyrotropin levels
= 1.00). The THBR is the highest (mean value are markedly affected by the time of the day when
about 1.15) during the first few days of life, then the sample is collected for analysis (levels are
apart from a small decrease between the first and highest around midnight [during the so-called
fourth months, it is maintained at adult levels from acrophase] and lowest in the morning). This
the end of the first week of life. Since the serum pattern is opposite to the diurnal secretion of
total T concentration varies markedly with age cortisol and is likely caused by inhibition of
(peaks at a level about twice as high as the adult thyrotropin secretion. Cortisol also produces a
during the first few days of life, then gradually decrease in TBG and an increase in transthyretin
597
(TTR), with a likely fall in both total T4 and T3. as 8 weeks to occur [27,152,156]. Obviously,
Interestingly, compared to young adults, the elderly knowledge of this time-lag response is particularly
had a significantly diminished 24-hour circadian critical when adjusting thyroid hormone dosage.
rhythm assessed by either the amplitude or mesor Discordant thyroid function test results may also be
(24-hour mean value) for thyrotropin, but the fT4 related to the large difference between the half-
lives of T4 (1 week) and TSH (1 hour) and
mesor was not significantly different between the
older and younger groups (19.5 2.1 pmol/L versus sampling under non-standard conditions.
19.4 3.6 pmol/L). Although these observations are
intriguing, further studies are needed to verify the In terms of both diagnostic performance and cost,
findings and establish the possible underlying there is still an ongoing debate on the best approach
mechanism(s). A practical conclusion is that the to screen for, diagnose, and monitor thyroid
use of defined sampling times can improve the disease. Initially, TSH alone, fT4 alone, and TSH
diagnostic utility of free-thyroid-hormone and fT4 in combination, either simultaneously or
measurements. sequentially, were all proposed as a simplified
screening strategy [157-167]. There is now
Seasonal Changes argument that TSH alone could be potentially
The seasonal and annual fluctuations in the misleading in some patients [167]. Because of the
concentrations of free thyroid hormones are minor high prevalence of both thyroid disease and non-
[113]. thyroidal illness (NTI) in hospitalized patients
[168,169], newer schemes usually follow different
Physical Exercise approaches for healthy populations (e.g., newborns,
The fT4 concentrations are elevated in trained elderly subjects) [170-173] and ambulatory and
athletes and after prolonged physical exercise hospitalized patients [27,174,175]. Hence caution
[148]. This latter effect has been shown to be is advised in the interpretation of thyroid-function
related to the enhanced release of free fatty acids test results in hospitalized or acutely ill patients,
(FFA) and their competition with thyroid hormones since the results can be misleading. In addition to
for binding sites on transport proteins [148,149]. TSH and fT4, in selected cases, the measurement
of fT3 and other thyroid-related analyte may be
Interpretation also needed. Since the development of simple and
Based on the assessment of available diagnostic economic immunoassays for the measurement of
tests [51], current guidelines of the ATA [150,151] total T4, the THBR (formerly T3U or T4U) test by
recommend that the principal laboratory tests for
thyroid disease should be estimation of fT4 and itself is seldom used as a general screening test for
thyroid function, because the serum total T4
measurement of TSH with a sensitive assay. For
correct interpretation of test results, it is therefore concentrations show much less overlapping of
important to understand the relationship between values among euthyroid, hypothyroid, and
these two key laboratory tests of thyroid function. hyperthyroid states than the THBR values do.
In most patients with an intact hypothalamic- However, the THBR test can provide information
pituitary axis, there is a strong inverse log-linear regarding the status of serum binding proteins for
correlation between TSH and fT4 [27,115,152-155]. thyroid hormones, particularly if used in
The relationship is such that an approximately 100- combination with total T4 and the fT4I (or fT3I) is
fold change in serum TSH concentration is calculated [50,150,176-179]. In hyperthyroid
associated with only a twofold change in patients, serum total T4 concentrations are
circulating fT4 concentration. This indicates that if elevated. TBG is more saturated with T4, and thus
the hypothalamic-pituitary axis is normal, TSH is the THBR (T3U or T4U) is increased. On the other
generally a more sensitive indicator of thyroid hand, in hypothyroid patients, serum total T4
status than fT4 [115,121,153]. Subclinical thyroid
disease has been defined as when a patient has FT4 concentrations are abnormally low, TBG is less
within the population reference interval but an saturated with T4, and the THBR is reduced.
abnormal TSH concentration [154]. One Although the estimation of unoccupied binding
explanation for subclinical thyroid disease is the sites on T4-binding serum proteins (primarily
narrow intraindividual biological variation [155]. TBG) still may be of some value in assessing the
When the thyroid function spontaneously changes thyrometabolic status of the patient, the widespread
or the level of thyroid hormone therapy is altered, availability of relatively inexpensive, fast,
the pituitary TSH response may require several automated, non-isotopic assays for free thyroid
weeks for full expression. The time lag needed to hormones made this approach largely obsolete.
achieve full TSH response to changes in circulating TSH is considered the best screening test for
fT4 (and fT3) requires a minimum of 4 and as long hypothyroidism in ambulatory patients, and a
useful follow-up test for those with a confirmed
598
elevation of TSH is measurement of thyroid coefficient of variation) for FT4 ranging from
peroxidase antibodies [32]. approximately 3.8% to 10.8% at mean
concentrations ranging from 1.9 ng/dL to 2.7
Besides considerable physiological variations due ng/dL. In the case of FT3, the summary report
to age, pregnancy, and other conditions (see shows imprecision values ranging from
earlier), free thyroid hormone results can be approximately 3% to 9% at mean concentrations of
difficult to interpret because of patient overall 0.6 ng/dL to 1.0 ng/dL. The imprecision targets for
disease state, medications, and other less common FT4 and FT3 are 9.5% and 7.9%, respectively [32].
factors. Undoubtedly, the use of free hormone Thus most laboratories are attaining the desired
measurements during pregnancy and in the elderly imprecision target.
has gained prominence, but NTI, various drugs
such as heparin, binding-protein abnormalities, Of greater concern are the accuracy and validity of
human thyroid-hormone autoantibodies, and immunoassay methods. The development of LC-
heterophile, immunization-induced, or other human MS/MS and reference methods will aid in the
antibodies usually are the most challenging for the resolution of these issues. Thienpont [86] has
interpretation of free-thyroid-hormone results. A expressed concern at the potential proliferation of
number of strategies have been described for the free-thyroid-hormone methods using either UF or
identification or interpretation of discordant thyroid ED followed by LC-MS/MS procedures which
function tests [33,115,177-179]. have not been validated against a reference
measurement procedure. This is a critical point
Non-Thyroidal Illness because it should not be accepted that any
Non-thyroidal illness (NTI) remains a somewhat procedure using MS techniques automatically
ill-defined concept but introduces problems qualifies as an accepted method. It is likely that the
associated with the evaluation of true thyroid future routine measurement of free thyroid
status. NTI refers to the anomaly of patients who hormones lies with MS technology.
display abnormal results in their thyroid function
tests but do not actually appear to suffer directly References
from thyroid disease [180-187]. These include 1 Ekins R. Measurement of free hormones
those patients who are mildly to severely unwell or in blood. Endocrine Rev 1990;11:5-46.
stressed and those to whom certain 2 Ekins R. Analytical measurements of free
pharmacological agents have been administered. thyroxine. Clin Lab Med 1993;13:599-
Among the thyroid abnormalities, most prominent 630.
are low serum T3 (and often elevated reverse T3) 3 Mendel CM. The free hormone
concentrations, leading to the description low T3 hypothesis: a physiologically based
mathematical model [review]. Endocrine
syndrome [181]. In patients suffering from more
Rev 1989;10:232-274.
severe NTI, both serum T3 and T4 are depressed,
4 Mendel CM, Cavalieri RR, Weisiger RA.
and this condition was originally described as On plasma protein-mediated transport of
euthyroid sick syndrome and referred to by some as steroid and thyroid hormones [letter]. Am
low T4 syndrome. The fall in serum T4 may be J Physiol 1988;255(Endocrinol Metab
severe and possibly indicates a poor outcome 2):E221-227.
[185,188]. In mild cases of NTI, serum free and 5 Pittman CS, Chambers JB, Read VH. The
total T4 both tend to be higher than normal (high extrathyroidal conversion rate of thyroxine
T4 syndrome). Immunoassay FT4 results from to triiodothyronine in normal man. J Clin
Invest 1971;50:1187-1196.
different methods are quite variable in hospitalized
6 Sterling K, Hegedus A. Measurement of
patients, and it has been suggested that correct
free thyroxine concentration in human
interpretation of results should also take into
serum. J Clin Invest 1962;41:1031-1040.
account the assay method [188].
7 Oppenheimer JH, Squef R, Surks MI,
Hauer H. Binding of thyroxine by serum
NTI patients present two types of diagnostic
proteins evaluated by equilibrium dialysis
challenges. One is differentiation of abnormal
and electrophoresis techniques. Alterations
thyroid results caused by NTI from those caused by
in nonthyroidal illness. J Clin Invest
a treatable thyroid disorder. The other is the
1963;42:1769-1782.
treatment and monitoring of thyroid patients,
8 Oppenheimer JH, Surks MI.
particularly those with hypothyroidism whose
Determination of free thyroxine in human
disease is obscured by the presence of NTI.
serum: a theoretical and experimental
analysis. J Clin Endocrinol 1964;24:785-
Performance Goals 793.
Survey data from the 2007 CAP Participant
Summary Report show imprecision values (%
599
9 Surks MI, Oppenheimer JH. Postoperative Thyroid Hormones. Amsterdam: Excerpta

changes in the concentration of thyroxine- Medica; 1979:72-92.
binding prealbumin and serum free 22 Bunting J. Immunoassays for free
thyroxine. J Clin Endocrinol 1964;24:794- hormone measurement. In: Albertini A,
802. Ekins RP, eds. Free Hormones in Blood.
10 Ingbar SH, Braverman LE, Dawber NA, Amsterdam: Elsevier Biomedical;
Lee YG. A new method for measuring the 1982:139-149.
free thyroid hormone in human serum and 23 Ekins RP. The direct measurement of free
an analysis of the factors that influence its (non-protein bound) hormones in body
concentration. J Clin Invest 1965;44:1679- fluids. In: Hunter WM and Corrie JET,
1689. eds. Immunoassays for Clinical
11 Liewendahl K, Lamberg B-A. Free Chemistry. Edinburgh: Churchill
thyroxine in serum determined by dialysis Livingstone; 1983:319-337.
and Sephadex G-25 filtration. J Clin 24 Witherspoon LR, Schuler SE. Estimation
Endocrinol Metab 1965;25:991-993. of free thyroxine concentration: clinical
12 Sterling K, Brenner MA. Free thyroxine in methods and pitfalls. J Clin Immunoassay
human serum: simplified measurement 1984;7:192-205.
with the aid of magnesium precipitation. J 25 Wilke TJ. Estimation of free-thyroid-
Clin Invest 1966;45:153-163. hormone concentrations in the clinical
13 Schussler GC, Plager JE. Effect of laboratory. Clin Chem 1986;32:585-592.
preliminary purification of 131I thyroxine 26 White GH. Recent advances in routine
on the determination of free thyroxine in thyroid function testing. CRC Crit Rev
serum. J Clin Endocrinol 1967;27:242- Clin Lab Sci 1987;24:315-362.
250. 27 Spencer CA. Thyroid profiling for the
14 Volpert EM, Martinez M, Oppenheimer 1990s: free T4 estimate or sensitive TSH
JH. Radioiodinated impurities in measurement. J Clin Immunoassay
commercial preparation of 131I-thyroxine 1989;12:82-89.
and their effect on the measurement of 28 Liewendahl K. Assessment of thyroid
free thyroxine in human serum by status by laboratory methods:
equilibrium dialysis. J Clin Endocrinol developments and perspectives. Scand J
Metab 1967;27:421-428. Clin Lab Invest 19990;50(Suppl 20):83-
15 Yeo P. Clinical usefulness of serum free 92.
T4 and T3 measurements. Ligand Rev 29 Bayer MF. Effective laboratory evaluation
1968;3:36. of thyroid status. Med Clin North Am
16 Liewendahl K, Lamberg B-A. Free 1991;75:1-26.
thyroxine in serum and its use in clinical 30 Stockigt JR. Free thyroid hormone
diagnosis. Acta Endocrinol (Copenh) measurement. A critical appraisal.
1969;61:343-358. Endocrinol Metab Clin North Am
17 Robbins J, Rall JE. The iodine-containing 2001;30:265-89.
hormones. In: Gray CH, James VHT, eds. 31 Midgley JEM. Direct and indirect free
Hormones in Blood. 3rd ed. Vol 1. thyroxine assay methods: theory and
London: Academic Press; 1979:575-688. practice. Clin Chem 2001;47:1353-1363.
18 van der Walt A. Free thyroxine and free 32 Baloch Z, Carayon P, Conte-Devolx B,
triiodothyronine. In: Pesce AJ, Kaplan LA, Demers LM, Feldt-Rasmussen U, Henry
eds. Methods in Clinical Chemistry. St JF et al. Laboratory medicine practice
Louis: Mosby; 1987:277-291. guidelines. Laboratory support for the
19 Pearce CJ, Byfield PGH. Free-thyroid- diagnosis and monitoring of thyroid
hormone assays and thyroid function disease. Thyroid 2003;13:3-126.
[review]. Ann Clin Biochem 1986;23:230- 33 Dufour DR. Laboratory tests of thyroid
237. function: uses and limitations. Endocrinol
20 Ekins RP, Ellis SM. The Metab Clin N Am 2007;36:579-594.
radioimmunoassay of free thyroid 34 Halmolsky MW, Stein M, Freedberg AS.
hormones in serum. Robbins J, Braverman The thyroid hormone-plasma protein
LE, eds. Thyroid Research: Proceedings of complex in man. II. A new in vitro method
the Seventh International Thyroid for study of uptake of labeled hormonal
Conference (Boston). Amsterdam: components by human erythrocytes. J Clin
Excerpta Medica; 1975:597-600. Endocrinol Metab 1957;17:33-44.
21 Ekins RP. Methods for the measurement 35 Clark F, Horn DB. Assessment of thyroid
of free thyroid hormones. In: Ekins R, function by the combined use of the serum
Faglia G, Pennisi F, Pinchera A, eds. Free protein-bound iodine and resin uptake of
600
131I-triiodothyronine. J Clin Endocrinol 47 Horn K, Castieiras MJ, Ortol J, Kock R,

Metab 1965;25:39-45. Perriard FC, Bittner S et al. The
36 Welby ML, O'Halloran MW, Marshall JA. determination of thyroxine and thyroxine
A comparison of effective thyroxine ratio, uptake with new homogeneous enzyme
free thyroxine index and free thyroxine immunoassays using Boehringer
concentration in correcting for thyroxine- Mannheim/Hitachi analysis systems. Eur J
binding abnormalities in serum. J Clin Clin Chem Clin Biochem 1991;29:697-
Endocrinol Metab 1974;3:63. 703.
37 Lee ND, Pileggi VJ. Measurement of 48 Soldin SJ, Morales A, Albalos F, Lenherr
free thyroxine in serum. Clin Chem S, Rifai N. Pediatric reference ranges on
1971;17:166-173. the Abbott IMx for FSH, LH, prolactin,
38 Robbins J, Rall JE. The iodine-containing TSH, T4, T3, free T4, free T3, T-uptake,
hormones. In: Gray CH, James VHT, eds. IgE, and ferritin. Clin Biochem
Hormones in Blood. 3rd ed. Vol 4. 1995;28:603-606.
London: Academic Press; 1983:219-265. 49 Faix JD, Rosen HN, Velazquez FR.
39 Prince HP, Ramsden DB. A new Indirect estimation of thyroid hormone-
theoretical description of the binding of binding proteins to calculate free
thyroid hormones by serum proteins. Clin thyroxine index: comparison of
Endocrinol 1977;7:307-324. nonisotopic methods that use labeled
40 Robbins J, Johnson ML. Theoretical thyroxine (T uptake). Clin Chem
considerations in the transport of the 1995;41:41-47.
thyroid hormones in blood. In: Ekins R, 50 Nelson JC, Tomei RT. Dependence of the
Faglia G, Pennisi F, Pinchera A, eds. Free thyroxin/thyroxin-binding globulin (TBG)
Thyroid Hormones. Amsterdam: Excerpta ratio and the free thyroxin index on TBG
Medica; 1979:1-16. concentrations. Clin Chem 1989;35:541-
41 Hamada S, Nakagawa T, Mori T, Torizuka 544.
K. Re-evaluation of thyroxine binding and 51 Larsen PR, Alexander NM, Chopra IJ,
free thyroxine in human serum by paper Hay ID, Hershman JM, Kaplan MM et al.
electrophoresis and equilibrium dialysis, Committee on Nomenclature of the
and a new free thyroxine index. J Clin American Thyroid Association. Revised
Endocrinol Metab 1970;31:166-179. nomenclature for tests of thyroid
42 Fresco G, Curti G, Biggi A, Fontana B. hormones and thyroid-related proteins in
Comparison of calculated and measured serum. J Clin Endocrinol Metab
free thyroid hormones in serum in health 1987;64:1089-1094. (Reprinted in: Clin
and in abnormal states. Clin Chem Chem 1987;33:2214-2219.)
1982;20:1325-1329. 52 Tuttlebee JW, Bird R. A comparison of
43 Hamolsky MW, Golodetz A, Freedberg free thyroxine concentration and the free
AS. The plasma protein-thyroid complex thyroxine index as diagnostic tests of
in man. III. Further studies on the use of thyroid function. Ann Clin Biochem
the in vitro red blood cell uptake of 131I-I- 1981;18:88-92.
triiodothyronine as a diagnostic test of 53 Csako G, Zweig MH, Ruddel M,
thyroid function. J Clin Endocrinol Metab Glickman J, Kestner J. Direct and indirect
1959;19:103-116. techniques for free thyroxin compared in
44 Stein EA, Chen IW, Gau N, Kaplan L, patients with nonthyroidal illness. III.
Maxon H, Sperling M. Spectrophotometric Analysis of interference variables by
assays for serum thyroxine (T4) and stepwise regression. Clin Chem
unsaturated T4 binding globulin for 1990;36:645-650.
evaluation of thyroid function [abstract]. 54 Liewendahl K, Helenius T. Comparison of
Clin Chem 1981;27:1070. serum free thyroxine indices and
45 Bellet N, Winfrey L, Horton A, Syed A, corrected thyroxine tests. Clin Chim
Khanna P, Colvin W. Development of Acta 1975;64:263-272.
homogeneous fluorescence rate 55 Wilke TJ. Free thyroid hormone index,
immunoassays for thyroid function testing thyroid hormone/thyroxin-binding
[abstract]. Clin Chem 1981;27:1071. globulin ratio, triiodothyronine uptake,
46 King ME. The performance characteristics and thyroxin-binding globulin compared
of the thyroxine (T4 E) and thyronine (TU for diagnostic value regarding thyroid
E) methods for the Du Pont aca discrete function. Clin Chem 1983;29:74-79.
clinical analyzer. Technical Bulletin. 56 Konno N, Nakazato T, Hagiwara K,
Wilmington, DE: Du Pont Co; 1990. Taguchi H. A comparison of
measurements of serum free T3
601
concentration by equilibrium dialysis, free 69 Bayer MF. Free-thyroxine results are

T3 index, and T3: TBG ratio in thyroidal affected by albumin concentration and
and non-thyroidal illnesses. Acta nonthyroidal illness. Clin Chim Acta
Endocrinol 1983;103:501-508. 1983;130:391-396.
57 Liewendahl K, Tikanoja S, Helenius T. 70 Byfield PGH, Lalloz MRA, Pearce CJ,
Reference methods for serum free thyroid Himsworth RL. Free thyroid hormone
hormones. Scand J Clin Lab Invest concentrations in subjects with various
1995;55(Suppl 223):55. abnormalities of binding proteins:
58 Jiang N, Tue KA. Determination of free experience with Ameriex free T4 and free
thyroxine in serum by radioimmunoassay. T3 assays. Clin Endocrinol 1983;19:277-
Clin Chem 1977;23:1679-1683. 283.
59 Ekins RP, Filetti S, Kurtz AB, Dwyer K. 71 Stockigt JR, Stevens V, White EL, Barlow
A simple general method for the assay of JW. Unbound analog
free hormones (and drugs): its application radioimmunoassays for free thyroxin
to the measurement of serum-free measure the albumin-bound hormone
thyroxine levels and the bearing of assay fraction. Clin Chem 1983;29:1408-1410.
results on the free thyroxine concept. J 72 Bayer MF, McDougall IR. Clinical
Endocrinol 1980;85:29p-30p. assessment of a radioimmunoassay for
60 Rajan MGR, Samuel AM. A two-step free thyroxine using a modified tracer. J
radioimmunoassay for free Nucl Med 1984;25:402-405.
triiodothyronine in serum. Clin Chem 73 Ekins R, Edwards P, Jackson T, Geiseler
1987;33:372-376. D. Interpretation of labeled analog free
61 Wilkins TA, Midgley JEM. New methods hormone assay. Clin Chem 1984;30:491-
of free-thyroid-hormone assay. In: 493.
Bizollon CA, ed. Physiological Peptides 74 Wilkins T, Midgley JEM, Barron N.
and New Trends in Radioimmunology. Comprehensive study of a thyroxin-
Amsterdam: Elsevier; 1981:215. analog-based assay for free thyroxin
62 Wilkins TA, Midgley JEM, Whitworth (Amerlex fT4). Clin Chem
AS. Novel single-tube equilibrium solid- 1985;31:1644-1653.
phase free-T4 RIA. Ligand Q 1981;4:45. 75 Wenzel KW. Thyroxine analogue assay
63 Wilkins TA, Midgley JEM, Giles AF. kits in clinical medicine. Lancet
Theoretical basis, computer simulation, 1985;2:781-782.
optimization and technical validation of a 76 Alexander NM. Free thyroxin in serum:
new direct free ligand assay principle with labeled thyroxin-analog methods fall short
particular reference to the measurement of of their mark [editorial]. Clin Chem
free thyroxine. In: Radioimmunoassay and 1986;32:417-418.
Related Procedures in Medicine. Vienna: 77 Csako G, Zweig MH, Benson C, Ruddel
I.A.E.A.; 1982:221. M. On the albumin dependence of
64 Midgley JEM, Wilkins TA. A method for measurements of free thyroxin. ll. Patients
determining the free portions of with nonthyroidal illness. Clin Chem
substances in biological fluids. European 1987;33:87-92.
Pat. No. 0026103. 1985. 78 Sapin R, Schlienger J, Gasser F, Noel E,
65 Geiseler D, Ritter M. On the validity of Lioure B, Grunenberger F et al.
free hormone measurements. Anal Intermethod discordant free thyroxine
Biochem 1983;132:174-182. measurements in bone marrow-
66 Ekins R, Midgley JEM, Moon CR, transplanted patients. Clin Chem
Wilkins TA. Validity of analog free 2000;46:418-422.
thyroxin immunoassays. Part II and 79 Weeke J, Orskov H. Ultrasensitive
Responses. Clin Chem 1987;33:2137- radioimmunoassay for direct
2152. determination of free triiodothyronine
67 Midgley JEM, Ekins R. Continuation of concentration in serum. Scand J Clin Lab
misrepresentations of analogue free Invest 1975;35:237-244.
hormone assays. Ann Clin Biochem 80 Petersen BA, Hanson RN, Giese RW,
1990;27:388-391. Karger BL. Picogram analysis of free
68 Stockigt JR, de Garis M, Csicsmann J, triiodothyronine and free thyroxine
Barlow JW, White EL, Hurley DM. hormones in serum by equilibrium dialysis
Limitations of a new free thyroxine assay and electron capture gas chromatography.
(Amerlex free T4). Clin Endocrinol J Chromatogr 1976;126:503.
1981;15:313-318. 81 Petersen BA, Giese RW, Larsen PR,
Karger BL. Measurement of free thyroid
602
hormone in serum by dialysis and gas polyacrylamide gel filtration. Clin Chem
chromatography. Clin Chem 1978;24:652-656.
1977;23:1389-1396. 94 Irvine CHG. measurement of free
82 Christensen LK. A method for the thyroxine in human serum by a Sephadex
determination of free, non-protein bound binding method. J Clin Endocrinol Metab
thyroxine in serum. Scand J Clin Invest 1974;38:655-662.
1959;11:326-331. 95 Levinson S, Rieder SV. Parameters
83 Ross HA. A dialysis rate method for affecting a rapid method in which
measurement of free iodothyronine and Sephadex is used to determine the
steroid hormones in blood. Experientia percentage of free thyroxine in serum.
1978;34:538-539. Clin Chem 1974;20:1568-1572.
84 Ross HA, Visser JW, der Kinderen PJ, 96 Finucane JF, Griffiths RS. A rapid and
Tertoolen JF, Thijssen JH. A comparative simple method for simultaneous
study of free thyroxine estimations. Ann measurement of serum free thyroxine and
Clin Biochem 1982;19:108-113. triiodothyronine fractions. J Clin Pathol
85 Yue B, Rockwood AL, Sandrock T, 1976;29:949-954.
Laulu SL, Kushnir MM, Meikle AW. 97 Forest JC, Masse J, Lavoie R. Dosage de
Free thyroid hormones in serum by direct la thyroxine libre par adsorption sur gel de
equilibrium dialysis and on-line solid- Sephadex. Ann Biol Clin 1989;47:317-
phase extraction-liquid 322.
chromatography/tandem mass 98 Romelli PB, Pennisi F, Vancheri, L.
spectrometry. Clin Chem 2008;54:642- Measurement of free thyroid hormones in
651. serum by column adsorption
86 Thienpont LM. A major step forward in chromatography and radioimmunoassay. J
the routine measurement of serum free Endocrinol Invest 1979;2:25-40.
thyroid hormones. Clin Chem 99 Joseph LJ, Desai KB, Velumani A,
2008;54:625-626. Samuel AM. Determination of free
87 Weeke J, Boye N, Orskov H. thyroxine in serum using column
Ultrafiltration method for direct chromatography & radioimmunoassay.
radioimmunoassay measurement of free Indian J Med Res [B] 1991;94:36-40.
thyroxine and free triiodothyronine in 100 Ekins R. The free hormone hypothesis and
serum. Scand J Clin Lab Invest measurement of free hormones. Clin
1986;4:381-389. Chem 1992;38:1289-1293.
88 Tikanoja S. Ultrafiltration devices tested 101 Nelson JC, Wilcox RB. Analytical
for use in a free thyroxine assay validated performance of free and total thyroxine
by comparison with equilibrium dialysis. assays. Clin Chem 1996;42:146-154.
Scand J Clin Lab Invest 1990;50:663-669. 102 Ross HA, Benraad TJ. Is free thyroxine
89 Tikanoja SH, Liewendahl BK. New accurately measurable at room
ultrafiltration method for free thyroxin temperature? Clin Chem 1992;38:880-886.
compared with equilibrium dialysis in 103 Van der Sluijs Veer G, Verme I, Bonte
patients with thyroid dysfunction and HA, et al. Temperature effects on free-
nonthyroidal illness. Clin Chem thyroxine measurements: analytical and
1990;36:800-804. clinical consequences. Clin Chem
90 Thorson SC, Wilkins GE, Schaffrin M, 1992;38:1327-1331.
Morrison RT, McIntosh HW. Estimation 104 Witherspoon LR, Shuler SE, Garcia MD.
of serum free thyroxin concentration by The triiodothyronine uptake test: an
ultrafiltration. J Clin Lab Med assessment of methods. Clin Chem
1972;80:145-154. 1981;27:1272-1276.
91 Soldin SJ, Soukhova N, Janicic N, 105 Spaulding SW, Gregerman RI. Free
Jonklaas J, Soldin OP. The measurement thyroxine in serum by equilibrium
of free thyroxine by isotope dilution dialysis: effects of dilution, specific ions
tandem mass spectrometry. Clin Chim and inhibitors of binding. J Clin
Acta 2005;358:113-118. Endocrinol Metab 1972;34:974-82.
92 Gu J, Soldin OP, Soldin SJ. Simultaneous 106 Ingbar SH. Observations concerning the
quantification of free triiodothyronine and binding of thyroid hormones by human
free thyroxine by isotope dilution tandem serum prealbumin. J Clin Invest
mass spectrometry. Clin Biochem 1963;42:143-160.
2007;40:1386-1391. 107 Ekins R. Direct determination of free
93 McDonald LJ, Robin NI, Siegel L. Free thyroxin in undiluted serum by
thyroxine in serum as estimated by
603
equilibrium dialysis/radioimmunoassay electrochemiluminescent immunoassay.

[letter]. Clin Chem 1989;35:510-512. Internal Medicine 2007;46:1225-1229.
108 Holm SS, Hansen SH, Faber J, Staun- 120 Cranfield LM, Sullivan M, Tillbrook L.
Olsen P. Reference methods for the An unusual case of multi-parameter
measurement of free thyroid hormones in interference on Roche E170 immunoassay
blood: evaluation of potential reference system. Ann Clin Biochem 2008;45(Suppl
methods for free thyroxine. Clin Biochem 1):47.
2004;37:85-93. 121 Wu AHB. Quality specifications in thyroid
109 Thienpont LM, Beastall G, Christofides diseases. Clin Chim Acta 2004;346:73-77.
ND, Faix JD, Ieiri T, Jarrige V et al. 122 Ermlinger MW, Khnel W, Lambrecht H-
Proposal of a candidate international G, Ranke MB. Reference intervals from
conventional reference measurement birth to adulthood for serum thyroxine
procedure for free thyroxine in serum. (T4), triiodothyronine (T3), free T3, free T4,
Clin Chem Lab Med 2007;45;934-936. thyroxine-binding globulin (TBG) and
110 Thienpont LM, Beastall G, Christofides thyrotropin (TSH). Clin Chem Lab Med
ND, Faix JD, Ieiri T, Miller WG et al. 2001;39:973-979.
Measurement of free thyroxine in 123 Hbner U, Englisch C, Werkmann H, Butz
laboratory medicine: proposal of H, Georg T, Zabtansky S, Herrmann W.
measurand definition. Clin Chem Lab Med Continuous age-dependent reference
2007;45;563-564. ranges for thyroid hormones in neonates,
111 van Uytfanghe K, Stckl D, Ross HA, infants, children and adolescents
Thienpont LM. Use of frozen sera for fT4 established using the ADVIA Centaur
standardization: investigation by Analyzer. Clin Chem Lab Med
equilibrium dialysis combined with 2002;40:1040-1047.
isotope dilution mass spectrometry and 124 Hubl W, Schmieder J, Gladrow E and
immunoassay. Clin Chem 2006;52:1817- Demant T. Reference intervals for thyroid
1820. hormones on the ARCHITECT Analyzer.
112 Tanner M, Kent N, Smith B, Fletcher S, Clin Chem Lab Med 2002;40:165-166.
Lewer M. Stability of common 125 Gonzalez-Sagrado M, Martin-Gi FJ.
biochemical analytes in serum gel tubes Population-specific Reference values for
subjected to various storage temperatures thyroid hormones on the Abbott
and times pre-centrifugation. Ann Clin ARCHITECT i2000 analyzer. Clin Chem
Biochem 2008;45:375-379. Lab Med 2004;42:540-542.
113 Keffer JH. Preanalytical considerations in 126 dHerbomez M, Jarrige V, Darte C.
testing thyroid function. Clin Chem Reference intervals for serum thyrotropin
1996;42:125-134. (TSH) and free thyroxine (fT4) in adults
114 Mardell R, Gamlen TR. Discrepancies using the Access Immunoassay System.
results for fT4 by RIA and dialysis Clin Chem Lab Med 2005;43:102-105.
procedures explained. Clin Chem 127 Kratzsch J, Fiedler GM, Leichtle A,
1987;28:1989. Brugel M, Buchbinder S, Otto L et al.
115 Stockigt J. Assessment of thyroid New reference intervals for thyrotropin
function: towards an integrated laboratory and thyroid hormones based on National
clinical approach. Clin Biochem Rev Academy of Clinical Biochemistry criteria
2003;24:109-123. and regular ultrasonography of the thyroid.
116 Despres N, Grant AM. Antibody Clin Chem 2005;51:1480-1486.
interference in thyroid assays: a potential 128 Roti E, Gardini E, Minelli R, Bianconi L,
for clinical misinformation. Clin Chem Flisi M. Thyroid function evaluation by
1998;44:440-454. different commercially available free-
117 Zouwail SA, OToole AM, Clark PMS, thyroid-hormone measurement kits in term
Begley JP. Influence of thyroid hormone pregnant women and their newborns. J
autoantibodies on 7 thyroid hormone Endocrinol Invest 1991;14:1-9.
assays. Clin Chem 2008;54:927-928. 129 Franklyn JA, Ramsden DB, Sheppard MC:
118 van der Watt G, Haarburger D, Berman P. The influence of age and sex on tests of
Euthyroid patient with elevated serum free thyroid function. Ann Clin Biochem
thyroxine. Clin Chem 2008;54;1239-1247. 1985;22:502-505.
119 Ando T, Yasui J, Inokuchi N, Usa T, 130 John R, Bamforth FJ. Serum free
Ashizawa K, Kamihara S,Eguchi K. Non- thyroxine and free triiodothyronine
specific activities against ruthenium concentrations in healthy full term,
crosslinker as a new cause of assay preterm and sick preterm neonates. Ann
interference in an Clin Biochem 1987;24:461-465.
604
131 Harkavy KL, Enecio CE. Free thyroxine technical artifact. Clin Chem
levels in hospitalized newborns: depressed 1989;35:1891-1896.
levels in critical, nonthyroidal illness. J 144 Amino N, Nishi K, Nakatani K, Mizuta H,
Perinatol 1991;11:117-121. Ichihara K, Tanizawa O, Miyai K. Effect
132 Nelson JC, Clark SJ, Borut DL, Tomei of albumin concentration on the assay of
RT, Carlton EI. Age-related changes in serum free thyroxin by equilibrium
serum free thyroxine during childhood and radioimmunoassay with labeled thyroxin
adolescence. J Pediatr 1993;123:899-905. analog. Clin Chem 1983;29:321-325.
133 Wiedemann G, Jonetz-Mentzel L, Panse 145 Kahric-Janicic N, Soldin SJ, Soldin OP,
R. Establishment of reference ranges for West T, Gu J, Jonklaas J. Tandem mass
thyrotropin, triiodothyronine, thyroxine spectrometry improves the accuracy of
and free thyroxine in neonates, infants, free thyroxine measurements during
children and adolescents. Eur J Clin Chem pregnancy. Thyroid 2007;17:303-311.
Clin Biochem 1993;31:277-288. 146 Iwatani Y, Amino N, Tanizawa O, Mori
134 Fisher DA. Physiological variations in H, Kawashima M, Yabu Y, Miyai K.
thyroid hormones: physiological and Decreases of free thyroxin in serum of
pathophysiological considerations. Clin lactating women. Clin Chem
Chem 1996;42:135-139. 1987;33:1217-1279.
135 Tuttlebee JW. Further experience with free 147 Burrow GN, Fisher DA, Larsen PR.
thyroxine assays with particular reference Maternal and fetal thyroid function. N
to pregnancy. Ann Clin Biochem Engl J Med 1994;331:1072-1078.
1982;19:374-378. 148 Liewendahl K, Helenius T, Naveri H,
136 Franklyn JA, Sheppard MC, Ramsden DB, Tikkanen H. Fatty acid-induced increase
Hoffenberg R. Free triiodothyronine and in serum dialyzable free thyroxine after
free thyroxine in sera of pregnant women physical exercise: implication for
and subjects with congenitally increased nonthyroidal illness. J Clin Endocrinol
or decreased thyroxin-binding globulin. Metab 1992;74:1361-1365.
Clin Chem 1983;29:1527-1530. 149 Mendel C, Frost PH, Cavalieri RR. Effect
137 Ashwell K, Hopton MR, Harrop JS. of free fatty acids on the concentration of
Serum free thyroxine concentrations in free thyroxine in human serum: the role of
subjects with high circulating levels of albumin. J Clin Endocrinol Metab
thyroxine-binding globulin. Ann Clin 1986;63:1394-1399.
Biochem 1983;20:285-288. 150 Hay ID, Bayer MF, Kaplan MM, Klee
138 Wilke TJ. Diagnostic value of three GG, Larsen PR, Spencer CA. Committee
methods for assessing free thyroxine in on Nomenclature of the American Thyroid
pregnancy. Ann Clin Biochem Association. American Thyroid
1983;20:60-61. Association assessment of current free
139 Ramsden DB, Sheppard MC, Sawers RS, thyroid hormone and thyrotropin
Smith SC, Hoffenberg R. Serum free measurements and guidelines for future
thyroxine concentrations in normal clinical assays. Clin Chem 1991;37:2002-
euthyroid subjects and ones with high 2008.
serum thyroxine binding globulin 151 Surks MI, Chopra IJ, Mariash CN,
concentration. Clin Chim Acta Nicoloff JT, Solomon DH. American
1983;130:211-217. Thyroid Association guidelines for use of
140 Meinhold H, Wenzel KW. Measurement laboratory tests in thyroid disorders. J Am
of serum fT4 in pregnancy and binding Med Assoc 1990;263:1529-1532.
protein abnormalities by different 152 Nicoloff JT. Guide to Thyroid Function
methods. Nuc-Compact 1985;6:365-366. Testing for Laboratorians and Physicians.
141 Gow SM, Kellett HA, Seth J, Sweeting Abbott Park, IL: Abbott Diagnostic
VM, Toft AD, Beckett GJ. Limitations of Educational Services; 1991.
new thyroid function tests in pregnancy. 153 Stockigt JR. Case finding and screening
Clin Chim Acta 1985;152:325-333. strategies for thyroid dysfunction. Clin
142 Price A, Griffiths H, Morris BW. A Chim Acta 2002;315:111-124.
longitudinal study of thyroid function in 154 Surks MI, Ortiz E, Daniels GH, Sawin CT,
pregnancy. Clin Chem 1989;35:275-278. Col NF, Cobin RH et al. Subclinical
143 Ball R, Freedman DB, Holmes JC, thyroid disease: scientific review and
Midgley JE, Sheehan CP. Low-normal guidelines for diagnosis and management.
concentrations of free thyroxin in serum in JAMA 2004;291:228-38.
late pregnancy: physiological fact, not 155 Andersen S, Pedersen KM, Bruun NH,
Laurberg P. Narrow individual variations
605
in serum T4 and T3 in normal subjects: a 169 DeGroot LJ, Mayor G. Admission

clue to the understanding of subclinical screening by thyroid function tests in an
thyroid disease. J Clin Endocrinol Metab acute general care teaching hospital. Am J
2002;87:1068-1072. Med 1992;93:558-564.
156 Trundle DS, Spencer CA. Clinical 170 Joseph R, Aw TC, Tan KL. Free thyroxine
Chemistry News Laboratory Guide to as a supplement to thyrotropin in cord
Thyroid Testing [reference poster]. Clin screening for hypothyroidism. Ann Acad
Chem News Suppl. Washington, DC: Med (Singapore) 1993;22:549-552.
AACC; 1989. 171 Nystrom E, Petersen K, Lindstedt G,
157 Hopton MR, Harrop JS. Lundberg PA. Screening for thyroid
Immunoradiometric assay of thyrotropin disease in women >50 years of age
as a first-line thyroid function test in the seeking hospital care: influence of
routine laboratory. Clin Chem common nonthyroidal illness on serum
1986;32:691-3. free thyroxin as determined by analog
158 Penney MD, OSullivan DJ. Total or free radioimmunoassay. Clin Chem
thyroxin as a primary test of thyroid 1986;32:603-608.
function. Clin Chem 1987;33:170-171. 172 Simons RJ. Thyroid function in the
159 John R, Henley R, Lloyd G, Elder GH. elderly. J Clin Immunoassay 1989;12:100-
Evaluation of a new strategy for detection 105.
of thyroid dysfunction in the routine 173 Lewis GF, Alessi CA, Imperial JG,
laboratory. Clin Chem 1988;34:1110-114. Retetoff S. Low serum free thyroxine
160 Nuutila P, Irjala K, Viikari J, Prinssi VP, index in ambulating elderly is due to a
Kaihola HL. Comparative evaluation of resetting of the threshold of thyrotropin
serum thyroxine, free thyroxine and feedback suppression. J Clin Endocrinol
thyrotropin determinations in screening of Metab 1991;73:843-849.
thyroid function. Ann Clin Res 174 Spencer C Eigen A, Shen D, Duda M,
1988;20:158-163. Qualls S, Weiss S, Nicoloff JJ. Specificity
161 Beaman JM, Woodhead JS. A simplified of sensitive assays of thyrotropin (TSH)
strategy for testing thyroid function. Clin used to screen for thyroid disease in
Chem 1989;35:828-830. hospitalized patients. Clin Chem
162 Hullin DA, Pugh R, McGrane MTG, 1987;33:1391-1396.
Davis SJ. Simultaneous free thyroxine and 175 Hickman PE, Yu S, Price L, Ward G.
thyrotropin measurement as a strategy for Thyroid stimulating hormone secretion is
thyroid function testing. J Clin a dynamic process. Ann Clin Biochem
Immunoassay 1989;12:106-110. 2001;38:147-148.
163 Helfand M, Crapo LM. Screening for 176 Csako G, Zweig MH, Ruddel M,
thyroid disease. Ann Intern Med Glickman J, Kestner J. Direct and indirect
1990;112:840-849. techniques for free thyroxin compared in
164 Kaptein EM. Clinical application of free patients with nonthyroidal illness. III.
thyroxine determinations. Clinics Lab Analysis of interference variables by
Med 1993;13:653-672. stepwise regression. Clin Chem
165 Taimela E, Koskinen P, Nuutila P, 1990;36:645-650.
Nikkanen V, Saraste M, Taimela S, Irjala 177 Ward G, McKinnon L, Badrick T,
K. Free thyroid hormones and a third Hickman PE. Heterophilic antibodies
generation TSH assay in the detection of remain a problem for the immunoassay
hyperthyroidism during long-term laboratory. Am J Clin Pathol
thyroxine treatment in thyroid carcinoma 1997;108:417-421.
patients. Scand J Clin Lab Invest 178 Dayan CM. Interpretation of thyroid
1995;55:181 -186. function tests. Lancet 2001;357:619-624.
166 Braverman LE. Evaluation of thyroid 179 Topliss DJ, Eastman CJ. Diagnosis and
status in patients with thyrotoxicosis. Clin management of hyperthyroidism and
Chem 1996;42:174-178. hypothyroidism. Med J Aust
167 Beckett GJ, Toft AD. First-line thyroid 2004;180:186-193.
function tests: TSH alone is not enough. 180 Stockigt JR. Guidelines for diagnosis and
Clin Endocrinol 2003;58:20-21. monitoring of thyroid disease:
168 Vermaak WJH, Kalk W, Zakolski WJ. nonthyroidal illness. Clin Chem
Frequency of euthyroid sick syndrome as 1996;42:188-192.
assessed by free thyroxine index and a 181 Wartofsy L, Burman KD. Alterations in
direct free thyroxine assay. Lancet thyroid function in patients with systemic
1983;1:1373-1375.
606
illness: the euthyroid sick syndrome. altered thyroid hormone indices to

Endocr Rev 1982;3:164-217. survival in nonthyroidal illnesses. Clin
182 Kaplan MN, Hamburger JI. Nonthyroidal Endocrinol 1982;16:565-574.
causes of abnormal thyroid function test 186 Fliers E, Alkemade A, Wiersinga WM.
data. J Immunoassay 1989;12:90-99. The hypothalamic-pituitary-thyroid axis in
183 Cavalieri RR. The effects of nonthyroid critical illness. Best Pract Res Clin
disease and drugs on thyroid function Endocrinol Metab 2001;15:453-464.
tests. Med Clin North Am 1991;75:27-39. 187 Langton JE, Brent GA. Nonthyroidal
184 Docter R, Krenning EP, de Jong M, illness syndrome: evaluation of thyroid
Hennemann G. The sick euthyroid function in sick patients. Endocrinol
syndrome: changes in thyroid hormone Metab Clin N Am 2002;31:159-172.
serum parameters and hormone 188 Sapin R, dHerbomez M. Free thyroxine
metabolism. Clin Endocrinol measured by equilibrium dialysis and nine
1993;39:499-518. immunoassays in sera with various serum
185 Kaptein EM, Weiner JM, Robinson WJ, thyroxine-binding capacities. Clin Chem
Wheeler WS, Nicoloff JT. Relationship of 2003;49:1531-1535.
607
Tables
Table 1. Free Thyroxine and Free Triiodothyronine Estimate Methods
Method 1: Index methods using indirect (tracer) equilibrium dialysis
Principle of analysis: Semipermeable membrane used to separate free thyroid hormone, including
tracer amount of iodinated, from protein-bound thyroid hormone. Percent free determined by
proportion of tracer in dialysate. Levels of free thyroid hormone determined using proportion of
measured total T4 of dialysate.
Comments: Historical research tool. Unwieldy. Purification of iodinated thyroid hormone critical.
Method 2: Index methods using TBG measurement
Principle of analysis: Calculation of free thyroid hormone based on law of mass action using
measurements of total T4 and TBG
Comments: Number of derived models, requires accurate T4 and TBG. Historical, not in routine use.
Method 3: Index methods using free thyroid hormone indices
Principle of analysis: Calculation of free thyroid hormone index (fTI) based on product of THBR and
total T4
Comments: No unit, indirect estimate of free hormone level
Method 4: Two-step, labeled hormone immunoassay
Principle of analysis: First step is reaction of patient sample with solid-phase antibody. Solid phase is
washed to remove patient serum prior to addition of tracer dose of labeled hormone.
Comments: Binding of tracer to serum proteins is avoided, which should eliminate protein-based
interferences.
Method 5: Labeled hormone analog (one-step) immunoassay
Principle of analysis: Immunoassay using analog tracer which does not bind to thyroid binding
proteins or disturb equilibrium between free and protein-bound thyroid hormone
Comments: Early, low-molecular analog binds to albumin; these methods are not recommended.
Large-molecular-weight analogs do not bind albumin but have autoantibody interference.
Method 6: Labeled antibody immunoassay
Principle of analysis: Tracer amount of labeled antibody competes with free thyroid hormone and
solid phasebound thyroid hormone analog.
Comments: Autoantibodies interference
Method 7: Equilibrium dialysis
Principle of analysis: Semipermeable membrane used to separate free from protein-bound thyroid
hormone. Free hormone level in dialysate determined by either immunoassay or LC-MS/MS.
Comments: Initially cumbersome, but recent developments with dialysis plate technology make
method amenable for routine use.
Method 8: Ultrafiltration
Principle of analysis: Centrifugation using semipermeable membrane used to separate free from
protein-bound thyroid hormone. Free hormone level in filtrate determined by either immunoassay or
LC-MS/MS.
Comments: Ultrafiltration cumbersome, not amenable to large batches
Method 9: Gel filtration/gel-bead dialysis
Principle of analysis: Separation based on molecular sieving prior to assay
Comments: Historical
Method 10: Column adsorption chromatography
Principle of analysis: Separation based on molecular weight
Comments: Historical
608
609
Gammaglutamyl Transferase (GGT)

Danyal B. Syed
Name: Gammaglutamyl transferase, GGT, GGT1, GGT2, GGT3, -glutamyl

transpeptidase, GGTP, -GTP, a-glutamyl transpeptidase,
L--glutamyltransferase
Systematic name: (5L-glutamyl)-peptide:amino-acid 5-glutamyltransferase
Clinical significance: click here <page 5, line 21>
Molecular mass: 90,000 D to 110,000 D (gel filtration)
Number of subunits: Dimer
Chemical class: Enzyme, protein, glycoprotein
Known isoenzymes: Multiple forms are known, but their identity is still in question.
Chromosomal location: 22q11.1-11.2; 22[1]; 22q11.23 [1,2]
-Glutamyl-L-peptide + amino acid EC 2.3.2.2 -Glutamyl amino acid + peptide
NRSCL Reference Method: NCCLS RS17
Principles of Analysis and Current Usage (ammediol) buffer, pH 8.6, because enzyme activity is
not affected over a broad range of buffer concentrations,
-Glutamyltransferase (GGT) catalyzes the transfer of a and the donor substrate has a higher solubility in this
-glutamyl group to an amino acid or peptide (external buffer [4]. The glutamyl-group acceptor in this selected
transpeptidation), to another substrate molecule (internal method is glycylglycine, 50 mmol/L. Although other
transpeptidation) or to water (hydrolysis). The enzyme is substances, such as methionine, glutamine, and cystine,
involved in the following metabolic pathways: (1) can act as acceptor substrates, glycylglycine is the
arachidonic acid metabolism, (2) cyanoamino acid acceptor of choice [6].
metabolism, (3) glutathione metabolism, (4)
selenoamino acid metabolism, and (5) taurine and The temperature used in the selected method is 25C or
hypotaurine metabolism. Because of its involvement in 30C, with the reaction followed at 405 nm where the
the glutathione cycle, glutathione was used as the donor hydrolysis product, 4-nitroaniline, absorbs, but the
substrate in the earlier methods of GGT activity substrate does not [4]. The day-to-day imprecision of
measurement. Substrate disappearance or product this method in the manual mode measured as the
formation was detected by manometric, coefficient of variation (CV) is 5.5% at a level of 23.6
chromatographic, or ultraviolet techniques. U/L for 22 days.
Modern methods use either (-L-glutamyl)-4-nitroanilide A modification of the Szasz procedure was also
or (-L-glutamyl)-3-carboxy-4-nitroanilide as substrates published by the Committee on Enzymes of the
because these compounds allow the direct measurement Scandinavian Society for Clinical Chemistry and
of reaction rate without deproteinization or any chemical Clinical Physiology (Table 1, Method 2) [7]. The (-L-
treatment of the cleavage products (Table 1, Methods 1 glutamyl)-4-nitroanilide concentration for this method is
and 3). Orlowski and Meister introduced the use of (-L- 4 mmol/L. The buffer used is Tris-HCl, 100 mmol/L, pH
glutamyl)-4-nitroanilide in 1963 [3]. Szasz used this 7.6. MgCl2 is added to a concentration of 10 mmol/L to
substrate (4.0 mmol/L) in a kinetic method [4] that was increase the solubility of the substrate. Glycylglycine is
published as a selected method by the American used at 75 mmol/L, with the reaction carried out at 37C.
Association for Clinical Chemistry [5]. Szasz preferred
100 mmol/L of 2-amino-2-methyl-propane-1,3-diol Fluorometric methods for GGT analysis have also been
reported [8,9] (Table 1, Method 4). Imprecision is
i
similar to that of spectrophotometric GGT analyses, but
Gammaglutamyl Transferase sensitivity is 20 times greater. This allows the use of
Previous and current authors of this method: smaller sample volumes and shorter incubation times.
First edition: Steven Gendler These substrates have not come into common usage,
Methods edition: Steven Gendler, Amadeo J. Pesce probably because of the equipment requirements.
Third edition: Not updated Table 2 compares the various methods for GGT
Fourth edition: Steven Gendler, Amadeo J. Pesce determination.
Fifth edition: Danyal B. Syed
610
8.5, glycylglycine 129 to 250 mmol/L, and LGCN 6.6

Reference and Preferred Methods to 10.2 mmol/L, given a serum-to-reagent ratio of 1:11,
The reference method for GGT utilizes -L-glutamyl-3- 100 mmol/L Tris-HCl buffer, at a temperature of 30C.
carboxy-4-nitroanilide (LGCN) as substrate [10]. Most However, the latest recommendation of IFCC is to
of the problems encountered in GGT catalytic activity measure the enzymatic activity at 37C. Response
measurement include: surface methodology has shown that maximal activity is
1. The purity of the reagents; for example, some achieved using the given buffer, temperature, and
investigators report the presence of an impurity when fraction of sample volume with a pH of 8.16, a
the LGCN is synthesized by certain manufacturers glycylglycine concentration of 190 mmol/L, and a
[11]. This requires a pyridine extraction to remove the substrate concentration of 8.4 mmol/L. These conditions
impurity, which acts as a competitive inhibitor [12]. are proposed in the recommended method that follows
2. Because of the low solubility of LGCN, it is [16]. The IFCC Primary Reference Procedure utilizes pH
necessary to use a supersaturated solution to achieve of 7.70, the glycylglycine concentration of 150 mmol/L,
a maximum reaction rate. This supersaturated and a substrate (LGCN) concentration of 6 mmol/L
solution is stable for only a few hours [7]. [10].
3. Optimization of the assay is difficult because of the
inhibitory effect of excess donor or acceptor The absorbance of the hydrolysis product, 2-nitro-5-
substrates. This double inhibition is caused by the aminobenzoic acid, is followed at 410 nm. This is a
kinetic mechanism of GGT [13]. Donor substrate slightly longer wavelength than is used for the non-
inhibition with both 4-nitroanilide compounds is pH- carboxylated substrate, since the carboxy derivative has
dependent: maximal at acid pH, slight at pH 8.5, and a higher absorbance than the non-carboxylated substrate
absent at more alkaline pH [14]. Thus assay [17]. The longer wavelength reduces the blank
conditions must strike a balance that takes into absorbance.
account maximal substrate inhibition, which is both
pH and substrate-concentration dependent, and In order to overcome the concerns mentioned above, the
maximum reaction rate, which is also both pH and IFCC primary method has well-defined measurement
substrate-concentration dependent. conditions. Besides, there is an international effort
requiring the manufacturers of in-vitro diagnostic
The selected methods (Table 1, Methods 1 and 2) do not medical devices to meet the metrological traceability of
use the currently IFCC-preferred substrate, LGCN. This values assigned to the calibrators by using primary
chemical is the preferred substrate because it is much reference procedures and reference materials [18]. The
more soluble than (-L-glutamyl)-4-nitroanilide. standard ISO 15189 requires that when possible, the
Spontaneous hydrolysis of the carboxy derivative is not results should also be traceable to SI units. To meet this,
detectable. Further, kinetic constants and serum GGT an International Standard on traceability was put
activities with the new substrate are comparable to those forward, describing the order of measurement
of the old substrate. However, controls supplemented procedures and calibration materials [19]. Another
with beef kidney GGT are on average 19% lower with related standard, ISO 18153, deals with the traceability
the carboxylated substrate. Thus the (-L-glutamyl)-4- of the enzyme catalytic concentrations [20]. The IFCC
nitroanilide has been more frequently used, since higher Committee on Reference Systems for Enzymes (C-RSE)
GGT activation and greater sensitivities are achieved has published the recommendations for establishing a
with this substrate. The use of the carboxy derivative is global reference system for enzymes. The system
increasing, however. comprises primary reference procedures to measure the
catalytic enzyme concentrations, a network of well-
Several studies were done to optimize the GGT reaction established laboratories and primary calibrators (certified
by use of the carboxylated substrate [14-16]. Given the enzyme reference materials [ERMs]) [21]. For GGT, the
complicated kinetics of the enzyme, it is not surprising certified ERM, AD-452 /IFCC 452, is available for use
that several optimized methods have been proposed. as primary standard/calibrator [22]. Commercial
Some inhibition of GGT by LGCN was reported. The calibrators which can be traced to this primary reference
extent of the reported inhibition varied with the material are also available [23,24]. Some instrument and
investigator and the instrument used. Significant kit manufacturers use a calculation factor to provide
inhibition by the carboxylated donor substrate was traceability to the IFCC GGT reference measurement
reported with a concentration as low as 3.5 mmol/L [17], procedure.
but most data showed no inhibition until a concentration
of 10 to 12 mmol/L is achieved [14]. Specimen
Serum or plasma (lithium heparin or Na heparin) is the
In the past, equations that modeled the GGT assay were recommended specimen for the determination of the
developed. Equations derived using kinetic theory [15] catalytic concentration of GGT [5,7]. Either specimen
and equations derived using a computerized optimization can be collected by employing normal phlebotomy
scheme (called response surface methodology) have procedures. The plasma or serum should be separated
been shown to accurately model the enzyme assay over a from the contact with cells within 2 to 4 hours of
broad range of GGT concentrations. Response surface specimen collection.
methodology has shown that maximal GGT activity is These specimens give the same results. EDTA and
achieved within a range of assay conditions: pH 7.8 to citrate do not interfere with analysis, and serum GGT
611
activity is stable at room temperature or 2C to 8C for mass and the hydrophilic properties are similar to those
at least 7 days and for at least 2 months when frozen at of papain-solubilized liver GGT. Serum GGT shows an
or below 18C [25]. When refrigerated or frozen, the electrophoretic mobility and lectin-affinity reaction
specimen must first be brought to room temperature and identical to those of the liver enzyme but different from
then mixed well prior to analysis. those of GGT from the kidney, urine, and pancreas.
However, an antibody raised against liver GGT has the
Interferences same reactivity against liver GGT as it does against
The presence of hemolysis and prolonged contact with kidney and pancreatic GGT. The data are consistent with
cells does not influence GGT values, according to one (1) the isoenzymes having similar structures but
review [25]. However, some methods require that differing primarily in their carbohydrate content, similar
hemolyzed specimens with hemoglobin concentration of to the alkaline phosphatase isoenzymes, and (2) serum
150 mg/dL should be avoided, whereas others have set GGT originating from the hepatobiliary system [38].
this limit to 500 mg/dL or higher. Some investigators Larger forms of GGT are reported with molecular
have shown fluoride, oxalate, and citrate at a masses greater than 200,000 D. This varies according to
concentration of 1 g/L to inhibit GGT by approximately the detergent used for solubilization and molecular mass
15%. The presence of lipemia or icterus has been noted determination, since the enzyme is known to aggregate
to cause decreased activity with some methods but only with lipoproteins and membrane fragments. Given the
above 2000 mg/dL of triglycerides and 40 mg/dL of total confusion in GGT characterization, it is prudent to wait
bilirubin, respectively. Divalent ions like calcium and for more evidence before any conclusions are drawn
magnesium and monovalent ions like sodium and [39].
potassium have no effect on the catalytic activity of
GGT. It has been proposed that GGT participates in the
transport of amino acids into cells. This accounts for its
GGT Reference Interval membrane-bound structure, and because transport into
It is recommended that each laboratory establish its own the renal proximal tubules is important to conserve
reference interval based on methodology, temperature of amino acids, this concept fits its high activity in this
GGT activity measurement and population. The organ. The enzyme is universally distributed in tissue but
reference interval is sex and age dependent [25]. with considerable variance in content per gram of tissue,
Activities in women are typically less than those in men as shown in Table 5 [40,41]. The problem of whether
of the same age. Individuals who have increased body this reaction and enzyme is the major enzyme pathway
mass index (BMI) by 30% or more typically show higher of amino acid transport into the cell is not completely
enzyme activities [26,27]. The within-day variation has settled [42].
been reported to be between 10% and 15%, which is
effectively independent of age and gender [28-30]. The enzyme appears in serum after chronic, excessive
Americans of African origin have been shown to exhibit ethanol ingestion and drug intake, as well as in the
GGT activity almost twice that of the corresponding presence of hepatobiliary disease. The reference interval
European-origin population [31]. Smoking a pack of population was selected to eliminate these factors, since
cigarettes a day will increase the GGT activity by 10% it is well known that ethanol ingestion increases serum
[27,32], and alcohol consumption has a direct GGT. The enzyme by itself has poor specificity for the
relationship with increased GGT activity [27,32,33]. The differential diagnosis of liver disease. For example,
values presented in Table 3 are from a selected when 1040 adult hospital patients were studied, 139
population that excluded persons who consumed more were found to have increased serum GGT levels [43]. Of
than 1 L/day of wine or beer and women on oral the 139 cases, only 45 had primary hepatobiliary disease.
contraceptives. GGT activities are much higher during Twenty-one percent of the elevated GGT values
the first months of life as compared with older children, occurred in patients with malignancy and 16% in
as seen in Table 4. Oral contraceptives resulted in a 20% patients with cardiac disease. Thus elevations of the
increase in GGT activity, and a large increase in GGT enzyme are not sufficiently specific to render it useful as
activity was noted for children (400%) and adults the sole criterion for establishing hepatobiliary disease.
(200%) on anticonvulsant drug (phenytoin and
phenobarbital) therapy. Other commonly used drugs GGT levels were reported to be increased an average of
associated with increased GGT activity include 12-fold in obstructive liver disease when compared to
carbamazepine, cimetidine, furosemide, methotrexate, ALP, which increased only 3-fold, so GGT is slightly
heparin, isotretinoin, and valproic acid [34]. Full-term more sensitive than ALP in this regard. GGT activity
neonates exhibit GGT activity levels six to seven times increases in cholestasis via the same mechanism as does
the upper limit of the adult reference range; values ALP [44-46]. It has been reported that increased GGT
decline and approach adult levels by the age of 5 to 7 activity level in children may be a reliable index of bile
months [35]. duct damage and that the serum GGT may provide a
useful indicator in separating the two forms of idiopathic
Interpretation cholestasis, with or without a bile duct involvement [47].
Gamma-glutamyltransferase is an enzyme present It has been noted that in infants diagnosed with biliary
primarily in the kidney, pancreas, liver, and prostate atresia and managed surgically, the GGT levels stay high
[36]. GGT in serum has a molecular mass of 90,000 D, in the blood if the infant is breastfed. This is due to the
as measured by electrophoresis [37]. This molecular
612
high level of GGT in human breast milk for at least 4 <http://www.ncbi.nlm.nih.gov/sites/entrez?db=

weeks postpartum [48]. gene&cmd=retrieve&dopt=default&rn=1&li>
2 Collins JE, Mungall AJ, Badcock KL, Fay JM,
In three comprehensive European cohort studies Dunham I. The organization of the -Glutamyl
involved in EUROSTROKE, the association of GGT as transferase genes and other low copy repeats in
a marker for alcohol consumption related to fatal and human chromosome 22q11. Genome Research
nonfatal hemorrhagic and ischemic stroke was evaluated. 1997;7:522-31.
The results showed that an increased GGT, as a marker 3 Orlowski M, Meister A. -Glutamyl-p-
of alcohol consumption, is associated with increased risk nitroanilide: a new convenient substrate for
of stroke, especially hemorrhagic stroke [49]. determination and study of L- and D--glutamyl
transpeptidase activities. Biochim Biophys Acta
GGT activity measurement has been found to be useful 1963;73:679-81.
in differentiating the source of ALP activity increase, 4 Szasz, G. A kinetic photometric method for
whether it is of bone or liver origin, because GGT serum -glutamyltranspeptidase. Clin Chem
activity is increased only in cases of hepatic injury, 1969;15:124-36.
whereas ALP is increased when either liver, bone, or 5 Szasz, G. Reaction-rate method for -
both are involved. It can also be used for similar glutamyltransferase activity in serum. Clin.
differentiation among Duchenne muscular dystrophy Chem. 1976;22:2051-55.
patients who are treated with potentially hepatotoxic 6 Shaw LM, London JW, Petersen LE. Isolation
drugs [50]. of -glutamyltransferase from human liver and
comparison with the enzyme from human
In a perspective study, the association between GGT and kidney. Clin Chem1978;24:905-15.
the risk for chronic kidney disease (CKD) in nondiabetic 7 Stromme JH. Chairman of the Committee on
and nonhypertensive Korean workers was evaluated. Enzymes of the Scandinavian Society for
From the findings of this study, it was concluded that Clinical Chemistry and Clinical Physiology:
GGT may be an early predictor of CKD independent of Recommended method for the determination of
other well-known baseline factors [51].
-glutamyltransferase in blood. Scand J Clin
Lab Invest 1976;36:119-25.
Gammaglutamyl Transferase Performance Goals
8 MacQueen JD, Driscoll RC, Gargiulo RJ. New
The short-term variation of GGT in healthy subjects has
substrate for fluorometric determination of -
been shown to be approximately 15% (within-subject)
glutamyltransferase activity in blood serum.
[52]. Performance goals for GGT are based on biological
Clin Chem 1977;23:879-881.
variation and analytical precision. The total error, based
9 Prusak E, Siewinski M, Szewczuk A. A new
on biological variation and analytical imprecision and
bias, are all approximately 20% to 22% [53,54]. The fluorometric method for the determination of -
National Academy of Clinical Biochemistry (NACB) glutamyltransferase activity in blood serum,
recommends that (1) assays for GGT should have a total Clin Chim Acta 1980;107:21-26.
error of 20% at the upper reference limit, (2) fasting 10 Schumann G, Bonora R, Ceriotti F, Ferard G,
blood sample is recommended, (3) single upper Ferrero CA, Franck P FH et al. IFCC primary
reference limit is appropriate for adult men, but for reference procedures for the measurements of
children and adult women, separate reference limits catalytic activity concentrations of enzymes at
based on age are needed, and (4) GGT activity 37C: part 6 reference procedure for the
measurement should be reserved for specific indications measurement of catalytic concentration of -
to determine the source of high ALP activity level [54]. glutamyltransferase. Clin Chem Lab Med
GGT is not a regulated analyte according to the Clinical 2002;40:734-38.
Laboratory Improvement Amendments (CLIA) 11 Rowe JA, Tarlow D, Rosalki SB. Impurity in -
(493.931, 42 CFR Ch. IV PP1022-23[10-1-2004 glutamyl-p-nitroanilide used as a substrate for
edition]) rules and regulations. The College of American D-glutamyl transferase [letter]. Clin
Pathologists (CAP) acceptability criterion for Chem1973;19:435-36.
proficiency testing is a peer-group mean of 3 SD. With 12 Huseby NE, Stromme JH. Practical points
increasing use of the IFCC primary reference procedure regarding routine determination of -glutamyl
and the international reference material, as well as the transferase (-GT) to serum with a kinetic
efforts by instrument/kit manufacturers to trace their method at 37C. Scand J Clin Lab
calibration to the primary reference method, the Invest1974;34:357-63.
analytical intralaboratory variation has been reduced to 13 Stromme JH, Theodorsen L. -Glutamyl-
less than 3%, while interlaboratory variation is now less transferase: substrate inhibition, kinetic
than 5% (CAP Participants Summary C-B, mechanism, and assay conditions. Clin Chem
Chemistry/Therapeutic Drug monitoring, Surveys 2007). 1976;22:417-21.
14 Schiele F, Artur Y, Bagrel D, Bagrel C, Siest G.
References Measurement of plasma gamma
1 Entrez Gene: GGT1 gamma- glutamyltransferase in clinical chemistry:
glutamyltransferase 1 [Homo sapiens] Gene Id kinetic basis and standardization propositions.
2678. Available at Clin Chim Acta 1981;112:187-95.
613
15 Shaw LM, London JW, Fetterolf D, Garfinkel 26 Salaspuro M. Characteristics of laboratory

D. -Glutamyltransferase: kinetic properties and markers in alcohol-related organ damage. Scand
assay conditions when -glutamyl-4-nitroanilide J Gastroenterol 1989;24:769-80.
and its 3-carboxy derivative are used as donor 27 Skurtveit S, Tverdal A. Sex differences in
substrates. Clin Chem 1977;23:79-85. gamma-glutamyltransferase in people aged 40-
16 London JW, Shaw LM, Theodorsen L, 42 years in two Norwegian counties. Drug
Stromme JH. Application of response surface Alcohol Depend 2002;67:95-98.
methodology to the assay of gamma- 28 Fraser CG. Biological variation in clinical
glutamyltransferase. Clin Chem 1982;28:1140- chemistry. An update: collated data, 1988-1991.
3. Arch Pathol Lab Med 1992;116:916-23.
17 Theodorsen L, Stromme JH. -Glutamyl-3- 29 Holzel WGE. Intra-individual variation in
carboxy-4-nitroanilide: the substrate of choice serum from patients with chronic liver diseases.
for routine determinations of -glutamyl- Clin Chem 1987;33:1133-36.
transferase activity in serum? Clin Chim Acta 30 Fraser CG, Cummings ST, Wilkinson SP,
1976;72:205-10. Neville RG, Knox JD, Ho O, MacWalter RS.
18 International Organization for Standardization. Biological variability of 26 clinical chemistry
Medical Laboratories: Particular Requirements analytes in elderly people. Clin Chem
for Quality and Competence. ISO 15189. 1989;35:783-86.
Geneva: ISO; 2003. 31 Manolio TA, Burke GL, Savage PJ, Jacobs DR
19 International Organization for Standardization. Jr, Sidney S, Wagenknecht LE et al. Sex- and
In-Vitro Diagnostic Medical Devices. race-related differences in liver-associated
Measurement of Quantities in Biological serum chemistry tests in young adults in the
Samples: Metrological Traceability of Values CARDIA study. Clin Chem 1992;38:1853-59.
Assigned to Calibrators and Control Materials. 32 Nilssen O, Helge-Forde O, Brenn T. The
ISO17511. Geneva: ISO; 2003. Tromso Study: distribution and population
20 International Organization for Standardization. determinants of -glutamyltransferase. Am J
In-Vitro Diagnostic Medical Devices. Epidemiol 1990;132:318-26.
Measurement of Quantities in Biological 33 Moussavian SN, Becker RC, Piepmeyer JL,
Samples: Metrological Traceability of Values Mezey E, Bozian RC. Serum -glutamyl
for Catalytic Concentration of Enzymes transpeptidase and chronic alcoholism-influence
Assigned to Calibrators and Control Materials. of alcohol ingestion and liver disease. Dig Dis
ISO 18153, Geneva: ISO; 2003. Sci 1985;30:211-4.
21 Siekmann L, Bonora R, Burtis CA, Ceriotti F, 34 Young DS. Effect of Drugs on Clinical
Clerc-Renaud P, Frard G et al. IFCC primary Laboratory Tests. 5th ed. Washington DC:
reference procedures for the measurement of AACC Press; 2000:2200.
catalytic activity concentrations of enzymes at 35 Cabrera-Abreu JC, Green A. Gamma-
37C. Part 1. The concept of reference glutamyltransferase: value of its measurement
procedures for the measurement of catalytic in paediatrics. Ann Clin Biochem 2002;39:22-5.
activity concentrations of enzymes. Clin Chem 36 Rosalki SB. Gamma-glutamyl transferase. Adv
Lab Med 2002;40:631-634. Clin Chem 1975;17:53-10.
22 Linsinger T, Kristiansen N, Schimmel H, 37 Tsuji A, Matsuda Y, Katunuma N.
Pauwels J, Siekmann L, Ceriotti F et al. Characterization of human serum -
Certification Report: Catalytic Concentration of glutamyltransferase. Clin Chim Acta
-Glutamyltransferase (GGT) Determined by 1980;104:361-6.
IFCC Method at 37 C. Certified Reference 38 Huseby NE. Separation and characterization of
Material ERM-AD452/IFCC. Report EUR human -glutamyltransferases. Clin Chim Acta
19577 EN. Luxembourg: Office for Official 1981;111:39-45.
Publications of the European Committees; 39 Henderson AR. Clinical enzymology. Annu
2005. Rev Clin Biochem 1980;1:233-70.
23 Canalias F, Camprubi S, Snchez M, Gella FJ. 40 Goldberg JA, Friedman EM, Pineda EP, Smith
Metrological traceability of values for catalytic EE, Chatterji R, Stein EM, Ruttenburg AM. The
concentration of enzymes assigned to a colorimetric determination of -glutamyl
calibration material. Clin Chem Lab Med transpeptidase with a synthetic substrate. Arch
2006;44:333-9. Biochem Biophys 1960;91:61-70.
24 Infusino I, Bonora R, Panteghini M. 41 Orlowski M, Szewczuk A. Colorimetric
Traceability in clinical enzymology. Clin determination of -glutamyl transpeptidase
Biochem Rev 2007;28:155-161. activity in human serum and tissues with
25 Ladenson JH. Nonanalytical sources of synthetic substrate. Acta Biochemica Polonica
variation in clinical chemistry results. In: 1961;8:189-200.
Sonnenwirth AC, Jarett L, eds. Gradwohls 42 Goldberg DM. Structural, functional and
Clinical Laboratory Methods and Diagnosis. clinical aspects of -glutamyl transferase. Crit
Vol. 1. 8th ed. St Louis: Mosby; 1980. Rev Clin Lab Sci 1980;121:1-58.
614
43 Burrows S, Feldman W, McBride F. Serum 50 Rosales XQ, Chu ML, Shilling C, Wall C,
gamma-glutamyl transpeptidase: evaluation in Pastores GM, Mendell JR. Fidelity of gamma-
screening of hospitalized patients. Am J Clin glutamyltransferase (GGT) in differentiating
Pathol 1975;64:311-314. skeletal muscle from liver damage. J Child
44 Anciaux ML, Pelletier AG, Attali P, Meduri B, Neurol 2008;23(7):748-51.
Liguory C, Etienne JP. Prospective study of 51 Ryu S, Chang Y, Kim D-I, Kim WS, Suh B-S.
clinical and biochemical features of -Glutamyltransferase as a predictor of chronic
symptomatic choledocholithiasis. Dig Dis Sci kidney disease in nonhypertensive and
1986;31:449-53. nondiabetic Korean men. Clin Chem
45 Sheeman M, Maythorn P. Predictive value of - 2007;53:71-7.
glutamyl transpeptidase in various diseases. In: 52 Ricos C, Arumi EG, Rubio RR, Schwartz S.
Goldberg DM, Werner M, eds. Progress in Eficia de un programa interno de control de
Clinical Enzymology. New York: Masson; calidad. QuimClin 1986;5:159-65.
1979:184-7. 53 Rics C, Alvarez V, Cava F, Garcia-Lario JV,
46 Ratanasavanh D, Tazi A, Gaspart E. Hepatic - Hernandez A, Jimnez CV et al. Current
glutamyl transpeptidase release: effect of the databases on biologic variation: pros, cons, and
bile salt and membrane structure modification. progress. Scand J Clin Lab Invest 1999;59:491-
Adv Biochem Pharmacol 1982;3:93-103. 500.
47 Maggiore G, Bernard O, Hadchouel M, 54 Dufour RD, Lott JA, Nolte FS, Gretch DR,
Lemonnier A, Alagille D. Diagnostic value of Koff RS, Seefe LB. Diagnosis and monitoring
serum gamma-glutamyl transpeptidase activity of hepatic injury. 1. Performance characteristics
in children. J Pediatr Gastroenterol Nutr of laboratory tests. Clin Chem 2000;46:2027-
1991;12:21-6. 2049.
48 Colagiovanni DB, Meyer DJ, Wolf J, Hart K, 55 Moss D, Henderson R. Clinical enzymology. In:
Abbott E, Santos C. High -glutamyltransferase Burtis CA, Ashwood ER, eds. Tietz Textbook
(GGT) activity in human breast milk confounds of Clinical Chemistry. 3rd ed. Philadelphia:
interpretation of high serum GGT activity in a Saunders; 1999:686-689.
nursing infant with liver disease. Clin Chem 56 Schiele F, Guilmin AM, Detienne H, Siest G.
2005;51:1750-51 Gamma-glutamyltransferase activity in plasma:
49 Bots ML, Salonen JT, Elwood PC, Nikitin Y, statistical distributions, individual variations,
Freire de Concalves A et al. Gamma- and reference intervals. Clin Chem
glutamyltransferase and risk of stroke: the 1977;23:1023-8.
EUROSTROKE project. J Epidemiol
Community Health 2002;56(Suppl 1):i25-29.
Tables
Table 1: Methods for -Glutamyltransferase (GGT) Determination
Method 1: (Szasz [5]) Reaction-rate method for -glutamyltransferase activity in serum; quantitative, kinetic
Principle of analysis: (-L-Glutamyl)-4-nitroanilide 2-nitroaniline (405 nm)
Comments: Most frequently used; selected method of American Association for Clinical Chemistry; donor-
substrate at supersaturating concentration
Method 2: (Scandinavian Society for Clinical Chemistry [SSCC] [7]) Modified Szasz; quantitative, kinetic
Principle of analysis: As for Method 1
Comments: Similar to Method 1; different pH, buffer, and substrate concentrations
Method 3: Substrate, modified SSCC [14]; quantitative, kinetic

Principle of analysis: (-L-Glutamyl)-3-carboxy-4-nitroanilide 2-nitro-5-aminobenzoic acid (410 nm)
Comments: Also the principle of analysis for the IFCC Primary Reference Method*; recommended method
because of more soluble substrate donor and well-defined assay conditions
Method 4: Fluorometric; quantitative [8,9]

Comments: Not in current use; most rapid, sensitive technique; requires specialized instrumentation
615
Table 2: Comparison of Methods for GGT Determination
Reaction condition: Temperature SSCC-Szasz: 0.0909

AACC-Szasz*: 25C or 30C Recommended method: 0.0909
SSCC-Szasz: 37C IFCC Primary Reference Method: 0.0909
Recommended method: 30C (1:11)
IFCC Primary Reference Method: 37.0C
0.1C Reaction condition: Wavelength
Reaction condition: Final concentration of AACC-Szasz*: 405 nm
reagents SSCC-Szasz: 405 nm
a. Substrate donor Recommended method: 410 nm
b. Substrate acceptor IFCC Primary Reference Method: 410 nm 1
c. Buffer nm
d. pH
e. Other Reaction condition: Interferences
AACC-Szasz*: AACC-Szasz*: Gross hemolysis
a. (-L-Glutamyl)-4-nitroanilide: 4 SSCC-Szasz: Gross hemolysis
mmol/L Recommended method: Gross hemolysis
b. Glycylglycine: 50 mmol/L IFCC Primary Reference Method: Gross
c. 2-Amino-2-methyl-propane-1,3-diol: hemolysis
100 mmol/L
d. 8.6 Reaction condition: Precision (at upper limit of
SSCC-Szasz: normal) (%CV)
a. (-L-Glutamyl)-4-nitroanilide: 4.0 AACC-Szasz*: 5.5%
mmol/L SSCC-Szasz:
b. Glycylglycine: 75 mmol/L Recommended method:

c. Tris-HCl: 100 mmol/L IFCC Primary Reference Method :< 2.5
d. 7.6
e. MgCl2: 10 mmol/L Reference interval
Recommended method: a. Men
a. (-Glutamyl)-5-carboxy-4-nitroanilide: b. Women
8.4 mmol/L AACC-Szasz*:
b. Glycylglycine: 190 mmol/L a. 5-30 U/L
c. Tris-HCl: 100 mmol/L b. 3-20 U/L
d. 8.16 SSCC-Szasz:
IFCC Primary Reference Method: a.
a. -L-Glutamyl-3-carboxy-4-nitroanilide: 6 b.
mmol/L Recommended method:
b. Glycylglycine: 150 mmol/L a.
c. Sodium hydroxide; 2 mol/L (for adjusting b.
IFCC Primary Reference Method
pH)
a. 55 U/L or 0.92 kat/L (upper 97.5 percentiles)
d. pH 7.70 0.05
b. 38 U/L or 0.63kat/L (upper 97.5 percentiles)
Reaction condition: Fraction of sample
volume
AACC-Szasz*: 0.0909 (1:11)
*Selected method, American Association for Clinical Chemistry. From Szasz G. Reaction-rate method for -
glutamyltransferase activity in serum. Clin Chem 1976;22:2051-2055.
Committee on Enzymes of the Scandinavian Society for Clinical Chemistry.[7]
Kaplan and Pesces Clinical Chemistry
International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) , Scientific Division, Committee on
Reference Systems on Enzymes(C-RSE)[10]
Certification Report: Certified Reference Material ERM-AD452/IFCC European Commission, Directorate general
Joint Research Center Institute for Research Materials and Measurements (IRMM) Geel (BE).[22]
616
Table 3: -Glutamyltransferase Activity as a Function of Sex and Age for a Selected Population
Age Males (Percentiles) Females (Percentiles)
in Years
2.5 50 90 95 97.5 2.5 50 90 95 97.5
Activity, U/L
8-12 8.4 14.0 19.8 23.8 27.8 8.0 13.9 19.4 21.7 24.7
12-16 10.1 15.4 20.0 23.6 26.4 6.4 13.7 19.8 23.9 31.0
16-20 9.2 17.5 29.1 35.0 45 7.0 13.4 19.5 21.6 24.4
20-30 11.5 23.6 48.4 63.8 76.1 7.0 15.7 25.8 32.2 41.5
30-40 12.1 28.6 64.4 81.2 94.0 9.0 17.4 33.9 43.9 55.0
40-50 13.1 32.4 68.1 85.6 105 8.9 18.5 42.2 53.2 71.3
50-60 14.6 32.6 66.7 86.3 103 10.3 20.2 47.4 59.4 83.2
>60 13.8 28.6 61.7 72.5 83.6 10.7 22.5 40.5 48.5 60.2
From [56] With permission of the American Association for Clinical Chemistry.
Table 4: -Glutamyltransferase Activity During Early Childhood Development
Age Mean (U/L) Range (U/L)

0-1 month 71 0-151
1-2 months 48 0-114
2-4 months 33 0-81
4-7 months 17 0-34
7 months-2 years 11 0-24
Ref: [35,55]
Table 5: Tissue Distribution of -Glutamyltransferase

Tissue Units/mg of Wet Units/g of Wet Ratio to Serum
Tissue Tissue
Kidney 44 2225 7420
Pancreas 13 184.5 614
Liver 4 86.5 288
Spleen 2 34 113
Heart * 1 3
Skeletal muscle * 1.5 5
Lung 6 7 23
Brain * 11 37
Intestine 1 21 70
Sera 0.3 1
*<1 unit/mg tissue.
Ref:[40,42]
617
Procedure: Kinetic Measurement of GGT Activity Calculations

Using the molar absorptivity, , of 2-nitro-5-
Principle
aminobenzoic acid as 9500 Lcm-1mol-1, one can
(-L-Glutamyl)-3-carboxy-4-nitroanilide + glycylglycine
calculate the enzyme activity for a 1-cm pathlength with
EC 2.3.2.2 the following formula:
(-L-glutamyl)glycylglycine + 2-nitro-5-aminobenzoic acid
or A for t min total volume 1 106 mol
(-L-Glutamyl)-3-carboxy-4-nitroanilide + H2O U/L
t min sample volume mole
EC 2.3.2.2
glutamic acid + 2-nitro-5-aminobenzoic acid where A = difference between the initial and final
absorbance over analysis time,
This reaction is monitored at 410 nm. t, in minutes
= molar absorptivity of 2-nitro-5-aminobenzoic acid
Reagents 106 = factor to convert concentration to micromoles per
1. Buffer/substrate solution. Tris (100 liter
mmol/L), 12.11 g; glycylglycine (190 mmol/L), 25.10 g;
and (-L-glutamyl)-3-carboxy-4-nitroanilide (8.4 For the assay, the values are as follows:
mmol/L), 2.91 g are added to about 800 mL of distilled
water. Adjust pH to 8.16 with 1 mol/L of HCl, and dilute A for t min 2.20 mL 1 106
U/L
to 1000 mL. This solution is stable 2 days at room 5 min 0.2 mL 9500(L/mol cm)
temperature or 5 days at 2C to 8C. U / L A 231.6
Assay Note: The use of enzyme reference materials (ERM) as

Equipment: spectrophotometer (10 nm bandpass) calibrators with well-defined assay values is being
equipped with cuvette heated at 37C and a 410 nm recommended by IFCC and ISO 18153 for uniformity of
filter. results across methods and instruments and for
1. Place 2 mL of buffer/substrate solution and traceability of the results to the primary reference system
0.20 mL of specimen (control, enzyme comprising primary reference method and ERM with
reference material/standard/calibrator or assigned value. See Tables 1 and 2 for IFCC primary
unknown patient serum or plasma) in a 3-mL reference procedure principle and reaction conditions.
cuvette. For details of IFCC primary reference procedure for the
2. After 3 minutes of preincubation at 37C, measurement of catalytic activity of GGT, reader is
follow absorbance at 410 nm for 5 minutes. referred to Clin Chem Lab Med 2002;40:734-738.
3. The serum sample volume fraction is 0.091
(1:11).
618
Gastric Fluid Analysis
Lawrence A. Kaplan
Name: Gastric fluid

Clinical significance: Refer to Chapter 34, The Pancreas: Function and Chemical Pathology,
and Chapter 35 Gastrointestinal Function and Digestive Disease in the 5th edition of Clinical
titration to a neutral pH (or pH 7.4) using phenol red
Principles of Analysis and Current Usage indicator (1 g/L)[5] or a pH meter.[3,5,6]
Gastric fluid analysis is a vague term that can be used to
include analysis of a number of electrolyte and protein As an alternative to titration methods, Moore and
(enzymes) components of gastric juices. The most Scarlata[3] determined total gastric acid content by
commonly performed analysis of gastric fluid is the direct measurement with a pH glass electrode (method 2,
measurement of the total acid output. Gastric Fluid Analysis Methods Summary Table). They
determined the actual hydrogen-ion concentration by
Historically,[1] total gastric acid was defined as the sum dividing the pH by the hydrogen-ion activity
of the combined acid (that acid combined with
coefficient.[7] The H+ activity coefficient, which is the
physiological buffers, such as proteins and salts) and the
thermodynamic activity of the hydrogen ions,[7] is
free acid (the acid that exceeded the buffering capacity
dependent on the molar concentration of other ions
of gastric fluid). Subsequent investigations have
present in the solution, such as sodium or potassium
demonstrated that most pure gastric fluid closely
resembles hydrochloric acid[2] but has some additional ions. Tables providing the relationship between the H+
minor buffering activity, perhaps reflecting weak organic activity coefficient and the concentration of H+ and
acids such as lactic, carbonic, and others.[3] other ions are available.[8] As an approximation, one
can assume an average total concentration of Na and K
Because of the division of total gastric acid into of 50 mEq/L and perform a simplified calculation.[4]
combined (the weak organic acids) and free (HCl),
the earlier analyses used a two-stage titration (method 1, The measurement of total gastric acidity by methods 1
Gastric Fluid Analysis Methods Analysis Table). A and 2 (Gastric Fluid Analysis Methods Summary Table)
mixed indicator solution, consisting of Topfers reagent requires intubation of the patient with a gastric tube and
(5 g/L alcoholic solution of p- aspiration of gastric fluids. To avoid the necessity for
dimethylaminoazobenzene) and phenolphthalein (10 g/L intubations, indirect measurements of gastric acidity
in ethanol), was added to the gastric fluid sample.[4] have been suggestedthe tubeless gastric analysis.
Sodium hydroxide (0.1 mol/L) was added from a buret These tests employ a carboxylic-cationic resin in which
to titrate both of these indicators; Topfers reagent was some of the hydrogen ions have been replaced by a
titrated to a pH range of 2.8 to 3.5 (red to colorless) and water-soluble dye (method 3, Gastric Fluid Analysis
the phenolphthalein was titrated to a pH range of 8.2 to Methods Summary Table). The resin is ingested, and
10 (colorless to red). Results were often reported as the some of the dye is displaced by gastric acid, absorbed by
number of milliliters of 0.1 M NaOH needed to titrate the small intestines, and excreted in urine. The amount
100 mL of gastric fluid. Alternatively, the more of dye excreted in the urine, measured
meaningful total milliequivalents (mEq) of acid secreted spectrophotometrically, is proportional to the total
can be calculated as follows: gastric acidity.[1] The most commonly used dye is
azure A (3-amino-7-dimethylaminophenazathionium
chloride). The presence of this dye can be measured by
spectrophotometry at 630 nm.[9,10] Heating the urine in
the presence of CuSO4 has been recommended so that
colorless, reduced forms of azure A can be converted
back to the blue form.[10]
It is important to measure gastric acid secretion both

under a basal, resting, condition (basal acid output,
BAO) and under a stimulation condition (maximum acid
output, MAO). Various foodstuffs have historically been
where Vg = volume of gastric fluid titrated (mL) employed to provide such stimulation. These test meals
VNaOH = volume of NaOH needed to titrate have varied considerably.[1] refined chemical
the gastric fluid (mL) stimulants include histamine, insulin-induced
M = molarity of NaOH (usually 0.1 mol/L or 0.1 hypoglycemia, betazole (Histalog, a histamine analog),
Eq/L) porcine gastrin, and recently pentagastrin, the artificially
synthesized active portion of the gastrin molecule.
Other alternative titration methods include the titration
of total acidity using a phenolphthalein end point and The most commonly used procedures for measurement
of gastric acid content are the titration methods.
619
Reference and Preferred Methods through the nose) into the lowest portion of the stomach.
The concept of combined and free gastric acid is After removal of residual gastric fluid, sampling of
presently considered inaccurate. The procedure gastric fluid from the stomach continues for the next 60
employing both Topfers reagent and phenolphthalein is min, with samples collected in four 15-min sampling
not considered to provide any additional useful periods. After the collection of the basal gastric
information. For measurement of total acidity, one can secretions, the pentagastrin (usually 5 to 6 g/kg of body
use either a titration method or the pH electrode method weight) is administered, and gastric secretions are again
of Moore and Scarlata.[3] Titrating to neutrality (or pH collected for 60 minutes in four 15-min sampling
7.4) with a pH meter or phenol red indicator, or to pH 8 periods. Each gastric fluid sample should be placed in a
to 10 with phenolphthalein, provides sufficient clinical suitable, clean container, closed, and sent to the
information and employs readily available laboratory laboratory for analysis. The specimen should be clear
equipment. Although the method of Moore and Scarlata and without sediment. Centrifugation of specimens may
may provide a more exact measurement of gastric be necessary to remove particulate matter. The samples
acidity, the greater accuracy may not be clinically are stable for several days at room or refrigerated
necessary. In addition, a pH meter able to measure pH to temperatures.
the nearest thousandth (0.001) will probably not be
available in many laboratories. Also, to be most Gastric Fluid Reference Interval
accurate, one needs to measure sodium and potassium Gastric residue (stomach contents after a 12-h fast) has a
as well as pH, or else one must make assumptions volume of 20 to 100 mL. A volume greater than 100 mL
concerning the concentration of Na+ and K+ that may may be considered abnormal. This can be caused by
not always be valid. delayed emptying, increased secretion, duodenal ulcer,
ZollingerEllison syndrome, or a mixture of regurgitated
Tubeless gastric analysis using azure A has not been material containing bile from the duodenum. Gastric
widely accepted because of the large number of false- juice may be fluid or viscous (mucus increases the
positive (excess acid production, 1% to 3%) and false- viscosity) and usually has a sour odor. It is usually
negative results (10% to 15%) obtained with this colorless. If it is yellow green, bile is probably present. If
method.[1,11] A large number of physiological causes brown or red, the presence of blood should be suspected.
of erroneous results have been reviewed,[1] though Acid content should increase after stimulation.
false-negative results may be reduced if one employs a
proper gastric stimulator before the patient ingests the The pH of the basal gastric fluid should be less than 3,
resin. Nevertheless, direct comparisons of the azure A though very low basal pH values are also associated with
procedure to a quantitative assessment of total acid excessive parietal cell activity. A basal pH greater than 6
output have shown the azure A test to be is almost certainly caused by anacidity, a condition of
unreliable.[12,13] subnormal parietal cell secretion. The pH of gastric
secretions after gastric stimulation should fall to less
The recommended procedure for routine use is a than 2 with normally functioning parietal cells.
titrimetric quantitation of total acidity or the direct
measurement of gastric acidity with a pH meter. Values for maximal acid output are listed in Gastric
Fluid Analysis Table: Computed range of peak acid
The recommended gastric stimulant is pentagastrin.[14] output as determined by age, weight, and sex.[18]
This substance provides a more reproducible stimulus
than histamine or betazole (Histalog), is more potent Interpretation
than these two, and is virtually free of undesired side Anacidity (achlorhydria) is associated with pernicious
effects.[15] Histamine can cause drowsiness, headache, anemia, and the presence of sufficient gastric acidity
tachycardia, erythema, and hypertension, even in the rules out this disease. Insufficient HCl production by the
presence of antihistamines.[16] Although the use of parietal cell is denoted by an elevated pH of the basal
betazole (Histalog) can avoid most of these side effects, gastric secretions and by the failure of the gastric fluid
cases of life-threatening syncope and shock have been pH to fall below 2 after pentagastrin stimulation. Other
reported.[17] Since the use of pentagastrin affords an diseases associated with achlorhydria include gastric
inexpensive, reliable alternative, it is used almost carcinoma, hypochronic anemia, rheumatoid arthritis,
exclusively as the stimulator for gastric secretion. The and myxedema.
use of intragastric electrodes is the latest technique to
measurement of gastric pH. These systems can record Ulcers, on the other hand, are associated with excessive
the pH at defined intervals and allow changes in gastric HCl production in the basal period (pH < 3) and an
acid pH due to the effects of various substances to be MAO greater than 35 (Gastric Fluid Analysis Table:
monitored. Pentagastrin stimulation test), especially in a patient
older than 45 years of age.
Specimen
A gastric sample is usually obtained from a patient who The ZollingerEllison syndrome is an extreme form of
has fasted overnight (12 h). It is important to prevent excessive acid production resulting from a gastrin-
dilution of gastric fluid with saliva or water during the secreting tumor, usually of the pancreas (gastrinoma).
procedure. The intubation tube is introduced (usually Diagnosis of a gastrinoma can be made by
620
demonstration of very high gastric acidity (10 to 100 9 Klein B, Weisman M. Quantitative
BAO or a MAO that is only 40% to 60% higher than the determination of azure A in urine. Clin Chem
BAO). If the BAO/MAO ratio is greater than 60%, the 1959;5:115-118.
person is almost certain to have the ZollingerEllison 10 Rosenthal HL, Bascaglia S. Tubeless gastric
syndrome. analysis. JAMA. 1958;168:409.
11 Correia JP, DeMoura MC, Da Costa CG.
References Tubeless gastric analysis. Br J Clin Pract
1 Cannon DC, Winkleman JW, editors. Henrys 1967;211:227-31.
Clinical chemistry: principles and techniques. 12 Galambos JT, Kirsner JB. Tubeless gastric
2nd ed. New York: Harper & Row; 1974. p analysis. Arch Intern Med 1955;96:752-6.
1557-63. 13 Segal HL, Miller LL, Plumb EJ. Tubeless
2 Bock OA. The concepts of free acid and gastric analysis with an azure A ion-exchange
total acid of the gastric juice. Lancet compound. Gastroenterology 1955;28:402-8.
1962;2:1101-2. 14 Abernathy RJ, Gillespie IE, Lowrie JH.
3 Moore EW, Scarlata RW. The determination of Pentagastrin as a stimulant of maximal gastric
gastric acidity by the glass electrode. acid response in men: a multicentre pilot study.
Gastroenterology 1965;49:178-88. Lancet 1967;1:291-5.
4 Bauer JD. Clinical laboratory methods. 9th ed. 9, 15 Konturek SJ, Lenkosz J. Pentapeptide infusion
St. Louis: C.V. Mosby Co; 1982. p. 780-2. test. Scand J Gastroenterol 1967;2:112-7.
5 Lubran M. Measurement of gastric activity. 16 Callender ST, Retief FP, Witts LJ. The
Lancet 1966;2:1070-1. augmented histamine test with special reference
6 Baron JH. Measurement and nomenclature of to achlorhydria. Gut 1960;1:326-36.
gastric acid. Gastroenterology 1963;45:118-21 17 Blum NI, Mayoral LG, Kalser MH.
7 Weber SG. Electrode-based measurements. In: Augmented gastric analysisa word of
Kaplan LA, Pesce AJ. editors: Clinical caution. JAMA. 1965;19:339-41.
chemistry: theory, analysis, and correlation, 3rd 18 Blackman AH, Lambert DL, Thayer WR,
ed. St. Louis: C.V. Mosby Co; 1984 p. 253. Martin HF. Computed normal values for peak
8 Moore EW. Determination of pH by the glass acid output based on age, sex, and body weight.
electrode: pH meter calibration for gastric Am J Dig Dis 1970;15:783-9.
analysis. Gastroenterology 1968;54:501-7.
Tables
Gastric Fluid Analysis: Methods of Gastric Acid Output Analysis

Method 1: Titration
Principle of analysis: Acid is titrated using a pH indicator to indicate end point; Tpfers reagent for pH ~3,
phenol red for pH 77.4, or phenolphthalein for pH 812. Amount of total acid present calculated from volume of
base consumed
Comments: Most common; use of phenolphthalein indicator is sufficient; recommended method
Method 2: pH electrode
Principle of analysis: Acid concentration determined by pH electrode and corrected for presence of other ions by
calculation or estimation of the H+ activity coefficient
Comments: Rare; high degree of accuracy; probably greater than needed clinically
Method 3: Tubeless dye test
Principle of analysis: Dye, bound to ingested cationic resin, is displaced by gastric acid, absorbed by intestine,
and excreted in urine. Amount of dye, measured spectrophotometrically, is related to total gastric activity
Comments: Common; poor accuracy; not recommended
621
Gastric Fluid Analysis: Computed Range of Peak Acid Output (PAO, mEq/hour) as Determined by Age,
Weight, and Sex
Age Weight Weight Male Female

(yr) kg lb
1540 45.4 100 9.9 18.7 5.9 18.7
1540 50.0 110 11.1 19.9 6.6 15.4
15-40 54.5 120 12.3 21.1 7.4 16.2
1540 59.0 130 13.5 22.3 8.2 16.9
1540 63.6 140 14.7 23.4 9.0 17.8
1540 68.2 150 15.9 24.7 9.8 18.6
1540 72.7 160 17.1 25.9 10.6 19.4
1540 77.3 170 18.2 27.0 11.4 20.1
1540 81.8 180 19.4 28.2 12.1 20.9
1540 86.4 190 20.6 29.4 12.9 21.7
1540 90.9 200 21.8 30.6 13.7 22.5
1540 95.5 210 23.0 31.8 14.5 23.3
1540 100.0 220 24.4 32.9 15.3 24.0
1540 104.5 230 16.1 34.2 16.1 24.9
1540 109.1 240 26.6 35.4 16.9 25.6
1540 113.6 250 27.8 36.6 17.7 26.4
1540 118.8 260 28.9 37.7 18.4 27.2
1540 122.7 270 30.2 38.9 19.2 28.0

(yr) kg lb
45 45.4 100 6.6 16.6 4.5 13.3

45 50.0 110 7.6 17.6 5.2 13.9
45 54.5 120 8.6 18.6 5.8 14.6
45 59.0 130 9.5 19.6 6.5 15.2
45 63.6 140 10.5 20.6 7.1 15.9
45 68.2 150 11.5 21.6 7.8 16.5
45 72.7 160 12.5 22.5 8.4 17.2
45 77.3 170 13.5 23.6 9.1 17.9
45 81.8 180 14.5 24.5 9.7 18.5
45 86.4 190 15.5 25.5 10.4 19.2
45 90.9 200 16.5 26.5 11.0 19.8
45 95.5 210 17.4 27.5 11.7 20.5
45 100.0 220 18.4 28.5 12.3 21.1
45 104.5 230 19.4 29.5 12.9 21.8
45 109.1 240 20.4 30.5 13.6 22.4
45 113.6 250 21.4 31.5 14.3 23.1
45 118.8 260 22.4 32.4 14.9 23.7
45 122.7 270 23.4 33.4 15.6 24.4
622

(yr) kg lb
50 45.4 100 5.6 15.7 3.9 12.7

50 50.0 110 6.5 16.6 4.5 13.3
50 54.5 120 7.4 17.5 5.1 13.8
50 59.0 130 8.3 18.4 5.7 14.4
50 63.6 140 9.2 19.3 6.3 15.0
50 68.2 150 10.1 20.2 6.8 15.6
50 72.7 160 11.0 21.1 7.4 16.2
50 77.3 170 11.9 21.9 8.0 16.8
50 81.8 180 12.8 22.9 8.6 17.4
50 86.4 190 13.7 23.7 9.2 17.9
50 90.9 200 14.6 24.6 9.8 18.6
50 95.5 210 15.5 25.5 10.4 19.2
50 100.0 220 16.4 26.4 10.9 19.7
50 104.5 230 17.3 27.3 11.6 20.3
50 109.1 240 18.1 28.2 12.1 20.9
50 113.6 250 19.0 29.1 12.7 21.5
50 118.8 260 19.9 30.0 13.3 22.1
50 122.7 270 20.8 30.9 13.9 22.7

(yr) kg lb
55 45.4 100 4.4 14.5 3.1 11.9

55 50.0 110 5.2 15.3 3.6 12.4
55 54.5 120 5.9 16.0 4.1 12.9
55 59.0 130 6.8 16.8 4.6 13.4
55 63.6 140 7.5 17.6 5.1 13.9
55 68.2 150 8.3 18.4 5.6 14.4
55 72.7 160 9.7 19.1 6.2 14.9
55 77.3 170 9.8 19.9 6.7 15.4
55 81.8 180 10.6 20.7 7.2 15.9
55 86.4 190 11.4 21.5 7.7 16.5
55 90.9 200 12.2 22.2 8.2 16.9
55 95.5 210 12.9 23.0 8.7 17.5
55 100.0 220 13.7 23.8 9.2 17.9
55 104.5 230 14.5 24.6 9.7 18.5
55 109.1 240 15.3 25.3 10.2 19.0
55 113.6 250 16.0 26.1 10.8 19.5
55 118.8 260 16.8 - 26.9 11.3 20.0
55 122.7 270 17.6 27.6 11.8 20.5
623

(yr) kg lb
60 45.4 100 2.9 13.0 2.1 10.9

60 50.0 110 3.6 13.7 2.5 11.3
60 54.5 120 4.2 14.3 2.9 11.7
60 59.0 130 4.9 14.9 3.4 12.1
60 63.6 140 5.5 15.5 3.8 12.6
60 68.2 150 6.1 16.2 4.2 12.9
60 72.7 160 6.7 16.8 4.6 13.4
60 77.3 170 7.4 17.4 5.0 13.8
60 81.8 180 7.9 18.1 5.4 14.2
60 86.4 190 8.6 18.7 5.9 14.6
60 90.9 200 9.2 19.3 6.3 15.0
60 95.5 210 9.9 19.9 6.7 15.5
60 100.0 220 10.5 20.6 7.1 15.9
60 104.5 230 11.1 21.22 7.5 16.3
60 109.1 240 11.7 21.9 7.9 16.7
60 113.6 250 12.4 22.4 8.3 17.1
60 118.8 260 12.9 23.0 8.7 17.5
60 122.7 270 13.6 23.7 9.2 17.9

(yr) kg lb
65 45.4 100 1.2 11.3 0.9 9.8

65 50.0 110 1.7 11.8 1.3 10.1
65 54.5 120 2.2 12.2 1.6 10.4
65 59.0 130 2.6 12.7 1.9 10.7
65 63.6 140 3.1 13.1 2.2 10.9
65 68.2 150 3.5 13.6 2.5 11.3
65 72.7 160 3.9 14.0 2.8 11.6
65 77.3 170 4.4 14.5 3.1 11.9
65 81.8 180 4.9 14.9 3.4 12.2
65 86.4 190 5.3 15.4 3.7 12.5
65 90.9 200 5.8 15.8 3.9 12.8
65 95.5 210 6.2 16.3 4.3 13.1
65 100.0 220 6.7 16.8 4.6 13.4
65 104.5 230 7.1 17.2 4.9 13.7
65 109.1 240 7.6 17.7 5.2 13.9
65 113.6 250 8.1 18.1 5.5 14.3
65 118.8 260 8.5 18.6 5.8 14.6
65 122.7 270 8.9 19.0 6.1 14.9
From Blackman, A.H., et al.: Am. J. Dig. Dis. 15:783-789, 1970.
624
Gastric Fluid Analysis: Pentagastrin Stimulation Test

Adult men reference interval:
Basal acid output (BAO) (mmol/h):
Under age 30 2.22.7
Over age 30 2.22.7
Maximal acid output (MAO) (mmol/h):
Under age 30 1442
Over age 30 333
Adult women reference interval:
Basal acid output (BAO) (mmol/h): 11.5
Maximal acid output (MAO) (mmol/h): 720
ZollingerEllison syndrome
Basal acid output (BAO) (mmol/h): 10100 (or more)
Maximal acid output (MAO) (mmol/h): 40%-60% above BAO
Ulcer predisposition
Maximal acid output (MAO) (mmol/h):
Likely >35
Highly likely >45
Low risk<11
5. If the specimen volume is less than 10 mL,
Procedure: Measurement of Total Gastric Acidity titrate a 1 mL aliquot with 0.01 M NaOH
Principle (dilute 0.1 M NaOH 1:10 with distilled water).
Pentagastrin is a synthetic product that has the Calculations
same physiological action as gastrin. It is a powerful
stimulus for HCl secretion in the stomach. Acid present
in the stomach contents is titrated with 0.1 M NaOH
with phenolphthalein as a pH indicator. Total gastric
acidity is calculated under basal (basal acid output,
BAO) and maximal (maximum acid output, MAO)
stimulation conditions.
For a pentagastrin stimulation, record each acid value,
Reagents and then report the following:
1. 0.1 M NaOH. Place 4 g of NaOH in a 1 L
volumetric flask, and add approximately 900 mL of Basal I + Basal II = mEq of acid per h in basal juice
distilled water to dissolve the pellets. When the solution 0.5
cools to room temperature, bring the volume to 1 L with
additional distilled water, and mix. Stable in plastic where basal I and II are the two highest values obtained.
bottle for 2 years at room temperature. This is the BAO.
2. 1% phenolphthalein, 10 g/L (31.2 mmol/L).
Dissolve 100 mg of phenolphthalein in 10 mL of Stim I + Stim II + Stim III + Stim IV =
absolute ethanol. Stable for 1 year at room temperature. mEq of acid per h in stimulated juice
Assay Alternatively, one can use the two highest consecutive

Equipment: pH meter, buret. 15-min stimulated samples and multiply the results by 2.
1. Measure and record the volume of each This is the MAO.
specimen. Use a graduated cylinder.
2. Measure the pH of each specimen using a pH Notes
meter that has been calibrated at pH 4 and 7. 1. The color of the gastric fluid should be noted on
the report form. Normal gastric fluid is almost colorless.
3. If the specimen contains food particles,
Contamination of fluid by bile will result in a yellow-
centrifuge at 500 g for 15 min. Discard pellet.
green color, whereas blood contamination will color the
4. Put a 10 mL aliquot of the specimen in a 50 mL gastric fluid a red to coffee-ground brown color.
Erlenmeyer flask. Add 3 drops of 2. Mucus can interfere in the analysis. Excess
phenolphthalein, and mix. Titrate, with 0.1 M mucus, found in cases of gastritis and pyloric
NaOH delivered from a buret, to the first pink obstruction, can be removed by centrifugation of the
color that remains. Record the volume (in sample at 1500 g for 10 min.
milliliters) of NaOH used.
Gastric Fluid Reference Interval
click here
625
Gentamicin and other Aminoglycosides

Danni L. Meany and William Clarke
Name: Gentamicin and other aminoglycosides

Clinical significance: Refer to Chapter 14, Therapeutic Drug Monitoring, in the 5th edition of Clinical
Chemistry: Theory, Analysis, Correlation
Molecular formula: See Table 1
Molecular weight: See Table 1
Merck Index: See Table 1
Chemical class: Aminoglycosides
Structure:
NH
O
NH2
H 2N NH2
O
O
HO
HN
HO OH

Aminoglycosides are polycationic agents that are used Monitoring of serum aminoglycoside concentration is
primarily to treat serious infections caused by aerobic essential for both efficacy and avoidance of toxicity, since
gram-negative bacilli [1-4]. Nine aminoglycosides aminoglycosides have clearly defined therapeutic ranges,
(gentamicin, tobramycin, amikacin, streptomycin, and nephrotoxicity and ototoxicity are frequently
neomycin, kanamycin, paromomycin, netilmicin, and experienced after exposure to high concentrations. Many
spectinomycin) have been approved for clinical use in the assays have been developed for quantitative
United States. Gentamicin is the aminoglycoside used aminoglycoside measurement. In this chapter, current
most often because of its low cost and reliable activity methodologies for quantitative gentamicin measurement
against gram-negative aerobes [4]. Tobramycin and will be described, which serves as a prototype for
amikacin are also frequently prescribed. In general, measurement of other aminoglycosides.
gentamicin, tobramycin, and amikacin are often used
interchangeably in similar circumstances [3]. Other Microbiological Assay
aminoglycosides, however, are used only for selected Microbiological assays were commonly used for analysis
indications (e.g., paromomycin for protozoa and of gentamicin prior to the 1970s. Among them, agar disk
spectinomycin for Neisseria gonorrhoeae). diffusion assay was the most frequently used [4-8]. In this
assay, gentamicin standards prepared in pooled human sera
or serum specimens containing gentamicin were pipetted
i onto sterile paper disks and placed onto an agar plate that
Gentamicin and Other Aminoglycosides
Previous and current authors of this method: was prepared with antibiotic medium and seeded with
First edition: Joseph R. DiPersio indicator bacteria (e.g., Klebsiella pneumoniae) [8]. The
Methods edition: Joseph R. DiPersio agar plate was then incubated at 37C for 4 to 18 hours,
Second edition: Joseph R. DiPersio depending on the antibiotic medium and indicator
Third edition: Joseph R. DiPersio organism used. During this period, the indicator organism
Fourth edition: Joseph R. DiPersio grew throughout the agar medium, producing a visual haze
Fifth edition: Danni L. Meany, William Clarke of growth. However, zones of growth inhibition formed
around the antibiotic-containing disks because of diffusion
626
of drug into the agar. After incubation, the diameters of it competes with radio-labeled gentamicin and thus reduces
each zone of inhibition were measured with a vernier precipitation of radioactivity [14-16]. The technical
caliper. A calibration curve was established using the simplicity, sensitivity, and specificity of RIA made it
diameters of zone of inhibition versus the gentamicin superior to microbiological assays. For the first time,
concentration in standards, which was then used to gentamicin could be measured in patients receiving
calculate the gentamicin concentration in a clinical multiple antibiotics. However, because of continuous
specimen. changes in the counting rate of radioisotope standards due
to the short half-life of the isotope, a calibration curve
Microbiological assays were inexpensive, easy to set up, must be generated with each run. This inconvenience has
and required no instrumentation. However, their caused replacement of RIA by non-isotopic homogenous
turnaround time was long (e.g., 17 hrs), and their accuracy immunoassays.
varied substantially [6,9]. In addition, they are known to be
affected by the presence of other antimicrobial agents such Radioenzymatic Assay
as colistin, nalidixic acid, polymyxin B, amikacin, and Radioenzymatic assays (REAs) were introduced in an
tobramycin [8]. To minimize these interferences, an attempt to overcome the drawbacks of widely used
indicator organism resistant to the antibiotics or techniques microbiological assays and radioimmunoassays in the
that inactivate these interfering drugs could be used. 1970s [17]. An REA utilized enzymes that are known to
Microbiological assays only measure biological activity of inactivate gentamicin by transfer of radio-labeled
gentamicin against the indicator organism, not necessarily substrates to gentamicin through adenylation [18,19] or
against bacteria infecting patients. Because of these acetylation [17]. One example of an REA is the gentamicin
shortcomings, microbiological assays have been largely 3-transferase assay [17]. In this assay, serum specimens
replaced by immunoassays for monitoring serum containing gentamicin were incubated in a buffer solution
aminoglycoside concentrations. Nevertheless, they are still with [acetyl-14C] acetyl-coenzyme A and gentamicin 3-
used in microbiology laboratories for aminoglycoside transferase. This enzyme transferred the radio-labeled
susceptibility testing. acetyl group to gentamicin. After incubation, the reaction
mixture was pipetted onto phosphocellulose paper that the
Bioluminescent Assay radio-labeled gentamicin bound very tightly. The excess
Bioluminescent assays utilize the rapid effect of [acetyl-14C] acetyl-coenzyme A was washed out, and the
gentamicin on bacterial ATP levels. This antibiotic causes radioactivity left on the paper was measured by a liquid
dose-dependent decreases in total and intracellular ATP scintillation counter. A control for non-specific binding of
and dose-dependent increases in extracellular ATP [10- [acetyl-14C] acetyl-coenzyme A to phosphocellulose paper
13]. In the bioluminescent assay that measures was run in the absence of gentamicin 3-transferase. This
extracellular ATP [10], serum specimens containing assay was known to be affected by sisomicin, a structural
gentamicin are incubated with a day-old bacterial culture analog of gentamycin; netilmicin and tobramycin are also
at 37C for 45 minutes. After incubation, luciferase is detected to a small extent [17]. REA bears the same
added directly into the culture. It reacts with extracellular drawback as RIA because of the short half-life of the
ATP and emits luminescence. In this case, the recorded radioisotopes, and REA has also been replaced by non-
light intensity is proportional to the gentamicin isotopic homogeneous immunoassays.
concentration. In the bioluminescent assay that measures
total [13] or intracellular ATP [12], extraction of Non-Isotopic Homogeneous Immunoassays
intracellular ATP is required before addition of luciferase, Non-isotopic homogenous immunoassays were first
and the recorded light intensity is inversely proportional to developed in the late 1970s [20]. Now they have become
the gentamicin concentration. Bioluminescent assays have the most popular methods for measuring gentamicin and
the same drawbacks as microbiological assays because of other aminoglycosides in rapid-response and core
their similarities. Likewise, they are still used in laboratories, because they are easily coupled with
microbiology laboratories for aminoglycoside automated multiple-channel analyzers. All non-isotopic
susceptibility testing. Because of the chemiluminescence homogenous immunoassays are competitive binding
detection scheme employed, they offer better sensitivity immunoassays.
than microbiological assays. In addition, bioluminescent
assays require shorter incubation times [10]. Particle-Enhanced Turbidimetric Inhibition
Immunoassay
Radioimmunoassay The particle-enhanced turbidimetric inhibition
Radioimmunoassays (RIAs) were developed to measure immunoassay (PETINIA) is based on the coupling of
gentamicin in 1972 [14]. In an RIA, unlabeled gentamicin gentamicin to a latex particle which reacts with anti-
competes with radio-labeled gentamicin for antibody. Then gentamicin antibody to form aggregates that increase
antibody-bound radioactivity is precipitated by addition of turbidity of the solution [21]. Unlabeled gentamicin from a
a second antibody and is measured in the precipitate. sample competes for antibody and thus decreases the rate
Unlabeled gentamicin is determined by the extent to which of aggregation. Since turbidity of the solution is measured
627
by a turbidimeter as the intensity of light transmitted gentamicin in the sample regulates the reassembly of ED
through the medium, the lower the turbidity is, the higher and EA by competing for a limited amount of antibody.
the recorded light intensity and the higher the unlabeled Therefore, the amount of signal produced is proportional to
gentamicin concentration in the sample. In other words, the concentration of unlabeled gentamicin.
the recorded light intensity is proportional to the unlabeled
gentamicin concentration. Direct Chemiluminescence Immunoassay
A direct chemiluminescence immunoassay is analogous to
A nephelometer can also be used to measure turbidity of a RIA, except the label is a chemiluminescent acridinium
solution [22]. However, because it measures the scattered ester [4]. In this assay, solid-phase antibody is incubated
light at right angles to the incident light rather than the with unlabeled and acridinium-labeled gentamicin. After
transmitted light in the direct path, nephelometric incubation, antibody-coated beads are separated using a
measurements cannot be as easily integrated with magnetic field, and the amount of antibody-bound
automated analyzers as turbidimetric measurements. For acridinium ester is measured by photomultiplier detection
this reason, turbidimetric inhibition immunoassays are of flash emission upon addition of an oxidizing reagent.
more popular than nephelometric inhibition
immunoassays. Fluorescence Immunoassay
A fluorescence immunoassay (FIA) is based on
Fluorescence Polarization Immunoassay competitive binding between unlabeled gentamicin in a
A fluorescence polarization immunoassay (FPIA) sample and a known amount of gentamicin labeled with a
measures polarized fluorescence intensity from fluorogenic substrate (e.g., a derivative of umbelliferyl--
competitive binding of unlabeled and fluorescein-labeled D-galactoside) for an antibody [9,10]. Binding of the
gentamicin to anti-gentamicin antibodies [23-30]. When labeled gentamicin to the antibody prevents hydrolysis of
fluorescein-labeled gentamicin binds to an antibody, the the fluorogenic substrate by -galactosidase. Therefore, -
complex has high polarized emission upon excitation by galactosidase only hydrolyses the unbound substrate to
polarized laser because of its slow molecular rotation. produce a fluorescent product, which is measured
However, when fluorescein-labeled drug is free in fluorometrically. FIA is similar to EMIT, except it is the
solution, it has low polarized emission because of its fast enzyme substrate that is linked to gentamicin.
rotation. Therefore, the greater the amount of unlabeled
gentamicin in the sample, the less polarization of emitted High-Performance Liquid Chromatography
lightthat is, polarized fluorescence intensity is inversely High-performance liquid chromatography (HPLC) has
proportional to the unlabeled gentamicin concentration in been used for determination of gentamicin mostly in
the sample. pharmacokinetic and drug stability studies [4]. It is not
frequently used in clinical laboratories, because the
Enzyme-Multiplied Immunoassay Technology protocol is labor intensive: (1) gentamicin needs to be
The enzyme-multiplied immunoassay (EMIT) was the first extracted from serum, and (2) ultraviolet absorption of
homogeneous immunoassay to be developed for clinical gentamicin is not sensitive enough for detection, so it
use [20]. In this assay, unlabeled gentamicin competes for needs to be fluorescently derivatized with o-
anti-gentamicin antibody with gentamicin linked to phthalaldehyde [7,32]. Recently, HPLC coupled with mass
glucose-6-phosphate dehydrogenase [24,26-30]. When the spectrometry has been used for gentamicin analysis, which
enzyme-linked gentamicin binds to an antibody, activity of has become the most powerful method for confirmation of
the enzyme is reduced. Consequently, only when the antibiotic residues in food [4,33].
enzyme-linked gentamicin is free in solution can it convert
NAD+ to NADH, which is measured photometrically at A summary of gentamicin methods is presented in Table 2.
340 nm. The absorbance at 340 nm is proportional to the
unlabeled gentamicin in the sample. Reference and Preferred Methods
Currently there is no reference method for determination
Cloned-Enzyme Donor Immunoassay of gentamicin in serum or plasma. Non-isotopic
The cloned-enzyme donor immunoassay (CEDIA) is based homogeneous immunoassays are the most widely used
on bacterial enzyme -galactosidase which has been methods because they are rapid, sensitive, accurate,
genetically engineered into two inactive fragments: precise, and easily automated. Radioimmunoassay and
enzyme donor (ED) and enzyme acceptor (EA) [31]. These radioenzymatic assay are not practical in rapid-response
fragments spontaneously reassemble to form fully active and core laboratories for two reasons: (1) a calibration
enzyme that cleaves a galactopyranoside substrate, curve should be generated with each run due to decay of
generating a color change. In CEDIA, gentamicin is tagged radioisotopes, and (2) gentamicin-specific radioactivity
to ED, which does not interfere with enzyme activity. should be separated (e.g., precipitation or phosphocellulose
However, when ED-linked gentamicin binds to an paper) from nonspecific radioactivity. In comparison,
antibody, it interferes with reassembly of ED and EA and calibration curves for homogeneous non-isotopic
thus eliminates the activity of the enzyme. Unlabeled immunoassays are stable for up to 2 weeks, and they do
628
not require separation steps. Among all the non-isotopic traditional dosing, aminoglycosides are administered every
immunoassays, PETINIA is the most popular, followed by 8 hrs in patients with normal renal function; in patients
FPIA. According to the 2007 College of American with impaired renal function, loading and maintenance
Pathologists (CAP) Participant Summary Report, 40% of dose are reduced and/or the dosing interval extended
participants use the former, and 19% use the latter. because aminoglycosides have prolonged half-life in them.
In the case of gentamicin, the loading dose for adults is
Specimen typically 2 to 3 mg per kilogram of lean body weight, and
Serum and plasma are the most commonly used specimens the maintenance dosing is 1.7 mg per kilogram. The
for therapeutic drug monitoring of gentamicin, which are recommended dosing interval is 8 hrs with creatinine
obtained after patients have received 2 to 3 doses after clearance > 90 mL/min, 12 hrs with creatinine clearance
initiation of therapy or after adjustment of the dose. between 50 and 90 mL/min, and 24 to 48 hrs with
Trough concentrations are measured within 30 min of the creatinine clearance between 10 and 50 mL/min [3]. In the
next dose, and peak concentrations 30 to 45 min after the traditional dosing, maintenance dose is selected based
end of an intravenous infusion or approximately 60 min upon a percentage of the loading dose to maintain effective
after an intramuscular injection. Gentamicin administration peak concentration. Trough gentamicin concentrations
times and the time that specimens are obtained are should also be monitored to avoid concentration-related
essential in interpreting the results. Serum gentamicin toxicity (see Table 3).
concentrations should be monitored periodically
throughout therapy. Routine measurement is not necessary Consolidated dosing, however, administers
with prophylactic therapy given for less than 48 to 72 hrs. aminoglycosides in higher dose (e.g., 7mg/kg) every 24 hrs
in patients with normal renal function or longer in patients
Microbiological assays and bioluminescent assays can be with impaired renal function. It targets a peak
used for antimicrobial susceptibility testing in any body concentration of ~ 20 g/mL, and trough concentration is
fluids. HPLC may be used to assay gentamicin in other often undetectable because of the short half-life of the
body fluids. However, there are no concrete data on the drugs compared to the long dosing interval [34]. In this
general use of immunoassays in fluids other than serum or dosing regime, a specimen for drug monitoring is obtained
plasma. 6 to 14 hours after the first dose, and the result from this
measurement is then used to predict the necessary dosing
Interferences interval [35]. See Table 4 for consolidated once-daily
Microbiological and bioluminescent assays are prone to dosing and desired serum concentrations for
interferences by antimicrobial agents present in samples aminoglycosides.
[4,8,9,17]. Therefore, they cannot be used for patients
receiving multiple antibiotics. Compounds that are Consolidated dosing utilizes two pharmacodynamic
structurally related to gentamicin, such as netilmicin, properties of aminoglycosides: concentration-dependent
sisomicin, and sagamicin, may cross-react with antibodies killing and post-antibiotic effect, which is defined as the
to gentamicin in immunoassays. However, with time required for an organism to demonstrate viable
improvement of antibody specificity, minimum cross- regrowth following the removal of an antibiotic. Higher
reactivity with other aminoglycoside (e.g., less than 10%) dosage of aminoglycosides results in higher peak
has been achieved by most available immunoassays concentration, which increases both the rate and extent of
[21,25,29]. In addition, they are slightly affected by bacterial cell death. In addition, the higher the dosage, the
hemolysis, lipemia, icterus, rheumatoid factor, and greater the post-antibiotic effect, up to a certain maximal
heterophilic antibodies (e.g., less than 10%) [21]. response [36]. The potential benefits of the consolidated
dosing over the traditional one are (1) possibility of
Gentamicin and Other Aminoglycosides Therapeutic decreased nephrotoxicity, (2) ease of administration and
Range therefore less expensive, and (3) a lesser intensity of
Since gentamicin and other aminoglycosides display peak monitoring of serum drug concentrations [37]. Currently,
concentrationdependent killing of microorganisms, the consolidated dosing may be considered an option in
therapeutic goal is to achieve a concentration in serum treatment of moderate and severe infections due to gram-
such that the indicated organisms are killed, but the host negative aerobic bacteria.
remains undamaged. Because the indicated organisms are
variable and become resistant to certain drugs, treatment Interpretation
with specific aminoglycosides agent should always be Aminoglycosides are used as the treatment for a wide
directed by susceptibility testing [2]. Therefore, desired range of aerobic gram-negative bacilli, as well as
peak concentrations vary with indicated organisms. staphylococci and enterococci when used in combination
with beta-lactams [2,3]. The antibacterial properties of
Two different approaches have been used for aminoglycosides result from (1) irreversible binding to the
administration of aminoglycosides: traditional multiple- 30S bacterial ribosome, thus inhibiting bacterial protein
daily dosing and consolidated once-daily dosing. In the synthesis, and (2) creating fissures in the outer bacterial
629
membrane, resulting in the leaking of intracellular contents vertigo, disequilibrium, light-headedness, nausea,
and increasing antibiotic uptake [38,40]. Since the uptake vomiting, and ataxia; the symptoms of the latter are
of aminoglycosides is energy dependent, and anaerobes tinnitus and hearing loss. Subclinical ototoxicity may be
have less energy available, they are intrinsically resistant detected in 10% to 20% of patients treated with
to aminoglycosides. As a result, beta-lactams are used aminoglycosides for long periods, but clinically overt
synergistically to facilitate uptake of aminoglycosides into ototoxic reactions may occur in only 1% to 3% of patients.
anaerobes such as enterococci. In contrast to the intrinsic
resistance is acquired resistance of aminoglycosides. Two The pathogenesis of nephrotoxicity is mainly attributed to
major mechanisms account for this: (1) bacterial the cationic amino groups on aminoglycosides. When
production of enzymes that inactivate the drug [41-43] and taken up into proximal renal tubule cells, aminoglycosides
(2) decreased accumulation of the drug due to an efflux are accumulated within lysosomes. This may be a charge-
system [44]. Because of structural differences, amikacin is mediated process. Inhibition of lysosomal function may be
not inactivated by the enzymes that inactivate gentamicin responsible for subsequent aminoglycoside-induced
and tobramycin. Therefore, amikacin is usually reserved cellular injury [45]. Another hypothesis of nephrotoxicity
for serious gram-negative infections that are resistant to postulates that aminoglycosides induce cell death by
gentamycin and tobramycin. Although acquired resistance stimulating the extracellular calcium-sensing receptor on
to aminoglycosides is becoming increasingly prevalent the surface of proximal tubular cells [46,47]. On the other
owing to repeated and long periods of use of one type, it is hand, the pathogenesis of induced ototoxicity with hearing
still less frequently encountered than other antibiotics, loss is less well understood. Sustained or excessive peak
which confirms their clinical importance despite the serum concentrations are thought to be a risk factor. In
introduction of newer and less toxic antimicrobials. order to minimize toxicity, potential risk factors that
predispose to aminoglycoside toxicity should be identified
Because aminoglycosides are very polar and poorly and when possible corrected (Table 6). Aminoglycosides
absorbed through the gastrointestinal tract, they are should only be used when their unique antibiotic potency
administered intravenously or intramuscularly. Peak serum is needed, and once the infecting organism has been
concentrations are measured 30 to 60 min after determined, it should be replaced by a potentially less
intravenously infusions or 30 to 90 min after intramuscular toxic antibiotic.
injections. In addition to blood, adequate drug
concentrations are also found in bone, synovial fluid, Gentamicin Performance Goals
peritoneal fluid, and other body fluids [3]. Since Acceptable performance (accuracy) for gentamicin assays
aminoglycosides are primarily distributed within the according to the Clinical Laboratory Improvement
extracellular fluid, disease states and drugs that alter fluid Amendments of 1988 (CLIA-88) is the target value 25%
balance (e.g., renal diseases and diuretics) require of the peer group mean. The same criteria are applied to
modifications to dosage to avoid adverse effects of amikacin and tobramycin. According to the 2007 CAP
aminoglycosides [3]. survey, the imprecision for all the gentamicin
measurements ranged from 4% to 11% for specimens, with
The main adverse effects of aminoglycosides are an average gentamicin concentration of 5.05 g/mL. The
nephrotoxicity and ototoxicity (Table 5). Aminoglycosides imprecision for tobramycin and amikacin ranged from 3%
are not metabolized in human bodies, and they are quickly to 9% for an average tobramycin concentration of 3.07
excreted by glomerular filtration, with an average half-life g/mL and 4% to 15% for an average amikacin
of 2.5 hrs with normal renal function [2]. However, their concentration of 5.83 g/mL, respectively. Based on the
half-life in the renal cortex is approximately 100 hrs [3]. data, current aminoglycoside assays demonstrate
Therefore, repetitive dosing may result in renal acceptable performance with regard to laboratory
accumulation of aminoglycosides and hence regulations.
nephrotoxicity. Another important aspect of nephrotoxicity
of aminoglycosides comes from their high intracellular References
concentrations in proximal renal tubule cells. 1 Kumana CR, Yuen KY. Parenteral
Aminoglycosides are absorbed from the glomerular filtrate aminoglycoside therapy. Selection, administration
by proximal renal tubular cells. The very high intracellular and monitoring. Drugs. 1994;47:902-913.
levels lead to subsequent tubular necrosis. 2 Moyer TP, Shaw LM. Therapeutic drugs and their
Aminoglycosides may produce clinically evident management. In: Burtis CA, Ashwood ER, Bruns
nephrotoxicity in 2% to 8% of patients, which is usually DE, eds. Tietz Textbook of Clinical Chemistry and
reversible with discontinuation of therapy. Moreover, Molecular Diagnostics. St Louis, Mosby;
periodic monitoring of serum creatinine concentrations 2006:1237-1286.
may alert to developing nephrotoxicity. By comparison, 3 Gonzalez LS 3rd, Spencer JP. Aminoglycosides:
ototoxicity is usually irreversible. Aminoglycoside- a practical review. Am Fam Physician.
induced ototoxicity may result in either vestibular or 1998;58:1811-1820.
cochlear damage. Manifestations of the former include
630
4 Stead DA. Current methodologies for the analysis 18 Butcher RH. Rapid serum gentamicin assay by
of aminoglycosides. J Chromatogr B Biomed Sci enzymatic adenylation. Am J Clin Pathol.
Appl. 2000;747:69-93. 1977;68:566-569.
5 Sabath LD, Casey JI, Ruch PA, Stumpf LL, 19 Benveniste R, Davies J. R-factor mediated
Finland M. Rapid microassay of gentamicin, gentamicin resistance: A new enzyme which
kanamycin, neomycin, streptomycin, and modifies aminoglycoside antibiotics. FEBS Lett.
vancomycin in serum or plasma. J Lab Clin Med. 1971;14:293-296.
1971;78:457-463. 20 Bastiani RJ. The EMIT system: a commercially
6 Ratcliff RM, Mirelli C, Moran E, OLeary D, successful innovation. Antibiot Chemother.
White R. Comparison of five methods for the 1979;26:89-97.
assay of serum gentamicin. Antimicrob Agents 21 Wei TQ, Chu VP, Craig AR, Duffy JE, Obzansky
Chemother. 1981;19:508-512. DM, Kilgore D et al. Automated homogeneous
7 Delaney CJ, Opheim KE, Smith AL, Plorde JJ. immunoassay for gentamicin on the dimension
Performance characteristics of bioassay, clinical chemistry system. Clin Chem.
radioenzymatic assay, homogeneous enzyme 1999;45:388-393.
immunoassay, and high-performance liquid 22 Kricka LJ. Optical Techniques. In: Burtis CA,
chromatographic determination of serum Ashwood ER, Bruns DE, eds. Tietz Textbook of
gentamicin. Antimicrob Agents Chemother. Clinical Chemistry and Molecular Diagnostics. St
1982;21:19-25. Louis: Mosby; 2006:61-91.
8 Lund ME, Blazevic DJ, Matsen JM. Rapid 23 Jolley ME, Stroupe SD, Schwenzer KS, Wang CJ,
gentamicin bioassay using a multiple-antibiotic- Lu-Steffes M, Hill HD et al. Fluorescence
resistant strain of Klebsiella pneumoniae. polarization immunoassay. iii. an automated
Antimicrob Agents Chemother. 1973;4:569-573. system for therapeutic drug determination. Clin
9 Selepak ST, Witebsky FG, Robertson EA, Chem. 1981;27:1575-1579.
MacLowry JD. Evaluation of five gentamicin 24 Cheng AF, Lam AW, French GL. Comparative
assay procedures for clinical microbiology evaluation of the Abbott TDX, the Abbott
laboratories. J Clin Microbiol. 1981;13:742-749. ABA200, and the Syva LAB5000 for assay of
10 Nilsson L. Correlation of bioluminescent assay of serum gentamicin. J Antimicrob Chemother.
gentamicin in serum with agar diffusion assay, 1987;19:127-33.
latex agglutination inhibition card test, enzyme 25 Oeltgen PR, Shank WA, Jr., Blouin RA, Clark T.
immunoassay, and fluorescence immunoassay. J Clinical evaluation of the Abbott TDx
Clin Microbiol. 1984;20:396-369. fluorescence polarization immunoassay analyzer.
11 Nilsson L. New rapid bioassay of gentamicin Ther Drug Monit. 1984;6:360-367.
based on luciferase assay of extracellular ATP in 26 Seifart HI, van der Bijl P, Brits DA, Kotze TJ.
bacterial cultures. Antimicrob Agents Chemother. Serum gentamicin assay: an assessment of two
1978;14:812-816. different Syva EMIT and Abbot TDx instruments.
12 Nilsson L, Hojer H, Ansehn S, Thore A. A rapid J Antimicrob Chemother. 1990;25:665-672.
semiautomated bioassay of gentamicin based on 27 Ngui-Yen JH, Doyle PW, Smith JA. Comparative
luciferase assay of bacterial adenosine analysis of two rapid, automated methods for
triphosphate. Scand J Infect Dis. 1977;9:232-236. determining aminoglycoside levels. J Clin
13 Harber MJ, Asscher AW. A new method for Microbiol. 1984;20:962-965.
antibiotic assay based on measurement of 28 Pohlod DJ, Saravolatz LD, Somerville MM.
bacterial adenosine triphosphate using the firefly Comparison of enzyme-multiplied immunoassay
bioluminescence system. J Antimicrob technique with fluorescence polarization
Chemother. 1977;3:35-41. immunoassay for determination of gentamicin
14 Lewis JE, Nelson JC, Elder HA. and tobramycin levels in serum. J Clin Microbiol.
Radioimmunoassay of an antibiotic: gentamicin. 1984;20:866-868.
Nat New Biol. 1972;239:214-216. 29 Andrews JM, Wise R. A comparison of the
15 Mahon WA, Ezer J, Wilson TW. homogeneous enzyme immunoassay and
Radioimmunoassay for measurement of polarization fluoroimmunoassay of gentamicin. J
gentamicin in blood. Antimicrob Agents Antimicrob Chemother. 1984;14:509-520.
Chemother. 1973;3:585-589. 30 Araj GF, Khattar MA, Thulesius O, Pazhoor A.
16 Casley DJ. An improved radioimmunoassay for Measurements of serum gentamicin
serum gentamicin levels using 125I-labelled concentrations by a biological method,
gentamicin. Clin Exp Pharmacol Physiol. fluorescence polarization immunoassay and
1977;4:525-532. enzyme multiplied immunoassay. Int J Clin
17 Case RV, Mezei LM. An enzymatic radioassay Pharmacol Ther Toxicol. 1986;24:542-545.
for gentamicin. Clin Chem. 1978;24:2145-2150.
631
31 Henderson DR, Friedman SB, Harris JD, 40 Fourmy D, Recht MI, Blanchard SC, Puglisi JD.
Manning WB, Zoccoli MA. CEDIA, a new Structure of the A site of Escherichia coli 16S
homogeneous immunoassay system. Clin Chem. ribosomal RNA complexed with an
1986;32:1637-1641. aminoglycoside antibiotic. Science.
32 Anhalt JP. Assay of gentamicin in serum by high- 1996;274:1367-1371.
pressure liquid chromatography. Antimicrob 41 Davies J, Wright GD. Bacterial resistance to
Agents Chemother. 1977;11:651-655. aminoglycoside antibiotics. Trends Microbiol.
33 Lecaroz C, Campanero MA, Gamazo C, Blanco- 1997;5:234-240.
Prieto MJ. Determination of gentamicin in 42 Hon WC, McKay GA, Thompson PR, Sweet RM,
different matrices by a new sensitive high- Yang DS, Wright GD et al. Structure of an
performance liquid chromatography-mass enzyme required for aminoglycoside antibiotic
spectrometric method. J Antimicrob Chemother. resistance reveals homology to eukaryotic protein
2006;58:557-563. kinases. Cell. 1997;89:887-895.
34 Nicolau DP, Wu AH, Finocchiaro S, Udeh E, 43 Daigle DM, McKay GA, Wright GD. Inhibition
Chow MS, Quintiliani R et al. Once-daily of aminoglycoside antibiotic resistance enzymes
aminoglycoside dosing: impact on requests and by protein kinase inhibitors. J Biol Chem.
costs for therapeutic drug monitoring. Ther Drug 1997;272:24755-24758.
Monit. 1996;18:263-266. 44 Westbrock-Wadman S, Sherman DR, Hickey MJ,
35 Nicolau DP, Freeman CD, Belliveau PP, Coulter SN, Zhu YQ, Warrener P et al.
Nightingale CH, Ross JW, Quintiliani R. Characterization of a Pseudomonas aeruginosa
Experience with a once-daily aminoglycoside efflux pump contributing to aminoglycoside
program administered to 2184 adult patients. impermeability. Antimicrob Agents Chemother.
Antimicrob Agents Chemother. 1995;39:650-655. 1999;43:2975-2983.
36 Ferriols-Lisart R, Alos-Alminana M. 45 Olbricht CJ, Fink M, Gutjahr E. Alterations in
Effectiveness and safety of once-daily lysosomal enzymes of the proximal tubule in
aminoglycosides: a meta-analysis. Am J Health gentamicin nephrotoxicity. Kidney Int.
Syst Pharm. 1996;53:1141-1150. 1991;39:639-646.
37 Barza M, Ioannidis JP, Cappelleri JC, Lau J. 46 Ward DT, McLarnon SJ, Riccardi D.
Single or multiple daily doses of Aminoglycosides increase intracellular calcium
aminoglycosides: a meta-analysis. BMJ (Clin levels and ERK activity in proximal tubular OK
Res). 1996;312:338-345. cells expressing the extracellular calcium-sensing
38 Montie T, Patamasucon P. Aminoglycosides: the receptor. J Am Soc Nephrol. 2002;13:1481-1419.
complex problem of antibiotic mechanisms and 47 Ward DT, Maldonado-Perez D, Hollins L,
clinical applications. Eur J Clin Microbiol Infect Riccardi D. Aminoglycosides induce acute cell
Dis. 1995;14:85-87. signaling and chronic cell death in renal cells that
39 Mingeot-Leclercq MP, Glupczynski Y, Tulkens express the calcium-sensing receptor. J Am Soc
PM. Aminoglycosides: activity and resistance. Nephrol. 2005;16:1236-1244.
Antimicrob Agents Chemother. 1999;43:727-737.
632
Table 1: Commonly Used Aminoglycoside Antibiotics

Merck Route of
Discovery Molecular Molecular Index Administration*
Name Date Formula Mass Number
Streptomycin 1944 C21H39N7O12 581.58 8685 IM
Neomycin 1949 6300 PO
Neomycin A C12H26N4O6322.36 6283
Neomycin B C23H46N6O13 614.65 6300
Neomycin C C23H46N6O13 614.65 6300
Kanamycin 1957 5118 IM,IV
Kanamycin A C18H36N4O11 484.50 5118
Kanamycin B C18H37N5O10 483.52 5118
Kanamycin C C18H36N4O11 484.50 5118
Gentamicin 1963 4251 IM,IV
Gentamicin A C18H36N4O11 484.50 4251
Gentamicin C1 C21H43N5O7477.60 4251
Gentamicin C2 C20H41N5O7463.57 4251
Gentamicin C1a C19H39N5O7449.55 4251
Tobramycin 1967 C18H37N5O9467.54 9318 IM,IV
Amikacin 1972 C22H43N5O13 585.62 405 IM,IV
Netilmicin 1976 C21H41N5O7475.60 6322 IM,IV
* IM, Intramuscular; PO, by mouth; IV, intravenous.

Neomycin is no longer used parenterally because of toxicity.
633
Table 2: Gentamicin Methods Summary Table

Method 1: Microbiological assay (agar plate diffusion)
Principle of analysis: Diameter of bacterial growthinhibition zone is proportional to gentamicin concentrations.
Advantages: Inexpensive, easy to set up, and requires no instrumentation
Disadvantages: Long turnaround time (4 to 18 hrs), variable accuracy, interfered by other antibiotics
Method 2: Bioluminescent assay; photometry
Principle of analysis: Dose-dependent effects of gentamicin on bacterial ATP levels that are measured upon addition
of luciferase
Advantages: Inexpensive, easy to set up
Disadvantages: Interfered by other antibiotics
Method 3: Radioimmunoassay (RIA); radiometry
Principle of analysis: Competitive binding of radio-labeled gentamicin and unknown for antibody
Advantages: Sensitive, specific, accurate
Disadvantages: Labor-intensive, establishment of calibration curves every run, centrifugation needed, specific
instrumentation, not appropriate in rapid response and core laboratories
Method 4: Radioenzymatic assay (REA); radiometry
Principle of analysis: Enzymatic transfer of radio-labeled substrate to gentamicin
Disadvantages: Labor-intensive, establishment of calibration curves every run, centrifugation needed, special
instrument, not appropriate in rapid-response and core laboratories
Method 5: Particle-enhanced turbidimetric inhibition immunoassay (PETINIA); turbidimetry
Principle of analysis: Competitive binding of particle-labeled gentamycin and unknown for antibody
Advantages: Sensitive, specific, accurate, automated instrumentation
Disadvantages: Expensive instrumentation, not suitable for low workload
Method 6: Fluorescence-polarization immunoassay (FPIA); polarization analyzer
Principle of analysis: Competitive binding of fluorescein-labeled gentamicin and unknown for antibody
Method 7: Enzyme-multiplied immunoassay technique (EMIT); spectrophotometry
Principle of analysis: Competitive binding of enzyme-labeled gentamicin and unknown for antibody
Method 8: Cloned-enzyme donor immunoassay (CEDIA); spectrophotometry
Principle of analysis: Competitive binding of enzyme-donor-labeled gentamicin and unknown for antibody;
reassembly of enzyme donor and receptor gains the full enzyme activity
Method 9: Direct chemiluminescence immunoassay; photometry
Principle of analysis: Competitive binding of acridinium ester-labeled gentamicin and unknown for antibody
Disadvantages: Moderate to expensive equipment
Method 10: Fluorescence immunoassay; spectrophotometry
Principle of analysis: Competitive binding of fluorogenic substrate-labeled gentamicin and unknown for antibody
Disadvantages: Moderate to expensive equipment
Method 11: High-performance liquid chromatography (HPLC); chromatography, fluorometry
Principle of analysis: Chromatographic separation followed by derivatization of gentamicin to fluorescent product
for detection
Advantages: Sensitive, highly specific, accurate, versatile; can handle small workloads
Disadvantages: Not appropriate for large workloads, expensive equipment, labor-insensitive sample preparation
634
Table 3: Traditional Multiple-Daily Doses and Desired Serum Concentrations for Aminoglycosides
Gentamicin, Tobramycin
Loading dose*: 2-3 mg/kg
Maintenance dose*: 1.7 mg/kg
Dosing interval*: every 8 hrs
Peak: 5-8 g/mL
Trough: <2 g/mL
Toxic peak range: >10 g/mL
Toxic trough range: >2 g/mL
Amikacin
Loading dose*: 7.5 mg/kg
Peak: 15-25 g/mL
Trough: 5-10 g/mL
Toxic range peak: >30-55 g/mL
Toxic range trough: >10 g/mL
Netilmicin
Peak: 4-10 g/mL
Trough: <2 g/mL
Streptomycin
Loading dose*: 7.5 mg/kg
Maintenance dose*: 7.5
Peak: 15-30 g/mL
Trough: 5-10 g/mL
* For adults with normal renal function.
Peak concentration determined 1 hr after intramuscular dose or 30-45 min after intravenous dose.
Trough determined 30 min before the next dose.
Table 4: Consolidated Once-Daily Doses and Desired Serum Concentrations for Aminoglycosides
Gentamicin, Netilmicin, and Tobramycin
Serum concentration for dosing every 24 hrs: <3 g/mL
Trough: 0.5-1.0 g/mL
Amikacin
Loading dose*: 15 mg/kg
Serum concentration for dosing every 24 hrs: <8 g/mL
Trough: <5 g/mL
* For adults with normal renal function.

Serum concentration determined 12 hrs after the start of the 60-min transfusion.
Trough determined 30 min before the next dose.
From Gonzalez LS 3rd, Spencer JP [3]. Am Fam Physician. 1998;58:1811
635
Table 5: Relative Aminoglycoside Toxicity

Relative Ototoxicity Relative
Aminoglycoside Vestibular Cochlear Nephrotoxicity
Streptomycin 3+ 1+ 2+
Neomycin 1+ 3+ 3+
Kanamycin 1+ 2+ 2+
Gentamicin 2+ 1+ 2+
Tobramycin 1-2+
Amikacin 2+ 1+ 2+
Netilmicin 1-2+
From Korzeniowski OM, Hook EW. In: Mandell GL, Douglas RG, Bennet JE, eds. Principles and Practice of Infectious
Diseases. New York: Wiley; 1979. .
Table 6: Risk Factors Predisposing Aminoglycoside Toxicity

Potentially Alterable Factors
Use of diuretics
Radiographic contrast exposure
Effective circulating volume depletion
Use of ACE inhibitors
Use of NSAIDs
Use of other nephrotoxic medications
Concomitant use of amphotericin (Fungizone IV)
Use of cisplatin (Platinol)
Unalterable Factors
Age
Preexisting renal diseases
From Gonzalez LS 3rd, Spencer JP [3]. Am Fam Physician 1998;58:1811.

636
METHODS in CLINICAL
CHEMISTRY

Editors
Peter E. Hickman, MB BS, PhD, FRCPA
Methods Editor
Associate Professor
Australian National University Medical School;
Director of Chemical Pathology
The Canberra Hospital
Australian Capital Territory, Australia

Methods Associate Editor
Principal Scientist ACT Pathology
Adjunct Professional Associate
University of Canberra
ACT Pathology
The Canberra Hospital,
Garran, Australian Capital Territory, Australia
A work of 144 Methods of Analysis describing current methodology.
*Published by Mosby, an affiliate of Elsevier; 2009

637
METHODS in CLINICAL
CHEMISTRY

Volume II
638
Pesce Kaplan Publishers

Methods Clinical Chemistry
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Permissions may be sought directly
from Pesce Kpalan Publishers: phone: (1) 212 864 4403 or 858 278 4587 (US); e-mail:
http://www.pescekaplan.com/contact_us.htm#Feedback%20Form.
Notice
Knowledge and best practice in this field are constantly changing. As new research and experience broaden
our knowledge, changes in (i) on state-of-the-art methodology and measurement technique, (ii) effect of
interferences, or (iii) interpretation of results may become necessary or appropriate. Readers are advised
to check the most current information in the relevant scientific literature. It is the responsibility of
the clinical laboratory practitioner, relying on their own experience and knowledge of the patient, to
determine the best procedure to take all appropriate safety precautions. To the fullest extent of the law,
neither the Publisher nor the Editors assumes any liability for any injury and/or damage to persons or
property arising out of or related to any use of the material contained in this book.
The Publisher
Previous copyrighted printed editions: 1987, electronic editions 1996, 2003
Library of Congress Cataloging-in-Publication Data

Methods Clinical chemistry [edited by] Lawrence A. Kaplan, Amadeo J. Pesce.
Bound version and electronic version.
Printed in the United States of America

639
Contributors to Methods of Analysis

Zakaria Ahmed, PhD Department of Pathology
Clinical Chemist Johns Hopkins School of Medicine
Department of Pathology and Laboratory Baltimore, Maryland
Medicine
Rochester General Hospital Paul F. Coleman, PhD
Rochester, New York Research Fellow
Infectious Disease Core R&D
Hassan M. E. Azzazy, PhD, DABCC, FACB Abbott Diagnostics
Chairman and Associate Professor Abbott Park, Illinois
Department of Chemistry
The American University in Cairo Joe DAgostino, BSc, Grad Dip FMI,
Cairo, Egypt MAACB
Senior Scientist
Tony Badrick, BAppSc, BSc, BA, MLitSt Clinical Biochemistry Unit
(Math), MBA, PhD, FAIMS, FAACB, FQSA, Alfred Pathology Service
FAIM, FACB, FRCPA (Hon) Alfred Hospital, Melbourne, Australia
Executive ManagerLaboratories
Sullivan Nicolades Pathology Sheila Dawling, PhD, CChem, FRSC
Taringa, Queensland, Australia Associate Professor of Pathology
Director, Toxicology
John Beilby, BSc(Hons), PhD, FAACB, TDM Laboratory,
MHGSA, ARCPA Associate Director
Principal Scientist Clinical Chemistry
PathWest The Vanderbilt Clinic
Nedlands, Western Australia, Australia Nashville, Tennessee
Marion Black, BSc(Hons), Dip. Ed, MAACB Joris R. Delanghe

Senior Scientist Professor of Clinical Chemistry
Clinical Biochemistry, Alfred Pathology Department of Clinical Chemistry
Service Ghent University Hospital
Victoria, Australia Gent, Belgium
John R. Burnett, MB ChB, MD, PhD, Goce Dimeski, BSc

FRCPA, FAHA Supervising Scientist
Clinical Professor Pathology QLD
PathWest Laboratory Medicine Princess Alexandra Hospital
Royal Perth Hospital Woolloongabba, Brisbane, Australia
Perth, Western Australia, Australia
Angela Ferguson, PhD
Kevin Carpenter, PhD, FHGSA Clinical Chemistry Fellow
Principal Scientist and Head of Washington University School of
Department Medicine
NSW Biochemical Genetics Service Department of Pathology and Immunology
The Childrens Hospital at Westmead St. Louis, Missouri
Westmead, New South Wales, Australia
Michael J. Figursk, PhD
Kee Cheung, BSc(Hons), PhD, GradCert. Research Associate
Mgt. University of Pennsylvania
Manager Hospital of University of Pennsylvania
Pathology QueenslandPrincess Alexandra Philadelphia, Pennsylvania
Hospital
Woolloongabba, Queensland, Australia John Galligan
Supervising Scientist
William Clarke, PhD, MBA, DABCC Pathology Queensland Central Laboratory
Director, TDM and Toxicology, Director, Royal Brisbane Hospital
CPOCT Herston, Queensland, Australia
640
Atholl Johnston, BSc, MSc, PhD,

Karen Golemboski, PhD, MT(ASCP) FBPharmacolS, FRCPath
Program Director and Chair Professor of Clinical Pharmacology
Clinical Laboratory Science William Harvey Research Institute
Bellarmine University (School of Medicine and Dentistry)
Louisville, Kentucky Queen Mary
University of London
Ronda F. Greaves, BSc, Grad Dip Ed, London, United Kingdom
MAppSc, PhD, MAACB
Senior Scientist Graham Jones, MBBS, DPhil, FRCPA, FAACB
Complex Biochemistry Department Staff Specialist in Chemical Pathology
The Royal Childrens Hospital Department of Chemical Pathology
RCH Laboratory Services St. Vincents Hospital
Parkville, Victoria, Australia Darlinghurst, New South Wales,
Australia
Kathryn Green, BSc(Hons), MSc(Med)
Senior Scientist Saeed A. Jortani, PhD, DABCC, FACB
NSW Biochemical Genetics Service Associate Professor of Pathology and
The Childrens Hospital at Westmead Laboratory Medicine
Westmead, New South Wales, Australia University of Louisville
School of Medicine
Elizabeth M. Hall, BSc, MSc, FRCPath Louisville, Kentucky
Principal Clinical Scientist
Department of Clinical Biochemistry Lawrence A. Kaplan, PhD, DABCC
East Kent Hospitals University New York, New York
NHS Trust
Kent and Canterbury Hospital Steven C. Kazmierczak, PhD, DABCC
Canterbury, Kent, United Kingdom Professor of Pathology
Director of Clinical Chemistry and
Peter E. Hickman, MB BS, PhD, MPH, Toxicology
MAACB, FRCPA Oregon Health and Science University
Associate Professor Department of Pathology
Australian National University Medical Portland, Oregon
School;
Director of Chemical Pathology Sandra Klingberg, B App Sci
The Canberra Hospital Supervising Scientist, Protein
Australian Capital Territory, Australia Laboratory
Pathology Queensland, Central
Gregory A. Hobbs, PhD, DABCC Laboratory
Clinical Laboratory Science Royal Brisbane Hospital
Bellarmine University Herston, Queensland, Australia
Louisville, Kentucky
David W. Holt, BSc, PhD, DSc(Med), Principal Scientist ACT Pathology
CSci, EurClin Chem, FESC, FRCPath Adjunct Professional Associate
Professor of Bioanalytics University of Canberra
ASI, Ltd ACT Pathology
London, United Kingdom The Canberra Hospital
Garran, Australian Capital Territory,
David G. Hughes, BAppSci, Grad Dip Sci Australia
Scientist
Clinical Chemistry, ACT Pathology Magdalena Korecka, PhD
The Canberra Hospital Senior Research Investigator
Australia University of Pennsylvania
Mind Jin, PhD
Temple University
641
William J. Korzun, PhD, DABCC, MT(ASCP) Brett McWhinney, BSc, MSc, MBA, MPhil
Associate Professor Supervising Scientist
Department of Clinical Laboratory HPLC Section, Department of Chemical
Sciences Pathology
Virginia Commonwealth University Royal Brisbane Hospital
Richmond, Virginia Herston, Queensland, Australia
Edmund Lamb, PhD, FRCPath Danni L. Meany, PhD

Clinical Scientist (Biochemistry) and Clinical Chemistry Fellow
Head of Department Johns Hopkins Medical Institutions
Department of Clinical Biochemistry Baltimore, Maryland
East Kent Hospitals
NHS Trust James J. Miller, PhD, DABCC, FACB
Kent and Canterbury Hospital Professor
Canterbury, Kent, United Kingdom University of Louisville
School of Medicine
Stanley S. Levinson, PhD, DABCC Department of Pathology and Laboratory
Professor of Pathology and Laboratory Medicine
Medicine Louisville, Kentucky
University of Louisville
Director of Clinical Chemistry and Michael Milone, MD, PhD
Immunochemistry Assistant Professor of Pathology and
Department of Veteran Affairs Medical Laboratory Medicine
Center Associate Director, Toxicology
Louisville, Kentucky Laboratory
School of Medicine
Barry Lewis, MD, FRCPA, FHGSA University of Pennsylvania
Head, Department of Clinical Philadelphia, Pennsylvania
Biochemistry
PathWest Laboratory Medicine WA Gerald J. Mizejewski, BS, MS, PhD
Princess Margaret Hospital Senior Research Scientist
Perth, Western Australia, Australia Wadsworth Center
New York State Department of Health
Jinong Li, PhD Albany, New York
Clinical Chemistry Fellow
Johns Hopkins Medical Institutions Scott A. Muerhoff, PhD
Baltimore, Maryland Volwiler Research Fellow
Infectious Diseases Research and
Greg Maine, PhD Development
Manager, Global Scientific Affairs Abbott Diagnostics
Associate Research Fellow Abbott Laboratories
Abbott Laboratories Abbott Park, Illinois
Abbott Park, Illinois
Anthony O. Okorodudu, PhD, MBA
Christopher R. McCudden, PhD, DABCC, Professor
NRCC Director, Clinical Chemistry Division
Assistant Professor UTMB/CMC Outreach Laboratory Services
Department of Pathology and Laboratory Department of Pathology
Medicine University of Texas Medical Branch
School of Medicine Galveston, Texas
University of North Carolina
Chapel Hill, North Carolina
Denise A. McKeown, MSci, AMRSC

Senior Analyst
St Georges, University of London
Analytical Unit
Department of Cardiac and Vascular
Sciences
642
Peter OLeary, BSc, MAACB, AFACHSE, Terry Pry, PhD

ARCPA, PhD Retired-Manager Scientific Affairs,
Adjunct Professor, School of Public Asia-Pacific
Health (Curtin) Abbott Diagnostic Division
Adjunct Associate Professor Abbott Laboratories
School of Womens & Infants Health Auckland, New Zealand
(UWA),
Director, Office of Population Health Kishor Raja, BSc(Hons), MSc, PhD, CSci
Genomics Principal Clinical Scientist/Honorary
Public Health Division Senior Lecturer
Health Department of Western Australia Clinical Biochemistry Department
Perth, Western Australia, Australia Kings College Hospital
Matthew T. Olson, MD
House Officer, Department of Pathology Jordan Reynolds, MD
Johns Hopkins Medical Institutions Resident Physician
Baltimore, Maryland University of Cincinnati
Department of Pathology and Laboratory
Felix O. Omoruyi, PhD Medicine
Fellow Cincinnati, Ohio
Department of Pathology
Clinical Chemistry Ken Robertson, BSc, AAIMS
University of Texas Medical Branch Senior Scientist in Charge (Research)
Galveston, Texas PathWest Laboratory Medicine
Royal Perth Hospital
Mauro Panteghini, MD Wellington St. Perth, Western
Professor of Clinical Biochemistry and Australia, Australia
Clinical Molecular Biology
University of Milan Medical School Andrea M. Rose, PhD, MBA
Laboratorio Analisi Senior Clinical Support Consultant
Milan, Italy Roche Diagnostics
Indianapolis, Indiana
Gerardo Perotta, MPA
Interim-Coordinator, Pathology Enrico Rossi, PhD, MAACB
Education Research Biochemist
Department of Pathology and Laboratory PathWest Laboratory Medicine
Medicine Nedlands, Western Australia, Australia
University of Cincinnati College of
Medicine Randal J. Schneider, MS, PhD
Cincinnati, Ohio Director of Clinical Chemistry and
Toxicology
Michael A. Pesce, PhD ProHealth Care Laboratories
Professor Emeritus of Pathology and Waukesha, Wisconsin
Cell Biology
Columbia University Medical Center Les Shaw, BS, PhD
New York Presbyterian Hospital Professor
New York, New York University of Pennsylvania
School of Medicine
Julia M. Potter, B Med Sc(Hons), MB BS, Philadelphia, Pennsylvania
PhD, FRCPA
Professor of Pathology Run Zhang Shi, PhD
Australian National University Medical Instructor
School Assistant Director of Clinical
Executive Director, ACT Pathology Chemistry and Immunology
The Canberra Hospital Department of Pathology
Australia Stanford University
Stanford, California
643
Ravinder Jit Singh, PhD John G. Toffaletti, PhD

Co-Director, Endocrine Laboratory Professor in Pathology
Organization Clinical Laboratories and Department of
Mayo Clinic Pathology
Rochester, Minnesota Duke University Medical Center
Durham, North Carolina
Patricia Slev, PhD
Assistant Professor of Pathology Susan Vickery, MSc, PhD
(Clinical) Senior Clinical Scientist
University of Utah, East Kent Hospitals University
Medical Director NHS Trust
Serologic Hepatitis and Retrovirus Kent and Canterbury Hospital
Laboratory Canterbury, Kent, United Kingdom
ARUP Laboratories
Salt Lake City, Utah Ping Wang, PhD, DABCC
Medical Director of Clinical Chemistry
Ramasamyiyer Swaminathan, MBBS, MSc, The Methodist Hospital
PhD, FRCPath Houston, Texas
Professor and Head of Department of
Chemical Pathology Gregory Ward, BSc(Hons), MSc, MAACB,
St. Thomas Hospital FAACB
London, United Kingdom Head, Biochemistry and Endocrinology
Sullivan Nicolades Pathology (Sonic
Danyal B. Syed, BSc, MA, PhD, C(ASCP), Healthcare)
CC(NRCC), DABCC, FACB Brisbane, Queensland, Australia
Laboratory Director
William F. Ryan Community Health Center Alan H. B. Wu, PhD
New York, New York Professor, Laboratory Medicine
University of California, San Francisco
Danyel H. Tacker, PhD, FACB Clinical Chemistry and Toxicology
Clinical Chemist Laboratories
Ochsner Medical CenterNew Orleans San Francisco General Hospital
New Orleans, Louisiana San Francisco, California
Jillian R. Tate, BSc(Hons), MSc Odette Youdell, BAppSci(Hons), MAACB

Senior Scientist Senior Scientist
Pathology Queensland Clinical Biochemistry
Chemical Pathology Department Alfred Pathology Service
Royal Brisbane and Womens Hospital Melbourne, Victoria
Brisbane, Queensland, Australia Australia
644
Foreword
Foreword for the 2009 edition of Methods in Clinical Chemistry
In the mid-1980s we perceived a need for an extensive, up-to-date, compilation of methods available for
use in clinical chemistry laboratories. To meet this need, we published in 1987 Methods in Clinical
Chemistry with C.V. Mosby. This volume provided not only a review of extant methodologies, but also a
critique of each method. This enabled the authors, when appropriate, to suggest one technique as a
recommended method. Since its initial publication, Methods in Clinical Chemistry has been repeatedly
updated and made available in electronic form (CD-ROM or Internet) by Pesce Kaplan Publishers.
Like the previous version, this edition is published in parallel with the current edition of our textbook,
Clinical Chemistry: Theory, Analysis and Correlation (5th edition; Elsevier, 2010). The editors of this
work, Peter Hickman and Gus Koerbin, have assembled an international group of expert clinical chemists
from the United State, Europe, and Australia/New Zealand. We have retained the scope of previous
editions, including:
144 revised method reviews of available technologies for the analysis of each analyte,
a critique of each methodology,
analytical quality goals (when available),
recent references,
a suggested procedure for manual methods.
This edition will be available in both electronic (CD-ROM) and printed (two volumes) formats. It is our
hope that this edition will be widely used and vigorously reviewed by its users. Using new software
technology, we will provide a mechanism for input from readers for future versions of this edition. We
have created an Internet site (http://www.pescekaplan.com/) where individuals can publicly post their
comments. Eventually, the Editors will redact the suggestions into changes incorporated into Methods in
Clinical Chemistry.
645
Preface for the 2009 edition of Methods in Clinical Chemistry
It was certainly an honor and privilege to be invited by Larry Kaplan and Amadeo Pesce to undertake the
editorial role for the 2009 edition of Methods in Clinical Chemistry. The challenge for us was to provide a
text that was contemporary, detailed but readable, and of paramount importance, a useful resource and
laboratory tool for both students and professionals as we near the end of the first decade of this
millennium.
Much in clinical chemistry has changed since the first edition was published in 1987: the breadth and
depth of our knowledge base, the development of laboratory technology, and the widespread use of the
internet and digital technology. Review of each chapter and final publication for this edition of Methods
in Clinical Chemistry has been vastly different from the original compilation, with Internet and digital
technology significantly streamlining the process. In 2003, the methods portion of the text was removed
from the 4th edition of Kaplan and Pesces Clinical Chemistry Theory, Analysis, Correlation and provided
as a CD-ROM. This edition includes 131 new and revised methods plus 13 older methods that were not
revised. Each method contains a critique of alternate methodologies; contemporary and historical,
analytical quality goals, and performance data. Methods in Clinical Chemistry is not only offered in a CD-
ROM/DVD format, but also in a two volume printed format. In addition, the methods will be accessible to
purchasers of the 5th edition of Kaplan and Pesces Clinical Chemistry Theory, Analysis, Correlation via
the publishers (Elsevier) Internet site, Evolve.
We have been very fortunate to collaborate with an outstanding team of 76 authors from all over the
world. Our clinical chemistry experts reside in 16 states of the United States, most states of Australia, and
in New Zealand, Egypt, Italy, Belgium and England, making this a truly international project. Without the
contributions of these industry leaders, this volume could not have been completed and to the team we say
a heartfelt thank you. It is somewhat clich to say that it takes a cast of thousands to produce a body of
work such as this, but without these authors and the help of the following individuals, we would not have
succeeded: Professor Julia Potter, Executive Director of ACT Pathology, Canberra, Australia for
supporting our undertaking of this project, Priscilla Delatorre for her patience, along with her formatting
and word processing skills, Nataliya Polyakov of the College of American Pathologists Surveys team for
her assistance in accessing proficiency data (usually at very short notice), Elseviers Ellen Wurm-Cutter
for her valued advice and particularly for her patience and tenacity in keeping us on schedule and to her
editorial assistant, Jennifer Hermes, for her behind-the-scenes contribution. No list of acknowledgements
would be complete without a mention of the respected and appreciated guidance, praise, constructive
criticisms and most of all encouragement that Larry Kaplan and Amadeo Pesce have provided throughout
this process, which commenced in early 2007.
We trust that this edition will be widely used and reviewed by you the reader. We invite you to use the
internet site (http://www.pescekaplan.com/) created to accept your valued input, comments and
suggestions so that future editions of Methods in Clinical Chemistry can respond to the ever changing
needs of the laboratory professional into the second decade of this millennium and beyond.
We intend to continue with this project, progressively updating methods and releasing a new version as
dictated by progress in the field.
646
We also intend to include more international contributors, with the hope that it would become the world
methods reference. Finding appropriate authors from diverse countries has proven to be a challenge and
we ask for recommendations from our colleagues.
.
Peter Hickman
Gus Koerbin
Australia, 2009
647
Methods
1. 25-OH-Vitamin D 24. Barbiturates* Page 205
Ravinder Jit Singh Page 17 25. Bence Jones Protein
2. 1-Antitrypsin Stanley S. Levinson Page 219
John Beilby Page 28 26. Benzodiazepines* Page 232
3. Acetaminophen 27. Beta-hCG (beta-human chorionic
Gus Koerbin, gonadotropin)
David G. Hughes, James J. Miller Page 251
Julia M. Potter Page 33 28. Beta-2-Microglobulin
4. Adrenocorticotropic Hormone (ACTH) James J. Miller Page 257
Hassan M. E. Azzazy Page 43 29. Bilirubin
5. Alanine Aminotransferase R. Swaminathan Page 261
James J. Miller Page 46 30. Blood Gas Analysis and Oxygen
6. Albumin Saturation
Kee Cheung Page 52 Goce Dimeski Page 275
7. Albumin in Urine 31. C-Reactive Protein (CRP)
Graham Jones Page 61 Odette Youdell Page 288
8. Alcohol 32. Calcium
Ping Wang Page 67 Randal J. Schneider Page 296
9. Aldolase* Page 81 33. Cancer Antigen 125 (CA 125)
10. Aldosterone Hassan M. E. Azzazy Page 304
Hassan M. E. Azzazy Page 85 34. Carbamazepine
11. Alkaline PhosphataseTotal Gus Koerbin,
Danyal B. Syed Page 90 Julia M. Potter Page 307
12. Alpha-Fetoprotein 35. Carbohydrate Antigen 15-3 (CA 15-3)
Gerald J. Mizejewski Page 101 Gregory A. Hobbs Page 317
13. Aluminum (Aluminium) 36. Carbohydrate Antigen 19-9 (CA 19-9)
Tony Badrick Page 117 Hassan M. E. Azzazy Page 320
14. Amino Acid Screening 37. Carbon Dioxide and Bicarbonate
Kevin Carpenter Page 124 William J. Korzun Page 323
15. Ammonia 38. Carcinoembryonic Antigens (CEA)
Elizabeth M. Hall Page 128 Gregory A. Hobbs Page 327
16. Amniotic Fluid Phospholipids 39. CatecholaminesPlasma
(AFPL) LS Ratio and PG Brett McWhinney Page 337
Hassan M. E. Azzazy Page 132 40. CatecholaminesUrine
17. Amylase Brett McWhinney Page 344
Ming Jin Page 150 41. Cerebrospinal Fluid (CSF) Protein
18. Angiotensin Converting Enzyme Quantitation
(ACE) Danyel H. Tacker,
Hassan M. E. Azzazy Page 157 Anthony O. Okorodudu Page 363
19. Anion Gap 42. Ceruloplasmin
Tony Badrick Page 162 Ahmed Zakaria Page 372
20. Anticonvulsive Drugs* Page 165 43. Chloride
21. Apolipoproteins A-1 and B William J. Korzun Page 377
Jillian R. Tate Page 181 44. Cholesterol
22. Aspartate Aminotransferase John R. Burnett,
James J. Miller Page 190 Ken Robertson Page 382
23. B-Type Natriuretic Peptide, 45. Cholinesterase
Danyal B. Syed Page 393
NT-proBNP, and ProBNP
Alan H.B. Wu Page 197 46. CMV (Cytomegalovirus)
Terry Pry, Greg Maine Page 405
648
47. Copper The following analytes are in Volume

K. Raja, R. Swaminathan Page 410 II
48. Cortisol
R. Swaminathan Page 420 70. Glucose
49. Creatine Kinase John Beilby Page 651
Mauro Panteghini Page 430 71. Glycated Hemoglobin
50. Creatine Kinase Isoenzymes Andrea M. Rose Page 662
Mauro Panteghini Page 436 72. Haptoglobin
51. Creatinine Joris R. Delanghe Page 670
Edmund J. Lamb Page 440 73. Hepatitis B
52. Cyclosporin (Cyclosporine A) Paul Coleman Page 674
David W. Holt, 74. Hepatitis C Virus
Denise A. McKeown, A. Scott Muerhoff Page 680
Atholl Johnston Page 452
75. High-Density Lipoprotein (HDL)
53. D-Xylose* Page 470 Cholesterol
54. Dehydroepiandrosterone and its John R. Burnett,
sulfate (DHEA and DHEA-S) Ken Robertson Page 686
Gus Koerbin Page 481 76. Holotranscobalamin
55. Digoxin and Digitoxin Marion Black Page 698
Randal J. Schneider Page 489 77. Homocysteine
56. Drug Screen Sheila Dawling Page 703
Christopher R. McCudden Page 495 78. Homovanillic Acid
57. Estradiol Lawrence A. Kaplan Page 718
Greg Ward Page 516 79. Human Immunodeficiency Virus (HIV)
58. Estriol* Page 520 Patricia Slev Page 728
59. Ethylene Glycol 80. Immunoelectrophoresis
Ping Wang Page 535 Stanley S. Levinson Page 740
60. Fecal Electrolytes and Osmolality 81. Immunoglobulin Quantitation
Felix O. Omoruyi, Karen Golemboski Page 753
Anthony O. Okorodudu Page 544 82. Insulin and C-Peptide
61. Fecal Fat and Fat Absorption Steven C. Kazmierczak Page 762
Lawrence A. Kaplan Page 547 83. Ionized Calcium
62. Fecal Occult Blood John G. Toffaletti Page 769
R. Swaminathan Page 559 84. Iron and Iron-Binding Capacity
63. Ferritin Gerardo Perrotta,
Hassan M. E. Azzazy Page 567 Jordan Reynolds Page 781
64. Folic Acid 85. Ketones
Sheila Dawling Page 573 Lawrence A. Kaplan Page 785
65. Follicle-Stimulating Hormone (FSH) 86. Lactate Dehydrogenase and Lactate
Angela Ferguson Page 586 Dehydrogenase Isoenzymes
66. Free Thyroxine and Free Mauro Panteghini Page 793
Triiodothyronine 87. Lactic Acid
Greg Ward Page 589 Steven C. Kazmierczak Page 797
67. Gamma-Glutamyl Transferase (GGT) 88. Lead
Danyal B. Syed Page 609 Gus Koerbin Page 803
68. Gastric Fluid Analysis* Page 618 89. Lipase
69. Gentamicin and Other Ming Jin Page 814
Aminoglycosides 90. Lipoprotein (a)
Danni L. Meany, Gregory A. Hobbs Page 820
91. Lithium
Danni L. Meany,
649
92. Luteinizing Hormone 115. Pyruvic Acid

Gregory A. Hobbs Page 828 Steven C. Kazmierczak Page 1019
93. Lysozyme* Page 834 116. Renin
94. Magnesium Greg Ward Page 1025
Steven C. Kazmierczak Page 840 117. Rheumatoid Factor
95. Maternal Fetal Screening Terry Pry Page 1030
Peter OLeary, 118. Rubella
Barry Lewis Page 846 Terry Pry, Greg Maine Page 1035
96. MetanephrinesUrine 119. Salicylates
Brett McWhinney Page 856 Gus Koerbin,
97. Methotrexate Julia M. Potter Page 1043
Michael A. Pesce Page 865 120. Serum Protein Electrophoresis
98. Methylmalonic Acid Sandra Klingberg Page 1052
Kevin Carpenter, 121. Sirolimus
Kathryn Green Page 875 Magdalena Korecka,
99. Mycophenolic Acid Michael Milone,
Michal J. Figurski, Leslie M. Shaw Page 1062
Magdalena Korecka, 122. Sodium and Potassium
Leslie M. Shaw Page 880 William J. Korzun Page 1070
100. Myoglobin 123. Steroid Hormone
Alan H.B. Wu Page 892 Receptors* Page 1078
101. Opiates* Page 899 124. Sweat Electrolytes: The Sweat Test
102. Organic Acid Screening Ronda F. Greaves Page 1104
Kevin Carpenter Page 912 125. T3 Uptake
103. Osmolality Run Zhang Shi Page 1115
Goce Dimeski Page 916 126. Tacrolimus
104. Oximetry Michael C. Milone,
Goce Dimeski Page 921 Michal Figurski,
105. Parathyroid Hormone (Parathyrin) Magda Korecka,
(PTH) Leslie M. J. Shaw Page 1120
Susan Vickery, 127. Testosterone
Edmund J. Lamb Page 934 Greg Ward, Gus Koerbin,
106. Phenylalanine* Page 939 Peter E. Hickman Page 1127
107. Phenytoin 128. Theophylline and Caffeine
Gus Koerbin Page 946 Saeed A. Jortani Page 1140
108. Phosphorus and Phosphate 129. Thyroglobulin (Tg)
Steven C. Kazmierczak Page 957 Run Zhang Shi Page 1151
109. Plasma Free Metanephrines 130. Thyroid Autoantibodies
Brett McWhinney Page 964 Run Zhang Shi Page 1158
110. Porphobilinogen Screening and 131. Thyroid-Stimulating Hormone (TSH)
Quantitation Greg Ward Page 1168
Enrico Rossi Page 974 132. Thyroxine (Total)
111. Procainamide and N- Greg Ward Page 1175
Acetylprocainamide 133. Total Serum Protein
Matthew T. Olson, Kee Cheung Page 1182
William Clarke Page 982 134. Transferrin and Carbohydrate-
112. Progesterone Deficient Transferrin
John Galligan Page 991 Sandra Klingberg Page 1191
113. Prolactin 135. Transthyretin (Prealbumin)
Sheila Dawling Page 996 Danyel H. Tacker,
114. Prostate Specific Antigen (PSA) Anthony O. Okorodudu Page 1201
Hassan M. E. Azzazy Page 1013 136. Tricyclic Antidepressants
Jinong Li, William Clarke Page 1206
650
137. Triglycerides 141. Urine Porphyrin Quantitation

John R. Burnett, Enrico Rossi Page 1258
Ken Robertson Page 1213 142. Urine Protein, Total
138. Troponins Susan Vickery,
Jillian R. Tate, Edmund J. Lamb Page 1264
Mauro Panteghini Page 1224 143. Vitamin B12
139. Urea Joe DAgostino Page 1270
Elizabeth M. Hall Page 1246 144. Zinc
140. Uric Acid Tony Badrick Page 1276
Elizabeth M. Hall Page 1252
*Updated in the last edition.
651
Glucose
Glucose
John Beilby
Name: Glucose, dextrose
Clinical significance: Refer to Chapter 38, Diabetes, in the 5th edition of Clinical
Merck Index: 4319
Chemical class: Carbohydrate
Structure:
NRSCL Definitive and Reference Methods: NCCLS RS1-A

i commonly used to measure glucose in body fluids.
Survey data from the 2007 College of American
Chemically, glucose is an aldohexose, with the Pathologists (CAP) Participant Summary Report
aldehyde form in equilibrium with the showed that 58% of the participating laboratories
glucopyranose form (which is shown above). The used hexokinase-based methods, with 92% of these
latter is the favored structure at physiological pH. laboratories measuring the product at 340 nm.
The aldehyde/enediol equilibrium allows glucose to Glucose oxidasebased methods were used by 42%
be reduced and oxidized with facility. of laboratories. Of these laboratories, 60% used a
glucose oxidaseoxygen electrode combination.
All modern methods for glucose analysis employ Less than 1% of laboratories used glucose
enzymes as reagents to increase analytical dehydrogenase methods.
specificity. The enzymes used are the hexokinase,
glucose oxidase, and glucose dehydrogenase Whole blood samples, anticoagulated with lithium
reactions. All method types have been automated, heparin are frequently analyzed for glucose as part
with resulting high specificity and precision. A of the profile on blood-gas instruments. These
specimen blank is used to correct for interfering instruments use a glucose oxidaseencased oxygen
substances that absorb at 340 nm. There are many electrode method. They can be laboratory-based or
suppliers of commercial reagents, and therefore the used for point-of-care testing [1] in, for example,
general principles are described here. emergency rooms and intensive care units.
Nonlaboratory-based methods are finger-prick
Current Methods In Use Hexokinase- and blood methods for monitoring glucose
glucose oxidasebased methods are the two most concentrations in diabetic patients and urine
dipsticks for semi-quantitative measurement of
glucose.
i
Glucose
Previous and current authors of this method: Hexokinase Methods
First edition: Lawrence A. Kaplan The hexokinase method involves two coupled
Methods edition: Lawrence A. Kaplan reactions. The hexokinase reaction phosphorylates
Second edition: Lawrence A. Kaplan hexoses with ATP and Mg2+. Glucose 6-phosphate
Third edition: Steven C. Kazmierczak dehydrogenase (G6PD) reacts specifically with
Fourth edition: Lawrence A. Kaplan glucose-6-phosphate produced by the hexokinase
Fifth edition: John Beilby reaction to yield 1 mole of nicotinamide adenine
652
Glucose
dinucleotide phosphate (NADP+ or NAD+) for each 2H2O2 + 4-aminoantipyrine + 1,7-

mole of glucose that is oxidized. The amount of
dihydroxynaphthalene peroxidase
NADPH produced is directly proportional to the
amount of glucose in the sample and is measured
by absorbance at 340 nm. G6PD derived from yeast red dye
is used with the cofactor NADP+. Most assays use
an enzyme derived from bacteria that employs
NAD+ as a substrate rather than the NADP+ used Many POCT analyzers, including blood-gas
by mammalian enzymes. The advantage of using instruments, measure glucose concentrations using
the bacterial enzyme is that red blood cell G6PD the GO method. In these instruments, a silver
and 6-phosphogluconate dehydrogenase, which use cathode and a platinum anode are surrounded by an
NADP as a substrate, do not interfere in the electrolyte solution and a multi-layer membrane
analysis of glucose. This reduces the interference in mounted at the tip. The membrane consists of an
the assay resulting from hemolysis. Although other outer membrane permeable to glucose, a middle
hexoses can enter into the hexokinase reaction, enzyme layer which contains immobilized glucose
normal serum concentrations of these sugars do not oxidase, and an inner membrane layer permeable to
cause significant interference. H2O2. The H2O2 is transported across the inner
membrane to the platinum anode at a negative
Glucose + ATP hexokinase potential of 675 mV and is oxidized. The electric
current produced is proportional to the amount of
glucose-6-phosphate + ADP H2O2, which in turn is directly related to the
amount of glucose in the sample. To complete the
glucose-6- electrical circuit, a reaction at the cathode converts
Glucose-6-P + NADP+ (NAD+) phosphate two Ag+ (from AgCl) to Ag for each molecule of
H2O2 oxidized.
dehydrogenase
Glucose
6-phosphogluconate + NADPH (NADH) + H+ Glucose + O2 gluconic acid + H2O2
H2O2 2H + O2 + 2e-
+
The hexokinase assay is not commonly performed
as a kinetic procedure because of the requirement Glucose Dehydrogenase Methods
for a rapid initial reading. However, it is often In 1975, a relatively pure form of glucose
performed as an end-point reaction using dehydrogenase (GDH) was isolated [2]. This
bichromatic measurements to help correct for allowed the development of a glucose method
interferences. based solely on this highly specific enzymatic
reaction:
Glucose Oxidase Methods glucose dehydrogenase
The glucose oxidase method uses two coupled -D-glucose + NAD+
+
enzyme reactions. In this case, the initial reaction is D-glucono--lactone + NADH + H
the specific one, and the indicator reaction is
nonspecific. The first reaction employs glucose Mutarotase is added to convert glucose from the
oxidase (GO) to oxidize glucose to glucuronic acid to the form of glucose. The reaction is monitored
and hydrogen peroxide. Glucose in solution is in at 340 nm as either a kinetic or an end-point
the ratio of 36% in the form and 64% in the reaction. Glucose dehydrogenase isolated from
form. Since GO is highly specific for -D-glucose, Bacillus cereus is highly specific for glucose, and
glucose requires mutarotation to the form, and the results produced are in close agreement with
most preparations of GO contain the enzyme the hexokinase method. This GD method has been
mutarotase to catalyze the conversion of -D- adapted for use on several automated instruments
glucose to the form. The hydrogen peroxide from [3,4].
the GO reaction can be measured by several
different techniques. In one method using dry Measurement of Urine Glucose
chemistry techniques (Vitros 250, Ortho Clinical Urine dipsticks are widely used to screen for
Diagnostic), the hydrogen peroxide reacts with a glucose in the urine. All strips use glucose oxidase
non-colored dye precursor to produce a colored with a chromogenic assay. An example of a
dye. The intensity of the dye is monitored reaction used in this method is to detect the H2O2
photometrically at 540 nm using reflected light. produced from the GO reaction, where potassium
iodide acts as a final oxygen acceptor, resulting in a
change of color.
-D-glucose + O2 glucose oxidase
D-gluconic acid + H2O2
653
Glucose
glucose oxidase reagent containing hexokinase, G6PD, NAD+, and

Glucose + 2H2O + O2 ATP until the reaction is complete. The final step is
to measure the amount of NADH produced.
glucuronic acid + 2H2O2 Blanks, quality control samples, and calibrators are
carried through the entire procedure.
H2O2 + 2KI peroxidase Glucose concentrations are commonly used for the
I2 (green to brown) + KOH diagnosis and monitoring of diabetes, and all the
methods need to be precise and accurate, especially
since cutoff values used for the diagnosis of
Other compounds, such as 3,3,5,5- diabetes must apply to all methods.
tetramethylbenzidine and 1,7-
dihydroxynaphthalene, whose oxidized forms are All of the enzymatic methods discussed under
colored, are also used. However, testing for urine Current Methods (above) show adequate
glucose cannot be used for the diagnosis or performance.
monitoring of diabetic patients for the following
reasons: Specimen Collection and Storage
1. Urine dipstick methods are semi-quantitative The choice of specimen used for glucose
techniques that have a low sensitivity for the determination depends on the analytical method to
diagnosis of diabetes and are not be used. Serum or plasma, free of hemolysis, is the
recommended for routine care of patients with specimen of choice for automated enzymatic
diabetes or as a screening tool for diabetes methods. Heparinized whole blood specimens can
[5,6]. be used on instruments using ion-selective
2. The renal threshold, the blood glucose level at electrodes. However, the glucose concentration in
which glucose is detected in urine, averages whole blood is approximately 12% to 15% lower as
160 to 180 mg/dL (8.9 to 10.0 mmol/L) but compared with plasma because of the higher water
varies widely between individuals. It can be content of plasma. Plasma is recommended for the
lower in children and during pregnancy and diagnosis of diabetes, since the diagnostic cutoff
higher in longstanding diabetes. points have been derived using plasma samples.
3. A negative test does not distinguish between Glucose concentrations in heparinized plasma are
euglycemia or hypoglycemia. approximately 5% lower than serum [10]. Other
4. The method is not specific for glucose and body fluids, such as cerebrospinal fluid (CSF),
may suffer from interferences by fluid intake, pleural fluid, and urine can also be analyzed. CSF
ascorbic acid, and urinary tract infections. may contain bacteria or excess numbers of cells
that metabolize glucose and should be analyzed
Reference and Preferred Methods immediately after collection. If a delay is
The definitive method for glucose determination is unavoidable, the sample must be stored at either
isotope dilution mass spectrometry (ID-MS). This 4C or frozen at 20C or collected into a
technique measures the analyte concentration with fluoride/oxalate specimen tube. The CSF must be
the greatest accuracy, so the systematic errors are clear for assay. Centrifugation may be necessary if
negligible, and there is a high level of precision the sample is not clear.
with the coefficient of variation (CV) is < 0.5%.
The technique for glucose is based on the addition A plasma sample drawn after an overnight fast of 8
to the sample of [13C6]glucose, a stable isotope, as to 16 hours is used for the diagnosis of diabetes.
the internal standard. The chemical behavior of the The plasma should be separated from the cells
internal standard and the analyte are the same. The within 60 minutes of collection unless the tube
ratio between the unlabelled and labeled substances contains a glycolysis inhibitor [11]. Serum samples
remains unchanged during the overall procedure, are appropriate for glucose analysis, providing the
and no correction for sample recovery is needed serum is not in contact with the cells for longer
[7]. External quality assurance programs use this than 90 minutes. Glucose in whole blood at room
technique to set the target values for their temperature can undergo glycolysis at a rate of
lyophilized sera [8]. Manufacturers also use this approximately 5% to 7% (10 mg/dL or ~ 0.6
technique to accurately determine glucose mmol/L per hour). The sample should be
concentrations in calibrators for glucose assays. centrifuged and removed from clot or cells as soon
as possible. In patients with leukocytosis or in
A reference method for glucose based on the samples that are contaminated with bacteria, the
hexokinase reaction has been developed and rate of glycolysis can be even higher. To preserve
validated [9]. This technique is accurate and blood that cannot be separated rapidly, samples
precise but is too time consuming for routine use. should contain the glycolysis inhibitor sodium
Plasma or serum samples are deproteinized using fluoride (which inhibits the enolase enzyme) at 2.5
solutions of barium hydroxide and zinc sulfate. The mg fluoride/mL of blood. This reagent is used in
resulting supernatant is incubated at 25C with a
654
Glucose
combination with anticoagulants such as potassium kinetic analysis of the reaction. By selection of the
oxalate. The glucose concentrations are stable for most appropriate initial and subsequent times for
72 hours at room temperature with fluoride, but spectrophotometric reading, a reduction in both
there is a 10% drop in the concentration due to a direct photometric and chemical interferences can
water shift from the cells. The glucose be made.
concentration does not change in serum or plasma
samples collected in tubes that contain gel The semiquantitative dipsticks employing glucose
separators. The glucose in these tubes is stable for oxidase and peroxidase are highly specific for
at least 1 week when stored at 4C. glucose, with no other sugar producing a reaction.
Strong oxidizing substances such as hypochlorite
Other fluids such as ascitic, pleural, peritoneal, and and chlorine bleach can produce a positive
drainage fluids can be analyzed for glucose. These reaction, and ascorbic acid at high levels can
samples should be collected into tubes containing interfere with the peroxidase step, reacting with the
fluoride/oxalate preservative. If the samples are hydrogen peroxide to give erroneously low glucose
turbid, this suggests the presence of cells, and concentrations.
samples must be centrifuged prior to analysis.
Glucose Reference Intervals and Interpretation
Capillary blood glucose measurement is widely There are many conditions that lead to low or
used to monitor glucose levels in diabetic patients. elevated blood glucose levels (see Chapter 38,
Concentrations are approximately 2 to 5 mg/dL Diabetes, in the 5th edition of Clinical Chemistry:
(0.11 to 0.28 mmol/L) higher than venous blood Theory, Analysis, Correlation). However, the
glucose concentrations in fasting patients. primary reason to measure glucose is for the
However, following a glucose load, the difference diagnosis of diabetes.
can be as great as 70 mg/dL (3.9 mmol/L).
Reference intervals are of use for neonates and
Interferences children, but the diagnosis of pre-diabetes and
The disadvantage of the coupled glucose oxidase diabetes is done using specific criteria published by
assays is that many compounds present in serum or various Diabetes Associations. The new cut-points
urine (such as bilirubin, ascorbic acid, and uric recommended by the American Diabetes
acid) can be oxidized by the hydrogen peroxide Association (ADA) [15] are as follows:
produced by the glucose oxidase reaction, resulting
in a negative bias. The negative bias resulting from Fasting blood glucose:
the reaction between the hydrogen peroxide and Normal glucose: < 100 mg/dL (<5.6 mmol/L)
ascorbic acid can often lead to strikingly low Pre-diabetic: 100 to 125 mg/dL (5.6 to 6.9 mmol/L)
values in cerebrospinal fluid (CSF) [12]. This is Diabetic: 126 mg/dL (7.0 mmol/L)
because glucose concentrations are relatively low
in CSF, and ascorbate concentrations in CSF are The ADA does not recommend the use of an oral
higher than in serum [13]. Since the extent of the glucose tolerance test (OGTT) for the routine
interference is inversely related to the glucose diagnosis of type 1 or 2 diabetes. However, it does
concentration, CSF samples from neonates are recommend the use of the glucose challenge test
particularly vulnerable to this form of interference. for the diagnosis of gestational diabetes mellitus
The coupled glucose oxidase reactions performed (GDM) [16]. Other expert bodies such as the World
as a kinetic analysis are especially sensitive to Health Organization and the Australian Diabetes
ascorbate interference [12]. Also, because Society recommend the use of an OGTT to screen
ascorbate is very unstable and breaks down rapidly, for type 2 diabetes [17,18]. Important limitations of
the degree of interference will vary with time and the OGTT are poor reproducibility, cost, and
temperature of storage before analysis. This inconvenience. However, it may have a higher
interference can be avoided if all CSF specimens sensitivity than the fasting plasma glucose (FPG)
are analyzed by an alternative technique that does for detecting diabetes [19]. This is a controversial
not use a glucose oxidasehydrogen peroxidase area, and it is likely that further changes will be
generating method. These include the hexokinase, made to the diagnostic criteria for diabetes as more
rate oxygen consumption, and glucose clinical studies are completed.
dehydrogenase methods [14].
The Australian Diabetes Society criteria for
In addition, there may be compounds that react by diagnosis of diabetes using a fasting sample are
oxidizing the indicator dye, resulting in positive [18]:
biases. The specificity and accuracy of the coupled Normal glucose- 3.0-6.0 mmol/L
glucose oxidase procedures have been increased by Impaired fasting glucose (IFG)- 6.1-6.9 mmol/L
a choice of reaction conditions that minimize this Diabetic- 7.0 mmol/L
bias. The most common approach is to employ
655
Glucose
A fasting blood glucose result 7.0 mmol/L in

subjects with typical symptoms is diagnostic of Performance Goals
diabetes. In subjects without symptoms, a Survey data from the 2007 CAP Participant
confirmatory test, preferably a fasting plasma Summary Report showed that 53% of laboratories
glucose concentration should be performed on a used hexokinase methods measured at 340 nm,
separate day. In addition, if a formal laboratory 25% used glucose oxidaseoxygen electrode
FPG measurement is between 5.5 and 6.9 mmol/L, methods, 17% glucose oxidase colorimetric
then a 75 g OGTT should be performed to exclude methods, 5% hexokinase colorimetric methods, and
diabetes [18]. less than 1% of laboratories use glucose
dehydrogenase (Table 2). The imprecision of these
Cord blood is a mixture of neonatal arterial and methods at a mean glucose concentration of 153.6
venous blood that is removed from the umbilical mg/dL (8.45 mmol/L) demonstrated that the
cord at the time of birth. Cord blood can be used glucose oxidase methods were the most precise,
for a variety of measurements, including blood with a mean CV of 2.0% to 2.4%, followed by the
gases, bilirubin, and glucose. The fetuss blood hexokinase UV methods, with a mean CV of 2.5%.
glucose concentration in cord blood is usually The glucose dehydrogenase methods have a mean
about 10% lower than the maternal blood at the CV of 2.7%.
time of birth [20].
Table 2. Methods Used for Glucose Analysis in 5259
Most new cases of type 1 (juvenile or insulin- Laboratories Enrolled in CAP 2007
dependent) diabetes are diagnosed in the acute Method Total Number of CV (%)
stage in the emergency room. In these cases, the (%) Laboratories
diabetic patient presents with hyperglycemia and Hexokinase,
ketoacidosis. Abnormalities of electrolytes and measured at 53% 2787 2.5% at mean
renal function tests are also common findings in 340 nm 153.9 mg/dL
type 1 diabetics acutely ill with ketoacidosis. (8.54 mmol/L)
The 95% confidence intervals in the glucose Hexokinase, 5% 250 2.8% at mean
concentrations of important cutoff points can be colorimetric 153.5 mg/dL
calculated by multiplying the analytical CV for the (8.52 mmol/L)
assay (CVa) by 1.96 [21]. If the CVa for the assay
were 2.0%, then for a true plasma glucose Glucose
concentration of 126 mg/dL (7.0 mmol/L), the 95% oxidase, 17% 878 2.4% at mean
confidence intervals (CI) for that result would be color 146.8 mg/dL
121 to 131 mg/dL (6.7 to 7.3 mmol/L). At a (8.15 mmol/L)
glucose concentration of 200 mg/dL (11.1 Glucose 25% 1316 2.0% at mean
mmol/L), the 95% confidence intervals would be oxidase 157.5 mg/dL
192 to 208 mg/dL (10.7 to 11.5 mmol/L). Table 1 oxygen (8.74 mmol/L)
lists the 95% CI around the major glucose cut- electrode
points for a CVa of 2.0% and 4.0%, which are the
approximate low and high CVa data for the 2007 Glucose <1% 17 2.7% at mean
CAP results. dehydroge 153.9 mg/dL
nase (8.54 mmol/L)
As can be seen from these results, the larger the
analytical CV, the greater uncertainty exists around CV, Coefficient of variation as published by the
the cutoff value. Therefore, it is important to keep 2007 College of American Pathologists Participant
the CVa as tight as possible. Summary Report.
Table 1
95% Confidence Intervals at Several Plasma Glucose Concentrations
Plasma 100 110 126 mg/dL 200 mg/dL (11.1 mmol/L)
glucose mg/dL mg/dL (7.0
(5.6 (6.1 mmol/L)
mmol/L) mmol/L)
Analytical 2.0 96-104 106-114 121-131 192-208
CV (%) (5.4-5.8) (5.9-6.3) (6.7-7.3) (10.7-11.5)
4.0 92-108 101-119 116-136 184-216
(5.2-6.0) (5.6-6.6) (6.5-7.5) (10.2-12.0)
656
Glucose
Acceptable Clinical Laboratory Improvement performance characteristics. All meters require

Amendments performance criteria (CLIA-88) for calibration. This can be automatic or done
measurement of glucose require that laboratories be manually with each lot change of reagents. The
accurate to within 6 mg/dL or 10% (greater) of manufacturers supply quality controls that must be
the peer-group mean. The intraindividual variation regularly checked to ensure the performance of the
of glucose in blood in healthy adults has been meter. A number of the larger manufacturers also
determined to be approximately 6.5%. Desirable provide external quality assurance programs. There
specifications for analytical imprecision derived are multiple analytical goals proposed for the
from studies of biological variation indicate an performance of glucose meters.
assay imprecision of no greater than 3.3% and a
total error of no greater than 7.9% [22]. All glucose The results from glucose meters are not as accurate
methods are within the desired performance criteria as those obtained using laboratory methods,
as defined by CLIA-88 and the analytical especially in the hypoglycemic range, where
imprecision specification if biological variability accurate measurement is crucial and can
defines required performance. overestimate glucose concentrations in the higher
range [25-28]. Many analytical goals have been
Glucose Meters proposed for the performance of glucose meters.
Management of glycemia in diabetes is crucially The ADA recommends the analytical error for
important for the prevention glucose results should be within 5% of the
of both acute and long-term complications. Self- reference values [29]. The Clinical and Laboratory
monitoring of blood glucose (SMBG) is effective Standards Institute (CLSI) uses the ISO 15197
for patients with type 1 and type 2 diabetes using guidelines that recommend results for meters be
insulin. The ADA recommends the use of SMBG within 15 mg/dL (0.83 mmol/L) for glucose
for all insulin-treated patients with diabetes, those concentrations less than 75 mg/dL (4.2 mmol/L)
treated with sulfonylureas or other insulin and within 20% for higher blood glucose values
secretagogues, and in patients not achieving when compared with compared with the central
glycemic goals [23]. However, there is debate laboratory [19]. A meter is considered within the
regarding the effectiveness of SMBG as a self- guidelines if 95% of pairs meet these criteria in
management tool for type 2 diabetics who are not both glucose ranges.
using insulin. Welschen et al. reviewed the
literature and concluded that two of six studies There are many preanalytical and analytical factors
reported a significant lowering of HbA1c in that can reduce the accuracy of the meter results.
patients using SMBG [24]. It was concluded that Preanalytical variables that may affect the
SMBG may be effective in improving glycemic performance of SMBG are hematocrit, temperature,
control in patients with type 2 diabetes who are not hypoxia, low systolic blood pressure, elevated
on insulin. However, a large, well-conducted triglycerides, and some drugs such as ascorbic acid
random-controlled trial is required. and acetaminophen. Other causes of poor results
include improper calibration and use of controls,
Individuals with diabetes use SMBG in an attempt poor storage of test strips, lack of maintenance, and
to maintain their glucose levels as close to those of the differences in the technology of the
a nondiabetic as is possible. Due to the imprecision instruments. Kost suggests a number of strategies
of the meters and the difference in results between that may help to improve the performance of
different manufacturers of meters, they are not meters [30]. These include standardized calibration,
recommended for the diagnosis of diabetes but only proficiency testing based on the same global
for the monitoring of treatment. standard, and improved technologies that eliminate
the effects of co-founding factors.
The methodology used in the strips is the same as
described for laboratory glucose analysis. A drop Whole blood glucose levels are 10% to 15% lower
of blood is applied to the strip that contains all the than the values obtained from plasma samples. In
necessary reagents. Some strips use current clinical practice, plasma and blood glucose
electrochemistry to detect the flow of electrons are used interchangeably, with a consequent risk of
from the enzymatic reaction by an electrode clinical misinterpretation. In human blood, glucose
incorporated in the strip. Others use a dye that is distributed (like water) between erythrocytes and
reacts with the glucose oxidase-peroxidase plasma. Many of the modern meters will
chromogenic reaction, and the color intensity is automatically adjust the glucose reading by
proportional to the glucose concentration and is multiplying the whole blood glucose concentration
quantified by reflectance photometry that measures by a constant factor of 1.11 to convert the
the rate of reaction or the final concentration of the concentration in whole blood to the equivalent
product. There are a wide variety of meters concentration in plasma [1]. The conversion will
available, and they all have slightly different provide harmonized results, facilitating the
657
Glucose
classification and care of patients, and lead to fewer fluid underneath the skin by reverse iontophoresis.
therapeutic misjudgments. In a randomized clinical trial of 200 children with
type 1 diabetes mellitus, the use of the GW2B with
A recent study of glucose-meter accuracy was conventional glucose meter monitoring did not
reported comparing two of the newer generation lower HbA1c levels or the frequency of
meters for the detection of subsequent nocturnal hypoglycemia, compared with conventional
hypoglycemia. Glucose concentrations were glucose monitoring alone [36,37].
measured throughout the day and night and every
15 to 20 min during a standardized exercise A double-blind crossover study of insulin-treated
protocol [27]. Reference samples were assayed in a type 2 diabetes compared the use of the MiniMed
central laboratory, using a hexokinase enzymatic CGMS with SMBG in 160 individuals. Over the
method. Sensitivities for detection of hypoglycemia 16-week treatment period, it showed that the
were 96% and 100% for the two methods, and CGMS was superior to SMBG in assessing daily
corresponding false-positive rates were both 5%. glucose fluctuations and in detecting unrecognized
Thus in a controlled clinical setting using venous hypoglycemic episodes, especially at night [38].
blood samples, both meters had a high degree of
accuracy compared with the laboratory reference A study that compared CGMS with SMBG
over a broad range of glucose concentrations in techniques in patients with gestational diabetes
children with type 1 diabetes. However, a number mellitus showed that CGMS detected a higher
of other studies have not shown such good proportion of mothers needing antihyperglycemic
correlation [30-32]. medication, compared with self-monitoring of
plasma glucose [39].
Minimally Invasive Glucose Monitoring
Continuous glucose monitoring system (CGMS) CGMS techniques appear to hold considerable
technology is in a developmental phase, but in the promise, and once the technology becomes more
future will significantly change diabetes robust they are likely to be widely used.
management. The use of insulin-pump therapy
offers better diabetic control with reduced Spectroscopy-based and fluorescence-based
postprandial hyperglycemia and reduced risk of noninvasive sensors are under development and
severe hypoglycemia, especially during the night hold promise for the future, but the technology
[33]. However, for this kind of therapy to be really needs further improvement before it can be used
successful, a system that continuously monitors clinically [40,41].
glucose and is in real-time communication with the
infusion pump to provide the amount of insulin References
needed for the maintenance of normal glucose 1. D'Orazio PBR, Fogh-Andersen N, Jacobs
levels is required. E, Kuwa K, Klpmann WR,, Larsson LLA
et al. IFCC-SD-WG-SEPOCT. Approved
There are a number of CGMS on the market, and IFCC recommendation on reporting
more are under development [34]. These systems results for blood glucose: International
measure glucose levels in subcutaneous interstitial Federation of Clinical Chemistry and
fluid. Recent reports describe the use of CGM in Laboratory Medicine Scientific Division,
clinical settings [35], but the monitors are not Working Group on Selective Electrodes
easily used on a routine, clinical, long-term basis. and Point-of-Care Testing (IFCC-SD-
WG-SEPOCT). Clin Chem Lab Med.
The CGMS developed by Medtronic, Medtronic 2006;44:1486-90.
MiniMed (Northridge, California), was the first 2. Banauch D, Brummer W, Ebeling W,
continuous glucose monitoring device approved by Metz H, Rindfrey H, Lang H, Leybold K,
the U.S. Food and Drug Administration. The Rick W, Staudinger HJ. [A glucose
CGMS sensor, consisting of glucose oxidase dehydrogenase for the determination of
plated platinum electrode is inserted through a glucose concentrations in body fluids
needle into the subcutaneous tissue of the anterior (author's transl)]. Z Klin Chem Klin
abdominal wall and measures interstitial glucose Biochem. 1975;13:101-7.
concentrations. Real-time glucose measurements 3. Banauch D, Rindfrey H, Helger R. A
can also be continuously transmitted using wireless glucose-determination on mechanized
technology to a glucose sensor. analyzers using glucose-dehydrogenase. Z
Klin Chem Klin Biochem. 1974;12:265-6.
Another FDA-approved glucose sensing system is 4. Maguire GA, Price CP. Kinetic glucose
the GlucoWatch 2 Biographer (GW2B) (Cygnus, dehydrogenase method for glucose
Redwood City, California). It is worn on the measurement with a discrete kinetic
forearm, and glucose is pulled from the interstitial
658
Glucose
analyzer overcomes interference by on Gestational Diabetes Mellitus. Diabetes

ascorbate. Clin Chem. 1984;30:157-8. Care. 2007;30 Suppl 2:S251-60.
5. American Diabetes Association: Tests of 17. World Health Organization: Definition,
glycemia in diabetes. Diabetes Care. Diagnosis and Classification of Diabetes
1999;22:S77-9. Mellitus and its Complications: Report of
6. Goldstein DE, Little RR, Lorenz RA, a WHO Consultation. Part 1: Diagnosis
Malone JI, Nathan D, Peterson CM. Tests and Classification of Diabetes Mellitus.
of glycemia in diabetes. Diabetes Care. Geneva: World Health Org; 1999.
1995;18:896-909. 18. Twigg SM, Kamp MC, Davis TM, Neylon
7. Magni F, Paroni R, Bonini PA, Kienle EK, Flack JR. Prediabetes: a position
MG. Determination of serum glucose by statement from the Australian Diabetes
isotope dilution mass spectrometry: Society and Australian Diabetes Educators
candidate definitive method. Clin Chem. Association. Med J Aust. 2007;186:461-5.
1992;38:381-5. 19. Sacks DB, Bruns DE, Goldstein DE,
8. Thienpont LM, Leenheer AP, Stockl D, Maclaren NK, McDonald JM, Parrott M.
Reinauer H. Candidate reference methods Guidelines and recommendations for
for determining target values for laboratory analysis in the diagnosis and
cholesterol, creatinine, uric acid, and management of diabetes mellitus. Clin
glucose in external quality assessment and Chem. 2002;48:436-72.
internal accuracy control. II. Method 20. Bossolan G, Trindade CE, Barreiros RC.
transfer. Clin Chem. 1993;39:1001-6. Blood galactose and glucose levels in
9. Neese JW DP, Bayse D, Robinson M, mothers, cord blood, and 48-hour-old
Cooper T, Stewart C. Development and breast-fed full-term infants. Neonatology.
Evaluation of a Hexokinase/Glucose-6- 2007;91:121-6.
phosphate Dehydrogenase Procedure for 21. Fraser C. Biological Variation: From
Use as a National Glucose Reference Principles to Practice. Washington DC:
Method. HEW Publication No. (CDC) 77- AACC Press; 2001:74-75.
8330. Atlanta: Centers for Disease Control 22. Ricos C, Alvarez V, Cava F, Garcia-Lario
and Prevention; 1976. JV, Hernandez A, Jimenez CV,
10. Ladenson JH, Tsai LM, Michael JM, Minchinela J, Perich C, Simon M. Current
Kessler G, Joist JH. Serum versus databases on biological variation: pros,
heparinized plasma for eighteen common cons and progress. Scand J Clin Lab
chemistry tests: is serum the appropriate Invest. 1999;59:491-500.
specimen? Am J Clin Pathol. 23. Standards of medical care for patients with
1974;62:545-52. diabetes mellitus. Diabetes Care. 2000;23
11. Chan AY, Swaminathan R, Cockram CS. Suppl 1:S32-42.
Effectiveness of sodium fluoride as a 24. Welschen LM, Bloemendal E, Nijpels G,
preservative of glucose in blood. Clin Dekker JM, Heine RJ, Stalman WA,
Chem. 1989;35:315-7. Bouter LM. Self-monitoring of blood
12. Maguire GA, Price CP. Evidence for glucose in patients with type 2 diabetes
interference by ascorbate in the who are not using insulin: a systematic
measurement of cerebrospinal fluid review. Diabetes Care. 2005;28:1510-7.
glucose by a kinetic glucose 25. Brunner GA, Ellmerer M, Sendlhofer G,
oxidase/peroxidase procedure. Clin Chem. Wutte A, Trajanoski Z, Schaupp L,
1983;29:1810-2. Quehenberger F, Wach P, Krejs GJ,
13. Spector R. Vitamin homeostasis in the Pieber TR. Validation of home blood
central nervous system. N Engl J Med. glucose meters with respect to clinical and
1977;296:1393-8. analytical approaches. Diabetes Care.
14. Price CP, Spencer K. A rapid kinetic assay 1998;21:585-90.
for glucose using glucose dehydrogenase. 26. Johnson RN, Baker JR. Error detection
Ann Clin Biochem. 1979;16:100-5. and measurement in glucose monitors.
15. Genuth S, Alberti KG, Bennett P, Buse J, Clin Chim Acta. 2001;307:61-7.
Defronzo R, Kahn R et al. Follow-up 27. Kost GJ, Tran NK, Abad VJ, Louie RF.
report on the diagnosis of diabetes Global standardization and improved
mellitus. Diabetes Care. 2003;26:3160-7. technologies for point-of-care glucose
16. Metzger BE, Buchanan TA, Coustan DR, testing. Clin Chim Acta. 2008;391:127-8.
de Leiva A, Dunger DB, Hadden DR, Hod 28. Wehmeier M, Arndt BT, Schumann G,
M, Kitzmiller JL, Kjos SL, Oats JN et al. Kulpmann WR. Evaluation and quality
Summary and recommendations of the assessment of glucose concentration
Fifth International Workshop-Conference measurement in blood by point-of-care
659
Glucose
testing devices. Clin Chem Lab Med. 36. Chase HP, Beck R, Tamborlane W,
2006;44:888-93. Buckingham B, Mauras N, Tsalikian E,
29. American Diabetes Association: Self Wysocki T, Weinzimer S, Kollman C,
monitoring of blood glucose [consensus Ruedy K et al. A randomized multicenter
Statement]. Diabetes Care. 1996;19:S62- trial comparing the GlucoWatch
S6. Biographer with standard glucose
30. Kost GJ, Tran NK, Abad VJ, Louie RF. monitoring in children with type 1
Evaluation of point-of-care glucose testing diabetes. Diabetes Care. 2005;28:1101-6.
accuracy using locally-smoothed median 37. Tansey MJ, Beck RW, Buckingham BA,
absolute difference curves. Clin Chim Mauras N, Fiallo-Scharer R, Xing D,
Acta. 2008;389:31-9. Killman C, Tamborlane WV, Ruedy KJ.
31. Ghys T, Goedhuys W, Spincemaille K, Accuracy of the modified Continuous
Gorus F, Gerlo E. Plasma-equivalent Glucose Monitoring System (CGMS)
glucose at the point-of-care: evaluation of sensor in an outpatient setting: results
Roche Accu-Chek Inform and Abbott from a diabetes research in children
Precision PCx glucose meters. Clin Chim network (DirecNet) study. Diabetes
Acta. 2007;386:63-8. Technol Ther. 2005;7:109-14.
32. Vanavanan S, Chittamma A. Performance 38. McNally PG, Dean JD, Morris AD,
of a multi-profile critical care testing Wilkinson PD, Compion G, Heller SR.
analyzer. Clin Chem Lab Med. 2008;46:9- Using continuous glucose monitoring to
14. measure the frequency of low glucose
33. Davis EA, Keating B, Byrne GC, Russell values when using biphasic insulin aspart
M, Jones TW. Hypoglycemia: incidence 30 compared with biphasic human insulin
and clinical predictors in a large 30: a double-blind crossover study in
population-based sample of children and individuals with type 2 diabetes. Diabetes
adolescents with IDDM. Diabetes Care. Care. 2007;30:1044-8.
1997;20:22-5. 39. Kestila KK, Ekblad UU, Ronnemaa T.
34. Chia CW, Saudek CD. Glucose sensors: Continuous glucose monitoring versus
toward closed loop insulin delivery. self-monitoring of blood glucose in the
Endocrinol Metab Clin North Am. treatment of gestational diabetes mellitus.
2004;33:175-95, xi. Diabetes Res Clin Pract. 2007;77:174-9.
35. Tanenberg R, Bode B, Lane W, Levetan 40. Cohen O, Fine I, Monashkin E, Karasik A.
C, Mestman J, Harmel AP, Tobian J, Glucose correlation with light scattering
Gross T, Mastrototaro J. Use of the patternsa novel method for noninvasive
Continuous Glucose Monitoring System glucose measurements. Diabetes Technol
to guide therapy in patients with insulin- Ther. 2003;5:11-7.
treated diabetes: a randomized controlled 41. Pickup JC, Hussain F, Evans ND, Rolinski
trial. Mayo Clin Proc. 2004;79:1521-6. OJ, Birch DJ. Fluorescence-based glucose
sensors. Biosens Bioelectron.
2005;20:2555-65.
660
Glucose
Table 3: Methods of Glucose Analysis

Method 1: Hexokinase (HK); quantitative, spectrophotometric K or EP
Glucose + ATP HK glucose 6-phosphate + ADP
Glucose 6-phosphate + NADP+ G6PD 6-phosphogluconate + NADPH + H+
Increased absorbance at 340 nm related to glucose concentration
Comments: Serum, CSF, urine; automated; most commonly used method; has been proposed as basis of
reference method; very good accuracy and precision
Method 2: Glucose oxidase coupled reaction (Trinder)
a. Quantitative, using various types of dyes as final O2 acceptor; K or EP
b. Quantitative or semiquantitative in dipstick screen, visual or reflectance photometry
Glucose + O2 glucose oxidase gluconic acid + H2O2
H2O2 + reduced dye horseradish peroxidase oxidized dye + H O
2
(colored)
Peroxidase indicator reaction increased absorbance related to glucose concentration
Comments:
a. Serum, urine, CSF; easily and usually adapted to automated analysis; second indicator reaction
susceptible to false-positive interferences from a variety of compounds; good accuracy and precision
b. Used in all dipstick screens; serum, urine
Method 3: Glucose dehydrogenase (GDH); quantitative, EP, K

GDH
Glucose + NAD+ D-gluconolactone + NADH + H
+
Increased absorbance at 340 nm is related to glucose concentration

Comments: Rare, serum, CSF; can be adapted to automated instruments
Method 4: Glucose oxidase (GO) oxygen consumption; quantitative, polarographic measurement using O2
electrode; K
Glucose + O2 glucose oxidase gluconic acid + H2O2
H2O2 consumed in side reactions
O2 consumption measured polarographically by oxygen electrode
Comments: Serum, CSF; semiautomated and fully automated systems; correlates best with the proposed
reference method; very good accuracy and precision
EP, End-point analysis mode; CSF, cerebrospinal fluid; GDH, glucose dehydrogenase; K, kinetic analysis mode.
Historic Methods method used for glucose analysis. Although it is a

more sensitive procedure than the other copper-
Most older methods for measurement of serum reduction methods, this method lacked specificity.
glucose were based on the ability of glucose to Benedicts modification of the copper-reduction
directly reduce cupric ions (Cu 2+) to monovalent methods is still used today but only as a
semiquantitative method for estimation of urine
cuprous ions (Cu+). In the presence of heat, the
glucose. This procedure, sensitive to total reducing
reduced cuprous ions can form cuprous oxide
compounds present in urine, yields red Cu2O and
(Cu2O), which can be detected by a variety of
yellow CuOH precipitates. The greater the
methods. The most popular method was the
concentration of glucose, the redder the final color.
reduction of phosphomolybdate (FolinWu) or
arsenomolybdate (SomogyiNelson) to form blue
The alkaline ferricyanide reaction involves the
molybdenum compounds. Neocuproine (2,9-
reduction of yellow ferricyanide, Fe(CN) 63, to
dimethyl-1,10-phenanthroline) reduced by Cu+ colorless ferrocyanide, Fe(CN) 64 , by glucose in
ions to form a highly colored complex was a alkaline conditions. When this method was initially
661
Glucose
automated, the reaction was monitored by reaction is based on the ability of many aromatic
measurement of the decrease in yellow color. Later amines in acid solutions to condense with the
adaptations of this method in the 1960s and 1970s aldehyde group of glucose to form glycosamines.
to AutoAnalyzer equipment utilized indirect The most widely used aromatic amine, o-toluidine,
measurement of ferrocyanide. The ferrocyanide is believed to be a carcinogen and is no longer
was reacted with excess ferric ions to form ferric used. These methods are no longer used and are
ferrocyanide (Prussian blue). The o-toluidine listed only for historic interest.
662
Glycated Hemoglobin
Glycated Hemoglobin
Andrea M. Rose
Name: Glycated hemoglobin, glycohemoglobin, GHb

Clinical significance: Refer to Chapter 38, Diabetes, in the 5th edition of Clinical Chemistry:
Merck Index: 4538 (hemoglobin)
Chemical class: Glycated protein
Reaction: See Figure 1 at end
Principles of Analysis and Current Usage differences that recognize glucose-modified from
The IUPAC-IUB Joint Commission on Biochem i cal nonglucose-modified hemoglobin (boronate affinity,
Nomenclature suggests glycation as the term that should immunoassay, enzymatic). Standardization effortswhich
be used to describe the nonenzymatic addition of glucose have resulted in the formation of the National
or other sugars to the free amino groups of proteins [1]. Glycohemoglobin Standardization Program (NGSP)and
Since the middle to late 1970s, the glycated protein used to treatment goals based on outcomes from the Diabetes
monitor glycemic control in patients with diabetes mellitus Control and Complications Trial (DCCT) [4] and the
has been glycated hemoglobin (GHb). The major United Kingdom Prospective Diabetes Study (UKPDS) [5]
component of adult hemoglobin is HbA, a tetramer have led the American Diabetes Association (ADA) to
consisting of two and two subunits. The best recommend that regardless of methodology or specific
characterized and most abundant adduct of HbA is termed analyte quantified, results should be reported in terms of %
HbA1c and describes a hemoglobin chain with a glucose HbA1c or % HbA1c equivalents [6].
bound to the N-terminal valine of this chain.
Ion-exchange HPLC elutes hemoglobin species based on
Sugars other than glucose have been shown to modify the differences in their charge. The sample is automatically or
amino terminus of hemoglobin and share a nomenclature manually injected into a column filled with resin or gel
with HbA1c based on the order of their elution from an having ionic functional groups of opposite charge from the
ion-exchange column. For example, HbA1a, A1b, A1c, analyte ions. Buffers of increasing ionic strength pass
collectively termed HbA1, are also called the fast through the column, displacing the hemoglobin species.
hemoglobins, owing to their more negative charge and The separated hemoglobin fractions pass through a flow
earlier elution compared to HbA0. Additionally, lysine cell, where absorbance is measured spectrophotometrically.
residues located throughout the and chains can serve as Background noise can be reduced by using a secondary
glycation sites. Collectively, glucose residues at the amino wavelength. Software integrates the raw data, and time in
terminus and/or the lysine residues of both and minutes from injection to the maximum point of the
subunits are referred to as total glycated hemoglobin elution peak is compared to an internal library of normal
(TGlyHb) [2,3]. hemoglobin and variant fractions. Current methods use
whole blood samples in their primary tubes and offer
Measuring glycemic control in patients with diabetes automatic mixing, dilution, and injection. HPLC offers
mellitus has improved in both methodology and excellent precision and the ability to detect the presence of
standardization since the previous edition of this chapter. hemoglobin variants [7,8].
Current methods in routine use still fall into two broad
categories based on (1) charge differences between The 2007 College of American Pathologists (CAP)
modified and unmodified hemoglobin (ion-exchange participant summary, GH2-A, showed no participants
chromatography, electrophoresis), and (2) structural reporting GHb results by electrophoresis. The June 2007
NGSP list of certified methods lists one manufacturer with
i a semiautomated electrophoretic method certified in 2006.
Glycated hemoglobin
Electrophoresis involves the separation of hemoglobin
species based on charge differences. The sample is applied
to an agarose gel, and the hemoglobins migrate across an
Methods edition: Mary Ellen King
electric field. After electrophoresis, bands are scanned by a
Second edition: Mary Ellen King
densitometer and compared to a standard for quantification.
Third edition: Andrea Rose
Fourth edition: Andrea Rose
Boronate affinity chromatography employs a binding
Fifth edition: Andrea Rose
ligand (boronic acid) immobilized to an inert, insoluble
663
Glycated Hemoglobin
matrix that selectively interacts with cis-diol groups laboratories could report values traceable to the DCCT.
present in sugars such as glucose to form a strong but The NGSP, formed in 1996, still uses the DCCT reference
reversible covalent bond [9]. Glycated hemoglobin method (HPLC using Bio-Rex 70 resin) to provide the
remains bound, while nonglycated hemoglobin is eluted anchor value for all primary and secondary reference labs
with wash buffer. A soluble diol-containing agent such as (PRL and SRL respectively) that are part of the NGSP
sorbital is then used to displace the glycated hemoglobin. network [12]. This network provides the framework for
Both species are detected photometrically at 414 nm. manufacturers and laboratories to obtain certification that
Boronate affinity chromatography detects total glycated their methods are traceable to the DCCT. Bio-Rex 70
hemoglobin, meaning the binding ligand can attach to HPLC, used in the Central Primary Reference Laboratory
glucoses at the amino terminus, as well as lysine residues (CPRL) and PRLs, is considered a Designated Comparison
downstream of both the and hemoglobin chains. The Method (DCM). Although the DCCT/EDIC Research
method, however, can be calibrated to reflect HbA1c Group reports that the Bio-Rex 70 HPLC and other HPLC
equivalent values. The ligand preferentially binds 1- methods used to support the DCCT and EDIC trials have
deoxyfructosyl derivatives [3]. been stable and precise long term [13], there are known
interferences associated with this methodology [12,14].
Immunoassays use antibodies that recognize the glycated This led the International Federation of Clinical Chemistry
amino terminus of the Hb chains including four or more to begin work in 1995 to develop a primary reference
amino acids. Commercial methods use antibody-mediated material and reference method that specifically measures
inhibition of latex agglutination or immunoturbidimetry to HbA1c.
obtain a HbA1c mass value [9]. HbA1c is expressed as a
percent after Hb is measured spectrophotometrically and In 1994, in order to achieve uniform, international
the ratio HbA1c/Hb is determined. standardization, the IFCC established a working group on
standardization of HbA1c [15]. The first step was to define
According to the NGSP website, one enzymatic procedure the analyte based on molecular structure rather than peak
was certified in April 2007 on an automated analyzer. (retention time) in an HPLC. The nomenclature
With this method, whole blood samples are first lysed, and recommended by the IFCC/IUPAC is N-(1-deoxyfructos-
then the lysate is subjected to extensive protease digestion. 1-yl) hemoglobin chain [16]. HbA1c is defined as the
Glycated valines serve as substrate for the recombinant stable glucose adduct to the N-terminal group of one or
enzyme, fructosyl valine oxidase. Hydrogen peroxide both of the -chains of hemoglobin A0, which may or may
formed in the initial reaction, plus an added chromagen, not include glucoses attached at lysine residues further
are catalyzed by horseradish peroxidase to produce a downstream. Glucoses attached at lysine residues without
colored agent that is proportional to the concentration of attachment at the N-terminal valine of the hemoglobin
GHb. Both the color formed from the chromagen and Hb chain are not defined as HbA1c [14]. The working group
are measured spectrophotometrically [10,11]. was also charged with isolating pure HbA1c and HbA0,
developing a reference method specific for HbA1c,
Reference and Preferred Methods establishing a Reference Laboratory Network, and
In 1993, a survey of CAP participants reporting GlyHb preparing secondary reference calibrators and controls to
results gave the following breakdown of methods in aid manufacturers in standardizing their methods [15]. In
routine use: affinity chromatography 35%, ion-exchange 2001, the following final method was adopted by the
HPLC 35%, electrophoresis 30%. The CAP 2007 GH2-A national member societies of the IFCC and published in
participant summary gave the following breakdown of 2002 [17]. Initially, hemoglobin isolated from lysed
survey participants by methodology for laboratories erythrocytes is proteolytically digested with
reporting %HbA1c or %HbA1c equivalents: immunoassay endoproteinase Glu-C, which releases either a glycated or
63.7%, HPLC 35.0%, boronate affinity chromatography nonglycated hexapeptide, C4,H9,O4-CO-CH2-NH-Val-His-
1.3%. In 2007, 99% of methods were NGSP certified at the Leu-Thr-Pro-Glu-COOH or NH2-Val-His-Leu-Thr-Pro-
time of the survey. Only 25 laboratories were reporting Glu-COOH, respectively. Note that cleavage occurs
TGlyHb using either of two manufacturers. between the sixth and seventh glutamic acid residues in the
hemoglobin chain. Therefore, the variant hemoglobins S
The Diabetes Control and Complications Trial completed and C, with substitutions of valine and lysine respectively
in 1993, and subsequent studies such as the Epidemiology at position six, are not cleaved by endoproteinase Glu-C
of Diabetes Interventions and Complications (EDIC) and [17]. Post proteolytic cleavage, the glycated and
the UKPDS, showed that the complications associated nonglycated hexapeptides are separated from other
with uncontrolled diabetes could be avoided in patients digested proteins and quantified by HPLC followed by
who maintained strict glycemic control. However, lack of either electrospray mass spectrometry or capillary
standardization among laboratories performing GHb electrophoresis with UV detection. Both systems have
testing led the American Association for Clinical been shown to give identical results. In 2004, members of
Chemistry (AACC) to form a subcommittee aimed at the IFCC Working Group on HbA1c Standardization
developing a standardization program so that all published a study documenting the linear relationship
664
Glycated Hemoglobin
between the IFCC method and each of three national Synchron CX7 [28], Bayer DCA2000 [32], Roche Tina-
HbA1c standardization schemes from the United States Quant immunoassay (Hitachi) [25,32], and COBAS
(NGSP), Japan (Japanese Diabetes Society/Japanese Integra [33-36] gave accurate recovery with these variants.
Society of Clinical Chemistry), and Sweden. While each of Immunoassay methods do not recognize nonglucose
the DCMs differs in recovery from each other and the adducts, which may also nonenzymatically attach to
IFCC Reference method (ion-exchange HPLC using Bio- hemoglobin as a result of uremia [37,38], high doses of
Rex 70 resin, high-resolution ion-exchange HPLC and aspirin [39], or alcoholism [40].
national calibrators, Mono S HPLC respectively), with the
linear relationships observed, the lower, more specific Cation-exchange chromatography and electrophoresis
IFCC values can be converted to the respective DCM separate hemoglobin species based on charge differences.
values via three distinct regression equations, allowing for Inaccurate results occur when a glycated or nonglycated
continued use of associated clinical decision points [18]. hemoglobin variant or chemical modification such as
acetylated or carbamylated hemoglobin cannot be
Specimen separated from HbA or HbA1c [9]. A seminal review by
Whole blood drawn in EDTA or other method-specific Bry et al. [9], as well as other recent reports, describe
anticoagulant may be used. Some methods require cell variants giving inaccurate results with HPLC [26,31,40-
lysis with hemolyzing reagent provided by the 47]. Interference is HPLC method and variant dependent.
manufacturer prior to analysis. Finger stick, as opposed to
venipuncture, may be used with point-of care devices. Boronate affinity chromatography measures total glycated
Samples are generally stable for one week at 2C to 8C, hemoglobin via attachment of the cis-diol groups of
but stability may be method specific [19,20]. The literature glucose to boronic acid. Because recognition is based on
suggests that storage at 20C is generally not structure and not charge, affinity chromatography does not
recommended [18]; however, the reader should consider recognize nonglucose adducts and is impervious to most
manufacturers guidelines. Whole blood stored at 70C variants. However, some method-specific instances of
has been shown to be stable for one year [21]. inaccurate results with hemoglobin variants have been
described [48], including with the common variants S and
Interferences C [30,49]. A mutation that results in acetylation of the
The accuracy of HbA1c measurements has been shown to amino terminus of the hemoglobin chain and a
be affected by the following: (1) conditions that cause consequent inability to glycate the amino terminus has
anemia or shortened erythrocyte survival, (2) the presence been described and would be expected to give decreased
of hemoglobin variants, and (3) nonglucose adducts to the reactivity with boronate affinity, as well as immunoassay
hemoglobin chain via carbamylation; acetylation; [26]. The CLC330 boronate affinity HPLC method from
vitamin C, E, or alcohol; or elevated concentrations of HbF. Primus Corporation has been used in several studies, with
Red blood cells (RBCs) are freely permeable to glucose hemoglobin variants as the comparative method. Recently
and other substances. Since the average RBC life span is this method was compared to the IFCC mass
120 days, the % HbA1c value reflects the nonenzymatic spectrophotometric reference method in support of its
addition of glucose adducts over the past 2 to 3 months [6]. consistent relative recovery in patients with hemoglobin S
Conditions that increase hemoglobin turnover will and C traits [29]. As noted under Reference and Preferred
decrease exposure of hemoglobin to glucose [22]. Methods, hemoglobins with amino acid substitutions at
Additionally, hemodialysis patients treated with position 6, including HbS and HbC, are not recognized in
erythropoietin to stimulate RBC production may have the IFCC reference method because of lack of enzymatic
HbA1c values that do not reflect glycemic control, owing cleavage in the initial step. Therefore, the consistent
to an increased proportion of young erythrocytes [23]. recovery of the IFCC method (measures the ratio
HbA1c/total HbA) versus the CLC330 boronate affinity
Amino acid substitutions in the and chains of adult method (measures total glycated Hb/total Hb) assumes no
hemoglobin give rise to hemoglobin variants that can difference in glycation rate for HbS and HbC [29]. While
affect the accuracy of diabetes monitoring in a variant and this author is aware of no publications citing a difference
method-dependent manner. For example, immunoassay in glycation rate for S and C variants, the literature does
methods use antibodies that recognize 4 to 10 N-terminal support such a difference with some variants [50,51], and
amino acids plus the ketoamine linkage and glucose readers are cautioned that some methodologies may not be
[24,25]. Amino acid substitutions close to the amino suitable for monitoring select patients.
terminus might be expected to impact antibody recognition
[26-28]. Clinically significant bias compared to CLC330 Substances such as ascorbic acid and other adducts that
boronate affinity, generally accepted to provide accurate compete for glucose binding to hemoglobin may result in
recovery with most hemoglobin variants [9,29], was an inaccurate and conflicting reflection of the patients
observed with both HbS (6Glu-Val) and HbC (6Glu- mean blood glucose level as indicated by the HbA1c value,
Lys) variants, using the Roche COBAS Integra [30] and which may differ with different methodologies [52].
the Metrika A1cNow [31] immunoassays. The Beckman Finally, both charge-based and structure-based methods for
665
Glycated Hemoglobin
diabetes monitoring may underestimate the HbA1c value monitored group having a higher MPG at each HbA1c
in patients with persistent elevations in HbF and may not level than the intensively treated group [60]. The picture is
accurately reflect the patients level of glycemic control further complicated by reports that more recent glucose
[53,54]. levels influence HbA1c levels more than earlier ones [61],
HbA1c levels may be partly genetically determined [62],
Reference Interval and that some individuals may be high versus low
Results from the DCCT and UKPDS clearly demonstrated glycators [62].
that lowering mean blood glucose values, and
consequently HbA1c, could reduce the complications With pressure growing to eliminate differences in HbA1c
associated with uncontrolled diabetes. Based on these trials, reporting globally, in May 2007, the Consensus Committee
the American Diabetes Association currently recommends of the American Diabetes Association, the European
that in general patients should achieve an A1c goal of < Association for the Study of Diabetes, the International
7%. The individual goal recommended for each patient is Diabetes Federation, and the International Federation of
to achieve an A1c value as close to the high end of normal Clinical Chemistry and Laboratory Medicine reached
(i.e., < 6 %) as possible without significant hypoglycemia agreement and issued the following directives regarding
[55]. The high end of normal in this case refers to the the worldwide standardization of the measurement of
reference interval reported in NGSP values which is ~ 4% HbA1c:
to 6% HbA1c and is based on the outcomes and 1 A1C test results should be standardized worldwide,
instrumentation used in the DCCT and UKPDS outcomes including the reference system and results reporting.
studies. 2 The new IFCC reference system for A1C represents
the only valid anchor to implement standardization of
The development of the IFCC reference system and the measurement.
subsequent joint effort to determine the relationship 3 A1C results are to be reported world-wide in IFCC
between the DCM values of Japan, the United States, and units (mmol/mol) and derived NGSP units (%) using
Sweden showed a strong correlation between each DCM the IFCC-NGSP master equation.
and the IFCC method. However, the increased specificity 4 If the ongoing average plasma glucose study fulfills
of the IFCC reference method means that IFCC HbA1c its a-priori-specified criteria, an A1C-derived average
values are significantly lowerfor example, 1.5% to 2.0% glucose (ADAG) value calculated from the A1C result
HbA1c compared to NGSP [56]. The IFCC method will also be reported as an interpretation of the A1C
reference interval is 2.8% to 3.8% HbA1c [57]. Currently, results.
manufacturers use the IFCC method to provide traceability 5 Glycemic goals appearing in clinical guidelines should
and an anchor to a higher-order reference method and be expressed in IFCC units, derived NGSP units, and
materials, as is required for manufacturers selling to as ADAG.
European Union countries. For patients and clinicians used Furthermore, the committee urges that these
to NGSP values, a conversion formula is then used to recommendations be implemented globally as soon as
convert from IFCC to NGSP. This has led to debate in the possible [64].
international clinical and scientific communities as to the
following: (1) Should the medical and scientific Interpretation
communities adopt the lower IFCC values globally, or Glycated hemoglobins (hemoglobins with covalently
should an IFCC to NGSP or other DCM conversion factor bound sugars) form nonenzymatically in red blood cells in
continue to be used to anchor the values to major clinical amounts mostly proportional (see above) to the average
trials; (2) Should the name HbA1c be changed to the blood glucose level over the 120-day lifespan of the
metrologically correct name, with a shortened version for erythrocyte [65]. Hemoglobin A1, the major glycated
general use, and the results reported in the standard hemoglobin fraction, can be further separated into
scientific unit of mmol/mol as supported by the IFCC; or subfractions, of which HbA1c is the predominant and best
(3) If an international study currently underway proves characterized component. HbA1c differs from its parent
there is a linear relationship between HbA1c values and nonglycated hemoglobin, A0, in that it has a glucose
mean blood glucose in both type 1 and 2 diabetics and residue covalently linked to the amino group of the N-
across ethnic groups, should laboratories report results of terminal amino acid of the chain. An unstable Schiffs
diabetes monitoring in terms of mean blood glucose base of glucose and hemoglobin (also termed the labile
[16,57-59]? On average, each 1% change in HbA1c has component) forms initially and rearranges to the stable
traditionally equated to approximately a 2 mmol/L (35 ketoamine product, A1c [66]. Also present are smaller
mg/dL) change in mean plasma glucose [55]. A very amounts of glycated hemoglobins, which are products of
recent reassessment of the DCCT data indicates that the nonenzymatic glycation of the hemoglobin chain by
relationship between MPG and HbA1c is not constant, as other sugars [67], as well as glucoses attached to lysine
previously thought. Moreover, the relationship appears residues of both and chains and at the -chain N-
dependent on whether the patient was treated intensively terminal group [68].
or monitored more conventionally, with the less closely
666
Glycated Hemoglobin
Data from the DCCT and UKPDS showed a reduction in 6 American Diabetes Association. Tests of
the microvascular and macrovascular consequences of Glycemia in Diabetes. Diabetes Care
uncontrolled diabetes as % HbA1c values approached 2004;27:S91-3.
nondiabetic levels in the intensively treated group. 7 Moridani M, Verjee Z, Allen L. Analytical
However, previous to the formation of and continued evaluation of hemoglobin A1c dual kit assay on
standardization efforts by the NGSP, lack of consensus on Bio-Rad Variant II: an automated HPLC
a comparative method for assay standardization, or even hemoglobin analyzer for the management of
which analyte to measure (i.e., HbA1 [A1a, Alb, A1c], diabetic patients. Clin Biochem 2003;36:317-20.
HbA1c, or total glycated Hb) affected the clinicians 8 Joutovsky A, Hadzi-Nesic J, Nardi M. HPLC
ability to effectively monitor patients and the ADAs Retention Time as a Diagnostic Tool for
ability to implement recommendations for patient care. As Hemoglobin Variants and Hemoglobinopathies:
of the May 2007 survey, CAP proficiency data indicated A Study of 60 000 Samples in a Clinical
99% of laboratories participating were reporting results as Diagnostic Laboratory. Clin Chem 2004;50:1736-
HbA1c or equivalent, and 99% of the methods used were 47.
certified by the NGSP. 9 Bry L, Chen P, Sacks D: Effects of Hemoglobin
Performance Goals Variants and Chemically Modified Derivatives on
It is generally accepted that U.S. laboratories should use Assays for Glycohemoglobin. Clin Chem
methods that are certified by the NGSP. Beginning 2001;47:153-63.
September 2007, NGSP certification requirements for 10 Sakurabayashi I, Watano T, Yonehara S, Ishimaru
manufacturers and laboratories desiring level II K, Hirai K, Komori T, Yagi M. New Enzymatic
certification tightened from 1% HbA1c for the 95% Assay for Glycohemoglobin. Clin Chem
confidence interval of the differences to NGSP to 0.85% 2003;49:269-74.
HbA1c. The precision criteria will be eliminated. CAP has 11 www.diazyme.com Direct Enzymatic HbA1c
in the past utilized a peer group grading system for HbA1c, Test Kit
but starting with the 2007 GH2-A survey, proficiency 12 Little R, Rohlfing C, Wiedmeyer H, Myers G,
testing participants were required to recover assigned Sacks D, Goldstein D. The National
NGSP target values within a 15% range limit [56]. For Glycohemoglobin Standardization Program: A
example, acceptable performance at a target value of 6.0 % Five-Year Progress Report. Clin Chem
HbA1c would be 5.1% to 6.9%. It is recommended that 2001;47:1985-92.
laboratories achieve an intralaboratory CV of < 3% and an 13 Steffes M, Cleary P, Goldstein D, Little R,
interlaboratory CV of < 5% [69]. Recent CAP data Wiedmeyer H, Rohlfing C, et al and the
suggests that most but not all methods have sufficient DCCT/EDIC Research Group. Hemoglobin A1c
precision to achieve this goal. Measurements over Nearly Two Decades:
Sustaining Comparable Values throughout the
References Diabetes Control and Complications Trial and the
1 IUPAC-IUB Joint Commission of Biochemical Epidemiology of Diabetes Interventions and
Nomenclature. Nomenclature of Glycoproteins, Complications Study. Clin Chem 2005;51:753-8.
Glycopeptides and Peptidoglycans. Pure Appl 14 Kobold U, Jeppsson J, Dulffer T, Finke A,
Chem 1988;60:1389-94. Hoelzel W, Miedema K. Candidate reference
2 Little R, Goldstein D. Endocrine Standardization methods for hemoglobin A1c based on peptide
of Glycohemoglobin Measurement. Anal Chem mapping. Clin Chem 1997;43:1944-51.
1995;67:393R-7R. 15 Goodall I. HbA1c Standardisation Destination
3 Cohen M, Clements R. Measuring Glycated Global IFCC Standardisation How, Why, Where
Proteins: Clinical and Methodological Aspects. and When, a Tortuous Pathway From Kit
Diabetes Technol Therapeutics 1999;1:57-69. Manufacturers, via Inter-laboratory Lyophilized
4 The Diabetes Control and Complications Trial and Whole Blood Comparisons to Designated
Research Group. The Effect Of Intensive National Comparison Schemes. Clin Biochem
Treatment Of Diabetes On the Development and Revs 2005;26:5-19.
Progression of Long-Term Complications In 16 Bruns D. The Clinical Chemist, Position
Insulin-Dependent Diabetes Mellitus. N Engl J Statement. Clin Chem 2006;52:232.
Med 1993; 329: 977-86. 17 Jeppsson J, Kobold U, Barr J, Finke A, Hoelzel
5 UK Prospective Diabetes Study (UKPDS) Group. W, Hoshino T, et. al. Approved IFCC Reference
Effect of intensive blood-glucose control with Method for the Measurement of HbA1c in Human
sulphonylureas or insulin compared with Blood. Clin Chem Lab Med 2002;40:78-89.
conventional treatment and risk of complications 18 Hoelzel W, Weykamp C, Jeppsson J, Miedema K,
in patients with type 2 diabetes. Lancet Barr J, Goodall I, et al. IFCC Reference System
1998;352:837-853. for Measurement of Hemoglobin A1c in Human
Blood and the National Standardization Schemes
667
Glycated Hemoglobin
in the United States, Japan, and Sweden: A Glycohemoglobin in the Presence of Hemoglobin
Method-Comparison Study. Clin Chem S and C Traits. Clin Chem 2005;51:264-5.
2004;50:166-74. 30 Roberts W, Brown D, Hanbury C, Hoyer J, John
19 Little R, England J, Wiedmeyer H, Goldstein D. W, et al. Effects of Hemoglobin C and S Traits on
Glycosylated Hemoglobin Measured by Affinity Eight Glycohemoglobin Methods. Clin Chem
Chromatography: Micro-Sample Collection and 2002;48:383-5.
Room-Temperature Storage. Clin Chem 31 Roberts W, Safar-Pour S, De B, Rohlfing C,
1983;29:1080-2. Weykamp C, Little R. Effects of Hemoglobin C
20 Little R, England J, Wiedmeyer H, Goldstein D. and S Traits on Glycohemoglobin Measurements
Effects of Whole Blood Storage on Results for by Eleven Methods. Clin Chem 2005;51:776-8.
Glycosylated Hemoglobin as Measured by Ion- 32 Frank E, Moulton L, Little R, Wiedmeyer H,
Exchange Chromatography, Affinity Rohlfing C, Roberts W. Effects of Hemoglobin C
Chromatography, and Colorimetry. Clin Chem and S Traits on Seven Glycohemoglobin Methods.
1983;29:1113-5. Clin Chem 2000;46:864-7.
21 Jones W, Scott J, Leary S, Stratton F, Smith S, 33 Fleming J. Evaluation of HbA1c on the Roche
Jones R, et. al. Stability of Whole Blood at -70C COBAS Integra Closed Tube System. Clin
for Measurement of Hemoglobin A1c in Healthy Biochem. 2007;40:822-7.
Individuals. Clin Chem 2004;50:2460-1. 34 Luzzi V, McKenna M, Landry C, Karsteter L,
22 Weykamp C, Martina W, van der Dijs F, Penders Mongia S, Roberts W, et al. COBAS Integra
T, van der Slik, W, Muskiet F. Hemoglobins S Tina-Quant Hemoglobin A1c Gen.2 Assay
and C: reference values for glycohemoglobin in Accurately Monitors HbA1c in Patients With S,
heterozygous, double-heterozygous and C, and E traits. Clin Chem Lab Med.
homozygous subjects, as established by 13 2007;45:S197-8
methods. Clin Chim Acta 1994;231:161-71. 35 Luzzi V, McKenna M, Landry C, Karsteter L,
23 Inaba M, Okuno S, Kumeda Y, Yamada S, Mongia S, Roberts W, et al. COBAS Integra
Imanishi Y, Tabata T, et al. Glycated Albumin Is Tina-Quant Hemoglobin A1c Gen.2 Assay:
a Better Glycemic Indicator than Glycated Accurate and Precise with no interference from
Hemoglobin Values in Hemodialysis Patients Hb variants. Clin Chem. 2007;53:A231
with Diabetes: Effect of Anemia and 36 Lenters-Westra E, Miedema K, Slingerland R.
Erythropoietin Injection. J Am Soc Nephrol Evaluation Of A New HbA1c method on COBAS
2007;18:896-903. INTEGRA 800 and its certification as a NGSP
24 Sacks D. Hemoglobin Variants and Hemoglobin secondary reference method. Clin Chem
A1c Analysis: Problem Solved? Clin Chem 2007;53(S6):A183.
2003;49:1245-7. 37 Rose A, Tongate C, Valdes R. A Hemoglobin
25 Chang J, Hoke C, Ettinger B, Penerian G. A1c Immunoassay Method Not Affected By
Evaluation and Interference Study of Hemoglobin Carbamylated Hemoglobin. Ann Clin Lab Sci
A1c Measured by Turbidimetric Inhibition 1995;25:13-9.
Immunoassay. Am J Clin Pathol 1998;109:274-8. 38 Weykamp C, Penders T, Muskiet F, van der Slik
26 Chen D, Crimmins D, Hsu F, Lindberg F, Scott M. W. Influence of Hemoglobin Variants and
Hemoglobin Raleigh as the cause of a falsely Derivatives on Glycohemoglobin Determinations,
increased hemoglobin A1c in an automated ion- as Investigated by 102 Laboratories Using 16
exchange HPLC method. Clin Chem Methods. Clin Chem 1993;39:1717-23.
1998;44:1296-301. 39 Weykamp C, Penders T, Siebelder C, Muskiet, F,
27 Watanabe T, Kato K, Yamada D, Midorikawa S, van der Slik W. Interference of Carbamylated and
Sato W, Shiga M, et al. A Nondiabetic Case of Acetylated Hemoglobins in Assays of
Hemoglobin Variant (Hb Niigata) with Glycohemoglobin by HPLC, Electrophoresis,
Inappropriately High and Low HbA1c Titers Affinity Chromatography, and Enzyme
Detected by Different Methods. Clin Chem Immunoassay. Clin Chem 1993;39:138-42.
1998;44:1562-4. 40 Stevens V, Fantl W, Newman C, Sims R, Cerami
28 Bisse E, Schauber C, Zorn N, Epting T, Eigel A, A, Peterson C. Acetaldehyde Adducts with
Van Dorsselaer A, et al. Hemoglobin Grwihl Hemoglobin. J Clin Invest 1981;67:361-9.
[225(A2)ProAla], an Electrophoretically 41 Schnedl W, Liebminger A, Roller R, Lipp R,
Silent Variant with Impaired Glycation. Clin Krejs G. Hemoglobin variants and determination
Chem 2003;49:137-43. of glycated hemoglobin (HbA1c). Diab Metab
29 Little R, Vesper H, Rohlfing C, Ospina M, Safar- Res Rev 2001;17:94-8.
Pour S, Roberts W. Validation by a Mass 42 Tsai L, Tsai S, Lin M, Liu S. Effect of
Spectrometric Reference Method of Use of Hemoglobin Variants (Hb J, Hb G, and HbE) on
Boronate Affinity Chromatography to Measure
668
Glycated Hemoglobin
HbA1c Values as Measured by Cation-Exchange Measurement in a Korean Population. Clin Chem

HPLC (Diamat). Clin Chem 2001;47:756-8. 2006;52:1445-6.
43 Wong S, Aw T. Hb E1c as an Indicator for the 55 American Diabetes Association. Standards of
Presence of Hb AE Phenotype in Diabetic Medical Care in Diabetes 2007. Diabetes Care
Patients. Clin Chem 1998;44:660-2. 2007;30:S4-S41.
44 Lahousen T, Roller R, Lipp R, Schnedl W. Silent 56 http://www.ngsp.org/
Haemoglobin variants and determination of 57 Pizzi R. The Quest for Glycohemoglobin
HbA1c with the HPLC Bio-Rad Variant II. J Clin Standardization. Clin Lab News October
Pathol 2002;55:699-703. 2006;32:1-4.
45 Friess U, Beck A, Kohne E, Lehmann R, Koch S, 58 Hicks J. Why the Quest for Glycohemoglobin
Haring H, et al. Novel Hemoglobin Variant Standardization Should Move Forward. Clin Lab
[66(E10) LysAsn], with Decreased Oxygen News December 2006.
Affinity, Causes Falsely Low Hemoglobin A1c 59 Sacks D. Global Harmonization of Hemoglobin
Values by HPLC. Clin Chem 2003;49:1412-5. A1c. Clin Chem 2005;51:681-3.
46 Camargo J, Gross J. Conditions associated with 60 Kilpatrick E, Rigby A, Atkin S. Variability in the
very low values of glycohaemoglobin measured Relationship between Mean Plasma Glucose and
by an HPLC method. J Clin Pathol 2004;57:346-9. HbA1c: Implications for the Assessment of
47 Ostendorf N, Glosemeyer-Allhoff T, Schauerte K, Glycemic Control. Clin Chem 2007;53:897-901.
Yamamoto R. Hemoglobin (Hb) Stanleyville II 61 Goldstein D, Little R, Lorenz R, Malone J,
Causes Gross Overestimation of the HbA1c Nathan D, Peterson C. Tests of Glycemia in
Proportion in Routine HPLC. Clin Chem Diabetes. Diabetes Care 1995;18:896-909.
2005;51:1752-4. 62 Snieder H, Sawtell P, Ross L, Walker J, Spector T,
48 Roberts, W. Hemoglobin Constant Spring Can Leslie R. HbA1c Levels Are Genetically
Interfere with Glycated Hemoglobin Determined Even in Type 1 Diabetes. Diabetes
Measurements by Boronate Affinity 2001;50:2858-63.
Chromatography. Clin Chem 2007;53:142-3. 63 Yudkin S, Forrest R, Jackson C, Ryle A, Davie S,
49 De B, Brown D, Hanbury M, Hoyer J, John G, Gould B. Unexplained variability of glycated
Lambert T, et al. Effects of hemoglobin C and S haemoglobin in non-diabetic subjects not related
traits on glycohemogobin measurements by six to glycaemia. Diabetologia 1990;33:208-15.
methods. Clin Chem 2001; 46(S6): A18. 64 Consensus Committee. Consensus Statement on
50 Biss E, Zorn N, Eigel A, Lizama M, Huaman- the Worldwide Standardization of the
Guillen P, Mrz W, et al. Hemoglobin Rambam Hemoglobin A1C Measurement. Diabetes Care
(69[E13]GlyAsp), a pitfall in the assessment 2007;30:2399-400.
of diabetic control: characterization by 65 Bunn H, Haney D, Kamin S, Gabbay K, Gallop P.
electrospray mass spectrometry and HPLC. Clin The Biosynthesis of Human Hemoglobin A1c
Chem 1998;44:2172-7. Slow Glycosylation of Hemoglobin in Vivo. J
51 Nakanishi, T, Miyazaki A, Shimizu A, Clin Invest 1976;57:1652-9.
Yamaguchi A, Nishimura S. Assessment of the 66 Bunn H, Haney D, Gabbay K, Gallop P. Further
effect of hemoglobin variants on routine HbA1c identification of the nature and linkage of the
measurements by electrospray ionization mass carbohydrate in hemoglobin A1c. Biochem
spectrometry. Clin Chim Acta 2002;323:89-101. Biophys Res Communs 1975;67:103-9.
52 Davie S, Gould B, Yudkin J. Effect of Vitamin C 67 Bunn H, Gabbay K, Gallop P. The glycosylation
on Glycosylation of Proteins. Diabetes of hemoglobin: relevance to diabetes mellitus.
1992;41:167-73. Science 1978;200:21-27.
53 Rohlfing C, Connolly S, England J, Little R. 68 Shapiro R, McManua M, Zalut C, Bunn H. Sites
Effect of Elevated Fetal Hemoglobin on HbA1c of Nonenzymatic Glycosylation of Human
Measurements: Four Common Assay Methods Hemoglobin A. J Biol Chem 1980;255:3120-7.
compared to the IFCC Reference Method. Clin 69 Sacks D, Bruns D, Goldstein D, Maclaren N,
Chem 2006;52(S6):A36. McDonald J, Parrott M. Guidelines and
54 Lee S, Kim M, Choi D, Kim S, Ki C. Incidence of Recommendations for Laboratory Analysis in the
Variant Hemoglobin (Hb) and Increased Fetal Hb Diagnosis and Management of Diabetes Mellitus.
Concentrations and Their Effect on HbA1c Clin Chem 2002;48:436-72.
669
Glycated Hemoglobin
Table 1: Glycated HemoglobinMethods of Analysis
Method 1: Boronate Affinity Column Chromatography

Principle of analysis: Spectral absorbance post separation of glycated and nonglycated hemoglobin. The
glycated component binds to boronate, while the nonglycated component passes through the column.
Comments: Measures total glycated hemoglobin (cis-diol groups present in sugars at the amino terminus,
as well as lysine residues downstream on both the and hemoglobin chains). Can be calibrated to reflect
HbA1c equivalent values. Preferentially binds 1-deoxyfructosyl derivatives.
Method 2: Electrophoresis
Principle of analysis: Spectral absorbance of separated hemoglobin components
Comments: Separation of hemoglobin species is based on charge differences. Any substance with the same
charge as HbA1c is measured as HbA1c.
Method 3: Enzymatic
Principle of analysis: Glycated valines serve as substrate for fructosyl valine oxidase in a coupled reaction
with horseradish peroxidase. The resulting colored agent, together with hemoglobin, is measured
spectrophotometrically.
Comments: New method.
Method 4: Immunoassay
Principle of analysis: Antibody-mediated inhibition of latex agglutination or immunoturbidimetry.
Comments: Measures HbA1c (four-plus N-terminal amino acids of hemoglobin and the ketoamine
linkage and glucose). Amino acid substitutions in this region may affect antibody recognition.
Method 5: Ion-Exchange HPLC
Principle of analysis: Spectral absorbance of separated hemoglobin components
Comments: Separation of hemoglobin species is based on charge differences. Any substance with the same
charge as HbA1c is measured as HbA1c.
Table 2 Comparison of Methods

Method *CLIA Complexity Type of Separation
Boronate Affinity Moderate (manual columns high) Structural differences
Automated HPLC
Electrophoresis High Charge differences
Enzymatic Moderate Structural differences
Immunoassay Moderate (POC waived) Structural differences
Ion-Exchange HPLC Moderate (manual columns high) Charge differences
*CLIA Complexity from www.accessdata.fda.gov
Figure 1. Formation of stable product.

670
Haptoglobin
Haptoglobin
Joris R. Delanghe
Name: Haptoglobin, Hp
Clinical significance: Refer to Chapter 40, Hemoglobin, in the 5th edition of Clinical Chemistry:
Molecular mass: 85,000 D (type 1-1); other genetically determined types (2-1, 2-2)
form a series of higher-molecular-weight polymers

i spectrophotometrically as reduced absorbance. The
Early methods for quantitation of serum haptoglobin intensity of the incident light beam is reduced because of
were based on formation of a complex between the light-scattering effects (Rayleigh and Mie scattering)
haptoglobin (Hp) and hemoglobin (Hb). A known of the immune complexes [2]. Addition of polyethylene
amount of hemoglobin was added to serum. If no glycol is used to accelerate the formulation of the
complex was formed, no Hp was present; if all the Hb complex. Readings usually occur at 340 nm. The
formed Hb-Hp complexes, the concentration of Hp was absorbance is plotted against Hp concentration on a
greater than the concentration of Hb; and if complexes calibration curve; the absorbance is inversely
formed, leaving free Hb, the concentration of Hp was proportional to the concentration of the haptoglobin.
less than the concentration of added Hb. The free Hb
was separated from the Hb-Hp complex by use of Enzyme-linked immunoadsorbent assays (ELISA, Table
electrophoresis or gel filtration, and the hemoglobin in 1, Method 2d) are the most recent development in the
each component was measured by use of its peroxidase measurement of haptoglobin. In one of these techniques
activity (Table 1, Method 1). The amount of total Hb [4], a specific monoclonal antibody (MoAb) to
complexed with Hp was the hemoglobin-binding haptoglobin was produced using pooled fecal extracts
capacity and was taken as a measure of the Hp level. containing the Hp antigen from patients with colorectal
Reference intervals for methods based on binding of Hb cancer. The MoAb developed was coated on polyvinyl
were expressed as milligrams of Hb bound per liter. chloride (PVC) microtiter plates. The extracts from feces
are added and incubated at 4C overnight, allowing
Current methods for measuring haptoglobin are haptoglobin to bind to the MoAb. This complex is
immunological, including nephelometry (Table 1, reacted with a streptavidinhorseradish peroxidase
Method 2a), turbidity (Table 1, Method 2c), radial conjugate to form a sandwich. After washing, the plates
immunodiffusion (RID) (Table 1, Method 2b), and are developed by adding substrate (2,2-azino; 3-
enzyme-linked immunosorbent assays (ELISA, Table 1, bisethylbenzthiazoline-6-sulfonate [ABTS]) and citrate
Method 2d). In the nephelometric method, serum buffer containing 0.02% hydrogen peroxide. The
containing an unknown amount of haptoglobin is mixed absorbance at 405 nm is directly related to the amount of
with anti-haptoglobin antibody. The formation of an Hp.
antigen-antibody complex results in light scattering
when a beam of monochromatic light is passed through Radial immunodiffusion is no longer routinely used,
the sample. The degree of light scattering is proportional because the method is not very precise. Furthermore,
to the concentration of haptoglobin [1,2]. An alternative serum concentrations of the Hp2-2 phenotype are
nephelometric procedure that measures the rate of underestimated because of the difficulty of migrating in
formation of the antigen-antibody complex is available agar. In the RID method, an antibody to haptoglobin is
[3]. This method has been shown to correlate closely evenly dispersed in a thin, two-dimensional gel. Small
with a hemoglobin-binding assay [3]. holes, or wells, are cut into the gel. A small amount of
serum (typically 5 L) is placed into the well and
Turbidimetric methods (Table 1, Method 2c) are similar allowed to diffuse radially into the gel surrounding the
to nephelometric assays in that haptoglobin binds to an well, forming a precipitin ring. The concentration of
anti-haptoglobin antibody forming a complex, which haptoglobin in the sample is directly related to the
causes turbidity. The turbidity is measured diameter or area of the ring.
Another ELISA technique [5,6] is based on the reaction

i
Haptoglobin of the carbohydrate moiety of haptoglobin with a
Previous and current authors of this method: mannosyl-specific lectin, concanavalin A (Con A). Con
First edition: Robert S. Franco A is immobilized to polystyrene microtiter plates.
Methods edition: Robert S. Franco Haptoglobin reacts quantitatively with the Con A,
Second edition: Not updated forming a stable complex. Polyclonal or monoclonal Hp
Third edition: Steven C. Kazmierczak antibodies [5,6] conjugated with horseradish peroxidase
Fourth edition: Harini Patel and Joris R. Delanghe are allowed to react with the haptoglobin complex,
Fifth edition: Joris R. Delanghe forming a sandwich. Again, substrate is added to
produce a colored product measured
671
Haptoglobin
spectrophotometrically. The absorbance at 405 nm is Haptoglobin Reference Interval

directly related to the amount of Hp. The range of Hp An international reference preparation for plasma
binding to Con A is 25 to 300 g/L using polyclonal proteins, lot CRM 470, certified by the Bureau of
antibodies, and 50 to 600 g/L using monoclonal anti- Reference of the European Community (BCR), has been
haptoglobin antibody. Streptococcal cells with adopted worldwide [2]. Calibrated against this standard,
haptoglobin binding receptors have also been used as the reference interval for haptoglobin in serums of adults
Hp-capture proteins in an ELISA system [7]. is between 0.3 and 2.0 g/L [11]. This reference interval
includes all three main subtypes of haptoglobin.
A horizontal electrophoresis method (Table 1, Method 3) Newborns have little or no haptoglobin; their levels
using a discontinuous polyacrylamide gel has been approach those of adults by about 4 months [12]. Males
described for measurement of haptoglobin polymorphs reportedly show haptoglobin concentrations that are 20%
from autopsy blood samples. This method has to 25% greater than those found in females [13].
advantages over the traditional vertical gel method,
including smaller sample requirement, easier gel There are two allelic, autosomal genes (Hp1 and Hp2),
preparation, and shorter run time [8]. resulting in three major phenotypes (Hp1-1, Hp2-1, and
Hp2-2). Haptoglobin reference intervals are phenotype-
Reference and Preferred Methods dependent: 0.57 to 2.27 g/L (Hp1-1), 0.44 to 1.83 g/L
There is no reference method for measuring haptoglobin (Hp2-1), and 0.38 to 1.5 g/L (Hp2-2) [2,14,15].
concentration. Nephelometric and turbidimetric methods
are to be preferred because of their suitability for Expression of the Hp gene is absent in
automated analysis in combination with a good anhaptoglobinemia (Hp0-0 phenotype), a condition
analytical performance. Nephelometry is often used present in approximately 1 in 1000 Caucasians. In
when large numbers of samples must be analyzed. This blacks, especially of West African origin (Nigeria,
method is available on automated instruments. Cameroon), anhaptoglobinemia is more frequent
Turbidimetry may be an attractive alternative if this is (>30%). In the United States, frequency of Hp0-0 in
automated on random-access chemistry analyzers. blacks is considerably less: 4%. Hypohaptoglobinemia
Advantages of this method include a small sample size, a has also been reported in a few nonblack families
fast turnaround time with results available the same day, carrying a silent allele with no gene product, Hp0
and relatively high precision. Disadvantages are [2,15].
relatively high reagent costs and the need for a
specialized piece of equipment. Interpretation
Haptoglobin is a genetically polymorphic serum protein
Radial immunodiffusion features a very small sample that binds free hemoglobin and thereby facilitates
size, extreme simplicity when commercially prepared hemoglobin removal by the reticuloendothelial system.
plates are used, and essentially no capital investment. A decreased haptoglobin concentration is therefore
One disadvantage of RID is the relatively long time evidence of increased erythrocyte destruction.
before the precipitin rings can be measured, typically 18
to 48 hours. Several studies indicate that for European and American
whites, type 1-1 occurs in 13% to 21% of the population,
The sensitivity of the concanavalin A ELISA assay is type 2-1 in 47% to 59%, and type 2-2 in 28% to 36%.
higher than that of affinity immunoelectrophoresis by an Type 1-1 is more common in blacks, whereas type 2-2
order of 1000 [9]. Binding of the Hp in the ConA ELISA predominates in Asian populations.
depends on the mannose-type carbohydrate moiety of
haptoglobin. Thus changes in the concentration of Haptoglobin is a positive acute-phase protein; that is, it
haptoglobin in various pathological conditions could be becomes elevated in blood after an acute disease or
related to quantitative differences in hepatic haptoglobin shock to the body. It may be especially elevated in
glycosylation [9]. The comparison of the reaction seriously burned patients. In severe psychiatric
conditions for RID, nephelometry, and turbidimetry of conditions, such as major depression, very high Hp
haptoglobin is found in Table 2. concentrations are found.
Specimen Haptoglobin is decreased in intravascular hemolysis as a

Serum from an unhemolyzed blood specimen is used for result of binding with hemoglobin, with subsequent
haptoglobin determination. The serum may be stored for removal of the hemoglobin-haptoglobin complex by the
1 week at 2C to 8C or frozen at 20C for longer liver. The rate of removal of these complexes by the
storage [10]. Avoid repeated freeze/thaw cycles. liver is very high (serum Hb-Hp complexes show a half-
life of a few minutes). Serum Hb levels also decrease in
Interferences extravascular hemolysis if the capacity of removal by the
Specimens should not be hemolyzed. Samples that are reticulo-endothelial system is insufficient and the free
grossly lipemic or turbid may interfere with some hemoglobin enters the bloodstream. Note that
nephelometric or turbidimetric methods. simultaneous hemolysis and inflammation may result in
a low, normal, or high serum haptoglobin level.
672
Haptoglobin
Because of the limited hemoglobin-binding capacity (1 blood testing based on the detection of
mole of Hp binds 1 mole of Hb) [16], haptoglobin can haptoglobin. Cancer 1996;78:48-56.
only be used as a parameter of mild or moderate 5 Katnik I, Dobryszcka W. Development of
hemolysis. Hemopexin is a heme-binding plasma protein concanavalin Aenzyme immunosorbant assay
which, after Hp, forms the second line of defense against for glycated haptoglobin using polyclonal and
hemoglobin-mediated oxidative damage during monoclonal antibodies. J. Immunoassay
hemolysis. A decrease in plasma hemopexin 1992;13:145-162.
concentration reflects a recent release of heme 6 Katnik I, Judach J, Kmieciak K, Gerber J,
compounds in the extracellular compartment. Heme- Dobryszcka W. Measurement of haptoglobin by
hemopexin complexes are cleared by receptor-mediated the reaction with concanavaline a in serum of
endocytosis. Hemopexin determination in tandem with patients with ovarian tumors. Eur J Clin Chem
Hp is useful in the assessment of hemolysis and allows Clin Biochem 1995;33:727-732.
for the monitoring of the severity of hemolysis after 7 Katnik I, Lammeler C, Guszczynski T,
depletion of Hp. After saturation of the Hb binding Dobryszycka W. Quantitation of human
capacity, concentrations of serum hemopexin start to haptoglobin by ELISA system based on
decrease, whereas the Hp concentration remains low streptococcal haptoglobin receptors. Arch
(<0.3g/L). During intense hemolytic diseases, Immunol Ther Exp 1993;41:105-109.
monitoring of serum hemopexin should be preferred to 8 Quarino L, Samples M, San Pietro D, Shaler R,
monitoring Hp. In contrast to other markers for Orta A, Jack D. Haptoglobin typing of
hemolysis (e.g., lactate dehydrogenase activity, bloodstains using horizontal discontinuous
potassium), Hp and hemopexin concentrations are not polyacrylamide gel electrophoresis. Sci Justice
influenced by in vitro hemolysis [15]. 1995;35:213-216.
9 Valette I, Pointis J, Rondeau, Y, et al. Is
One report describes the measurements of haptoglobin immunochemical determination of haptoglobin
concentration in the serum to differentiate between phenotype dependent? Clin Chim Acta
malignant and nonmalignant tumors of the ovary [6]. 1979;99:1-6.
Another report describes the detection of haptoglobin in 10 Dale JC, Pruett SK. Phlebotomy a minimalist
fecal matter by use of an ELISA technique [4]. The approach. Mayo Clin Proc 1993;68:249-255.
authors argue that detection of haptoglobin has many 11 Dati F, Schumann G, Thomas L, et al. A new
advantages compared to the current fecal occult blood international reference preparation for proteins
test for early detection of colorectal carcinoma. These in human serum (RPPHS). Eur J Clin Chem
advantages include no requirement for dietary restriction Clin Biochem 1996;34:517-520.
and the ability to differentiate between upper and lower 12 Davis ML, Austin C, Messmer BI, Nichols
gastrointestinal bleeding with the same sensitivity and WK, Bonin AP, Bennett MJ. IFCC-
specificity of the current test. standardized pediatric reference intervals for 10
serum proteins using the Beckman Array 360
References system. Clin Biochem 1996;29:489-492.
1 Van Lente F, Marchand A, Galen RS. 13 Hashimoto S, Miwa M, Akasofu K, et al.
Evaluation of a nephelometric assay for Changes in serum proteins of post-menopausal
haptoglobin and its clinical usefulness. Clin women. Maturitas 1991;13:23-33.
Chem 1979;25:2007. 14 Van Rijn HJ, Vander Wilt W, Stroes JW,
2 Fink PC, Roemer M, Haeckel R, et al. Schrijver J. Is turbidimetric immunoassay of
Measurements of proteins with the Behring haptoglobin phenotype dependent? Clin
Nephelometer: a multicentre evaluation. J Clin Biochem 1987;20:245-248.
Chem Clin Biochem 1989;27:261-276. 15 Langlois MR, Delange JR. Biological and
3 Sternberg JC. A rate nephelometer for clinical significance of haptoglobin
measuring specific proteins by polymorphism in humans. Clin Chem
immunoprecipitin reactions. Clin Chem 1996;42:1589-1600.
1977;23:1456. 16 Nyman M. Serum haptoglobin: methodological
4 Xing PX, Young GP, Ho D, Sinatra MA, and clinical studies. Scand J Clin Lab Invest
McKenzie IF. A new approach to fecal occult 1959;11:suppl 39.
673
Haptoglobin
Table 1: Methods of Haptoglobin Analysis

Method 1: Hemoglobin-binding
a. Electrophoresis
b. Gel filtration
a. Hb separated from Hb-haptoglobin by difference in electrophoretic mobility; peroxidase stain for
hemoglobin sued to visualize bands
b. Hb separated from Hb-haptoglobin by molecular weight difference; quantitation by measurement of
A415 of hemoglobin
Comments:
a. Historical; semiquantitative
b. Not suitable for clinical use; quantitative
Method 2: Immunochemical
a. Nephelometry
b. Radial immunodiffusion
c. Turbidimetry
d. ELISA
a. Antibody reacts with haptoglobin, forming complex that scatters light
b. Haptoglobin diffuses into gel containing anti-haptoglobin antibody; precipitin ring diameter
proportional to concentration
c. HpAb reacts with Hp to form complex which results in turbidity; the turbidity is proportional to the
concentration of Hp
d. Hp reacts with a binding agent (Ab, lectin) coated on a solid phase to form a complex; this reacts with
anti-HpAb on a second site on Hp to form a sandwich; this complex is proportional to the concentration of Hp
Comments:
a. Currently in clinical use; suitable for large number of samples
b. Currently in clinical use; suitable for small number of samples
c. Currently in clinical use; suitable for labs with automated equipment
d. Currently in clinical use; suitable for labs with automated equipment
Method 3: Electrophoretic
Various Hp types are separated by charge
Comments:
Not currently in use; small sample volume
Table 2: Comparison of Conditions for Haptoglobin Analysis

Condition RID* Nephelometry Turbidimetry
Temperature 23 2C (RT) RT RT
pH 7.4
Time of reaction 18 h 6.0 min 20 sec
Sample Serum Serum Serum, plasma
Sample size 5 L 10 L 50 L, diluted
Reportable range 25-2760 mg/L 42-1300 mg/dL 20-1200 mg/dL
Precision: Mean (% CV) 143 mg/L (7.0%) 178 (3.9%) 86 (8.3%)
*Kallestad Endoplate, Sanofi Diagnostics Inc., Chaska, MN

Behring Nephelometer Analyzer (BNA), Dade Behring Diagnostics Inc., San Jose, CA
Behring Turbitimer (Turbitimer), Dade Behring Diagnostics Inc., Westwood, MA
674
Hepatits B Virus
Hepatitis B Virus
Paul Coleman i
Name: Hepatitis B virus (HBV)

Principles of Analysis and Current Usage genes. Examples are the recent emergence of vaccine
The hepatitis B virus (HBV) was discovered in the early escape mutants with altered HBsAg [6] and antiviral-
1960s, using a blood sample from an Australian resistance mutants with altered polymerase proteins [7].
aborigine (therefore, HBV was initially called Australia Phylogenetic analysis of genome sequences has led to
antigen) [1]. The discovery of Australia antigen quickly the classification of HBV into eight genotypes (A to H)
led to the development of diagnostic assays for hepatitis based on a nucleotide divergence of approximately 25%
B surface antigen (HBsAg), which continues to be a key [8].
marker for HBV infection today. Forty years after its
discovery, HBV remains a global health care challenge, Acute infection with hepatitis B can result in a
with an estimated 350 million people currently infected distinctive yellow jaundice and may be accompanied by
[2]. The worldwide prevalence of HBV infection is fever, chills, fatigue, nausea, loss of appetite, and
about 10-fold higher than the global AIDS cases abdominal pain. These symptoms usually subside after
estimated by the UN AIDS agency. As a consequence of several weeks. Acute infection outcomes can range from
HBV infection, about 80 million chronic carriers have asymptomatic infection to a fulminant hepatitis that is
progressed to hepatocellular carcinoma, and over rapidly fatal. Chronic infection is usually asymptomatic.
500,000 deaths occur each year. Infection can be HBV transmitted vertically from mother to child usually
controlled by such measures as universal vaccination, results in chronic infection 80% of the time, where the
passive immunization, and more recently antiviral child lacks a robust immune response to neutralize and
therapy. Early diagnostic detection coupled with clear the virus. Chronic perinatal infection accounts for a
infection-control measures has led to a significant disproportionately large share of worldwide morbidity
reduction in HBV infection rates in developed countries and mortality. Chronic infection over a period of many
[3]. years often leads to cirrhosis, progression to liver cancer,
and death. In contrast, horizontal transmission to adults
HBV is classified in the genus Orthohepadnavirus of the through sexual contact, unsafe injections, household
Hepadnaviridae family. It has a small, partially single- contact, or other percutaneous transmission generally
stranded, circular DNA genome of approximately 3.2 results in about 90% of cases in an acute resolving
kilobases that is unique in structure. The minus-strand infection. An HBV vaccine is available (consisting of
DNA is the size of the entire genome, whereas the plus- recombinant HBsAg) which can be initiated at birth; it
strand DNA is about half this length [4]. The provides long-term protection in more than 90% of
encapsulated viral DNA contains a covalently linked healthy people. These infection profiles can be further
polymerase and an RNA oligomer to prime the complicated by co-infection with the hepatitis D virus
completion of plus-strand DNA [5]. The HBV genome that requires HBV for its replication cycle.
contains four genes with partially overlapping, open
reading frames (ORFs) that encode seven proteins: the The generalized appearance of HBV seromarkers in an
polymerase protein (Pol gene); core antigen and e acutely infected patient is presented in Figure 1. Note
antigen (C gene); large, medium, and small surface that a window period is present between the onset of
antigen proteins (S gene); and the X protein (X gene). infection and the detection of HBV seromarkers. The
Surface antigen (HBsAg) forms the outer envelope of acute HBV infection-window period extends from 1 to 3
the virion that surrounds the inner core (HBcAg), months (mean 59 days), as shown by an extensive
protecting the single-copy genome. The X protein has clinical study [9]. Improvements in diagnostic
been implicated in the integration of HBV DNA into the technology continue to narrow this window period
host genome and can also suppress viral replication; the during which an infectious, asymptomatic patient is
multifunctional polymerase protein directs the capable of spreading disease [10]. Closing the detection
replication of HBV DNA. window also allows earlier therapeutic intervention with
improved prognosis for the patient.
The HBV DNA replication cycle is unique as it proceeds
through an error-prone RNA reverse transcriptase During the infection cycle, HBV secretes large amounts
intermediary step where misincorporation of nucleotides of incomplete virions consisting of HBsAg only and also
into the copy genome results in variants. Although gene secretes e antigen (HBeAg) in a process that is thought
overlap constrains some viral variability, mutant or to challenge the host immune response and permit
variant forms have been identified for all four HBV complete virions to continue the infection cycle.
i
Hepatitis B
New method:
Fifth edition: Paul Coleman
675
Hepatits B Virus
Diagnostic monitoring of HBV antigens and the confirmatory assay was also introduced which utilized
subsequent host antibody response can differentiate both high-titered human anti-HBs to reduce the RIA signal of
the stage and severity of infection, as indicated in Table true positives while false-positive signal remained
1. The sentinel diagnostic markers for detecting HBV unchanged. In this manner, the investigator could rapidly
infection are HBsAg and HBV DNA. A standardized distinguish true-positive results from false positives.
testing approach has recently been proposed [11]. While RIA technology was the mainstay of HBV testing
for decades, the relatively short half-life of the
Table 1: Description of HBV Seromarkers radioisotope and radioactive disposal concerns led to the
and Their Diagnostic Outcome emergence of enzyme and chemiluminescent
immunoassays.
HBV HBV infection outcome
Seromarker Immunoassay
Immunoassays to detect HBsAg primarily utilize one or
HBsAg Persistence for longer than 6
more monoclonal antibodies directed against the a
months can indicate chronic
determinant of HBsAg to immobilize the antigen on a
infection
solid phase (Table 3, Method 3). The a determinant is
Anti-HBs Seroconversion to anti-HBs a highly conformational, hydrophilic domain between
can indicate immunity HBsAg amino-acid positions 100 to 160 that is stabilized
HBeAg Determines relative by a backbone of conserved, disulfide-bonded cysteine
infectivity residues. Bound HBsAg is detected by a second mono-
or polyclonal antibody to which either an enzyme
Anti-HBe Seroconversion to anti-HBe
(horseradish peroxidase, alkaline phosphatase) or a
indicates progression
chemiluminescent reporter (acridinium, ruthenium) is
towards infection resolution
conjugated. Sensitivity of HBsAg assays is evaluated
Anti-HBc IgM Differentiates acute/recent
using HBV seroconversion panels and antigen end-point
infection from chronic
detection. The sensitivity of blood-screening HBsAg
carrier state or resolved
assays is now less than 0.1 ng/mL and have been shown
HBV infection
to detect a wide range of viral variants [13]. Assay
Anti-HBc Helps to establish the stage
sensitivity at this low sensitivity level closes the acute-
of chronic infection
window period of infection by about 9 days [14].
HBV DNA Persistence without Confirmatory assays are available to identify false-
seromarkers can indicate positive test results. Recently there is an emerging
occult infection interest in quantitating the patient HBsAg level in order
to monitor response to antiviral therapy. The non-US
Agar Gel Diffusion Test Architect HBsAg assay (Abbott) is currently the only
The first screening assay for HBsAg was gel diffusion commercial immunoassay with an HBsAg quantitative
based (Table 3, Method 1). This technology involved claim.
adding high-titered guinea pig anti-HBs to a central well
that was ringed by test wells to which test serum samples Nucleic Acid Testing
were added. The samples were allowed to diffuse Nucleic acid testing (NAT) HBV assays amplify target
together. If HBsAg was present in the test serum sample, regions of circulating viral genomes in a patient blood
the binding of antigen by the guinea pig anti-HBs caused sample (Table 3, Method 4). The amplified viral product
the formation of a visible aggregate called a precipitin can be detected (and in some cases quantified) by
line. There are three characteristic precipitin line hybridizing with specific oligonucleotide probes
reactions: a curved identity line, a crossed nonidentity conjugated with a visualization ligand. A colorimetric or
line, and partial identity when common and unique luminescent reaction is then read. Multiple NAT targets
antigenic determinants are present. These tests required can be amplified from one sample extraction. NAT
large amounts of antigen and antibody for proper testing requires dedicated areas for sample extraction
precipitin performance, making them relatively versus detection to reduce the risk of contamination and
insensitive compared to more modern tests. This test is is generally regarded as technically more complex than
rarely used today, but it did set a precedent to EIA technology. Two NAT technologies for HBV
differentiate true-positive from false-positive results. detection are in use in blood bank settings: polymerase
chain reaction (PCR) and transcription-mediated
Radioimmunoassay amplification (TMA).
Radioimmunoassays (RIA) to detect HBsAg were
introduced in the early 1970s with the Ausria-125I test The COBAS AmpliScreen HBV test (Roche) is an
(Abbott) (Table 3, Method 2) [12]. Sensitivity of the example of a PCR qualitative in-vitro test for the direct
RIA was 30 times that of the gel-diffusion assay. A detection of HBV DNA in human plasma pools. PCR
serum sample was incubated in polypropylene tubes produces DNA amplicons from a target template. The
coated with guinea pig hyperimmune anti-HBs. Bound COBAS test includes an internal control for extraction
antigen (if present) was detected in a second incubation and amplification and has a reagent to prevent
by forming a sandwich with 125I-labelled guinea pig amplification of preexisting amplicons. An example of
anti-HBs and then measured in a gamma counter. A TMA technology is the qualitative PROCLEIX ULTRIO
676
Hepatits B Virus
HBV assay (GenProbe) that produces RNA amplicons Specimen

from a target template. These NAT technologies have a HBV test specimens can be serum, plasma, or cadaveric
limit of HBV detection at approximately 30 samples. Tissue samples such as liver biopsies can also
genomes/mL in their most sensitive procedures, so be tested for HBV. Studies have shown NAT sensitivity
clinical sensitivity will vary depending on the specimen to be unaffected by eight short-term freeze/thaw cycles
pooling formatwhether the test serum pool is [19]. HBsAg is an multimeric antigen that presents with
composed of 24, 16, or 8 donor specimens. Individual three-dimensional epitopes. The Abbott PRISM HBsAg
donor testing (ID NAT) is the most sensitive testing assay permits shipping at ambient temperature (30C or
format. colder) for a period not to exceed 7 days, then storage
Closure of the HBV infection-window period by NAT for up to 14 days at 2C to 8C. If storage periods
testing is tied to the size of the test specimen pool. Pools greater than 14 days are anticipated, the serum or plasma
of 24, 16, or 8 specimens show less, equivalent, or a should be removed from the clot or red blood cells to
slight advantage (~3 days), respectively, in window- avoid hemolysis. Serum or plasma should be stored
period closure compared to best-in-class HBsAg blood frozen (20C or colder). For cadaveric specimens,
screening immunoassays, while ID NAT closes the storage conditions are more restrictive, given the nature
window period by 8 to 14 days [14,15]. Still, some HBV of the sample. Previously frozen specimens must be
infections are not detected by HBsAg or NAT assays in centrifuged to remove particulate material. Heparinized
blood samples but can be detected in liver biopsies [16]. blood tubes should be avoided because they can interfere
with both HBsAg and NAT assays.
Rapid Assays
Rapid immunochromatographic tests for HBsAg are Interferences
available (Table 3, Method 5). The advantages of these Blood is a variable sample, and manufactures describe
assays are room-temperature storage and a visual readout potentially interfering blood substances in their package
in 10 to 20 min. The relative insensitivity of this format inserts. Specimens with rheumatoid factor or elevated
limits their application to health care settings in triglycerides, bilirubin, or hemoglobin are evaluated for
developing countries [17]. assay interference. Various microorganisms are also
evaluated. In addition, some specimens contain human
HBV Antibody Assays anti-mouse antibodies that can affect monoclonal
Monitoring the patients immune response to HBV antibody assays by causing false-positive results.
antigens is a key diagnostic tool in establishing disease Optimizing reagent diluents and washes can reduce
susceptibility and disease state. Two antibody responses nonspecific assay interference. The package insert
are routinely measured: antibody to HBV surface antigen specificity data for high-sensitivity blood-screening
(anti-HBs) (Table 3, Method 6) and antibody to HBV HBsAg immunoassays is 99.99%, demonstrating that
core antigen (anti-HBc) (Table 3, Method 7). Anti-HBs specificity interferences have been reduced to very low
can be induced by immunization (or past infection), and levels in certain assays.
antibody levels > 10 mIU/mL indicate that the patient is
protected from further HBV infection. Anti-HBc is a Interpretation
marker of past HBV infection, and this response can The U.S. Centers for Disease Control and Prevention
persist for years after infection. Antibody to HBV e (CDC) recommendations for interpreting hepatitis B
antigen (anti-HBe) is less often monitored and can be serological markers is shown in Table 2. Additional data
used to establish the infection stage of chronic carriers. on marker interpretations is available on the CDC
website:
Reference and Preferred Methods (http://www.cdc.gov/NCIDOD/DISEASES/HEPATITIS
There is no standardized reference method for HBV /b/faqb.htm).
detection, but there are standards for evaluating various
HBV assay formats. The World Health Organization
(WHO) offers standards for HBsAg immunoassay Table 2: Interpretation of Hepatitis
testing: WHO International Standard for HBsAg B Serological Test Results
(00/588) and for NAT testing, WHO HBV DNA
International Standard (97/746) [18]. The WHO website Tests Results Interpretation
(http://www.who.int/biologicals) describes the available
reference standards. There are national standards from
HBsAg
regulatory agencies such as the U.S. Food and Drug
anti- negative
Administration (FDA), Center for Biologics Evaluation
HBc negative Susceptible
and Research (CBER), HBsAg Lot-Release Panel#12.
anti- negative
Similar HBsAg panels for Germany are available from
HBs
the Paul-Ehrlich-Institut (http://www.pei.de), and French
HBsAg standards are available (http://afssaps.sante.fr).
Lastly, there are seroconversion panels available from HBsAg
private firms to compare assay performance, such as the anti- negative Immune due to
HBsAg Sensitivity Panel (BBI Diagnostics), described HBc positive natural
on their website (http://www.seracare.com). anti- positive infection
HBs
677
Hepatits B Virus
reviews are published so the user of these tests can make

Table 2: Interpretation of Hepatitis an informed decision as to which assay performance
B Serological Test Results goals best meet their needs. These performance goals
change (and will continue to change) in response to
HBsAg factors affecting infection-marker interpretation, such as
anti- negative Immune due to universal neonatal vaccination or co-infection with
HBc negative hepatitis B another infectious agent. Since diagnostic HBV testing
anti- positive vaccination can reduce but not totally eliminate the risk of
HBs transfusion-transmitted disease, future programs to
improve blood safety must consider strategies that
couple diagnostics with a second intervention such as
HBsAg
pathogen reduction [24].
anti-
HBc positive
References
IgM positive Acutely
1 Blumberg B S, Alter H J, Visnich S. A "new"
anti- positive infected
antigen in leukemia serum. JAMA
HBc negative
1965;191:541-546.
anti-
2 Kao JH, Chen DS. Global control of hepatitis B
HBs
virus. Lancet Infect Dis 2002;2:395-403.
3 Shepard CW, Simard EP, Finelli L, Fiore AE,
HBsAg Bell BP. Hepatitis B virus infection:
anti- epidemiology and vaccination. Epidemiol Rev
HBc positive 2006;28:112-25.
IgM positive Chronically 4 Summers J, O'Connell A, Millman I. Genome
anti- negative infected of hepatitis B virus: restriction enzyme
HBc negative cleavage and structure of DNA extracted from
anti- Dane particles. Proc Natl Acad Sci USA
HBs 1975;72:4597-601.
5 Block TM, Guo H, Guo JT. Molecular virology
of hepatitis B virus for clinicians. Clin Liver
The inclusion of HBV NAT testing in these Dis 2007;11:685-706
interpretations is still under discussion. HBV NAT 6 Zuckerman J, Zuckerman A. Mutations of the
testing in blood-donor screening is currently considered surface protein of hepatitis B virus. Antiviral
optional by the FDA because the estimated benefits of Res 2003;60:75-8.
adding this test to available screening tests are thought to 7 Locarnini S. Hepatitis B viral resistance:
be very limited [20]. However, there is a clear advantage mechanisms and diagnosis. J Hep
to using NAT sequence data to identify mutants and 2003;39(Suppl):S124-32.
investigate the contribution of genotype to pathogenesis 8 Norder H, Courouc A, Coursaget P,
[21]. Improvements in NAT analysis will lead to new Echevarria J, Lee S, Mushahwar I et al. Genetic
marker interpretations in the future. diversity of hepatitis B virus strains derived
worldwide: genotypes, subgenotypes, and
HBV Performance Goals HBsAg subtypes. Intervirology 2004;47:289-
The performance characteristics of HBV assays are 309.
established in clinical evaluations of acute and chronic 9 Schreiber GB, Busch MP, Kleinman SH,
infection populations. Recommendations on how to Korelitz JJ. The risk of transfusion-transmitted
establish these performance characteristics vary by viral infections. The Retrovirus Epidemiology
country [22,23]. In the United States, for example, Donor Study. N Engl J Med 1996;334:1685-90.
CBER oversees blood screening, while diagnostic testing 10 Wendel S, Levi JE, Biagini S, Candotti D,
is under separate FDA review. In July 2007, CBER Allain JP. A probable case of hepatitis B virus
recommended that HBsAg assays for screening whole transfusion transmission revealed after a 13-
blood and blood components have an end-point month-long window period [electronic version].
sensitivity of at least 0.5 ng/mL Transfusion 2008;48(8)1602-8. Epub 2008 May
(http:www.fda.gov/cber/guidelines.htm.). This aligns 2. doi:10.1111/j.1537-2995.2008.01723.x
with the Commission of the European Communities, 11 Pawlotsky JM, Dusheiko G, Hatzakis A, Lau D,
which issued common technical specifications for in- Lau G, Liang TJ et al. Virologic monitoring of
vitro diagnostic medical devices (notified under hepatitis B virus therapy in clinical trials and
document number C[2002]1344) in May 2002 that practice: recommendations for a standardized
recommended the same end-point sensitivity of 0.5 approach. Gastroenterology 2008;134:405-15.
ng/mL HBsAg. This example reflects the initiative of 12 Ling C M, Overby LR. Prevalence of hepatitis
government regulatory agencies to align performance B virus antigen as revealed by direct
goals. Regulatory agencies routinely review the radioimmune assay with 125I-antibody. J
performance of commercially available assays with Immunol 1972;109:834-841.
HBV reference materials, since not all assays perform to
the same levels of sensitivity and specificity. The
678
Hepatits B Virus
13 Scheiblauer H, Soboll H, Nick S. Evaluation of Study Group. An international collaborative

17 CE-marked HBsAg assays with respect to study to establish a World Health Organization
clinical sensitivity, analytical sensitivity, and international standard for hepatitis B virus DNA
hepatitis B virus mutant detection. J Med Virol nucleic acid amplification techniques. Vox
2006;78(Suppl 1):S66-70. Sang 2001;80:63-71.
14 Biswas R, Tabor E, Hsia CC, Wright DJ, 19 Krajden M, Minor JM, Rifkin O, Comanor L.
Laycock ME, Fiebig EW et al. Comparative Effect of multiple freeze-thaw cycles on
sensitivity of HBV NATs and HBsAg assays hepatitis B virus DNA and hepatitis C virus
for detection of acute HBV infection. RNA quantification as measured with branched-
Transfusion 2003;43:788-98. DNA technology. J Clin Microbiol
15 Koppelman MH, Assal A, Chudy M, Torres P, 1999;37:1683-6.
de Villaescusa RG, Reesink HW et al. 20 Stramer SL. Current risks of transfusion-
Multicenter performance evaluation of a transmitted agents: a review. Arch Pathol Lab
transcription-mediated amplification assay for Med 2007;131:702-7.
screening of human immunodeficiency virus-1 21 Valsamakis A. Molecular testing in the
RNA, hepatitis C virus RNA, and hepatitis B diagnosis and management of chronic hepatitis
virus DNA in blood donations. Transfusion B. Clin Microbiol Rev 2007;20:426-39.
2005;45:1258-66. 22 Cornberg M, Protzer U, Dollinger MM,
16 Hollinger FB. Hepatitis B virus infection and Petersen J, Wedemeyer H, Berg T et al.
transfusion medicine: science and the occult. Prophylaxis, diagnosis and therapy of hepatitis
Transfusion 2008;48:1001-26. B virus (HBV) infection: the German guidelines
17 Randrianirina F, Carod JF, Ratsima E, Chrtien for the management of HBV infection. Z
JB, Richard V, Talarmin A. Evaluation of the Gastroenterol 2007;45:1281-328.
performance of four rapid tests for detection of 23 Gish RG, Locarnini SA. Chronic hepatitis B:
hepatitis B surface antigen in Antananarivo, current testing strategies. Clin Gastroenterol
Madagascar [electronic version]. J Virol Hepatol 2006;4:666-76.
Methods 2008;151(2):294-7. Epub 2008 May 6. 24 Alter HJ. Pathogen reduction: a precautionary
doi:10.1016/j.jviromet.2008.03.019 principle paradigm. Transfusion Med Rev
18 Saldanha J, Gerlich W, Lelie N, Dawson P, 2008;22:97-102.
Heermann K, Heath A. WHO Collaborative
679
Hepatits B Virus
Table 3: Methods for Hepatitis B Virus

Method 1. HBV Agar Gel Diffusion Test
Principle of analysis: Immunodetection of viral capsid
Usage: Infrequently used
Comments: Precipitin line formation, low sensitivity
Method 2. HBsAg Radioimmunoassay
Principle of analysis: Capsid antigen detected by antibody sandwich assay
Comments: Easy to use but short shelf life
Method 3. HBsAg Immunoassay
Principle of analysis: Solid-phase antigen capture detected by antibody-enzyme conjugate
Comments: Detects all HBV genotypes, moderate to good sensitivity, less expensive, not overly complex to
run, confirmatory assay available
Method 4. HBV Nucleic Acid Testing
Principle of analysis: Amplified genome targets detected by probe hybridization
Usage: Frequently used in high-volume settings
Comments: High sensitivity in individual donor testing, complex, expensive
Method 5. HBsAg Rapid Assays
Principle of analysis: HBsAg wicks across support strip, conjugate binding causes stripe to appear
Comments: Limited sensitivity, subjective visual read
Method 6. Anti-HBs Assay
Principle of analysis: Antibody to HBsAg detected by antigen sandwich assay
Usage: Frequently used to establish immune state
Comments: Antibody result of 10 mIU/mL or greater indicates protective status
Method 7. Anti-HBc Assay
Principle of analysis: Antibody to HBcAg detected by inhibition assay format
Comments: Indicates stage of infection, new assays are more specific
Figure 1: Appearance of HBV seromarkers in an acute infection profile.

680
Hepatitis C Virus
Hepatitis C Virus
A. Scott Muerhoff
Name: Hepatitis C Virus (HCV)

Clinical significance: Refer to Chapter 31, Liver Function, and Chapter 32, Viral Hepatitis: Diagnosis and
Monitoring, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.

i Currently, antibody screening test results are used in conjunction
Hepatitis C virus (HCV) is one of the five human viruses (known with other clinical parameters such as liver enzyme activity, in
as hepatitis A, B, C, D, and E) that infect the liver, causing particular ALT and AST, to aid in the diagnosis of HCV liver
inflammation. All of these viruses cause acute disease, but disease. Since antibody tests do not distinguish active from
hepatitis B and C can result in chronic infection. It is estimated by resolved infection, additional tests are used to confirm the
the World Health Organization that about 180 million people presence of HCV antibodies in serum or plasma or to detect
worldwide are infected with HCV and that 130 million are ongoing (i.e., persistent) infection. Among the latter are nucleic
chronically infected. HCV is a leading cause of liver disease in acid tests (NAT), the recently developed core antigen assay, and
developing countries and is a significant risk factor for the the even more recently developed HCV combination
development of chronic liver disease (cirrhosis) and liver cancer. antigen/antibody test. In addition, there are tests for the
In the United States, after chronic alcohol abuse, infection with determination of the HCV genotype. HCV exists as a collection
HCV is the leading indication for liver transplantation. Unlike of six major genotypes that differ in their response to antiviral
hepatitis B virus, there is no vaccine for HCV. treatment. It is therefore imperative that the infecting genotype be
determined prior to commencement of treatment. Genotyping
Hepatitis C virus was first described in 1989 as the causative assays require isolation of viral RNA from serum or plasma,
agent of non-A, non-B (NANB) post-transfusion hepatitis [1]. followed by RT-PCR to amplify specific regions of the genome
HCV is an enveloped, positive-strand RNA virus and is classified for hybridization to genotype-specific probes or for direct
as a member of the Flaviridae family. The single-strand RNA sequencing and subsequent comparison to a database of known
genome is approximately 9400 nucleotides in length and encodes HCV genotypes.
a large polyprotein precursor that is cleaved by virus-encoded
proteases and the host cellular machinery into structural Antibody Tests
components (nucleocapsid or core and envelope proteins 1 and 2) HCV antibody tests utilize 2-4 recombinant antigens or peptides
and enzymes required for processing and replication (e.g., derived from the core (c22 protein or peptide), serine
nonstructural region-3 serine protease and helicase and protease/helicase (NS3/4 region, C33c or C200 antigen), and
nonstructural region-5b RNA polymerase). Identification of this replicase (NS5) genes. Most commercial tests detect only IgG, the
virus and the subsequent introduction of serological tests for exception being the ARCHITECT anti-HCV assay (Abbott
detecting circulating antibodies resulted in a dramatic reduction in Laboratories) which detects both IgM and IgG. A prototype
the rate of post-transfusion hepatitis in the United States and research assay (Immunomatrix Inc., Gaithersburg, MD) for
worldwide [2]. detection of HCV antibodies in human oral fluid utilizes
conjugates for detection of IgA and IgM in addition to IgG [3];
Most new infections in the developed world are the result of however, this assay is not yet commercially available. Some
needle sharing. Diagnosis of acute infection is often difficult, assays utilize an enzyme conjugate, whereas others tests use
since as many as 60% to 70% of patients have no distinct chemiluminescent detection and therefore have greater dynamic
symptoms, while the remainder may exhibit jaundice, fatigue, range and often greater sensitivity. The immunoassays are
loss of appetite and abdominal pain. Antibodies to HCV are validated for serum or plasma, although some manufacturers also
detected by immunoassay in only 50% to 70% of patients at the validate their assays for detection of antibodies in cadaveric
onset of symptoms, increasing to more than 90% after 3 months. specimens, cord blood, or other bodily fluids.
Viral RNA can be detected within 1 to 3 weeks post-exposure.
Individuals who report to the physician or hospital with Confirmation of HCV infection uses an FDA-approved
symptoms related to acute infection typically have elevated serum immunostrip blot assay produced by Chiron Corporation, known
bilirubin concentration and alanine aminotransferase (ALT) as RIBA (recombinant immunoblot assay). RIBA results are
activity. Only a minority of patients require hospitalization. reported as positive (2 or more markers detected), indeterminate
Fulminant hepatitis due to HCV infection occurs rarely but can be (1 HCV marker detected), or negative (no HCV markers
present in patients with hepatitis B virus co-infection. detected). Current testing algorithms for determination of acute or
chronic infection utilize the primary antibody detection assay in
conjunction with the RIBA test and/or nucleic acid tests
(described below). The current Centers for Disease Control and
Prevention (CDC) recommendations [4] indicate that tests results
i
Hepatitis C virus from third-generation screening enzyme immunoassays (EIAs)
Previous and current authors of this method: with signal-to-cutoff (S/CO) ratios above 3.8 do not require
New method supplemental testing using RIBA, since the vast majority of these
Fifth edition: A. Scott Muerhoff individuals will confirm positive, and most are positive for
circulating RNA (see Interpretation section). Confirmation of
681
Hepatitis C Virus
infection can also be achieved by use of nucleic acid tests which tests [14,15], but the former are very useful for monitoring
detect viral genetic material (see below). antiviral therapy outcomes. The lower-limit of detection of
qualitative assays range from 5 IU/mL (TMA-based method such
Core Antigen Test as Procleix HIV-1/HCV) to 50 IU/mL (AMPLICOR 2.0 or
An assay for detection of circulating HCV antigen has been Ampliscreen). Quantitative assays exhibit low-end quantitation
recently developed. The method utilizes monoclonal antibodies to ranges of 10 IU/mL (TaqMan real-time PCR assays) to 615
capture and detect the core antigen. Two versions of this test have IU/mL (Quantiplex bDNA 3.0) [16].
been reported. One version will detect antigen only during the
window period of infection, that is, prior to the detection of Genotyping
antibodies [5-7]. Another version detects core antigen prior to and HCV exists as collection of genotypes designated 1 thru 6
after seroconversion to antibody positive status [8-10]. A [17,18]. Genotypes are defined as viral variants exhibiting more
commercial total core antigen assay, the Ortho trak-CTM test, than 30% sequence divergence from other strains. Individuals
was introduced in 2001 by Ortho Clinical Diagnostics, but this infected with genotype 1 are typically more refractory to antiviral
test is not FDA-approved. An evaluation of this assay [11] treatment (combination therapy using Ribavirin and Pegylated-
concluded that it has utility in viral load monitoring, including interferon alpha). Individuals infected with genotypes 4, 5, and 6
determination of baseline viral load, the assessment of response to exhibit better treatment outcomes but typically require more
treatment, and the examination of early viral kinetics during aggressive therapy than individuals infected with genotype 2 or 3
therapy. However, the authors concluded that the total HCV core [19]. Response to therapy is determined via quantitative
Ag assay cannot be used as a marker of viral replication for HCV measurement of viremia levels before, during, and at the end of
RNA values below 20,000 IU/mL, thus limiting its use in treatment. The treatment goal is to reduce the circulating virus to
monitoring long-term response to treatment. The routine undetectable levels at the end of therapy.
application of the core antigen assay has not yet become
established in clinical practice; however, it may prove Genotyping tests are specialized and are typically performed at
useful/economical for detection of primary infection in blood reference laboratories. Four types of tests are used: RFLP
donors in low-incidence populations where RNA testing has not (restriction fragment length polymorphism), RT-PCR with probe
been implemented and in groups at high risk for infection, such as hybridization, RT-PCR with sequencing, and the InnoGenetics
hospital personnel, laboratory workers, hemodialysis patients, Line-Probe Assay, or INNO-LiPA HCV test [15]. The latter
organ donors, and prison populations, where frequent involves the generation of PCR-amplified target sequences of the
testing/monitoring would be beneficial and more cost effective viral genomic RNA (specifically, the 5-untranslated region) that
compared to nucleic acid tests. It may also be useful for is then hybridized to type-specific probes immobilized onto solid
confirmation of infection, since there appears to be good filter paper strips. The hybridized amplicons are detected via a
correlation between core antigenemia and viremia (i.e., RNA colorimetric system that produces banding patterns characteristic
titers). of specific genotypes [20]. Because these methods require
extraction and amplification of nucleic acids, the use of
Combination Antibody-Antigen Test heparinized tubes for blood collection is not recommended.
An assay for detection of both antibodies and core antigen is
available outside of the United States and Europe in a microtiter Confirmatory Antibody Test
assay format (Abbott-Murex HCV Ag/Ab Combination). Other The only FDA-approved assay for confirmation of anti-HCV
automated methods have been reported in the literature [12] but status among EIA repeat-reactive donors is the Chiron RIBA strip
are not yet commercially available. An antibody/antigen immunoblot assay [21]. This test utilizes three peptides, one from
combination assay will be useful for improvement of blood core and two from the NS4 region, and two recombinant antigens,
supply safety in that it will detect seroconverters but, also a one from NS3 and one from the NS5 region, immobilized onto a
portion of those who are infected but have yet to develop paper membrane support (per Chiron RIBA3 Package Insert,
antibodies. While not as sensitive at nucleic acid testing (NAT; December 1998). The two NS4 region peptides are co-
see below), the assay would be useful for screening blood in those immobilized into a single band; thus there are four HCV-specific
regions/countries where HCV endemicity is low. Clinical utility antigen bands on each strip. Visualization of antibody reactivity is
of the automated combination assay has not been established. done using an anti-human IgG enzyme conjugate and colorimetric
substrate. Since the recombinant antigens are fused to human
Nucleic Acid Testing (NAT) superoxide dismutase (hSOD), a control band of hSOD is
Infection with hepatitis C virus can be determined by detection of included. Also included are two human IgG control bands, one at
viral RNA in serum or plasma [13]. Blood donations in the each end of the strip and at different levels, to control for the
United States and many other countries are tested for HCV RNA detection step and to determine assay background. For the test to
by using nucleic acid amplification testsspecifically, reverse- be valid, there must be reactivity to both human IgG control
transcription polymerase chain reaction (RT-PCR) or an bands, and the reactivity to the hSOD must be lower than the
isothermal method known as transcription-mediated Level 1 hIgG control. Interpretation of assay results are based
amplification (TMA). These tests are also used for confirmation upon band intensity and banding pattern (see Table below). The
of HCV infection among individuals who are antibody positive. use of the RIBA confirmatory (also known as supplemental
In addition, they can be used to monitor treatment or disease testing) has diminished in recent years with the advent of nucleic
outcome in infected individuals. The assays have different acid testing and with the recommendation by the CDC that
sensitivity end-points typically measured in International Units confirmatory testing need only be done when S/CO ratios from
per mL (IU/mL) and are either qualitative or quantitative. screening or diagnostic tests are low (see algorithm in Figure 1).
Quantitative assays typically are less sensitive than qualitative
682
Hepatitis C Virus
Scoring and Interpretation of RIBA test results. Specimen

RIBA Band Intensity Specimens for testing in HCV antibody assays should ideally be
Score collected as serum, citrated plasma, or EDTA plasma. Some
Absent assays allow for use of lithium or sodium heparin plasma tubes,
(minus) but some do not (e.g., Abbott PRISM anti-HCV). Samples can
Less than intensity of Level 1 IgG control band be stored at room temperature for 1 to 3 days and up to 7 days at
+/ 4C. For long-term storage, temperatures below 10C are
Equal to intensity of Level 1 IgG control band recommended. It is best to avoid repeated freeze/thaw cycles. For
1+ core antigen tests, specimens can be stored at 2C to 8C for 3 to
Greater than intensity of Level 1 IgG control band and 5 days but should be stored frozen (<15C) if testing will be
less than that of Level 2 IgG control band 2+ done beyond 5 to 7 days. Repeated freeze/thaw cycles can reduce
Equal to intensity of Level 2 IgG control band core antigen detection sensitivity by up to 25% [23].
3+ Manufacturers of the test kits provide detailed information
Greater than intensity of Level 2 IgG control band 4+ regarding sample collection, storage, and handling in their
package inserts. Heparinized tubes should be avoided if
RIBA Antigen Band Pattern specimens are to be used for nucleic acid testing.
Interpretation
No bands with 1+ or greater reactivity Interferences
Or Analytical specificity among individuals infected with other
hSOD band alone with 1+ or greater reactivity microorganisms is excellent (usually >98.5%), with most positive
NEGATIVE results confirmed as true HCV antibody positives. Elevated levels
Any single HCV band with 1+ or greater reactivity of some substances such as bilirubin, hemoglobin, serum lipids,
Or and red blood cells can interfere but only when present at very
hSOD band with 1+ intensity AND one or more HCV high levels. Frozen samples should be centrifuged prior to testing.
bands with 1+ or greater intensity Manufacturers of the test kits provide detailed information
INDETERMINATE regarding interfering substances in their package inserts.
At least two HCV bands with 1+ or greater reactivity
POSITIVE Interpretation
An algorithm for hepatitis C virus antibody laboratory testing and
Reference and Preferred Methods result reporting, as recommended by the CDC, is shown in Figure
There are no currently accepted reference methods for 1. This algorithm is applicable to the diagnostic and public health
determination of HCV antibodies. HCV RNA detection assays setting and requires that RIBA testing be done for those who are
that are approved for routine clinical use are calibrated to the EIA-repeat reactive but NAT-nonreactive [4,24]. The FDA has
World Health Organization International Standard. IgG antibodies recently allowed blood banks variances to allow similar methods
are detected within 3 months of onset of symptoms in more than for blood donors [25,26]. Results from third-generation antibody
90 percent of cases. The average window period between RNA screening tests are interpreted based upon S/CO ratios.
and IgG detection is estimated at about 60 days [22]. Due to the Individuals with S/CO ratios > 3.8 will return positive results via
rapid ramp-up in viremia, HCV RNA is often detectable within 1 confirmatory immunoblot assay > 95% of the time, hence
to 3 days of exposure. If acute infection is suspected, HCV RNA supplemental testing via RIBA is not required, and individuals are
should be checked within 4 to 6 weeks of exposure. HCV core reported as HCV infected. If S/CO ratios are < 3.8, supplemental
antigen is also detected within the window period but typically a testing via RIBA3 and/or NAT is recommended. Interpretation of
few days after RNA. In chronically infected individuals, results from the RIBA confirmatory test are as discussed above.
antibodies tend to persist, though viremia can fluctuate between Nucleic acid test results are typically reported as negative (below
low, even undetectable, levels and very high titers (>800,000 the limit of detection) or positive for qualitative tests.
IU/mL). For children born to HCV-positive mothers, presence of Quantitative test results are reported in IU/mL, though they are
antibodies should be checked at regular intervalsif still positive not typically used for diagnosis but rather therapeutic monitoring.
at 18 months, qualitative NAT is preferred; if positive, infant is Interpretation of screening test results can include consideration
chronically infected; if negative, NAT should be repeated in 6 to of supplemental test results (e.g., RIBA).
12 months [16].
Diagnostic antibody test results are also reported as S/CO ratios.
Typically, values > 1.00 are considered positive, although some
manufacturers include a gray zone (e.g., 0.80 to 1.20) wherein
results are taken to be equivocal until the sample is retested in
Interpretation of HCV markers duplicate. If only one of the duplicate tests is positive, the result is
Anti-HCV HCV NAT Interpretation reported as equivocal, and another specimen should be obtained
- - No acute infection from the individual for further testing. Diagnosis of acute HCV
- + Acute HCV infection infection relies on a combination of antibody and NAT results
+ - Chronic or resolved (see Table below) [14]. For diagnostic purposes, results of
+ + Early seroconversion or serological tests should be used in conjunction with patient
chronic history and other hepatitis markers for diagnosis of acute and
HCV NAT lower limit @ 50 IU/mL. chronic infection.
683
Hepatitis C Virus
9 Kashiwakuma T, Hasegawa A, Kajita T, Takata A, Mori

HCV Performance Goals H, Ohta Y, et al. Detection of hepatitis C virus specific
The following recommendations for the performance and utility core protein in serum of patients by a sensitive
of HCV tests were recently published in a report from the fluorescence enzyme immunoassay. J Immunol Methods
National Academy of Clinical Biochemistry [27]. Performance 1996;190:79-89.
characteristics for each assay (i.e., sensitivity, specificity, percent 10 Aoyagi K, Ohue C, Iida K, Kimura T, Tanaka E,
CV, etc.) are established by manufacturers of commercial assays Kiyosawa K, et al. Development of a simple and highly
in accordance with the licensing body guidelines of the country in sensitive enzyme immunoassay for hepatitis C virus core
which it is to be used. antigen. J Clin Microbiol 1999;37:1802-8.
11 Bouvier-Alias M, Patel K, Dahari H, Beaucourt S,
1. EIA screening tests for HCV antibody are adequate for Larderie P, Blatt L, et al. Clinical utility of total HCV
diagnosis of past or current infection in a patient core antigen quantification: a new indirect marker of
population with a high prevalence of disease; HCV replication. Hepatology 2002;36:211-218.
supplemental testing is not needed in such patients. If 12 Shah DO, Chang CD, Jiang LX, Cheng KY, Muerhoff
confirmation is required, HCV RNA should be used. AS, Gutierrez RA, et al. Combination HCV core antigen
2. Supplemental anti-HCV tests (e.g., RIBA) should be and antibody assay on a fully automated
used in populations with low prevalence of disease or to chemiluminescence analyzer. Transfusion
confirm infection in a patient who is RNA negative. 2003;43:1067-1074.
3. HCV RNA detection and quantitation methods should be 13 Scott JD, Gretch DR. Molecular diagnostics of hepatitis
calibrated to the World Health Organization C virus infection: a systematic review. JAMA
international standard. 2007;297:724-732.
4. Specimens for HCV RNA testing should either be 14 Chevaliez S, Pawlotsky JM. Use of virologic assays in
collected as EDTA or citrated plasma, or be promptly the diagnosis and management of hepatitis C virus
centrifuged to prevent falsely low results. infection. Clin Liver Dis 2005;9:371-82.
5. Assays for HCV RNA detection should have a dynamic 15 Ferreira-Gonzalez A, Shiffman ML. Use of diagnostic
range from < 1000 copies/mL to > 3.2 106 copies/mL. testing for managing hepatitis C virus infection. Semin
6. Genotyping tests should reliably differentiate all six Liver Dis 2004;24 Suppl 2:9-18.
major genotypes and also distinguish genotype 1a from 16 Scott JD, Gretch DR. Molecular diagnostics of hepatitis
1b. C virus infection: a systematic review. JAMA
2007;297:724-732.
References 17 Simmonds P, Alberti A, Alter HJ, Bonino F, Bradley
1 Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, DW, Brechot C, et al. A proposed system for the
Houghton M. Isolation of a cDNA clone derived from a nomenclature of hepatitis C viral genotypes. Hepatology
blood-borne non-A, non-B viral hepatitis genome. 1994;19:1321-1324.
Science 1989;244:359-362. 18 Simmonds P, Smith DB, McOmish F, Yap PL, Kolberg
2 Tobler LH, Busch MP. History of posttransfusion J, Urdea MS, et al. Identification of genotypes of
hepatitis. Clin Chem 1997;43:1487-1493. hepatitis C virus by sequence comparisons in the core,
3 Zmuda JF, Wagoneer B, Liotta L, Whiteley G. E1 and NS-5 regions. J Gen Virol 1994;75:1053-61.
Recognition of multiple classes of Hepatitis C antibodies 19 Fried MW, Hadziyannis SJ. Treatment of chronic
increases detection sensitivity in oral fluid. Clin Diag hepatitis C infection with peginterferons plus ribavirin.
Lab Immunol 2001;8:1267-1270. Semin Liver Dis 2004;24 Suppl 2:47-54.
4 Alter MJ, Kuhnert WL, Finelli L. Guidelines for 20 Bouchardeau F, Cantaloube JF, Chevaliez S, Portal C,
laboratory testing and result reporting of antibody to Razer A, Lefrere JJ, et al. Improvement of hepatitis C
hepatitis C virus. MMWR Recomm Rep 2003;52:1-13. virus (HCV) genotype determination with the new
5 Muerhoff AS, Jiang L, Shah DO, Gutierrez RA, Patel J, version of the INNO-LiPA HCV assay. J Clin Microbiol
Garolis C, et al. Detection of HCV core antigen in 2007;45:1140-1145.
human serum and plasma with an automated 21 Revised recommendations for testing whole blood, blood
chemiluminescent immunoassay. Transfusion components, source plasma and source leukocytes for
2002;42:349-356. antibody to hepatitis C virus encoded antigen (anti-
6 Leary TP, Gutierrez RA, Muerhoff AS, Birkenmeyer HCV). FDA memorandum to all registered blood
LG, Desai SM, Dawson GJ. A chemiluminescent, establishments. Rockville, MD: U.S. Food and Drug
magnetic particle-based immunoassay for the detection Administration; 1993 Aug 5.
of hepatitis C virus core antigen in human serum or 22 Lynn SA, Wright DJ, Kleinman SH, Hirschkorn D, Tu
plasma. J Med Virol 2006;78:1436-1440. Y, Heldebrant C, et al. Dynamics of viremia in early
7 Courouce AM, Le Marrec N, Bouchardeau F, Razer A, hepatitis C virus infection. Transfusion 2005;45:994-
Maniez M, Laperche S, et al. Efficacy of HCV core 1002.
antigen detection during the preseroconversion period. 23 Abbott/Murex HCV Ag/Ab Combination Assay Package
Transfusion 2000;40:1198-1202. Insert, Version 2006.
8 Tanaka T, Lau JY, Mizokami M, Orito E, Tanaka E, 24 Centers for Disease Control and Prevention.
Kiyosawa K, et al. Simple fluorescent enzyme Recommendations for prevention and control of hepatitis
immunoassay for detection and quantification of C virus (HCV) infection and HCV-related chronic
hepatitis C viremia. J Hepatol 1995;23:742-5. disease. MMWR Recomm Rep 1998;47(RR-19):1-33.
684
Hepatitis C Virus
25 Kleinman SH, Stramer SL, Brodsky JP, Caglioti S, circumstances [monograph on the Internet]. Association
Busch MP. Integration of nucleic acid amplification test Bulletin 05-03. Bethesda: AABB; 2005. Available from:
results into hepatitis C virus supplemental serologic http://www.aabb.org/content).
testing algorithms: implications for donor counseling 27 Dufour DR, Lott JA, Nolte FS, Gretch DR, Koff RS,
and revision of existing algorithms. Transfusion Seeff LB. Diagnosis and monitoring of hepatic injury. I.
XXXX;46:695-702. Performance characteristics of laboratory tests. Clin
26 Variance request to use NAT results and be exempted Chem 2000;46:2027-2049.
from HIV and HCV supplemental testing in specific
Table 3: Methods for Hepatitis C Virus
Method 1. Antibody test

Principle: Immunoassay for antibody detection
Usage: Diagnosis or screening blood donations
Comments: Most tests detect IgG only, some IgG and IgM
Method 2. Nucleic acid tests
Principle: Amplify and detect viral RNA in serum and plasma
Usage: Qualitative for diagnostic confirmation, quantitative for
therapeutic monitoring
Comments: Detects evidence of ongoing viral replication
Method 3. Confirmatory antibody test
Principle: Western blot format, typically strip-blot
Usage: To confirm antibody positivity if S/CO is low
Comments: Detection of at least 2 markers is needed to confirm infection
Method 4. Genotyping test
Principle: Uses RT-PCR to amplify portion of virus genome, determine genotype
by hybridization to genotype-specific probes or by sequencing.
Usage: Identification of genotype in individuals known to be infected with
HCV
Comments: Genotype identification needed to determine treatment regime and
predict outcome.
Method 5. Core or nucleocapsid antigen test
Principle: Immunoassay (sandwich format) to detect circulating core antigen.
Usage: Detection of window period blood donations, as possible
confirmatory test, or quantitative assay can be used for
therapeutic monitoring.
Comments: Commercial tests for detection of core antigen are not yet available in
the USA.
Method 6. Combination antibody/antigen test
Principle: Simultaneous detection of core antigen and HCV antibodies
Usage: Detection of window period blood donations, routine and
frequent testing of individuals in high risk groups (injection
drug users, prison inmates, health care worker, hemodialysis
patients).
Comments: Commercial combination tests are not yet available in the USA.
685
Hepatitis C Virus
Figure 1: Laboratory algorithm for antibody to hepatitis C virus (anti-HCV) testing and result reporting (3).
686
High-Density Lipoprotein (HDL) Cholesterol

John R. Burnett and Ken Robertson
Name: High-density lipoprotein (HDL) cholesterol
Clinical significance: Refer to Chapter 37, Coronary Artery Disease: Lipid Metabolism, in the 5th
Molecular mass: 170,000 to 360,000 D (HDL)
Chemical class: Lipoprotein
Chemical structure:
i sphingomyelin and lysophosphatides [3]. Cholesterol is

the next major lipid found in HDL. The ratio of
Evidence suggests that high-density lipoprotein (HDL) esterified to free cholesterol is about 3 to 1. HDL is a
cholesterol (C) is a primary and independent coronary heterogeneous group of compounds that have been
heart disease (CHD) risk factor [1,2]. The National divided into two density classes: HDL2 (d = 1.063 to
Institutes of Health (NIH) National Cholesterol 1.125 g/mL) and HDL3 (d = 1.125 to 1.210 g/mL).
Education Program (NCEP) has periodically released Using the techniques of density-gradient
guidelines concerning the clinical importance of HDL-C ultracentrifugation and gradient polyacrylamide gel
and CHD. These new guidelines place more attention on electrophoresis, HDL2 has been further subdivided into
HDL-C for (1) assessment of CHD risk, (2) inclusion
during initial total cholesterol screening, and (3) two subspecies (HDL2a and HDL2b), and HDL3 has
consideration of choice of drug therapy. The widespread been subdivided into at least three subspecies [4,5].
use of HDL-C values in clinical medicine warrants a These subspecies appear to be discrete molecular entities
critical analysis of the methods currently available for with characteristic sizes, densities, and distribution.
cholesterol and apolipoprotein-lipoprotein
quantification. Column chromatography [6,7,8], isoelectric focusing [9],
and high-performance liquid chromatography [11] have
By definition, HDL is the fraction of plasma lipoproteins been employed to further investigate the heterogeneity
with a hydrated density of 1.063 to 1.21 g/mL in a within the HDL distribution.
preparative ultracentrifuge. Electrophoretically, it has the
mobility of 1 globulins. The most dense of the There are two direct ways to measure HDL: (1) by
lipoproteins, it is composed of approximately 50% analytical ultracentrifugation and (2) by isolation of
protein, 25% phospholipid, 20% cholesterol, and 5% HDL and measurement of the particles gravimetrically.
triglyceride. The major lipid in HDL is phospholipid, of However, there are only a few clinical laboratories that
which lecithin is the major fraction, followed by can afford an analytical ultracentrifuge and have the
expertise and time to perform the laborious and tedious
steps. To isolate the lipoproteins by column
i
High-Density Lipoprotein (HDL) Cholesterol chromatography, electrophoresis, preparative
Previous and current authors of this method: ultracentrifugation, or a polyanion precipitation
First edition: Herbert K. Naito technique followed by gravimetric determination of the
Methods edition: Herbert K. Naito amount of HDL after the salts and water have been
Second edition: Herbert K. Naito removed is time consuming and requires expertise.
Third edition: Herbert K. Naito
Fourth edition: Herbert K. Naito To circumvent some of these problems, the isolated
Fifth edition: John R. Burnett, Ken Robertson HDL fraction is not measured in its entirety. Instead,
either the protein moiety or the lipid moiety is measured
687
as an indirect means of quantitating HDL. HDL is values in the fractions with density d > 1.063 g/mL by
composed of approximately 50% protein, of which correcting for manipulative loss and cholesterol derived
apolipoprotein (apo) A-I and apoA-II are the major from apoB-containing lipoproteins, including Lp(a).
apoproteins, with apoA-I being quantitatively the more Both Warnick et al. [26] and Srinivasan et al. [27]
important. The phospholipid component is the largest reported that they found substantial amounts of apoB-
lipid constituent by mass, but since cholesterol is simpler associated cholesterol in the fraction with d > 1.063
to measure than phospholipids, the HDL-C has prevailed g/mL in samples drawn from a limited number of
as an indirect means of determining HDL concentration. children and adult men and women with normal serum
concentrations of lipids. Srinivasan et al. [27] found
Numerous techniques are available for HDL-C contamination of the apoB-containing cholesterol in the
quantitation. They are based on two steps: (1) isolation HDL fraction especially in hyperlipoproteinemic (types
of the HDL and (2) quantitation of the cholesterol in the IIa, IIb, and IV) persons. Hutt et al. [28] found that in
isolated HDL. The various methods differ primarily on addition to Lp(a), apoE-containing HDL also contributed
how the HDL fraction is isolated, with the cholesterol significantly to the positive bias of the ultracentrifugal
analysis usually being done by one of the many method as compared to precipitation and column
acceptable cholesterol methods available. The method of chromatography techniques. In addition, recovery of
Albers et al. [10] can be recommended, though certain lipoproteins is often less than 90%. Thus both studies
compatible enzymatic procedures would be preferable. [26,27] emphasize that without appropriate corrections,
the lipoprotein concentrations obtained by preparative
The following HDL isolation procedures are used today ultracentrifugation may not serve as appropriate
in the clinical laboratory [9,11-21]. reference intervals.
Preparative Ultracentrifugation (Table 1, Method 1) Column Chromatography (Table 1, Method 2)

For the simultaneous separation of both very-low- Ion-exchange chromatography and gel-permeation
density lipoproteins (VLDL) and low-density chromatography have both been used in the isolation of
lipoproteins (LDL), the plasma or serum is adjusted to a the major lipoprotein groups and the subpopulations
density of 1.063 g/mL by overlaying of the sample with within each group. These procedures separate HDL
sodium or potassium bromide solution (415 mg of KBr subclasses based on differences in charge or molecular
per 5 mL) and centrifugation of the sample at 105,000 g size, respectively. For instance, using hydroxyapatite
for 24 hours at 16C. After the supernatant solution packing, the solution with a density of 1.063 to 1.210
containing the VLDL and LDL is removed, the g/mL can be subfractionated into 12 to 14 HDL
infranatant solution can be analyzed for the cholesterol subclasses. The significance of each of these
concentration. Technically, the infranatant solution subfractions is not well defined at the present time, but
should be adjusted to a density of 1.210 g/mL so that one as more information is gathered about the metabolic role
can obtain a solution that truly reflects only HDL. of each of these subclasses of lipoproteins, their
However, there are only minuscule amounts of other distinction will be more important. These methods are
lipoproteins besides HDL with a density greater than not widely used today because of the critical care and
1.063 g/mL. In addition, the extra step of adjusting the attention necessary in standardizing the columns, the
solution with a density of 1.063 g/mL to a density of need to concentrate the eluent for cholesterol analysis,
1.210 g/mL introduces greater error than direct analysis and the need for large sample size. Similarly, the
of the fraction with a density greater than 1.063 g/mL, preparative block electrophoresis methods are almost
which for the most part represents all the HDL. This never used for routine analysis. These methods also
method of isolating the HDL and measuring the require large samples and are very time consuming.
cholesterol content for HDL cholesterol estimation is a Their applications are more suited for the collection of
classical procedure and is often regarded as a reference large amounts of lipoproteins.
procedure. However, it should be emphasized that
caution is warranted when the preparative ultracentrifuge Preparative Block Electrophoresis (Table 1, Method 3)
is used as a reference method for the isolation of Both starch block and Geon-Pevikon block
lipoprotein fractions. This method has been operationally electrophoresis methods are employed in the isolation of
defined in terms of hydrated density and is generally major lipoprotein fractions and their subfractions. These
used to isolate the major lipoprotein fractions (VLDL, electrophoresis procedures separate the lipoprotein
LDL, and HDL). In more heterogeneous samples, classes on the basis of their net charge and size. The
however, there is known to be an overlap of lipoproteins smaller HDL molecules have the highest mobility
within an operationally defined density range. For toward the anode. These methods are mainly used as
example, occurrence of floating -lipoproteins (- preparative techniques.
VLDL), HDL, and sinking pre--lipoprotein (Lp[a]),
respectively, in density ranges commonly used to Agarose Gel Electrophoresis (Table 1, Method 4)
separate VLDL, LDL, and HDL is well documented [22- This method uses the standard lipoprotein
25]. To make the preparative ultracentrifugation method electrophoresis procedure on agarose medium, followed
a standard method for HDL-C, one must correct for the by the overlaying of the electrophoresed sample with
occurrence of apoB-containing lipoproteins in the enzymatic cholesterol reagent. The enzyme reagents
density range 1.063 to 1.21 g/mL. Warnick et al. [26] used include cholesterol esterase and oxidase with a
have successfully improved the accuracy of HDL-C peroxidase indicator. More recently, cholesterol
688
dehydrogenase rather than cholesterol oxidase has been The use of sodium phosphotungstate is another method
used with the dye nitroblue tetrazolium. This dye is of precipitating apoB-containing lipoproteins. This
insoluble and stable following reduction. A densitometer method also underestimates HDL-C, but not so much as
with an automated integrator is used to scan the agarose the polyethylene glycol method does. This method is
strip after color development and quantitate each particularly sensitive to temperature fluctuations and
lipoprotein fraction. reagent concentrations, and these can be a major source
of error. According to Assmanns group [32],
The agarose gel electrophoresis method is not often used phosphotungstic acidMgCl2 can also be used for
for routine analysis. The presence of -VLDL and Lp(a) complete precipitation of apoB-containing lipoproteins
in the sample can introduce bias in the measurement of in hypertriglyceridemic sera (up to 155 mg/dL; 4.0
pre- and lipoproteins, because the measurements are mmol/L) with only marginal coprecipitation of HDL.
based on electrophoretic distributions of and pre- However, one disadvantage of this method is the
lipoproteins. If the calculations are based on the percent unfavorable ratio of sample to precipitation reagents
distribution of the , pre-, and lipoproteins (derived (10:1). Another disadvantage is the instability of the
from the densitometric scans) multiplied by the total reagent mixture containing phosphotungstic acidMgCl2
cholesterol, it is conceivable that the presence of - because of the formation of isopolytungstic acid after
VLDL and Lp(a) would falsely elevate and pre- prolonged storage [33].
lipoprotein cholesterols and falsely depress -lipoprotein
cholesterol. Use of the modified reagent offers considerable
advantages over the use of the customary reagent: the
Serfontein et al. [29] reported that HDL-C is selectively modified precipitation reagent remains fully functional
underestimated by 10% when sodium phosphotungstate- after a 12-month storage of the reagent at +50C [33],
MgCl2 precipitation is used to visualize lipoprotein whereas prolonged storage of the customary reagent
bands separated by agarose electrophoresis. They results in deposits of isophosphotungstic acid. Moreover,
postulated that since the cholesterol is recovered from when the modified precipitation reagent was used in sera
the precipitation solutions used to visualize the containing as much as 397 mg/dL (10.26 mmol/L) of
lipoprotein bands, the HDL, because of its relatively triglyceride, none of the precipitate floated to the surface
smaller molecular mass and relative solubility in the [22,23], whereas with the customary reagent, one can
precipitation medium, diffuses out of the gel into the expect floating of the precipitate in sera with triglyceride
precipitation medium. concentrations greater than 220 mg/dL (5.7 mmol/L).
The proportional composition of sample to reagent of
Precipitation With Polyanion Solutions (Table 1, 1:2 allows for the addition of diluents to make the
Method 5) precipitation procedure easier and improve the precision
of the analysis. Niedmann et al. [24] have suggested that
Differential precipitation of lipoproteins with various cholesterol oxidase-resistant 3,5-cholestadiene may
polyanion solutions is common practice and is suitable form under certain conditions with heparin-MnCl2 or
for the clinical laboratory because of its simplicity, phosphotungstic acidMgCl2 precipitation methods for
elimination of expensive instrumentation, speed, and low
cost. These techniques are all based on the ability of HDL cholesterol quantification. With their highly
various agents to precipitate selectively the major sensitive chromatographic techniques, they confirm that
lipoprotein fractions, except HDL. HDL is left in the definitely no inert 3,5-cholestadiene forms when the
supernatant solution for cholesterol quantitation. The present precipitation reagent is used.
agents most frequently employed include heparin
manganese chloride (LRC method), dextran sulfate Polyethylene glycol of various molecular weights has
magnesium chloride, sodium phosphotungstate, and been used occasionally for HDL-C determinations. Of
polyethylene glycol. all the precipitation techniques, this method has the most
serious problems with accuracy and precision. Wiebe
Dextran sulfate of various molecular mass with calcium and Smith [25], however, concluded that their method
or magnesium salts has been used for HDL-C was both accurate and precise.
quantitation. Most dextran sulfate procedures tend to
underestimate HDL-C, though this method does not There are a few studies that were extensive examinations
appear to be sensitive to variations in incubation and of different HDL-isolation methods. Warnick et al. [34]
centrifugation temperatures. Higher-molecular-mass evaluated improved, or second-generation, HDL-C
(50,000 D) dextran sulfates tend to produce HDL-C methods. The methods were compared on specimens
values closer to those obtained using the heparin with a wide range of total cholesterol, triglyceride, and
magnesium chloride methods. Warnick et al. [30] have HDL-C values. They also tested specimens to which
described a dextran sulfatemagnesium chloride were added either sodium chloride, to increase the ionic
procedure that is compatible with enzymatic cholesterol strength, or glucose, to approximate a specimen from a
determination. The method, when compared to the LRC diabetic patientfactors that reportedly affect
method [11], is considered simple, rapid, accurate, and lipoprotein precipitation. Methods were compared for
precise [31]. completeness of lipoprotein sedimentation by assessing
the proportion of specimens with turbid supernatants.
Cholesterol was measured in all supernatants. They also
689
measured total protein in a subset of supernatants and attaches to the LDL and VLDL, a magnet is used to
precipitates to determine whether the effectiveness of pull the lipoprotein complex to the bottom of the tube.
sedimentation of hypertriglyceridemic specimens might This approach eliminates the need for centrifugation to
be attributable to coprecipitation of other plasma obtain a supernatant solution containing the HDL. This
proteins. method also has the distinct advantage of being less
sensitive to lipemic specimens, which is a major
The dextran sulfateMg2+ method was selected as the interferent that can cause incomplete precipitation of
comparison method for this report because previous LDL and VLDL (and chylomicrons) with other
studies [30,35] demonstrated good specificity for HDL precipitation methods. Triglyceride concentrations can
be several thousand mg/dL before problems arise. The
separation. In the earlier works, dextran sulfateMg2+
method appears to be precise and accurate, although a
supernatants of EDTA-treated plasma contained virtually
small negative bias has been observed with conventional
no apoB, indicating specific separation of LDL. On the
dextran sulfate and a positive bias with phosphotungstate
other hand, the dextran sulfateMg2+ precipitate precipitation.
fractions contained little apoA-I, an indication that
precipitation of HDL may not have been excessive. Homogenous Assays
Recently, several so-called homogenous methods have
All the modified precipitation methods tested gave been developed that are capable of full automation
similar results for HDL-C, an indication that (Table 1, Method 7). These methods do not require off-
modifications of the methods may have led to improved line pretreatment and separation and are considered to be
accuracy. Results by the heparin-Mn2+ (92 mmol/L) and third-generation assays [39,40]. These assays may be
phosphotungstate-Mg2+ methods agreed best with those immunologic, making use of antibodies directed against
apoB and/or apoC-III, or they may involve reaction with
by the dextran sulfateMg2+ procedure. The heparin-
various precipitating or complexing agents to form
Mn2+ (46 mmol/L) and the two polyethylene glycol soluble or insoluble complexes of chylomicrons and
methods gave slightly higher results. Observed LDL and VLDL cholesterol. The majority of
differences were largest for specimens with high HDL-C laboratories have adopted the homogenous assays, even
values. Addition of either NaCl or glucose within the though questions have been raised regarding the
expected physiological range did not significantly affect specificity of these methods, especially in specimens
lipoprotein precipitation. In terms of sedimentation with unusual lipoprotein compositions.
effectiveness, the methods were ranged in the following
order: polyethylene glycol (100 g/L, pH 10) > dextran Miscellaneous Method Considerations
sulfateMg2+ > heparin-Mn2+ (92 mmol/L) = Quantitation of HDL2- and HDL3-Cholesterol
polyethylene glycol (75 g/L) > phosphotungstate-Mg2+ Subfractions
> heparin-Mn2+ (46 mmol/L).
The HDL2 subfraction may be better correlated with risk
Although no ideal method for measuring serum HDL-C of coronary heart disease than total HDL cholesterol.
is available, it is important to recognize the limitations of The HDL2/HDL3 ratio was reported to be the best
all procedures. Particular caution is urged for those biochemical marker of coronary heart disease risk of 22
setting up this test for clinical evaluation of persons at that were tested [41]. Time-consuming zonal
risk for CHD. Precision is also important; remember that ultracentrifugation, column chromatography, and
the mean difference in HDL-C concentration in the gradient-PAGE have all been used to isolate HDL2 and
Framingham Study populations with and without HDL3. Selective precipitation of the HDL subfractions
myocardial infarction (MI) is only 40 mg/dL. As more
work is done using different polyanion solutions for the has been reported. Double precipitation methods with
selective precipitation of lipoproteins, a more sensitive, polyanions to separate HDL subfractions offer a more
accurate, and precise method will probably evolve. practical procedure. The methods are relatively rapid,
economical, and simple. The method of Martini et al.
Polyacrylamide Gel Electrophoresis (Table 1, Method [42,43] uses heparin-MnCl2 to precipitate the VLDL and
6) LDL, followed by dextran sulfate (15,000 D) to
Roche et al. [36] and Muiz [37] reported the use of precipitate the HDL2. The supernatant solution contains
polyacrylamide gel electrophoresis (PAGE) for the HDL3 fraction used for cholesterol measurement in
quantitating HDL cholesterol. According to Roche et al.
the following relationship:
[36], the PAGE method correlated well (r = 0.96) with
the ultracentrifugal method. This was a higher
HDL2-C= Total HDL-C HDL3-C
correlation than that found by Muiz [37], who reported
a correlation coefficient of 0.84.
The methods appear to be compatible with both
Magnetic Precipitation chemical and enzymatic cholesterol procedures [44].
Another HDL-isolation technique is based on the Lundberg et al. [45] reported the use of polyethylene
isolation of HDL by selective precipitation of LDL and glycol 6000 to precipitate the apoB-containing
VLDL using dextran sulfate (50,000 D) that has iron lipoproteins, after which HDL2 was precipitated with
attached to it [38]. Thus when the dextran sulfate
690
dextran sulfateMgCl2. Estimation of total HDL-C and polymer mixture and acted upon to produce colorless
HDL2 and HDL3 fractions with dextran sulfateMgCl2 products. The second detergent reagent then releases the
cholesterol from LDL cholesterol, which then forms a
has been published and appears to be accurate and
colored product which is measured
precise [31].
spectrophotometrically.
Hirano et al. [47] have published a single-step
Reference Diagnostics distributes a product using Denka
precipitation procedure for the measurement of HDL-
Seiken reagents. This assay stabilizes the LDL
cholesterol subfractions. This technique involves
molecules by use of a surfactant. This means that the
precipitating apoB-containing lipoproteins and the HDL2
LDL cholesterol is nonreactive. The first reagent reacts
using a heparin/manganous chloride/dextran sulfate
with all non-LDL cholesterol to produce a colorless
reagent. The HDL3 in the supernatant is then measured
product. A second surfactant in the second reagent then
using a homogenous HDL-C assay. The HDL3 value is
releases the cholesterol from LDL cholesterol, which
then obtained by subtracting the HDL2 value (corrected
then is measured spectrophotometrically after formation
for volume) from the total HDL-C value (measured
of a colored compound.
using the same Denka Seiken reagent).
The Wako reagents are distributed in the United States
Quantitation of LDL Cholesterol by Sigma Diagnostics. In this technique, an enzymatic
Most clinical laboratories utilize the convenient method complex is used to stabilize the LDL molecules and so
of Friedewald to estimate LDL cholesterol. Since LDL is render the cholesterol nonreactive. Non-LDL cholesterol
known to be the critical atherogenic lipoprotein, it is is then reacted upon, again producing a colorless
natural that methods be developed for a convenient, product. The addition of the second reagent, which
direct, and rapid assay for LDL cholesterol. The majority contains a detergent, again releases the LDL cholesterol,
of the precipitation techniques involve the use of which is measured spectrophotometrically after
complex carbohydrates such as heparin, dextran sulfate, formation of a colored compound.
amylopectin, and polyvinyl sulfate to precipitate LDL
directly. The specificity of polyanion precipitation There are a number of other suppliers and manufacturers
depends on experimental variables such as polyanion of similar reagent kits who use a variety of other novel
concentration, polymer length, pH, ionic strength, and techniques to achieve the same end [48,49].
the concentration of divalent cations.
The introduction of immunologic-based LDL cholesterol
The Joint Committee for Traceability in Laboratory
procedures has led to the routine use of these methods
Medicine (JCTLM) accepts the CDC Reference Method
for LDL measurement. This method is based on
for HDL [50]. This technique involves
polyclonal antisera specific for HDL and VLDL; after
ultracentrifugation to remove everything of density less
centrifugation the supernatant solution contains the LDL
than 1.006, followed by use of a heparin/manganous
fraction, which is analyzed for cholesterol. Empirically,
chloride precipitation to remove non-HDL. The HDL-C
the direct LDL cholesterol procedure should be more
in the supernatant is then measured
precise and accurate than the estimated LDL cholesterol
spectrophotometrically. This technique is also the
by the Friedewald formula [46,48,49] because of three
reference method for LDL cholesterol. (LDL-C = Total
analytical variables (total cholesterol, triglyceride, and
Cholesterol post ultracentrifugation HDL-C).
HDL cholesterol measurements) compared to one (LDL
cholesterol).
There is no validated definitive method for HDL-C. The
accepted reference method has not been fully validated,
In recent years, a number of suppliers have released
as is the case for the cholesterol reference method.
direct assays for the measurement of LDL cholesterol.
Instead, the accepted accuracy target is a procedure used
These avoid the necessity for the precipitation step, so
at the CDC to assign HDL-C target values for human-
make them suitable for application as automated
based serum pools.
procedures. The LDL-C plus second-generation assay
distributed by Roche (using Kyowa reagents) utilizes the
The NCEP guidelines place more emphasis on the
process of selective micellary solubilization of LDL
measurement of HDL-C for the assessment of CHD risk.
cholesterol by a nonionic detergent. When this detergent
The medical decision values of less than 35 mg/dL and
is added to the standard cholesterol esterase/cholesterol
60 mg/dL or greater demand greater accuracy of
oxidase reagents, the relative reactivities of the
measurement to minimize the false positives or negatives
lipoproteins are altered so that HDL < chylomicrons <
[51].
VLDL < LDL. In the presence of Mg2+, a sugar
compound in the reagent further reduces the cholesterol
The direct gravimetric and ultracentrifugation techniques
enzymatic reaction of VLDL and chylomicrons, enabling
for HDL quantitation are rarely used for routine work.
the determination of LDL cholesterol.
These methods are too laborious and require highly
specialized techniques. Nevertheless, an
The Daiichi reagent, distributed in the United States by
ultracentrifugation-based method is considered the
Genzyme Diagnostics, uses a two-reagent method.
reference method.
During the first step, cholesterol from non-LDL
cholesterol components is released by a detergent
691
Although the heparinmanganese chloride procedure is the anticoagulant of choice, and the final concentration
not commonly used, it is considered one of the methods should be 1 mg of EDTA per mL of blood. The sample
of choice. The CDCs reference method is used to should be removed from the blood clot within 2 hours
standardize the participating laboratories in the Lipid and may be stored at 4C for up to 2 days. If specimens
Standardization Program. are to be kept for longer than 48 hours, they should be
frozen.
However, the heparinmanganese chloride method is not
without faults. Because of the presence of manganese HDL is relatively labile, and freeze/thaw cycles have
ions in the sample, this method is not compatible with been shown to affect some of the precipitation methods.
most cholesterol enzymatic systems, giving a consistent However, if specimens are frozen, they should be kept at
false-positive bias. However, if a reliable and accurate temperatures below 50C, at which temperature they
Liebermann-Burchard (L-B) method is used, this are stable for up to 2 years. Once thawed, specimens
becomes an excellent method. Hypertriglyceridemic should be gently mixed prior to analysis. It is
samples greater than 4000 mg/L (4.5 mmol/L) tend to recommended that tubes specifically designed for low-
cause incomplete precipitation of the apoB-containing temperature storage should be used.
lipoproteins, resulting in turbid solutions that contain
substantial quantities of VLDL and LDL. This problem Interferences
can be minimized by: The homogenous assays for HDL-C differ somewhat in
(1) Dilution of the samples with 0.9% saline and the effect of various interfering substances. Triglycerides
reprecipitation with heparinmanganese chloride (92 less than 900 mg/dL generally do not interfere with any
mM of Mn2+) of the homogenous methods. Hemoglobin < 2 g/L and
(2) Centrifugation at a higher gravity force, such as bilirubin < 100 mg/L also do not interfere with
12,000 g for 10 min homogenous assays, and some instrument manufacturers
(3) Removal of the chylomicron or VLDL fractions first, claim even higher limits for TG, hemoglobin, and
using preparative ultracentrifugation at a density of bilirubin. Reports of interference in these assays from
1.006 g/mL specimens with atypical lipoprotein patterns indicate the
(4) Ultrafiltration of the turbid supernatant solution with need for careful assessment of results from these types of
a 0.22 m ultrafilter specimens. There have also been reports noting
(5) Use of undiluted samples but with twice the volume unreliable results in samples from subjects with
of heparinmanganese chloride solution. In practice, Waldenstroms macroglobulinemia and in a case of
options 1 and 5 are preferred. polyclonal gammopathy.
Another problem associated with the heparinmanganese HDL Reference Interval

chloride method is the interval between sample Just as population-based reference intervals for total
collection and precipitation. Because the HDL-C values cholesterol were abandoned with the implementation of
decrease with an increase in time interval, it is the NCEP guidelines, HDL cholesterol population-based
recommended that the analysis be performed on the reference intervals are being abandoned for
sample soon after blood collection. interpretation of patients risk for CHD. The medical
decision points selected by the NCEP Adult Treatment
A detailed description of the heparinmanganese Panel (ATP) were based on specific end-points in
chloride procedure is included (click here). Since the clinical trials (heart attacks, deaths due to CHD,
mean difference in HDL-C concentration between abnormal EKG, etc.) and on simplicity. The NCEP
persons at normal risk for CHD and persons at high risk recommended only two medical decision points. In
is small (about 4 to 5 mg/dL), it is suggested that the 2001, the NCEP increased the high-risk medical decision
minimum precision for this method should be less than 3 level to < 40 mg/dL for HDL-C [51,52].
mg/dL for within-day and day-to-day variance.
Interpretation
The homogenous methods for HDL-C quantitation are During the last 3 decades, an increasing amount of basic,
steadily replacing the conventional precipitation clinical, and epidemiological research has focused on the
methods. These methods allow full automation of role of plasma lipoprotein level as a primary risk factor
previously tedious manual treatment of samples. for CHD. Since 1965, when Fredrickson and Lees [54]
Specimen developed a system for phenotyping
In brief, the patient must be fasting for at least 12 hours hyperlipoproteinemia, the major emphasis has been on
before the blood is drawn. It should be emphasized that chylomicrons, VLDL, and LDL. Although reports
although nonfasting conditions do not appear to linking cholesterol in the lipoprotein or HDL to lower
influence the blood HDL-C levels in most people, incidence of MI were available in the 1950s [53,55],
postprandial lipemia has the potential of interfering with interest in and recognition of this information waned as
many of the analytical methods. To minimize this the emphasis was placed on the seemingly atherogenic
analytical problem, it is always good laboratory practice lipoprotein lipoprotein. There is no longer any doubt
to request fasting specimens. Serum or plasma can be that LDL plays an important role in the development and
used as the sample, but since plasma is usually preferred progression of coronary atherosclerosis. Although
for lipid analysis, HDL analysis will most frequently be several investigators have clearly shown the deposition
performed on the same plasma specimen. EDTA is again of LDL and apoB in the intima and media of the vascular
692
wall [56,57], the precise mechanism by which LDL is Performance Goals for HDL-Cholesterol
involved with atherogenesis is not completely Measurements
understood. The NCEP Laboratory Standardization Panel
recommends that total error be within 13% of the true
Lipid and lipoprotein studies in the past have value. The total error comprises imprecision (random
emphasized the positive relationship among plasma total error) and inaccuracy or bias (systematic error). The
cholesterol, LDL, and the increased risk for CHD. imprecision of any assay should be < 4%. The limits for
However, other lipoproteinsthat is, chylomicrons, acceptable performance established by CLIA require that
VLDL (or VLDL-like particles), intermediate-density laboratories be within 30% of the peer-group mean.
lipoproteins (IDL), -VLDL, Lp(a), oxidized LDL,
homocysteine, and apoEhave now been implicated in Survey data from the 2007 College of American
this multifactorial and complicated disease. It is now Pathologists (CAP) Participant Summary Report [65]
well established that there is an inverse relationship shows a wide range of imprecision values (% coefficient
between hypertriglyceridemia and HDL (and HDL-C) of variation [CV]) for HDL cholesterol, ranging from
concentration. Thus although epidemiologists have not 4% to 7% in most cases. There are a few methods,
shown a clear relationship between elevated triglycerides however, which still show CVs of 8% to 12% or higher.
and increased risk for CHD, a distinct relationship does Overall, the results indicate good, acceptable
exist between HDL-C levels and CHD; that is, the lower performance. Even techniques with very poor precision
the HDL concentration, the greater the risk for CHD. appear to satisfy the Clinical Laboratory Improvement
The relationship between elevated triglyceride-rich Amendments (CLIA) criteria
particles (chylomicrons, VLDL, and IDL) and increased
risk for CHD is to a large extent probably attributable to References
the depressed HDL (HDL-C) levels. Partitioning the 1 Gordon T, Castelli WP, Hjortland MC, Kannel
total cholesterol measurement into the atherogenic factor WB, Dawber TR. High-density lipoprotein as a
(LDL cholesterol) and the antiatherogenic factor (HDL protective factor against coronary heart disease:
cholesterol) provides more meaningful information that the Framingham Study. Am J Med
shows a better statistical relationship with CHD than the 1977;62:707-14.
earlier method. There are numerous studies that have 2 Heiss G, Johnson N, Reiland S, Davis CE,
clearly shown the inverse and independent relationship Tyroler HA. The epidemiology of plasma high-
between HDL or HDL cholesterol and risk for CHD; a density lipoprotein cholesterol levels.
comprehensive review on this subject can be obtained Circulation 1980;62(suppl.):IV116-36.
elsewhere [58-60]. 3 Scanu AM, Lim CT, Edelstein C. On the
subunit structure of the protein of human serum
HDL-C is only a reflection of HDL concentration. There high density lipoprotein. II. A study of
are other markers that might better reflect the HDL Sephadex fraction IV. J Biol Chem
concentration, such as HDL phospholipid [61] or 1972;247:5850-5.
apoproteins [41]. Many investigators showed that MI 4 Anderson DW, Nichols AV, Pan SS, Lindgren
survivors have lower levels of apoA-I and apoA-II and FT. High-density lipoprotein distribution:
HDL-C than healthy adult males do. They postulated resolution and determination of three major
that since hypertriglyceridemia is common in CHD components in a normal population sample.
subjects, and since a reciprocal relationship between Atherosclerosis 1978;29:161-9.
HDL-C and triglyceride levels exists, the lower HDL-C 5 Patsch JR, Gotto AM. Separation and analysis
concentrations could in part reflect the higher of HDL subclasses by zonal ultracentrifugation.
triglyceride levels in MI subjects, as suggested by In: Lippel K, ed. Reports of the High-Density
Carlson and Ericsson in 1975 [62]. In addition, they also Lipoprotein Methodology Workshop, DHEW,
suggested that the small but statistically significant NIH Publ., No. 79-1661. Washington, DC: U.S.
difference in total apoprotein composition could reflect a Government Printing Office; 1979:310.
redistribution of HDL subclasses (HDL2 versus HDL3), 6 Laggner P. Physicochemical characterization of
a compositional alteration of the HDL molecules, or both high density lipoproteins. In: Day CE, ed. High-
[63]. Density Lipoproteins. New York: Marcel
Dekker; 1981:43-72.
Like other major classes of lipoproteins, HDLs are a 7 Lux SE, John KM. Further characterization of
heterogeneous group of macromolecules with different the polymorphic forms of a human high density
chemical and physical properties and, more than likely, apolipoprotein, apo LP-Gln-I (apoA-1).
with different metabolic functions. Thus three subclasses Biochim Biophys Acta 1971;278:266-70.
(HDL1, HDL2, and HDL3) of HDL have been 8 Edelstein C, Lim CT, Scanu AM. On the
subunit structure of the protein of human serum
identified. It appears that the HDL2 but not the HDL3 high density lipoprotein: a study of its major
has clinical significance from the standpoint of polypeptide component (Sephadex fraction III).
association with CHD risk. Thus partitioning the major J Biol Chem 1972;247:5842-9.
lipoprotein classes into IDL and LDL or HDL2 and 9 Allen JK, Hensley WJ, Nichols AV, Whitfield
HDL3 appears to increase their sensitivity, specificity, JB. An enzymic and centrifugal method for
and hence their predictive value for CHD risk [63,64].
693
estimating high density lipoprotein cholesterol. 23 Funke H, Assmann G. Influence of triglyceride-

Clin Chem 1979;25:325-7. rich lipoproteins on nephelometry of
10 Albers JJ, Warnick GR, Wiebe D, King P, apolipoprotein A-I. Clin Chem 1982;28:1153-8.
Steiner P, Smith L et al. Multi-laboratory 24 Niedmann PD, Luthe H, Wieland H, Schaper G,
comparison of three heparin-Mn2+ Seidel D. Accuracy of HDL cholesterol
precipitation procedures for estimating determination. Klin Wochenschr 1983;61:133-
cholesterol in high-density lipoprotein. Clin 8.
Chem 1978;24:853-6. 25 Wiebe PA, Smith SJ. Six methods for isolating
11 Manual of laboratory operations, Lipid high-density lipoprotein compared with use of
Research Clinics Program, Lipid and the reference method for quantifying
Lipoprotein Analysis, DHEW Pub. NIH Publ., cholesterol in serum. Clin Chem 1985;31:746-
No. 751-628. Washington, DC: U.S. 50.
Government Printing Office; 1974 May. 26 Warnick GR, Cheung MC, Albers JJ.
12 Bachorik PS, Wood PD, Albers JJ, Steiner P, Comparison of current methods for high-
Dempsey M, Kuba K et al. Plasma high-density density lipoprotein cholesterol quantitation.
lipoprotein cholesterol concentrations Clin Chem 1979;25:596-604.
determined after removal of other lipoproteins 27 Srinivasan SR, Webber LS, Whitaker CF,
by heparinmanganese precipitation or by Berenson GS. Quantification of lipoprotein
ultracentrifugation. Clin Chem 1976;22:1828- cholesterol in serum from children with
34. different lipoprotein profiles: heparin-calcium
13 Burstein M, Scholnick MR. Lipoprotein precipitation and ultracentrifugation compared.
polyanionmetal interactions. Adv Lipid Res Clin Chem 1983;29:481-5.
1973;11:67-108. 28 Hutt V, Klr HU, Wechsler JG, Ditschuneit H.
14 Burstein M, Scholnick HR, Morgin R. Rapid Three different methods for estimating high
method for the isolation of lipoproteins from density lipoprotein cholesterol. Clin Chim Acta
human serum by precipitation with polyanions. 1983;128:173-6.
J Lipid Res 1970;11:583-595. 29 Serfontein WJ, Ubbink JB, deVilliers LS.
15 Warnick GR, Albers JJ. A comprehensive Unreliability of enzymatic cholesterol analyses
evaluation of the heparinmanganese for the determination of electrophoretically
precipitation procedure for estimating high separated lipoproteins: selective loss of HDL
density lipoprotein cholesterol. J Lipid Res cholesterol. Clin Chim Acta 1983;130:123-8.
1978;19:65-76. 30 Warnick GR, Benderson J, Albers JJ. Dextran
16 Finley PR, Schifman RB, Williams RJ, Lichti sulfate-Mg2+ precipitation procedure for
DA. Cholesterol in high-density lipoprotein: quantitation of high-density-lipoprotein
use of Mg2+/dextran sulfate in its enzymatic cholesterol. Clin Chem 1982;28:1379-88.
measurement. Clin Chem 1978;24:931-3. 31 Talameh Y, Wei R, Naito H. Measurement of
17 Kostner GM. Enzymatic determination of total HDL, HDL2, and HDL3 by dextran
cholesterol in high-density lipoprotein fractions sulfate-MgCl2 precipitation technique in human
prepared by polyanion precipitation. Clin Chem serum. Clin Chim Acta 1986;158:33-41.
1976;22:695. 32 Assmann G, Schriewer H, Schmitz G, Hgele
18 Hatch FT, Lindgren FT, Adamson GL, Jensen EO. Quantification of high-density-lipoprotein
LC, Wong AW, Levy RI. Quantitative agarose cholesterol by precipitation with
gel electrophoresis of plasma lipoproteins: a phosphotungstic acid/MgCl2. Clin Chem 1983;
sample technique and two methods for 29:2026-30.
standardization. J Lab Clin Med 1973;81:946- 33 Draeger B, Wahlefeld AW, Ziegenhorn JA. A
60. practical method for HDL cholesterol
19 Vilkari J. Precipitation of plasma lipoproteins quantitation. Lab Med 1982;6:198-202.
by PEG-6000 and its evaluation with 34 Warnick GR, Nguyen T, Albers AA.
electrophoresis and ultracentrifugation. Scand J Comparison of improved precipitation methods
Clin Lab Invest 1976;36:265-8. for quantification of high-density lipoprotein
20 Lopes-Virella MF, Stone P, Ellis S, Colwell JA. cholesterol. Clin Chem 1985;31:217-22.
Cholesterol determination in high-density 35 Warnick GR, Benderson J, Albers JJ. Dextran
lipoproteins separated by three different sulfate-Mg2+ precipitation procedure for
methods. Clin Chem 1977;23:882-4. quantitation of high-density-lipoprotein
21 Abell LL, Levy BB, Brodie BB, Kendall FE. A cholesterol. Selected Methods Clin Chem
simplified method for the estimation of total 1983;10:91-9.
cholesterol in serum and demonstration of its 36 Roche D, Atger V, Le Quang NT, Girard A,
specificity. J Biol Chem. 1952;195:357-66. Ekindjian OG. Polyacrylamide gel
22 Draeger B, Ziegenhorn J, Wahlefeld AW. electrophoresis in quantification of high-density
Improved method for the precipitation of serum lipoprotein cholesterol. Clin Chem
high-density lipoprotein cholesterol (HDL 1985;31:1893-5.
cholesterol). Clin Chem 1982;28:1574-8.
694
37 Muniz N. Measurement of plasma lipoproteins 50 Joint Committee for Traceability in Laboratory

by electrophoresis on polyacrylamide gel. Clin Medicine-DB website. Available from:
Chem 1977;23:1826-33. http://www.bipm.org/jctlm/
38 Harris N, Galpchian V, Rifai N. Three routine 51 Summary of the second report of the National
methods for measuring high-density lipoprotein Cholesterol Education Program (NCEP) Expert
cholesterol compared with the reference Panel on detection, evaluation, and treatment of
method. Clin Chem 1996;42:738-43. high blood cholesterol in adults (Adult
39 Warnick GR, Nauck M, Rifai N. Evolution of Treatment Panel II). JAMA 1993;269:3015-23.
methods for measurement of HDL-cholesterol: 52 Executive summary of the third report of the
from ultracentrifugation to homogenous assays. National Cholesterol Education Program
Clin Chem 2001;47:1579-96. (NCEP) expert panel on detection, evaluation,
40 Langlois MR, Blaton VH. Historical milestones and treatment of high blood cholesterol in
in the measurement of HDL-cholesterol: impact adults (Adult Treatment Panel III). The Expert
on clinical and laboratory practice. Clin Chim Panel. JAMA 2001;285:2486-97.
Acta 2006;369:168-78. 53 Nikkil E. Studies on the lipid-protein
41 Naito HK. The association of serum lipids, relationships in normal and pathological sera
lipoproteins, and apolipoproteins with coronary and the effect of heparin on serum lipoproteins.
artery disease assessed by coronary Scand J Clin Lab Invest 1953;5(suppl. 8):1-101.
arteriographs. Ann NY Acad Sci 1985;454:230- 54 Fredrickson DS, Lees RS. A system for
8. phenotyping hyperlipoproteinemia. Circulation
42 Martini S, Baggio G, Baroni L, Enzi GB, Fellin 1965;31:321-7.
R, Baiocchi MR et al. Evaluation of HDL2 and 55 Barr DP, Russ EM, Eder HA. Plasma-lipid
HDL3 cholesterol by a precipitation procedure relationships in human plasma. II. In
in a normal population and in different atherosclerosis and related conditions. Am J
hyperlipidemic phenotypes. Clin Chim Acta Med 1951;11:480-93.
1984;137:291-8. 56 Hoff HF, Titus JL, Bajardo RJ, Jackson RL,
43 Farish E, Fletcher CD. A comparison of two Gotto AM, DeBakey ME et al. Lipoproteins in
micro-methods for the determination of HDL2 atherosclerotic lesions localization by
and HDL3 cholesterol. Clin Chim Acta immunofluorescence of apo-low density
1983;129:221-8. lipoproteins in human atherosclerotic arteries
44 Baldo Enzi G, Baroni L, Piccoli A, Baiocchi from normal and hyperlipoproteinemics. Arch
MR, Fellin R. Cholesterol determination in Pathol 1975;99:253-8.
HDL, HDL2 and HDL3 fractions after 57 Hoff HF, Heideman CL, Gotto AM Jr, Gaubatz
polyanion precipitation: a comparison between JW. Apolipoprotein B retention in the grossly
chemical extractive and totally enzymatic normal and atherosclerotic human aorta. Circ
procedures. Clin Chim Acta 1985;146:81-6. Res 1977;4:684-90.
45 Lundberg B, Hgstrm S, Pietilinen P, 58 Day CE, ed. High-Density Lipoproteins. New
Suominen M, Viikari J. Separation of plasma York: Marcel Dekker; 1981.
high-density lipoprotein subclasses by a 59 Naito HK. HDL cholesterol: metabolism,
combined precipitation method using clinical significance, and laboratory
polyethylene glycol 6000 and dextran sulphate. considerations. In: Homburger HA, Batsakis
Scand J Clin Lab Invest 1984;44:305-9. JG, eds. Clinical Laboratory Annual. Norwalk,
46 Friedewald WT, Levy RJ, Fredrickson DS. CT: Appleton-Century-Crofts; 1984:271-336.
Estimation of the concentration of low-density 60 Miller NE. High density lipoprotein,
lipoprotein cholesterol in plasma without use of atherosclerosis, and ischemic heart disease. In:
the preparative ultracentrifuge. Clin Chem Miller NE, ed. Atherosclerosis: Mechanisms
1972;18:499-509. and Approaches to Therapy. New York: Raven
47 Hirano T, Nohtomi K, Koba S, Muroi A, Ito Y. Press; 1984:153-68.
A simple and precise method for measuring 61 Naito HK, Greenstreet RL, David JA, Sheldon
HDL-cholesterol subfractions by a single WL, Shirey EK, Lewis RC et al. HDL
precipitation followed by homogenous HDL- cholesterol concentration and severity of
cholesterol assay. J Lipid Res 2008 Jan 27 (E- coronary atherosclerosis determined by
pub ahead of print). cineangiography. Artery 1980;8:101-12.
48 Miller WG, Waymack PP, Anderson FP, 62 Carlson LA, Ericsson M. Quantitative and
Ethridge SF, Jayne EC. Performance of four qualitative serum lipoprotein analysis. Part 2.
homogeneous direct methods for LDL- Studies in male survivors of myocardial
cholesterol. Clin Chem 2002;49(3):489-98. infarction. Atherosclerosis 1975;21:435-50.
49 Sugiuchi H, Uji Y, Okabe H, Irie T, Uekama K, 63 Kostner GM. Studies of the composition and
Kayahara N et al. Direct measurement of high- structure of human serum lipoproteins: isolation
density lipoprotein cholesterol in serum with and partial characterization of apolipoprotein
polyethylene glycol-modified enzymes and AIII. Biochim Biophys Acta 1974;336:383-95.
sulfated a-cyclodextrin. Clin Chem 64 Miller NE, Hammett F, Saltissi S, Rao S, van
1995;41:717-23. Zeller H, Coltart J et al. Relation of
695
angiographically defined coronary artery 66 Cooper GR. High-density lipoprotein reference

disease to plasma lipoprotein subfractions and materials. In: Lippel K, ed. Reports of the High
apolipoproteins. Br Med J 1981;282:1741-44. Density Lipoprotein Methodology Workshop,
65 C-A Chemistry/Therapeutic Drug Monitoring DHEW, NIH Publ., No. 79-1661. Washington,
Participant Summary, Surveys 2007. DC: U.S. Government Printing Office;
Washington, DC: College of American 1979:178-88.
Pathologists; 2007.
Table
Table 1: Methods for Isolation of High-Density Lipoprotein (HDL)

Method 1: Ultracentrifugation
Principle of analysis: Plasma or serum adjusted to density of 1.063 g/mL with potassium bromide and centrifuged at high
speeds for 24 h; all lipoproteins separated by density, with HDL fraction in 1.063 to 1.21 g/mL range
Comments: Specialized research laboratories; reference method; laborious and time consuming
Method 2: Column chromatography
Principle of analysis: HDL isolated and separated into subclasses on basis of charge (ion-exchange) or molecular size (gel
permeation)
Comments: Specialized, research laboratories; rarely used; difficult to properly control chromatographic conditions
Method 3: Starch block electrophoresis
Principle of analysis: HDL separated from other lipoproteins on basis of charge and size
Comments: Rarely used; used for isolation of large amounts of HDL, not for quantitative purposes
Method 4: Agarose gel electrophoresis
Principle of analysis: (See Method 3)
Comments: Frequently used; precision not adequate for clinical use
Method 5: Precipitation
a. Heparinmanganese chloride
b. Dextran sulfate
c. Phosphotungstate
d. Polyethylene glycol
Principle of analysis: Polyanions (heparin, dextran sulfate), phosphotungstate, polyethylene glycol, in presence of divalent
cations, used to precipitate larger, less dense lipoproteins; HDL quantitated in supernatant as HDL-C
Comments:
a. Frequently used reagent; current method of choice; not compatible with all cholesterol procedures
b. Used occasionally; use of lower molecular mass dextran can produce biased results; higher molecular mass
dextrans produce excellent results; compatible with enzymatic procedures
c. Most commonly used; underestimates HDL; sensitive to temperature fluctuations
d. Used infrequently; poorest accuracy and precision; not recommended
Method 6: Polyacrylamide gel electrophoresis
Principle of analysis: (See Method 3)
Comments: Used infrequently; probably underestimates HDL levels
Method 7: Homogenous methods
Synthetic polymer/detergent
Polyethylene glycol
2+
Principle of analysis: Cyclodextrin and Mg or synthetic polymer with a polyanion selectively blocks, but does not
precipitate, non-HDL lipoproteins; HDL free to react with reagent enzymes; immunological
Principle of analysis: Antibody to human apoB in first reagent reacts with apoB-containing lipoproteins, chylomicrons, LDL
and VLDL; HDL free to react with enzymes added with second reagent
Comments: Most common methods in use today; better precision than manual precipitation methods
696
Procedure: Isolations of High-Density Lipoproteins values will be slightly lower than results obtained using
by HeparinManganese Chloride the ultracentrifugal procedure (the reference procedure).
2. Purity check. If the sample shows an unusual
Principle value (HDL cholesterol concentrations less than 20 or
The larger, less dense apoB-containing lipoproteins greater than 80 mg/dL) or the -lipoprotein bands on the
(those with low, very low, and intermediate density) are electrophoresis medium do not agree with the serum or
precipitated overnight by heparinmanganese chloride. plasma HDL cholesterol value, perform a purity check or
After centrifugation to separate the precipitated repeat the analysis or do both.
lipoproteins, HDL cholesterol in the supernatant solution
is quantitated. A purity check on the supernatant solution should be
done if there is the possibility of incomplete
Reagents
precipitation of apoB-containing lipoproteins. To do this,
1. Manganese chloride solution, 1.0 mol/L.
electrophorese the supernatant solution, and stain with a
Dissolve 19.791 g of MnCl24H2O in distilled water,
lipid dye to show whether only one band exists. If there
and bring volume to 100 mL. This is stable for 3 months is more than one lipid-staining band and the second band
at 4C. is in the -globulin area or 2-globulin area, an apoB-
2. Heparin solution, 4000 U/mL. Dilute 1 mL of
heparin (A.H. Robins; 10,000 USP units/mL) with 1.5 containing lipoprotein is probably present. One can
mL of 0.9% saline solution. This is stable for 1 week at confirm this by doing immunological studies using apoB
4C. antisera.
3. Double-concentration precipitation. If the
Assay serum is turbid or hypertriglyceridemic (greater than 400
Equipment: Refrigerated centrifuge capable of mg/dL), a double-strength solution should be used to
maintaining temperature of 4C at 2000 g. precipitate the beta-containing lipoproteins. To do this,
1. Pipet 1 mL of serum into a 13 100 mm use 1 mL of serum, 100 L of heparin, and 100 L of
disposable culture tube. manganese chloride solution. Multiply the result by 1.2
to account for the dilution. A double-strength solution
2. Cap tubes with polyethylene stoppers, and place
should also be used for a serum sample if, after
in refrigerator for 30 min at 4C.
conventional manganese chlorideheparin precipitation,
3. Add 50 mL of heparin solution; mix on a vortex a floating lipid precipitate is present and the purity check
mixer for 1 min. of the supernatant shows contamination.
4. Add 50 mL of manganese chloride solution; 4. Pooled human sera for quality control.
mix on a vortex mixer. Reference materials are needed for calibration,
5. Cap tubes with polyethylene stoppers, and place standardization, and quality control of HDL-cholesterol
in refrigerator for 30 min at 4C. analyses. The calibration and control materials must be
6. Centrifuge the heparin-lipoprotein precipitate at suitable for determining cholesterol in the HDL fraction
2000 g for 1 hr at 4C. when specimens contain relatively low concentrations of
7. Using a transfer pipet, transfer supernatant cholesterol and when nonspecific and interfering
solution into a clean test tube. Store in a substances may be present. Calibration materials
refrigerator overnight to ensure complete applicable to low cholesterol level analyses can be
precipitation of all apoB-containing standards containing about 25 and 60 mg/dL of
lipoproteins. cholesterol. Serum control materials for these cholesterol
analyses can be sera diluted to contain about 25 and 60
8. The next morning, sediment any remaining
mg/dL of cholesterol. These levels cover the range of
precipitate at 2000 g for 30 min at 4C.
values of most HDL specimens. An alternative
9. Using a transfer pipet, transfer the clear procedure is to use a normal-range cholesterol serum
supernatant solution into a clean test tube for pool for calibration and to use both primary standard
measurement of cholesterol, using a standard solutions and serum pools of low concentration levels
method compatible with heparinmanganese for monitoring. Since a major cause of discrepancies
chloride. The L-B method is recommended. between HDL-C cholesterol results among laboratories
Notes can be attributed to lack of appropriate standardization
1. Biological sample. If one uses plasma samples, of the total cholesterol method, an effort should be made
first to standardize the cholesterol method. Reference
the 92 mmol/L Mn2+ salt concentration should be materials, according to Cooper [66], that appear suitable
selected. However, if one uses serum, the 46 mmol/L for calibrating and monitoring the determination of
Mn2+ concentration is more ideally suited for HDL cholesterol in HDL isolated fractions are: (1) primary
cholesterol determinations. If EDTA plasma samples are cholesterol standards, (2) serum calibrator pools, (3)
used with the heparin 46 mmol/L Mn2+ method, slightly serum quality control pools, and (4) fractions of HDL
higher HDL cholesterol values will be obtained because isolated by ultracentrifugation or precipitation methods
of incomplete precipitation of the apoB-containing and labeled accurately with total cholesterol content.
lipoproteins. If one uses the 92 mmol/L Mn2+ salt
preparation on serum samples, the HDL cholesterol
697
The calibration and control materials produced for use in Studies by Cooper [66] suggest that for long-term
the first steppreparation of the HDL sample for storage of HDL cholesterol pools, a temperature less
cholesterol analysisare often unsuitable because of an than 60C, rather than 20C or 5C, is necessary.
unpredictable instability of HDL particles, varying
composition, and flux of lipids and apoproteins among 5. Calculation of VLDL and LDL cholesterol.
lipoproteins [2] or of apoB from HDL particles [51]. Although one can isolate VLDL (density less than
Compromises, therefore, have to be accepted. Under 1.006 g/mL) and determine the cholesterol content,
these circumstances, it appears practical to seek HDL VLDL cholesterol can be estimated with a fair
reference materials that are homogeneous, are stable up amount of accuracy by the following
to 2 years, and possess a suitable similar-to-serum calculation:[46]
matrix. The HDL reference materials that might meet Triglyceride (mg/dL) = VLDL-C (mg/dL)
these criteria for control of the HDL isolation procedure 5
are (1) serum with low triglyceride or VLDL levels, (2)
bottom fraction 1.006 g/mL density prepared by This holds true only under the condition that serum
ultracentrifugation, and (3) bottom fraction 1.063 g/mL triglyceride concentration is less than 400 mg/dL.
density prepared by ultracentrifugation. Furthermore:
LDL-C = Total cholesterol (VLDL-C + HDL-C)

698
Holotranscobalamin
Holotranscobalamin
Marion Black i
Name: Holotranscobalamin, holoTC, active B12

Clinical significance: Refer to Chapter 43, Vitamins, in the 5th edition of Clinical Chemistry: Theory,
Principles of Analysis and Current Usage deficiency. There have been a number of recent clinical
The diagnosis of Vitamin B12 deficiency is complex studies outlining the diagnostic value of holoTC in
because of the poor sensitivity and specificity of the predicting vitamin B12 status in various clinical settings,
conventional assays [1,2]. in particular as an early marker of vitamin B12 deficiency
[4,8,9-13].
Vitamin B12 (cobalamin, Cbl), the current biochemical
marker for investigation of vitamin B12 deficiency, has Serum holoTC has proved challenging to measure,
been shown to have limited specificity, to be falsely because it accounts for only 10% to 30% of the
positive if the concentration of haptocorrin (HC), which circulating vitamin B12, and because the major part of
is a vitamin B12-binding protein, is reduced, and falsely TC circulates unsaturated with vitamin B12 (apo-
negative if haptocorrin is increased. The many different transcobalamin [apoTC]) [14-16]. Early attempts to
methodologies available for measuring Cbl have resulted measure holoTC involved separating TC from HC prior
in a range of method-dependent reference intervals to quantification of vitamin B12, allowing direct
[2,3,4]. measurement of the cobalamin attached to TC [17-19],
or an indirect calculation of holoTC concentration was
The concentrations of the metabolic markers made by measurement of total plasma cobalamins and
methylmalonic acid (MMA) and homocysteine (Hcy) are the plasma cobalamins not attached to TC [20-22].
considered to be more sensitive indicators of vitamin B12
status than total vitamin B12 (Cbl). Both MMA and Hcy Nexo et al. [23] combined a sensitive enzyme-linked
increase in vitamin B12 deficiency. However, Hcy has immunosorbent assay (ELISA) for TC measurement
been shown to have low specificity, being influenced by with a simple procedure for removal of apoTC and
lifestyle factors such as smoking and alcohol intake, and developed an assay to determine both holoTC and total
increasing in patients with folate deficiency and renal TC in < 200 L of serum. They reported a method in
impairment. MMA, while considered a more sensitive which magnetic beads coated with vitamin B12 (Cbl)
indicator of vitamin B12 status than Cbl, is an expensive precipitated apoTC, then measurement of the holoTC
assay requiring specialized instrumentation not readily present in the supernatant was done by ELISA.
available in most clinical laboratories.
More recently, several specific and reproducible
Recently a new biochemical marker, holotranscobalamin commercial assays have become available for the direct
(holoTC), has been proposed as a sensitive early marker measurement of holoTC. The holoTC competitive-
for diagnosing vitamin B12 deficiency. binding radioimmunoassay (holoTC RIA) described by
Ulleland et al. [24] was the first commercially available
Vitamin B12 circulates in plasma bound to two proteins, method for holoTC measurement. HoloTC is measured
HC and transcobalamin (TC). HC, a glycoprotein with a as a two-step process. First, total TC is captured from the
half-life of 240 hours, binds 70% to 90% of total Cbl and sample by magnetic particles (microspheres) coated with
is mostly saturated with Cbl [5,6]. HC has no known monoclonal anti-human TC antibodies. Then the Cbl
functions other than its ability to bind metabolically inert bound to holoTC is released and assayed by competitive-
forms of vitamin B12 [7]. Vitamin B12 bound to TC, binding RIA standardized with recombinant human
referred to as holotranscobalamin (holoTC), has a half- holoTC [24,25].
life of 1 to 2 hours and is the residual 10% to 30% of
total vitamin B12. HoloTC is the physiologically active Following characterization of a monoclonal antibody
vitamin B12 fraction which promotes the specific uptake with > 100-fold specificity for holoTC over apoTC, a
of its vitamin B12 by all cells through receptor-mediated direct immunoassay was developed in the ELISA format
endocytosis involving a specific TC receptor found in all [26].
tissues.
Axis-Shield Diagnostics Ltd subsequently developed a
The much shorter half-life of holoTC (1 to 2 hours) has direct enzyme immunoassay utilizing the highly specific
led to the view that measurement of holoTC would be mAb for holoTC. This assay was later adapted for use on
useful as a sensitive early marker of vitamin B12 the Abbott AxSYM platform. The Abbott AxSYM
i
Holotranscobalamin
New method
Fifth edition: Marion Black
699
Holotranscobalamin
active-B12 (previously holoTC) assay [27] is a precise, to light should be avoided for both the MEIA and RIA
rapid, automated two-step, sandwich microparticle assays [27,29].
enzyme immunoassay (MEIA) for the quantitative
determination of human holoTC in serum. The active- Both the RIA and ELISA methods determined that
B12 assay is based on two monoclonal antibodies. One is holoTC concentrations were higher in EDTA plasma
a mouse monoclonal antibody specific for active B12 than serum, but the MBA assay showed no differences
(holoTC), immobilized on latex microparticles. The between plasma (EDTA) or serum, fasting or nonfasting,
second binder is a mouse monoclonal antibody to TC. specimens [28].
The assay directly quantitates active B12 (holoTC),
avoids the pretreatment steps common to all vitamin B12 Interferences
assays, and shows good correlation with the Axis-Shield Trace amounts of hemolysis (hemoglobin < 500 mg/dL),
active-B12 (holoTC) RIA assay. icterus (bilirubin < 40 mg/dL), and lipemia (triglycerides
< 1500 mg/dL) do not generally interfere with holoTC
Since the commercial RIA assay for holoTC requires a measurement in either the RIA or MEIA assays[27,29].
large sample volume (400 to 800L), Refsum et al.
developed a microbiological assay (MBA) requiring less Holotranscobalamin Reference Interval
plasma volume (100L) that could measure both holoTC The holoTC reference interval is dependent on the
and total TC [28]. methodology. There is currently no consensus regarding
reference intervals for holoTC [31].
In their MBA assay, holoTC is captured with magnetic
beads coated with TC antibodies, followed by a For the RIA assay, a study performed by Ulleland et al.
conventional MBA for cobalamin measurement [24] involving 105 healthy volunteers (20 to 80 years of
(Lactobacillus leichmannii).The MBA assay showed age) reported a reference range of 24 to 157 pmol/L. A
good precision and a strong correlation with holoTC by later study by Loikas et al. [25] utilized a more carefully
RIA. selected reference group (n = 303 healthy adults, 22 to
88 years of age), in which they chose to exclude
Reference and Preferred Methods individuals with any conditions or medications that
There is no reference method for the measurement of might cause Cbl deficiency. Loikas et al. determined the
holoTC. The methods currently in use are the ELISA 95% central reference interval to be 37 to 171 pmol/L.
[23] and MBA [28] assays, the commercially available
RIA assay [29], and the automated MEIA assay [27] on The MEIA assay quotes the 95% central reference
the Abbott AxSYM immunoassay analyzer. All assays interval as 19 to 119 pmol/L, based on a study involving
have been shown to have acceptable performance 281 healthy donors [27].
[23,27-29]. Assay selection will depend on the
individual laboratorys needs. The automated MEIA For ELISA, Nexo et al. determined the central 95%
assaywith no preanalytical treatment required, a reference interval to be 40 to 150 pmol/L (n = 137, 21 to
turnaround time of 20 minutes, and an assay throughput 65 yrs of age, healthy blood donors) [23].
of 45 tests/houris suitable for testing large numbers of
samples in a clinical diagnostic setting [30]. The MBA reference interval, 42 to 157 pmol/L, was
determined by Refsum et al. [28] in their study (n = 500,
Specimen 18 to 69 yrs of age, healthy, nonfasting blood donors),
No significant circadian variation of serum holoTC has which was performed on serum samples. They also
been found in healthy, vitamin-replete subjects on a proposed that separate reference intervals be considered
standard, nonvegan nonvegetarian Danish normal diet. in younger women (45 years), holo TC being lower in
This supports the view that holoTC is a marker of long- women than men in this age group.
term vitamin B12 status and that samples may be
collected from nonfasting subjects [7]. A number of clinical studies have considered the
biological determinants of holoTC. Several studies have
Only serum or lithium-heparin plasma is recommended reported a strong relationship between serum creatinine
for the MEIA assay. Separated samples may be stored and holoTC [8,32], but others have found no association
for up to 28 days at 2C to 8C after collection. If testing [4,33].
is delayed more than 28 days, specimens may be stored
for up to 6 months at 20C or colder [27]. No major differences in holoTC concentrations between
males and females have been observed [23-25].
Both serum and EDTA plasma may be used for RIA and
ELISA, although plasma values are a little higher (6% to However, there may be hormonal regulation of holoTC;
8%) than for serum [23,24]. younger women who were current users of oral
contraceptives (OC) were shown by Riedel et al. [34], to
The RIA assay [24,29] requires 400 L of sample (800 have Cbl and holoTC levels 25% lower than non-users.
L if analyzed in duplicate). Samples may be stored for This was not associated with significantly higher
up to 7 days at 2C to 8C and then frozen. Repeated concentrations of the metabolic markers MMA and Hcy,
freeze/thaw cycles and extended exposure of specimens suggesting redistribution rather than depletion of
intracellular Cbl. Riedel et al. concluded that these
700
Holotranscobalamin
hormonal effects may lessen the diagnostic utility of availability of Schillings test (difficulties in obtaining
total Cbl and holoTC in OC users. Further studies are labeled vitamin B12 and native human intrinsic factor)
required to determine if OC users with borderline Cbl [35], it is difficult to assess the clinical usefulness
are likely to develop Cbl deficiency clinically [34]. This (sensitivity and specificity) of holoTC.
study also found that hormone replacement therapy
(HRT) had no noticeable effect on circulating Cbl, Hvas and Nexo [2] concluded that it is still debatable
holoTC or the metabolic markers MMA and Hcy. how much additional value the measurement of holoTC
provides beyond the already existing tests for diagnosing
Interpretation vitamin B12 deficiency. They suggested further studies
A number of studies have been published to support the were required to decide whether holoTC can be used
view that holoTC would be a better indicator of vitamin alone or should be used in combination with one or the
B12 status than total serum Cbl: other vitamin B12 markers.
HoloTC values have been shown to be low in
patients with biochemical signs of vitamin B12 Miller et al. [4], found that holoTC and vitamin B12 have
deficiency [9] equal diagnostic accuracy in screening for metabolic
Low values of holoTC have been reported in vitamin B12 deficiency and that measurement of both
both vegetarians and vegans and in populations holoTC and total vitamin B12 provides a better screen for
with a low intake of vitamin B12 [4,8,10] vitamin B12 deficiency than either assay alone.
Low holoTC (but not vitamin B12) has been
reported in patients with Alzheimers disease A recent study by Clarke et al. [13] confirms that
[11] holoTC has a modestly superior diagnostic accuracy
HoloTC has also been shown to reflect vitamin compared with conventional vitamin B12 for the
B12 absorption better than serum vitamin B12 detection of vitamin B12 deficiency, but neither test can
[12] be recommended to screen asymptomatic populations.
Data from Hermann et al. [8] supported the view that Holotranscobalamin Performance Goals
measurement of holoTC and MMA provide a better HoloTC is not currently regulated for proficiency
indication of Cbl status than the measurement of total testing. Discussions are underway in Europe with a view
vitamin B12, holoTC being a more sensitive marker than to introducing a holoTC proficiency testing scheme.
the functional marker of vitamin B12 metabolism,
methylmalonic acid (MMA). They found that the use of Clinical diagnostic laboratories tend to perform either
holoTC and MMA enabled them to differentiate between the automated (no pretreatment of sample) MEIA assay
storage depletion and functional vitamin B12 deficiency. [27] or the manual RIA assay [29]. The total precision
for the RIA assay (using manual pipetting steps) as
The effect of diminished renal function on holoTC determined using the National Committee for Clinical
concentration is unclear. Hermann et al. [8] expressed a Laboratory Standards (NCCLS) Protocol EP5-T2 [36]
cautionary note: because renal failure may impair was shown to be
cellular holoTC uptake, renal patients will have a higher 12% to 8%, across a range of holoTC concentrations
requirement of circulating holoTC. They concluded that from 14 to 139 pmol/L [29].
holoTC could not be used as a marker of vitamin B12
status in patients with renal dysfunction [8]. The precision for the AxSYM active B12 automated
MEIA assay [27] was determined using the NCCLS
Likewise, Hvas and Nexo [32] found that an increase in Protocol EP5-A2 [37]. The AxSYM active-B12 assay
holoTC and vitamin B12 were significantly associated [27] is designed to have a precision of less than 10%
with plasma creatinine: the higher the plasma creatinine, total CV for holoTC (active B12) in the range of 23 to 50
the higher the holoTC and vitamin B12 values. pmol/L. This performance goal is achievable as
demonstrated by Brady et al. [30], who showed total
Miller et al. [4] found no correlation between creatinine imprecision of 6.3% to 8.5%, and intraassay imprecision
and holoTC. More recently, Loikas et al. [33] examined of 3.4% to 5.1%. The ELISA holoTC assay has been
the effect of renal impairment on the vitamin B12 shown to have a total precision (CV) of 8% at 33 pmol/L
deficiency markers (vitamin B12, Hcy, MMA, holoTC) in and 7% at 88 pmol/L [23]. For the MBA assay, Refsum
1011 aged subjects. Using serum cystatin C as the et al. [28] assessed between-assay imprecision for
indicator of renal impairment, they found that the holoTC using five human plasma pools (27 to 167
metabolic markers Hcy and MMA, but not vitamin B12 pmol/L). Between-day CVs for holoTC were 4% to 9%.
or holoTC, were affected by renal function impairment.
All of the above assays have demonstrated acceptable
The effect of renal impairment on holoTC needs to be holoTC assay performance.
further clarified, particularly since vitamin B12 deficiency
and renal impairment are common in the aged References
population. 1 Schneede J, Ueland PM. Novel and established
markers of cobalamin deficiency:
Since there is no gold standard method available to complementary or exclusive diagnostic
diagnose vitamin B12 deficiency, owing to the decreased strategies. Semin Vasc Med 2005;5:140-55.
701
Holotranscobalamin
2 Hvas AM, Nexo E. Diagnosis and treatment of 17 Lindgren A, Kilander A, Bagge E, Nexo E.
vitamin B12 deficiency: an update. Holotranscobalamin-a sensitive marker of
Haematologica 2006;91:1506-1512. cobalamin malabsorption. Eur J Clin Invest
3 Green R. Metabolite assays in cobalamin and 1999;29:321-9.
folate deficiency. Baillieres Clin Haematol 18 Benhayoun S, Adjalla C, Nicolas JP, Gueant JL,
1995;8:533-66. Lambert D. Method for the direct specific
4 Miller JW, Garrod MG, Rockwood AL, measurement of vitamin B12 bound to
Kushmir MM, Allen LH, Haan MN, Green R. transcobalamin II in plasma. Acta Haematol
Measurement of total vitamin B12 and 1993;89:195-9.
holotranscobalamin, singly and in combination, 19 van Kapel J, Wouters NM, Lindemans J.
in screening for metabolic vitamin B12 Application of heparin-conjugated sepharose
deficiency. Clin Chem 2006;52:278-85. for the measurement of cobalamin-saturated and
5 Seetharam B, Li N. Transcobalamin II and its unsaturated transcobalamin II. Clin Chim Acta
cell surface receptor. Vitam Horm 2000;59:337- 1988;172:297-310.
66. 20 Vu T, Amin J, Ramos M, Flener V, Van yo L,
6 Obeid R, Morkbak A, Munz W, Nexo E, Tisman G. New assay for the rapid
Hermann W. The cobalamin-binding proteins determination of plasma holotranscobalamin II
transcobalamin and haptocorrin in maternal and levels: preliminary evaluation in cancer
cord-blood sera at birth. Clin Chem patients. Am J Hematol 1993;42:202-211.
2006;52:263-269. 21 Herzlich B, Herbert V. Depletion of serum
7 Hvas A, Gravholt C, Nexo E. Circadian holotranscobalamin II. An early sign of
variation of holotranscobalamin (holoTC) and negative vitamin B12 balance. Lab Invest
related markers. Clin Chem Lab Med 1988;58;332-337.
2005;43:760-764. 22 Lindemans J, Schoester M, van Kapel J.
8 Hermann W, Obeid R, Schorr H, Geisel J. Application of a simple immunoadsorption
Functional vitamin B12 deficiency and assay for the measurement of saturated and
determination of holotranscobalamin in unsaturated transcobalamin II and R-binders.
populations at risk. Clin Chem Lab Clin Chim Acta 1983;132;53-61.
Med;2003;41:1478-1488. 23 Nexo E, Christensen AL, Hvas AM, Petersen
9 Obeid R, Journa M, Hermann W. Cobalamin TE, Fedosov SN. Quantitation of holo-
status (holotranscobalamin methylmalonic acid) transcobalamin, a marker of vitamin B12
and folate as determinants of homocysteine deficiency. Clin Chem 2002;48:561-562.
concentration. Clin Chem 2002;48:2064-2065. 24 Ulleland M, Eilertsen I, Quadros E, Rothenberg
S, Fedosov S, Sundehagen E, Orning L. Direct
10 Lloyd-Wright Z, Hvas AM, Moller J, Sanders assay for cobalamin bound to transcobalamin
T, Nexo E. Holotranscobalamin as an indicator (holo-transcobalamin) in serum. Clin Chem
of dietary vitamin B12 deficiency. Clin Chem 2002;48:526-532.
2003;49:2076-2078. 25 Loikas S, Loppenen M, Suominen P, Moller J,
11 Refsum H, Smith AD. Low vitamin B12 status Irjala K, Isoaho R et al. RIA for serum holo-
in confirmed Alzheimers disease as revealed transcobalamin: method evaluation in the
by serum holotranscobalamin. J Neurol clinical laboratory and reference interval. Clin
Neurosurg Psychiatry 2003;74:959- 961. Chem 2003;49:455-462.
12 Bor M, Nexo E, Hvas AM. Holo- 26 Orning L, Rian A, Campbell A, Brady J,
transcobalamin concentration and Fedosov S, Bramlage B, Thompson K, Quadros
transcobalamin saturation reflect recent vitamin E. Characterization of a monoclonal antibody
B12 absorption better than does serum vitamin with specificity for holo-transcobalamin. Nutr
B12. Clin Chem 2004;50:1043-1049. Metab 2006;3:3
13 Clarke R, Sherliker P, Hin H, Nexo E, Hvas 27 Abbott AxSYM Active-B12 Assay package
AM, Schneede J et al. Detection of vitamin B12 insert. REF 1P43-20. January 2007.
deficiency in older people by measuring 28 Refsum H, Johnston C, Guttormsen AB, Nexo
vitamin B12 or the active fraction of vitamin B12 E. Holotranscobalamin and total transcobalamin
, holotranscobalamin. Clin Chem 2007;53:963- in human plasma: determination, determinants,
970. and reference values in healthy adults. Clin
14 Markle HV. Cobalamin. Crit Rev Clin Lab Sci Chem 2006;52:129-137.
1996;33:247-356. 29 Axis-Shield Diagnostics Ltd. HoloTC RIA. Ref
15 Herbert V. Staging vitamin B12 (cobalamin) FHTC100.
status in vegetarians. Am J Clin Nutr 30 Brady J, Wilson L, McGregor L, Valente E,
1994;59(Suppl 5):1213S-22S. Orning L. Active B12: a rapid, automated assay
16 Nexo E, Hansen M, Rasmussen K, Lindgren A, for holotranscobalamin on the Abbott AxSYM
Grasbeck R. How to diagnose cobalamin Analyzer. Clin Chem 2008;54:567-573.
deficiency. Scand J Clin Lab Invest 31 Morbak AL, Heimdal R, Emmens K, Molloy A,
1994;54(Suppl 219):61-76. Hvas AM, Schneede J et al. Evaluation of the
technical performance of novel
702
Holotranscobalamin
holotranscobalamin (holoTC) assays in a 35 Schilling RF. Intrinsic factor studies II. The
multicenter European demonstration project. effect of gastric juice on the urinary excretion of
Clin Chem Lab Med 2005;43:1058-1064. radioactivity after the oral administration of
32 Hvas AM, Nexo E. Holotranscobalamin: a first- radioactive vitamin B12. J Lab Clin Med
choice assay for diagnosing early vitamin B12 1953;42:860 -5.
deficiency. J Intern Med 2005;257:289-98. 36 National Committee for Clinical Laboratory
33 Loikas S, Koskinen P Irjala K, Lopponen M, Standards. Evaluation of Precision Performance
Isoaho R, Kivela S-L, Pelliniemi T-T. Renal of Clinical Devices: Tentative Guideline.
impairment compromises the use of total NCCLS Document EP5-T2. Villanova, PA:
homocysteine and methylmalonic acid but not NCCLS; March 1992.
total vitamin B12 and holotranscobalamin in 37 National Committee for Clinical Laboratory
screening for vitamin B12 deficiency in the Standards. Evaluation of
aged. Clin Chem Lab Med 2007;45:197-201. Precision Performance of Quantitative
34 Riedel B, Bjorke Monsen AL, Ueland P, Measurement Methods. Approved Guideline.
Schneede J. Effects of oral contraceptives and NCCLS Document EP5-A2. Wayne, PA:
hormone replacement therapy on markers of NCCLS; 2004.
cobalamin status. Clin Chem 2005;51:778 -81.
Tables
Holotranscobalamin Methods Summary Table
Method 1: RIA
Principle of analysis: Radioimmunoassay (competitive binding)
Comments: Serum, EDTA plasma, large sample volume required (400 L), sensitive, precise, time consuming,
multiple manual steps
Suitable for use in clinical laboratories
Method 2: ELISA
Principle of analysis: Magnetic beads coated with vitamin B12 (cobalamins) precipitate apo-transcobalamin
(apoTC), and the protein moiety of holoTC present in the supernatant is measured by ELISA.
Comments: Sensitive, precise, time consuming, requiring skill to pretreat the samples prior to running on an
automated platform. 100 L sample.
Method 3: MEIA
Principle of analysis: Microparticle enzyme immunoassay (MEIA). Two-step sandwich microparticle enzyme
immunoassay
Comments: Sensitive, precise, serum or lithium heparin plasma, automated, rapid assay suitable for use in clinical
laboratories
Method 4: MBA
Principle of analysis: Microbiological assay (MBA); holoTC captured with magnetic beads coated with TC
antibodies; followed by a conventional MBA for cobalamin measurement (Lactobacillus leichmannii)
Comments: Precise, time consuming, useful for studies with limited sample volume
703
Homocysteine
Homocysteine
Sheila Dawling
Name: Homocysteine or HcyH, total homocyst(e)ine or tHcy
Clinical significance: Refer to Chapter 37, Coronary Artery disease in the 5th edition of Clinical
Molecular mass: 135.2: convert molar to mass units: g/mL or mg/L = mol/L 0.1352
Chemical class: Amino acid
Structures:
O NH2 O O
HS HO S
OH S OH HS OH
NH2 O NH2 NH2
Homocysteine Homocystine Cysteine

(HcyH) (Hcy-Hcy) (Cys)
O O O
S S
H3C OH HO S OH O
NH2 NH2 NH2 H2N
S
Methionine Homocysteine Mixed Disulfide
(Met) (Hcy-Cys) Homocysteine thiolactone
i exists in a variety of oxidized forms, mostly through
disulfide links with other sulfur-containing amino acids
Homocysteine (2-amino-4-mercaptobutyric acid [Hcy]) or those contained in proteins such as albumin and the
was first described in 1932. It is a nonessential, sulfur- globulin proteins IgG, 1-acid glycoprotein, and HDL. A
containing amino acid that is not required for protein smaller fraction becomes irreversibly bound through
synthesis, but its clinical relevance became apparent amide linkage via reaction with Hcy-thiolactone [2].
some 30 years later when McCully observed the Presence of an excess of some of these forms is
relationship between grossly elevated plasma associated with a variety of adverse effects. These
homocysteine and accelerated cardiovascular disease [1]. include but are not limited to folate and B12 deficiency
Its importance derives from two key factors. First is its [3,4], cardiovascular disease [5,6], birth defects [7],
position as an intermediary metabolic product that pregnancy complications [8], psychiatric disorders [9],
derives from the demethylation of the essential amino cognitive impairment in the elderly [10], and
acid methionine, making it a key player in the osteoporosis [11]. Whether homocysteine is directly
folate/vitamin B12 cycles. Once formed, homocysteine is responsible for these effects or is merely the tangible
either remethylated to methionine in a step involving evidence of perturbed one-carbon metabolism or other
vitamin B12 and folate, or it is metabolized to cysteine in thiol pathways of different etiologies is still highly
two vitamin B6dependent reactions (Figure 1). debated [12].
Defective homocysteine metabolism results in high Plasma Hcy Content Protein Distribution %
plasma concentrations of homocysteine, accompanied by Form mol/L % of Total Hcy Cys Cys-Gly
either elevated or decreased levels of methionine _____________________________________________
(cystathionine -synthase or methionine synthase Protein-bound 8 80
deficiency, respectively). Secondly, because of its strong forms
chemical reactivity as a reducing agent, its presence in Albumin
plasma in its native form is short-lived. In plasma, it 88 79 67
Globulins
12 21 33
i
Homocysteine Disulfides 2 20
Previous and current authors of this method: Hcy-Cys
First edition: Not done Hcy-Hcy
Methods edition: Not done Hcy 0.2 0.2
Second edition: Not done
Third edition: Not done tHcy 10 100
Fourth edition: Kirby Marsh, Gregory Rezinger
Fifth edition: Sheila Dawling Hcy, Free homocysteine; tHcy, Total homocysteine
704
Homocysteine
Accurate measurement of homocysteine poses These take the form of an initial supply of the
significant analytical problems, since the small instrumentation, with purchase of additional reagents
homocysteine molecule must be measured specifically and consumables as necessary. The Bio-Rad assay [16]
against a background of a 20- to 50-fold excess of involves a 5-minute (50C) incubation of sample and
structurally similar molecules. Two approaches are taken internal standard with trialkylphosphine reducing agent
for the measurement of homocysteine in plasma: (1) and derivatizing reagent (4-aminosulfonyl-7-fluoro-
discrete analysis for the free homocysteine together with
2,1,3-benzoxdiazole, ABD-F). After precipitation of
some or all of the other oxidized species or (2) the
proteins with trichloroacetic acid, the supernate is
conversion of all the plasma forms to a single entity
(known as total homocysteine, or total homocyst(e)ine or chromatographed by reverse phase, and compounds are
tHcy). Selection of the method depends on the clinical detected and quantified by fluorescence detection (385
application, although for routine screening, the second nm excitation, 515 nm emission). Within less than 5
approach is the one most usually taken, since this can minutes, there is good separation of glutathione,
most often be accomplished with commercially available cysteine, homocysteine, and Cys-Gly. The Drew S30
reagents which run on automated chemistry platforms. system [17] is available on similar terms. This assay
involves a short incubation of sample and internal
Initial methods for measuring homocysteine used an standard (2-mercaptoethylamine) with TCEP (tris[2-
amino-acid analyzer (Table 1, Method 1). In these carboxyethyl]) phosphate reducing agent. After
procedures, the heparinized plasma samples were precipitation of proteins with trichloroacetic acid, the
deproteinized and the free homocysteine (as well as supernate is derivatized with ammonium SBDF (7-
other amino acids) sequestered in the filtrate. These were fluorobenzo-2-oxa-1,3-diazole-4-silfonate) for 50
then separated by cation-exchange chromatography and
minutes (60C). The thiol derivatives are subsequently
quantified by post-column derivatization with
chromatographed by reverse phase, and compounds are
Ninhydrin, using a dedicated amino acid analyzer [13].
An alternative approach is to prepare fluorescent detected and quantified by fluorescence detection. While
derivatives of the amino acids in the filtrate and then the Bio-Rad assay is a conventional HPLC system, the
separate these by HPLC with quantitation by Drew S30 takes the form of a small black-box
fluorimetric detection [14]. These methods require analyzer, producing a number rather than a
specialized, expensive equipment and are extremely chromatogram printout. The instrument is configured to
slow, with analysis times of 2 to 3 hours per sample. identify cysteine and Cys-Gly but only reports the value
Until recently however, these were the only HPLC for homocysteine. Thus specimens that require dilution
methods available with sufficient sensitivity (1 mol/L) must be diluted in a mixture of 200 mol/L cysteine and
to determine the free form of homocysteine in plasma. 30 mol/L Cys-Gly. Both assays compare well to the
These are still widely used for specialized monitoring of CDC method, perform well, and are suited to high-
patients with homocystinuria, where both plasma and throughput analysis, with run times of 5 to 6 minutes per
urine concentrations may be measured, together with sample. The ChromSystems kit has been the most recent
methionine and other essential amino acids, to monitor
entry to the market. This is run on a conventional HPLC
the patients dietary status.
system with fluorimetric detection. The assay range is
Reduction of all plasma species of oxidized 0.5 to 200 mol/L and compares well to other HPLC
homocysteine to yield free, reduced homocysteine is methods [18].
used by most procedures to yield total plasma
homocysteine (tHcy). Total Hcy is then either quantified Electrochemical detection for HPLC has the distinct
by HPLC (Table 1, Method 2) or is converted to S- advantage over fluorimetric detection in that the thiols
adenosyl-homocysteine (SAH) and subsequently can be detected at the same sensitivity without prior
monitored traditionally with coupled enzymatic reactions derivatization, thus eliminating what can be a time-
(Table 1, Method 4) or reacted with an SAH antibody consuming step. Martin et al. [19] described a method
(Table 1, Method 3). These methods are described that used penicillamine as an internal standard and
further. dithiothreitol as a reducing agent for 15 min at 37C.
Proteins were precipitated with sulfosalicylic acid and
Stabler et al. [15], described a capillary gas the supernate injected directly onto a C18 column with a
chromatography-mass spectrometry method using t-
mobile phase of 10 mM sodium dihydrogen phosphate,
butyldimethylsilyl derivatives and tritiated internal
standards to quantify plasma concentrations of total pH 2.75 with phosphoric acid, 15% methanol, and 12
homocysteine, cysteine, and methionine. They reduced mM octane sulfonic acid modifier. The run time was
all plasma forms of the three compounds by reduction approximately 8 min. The electrochemical detector was
with 2-mercaptoethanol and used selected ion operated at a screening potential of +0.4 V and detection
monitoring for identification and quantification of the at +1.0 V. ESA diagnostics promotes a method for use
amino acids. This method has not received much on its eight-channel CoulArray detector. While not a
attention. kit as such, the method is fully validated and U.S.
Food and Drug Administration (FDA)-approved. The
Three HPLC kits are currently available (Bio-Rad, Drew porous graphite electrodes require minimal maintenance
Scientific, and ChromSystems) for total homocysteine. compared to other detectors of this type, and the assay,
705
Homocysteine
with a run time of about 15 minutes, gives results that homocysteinefluorescein tracer for sites on a
correlate well with the CDC reference method [20]. A monoclonal antibody specific for the S-adenosyl-L-
30-minute incubation with penicillamine internal homocysteine moiety. The amount of tracer bound is
standard and TCEP (tris-2[carboxyethyl]-phosphine quantified by measuring the fluorescence polarization of
hydrochloride) reducing agent is required. Proteins are the resultant mixture, with this being inversely
precipitated with perchloric acid, and the supernate is proportional to the free S-adenosyl-L-homocysteine
concentration and hence inversely to the tHcy in the
applied to the C18 column. The mobile phase is 0.15 M
specimen. The IMx method is a batch procedure (20
sodium dihydrogen phosphate with 1 mM sodium
specimens in 1 hour), whereas the AxSYM is a random-
dodecylphosphate modifier, pH 2.8 with 10% access analyzer (35-minute assay). This method is still
acetonitrile. Four of the eight channels are used for this widely used and compares very favorably with the CDC
application, at settings of 520, 750, 800 and 820 mV. reference HPLC method, and has excellent precision
The assay is linear to 100 mol/L but measures only the (Table 2) [26]. The method is also available from Bio-
total homocysteine. Rad as a microtiter plate assay (Axis) [27].
HPLC with tandem mass spectrometry has been applied The Bayer ADVIA Centaur assay uses the same
to the analysis of total homocysteine. Sample chemistry as described for the FPIA assay. (Table 1,
preparation is as described above for electrochemical Method 3b, and Figure 3). Dithiothreitol is used as a
detection, and the column effluent is monitored in SRM reducing agent, and reduced forms of homocysteine are
(selected reaction monitoring mode) (136.1 90.0 amu converted enzymatically to SAH. The solid-phase
for homocysteine and 140.0 94.0 amu for d4- (paramagnetic particles covalently linked to SAH) is
homoysteine) by neutral loss of formic acid. Although added, followed shortly by the Lite Reagent (monoclonal
the method benefits from the use of deuterated internal mouse antibodies to SAH labeled with acridinium ester).
standards, it offers little advantage for the additional Any SAH produced competes with SAHmagnetic
expense incurred [21]. This procedure can be simplified particle conjugates for acridinium-labeled antibodies.
and adapted to 96-well plates for high throughput by After separation of the magnetic particles, free tracer is
analyzing reduced specimens without derivatization and liberated with the trigger reagent, and a
monitored as above [22]. Tandem mass spectrometry is chemiluminescent signal is produced in inverse
used for high-throughput testing of dried blood spots for proportion to the amount of homocysteine present in the
the presence of homocystinuria in newborn screening sample. Results are available in 18 minutes and
programs. However, for this specialist application, the compared favorably with both FPIA and HPLC [28].
butylated derivatized specimen is aspirated directly into This is a widely used method, but precision is rather
the mass spectrometer without any chromatography, and disappointing (Table 2).
the amino acid methionine is monitored at mass 209.2 in
neutral loss mode. Neither of these immunoassays is susceptible to a hook
effect because of the relatively narrow range of
Although now obsolete, the introduction of a homocysteine concentrations determined, but both may
radioenzyme assay for homocysteine paved the way for be the subject of HAMA or heterophilic interference.
the chemical analysis of this compound [23]. Central to Interference by heterophile antibodies should be
this assay was the reversibility of the reaction catalyzed suspected when serial specimens from the same patient
by S-adenosyl-L-homocysteine hydrolase (SAHH), or repeat analysis of the same sample produce wildly
which allowed both non-radiolabeled S-adenosyl-L- different results, or when recovery on dilution is not
homocysteine and L-homocysteine to act as L- linear. These polyclonal autoantibodies, either IgG or
homocysteine donors in the synthesis of radiolabeled S- IgM, bind across a broad spectrum of different animal
[14C]-adenosyl- L-homocysteine from radiolabeled antibodies used in kit manufacture and thus form
[14C]-adenosine. The product was quantified by HPLC. sandwiches between the capture and detection
antibodies, producing a spurious signal. Analysis by an
The first of the assays to be made available for a alternative method is usually sufficient to align the
commercial clinical chemistry platform was in the early measured result with the clinical picture. Should it be
1990s, with the development of the fluorescence necessary to prove that these antibodies exist, a broad
polarization immunoassay (FPIA) for the Abbott IMx spectrum passive precipitant such as PEG can be used,
[24] and subsequently the AxSYM platform [25]. In this but it is preferable to use an active blocking agent such
procedure (Table 1, Method 3a, and Figure 2) all as The Scantibodies Heterophilic Blocking Reagent
additions are performed automatically by the Tubes (HBRT). This proprietary mix of lyophilized
instruments pipetting system. The homocystine, mixed mouse anti-human IgM has high affinity for human anti-
disulfide, and protein-bound forms of homocysteine in animal antibodies commonly used in reagent
the serum or plasma are reduced with dithiothreitol to manufacture (mouse, goat, sheep, and rabbit) and to
form the reduced, free form of homocysteine (tHcy). rheumatoid factor. Each tube contains sufficient reagent
This product is then converted to S-adenosyl-L- to inactivate 500 L of patient sera and is incubated for
homocysteine by the catalytic activity of the enzyme 1 hour at room temperature. Although FDA approved for
SAHH in the presence of excess adenosine. The S- this purpose, the value obtained from a sample treated
adenosyl-L-homocysteine produced is quantified in a with HBR should not be used as a reportable result.
competitive reaction between it and S-adenosyl-L-
706
Homocysteine
With recent renewed interest in homocysteine, enzymic homocysteine using TCEP (tris-2[carboxyethyl]-
assays for chemistry platforms have been developed phosphine hydrochloride).
(Table 1, Method 4). These are often dealt with as a
group, although as shown by the three examples below, Reduced homocysteine is reacted with added S-
their chemistries are quite varied, and consequently they adenosyl-L-methionine and the enzyme homocysteine
often show poor precision when considered together methyltransferase, producing S-adenosyl-L-
(Table 2). homocysteine and methionine.
Vitros MicroTip ECi Assay (Table 1, Method 4a). Homocysteine + SAM HMT SAH + methionine
This is a three-reagent homogeneous enzyme assay.
These same reagents are available as the Catch Added SAHH then cleaves S-adenosyl-L-homocysteine
Homogeneous Enzymatic Assay (Catch Inc, Bothell, to adenosine and regenerates homocysteine, which
WA), which can be run on a number of automated perpetuates the reaction.
chemistry platforms [29]. Reagent 1 contains serine,
NADH, and lactate dehydrogenase (LD); Reagent 2
contains TCEP (tris-2[carboxyethyl]-phosphine SAH SAHH homocysteine + adenosine
hydrochloride) reducing agent; Reagent 3 contains
cystathionine -synthase (CBS) and cystathionine - This adenosine is enzymatically converted to inosine and
lyase (CBL). The first step reduces all plasma forms of ammonia by added adenosine deaminase.
homocysteine to free homocysteine using TCEP (tris-
2[carboxyethyl]-phosphine hydrochloride). This reacts Adenosine ADA inosine + NH3
with added serine and the enzyme cystathionine -
synthase (CBS) to form L-cystathionine: The final step in the reaction is to quantify the liberated
ammonia with the enzyme Glutamate Dehydrogenase in
the presence of added 2-oxoglutarate.
Serine + homocysteine CBS L-cystathionine
Cystathionine -lyase (CBL) is then added, which NH3 + 2-oxoglutarate + NADH GLDH glutamate +
catalyses the conversion of L-cystathionine to NAD+ + H2O
homocysteine, pyruvate, and ammonia:
The NAD+ generated can be monitored by the rate of
CBL homocysteine + pyruvate + change in absorbance at 340 nm, which is proportional
L-Cystathionine
to the concentration of homocysteine in the plasma
NH3
sample.
The final reaction is the reduction of pyruvate to lactate
A third enzymatic approach (Table 1, Method 4c) is
by added lactate dehydrogenase (LD):
based on a single, highly specific enzyme, homocysteine
,-lyase (rHCYase). As described by Tan et al. [31], the
Pyruvate + NADH LD lactate + NAD+ reactions are performed in 96-well plates. In the first
step, all forms of homocysteine are reduced with
The NAD+ generated can be monitored by the rate of dithiothreitol, and the reduced homocysteine is
change in absorbance at 340 nm, which is proportional converted to 2-oxobutyrate, ammonia, and hydrogen
to the concentration of homocysteine in the plasma sulfide.
sample.
Homocysteine + SAM rHCYase 2-oxobutyrate + NH3
Results compare favorably with both FPIA and the Bio- Vit B6 + H2S
Rad HPLC [29]. There is negative interference from In the second step, the reaction is terminated by the
elevated triglycerides (>500 mg/dL). The assay also addition of DBPDA (N,N-dibutyl phenylene diamine)
measures endogenous cystathionine, but this is usually and oxidizing agent (potassium ferricyanide) in sulfuric
an insignificant amount. The major drawback with this acid. DBPDA combines with hydrogen sulfide to form
assay is that it is not compatible with some of the other 3,7-bis(dibutyl amino)phenothiazine-5-ium chloride,
tests typically run on chemistry platforms. Carryover on which is strongly fluorescent (665 nm excitation, 690 nm
reagent probes and reaction cuvettes from emission).
hydroxylamine used in some colorimetric iron assays
can falsely lower homocysteine results. In addition, these H2S + DBPDA potassium 3,7-bis(dibutyl
reagents contain a significant amount of lipase and amino)phenothiazine-5-ium
should not be used concurrently with plasma lipase chloride ferricyanide
testing.
One 96-well plate can be processed in about 1 hour, but
Diazyme Assay (Table 1, Method 4b). this application has not yet been used commercially.
This two- or three-reagent system can be run on a Results compared well with HPLC.
number of automated chemistry platforms [30]. The first
step reduces all plasma forms of homocysteine to free
707
Homocysteine
Reference and Preferred Methods Interferences

The reference method is provided by the Centers for Hemolysis has not been found to interfere with
Disease Control and Prevention (CDC) [32]. This HPLC homocysteine determinations. Homocysteine is stable to
procedure uses reduction, precipitation, derivatization, repeated freeze/thaw cycles, but the matrix often
and fluorometric detection, as described at the end of degenerates, making analysis difficult [33]. Adenosine
this chapter. This method is not applicable to routine analogues such as 3-deazaadenosine are effective in
testing in the clinical setting because of time constraints preventing leakage of homocysteine from red cells but
and the dedicated equipment required. However, it does are not compatible with assays that use enzymatic
have value in the investigation of unusual or complicated reactions with SAHH [35].
patients. Review of College of American Pathologists
(CAP) survey data (Table 2) shows almost 50% of labs Heterophile antibody or HAMA interference with
reporting results to CAP are using chemiluminescence immunoassay tests should be suspected when repeated
and 30% fluorescence polarization. Only a handful of analysis produces poor replication, there is poor
labs are using HPLC, although the incidence is rising recovery on dilution, or different methodologies provide
slightly as kit assays are becoming available. In contrast, discrepant results, as discussed earlier.
over half the participants in the United Kingdom
National External Quality Assessment Service (NEQAS) Homocysteine Reference Interval
are using FPIA and a quarter HPLC methods; Rasmussen et al. proposed age- and gender-specific
chemiluminescence is not widely used. FPIA reference intervals for total homocysteine in plasma
consistently outperforms other methods in terms of [38], but given the poor precision of some assays, these
variance. The comparative results are shown in Table 2 differences are probably not reproducible in the clinical
and discussed in the section on Performance Goals. setting. The following upper reference limits have been
proposed by an expert committee for different groups,
Specimen
After blood collection but before removal of the red based on the major determinants: age, pregnancy, folate
cells, there is a time- and temperature-dependent supplementation, and following methionine challenge
increase in the serum/plasma homocysteine which is [12]. Concentrations in newborns are approximately half
attributed to its continued release from erythrocytes. those seen in healthy adults.
This increase can be as much as 100% in 8 hours at Group Folate
room temperature [33,34]. Gel separators can help if
centrifugation is prompt. Therefore, most methods use Supplemented Not Supplemented
EDTA plasma as the specimen of choice; serum Fasting tHcy mol/L
produces slightly higher values because of the time
required for the clotting process. Heparin may be used if Pregnancy 8 10
EDTA interferes with the analytical method used. All Children < 15 yr 8 10
specimens should be cooled on ice and transported to the Adults 15 65 yr 12 15
lab for immediate centrifugation. Once plasma is
separated, homocysteine is relatively stable (>4 days at Elderly > 65 yr 16 20
room temperature, several weeks refrigerated, several Post-methionine-load 5 times fasting level or
years at 20C). Homocysteine is stable to repeated
freeze/thaw cycles, but the matrix often degenerates, 4 6 hr 40 mol/L above fasting level
making analysis difficult [33].
In patients with homocystinuria, concentrations of total
Adenosine analogues such as 3-deazaadenosine are homocysteine are often > 100 mol/L on treatment and
effective in preventing leakage of homocysteine from
> 250 mol/L untreated. Free, unbound homocysteine is
red cells but are not compatible with assays that use seen in the plasma of these patients, usually 1 to 10
enzymatic reactions with SAHH [35]. In any event, the mol/L, and larger amounts of homocysteine can be
use of a specialist blood collection tube is not practical
detected in their urine.
in a routine clinical setting.
Venous stasis does not affect the homocysteine content Interpretation

of plasma, but samples collected in the supine position Homocystinuria, a rare autosomal recessive disease, is a
are 10% lower than those with the patient upright, consequence of the formation of an enzyme deficiency
presumably because of the decline in albumin, its main (usually cystathionine -synthase, less frequently
serum binding protein. Plasma homocysteine methionine synthase or one of the other enzymes; see
concentrations are usually measured fasting. Figure 1) involved in the homocysteine metabolic
Homocysteine concentration may decrease 10% to 15% pathway. Extremely high concentrations (20 to 50 times
over the first few hours following eating and then rise normal) of homocysteine are present in the serum and
again by 6 hours following eating, in proportion to the urine. Clinical sequelae include unexplained mental
methionine (protein) content of the meal [36]. A retardation, thromboembolism, lens dislocation,
methionine-loading test has been used to identify progressive myopia, osteoporosis, Marfan-like features,
persons heterozygous for CBS deficiency [37], which psychiatric disorders, or megaloblastic anemia, as well
involves a plasma homocysteine measurement 4 to 6 as affected siblings [39]. Because of the relationship
hours after intake of 100 mg methionine/kg body weight. between high homocysteine concentrations and vascular
708
Homocysteine
disease, McCully, in 1969, suggested that mild to randomization can be thought of as a natural RCT.
moderate hyperhomocysteinemia might be a factor The method relies on getting good estimates from
involved in premature vascular disease [1]. At that time, genetic association studies. Misleading conclusions can
however, no methods were available to measure also be drawn in the presence of linkage disequilibrium,
homocysteine at the low concentrations normally present genetic heterogeneity, pleiotropy, or population
in plasma. Recent improvements in methods, as stratification.) Plasma homocysteine can be lowered with
described earlier, have provided a means to detect B vitamins and folic acid, and persons with high plasma
homocysteine at the concentrations present in normal levels or dietary folate and B12 intake have a lower risk
plasma. of CHD [47-50]. In contrast to what was expected on the
basis of epidemiological evidence, large randomized
It transpires that many factors influence plasma studies were unable to show that lowering homocysteine
homocysteine concentrations, more commonly an with B vitamins prevented recurrent stroke, myocardial
elevation rather than a decline. These should be taken infarction, or death in patients who had previously had a
into account when interpreting the result and can be stroke [51] or myocardial infarction [52]. Studies on
summarized in the following categories: genetic, restenosis after percutaneous intervention also yielded
physiological, lifestyle-related, clinical conditions, and inconsistent results [19,20].
drugs. There are excellent summaries of these in Tables
3, 4, and 5 of Refsum et al. [12] The most likely cause of As expected, many pathological events appear to be
hyperhomocysteinemia depends on the age of the person mediated by changing the activity or blood concentration
and magnitude of the elevation. of the B vitamins, in particular folate and/or cobalamin,
since these vitamins are required for homocysteine
Folate and/or cobalamin deficiency and/or renal metabolism (Figure 1). Other mechanisms may arise
impairment account for the majority of cases in a from altering renal function and more rarely by
population where food is not fortified with folic acid. In influencing enzyme activity. An interesting suggestion is
folate-supplemented populations, renal impairment and that individual differences in the distribution of Hcy
cobalamin deficiency prevail [12]. Homozygosity for amongst the plasma proteins may help explain some of
MTHFR (677C > T) polymorphism (see Figure 1) is the the inconsistency between tHcy and clinical
most common genetic determinant. These individuals manifestations. This is because the Hcy bound to
usually have an elevation of some 2.5 mol/L, although globulins (such as HDL) is not as easily exchanged with
this may not be so relevant in persons with adequate the aqueous phase as that bound to albumin and thus
folate/B12 and riboflavin intake [40]. Mandatory folate represents a more permanent protein modification [53].
supplementation of grain products in the United States Homocysteine has been shown to have a number of
since 1998 has resulted in a decrease of about 30% in important actions at the cellular level, which can help
homocysteine concentrations in the population [41]. explain some of its pathological effects. It promotes
endothelial growth factors, stimulates platelet factor V
Studies that have sought to understand the role and XII activity, induces tissue factor expression,
hyperhomocysteinemia plays in the pathogenesis of both reduces protein C activity, suppresses expression of
arterial occlusive disease and recurrent venous thrombomodulin and heparan sulfate, directly induces
thrombosis have driven much of the analytical vascular smooth muscle proliferation, inhibits nitric
improvement and molecular investigations [12]. Case- oxide synthesis, and stimulates free radical production
controlled as well as prospective studies have and oxidation of LDL cholesterol [54]. Non-
demonstrated that plasma total homocysteine is a strong, cardiovascular effects include but are not limited to birth
graded, and independent risk factor for coronary heart defects [7], pregnancy complications [8], psychiatric
disease and stroke [6,43,44]. Evidence from Mendelian disorders [9], cognitive impairment in the elderly [10],
randomization, demonstrating an association between and osteoporosis [11]. Whether homocysteine is directly
coronary heart disease (CHD) and the methylene responsible for these effects or is merely the tangible
tetrahydrofolate reductase (MTHFR) polymorphism evidence of perturbed one-carbon metabolism or other
(677C > T), provided additional support for a causal thiol pathways of different etiologies is still highly
relationship [45,46]. (NOTE: Mendelian randomization is debated [12].
a method that allows one to test for, or in certain cases to
estimate, a causal effect from observational data in the Treatment of hyperhomocysteinemia is usually
presence of confounding factors. It uses common genetic adequately achieved inexpensively through
polymorphisms with well-understood effects on supplementation with vitamins B6, B12, and/or folate.
exposure patterns [e.g., propensity to drink alcohol] or Extreme cases, such as homocystinuria, may require a
effects that mimic those produced by modifiable methionine- and protein-restricted diet or betaine
exposures [e.g., raised blood cholesterol]. Importantly, supplementation.
the genotype must only affect the disease status
Homocysteine Performance Goals
indirectly via its effect on the exposure of interest. Desirable analytical specifications derived from studies
Because genotypes are assigned randomly when passed of biological variation indicate a desired assay bias of <
from parents to offspring during meiosis, the population
10% (0.375 interindividual coefficient of variation
genotype distribution should be unrelated to the [CV]) and an imprecision of < 5% (0.75
confounders that typically plague observational intraindividual CV) [55,56].
epidemiology studies. In this regard, Mendelian
709
Homocysteine
The analytical range should cover the 5th to 95th Hyperhomocystinaemia a common finding in
percentile of the general population (~3 to 40 mol/L) a psychogeriatric population. Eur J Clin Invest
without the need for dilution or concentration. 1996;26:853-859.
10 Smith AD. Homocysteine, B vitamins, and
Acceptable performance criteria defined by the Clinical cognitive deficit in the elderly [editorial]. Am J
Laboratory Improvement Amendments (1988) for Clin Nutr 2002;75:785-786.
measurement of plasma total homocysteine requires that 11 Herrmann M, Widmann T, Colaianni G,
laboratories be accurate to within 3 SD of the peer Colucci S, Zallone A, Herrmann W. Increased
group mean. Precision data from the CAP 2006-7 osteoclast activity in the presence of increased
participant summary reports are shown in Table 2 for homocysteine concentrations. Clin Chem
approximately 750 and 33 laboratories returning results 2005;51:2348-2353.
for tHcy assays to CAP and NEQAS, respectively. Data 12 Refsum H, Smith AD, Ueland PM, Nexo E,
clearly show FPIA outperforms other methods in Clarke R, McPartlin J, Johnston C, Engbaek F,
precision, and there is a strong negative bias for Schneede J, McPartlin C, Scott JM. Facts and
chemiluminescence compared to other methods, ranging recommendations about total homocysteine
from 20% at 5 mol/L to 35% at 70 mol/L. This bias is determinations: an expert opinion. Clin Chem
surprising, given that the assays correlate well with 2004;50:3-32.
patient specimens [28], and is most likely attributable to 13 Le Boucher J, Charret C, Coudray-Lucas C,
a matrix effect introduced by the survey materials. Giboudeau J, Cynober L. Amino acid
Certified reference material is lacking, and this has been determination in biological fluids by automated
blamed for some of the poor correlation between ion-exchange chromatography: performance of
methods. A recently introduced calibration/linearity Hitachi L-8500A. Clin Chem1997;43:1421-
verification survey at CAP has not been widely 1428.
subscribed. 14 Araki A, Sako Y. Determination of free and
total homocysteine in human plasma by HPLC
with fluorescence detection. J Chromatogr
References
1 McCully KS. Vascular pathology of 1987;422:43-52.
homocystinemia: implications for the 15 Stabler SP, Marcell PD, Podell ER, Allen RH.
Quantitation of total homocysteine, total
pathogenesis of arteriosclerosis. Am J Pathol
1969;56:111-128. cysteine, and methionine in normal serum and
2 Hortin GL, Seam N, Hoehn GT. Bound urine using capillary gas chromatography-mass
spectrometry. Anal Biochem 1987;162:185-
homocysteine, cysteine, and cysteinylglycine
distribution between albumin and globulins. 196.
16 Dias VC, Bamforth FJ, Tesanovic M, Hyndman
Clin Chem 2006;52:2258-2264.
ME, Parsons HG, Cembrowski GS. Evaluation
3 Savage DG, Lindenbaum J, Stabler SP, Allen
RH. Sensitivity of serum methylmalonic acid and intermethod comparison of the Bio-Rad
high-performance liquid chromatographic
and total homocysteine determinations for
method for plasma total homocysteine. Clin
diagnosing cobalamin and folate deficiencies.
Am J Med 1994;96:239-246. Chem 1998;44:2199-2201.
17 Zhang M, Gunter EW, Pfeiffer CM. Evaluation
4 Klee GG. Cobalamin and folate evaluation;
measurement of methylmalonic acid and of the Drew Scientific DS30 homocysteine
homocysteine vs vitamin B12 and folate. Clin assay in comparison with the Centers for
Chem 2000;239-246. Disease Control and Prevention reference
5 Refsum H, Ueland PM, Nygard O, Vollset SE. HPLC method. Clin Chem 2001;47:966-967.
Homocysteine and cardiovascular disease. Ann 18 Akgul EO, Cakir E, Ozcan O, Yaman H, Bilgi
Rev Med 1998;49:31-62. C, Erbil MK. A comparison of three high-
6 The Homocysteine Studies Collaboration. performance liquid chromatographic (HPLC)
Homocysteine and risk of ischemic heart methods for measurement of plasma total
disease and stroke: a meta-analysis. JAMA homocysteine. Turk J Med Sci 2005;35:289-
2002;288:2015-2022. 295.
7 Ray JG, Laskin CA. Folic acid and 19 Martin SC, Tsakas-Ampatzis I, Bartlett WA,
homocyst(e)ine metabolic defects and the risk Jones AF. Measurement of plasma total
of placental abruption, pre-eclampsia and homocysteine by HPLC with coulometric
spontaneous pregnancy loss: a systematic detection. Clin Chem 1999;45:150-152.
review. Placenta 1999;20:519-529. 20 Zhang M, Pfeiffer CM. Comparing the ESA
8 Vollset SE, Refsum H, Irgens LM, Emblem HPLC total homocysteine assay with
BM, Tverdal A, Gjessing HK. Plasma total electrochemical detection to the CDC in-house
homocysteine, pregnancy complications, and HPLC assay with fluorescence detection. Clin
adverse outcomes: the Hordaland Chim Acta 2004;340:195-200.
Homocysteine Study. Am J Clin Nutr 21 Kuhn J, Gotting C, Kleesiek K. Rapid micro-
2000;71:962-968. scale assay for homocysteine by liquid
9 Nilsson K, Gustafson L, Faldt R, Andersson A, chromatography-tandem mass spectrometry.
Brattstrom L, Lindgren A. Clin Biochem 2005;39:164-166.
710
Homocysteine
22 Arndt T, Guessregen B, Hohl A, Heicke B. 37 Den Heijer M, Blom HJ, Gerrits WBJ,
Total plasma homocysteine measured by liquid Rosendaal FR, Haak HL, Wijermans PW, Bos
chromatography-tandem mass spectrometry GMJ. Is hyperhomocysteinaemia a risk factor
with use of 96-well plates. Clin Chem for recurrent venous thrombosis? Lancet
2004;50:755-757. 1995;345:882-885.
23 Kredich NM, Kendall HE, Spence FJ Jr. A 38 Rasmussen K, Moller J, Lyngbak M, Holm
sensitive radiochemical enzyme assay for S- Pedersen A-M, Dybkjaer L. Age and gender
adenosyl-L-homocysteine and L-homocysteine. specific reference intervals for total
Anal Biochem 1981;116:503-510. homocysteine and methylmalonic acid in
24 Shipchandler MT, Moore EG. Rapid, fully plasma before and after vitamin
automated measurement of plasma supplementation. Clin Chem 1996;42:630-636.
homocys(e)ine with the Abbott IMx analyzer. 39 Cruysberg JRM, Boers GH, Trijbels JMF,
Clin Chem 1995;41:991-994. Deutman AF. Delay in diagnosis of
25 Pernet P, Lasnier E, Vaubourdolle M. homocystinuria: retrospective study of
Evaluation of the AxSYM homocysteine assay consecutive patients. Br Med J 1996;313:1037-
and comparison with the IMx homocysteine 40.
assay. Clin Chem 2000;46:1140-1141. 40 Zee RYL, Mora S, Cheng S, Erlich HA,
26 Pfeiffer CM, Twite D, Shih J, Holets- Lindpainter K, Rifal N et al. Homocysteine,
McCormack SR, Gunter EW. Method 5,10-methyltetrahydrofolate reductase 677C>T
comparison for total plasma homocysteine polymorphism, nutrient intake, and incident
between the Abbott IMx analyser and an HPLC cardiovascular disease in 24 968 initially
assay with internal standardization. Clin Chem healthy women. Clin Chem 2007;53:845-851.
1999;45:152-153. 41 Selhub J, Jacques PF, Bostom AG, Wilson PW,
27 Frantzen F, Faaren AL, Alfheim I, Nordhei AK. Rosenberg IH. Relationship between plasma
Enzyme conversion immunoassay for homocysteine concentration and vitamin status
determining total homocysteine in plasma or in the Framingham study population. Impact of
serum. Clin Chem 1998;44:311-316. folic acid fortification. Public Health Rev
28 Tewari PC, Zhang B, Bluestein BI. Analytical 2000;28:117-145.
and clinical evaluation of the Bayer ADVIA 42 Ueland PM, Clarke R. Homocysteine and
Centaur homocysteine assay. Clin Chim Acta cardiovascular risk: considering the evidence in
2004;342:171-178. the context of study design, folate fortification,
29 Kellogg MD, Parker R, Ricupero A, Rifai N. and statistical power. Clin Chem 2007;53:807-
Evaluation of an enzymatic homocysteine assay 809.
for the Hitachi series chemistry analyzer. Cin 43 Arnesen E, Refsum H, Bonaa KH, Ueland PM,
Chim Acta 2005;354:117-122. Forde OH, Nordrehaug JE. Serum total
30 Dou C, Xia D, Zhang L, Chen X, Flores P, homocysteine and coronary heart disease. Int J
Datta A, Yuan C. Development of a novel Epidemol 1995;24:704-709.
enzymatic cycling assay for total homocysteine. 44 Nygard O, Nordrehaug JE, Refsum H, Ueland
Clin Chem 2005;51:1987-1989. PM, Farstad M, Vollset SE. Plasma
31 Tan Y, Tang L, Sun X, Zhang N, Han Q, Xu M homocysteine levels and mortality in patients
et al. Total-homocysteine enzymatic assay. Clin with coronary artery disease. N Engl J Med
Chem 2000;46:1686-1688. 1997;337:230-236.
32 Pfeiffer CM, Huff DL, Gunter E. Rapid and 45 Wald DS, Law M, Morris JK. Homocysteine
accurate HPLC assay for plasma total and cardiovascular disease: evidence on
homocysteine and cysteine in a clinical causality from a meta-analysis. BMJ 2003;32:1-
laboratory setting. Clin Chem 1999;45:290-292. 22
33 Fiskerstrand T, Refsum H, Kvalheim G, Ueland 46 Casas JP, Bautista LE, Smeeth L, Sharma P,
PM. Homocysteine and other thiols in plasma Hingorani AD. Homocysteine and stroke:
and urine: automated determination and sample evidence on a causal link from mendelian
stability. Clin Chem 1993;39:263-271. randomisation. Lancet 2005;365:224-232.
34 Andersson A, Isaksson A, Hultberg B. 47 Verhaar MC, Stroes E, Rabelink TJ, Folates and
Homocysteine export form erythrocytes and its cardiovascular disease. Arterioscler Thromb
implication for plasma sampling. Clin Chem Vasc Biol 2002;22:6-13.
1992;38:1311-1315. 48 Robinson K, Arheart K, Refsum H, Brattstrom
35 Woltersdorf WW, Bowron A, Day AP, Scott J, L, Boers G, Ueland P et al. Low circulating
Stansbie D. Abbott IMx homocysteine assay: folate and vitamin B6 concentrations: risk
significant interference by 3-deazaadenosine. factors for stroke, peripheral vascular disease,
Ann Clin Biochem 1999;36:533. and coronary artery disease. Circulation
36 Ueland PM, Refsum H, Stabler SP, Manilow 1998;97:437-43; Erratum 1999;99:983.
MR, Andersson A, Allen RH. Total 49 Folsom AR, Nieto FJ, McGovern PG, Tsai MY,
homocysteine in plasma or serum: methods and Malinow MR, Eckfeldt JH, Davis CE.
clinical applications. Clin Chem 1993;39:1764- Prospective study of coronary heart disease
1779. incidence in relation to fasting total
711
Homocysteine
homocysteine, related genetic polymorphisms, myocardial infarction. N Engl J Med

and B vitamins: the Atherosclerosis Risk in 2006;354:1578-1588.
Communities (ARIC) study. Circulation 53 Hortin GL, Seam N, Hoehn GT. Bound
1998;98:204-210. homocysteine, cysteine, and cysteinylglycine
50 Voutilainen S, Rissanaen TH, Virtanen J, Lakka distribution between albumin and globulins.
TA, Salomen JT. Low dietary folate intake is Clin Chem 2006;52:2258-2264.
associated with an excess incidence of coronary 54 Zinellu A, Zinellu E, Sotgia S, Formato M,
events: the Kuopio Ischemic Heart Disease Risk Cherchi GM, Deiana L, Carru C. Factors
Factor Study. Circulation 2001;103:2674-2680. affecting S-homocysteinylation of LDL
51 Tool JF, Manilow MR, Chambless LE, Spence apoprotein B. Clin Chem 2006;52:2054-2059.
JD, Pettigrew LC, Howard VJ et al. Lowering 55 Clarke R, Woodhouse P, Ulvik A, Frost C,
homocysteine in patients with ischemic stroke Scherliker P, Refsum H et al. Variability and
to prevent recurrent stroke, myocardial determinants of total homocysteine
infarction and death: the Vitamin Intervention concentrations in plasma in an elderly
for Stoke Prevention (VISP) randomized population. Clin Chem 1998;44:102-107.
controlled trial. JAMA 2004;291:565-575. 56 McKinley MC, Strain JJ, McPartlin J, Scott JM,
52 Bonaa KH, Njolstad I, Ueland PM, Schirmer H, McNulty H. Plasma homocysteine is not subject
Tverdal A, Steigen T et al. Homocysteine to seasonal variation. Clin Chem 2001;47:1430-
lowering and cardiovascular events after acute 1436.
712
Homocysteine
Table 1: Homocysteine Summary Methods

Method 1: Amino acid analyzer for free homocysteine
Principle of analysis: Plasma or serum is deproteinated, and free amino acids are sequestered in the filtrate.
Either: a. Amino acids are separated by ion-exchange chromatography and detected by post-column derivatization with
Ninhydrin, yielding a mauve color for monobasic amino acids proportional to their concentrations.
Or: b. Amino acids are derivatized with fluorescent probe and the products separated by ion-exchange
chromatography. Fluorescence detection of individual amino acids proportional to their concentrations.
Comments: Used for profiling patients with homocystinuria; expensive to operate and limited capacity.
Analytical times are on the order of 1 to 3 hours or more.
Method 2: High-performance liquid chromatography of total reduced homocysteine
Principle of analysis: Plasma is treated with reducing agent and deproteinated. Free, reduced homocysteine and
other thiols are sequestered in the filtrate.
Either: a. Reduced thiols are derivatized to fluorescent compounds which are separated chromatographically. The amount
of homocysteine present is proportional to the fluorimetric signal.
Or: b. The compounds are separated chromatographically without derivatization and detected by electrochemical
oxidation.
Or: c. The compounds are separated chromatographically with or without derivatization and detected by tandem mass
spectrometry in SRM mode.
Comments: HPLC methods require a great deal of technical skill. In addition, throughput is limited. Method 2a
is the CDC reference method. Direct aspiration LC/MS/MS of derivatized amino acids (without chromatography)
is the method of choice for high-throughput newborn screening programs
Method 3: Enzymic product immunoassays of tHcy (fluorescence polarization or chemiluminescence)
Principle of analysis: All plasma forms of homocystine and mixed disulfide are reduced with dithiothreitol to
form reduced free homocysteine. This is converted enzymatically to S-adenosyl-homocysteine by S-adenosyl-
homocysteine hydrolase. The S-adenosyl-homocysteine produced competes directly with added S-adenosyl-
homocysteine-tracer conjugate for sites on a monoclonal antibody specific for S-adenosyl-homocysteine. The
amount of tracer bound is quantified by measuring either:
a. The fluorescence polarization of the resultant mixture. The amount of polarization is inversely proportional to
the plasma homocysteine concentration, or
b. Chemiluminescence after addition of trigger reagents. The luminescence is inversely proportional to the plasma
homocysteine concentration.
Comments: This assay is fairly simple to perform, is automated, and takes 30 to 60 min.
Method 4: Coupled enzymatic assays of tHcy for automated chemistry platforms
Principle of analysis: Homocystine, mixed disulfide, and protein-bound forms of homocysteine in the serum or
plasma are reduced with TECP or dithiothreitol to form the reduced free form of homocysteine. The total free
homocysteine is reacted in a series of coupled enzymatic reactions:
a. With serine and CBS produces cystathionine; cystathionine with CBL produces HCY and pyruvate; pyruvate
with LDH; monitored at 340 nm
b. With SAM and HMT produces SAH and methionine; SAH with SAHH produces HCY and adenosine;
adenosine with ADA produces inosine and NH3; NH3 reacted with GLDH; monitored at 340 nm
c. rHCYase with Vit B6 produces H2S; H2S reacted with DBPDA and ferricyanide; fluorescence detection
Comments: Automated, simple to perform and less expensive than immunoassays; reaction times vary from
about 10 to 20 minutes. Methods a and b available for many platforms.
713
Homocysteine
Figure 1
Diagram of homocysteine metabolism showing the interrelationship of the methionine and folate cycles with
methylation
CBS, Cystathionine--synthase; MAT, methionine adenosyl transferase; MS, methionine synthase; MTHFR, methylene
tetrahydrofolate reductase; SAH, S-adenosyl homocysteine; SAHH, S-adenosyl homocysteine hydrolase; SAM, S-adenosyl
methionine.
714
Homocysteine
Table 2: Performance of Analytical Methods in External Quality Assurance Programs

CAP 2006 - 7 Spot Challenges
%
Method of Labs Individual Specimen Mean in mol/L and (CV %)
Specimen HMS-04 HMS-05 HMS-06 HMS-01 HMS-02 HMS-03 Mean Value CV %
Chemiluminescence 48.8 4.15 (16.3) 48.68 (18.3) 10.13 (15.5) 3.71 (14.7) 8.95 (10.2) 9.01 (10.1) 14.1 14.2
Enzymatic 16.5 5.19 (21.2) 68.18 (13.1) 13.16 (6.4) 4.27 (15.0) 11.16 (7.6) 11.14 (7.5) 18.9 11.8
FPIA 28.7 5.81 (7.4) 70.65 (7.5) 13.15 (5.9) 5.4 (6.8) 11.59 (5.0) 11.73 (4.6) 19.8 6.2
HPLC All 2.0 --- --- --- 4.89 (14.7) 11.05 (10.9) 11.09 (11.3) ---
Others 4.0
Assigned Value* 5.7 75.6 14.2 4.7 11.6 11.6 20.5
All Results %CV n = 740 21.0 22.4 16.9 20.7 14.8 14.5 --- ---
NEQAS 2006 - 4 Spot Challenges

Individual Specimen Median (Range) in
Method % of Labs mol/L
HO
Specimen HO 06:04 HO 07:01 07:02 HO 07:03
2.6 (2.2 -
Chemiluminescence 6.1 15.5 (15.0 - 15.9) --- 2.9) 9.8 (7.4 - 12.2)
Enzymatic 15.2 16 (15.2 - 19.8) 6.7 (5.8 - 8.0) --- 12.3 (6.4 - 18.1)
4.0 (2.4 -
FPIA 54.4 14.4 (12.1 - 15.7) 6.2 (5.3 - 7.9) 4.3) 15.5 (14.5 - 17.4)
7.3 (5.2 - 6.1 (3.0 -
HPLC All 24.2 14.3 (6.1 - 18.3) 11.7) 133) 15.5 (13.6 - 17.8)
Median Value 14.7 6.3 4.0 15.3
All Results %CV n = 33 15.0 18.5 69.8 18.1

* Based on reference labs performing CDC assay [32].
715
Homocysteine
Figure 2
Fluorescence polarization immunoassay (FPIA) (Method 3a in Table 1) for the Abbott IMx and
AxSYM platforms.
No SAH
YYY YYY
YYY YYY in sample
High P
SAH in
sample
Low P
All forms of homocysteine are reduced with dithiothreitol, which is then converted enzymatically to S-
adenosyl-L-homocysteine (SAH). This product is quantified in a competitive reaction between it and
SAHfluorescein tracer conjugate for sites on a monoclonal antibody specific for the SAH moiety. The
amount of tracer bound is quantified by measuring the fluorescence polarization of the resultant
mixture, with this being inversely proportional to the free SAH concentration and hence inversely to
the tHcy in the specimen. Direction of applied and emitted light is shown by the yellow arrows.
Figure 3
Chemiluminescent magnetic microparticle immunoassay (Method 3b in Table 1).
Capture Ab-
SAH in Acridinium label
patient sample
Y
YY
Incubate All forms of homocysteine are first

reduced with dithiothreitol, which is then
converted enzymatically to S-adenosyl-L-
homocysteine (SAH). A competitive
reaction is set up in which patient-derived
SAH and SAH covalently linked to
Magnetic micro-
particles coated
Y paramagnetic particles compete for binding
sites on a monoclonal antibody specific for
the SAH moiety. Pre-trigger (acid) solution
with SAH releases the complexed SAH-labeled
detection antibody from the microparticles,
Magnetic which are drawn towards the magnet.
Trigger solution (sodium hydroxide)
Capture
optimizes the pH for the release of
chemiluminescence, which is captured and
quantified in the photometer.
Y Pre-trigger
Addition
RLU
Trigger Measurement
Addition
Y Y
716
Homocysteine
Appendix
Detailed Method for Total Homocysteine and 100g/L TCA (Trichloroacetic acid)
Cysteine published from the Centers for containing 1 mmol/L EDTA:
Disease Control (CDC) [32] Weigh 1.0 g TCA and 3.72 mg EDTA-Na2
(FW=372.2). QS to 10 mL with H2O.
Principle Ultrasonicate solution until EDTA is
Homocysteine, a thio-containing amino acid dissolved. Stable at room temperature.
derived from the metabolism of methionine, is
readily oxidized in body fluids to the disulfides 1.55 mol/L Sodium Hydroxide:
homocystine and cysteine-homocysteine. Weigh 6.2 g of NaOH and QS to 100 mL with
Reduction of homocystine and these disulfides, H2O. Stable at room temperature.
as well as the release of protein-bound
homocysteine with tris(2- 0.125 mol/L Sodium Borate Buffer (pH 9.5)
carboxyethyl)phosphine (TCEP) followed by containing 4 mmol/L EDTA:
trichloroacetic acid, allows this method to Weigh 4.767 g sodium borate decahydrate and
measure total plasma homocysteine and 148 mg EDTA-Na2 into a beaker. Add 90 mL
cysteine concentrations. After derivatization H2O. Ultrasonicate solution until crystals are
with a thiospecific fluorogenic reagent, dissolved. QS to 100 mL with H2O (pH should
ammonium 7-fluorbenzo-2-oxa-1,3-diazole-4- be 9.5). Stable at room temperature.
sulphonate (SBD-F), the thiols are separated
by isocratic reversed-phase high-performance 1 g/L SBD-F (ammonium 7-fluorbenzo-2-
liquid chromatography (HPLC) using oxa-1,3-diazole-4-sulphonate) in Sodium
fluorescence detection. A calibration curve is Borate Buffer:
carried through the derivatization procedure Wako Chemicals.
along with an internal standard, cysteamine. Prepare sufficient SBD-F solution for the assay
tubes (50 L is used per tube, 20 tubes = 1
Specimen mL). Weigh appropriate amount (mg) of SBD-
Whole blood collected into EDTA F into a light-protected (amber) vial. Add
anticoagulant is immediately placed in wet ice volume of sodium borate buffer needed to
for transport and centrifuged as soon as make a 1 g/L (1 mg/mL) solution. Vortex.
possible at 3000 rpm for 10 min. The plasma is Stock SBD-F is stored at 4C. Solution must
separated from the cells and refrigerated (2C be prepared fresh for derivatization and can be
to 8C) until analysis. stored at room temperature until used.
Reagents Internal Standard 40 mmol/L Cysteamine:

HPLC Mobile Phase: 0.1 mol/L acetic acid- Sigma Chemical Co.
acetate buffer, pH 5.0, containing 30 mL/L Weigh 4.54 mg cysteamine hydrochloride (FW
methanol. 113.61) into a vial. Add 1 mL MPB.
Mobile Phase Buffer (MPB): Refrigerate until used in assay. Prepare fresh
Weigh out 13.6 g sodium acetate trihydrate and on day of use. Not stable overnight.
QS to 1 L with H2O. Adjust the pH by addition
of glacial acetic acid to pH 5 (0.1). Stable for 1 0.1 mmol/L L-Homocysteine Stock:
month at room temperature. Sigma Chemical Co.
Working Mobile Phase: Weigh 1.352 mg homocysteine (FW 135.18)
Combine 970 mL of this acetate buffer with 30 into a vial. Add 10 mL of MPB. Refrigerate
mL methanol. Mix well. Stable at room until used in assay. Prepare fresh on day of
temperature. use. Not stable overnight.
Phosphate Buffered Saline (PBS): 0.8 mmol/L L-Cysteine Stock:

Sigma Chemical Co. Stable at room Sigma Chemical Co.
temperature. Weigh 12.6 mg cysteine hydrochloride (FW
157.62) into a vial. Add 10 mL of MPB.
100 g/L TCEP (tris(2- Refrigerate until used in assay. Prepare fresh
carboxyethyl)phosphine Hydrochloride: on day of use. Not stable overnight.
Pierce Chemical Co.
Weigh 1 g of TCEP hydrochloride and QS to Calibration Standards:
10 mL with H2O. Stable at room temperature. Prepare in thiol-free EDTA plasma, at
concentrations of 0, 5, 10, 25, and 50 mol/L
717
Homocysteine
homocysteine and 0, 80, 160, 400, and 800 Column oven temperature 29C
mol/L cysteine using the dilutions below: Excitation wavelength 385 nm, emission 515
0, 0 mol/L MPB only nm
5, 80 1 in 200 dilution of each
stock solution
stock solution
stock solution
stock solution
Add MPB to each standard so that the total
amount of MPB plus standard stock solution is
the same in each tube.
If a small mount of calibrator is needed, an
intermediate dilution of stock can be prepared
(e.g., 1 nmol/mL by a 100-fold dilution of the
0.1 mol/mL solution) in MPB Chromatogram showing separation of thiols
from a patient serum containing 12.5 mol/L
Control Samples: tHcy. (From Pfeiffer CM, Huff DL, Gunter E.
Commercially available materials e.g., Utak, Clin Chem 1999;45:290-2, with permission).
Chromsystems, Bio-Rad. Refer to package Cys, Cysteine; Hcy, homocysteine; Cys-Gly,
insert for reconstitution instructions and cysteinylglycine; GSH, glutathione; CysNH2,
stability. cysteamine internal standard.
Calculations:
Procedure:
Pipet 50 L patient, calibrator, or control On each patient chromatogram, five peaks of
plasma into polypropylene tubes. interest can be found within a 6-min analysis
time, as shown in the specimen chromatogram.
Add 25 L of internal standard and 25 L Identify the peaks corresponding to
PBS, followed by 10 L TCEP. Cap and homocysteine and cysteamine. Their
incubate for 30 min at room temperature. concentrations in plasma samples are
calculated using the internal standard method.
Add 90 L trichloroacetic acid / EDTA, vortex Calculate the ratio of homocysteine and
for 30 sec, and centrifuge for 10 min at 13000 cysteine peaks height (or area) to cysteamine
g. peak height (or area) for each standard. Plot
these ratios versus the concentration of the
Transfer 50 L of the supernatant to an standard, and calculate the linear least squares
autosampler vial containing 10 L NaOH, 125 regression equation of the relationship.
L borate buffer/EDTA and 50 L SBD-F Calculate the same ratios for each patient
solution. Cap and incubate for 60 min at 60C. sample and control. Extrapolate from standard
curve to calculate the homocysteine and
Derivatized samples may be stored at 20C cysteine concentration in each patient sample
up to 1 week until analysis by HPLC. and control.
HPLC Analysis: Reference Interval:

Total homocysteine < 13.0 mol/L for adults
Waters Alliance Model 2690 HPLC pump and < 10.0 mol/L for children
Autosampler WISP 710B with 10 L injection Critical high levels of homocysteine are >100
loop mol/L.
Waters Model 474 Scanning Fluorescence Total cysteine < 360 mol/L for adults.
Detector
Altech Associates Adsorbosphere C18 pre- Analytical Measurement Range:
column Total homocysteine 1.6 to 50 mol/L; total
Phenomenex Prodigy ODS2 analytical column cysteine 1.6 to 800 mol/L
(5 m particle size, 3.2 mm 150 mm) Patient samples exceeding the calibration
Millennium32 software range are repeated using a smaller volume of
plasma diluted into PBS.
Column flow rate 0.7 mL/min
718
Homovanillic Acid
Homovanillic Acid
Lawrence A. Kaplan
Name: Homovanillic acid, 3-methoxy-4-hydroxyphenylacetic acid, HVA
Clinical significance: Refer to Chapter 53, Neoplasia, in the 5th Edition of Clinical Chemistry: Theory,
Merck Index: 638
Chemical class: Substituted aromatic carboxylic acid (catecholamine metabolite)
Structure: see Figure 3
i With sensitivities around 0.5 mol/L, these assays could be

used to detect neuroblastomas.
High-performance liquid chromatography (HPLC) for the
measurement of urinary HVA (Table 1, Method 1) usually Although there are no reports or surveys comparing the
requires extraction from the biological sample. Usually the frequency of use of the methods employed for the
biogenic amines are first separated by anion-exchange measurement of HVA, most recent literature citations
chromatography on XAD-X4, DEAE-cellulose, or an report using an HPLC method. The other historical
equivalent resin, and then HPLC analysis is performed [1- methods are reviewed below.
11]. In one report [12], diluted urine was employed as the
sample to measure hydroxyl-methoxy mandelic acid Historically, colorimetric analysis of HVA utilized the
(HMMA), also known as vanillylmandelic acid (VMA), 5- reaction of 1-nitroso-2-naphthol with biogenic amines
hydroxyindoleacetic acid (5-HIAA), and homovanillic acid (Table 1, Method 4). HVA formed a red chromogen,
(HVA) in an isocratic HPLC system with amperometric which was measured at 500 nm [22]. For preparation of a
detection. In all cases, final separation was accomplished sample blank, potassium thiocyanate was added to remove
by reversed-phase, ion-paired (C18) chromatography, and the color that resulted from the reaction of HVA [22]. An
improved colorimetric reaction was proposed by Knight
detection was accomplished by ultraviolet
and Hammond, who substituted 1-nitroso-2-naphthol-4-
spectrophotometry (see Figure 1) or amperometry. Many
sulfonic acid for the previously used 1-nitroso-2-naphthol
procedures allow the simultaneous measurement of HVA
[23]. The assay can also be performed after the partial
and VMA [2,3,5,6,7,9,10].
purification of HVA by anion-exchange chromatography.
Capillary electrophoresis methods (Table 1, Method 2)
Another approach to quantitation was the use of thin-layer
have been reported for HVA analysis [13-18]. These
chromatography (TLC) performed on silica gel. The
procedures were used to measure HVA and VMA using a
biogenic amines were measured by scanning reflectometry
fused-silica capillary and an ion-paired detergent.
at 400 nm (Table 1, Method 5). The compounds, which
Detection was by ultraviolet absorbance, either fixed at
were allowed to photo-oxidize for 2 days, developed a red
245 nm [15-17] or with use of a diode array [13,19] or by
color [24].
amperometry [18].
A gas chromatographic (GC) method was developed with
Competitive-binding enzyme immunoassays (EIA) for
p-hydroxyphenyl acetic acid used as an internal standard
HVA have been reported for measurement of urinary HVA
(Table 1, Method 6). The sample was extracted with ether
(Table 1, Method 3) [20,21]. These assays employ
and derivatized with TRI-SII/TBT (Pierce Chemical Co.,
monoclonal antibodies to achieve the desired specificity.
Rockford, IL) forming the silyl derivative. The compounds
were separated by gas chromatography with 3% OV-1 as
i
Homovanillic Acid the stationary phase, and the HVA was detected with a
Previous and current authors of this method: flame-ionization detector [25]. The HVA could also be
First edition: Not done measured by electron-capture detection [26].
Methods edition: Steven J. Soldin, Lawrence A. Kaplan
Second edition: Not updated Gas chromatographymass spectrometry was also used for
Third edition: Not updated quantitation of serum and urine levels (Table 1, Method 7)
Fourth edition: Steven J. Soldin, Lawrence A. Kaplan [27]. The HVA in the samples was first separated by use of
Fifth edition: Lawrence A. Kaplan Amberlite XAD-4 anion-exchange columns. The eluted
719
Homovanillic Acid
compounds were converted to their trifluoroacetyl with elevated HVA concentrations [23]. The use of the 4-
hexafluoroisopropanol ester derivatives, the derivatives sulfonic acid derivative improves the stability of the
were separated by GC, and the HVA derivative was chromogen formed upon reaction of HVA [23]. The 4-
detected by a mass spectrophotometer. The limit of sulfonate substrate does react with 5-HIAA and VMA.
detection was 2 mg/mL for plasma and 120 ng/mL for Interference from 5-HIAA is minimized because its
urine. reaction product has a significantly different absorbance
maximum. VMA interference is reduced by use of an
Reference and Preferred Methods extraction solvent (CHCl3) that minimizes VMA recovery.
There is no formal reference method for HVA analysis. After partial purification of HVA by ion-exchange
chromatography, the accuracy of the method is much
Because of its superior sensitivity and specificity, and improved, and small increases in HVA excretion can be
because of its ease of use, high-performance liquid detected.
chromatography (HPLC) has become the method of choice
for the measurement of HVA. The HPLC methods allow The TLC method is slow and requires specialized
separation of HVA within 20 to 30 min. An HPLC equipment. The GC methods do work well for HVA, but
method, with either ultraviolet or amperometric detector, is the need for derivatization and the relatively poor
the preferred method for HVA analysis. The procedures sensitivity has made them less likely to be adopted.
are flexible to run a few or many samples at a time.
However, because of its superior sensitivity, amperometric Specimen
detection is preferred to ultraviolet spectroscopy. The The preferred sample for HVA analysis is a 24-hour urine
HPLC-amperometric methods are similar to the HPLC collected with sufficient acid to maintain a pH of less than
analysis of urinary catecholamines employing 3 during the collection period. While a complete 24-hour
amperometric detection [1,8,10,28]. urine collection is recommended, shorter collection times
may be acceptable if the result is expressed per milligram
The ability to quantitate simultaneously VMA and HVA of creatinine. Usually 15 mL of 6 M HCl added to the
may not have much practical advantage. Urine HVA is collection vessel before collection of urine is sufficient to
rarely elevated in pheochromocytomas, whereas urinary maintain the proper pH. Specimens should be kept
VMA is almost always increased in neuroblastomas. A rise refrigerated during collection. If analysis is delayed,
in urinary HVA output is used to diagnose neuroblastomas. specimens may be frozen indefinitely. An investigation of
By concentrating the HPLC assay on the analysis of either the diurnal variation of urinary HVA secretion indicates
VMA or HVA, one can increase the speed of analysis for that a random urine sample may be satisfactory for the
the specific tumor that is being investigated. Analyzing for diagnosis of neuroblastomas, especially when the excretion
only VMA allows washout of the HPLC column as soon as rate is normalized for creatinine excretion (mg of HVA/g
the VMA peak is eluted, whereas assaying for only HVA of creatinine) [29]. However, these results should be
allows elution to start with a stronger mobile phase and validated in each laboratory.
HVA eluting more rapidly.
The use of filter paper to collect and store urine samples
The capillary electrophoresis assays require equipment and has been successfully demonstrated [30]. HVA was eluted
knowledge that may not be as commonly available as for from the filter paper with tartrate buffer, allowing both
the HPLC technique. Since there have been few reports HVA and creatinine to be measured on the urine sample.
using this technique, it is difficult to judge its efficacy.
Similarly, there have only been two reports of the use of Interferences
the competitive binding enzyme immunoassay, which Catecholamine metabolites, vanillin, vanillic acid, glyceryl
carries the burden of requiring a specialized monoclonal guaiacolate, and many other compounds are known to
antibody. The EIA does have the advantages of not interfere in colorimetric assays. The HPLC and capillary
requiring any extraction of the urine sample and the ability electrophoresis methods show little interference from other
to perform many analyses at once, either in an automated endogenous compounds or exogenous drugs. The
spectrophotometer or in a multi-well reaction plate. monoclonal EIA showed 0.5% cross-reactivity with VMA
However, unless research is being performed, the need for and < 8% cross-reactivity with other structurally related
the HVA assay is rather infrequent, obviating need for catecholamine metabolites.
analysis of many samples. In any case, neither assay seems
to have any advantage in terms of sensitivity or specificity Homovanillic Acid Reference Interval
over the HPLC assays. The age-dependent percentile values for urinary HVA, as
mg of HVA/24 h or mg of HVA/g of creatinine, are
The colorimetric methods are not specific for HVA. In shown in Table 3. There appears to be no difference in
addition, the reaction of 1-nitro-2-naphthol with HVA excretion rates between males and females.
results in the formation of an unstable chromogen. The
instability of this chromogen increases with increasing Seasonal variations in urinary levels of VMA and HVA
HVA concentration, resulting in a rapid fading of the color have been reported using samples collected on filter paper
720
Homovanillic Acid
and analyzed by HPLC [31]. Although statistically neuroblastomas, several nations (Austria, Canada, France,
detectable, the differences were not considered clinically Germany, Japan, and the United Kingdom) proposed
important. The fluctuations in the concentrations of VMA general population screening for this disease in newborns.
and HVA were more apparent in a small study that used The underlying belief was that once detected, the disease
random urine samples [32]. could be eliminated. The screening programs were beset
with both false-positive and false-negative results, and it
Interpretation was found that many neonatal neuroblastomas regress and
The most important endogenous catecholamines are disappear. Statistical analysis of the results of these
epinephrine, norepinephrine, and dopamine. The screening programs could not support the continuation of
catecholamines are produced primarily in the chromaffin screening [27,38]. In July 2006 the National Screening
cells of the adrenal medulla, the brain, and the sympathetic Committee policy on neuroblastoma screening reaffirmed
nervous system. Dopamine is primarily synthesized in the this conclusion [39].
central nervous system (CNS) and acts as a
neurotransmitter. This compound is metabolized to 3- Measurement of urinary HVA and VMA can be useful in
methoxy-4-hydroxyphenylacetic acid (homovanillic acid, the diagnosis of familial dysautonomia (Riley-Day
HVA) by monoamine oxidase (MAO) and catechol-O- syndrome). This neurological disorder, present with
methyl transferase (see Figure 3). Work in rhesus monkeys highest frequency among Ashkenazic (Eastern European)
suggests that HVA is not substantially further metabolized Jews, is noted for a defect in the metabolism of
in primates and is excreted only by the kidney [33]. catecholamines [40,41]. In this disease, the urinary
excretion of VMA (expressed as g of HVA or VMA per
In 1865, Virchow described a highly malignant tumor mg of creatinine) is decreased, while the excretion of HVA
appearing in infancy and childhood, which has today been is increased. The ratio of HVA to VMA excretion rates is
termed neuroblastoma. Neuroblastoma is the most thus greatly increased in children with familial
common (approximately one quarter of all cancers) type of dysautonomia. The laboratory analysis of urinary HVA
cancer in the first year of life, having an overall incidence and VMA can help to confirm the diagnosis of this
of 5% to 7% in all childhood malignancies [34]. Males are syndrome.
affected slightly more commonly than females [35]. As
many as 14 cases per million children from birth to 9 years HVA analysis of cerebrospinal fluid (CSF) has been used
of age are reported annually in the United States [35,36]. to assess a number of mental disorders, including drug
abuse [42], depression, and alcoholism [43], as well as
Neuroblastomas are tumors that secrete mainly the changes related to aging [44].
precursors of norepinephrine, that is, dopamine and its
metabolites. In recent studies [2,7] it has been shown that Homovanillic Acid Performance Goals
urinary VMA and HVA were the analytes of choice for the There are no published performance goals for the analysis
laboratory diagnosis of neuroblastoma, since urinary of HVA. However, data for the College of American
dopamine is somewhat less reliable. Pathologists (CAP) proficiency surveys provide some
guidance. For urine HVA measured by HPLC, the target
Measurement of HVA excretion was of no value in ranges (mean 3SDs) for acceptable proficiency results
identifying patients with pheochromocytoma but was of were:
great use in identifying patients with neuroblastoma (Table Mean (mg/L) SD, range (mg/L)
4). The age-related 100th percentile value was exceeded in 2.6 0.6, 2.42-2.78
14 out of 15 patients, and all neuroblastoma patients had 105 13.3, 39.9-144.9
HVA values exceeding the age-related 95th percentile
[2,11]. References
1 Soldin SJ, Lam G, Pollard A, Allen LC, Logan
An evaluation of 18 additional patients with neural crest AG. High-performance liquid chromatographic
tumors (15 with a neuroblastoma and three with a analysis of urinary catecholamines employing
pheochromocytoma), not included in Table 4, showed an amperometric detection: reference values and use
elevation of VMA above the 100th percentile in 17 of 18 in laboratory diagnosis of neural crest tumors.
patients, whereas with 1 patient, a normal value was Clin Biochem 1980;13:285-291.
recorded. In 13 of the 15 additional patients with 2 Soldin SJ, Hill JG. Liquid-chromatographic
neuroblastoma, HVA was elevated above the 100th analysis for urinary 4-hydroxy-3-
percentile. In one case, the HVA value lay between the methoxymandelic acid and 4-hydroxy-3-
95th and 100th percentile, and in the other a value less methoxyphenylacetic acid and its use in the
than the 95th percentile was recorded. The 3 additional investigation of neural crest tumors. Clin Chem
patients with pheochromocytoma had normal HVA values. 1981;27:502-503.
3 Yoshida A, Sakai T, Tamura Z. Simple method
Because the quantitation of urinary HVA afforded a for the determination of homovanillic acid and
reasonable, simple, and inexpensive way to diagnose vanillylmandelic acid in urine by high-
721
Homovanillic Acid
performance liquid chromatography. J chromatography. J Chromatogr B 1997;693:463-

Chromatogr 1982;227:162-167. 467.
4 Krstulovic AM, Zakaria M, Lohse L, Bertani- 16 Shirao MK, Suzuki S, Kobayashi J, Nakazawa H,
Dziedzic L. Diagnosis of neural crest tumors by Mochizuki E. Analysis of creatinine,
reversed-phase high-performance liquid vanilmandelic acid, homovanillic acid and uric
chromatographic determination of urinary acid in urine by micellar electrokinetic
catecholamine metabolites. J Chromatogr chromatography. J Chromatogr B 1997;693:463-
1979;186:733-748. 467.
5 Frattini P, Santagostino G, Schinelli S, Cucchi 17 Garcia A, Heinanen M, Jimenez LM, Barbas C.
ML, Corona GL. Assay of urinary vanilmandelic, Direct measurement of homovanillic,
homovanillic and 5-hydroxyindole acetic acids by vanillylmandelic and 5-hydroxyindoleacetic acids
liquid chromatography with electrochemical in urine by capillary electrophoresis. J
detection. J Pharmacol Methods 1983;10:193- Chromatogr A 2000;871:341-50.
198. 18 Li X, Jin W, Weng Q. Separation and
6 Yoshida A, Yoshioka M, Tanimura T, Tamura Z. determination of homovanillic acid and
Determination of vanilmandelic acid and vanillylmandelic acid by capillary electrophoresis
homovanillic acid in urine by high speed liquid Analytica Chimica Acta 2002;461:123-130.
chromatography. J Chromatogr 1976;116:204- 19 Kati B, Elgstoen P, Jellum E. Capillary
243. electrophoresis for diagnosis of metabolic
7 Yoshida A, Yoshioka M, Yamazaki T, Sakai T, disease. Electrophoresis 1997;18:1857-1860.
Tamura Z. Urinary levels of vanilmandelic acid 20 Taran F, Frobert Y, Crminon C, Grassi J,
and homovanillic acid determined by high-speed Olichon D, Mioskowski C, Pradelles P.
liquid chromatography. Clin Chim Acta Competitive enzyme immunoassay with
1976;73:315-320. monoclonal antibody for homovanillic acid
8 Morrisey JL, Shihabi ZK. Assay of 4-hydroxy-3- measurement in human urine samples. Clin Chem
methoxyphenylacetic (homovanillic) acid by 1997;43:363-368.
liquid chromatography with electrochemical 21 Shi RZ, Ho YP, Yeung JH, Or PM, To KK, Lau
detection. Clin Chem 1979;25:2045-2047. MW, Arumanayagam M. Development of an
9 Soldin SJ, Hill JG. Simultaneous liquid- enzyme-linked immunosorbent assay with
chromatographic analysis of 4-hydroxy-3- monoclonal antibody for quantification of
methoxymandelic acid and 4-hydroxy-3- homovanillic acid in human urine samples. Clin
methoxyphenylacetic acid in urine. Clin Chem Chem 1998;44:1674-9.
1980;26:291-294. 22 Goldenberg H. Specificity of the nitrosonaphthol
10 Baursfeld W, Diener U, Knoll E, Ratge D, Wisser reaction: detection of metanephrine and other
H. Determination of urinary vanilmandelic acid guaiacol derivatives [abstract]. Clin Chem
and homovanillic acid by high performance liquid 1967;13:698.
chromatography with amperometric detection. J 23 Knight JA, Hammond RE. Improved colorimetry
Clin Chem Clin Biochem 1982;20:217-220. of urinary 3-methoxy-4-hydroxyphenylacetic acid
11 Soldin SJ. High-performance liquid (homovanillic acid). Clin Chem 1977;23:2007-
chromatography: application in a childrens 2010.
hospital. Adv Chromatogr 1982;20:139-163. 24 Huck H, Dworzak E. Quantitative analysis of
12 Bonfigli AR, Coppa G, Testa R, Testa I, De Sio catecholamine and serotonin metabolites on thin-
G. Determination of vanillylmandelic, 5- layer plates. II. reflectance measurements after a
hydroxyindoleacetic and homovanillic acid in specific photochemical reaction. J Chromatogr
urine by isocratic liquid chromatography. Eur J 1972;74:303-310.
Clin Chem Clin Biochem. 1997;35:57-61. 25 Brewster MA, Berry DH, Moriarty M. Urinary 3-
13 Caslavaka J, Gassmann E, Thormann W. methoxy-4-hydroxyphenylacetic (homovanillic)
Modification of a tunable UV-visible capillary and 3-methoxy-4-hydroxymandelic
electrophoresis detector for simultaneous (vanillylmandelic) acids: gas-liquid
absorbance and fluorescence detection :profiling chromatographic methods and experience with 13
of body fluids for drugs and endogenous cases of neuroblastoma. Clin Chem
compounds. J Chromatogr A 1995;709:147-156. 1977;23:2247-2249.
14 Kati B, Elgstoen P, Jellum E. Capillary 26 Cahuhan J, Darbre A. Determination of
electrophoresis for diagnosis of metabolic homovanillic, isohomovanillic and
disease. Electrophoresis 1997;18:1857-1860. vanillylmandelic acids in human urine by means
15 Jensen JS, Harding S, Roulund H, Rogers of glass capillary gas-liquid chromatography with
IS, Emmett PM, Golding J et al. Analysis of temperature programmed electron-capture
creatinine, vanilmandelic acid, homovanillic acid detection. J Chromatogr 1980;183:391-401.
and uric acid in urine by micellar electrokinetic 27 Takahashi S, Yoshioka M, Yoshiue S, Tamura Z.
722
Homovanillic Acid
Mass fragmentographic determination of <http://www.cancer.gov/cancertopics/pdq/treatme

vanilmandelic acid, homovanillic acid and nt/neuroblastoma/healthprofessional>
isohomovanillic acid in human body fluids. J 36 Miller RW, Fraumeni JF, Hill J. A neuroblastoma
Chromatogr 1978;145:1-9. epidemiologic approach to its origin. Am J Dis
28 Frattini P, Santagostino G, Schinelli S, Cucchi Child 1968;115:253-261.
ML, Corona GL. Assay of urinary vanilmandelic, 37 Schilling FH, Spix C, Berthold F, Erttmann R,
homovanillic and 5-hydroxyindole acetic acids by Fehse N, Hero B et al. Neuroblastoma screening
liquid chromatography with electrochemical at one year of age. N Engl J Med 2002;346:1047-
detection. J Pharmacol Methods 1983;10:193- 53.
198. 38 Woods WG, Gao RN, Shuster JJ, Robison LL
29 Tuchman R, Robison LL, Maynard RC, Bernstein M, Weitzman S et al. Screening of
Ramnaraine ML, Krivit W. Assessment of the infants and mortality due to neuroblastoma. N
diurnal variations in urinary homovanillic and Engl J Med. 2002;346:1041-1046.
vanillylmandelic acid excretion for the diagnosis 39 National Library for Health. National Screening
and follow-up of patients with neuroblastoma. Committee Policy: Neuroblastoma Screening.
Clin Biochem 1985;18:176-179. Available at
30 Hanai J, Kawai T, Sato Y, Takasugi N, Nishi M, <http://www.library.nhs.uk/screening/ViewResou
Takeda T. Simple liquid-chromatographic rce.aspx?resID=57184>
measurement of vanillylmandelic acid and 40 Smith AA, Taylor T, Wortis SB. Abnormal
homovanillic acid in urine on filter paper for mass catecholamine metabolism in familial
screening of neuroblastoma in infants. Clin Chem dysautonomia. N Engl J Med 1963;268:705-707.
1987;33:2043-2046. 41 Geltzer AI, Gluck L, Talner NS, Polesky HF.
31 Nishi M, Miyake H, Takeda T, Takasugi N, Sato Familial dysautonomia. N Engl J Med
Y, Hanai J. Seasonal variations in 1964;271:436-440.
vanillylmandelic acid and homovanillic acid in 42 Roy A, Berman J, Williams R, Kuhn C. Higher
infants urine. Jpn J Clin Oncol 1987;17:151-155. levels of CSF homovanillic acid in recently
32 Nishi M, Miyake H, Takeda T, Yamashiro K, abstinent cocaine-dependent patients. Am J
Takasugi N, Hanai J, Kawai T. Fluctuation in the Psychiatry 2002;159:1053-1055.
concentrations of vanillylmandelic acid and 43 Sher L, Oquendo MA, Li S, Huang Y-y,
homovanillic acid in mass screening for Grunebaum MF, Burke AK et al. Lower CSF
neuroblastoma. Eur J Pediatr 1990;149:859-861. homovanillic acid levels in depressed patients
33 Miller AL, Keenan RW, Maas JW, Asch RH. with a history of alcoholism.
Disposition of homovanillic acid in the primate. Neuropsychopharmacology 2003;28:1712 -
Metab Brain Dis 1987;2:207-212. 1719.
34 Gurney JG, Severson RK, Davis S, Robinson LL. 44 Rapoport SI, Schapiro MB, May C. Reduced
Incidence of cancer in children in the United brain delivery of homovanillic acid to
States: sex-, race-, and 1-year age-specific rates cerebrospinal fluid during human aging. Arch
by histologic type. Cancer 1995;75:2186-2195. Neurol 2004;61:1721-4.
35 National Cancer Institute. Neuroblastoma
Treatment PDQ. Available at
723
Homovanillic Acid
Tables
Table 1: Methods of Homovanillic Acid (HVA) Analysis

Method 1: High-performance liquid chromatography (HPLC)
Principle of analysis: HVA is extracted with an ion-exchange resin and separated by reversed-phase
chromatography. Detection by amperometry.
Comments: Most common; preferred method
Method 2. Capillary electrophoresis
Principle of analysis: HVA is extracted with solvent and separated by electrophoresis. Detection by ultraviolet
spectrophotometry or amperometry.
Comments: Rarely reported method.
Method 3. Competitive-binding enzyme immunoassay (EIA)
Principle of analysis: Using monoclonal antibodies, EIA is performed directly on urine sample, with colorimetric
measurement of enzyme activity.
Comments: Rarely performed method.
Method 4. Colorimetric
Principle of analysis: Reaction of nitrosonaphthol to form blue compound with absorption maximum at 500 nm
Comments: Historical; lacks specificity
Principle of analysis: HVA is extracted and separated on silica gel; color developed by photooxidation, quantified
by reflectance spectrophotometry
Comments: Historical; slow, not suitable for clinical laboratory
Principle of analysis: HVA is extracted and the silyl derivative formed; detected by flame ionization
Comments: Rare; slow, research method
Method 7: Gas chromatographymass spectrometry (GC/MS)
Principle of analysis: HVA is extracted with an ion-exchange resin, converted to the trifluoroacetyl
hexafluoroisopropanol ester derivatives, and separated by gas chromatography with detection by mass spectrometry
Comments: Rare; slow, research method
Table 2: Assay Conditions for HPLC Analysis of HVA*

Parameter Values
Sample volume 4 mL
Fraction sample volume 0.1 (anion-exchange chromatography step)
0.2 (anion-exchange elution step)
Final concentration and reagents TrisHCl, pH 7.0, 9 mmol/L
EDTA, 0.09 mmol/L (anion-exchange
chromatography step)
NaCl, 0.2 mol/L (anion-exchange elution step)
Assay time Approximately 60 min
Sensitivity 0.02 g (at 0.05 AUFS)
AUFS, Absorbance units at full scale.

*For assay presented in this chapter; for simultaneous HPLC analysis of VMA and HVA.
724
Homovanillic Acid
Table 3: Percentile Reference Values for Urinary HVA Excretion

(Using Method Described in VMA Chapter method review)
HVA
95th 100th
Age n mg/24 hr (mol/24 h)*
01 48 2.8 (15.4) 3.5 (19.2)
24 34 4.7 (25.8) 6.7 (36.8)
59 20 5.4 (29.6) 5.7 (31.3)
1019 40 7.2 (39.5) 8.5 (46.7)
>19 56 8.3 (45.6) 10.3 (56.5)
mg/g of creatinine (mmol/nmol of creatinine)
01 37 32.6 (20.3) 76.9 (49.4)
24 49 22.0 (13.7) 58.8 (36.5)
59 79 15.1 (9.4) 33.2 (20.6)
1019 55 12.8 (7.9) 44.2 (27.4)
>19 56 7.6 (4.7) 9.4 (5.8)
*Analyses performed on timed urine samples obtained from patients under investigation for hypertension.
Analyses performed on untimed urine specimens obtained from hospital patients not suspected of having a neural crest tumor.
Table 4: HVA Excretion by Patients with Neural Crest Tumors

(Using Method Described in VMA Chapter, click here)
HVA
(mmol/mol
Age mg/24 (mol/24 mg/g of of
(years) Sex hour hour) creatinine creatinine)
Neuroblastoma
<1 M 36.8 (202) 1421 (882)
<1 F 14.2 (78) 408 (253)
<1 F 14.1 (77) 190 (118)
<1 F 15.1 (83) 158 (198)
1 F 41.2 (226) 496 (308)
1 F 9.4 (52) 68 (42)
1 M 10.0 (55) 57 (35)
2 F 21.6 (118) 140 (87)
2 M 6.4 (35) 90 (56)
3 M 70 (384) 341 (212)
3 M 96.5 (530) 200 (124)
3 M 42.2 (232) 313 (194)
3 M 61.7 (339) 371 (230)
12 M 31.1 (171) 55 (34)
Pheochromocytoma
11 M 2.2 (12) 2.1 (1.3)
12 M 2.0 (11) 9.4 (58)
17 M <2.0 (<11) <10.0 (<6.2)
8 F <2.0 (<11) 2.7 (1.7)
30 F 17.1 (94) 8.1 (5.0)
28 M <2.0 (<11) 1.7 (1.1)
45 M 4.7 (26) 5.3 (3.3)
43 F 1.2 (<11) 1.2 (0.75)
22 F <2.0 (<11) 2.6 (1.6)
19 M 4.0 (22) 2.9 (1.8)
33 F 8.5 (47) 7.9 (4.9)
56 F <2.0 (<11) 2.3 (1.4)
70 M 4.0 (22) 3.9 (2.4)
43 F 3.2 (18) 7.4 (4.6)
62 M <2.0 (<11) <7.0 (<4.3)
34 F 6.4 (35) 7.5 (4.7)
725
Homovanillic Acid
Figures
Homovanillic Acid: Figure 1
Absorption spectrum of homovanillic acid (HVA) in 1% acetic acid.
HPLC analysis of homovanillic acid (HVA) with ultraviolet spectrophotometric detection. A, 0.5 g HVA standard. B,
Unspiked patient sample. C, Patient sample spiked with 50 g of HVA.
726
Homovanillic Acid
Metabolism of dopamine to homovanillic acid. MAO, Monoamine oxidase; COMT, catechol-O-methyltransferase.
1. NaCl, 0.2 mol/L. Place 11.7 g of NaCl in a 1-L

Procedure: HPLC Methods for Urinary HVA Using volumetric flask. Dissolve salt in approximately 900 mL of
Electrochemical Detection distilled water, and bring to mark with additional distilled
The procedure for VMA (vanillymandelic acid) can be water. Stable 6 months at room temperature.
used for the quantitation of urinary HVA because this 2. 1% acetic acid, 0.18 mol/L. Dilute 10 mL of
method allows for the simultaneous measurement of VMA concentrated acetic acid to a final volume of 1 L with
and HVA in urine samples. The alternative method, distilled water. Stable 2 years at room temperature.
presented below, employs ultraviolet spectrometry for 3. Methanol. HPLC grade, filtered through a 0.22
detection and either isocratic or gradient elution of the m filter.
HVA (Homovanillic Acid Table: Assay Conditions for 4. DEAE-Sephadex G-25. Suspend dry resin in
HPLC Analysis). excess Tris buffer, and decant fines. Store in refrigerator
when not in use.
5. DEAE-Sephadex columns. Polypropylene
Procedure: Homovanillic Acid Analysis by HPLC and
columns consisting of a separate barrel, a plastic filter disk,
Spectrophotometric Detection
and a reservoir (total column volume, approximately 17
Lawrence A. Kaplan
mL) can be purchased from BioRad Labs. The suspension
Principle of DEAE-Sephadex is poured into the assembled column
This method involves a preliminary purification of the until the lower barrel of the column (approximately 3 mL)
urine by DEAE-Sephadex (G-25) column chromatography is filled with the beads. The columns should be washed
and subsequent separation and quantitation by reversed- with 5 mL of Tris-EDTA buffer. These columns are
phase HPLC with detection of the HVA peak by ultraviolet prepared fresh on the day of use.
spectrophotometry at 260 nm. 5. Mobile phase. Add 200 mL of HPLC-grade
methanol to 800 mL of 1% acetic acid. Mix, and filter
Reagents through an aqueous solvent filter (0.45 m). Degas for 15
Place 1.21 g of Tris and 37.2 mg of disodium min, putting the mobile phase under negative pressure.
ethylenediaminetetraacetate in a 1-L volumetric flask. 6. HVA stock standard, 1 g/L (5.49 mmol/L).
Dissolve the salts in approximately 900 mL of distilled Place 50 mg of HVA in a 50-mL volumetric flask.
water. Adjust the pH to 7.0 with HCl, and bring volume to Dissolve HVA in approximately 40 mL of 1% acetic acid,
mark with distilled water. Stable for 3 months at 4C to and bring volume to mark with additional 1% acetic acid.
8C. Stable 1 year at 4C to 8C.
727
Homovanillic Acid
7. HVA working standard, 50 mg/L (0.27 Calculation

mmol/L). Dilute 5 mL of HVA stock standard to a final % recovery: The extraction efficiency for each
volume of 100 mL with 1% acetic acid. Stable for 3 specimen is determined by calculation of the recovery of
months at 4C to 8C. the HVA spike. This equation can be used only when
Vsx = Vx.
Assay (see Table 2)
Equipment: HPLC system including an ultraviolet detector
Mstd (PHsx - PHx)1000 = % recovery
capable of reading at 254 or 260 nm, an injector (manual
or automated), two pumps, a recorder integrator, and a PHstd (Vx) Msp
solvent programmer. This method can be performed by use
where
of a C18 Bondapak from Waters, Inc. (Milford, MA), or
equivalent. 1000 = Factor to account for sample volume from
1. Add 4 mL of centrifuged urine to 36 mL of 0.01 extraction (10 mL) and for 100% recovery (10 100 =
M Tris0.1 mM EDTA solution. Adjust pH to 7.0
0.05 with 1 M NaOH. 1000)
2. Remove a 20-mL aliquot, and place in labeled test PHstd = Peak height of standard
tube (unspiked sample).
Mstd = Amount in micrograms of HVA standard injected
3. Add 50 L of HVA standard (1 mg/mL) to
remaining solution, and readjust pH to 7.0 (spiked PHx = Peak height of unknown
sample). Vx = Volume of unknown injected
4. Spiked and unspiked solutions are carefully PHsx = Peak height of spiked sample
applied to DEAE-Sephadex columns so as not to Vsx = Vx = Volume of spiked sample injected
disturb resin bed. (Use plastic columns with lower
Msp = Amount in micrograms of spike HVA added to
portion filled with resin. Before applying
samples, wash resin beds with a one-column urine
volume of Tris buffer.)
5. Allow samples to drain into column. Wash Urinary excretion
column with 10 mL of water, and discard eluate.
1. Mstd (PHx) x 10 x 100% = g of HVA/mL of urine
6. Place a clean test tube under column. Elute HVA
with 10 mL of 0.2 M NaCl, allowing eluate to PHstd (Vx) (% recovery) 2
drain into the clean test tube.
7. Inject 40 to 50 L of spiked and unspiked sample
into HPLC system. Homovanillic Acid: Figure 2 2. (g of HVA/mL of urine) TV (mL)
illustrates typical chromatograms for a standard = mg of HVA/TV, or
and a patient sample.
8. Chromatographic conditions: mobile phase (20% mg of HVA/24 hours
methanol in 1% acetic acid) is run through pump
A, 80% methanol in pump B. Run HPLC
isocratically at 100% pump A until peak HVA is 3. g of HVA/mL of urine = g of HVA/mg of creatinine
eluted, and then switch flow to the 80% methanol mg of creatinine/mL of urine
in pump B, and increase flow rate to 2.5 mL/min.
After approximately 5 more min, switch mobile where TV is total volume.
phase back to 20% methanol and reequilibrate for Note
5 min. Change flow rate to 1.0 mL/min, and insert
next sample. This procedure should quickly elute Alternatively, one could perform the analysis by running a
the more strongly retained compounds and linear gradient elution from 0% to 20% methanol in 1%
prevent interference during subsequent injections. acetic acid over 10 minutes, with 5 minutes for
Wavelength: 254 nm (absorbance units reequilibration at 1% acetic acid after washing the column
at full scale, 0.05) with 80% methanol to elute material retained on the
Flow rate: 1 mL/min column. This procedure is very useful to separate peaks
Chart speed: 0.5 cm/min that closely elute with the HVA peak and interfere in its
analysis.
728
Human Immunodeficiency Virus (HIV)

Patricia Slev
Name: Human Immunodeficiency Virus (HIV)

Clinical significance: Refer to Chapter 32, Viral Hepatitis: Diagnosis and Monitoring, in the 5th edition
Structure: See Figure below.
Courtesy of Bio-Rad Laboratories [1].
Principles of Analysis and Current Usage States [5]. A significant proportion remain undiagnosed
Acquired Immunodeficiency Syndrome (AIDS) was first and unaware they are infected [5,6]. This is unfortunate
recognized as a clinical entity in 1981. The syndrome was because with the advent of extensive and effective
associated with abnormalities in the immune response, treatment options such as highly active antiretroviral
particularly immunodeficiency, neuropathology, treatment (HAART), HIV has now become a treatable,
opportunistic infections, and rare forms of cancer. chronic disease. However, individuals who are unaware
Epidemiological studies suggested that the culprit was an they are infected cannot be treated promptly and
infectious agent that could be transmitted via intercourse effectively; they continue to spread HIV infection
or blood products [2]. It was not until 1983 that the virus disproportionately and unknowingly. Therefore,
responsible for AIDS was isolated from infected patients. diagnosing, screening, and monitoring HIV infection is
Originally named HTLV-III because it was a cytopathic T- more important than ever. Although technology has been
cell lymphotropic virus, further analysis of the biology of rapidly advancing to allow for earlier, improved detection
the virus revealed that although HTLV-III was a member and diagnosis of HIV infection, understanding the
of the lentivirus genus of the Retroviridae family of viruses advantages, limitations and appropriate use of each of the
and therefore in the same family as HTLV-I and HTLV-II, multitude of currently available assays is crucial.
it was significantly different from the HTLV viral entities.
The virus was renamed Human Immunodeficiency Virus, The two broad methodological categories that have been
or HIV [3]. developed for donor screening, diagnosing, and monitoring
Today it is estimated that a total of 39.5 million people HIV infection are serological and nucleic acid testing
worldwide are living with HIV, [4] and approximately 1 (NAT). Both serological and NAT assays are used to
million of these HIV-infected individuals are in the United screen for HIV infection in blood or organ donors.
However, serological methods have been primarily utilized
i
Human Immunodeficiency Virus (HIV) for diagnosing HIV infection, and molecular assays have
New method: been reserved for monitoring HIV disease progression.
5th edition: Patricia Slev Two exceptions to this general approach are a suspected
acute HIV infection and HIV diagnosis in infants. In
729
suspected acute HIV infections due to recent exposure, a

serological screen may be followed by a viral-load assay. As of October 2006, the FDA approved the APTIMA
In infants, guidelines suggest a combination of both RNA HIV-1 RNA Qualitative Assay Gen-Probe Incorporated
and DNA HIV NAT for diagnosis [7]. (San Diego, CA) as the first viral-load assay for diagnosis
Immunophenotyping or T-cell lymphocyte count assays of HIV [11]. This represents a significant departure from
based on flow cytometry technology have also been previous guidelines in that the APTIMA assay, a NAT
employed for monitoring HIV disease progression but will method, can now be used not only for monitoring but also
not be discussed in detail in this section. HIV genotyping for confirming HIV infection.
and phenotyping for the purpose of monitoring
development of resistance to available antiretroviral HIV assays used in the United States must be licensed by
treatments are also available but will be only briefly the FDA for the specific use, whether it is blood/organ
addressed in this chapter. screening, diagnosing, or monitoring HIV infection. For
example, although there are over 40 available ELISA
Regardless of the method employed, one problem that assays worldwide, only 10 are approved for use in the
must be taken into account is the ability of the test to United States. There are many different algorithms for
detect the diverse HIV strains that have been identified to diagnosing HIV infection, screening potential blood and
date. All assays, but particularly the serological screens, organ donors, and monitoring HIV disease progression
are optimized and can be categorized according to the worldwide, depending on the local availability of assays.
types and groups of HIV detected. Based on molecular This chapter will focus on FDA-approved methods
analysis of the envelope gene, HIV can be classified available in the Unites States, with an emphasis on
phylogenetically into two types: HIV-1 and HIV-2. Within laboratory diagnosis and disease-progression monitoring,
HIV-1 there are at least three notable groups: M (major), O rather than blood or organ-donor screening.
(outlier) and N (new). The majority of the HIV-1 strains
circulating in the United States are group M. Group M Antibody Tests
strains can be further categorized into subtypes or clades, Antibody tests for HIV can be categorized by clinical
A-J, based on sequence. Subtype B strains predominate in functionnamely, screening or confirmation. In terms of
the United States, Japan, Australia, and Western Europe. screening for HIV infection, both rapid immunoassays and
Subtype A is prevalent in Russia, as well as West and traditional EIAs are available. Rapid immunoassays have
Central Africa, while subtype C is most common in been developed in an effort to aid in the early
Southern and East Africa, India, and Nepal [8]. Group O identification of HIV-infected individuals. Early
strains were first identified in the early 1990s and are low identification can reduce viral spread by individuals who
in prevalence in the United States [9]. The N group was are not aware that they are infected and improve care for
discovered in the late 90s in Cameroon. HIV-2 is primarily those infected. As the name implies, rapid immunoassays
restricted to Western Africa and rare in the United States provide results quickly, usually less than 30 minutes, and
[10]. In summary, nongroup M strains and HIV 2 are rare are relatively inexpensive and do not require
in the United States and Europe. The ability to detect these instrumentation, although most are still performed in a
rare strains is limited and depends on the assay used. laboratory. Rapid immunoassay testing is recommended
for screening high-risk individuals in high-prevalence
Functionally, serological methods can be further classified areas and women in labor and delivery who have not had
according to whether they are utilized as a screen or as a access to antenatal care. As such, rapid immunoassays
confirmatory (supplemental) test for HIV diagnosis. have been employed in a variety of clinical settings,
Screening assays include rapid immunoassays (results including emergency departments, labor and delivery, and
available in less than 30 minutes) and the more traditional ambulatory clinical sites.
laboratory enzyme immunoassay (EIA) or enzyme-linked
immunosorbent assay (ELISA). Screening assays are In the last few years, there has been a marked increase in
designed to detect individuals potentially infected with the number of rapid immunoassays available. The first
HIV. Although these assays are very sensitive, follow-up rapid immunoassay was approved by the FDA in 2002
testing must be performed to confirm that the antibodies [12]. Currently there are six FDA-approved screening
detected in the screening assay are indeed HIV-specific. rapid immunoassays kits for diagnosis available in the
The most widely utilized confirmatory assay for HIV-1 is United States (Tables 2 and 3). Three of these are
the Western blot, although the indirect approved for point-of-care testing and therefore use
immunofluorescence assay (IFA) is also U.S. Food and unprocessed specimens such as whole blood and oral fluid
Drug Administration (FDA) approved for this purpose. and have waived status under the Clinical Laboratory
Once HIV infection is confirmed in an individual, some Improvement Amendments of 1988 (CLIA-88) [13]. The
form of NAT is recommended to establish a baseline viral specifics of each of these tests are beyond the scope of this
load in the patient. This can be used to predict disease discussion, but rapid immunoassays all have certain
progression and monitor response to antiretroviral characteristics in common. All require no instrumentation
treatment. and are interpreted visually. Periodic testing of both known
730
HIV-negative and HIV-positive specimens (controls) is HIV ELISA assays are qualitative (meaning
required under the following conditions: (1) new operator, presence/absence) and run in triplicate if the initial result is
(2) new kit shipment, (3) new kit lots, and (4) specified reactive.
periodic intervals. The most common sample type for rapid
immunoassays is serum or plasma, although the Antigen Tests
development of oral fluid as an acceptable sample type has EIA assays that detect HIV antigens, primarily p24, are
increased the popularity of rapid immunoassays. However, also available but are no longer used for diagnosis in the
oral fluid samples have slightly decreased sensitivity and United States; they are occasionally used for donor-
specificity when compared to whole blood samples using screening purposes.
the same kit [13]. The most common antigens tested are
HIV-1, gp 41, gp 120, and gp 160 (envelope proteins); Combination Antibody-Antigen Tests
occasionally tested are HIV-1 p24 (core antigen) and HIV- Fourth-generation HIV tests that detect both HIV antigens
2 specific gp 36 (envelope protein) (see Anatomy of HIV (p24) and anti-HIV antibody have been developed, but
on page 1). Sensitivity and specificity are quite good for none have been approved for use in the United States to
group-M subtypes but vary for group O and HIV-2 [12]. date.
Rapid immunoassays are screening tests that are growing
in popularity in both the United States and abroad. All rapid immunoassays and traditional EIAs (antibody,
antigen) are considered to be screening tests. Because the
Traditional EIA/ELISA still remains the most common purpose of screening assays is to detect any potentially
screening method used in the United States. Similar to infected individuals, these tests have very high sensitivity
rapid immunoassays, EIAs can be categorized based on the and good specificity but low positive-predictive value in
HIV strains they can detect. There are HIV-1- and HIV-2- populations with low HIV prevalence. Samples from
specific assays, as well as combination assays that detect noninfected individuals have been known to be repeatedly
both (Tables 4, 5, and 6). Anti-HIV antibodies develop reactive in screening assays for a variety of reasons,
approximately 2 to 12 weeks after HIV infection and including reaction with HLA antigens that are present
depending on the EIA assay used, may be detected as early owing to the manner in which the virus was propagated
as 2 to 6 weeks post-infection. and bacterial contamination due to generation of
recombinant proteins, both used in manufacturing the
Several generations of EIA or ELISA have been developed antigens for the EIA. Recent reports suggest that flu
since the original in 1985. First-generation assays relied on vaccination can also result in an occasional positive HIV
viral lysates, whereas second-generation assays were screen [15]. Therefore, algorithms for HIV testing in the
improved by the addition of synthetic peptide and United States require supplemental testing with a more
recombinant proteins as antigens but detected primarily specific test for confirmation of HIV-1 infection.
IgG. The third-generation assays, which are currently on
the market, developed the antigen sandwich technique. Confirmatory Antibody Tests
Third-generation EIA are considered the most sensitive, There are two commonly available methods in the United
because they detect all immunoglobulin isotypes, including States for confirmation of HIV-1 infection, namely
IgG and IgM. There are several third-generation-plus Western blot and Indirect Immunofluorescence Assay
assays that detect HIV-1 group O as well. (IFA) (Table 7). A Western blot is based on the principle
of electrophoresing proteins, which migrate according to
The principle for an antigen sandwich EIA, which is the their molecular weight in a sodium dodecyl sulfate (SDS)
most common third-generation assay format in the United polyacrylamide gel, and then blotting the results onto a
States is as follows. HIV-specific antigens are fixed onto a nitrocellulose strip. In this case, the bands on the
solid support. The patients sample, possibly containing nitrocellulose blot are discrete and correspond to different
the anti-HIV antibody, and then another HIV antigen HIV-1 proteins. From top to bottom, the following HIV-1
labeled with an enzyme are added in that order. The anti- bands are present: gp 160 (env), gp 120 (env), p65 (pol),
HIV antibody is sandwiched between two antigen p55 (pol), gp 41 (env), p 40 (gag), p3 (pol), p24 (gag) and
molecules, hence the name of the format. Subsequently, a p18 (gag) (see Anatomy of HIV on page 1) [1]. The HIV
chromogen/substrate is added, and the intensity of the antigen strip is subsequently incubated with patient
color change is measured spectrophotometrically. The sample, followed by the addition of an enzyme-labeled
optical density (OD) values obtained are directly (horseradish peroxidase or alkaline phosphatase)
proportional to the amount of anti-HIV antibody present in antihuman immunoglobulin. A relevant colorless substrate
the patient sample, and an OD cutoff > 1.00 is considered is subsequently added that produces an insoluble colored
reactive (positive) [14]. A recent development has been the band on the strip wherever an antigen-antibody complex is
FDA approval of the first automated HIV-1/O/2 EIA present. The band pattern and intensity are reported for
offered by Bayer (Tarrytown, NY)) on the ADVIA each patient. Thus each patient has a relatively unique
Centaur platform, using a microparticle Western-blot profile, depending on the type and amount of
chemiluminometric immunoassay format (Table 5). All antibody they have directed against different HIV antigens.
731
therefore generally referred to as viral-load tests. Viral-

There are three possible interpretations based on the load assays are critical for monitoring HIV infection in
Western-blot profile (band pattern and intensity): positive, response to therapy and predicting disease progression. As
indeterminate, and negative. The cardinal bands, which are stated previously, NAT was used almost exclusively for
of diagnostic significance and the first bands to become monitoring HIV infection or screening donors for potential
positive, are HIV-1 gp 41, gp 120, and gp 160, as well as HIV infection until 2006, when the FDA approved the
p24. The Centers for Disease Control and APTIMA HIV-1 RNA Qualitative Assay as the first viral-
Prevention/Association of State and Territorial Public load test that could be used to diagnose or confirm HIV
Health Laboratory Directors (CDC/ASTPHLD) criteria infection.
require that two of the three cardinal bands must be present
for the sample to be interpreted as positive. Presence of There are two basic formats for viral-load assays for HIV:
any bands on the blot that do not meet positive criteria, nucleic acid amplification and signal amplification. The
including nonviral bands alone, is interpreted as three nucleic acid amplification technologies are reverse-
indeterminate. For the blot to be considered negative, no transcriptase polymerase chain reaction (RT-PCR), nucleic
bands can be present [16]. Occasionally, Western blots acid sequence-based amplification (NASBA) and
exhibit a nonspecific staining (NSS) pattern, which is transcription-mediated amplification (TMA). Branched-
defined by a smear on the strip that obscures all or some of DNA amplification (b-DNA) is the only signal
the bands and is associated with heterophile antibodies amplification assay available. Although nucleic acid
[17]. This pattern is uncommon and rarely associated with amplification tests are the most complex and sensitive of
HIV infection. the nucleic acid technologies for viral detection, all NAT
technologies have wide and comparable dynamic ranges
Indeterminate Western-blot results are clinically (40 to > 1 million copies/mL). However, most formats
challenging because of the multifactorial causes. An have a standard and an ultrasensitive test version, which
indeterminate Western blot may be due to cross-reacting usually differ in their dynamic range. Generally, the
antibodies (e.g., influenza vaccine), advanced HIV (low qualitative or ultrasensitive versions of a method are
antibody titers), orrarelyinfection with HIV-2. The geared towards the low-end range of viral copy numbers
patient status and risk factors, as well as the specific and require initial centrifugation of the sample to
Western-blot profile, must be considered together. Any concentrate the virus. Because viral-load assays amplify
indeterminate Western-blot result should be followed up different regions of the HIV genome, like EIA assays, they
with a repeat blot 4 to 6 weeks later and another after 6 vary in their ability to detect different HIV strains,
months to rule out any potential seroconverters. There are including groups N, O, and HIV-2. Most are optimized for
a limited number of options for confirming HIV-2 HIV-1 group M, particularly subtype B detection, which
infection, and none is FDA approved. There is, however, a represents the most prevalent circulating strains in the
typical indeterminate pattern on an HIV-1 Western blot United States. Most importantly, although plasma viral
that an HIV-2 infection can produce, and it consists of the concentrations generally correlate between viral-load
following: presence of gag (p55, p24, or p17) plus pol assays, they can vary by at least twofold [19]. Therefore, it
(p66, p51, p32) in the absence of env (gp 160, gp 120, or is recommended that when patients are monitored, the
gp 41) [10]. same viral-load assay be routinely employed [20].
IFA is a qualitative confirmatory method for HIV-1 that PCR (Roche Molecular Systems, Pleasanton, CA) is the
requires HIV-1-infected T-lymphocytes and a fluorescent most established and well known technology. Briefly, viral
microscope. First, uninfected (control) and HIV-1-infected nucleic acid is isolated, and a discrete nucleic-acid portion
T-lymphocytes are fixed on a slide in two separate wells. is exponentially amplified using a pair of target-specific
Dilute patient sample is then added to both wells. primers and a thermostable polymerase, resulting in
Following a wash, the FITC (fluorescein isothiocyanate) million copies of the original sequence, or amplicons. In
anti-human IgG antibody conjugate is added. Subsequent the case of HIV, which is an RNA virus, complementary
to another wash, the results are visualized using a or c-DNA template must first be generated using the
fluorescent microscope [18]. This technique requires reverse transcriptase enzyme. In addition, for the HIV PCR
interpretation expertise and an expensive fluorescent assays (nonreal time) biotinylated primers, as well as
microscope but can be used to resolve indeterminate AmpErase (uracil-N-glycosylase), which functions to
Western blots. For specifically confirming HIV-2 prevent amplicon contamination, are used in the PCR
infection, research-use-only methods are available. reaction. Subsequent to the generation of multiple
amplicon copies, the amplicon is denatured to single-
Nucleic Acid Testing (NAT) strand copies and hybridized with target-specific
Nucleic acid testing for HIV is available in a variety of oligonucleotide probes. Unbound material is washed away,
formats (Table 8). All forms of NAT testing rely on and the Avidin enzymelabeled conjugate is allowed to
detecting HIV-1 RNA in the sample, and although a few bind to the biotin-labeled amplicon, which has been
are qualitative, the majority are quantitative and are hybridized with the target-specific oligonucleotide probes,
732
followed by the addition of the substrate. Detection is Genotyping and Phenotyping Assays
based on the color change that is read HIV drug resistance can be monitored by genotyping or
spectrophotometrically [21]. phenotyping assays. The genotyping assays are the most
common and usually based on sequencing of the HIV
There are at least three different versions of PCR-based reverse transcriptase and protease genes. Non-sequence-
HIV assays, all offered by Roche, that are variations that based genotyping assays that depend on line probe
differ in the degree of automation, specific equipment technology are also available. Phenotyping assays rely on
used, and sample preparation. The Amplicor HIV-1 genetic engineering. First, a recombinant virus which
Monitor Assay is the least automated; the most automated expresses the reverse transcriptase and protease genes
platform is AmpliPrep/COBAS (comprehensive amplified from the patient sample (via a vector) is
bioanalytical system) in which both sample extraction and generated and then tested in cell culture with varying
the PCR reaction are automated [22]. concentrations of the drug to be evaluated. Simultaneously,
wild-type virus is grown as a control. Resistance is
In May 2007, the FDA approved the first real-time PCR measured by determining the concentration of the drug that
assays for HIV viral loads: the Roche Cobas inhibits replication of the virus by 50% as compared to the
AmpliPrep/Cobas TaqMan HIV-1Test (Pleasanton, CA) wild-type. Assays that combine genotype and phenotype
and Abbott RealTime HIV-1 assay (Abbott Park, Illinois). database information (Virtual Phenotype Assay) are
Real-time PCR assay is a target amplification technique another option [20].
that allows for real-time detection of accumulating
product, as opposed to the traditional end-point PCR Reference and Preferred Methods
assays; it represents an important advance in current There are no formal reference methods for HIV testing.
technological methods. The preferred procedure
for HIV diagnostic testing has been screening a plasma or
NASBA and TMA are similar technologies. NASBA is serum sample with an EIA, followed by confirmatory
marketed as the Nuclisens assay by bioMerieux (Durham, testing with a Western blot. With the advent of the rapid
NC) and is an isothermal reaction useful for amplification immunoassays and particularly the ability to test oral fluid
of RNA targets. It employs two analyte-specific primers and fingerstick specimens for HIV antibodies, rapid
and three enzymes: reverse transcriptase, RNase H, and T7 immunoassays have gained popularity as screening assays.
RNA polymerase. In contrast, TMA offered by Gen-Probe It remains to be determined what impact the APTIMA HIV
(San Diego, CA) is also an isothermal reaction that can assay will have on HIV diagnostics as a screening and/or
amplify RNA or DNA and uses two analyte-specific supplemental test option. For monitoring progression of
primers and two enzymes, namely a T7 RNA polymerase HIV disease, the preferred viral-load method is a PCR-
and reverse transcriptase. In both cases, a double stranded based assay. It is also likely that the recently approved
c-DNA template that includes the T7-promoter sequence is Roche and Abbott real-time PCR-based viral-load assays
synthesized by the reverse transcriptase and the primer may revolutionize viral-load tests because of their rapidity
pair. Secondly, the DNA template is transcribed by the and automation.
RNA polymerase, which recognizes the T7 promoter
sequence to generate 100 to 1000 RNA copies. The newly Specimen
transcribed RNA amplicons reenter the cycle as templates Acceptable sample types approved by the FDA vary,
for the original reverse transcriptase reaction. Detection of depending on the screening or confirmatory assay
amplicons is achieved by acridinium esterlabeled DNA employed. Acceptable sample types for rapid
probes that are specific for the target and give off a immunoassays include plasma, serum, whole blood
chemiluminescent signal when they hybridize to the (venous or fingerstick), and oral fluid. For traditional EIA,
amplicon [22,23]. acceptable sample types include serum (cadaveric is
acceptable for certain assays), plasma, dried blood spots,
B-DNA technology is a signal amplification technology urine, and oral fluid. IFA-acceptable specimens include
that does not require RNA purification or nucleic acid serum, plasma, and dried blood spots. For the Western-blot
amplification. A plasma sample is first centrifuged to method, all IFA-approved sample types, plus urine and
concentrate the HIV virus, then the cells are lysed, and the oral fluid, are acceptable (Tables 1-6).
viral RNA is captured by HIV-specific probes coated onto
the microwell. Target probes that bind to different regions Plasma is the preferred sample type for viral-load assays
of the viral RNA, as well as to the preamplifier probes, are (Table 8). With viral-load assays, the choice of
then allowed to hybridize. Multiple copies of the alkaline anticoagulant is important. In particular, heparin must not
phosphataselabeled amplifier probe are subsequently be used for any PCR-based viral-load tests, although it is
added; upon binding and addition of the substrate, they acceptable for NASBA and bDNA technologies [25].
produce a chemiluminescent signal [24].
733
Interferences Once the diagnosis of HIV has been established, viral-load

As with any immunoassay, heterophile antibodies can be a assays are recommended for monitoring HIV disease
source of interference [26]. Heterophile antibodies may progression. The goal of successful HIV therapy is an
interfere not only with EIA but also with Western-blot undetectable HIV viral load. Although viral-load assays
interpretation by producing a nonspecific staining (NSS) correlate well, it is recommended that only one method be
pattern [17]. In addition, flu vaccination has been used for serial monitoring. A change in HIV viral load that
associated with false-positive EIA results (see above). In is > 0.5 log 10 copies/mL is clinically significant [20].
regard to NAT, substances that can inhibit PCR reactions
can be an issue. This is why an internal or external control Performance Goals
for each sample, depending on the assay type, is always an In general, performance characteristics for each assay,
integral component of any viral-load assay. Therefore, a such as specificity and sensitivity, are dictated by the
result of none detected on a viral-load assay is not manufacturers of the commercial assays in accordance
synonymous with negative for HIV infection. In fact, the with recommended guidelines of the licensing bodies for
goal of HIV antiretroviral therapy is to achieve a viral load the country in which the assay is utilized. The CLIA-88
that is below the limit of detection (using < 50 copies/mL require that laboratories performing non-waived HIV
as a cutoff) [27]. antibody testing participate in a proficiency testing (PT)
program. Since most of the HIV antibody testing methods
Replication-competent HIV-1 can persist in the resting T- are deemed at least moderately complex, with a only a
cells of these patients, although they have an couple of rapid immunoassay exceptions, most hospital
undetectable viral-load result. Nevertheless, untreated and reference laboratories participate in a PT program.
individuals rarely have undetectable HIV viral loads. Currently there are a number of active programs for HIV
antibody testing, including the one offered by the College
Interpretation of American Pathologists (CAP), which has the highest
Studies suggest that 25% to 45% of HIV-infected enrollment. Challenge samples are sent out periodically,
individuals are unaware they are infected, and the majority and a score of 80% must be achieved to receive a passing
of these are not diagnosed with HIV infection until they grade, as is typical with all analytes and the CAP PT
are within a year of clinical disease onset (AIDS) [5,6]. program. In addition, many of these programs have
Therefore, HIV screening is now recommended by the expanded to include rapid immunoassays, although a
CDC for all individuals 13 to 64 years of age in all routine proportion of the currently available rapid immunoassays
health care settings [28]. are CLIA waived.
Individuals are screened with either a rapid immunoassay A recent survey evaluating HIV-1 antibody testing
or EIA. Repeatedly reactive screening assay results must performance results showed that overall performance or
be confirmed with one of the available confirmatory tests, efficiency (TP+TN/all results) for HIV-1 EIA was 99%,
most commonly the Western blot [12,29]. Any repeatedly HIV-1 Western blot 99.5%, and HIV-1 IFA 98.7%. Only
reactive EIA (rapid or traditional) result that is not 4.3% utilized IFA for confirmation of HIV infection, and
confirmed by supplemental testing suggests a false- approximately 1.2% used the Gen-Probe APTIMA HIV-1
positive screen, as long as recent exposure and the RNA qualitative assay as a form of confirmation for
seroconversion window can be ruled out. Indeterminate repeatedly reactive EIA results [31]. The same program is
confirmatory test results should be repeated. According to no longer being offered for HIV-1 RNA determinations,
CDC guidelines, individuals with two indeterminate but the results from the February 2006 report indicate that
Western-blot results spanning at least 6 months are the performance for quantitative and qualitative tests
considered negative for HIV infection, provided recent combined was 96.5%; more specifically, the performance
exposure has been ruled out. Positive results for for quantitative was 96.3% and 100% for the qualitative
confirmatory testing should be followed up with a viral- tests [32].
load assay to establish baseline plasma HIV concentration.
Technically, the APTIMA HIV assay can now be used as a References
confirmatory test, although it has not yet been widely 1 Human Immunodeficiency Virus Type 1, Genetic
adopted for this purpose. Systems HIV-1 Western Blot [package insert].
Bio-Rad. Redmond, WA.2004
HIV diagnosis is particularly difficult in neonates that have 2 Fields BN, Knipe DM, Howley PM .
been exposed to the virus in utero. Passively acquired HIV Fundamental Virology. 3rd edition. New York:
specific antibodies due to transplacental transmission can Lippincott-Raven.1996.
be present in these infants until 18 months of age. HIV 3 Coffin J, Haase A, Levy J, Montagnier L,
infection should not be diagnosed in neonates based on Oroszlan S, Teich N et al. What to call the AIDS
serological assays but rather by a combination of viral-load virus? Nature 1986; 321: 10.
assays (HIV RNA and HIV DNA) [7,12]. 4 http://www.unaids.org/en/HIV_data/epi2006
734
5 Hammer SM. Management of Newly Diagnosed

HIV Infection. N Engl J Med 2006; 353: 1702-
1709. 21 AMPLICOR HIV-DNA Test, version 1.5RT-
6 Marks G, Crepaz N, Janssen RS. Estimating PCR [package insert], Roche Molecular Systems,
sexual transmission of HIV from persons aware Branchburg, NJ, 2006.
and unaware that they are infected with the virus 22 Junkind D. Molecular Testing for Infectious
in the USA. AIDS 2006; 20: 1447-1450. Disease. Science 2001; 94: 1553-1554.
7 Guidelines for the use of Antiretroviral Agents in 23 Chan AB, Fox JD. NASBA and other
Pediatric HIV infection. January 20, 2004, transcription based amplification methods for
http://aidsinfo.nih.gov. research and diagnostic microbiology. Reviews
8 http://www.avert.org/hivtypes.htm Med Microbiol 1999; 10: 185-196.
9 Lin HH, Gaschen BK, Collie M, El-Fishaway M, 24 VERSANT HIV-1 RNA 3.0 Assay (b-DNA)
Chen Z, Korber BT et al. Genetic characterization [package insert]. Tarrytown, NY. Bayer
of diverse HIV-1 strains in an immigrant HealthCare. 2005.
population living in NewYork City. J AIDS 2006; 25 Liegler TJ, Grant RM. Nucleic Acid-Based HIV-
41: 399-404. 1 Viral Load Assays. HIV InSite Knowledge
10 Centers for Disease Control and Prevention. Base UCSF chapter, May 2006.
Human immunodeficiency virus type 2. 1998. http://hivinsite.ucsf.edu/InSite.
http://www.cdc.gov/hiv/pubs/facts/hiv2.jsp. 26 Kricka LJ. Human anti-animal antibody
11 interferences in immunological assays. Clin
http://www.fda.gov/bbs/topics/NEWS/2006/NEW Chem 1999; 45: 942-956.
01479. 27 Hammer SM, Saag MS, Schechter M, Montaner
12 Branson BM. Point-of-care rapid tests for HIV JS, Schooley RT, Jacobsen DM et al. Treatment
antibodies, J Lab Med 2003;27: 288-295. for adult HIV infection: 2006 recommendations
13 Greenwald JL, Burstein GR, Pincus J, Branson B. of the International AIDS Society-USA panel.
A Rapid Review of Rapid HIV Antibody Tests. JAMA 2006 16;296 (7): 827-843.
Current Inf Dis Reps 2006, 8: 125-131. 28 Branson BM, Handsfield HH, Lampe, MA,
14 Constantine NT. HIV Antibody Assays. HIV Janssen RS, Taylor AW, Lyss B, et al. Revised
InSite Knowledge Base. UCSF. Chapter. May Recommendations for HIV Testing of Adults,
2006. http://hivinsite.ucsf.edu/InSite Adolescents, and Pregnant Women in Health-
15 Erickson, CP; McNiff, T, Klausner JD. Influenza Care Settings. MMWR September 2006;
Vaccination and False-Positive HIV Results. N 55(RR14); 1-17.
Engl J Med 2006; 354, 13. 29 Centers for Disease Control and Prevention.
16 Centers for Disease Control and Prevention. Notice to readers: protocols for confirmation of
Interpretation and use of the Western blot assay reactive rapid HIV tests. MMWR 2004; 53: 221-
for serodiagnosis of human immunodeficiency 222.
virus type 1 infections. MMWR1989; 38: 1-7. 30 New York State Department of Health. HIV
17 Willman JH, Martins TB, Jaskowski TD, Hill testing and diagnosis in infants and children. New
HrR, Litwin CM . Multiplex Analysis of York (NY): New York State Department of
heterophile antibodies in patients with Health; 2005 Feb. 13p.
indeterminate HIV immunoassay results. Am J 31 Centers for Disease Control and Prevention
Clin Pathol 2001; 115: 764-769. Model Performance Evaluation Program (MPEP).
18 Human Immunodeficiency Virus Type 2, Human Immunodeficiency Virus Type 1 (HIV-1)
Fluorognost HIV-2 IFA [package insert]. Bio- Antibody Testing. Report of Results for the
Rad , Redmond, WA. 2004. Performance Evaluation Survey Conducted
19 Schurman R, Descamps D, Weverling GS, Kaye During January 2007.
S, Tijnagel J, William I, et al. Multicenter 32 Centers for Disease Control and Prevention
comparison of three commercial methods for Model Performance Evaluation Program
quantification of human immunodeficiency virus Human Immunodeficiency Virus Type 1
type 1 RNA in plasma. J Clin Microbiol 1996; 34: ribonucleic Acid (RNA) Determinations. Report
3016-3022. Results for the Performance Evaluation Survey
20 Richman DD. Benefits and limitations of testing Conducted during February 2006 ).
for resistance to HIV drugs. J Antimicrob 33 http://www.fda.gov/cber - TABLES (accessed
Chemother 2004; 53: 555-557. June 2007)
735
Table 1: HIV methods summary

Method 1. Rapid Immunoassay
Principle: Detects antibodies against HIV
Usage: Screening for HIV infection
Comments: Many kits approved in the US for various sample types including whole
blood & oral fluid; increasing in popularity
Method 2. Enzyme Immunoassay / Enzyme Linked Immunosorbent Assay
( EIA/ELISA)
Principle: Detects antibodies against HIV or HIV antigens on asolid support
Usage: Screening for HIV infection
Comments: Many kits available, differ in the ability to detect HIV types, groups and subtypes
Method 3. Western Blot
Principle: Detect antibodies against HIV antigens electrophoresed on a strip
Usage: Confirming HIV infection
Comments: Most common confirmatory test for HIV infection
Method 4. Indirect Immunofluorescence Assay (IFA)
Principle: Detect antibodies against HIV antigens on infected cells
Usage: Confirming HIV infection
Comments: Less common confirmatory test for HIV infection
Method 5. Reverse Transcription Polymerase Chain Reaction (RT-PCR)
Principle: Determine HIV plasma concentration based on PCR nucleic acid amplification
Usage: Monitoring HIV infection
Comments: Common method for monitoring HIV infection
Method 6. Branched DNA (b-DNA)
Principle: Determine HIV plasma concentration based on nucleic acid signal amplification
Comments: Does not require nucleic acid extraction, purification
Method 7. Nucleic Amplication (NASBA)
Transcription Mediated Amplification (TMA)
Principle: Determine HIV plasma concentration based on non-PCR based nucleic acid amplification
Comments: Both technologies are similar in principle but use different enzymes
Table 2 - HIV-1 Rapid Immunoassays (Antibody)

Trade Name Format Sample Use Manufacturer Approval
Date
Murex SUDS HIV-1 Test Rapid Serum / Plasma Donor Murex Diagnostics, 5/22/1992
Immunoassay Screen Inc
Norcross, GA
US License 1152
Reveal Rapid HIV-1 Rapid Serum / Plasma Diagnostic MedMira 4/16/2003

Antibody Test Immunoassay Laboratories, Inc
Halifax, Nova Scotia
Canada B3S 1B3
Uni-Gold Recombigen Rapid Serum / Plasma / Diagnostic Trinity Biotech, plc 12/23/2003
HIV Immunoassay Whole Blood Bray Co., Wicklow
(venipuncture and Ireland
fingerstick)
736
Table 3 - HIV-1,2 Rapid Immunoassays (Antibody)

Date
Multispot HIV-1/HIV-2 Rapid Test Rapid Plasma / Serum Diagnostic Bio-Rad Laboratories 11/12/2004
Immunoassay Redmond, WA
OraQuick ADVANCE Rapid HIV- Rapid Whole Blood, Diagnostic OraSure Technologies 6/22/2004
1/2 Antibody Test Immunoassay Plasma, Oral Fluid Bethlehem, PA
18015-1360
HIV 1/2 STAT-PAK ASSAY Rapid Fingerstick & Diagnostic Chembio Diagnostic 5/25/2006
SURE CHECK HIV 1/2 ASSAY Immunoassay venous whole blood, Systems, Inc
serum, plasma Medford, NY
Table 4 - HIV-1 EIA Antibody Assays

Trade Name Format Sample Use Manufacturer Approval Date
HIVAB HIV-1 EIA EIA Serum / Donor Screen Abbott Laboratories 3/1/1985
Plasma Abbott Park, IL
US License 0043
Genetic Systems EIA Serum / Donor Screen Bio-Rad Laboratories Blood 6/29/1998
rLAV EIA Plasma Virus Division
Redmond, WA
US License 1109
Vironostika HIV-1 EIA Serum / Donor Screen bioMerieux, Inc 12/18/1987

Microelisa System Plasma Durham, NC 27712
US License 1624
Vironostika HIV-1 EIA Plasma / Diagnostic bioMerieux, Inc 6/6/2003

Plus O Microelisa Serum / Durham, NC 27712
System Dried Blood US License 1624
Spots
HIVAB HIV-1 EIA EIA Dried Blood Diagnostic Abbott Laboratories 4/22/1992
Spot
HIV-1 Urine EIA EIA Urine Screen Diagnostic Calypte Biomedical Corp 8/6/1996
Genetic Systems EIA Dried Blood Diagnostic Bio-Rad Laboratories Blood 6/29/1998
rLAV EIA Spot Virus Division
Vironostika HIV-1 EIA Dried Blood Diagnostic bioMerieux, Inc 4/11/1990

Microelisa System Spot Durham, NC 27712
Oral Fluid EIA Oral Fluid Diagnostic bioMerieux, Inc 12/23/1994

Vironostika HIV-1 Durham, NC 27712
Microelisa System
737
Table 5 - HIV-1,2 EIA Antibody Assays

Trade Format Sample Use Manufacturer Approval
Name Date
Abbott EIA Serum / Plasma / Donor Screen Abbott 2/14/1992
HIVAB Cadaveric Serum Laboratories
HIV-1/HIV- Abbott Park, IL
2 (rDNA) US License 0043
EIA
Genetic EIA Serum / Plasma / Donor Screen Bio-Rad 12/9/2000

Systems Cadaveric Serum Laboratories
HIV-1/HIV- Blood Virus
2 Peptide Division
EIA Redmond, WA
US License 1109
Genetic EIA Serum / Plasma / Donor Screen Bio-Rad 8/5/2003

Systems Cadaveric Serum Laboratories Inc
HIV-1/HIV- Redmond, WA
2 Plus O US License 1109
EIA
ADVIA Microparticle Plasma/Serum Diagnostic for Bayer Healthcare, 5/18/2006

Centaur Chemilumin- qualitative LLC
HIV 1/O/2 ometric determination of Tarrytown, NY
Enhanced Immunoassay antibodies to the
ReadyPack human
Reagents immunodeficiency
virus type 1,
including Group O,
and/or type 2 in
serum or plasma
Table 6 - HIV-2 EIA Antibody Assays

Date
Genetic Systems EIA Serum / Donor Bio-Rad Laboratories 4/25/1990
HIV-2 EIA Plasma Screen Blood
Virus Division
Redmond, WA
US License 1109
738
Table 7 - Confirmatory Assays

Date
Cambridge Biotech HIV-1 WB Serum / Donor Calypte Biomedical Corp 1/3/1991
Western Blot Kit Plasma Supplemental Berkeley, CA
US License 1207
Genetic Systems HIV-1 WB Serum / Donor Bio-Rad Laboratories 11/13/1998

Western Blot Plasma Supplemental Blood Virus Division
Fluorognost HIV-1 IFA IFA Serum / Donor Waldheim Pharmazeutika 2/5/1992

Plasma Supplemental GmbH
Vienna, Austria
US License 1150
Cambridge Biotech HIV-1 WB Urine Diagnostic Calypte Biomedical Corp 5/28/1998

Western Blot Kit Supplemental
Genetic Systems HIV-1 WB Dried Diagnostic Bio-Rad Laboratories 11/13/1998

Western Blot Blood Spot Supplemental Blood Virus Division
OraSure HIV-1 Western WB Oral Fluid Diagnostic Epitope, Inc 6/3/1996

Blot Kit Supplemental
Fluorognost HIV-1 IFA IFA Dried Diagnostic Waldheim Pharmazeutika 5/14/1996

Blood Spot Supplemental GmbH
739
Table 8 - HIV Nucleic Acid Testing (HIV-1 NAT)

Trade Format Sample Use Manufacturer Approval
Name(s) Date
Roche PCR Plasma Prognosis / Roche Molecular 3/2/1999
Amplicor Patient Systems, Inc
HIV-1 Management Branchburg
Monitor HIV-1 Viral Township, NJ
Test Load Assay
NucliSens NASBA Plasma Prognosis / bioMerieux, Inc 11/19/2001

HIV-1 QT Patient Durham, NC
Management 27712
HIV-1 Viral
Load Assay
COBAS PCR Plasma Donor Screen Roche Molecular 12/20/2002

Ampliscreen Expanded Systems, Inc
HIV-1 Test Indications For Pleasanton, CA
Use: Source 94566-0990
Plasma donors,
other living
donors, and
organ donors
Procleix HIV-1/HCV Plasma Donor Screen Gen-Probe 2/8/2002

Nucleic Acid Expanded San Diego, CA
Test (TMA) Indications For 92121
Use: Source
Plasma donors,
living organ
donors and
cadaveric
samples
UltraQual PCR Plasma Donor Screen National Genetics 9/18/2001

HIV-1 Institute
RT-PCR Los Angeles, CA
Assay 92121
Versant Signal Plasma Patient Bayer Corp 9/11/2002

HIV-1 amplification Monitoring Berkeley, CA
RNA 3.0 nucleic acid probe 94702
(bDNA)
All tables were adapted from the FDA CBER website [33].
740
Immunoelectrophoresis
Stanley S. Levinson
Name: Immunoelectrophoresis
Clinical significance: Refer to Chapter 53, Neoplasia, in the 5th edition of Clinical Chemistry:

i applications for immunoelectrophoresis, the greatest
Electrophoresis demand exists in the medical laboratory for the
Protein electrophoresis is based on the principle that detection and identification of monoclonal
proteins that are composed of amino acids show immunoglobulin gammopathies in serum and urine.
charges that depend on the properties of the amino Agarose is chosen because it is sufficiently porous to
acids, the configuration of the protein, and the pH of allow antibodies to enter the gel but sufficiently
the buffer and medium. The migration of proteins in dense to trap precipitated immune complexes within
an electric field is largely determined by the charge the gel. Antibodies cannot enter more tightly cross-
on the molecule and the pH of the electrophoretic linked media such as acrylamide. As a result, specific
medium. Different proteins will separate from one antibody identification of proteins after separation on
another in accordance with the molecular charge. In less porous media requires removal of the proteins
gels containing a cross-linked matrix, such as from the gel onto membranes (Western blotting),
acrylamide or starch, the separation may be which is more tedious but widely used in research
substantially influenced by the size of the molecules. laboratories.
Immunoelectrophoresis is usually conducted in
agarose gel that contains minimal cross-linking, so Precipitation of the complexes after reaction with
the charge on the molecule largely determines the specific antibodies is dependent on the antigens
migration distance. (proteins that are themselves most often antibodies)
reacting with antigen-specific antibodies (also
In clinical laboratories, agarose electrophoresis is proteins), which combine to form a lattice (immune
optimized for the identification of monoclonal complex). According to the Heidelberg principle, if
immunoglobulins. This is called high resolution the antibody concentration is held constant, the
protein electrophoresis (HRPE). All kits presently amount of precipitate will increase as the antigen
available for the clinical laboratory use such an concentration increases until a zone of equivalency is
optimized approach. reached. As the antigen concentration decreases
Immunoprecipitation below equivalency, there will be less precipitate
After electrophoretic separation, the proteins are (antigen and antibody will become increasingly
treated with specific antibodies. Except for capillary soluble). This phenomenon is called prezoning. Also,
zone electrophoresis, where there is no gel, the as the antigen increases above equivalency, solubility
resultant antigen-antibody complex(es) become will also increase (prozoning). As will be discussed
insoluble and precipitate in the gel. The rate of below, these processes may become important when
precipitation depends on the proportions of the interpreting the results of the immunoelectrophoresis,
reactants, temperature, salt concentration, and pH of since prozoning may cause fading of the precipitant
the solution. Proteins that have not reacted with the that must be recognized by the interpreter. Except for
specific antibodies are removed from the gel by capillary zone electrophoresis (immunosubtraction
washing. The antigen-antibody complex is best where the electrophoresis is performed in free
visualized after staining. Although there are other solution without a solid gel) (Table 1, Method 4),
after precipitation, the gels are washed to remove the
i nonspecific proteins that have not precipitated in the
Immunoelectrophoresis agarose, then the agarose gel is stained with a stain
Previous and current authors of this method: that binds proteins. The stain allows the specifically
First edition: Gayle Birkbeck precipitated protein to be better visualized.
Methods edition: Gayle B. Jackson
Second edition: Gayle B. Jackson Although immunoelectrophoretic techniques can be
Third edition: Steven C. Kazmierczak used to identify any protein, the most common use is
Fourth edition: Gayle B. Jackson for typing monoclonal immunoglobulins. Generally,
Fifth edition: Stanley S. Levinson
741
in clinical laboratories, four types of

immunoelectrophoresis have been used: Immunofixation Electrophoresis (IFE)
A brief history of this technique (Table 1, Method 3)
Electroimmunoassay can be found in Levinson and Keren [6]. The
Electroimmunoassay (EIA) has also been called the technique was first described for clinical laboratory
rocket or Laurell rocket [1,2] (Table 1, Method use in 1976 by Ritchie and Smith [7] and Cawley and
1) technique. In this method, the antibody is mixed associates [8] for identifying monoclonal
into the gel, and the specific protein migrates to the immunoglobulins in serum, urine, and cerebrospinal
point of equivalency, where it precipitates. The fluid. These papers illustrate exact methods well. The
precipitate appears as an arrowhead or rocket. major change in currently used methods is the
Known calibrators with known amounts of protein availability of the complete procedure as a kit.
are run in some lanes. Unknown amounts of protein
run in others. This technique quantifies the amount of As illustrated in Figure 1, replicate electrophoresis of
unknown protein by comparing the length of the samples is performed in multiple lanes. Following
rocket from the origin with the calibrators that are this, the fractionated proteins in the electrophoretic
graphed on a curve. Generally, this technique has lanes are precipitated (fixed) into the agarose by
been supplanted by more accurate and overlaying with antisera specific for each type of
methodologically sensitive two-site solid-phase immunoglobulin to be identified. After a wash step to
immunoassay techniques. Some kits are available for remove unfixed proteins, the strips are stained with a
quantifying the coagulation factors Protein C and protein stain and washed to remove the excess stain.
Protein S by this technique. The separated proteins precipitate in a pattern that is
easily related to the pattern on routine HRPEshown
Classical Immunoelectrophoresis on the strip on the left in Figure 1, designated serum
Immunoelectrophoresis (IEP) was first described by protein electrophoresis (SPE). IFE shows three to
Grabar and Williams [3]. (Table 1, Method 2). five times more sensitivity than routine HRPE,
Electrophoretic separation of replicate samples in because the antibodies that are bound to the selected
agarose gels is conducted in multiple lanes in the first proteins increase the overall protein mass that is
step. In the second step, immunoglobulin-specific stained [9]. In clinical laboratories, IFE is currently
antiserum is placed in preformed troughs cut parallel the predominant technique used for identifying
to each lane. The gels are allowed to incubate in a proteins of interest and is the recommended method
closed chamber for 24 to 48 hours, while the antisera for the identification of monoclonal proteins (M
and antigen diffuses into the gel. The proteins from proteins) [10].
the patients serum which were separated by
electrophoresis in the prior step migrate by diffusion The advantages of IFE over IEP are several-fold,
towards the antisera. A precipitate develops, fixing particularly with regard to interpretation.
the immunoglobulins in the form of an arc. The gel is With IEP, interpretation depends largely on the
then washed and stained. Examples of this technique skill of the examiner, and subtle distortions of the
can be found in Ritzmann and Lawrence [4] and Penn arcs may be overlooked, whereas with IFE, the
[5], as well as Gayle Jacksons description in abnormal proteins migrate as restricted bands that
previous editions of this textbook. line up with the bands seen on routine SPE,
allowing for simple direct comparisons, as seen in
Considerable skill is needed to interpret the results, Figure 1 for the IgG kappa paraprotein.
which are difficult to relate to routine protein IgM interference is eliminated. Monoclonal IgM
electrophoresis. A shorter, narrow arc more distant is particularly difficult to identify with IEP due to
from the trough and close to the antigen source an umbrella effect in which polyclonal IgG
indicates a decrease in antigen, while a longer, interferes with the identification of monoclonal
thicker band close to the trough indicates an IgM paraproteins. This effect occurs because the
increased antigen concentration. However, if too high smaller 7S IgG molecules diffuse faster than the
a concentration of protein is present, the precipitate larger 19 S IgM molecules towards the trough. As
may form very close to the trough or within the a result, the light chain antisera is entirely
trough. Also, prozoning may occur when the antigen complexed by the intact light chain bound to the
is present in high concentration. The presence of an polyclonal IgG prior to the arrival of the larger
abnormal immunoglobulin is inferred from monoclonal IgM. The light chain associated with
comparing the size, shape, and distortions of the arc the slower moving IgM does not react with the
with the control. The identification of small abnormal depleted light chain antisera, and it is not
components is particularly difficult. precipitated. As a result, the monoclonal IgM
742
appears to be polyclonal. When a monoclonal Following electrophoresis, immunotyping can be

IgM abnormality is suspected, it is necessary to accomplished by use of immunosubtraction that is
reduce the IgM by treating the serum with a performed by incubating the serum sample with
reducing agent such as mercaptoethanol prior to beads coupled to anti-gamma, alpha, mu, kappa, and
electrophoresis, which breaks the 19S molecule lambda antisera. Following incubation with each of
down to 7S. This allows it to migrate more the heavy- and light-chain antisera solid phase
rapidly and compete equally with the IgG. IFE is reagents, the supernatants are reanalyzed by CZE to
not affected by this phenomenon. determine which of the immunoglobulin fractions has
Biclonal gammopathies are identified with IFE as been removed.
two bands, one staining for kappa and the other
lambda. With IEP, an increase in the density of Although CZE is useful for high volume SPE, it is
both lambda and kappa arcs may be not recommended for paraprotein determinations on
misinterpreted as a polyclonal increase. urine or CSF analysis [11], and even paraproteins that
IEP is insensitive for picking up very small are identified by CZE in serum are generally typed by
monoclonal proteins, those less than 1 g/L, IFE.
because diffusion of the antigen causes the
concentration to be reduced. With IFE, the Reference and Preferred Methods and Specimens
antigen is concentrated into a narrow band. IFE is Although there is no definitive reference method, IFE
very sensitive, detecting proteins to (Table 1, Method 3) has been defined by the College
concentrations as low as 100 mg/L. This of American Pathologists (CAP) Expert Panel as the
sensitivity can be substantially increased with preferred method for definitive identification and
cerebrospinal fluid (CSF) and urine by typing of monoclonal proteins [10]. This expert panel
mechanically concentrating the fluid. was selected and sponsored by the American Society
Serum samples run in antigen excess are usually of Hematology, the American College of
still interpretable with IFE, whereas with IEP, Rheumatology, the Clinical Center of the National
prozoning may occur in the region of antibody Institutes of Health, and the Clinical Immunology
excess, causing a failure of arc formation. The Society. The panel discouraged the use of the older
reason for this is because comparison of the SPE IEP method (Table 1, Method 2). The panel noted
with the IFE pattern provides a simple check for that because of its greater sensitivity, IFE may be
identifying prozoning, as will be discussed below useful for identifying a very small monoclonal
under Interpretation. protein, even when agarose gel HREP is negative,
The turn around time for IFE is much more rapid especially for identifying monoclonal kappa and
(2 to 4 hours) than IEP, which requires a 24- to lambda free light chains [10]. An International
48-hour incubation for the diffusion step. Federation of Clinical Chemistry Committee on
plasma proteins and SIbioC Study Group on proteins
Capillary Zone Electrophoresis also recommended IFE as the choice for identifying
Capillary Zone Electrophoresis (CZE) (Table 1, monoclonal free light chains in urine [13].
Method 4) measures proteins by absorbance [11]. No
protein stains are needed for this technique. Specimen
Molecules are separated in free buffer in thin, fused- Serum: Serum samples should be screened by HRPE
silica capillaries, generally of 20 to 100 mm internal and those with paraproteins typed by IFE.
diameter, by the application of a very high voltage.
Generally, CZE gives results similar to SPE by CSF: IFE is also very sensitive for identifying
HRPE. The major advantage of this technique is that oligoclonal banding in CSF (see Interpretation
many samples can be run rapidly, so for high-volume below), although many laboratories now use
laboratories, it offers a quick alternative compared to isoelectric focusing techniques. CSF should be
agarose electrophoresis. However, the sensitivity is concentrated 100 for initial examination, although
somewhat greater so that fibrinogen that migrates in repeats with dilutions of the concentrate may be
the beta-region of the gel, and is present in some necessary.
serum samples at very low concentrations, sometimes
appears to be a monoclonal protein (about 3% to 5% Urine: Urine samples may be screened by HRPE, but
of the time), which requires rerunning the sample by IFE is more sensitive for identifying monoclonal free
HRPE. Nevertheless, with high volume, there are light chains in urine (Bence Jones Proteins), and IFE
great savings in being able to run so many samples should be applied to all suspect samples by screening
more rapidly. with kappa and lambda antisera for free and bound
immunoglobulins (see Interpretation below) [6,10]. It
743
is important to concentrate the urine 100 to 150 the appearance of low-concentration paraproteins shown
prior to HRPE and IFE so that very low concentration in Figure 2. Notice that in Figure 2, both SPE and IFE
Bence Jones protein (BJP) can be identified (see show substantial polyclonal (diffuse) staining near the
Interpretations below). It is also important to perform monoclonal IgG proteins. This indicates that there are
the testing on 24-hour collections. Studies have substantial amounts of polyclonal Ig. Most commonly,
shown that testing on 2 or 3 consecutive 24-hour these low concentrations of paraproteins are not
urines essentially rules out Bence Jones proteinuria associated with disease but are monoclonal gammopathies
(BJPuria) [12]. of unknown significance (MGUS). Visual inspection of
the SPE gels in Figure 2 indicates that the monoclonal
Two types of antisera are available for fixation of proteins are less than 10% of the total staining in the
light chains: antisera that react with free light chains gamma region or 1000 mg/L. Kyle and associate have
only (FLC-only) and antisera against both free light found that monoclonal proteins < 5 g/L will develop into
chains and intact immunoglobulins (light chains myeloma only about 6% of the time, < 10 g/L about 7%,
bound to heavy chains), often called antisera against < 15 g/L about 11%, < 20 g/L about 20%, < 25 g/L about
free and bound (F & B). The former is produced by 24%, and < 30 g/L about 34% of the time [14]. In some
affinity purification, adsorbing out light chain cases, it has been shown that these MGUS disappear with
specific antibodies that react with intact time [14,15]. In other cases, low concentration
immunoglobulins, leaving only those that react with monoclonal bands turn out to be due to myeloma upon
FLC in solution. It usually has less avidity than bone marrow biopsy, especially when they are > 20 g/L,
antisera against F & B. As such, it is more and there is little diffuse polyclonal staining associated
susceptible to prozoning (see Interpretation below). with them (hypogammaglobulinemia of the normal
Some preparations of antisera against FLC-only may immunoglobulins). Nevertheless, there is generally no
also show somewhat diffuse band patterns on IFE, advantage in using HRPE that identifies very tiny
but in recent years, most preparations produce monoclonal proteins in serum as compared to HRPE
distinct patterns. Due to its lower avidity, antisera methods that are not quite as sensitive, since the great
against free should not be used for initial screening majority of tiny bands will be MGUS.
but only in conjunction with antisera against F & B
as an adjunct for better definition when the BJP is Position of Migration and Multiple Monoclonal Bands
migrating close to an intact monoclonal protein The IFE in Figure 2 indicates that on the average, the
containing the same light chain type [13]. polyclonal IgA migrates most rapidly toward the anode,
since IgA is the most negative of the three major
Reference Intervals immunoglobulins, whereas the polyclonal 19S IgM
See Interpretation below. tetramers that are too large tend to stay near the origin. As
seen in Figure 2, of the three major immunoglobulin
Interpretation classes, polyclonal IgG migrates most towards the
Serum Electrophoresis cathode. Because of this more rapid migration, IgA M
For serum samples, current guidelines suggest screening proteins are often found in the beta region, at about the
for monoclonal proteins using HRPE followed by IFE for beta-2 position, and low level IgA monoclonals are
definitive identification [10]. Figure 1 illustrates a sometimes masked by beta-2-migrating serum proteins
classical picture for multiple myeloma, where SPE by such as transferrin. Nevertheless, because of the wide
HRPE show a paraprotein band (restriction migrating as a charge variability in the Fab region of the molecule, there
band) in the gamma region of the gel, and IFE indicates is substantial overlap for the migration position of any
that it is an IgG- monoclonal protein (M-protein). The single immunoglobulin. As a result, IgA M proteins may
SPE indicates there is very little immunoglobulin except be found in the more cathodal gamma region and IgG M
for the monoclonal band (clear gel on both sides of the proteins in the beta region. Some laboratories test SPE
band). The IFE also shows little or no polyclonal gels that show an elevation in the beta-2 fraction by
immunoglobulin (little diffuse staining on the IFE lanes). testing for / ratios by immunonephelometry or
In many cases of multiple myeloma, the uninvolved immunoturbidimetry; others test by running IFE using
immunoglobulins are very low because the normal plasma trivalent antisera against IgA, IgM, and IgG. Figure 3
cells are decreased, while the abnormal clone is shows a biclonal pattern with such a rapidly migrating
substantially increased (see the chapter on Bence Jones monoclonal IgG- in the beta region of the gel, as well as
Proteins). an IgG- running cathodally.
Classical Profile versus Monoclonal Gammopathies of As shown in Figure 3, in some cases two or more
Unknown Significance monoclonal bands are seen. IFE may show the bands are
The classical profile in Figure 1 can be contrasted with an intact monoclonal immunoglobulin and a free light
744
chain. Generally, the free light chain will not be seen on cause the band to appear as a donut with a faded
SPE, either because it is too small to show up on the less center, as shown in Figure 4, that will appear as a
sensitive SPE or because it is migrating in the beta region solid band with a 3 to 5 dilution, as shown.
and masked by normal beta-migrating proteins. In other Prozoning may be a particular problem with CSF and
cases, the bands turn out to represent two or more intact especially urine, because concentration of the sample
monoclonal proteins (so-called biclonal or triclonal (approximately 100) prior to electrophoresis may
gammopathies). Occasionally, one band turns out to be a result in concentrations of the M protein of the order
heavy chain without a light chain. of 60 g/L. In these cases, multiple dilutions with
repeat analysis may be necessary. These problems
When two intact M proteins are seen, and one band is will be discussed below.
kappa and the other lambda, as shown in Figure 3, the
pattern is a true biclonal gammopathy and indicates that Proteins Other Than Immunoglobulins
the M proteins originated from two different clones of B IFE is useful for identifying other proteins besides
cells. If the two bands are both IgG- or one is IgG- and monoclonals. For example, in some cases, samples
the other IgA-, or similarly both are lambda, it is unclear that are thought to be serum may contain some
that there are two abnormal clones, since it is more likely fibrinogen [16]. Figure 5 illustrates the identification
that cells from a single clone are class switching, or that of fibrinogen by IFE. We have found IFE useful for
an abnormal clone has given rise to a progeny that is identifying lysozyme in urine as well (see the chapter
producing a different class of immunoglobulin or a on Bence Jones proteinuria) [9]. If desired, exact
fragment, in which case both monoclonal proteins may identification can be made by IFE for the specific
not be from different clones. In these cases, the pattern proteins, as shown in Figure 5, for fibrinogen. If
may not represent a true biclonal gammopathy. necessary, highly purified antisera for many proteins
are available in laboratories performing
Multiple monoclonal bands may also be seen in cases of immunonephelometry or turbidimetry for protein
IgA and IgM monoclonal gammopathies. Secretory IgA is quantification, and these antisera can be used for IFE.
a dimer, and serum IgA is a monomer. A dysplastic cell
may produce both, so that two bands may be seen, both Urine Protein Electrophoresis
typing for the same light chain type. The faster migrating Figure 6 illustrates extreme prozoning that may occur
band is the monomer. IgM myeloma is very rare. Usually when examining very concentrated urine when a high
IgM monoclonal proteins are associated with level BJP is present. HRPE of urine (UPE) shows a
lymphoproliferative diseases such as Waldenstroms very large monoclonal is present. The initial IFE was
syndrome. Normally serum IgM is a pentamer, but performed on the undiluted 100 concentrate. A BJP
abnormal cells may produce monomers or even dimers, so is clearly apparent from analysis of the initial 100
that two or three IgM M-protein (always the same light- concentrate, since the dense area on UPE combined
chain type) bands may be seen. with extensive prozoning when fixed with F & B
kappa, but not lambda, implicates a -BJP [17]. For
IgD and IgE demonstration, the IFE was repeated with dilutions.
If a monoclonal protein is identified on serum HRPE With dilution, a discrete monoclonal band is seen.
and stains for kappa or lambda but not for IgG, IgA, Also, notice that antibody against kappa FLC-only
or IgM with IFE, it may be a monoclonal free light does not show a reaction with the 100 concentrate.
chain, or it may be an intact IgD or IgE. Under this This occurs because of its lower avidity. These
circumstance, testing for IgD and IgE should be problems will be discussed in more detail in the
performed. It is very unusual for a monoclonal free chapter dealing with Bence Jones proteins.
light chain to reach high enough concentrations in
serum to be identified by SPE, because most of the CSF Protein Electrophoresis
free light chain spills over into the urine. Moreover, if Multiple sclerosis (MS) is a chronic, inflammatory,
urine from the same patient is analyzed and a BJP is demyelinating disease that affects the CNS. Magnetic
identified, this does rule out an IgD or IgE, since IgD resonance imaging (MRI) of the brain and spine is
paraproteins frequently have an associated BJP. often used to evaluate individuals with suspected MS.
MRI shows areas of demyelination as bright signal
Prozoning on T2-weighted images or FLAIR (fluid attenuated
One difficulty with IFE is that when the M proteins inversion recovery) distributed in the white matter of
are at high concentration, they exhibit prozoning. the corpus callosum, optic nerves, brainstem, and
With serum proteins, this rarely causes a problem spinal cord [18]. Because MRI can reveal lesions that
because of the limited range of concentrations occurred previously but produced no clinical
encountered (about 1 g/L to about 6 g/L). It may symptoms, it can provide the evidence of chronicity
745
needed for a definite diagnosis of MS. When or polyclonal IgM. Since the CIC are very large, they
oligoclonal banding is found in CSF but not serum, are trapped in the gel and do not wash out. They
this indicates intrathecal synthesis (within the spinal appear to be a paraprotein on SPE. With IFE, they
theca) of immunoglobulins. This may provide remain trapped in the gel and may appear as shown in
supporting evidence of MS when the diagnosis is Figure 9, where each strip shows a band regardless of
uncertain. Oligoclonal banding is also associated with the fixation antibody. In some cases, the paraprotein
diseases such as neurosyphilis, acute bacterial or viral appears as a band on the strips treated with IgG, IgM,
meningitis, progressive multifocal kappa, and lambda but not IgA. This latter type of
leukoencephalopathy, subacute sclerosing profile is usually due to polyclonal IgM RF binding
panencephalitis, progressive rubella panencephalitis, IgG, whereas bands in all lanes are more often due to
polyneuritis, optic neuritis, trypanosomiasis, and monoclonal RF.
other infectious and autoimmune diseases.
Nevertheless, in the presence of MRI and clinical Components of the CIC are best determined by
evidence, it is especially helpful in identifying active treating the serum with 2-mercaptoethanol (see the
MS. Methods Section, G. Mercaptoethanol treatment).
This breaks the sulfhydryl bonds so that the RF loses
CSF samples are initially concentrated 80 to 100 for its affinity for the polyclonal IgG. Treatment with
analysis by HRPE and IFE. Oligoclonal banding mercaptoethanol may be for 5 hours or 24 hours at
implies multiple bands (often only two) seen in 37C. The IFE is repeated after treatment. The
serum and CSF which are from a few clones sample shown in Figure 9 required 24 hours of
containing both kappa and lambda Ig, but they are treatment because a small band was seen in all lanes
not monoclonal. Generally, oligoclonal banding is after an initial 5-hour treatment. Sometimes 12-hour
superimposed on a diffuse polyclonal pattern, as treatment will break the immunoglobulins into
shown in the HRPE of CSF in Figure 7. Notice that fragments, so it is best to treat first for 5 hours, and
in this case, the bands are more distinct on CSF then repeat with a 12-hour treatment if the pattern is
electrophoresis than on IFE using IgG antibody, not clear. Notice after treatment, it can be seen that
because the IFE, which shows greater sensitivity, is there is a small monoclonal IgG- and polyclonal
overloaded. If the sample used for IFE were diluted, IgG. The best interpretation is IgG- monoclonal
the bands would show up better. But in the case with RF activity complexed with polyclonal IgG.
shown, the gel was purposely overloaded for
demonstration purposes to show that the bands are Figure 10 shows a broad restriction on SPE that with
not monoclonal but exhibit characteristics of both IFE stains mildly for IgG and lambda but more
kappa and lambda (notice the bands stain for kappa strongly for IgM and kappa. After a 5-hour treatment
and lambda). In general practice, it is not necessary to with mercaptoethanol, a monoclonal IgM- band can
stain CSF for kappa and lambda but only for IgG. In be seen, while the IgG and lambda appear diffuse.
the sample shown in Figure 7, it would not be The best interpretation would be monoclonal IgM-
necessary to run the IFE, because oligoclonal RF that was complexed with polyclonal IgG. Notice
banding is seen on HRPE of CSF. In Figure 8, after treatment, there is a slight precipitate at the
oligoclonal banding is not seen on the HRPE of CSF, origin. This could have been removed by centrifuging
but they are seen on the more sensitive IFE after 10 the sample before IFE.
dilution of the 100 concentrate. The labor associated
with dilution and re-assay is probably the main Immunofixation Performance Goals and Quality
reason many laboratories have switched to isoelectric Control.
focusing for identifying oligoclonal bands. Because The single most important characteristic of correct
of its expense and because most laboratories analyze daily performance is that a sample with a monoclonal
very few CSF samples for oligoclonal banding, protein should show a sharp band when fixed with an
isoelectric focusing is only run in specialized appropriate specific antisera and that there is little
laboratories. nonspecific staining. Appropriate banding patterns
are shown in Figures 1 through 4. Diffuse staining
Immune Complex Interference (Figures 2 through 4) due to polyclonal
In some rare cases, circulating immune complexes immunoglobulins migrating near the monoclonal is
(CIC) may appear as paraproteins on agarose appropriate.
electrophoresis. The CIC consist of IgM or IgG
rheumatoid factors and very rarely IgA that are As with most testing, performance goals are
bound to the Fc portion of polyclonal IgG. The periodically assessed by proficiency testing. The
rheumatoid factors may be monoclonal IgG or IgM CAP provides testing assessments for monoclonal
746
gammopathies in serum and urine (Bence Jones 8 Cawley LP, Minard BJ, Tourtellotte, M,
Protein specimens) and for oligoclonal banding in Chelle C. Immunofixation electrophoretic
CSF. These require identification as to whether or not techniques applied to identification of
there is a monoclonal immunoglobulin and whether proteins in serum and cerebrospinal fluid.
or not there is oligoclonal banding. The type of Clin Chem 1976; 22:1262-1268.
monoclonal must also be determined. 9 Levinson SS, Elin RJ, Yam L. Light chain
proteinuria and lysozymuria in a patient with
Since the results of IFE are interpretive, exact control acute monocytic leukemia. Clin Chem 2002;
limits for various applications are not defined. Most 48: 1131-1132.
electrophoretic plates contain wells at the bottom of 10 Keren DF, Alexanian R, Goeken JA,
the plates for control sera that qualitatively indicates Gorevic PD, Kyle RA, MD, Tomar RH.
whether the reaction has occurred: a precipitant ring Guidelines for Clinical and Laboratory
indicates a positive reaction (Figure 1). Evaluation of Patients With Monoclonal
Gammopathies. Arch Pathol Lab Med 1999;
New lots of antisera should be tested against patient 123: 106-107.
samples with known monoclonal gammopathies or 11 Sheldon J, Riches P. Capillary
controls containing monoclonal proteins. The best electrophoresis for investigation of proteins
negative control is to test new lots against normal in biological fluids. J Clin Ligand Assay
sera, where only diffuse polyclonal patterns should 2004; 27: 227-233.
be seen. 12 Bridgden ML, Neal ED, McNeely MD,
Hoag GN. The optimum urine collections
References and monitoring of Bence-Jones proteinuria.
1 Laurell CB. Quantitative estimation of Am J Clin Pathol 1990; 93: 689-693.
proteins by electrophoresis in agarose gel 13 Graziani M, Merlini G, Petrini C; IFCC
containing antibodies. Anal Biochem 1966; Committee on Plasma Proteins; SIBioC
15: 45-52. Study Group on Proteins. Guidelines for the
2 Laurell CB. Composition and variation of analysis of Bence Jones protein. Clin Chem
the gel electrophoretic fractions of plasma, Lab Med. 2003; 41:338-346.
cerebrospinal fluid and urine. J Clin Lab 14 Kyle RA, Therneau TM, Rajkumar SV,
Invest 1972; 29 suppl 124: 71-82. Larson DR, Plevak MF, Offord JR et al.
3 Grabar P, Williams LA Jr: Mthode Prevalence of monoclonal gammopathy of
permettant ltude conjugue des proprits undetermined significance.N Engl J Med
lectrophortiques et immunochemiques 2002; 346: 564-569.
dun mlange de protines: application au 15 Strobel SL.Transient paraproteinemia: an
serum sanguine. Biochim Biophys Acta intriguing immunological anomaly. Ann
1953; 10:193-94. Clin Lab Sci 2003; 33: 265-270.
4 Ritzmann SE, Lawrence M. Qualitative 16 Qiu LL, Keeling KL, Levinson SS, Elin RJ.
Immunoelectrophoresis. In: Ritzmann SE, A Convenient and Effective Method for
Daniels JC, eds. Serum protein Removing Fibrinogen from Contaminated
abnormalities. Boston, MA: Little Brown, Serum Specimens for Protein
1975:44. Electrophoresis. Clin Chem 2003; 49: 868-
5 Penn G, Batya J. In: Interpretation of 872.
immunoelectrophoretic patterns. Chicago,IL: 17 Levinson SS. Urine protein electroporesis
Am Soc Clin Pathol, 1978. and immunofixation electrophoresis
6 Levinson SS, Keren DF. Free Light Chains supplement one another in characterizing
of Immunoglobulins: Clinical Laboratory proteinuria. Ann Clin Lab Sci 2000; 30: 79-
Analysis: Crit Rev Clin Chem 1994; 40: 84.
1869-1878. 18 Rudick RA. A 29-year old man with
7 Ritchie RF, Smith R. Immunofixation. III. multiple sclerosis. JAMA 1998; 280: 1432-
Application to the study of monoclonal 1439.
proteins. Clin Chem 1976;22:1982-1985.
747
Table 1: Immunoelectrophoresis Methods Summary
Method 1: Electroimmunodiffusion (EID) (Laurell rocket); quantification by size of immunoprecipitate in gel.

Principle of analysis: Electrophoresis of antigens into agarose gel containing monospecific antibodies for
the protein of interest; the length of gel pattern (rocket) from the origin is proportioned to antigen
concentration.
Comments: Used for special tests such as Protein C and Protein S.
Method 2: Immunoelectrophoresis (IEP); identification of monoclonal proteins through precipitin arc formation.
Principle of analysis: Proteins separated in agarose by zone electrophoresis; After electrophoresis,
antiserum is placed in a trough horizontal to the electrophoretogram and allowed to diffuse for 24 hours;
precipitin arcs form where antigen and antibody meet in the zone of equivalency; monoclonality can be
determined from shape and mobility of arc
Comments: No longer recommended for use in clinical laboratories.
Method 3: Immunofixation; identification of protein bands on zone electrophoresis by formation of precipitin bands
after overlaying the gel with highly purified antisera.
Principle of analysis: Proteins are separated by zone electrophoresis; area containing the protein to be
identified is overlain with a monospecific antibody, and its corresponding antigen precipitates.
Comments: Serum, urine, and CSF; can be used with the IFE to identify monoclonal proteins and
oligoclonal banding. Because of its greater speed of completion, parallel similarity of banding to SPE,
UPE, and CSF electrophoresis, this method lends itself to simplicity in interpretations and great sensitivity.
This is now the recommended method for SPE and UPE identification.
Method 4: Capillary electrophoresis and immunosubtraction
Principle of analysis: Protein fractions measured by absorbance following electrophoretic separation in a
small-bore capillary tube in free solution; following electrophoresis, serum sample is incubated with beads
coupled with anti-IgG, IgA, IgM, kappa, and lambda antibodies. The proteins of interest are captured by the
bead, and CZE is repeated. Disappearance of a peak is evidence of its type.
Comments: Although useful in laboratories that analyze very large quantities of samples for SPE, it has
not been perfected for immunoglobulin typing in either serum or urine, nor is there good evidence that it is
useful for analysis of BJP in urine. Typing after CZE is routinely performed by IFE.
Methods Overview such as Minicon with pore size that does not pass >
(generally obtained as a kit from a commercial 20,000 kD proteins. Most mechanical concentrators
source). concentrate to about 100. This usually takes 2 to 6
A. Ingredients hours. Try to avoid the sample becoming dry, since
1. Agarose gel, usually in buffer (tris-barbital much protein may be lost on the containers walls.
solution with preservatives (sodium azide). To achieve 150 concentration, the concentrator
2. Buffer pH 8.4 to 8.8. Often barbital and would have to be refilled with sample. In some cases,
sodium barbital, with 0.1%sodium azide as the urine may be very dense, so that even 100
preservative concentration cannot be easily achieved. Centrifugal
3. Stain: Usually amido black, Coomassie blue, concentrators are more rapid but will not concentrate
or acid blue stain in 5% glacial acetic acid. to 100 and are therefore not adequate. Dilution of
4. Protein fixative (precipitates all proteins) the concentrate with saline may be necessary if the
10% sulfosalicylic acid and 10% acetic acid. initial pattern is confusing (see Interpretations
5. Highly purified monospecific antisera to IgG, above).
IgA, IgM, F & B kappa and lambda, and FLC-only C. CSF specimen preparation
when needed for identification of BJP (see chapter on 1. Concentrate CSF to 80 to 100 for typing
BJP). These antisera are especially prepared for IFE, oligoclonal bands in CSF. Always perform IFE on a
lesser grade antisera will show artefacts and non- serum specimen from the same patient for
specific staining that may confuse the interpretation. comparison with the CSF IFE. Dilution of the
B. Urine specimen concentration concentrate with saline and repeat assay may be
1. For the detection of BJP, concentrate the necessary if the pattern appears overloaded (see
urine to 100 to 150 using stationary concentrators Interpretations above)
748
D. Electrophoresis step antisera with urines containing known kappa and

1. Samples are usually in the 1 to 10 L range lambda BJP that have been saved
and applied into wells or into the manufacturers frozen in aliquots. It is also worthwhile to test
template on HRPE. new FLC-only antisera with a normal serum sample
2. Prior to sample application, concentrate urine or control to be sure FLC-only antisera is not reacting
and CSF appropriately. with intact immunoglobulins (no reaction should be
3. Perform electrophoresis. seen, since there is so little FLC in normal serum).
E. Immunofixation G. Mercaptoethanol treatment
1. After the gel has been removed from the In some cases, after IFE, band restrictions will
electrophoresis chamber, align the antisera template be seen in the same position in multiple lanes. This
on the gel so that the slits in the template are aligned indicates that rheumatoid factor (RF) is present and
over the application area of the gel, ensuring good has bound IgG into immune complexes. The RF is
contact between the template and the agarose. In usually IgM, but may be IgG or very rarely IgA. RF
some cases, the antisera may be applied on filter is an endogenous antibody that binds to the Fc
paper by laying it on the gel. Most kits now use portion of endogenous IgG and forms immune
templates, since less antisera is used, and there is no complexes. These are so large that they are fixed in
diffusion onto adjacent strips. the gel prior to immunofixation, so they do not wash
2. Many gels contain control wells near the out during the wash procedure. Bands may appear in
bottom (see Figure 1). Apply an appropriate sample, all lanes or the IgG and IgM lanes and very rarely in
usually 1 to 10 L, into the control wells (or as the IgA and IgG lanes. The tip-off is restricted
specified by the manufacturer). The staining at the same position in both the kappa and
immunoglobulins in the control will react with the lambda lanes.
antisera applied in Step 4 so that a control reaction
should occur. To prove an immune complex is present and to
3. Usually there is a blotting step at this point. identify which immunoglobulins are involved, it is
4. With the gel slanted, apply onto the gel 1 or 2 necessary to break the sulfide bonds binding the
drops of the Protein Fixative to the appropriate lanes immunoglobulins together. This causes the RF to
and 1 or 2 drops of appropriate antisera to the loose its antibody activity and migrate separately.
appropriate lanes so that the drops flow quickly and The bonds are broken by treatment with the reducing
evenly down each lane. The protein fixative agent mercaptoethanol, and the IFE is repeated.
precipitates all of the proteins in the SPE, UPE, or
CSF protein lane, whereas the antisera fixes only the 1. Make a 1:10 dilution of 2-mercaptoethanol with
specific immunoglobulin or other specific protein isotonic saline. Make a mixture by adding 10 L of
(see Figures 1, 2, and 5). the 1:10 dilution to 100 L of the patients serum.
5. Incubate the gel in a closed incubation
chamber. 2. Allow the mixture to incubate in a tightly capped
6. At this point, the gel is usually washed with tube for 5 hours or overnight at 37C. Use the
saline. resulting serum directly for IFE. If a precipitate
7. Staining is now performed, after which appears, centrifuge and use the supernatant for IFE. If
excess stain is removed from the gel with glacial further storage is necessary, keep the mixture at
acetic acid washes. refrigerated temperature, but be sure to warm it well
F. Quality control and shake it before electrophoresis. (Usually an
1. For serum samples, the control wells at the overnight incubation is preferred, but in some cases,
bottom of the strip found with many kits help insure the immunoglobulins are destroyed, and a 5-hour
that the antisera is effective; a circular precipitate is incubation is required).
seen, indicating the reaction was appropriate (see
Figure 2). Although these controls contain intact light 3. If a precipitate appears after mercaptoethanol
chain, they usually do not contain FLC that are the treatment, centrifuge the tube to collect the
most important monoclonal protein assayed in urine. precipitate, and use the supernatant for
3. Therefore, before being put into use, it is electrophoresis. The immunoglobulins are still in
important to test each new batch of FLC-only solution.
749
Figures
Figure 1. Serum protein electrophoresis (SPE) and

IFE. SPE (channel to the far left) indicates the lane
containing the SPE. This lane was fixed to
precipitate all proteins. The remainder of the
channels to the right were fixed with specific
antisera for IgG, IgA, IgM, kappa, and lambda
serum proteins as indicated. Migration is towards the
anode. In the SPE lane, albumin is the fastest
migrating band. Bands in order of migration are:
albumin, alpha-1-globulin, beta-1-globulin, beta-2-
globulin, and gamma-globulin. Normal polyclonal
gamma globulin shows diffuse staining, as shown in
Figure 2. In Figure 1, a band restriction is shown.
This indicates a paraprotein. IFE shows the paraprotein is an IgG-kappa monoclonal protein. Note that very little or
no staining is seen in the IgA, IgM, and lambda lanes, indicating that these proteins are decreased (compare with the
polyclonal staining in Figure 2). Wells at the bottom of the gel are control wells; a precipitate in these wells ensures
that a reaction has occurred. Arrow pointing right (O) indicates the origin (placement of the sample). Arrow pointing
left indicates the position of the monoclonal band.
Figure 2. Low concentration paraproteins on

SPE and IFE. Arrows indicate the position of
the M proteins. The monoclonal protein in the
gel above is an IgG-kappa, whereas that
below is an IgG-lambda. Note the darker,
diffuse, polyclonal staining about the band, as
compared to Figure 1, indicating that there
are substantial amounts of normal
immunoglobulins, as well as monoclonals, in
the gels shown in Figure 2. For other details
see Figure 1.
Figure 3. Biclonal gammopathy. Arrows point to the two different monoclonal bands. Otherwise, details are similar
to Figure 1.
750
Figure 4. Prozoning in serum. The band may be shaped

like a donut. Note how the band becomes solid with a 5
dilution. Otherwise, details are similar to Figure 1.
Figure 5. IFE of fibrinogen. In some cases, fibrinogen is not completely

removed, or a plasma sample is thought to be serum (often a sample with
anticoagulant is pored into a red top [serum] tube). Contamination by
fibrinogen can be mistaken for a monoclonal protein on SPE. Fibrinogen
or other serum proteins can be identified by IFE. In the case shown, the
direction of migration is up, toward the anode. Arrow indicates the
fibrinogen band.
Figure 6. Extreme prozoning with urine electrophoresis. 100 concentrated urine was diluted as indicated for IFE. F
& B indicates antisera against free and bound light chain, and F indicates antisera against FLC-only. UPE was fixed
with total protein fixative. The number 100 along the bottom indicates the amount of mechanical concentration.
After concentration, samples were diluted with saline to lesser concentrates as indicated. Zero indicates
unconcentrated urine and 1:10 indicates a 10-fold dilution of the unconcentrated urine. For explanation, see the text.
For other details of the Figure, see Figure 1.
751
Figure 7. Oligoclonal banding in CSF is both kappa

and lambda. HRPE of serum and CSF with IFE is
shown. Oligoclonal banding is seen in CSF protein
electrophoresis and on the IFE fixed with IgG,
kappa, and lambda but not in the serum protein
electrophoresis. In this case, oligoclonal banding
was apparent from the CSF electrophoresis without
IFE. The IFE was performed to demonstrate the
bands are both kappa and lambda. Details of the gels
are the same as in Figure 1.
Figure 8. Oligoclonal banding in IFE but not

CSF protein electrophoresis. In this case, the
oligoclonal banding is not seen on routine CSF
electrophoresis but is seen on the more
sensitive IFE after diluting the 100
concentrate to 10. Details of the gels are
similar to Figure 1.
Figure 9. Interference by IgG RF circulating

immune complexes. Above, SPE and IFE before
mercaptoethanol treatment and below, after
treatment. Notice (above) that the SPE (gel to
the far left) shows a paraprotein in the beta-
gamma region. IFE shows bands in all lanes.
After treatment, below, a rapidly migrating IgG-
kappa monoclonal protein (arrow) is seen in the
lane fixed for IgG, along with diffuse staining.
The paraprotein in the other lanes is no longer
visible. The interpretation is IgG monoclonal
protein with rheumatoid factor activity forming
an immune complex by binding to polyclonal
IgG. Details of the gel are the same as in Figure
1.
752
Figure 10. Interference by IgM RF circulating immune complexes. Above, SPE and IFE before mercaptoethanol
treatment and below, after treatment. The arrows point to the location of the IC that appears as a broad band before
treatment. After treatment (below) an IgM-kappa monoclonal protein is seen, along with diffuse staining on the lane
stained for IgG. Other details are similar to the legend in Figure 1.
753
Immunoglobulin Quantitation
Karen Golemboski
Name: Immunoglobulin quantitation

Clinical significance: Refer to Chapter 53, Neoplasia, in the 5th edition of Clinical
Molecular formula and weight: See Immunoglobulin Table: Physical Properties
i
Historically, measurement of immunoglobulins (Ig) was IgM, IgG, IgA
included as part of the total globulins present in human Single radial immunodiffusion (RID; Table 1, Method 4)
serum (Table 1, Methods 1 and 2). Globulin isolation was an early method of Ig quantitation used in clinical
was achieved by salt or gold precipitation and settings. It consists of an agar plate with antibody
quantitation by the biuret assay. The introduction of incorporated throughout the agar. Test samples, placed
electrophoretic separation techniques allowed in antigen wells, diffuse into the agar and form a
fractionation of the globulins into alpha, beta, and precipitin ring of antigen-antibody (Iganti-Ig)
gamma components, the gamma fraction consisting complexes around the well. The diameter of the
almost entirely of immunoglobulins. Qualitative and precipitin ring is directly proportional to the amount of
semiquantitative estimates of total immunoglobulin antigen placed in the well. This procedure is used today
levels could be obtained from immunoelectrophoresis in laboratories performing low volumes of IgD or IgG
and protein electrophoresis, respectively (Table 1, subclass determinations, and the method is described in
Method 3). detail at the end of this chapter. A variation of the RID
technique, which shortened the lengthy incubation time,
Knowledge of immunoglobulin structure and the was electro-immunodiffusion (also known as Laurell
availability of monoclonal antibodies to each rocket electrophoresis). Automated methods of
immunoglobulin class (that is, specific for the constant immunoglobulin quantitation using nephelometry (Table
regions of the heavy and light chains) has made it 1, Method 5) or immunoturbidimetry (Table 1, Method
possible to quantitate each immunoglobulin class. The 6) have replaced RID techniques for measurement of
resulting complex of immunoglobulin (as antigen) and IgM, IgG, and IgA, providing faster results. The
antibody can be detected by physical means (either by nephelometric method measures light scattering
precipitation of immune complexes or by produced by the antigen-antibody complexes formed in
immunonephelometry, which detects the light-scattering the test sample tube. The relative light-scattering values
properties of immune complexes) or by measurement of are transformed into reportable units (that is, mg/L or
a label attached to the anti-immunoglobulin antibody International Units [IU]/L). Both end-point and kinetic
(e.g., fluorescent, enzymatic, or chemiluminescent label) methods of measurement are used. The end-point
(Table 1, Methods 4 through 9). method measures the light scattered by the complexes
formed after a steady state has been achieved. Thus
Differences in structure and relative predominance in antibody is added to the patient sample, and the sample
serum and other body fluids (for example, there is a is incubated for 15 min to 1 hour before quantitation.
millionfold difference between the serum concentrations The kinetic or rate method measures the maximum rate
of IgG and IgE) necessitate different methods of of increase of forward light scatter produced by the
quantitation for the different classes of formation of antibody-antigen complexes. With this
immunoglobulins. Igs are generally assayed as separate method, a lengthy incubation of the sample is not
classes, and not all Ig classes are measured in every required before quantitation. Automated nephelometric
clinical situation. Serum levels of IgD generally are not methods are available (Beckman Instruments, Brea CA;
of clinical importance unless a plasma-cell malignancy is Siemens Healthcare Diagnostics/Dade Behring, Inc.,
producing a monoclonal IgD protein. Quantitation of IgE Deerfield, IL). Turbidimetric methods measure the
is usually performed as part of an evaluation of allergic increase in absorbance resulting from the presence of the
conditions. precipitate. The absorbance reading can be taken at
equilibrium or on a kinetic basis (at selected time
intervals during the reaction). Automated turbidimetric
i
Immunoglobulin Quantitation methods include those from Abbott Laboratories (Irving,
Previous and current authors of this method: TX), Beckman Instruments, Olympus America
First edition: Gayle Birkbeck (Melville, NY), Roche Diagnostics (Indianapolis, IN),
Methods edition: Gayle B. Jackson and Siemens Healthcare Diagnostics/Dade Behring.
Second edition: Gayle B. Jackson Results in both methodologies are compared with a
Third edition: Gayle B. Jackson standard curve.
Fourth edition: Gayle B. Jackson
Fifth edition: Karen Golemboski
754
IgD chemiluminescent detection of an enzyme-labeled

IgD, whose physiological function is still unknown, is second antibody. Multiple antigen-coated threads may be
present in the serum at a concentration less than combined in the test chamber to allow simultaneous
IgM/A/G but higher than that of IgE. Serum levels of testing for IgE antibodies to multiple antigens. Rolling-
IgD are usually measured when production of a circle amplification, a highly sensitive detection
monoclonal IgD protein is detected by immunofixation. technique, utilizes labeled DNA for detection, with a
Methods of IgD quantitation include radial DNA primer being attached to the reagent antibody (in
immunodiffusion (particularly, the Mancini end-point this case, anti-human IgE). When circular DNA, DNA
method), which is of relatively poor sensitivity, and the polymerase, and nucleotides are added, linked copies of
more sensitive but less accessible methods of labeled circular DNA are produced, which can then be
immunoassay. In addition, reagents for nephelometric hybridized to labeled complementary oligonucleotide
IgD quantitation are available for use on the Behring probes (Qiagen, Valencia, CA).
BNII Analyzer (Siemens/Dade-Behring, Deerfield, IL).
Current in-vitro allergy testing methods are much
IgE improved compared with original radiolabeled methods
Of the five antibody classes, IgE is present in serum in (radiolabeled allergosorbent testing [RAST]; Table 1,
the smallest quantity. Sensitive methods such as Method 8). Many methods now calibrate results to
immunonephelometry, using an antibody specific for the World Health Organization (WHO) reference material,
heavy chain, are required to detect these low levels of and preparation of antigens has become more
protein. One example is the Dimension (Dade-Behring, standardized, although reference antigen preparations are
Deerfield, IL), which measures light scattered by not yet available. Despite advances in allergen-specific
immune complexes (nephelometry), the intensity of light IgE quantitation, assays are still less sensitive than
scatter being proportional to the concentration of IgE traditional skin testing [1].
antibodies in the sample. Another automated method is
the Elecsys II IgE system (Hoffman-La Roche, Nutley, Light-Chain Quantitation
NJ), a sandwich immunoassay in which the anti-IgE In certain lymphoproliferative disorders, unbalanced
antibody is bound to streptavidin. A biotin-labeled synthesis of immunoglobulin subunits results in excess
enzyme binds to the streptavidin-IgE complex, allowing production of or light chains. These free light chains
chemiluminescent detection of IgE in the patient may circulate in serum or be excreted in urine.
specimen (Table 1, Method 9). The ImmuneTech Total Quantitation of free light chains in serum is more
IgE system (ImmuneTech Corporation, Menlo Park, CA) sensitive than urine immunoglobulin quantitation
is also biotin-avidin based, with the specimen being methods and also allows detection and monitoring of
combined with anti-IgE-coated microspheres. A biotin- monoclonal gammopathies which are non-secretory or
labeled anti-human IgE antibody is followed with which secrete only light chains. Immunonephelometric
fluorescent-labeled streptavidin, and the fluorescent and immunoturbidimetric assays use antibodies specific
signal from the resulting complex is directly proportional for light chains not associated with heavy chains (The
to the amount of IgE in the sample. Fluorescence is Binding Site, San Diego, CA) [2].
measured with a flow cytometer. Biokit S.A. (Barcelona,
Spain) markets the Quantia IgE, an immunoturbidimetric Standardization and Quantitation
assay enhanced with latex particles. The UniCAP Total The immunoglobulins are divided into five major classes
IgE method from Pharmacia/Pfizer (New York, NY) is a with four immunologically characterized subclasses of
fluoroimmunoassay (Table 1, Method 7). IgG and two of IgA, differing in heavy-chain structure
and glycosylation. The ratio of these subclasses may
Quantitation of total serum IgE is usually performed to vary with some infectious and autoimmune diseases and
diagnose or confirm hypersensitivity (allergy). However, will certainly vary with monoclonal gammopathies.
since many patients with allergy or atopic disease may Immunoglobulins also vary in size: IgM is usually
have normal or only slightly elevated total IgE, this present as the stable 19S pentamer of 7S subunits but
measurement may or may not be sufficient. Much of the occasionally may be found in free 7S form. IgA is 7S in
development in immunoglobulin testing over the last 2 the monomeric form but 11S in the dimeric form (see
decades has been directed at measurement of specific Table 2).
IgE antibody levels for individual allergens, as an
adjunct to skin testing for allergy. Specific IgE Reference Materials
antibodies are assayed by coating a solid surface with the The WHO first introduced a reference preparation for
antigen in question, allowing patient antibodies to bind IgG, IgA, and IgM in 1970 (NIBSC 67/86), allowing
and detecting IgE with a labeled anti-IgE antibody. The conversion of mg/dL levels to IU/L. The CAP and the
Immulite 2000 system (Diagnostic Products European Community Bureau of References (BCR)
Corporation, Los Angeles, CA) uses allergens bound to jointly produced a subsequent preparation in 1993, CRM
beads; IgE is detected with alkaline phosphataselabeled 470 (BCR)/ RPPHS (CAP) [3]. IgE methods are
anti-IgE. A chemiluminescent substrate emits light in currently standardized against the 2nd WHO
direct proportion to the amount of allergen-specific IgE International Reference preparation, NIBSC 75/502.
in the sample. The AP 1800 CLA (Hitachi Chemical Standard reference material for IgD is also available
Diagnostics, Inc., Mountain View, CA) employs (British Research Standard 67/37).
cellulose threads as the solid surface and
755
Reference and Preferred Methods urine (Bence-Jones proteins), although concentration

The International Federation of Clinical Chemistry and may be required.
Laboratory Medicines Committee on Plasma Proteins
has designated both immunonephelometry and CSF
immunoturbidimetry as reference methods for An increased level of intrathecal immunoglobulin
quantitation of immunoglobulins IgG, IgA, and IgM in production is associated with demyelinating diseases
serum or plasma [4]. Automated versions of both such as multiple sclerosis (MS). IgG levels are compared
methods are readily available, and according to 2007 with CSF albumin to control for increased plasma levels
CAP survey data, approximately equal numbers of each or damage to the blood-brain barrier. Measurement of
are in use, and no respondents indicated use of any other CSF IgM or IgA may be also be indicated if conditions
methods for quantitation of these immunoglobulins. such as neurotuberculosis, mumps meningitis, or
Both methods performed consistently within acceptable adrenoleukodystrophy are suspected. Automated
tolerances on the survey. nephelometric methods (Dade Behring BNII System and
Beckman Coulter Immage) have demonstrated sufficient
No methods for quantitation of IgE have been designated sensitivity and acceptable linearity for immunoglobulins
as reference methods to date. Respondents to 2007 CAP in CSF [8].
surveys indicated a wide variety of methods in use; the
two most commonly reported were nephelometry (65) Tears
and chemiluminescence (175). The remaining Detection of elevated levels of IgE in tears indicates an
respondents used labeled immunoassay (fluorescent-, ocular allergic response. A solid-phase
enzyme-, or radiolabeled). immunochromatographic test, Lacrytest, is designed as a
strip which is placed in the eye until sufficient tears are
High-resolution serum (and urine) protein absorbed (Adiatec SA; Loire-Atlantique, France). The
electrophoresis is the recommended method for test detects IgE above 2.5 KIU/L (3 ng/mL), which is
detection of a monoclonal (M) protein. Once detected, considered the upper limit of normal levels.
densitometry should be used to quantitate M proteins,
and the heavy- and light-chain classes of the M protein Other
should be defined by immunofixation (the use of Normal levels for immunoglobulin in thoracentesis fluid,
immunoelectrophoresis is discouraged). If the level of synovial fluid, and other serous exudates are not well
secreted monoclonal protein is low, such that the defined, and the clinical usefulness of these fluid
densitometer peak may be obscured by normal levels of quantitations is limited.
other immunoglobulins (e.g., monoclonal IgA < 2.0
g/dL), then quantitation by nephelometry is indicated. Interferences
For high levels of M protein, however, values Since light scattering produced by contaminating
determined by nephelometry (especially IgM and IgA) particles in the antiserum can interfere with quantitation,
are often falsely elevated [5,6]. antiserum that is free of contamination must be used.
Turbid patient samples may have to be filtered before
Specimen use on the nephelometer. The presence of rheumatoid
Immunoglobulin quantitation, for routine diagnostic factor or heterophilic antibodies which bind to reagent
purposes, is performed most commonly on serum antibodies may interfere with immunoassay. Prozone, or
specimens. Samples for quantitation may be stored for hook, effects in the presence of high levels of
up to 5 days at 2C to 8C if they are protected from immunoglobulin (e.g., in multiple myeloma) may lead to
contamination and evaporation. For sample shipment or falsely low results and require dilution of sample for
longer storage periods, samples should be frozen at accurate quantitation [9].
20C or colder. Once thawed, samples generally should
not be refrozen, because repeated freezing and thawing Immunoglobulin Quantitation Reference Interval
may cause deterioration of proteins. Immunoglobulin concentrations decrease during
pregnancy and reach lowest levels in the early
Immunoglobulins in Other Body Fluids postpartum period. Levels vary widely in children and
Urine adults; immunoglobulins present in infants as a result of
Under normal conditions, immunoglobulins are not transplacental transfer decline until about 6 months of
found in urine. The presence of Igs in urine generally age. Normal levels in children increase through
indicates an overload proteinuria resulting from excess adolescence to adult levels (Table 2). Declining levels of
production, which may be due to a monoclonal IgG and IgM in aging individuals have been reported
gammopathy or in some cases polyclonal Ig production [10,11].
during an infectious disease. Urine specimens generally
are concentrated 20 to 50 times before they are evaluated Interpretation
by electrophoresis or by immunofixation. Detection of Immunoglobulin abnormalities can be divided into three
IgD in urine, even after concentration, has not been groups: (1) increased levels of several different
reported [7]. immunoglobulins (polyclonal gammopathy), (2)
increased levels of a single immunoglobulin type or of
Free light chains are small enough to pass through the proteins related to immunoglobulins (monoclonal
glomerular membrane and therefore may be excreted in
756
gammopathy), and (3) decreased levels of one or more lymphomas, monoclonal gammopathy of undetermined
immunoglobulins. significance (MGUS), and old age.
Human Serum Immunoglobulin Reference Intervals (mg/mL)

Immunoglobulin
class IgG IgM IgA IgD IgE
Cord specimen 7.6616.93 0.040.26 0.00040.09
0.53 months 2.998.52 0.151.49 0.030.66
36 months 1.429.88 0.181.18 0.040.90
612 months 4.1811.42 0.432.23 0.0140.95
12 years 3.5612.04 0.372.39 0.131.18 116122 ng/mL
23 years 4.9212.69 0.492.04 0.231.37 80122 ng/mL
36 years 5.6413.81 0.512.14 0.352.09
47 years 140442 ng/mL
69 years 6.5815.35 0.502.28 0.293.84
1014 years 374674 ng/mL
1216 years 6.8015.48 0.452.56 0.812.52
Adult 8.0016.00 0.502.00 1.403.50 00.14 22000 ng/mL
Data from:
Meites S, editor. Pediatric clinical chemistry; 2nd edition. Chicago: American Association for Clinical Chemistry;
1981.
A normal human serum immunoglobulin class is

composed of a heterogeneous mixture because of
Serum protein electrophoresis, followed by
subclass differences and the presence of numerous,
immunofixation electrophoresis, may be used to classify
different, heavy- and light-chain variable regions. Many
the increase as polyclonal or monoclonal and to identify
different clones of plasma cells synthesize and secrete
the monoclonal protein, if present. Monoclonal proteins
any one of the different class, subclass, and type
may be produced in a variety of diseases classified as
combinations, such as IgA, kappa. However, only one
monoclonal gammopathies, including multiple myeloma
clone synthesizes an IgA, kappa protein with a given
(producing monoclonal immunoglobulins which may be
heavy- and light-chain variable regionamino acid
of any class or may secrete only light chains),
sequence (antigen specificity).When many different
Waldenstrms macroglobulinemia (monoclonal IgM),
families or clones of plasma cells are responsible for an
heavy-chain disease (monoclonal production of
increased production of immunoglobulins, the increase is
immunoglobulin heavy chains only), and amyloidosis, in
termed polyclonal. When only one clone is responsible
which the monoclonal immunoglobulin deposits fibrils
for the increase, the increase is homogeneous and is
in multiple organs [12].
termed monoclonal. Immunoglobulin quantitation alone
cannot determine whether an immunoglobulin elevation
Quantitation of a monoclonal protein is important both
is monoclonal or polyclonal. For this purpose, protein
in the initial diagnosis and staging of disease and in
electrophoresis and/or immunofixation should be used in
following the disease course. In multiple myeloma, the
conjunction with immunoglobulin quantitation. A
amount of monoclonal protein is directly related to the
monoclonal protein will appear as a narrow band on the
size of the tumor burden, and quantitation of the
protein electrophoresis. Immunofixation will
monoclonal protein is one factor used in following the
demonstrate an increase in just one heavy chain and one
therapy. Malignant monoclonal gammopathies are often
light chain class. In a polyclonal increase, the band will
accompanied by decreases in the levels of normal
be diffuse on protein electrophoresis, and
polyclonal immunoglobulins.
immunofixation will demonstrate increases in more than
one antibody class.
Immunoglobulin deficiencies are best demonstrated by
Increased immunoglobulins may indicate a normal
quantitation of the immunoglobulin classes. Protein
polyclonal response to infection, chronic inflammation, a
electrophoresis can be used as a screen for
hypersensitivity response (allergic disease), or
hypogammaglobulinemia, in which B-cell activity is
production of an abnormal monoclonal immunoglobulin.
deficient or absent, and there is virtually no IgG detected
Polyclonal increases are nonspecific and may be found
in serum. However, the other immunoglobulin classes
with infections, autoimmune diseases, advanced
contribute so little to total levels that the densitometer
cirrhosis, chronic active hepatitis, biliary cirrhosis,
curve shows little change if they are decreased or absent.
sarcoidosis, narcotics addiction, amyloidosis, and many
parasitic diseases. Monoclonal gammopathies are found
Decreased total or individual immunoglobulin levels
with multiple myeloma, Waldenstrms
suggest an immunodeficiency disorder, which may be
macroglobulinemia, heavy-chain disease, essential
congenital or acquired. Most immunodeficiencies
cryoglobulinemia, chronic lymphocytic leukemia,
emerge in childhood, usually after maternal antibodies
are lost at about 6 to 8 months of age, but some may first
757
become evident in adulthood. Altered immunoglobulin require that laboratories be accurate to within 3SD of
levels may be the result of a defect in B cells, T cells, or the peer-group mean, and for immunoglobulin G,
both, since responses of B cells are not independent of T laboratories must obtain results that are within 25% of
cells. the peer-group mean. Maximum SD values for
immunoglobulins at medically significant decisions
About half of primary immunodeficiencies are the levels are as follows [15]:
consequence of B-cell defects and result in decreased
immunoglobulin levels. Individuals with these disorders Immunoglobulin Decision level Maximum SD
generally suffer repeated infections with encapsulated IgA 400 mg/L 17 mg/dL
bacteria such as Streptococcus pneumoniae or IgE 200 IU/mL 38 IU/mL
Haemophilus influenzae. A defect in B-cell maturation IgG 500 mg/dL 13 mg/dL
results in Brutons (X-linked) agammaglobulinemia, IgG 2000 mg/dL 125 mg/dL
resulting in markedly decreased or absent B cells and IgM 300 IU/mL 9 IU/mL
immunoglobulins. Common variable immunodeficiency
(CVI) represents a group of disorders, all of which result
in hypogammaglobulinemia; these patients may present The intraindividual variability of immunoglobulins G, A,
in childhood, but the peak age of onset is in the third and M, measured in healthy adults over a 3- to 5-month
decade of life. period, were found to range from 4% to 6% [15]. Ricos
et al. established desirable specifications for analytical
Alterations may also occur in specific immunoglobulin imprecision derived from studies of biological variation,
classes. Selective IgA deficiency is the most common finding an assay imprecision for IgA, IgG, and IgM of
immunodeficiency disorder, with an incidence of about 1 no greater than 2.7%, 2.3%, and 3.0% and total error of
in 500-700 individuals, although many are no greater than approximately 14%, 8%, and 17%,
asymptomatic. These individuals have IgG and IgM respectively [16].
levels within normal limits and respond well to vaccines,
but they may suffer increased respiratory and Current automated methods for IgG, IgA, and IgM
gastrointestinal infections. IgA-deficient individuals may quantitation appear to perform well within established
also be deficient in one or more subclasses of IgG, performance ranges; CAP data indicated only one IgA
especially IgG2. Since antibodies to polysaccharide high value result that exceeded the CLIA +3SD limit
antigens (such as bacterial capsular antigens) are usually (out of nearly 3000 reports). Quantitation methods for
of the IgG2 subclass, individuals deficient in IgG2 may IgE are also consistent, with CV values averaging
show increased susceptibility to bacterial infection, even around 7.5%.
though total IgG levels may be within normal limits. A
decrease in IgG and IgA with increased IgM References
concentration is seen in hyper-IgM syndrome, in which 1 Hamilton RG, Adkinson NF Jr. In-vitro assays
T cells are unable to produce the cytokines necessary to for the diagnosis of IgE-mediated disorders. J
induce B cells to switch from IgM to production of other Allergy Clin Immunol 2004;114:213-115.
immunoglobulin classes. 2 Katzmann JA, Clark RJ, Abraham RS, Bryant
S, Lymp JF, Bradwell AR, Kyle RA. Serum
Combined immunodeficiency disorders, affecting both B reference intervals and diagnostic ranges for
and T cells, may also result in abnormal immunoglobulin free and free immunoglobulin light
levels. These include severe combined chains: relative sensitivity for detection of
immunodeficiency syndrome (SCIDS; markedly reduced monoclonal light chains. Clin Chem
IgG/A/M), Wiskott-Aldrich syndrome (decreased IgM, 2002;48:1437-1444.
elevated IgA and IgE), and ataxia-telangiectasia 3 Johnson AM. Effect of a new international
(decreased or absent IgA; decreased IgE, IgG subclasses, reference preparation for proteins in human
or a combination) [13]. serum (CRM 470) on results of the College of
American Pathologists surveys for plasma
Acquired immunodeficiency may be secondary to proteins. Arch Pathol Lab Med 2000;124:1496-
concomitant malignancy or other diseases or may result 1501.
from protein-calorie malnutrition or from infection with 4 Blirup-Jensen S. Protein standardization III:
human immunodeficiency virus (HIV) [14]. method optimization. Basic principles for
quantitative determination of human serum
Immunoglobulin Performance Goals proteins on automated instruments based on
Survey data from the 2007 College of American turbidimetry or nephelometry. Clin Chem Lab
Pathologists (CAP) Participant Summary Report show Med 2001;11:1098-1109.
imprecision values (% coefficient of variation [%CV]) 5 Keren DF, Alexanian MD, Goeken JA, Gorevic
for immunoglobulins G, A, and M to range from PD, Kyle RA, Tomar RH. Guidelines for
approximately 2% to 8% in samples with clinical and laboratory evaluation of patients
immunoglobulin concentrations at the upper end of the with monoclonal gammopathies. Arch Pathol
normal reference interval [5]. Acceptable performance Lab Med 1999;123:106-107.
criteria (Clinical Laboratory Improvement Amendments
88) for measurement of immunoglobulins A, E, and M
758
6 Alexanian R, Weber D, Liu F. Differential eds. Manual of Molecular and Clinical

diagnosis of monoclonal gammopathies. Arch Laboratory Immunology. 7th ed. Washington,
Pathol Lab Med 1999;123:108-113. DC: ASM Press; 2006:906-913.
7 Vladutiu AO. Immunoglobulin D: properties, 14 Stevens RA, Lempicki RA, Natarajan V,
measurement, and clinical relevance. Clin Higgins JH, Adelsberger JW, Metcalf JA.
Diagn Lab Immunol 2000;7:131-140. General immunologic evaluation of patients
8 Owen WE, Roberts WL. Performance with human immunodeficiency virus infection.
characteristics of four immunonephelometric In: Detrick B, Hamilton G, Folds J, eds. Manual
assays for the quantitative determination of IgA of Molecular and Clinical Laboratory
and IgM cerebrospinal fluid. Am J Clin Pathol Immunology. 7th ed. Washington, DC: ASM
2003;119:689-693. Press; 2006:847-861.
9 Butch AW. Dilution protocols for detection of 15 Ford RP, Mitchell PEG, Fraser CG. Desirable
hook effects/prozone phenomenon. Clin Chem performance characteristics and clinical utility
2000;46:1719-1720. of immunoglobulin and light chain assays
10 Warren JS. Immunoglobulin quantification and derived from data on biological variation. Clin
viscosity measurement. In: Detrick B, Hamilton Chem 1988;34:1733-1736.
G, Folds J, eds. Manual of Molecular and 16 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
Clinical Laboratory Immunology. 7th ed. Hernandez A, Jimenez CV et al. Current
Washington, DC: ASM Press; 2006:69-74. databases on biologic variation: pros, cons and
11 Lock RJ. Immunoglobulins and progress. Scand J Clin Lab Invest 1999;59:491-
immunoglobulin subclasses in the elderly. Ann 500.
Clin Biochem 2003;40:143-148. 17 Ritzman SE. Immunoglobulin abnormalities. In:
12 Katzmann JA, Kyle R. Immunochemical Ritzmann SE, Daniels JC, eds. Serum protein
characterization of immunoglobulins in serum, abnormalities, diagnostic and clinical aspects.
urine, and cerebrospinal fluid. In: Detrick B, Boston: Little, Brown; 1975:351-485.
Hamilton G, Folds J, eds. Manual of Molecular 18 Berne BH. Differing methodology and
and Clinical Laboratory Immunology. 7th ed. equations used in quantitating immunoglobulins
Washington, DC: ASM Press; 2006:88-100. by radial immunodiffusion in a comparative
13 Conley ME. Primary antibody deficiency evaluation of reported and commercial
diseases. In: Detrick B, Hamilton G, Folds J, techniques. Clin Chem 1974;20:61-68.
759
Table 1: Methods of Immunoglobulin Quantitation

Method 1: Salt precipitation; isolation of total immunoglobulin by salt precipitation
Principle of analysis: Immunoglobulins are less soluble in certain salt solutions and precipitate; amount of
immunoglobulins quantified by total protein assay (such as biuret)
Comments: Serum; historical, nonspecific
Method 2: Gold precipitation; isolation of total immunoglobulin by gold precipitation
Principle of analysis: Similar to salt precipitation; gold used as precipitant
Comments: Serum; historical, nonspecific
Method 3: Electrophoresis; estimation of total immunoglobulin by physical separation
Principle of analysis: Proteins separate based on class (albumin, gamma globulin), calculation of each class as
percentage of total protein
Comments: Serum; good screening method for detection of monoclonal immunoglobulins
Method 4: Radial immunodiffusion (RID); quantitation by immunoprecipitation in gel
Principle of analysis: Immunoglobulin diffuses into gel containing antibody, forming a ring-shaped
immunoprecipitate; diameter of ring proportional to concentration
Comments: Serum, body fluids; accurate, slow, expensive; used for IgG subclass and IgD determination.
Variation: Electroimmunodiffusion (Laurell rocket); immunoglobulin electrophoresis into antibody-containing
agarose gel; height of gel pattern (rocket) proportional to immunoglobulin concentration.
Method 5: Nephelometry; quantitation by immunoprecipitation in solution
Principle of analysis: Reaction of immunoglobulin with its specific antibody results in immunoprecipitate,
which has light-scattering properties; amount of light scatter proportional to immunoglobulin concentration
Comments: Serum, body fluids; accurate, rapid, adaptable to automation; most commonly used
Method 6: Turbidimetry; quantitation by size of immunoprecipitate in gel
Principle of analysis: Immunoglobulin reacts with antibody or other precipitation agent; turbidity proportional to
immunoglobulin concentration
Comments: Serum, body fluids; accurate, rapid, adaptable to automation; commonly used
Method 7: Immunofluorometry; quantitation by competition for labeled antibody on a solid phase
Principle of analysis: Immunoglobulin adsorbed onto solid surface competes for fluorescent-labeled antibody
with immunoglobulin sample; fluorescence signal is inversely proportional to immunoglobulin concentration
Comments: Serum, some body fluids; precision slightly less than nephelometry; adaptable to automation
Method 8: Labeled immunoassay, competitive
Principle of analysis: Immunoglobulin reaction with antibody displaces labeled immunoglobulin. Label may be
detected by colorimetric, fluorometric, or chemiluminescent method. Amount of label detected is inversely
proportional to immunoglobulin concentration in specimen.
Comments: Any body fluid; sensitive
Method 9: Labeled immunoassay, noncompetitive
Principle of analysis: Immunoglobulin reacts with antibody on solid surface; reaction quantitated by second,
labeled antibody. Label may be detected by colorimetric, fluorometric, or chemiluminescent method. Amount of
label detected is proportional to immunoglobulin concentration in specimen.
Comments: Any body fluid; sensitive
760
Table 2: Physical Properties of Human Immunoglobulins

Immunoglobulin
class IgG IgM IgA IgD IgE
Molecular 150,000 900,000 160,000 (monomer) 185,000 200,000
mass (Daltons) 320,000 (dimer)
Sedimentation 6.6 18.019.0 6.2510.9 6.27.0 7.867.92
coefficient, S
Heavy chains
Heavy-chain 1 , 2 , 3 , 4 1, 2 1, 2
subclasses
Light chains or or or or or
Molecular IgG()22 IgM()(22 )5 IgA()(22 )1-3 IgD()22 IgE()22
formula IgG()22 IgM()(22)5 IgA()(22)1-3 IgD()22 IgE()22
Reference Intervals Serum Concentrations (mg/mL) (by age)
Cord specimen 7.6616.93 0.040.26 0.00040.09
0.53 months 2.998.52 0.151.49 0.030.66
36 months 1.429.88 0.181.18 0.040.90
612 months 4.1811.42 0.432.23 0.0140.95
12 years 3.5612.04 0.372.39 0.131.18 116122 ng/mL
23 years 4.9212.69 0.492.04 0.231.37 80122 ng/mL
36 years 5.6413.81 0.512.14 0.352.09
47 years 140442 ng/mL
69 years 6.5815.35 0.502.28 0.293.84
1014 years 374674 ng/mL
1216 years 6.8015.48 0.452.56 0.812.52
Adult 8.0016.00 0.502.00 1.403.50 00.14 22000 ng/mL
Data from Seligson O, ed. Immunology. In: Handbook Series in Clinical Laboratory Science. Vol 1. Section F. Part I. Boca
Raton, FL: CRC Press; 1978; and Meites S, ed. Pediatric Clinical Chemistry. 2nd ed. Chicago: American Association for
Clinical Chemistry; 1981.
IgG Subclass Quantitation by Radial and precision are somewhat less than those obtained
Immunodiffusion with the end-point technique [17,18]. Commercial plates
are available for radial immunodiffusion quantitation.
Introduction Many plates can be used with either the Fahey or
For laboratories performing few IgG subclass Mancini technique.
quantitations, the technique of radial immunodiffusion
yields appropriate quantitative information. Commercial
kits such as those from the Binding Site are available.
There are several methods of obtaining the quantitative Reagents and Materials Required
result by this technique. The method described in this Kits (includes calibrators and 7% BSA)
procedure uses an end-point calibration curve in which
standards and unknowns are allowed to diffuse to end- IgG subclass 1
point before a calibration curve is constructed. IgG subclass 2
IgG subclass 3
Two radial immunodiffusion methods have been used in IgG subclass 4
clinical laboratories. The Mancini method is based on an
end-point (equivalence-zone) reading. The plates are Reagent Preparation, Storage, and Stability
incubated for about 48 hr (72 hr for IgM). The diameter Plates: If plates are kept closed and properly stored (4C
of the precipitin ring is measured, and then the antigen to 6C), they may be used until expiration date. They
level is determined from a standard curve. A standard must not be frozen.
curve is prepared by plotting on linear graph paper the Calibrators: Each kit contains four calibrators (one for
concentrations of three or four reference sera against the each subclass) and 0.01% thimerosal as preservative.
ring diameter squared (mm2). The Fahey-McKelvey The concentrations of the calibrators are indicated on the
method measures the ring diameter before equivalence labels. Note that this is expressed in mg/liter. Medium
(such as 18 2 hr). One constructs the standard curve by and low-level dilutions of the calibrators are required to
plotting ring diameter (mm) against the concentration of provide calibration curves. These can be made by
the reference standards on two-cycle semilogarithmic diluting the calibrators to 60% and 10%. It is
graph paper. The Fahey technique has the advantage of recommended that 120 L of calibrator be mixed with
producing results more rapidly; however, the accuracy
761
80 L of BSA for the 60% dilution and that 25 L of quoted on the control is 6100 mg/L, this is equivalent to
calibrator be mixed with 225 L of BSA for the 10% a ring diameter of 6.4 mm (from the RID reference
dilution. table). The control serum range should be the values
Human Serum Control: Store at 4C. obtained for 6.1 to 6.7 mm, which reads as 5370 to 6880
mg/L. Control values must be within stated ranges
Specimens before patient results can be reported. Notify supervisor
Use serum collected by normal means. Perform IgG if controls are out of range.
quantitation before setting up subclasses. This is to be
used as a guideline for making sample dilutions. Reporting Results
Compare patient values with expected values based on
Procedure previous IgG quantitation and the reference interval for
1. Dilute control and patient samples in 7% BSA. that age. If these correlate, answer test in computer. If no
a. Each subclass requires a different ring was detected, answer NONE. If ring was smaller
sample dilution. The standard than lowest standard, repeat patient at a lower dilution or
dilutions are: neat. If ring at neat dilution is still smaller than lowest
standard, answer < _____ (lowest standard mg/dL
IgG1 1:10* value).
25 L sample + 225 L of 7% BSA If a sample must be repeated for any subgroup, also
IgG2 1:10* repeat it for the other subgroups to keep the use of the
25 L sample + 225 L of 7% BSA wells consistent on all plates.
IgG3 Neat Expected Values
IgG4 Neat
IgG Subclass Reference Intervals
* For control, a smaller volume may IgG1 IgG2 IgG3 IgG4
be needed for 1:10 dilution. Age (years) mg/dL mg/dL mg/dL mg/dL
b. Because of the wide variation in less than 1 190620 30140 962 613
normal values based on patient age, it 1 230710 30170 1198 443
may be appropriate to use other 2 80830 40240 6130 3120
dilutions.
3 350790 50260 998 5180
c. Dilute Human Serum Control for use
45 360810 60320 9160 9160
as a control, using 1:10 dilution for
IgG 1 and IgG 2; neat control for IgG 67 2801120 30630 4250 11620
3 and IgG 4. 89 2801740 80550 22320 10170
2. Label worksheets to indicate the proper wells 1012 2701290 110550 13250 7530
and dilutions for each sample. 1315 2801020 60790 14240 1330
3. Fill appropriate wells with 5 L of samples 16 to adult
(neat or diluted, as appropriate). 327855 52556 4117 1114
4. Tightly close the plates, and incubate at room
temperature on a flat surface until end-point is In healthy adults, IgG constitutes approximately 75% of
reached (72 hr). the total serum immunoglobulins. Within the IgG class,
5. Using the RID viewer, read and record ring the concentrations of the four subclasses are
diameters to 0.1 mm. approximately as follows: IgG1, 60%; IgG2, 30%; IgG3,
6. To construct the standard curve, plot the square 5%; and IgG4, 4% (see table above). These figures vary
of the diameters against the concentration for somewhat from individual to individual, and this may be
each standard. At end-point, this should result under genetic control.
in a straight line. Construct a best-fit straight
line through adjacent points. IgG is the only class of immunoglobulins that can pass
7. Determine the concentration of the patient the placenta in humans and is responsible for protection
sample from the standard curve. Multiply by of the newborn during the first few months of life. IgG2
the appropriate dilution factor. is transferred more slowly than the other subclasses. The
IgG subclasses are also capable of fixing complement in
Quality Control the following order of descending efficiency: IgG3 >
Record control values for each subclass on QC sheet. IgG1 > IgG2 > IgG4. IgG4 is unable to fix complement
The ranges for the Human Serum Control are based on by the classic pathway but may be active in the alternate
the RID reference table enclosed in the kit. The pathway.
concentration mean of the control is listed on the control
bottle. Take this mean and determine the concentration Macrophages bear surface receptors which bind IgG1
on the table closest to the mean value. Then determine and IgG3 and their Fc fragments which are responsible
the ring diameter of this concentration as listed in the for arming the macrophages. Many infections in infants
reference Table. The range of the control should be may be due to failure to produce the normal proportions
within +0.3 mm; for example, if the IgG1 concentration of the subclasses, particularly IgG2.
762
Insulin and C-Peptide

Name: Insulin and C-peptide

Clinical significance: Refer to Chapter 38, Diabetes, in the 5th edition of Clinical Chemistry:
Molecular mass: Insulin, 6000 D
C-peptide, 3000 D
Proinsulin, 9000 D
Merck Index: 4866
Chemical class: Protein (hormone)
immunoassays. Enzyme immunoassay (Table 1, Method 2)

Principles of Analysis and Current Usage i is one of the most popular techniques. Other assays include
Insulin and C-peptide are synthesized in the rough insulin radio-receptor methods using cultured leukocytes,
endoplasmic reticulum of the beta cells of the pancreatic and bioassays that measure the stimulation of glucose
islets of Langerhans. Like other polypeptide hormones and oxidation.
related proteins that enter the endoplasmic reticulum,
insulin and C-peptide are synthesized as part of a larger The quality of the various assays used for measurement of
prohormone. Insulin is stored in secretory granules as an insulin is a matter of concern. Problems include lack of
inactive 9000-D precursor molecule known as proinsulin. standardization, nonspecificity of the antibodies used, and
Proinsulin contains the insulin molecule covalently linked the presence of anti-insulin antibodies and insulin auto-
to a single polypeptide chain of 3000 D known as C- antibodies in patient samples [3]. Problems due to widely
peptide. Proteolytic cleavage of the proinsulin polypeptide disparate results obtained with different insulin
releases one molecule of insulin and one molecule of the immunoassays have been recently confirmed, and these
C-peptide. Normally, 90% to 97% of the product released problems hinder efforts towards achieving consistent
from the beta cells is C-peptide and insulin in measures for treatment guidelines [4].
equimolecular amounts. The remainder is primarily
proinsulin [1]. C-peptide is not biologically active. The Many persons who have received diabetic therapy have
half-life of C-peptide is significantly longer than that of circulating endogenous antibodies to insulin. These
insulin, resulting in fasting concentrations of C-peptide antibodies may increase or decrease the apparent insulin
that are 5- to 10-fold higher than those of insulin. concentration because of an effect on the binding antibody
or on the separation step [5]. The endogenous antibodies
Active insulin is a 51amino acid protein of 6000 D. It is can be measured and then removed by precipitation (see
composed of two chains (A and B) that are joined by following discussion).
disulfide bridges. The insulin molecules of humans,
horses, cattle, sheep, pigs, and rabbits are biologically and For assay purposes, antibodies to insulin are usually raised
(to a lesser extent) immunologically similar. in guinea pigs, using purified porcine insulin. Because
human and rabbit insulin are very similar, antisera
The analysis of insulin holds a position of singular developed in rabbits generally do not have a high affinity
importance in clinical chemistry, since it was this for human insulin. Insulin is sufficiently immunogenic to
substance that was measured in the first radioimmunoassay initiate an antibody response without being covalently
(Table 1, Method 1) [2]. In the modern clinical laboratory, linked to a carrier. However, the hypoglycemia caused by
measurement of insulin is accomplished using a variety of injection of pure insulin is a limiting factor when one is
designing an immunization schedule. A good insulin assay
will allow the detection of 10 to 30 pg (24 to 72 U) of
iInsulin
insulin per assay tube. Antibody cross-reactivity with
Previous and current authors of this method: human, porcine, and bovine insulin; human growth
First edition: Michael D.D. McNeely hormone (HGH); C-peptide; and proinsulin should be
Methods edition: Michael D.D. McNeely determined for all assays. Cross-reactivity will be nearly
Second edition: Not updated 100% with various animal insulins, up to 60% with
proinsulin, but less than 1% with C-peptide and HGH.
Fourth edition: Steven C. Kazmierczak Immunometric assays that are specific for insulin are
Fifth edition: Steven C. Kazmierczak commercially available for a large number of instrument
763
platforms. Unfortunately, these assays may not recognize proportional to the insulin concentration within the sample.
insulin analogs used for treatment of diabetes [6]. A recent
study that compared 11 different insulin assays found that Another nonisotopic immunoassay procedure for insulin is
measured values varied by a factor of two. These that based on microparticle enzyme immunoassay (MEIA)
commercial insulin assays were standardized against the technology (Table 1, Method 4). In this procedure, insulin
first International Reference Preparation 66/304 (IRP is incubated with monoclonal (mouse) anti-insulin-coated
66/304) [7]. microparticles, producing an antibody-insulin complex. An
aliquot of this reaction mixture containing insulin bound to
Insulin is readily labeled with radioactive iodine (125I) the anti-insulin-coated microparticles is transferred to a
using the chloramine-T procedure. Separation of bound glass fiber matrix. The matrix is washed to remove
and free labeled insulin has been carried out by use of most unbound materials and a second anti-insulin antibody
available techniques. These include dextran-coated conjugated with alkaline phosphatase is dispensed onto the
charcoal, double-antibody precipitation, and polyethylene glass fiber matrix, where it binds to the antibody-antigen
glycol. Endogenous insulin antibodies influence the complex. Unbound materials are removed from the matrix
separation. using a second wash step, and the substrate, 4-
methylumbelliferyl phosphate, is added to the matrix. The
The original enzyme immunoassays (EIAs) for insulin fluorescence produced is directly related to insulin
showed good correlation with RIA procedures [8,9] but concentrations within the specimen.
were not so easily performed. The EIAs employ a
sandwich technique. In this technique, antibodies to insulin A variety of high-performance liquid chromatography
are adsorbed onto a solid phase (such as disks or beads). (HPLC)-based procedures for insulin have been described,
Insulin from the sample remains bound to the solid-phase but these are primarily used in research settings. Recently,
antibodies after a washing step. Anti-insulin antibodies mass spectrophotometric methods for insulin have been
directed to a second site on the insulin molecule and described [15].
conjugated to an enzyme are added to the solid phase.
After washing, the amount of bound antibodyenzyme is Assays for C-peptide are used for detecting self-induced
determined by incubation of the solid phase with enzyme hyperinsulinism, since pharmaceutical preparations of
substrate. The enzyme most often used is horseradish insulin do not contain C-peptide. In addition, measurement
peroxidase, which reacts with substrates such as 5- of C-peptide can be used to monitor pancreatic surgery.
aminosalicylic acid[9] and p-hydroxyphenylacetic acid Levels should be undetectable following radical
[10]. The products of the enzyme reaction are measured pancreatectomy and should increase following successful
spectrophotometrically (at 500 nm) [9] or fluorometrically islet-cell or pancreas transplantation. C-peptide RIA
[10] (Dexcitation, 317 or 325 nm; Demission, 414 nm). The development is difficult, since the molecule does not
amount of product formed is directly proportional to the contain tyrosine, and synthetic or tyrosylated molecules
amount of insulin present in the sample. A homogeneous must be used. Standardization of the C-peptide assay is a
EIA for insulin has not been developed. source of interassay variability. It is generally agreed
[10,16] that prior precipitation of endogenous antibodies is
A number of non-isotopic immunoassays for insulin have mandatory before a C-peptide assay is performed.
also been developed. In these competitive insulin
immunoassays, enzyme labels employing fluorometric or Although measurement of C-peptide offers several
luminometric measurements of activity, as well as advantages over insulin determinations, several problems
fluorescence labels for use in fluorescence polarization need to be resolved. The issue of standardization must be
immunoassays, have been employed [11-13]. addressed, as various antisera used are known to differ in
their antigenic domains, and different materials have been
The non-isotopic immunoassays provide good sensitivity used as standards [17,18]. In addition, even minor cross-
and specificity, and the reagents used are typically stable reactivity of C-peptide antisera with human proinsulin can
for longer periods of time when compared to RIA significantly interfere in C-peptide immunoassays if
techniques [14]. proinsulin secretion still persists in a diabetic patient who
has circulating insulin antibodies induced by insulin
One common non-isotopic technique used for measuring therapy [17,19].
insulin is a two-site immunoenzymometric assay (Table 1,
Method 3). Insulin is bound with mouse monoclonal Routine analysis of C-peptide in the clinical laboratory can
antibody that has been immobilized on a magnetic solid- be performed using a variety of commercial
phase bead. A second enzyme-labeled mouse monoclonal chemiluminescence and enzyme-linked immunoassays [3].
antibody is added, which also binds to the insulin, forming Chromatographic methods have been described but are
an antibody-insulin-labeled antibody sandwich. The solid used primarily in the pharmaceutical industry and in
phase beads are used to remove any unbound enzyme- research. Procedures based on HPLC-MS following solid-
labeled antibody and then incubated with a fluorogenic phase extraction or immunoaffinity chromatography have
substrate. The amount of fluorescence produced is directly been described [20,21].
764
Reference and Preferred Methods Insulin Reference Interval

There are no internationally recognized reference methods There is no single range for insulin in serum. All values
for insulin or C-peptide. must be considered in relation to the clinical state of the
The gold standard method for insulin is a bioassay, since patient, particularly the dynamics of the plasma glucose.
this measures biological activity and thus is of greater
physiological significance. Such methods are used to Insulin concentrations in fasting individuals are usually
standardize pharmaceutical insulin preparations and are less than 25 U/mL (1042 pg/mL, 0.17 pmol/mL). During
only available in specialized laboratories. a glucose tolerance test, the insulin value usually peaks
between 50 and 100 U/mL (2083 to 4167 pg/mL, 0.35 to
For routine insulin assay, non-isotopic immunoassays are 0.7 pmol/L) and follows the glucose curve by about 15
most commonly used. The selected method should not min.
employ an antibody that cross-reacts more than 1% with
C-peptide or growth hormone. With the isotopic assays, if When the serum glucose concentration falls below 500
dextran-coated charcoal is used to separate bound from mg/L (2.8 mmol/L), insulin concentration should be
free insulin, endogenous antibodies bind the tracer, falsely suppressed below 20 U/mL (833.4 pg/mL). During a
lowering the values. Double-antibody separation hypoglycemic episode, the glucose/insulin ratio in normal
techniques may give falsely increased values in the persons is usually greater than 25 U/mL. If the ratio is
presence of endogenous antibodies [22]. For this reason, higher, an insulin-secreting tumor should be suspected.
an appropriate method for removal of endogenous
antibodies should be available. The sensitivity of the assay Interpretation
should be equivalent to at least 10 U/mL (416.7 pg/mL) Serum insulin concentrations can be increased in patients
of sample. For C-peptide measurements, a technique with with noninsulin dependent diabetes mellitus (NIDDM),
a preliminary antibody precipitation is the method of hyperthyroidism, type IV hyperlipoproteinemia, renal
choice. Before selecting a method for routine use, consult failure, liver cirrhosis, and generalized acute illnesses.
van Rijn et al. [16] Decreased serum insulin concentrations are found in
patients with insulin-dependent diabetes mellitus (IDDM)
A method for the detection of insulin antibodies is and can be associated with pheochromocytoma,
sometimes required. For this purpose, the method of malnutrition, and cystic fibrosis. However, the findings of
Gerbitz and Kemmier [23] is recommended. The 2007 increased or decreased insulin levels in these pathological
College of American Pathologists (CAP) Special Ligand states are incidental findings and rarely of diagnostic use.
quality-assurance survey showed that essentially all
laboratories utilized a non-isotopic method for insulin The primary reason for the measurement of serum insulin
analysis. Non-isotopic assays provide good precision and is to evaluate fasting hypoglycemia. Neurological and
accuracy for determination of insulin and C-peptide. behavioral symptoms are frequently seen in patients with
insulin-induced hypoglycemia. The neurological
Specimen symptoms (excessive sweating, palpitation, tachycardia,
Insulin concentrations are identical in serum and tremor, and weakness) are the result of increased
heparinized plasma. The hormone is stable for at least 12 sympathetic nervous system activity with increased
hours at room temperature, for a week at 4C, and for a epinephrine levels. Patients with severe hypoglycemia may
month at 10C. Proinsulin and its cleaved metabolites present in a coma or experience convulsions.
have been reported to have limited stability in human
serum. Use of protease inhibitors to promote stability of The most important cause of repetitive episodes of
the proinsulin-related peptides has been recommended [4]. hypoglycemia is an insulin-secreting tumor (insulinoma)
Increases of up to 75% have been seen in frozen samples or pancreatic hyperplasia. Because surgical procedures can
following long-term storage (>2 years) [24]. Random easily correct this life-threatening condition, it is important
insulin assays are not useful. Specimens for insulin to make the diagnosis quickly. Another frequently seen
analysis should always be collected as part of a planned cause of insulin-induced hypoglycemia is factitious
clinical protocol. hypoglycemia, that is, a self-induced hypoglycemia. In this
Interferences situation, the patients are either purposefully or
The presence of an insulin-degrading enzyme in accidentally overtreating themselves with pharmaceutical
erythrocytes may result in falsely decreased insulin preparations of insulin.
concentrations in specimens that are hemolyzed [25].
Falsely increased results caused by heterophile antibodies Since insulin release from insulinomas may be episodic in
or rheumatoid factors have been observed in nature, the most accurate procedure for evaluation of
immunoassays which use a sandwich technique [26,27]. suspected insulin-induced hypoglycemia is to perform
Serum for insulin determination should be separated from multiple serum insulin and glucose determinations while
the red blood cells within 5 hours following collection the patient is fasting. Fasting periods as long as 72 hours
[28]. have been used. The immunoreactive insulin/glucose
765
(II/G) ratio is then calculated. In healthy, nonobese insulin Immunoassays: Report of the American
persons, this ratio is normally 0.3 (insulin expressed as Diabetes Association workgroup. Clin Chem
U/mL, and glucose in mg/100 mL). If the ratio is greater 2007;53:711-716.
than 0.3, or in obese persons greater than 0.3 and a serum 5 Nisselbaum JS, Fleisher M, Schwartz MK.
glucose less than 600 mg/L, the insulin level is Discrepancies between serum insulin RIA
inappropriately elevated for the corresponding glucose methods in the presence of endogenous
level, and an insulinoma should be considered. After a 48- antibodies. Clin Chem 1977; 23: 1167. (abstract).
hour fast, the II/G ratio has a sensitivity of 85% for 6 Heald AH, Bhattacharya B, Cooper H, Ullah A,
insulinomas. After a 72-hour fast, the sensitivity of the McCulloch A, Smellie S, et al. Most commercial
ratio rises to 95% [29]. insulin assays fail to detect recombinant insulin
analogues. Ann Clin Biochem 2006; 43; 306-308.
There are multiple clinical indications for measurement of 7 Manley SE, Stratton IM, Clark PM, Luzio SD.
C-peptide. Although C-peptide is not used for the routine Comparison of 11 human insulin assays:
monitoring of diabetes, it can be an important tool in implications of clinical investigation and research.
certain clinical situations. In some countries, C-peptide Clin Chem 2007; 53: 922-932.
measurements are used to qualify medical insurance 8 Hinsberg WD III, Milby KH, Zare RN.
coverage for insulin-pump therapy. Measurement of C- Determination of insulin in serum by enzyme
peptide can aid in the differential diagnosis of factitious immunoassay with fluorimetric detection. Anal
hypoglycemia and hypoglycemia caused by Chem 1981; 53: 1509-1512.
hyperinsulinism. Since pharmaceutical insulin preparations 9 Yoshioka M, Taniguchi H, Kawaguchi A,
do not contain C-peptide, an increased serum insulin Kobayashi T. Murakami K. Seki M, et al.
concentration accompanied by a normal C-peptide level Evaluation of a commercial enzyme immunoassay
should raise the suspicion of self-induced hypoglycemia. for insulin in human serum, and its clinical
C-peptide measurements can help assess the success of application. Clin Chem 1979; 25: 35-38.
surgical removal of all or part of the pancreas and islet 10 Meistas MT, Kumar MS, Schumacher OP.
transplantation. Decreased C-peptide concentrations have Diagnosis of self-induced hyperinsulinism in an
been observed in starvation and in Addisons disease. insulin-dependent diabetic patient by
radioimmunoassay of free C-peptide. Clin Chem
Insulin Performance Goals 1981; 27: 184-186.
The Clinical Laboratory Improvement Amendments of 11 Tsuji A, Maeda M, Arakawa H, Shimizu T,
1988 (CLIA-88) evaluation criteria for acceptable Ikegami Y, Sudo, H, et al. Enzyme immunoassay
performance for insulin measurements require that of hormones and drugs by using fluorescence and
laboratories achieve results that are within 3 SD of the chemiluminescence reaction. In: Dal SB, ed.
mean value of all reporting laboratories. The precision of Enzyme labelled immunoassay of hormones and
the various methods commercially available can be drugs. Berlin-New York: Walter de Gruyter &
strikingly different; the interlaboratory coefficient of Co., 327-39, 1978.
variation (CV) ranges from approximately 3% to 25% 12 Yamajuchi Y, Hayaski C, Miyai K: Fluorescence
[30]. polarization immunoassay for insulin
preparations. Anal Letts 1982; 15: 731-737.
The within-individual biological variability for insulin has 13 Toivonen E, Hemmila I, Marniemi J, Jorgensen
been reported as 21% CV, and the within-group variability PH, Zeuthen J and Lovgren T. Two-side time-
for healthy individuals has been reported as 58% CV [31]. resolved immunofluorometric assay of human
Based on these values for biological variability, one can insulin. Clin Chem 1986; 32: 637-640.
derive a desirable allowable measurement bias of 15.5%, 14 Andersen L, Dinesen B, Jorgensen PN, Poulsen F,
imprecision of 10.6%, and total analytical error of 32.0% Roder ME. Enzyme immunoassay for intact
for a single reportable result at insulin concentrations human insulin in serum or plasma. Clin Chem
within or near the normal reference interval. The precision 1993; 39: 578-582.
of commercially available insulin methods meet the 15 Van Uytfanghe K, Rodriguez-Cabaleiro D, Stockl
criteria for acceptable performance. D, Thienpont LM. New liquid chromatography-
References electrospray ionization-tandem mass spectrometry
1 Wahren J, Ekberg K, Jornvall H. C-peptide is a measurement procedure for quantitative analysis
bioactive peptide. Diabetologia 2007;50:503-509. of human insulin in serum. Rapid Commun Mass
2 Yalow RS, Berson SA. Immunoassay of Spectrom 2007; 21: 1-3.
endogenous plasma insulin in man, J Clin Invest 16 van Rijn, H.J.M., Hoekstra, J.B.L., and Thijssen,
1960; 39: 1157-1175. J.H.H.: Evaluation of three commercially
3 Clark PM. Assays for insulin, proinsulin(s) and available C-peptide kits, Ann. Clin. Biochem.
C-peptide. Ann Clin Biochem 1999; 36: 541-564. 19:368-373, 1982.
4 Marcovina S, Bowsher RR, Miller WG, Staten M, 17 Koskinen P. Nontransferability of C-peptide
Myers G. Caudill SP, et al. Standardization of measurements with various commercial
766
radioimmunoassay reagents. Clin Chem 1988; 34: 25 Duckworth WC, Hamel FG, Bennett R, Ryan MP,
1575-1578. Roth RA. Human red blood cell insulin
18 Faber OK, Binder C, Markussen J, Heding LG, degrading enzyme and rat skeletal muscle insulin
Naithani VK, Kuzuya H, et al. Characterization protease share antigenic sites and generate
of seven C-peptide antisera. Diabetes 1978; identical products from insulin. J Biol Chem
27(Suppl 1): 170-177. 1990; 265: 2984-2987.
19 Myrick JE, Gunter EW, Maggio VL, Miller DT, 26 Boscato LM, Stuart C. Heterophilic antibodies: a
Hannon WH. An improved radioimmunoassay of problem for all immunoassays. Clin Chem 1988;
C-peptide and its application in a multi-year 34: 27-33.
study. Clin Chem 1989; 35: 37-42. 27 Highton J, Hessian PA. A solid-phase enzyme
20 Darby SM, Miller ML, Allen RO, LeBeau M. A immunoassay for C-reactive protein: clinical
mass spectrophotometric method for quantitation value and the effect of rheumatoid factor. J
of intact insulin in blood samples. J Anal Toxicol Immunol Methods 1986; 68: 185-192.
2001; 25: 8-14. 28 Walters E, Henley R, Barnes I. Stability of
21 Rogatsky E, Balent B, Goswami G, Tomuta V, insulin in normal whole blood. Clin Chem 1986;
Jayatillake H, Cruikshank G et al. Sensitive 32: 224 (letter).
quantitative analysis of C-peptide in human 29 Ravel, R. Clinical laboratory medicine, ed. 3,
plasma by two-dimensional liquid Chicago, 1978, Year Book Medical Publishers, p.
chromatography-mass spectrometry isotope- 330.
dilution assay. Clin Chem 2006; 52: 872-879. 30 College of American Pathologists Special Ligand
22 Feldkamp CS, Chapin E, Shearer G. Blanking in Participant Summary Report Y-B, Washington,
insulin radioimmunoassay. Clin Chem 1977; 23: DC, 2001, CAP.
1167 (abstract). 31 Fraser CE. Biological variation: From principles
23 Gerbitz KD, Kemmier W. Method for rapid to practice. Washington, DC: AACC Press, 2001.
quantitation and characterization of insulin 32 Greenwood FC, Hunter WM, Glover JS. The
antibodies. Clin Chem 1978; 24: 890-894. preparation of I-131-labeled human growth
24 Feldman JM, Chapman BA. Radioimmunoassay hormone of high specific radioactivity. Biochem J
of insulin in serum and plasma. Clin Chem 1973; 1963; 89: 114-123.
19: 1250-4.
Tables
Table 1: Methods for Insulin Analysis
Principle of analysis: Radioactive iodine (125I)-labeled insulin competes with sample insulin for binding sites on
anti-insulin antibodies. Separation of bound from free ligand can be accomplished by double-antibody precipitation,
solid phase, and so on.
Comments: Most frequently used; can be affected by endogenous anti-insulin antibodies
Method 2: Enzyme immunoassay (EIA)
Principle of analysis: Employs sandwich technique, in which second antibody is coupled to horseradish
peroxidase. Reaction products can be measured spectrophotometrically (500 nm) or fluorometrically (excitation, 317
nm; emission, 414 nm)
Comments: Research; as sensitive as RIA
Method 3: Immunoenzymometric assay (IEMA)
Principle of analysis: Insulin bound to antibody immobilized to magnetic solid phase support. Addition of second
antibody labeled with ALP added to form sandwich. Fluorogenic substrate added following wash step to remove
unbound labeled antibody
Comments: Frequently used; interference from endogenous anti-insulin antibodies and human anti-mouse antibodies
possible
Method 4: Microparticle enzyme immunoassay (MEIA)
Principle of analysis: Similar to Method 2, except insulin bound to anti-insulin-coated microparticles transferred to
glass fiber matrix where anti-insulin - ALP conjugate is added to form sandwich complex. 4-methylumbelliferyl
added as substrate for ALP
Comments: Frequently used; same as for Method 3
767
Table 2: Assay Conditions for Radioimmunoassay of Insulin (Ref 16)

Condition RIA
Reaction temperature 4C
Sample volume 100 L
Assay time Approximately 20 h
Final concentration of reagents 125I-insulin: 0.2 ng/mL (40,000 cpm)
Phosphate buffer (pH 7.5): 40 mol/mL

Albumin: 2 mg/mL
Sensitivity Approximately 10 U/mL
Linearity 10288 U/mL
Interferences Anti-insulin antibodies
periodically, and adjust if necessary.
Procedure: Radioimmunoassay for Insulin Dissolve 1195 g of NaH2PO42H2O in about 200 mL of
Antibody Precipitation
deionized water and bring to a volume of 250 mL. Use this
Principle solution to adjust the pH of the phosphate-BSA solution to
Polyethylene glycol is added to serum or plasma in a exactly 7.50.
concentration that causes precipitation of IgG (and 2. Dextran-coated charcoal. Click here for
therefore endogenous antibodies). description.
3. Insulin antiserum. Antiserum to porcine insulin,
produced in guinea pigs, is available from New England
Reagent Nuclear (549 Albany St., Boston, MA 02118). For use,
1. Polyethylene glycol (250 g/L). Dissolve 2.5 g of dilute the antiserum in phosphate-BSA buffer to produce
PEG 8000 (7000 to 9000 D) (Fisher Scientific Carbowax 40% to 60% binding of tracer.
PEG 8000), and bring to 10 mL with deionized water. 4. Stock standard (10 g/mL; 0.24 U/mL, 1.67
nmol/mL). Dissolve 10.0 mg of crystalline porcine insulin
(Sigma 13505, 24 U/mg) in 0.01 mol/L of HCl, and dilute
Assay to 10 mL with phosphate-albumin buffer. Remove 1.00
1. Combine 1.0 mL of serum (or plasma) with 1.0 mL of this solution, and bring to 100 mL with phosphate-
mL of PEG solution. albumin buffer. This solution is stable for 1 month stored
2. Mix on a vortex mixer. at 70C in 1 mL aliquots and for 7 days at 4C after
3. Centrifuge at 3000 rpm (approximately 1000 g) at thawing.
4C for 15 min. 5. Working standards. Each day, prepare a series of
standards by diluting 250 L of stock standard to 50 mL
4. Remove supernatant, which may now be used as a
with phosphate-BSA to produce a 1200 U/mL (50
sample in the insulin assay. The insulin level
ng/mL) standard. The 1200 U/mL standard is then diluted
obtained on the supernatant must be multiplied by
with phosphate-BSA according to the following values:
2.
Insulin Assay [10,14,16,17] Concentration mL of 1200 Final Volume

(U/mL) U/mL standard (mL)
Principle 9.6 0.200 25
A guinea pig antiserum to porcine insulin is incubated with 19.2 0.400 25
the test serum and 125I-labeled human insulin. Separation 28.8 0.600 25
is achieved with dextran-coated charcoal. 48.0 0.400 10
72.0 0.600 10
96.0 0.800 10
Reagents 144.0 1.200 10
1. PhosphateBSA buffer, 0.05 mol/L, pH 7.5. 192.0 1.600 10
Dissolve 8.90 g of Na2HPO42H2O in about 500 mL of 288.0 2.400 10
deionized water using mild heating. Add 2.5 g of bovine
serum albumin (BSA), and slowly mix to dissolve. Adjust 6. Tracer (125I-labeled insulin). Obtain iodinated
pH to 7.5 using NaH2PO4 solution described next. Bring insulin from New England Nuclear (Boston, MA) (NEX-
volume to 1000 mL with deionized water. Stable for 3 104), or label porcine insulin using the method of
months at refrigerated temperatures. Check pH Greenwood et al. [32]. Dilute the tracer in phosphate-BSA
to produce a minimum of 20,000 cpm (counts per min) in
768
200 L with approximately 0.1 ng of insulin. This is stable 6. Add 2.0 mL of well-mixed dextran-coated
for 48 h at 4C when diluted. charcoal. Mix, and centrifuge at 1000 g for 15 min at 4C.
Decant, and count the supernatant.
Assay
Equipment: A gamma-ray counter that will record 125I Calculations
and a refrigerated centrifuge capable of generating 1000 g Subtract the NSB counts from all tubes. Compute for each
force. specimen and standard as follows:
1. Combine 200 L of antiserum and 200 L of
labeled insulin tracer in a series of 12 100 mm tubes __Counts NSB_ 100% = B/T
sufficient for standards, controls, unknown samples, and Total count NSB
background counts.
2. Add 100 L of standards, controls, or unknown where B = antibody-bound counts
sera. T = total counts nonspecific binding
3. Prepare total-count (TC) and nonspecific binding
(NSB) tubes by combining 200 L of tracer and 300 L of Plot % B/T versus concentration.
phosphate-BSA. Save TC tube for measurement of total
counts, and process NSB tube with samples. Note: If the value of the blank tubes is significantly greater
4. A blank tube may be prepared for any sample by than that of the NSB tubes, an endogenous insulin
substitution of 200 L of phosphate-BSA for antiserum. antibody should be suspected.
5. Incubate overnight at 4C (18 h).
769
Ionized Calcium
John G. Toffaletti
Name: Ionized calcium
Atomic symbol: Ca2+
Merck Index: 1613
Chemical class: Alkaline earth element, ionic form
Principles of Analysis and Current Usage All modern calcium-selective electrodes are based on
similar designs and principles. Differences between
Calcium in serum or plasma exists in a variety of forms instruments are found mostly in the flow systems,
as a result of the binding equilibria between calcium ions method of calibration, control materials, and types of
and both high- and low-molecular-mass anions. The ion-selective membranes used. A basic diagram of such
term complex-bound calcium describes the calcium an electrode is shown in Figure 1.
bound to low-molecular-mass anions such as
bicarbonate, citrate, and phosphate [1]. Complex-bound The ion-selective membrane is critical to the specificity
calcium constitutes from 10% to 15% of the total serum of the electrode. It must selectively interact with calcium
calcium. The protein-bound calcium exists ions in the presence of other ions such as magnesium,
predominately as calcium-albumin and constitutes potassium, hydrogen, and sodium. These membranes
approximately 40% of the total calcium [2]. The consist of a calcium-ion sensor and a solvent mediator.
proportions of each form of calcium detected by ionized, The sensor is a molecule with high selectivity for
ultrafiltrable, and total calcium methods is shown in calcium that is dissolved in the solvent mediator. The
Table 1. mediator functions to disperse and align the calcium
sensor to optimize the affinity and selectivity of the
The unassociated calcium ions, hereafter referred to as sensor for calcium ions. The sensor-solvent mediator is
ionized calcium, represent 45% to 50% of the total often incorporated into a polyvinyl chloride (PVC)
calcium. The equilibria between these forms are affected matrix that gives structural support to the membrane.
by pH, temperature, ionic strength, and magnesium-ion The sensor and solvent mediator must be insoluble in
concentration, as well as by the addition or removal of aqueous solutions.
calcium from blood. The pioneering work of McLean
and Hastings on the effect of calcium on frog heart The calcium sensors are usually either ion exchangers or
muscle contraction [3], established that the ionized neutral carriers. Ion exchangers are calcium salts, usually
calcium in body fluids is the most physiologically active of an organophosphate. Neutral carriers are organic
form of calcium [4]. The concentration of ionized molecules without net charge that form a molecular
calcium in blood affects many functions, including pocket with a favorable steric and electrostatic
excitation and contraction of cardiac and skeletal muscle environment for binding calcium ions. These sensors are
[5,6], neurotransmission, and secretion of parathyroid illustrated in Figure 2.
hormone (PTH) [7-9].
The ion-selective membrane functions by allowing
Although methods based on bioassay or ultrafiltration calcium ions to bind to the sensors in the membrane. If
have been developed in the past to measure ionized or calcium ions are unequally distributed on the internal
ultrafiltrable calcium, ion-selective electrodes (Table 2, and external surfaces of the membrane, an electric
Method 2) are virtually the only routine method in use potential will develop across the membrane. Since the
today. Current commercially-available electrodes are internal electrode solution has a constant ionized calcium
relatively durable, easily serviced, suitable for use with concentration, variations in the membrane potential are
serum and blood, and automated for precision, related to the ionized calcium concentration (activity) of
convenience, and rapid results in clinical work. the sample. An inert cellophane membrane has been
used to prolong electrode life by preventing direct
contact between protein and lipids in the sample and the
calcium sensor in the electrode. However, such
i membranes may increase the response time of the
Ionized Calcium
Previous and current authors of this method: electrode.
Methods edition: John G. Toffaletti A potential develops at the junction of the reference-
Second edition: Not updated electrode solution and the sample, called the liquid-
Third edition: Not updated junction potential, as shown in Figure 1. If all samples
Fourth edition: Not updated and calibrators produced the same junction potential, it
Fifth edition: John G. Toffaletti would be of no consequence. However, whole blood
770
samples can develop a liquid-junction potential that is Specimens should be collected anaerobically, because
different from that developed by serum or blood. CO2 is lost when blood is exposed to atmospheric air,
Manufacturers of ionized calcium analyzers have which causes pH to increase. As mentioned, a lower pH
designed their analyzers to minimize this effect. (more H+ ion) will increase the concentration of ionized
Examples include: (1) allowing time for the sample to calcium in serum, and a higher pH will have the opposite
2+
stabilize while in contact with the Ca electrode before effect [15]. If strictly anaerobic sampling or handling is
electrical contact is made with the reference electrode, not possible, reporting the calculated ionized calcium at
then reading the potential immediately, before the pH 7.4 is an alternative that gives results within 3% to
junction potential is fully developed; and (2) using a 4% of the uncorrected result on fresh sample if the pH
high concentration of anions and cations with similar correction is not too large.
ionic mobilities in the reference-electrode solution.
The patient should be seated or supine and relaxed when
Reference Methods blood is drawn, tourniquet application should be as brief
Although not designated as an official reference method as possible, and the tourniquet should remain on the arm
for ionized calcium, a designated comparison method until all blood is withdrawn. Although posture influences
was published and later adopted by the National total calcium more than ionized calcium, standing can
Committee on Clinical Laboratory Standards (NCCLS) increase ionized calcium concentration by 1% to 2%
as a means to achieve comparability of analytical results [16]. Prolonged venous stasis can also increase ionized
among the commercially available calcium ion-selective calcium by about 2%, and a few minutes of forearm
electrodes [10]. exercise with stasis can elevate ionized calcium by 8%
because of localized acid production [17]. Bedrest for 3
Specimen to 12 days has been shown to elevate ionized calcium
For measurement of ionized calcium in clinical samples, into the abnormal range [18]. This should be considered
the use of heparinized whole blood collected when one is interpreting results on patients who are
anaerobically is preferable and is far more prevalent than completely bedridden.
use of serum. Heparinized whole blood may be analyzed
immediately (no clotting and centrifugation is needed) Interferences
and provides nearly double the usable sample compared Anticoagulants that strongly bind calcium ions, such as
to serum, an especially important factor when analyzing citrate, oxalate, or EDTA, significantly lower the ionized
blood from neonates. A concern with ordinary sodium calcium concentration. Heparin, the most widely-used
heparin is that the heparin anion binds to calcium ions, anticoagulant for ionized calcium measurements, is a
which alters the ionized calcium concentration. weaker binder of calcium ions. Ordinary sodium or
However, modern syringes are readily available that lithium heparin at a concentration of 30 IU/mL in plasma
contain either a calcium-balanced heparin or a very low lowers ionized calcium by about 3% to 5%. To minimize
amount of heparin dispersed in a polysaccharide web, the binding effect of heparin, a variety of modified
either of which has minimal effect on the ionized heparin salts have been used. For many years, calcium-
calcium concentration. Although serum contains no balanced or electrolyte-balanced heparins have been
anticoagulant to bind calcium, the clotting process has available that virtually eliminate any effect on ionized
variable effects on the pH and ionized calcium [11,12]. calcium in either plasma or whole blood. This type of
Perhaps the only remaining advantage of serum is that it heparin has small amounts of calcium and other
is more stable in storage if analysis has to be delayed. electrolytes added that effectively occupy the binding
With modern blood-gas/electrolyte/metabolite analyzers sites on heparin and prevent further binding in the blood
that analyze whole blood, the need for a rapid ionized sample. Although this type of anticoagulant virtually
calcium result as part of this panel of tests has also made eliminates the effect of heparin over a wide range of
whole blood the predominant sample for clinical use. ionized calcium concentrations, samples with very low
ionized calcium concentrations will have a slight
Ideally, samples should be analyzed as soon as possible, positive bias, and samples with very high ionized
preferably within 1 hour after collection. If this is not calcium concentrations will have a slight negative bias
possible, the ionized calcium concentration in either [12]. Another approach has been to use a very low
clotted or heparinized blood stored in sealed tubes is amount of heparin dispersed in a soluble polysaccharide
stable for 6 hours or more at 4C if in vitro hemolysis is web which appears as a puff of anticoagulant in the
not evident [13]. This stability is fortuitous, however, syringe. The final concentration of heparin is so low
because the effects of lactate and acid production during (~2.5 IU/L) that the interference to ionized calcium is
storage apparently offset the effects of each other on the effectively eliminated at all concentrations. The
ionized calcium concentration: the hydrogen ion releases polysaccharide web dissolves very rapidly to quickly
calcium ions from albumin while lactate anions chelate disperse the heparin in the blood sample.
about the same amount of free calcium ions [14].
Therefore, although it is desirable to analyze samples as Heparin affects ionized calcium slightly less in whole
soon as possible, whole blood stored at 4C for several blood than in plasma because red cells can artifactually
hours may still be acceptable. Serum kept in tightly increase the ionized calcium concentration by their effect
sealed syringes, with no air bubbles present, is stable for on the liquid junction potential. Syringes and evacuated
1 to 2 hours at room temperature and over 24 hours at collection tubes of whatever size should be filled as
4C. completely as possible and kept tightly sealed to
771
minimize loss of CO2, which increases pH, and to the ICU. Hypocalcemia is common in ICU patients who
prevent an undesirably high concentration of heparin develop sepsis or renal, cardiac, or pulmonary failure or
anticoagulant. in patients with thermal burns. Total calcium
measurements frequently have little meaning, because
It should be noted again that serum is less than ideal for serum protein concentrations are reduced, acid-base
collection of ionized calcium samples. Serum has no disturbances are present, and citrated blood products
anticoagulant to bind calcium ions, but the clotting have been given. If calcium is administered,
process apparently has variable and unpredictable effects measurement of ionized calcium provides a guide to
on the pH and ionized calcium when compared to whole proper dosage in these patients [28].
blood collected without any anticoagulant and analyzed
immediately [11,12]. Critically ill patients, such as those with infections and
sepsis, pancreatitis, burns, and major trauma are all
Ionized Calcium Reference Intervals susceptible to hypocalcemia. Because of its importance
Depending on the type of anticoagulant used, the in maintaining cardiac output, arterial pressure, and
population studied, and to some extent the analyzer used, systemic vascular resistance, adequate calcium
the lower reference limit for ionized calcium may range concentrations are especially important, and guidelines
from 1.13 to 1.16 mmol/L and the upper reference range have been reported on administering calcium [29]. A low
from 1.30 to 1.32 mmol/L. The above values are for ionized calcium concentration has been associated with
adults. Values for children appear to be about 0.05 increased mortality in patients admitted to the ICU
mmol/L higher.
[30,31] and in patients with pancreatitis [32]. Patients
with sepsis commonly have decreased blood ionized
The above intervals are intended as guidelines that
should be confirmed by data for the specific instrument calcium concentrations [33], and the mechanism of
and population served by the hospital. Variables that hypocalcemia may be related to the production of
may influence reference intervals are the concentration various cytokines. These patients, especially if an
and type of anticoagulant used, sample handling infection is present, also have low or suppressed
practices, and the design and calibration system used by parathyroid hormone (PTH) levels. In sepsis, factors
the manufacturer of the electrode. Values for serum, such as TNF, IL-6, and CRP were all markedly elevated
plasma, and whole blood differ; whole blood is the usual on the first day of admission to the ICU [34]. It should
sample collected and analyzed. also be noted that yet another study reported that
albumin-corrected total calcium measurements are not
Interpretation suitable for diagnosis of calcium disorders in the
Changes in blood electrolyte concentrations are common critically ill [35]. Interestingly, one report claims that
during major surgical operations involving the heart, ionized magnesium can be inferred from total
liver, and abdomen. Ionized calcium is often measured
magnesium, but that ionized calcium must be measured
during these procedures to monitor and control blood
levels, with decisions about calcium supplementation directly [28].
based on these measurements. Administration of
heparin, citrate, bicarbonate, drugs, protein solutions, In neonates, ionized calcium measurements are
along with alterations in blood pH and body temperature, especially important for following their calcium status.
can significantly affect the calcium concentration in Neonatal hypocalcemia appears to be a normal
blood. Because of alterations in blood pH and protein occurrence during the first week of life and may act as a
concentration, total calcium measurements are of little stimulus to induce the infants parathyroid glands to
use in these circumstances [19]. Several recent studies become functional [36]. Typically, no calcium
emphasize the importance of monitoring ionized calcium supplementation is necessary or even beneficial [37].
instead of total calcium during surgery and recovery [20- However, in preterm infants or neonates with severe or
24]. With no supplemental calcium or magnesium given prolonged hypocalcemia, calcium supplementation may
during the procedure, ionized calcium was found to be beneficial [38,39]. In neonates with seizures,
decrease by up to 20% during cardiac surgery, and both hypocalcemia and hypomagnesemia should be
ionized and total magnesium decreased by considered as causes [40]. In critically ill children,
approximately 25% [25]. ionized hypocalcemia was both common and associated
with a higher mortality [41]. Ionized calcium measured
Hypocalcemia appears to be a frequent complication in by point-of-care analyzers may be of special importance
trauma patients, with contributing factors being colloid- in critically ill children when lab results are not available
induced hemodilution, severe shock, pH disturbances rapidly [42].
and elevated lactate [26]. More importantly, an ionized
calcium on admission of less than 1.00 mmol/L was In renal disease, after renal transplantation, or in patients
associated with greater injury severity, shock, longer on hemodialysis, calcium metabolism is often altered,
ICU stays, greater need for ventilation, and higher sometimes profoundly. During hemodialysis,
mortality [27]. maintaining a slight positive calcium balance is
important for maintaining good cardiac contractility and
Perhaps the most prevalent use of ionized calcium blood pressure [43,44], and ionized calcium
measurements is for monitoring critical care patients in measurements are probably the best means to monitor
772
this, since total or corrected total calcium values do not 4 McLean FC, Hastings AB. Clinical estimation
agree with directly measured ionized calcium in many and significance of calcium-ion concentrations
patients with renal disease [45]. The levels of PTH and in the blood. Am J Med Sci 1935;189:601-613.
ionized calcium may be important determinants in long- 5 Olinger GN, Hottenrott C, Mulder DG,
term survival of patients on dialysis [46,47]. Maloney JV Jr, Miller J, Patterson RW et al.
Acute clinical hypocalcemic myocardial
For the differential diagnosis of altered calcium depression during rapid blood transfusion and
metabolism in patients with hyperparathyroidism, postoperative hemodialysis: a preventable
hypoparathyroidism, malignancy, or renal disease, complication, J Thorac Cardiovasc Surg
ionized calcium measurements may be paired with PTH 1976;72:503-511.
measurements to provide an important differential for 6 Drop LJ, Scheidegger D. Plasma ionized
the cause of the calcium abnormality [48-51]. calcium concentration: important determinant
of the hemodynamic responses to calcium
Ionized calcium measurements can be useful in acute infusion. J Thorac Cardiovasc Surg
pancreatitis, during which ionized calcium typically 1980;79:425-431.
decreases during the first 24 hours of an attack and 7 Chambers DJ, Dunham J, Braimbridge MV,
return to normal by 48 hours [52]. Because both total Slavin B, Quiney J, Chayen J. The effect of
calcium and albumin also decrease, hypocalcemia may ionized calcium, pH, and temperature on
be missed by measurements of either the total or bioactive parathyroid hormone during and after
corrected total calcium. This is significant because open-heart operations. Ann Thorac Surg
persistently low ionized calcium has been associated 1983;36:306-313.
with severe pancreatitis [32]. 8 Toffaletti J, Nissenson R, Endres D, McGarry
E, Mogollon G. Influence of continuous
Ionized calcium has been reported to be increased in a infusion of citrate on responses of
higher percentage of patients with malignancy than total immunoreactive parathyroid hormone, calcium
calcium [53,54]. In general, whenever the cause of and magnesium components, and other
hypercalcemia cannot be readily determined, malignancy electrolytes in normal adults during platelet
must be considered. In this case, additional measurement apheresis. J Clin Endocrinol Metab
for PTH, cyclic AMP, or phosphate may be useful. 1985;60:874-879.
9 Toffaletti J, Cooper DL, Lobaugh B. The
Ionized Calcium Performance Goals response of parathyroid hormone to specific
Kost has published critical limits for ionized changes in either ionized calcium, ionized
hypocalcemia of 0.82 +0.14 mmol/L and for ionized magnesium, or protein-bound calcium in
hypercalcemia of 1.55 +0.19 mmol/L [55]. humans. Metabolism 1991;40:814-818.
Consequently, performance goals for ionized calcium 10 DOrazio P, Bowers GN Jr. Design and
would be standard deviations (SD) of 0.03 to 0.05 preliminary performance characteristics of a
mmol/L in the low and normal range and within 0.08 newly proposed reference cell for ionized
mmol/L in the upper range. Review of recent CAP calcium in serum. Clin Chem 1992;38:1332-
quarterly summaries for ionized calcium indicates that 1339.
each major brand of analyzer meets these goals. For four 11 Toffaletti J, Ernst P, Hunt P, Abrams B. Dry
control materials (AQ-01, AQ-02, AQ-11, and AQ-12) electrolyte-balanced heparin syringes evaluated
with means of 0.96, 0.73, 1.90, and 1.30 mmol/L, for determining ionized calcium and other
respectively, the SDs among the major brands of electrolytes in whole blood. Clin Chem
analyzers ranged from 0.027 to 0.052, and the CVs 1991;37:1730-1733.
ranged from 1.9% to 4.3%. As a specific example, for 12 Toffaletti J, Wildermann RF. The effects of
the AQ-12 material with a mean of 1.30 mmol/L, the heparin anticoagulants and fill volume in blood
SDs ranged from 0.026 to ,0.035 and the CVs ranged gas syringes on ionized calcium and
from 1.9% to 2.6%. Thus even for the analyzer with the magnesium measurements. Clin Chim Acta
highest CV, 99% of the results for this control material 2001;304:147-151.
would range from 1.20 to 1.40 mmol/L. 13 Toffaletti J, Blosser N, Kirvan K. Effects of
storage temperature and time before
References centrifugation on ionized calcium in blood
1 Toffaletti J, Savory J, Gitelman HJ. Separation collected in plain Vacutainer tubes and silicone-
and quantitation of serum constituents separator (SST) tubes. Clin Chem 1984;30:553-
associated with calcium by gel filtration. Clin 556.
Chem 1976;22:1968-1972. 14 Toffaletti J, Abrams B. Effects of in vivo and in
2 Toffaletti J, Savory J, Gitelman HJ. Use of gel vitro production of lactic acid on ionized,
filtration to examine the distribution of calcium protein-bound, and complex-bound calcium in
among serum proteins. Clin Chem blood. Clin Chem 1989;35:935-938.
1977;23:2306-2310. 15 Wang S, McDonnell EH, Sedor FA, Toffaletti
3 McLean FC, Hastings AB. A biological method JG. pH effects on measurements of ionized
for the estimation of calcium ion concentration. calcium and ionized magnesium in blood. Arch
J Biol Chem 1934;107:337-350. Pathol Lab Med 2002;126:947-950.
773
16 Renoe BW, McDonald JM, Ladenson JH. 30 Hastbacka J, Pettila V. Prevalence and
Influence of posture on free calcium and related predictive value of ionized hypocalcemia
variables. Clin Chem 1979;25:1766-1769. among critically ill patients. Acta Anaesthiol
17 Renoe BW, McDonald JM, Ladenson JH. The Scand 2003;47:1264-1269.
effects of stasis with and without exercise on 31 Ward RT, Colton DM, Meade PC., Henry J-C,
free calcium, various cations, and related Contreras L-M, Wilson OM, Fleming AW.
parameters. Clin Chim Acta 1980;103:91-100. Serum levels of calcium and albumin in
18 Heath H III, Earll JM, Shaaf N, Piechocki JT, survivors versus nonsurvivors after critical
Li TK. Serum ionized calcium during bed rest injury. J Crit Care 2004;19:54-64.
in fracture patients and normal man. 32 Kawa S, Mukawa K, Kiyosawa K.
Metabolism 1972;21:633-640. Hypocalcemia <7.5 mg/dL: early predictive
19 Jawan B, deVilla V, Luk HN, Chen YS, Chiang marker for multisystem organ failure in severe
YC, Wang CC et al. Ionized calcium changes acute necrotizing pancreatitis proposed by the
during living-donor liver transplantation in study analyzing post-ERCP pancreatitis. Am J
patients with and without administration of Gastroenterol 2000;95:1096-1097.
blood-bank products. Transpl Int 2003;16:510- 33 Zaloga G. Ionized calcium during sepsis. Crit
514. Care Med 2000;28:266-268.
20 Abbott TR. Changes in serum calcium fractions 34 Lind L, Carlstedt F, Rastad J, Stiernstrom H,
and citrate concentrations during massive blood Stridsberg M, Ljundggren O et al.
transfusions and cardiopulmonary bypass. Br J Hypocalcemia and parathyroid hormone
Anaesth 1983;55:753-760. secretion in critically ill patients. Crit Care Med
21 Jankowski S, Knighton J, Dunnill R, Dickson 2000;28:93-99.
D. Changes in ionized calcium concentrations 35 Slomp J, van der Voort PH, Gerritsen RT, Berk
and acid-base status during abdominal aortic JA, Bakker AJ. Albumin-adjusted calcium is
vascular surgery. Brit J Anaesth 1998;81:965- not suitable for diagnosis of hyper- and
967. hypocalcemia in the critically ill. Crit Care Med
22 Catinella FP, Cunningham JN, Strauss ED, 2003;31:1389-1393.
Adams PX, Laschinger JC, Spencer FC. 36 Wieland P, Duc G, Binswanger G, Fischer JA.
Variations in total and ionized calcium during Parathyroid hormone response in newborn
cardiac surgery. J Cardiovasc Surg infants during exchange transfusion with blood
1983;24:593-602. supplemented with citrate and phosphate: effect
23 Svagzdiene M, Sirvinskas E. Changes in serum of IV calcium. Pediatr Res 1979;13:963-968.
electrolyte levels and their influence on the 37 Brown DR, Sterauka BH, Taylor FH. Treatment
incidence of atrial fibrillation after coronary of early-onset neonatal hypocalcemia: effects
artery bypass grafting surgery. Medicina on serum calcium and ionized calcium. Am J
(Kaunas) 2006;42:208-214. Dis Child 1981;135:24-28.
24 Pedersen KO, Juhl O. Blood ionized calcium 38 Wandrup J. Critical analytical and clinical
measurements during aortocoronary bypass aspects of ionized calcium in neonates. Clin
graft operations. Scand J Clin Lab Invest Chem 1989;35:2027-2033.
1983;43(suppl 165):107-109. 39 Dilena BA, White GH. The response of plasma
25 Inoue S, Akazawa S, Nakaigawa Y. Changes in ionised calcium and intact parathyrin to calcium
plasma total and ionized magnesium supplementation in preterm infants. Acta
concentrations and factors affecting magnesium Paediatr Scand 1991;80:1098-1100.
concentrations during cardiac surgery. J Anesth 40 Sheth DP. Hypocalcemic seizures in neonates.
2004;18:216-219. Am J Emerg Med 1997;15:638-641.
26 Vivien B, Langeron O, Morell E, Devilliers C, 41 Singhi SC, Singh J, Prasad R. Hypocalcaemia
Carli PA, Coriat P, Riou B. Early hypocalcemia in a paediatric intensive care unit. J Trop
in severe trauma. Crit Care Med 2005;33:1946- Pediatr 2003;49:298-302.
1952. 42 Vos G, Engel M, Ramsay G, Waardenburg D.
27 Cherry RA, Bradburn E, Carney DE, Shaffer Point-of-care analyzer during the interhospital
ML, Gabbay RA, Cooney RN. Do early ionized transport of critically ill children. Eur J Emerg
calcium levels really matter in trauma patients? Med 2006;13:304-307.
J Trauma 2006;61:774-779. 43 Henrich WL, Hunt JM, Nixon JV. Increased
28 Koch SM, Warters RD, Mehlhorn U. The ionized calcium and left ventricular contractility
simultaneous measurement of ionized and total during hemodialysis. N Engl J Med
calcium and ionized and total magnesium in 1984;310:19-23.
intensive care unit patients. J Crit Care 44 Leunissen KML, van den Berg BW, van Hoof
2002;17:203-205. JP. Ionized calcium plays a pivotal role in
29 Vincent J-L, Jankowski S. Why should ionized controlling blood pressure during
calcium be determined in acutely ill patients? haemodialysis. Blood Purif 1989;7:233-239.
Acta Anaesthesiol Scand 1995;39(suppl 45 Clase CM, Norman GL, Beecroft ML,
107):281-286. Churchill DN. Albumin-corrected calcium and
ionized calcium in stable haemodialysis
774
Ionized Calcium
patients. Nephrol Dial Transplant and total calcium in hypercalcemic conditions. J

2000;15:1841-1846. Clin Endocrinol Metab 1979;48:393-397.
46 Avram MM, Sreedhara R, Avram DK, 51 Boyd JC, Lewis JW, Slatopolsky E, Ladenson
Muchnick RA. Enrollment PTH level is a new JH. Parathyrin measured concurrently with free
marker of survival in hemodialysis and or total calcium in the differential diagnosis of
peritoneal dialysis therapy for uremia. Am J hypercalcemia. Clin Chem 1981;27:574-579.
Kidney Dis 1996;28:924-930. 52 Croton RS, Warren RA, Stott A, Roberts NB.
47 Madelaine P, Aquiles J, Bover J, Rodriguez M, Ionized calcium in acute pancreatitis and its
Flesenfeld AJ. The set point of calcium and the relationships with total calcium and serum
reduction of parathyroid hormone in lipase. Brit J Surg 1981;68:241-244.
hemodialysis patients. Kidney Int 1996;49:226- 53 Toffaletti J. Total, ultrafiltrable, dialyzable, and
231. ionized calcium measurements in randomly
48 Monchick JM, Martin HF. Ionized calcium in selected samples compared with clinical
the diagnosis of primary hyperparathyroidism. diagnoses. Ann Clin Biochem 1983;20:349-
Surgery 1980;88:185-192. 352.
49 McLeod ML, Monchik JM, Martin HF. The 54 Shemerdiak WP, Kukreja SC, Lad TE, York
role of ionized calcium in diagnosis of subtle PAJ, Henderson WJ. Evaluation of routine
hypercalcemia in symptomatic primary ionized calcium determinations in cancer
hyperparathyroidism. Surgery 1984;95:667- patients. Clin Chem 1981;27:1621-1622.
673. 55 Kost GJ. The significance of ionized calcium in
50 Ladenson JH, Lewis J, McDonald JM, cardiac and critical care. Arch Pathol Lab Med
Slatopolsky E, Boyd JC. Relationship of free 1993;117:890-896.
775
Table 1: Detection of Calcium Components by Various Methods

Calcium Component Ion-Selective Electrode: Ultrafiltrable Total
Ionized Calcium (%) Calcium (%) Calcium (%)
Calcium ions 100 100 100
Calcium phosphate 0 100 100
Calcium bicarbonate 0 100 100
Calcium citrate 0 100 100
Calcium albumin 0 0 100
Table 2: Methods for Free Calcium Measurements

Serum, plasma, and whole blood:
Method 1: Biological assay; stimulation of muscle contraction
Principle of analysis: A heart is perfused with solutions containing Ca2+. The amplitude of
contraction is directly related to the concentration of Ca2+.
Comments: Historically important but an extremely tedious method
Method 2: Ion-selective electrode; potentiometry
Principle of analysis: Calcium ion-selective membrane specifically binds calcium ions, creating an
electrochemical potential across the membrane that is proportional to Ca2+ concentration.
Comments: A rapid and durable method for both stat and routine analyses; widely used in
commercially available instruments; some give pH corrected result
Serum or plasma:
Method 3: Ultrafiltrable; ultrafiltration with analysis by atomic absorption or other means
Principle of analysis: Pressure (mechanical or centrifugal) causes protein-free filtrate to pass through
ultrafilter; calcium measured in ultrafiltrate.
Comments: A time-consuming method that may be suited for laboratories without specialized
equipment for measuring ionized calcium; detects both calcium ions and all calcium complexes.
776
Figure 1: Basic design of calcium ion-specific electrode system.

The calcium-ion sensor develops a potential proportional to the ionized calcium concentration, while the reference sensor
maintains a stable potential independent of the sample composition. Both electrodes maintain a stable internal filling
solution so that the only variable read by the voltmeter is the potential across the ion-selective membrane in contact with
the blood sample. (Courtesy of Nova Biomedical.)
Figure 2: Calcium-ion sensors.

777
Figure 3: Typical series arrangement of analyte-specific electrodes in a modern blood gaselectrolyte analyzer.
Peristaltic pumps and valves automatically flow the blood sample to the electrodes while measurements are taken, pump
the blood out to waste, then clean the electrodes with a rinse solution. A reference electrode is used to measure the
potentials between it and each analyte-specific electrode.
Figure 4
Figure 4. Radiometer ionized calcium electrode design

(cross-sectional diagram)
Ag rod coated with AgCl

- provides electrical contact to buffer solution
- Ag/Ag+ equilibrium maintains a stable
potential
Electrolyte solution
Porous pin
- absorbs electrolyte solution
- holds PVC membrane
PVC membrane
- contains a Ca++ ion exchanger
Cellophane membrane
- protects PVC membrane
- prevents protein build-up
- held in place by white plastic gasket
Figure 4: Diagram of the Radiometer ionized calcium electrode.

A chloridized silver wire provides a stable potential within this half-cell. Note that this electrode design uses a cellophane
membrane to protect the ion-selective membrane (PVC with ion exchanger). (Courtesy of Radiometer America.)
778
Figure 5. IL GEM ionized calcium sensor

(cross-sectional diagram)
Card
Reference
Electrode
Figure 5: Diagram of the IL GEM ionized calcium electrode.

This electrode system uses the ionophore ETH 1001 in a PVC membrane deposited on a plastic card which contains
miniature sensors for other analytes in a multi-use disposable cartridge. A hydrogel electrolyte matrix is deposited over an
Ag/AgCl wire, followed by a coat of ion-selective membrane. (Courtesy of Instrumentation Laboratory.)
Figure 6a. i-STAT ionized calcium sensor and reference electrode

(side-view diagram)
iCa SENSOR
ION SELECTIVE MEMBRANE
INSULATION LAYER
SOLID ELECTROLYTE
AgCl
Ag
SILICON DIOXIDE
SILICON BASE
REFERENCE ELECTRODE
H2O VAPOR PERMEABLE LAYER INSULATION LAYER
SOLID ELECTROLYTE + KCl
AgCl AgCl
Ag
SILICON DIOXIDE
SILICON BASE
Ion-selective membrane
Solid Electrolyte
AgCl
SiO2 Layer
Insulating Layer
Ag
Ag contact
pad
Figure 6: Diagram and image of the i-STAT ionized calcium sensor on a single-use card.
A calcium ion-selective film is deposited over a solid electrolyte gel that covers the Ag/AgCl wire. The reference electrode
has a thin layer of KCl solution covered by a water-permeable, electrolyte-impermeable silicone layer. A small opening in
the silicone layer forms a liquid junction. The image shows a top view of these components as they are arranged on a
miniature test card. (Courtesy of i-STAT Corporation.)
779
Analysis of Ionized Calcium by Ion-Selective Electrode
Principle of Standard Laboratory Analyzers

The ion-selective electrodes, reagents, and flow systems of modern ionized calcium analyzers have many similarities. Most
measure ionized calcium within a panel of critical care tests that include blood gases, pH, Na, K, glucose, and lactate.
Some analyzers also include ionized magnesium and creatinine in this panel. A series of analyte-specific electrodes are in
solution contact with a reference electrode, and a potentiometer measures the specific potential that develops at each
analyte electrode, including the ionized calcium electrode, with a typical example shown in Figure 3. While the potential in
the ionized calcium (analyte) electrode changes in proportion to the ionized calcium concentration in the sample, the
reference electrode provides a stable, fixed potential that is not altered by the sample composition.
The calcium-ion electrode has an ionophore (a molecule that specifically binds calcium ions) imbedded within a stronger
polyvinyl chloride (PVC) or other type membrane. This ion-selective membrane may be covered with a cellophane-type
membrane that prevents contamination by proteins and other constituents in the blood samples being analyzed. An
electrolyte solution with a constant concentration of ionized calcium is contained within the electrode and is in contact with
the inner layer of the ion-selective membrane. As a blood sample contacts the outer layer of the ion-selective/cellophane
membrane, an electric potential develops across the membrane. A potentiometer or voltmeter reads the potential between
this ion-selective electrode and a constant-potential reference electrode, with this potential related to the analyte
concentration. Standard laboratory blood gaselectrolyte analyzers include those made by Bayer (Siemens),
Instrumentation Laboratory, Nova Biomedical, and Radiometer. Diagrams showing the design of ionized calcium
electrodes are shown for Radiometer (Figure 4) and Instrumentation Laboratory (Figure 5).
In most modern analyzers, once the sample is injected or aspirated into the analyzer, the remaining analysis is automated:
flowing the sample to the various analytical and reference electrodes, taking stable measurements and calculating
concentrations, then removing the sample to waste and rinsing for the next sample. A peristaltic pump and a series of flow
selector valves pull the sample through the analyzer, allow the rinse solution to flow, and open/close the injection and exit
ports. These actions are automatically coordinated once a sample is injected. The results are displayed approximately 60
sec later, and another sample may be analyzed 180 sec after injecting the previous sample.
In addition to calibrator fluids, ionized calcium analyzers use detergent rinse solutions between each sample to remove the
blood sample and clean the electrodes and flow system. Periodically, or when a clog is detected in the system, a stronger
cleaning solution (either bleach or alkaline solution) may be sent through the system.
Principle of Single-Use Test Cartridges for Ionized Calcium

Disposable, single-use cartridges are now in widespread use that analyze ionized calcium along with other blood gas and
electrolyte tests as part of a test panel. Although in a physically different form, the principles of analysis of these single-use
cartridges are very similar to those in standard laboratory analyzers. These cartridges are inserted and read in a small
(hand-held) test device and are typically used near the patient in point-of-care settings. The test system is largely within
the disposable cartridge itself and contains the sample injection port, the ion-selective and reference electrodes, a calibrant,
and the flow system.
One such system (i-STAT) has a calcium ionselective film of dioctylphenyl phosphonate plasticized polyvinyl chloride
containing the calcium salt of an organic phosphate compound that is coated over the active area of the electrode. The
reference electrode is an Ag/AgCl electrode covered by a thin KCl electrolyte layer, which in turn is covered with a water-
permeable, electrolyte-impermeable silicone layer. A small opening in the silicone layer allows formation of a liquid
junction and measurement of the electrochemical cell voltage between the sample and calibrant, which is related to the
concentration of the ionized calcium in the blood sample. Both a cross-section diagram and image of this single-use
electrode is shown in Figure 6.
Calibration
A typical calibration consists of analyzing one or more levels of calibrants or standards. Typically this is done initially
when an electrode or test pack is changed, then may be done manually or automatically at times necessary to satisfy
regulatory requirements.
Quality Control
Until relatively recently, quality of results were maintained by analysis of two or three external quality-control samples
provided by manufacturers in sealed glass ampules or other containers. Ionized calcium concentrations in these controls
should cover a range of concentrations from low to high levels. CAP and/or CLIA still require analysis of control materials
at least every 8 hours. However, over the past 10 to 15 years, quality-control procedures have undergone significant
evolution. In basic terms, these quality-control procedures are aimed at minimizing the need to analyze external control
material at frequent intervals, which are especially troublesome with near-patient use by caregivers. These new control
780
Ionized Calcium
procedures also depend on systems that are inherently stable over an extended period of time, from several hours to several
weeks.
Electronic quality-control devices. These devices are often used with point-of-care test systems. The electronic control
devices are inserted into the cartridge reading system to check the electronic and physical integrity of the system. Although
these devices should determine if the cartridge-reading system is functioning properly, they have no relevance to the
integrity of the analytical cartridge itself.
Internal calibrant/control solutions. After an initial calibration of the cartridge lot, to help ensure that the cartridge and
electrode systems are functioning properly during the life of the test pack lot, an internal calibrant/control solution, sealed
within the cartridge, is analyzed with each blood sample. If this internal solution gives a reading within established limits,
the clinical analysis is considered valid, and the results are reported. The single-use i-STAT cartridge incorporates a
calibrant solution within each disposable test cartridge. Instrumentation Laboratory uses a control system based on analysis
of several solutions sealed within gas-impermeable multi-use test packs to verify calibration stability. The responses of
these control fluids and the blood samples are also monitored in multiple ways to help identify failures of the instrument,
flow system, or electrodes, and also identify problems with specific samples. When problems are detected, corrective
actions by rinsing and recalibration may be automatically initiated.
781
Iron and Iron-Binding Capacity

Gerardo Perrotta and Jordan Reynolds
Name: Iron and Iron-Binding Capacity
Clinical significance: Refer to Chapter 39, Iron, Porphyrin, and Bilirubin Metabolism, in the 5th
Atomic symbol: Fe2+ (ferrous), Fe3+ (ferric)
Atomic mass: 56 D
Merck Index: 4942
Chemical class: Element, metal
i
Other methods, such as radiometry and atomic
There are many methods available for determining absorption spectrophotometry, complete the list of
serum iron. All colorimetric procedures have in common methods available for assaying serum iron. Most recent
a reaction in which ferric iron (Fe3+) is reduced to the CAP survey data [4] indicate that virtually all
ferrous state (Fe2+) by the addition of strong reducing laboratories utilize a colorimetric procedure, and none
agents such as hydrazine, ascorbic acid, thioglycolic are listed as using the atomic absorption method.
acid, or hydroxylamine [1-3].
Serum iron analysis is almost always accompanied by a
Once the reduction has taken place, colorimetric measurement of the total iron-binding capacity (TIBC).
reactions use an iron-complexing chromogenic agent to As indicated in Table 1, serum iron is bound to
bind the ferrous iron [2-4]. The more commonly used transferrin, but only a portion of the transferrin molecule
complexing agents, according to one of the recent 2007 is saturated with iron. The unsaturated iron-binding
College of American Pathologists (CAP) surveys, are capacity of transferrin (UIBC) denotes the available
5,5(3-(2-pyridyl)-1,2,4-triazine-5,6 diyl)-bis-2- iron-binding sites of serum. On the other hand, the
furansulfonic acid (Ferene Iron Reagent), with or amount of iron that serum transferrin can bind when
without prior protein removal, and 3-(2-pyridyl)-5,6- completely saturated is the total iron-binding capacity
bis(4-phenylsulfonic acid)-1,2,4-triazine (FerroZine Iron (TIBC).
Reagent).
In most colorimetric methods, TIBC is measured by first
In summary, all colorimetric procedures employ the saturating the transferrin with excess ferric iron. The
following reaction (Table 1, Methods 1 and 2): remaining iron that cannot be bound to transferrin is
removed by a chelator, such as magnesium carbonate,
which forms an insoluble complex with the excess ferric
iron. The total amount of iron saturating the transferrin is
then measured as described for total serum iron analysis.
Non-transferrin-bound iron (NTBI) and iron bound to
ferritin can cause apparent higher transferrin saturation
levels [6].
The use of coulometry for measurement of serum iron
has also been developed (Table 1, Method 3). The NTBI can also occur in patients in whom the binding
method first removes iron from transferrin by the capacity of transferrin is overloaded, such as those
addition of an alcoholic HCl solution. The freed iron is patients in an iron overloaded state (thalassemia and
exposed to different specific potentials in a multi- hemochromatosis). Not all of the iron is associated with
electrode sensor. The transfer of electrons between the transferrin.
2+ 3+
Fe and Fe generates a current. The instrument detects An ion-exchange technique for measurement of serum
the electrons in transition and relates the total electron iron may also be used. A sample of serum is added to an
flow to iron concentration [5]. aliquot of ion-exchange resin preloaded with iron. After
3+ 2+
Fe + e- Fe a short period of equilibration, the serum sample
(transferrin) is saturated with iron, since the affinity of
iron for the transferrin is greater than its affinity for the
i
Iron and Iron-Binding Capacity resin. An aliquot of this iron-saturated serum is then
Previous and current authors of this method: analyzed the same way as for total iron analysis.
First edition: Gerado Perotta
Methods edition: Gerado Perotta It is also possible to measure TIBC by an additive
Second edition: Gerado Perotta procedure. A known iron standard (5000 g/L) is
Third edition: Gerado Perotta incubated with the serum sample at a pH of 8.5, and the
Fourth edition: Gerado Perotta transferrin becomes saturated. The remaining unbound
Fifth edition: Gerado Perotta, Jordan Reynolds
782
(free) iron is measured by the regular iron procedure. has been found to interfere with ferrozine-based
The 5000 g/L standard minus the excess Fe not bound methods. Administration of oral contraceptives will
to transferrin is the UIBC, or unbound iron-binding elevate iron and TIBC levels. Drugs containing azo
capacity. groups may interfere with the coulometric analysis.
UIBC + Serum iron = TIBC Iron and Iron-Binding Capacity Reference Interval
The range for serum iron in healthy persons is very
Reference and Preferred Methods method dependent. Reference intervals range from 40 to
In 1978, the International Committee for Standardization 160 g/dL, to 20 to 250 g/dL in male adults and 20 to
in Hematology (ICSH) proposed a reference method 200 g/dL in female adults. Iron concentration in
with bathophenanthroline and revised it in 1990 by premenopausal women may be 10% to 15% lower than
replacing the chromogen with ferrozine or ferene [7]. for men, though product information for most methods
The Clinical and Laboratory Standards Institute (CLSI), cite one range for both sexes. Reports indicate that
formerly NCCLS, proposed its own method that differed postmenopausal women have serum iron concentrations
from the ICSH in terms of reagent concentration [8]. that are not significantly different from those of men of
Tietz et al. compared the ICSH methods to five similar age. Neonates have serum iron concentrations
commercial methods and found many discrepancies at approximately twice that of healthy adults.
low concentrations. Effort towards a more reliable and
perhaps definitive reference method will hopefully be In healthy persons, transferrin is usually approximately
available in the future [9]. 30% saturated with serum iron. TIBC reference values in
healthy adults are approximately 225 to 425 g/dL.
Current methods using either FerroZine or TPZ as the
iron complexing reagent yield consistently higher values Interpretation
than the bathophenanthroline method. The most common Measurement of serum iron alone is insufficient. It is
comparison is between colorimetric methods that use usual to make simultaneous measurement of iron, TIBC,
prior protein removal and those without prior protein and ferritin as an index of iron stores. The ratio of iron to
removal. Most laboratories reporting in a recent CAP TIBC is usually expressed as the transferrin saturation.
survey [4] used methods that fall into the category of
analysis without prior protein removal. Itano reviewed Iron deficiency is common and is characterized by low
the results of a questionnaire that was used to determine serum iron, low transferrin saturation, and low ferritin.
the cause of the differences reported in the CAP survey Iron deficiency is more common in women. Women
between the various methods listed. His findings during childbearing years lose about 600 to 800 mg of
suggested that significantly lower values were common iron during each pregnancy. This loss must be balanced
for the FerroZine and TPZ procedures not using prior by the absorption of an equivalent amount of dietary iron
protein removal [4,10]. However, there appeared to be [13]. There is also significant blood loss at parturition,
no significant bias between these two chromogenic and iron is lost during lactation.
reagents used for either category. His study reinforced
the need for some reference method. Iron excess may be as a result of ingestion of iron or
combination tablets, or it may occur in the condition of
Specimen hemochromatosis, where there is increased absorption of
No significant differences in apparent concentration iron from the GI tract.
have been reported between serum and plasma [11].
Phlebotomy should be accomplished with minimal stasis Iron toxicity occurs more frequently in children, though
to allow the free flow of blood into the collection device. it may occur in adults as well. Ingestion of therapeutic
A fasting, early-morning sample is optimal for analysis, iron tablets is the most common cause. Toxicity is
because a diurnal variation may decrease the iron usually defined as a serum iron result that exceeds the
concentration by 30% during the course of a day. Peak TIBC. In normal individuals, the transferrin is one-third
iron concentrations occur in the early- to late-morning saturated, but in patients with iron overload, the
hours. The specimen should be centrifuged as soon as transferrin is completely saturated [14]. Common
possible after collection. Storage of whole blood without symptoms of iron toxicity include vomiting, upper
separation of serum or plasma renders specimens abdominal pain, headaches, and diarrhea.
unsuitable for analysis. Serum iron is stable up to 1 week
at 4C to 8C and is stable indefinitely when frozen at Iron poisoning most commonly occurs in patients
20C. Transferrin is stable for at least a week at younger than 19 years of age. Serum iron concentration
refrigerated temperatures and at least a month when greater than 350 g/dL is considered to be indicative of
frozen. systemic toxicity; however, lower serum levels may also
be present in toxic patients [15]. Serum iron levels reach
Interferences a maximum 2 to 6 hours after ingestion, and
Hemolyzed samples must be rejected, since erythrocytes measurement at about 4 hours is recommended.
contain iron and therefore falsely increase apparent iron However, some methods of measurement in the
concentration. Response to hemolysis is both overdose situation may be inaccurate because of the high
heterogeneous and unpredictable, and therefore no concentrations observed [16].
reliable corrective measures can be made [12]. Copper
783
Iron overload occurs in conditions such as 6 Al-Refaie FN, De Silva CE, Wonke B,
hemochromatosis, ineffective erythropoiesis, and Hoffbrand AV. Changes in transferrin
transfusional iron overload, for example in -thalassemia saturation after treatment with the oral iron
major [17]. chelator deferiprone in patients with iron
overload. J Clin Pathol. 1995;48:110-114.
Iron is rarely measured in urine specimens; rather the 7 Lewis SM. International Committee for
urine is used for hemosiderin analysis. This iron-binding Standardization in Hematology proposed
protein may appear in urine as yellow-brown granules or recommendation for measurement of serum
in cells or casts. Hemosiderinuria is useful for detection iron in human blood. Am J Clin Pathol.
of hemochromatins, hemolytic anemias, and paroxysmal 1971;56:543-545.
nocturnal hemoglobinuria. 8 Valcour AA, Krzymowski G, Onoroski M,
Bowers GN Jr, McComb RB. Proposed
Iron and Iron-Binding Capacity Performance Goals reference method for iron in serum used to
Survey data from the 2007 CAP Summary Report shows evaluate two automated iron methods. Clin
imprecision values (% coefficient of variation) for Chem. 1990;36:1789-1792.
measurement of iron at a concentration of 167 g/dL to 9 Tietz N, Rinker A, Morrison S. When is a
range from 1.7% for ferene methods without serum iron really a serum iron? The status of
precipitation to 4.8% for the nitroso PSAP methods [4]. serum iron measurements. Clin Chem.
1994;40:546-551.
Desirable specifications for analytical imprecision 10 Itano M. CAP comprehensive chemistry, serum
derived from studies of biological variation indicate an iron survey. Am J Clin Pathol. 1978;70:516-
assay imprecision of 13.2% (0.5 within individual 522.
biological variation [CVw]) [18]. The acceptable 11 Romeo R, Boedeker M, Duncan C.
performance criteria according to the Clinical Comparisons between serum and plasma
Laboratory Improvement Amendments of 1988 (CLIA- analytes measured on the Baxter Paramex. Clin
88) require that laboratories be within 20% of their Chem. 1991;37:967.
respective peer group. 12 Lippe G, Salvagno GL, Montagnana M,
BroccoG, Guidi GC. Influence of hemolysis on
References routine clinical chemistry testing. Clin Chem
1 Peters T, Giovanniello T. Apt L, Ross JF. A Lab Med. 2006;44:311-316.
simple, improved method for the determination 13 Sonnenwirth AC, Jarett L. Gradwohls clinical
of serum iron II. J Lab Clin Med. 1956;48:280- laboratory methods and diagnosis. 8th ed. St
288. Louis: Mosby; 1980.
2 Levy A, Vitacce P. Direct determination and 14 Breuer W, Cabantchik Z. A fluorescence-based
binding capacity of serum iron. Clin Chem. one-step assay for serum non-transferrin-bound
1961;7:241-248. iron. Anal Biochem. 2001;299:194-202.
3 Goodwin J, Murphy B, Guillemette M. Direct 15 Mills KC, Curry SC. Acute iron poisoning.
measurement of serum iron and binding Emerg Med Clinics North Am. 1994;12;397-
capacity. Clin Chem. 1966;12:47-57. 411.
4 College of American Pathologists. CAP 16 Curry SC. Iron. In: Reisdorff EJ, Roberts MR,
Surveys 2007 Participant Summary, Chemistry Wiegenstein JG. Pediatric emergency medicine.
(C-A). Washington, DC: CAP, Chemistry Philadelphia: Saunders; 1993:673-679.
Resource Committee; 2007. 17 Olivieri NF, Brittenham GM. Iron-chelation
5 FerroChem Model 3050, Serum Iron and Total therapy and the treatment of thalassemia.
Iron Binding Capacity Analyzer Instruction Blood. 1997;89:739-761.
Manual. 2nd ed. Bedford, MA: Environmental 18 White GH, Farrance I. Uncertainty of
Sciences Associates; 1979. Revised February measurement in quantitative medical testing: a
1981. laboratory implementation guide. Clin Biochem
Rev. 2004;25:S1-4.
784
Tables
Table 1: Iron Methods Summary

Method 1: FerroZine iron reagent; colorimetric
Transferrin (Fe3+) buffer 2Fe2+ + transferrin
2
reducing agent
Fe2+ + 3 FerroZine Fe (FerroZine)2+(magenta color complex, 560 nm)
Fe2+ + 3 NSPAP Fe (NSPAP)2+ (tense green complex, 750 nm)
Comments: Good coefficient of variation; no prior protein removal is necessary; manual or automated
assay
Method 2: Du Pont aca; colorimetric
Transferrin (Fe3+) pH 4.2 Fe3+ + Transferrin
Fe3+ + Hydroxylamine Fe2+
Fe2+ + Bathophenanthroline Red complex (540 nm)
Comments: Good for stat tests; automated
Method 3: Electrochemical; coulometric
Principle of analysis: Fe3+ + e- Fe2+; sum of electrons in this transition is between two possible
valence states of Fe
Comments: Correlates well with FerroZine method; has very good accuracy and precision; excellent
for microsamples and stat tests; automated
Method 4: Spectroscopy; atomic absorption
Principle of analysis: Iron is concentrated by chelation with bathophenanthroline and is extracted into
methyl isobutyl ketone (MIBK); extract is analyzed by atomic absorption at 248.3 nm
Comments: Impractical and low sensitivity; mainly for research
Reaction between acetone and the sodium nitroferricyanide (nitroprusside) formulation found in Ketostix
(Ames, Division of Miles Laboratories, Elkhart, IN).
Figure
Iron: Figure 1
Absorption spectra of FerroZine [3-(2-pyridyl)-5,6-bis(4-phenylsylfonic acid)-1,2,4-triazine] reagent. Dashed

line, Reagent blank; solid line, reagent plus iron.
785
Ketones
Ketones
Lawrence A. Kaplan
Name: Ketones
Clinical significance: Refer to Chapter 38, Diabetes, in the 5th Edition of Clinical
Common name: Acetoacetic acid Acetone -Hydroxybutyric acid
Molecular formula: C4H6O3 C3H6O C4H8O3
Merck Index: 51 58 4727
Chemical class: Ketocarboxylic acid Ketone Hydroxycarboxylic acid
Structure: See Figure 1 See Figure 1 See Figure 1
Metabolome sites:
Acetoacetic acid: http://www.hmdb.ca/scripts/show_card.cgi?METABOCARD=HMDB00060.txt
Acetone: http://www.hmdb.ca/scripts/show_card.cgi?METABOCARD=HMDB01659.txt
-Hydroxybutyric acid: http://www.hmdb.ca/scripts/show_card.cgi?METABOCARD=HMDB00011.txt
Principles of Analysis and Current Usage i ketone bodies. A clinically meaningful assay that
specifically measures one of these biochemicals will lack
Ketones are biochemicals that have structural moieties of the ability to detect the others or to detect all the ketone
bodies. In addition, any clinically useful test for ketones
ketones (that is, ; see structures). should be able to be performed rapidly, simply, 24 hours a
day.
Commonly found serum ketones include pyruvate,
acetoacetic acid, and acetone. Their reduced forms of the Total Ketones; Urine and Serum
first two ketones are lactic acid and -hydroxybutyric acid, Total ketones are the sum of the measurable ketones found
respectivel in serum of diabetics, that is, acetoacetic acid and acetone.
Another term used to describe these biochemicals is The reaction between ketones and nitroprusside (sodium
ketone bodies, which are acetoacetic acid, acetone, and - nitroferricyanide) under alkaline conditions is the most
hydroxybutyric acid. This discussion will only review the widely used procedure for the detection of total ketones in
analysis of these ketone bodies, which are associated with diabetes care (Table 1, Method 1) [2,3]. Acetoacetic acid
diabetes, the clinical context in which ketones are most and acetone both form a purple color in this reaction in the
frequently measured. Less commonly, other ketoacids can presence of glycine and buffered by disodium phosphate
be produced in certain diseases such as maple syrup urine (Figure 1). This method is used most frequently as a
disease (see Chapter 52 and Reference 1). The discussion semiquantitative measure of total ketones in serum and
of methodologies for pyruvate and lactic acids and the urine to screen for diabetic ketoacidosis. This assay does
complex ketoacids are presented elsewhere. not detect -hydroxybutyric acid.
The analysis for ketones has been limited by the fact that A modification of the nitroprusside test is the
the proportions of specific ketone bodies present in incorporation of the reagents into a reagent strip (Ketostix)
diabetes can vary considerably and by the fact that no or a tablet (Acetest). This technique is also incorporated
single chemical or enzymatic method measures these into the multitest urine chemistry dipsticks that are used
for routine urinalysis. The dipsticks are most accurate for
urines with a specific gravity of 1.010 to 1.020. Both
methods can be used for serum or urine. The use of
i Ketones reflectance spectrophotometry has enabled quantitative
Previous and current authors of this method: measurement of total ketones in a multitest urine dipstick
First edition: Lawrence A. Kaplan [4,5].
Methods edition: Lawrence A. Kaplan and Nancy Gau
-Hydroxybutyrate and Acetoacetic acid
Second edition: Lawrence A. Kaplan
Several techniques have been developed that measure
Third edition: Steven C. Kazmierczak these ketone bodies with high specificity. Acetoacetic acid
Fourth edition: Steven C. Kazmierczak and -hydroxybutyric acid can be quantitated by an
Fifth edition: Lawrence A. Kaplan
786
Ketones
enzyme assay (Table 1, Method 2) which makes use of the The GM7 multi-assay analyzer (Analox Instruments
following reversible reaction: U.S.A., Lunenburg, MA
[http://www.analox.com/frame1.htm]), couples the -
hydroxybutyrate dehydrogenase (3-HBDH) reaction with a
-HBD, pH 7.0 peroxidase (POD) reaction and monitors the consumption
+
NADH + H+ + AcAc -HB + NAD of oxygen.
3HBDH
pH 8.5-9.5 -HB + NAD+ acetoacetate +NADH + H+
Where AcAc = acetoacetic acid, -HB = -hydroxybutyric POD
acid, and -HBD = -hydroxybutyric acid dehydrogenase. NADH + 02 + H+ NAD+ + H20
The reaction can be monitored by measuring the
absorbance of NADH at 340 nm [6,7]. At pH 8.5 to 9.5, This assay is available only on a small analyzer and is not
the reaction proceeds to the left as written, and the used for routine clinical situations.
concentration of -hydroxybutyric acid (-HB) is
quantitated when the increase in absorbance is monitored As was discussed earlier, the most frequent clinical
at 340 nm. When performed at pH 7.0, the reaction situation requiring the measurement of ketone bodies is
proceeds to the right, and the amount of acetoacetic acid diabetes, both for the diagnosis of acute diabetic
(AcAc) present is proportional to the decrease in ketoacidosis and for self-monitoring of diabetic therapy.
absorbance at 340 nm. Because the nitroprusside method for assessment of blood
and urine ketones is used for emergency room needs, for
A number of assays that employ the above reaction at pH physicians office screening, or for home use,
8.5 to 9.5 have been developed for specifically monitoring overwhelmingly, most tests for ketones, especially blood
-hydroxybutyric acid for diabetes in the home, clinic, or ketones (see below), will not be performed in the
acute-care setting [8,9]. The MediSense hand-held device traditional laboratory by technologists. The
(Abbott Corporation, Abbott Park, IL, USA) uses capillary semiquantitative nitroprusside test in the form of a dipstick
blood (5 L), which is added to a reagent strip, which is screen is currently the most simple to use in these clinical
inserted into a reading device. Using the - situations. Appropriately, this method is overwhelmingly
hydroxybutyrate dehydrogenase reaction shown above at the most frequently used test method.
alkaline pH, the NAD+ is reduced to NADH, which is
reoxidized to NAD+ by a redox mediator. The current that In the United States, the nitroprusside test and the hand-
is generated by the second reaction is detected by an held devices for measuring -hydroxybutyrate are waived
electrochemical detector and is directly proportional to the tests, that is, they can be performed by non-technologists
concentration of -hydroxybutyrate in the sample (Table 1, in physicians offices, in clinics, or in patients homes. All
Method 3a). the other test procedures listed in Table 1 are more
appropriate for research situations.
Other such hand-held devices for measurement of -
hydroxybutyric acid include the KetoSite assay (GDS There have been many other methods suggested for the
Technology, Elkhart, IN analysis of ketones, but these are more applicable for
[http://www.gdstech.com/products.htm]) and the research use, or they are historic in nature and not used at
Cardiochek ketone assay ( Polymer Technology Systems, present. These methods are listed in Appendix A.
Indianapolis, IN
[http://www.cardiochek.com/docs/pdf/Ket_english.pdf]). Reference and Preferred Method
The KetoSite and Cardiochek devices, which use 15 to 25 There are no accepted reference methods for the
L of whole blood, employ reagents incorporated into dry- measurement of acetoacetic acid, acetone, or -
film and reflectance photometry to measure the reaction in hydroxybutyric acid.
which the NADH reduces a tetrazolium salt to produce a
colored product (Table 1, Method 3b). In healthy persons, the nitroprusside test for acetoacetic
In the United States, the MediSense, KetoSite, and acid accurately reflects total ketones. However, this test
Cardiochek products are waived tests and are approved for can be misleading in diabetic ketoacidosis, when -
use in physicians offices and for self-testing in the home. hydroxybutyric acid is the predominant ketone body
formed. Tests for acetoacetic acid may be negative or
-HB weakly positive, thereby underestimating the total load of
+
-HB + NAD NADH + H+ + AcAc ketone bodies and the degree of ketosis. As the diabetic
diaphorase ketoacidosis is resolved, more acetoacetic acid can be seen
than is found in the initial stages of the crisis as -
NADH + tetrazolium dye hydroxybutyric acid is metabolized to acetoacetate.
However, there is no strong clinical data available that
NAD+ + reduced tetrazolium dye
indicates that measurement of -hydroxybutyric acid has
colored any significant advantage over the nitroprusside screen for
787
Ketones
the diagnosis of acute diabetic ketoacidosis. In most cases, acetone is a volatile substance and can be lost by
ketoacidosis is detected with the nitroprusside screen, and evaporation. Thus urine samples should be well sealed and
the treatment of the acidosis is usually monitored by refrigerated within 20 to 30 minutes of collection if not
measurements of blood pH. In one report [8], the area analyzed immediately. Some urine preservative can affect
under the ROC curves of serum total ketones and capillary test results.
-hydroxybutyric acid were not statistically different,
leading to the conclusion that serum ketone and blood - Interferences
hydroxybutyric acid measurement were equally effective Nitroprusside Reaction: Total Ketones
in diagnosing DKA among uncomplicated cases. This Ketostix can give false-positive results in the presence of
agrees with similar conclusions drawn by an earlier report large amounts of levodopa [13], and in general, the urinary
[10]. ketone assay is reported to give false-positive results in the
presence of drugs containing sulfhydryl groups, such as the
The primary advantage to the nitroprusside screen is its antihypertensive drug captopril, as well as Mesna, N-
relative ease of use and reagent stability, and it is the acetylcysteine, dimercaprol, and penicillamine [14]. It
appropriate test for ketone detection in most clinical should be noted that many of these drugs, such as
situations. The major drawback to this procedure is that it captopril, may be used to treat diabetics, and so this assay
cannot detect -hydroxybutyric acid at all and is five- to must be used with caution in such cases. The presence of
tenfold less sensitive for acetone than for acetoacetic acid. high concentrations of phenylpyruvic acid and
Acetone is rarely positive at serum concentrations less than acetaldehyde in urine can also result in false-positive
5 mmol/L. In addition, although low, undetectable results with the nitroprusside procedure.
amounts of ketones are often present in the urine of
healthy individuals, clinically false-positive results can be False-negative results have also been reported for acidic
associated with the urine of fasting individuals and up to urine containing large amounts of ascorbic acid [15].
30% of the urines of pregnant woman [11]. As with any of
the dry-chemical processes, however, the reagent can The presence of hemolysis may result in discoloration of
deteriorate unless properly protected from air. the Ketostix test strip and adversely affect interpretation,
whether by manual or automated procedure.
The reported sensitivity of the qualitative Ketostix is 50 to
100 mg/L of acetoacetate, whereas the sensitivity for -Hydroxybutyrate
Acetest is 25 to 50 mg/L. The dipstick methods are linear The urinary semiquantitative nitroprusside test is
up to 1600 mg/L of acetoacetate. acceptable for occasional self-monitoring of ketones, but
the measurement of -hydroxybutyric acid will provide
The enzymatic procedures are specific for acetoacetate or better clinical sensitivity than will urine testing for
-hydroxybutyrate and have been readily adapted to highly detecting ketosis [16]. The advent of simple hand-held
automated instruments. The reagents for these assays can devices applicable for self-testing for the measurement of
be easily prepared by most laboratories, and these methods blood -hydroxybutyrate will most likely make this assay
are recommended for those laboratories interested in the recommended one in the futurefor the detection of
specifically quantitating the individual serum and urine ketosis for home use and possibly for acute ketoacidosis as
ketone bodies. These laboratories would need to comply well. The assays for the quantitative measurement are
with any regulations regarding the use of home brew relatively free of interferences. High levels of acetoacetate
reagents (see CLIA 88 in the United States). have been found to falsely decrease -hydroxybutyrate
The easy-to-use hand-held devices (MediSense, KetoSite, concentrations [3,10]. Elevated blood ascorbic acid levels
Analox, and Cardiochek) for measurement of blood - may cause false positives in one of the hand-held assays
hydroxybutyrate makes measurement of this analyte more for whole blood -hydroxybutyrate acid concentrations
readily available for most clinical situations. These devices [12].
use small amounts of whole blood (10 to 25 L) and have
acceptable linearity (up to 21 mg/dL and 70 mg/dL for the Ketones Reference Interval
KetoSite and Cardiocheck devices, respectively). The The reference interval for serum and urinary total ketones
Cardiocheck device reports possible false-positive results is less than 50 mg/L (approximately < 500 mol/L).
in the presence of ascorbic acid [12].
The waived-test instruments report normal serum levels of
Specimen -hydroxybutyrate acid to be < 31 mg/L (<0.3 mmol/L),
Well-centrifuged urine, serum, or plasma can be used for while a range of < 52 mg/L (<0.5 mmol/L) was reported in
the semiquantitative nitroprusside test, and serum or the NACB document. Henry reports a reference interval
plasma can be used for the enzymatic assays as well. The for serum acetoacetate of 5 to 30 mg/L (49 to 294 mol/L)
hand-held devices that measure -hydroxybutyric acid can [17], which compares well with results of 3 to 23 mg/L (29
use whole blood anticoagulated with heparin or EDTA. to 225 mol/L) obtained by a laboratory using an
Bacterial contamination of urine can hasten the enzymatic procedure [18]. There do not appear to be any
disappearance of the ketones from urine. In addition, age-related differences in reference intervals for ketone
788
Ketones
bodies [19]. At these concentrations, serum and urine and amount of the ketone bodies formed depend on the
samples will be negative by the nitroprusside oxidative state of the cells. In early, severe diabetic
semiquantitative screening tests. See p 4, Reference ketoacidosis or in acute alcoholic ketoacidosis, the cells
Intevals section, for acetoacetic acid reference interval and are in a more reduced state (that is, decreased
-hydroxybutyric acid reference interval of the individual NAD/NADH ratio), and -hydroxybutyric acid is usually
ketone bodies. the predominant ketone body that forms. If these states are
associated with a concomitant lactic acidemia, there can be
Interpretation an even greater tendency toward production of -
Acetoacetic acid, acetone, and -hydroxybutyric acid are hydroxybutyric acid from acetoacetic acid. As the
the products of fatty-acid metabolism that accumulate in biochemical status of the diabetic ketoacidotic patient
large amounts when the livers capacity to metabolize the improves and the NAD/NADH ratio increases, the -
final oxidative catabolite of fatty acids, acetyl coenzyme A hydroxybutyric acid levels decrease, while the levels of
(CoA), is overwhelmed. This occurs in states of starvation, acetoacetic acid increase. Thus when measurements of
impaired carbohydrate metabolism (absence of insulin, total ketones are made with the nitroprusside method
such as diabetes; or low carbohydrate diet), or acute (urinary dipsticks), the test results may paradoxically
ethanol abuse. apparently worsen, indicating an increase in ketones even
as the ketoacidosis is resolving.
In these states, the oxidative capacity of liver cells is
diminished, and the ratio of NAD/NADH is greatly The ketone bodies appear in urine when the renal threshold
decreased, resulting in an inability to oxidize or transfer of ~ 25 mg/dL is exceeded.
acetyl CoA or to allow metabolism by the Embden-
Meyerhof pathway. To allow continued metabolism and Both the American Diabetes Association [11] and the
recycling of the coenzyme A cofactor, the hepatocytes National Association of Clinical Biochemistry (NACB)
begin active ketogenesis by the following series of [20] recommend that the measurement of blood ketones is
reactions: preferred only as an adjunct test for the diagnosis of acute
diabetic ketoacidosis. Blood ketone measurements may
serve as an adjunct test for assessing the severity and
resolution of diabetic ketoacidosis, although measurements
of blood pH are probably more reliable.
It has been recommended that urinary ketone

measurements not be used for the diagnosis of acute
ketoacidosis [13,14].
Generally, the reported proportions of total serum ketone

bodies are 2% acetone, 20% acetoacetic acid, and 78% -
hydroxybutyric acid [17]. The varying pattern of ketones
and ketone bodies produced in diabetics was reported in a
small series of patients (n = 22) with hyperglycemia
(serum glucose ranging from 1870 to 6140 mg/L, mean
3220 mg/L) [18]. None of the patients had an elevated total
CO2 result, and only 6 had a positive (trace, approximately
50 mg/L) result for ketones by the Ketostix test. Only 1
patient had a normal (<2 mmol/L) lactic acid (range 1.92
to 13.7 mmol/L), though all but 2 had normal (<0.15
mmol/L) pyruvic acid levels. -Hydroxybutyric acid levels
were elevated in 5 of the 22 patients (>0.7 mmol/L) and in
4 of the 6 patients with positive Ketostix results. All 6
patients with a positive Ketostix had acetoacetic acid
levels 0.2 mmol/L. These results are summarized in
Table 2.
Ketones Performance Goals

There are no established performance goals for the
quantitative measurement of total ketones or of the
individual ketone bodies.
Thus a series of ketones or ketone bodies (that is, the Coefficients of variation of less than 8% for levels within
ketones plus -hydroxybutyric acid) are formed. The type the normal range have been reported for the enzymatic
789
Ketones
procedures for acetoacetate, pyruvate, or - 13 Ketostix package insert. 1983. Ames, Division of
hydroxybutyrate [6,7]. Miles Laboratories, Inc., Elkhart, IN.
14 Csako G. Causes, consequences, and recognition
Quantitative total ketone measurement by reflectance of false-positive reactions for ketones. Clin Chem
photometry has precision that is acceptable for clinical use. 1990;36:1388-1389.
The reported within- and between-run reproducibility of 15 Rosenbloom AL, Malone JI. Recognition of
this assay was reported as 11.0% to 3.6% and 11.0% to impending ketoacidosis delayed by ketone reagent
5.8% for acetoacetate, 8.2% to 9.2% and 10.4% to 16.1% strip failure. JAMA 1978;240:2462-2464.
for acetone, for high- and low-concentration urine pools, 16 Guerci B, Benichou M, Floriot M, Bohme P,
respectively [4]. However, there is no convincing evidence Fougnot S, Franck P, Drouin P. Accuracy of an
that the quantitative assay yields more clinically useful electrochemical sensor for measuring capillary
results than the semiquantitative one. blood ketones by fingerstick samples during
metabolic deterioration after continuous
References subcutaneous insulin infusion interruption in type
1 Online Mendelian Inheritance in Man. Johns 1 diabetic patients. Diabetes Care 2003;26:1137-
Hopkins University. Available at 1141.
<http://www.ncbi.nlm.nih.gov/entrez/dispomim.c 17 Mayes PA, Robson W. The determination of
gi?id=248600> ketone bodies. Biochem J 1957;67:11-15.
2 Rothera ACH. Note on the sodium nitroprusside 18 Sacoolidge J, Donohoo R, Gau N. Unpublished
reaction for acetone. J Physiol 1908;37:491-494. results. Cincinnati, OH: Clinical Chemistry
3 Free AH, Free HM. Nature of nitroprusside Laboratory, University of Cincinnati Hospital.
reactive material in urine in ketosis. Am J Clin 19 Peden VH. Determination of individual serum
Pathol 1958;30:7-10. ketone bodies, with normal values in infants
4 Penders J, Fiers T, Delanghe JR. Quantitative and children. J Lab Clin Med 1964;63:332-343.
evaluation of urinalysis test strips. Clin Chem 20 NACB, LMPG. Guidelines and
2002;48:2236-2241. Recommendations for Laboratory Analysis in the
5 Penders J, Fiers T, Giri M, Wuyts B, Ysewyn L, Diagnosis and Management of Diabetes Mellitus.
Delanghe JR. Quantitative measurement of ketone Available at
bodies in urine using reflectometry. Clin Chem <http://www.nacb.org/lmpg/diabetes/5_diabetes_
Lab Med 2005;43:724-9. keytone.doc>
6 Li PL, Lee JT, MacGillivray MH, Schaefer PA, 21 Eriksson CJP. Micromethod of determination of
Siegel JH. Direct fixed-time kinetic assays for ketone bodies by dead space gas chromatography.
beta-hydroxybutyrate and acetoacetate with a Anal Biochem 1972;47:235-243.
centrifugal analyzer or a computer-backed 22 Kimura M, Kobayashi K, Matsuoka A, Hayashi
spectrophotometer. Clin Chem 1980;26:1713- K, Kimura Y. Head-space gas-chromatographic
1717. determination of 3-hydroxybutyrate in plasma
7 Hansen JL, Frier EF. Direct assays of lactate, after enzymic reactions, and the relationship
pyruvate, beta-hydroxybutyrate and acetoacetate among the three ketone bodies. Clin Chem
with a centrifugal analyzer. Clin Chem 1985;31:596-8.
1978;24:475-479. 23 Brega A, Villa P, Quadrini G, Quadri A, Lucarelli
8 Tantiwong P, Puavilai G, Ongphiphadhanakul B, C. High-performance liquid chromatographic
Bunnag P, Ngarmukos C. Capillary blood beta- determination of acetone in blood and urine in the
hydroxybutyrate measurement by reagent strip in clinical diagnostic laboratory. J Chromatography
diagnosing diabetic ketoacidosis. Clin Lab Sci 1991;553:249-54.
2005;18:139. 24 Scott-Wilson H. A method for estimating acetone
9 Umpierrez GE, Watts NB, Phillips LS. Clinical in animal fluids. J Physiol 1911;42:444-470.
utility of beta-hydroxybutyrate determined by 25 Van Slyke DD. Studies of acidosis. VII. The
reflectance meter in the management of diabetic determination of -hydroxybutyric acid,
ketoacidosis. Diabetes Care 1995;18:137-138. acetoacetic acid, and acetone in urine. J Biol
10 Porter WH, Yao HH, Karounos DG. Laboratory Chem 1917;32:455-493.
and clinical evaluation of assays for - 26 Procos J. Modification of the spectrophotometric
hydroxybutyrate. Am J Clin Pathol 1997;107:353- determination of ketone bodies in blood enabling
358. the total recovery of -hydroxybutyric acid. Clin
11 American Diabetes Association. Tests of Chem 1961;7:97-106.
glycemia in diabetes: position statement. Diabetes 27 Tsao MU, Lowrey GH, Graham EJ.
Care 2002;25:S97-S99. Microdetermination of acetone in biological
12 PTS package insert. Available at fluids. Anal Chem 1959;31:311-314.
<http://www.cardiochek.com/docs/pdf/Ket_englis 28 Nadeau G. The interference of acetone in blood
h.pdf> alcohol determinations: a simple method for the
790
Ketones
determination of blood acetone (Ravin modified). 30 Drews PA. Carbohydrate derivatives and
Can Med Assoc J 1952;67:158-159. metabolites. In: Henry RJ, Cannon DC,
29 Amlathe S, Gupta VK. Spectrophotometric Winkleman JW, eds. Clinical Chemistry:
determination of acetone using vanillin. Analyst Principles and Technics. 2nd ed. Hagerstown,
1990;115:1385-7. MD: Harper & Row; 1974:Chapter 26.
Tables
Table 1: Methods for Ketone Analysis
Method 1: Colorimetric; semiquantitative and quantitative
Principle of analysis: Na2Fe(CN)5NO + acetone/acetoacetate purple color
Comments: Most common; frequently used as a stat procedure; sensitivity for acetoacetate five times greater than for
acetone; does not detect -hydroxybutyric acid.
Method 2: Enzymatic; quantitative
__________pH 7.0__________
NADH + H+ + AcAc -hydroxybutyrate dehydrogenase -HB + NAD+

pH 8.5-9.5
Comments: Rare, not useful as a stat procedure; by adjustment of pH and cofactor concentration (that is, NADH or
NAD), reaction used to quantitate AcAc and -HB
Method 3: Enzymatic; quantitative
a. Principle of analysis:
NADH + H+ + AcAc + electrons -Hydroxybutyrate dehydrogenase -HB + NAD+
pH 8.5-9.5
Current generated by re-dox reaction detected by electrochemical detector.
Comments: Possibly useful as an office, stat, or home procedure.
b. Principle of analysis:
NADH + H+ + AcAc + electrons -Hydroxybutyrate dehydrogenase -HB + NAD+
pH 8.5-9.5
NADH + oxidized tetrazolium dye reduced tetrazolium dye + NAD+

Colorless colored
Historical and Research Methods for Acetone (see Appendix A)
Method 4: Gas chromatography; quantitative

Principle of analysis: Acetone is detected by flame ionization detector; AcAc is converted to acetone by heating.
Comments: Rare, not useful as stat test; acetoacetate quantitated by subtraction of nonheated acetone from heated
acetone
Method 5: Acetone; gravimetric or titrimetric, quantitative
Principle of analysis: Acetone is precipitated by mercury salts.
Comments: Historical; measures only acetone
Method 6: Acetone; colorimetric, quantitative
Principle of analysis: see Appendix A below
Comments: Rare; measures only acetone
AcAc, Acetoacetic acid; -HB, -hydroxybutyric acid.

791
Ketones
Table 2: Pattern of Clinical Results Found in 22 Hyperglycemic Patients [22]*

Glucose: Mean, 3320 mg/L; range, 18706140 mg/L
Ketones, by Ketostix: 6/22 positive (trace)
Lactate: 1/22 positive (>2 mmol/L)
Pyruvate: 2/22 positive (>0.15 mmol/L)
HBA: 5/22 positive (>0.7 mmol/L)
AcAc: 6/22 positive (>0.2 mmol/L)
AcAc, Acetoacetic acid; HBA, -hydroxybutyric acid.

*Individual ketones and ketone bodies measured by enzymatic assays are described elsewhere in the text. Values in
parentheses, Reference intervals.
Figure
Figure 1
Reaction between acetone and the sodium nitroferricyanide (nitroprusside) formulation found in Ketostix (Ames, Division of
Miles Laboratories, Elkhart, IN).
792
Ketones
Appendix A. Research and Historical Methods for include the reaction between acetone and vanillin at 50C
Measurement of Total Ketones. to 55C to form divanillalacetone (Table 1, Method 6),
which is red (Amax = 415 nm) under the alkaline
Acetone and acetoacetic acid have been quantitated by gas conditions of the reaction [28,29], and the reaction
chromatography (Table 1, Method 4). Acetone is between nitroferricyanide, ferric chloride, and 2,5-
quantified by a flame ionization detector, and acetoacetic dichlorobenzene diazonium chloride [30].
acid is estimated by measurement of acetone before and
after heating to convert the acetoacetic acid to acetone The historical methods for urinary ketones based on
[21,22]. The difference represents the amount of precipitation were time consuming and laborious and are
acetoacetic acid present in the sample. no longer in use. The dinitrophenylhydrazone procedure
requires extraction and is interfered with by ethanol and
High-performance liquid chromatography [23] has also acetaldehyde [17]. The salicylate method is not a useful
been used for quantitation of acetone and acetoacetic acid. quantitative test because of the color instability of the
reaction product. The Gerhardt ferric chloride test for
The oldest quantitative methods for the analysis of ketones acetoacetate is nonspecific, with false-positive results
employed precipitation of acetone with mercury salts, occurring with phenols and salicylates, and has poor
followed by gravimetric or titrimetric analysis (Table 1, sensitivity (250 to 500 mg/L).
Method 5) [24,25]. Colorimetric analyses that have been
used include the reaction of acetone with salicylaldehyde The vanillin method is quantitative and fairly specific but
to form a red color [19, 26] and the formation of measures only acetone. Its precision is reported to be 15%
hydrazones by the reaction between the ketones and for serum acetone levels within the normal range.
dinitrophenylhydrazine [27]. Other colorimetric assays
793
Lactate Dehydrogenase and Lactate Dehydrogenase Isoenzymes

Mauro Panteghini
Name: Lactate dehydrogenase (LD, L-lactate: NAD+ oxidoreductase)

Known isoenzymes: LD1, LD2, LD3, LD4, LD5
Biochemical reaction: (See below)
IFCC reference method: [1]
U N F 2 1 -0 2
L a c ta te
CH 3
D ehyd rogen ase CH 3
p H 8 .8 t o 9 . 8 NADH
H C OH + C O +
NAD
C p H 7 . 4 t o 7 .8 C H
O O O O
L -L a c ta te P y ru v a te

i and H (or B), each under separate genetic control. As a
Lactate dehydrogenase (LD) is a hydrogen transfer result, there are five isoenzymes: H4 (designated LD1
enzyme that catalyzes the oxidation of L-lactate (L) to because of its faster electrophoretic mobility toward the
pyruvate (P) with the mediation of NAD+ as a hydrogen anode), H3M1 (LD2), H2M2 (LD3), H1M3 (LD4), and
acceptor. The reaction is reversible, and the reaction M4 (LD5). A different, sixth LD isoenzyme, LD-X (also
equilibrium is pH dependent, with alkaline pH favoring called LDc), composed of four X (or C) subunits, is
the conversion PL and neutral pH favoring the reverse
reaction. Both P and L in excess inhibit enzyme activity, present in postpubertal human testes. A seventh LD,
although the effect of P is greater. The specificity of the called LD6, has been identified in the sera of severely ill
enzyme extends from L-lactate to various related 2- patients. The most common method used for separation
hydroxyacids and 2-oxo-acids [2]. The catalytic of LD isoenzymes is electrophoresis.
oxidation of 2-hydroxybutyrate, the next higher
homologue of L, to 2-oxobutyrate is referred to as 2- LD activity is present in all cells of the body and is
hydroxybutyrate dehydrogenase (HBD) activity. EDTA invariably found only in the cytoplasm of the cell.
inhibits the enzyme, perhaps by binding Zn2+; however, Enzyme concentrations in various tissues are about 1500
the postulated activator role for zinc ions is not fully to 5000 times greater than those physiologically found in
established. serum. Therefore, leakage of the enzyme from even a
The enzyme has a molecular weight of 134,000 D and is small mass of damaged tissue increases the observed
composed of four peptide chains of two types: M (or A) serum activity of LD to a significant extent. Different
tissues show different isoenzyme composition. In the
heart, kidneys (cortex), and erythrocytes, LD1 and LD2
i
Lactate Dehydrogenase and Lactate Dehydrogenase predominate, whereas in liver and skeletal muscle, LD4
Isoenzymes and LD5 isoenzymes predominatealthough skeletal
muscle damage may also result in anodic LD patterns.
First edition: Amadeo J. Pesce, John F. Chapman,
Isoenzymes of intermediate electrophoretic mobility
Linda L. Woodard, Lawrence M.
account for the LD activity from many sources (e.g.,
Silverman
spleen, lungs, lymph nodes, leukocytes, and platelets)
Methods book: Amadeo J. Pesce, John F. Chapman,
[3].
Linda L. Woodard, Lawrence M.
Silverman
Current routine methods for quantitation of total LD
Second edition: Steven C. Kazmierczak
activity use kinetic spectrophotometry to measure the
Third edition: Monica Payne, Peter E. Hickman
interconversion of the coenzyme NAD+ and NADH at
Fourth edition: Amadeo J. Pesce
340 nm [3-5]. Older reactions used an indicator dye after
Fifth edition: Mauro Panteghini
794
a set period of time (end-point) to determine the amount Interferences

of reactants consumed or formed [6]. For instance, the Any specimen with visible hemolysis will give falsely
tetrazolium reaction oxidizes NADH using a dye system high results because of release of LD from red cells [10-
to form a colored tetrazolium dye. This method is still 12].
used in some electrophoretic assays for isoenzyme
activity detection. LD Reference Intervals
Values for LD activity in serum vary considerably,
Reference and Preferred Methods depending on the direction of the enzyme reaction and
An LP reference method, optimized for LD1, was the method used. The reference interval in adult
developed by the International Federation of Clinical Caucasian subjects, determined with an assay traceable
Chemistry and Laboratory Medicine (IFCC) [7]. This to the IFCC reference procedure at 37C, was found to
method has recently been the basis for developing an be 125 to 220 U/L [13]. LD reference limits are higher in
IFCC primary reference procedure for LD at 37C [1]. children (180 to 360 U/L) [14].
The most widely used procedures for LD determination LD Isoenzymes Reference Intervals
employ the LP reaction, since it is claimed that there Using an agarose gel technique with fluorometric
is less dependence on the NAD+ and the L quantitation of the generated NADH, the following
concentrations and less contamination of NAD+ with reference intervals for isoenzymes were obtained [3].
inhibiting products [1,7].
Measurement Conditions [1] LD Fraction Activity, U/L* Percent

Temperature 37C Total LD 120-250 100
Wavelength 340 nm LD1 17-65 14-26
Buffer, concentration N-methyl-D-glucamine, LD2 35-98 29-39
325 mmol/L LD3 24-65 20-26
pH (at temperature) 9.4 (37C)
LD4 10-40 8-16
Volume fraction of sample 0.0435 (1:23)
Concentration of Reagents in the Final Complete LD5 7-40 6-16
Reaction Mixture *LP; 37C; pH 9.4
L-(+)-Lactate 10 mmol/L
-NAD+ 10 mmolL Interpretation
free acid 3.15 mmol/L Because of its wide distribution in all tissues, increases
lithium salt 6.85 mmol/L in serum LD activity occur in a variety of clinical
conditions, including myocardial infarction, hepatitis,
Electrophoretic separation on agarose gels or cellulose hemolysis, and disorders of kidneys, lung, and muscle. A
acetate membranes is the procedure most commonly systematic review of the literature concluded that serum
used to demonstrate LD isoenzymes [8]. After the LD is only relevant in hematology and oncology [15].
isoenzymes have been separated by electrophoresis, a
reaction mixture is layered over the separation medium. Hemolytic anemias significantly increase LD
The NADH generated over the LD zones is detected concentrations in serum. Marked elevations of the LD
either by its fluorescence when excited by long-wave activityup to 50 times the upper reference limithave
ultraviolet light or by its reduction of a tetrazolium salt been observed in the megaloblastic anemias. These
to form a colored formazan. anemias, usually resulting from the deficiency of folate
or vitamin B12, cause the erythrocyte precursor cell to
Specimen break down in the bone marrow (ineffective
Serum is the preferred specimen for measuring LD erythropoiesis), resulting in the release of large
activity. Plasma samples may be contaminated with quantities of LD1 and LD2 isoenzymes. These
platelets, which contain high concentrations of LD [9]. elevations rapidly return to normal after appropriate
Serum should be separated from the clot as soon as treatment. For monitoring purposes, LD is relevant in
possible after the specimen has been obtained. predicting the survival rate (probability of survival) and
Hemolyzed serum must not be used, because duration in Hodgkins disease and non-Hodgkins
erythrocytes contain 4000 times more LD activity than lymphoma.
serum. The different isoenzymes vary in their sensitivity
to cold, with LD4 and LD5 being especially labile. Patients with malignant disease often show increased LD
Activity of LD4 and LD5 is lost if the samples are stored activity in serum; up to 70% of patients with liver
at 20C. Thus serum specimens should be stored at metastases and 20% to 60% of patients with other
room temperature, at which no loss of activity occurs for nonhepatic metastases have increased total LD activity.
at least 3 days. Notably increased LD is observed in germ cell tumors
(60% of cases), such as teratoma, seminoma of the testis,
and dysgerminoma of the ovary [16]. The percent of
patients with increased LD depended on the stage of the
disease. LD appears to be a useful predictor of outcome
795
in patients with testicular germ cell tumors, melanoma, catalytic concentration of enzymes. Part 8.
and small cell lung cancer. IFCC method for lactate dehydrogenase (L-
Lactate: NAD+ oxidoreductase, EC 1.1.1.27).
LD Performance Goals International Federation of Clinical Chemistry
For a single laboratory, the analytical goals for desirable (IFCC). Eur J Clin Chem Clin Biochem 1994;
performance of LD assays, derived from biological 32: 639655.
variation of the enzyme, are as follows [17,18]: 8 Moses GC, Ross ML, Henderson AR. Ten
electrophoretic methods compared with a
Imprecision (CV): 4.3% selected method for quantifying lactate
Bias: 4.3% dehydrogenase isoenzymes in serum. Clin
Total error: 11.4% Chem 1988; 34: 18851890.
9 Bais R, Edwards JB. Plasma lactate
At LD activity around the upper reference limit, the dehydrogenase activity will be increased if
current interlaboratory coefficient of variation (CV) detergent and platelets are present. Clin Chem
from the 2007 College of American Pathologists CARB- 1977; 23:1056-1058.
B cardiac markers participants survey ranges from 10 Brydon WG, Roberts LG. The effect of
approximately 2% to 6.3% in different peer groups. An hemolysis on the determination of plasma
international study assessing the accuracy of LD results constituents. Clin Chim Acta 1972; 41: 435-
from six diagnostic companies has shown that the results 438.
of none of the companies are within acceptable limits 11 Frank JJ, Bermes EW, Bickel MS, Watkins BF.
when compared to the IFCC reference procedure, Effect of in vitro hemolysis on chemical values
implying that current analytical techniques probably for serum. Clin Chem 1978; 224: 1966-1970.
require significant improvement [19]. 12 Blank DW, Kroll MH, Ruddel ME, Elin RJ.
Hemoglobin interference from in vivo
References hemolysis. Clin Chem 1985; 31: 1566-1569.
1 Schumann G, Bonora R, Ceriotti F, Clerc- 13 Pagani F, Bonora R, Panteghini M. Reference
Renaud P, Ferrero CA, Frard G, et al. IFCC interval for lactate dehydrogenase catalytic
primary reference procedures for the activity in serum measured according to the
measurement of catalytic activity new IFCC recommendation. Clin Chem Lab
concentrations of enzymes at 37C. Part 3. Med 2003; 41: 970-971.
Reference procedure for the measurement of 14 Freer DE, Statland BE, Johnson M, Felton H.
catalytic concentration of lactate Reference values for selected enzyme activities
dehydrogenase. Clin Chem Lab Med and protein concentrations in serum and plasma
2002;40:643-648 derived from cord-blood specimens. Clin Chem
2 McComb RB. The measurement of lactate 1979; 25: 565-569.
dehydrogenase. In: Homburger HA, ed. Clinical 15 Huijgen HJ, Sanders GTB, Koster RW,
and analytical concepts in enzymology. Skokie, Vreeken J, Bossuyt PMM. The clinical value of
IL: College of American Pathologists, lactate dehydrogenase in serum: a quantitative
1983:157171. review. Eur J Clin Chem Clin Biochem 1997;
3 Panteghini M, Bais R, van Solinge WW. 35: 569-575.
Enzymes. In: Burtis CA, Ashwood ER, Bruns 16 Von Eyben FE. A systematic review of lactate
DE, eds. Tietz Textbook of Clinical Chemistry dehydrogenase isoenzyme 1 and germ cell
and Molecular Diagnostics. 4th ed. tumors. Clin Biochem 2001; 34: 441-54.
Philadelphia, PA: Elsevier Saunders, 2006:597- 17 Ricos C, Garcia-Lario JV, Alvarez V, Caval F,
643 Domenech M, Hernandez A, et al. Biological
4 Buhl SN, Jackson KY. Optimal conditions and variation database, and quality specifications
comparison of lactate dehydrogenase catalysis for imprecision, bias and total error (desirable
of the lactate-to-pyruvate and pyruvate-to- and minimum). The 2004 update.
lactate reactions in human serum at 25 , 30 www.westgard.com/guest26.htm
and 37 C. Clin Chem 1978; 24: 828-831. 18 Moses GC, Henderson AR. Biological variance
5 Howell BF, McClure S, Schaffer R. Lactate-to- of total lactate dehydrogenase and its
pyruvate or pyruvate-to-lactate assay for lactate isoenzymes in human serum. Clin Chem 1984;
dehydrogenase: re-examination. Clin Chem 30: 1737-1741
1979;25:269-272 19 Jansen R, Schumann G, Baadenhuijsen H,
6 Spiegel HE, Symington JA, Hordynsky WE, Franck P, Franzini C, Kruse R, et al. Trueness
Babson AL. Colorimetric determination of verification and traceability assessment of
lactate dehydrogenase (L-lactate: NAD results from commercial systems for
oxidoreductase) activity. In Standard Methods measurement of six enzyme activities in serum.
of Clinical Chemistry, . New York: Academic An international study in the EC4 framework of
Press, Inc.;7:43-46 the Calibration 2000 project. Clin Chim Acta.
7 Bais R, Philcox M. Approved recommendation 2006; 368: 160-167.
on IFCC methods for the measurement of
796
Procedure [1] Assay

Equipment: A recording spectrophotometer
Principle equipped with a temperature-controlled cuvette
_ compartment held at 37C.
Lactate + NAD+ lactate dehydrogenase pyruvate +
1. Place 2.0 mL of reaction solution into a test
H+ + NADH cuvette, and equilibrate in a water bath at 37C for 4 to 5
min.
The reaction of L and NAD+ is allowed to occur at an 2. Add 0.1 mL of serum, mix, and incubate for
alkaline pH. One follows the reaction by recording the 180 s.
rate of formation of NADH at 340 nm. The results are 3. Add 0.2 mL of start reagent solution, mix
expressed in U/L. thoroughly, wait 90 s, and start the automatic recording
of the change in absorbance at 340 nm for additional 180
Reagents s.
Reaction solution [7.30 g (373.8 mmol/L) N-
methyl-D-glucamine, 0.552 g (57.50 mmol/L) lactic Calculation
acid, monolithium salt]. Dissolve in about 80 mL Average the change of absorbance (A) per min
water. Adjust to pH 9.4 (37C) with 2 mol/L values. If the A values are decreasing or appear
hydrochloric acid. Transfer to a 100 mL volumetric inconsistent (e.g., not linear), repeat the assay with a
flask. Fill the water (20C) up to the calibration mark of diluted specimen.
the volumetric flask. This solution is stable in a
refrigerator for 1 month.
LD activity (U/L) = A/min ________1________
Start reagent solution [0.240 g (36.23
total vol_
mmol/L) NAD+, free acid, 0.556 g (78.78 mmol/L)
molar absorptivity
NAD+, lithium salt, dihydrate]. The NAD+ free acid/
sample vol
NAD+ lithium salt mixture dissolves rather slowly.
Raising the temperature (up to 40C) speeds up the
dissolving process. Dissolve in about 6 mL water. = A/min _1_ _2.30_ 1000
Transfer to a 10 mL volumetric flask. Fill water (20C) 6.30 0.10
up to the calibration mark of the volumetric flask. This
solution is stable in a refrigerator for 1 week. = A/min 3651
Note: The use of NAD+ free acid without NAD+ lithium
salt for the preparation of the start reagent solution Note: The molar absorptivity of NADH is 630 m2/mol
reasonably decreases the pH value of the final complete at 339 nm. The method should be linear to 600 U/L.
reaction mixture.
797
Lactic Acid
Lactic Acid
Name: L-Lactic acid, L-lactate
Clinical significance: Refer to Chapter 29, Acid-Base Control and Acid-Base Disorders, in the
Merck Index: 5172 to 5174
Chemical class: -Hydroxycarboxylic acid (glucose metabolite)
Structure:
i applications, employ electrochemical detection. The
Measurement of lactic acid, or lactate, in blood is advantage of using biosensors is that the sample can be
important because the concentration of this analyte is assayed immediately after collection without the need for
related to tissue oxygenation. Lactic acid is considered to sample processing [16]. The principle of measurement is
be a reliable indicator of severity, prognosis, and derived from older methods of lactate oxidation by
effectiveness of therapy in patients with sepsis and ferricyanide, catalyzed by yeast L(+)-lactate
multiple organ failure [1,2]. Lactic acid is most commonly dehydrogenase [17-22] (Table 1, Method 3). The
measured by following its oxidation reaction using either hexacyanoferrate(II) produced by the reaction is then
chemical or enzymatic means. Chemical oxidation oxidized at a platinum electrode against a silversilver
methods use either permanganate or manganese dioxide to chloride reference electrode. The amount of current
degrade lactate to acetaldehyde and CO2 or CO. The generated at the platinum electrode is a measure of the
concentration of hexacyanoferrate and thus of lactic acid.
amount of lactate is proportional to the amount of the
products formed. Acetaldehyde can be measured
A dedicated lactate analyzer that uses an enzyme electrode
colorimetrically [3,4,5], by titration [6,7], or by gas
sensor concept has been developed (Table 1, Method 4). A
chromatography [8,9]. The CO formed (Table 1, Method
platinum working electrode measures the hydrogen
lb) can be determined by titration [10] or gasometric
peroxide formed from the reaction, as follows:
analysis [11]. The CO2 formed (Table 1, Method lc) can
lactate oxidase
be quantified gasometrically by a Van Slyke manometer Lactate + O2H2O2 + pyruvate
[12,13].
platinum electrode
Enzymatic methods [14,15] employ lactate dehydrogenase H2O2 2H+ + O2 + 2e-
(LD) to oxidize lactic acid to pyruvic acid, with the
formation of NADH (Table 1, Method 2). Hydrazine or The platinum electrode is kept at an electrical potential of
semicarbazide is added to form a complex with the 0.70 V with a silversilver chloride reference electrode.
pyruvate and remove it as a reaction product, thereby The fabrication of microelectrodes for the measurement of
driving the reaction to completion. The reaction is lactate within single cells has also been described [23]. In
quantitated by measurement of the formation of NADH by this technique, lactate is measured using an enzyme-linked
either absorption spectroscopic or fluorometric methods. assay based upon lactate oxidase and the amperometric
The analysis can be monitored as either an end-point or a determination of hydrogen peroxide [23].
two-point kinetic reaction.
Dedicated lactate analyzers, often used in point-of-care Reference and Preferred Methods
There is no reference method for lactate.
Methods that use chemical oxidation are accurate and
i
Lactic Acid sensitive but time consuming and laborious. Excessive
Previous and current authors of this method: oxidation conditions can lead to further reaction of
First edition: Nancy Gau acetaldehyde and inaccurate results. Thus the reaction
Methods edition: Nancy Gau conditions of the methods measuring acetaldehyde must be
Second edition: Not updated carefully controlled. A gas chromatographic method for
Third edition: Steven C. Kazmierczak measuring the acetaldehyde formed is suitable if the proper
Fourth edition: Nancy Gau equipment is available. The gas-liquid chromatographic
Fifth edition: Steven C. Kazmierczak
798
Lactic Acid
columns, however, are easily contaminated, and care must sodium iodoacetate (final concentration, 0.5 g/L of blood)
be taken to avoid interfering substances such as propylene are stable for up to 2 hours at room temperature [31]. The
glycol and ethanol [9]. iodoacetate does not affect coagulation and thus yields a
suitable serum sample. Heparinized plasma can be used for
The gasometric procedures measuring CO2 produced by lactic acid analysis, but the specimen must be placed on
the oxidation of lactic acid have an advantage over the ice, and the plasma must be separated within an hour of
measurement of acetaldehyde in that excess oxidant, which collection. Plasma samples can also be collected into tubes
can cause further oxidation of the acetaldehyde, does not containing fluoride and oxalate (60 and 12 mmol/L,
affect the formation of CO2. The other gasometric respectively). Lactate is stable for up to 8 hours at room
temperature in tubes collected in this manner [29].
procedure, based on the analysis of CO, is less accurate
and shows a positive bias with both blood and urine
In terms of convenience and accuracy, the recommended
[10,24]. Both gasometric procedures require gasometers,
sample for lactic acid analysis is a serum sample drawn in
which are not routinely available in most laboratories.
a sodium iodoacetate phlebotomy tube. The sample should
be drawn with minimal stasis, preferably without a
Electrochemical devices for lactate analysis are very rapid
tourniquet. If a tourniquet must be used, it should be
(less than 3 min), are automated, and can be used with
released after venepuncture, and the phlebotomist should
whole blood.
then allow several minutes to pass before actually drawing
blood into the phlebotomy tube. Patients should avoid
Enzymatic methods allow direct measurement of lactic
strenuous use of the hand or arm prior to sampling.
acid, are simple to perform, and are adaptable to widely
available automated equipment. For these reasons, the
Interferences
enzymatic procedure employing lactic dehydrogenase is
The presence of hemolysis or high bilirubin concentrations
the method of choice. The Hohorst manual method is an
have been reported to interfere with some methods. High
end-point measurement in the presence of semicarbazide
concentrations of pyruvate and malic acid can interfere
[25]. These reactions generally require a long incubation
with some enzymatic methods [32].
period and are very dependent on the concentration of
various reactants [26].
Lactic Acid Reference Interval
A two-point kinetic method was introduced using an The definition of significant lactic acidosis is defined as a
alkaline pH with hydrazine added to consume the pyruvate blood lactate of 5.0 mmol/L or greater, with an arterial pH
of less than 7.25. In healthy, fasting individuals, less than 2
formed. The method is easily adapted to many automated
analyzers. One limitation of the method is the low upper mmol/L is present in venous blood. Lactate levels in
limit of linearity (5 mmol/L), which requires appropriate arterial blood can be half to two-thirds the level of venous
dilutions of those samples with a moderately elevated blood. These values assume that the patient is at rest;
lactate concentration. The fluorescence method, which higher levels can be seen otherwise. Lactic acid levels
detects NADH by its emission at 450 nm, is an acceptable increase 20% to 50% over baseline fasting values after a
meal [33]. Lactate in capillary blood obtained from
alternative to measurements using absorption
newborns is approximately 50% higher compared with
spectroscopy.
adults [34]. CSF lactate levels usually correlate with levels
seen in blood, although CNS disorders such as bacterial
Specimen
meningitis, cerebrovascular accidents, and intracranial
Rigorous sample processing is necessary for obtaining
hemorrhage can change CNS lactate independently of
accurate lactic acid measurements because blood cells will
those levels seen in blood [35,36]. Cerebrospinal fluid
metabolize glucose to lactic acid, resulting in a positive
lactic acid levels in children 0 to 16 years of age ranged
bias [27,28]. This metabolism is time and temperature
from 1.1 to 2.8 mmol/L [37]. Lactate may also be
dependent. Increases in lactate of 0.4 mmol/L when placed
measured in urine. The normal 24-hour excretion of lactate
on ice and 0.7 mmol/L at room temperature 1 hour
is 5.5 to 22 mmol/day.
following collection have been described [29].
Interpretation
Preservation of lactate concentrations by deproteinization
Lactic acid is a product of the metabolism of pyruvic acid.
of whole blood or serum is neither convenient nor reliable
Healthy individuals produce approximately 1400 mmol of
[30]. Studies on the effect of sodium fluoride and sodium
lactic acid per day, which is produced primarily from
iodoacetate showed an inhibitory effect on glycolysis with
glycolysis and deamination of alanine [38]. Analysis of
either compound and complete inhibition with a mixture of
lactic acid can be used to determine the metabolic basis for
1% sodium fluoride and 1% iodoacetate [28]. Since
an unexplained metabolic acidosis, as determined by a
sodium fluoride can partially inhibit coagulation, the
blood gas analysis. The presence of an elevated anion gap
serum specimens from sodium fluoride phlebotomy tubes
in the absence of an obvious cause of the acidosis, such as
tend to be contaminated with protein clots. One study
renal failure, hyperglycemic ketosis, or sepsis, would
showed that lactic acid levels in samples collected with
prompt a clinician to order a lactic acid analysis [39].
799
Lactic Acid
An elevated lactic acid level would indicate that the produced in humans. The production of D-lactic acid by
metabolic acidosis may be a lactic acidosis. Lactic acid is bacteria in the colon can cause D-lactic acidosis. Patients
formed in muscle cells to maintain cellular levels of NAD. who have undergone jejunoileal bypass surgery can
As the oxygen level in muscle cells decreases, the cells present with extremely high plasma concentrations of this
must increasingly depend on anaerobic metabolism by the metabolite during acidotic episodes as a result of increased
glycolytic pathway to generate ATP. To regenerate enough production by intestinal bacteria due to malabsorption of
NAD to allow this pathway to continue metabolizing carbohydrates [43]. Essentially all of the methods in use in
glucose, NADH+ is oxidized by reduction of pyruvic acid, clinical laboratories utilize L-lactate dehydrogenase, which
the final product of the Embden-Meyerhof pathway, to does not detect D-lactate. D-Lactate can be measured using
lactic acid by the action of lactate dehydrogenase. Muscle enzymatic assays that utilize D-lactate dehydrogenase.
lactic acid accumulated during periods of tissue anoxia is
released into the general circulation where it can be readily Lactic Acid Performance Goals
metabolized by the heart (rich in LD1) and liver. Participant summary reports from 2008 by the College of
American Pathologists indicate that the majority of
Serum lactic acid concentration can rise accompanying the laboratories measure lactic acid using an automated
tissue hypoxia associated with vigorous exercise. This can enzymatic assay. Less than 5% report using an electrode-
result in a mild lactic acidosis that disappears as soon as based system for measurement of lactate.
the exercise ceases. Muscle pain and cramping can
accompany this lactic acidosis. A life-threatening lactic Acceptable performance criteria for lactic acid requires
acidosis can be seen when the tissue hypoxia is caused by that laboratories obtain lactate values that are within 2 SD
respiratory failure or hypoperfusion states (i.e., shock or or 0.2 mmol/L of the peer-group mean, whichever is
circulatory collapse) [39]. Severe dehydration can also greater. Lactic acid is not currently regulated for
lead to reduced delivery of oxygen to muscle cells, and proficiency testing. Interindividual biological variation
lactic acidosis may often accompany diabetic ketoacidosis measured in 8 healthy individuals over a 5-day period was
and hyperglycemic type 2 diabetes mellitus. Increased determined to be 16.7% [44]. Desirable specifications for
tissue oxygen consumption, as seen in sepsis and analytical imprecision derived from studies of biological
malignancies, can also result in a lactic acidosis. In all variation indicate an assay imprecision of no greater than
these cases, the lactic acidosis will remain until the cause 13.6% [45]. Methods used in clinical laboratories for
of the reduced delivery of oxygen to tissues is corrected. measurement of lactate show analytical imprecision that is
This usually requires correction of the respiratory or well within the limits deemed necessary for medical
cardiac disorder or rehydration of the patient. decision making. The vast majority of lactate methods
used in clinical laboratories are enzymatic methods
The degree of acidemia due to lactate can often help involving the oxidation of lactate to pyruvate. A much
indicate the severity of the underlying disease. The smaller percentage, approximately 5%, utilize methods
survival rate for patients with very elevated blood lactic involving the oxidation of lactate to pyruvate.
acid levels (>10.5 mmol/L) was worse (30%) than for
patients with lower lactate levels (65%) [40]. There is a References
report of lactic acidosis (lactate, 18 mmol/L; pH, 7.25) 1. Bakker J, Coffernils M, Leon M, Gris P, Vincent
resulting from the ingestion of large amounts of propylene JL. Blood lactate levels are superior to oxygen-
glycol [40]. The high lactate level resulted from hepatic derived variables in predicting outcome in human
metabolism of the propylene glycol to lactate. septic shock. Chest 1991;99:956-962.
2. Pittard AJ. Does blood lactate measurement have
There have been several reports on the relationship a role in the critically ill patient? Ann Clin
between cerebrospinal fluid (CSF) lactic acid levels and Biochem 1999;36:401-407.
central nervous system (CNS) infections [41,42]. Increased 3. Hochella NJ, Weinhouse S. Automated lactic acid
CSF lactate is usually associated with decreased CSF determination in serum and tissue extracts. Anal
glucose concentrations because of increased anaerobic Biochem 1965;10:304.
glucose utilization. Elevated lactate levels (>4.0 mmol/L) 4. Barker SB, Summerson WH. Colorimetric
are associated with CNS infections of bacterial, fungal, determination of lactic acid in biological material.
and mycobacterial, but not viral, origin [42]. In known J Biol Chem 1941;138:535-554.
bacterial meningitis, decreasing CSF lactate levels reflect 5. Barker SB. Lactic acid. In: Seligson D, ed.
successful treatment of the disease. Increased CSF lactic Standard Methods in Clinical Chemistry. 3rd ed.
acid can also result from impaired CNS blood supply New York: Academic Press; 1961:167.
because of space-occupying tumors. Thus elevated CSF 6. Friedman TE, Graeser JB. The determination of
lactate has a high sensitivity for a CNS bacterial infection lactic acid. J Biol Chem 1933;100:291-308.
but a lower specificity for the disease. 7. Long C. The stabilization and estimation of lactic
acid in blood samples. Biochem J 1946;40:27-33.
Unless otherwise specified, lactic acidosis refers to 8. Hoffman NE, Barboriak JJ, Hardman HF. A
acidosis caused by L-lactic acid. This is the normal isomer sensitive gas chromatographic method for the
800
Lactic Acid
determination of lactic acid. Anal Biochem 25. Hohorst HJ. In: Bergmeyer HU, ed. Methods of
1964;9:175-179. Enzymatic Analysis. New York: Academic Press;
9. Savory J, Kaplan A. A gas chromatographic 1963:266.
method for the determination of lactic acid in 26. Olson GF. Optimal conditions for the enzymatic
blood. Clin Chem 1966;12:559-569. determination of L-lactic acid. Clin Chem
10. Ronzoni E, Wallen-Lawrence Z. Determination of 1962;8:1-10.
lactic acid in blood. J Biol Chem 1927;74:363- 27. Drewes PA. In: Clinical Chemistry: Principles
377. and Technics. 2nd ed. Hagerstown, MD: Harper
11. Schneyer J. Eine method zur quantitativen & Row;1974:1330-1334.
Milchsaurebestimmung in Harne. Biochem Z 28. Bueding E, Goldfarb W. The effect of sodium
1915;70:294-299. fluoride and sodium iodoacetate on glycolysis in
12. Avery BF, Hastings AB. A gasometric method for human blood. J Biol Chem 1941;141:539-544.
the determination of lactic acid in the blood. J 29. Astles R, Williams CP, Sedor F. Stability of
Biol Chem 1931;94:273-280. plasma lactate in vitro in the presence of
13. Baumberger JP, Field J. Manometric method for antiglycolytic agents. Clin Chem 1994;40:1327-
quantitative determination of lactic acid. Proc Soc 30.
Exp Biol 1927;25:87-98. 30. Pesce AM, Bodourian SH, Nicholson JF. Rapid
14. Pfeiderer G, Dose K. Eine enzymatiche kinetic measurement of lactate in plasma with a
Bestimmung der L(+) Milchsaure mit centrifugal analyzer. Clin Chem 1975;21:1932-
Milchsauredenhdrase. Biochem Z 1955;326:436- 1934.
439. 31. Kaplan LA, Gau N, Stein EA. Collection and
15. Parijs J, Barbier F. The enzymatic L(+)-lactate storage of serum lactic acid samples at room
determination in blood. Z Klin Chem 1965;3:74- temperature without deproteinization. Clin Chem
78. 1980;26:175-176.
16. Lowinsohn D, Bertotti M. Flow injection analysis 32. Fleisher WR. Enzymatic methods for lactic and
of blood L-lactate by using a Prussian Blue-based pyruvic acids. Stand Meth Clin Chem
biosensor as amperometric detector. Anal 1970;6:245-259.
Biochem 2007;365:260-265. 33. Friedman TE, Haugen GE, Kmieciak TC. Pyruvic
17. Williams DL, Doig AR, Korosi A. acid. J Biol Chem 1945;157:673-689.
Electrochemical analysis of blood glucose and 34. Nielsen J, Ytrebo LM, Borud O. Lactate and
lactate. Anal Chem 1970;42:118-121. pyruvate concentrations in capillary blood from
18. Durliat H, Comtat M, Baudras A. newborns. Acta Paediatr 1994;83:920-922.
Spectrophotometric and electrochemical 35. Pryce JD. A modification of the Barker-
determinations of L(+)-lactate in blood by use of Summerson method for the determination of
lactate dehydrogenase from yeast. Clin Chem lactic acid. Analyst 1969;94:1151-1152.
1976;22:1802-1805. 36. Glasser L. Body fluids: tapping the wealth of
19. Racine P, Engelhardt R, Higelin JC, Mindt W. An information in CSF. Diag Med 1981;Jan-Feb:23-
instrument for the rapid determination of L-lactate 33.
in biological fluids. Med Instrum 1975;9:11-14. 37. Knight JA, Dudek SM, Haymond RE. Increased
20. Racine P, Klenk HO, Kochsiek K. Rapid lactate cerebrospinal fluid lactate and early diagnosis of
determination with an electrochemical enzymatic bacterial meningitis. Clin Chem 1979;25:809-
sensor: clinical usability and comparative 810.
measurements. J Clin Chem Clin Biochem 38. Marko P, Gabrielli A, Caruso LJ, Mizock BA,
1975;13:533-539. Franklin C. Too much lactate or too little liver? J
21. Soutler WP, Sharp F, Clark DM. Bedside Clin Anesthesia 2004;16:389-395.
estimation of whole-blood lactate. Br J Anaesth 39. Narins RG, Rudnick MR, Bastl CP. Lactic
1978;50:445-450. acidosis and the elevated anion gap (II). Hosp
22. Piquard F, Schaefer A, Haberey P. Bedside Pract 1980;June:91-98.
estimation of plasma lactate. Clin Chem 40. Anderson CT, Westgard JO, Schlingen K,
1980;26:532-533. Birnbaum ML. Contribution of arterial blood
23. Cai X, Klauke N, Glidle A, Cobbold P, Smith lactate measurement to the care of critically ill
GL, Cooper JM. Ultra-low-volume, real-time patients. Am J Clin Pathol 1977;68:63-67.
measurements of lactate from the single heart cell 41. Bland RD, Lister RC, Reis JP. Cerebrospinal
using microsystems technology, Anal Chem fluid lactic acid level and pH in meningitis. Am J
2002;74:908-914. Dis Child 1974;128:151-156.
24. Maver MC. The Schneyer method for the 42. Brook I, Bricknell KS, Overturf GD, Finegold
determination of lactic acid in urine. J Biol Chem SM. Measurement of lactic acid in cerebrospinal
1917;32:71-76. fluid of patients with infections of the central
nervous system. J Infect Dis 1978;137:384-390.
801
Lactic Acid
43. Bongaerts G, Tolboom J, Naber T, Bakkeren J, 45. Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
Severijnen R, Willems H. D-lactic acidemia and Hernndez A, Jimnez CV et al. Current database
aciduria in pediatric and adult patients with short on biologic variation: pros, cons and progress.
bowel syndrome. Clin Chem 1995;41:107-110. Scand J Clin Lab Invest 1999;59:491-500.
44. Panteghini M, Pagani F. Biological variation of
lactate and pyruvate in blood. Clin Chem
1993;39:908.
Table
Table 1: Lactic Acid Methods Summary Table
Method 1: Chemical oxidation with formation of:
Principle of analysis: Permanganate or MnO2 oxidizes lactate to one of several products in one of the following
reactions
Comments: Historical; accurate, precise, and sensitive, but time consuming
Method 1a: Acetaldehyde (CH3CHO); quantitative, end-point
Lactate H2SO4, heat acetaldehyde + H2O + CO
Method 1b: CO (carbon monoxide); quantitative, end-point
Principle of analysis: As in Method 1a
Method 1c: CO2 (carbon dioxide); quantitative, end-point
Principle of analysis: 2 lactate + O2 2 acetaldehyde + 2CO2 + H2O
Method 2: Enzymatic; quantitative, end-point or kinetic
Principle of analysis: Lactate + NAD+ LD pyruvate + NADH + H+
Comments: Automated; accurate and precise; automatable and rapid
Method 3: Electrochemical; quantitative, end-point

Lactate + 2Fe(CN)63 pyruvate + 2H+ + 2Fe(CN)64
2Fe(CN)64 Pt. electrode 2FE(CN)63 + 2e
Comments: Automated; accurate, precise, and sensitive, rapid analysis, but equipment not readily available
Method 4: Enzyme electrode sensor; quantitative, end-point
Lactate + O2 LO H2O2 + pyruvate
H2O2 2H+ + O2 + 2e
Current measured is proportional to hydrogen peroxide concentration
Comments: Semiautomated
K, Kinetic analysis mode; EP, end point analysis mode; LD, lactate dehydrogenase; LO, lactate oxidase.
Procedure: Enzymatic Analysis for Lactic Acid removes the pyruvate as it is formed, further shifting the
Editors Note: While the following assay was developed equilibrium to the right. The rate of increase in absorption
for a specific analyzer, it can easily be adapted to any of the NADH formed, measured at 340 nm, is related to
the concentration of lactic acid in the sample. Instructions
automated spectrophotometer.
are given for an ABA-100 and a centrifugal analyzer.
However, this procedure is easily adapted to any random
Principle access analyzer.
This method is based on the enzymatic oxidation of lactate
to pyruvate by lactate dehydrogenase in the presence of Reagents
NAD. The reaction is as follows: 1. TrisEDTAhydrazine buffer (499, 11.9, and
LD, pH 9.8 226 mmol/L, respectively). Dissolve 60.5 g of
L-Lactate + NAD + pyruvate + NADH + H+ hydroxymethylaminomethane (Tris) and 4 g of disodium
ethylenediaminetetraacetate in 800 mL of distilled water.
At this pH, equilibrium favors oxidation of lactate to Add 11 mL of hydrazine hydrate. Adjust the pH to 9.8, and
pyruvate. If hydrazine is added, it reacts with pyruvate and dilute to 1 L with distilled water. Stable 6 months at 4C.
2. Beta-NAD (Product no. 7004, Sigma Chemical
802
Lactic Acid
Co., St. Louis, MO). Preweigh 65.4 mg (0.1 mmol) beta- 5. Set 0 absorbance on position 1 with zero knob.
NAD into test tubes. Cover, and store at 4C. Stable for at 6. Push calibrate button in, and set to 0.500 with
least 1 month. Dissolve preweighed NAD into 3 mL of scale knob.
distilled water before preparing the working reagent. 7. Return carousel to starting position, and push run
3. Lactate dehydrogenase (Boehringer button.
Mannheim Catalog no. 127230). An ammonium sulfate 8. When carousel begins to move, spin it back to the
suspension of purified rabbit muscle LD, approximately starting position. It will now print first absorbance
550 U/mg. readings.
4. Working reagent. Mix 27 mL of buffer, 3 mL of 9. After second print (5 min later), perform
dissolved NAD, and 40 L of LD solution. Mix. Stable 24 calculations.
h at 4C.
5. Stock lactate standard (20 mmol/L). Dissolve Calculation
192 mg of lithium L(+)-lactate (available from the 1. Calculate change of absorbance (A) for each
National Bureau of Standards) in 100 mL of distilled cuvette. Subtract first print from second print.
water. Stable 6 months at 4C. 2. Use the following formula:
6. Working standards
2 mmol/L. 1 mL of 20 mmol/L standard, diluted Absorbance UNK
volumetrically to 10 mL with distilled water. Conc STD Conc UNK (nmole / L)
5 mmol/L. 2.5 mL of 20 mmol/L standard, diluted AbsorbanceSTD
volumetrically to 10 mL with distilled water.
Both standards are stable for 2 months at 4C. 3. For unknowns (UNK) less than 5 mmol/L, use the
Assay (Reaction Conditions for Lactate Analysis on ABA- 2 mmol/L standard (STD) for the calculation. For
100 Table ) unknowns greater than 5 mmol/L, use the diluted sample
Equipment: ABA-100 (Abbott Diagnostics, for calculation, again either the 2 or 5 mmol/L standard
Chicago, IL) (click here).
1. Test parameters: Notes
Temperature 37C 1. The between-run coefficient of variation for the
Mode End-point ABA-100 enzymatic method is 8.9% at a lactate
Normal/fast reaction rate Normal concentration of 4 mmol/L.
Direction Up 2. This method has also been adapted to a
Time 5 min centrifugal analyzer (CentrifiChem 400, Baker
Revolutions 3 Instruments, Allentown, PA). Use the following
Filter 340/380 parameters on this instrument:
Syringe plate 1:51 Pipettor Analyzer
Decimal 0.000 10 L of sample 30C
Calibration factor 0.500 50 L of diluent 340 nm
Set after initial dispersing. 350 L of reagent t0 at 3 sec
Zero setting 0
(place reagent in 0 cuvette) t at 4 min
2. Follow normal operational procedures according
Automatic mode
to Abbott ABA-100 instrument manual.
Terminal Concentration
3. Analyze both the 2 and 5 mmol/L lactate
Follow normal pipetting and operational guidelines for the
standards in each run. Analyze patient samples, undiluted
CentrifiChem as outlined in the CentrifiChem procedure
and diluted fivefold with distilled water.
manual. Set print to 2 or 5 mmol/L standard, according to
4. After the last sample is aspirated and dispensed,
the approximate concentration of the unknown.
push stop button. Go to test mode.
803
Lead
Lead
Gus Koerbin
Name: Lead
Molecular formula: Pb
Merck Index: 5227
i urine sample [7,10]. After sample preparation, the
Lead, a periodic-system Group IVB metal, occurs in solution of ionic lead is aspirated into an air-acetylene
three oxidation states, 0, +2, and +4. The +2 oxidation flame, where it is reduced to the atomic state.
state is the most widely occurring form of the inorganic Quantitation is accomplished by measurement of the
salts of lead, whereas Pb(IV) is the most important state light absorbed by the lead at 283.3 nm.
in organolead compounds. A complete discussion on
lead regarding sources of occupational exposure has Although numerous variations in this AAS procedure
been published [1]. have been published, two are noteworthy. The first
features co-precipitation of the lead with bismuth
The method historically used to analyze lead in both hydroxide after sample digestion [8]. After
biological and environmental samples is the dithizone centrifugation and supernatant fluid removal, the
compleximetric method (Table 3, Method 1) [2-4]. The precipitate is dissolved in acid for aspiration into the
sample, 10 mL of blood or 25 mL of urine, tissues, or flame. The second method features chelation of the lead
particulates, is ashed by acid oxidation to remove with pyrrolidine dithiocarbamate after sample digestion
organic material. After ashing, the sample is redissolved [7,10]. This lead chelate is extracted into methyl isobutyl
in acidified water. The pH is adjusted to 9-10, cyanide is ketone (MIBK), and then the MIBK extract is aspirated
added to eliminate reaction with other metals, and then into the flame.
the sample is reacted with diphenylthiocarbazane to form
the red leaddithizonate complex. The complex is In addition to these variations, a blood micromethod for
extracted into chloroform, and the absorbance of the flame AAS is also widely used [12,13]. This procedure,
complex is measured spectrophotometrically at 510 nm the Delves microscale technique [9], does not use the
(Figure 1). acid oxidation sample-ashing procedure. In the Delves
cup method, 10 L of blood, obtained by a finger stick,
Many methods of sample preparation, concentration, and is added to a special nickel cup that contains hydrogen
spectrophotometric techniques for atomic absorption peroxide to partially oxidize the sample. After the
spectroscopy (AAS) have been reported [5-11]. sample preparation procedure, the cup is inserted into the
However, all have the common features of (1) a AAS flame for the lead determination (Table 3, Method
digestion or treatment of the sample to allow separation 2b).
of the lead from the sample matrix, (2) aspiration of the
ionic lead into a flame or heating of the ionic lead Other major variations in the classical flame AAS
solution by electrical means, where it is reduced to the procedures now widely used are the flameless or
atomic state, and (3) spectrophotometric quantitation by electrothermal AAS procedures that are otherwise
measurement of the light absorbed by the lead at its referred to as the graphite furnace [6] and carbon rod
characteristic resonance frequency of 283.3 nm [5-8,10]. [11] techniques (Table 3, Method 2c). Some
A sample-ashing procedure using acid oxidation similar electrothermal procedures [6] feature the sample-ashing
to that used for the dithizone method [2-4] (Table 3, by acid oxidation, as described for the classical AAS
Method 2a) typically requires a 10-mL blood or 50-mL methods, whereas in others a microsample, 5 L, is
introduced directly into the instrument [11]. The
electrothermal AAS procedures feature a graphite tube
i
Lead or carbon rod held on the optical axis of the AAS unit.
Previous and current authors of this method: The tube is protected from the atmosphere by flowing
nitrogen or argon. After sample introduction, the
First edition: M. Wilson Tabor
graphite tube or carbon rod is heated by electrical
Methods edition: M. Wilson Tabor
resistance successively to a drying temperature of
Second edition: M. Wilson Tabor 100C, followed by a charring temperature of 400C,
Third edition: M. Wilson Tabor and finally to an atomization (reduction) temperature of
Fourth edition: M. Wilson Tabor, Tania Carreon- 2000C to 2500C. During atomization, a high
Valencia
concentration of Pb0 is formed in the light path, but the
Fifth edition: Gus Koerbin
804
Lead
atoms quickly escape by diffusion. Lead absorbs light at the gamma rays emitted from the radionuclide 207Pb
both 280.2 and 283.3 nm, and the absorbance is
isomeric transition to the stable 206Pb isotope. The
proportional to the lead concentration in the sample.
theory and practical applications of this technique have
been described in numerous reviews [17].
The third general category of lead analytical methods is
electrochemical. These anodic stripping voltammetry Reference and Preferred Methods
(ASV) instrumental methods have become widely There is no reference method for the determination of
accepted methods for lead analysis [14-17] (Table 3, blood lead. There is a definitive method for lead
Method 3). In this methodology, ionic lead in blood or determination in both whole blood and urine, employing
urine samples is first reduced to elemental lead by a isotope dilution inductively coupled plasma mass
negative potential applied to a working mercury spectrometry (ICP/MS) [18] (Table 3, Method 4). This
electrode. The reduced lead is deposited on the mercury method has been employed by the Centers for Disease
electrode. Control and Prevention (CDC) for establishing lead
Pb+2 + 2 e- Pb0 (deposited on mercury electrode) concentrations in their samples for the licensure and
proficiency testing program [19].
After a preselected time, the working electrode potential The choice of methodology for the
is moved toward more positive values, thereby causing determination of lead in a particular laboratory is
the reoxidation of the lead from the mercury electrode. dependent on the following several factors: the
From this oxidation the resulting anodic current, which availability of equipment, the number of samples to be
is proportional to the concentration of lead, is measured. analyzed per day, the purpose for the analysis, and
Pb0 (from mercury electrode) Pb2+ + 2 e- experience of those undertaking the analysis.
The atomic absorption method has been
published as a selected method of the American
The ASV technique has become a reliable and
Association for Clinical Chemistry [8] and has shown
reproducible method for microliter-sized blood samples
accuracy, in terms of relative recovery, of 98.8% 1.0%
because of the design of an efficient mercurygraphite
for lead in blood, 99% 4.5% for lead in urine, and 97%
composite electrode. This electrode has been
3.6% for lead in hair.
incorporated into equipment for the semiautomated
determination of lead in blood, urine, water, foods, and
The dithizone procedure is subject to certain
reagents. Sample preparation for ASV is facilitated by
interferences, but the method has the advantage of being
the use of the Metexchange reagent, a chromate-based
thoroughly tested and evaluated. Disadvantages of the
digestion solution containing an antifoaming surfactant.
method include use of large quantities of scrupulously
The sample preparation is simplified to the addition of
cleaned glassware, large quantities of reagents, and the
sample to the Metexchange reagent. The instruments are
procedure is highly susceptible to contamination from
designed to measure the current generated during a
outside sources and is tedious and time consuming.
specific window of voltage during which lead is
stripped off the mercury-graphite electrode. More
Atomic absorption spectroscopy is recommended as a
accurate values may be obtainable if the current change
general procedure. The utility of this method is its wide
during analysis is plotted on a strip recorder. Usually the
applicability. AAS methods using an electrothermal
peak area of the lead peak is used for calculations, but
furnace (flameless AAS) have the advantage of this
peak height can be used as well. Two ASV methods are
greater sensitivity of one to two orders of magnitude
recommended by The National Institute for
over flame AAS procedures [6,11]. Disadvantages of the
Occupational Safety and health (NIOSH), one requiring
electrothermal AAS procedure are that sample matrix
100 L of blood and the other requiring 1.0 mL of urine
effects are more prominent than in flame atomic
[14-16]. Blood ASV methods are widely applicable to
absorption, analysis is time consuming, and the
routine monitoring of biological samples during
equipment is expensive compared to flame AAS. The
treatment for lead intoxication and for studies of lead
flame AAS procedure utilizing the MIBK extraction
exposure in children and in workers who are
technique offers several advantages [7], such as speed of
occupationally exposed, such as in the smelting and
analysis, no requirement for a high level of technical
battery industries.
skill, and no known interferences. The main
disadvantage is the requirement for larger sample sizes
There are several nonroutine methods for the analysis of
than those in some of the other AAS [9] and ASV [4,16]
lead in a variety of sample matrices [12,13]. These
procedures. The principal advantage of the bismuth
methods include electron microprobe, x-ray
hydroxide precipitation method [8] is its speed, making
fluorescence, neutron activation analysis, and mass
it practical for routine monitoring of biological samples
spectrometry. All require the use of expensive facilities
during treatment for lead intoxication.
and a high degree of expertise for the analyst. In neutron
activation analysis for lead, the sample is irradiated with
The Delves cup method [9] has the advantage of small
fast neutrons, energy greater than 1 MeV, converting the
sample size, speed, and the requirement for only a
stable 206Pb isotope to the metastable radioisotope minimum of technical competence. Disadvantages
207Pb. Quantitation is accomplished by measurement of include interferences from background absorptions and
matrix effects and variations in lot-to-lot cup quality.
805
Lead
because the currentvoltage peak for this element is not

ASV methods have high sensitivity and are less resolved from that for lead when the two metals are at
expensive per analysis than other methods in terms of comparable concentrations.
equipment cost and time required for analysis. The
principal disadvantage is that thallium interferes because Lead Reference Intervals
the currentvoltage peak for this element is not resolved There are two classes of persons who must be considered
from lead when the two metals are at comparable in regard to lead exposure: those exposed occupationally
concentrations. The ASV methods are recommended for
and those exposed environmentally. The Occupational
laboratories having a need to do lead analyses on a
Safety and Health Administration (OSHA) has two
routine basis, particularly for doing monitoring studies.
The AAS methods are recommended for laboratories standards for lead: 29 CFR 1926.62 covers construction
having a need to do not only lead but also other metals industry activities, and 29 CFR 1910.1025 covers
on a routine basis, particularly in doing analyses on general industry such as foundries and the manufacture
samples in addition to blood and urine. of lead-containing products [24,25]. These laws require
the employer to keep lead exposures at or below 50
Specimen g/m3 of air. This level is called the Permissible
Whole blood is the sample of choice, but urine can be Exposure Limit (PEL), based on an 8-hour time-
used for determining lead in humans. Blood must be weighted average (TWA).The World Health
collected in lead-free heparinized vacutainers. EDTA Organization (WHO) has stated that blood lead levels >
may also be used as an anticoagulant. Caution should be 30 g/dL indicate significant exposure. The maximum
exercised during sample handling to prevent any external allowable blood lead concentration in workers has been
contamination. No differences in blood lead set at 60 g/dL (2.9 mol/L), a level at which health
concentrations have been observed between properly effects are observed. The typical range for healthy non-
collected venous and capillary specimens. Heparinized, exposed adults is 10 to 20 g/dL (0.48 to 0.96 mol/L),
refrigerated blood samples are stable for 2 weeks, but with values for children being lower [1,22,26]. Blood
EDTA blood samples are stable for several months if and urine lead concentrations have been reported to be
frozen at 20C. The sample is suitable for analysis by about 10% to 20% higher in males than females [26].
all four methods listed in Table 3. If EDTA is used as an
anticoagulant, 1.4 mg of calcium chloride per milliliter Interpretation
of blood must be added to enhance lead recovery from The pathological effects of lead have been recognized
the samples [20]. for centuries. Many physiological systems, including
renal, nervous, reproductive, endocrine, immune, and
For the extraction/AAS procedure utilizing ammonium hemopoietic, are affected. These effects have been the
pyrrolidine dithiocarbamate as a chelating agent [7,10], subject of numerous reviews and research publications.
the calcium chloride should be added to the sample after For example, Damstra [27] has presented an overview of
the addition of the chelating agent [21]. the toxicological properties of lead, and Mahaffey [28]
has presented an overview of sources of human exposure
Urine samples must be collected in lead-free borosilicate
and bioavailability.
or polyethylene bottles. At least 50 mL should be Exposure to lead, either acute or chronic,
collected and the specific gravity measured. The sample presents a variety of signs, symptoms, and chemical
must be preserved by the addition of 500 mg of thymol evidence. The exposed person may be asymptomatic or
per liter of urine. The urine is stable for 1 week if symptomatic. A number of signs and symptoms of
refrigerated. The sample is suitable for analysis by all exposure have been observed. For exposed adults, Lane
three methods. et al. [29] listed tiredness, lack of energy, constipation,
slight abdominal discomfort, anorexia, altered sleep,
Interferences irritability, anemia, pallor, and less frequently diarrhea
Some lots of heparin used as an anticoagulant for blood and nausea. More severe signs and symptoms include
lead measurements have been reported to contain lead colic, reduction of muscle power, muscle tenderness,
[22]. Lead AAS measurement by direct aspiration of paresthesia, and symptoms or signs of neuropathy or
urine has been shown to give falsely elevated results. encephalopathy. A short list, in descending order of
This interference is caused by nonspecific absorption of frequency of presentation, has been published for
exposed adults [30]. These include abdominal pain,
urinary constituents such as calcium, creatinine, and urea
constipation, vomiting, non-abdominal pain, seizures,
and may falsely increase the result by up to 100 g/L.
paresthesia, psychological symptoms, and diarrhea. The
By subtracting the absorbance reading at 220 nm (non- symptoms most frequently seen in children are
lead peak) from the lead peak at 217 nm, correction for irritability, vomiting, abdominal pain, ataxia, anorexia,
this interference is achieved. This correction is behavioral changes, speech disturbances, seizures,
applicable to both direct aspiration of urine and intercurrent fever, and dehydration [31,32].
determinations where there is co-precipitation by A number of reviews of the clinical
bismuth [23]. The dithizone procedure is subject to presentation of lead exposure have been published
interference from tin(II), bismuth, and thallium. The [13,33]. Other biochemical parameters indicative of lead
ASV methods are subject to interference from thallium,
806
Lead
exposure include delta-aminolevulinic acid, delta- Blood lead concentration 20-24 g/dL (0.97-1.16
aminolevulinic acid dehydratase, and porphyrins. mol/L). Minimal data exists on chelation therapy for
Toxic Levels of Lead in Blood children in this range, and children probably should not
There is no safe blood lead concentration. Virtually any be chelated except in the context of approved clinical
amount of lead in blood is now considered to have some trials.
degree of toxicity. Table 1 below summarizes the effects
of lead on children as blood lead levels increase. Note Blood lead concentration 25-44 g/dL (1.21-2.12
that low levels of lead have very subtle effects; only at mol/L). In this blood lead range, treatments vary from
high levels are signs and symptoms obvious. clinic to clinic. Some practitioners will treat children at
this level with one or more of the drugs listed in the table
Table 1: Lowest Blood Levels at Which Clinical above, on an outpatient basis.
Effects are Observed in Children
Blood lead Blood lead concentration 45-69 g/dL (2.17-3.33
concentration (g/dL): Effect:
mol/L). Children with levels greater than 45 g/dL
All Lead deposited into bones
should be referred for immediate appropriate chelation
Less than 10 Production of red blood cells
slows down therapy.
1520 Abnormally high amount of
EP in the blood When chelation therapy ceases, some rebound may
Below 25 Subtle behavioral and occur. The action of the chelators has caused
learning changes mobilization of lead from tissue stores, and this process
5060 Loss of nerve signals from may continue after chelation therapy ceases
brain to muscle
80100 Brain damage (lead Lead Performance Goals
encephalopathy) Survey data from the 2007 College of American
Pathologists Participant Summary Report shows
Available Treatments for Lead Poisoning imprecision values (% coefficient of variation) for
The most effective drugs to treat lead poisoning are measurement of lead at a mean concentration of 21.2
chelating agents that attach to the lead in a form which g/dL range from 7.4% for anode stripping voltammetry
can leave the body through the urine. As lead is removed to 10.2% for methods employing ICP/MS.
from the blood, lead moves out of tissues into the blood
to be in turn bound and removed. All drugs, including Acceptable Clinical Laboratory Improvement
chelators, have potential side effects and are used with Amendments of 1988 (CLIA-88) performance criteria
caution. Several different drugs of this type are used in for measurement of lead in blood require that
the treatment of lead poisoning and are described below. laboratories be accurate to within 4 g/dL or 10% of
the peer group mean. A problem associated with
Table 2: Chelating Agents Used in Treating Children obtaining accuracy and precision in all methods of
With Lead Poisoning analysis is a lack of adherence to method protocols [12].
Product name: Calcium disodium EDTA, or versenate
In terms of accuracy, an overestimation of lead is
Generic name: Calcium disodium edetate, or
probably attributable to lead contamination of glassware,
edetate calcium
Chemical name: Calcium disodium reagents, or equipment and inadequate blank correction.
ethylenediaminetetra-acetic acid, [[N,N'-1,2- An underestimation is probably attributable to
ethanediyl-bis[N-(carboxymethyl)- incomplete analytical recovery in extraction or
glycinato](4)-N,N',O,O',ON,ON']-, disodium, concentration steps and nonlinearity of calibration
hydrate, (OC-6-21)-Calciate(2-) curves.
Abbr: CaNa2 EDTA
References
Product name: BAL in oil 1 U.S. Department of Health & Human Services.
Generic name: Dimercaprol Criteria for a recommended standard:
Chemical name: 2,3-dimercapto-1-propanal
occupational exposure to inorganic lead,
Abbr.: BAL
revised criteria. NIOSH, DHEW Publication
Product name: Cuprimine
Generic name: D-Penicillamine No. 78-158. Washington, DC: 1978; U.S.
Chemical name: 3-Mercapto-D-valine Government Printing Office.
Abbr.: D-penicillamine 2 Lead in blood and urine, Method no. P&CAM
Product name: Chemet 102. In: Taylor DG, manual coordinator.
Generic name: Succimer NIOSH Manual of Analytical Methods. 2nd ed.
Chemical name: Meso-2,3- Washington, DC: U.S. Government Printing
Dimercaptosuccinic acid Office; 1977:1.102-1 to 9.6.
Abbr: DMSA 3 American Society for Testing and Materials.
Standard test method for lead in the atmosphere
807
Lead
by colorimetric dithizone procedure: gaseous Washington, DC: U.S. Government Printing

fuels: coke: atmospheric analyses. Method D- Office; 1977:1.200-1 to 8.
3112-77. Philadelphia: American Society for 17 Filby RH, Shah KR. Activation analysis and
Testing and Materials; 1982;26:628-37. applications to environmental research. Toxicol
4 Rice EW, Fletcher DC, Stumpf A. Lead in Environ Chem Rev. 1984;2:1-44
blood and urine. Stand Methods Clin Chem. 18 Barnes IL, Murphy TJ, Gramlich JW, Shields
1965;5:121-9. WR. Lead separation by anodic deposition and
5 General procedure for metals, Method no. isotope ratio mass spectrometry of microgram
P&CAM 173. In: Taylor DG, manual and smaller samples. Anal Chem.
coordinator. NIOSH Manual of Analytical 1973;45:1881-4
Methods. 2nd ed. Washington, DC: U.S. 19 Licensure and Proficiency Testing Division,
Government Printing Office; 1977:1.173-1 to Bureau of Laboratories, Centers for Disease
10. Control and Prevention. Blood lead proficiency
6 Lead in air or blood, Method no. P&CAM 214. testing. Atlanta: Public Health Service, U.S.
In: Taylor DG, manual coordinator. NIOSH Department of Health and Human Services.
Manual of Analytical Methods. 2nd ed. 20 Centers for Disease Control and Prevention.
Washington, DC: US. Government Printing Blood lead analysis 1981. Document no.
Office; 1977:1.214-1 to 6. 34Q205158208. Atlanta: U.S. Department of
7 Lead in blood and urine, Method no. P&CAM Health and Human Services; May 1982.
262. In: Taylor DG, manual coordinator. 21 Zinterhofer LJM, Jatlow PI, Fappiano A.
NIOSH Manual of Analytical Methods. 2nd ed. Atomic absorption determination of lead in
Washington, DC: U.S. Government Printing blood and urine in the presence of EDTA. J Lab
Office; 1977:1.262-1 to 4. Clin Med. 1971;78:664-74
8 Kopito L, Schwachman H. Measurement of 22 Crick J, Flegal AR. Contaminant lead in blood-
lead in blood, urine and scalp hair by atomic collection tubes for trace-element studies. Clin
absorption spectrometry. Stand Methods Clin Chem. 1992;38:600-1
Chem. 1972;7:151-62. 23 Segal RJ. Nonspecificity of urinary lead
9 Marcus M, Hollander M, Lucas RE, Pfeiffer measurements by atomic absorption
NC. Micro-scale blood lead determinations in spectroscopy. Clin Chem. 1969;15:1124
screening: evaluation of factors affecting 24 U.S. Department of Labor, Occupational Safety
results. Clin Chem. 1975;21:533-556 and Health Administration. Regulations
10 Yeager DW, Cholak J, Henderson EW. (Standards - 29 CFR) Lead. 1926.62. 2006
Determination of lead in biological and related (online). Available at:
material by atomic absorption http://www.osha.gov/pls/oshaweb/owadisp.show_docum
spectrophotometry. Environ Sci Technol. ent?p_table=STANDARDS&p_id=10641 Accessed
1971;5:1020-1022. 2008 Mar 6.
11 Subramanian KS, Meranger JC. A rapid 25 U.S. Department of Labor, Occupational Safety
electrothermal atomic absorption and Health Administration. Regulations
spectrophotometric method for cadmium and (Standards - 29 CFR) Lead. 1910.1025. 2006
lead in human whole blood. Clin Chem. (online). Available at
1981;27:1866-71 http://www.osha.gov/pls/oshaweb/owadisp.sho
12 Boone J, Hearn J, Lewis S. Comparison of w_document?p_table=STANDARDS&p_id=10
interlaboratory results for blood lead with 030 Accessed 2008 Mar 6.
results from a definitive method. Clin Chem. 26 National Research Council, National Academy
1979;25:389-393 of Sciences. Airborne lead in perspective:
13 Christian GD. The biochemistry and analysis of report of the Committee on Biological Effects
lead. Adv Clin Chem. 1976;18:289-326 of Atmospheric Pollutants of the National
14 Searle B, Chan W, Davidow B. Determination Research Council, National Academy of
of lead in blood and urine by anodic stripping Sciences. Washington DC: National Academy
voltammetry. Clin Chem. 1973;19:76-80 of Sciences, Printing and Publishing Offices;
15 Lead in blood, Method no. P&CAM 195. In: 1972.
Taylor DG, manual coordinator. NIOSH 27 Damstra T. Toxicological properties of lead.
Manual of Analytical Methods. 2nd ed. Environ Health Perspect. 1977;19:297-307.
Washington, DC: U.S. Government Printing 28 Mahaffey KR. Quantities of lead producing
Office; 1977:1.195-1 to 7. health effects in humans: sources and
16 Lead in urine, method no. PP&CAM 200. In: bioavailability. Environ Health Perspect.
Taylor DG, manual coordinator. NIOSH 1977;19:285-295
Manual of Analytical Methods. 2nd ed. 29 Lane RE. Diagnosis of inorganic lead
poisoning: a statement. BMJ. 1968;4:501.
808
Lead
30 Dagg JH, Goldberg A, Smith JA, Lochhead A. 32 Sachs HK, Blanksma LA, Murray EF,
The relationship of lead poisoning to acute OConnell MJ. Ambulatory treatment of lead
intermittent porphyria. Q J Med. 1965;34:163- poisoning: report of 1155 cases. Pediatrics.
75. 1970;46: 389-96.
31 Centers for Disease Control and Prevention. 33 Granick JL, Sassa S, Kappas A. Some
Increased lead absorption and lead poisoning in biochemical and clinical aspects of lead
young children. J Pediatr. 1975;87:824. intoxication. Adv Clin Chem. 1978;20:287-339.
Tables
Table 3: Methods of Lead Analysis
Method 1: Spectrophotometric dithizone
Type of analysis: Colorimetric, quantitative
Principle: After ashing, Pb2+ + dithizone form a complex absorbing at 510 nm
Usage: Blood, urine, tissues, environmental
Comments: Widely used, very hard to do with good precision
Method 2a: Atomic absorption spectrometry (AAS)Classical AAS
Type of analysis: Spectrophotometric, quantitative
Principle: Ashing of lead sample, ionized lead is reduced in flame; absorbs light at 283.3 nm
Usage: Blood, urine, tissues, environmental
Comments: Straightforward; applicable to variety of macrosamples and sizes
Method 2b: Atomic absorption spectrometry (AAS)Delves microscale technique
Principle: Same as in (a), except sample is oxidized in microcup by H2O2
Usage: Blood, urine
Comments: As above and additionally applicable to microsamples
Method 2c: Atomic absorption spectrometry (AAS)Graphite furnace or carbon rod (FAAS)
Principle: Same as in (a), except sample is dried, ashed, and vaporized on a graphite rod
Usage: Blood, urine
Comments: As for 2b, widely used
Method 3: Anodic stripping voltammetry
Type of analysis: Electrochemical, quantitative
Principle: Ionic lead plates as Hg amalgam; at 430 mV Pb reoxidized with increased amperage proportional to
lead concentration
Usage: Blood, urine
Comments: Simple to do, most reliable over wide concentration ranges, widely used
Method 4: Isotope-dilution inductively coupled plasma mass spectrometry
Type of analysis: Mass spectrometry, quantitative
Principle: Extraction by anion exchange chromatography with isotopic analysis
Usage: Blood, urine
Comments: NIST definitive method
809
Lead
Table 4: Comparison of Assay Conditions for Lead Analysis

Assay 1: Dithizone
pH: >11
Final concentration of reagent components: Dithizone: 6.25 mol/L
Sample volume (blood): 10 mL
Linearity: 0 to 5 g/mL
Precision: 6%
Time of reaction: Hours
Major interferences: Tin, bismuth, thallium
Assay 2: Atomic absorption (Moero)
Temperature: 2000+
pH:
Final concentration of reagent components: Ammonium pyrrolidine dithiocarbamate: 0.12 mol/L
Sample volume (blood): 10 mL
Linearity: 01000 ng/mL
Precision: 5%
Time of reaction: Minutes
Major interferences:
Assay 3: Anodic stripping
pH: <1.0
Final concentration of reagent components: Perchloric acid: 0.35%
Sample volume (blood): 100 L
Linearity: 6 to 1000 ng
Precision: 5% at 600 ng/mL
Time of reaction: Minutes
Major interferences: Thallium
Figure
Lead: Figure 1
Spectra of dithizonelead complex; method described in text. Dashes, dithizone; dots with dashes, 5 g of extracted lead
added; solid line, 25 g of extracted lead.
810
Lead
Procedure: Preparation of Reagents, Standards, 5. Cool the sample, add an additional portion of
Glassware the acid, and repeat the heating procedure.
6. Finally, the residue will remain pale yellow or
Cleaning of Glassware
light brown after it has been heated for 5 to 10
Reagents minutes at the 400C temperature.
1. Hot chromic acid. Dissolve 25 g (0.25 mol) of
reagent-grade chromic trioxide (Caution: suspected Standards
carcinogen!) in 2.5 L of reagent-grade sulfuric acid. Stir
Stock aqueous lead solution
with a Teflon-coated stirring bar. Heat to 50C before
1. To prepare a 1 mg/mL lead standard stock
use. Stable for 6 months at room temperature. When the
solution, dry lead nitrate (reagent grade) at 100C for 2
solution changes color from red to green, it should be
hours, cool, dissolve 1.5980 g in nitrite acid (Ultrex
discarded. Dispose according to appropriate
grade) diluted 1:10 with water, and then dilute to 1 L
environmental guidelines.
with the 1:10 acid.
2. Nitric acid soaking solution. Dilute
2. From this 1 mg/mL solution, make working
concentrated nitric acid (reagent grade) with an equal
volume of glass- or double-distilled water. Stable for 1 standards of 1, 2, 5, and 10 g of lead per
year at room temperature. milliliter of solution by appropriate dilutions
with 1:10 nitric acid.
Method 3. The 1 mg/mL stock solution is stable for at least
All glassware must be rendered lead free before 6 months, whereas the working standard
use. solutions should be made fresh on a weekly
1. Wash in detergent in tap water, and follow with basis. Commercial standards sold are usually
tap- and distilled-water rinses. 1% in HNO3 and are stable only for 6 to 8
2. Soak in hot chromic acid, and follow with tap- months at room temperature. Our experience
and distilled-water rinses.
shows that standards made up in 10% HNO3 are
3. Soak in the nitric acid soaking solution for 30
stable for several years stored at room
minutes, and follow with tap-, distilled-, then double-
distilled-water rinses. temperature.
Notes. The entire procedure must be followed
Blood Standard
when glassware is being initially prepared for use in the 1. Obtain heparinized expired blood from the local
lead analysis. For each successive use of equipment that blood bank to be used as a working laboratory
has previously been subjected to the entire cleaning blood standard stock. The sample is preserved
procedure, it is necessary to use only the nitric acid by refrigeration.
cleaning procedure of step 3. However, glassware must
2. Each day that an analysis is to be performed,
be committed to lead analysis to avoid contamination
spike 6 10-mL aliquots to 0, 1, 2, 5, 10, and 20
from other sources. Also, polyethylene laboratory ware,
g lead per milliliter, respectively, using the 1
used for storage of standards and so on, can be
mg/mL aqueous stock solution.
effectively cleaned when it is subjected to the above
procedure. 3. Run these samples along with the aqueous
working standards and the blood specimens
Ashing of the Sample being analyzed.
It should be noted that accurate records must be
Reagent maintained as to the analytical results obtained from
1. Redistilled nitric acid (Ultrex grade). analysis of this blood standard. This continuous
Method laboratory record will allow for the long-term
1. Add 10 mL of a sample of heparinized blood to monitoring of method performance. Further, as one
a 125 mL borosilicate glass beaker, followed by supply of expired blood is being depleted, a new supply
7 mL of redistilled nitric acid. For urine, use a should be obtained and run so that an overlap of the
25 mL sample. blood standards is maintained.
2. Evaporate the sample at 130C, just to the point Urine Standard
of dryness, and cover the beaker with a lead- 1. Prepare a standard by obtaining urine from a
free watch glass. nonexposed person.
3. After cooling, add successive portions of the 2. Preserve the sample by the addition of Ultrex
nitric acid, that is, 2.0, 1.5, 1.0, and 0.5 mL, by nitric acid to a final concentration of 10% (v/v)
sliding the watch glass only sufficiently to and store at 4 C.
facilitate each new addition of acid.
3. On the day the analysis is to be performed,
4. After each addition and as soon as the residue spike 6 25-mL aliquots to 0, 1, 2, 5, 10, and 20
becomes light colored, heat the sample to g of lead per milliliter, respectively, using the
400C just long enough to blacken the sample. 1 mg/mL aqueous stock solution.
811
Lead
4. Run these samples along with the aqueous removed; add redistilled 6 M HCI until the indicator
working standards and the urine specimens turns pink; and then adjust the volume to 100 mL with
being analyzed. The same procedure described double distilled water. Stable for 6 months at room
for the recording of the laboratory blood temperature.
standard and the initiation of the use of a new 10. 10% potassium cyanide solution (1.53
urine stock should be applied to the laboratory mol/L). Dissolve 10 g in 100 mL of distilled water.
Extract as for sodium citrate. Bring to 500 mL volume.
urine standard.
Stable for 1 year at room temperature.
11. Potassium cyanideammonia solution (0.306
Procedure: Spectrophotometric Method [8] mol/L KCN, 2.25 mol/L NH4OH). Mix 200 mL of
Principle 10% KCN solution and 150 mL of concentrated
+ ammonium hydroxide. Dilute to 1 L. Stable for 1 year at
Ashed sample is solubilized in a KCN-NH4 solution,
and the lead is extracted into a dithizoneCHCl3 room temperature.
12. Dithizone reagent (6.25 mol/L). Dissolve 16
(chloroform) solution. The leaddithizonate complex is
mg of dithizone in 1 L of chloroform. Prepare fresh
back extracted into nitric acid, which is then made
daily.
alkaline with KCN-NH4+ and reextracted into dithizone
CHCl3 reagent. The complex is quantitated Assay
spectrophotometrically at 510 nm. Equipment: Spectrophotometer with a band pass < 10
nm, capable of reading at 510 nm; pH meter.
Reagents
1. The ash from each standard and sample is
1. Nitric acid. Double distilled, Ultrex grade.
Stable for 1 year at room temperature. warmed with 2.0 mL of redistilled,
2. Sodium diethyldithiocarbamate (0.177 concentrated nitric acid to dissolve the residue,
mol/L) 4%. Dissolve 4 g of sodium and then 25 mL of distilled water is added. The
diethyldithiocarbamate in 100 mL of distilled water. solution is heated at 50C to 60C until
Stable for 1 month at ambient temperature. clarified.
3. Ammonium hydroxide, concentrated 15.1 2. Add 1.0 mL of hydroxylamine solution,
mol/L. High-purity. Specific gravity, 0.92. followed by 4 mL of sodium citrate solution.
4. Hydrochloric acid (6 mol/L). Dilute redistilled Using a pH meter, adjust the titrated sample
concentrated HCl, Ultrex grade, with an equal volume of solution to pH 9 to 10 with ammonium
water. Stable for several years at room temperature. hydroxide.
5. Dithizone extraction solution (390 mol/L).
Dissolve 100 mg of dithizone in 1 L of chloroform (ACS 3. After a quantitative transfer to a 125 mL
reagent grade or better). Stable for 1 month at room separatory funnel containing 5 mL of potassium
temperature. cyanide solution, add 5 mL of the chloroform
6. Sodium citrate (0.548 mol/L). Dissolve 125 g solution of the dithizone reagent. The sample is
of sodium citrate eicosadecahydrate in 400 mL of extracted by being shaken for 2 min; then the
distilled water. Adjust to pH 9 to 10 with ammonium layers are allowed to separate.
hydroxide and bring to 500 mL volume. Extract for 50 4. The chloroform layer is drawn off into a second
min with a 10 mL portion of the dithizone extraction separatory funnel containing 30 mL of dilute (1
solution. Trace amounts of lead in the citrate solution part acid to 99 parts double distilled water)
will turn the green dithizoneCHCl3 solution red.
nitric acid. Successive 5 mL portions of the
Discard chloroform phase, and repeat extraction until the chloroform solution are added to the first
chloroform extract remains green. Extract the citrate funnel, and the extraction process is repeated
solution repeatedly with chloroform to remove dissolved until the extraction dithizone reagent remains
dithizone until the extract is colorless. Remove the last
green.
traces of chloroform by pipetting. Stable for 6 months at
4C. Check for pH change and microbial growth 5. Combine the chloroform extracts in the second
monthly. funnel, extract with the nitric acid, and allow
7. m-Cresol purple indicator. Dissolve 1 g in the layers to separate.
100 mL of distilled water. Stable for 1 year at room 6. Discard the chloroform (bottom) layer, and then
temperature. add 6.0 mL of the ammoniacyanide mixture to
8. Hydroxylamine solution (9.3 mol/L). Dissolve the aqueous layer (nitric acid), and add exactly
20 g of hydroxylamine in 65 mL of distilled water. 15.0 mL of the standard dithizonechloroform
Stable for 2 years at room temperature. reagent solution. Shake 2 minutes, and allow
9. Working hydroxylamine solution (6.06 the layers to separate.
mol/L). To 65 mL of hydroxylamine solution add a few
drops of m-cresol purple indicator. Then add ammonium 7. Carefully decant the chloroform solution into a
hydroxide until the indicator turns yellow; add 5 to 10 spectrophotometer tube and read at 510 nm.
mL of 4% solution of sodium diethyldithiocarbamate to
combine metal impurities; mix; then extract repeatedly Calculation
with chloroform until the excess carbamate has been
812
Lead
Plot A for each blood or urine standard (vertical axis) with water. Separate organic phase with separatory
510
funnel. Prepare fresh before use.
versus micrograms of lead per standard (horizontal axis).
Determine the slope of the curves, dividing the A by Assay
510
gPb: (A /gPb). Equipment: Any recent-model flame atomic absorption
510 spectrophotometer, preferably one equipped with an
(A Sample) (A Blank) automatic sampler.
510 510
Net amountof Pb in sample 1. Sample preparation. Ash a 10 mL sample of
A / g Pb
510 heparinized blood or a 50 mL sample of urine.
(See Ashing of the sample).
Net amount of Pb = g of Pb/mL 2. When all of the organic matter has been
Sample volume destroyed, add 1.0 mL of concentrated
redistilled nitric acid and 5.0 mL of distilled
For blood: water to the beaker and slowly apply heat at
gPb/mL 1000 = gPb/L of blood 50C to 60C. When the residue has dissolved,
transfer the solution to a clean 50-mL
For urine: volumetric flask.
0.024/ (specific gravity of urine 1.000) gPb/mL 3. Add two drops of phenol red indicator and then
1000 = gPb/L urine 5.0 mL of ammonium citrate buffer, pH 8.5.
Add concentrated ammonium hydroxide drop
(Note: corrected to specific gravity of 1.024) by drop until the solution turns red, and then
add 1.0 mL of 10% potassium cyanide solution.
Procedure: Atomic Absorption Spectroscopy Add 1 mL of 2% ammonium pyrrolidine
dithiocarbamate, and mix the entire solution
Principle well.
The blood or urine sample is ashed and extracted into
4. Add 4.0 mL of water-saturated methyl isobutyl
methyl isobutyl ketone, and the organic layer is passed
ketone, and extract the sample by shaking for
through the flame of the atomic absorption instrument
30 sec. Add distilled water until the organic
where the Pb++ is reduced to Pb0. The amount of light
layer is brought up into the neck of the flask.
from a hollow cathode lamp absorbed by the lead atoms
is proportional to the concentration of lead in the sample. 5. Set the atomic absorption spectrophotometer
(check manufacturers instructions) to measure
Reagents absorption at 283.3 nm, and start the air-
1. Nitric acid. Concentrated, redistilled, Ultrex acetylene flame. Reduce the acetylene flow so
grade. Stable for 1 year at room temperature. that an oxidizing flame is obtained while the
2. Phenol red indicator. Dissolve 0.1 g of water-saturated methyl isobutyl ketone layer is
phenol red in 100 mL of distilled water. Stable for 1 year being aspirated.
at room temperature.
6. Zero the instrument against a reagent blank.
3. Ammonium citrate buffer, pH 8.5 (4.16
Use at least two standards to calibrate the
mol/L citrate, 12.12 mol/L ammonia). Add 400 g of
citric acid (2.08 mol) to 100 mL of distilled water. instrument if using the calibrate mode.
Dissolve by slowly adding 400 mL of concentrated 7. Aspirate each sample and record data.
ammonium hydroxide (6.06 mol). Allow to cool, and 8. If instrument is used in the absorbance mode,
adjust to pH 8.5. Stable for 6 months at 4C. Check for record absorbance once a steady-state signal is
pH change and microbial growth monthly. achieved.
4. Dithizone extraction solution (390 mol/L).
Dissolve 100 mg of dithizone in 1 L of chloroform. Calculations
Stable for 2 months at room temperature. If the instrument is set in concentration mode, results are
5. 10% potassium cyanide solution (1.53 obtained in concentration units. If the absorption mode is
mol/L). Dissolve 10 g in 100 mL of distilled water. used, calculate concentration by direct proportions with
Extract repeatedly with successive 10 mL aliquots of one of the standards or directly from a calibration curve.
dithizone extraction solution until the chloroform
remains green. Extract repeatedly with chloroform until
Calibration curve. Pipet 0.5 mL of each working
the extract is colorless. Remove the last traces of
aqueous standard into individual 50 mL volumetric
chloroform by evaporation. Stable for several years at
room temperature. flasks. Process each diluted solution according to the
6. Ammonium pyrrolidine dithiocarbamate procedure, from step 2 (first evaporation). Construct a
(0.12 mol/L). Dissolve 2 g of ammonium pyrrolidine calibration curve of absorbance versus micrograms of
dithiocarbamate in 100 mL of distilled water. Stable at lead.
6C for 6 months. Blood: gPb/L blood = g from calibration curve (10
7. Methyl isobutyl ketone. Shake 500 mL of mL sample) 100
methyl isobutyl ketone (ACS reagent grade or better) Urine:g of Pb/L = g from calibration curve(50 mL
813
Lead
sample) 1000 0.024(SG of urine -1) evaporated. The digestion should be continued
for an additional 30 to 60 min until the acid is
Procedure: Anodic Stripping Voltammetry [14] almost completely evaporated. Remove the cell
from the hot plate, cool it, and add 4.5 mL of
Principle 0.35% perchloric acid.
Sample is ashed by nitric, perchloric, sulfuric acid
4. For urine samples, reconstitute urine ash with
mixture. The resultant solution is quantitated by
4.5 mL of 0.35% perchloric acid, and continue
measurement of the current at 430 mV after reduction
with the assay.
and plating of lead.
5. Following the manufacturers instructions for
Reagents the instrument, analyze the sample solution
1. Acids. Double-distilled, concentrated nitric under the following conditions (voltages given
acid (Ultrex grade); double-distilled, concentrated with respect to a Ag/AgCl reference electrode):
sulfuric acid (Ultrex grade); and double-distilled 70% plating potential, 780 mV; plating time, 10 to
perchloric acid. 30 min (being sure that plating time for samples
2. Acid ashing solution. Combine 32.7 mL of and standards are exactly the same); sweep rate
double-distilled concentrated nitric acid with 65.3 mL of of +60 mV per second; current range of 0.2 to
double-distilled 70% perchloric acid and 2.0 mL of 1.0 mA, full scale. The stirring rate for the
double-distilled concentrated sulfuric acid. Stable for 6 sample solution during plating should not be
months at room temperature. lower than 170 rpm/min, since at lower rates,
3. 0.35% perchloric acid (0.35 g/L). Add 1 mL
significant variations in the amount of lead
of double-distilled, concentrated (70%) perchloric acid
plated on the electrode may occur. As the
to 199 mL of water. Stable for 1 year at room
temperature. sample is stripped, a cadmium peak may occur
first at 600 mV, followed by the lead peak at
Assay 430 mV and a copper peak at 50 mV.
Equipment: Model 3010 atomic stripping voltameter
(Environmental Science Associates, or equivalent); Calculations
Ultrasonics mixer model W-140-E sonifier (Heat One prepares a standard curve for each sample type,
Systems, Ultrasonics Inc., or equivalent). urine or blood, by plotting lead peak height in
microamperes (A) (measured when one draws a
1. Mix the sample by rinsing the ultrasonic mixer
baseline to connect the flat portions of the tracing before
tip thoroughly in distilled deionized water, and
and after the lead peak at 430 mV and then draws a
then place the tip into a blood sample within 2
perpendicular line from the apex of the peak to the
cm of the bottom of the Vacutainer tube, and
baseline; the length of this perpendicular is the peak
mix at the maximum permissible power for 20
height) versus the nanograms of lead in the standard
to 30 sec. If a sample in a Vacutainer tube sits
sample. A response factor is calculated from the standard
for more than 10 min after it has been mixed,
curve, that is, ngPb/A. The amount of lead in a sample
remix it before taking an aliquot for analysis.
is then calculated as follows:
2. Pipet 100 L of the mixed blood or standard
ngPb = Peak height for sample (in A) Response
into an analysis cell. Rinse the pipet once with
factor (in ng/A)
distilled, deionized water, and add the rinse to
the analysis cell. To prepare the pipet for reuse, For urine:
rinse the pipet four times with distilled gPb/L = Amt. (in ng) of Pb in urine (Volume [in mL]
deionized water, and discard the rinse water. of urine) 1000
Since some batches of the micropipets may
have an appreciable contamination with lead, it Corrected gPb/L urine = gPb/L 0.024 (Specific
is best to use the same tip throughout a series of gravity of sample) 1.000
samples and to rinse it between each sample.
Duplicate 100 L blood samples should be For blood:
taken for analysis. gPb/L blood = Amt. (in ng) of Pb in blood (Volume [in
3. Pipet 300 L of acid ashing solution into the mL] of sample) 1000
analysis cell containing the sample. Place the
cell in the digestion rack on a hot plate at 220C Note: Using the ESA model 3010 stripping voltameter
to 240C. The sample will initially turn yellow without a strip recorder, some workers have observed a
but will clear in about 10 min. A reflux line will negative interference caused by high levels of copper in
be established about halfway up the cell. At the pregnant women. Measuring the lead peak with a
end of 30 minutes, there will be approximately recorder prevented this interference.
100 to 200 L of acid left in the bottom of the
cell, with most of the nitric acid having
814
Lipase
Lipase
Ming Jin
Name: Lipase, triacylglycerol acylhydrolase (LPS)

Clinical significance: Refer to Chapter 34, The Pancreas: Function and Chemical
Pathology, in the 5th edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular weight: Approximately 48 kDa
Enzyme commission number: EC 3.1.1.3
Chemical class: Enzyme
Isoforms: At least two; most likely posttranslational variants

i point measurements (Table 1, Method 1b) and developed
The measurement of serum lipase activity has long been a method using pH stat apparatus [3]. A more recently
recognized as useful for diagnosing pancreatitis. developed variation of the pH technique involves
Unfortunately, early methods for lipase were plagued by measurement of the rate of change of pH as the fatty
technical difficulties, making accurate measurement of acids are liberated [4]. This method requires a very
the enzyme problematic and its clinical utility unreliable. sensitive pH measurement deviceone that can measure
However, advances in our understanding of lipase action pH changes of less than 0.02 for values in the normal
have greatly improved the reliability of lipase range. This procedure employs olive oil as the substrate
measurements and subsequently have improved the but does not use bile acids or colipase to activate lipase.
clinical utility of lipase for diagnosing pancreatitis. The reaction is very rapid, with most analyses being
completed in 3 minutes.
Lipases are enzymes that hydrolyze glycerol esters of
long-chain fatty acids. They act on the glycerol bonds Colorimetric Method
(ester linkage) to liberate two molecules of fatty acid and The method of Massion and Seligson [5] quantitates the
liberated fatty acids after 1 h of incubation by extracting
one molecule of -monoglyceride (Figure 1).
them and allowing them to change the color of an
appropriately buffered methyl red indicator (Table 1,
Essential to the understanding of lipase methodology is
Method 2).
the fact that the enzyme acts only at an ester-water
interface. Thus lipase-assay substrates must be
Immunological Techniques
established as emulsions. The reaction rate will increase
A number of immunochemical assays have been
with the dispersion of the emulsion (surface area).
reported for measurement of lipase. These include both
Failure to achieve a suitable ester-water interface will
quantitative enzyme immunoassay procedures [6] and a
permit the action of nonlipase enzymes (carboxylic ester
semiquantitative latex-agglutination procedure (Table 1,
hydrolase, aryl-ester hydrolase, and lipoprotein lipase).
Method 3a). Antibodies to lipase are bound to latex
The use of short-chain fatty acid triglyceride substrates,
particles, and these particles are mixed with the sample.
which are water soluble, will also permit this false lipase
If lipase is present, agglutinated clumps of particles
reaction to occur.
appear on the reaction slide [7].
Titrimetric Method
A noncompetitive immunoactivation procedure has been
Titration methods were the first practical assays. The
described [8]. This assay uses Fab-antibody fragments
classical technique of Cherry and Crandall [1] was based
against pancreatic lipase that are covalently coupled to
on the incubation of serum for 24 hours at 37C, with a
horseradish peroxidase. In the presence of lipase, the
50% emulsion of olive oil and 5% gum acacia in a pH
peroxidase catalyzes a color-producing reaction, the rate
7.0 phosphate buffer. The liberated fatty acid was then
of which is proportional to the lipase concentration in the
quantified by titration against 0.05 M NaOH with a
sample (Table 1, Method 3b).
phenolphthalein indicator being used (Table 1, Method
1a).
Coupled Enzyme Spectrophotometric Method
Alternative detection methods to replace titration
Several coupled enzymatic assays have been developed
methods have been developed. For example, Tietz and
for the measurement of serum lipase. One enzymatic
Fiereck [2] did considerable work using pH meter end-
assay utilizes a micellar-solubilized long-chain fatty
acid, 1,2-diglyceride (Table 1, Method 4a) [9]. Lipase
i
Lipase cleaves the substrate to form 2-monoglyceride and fatty
Previous and current authors of this method: acids. The 2-monoglyceride is subsequently hydrolyzed
First edition: Michael D.D. McNeely to glycerol by a specific 2-monoglyceride lipase
Methods edition: Michael D.D. McNeely (MGLP). In the final reaction step, the chromogen
Second edition: Michael D.D. McNeely system (4-aminophenazone/N-ethyl-N-[2-hydroxy-3-
Third edition: Steven C. Kazmierczak sulfopropyl-m-toluidine] [TOOS]) is oxidized to a violet
Fourth edition: Steven C. Kazmierczak dye, with peak absorption at 550 nm. The reaction
Fifth edition: Ming Jin mixture also contains colipase and deoxycholate.
815
Lipase
been proposed as a useful reference measurement

A different enzymatic approach has been adapted to the procedure in the evaluation of new lipase methods [13],
dry-film technology (Table 1, Method 4b) [10]. This but the technique of this method is too complex for
method employs 1-oleyl-2,3-diacetoyl glycerol adsorbed routine use.
onto TiO2 particles as the substrate. The TiO2 particles
act to provide the surface needed for lipase activity. In Methods using substrates other than long-chain fatty acid
addition, an emulsifying agent and porcine colipase are triglycerides must be viewed with great suspicion. The
present to increase lipase activity. Lipase will act on the use of purified olive oil substrate, extraction of long-
asymmetric triglyceride to release the oleyl fatty acid chain fatty acid triglycerides, and quantitation against
and the water-soluble [11,12] diacetoyl glycerol. The pure standards is recommended as a standard technique.
latter molecule diffuses into another film layer of the The modified method of Massion and Seligson [5]
reaction slide, where an esterase clears both acetate (detailed in the following paragraphs) has these
groups to release glycerol, which can then be quantitated characteristics carried out with microsamples in less than
by an enzymatic method. When one employs a lag 2 hours.
period before measuring the glycerol formed,
endogenous glycerol is consumed and will not interfere The turbidimetric methods, by detecting emulsion
in the assay. clearing, are fast and can be automated [19], but the
stability and reproducibility of the substrate is extremely
A new synthetic substrate, 1,2-O-dilauryl-rac-glycero-3- critical. An important deficiency of these methods is that
glutaric acid-[4-methylo-resorufin]-ester) has been used because of instrument limitations, the substrate
for lipase measurement [13]. Lipase hydrolyzes the concentrations are far below the Km (Michaelis-Menten
substrate in an alkaline medium to an unstable constant) of the reaction. This makes the standardization
dicarbonic acid ester that spontaneously hydrolyzes to of these procedures more difficult and the reproducibility
glutaric acid and methylresorufin. Methylresorufin is poorer than the methods utilizing triglyceride or
measured at 570 nm (Table 1, Method 4c). diglyceride substrates. Modifications such as increasing
the concentrations of colipase and bile salts and
These coupled enzymatic assays based on the continuous reaction-rate monitoring have led to
measurement of free glycerol can be adversely affected increased assay performance [19]. The other deficiency
if the patient sample contains increased concentrations of is the limited range of linearity.
free glycerol. Certain drugs and medications have been
found to contain high concentrations of this compound. The optimized titrimetric method uses pH measurement
However, if a lag period is employed before the techniques [2,4] by titrating liberated fatty acid with a
measurement of formed glycerol, endogenous glycerol is pH stat instrument during the reaction. This procedure
consumed and does not interfere with the assay. appears to be reasonably precise and rapid. Good
correlation has been shown between the pH
Turbidimetric Method measurement and turbidimetric methods. This method is
A completely different analytical approach that utilized suggested to be a reference method.
clearing of an emulsion was introduced by Vogel and
Zieve [14] and has been modified successfully by others Colorimetric methods that employ either chromogenic
[15]. In this method, the turbidity of the oil-water long-chain fatty acid substrates or colorimetric
emulsion, monitored at 340 or 400 nm, is reduced as determination of glycerol are being used with increased
lipase hydrolyzes the triglyceride molecules. The frequency [9,10,21]. They have the advantage of
technique requires a highly stable, reproducible substrate minimum sample volume requirements (<25 L), good
with an appropriate initial absorbance. The sample is precision, wide dynamic range, and ease of automation.
added to the substrate, and kinetic measurements are Coupled enzyme spectrophotometric methods with
made turbidimetrically or nephelometrically (Table 1, different substrate are widely used in clinical
Method 5). laboratories and are run on automated instruments.
Most assays in use today employ colipase to enhance It is important that all new methods for lipase
lipase measurements. Colipase increases the reaction rate measurement be compared to a technique employing an
of lipase reactions by allowing the enzyme better access olive oil substrate with a long-chain fatty acid
to the substrate [16]. Although variable amounts of measurement of known clinical utility.
colipase have been reported to be released into
peripheral blood during acute pancreatitis [17], colipase In order to standardize methods, reference materials
is added to ensure sufficient activation of the pancreatic have been prepared from human pancreatic juice and
isoenzyme [18]. The suggested optimal reaction mixture recombinant technology. These have been standardized
for lipase includes triolein or olive oil substrate (100 using a titrimetric procedure [22].
mL/L), colipase (6 mg/L), CaCl2 (8.5 mmol/L), and a
bile salt, sodium glycocholate (35 mmol/L) [19]. Specimen
Serum and plasma samples are suitable for use, since
Reference and Preferred Methods both give equivalent results [23]. Specimens are stable
There is no reference method for lipase assay. The for at least 1 week at room temperature, up to 3 weeks at
titrimetric method using pH measurement techniques has 4C, and for several years at 20C [24]. Heating of
816
Lipase
specimens has been found to destroy enzyme activity Various drugs have been implicated in causing
[25]. Repeated freezing and thawing should be avoided. pancreatitis. A strong association between therapy with
valproic acid and pancreatitis has been described [36].
Interferences Therapy with 2,3-dideoxyinosine (ddI), a promising
The different methodologies used for lipase therapeutic agent for human immunodeficiency virus
measurements have their own associated interferences. infections, has been implicated in causing pancreatitis in
Sample hemolysis does not generally interfere with any approximately 10% of patients taking the drug [37]. In
of the described methods. Rheumatoid factor has been addition, infection with human immunodeficiency virus
shown to interfere with the turbidimetric procedures by itself is associated with acute pancreatitis. It appears
because of aggregation of sample components [26]. that the virus has a direct damaging effect on the
Approximately 3% to 5% of specimens analyzed by exocrine pancreas.
emulsion-clearing methods will display an anomalous
absorbance increase due to this phenomenon. By Renal insufficiency in patients without pancreatitis may
avoidance of freezing or by using polyethylene glycol result in increases of lipase of up to six times the upper
(PEG) to precipitate protein (mostly IgM), the frequency reference interval limit. Values greater than 10 times the
of these problems can be reduced [27]. upper reference interval limit are considered to be
pathognomonic for pancreatitis [30]. Obstruction of the
Coupled enzymatic assays based on the determination of pancreatic duct may also increase serum lipase activity.
glycerol can be adversely affected if the patient sample
contains abnormally increased concentrations of glycerol Lipase Performance Goals
that may be found in certain drugs and medications [28]. Lipase is not regulated under Clinical Laboratory
Improvement Amendments (CLIA). The criteria for
In addition to these analytical interferences, a variant acceptable performance for lipase assay by College of
form of lipase consisting of lipase bound to IgG has been American Pathologists (CAP) Survey standards is within
described [28]. This macrolipase was found to result in 30% of the mean value of laboratory peer groups. Data
greatly increased lipase activity; the clinical significance on lipase assay with different methodology from the
of this variant, if any, remains to be determined. CAP Survey Participant Summary Report in C-A 2007
indicate CVs generally below 10% at the mean of lipase
Lipase Reference Interval activities around 100 U/L.
Reference intervals for lipase activity in healthy
individuals are largely method dependent. The upper The variety of methods used to measure lipase activities
reference limit for turbidimetric procedures is results in a wide range of activities obtained from
approximately 200 U/L, and the upper reference limit for identical specimens. Up to 20-fold differences may be
the colorimetric assays is approximately 60 U/L. The seen in identical survey materials analyzed with different
upper reference limit for the coupled enzyme methods instrument/reagent combinations.
varies, depending upon the choice of substrate and the
measurement conditions. Lipase activities in serum References
remain constant until about age 60, when values may 1. Cherry IS, Crandall IA Jr. The specificity of
rise by as much as 20% [29]. There are no differences pancreatic lipase: its appearance in the blood
between males and females. after pancreatic injury. Am J Physiol 1932; 100:
266-273.
Interpretation 2 Tietz NW, Fiereck EA. Measurement of lipase
Serum lipase measurement is used to diagnose acute in serum. In: Cooper GR, editor. Standard
pancreatitis. Many recent studies have shown lipase to methods of clinical chemistry. New York:
have greater diagnostic utility in the evaluation of Academic Press; 1972. vol. 7, p. 19-31.
pancreatitis when compared to less specific amylase [30- 3 Tietz NW, Repique EV. Proposed standard
33]. Reported sensitivities and specificities of lipase for method for measuring lipase activity in serum
diagnosing pancreatitis range from 60% to 100% and by a continuous sampling technique. Clin Chem
81% to 97%, respectively [33-35]. After the onset of 1973; 19: 1268-1275.
acute pancreatitis, serum lipase increases within 4 to 8 4 Cerolotti F, Bonini PA, Murone M, Barenghi L,
hours, reaches a peak at approximately 24 hours and Luzzana M, Mosca A et al. Measurement of
may take 8 to14 days to return to normal levels. lipase activity by a differential technique. Clin
Chem 1985; 31: 257-260.
Lipase is present in at least two molecular forms, 5 Massion CG, Seligson D. Serum lipase: a rapid
designated L1 and L2, that are probably posttranslational photometric method. Am J Clin Pathol 1967;
variants. L2 is absent in normal persons but present in 48: 307-313.
high activity in patients with pancreatitis. The L2 6 Rood FW, Coleman PF. Solid-phase enzyme
isoform has been found to have better diagnostic immunoassay for the quantitation of pancreatic
efficiency than total lipase; however, further studies are lipase in human serum. Clin Chem 1986; 32:
needed to evaluate the clinical utility of this more 1129-1130.
specific marker. 7 Moller-Peterson J, Klaerke M, Dati F, Toth T.
Immunochemical qualitative latex agglutination
test for pancreatic lipase in serum evaluated for
817
Lipase
use in diagnosis of acute pancreatitis. Clin plasma and serum in tests performed in the
Chem 1985; 31: 1207-1210. Ektachem 700 XR analyzer, and evaluation of
8 Van Ingen HE, Sanders GTB. Clinical "plasma separator tubes (PST)". Clin Chem
evaluation of pancreatic lipase mass 1989; 35: 151-153.
concentration assay. Clin Chem 1992; 38: 2310- 24 Tietz NW, Shuey DF. Lipase in serum-the
2313. elusive enzyme: an overview. Clin Chem 1993;
9 Fassati P, Ponti M, Paris P, Berti G, Tarenghi 38: 1167-1185.
G. Kinetic colorimetric assay of lipase in serum. 25 Hussain S, Dasgupta A. Effect of heat treatment
Clin Chem 1992; 38: 211-215. on sera for deactivation of AIDS virus on
10 Mauck JC, Weaver MS, Stanton C. commonly monitored serum analytes. Clin
Development of a Kodak Ektachem clinical Chem 1991; 37: 932.
chemistry oxide for serum lipase. Clin Chem 26 Kannisto J, Lalla M, Lukkari E.
1984; 30: 1058. Characterization and elimination of a factor in
11 MacDonald RP, LeFave RO. Serum lipase serum that interferes with turbidimetry and
determination with an olive oil substrate using a nephelometry of lipase. Clin Chem 1983; 29:
three-hour incubation period. Clin Chem 1962; 96-99.
8: 509-519. 27 Bilodeau L, Grotte DA, Preese LM. Glycerol
12 Roe JH, Byler RE. Serum lipase determination interference in serum lipase assay falsely
using a one-hour period of hydrolysis Anal indicates pancreas injury. Gastroenterology
Biochem 1963; 6: 451-460. 1992; 103: 1066-1067.
13 Panteghini M, Bais R, van Solinge WW. 28 Stein W, Bohner J, Bahlinger M. Macrolipase-a
Enzymes. In: Burtis CA, Ashwood RA, Bruns new member of the family of immunoglobulin-
DE, editors. Tietz textbook of clinical chemistry linked enzymes. J Clin Chem Clin Biochem
and molecular diagnostics. 4th ed. St. Louis 1987; 25: 837-843.
MO: Elsevier Saunders; 2006. pp 597-643. 29 Mohiuddin J, Katrak A, Junglee D, Green MF,
14 Vogel WC, Zieve L. A rapid and sensitive Dandona P. Serum pancreatic enzymes in the
turbidimetric method for serum lipase based elderly. Ann Clin Biochem 1984; 21: 101-104.
upon differences between the lipases of normal 30 Lott JA, Lu CJ. Lipase isoforms and amylase
and pancreatitis serum. Clin Chem 1963; 9: isoenzymes: assays and application in the
168-181. diagnosis of acute pancreatitis. Clin Chem
15 Shipe JR, Savory J. The simultaneous kinetic 1991; 37: 361-368.
measurement of amylase and lipase using an 31 Werner M, Steinberg WM, Pauley C. Strategic
automatic sampling fluoronephelometer. Clin use of individual and combined enzyme
Chem 1973; 19: 645. indicators for acute pancreatitis analyzed by
16 Borgstrm B, Erlanson-Albertsson C, Weilock receiver-generator characteristics. Clin Chem
T. Pancreatic colipase: chemistry and 1989; 35: 967-971.
physiology. J Lipid Res 1979; 20: 805-816. 32 Van Lente F, Kazmierczak SC.
17 Junge W, Leybold K. Detection of colipase in Immunologically-derived pancreatic amylase,
serum and urine of pancreatitis patients. Clin pancreatic lipase, and total amylase compared
Chim Acta 1982; 123: 293-302. as predictors of pancreatic inflammation. Clin
18 Arzoglou PL, Lessinger JM, Frard G. Plasma Chem 1989; 35: 1542.
lipase properties as related to pancreatic 33 Kazmierczak SC, Van Lente F, Hodges ED.
condition. Clin Chem 1986; 32: 50-52. Diagnostic and prognostic utility of
19 Tietz NW, Astles JR, Shuey DF. Lipase activity phospholipase A activity in patients with acute
measured in serum by a continuous-monitoring pancreatitis: comparison with amylase and
pH-stat technique - an update. Clin Chem 1989; lipase. Clin Chem 1991; 37: 356-360.
35: 1688-1693. 34 Van Lente F, Kazmierczak SC.
20 Feld RD, Witte DL, Barrett DA. Kinetic Immunologically-derived pancreatic amylase,
determination of serum lipase activity with the pancreatic lipase, and total amylase compared
Abbott ABA-100. Clin Chem 1976; 22: 607- as predictors of pancreatic inflammation. Clin
610. Chem 1989; 35: 1542.
21 Robbrecht JH, DeBuyzere ML, Delanghe JR. 35 Panteghini M, Pagani F. Diagnostic value in
Wake UV colorimetric pancreatic lipase assay measuring pancreatic lipase and the P3 isoform
with 1,2-dilinoleoylglycerol as substrate of the pancreatic amylase isoenzyme in serum
evaluated. Clin Chem 1989; 35: 1540-1541. of hospitalized hyperamylasemic patients. Clin
22 Lessinger J-M, Parashou S, Arzoglou P, Ramos Chem 1989; 35: 417-421.
P, Chapus C, Dufaux J et al. Determination of 36 Lott JA, Bond LA, Bobo RC, McClung HJ,
lipase catalytic activity in two reference Murray RD. Valproic acid-associated
materials: BCR 693 and BCR 694 by titrimetry pancreatitis: report of three cases and a brief
at constant pH. Clin Chem Lab Med review. Clin Chem 1990; 36: 395-397.
2004;42:62-66 37 Lerch MM, Alder G. Acute pancreatitis. Curr
23 Doumas BT, Hause LL, Simuncak DM, Opin Gastroenterol 1992; 8:817-823.
Breitenfeld D. Differences between values for
818
Lipase
Table
Table 1: Methods of Lipase Analysis
Method 1: Titrimetric Method
Reaction A: Triglyceride lipasemonoglyceride + 2 fatty acids
a. Reaction A with fatty acid + NaOH titration to neutrality using phenolphthalein indicator
b. Reaction A with titration of released acids using a pH meter to assess end point
Comments:
a. Used; Cherry-Crandall method
b. Rarely used; requires a pH meter accurate to 0.01 units
Method 2: Colorimetric Method
Principle of analysis: Reaction A with released fatty acids changing the color of a pH indicator, methyl red
Comments: Rarely used
Method 3: Immunological Techniques
a. Latex agglutination
b. Immunoactivation
a. Antibody-coated latex beads aggregate if lipase is present in sufficient quantity
b. Competitive binding assay using radiolabeled lipase; anti-lipase Fab' fragments linked to horseradish
peroxidase (HRP); HRP activity increased by presence of lipase
Comments:
a. Rarely used; semiquantitative
b. Rarely used
Method 4: Coupled Enzyme Spectrophotometric Method
a. 1,2-Diglyceride + H2O lipase 2-monoglyceride + fatty acid
2-Monoglyceride + H2O 2-monoglyceride lipase glycerol + fatty acid
Glycerol + ATP glycerol kinase glycerol 3-phosphate + ADP
Glycerol 3-phosphate + O2 glycerol phosphate oxidase dihydroxyacetone phosphate + H2O2

2H2O2 + 4-aminophenazone + TOOS peroxidase quinone monoimine dye + 4 H2O
b. 1-Oleoyl-2,3-diacetoyl glycerol on TiO2 particles lipaseoleic acid + 2,3-diacetoyl glycerol
2,3-Diacetoyl glycerol diacetinase glycerol + acetate
Glycerol is then quantitated by enzymatic reaction
c. 1,2-O-dilauryl-rac-glycero-3-glutaric acid-(6-methylresorufin)-ester
lipase 1,2-o-dilauryl-rac-glycerol + glutaric acid-(6-methylresorufin)-ester
OH glutaric acid + methylresorufin (absorbs at 570 nm)
Comments:
Commonly used
Method 5: Turbidimetric Method
Principle of analysis: Turbidimetric or nephelometric monitoring of decrease in size of emulsion of substrate
after action of lipase
Comments: Widely used; very dependent on availability of stable, reproducible substrate
819
Lipase
Figure
Figure 1: Hydrolysis of triglyceride by lipase
Procedure: Turbidimetric Method for Serum Lipase cuvette, the oil must be passed through more alumina.
(Modified) [30] The oil is stable for at least 1 week at 4C to 8C.
Principle 3. Olive oil emulsion

The clearing of an emulsion of olive oil is measured Dissolve 1.0 g of purified olive oil in 100 mL of absolute
turbidimetrically at 340 nm. ethanol. Slowly add 4 mL of this solution, while stirring,
to 100 mL of Tris-deoxycholate buffer. The absorbance
Reagents of this reagent at 340 nm should be adjusted by the
1. Tris-deoxycholate buffer addition of more Tris-deoxycholate buffer until it is
Dissolve 3.0 g (0.025 mol) of within the measurable range of the photometer.
tris(hydroxymethyl)aminomethane and 6.0 g (0.014 mol)
of sodium deoxycholate in distilled water, and dilute to a 4. Standard
final volume in 1 L. Adjust to pH 8.8 with concentrated A lyophilized serum with known lipase activity is used.
HCl. Stable for 6 months at 4C to 8C.
Measurement Procedure
2. Olive oil (reagent grade, Fisher Scientific Co.) Equipment: A 10 nm band-pass spectrophotometer
Separate the olive oil from the fatty acids by passing it with temperature-controlled cuvette.
through alumina (80 to 200 mesh, catalog number A- 1. To 1 mL of emulsion, add 50 L of sample,
540, Fisher Scientific Co.) contained in a column or standard, or control. Mix immediately.
separatory funnel plugged with glass wool. One volume 2. Read the absorbance of each tube at 340 nm at
of oil is purified with an equal volume of alumina. 1 minute (A1) and again at 5 minutes (A5).
Check the fatty acid content of the oil by pipetting a
0.050 mL aliquot into 3 mL of methyl red reagent, Calculation
shaking, and centrifuging. If the absorbance (A502 nm) LPS unknown = (A1 A5) unknown x LPS standard
of the methyl red layer is more than 0.020 in a 10 mm (A1 A5) standard
820
Lipoprotein (a)
Lipoprotein (a)
Gregory A. Hobbs i
Name: Lipoprotein (a); Lp(a)

Clinical significance: Associated with coronary heart disease, Refer to Chapter 37, Coronary artery
disease, in the 5th edition if Clinical Chemistry: Theory, Analysis, Correlation.
Chemical class: Lipoprotein
Principles of Analysis and Current Usage was delayed for 36 hours [21]. It is probably wise to
Lipoprotein (a) (Lp[a]) was first discovered in 1963 [1]. assay for Lp(a) shortly after venipuncture or within 36
It is a distinct type of lipoprotein whose physiological hours after storage at 4C.
function is not well understood. Its protein component
consists of one molecule of apoB-100 attached to one Interferences
molecule of apo(a). The latter protein is found only in It is claimed [18] that hemoglobin (<500 mg/dL),
Lp(a). It is apo(a) that distinguishes the particleLp(a) conjugated bilirubin (<30 mg/dL), unconjugated
otherwise closely resembles low density lipoprotein bilirubin (<30 mg/dL), plasminogen (<200 mg/dL),
(LDL). Apo(a) in turn closely resembles plasminogen, apoB (<200 mg/dL), and triglycerides (<1500 mg/dL) do
sharing extensive amino-acid sequence homology [2]. not interfere with assays for Lp(a). No apo(a) size
Differences in the size of apo(a) within and between dependency was observed. Wako [19] found no
individuals, as well as variations in its carbohydrate significant interference from ascorbic acid (<50 mg/dL),
content and oxidation state [3], lead to significant triglyceride (Intralipid < 2%), and rheumatoid factor
variability in the size of Lp(a). (<510 U/mL). However, they did find some interference
from hemoglobin and bilirubin.
In density gradients, Lp(a) bands with LDL, being of
similar density, and may be detected by lipoprotein Reference Interval
electrophoresis. The most widely used assay for Lp(a) is Owing to the lack of standardization, each laboratory
an ELISA assay. must determine its own reference interval. This should
be done with a suitably large population and should be
Though numerous studies point to Lp(a) as an done in consultation with cardiologists who will be using
independent predictor of coronary heart disease [4-8], the results. A widely used reference interval is < 30
current usage is limited because of the lack of mg/dL, but the above-noted racial differences may affect
standardization of an Lp(a) assay. Standardization has clinical interpretation. The reference interval is method
been challenging, owing chiefly to the heterogeneity in specific, and if changing methods, a new interval must
particle size and composition. be established.
Reference and Preferred Methods The principle confounding variable among reference
There is no reference method listed by the International materials is the size of the apo(a) moiety [10]. Methods
Federation for Clinical Chemistry. The most specific insensitive to this variable are more easily calibrated. An
method is to assay amino-acid concentration of purified international reference standard has been developed [22].
Lp(a) [9,10]. Slight racial and sex differences have been Although a value has been assigned, there is
reported [11]. Immunometric analysis includes ELISA considerable variation in this value among assays. A
[12-15], DELPHIA, immunonephelometry, and value transfer procedure has been described [1]
turbidimetry [16]. Commercially available assays
include an immunoturbidimetric test [17-19], which can Interpretation
be run on automated chemistry analyzers. Some authors [23] recommend that Lp(a) only be
measured in certain high-risk groups and then only by
specialized lipid laboratories. An elevated Lp(a) is
Specimen associated with increased risk for coronary heart disease
Serum or plasma is generally acceptable as a specimen. [4-8] and angina pectoris [24].
Some workers [9] have claimed the preparation of a
primary standard stable at 70C for more than 1 year. It Lp(a) Performance Goals
has been shown that controls are stable at 70C for 5 Recovery and correlation experiments are complicated
years, but not plasma samples [20]. Others have found by the lack of standardization. In general, one should be
that Lp(a) is relatively stable at room temperature up to able to meet or exceed the claims of the assay
24 hours after venipuncture but less so when processing manufacturer.
i
Lipoprotein (a)
New method:
821
Lipoprotein (a)
Assays vary with respect to the upper limit of linearity, methods for the determination of plasma
from 80 to 150 mg/dL [17-19]. Evaluation of linearity lipoprotein(a). Clinical Chemistry
has been addressed elsewhere [22,25]; data should at a 2000;46:1956-1967.
minimum be graphed and appear to be linear on visual 11 Marcovina SM, Albers JJ, Wijsman E, Zhang
inspection. All calibrators should fall within the linear ZH, Chapman NH, Kennedy H. Differences in
portion of the curve. Lp(a) concentration and apo(a) polymorphs
between black and white Americans. J Lipid
A coefficient of variation (CV) of less than 5% should Res 1996;37:2569- 2585.
be achievable, but for some kits, imprecision will be 12 Labeur C, Michlels G, Bury J, Usher DC,
greater at low concentrations (usually using the low Rosseneu M. Lipoprotein(a) quantified by an
control). enzyme-linked immunosorbent assay with
monoclonal antibodies. Clin Chem
The limit of detection should be less than 5 mg/dL, 1989;35:1380-1384.
though some manufacturers claim 1 mg/dL [19]. 13 Abe A, Maeda S, Makino K, Seishima M,
Shiimokawa K, Noma A, Kawade M. Enzyme-
References linked immunosorbent assay of lipoprotein(a) in
1 Berg K. A new serum type system in man: the serum and cord blood. Clin Chim Acta
Lp system. Acta Pathol Microbiol Scand 1988;30:31-40.
1963;59:362-382 14 Yeo KH, Walmsley TA, Owen MC, George
2 McLean JW, Tomlinson JE, Kuang WJ, Eaton PM. A competitive ELISA for lipoprotein(a).
DL, Chen EY, Fless GM et al. cDNA sequence Clin Chim Acta 1992;205:213-222.
of human apolipoprotein(a) is homologous to 15 Morikawa W, Iki R, Terano T, Funatsu A,
plasminogen. Nature 1987;330:132-137. Sugiuchi H, Uji Y, Okabe H. Measurement of
3 Naruszewicze M, Giroux L-M, Davignon J. Lp(a) with a two-step monoclonal competitive
Oxidative modification of Lp(a) causes changes sandwich ELISA method. Clin Biochem
in the structure and biological properties of 1995;28:269-275.
apo(a). Chem Phys Lipids 1994;67/68:167-174. 16 Tiran A, Tiran B, Hojas S, Kostner GM,
4 Dahlen GH, Guyton JR, Mohammad A, Farmer Wilders-Truschnig MM. Immunoquantification
JA, Kautz JA, Gotto AM. Association of levels of lipoprotein(a): comparison of nephelometry
of lipoprotein Lp(a), plasma lipids, and other and electroimmunodiffusion. J Clin Lab Anal
lipoproteins with coronary artery disease 1993;7:256-262.
documented by angiography. Circulation 17 Lp(a), Package Insert. Rev. 2007-09. Kamiya
1986;74:758-765. Biomedical Company. Seattle, WA.
5 Armstrong VW, Cremer P, Eberle E, Manke E, 18 Lp(a), Package Insert. Rev. 2005-12. Point
Schulze F, Wieland H et al. The association Scientific Inc. Canton, MI.
between serum Lp(a) concentrations and 19 Autokit Lp(a), Package Insert. Rev.
angiographically assessed coronary 1/9903IBD00K. WAKO Chemicals. Richmond,
atherosclerosis. Atherosclerosis 1986;62:249- VA.
257. 20 Simo JM, Camps J, Vilella E, Gomez F, Paul
6 Hearn JA, DeMaio SJ Jr, Roubin GS, A, Joven J. Instability of lipoprotein(a) in
Hammarstrom M, Sgoutas D. Predictive value plasma stored at 70C: effects of
of lipoprotein(a) and other serum lipoproteins concentration, apolipoprotein(a) genotype, and
in the angiographic diagnosis of coronary artery donor cardiovascular disease. Clin Chem
disease. Am J Cardiol 1990;66:1176-1180. 2001;47:1673-1678.
7 Rosengren A, Wilhelmsen L, Eriksson E, 21 Pai JK, Curlan GC, Cannuscio CC, Rafai N,
Risberg B, Wedel H. Lipoprotein (a) and Ridker PM. Stability of novel plasma markers
coronary heart disease: a prospective case- associated with cardiovascular disease:
control study in a general population sample of processing within 36 hours of specimen
middle-aged men. Brit Med J 1990;301:1248- collection. Clin Chem 2002;48:1781-1784.
1251. 22 Passey RB, Bee DE, Caffo A, Erikson JM.
8 Kamstrup PR, Benn M, Tybjaerg-Hansen A, Evaluation of the Linearity of Quantitative
Nordestgaard BG. Extreme lipoprotein(a) levels Analytical Methods: Proposed Guideline.
and risk of myocardial infarction in the general NCCLS Document EP6-P. Villanova, PA:
population. Circulation 2008;117:176-184. National Committee for Clinical Laboratory
9 Kostner GM, Ibovnik A, Holzer H, Grillhofer Standards; 1986.
H. Preparation of a stable, fresh frozen primary 23 Marcovina SM, Levine DM, Lippe G.
lipoprotein(a) (Lp[a]) standard. J Lipid Res Lipoprotein(a). In: Rifai N, Warnick GR, eds.
1999;40:2255-2263. Structure, Measurement, and Clinical
10 Marcovina SM, Albers JJ, Scanu AM, Kennedy Significance in Laboratory Measurement of
H, Giacully F, Berg K et al. Use of a reference Lipids, Lipoproteins, and Apolipoproteins.
material proposed by the International Washington DC: AACC Press; 1994.
Federation of Clinical Chemistry and 24 Rifai N, Ma J, Sacks FM, Ridker PM,
Laboratory Medicine to evaluate analytical Hernandez WJL, Stampfer MJ, Marcovina SM.
822
Lipoprotein (a)
Apolipoprotein(a) size and liporotein(a) 25 Kroll MH, Emancipator K. A theoretical

concentration and future risk of angina pectoris evaluation of linearity. Clin Chem 1993.
with evidence of severe coronary 39:405-413.
atherosclerosis in men: the Physicians Health
Study. Clin Chem 2004;50:1364-1371.
823
Lithium
Lithium
Danni L. Meany and William Clarkei
Name: Lithium
Clinical significance: Treatment of acute mania and long-term prophylaxis of recurrent mania
and depressive episodes Refer to Chapter 47, Nervous System, in the 5th edition
Molecular formula: Li
Merck Index: 5343
Structure: Li
Reference method: FAAS
i
Principles of Analysis and Current Usage at 670.8 nm. When the emitted light from the lamp enters
Lithium has been used for treatment of acute mania and the flame, some of it is absorbed by the ground-state
long-term prophylaxis of recurrent mania and depressive lithium atoms in the flame, which results in a decrease of
episodes [1, 2]. Because it has a clearly defined therapeutic light intensity. FAAS is approximately 100 times more
range, monitoring of serum lithium concentrations has sensitive than FAES, because in FAAS most of the lithium
been well documented to improve its usage for therapeutic atoms are in the ground state, whereas in FAES only a
management of depression and for mood stabilization. small fraction of lithium atoms contribute to emission [3].
Current methodologies for lithium analysis include flame
atomic absorption spectroscopy (FAAS), flame atomic Ion-Selective Electrodes (ISEs)
emission spectroscopy (FAES), ion-selective electrodes ISE technology for lithium was established in the late
(ISEs), and colorimetric assays. Although FAAS and 1980s [4,5]. Before 1987, lithium was measured by FAES
FAES are still used in a few clinical laboratories (less than in about 90% of clinical laboratories and by FAAS in the
3% of the participants of the College of American rest. Because lithium analysis using ISEs is simple, rapid,
Pathologists [CAP] survey in 2007), they now have been and only requires a small serum sample without requiring
replaced largely by ISEs and colorimetric methods, so they prior manipulations, it gained in popularity very quickly.
are only briefly described here. Readers who desire more This was evidenced by the CAP participant summary
information on FAAS and FAES can refer to the previous report that the number of participants using ISEs for
version of this chapter. lithium analysis had increased from less than 20 in 1987 to
540, or 25% of the participants, in 1989. Currently, 25% of
Flame Atomic Emission Spectroscopy (FAES) the 2007 CAP participants use ISEs for lithium analysis.
Determination of lithium concentrations using FAES is
based on characteristic emission of lithium atoms at 670.8 Similar to sodium and potassium analysis by ISEs, an
nm when given sufficient energy by hot flame. This electrical potential that is proportional to lithium
characteristic wavelength is selected to provide sufficient concentration in the sample is generated between a lithium
sensitivity and specificity for lithium. Under controlled ISE and a reference electrode, and measured by a
conditions, the emission intensity at 670.8 nm is directly potentiometer. However, the introduction of lithium ISEs
proportional to the number of lithium atoms and therefore was much later than sodium and potassium ISEs, because
lithium concentration in a sample. This serves as the basis early lithium ISEs responded excessively to many common
for lithium quantification using FAES [3]. ions found in serum, such as Na+, K+, Mg++, Ca++, H+ and
NH4+ [6,7]. Interference from Na+ was particularly
Flame Atomic Absorption Spectroscopy (FAAS) troublesome because the molar ratio of lithium to sodium
FAAS uses the same characteristic wavelength as FAES
for lithium analysis. However, there is an important
difference between these two techniques. In FAES, lithium i
Lithium
atoms in the flame are excited to a high energy level so Previous and current authors of this method:
they emit light, and it is the emitted light that is measured. First edition: Robert L. Murray
In FAAS, lithium atoms in the flame are kept at a low Methods edition: Robert L. Murray
energy level so that they are capable of absorbing radiation Second edition: Robert L. Murray
at a very narrow bandwidth corresponding to lithiums Third edition: Steven C. Kazmierczak,
own characteristic wavelength. In FAAS, a hollow lithium Fourth edition: Steven C. Kazmierczak
cathode lamp is used to produce the characteristic emission Fifth edition: Danni L. Meany, William Clarke
824
Lithium
ions in serum is about 1:1500 [4]. One approach to method was acceptable.
mitigate the positive bias from Na+ was to use algorithms
in microprocessor circuits to correct for it [4,5]. However, Reference and Preferred Methods
a more definitive way to achieve a high sensitivity for The reference method for lithium analysis is based on
lithium was to change the composition of the ion-selective FAAS [15], although both FAES and FAAS methods are
membrane and add selectivity-enhancing additives in the equally precise at clinically significant concentrations [16].
ionophores [8,9]. For example, addition of The preferred method for serum lithium analysis has
trioctylphosphine oxide to the ionophore dodecylmethyl- changed from flame photometry (FAAS and FAES) to
14-crown-4, improves both sensitivity and selectivity of ISEs to the current interest in colorimetric methods
lithium ISEs [9]. The drawback of improving lithium ISEs because of the reasons described earlier in this chapter.
with additives may be its increased sensitivity to protons. The performance of all these methods meets acceptable
To prevent errors caused by loss of carbon dioxide, serum performance criteria for lithium by the Clinical Laboratory
samples for lithium analysis by ISEs should be handled Improvement Amendments of 1988 (CLIA-88), so the
anaerobically [4,7]. choice of methodology can be based on available
instrumentation, degree of automation available, and
Colorimetric Methods logistic considerations.
Although colorimetric methods are the most recently
introduced approach for lithium analysis, they are the most Specimen
widely used methods in clinical laboratories. Their use is Because lithium reaches its steady-state concentration in
promoted by the availability of automated multiple- serum approximately 10 to 12 hours after the last dose in a
channel analyzers in rapid-response and core laboratories, patient on chronic dosing, the standard draw time for
potential for workstation consolidation, and labor savings lithium therapeutic monitoring is at least 10 to 12 hours
if lithium analysis can be included on common, existing after the evening dose for patients on a twice-daily dose
analytic instrumentation. According to the 2007 CAP regime [1]. Serum and plasma are commonly used
participant summary report, over 75% of the participants specimens, although specimen requirement may be method
use one of the following colorimetric methods described dependent. If plasma is used, sodium heparin Vacutainer is
here: crown ether method, substituted porphyrin method, recommended, and lithium heparin Vacutainer should be
and crown formazan method. avoided. Serum and plasma should be separated from
cellular blood components if storage of more than 4 hours
In the crown ether method, lithium binds to a crown ether is anticipated. Once separated, lithium is stable for 8 hours
dye, which causes a colorimetric shift of peak absorbance at room temperature, for 2 days at 4C, and for months if
from 400 nm to 600 nm. The shift of the peak absorbance frozen. Serum separator tubes are unacceptable because
is then used quantitatively to correlate with lithium lithium concentration may be reduced when sample is
concentrations [10,11]. Overall, this method was capable stored in a separator tube for a prolonged period of time
of quantitating serum lithium with adequate precision and (>2 to 6 hrs). Hemolyzed specimens should be rejected for
correlated well with ISE-based methods. analysis.
In the substituted porphyrin method, lithium chelates with Interferences

a substituted porphyrin compound and decreases its Analytical interferences are method dependent and should
absorbance of visible light at alkaline pH. This method be determined on an individual basis. In general, FAAS
determines the lithium concentration by calculating the and FAES methods are less prone to interferences than
difference of absorption of an unknown sample at 505 and ISEs and colorimetric methods. FAES results are
480 nm against that of a calibrator [12]. A comparison of unreliable for urine analysis only when extreme
this method with an ISE-based method showed excellent concentrations of potassium and hemoglobin are present in
agreement. urine. ISEs and colorimetric methods are prone to
interferences from both endogenous and exogenous
Crown formazan dyes have been reported previously for sources. As stated previously, the endogenous interfering
colorimetric lithium analysis. The method discussed here ions for ISEs mainly include Na+ and H+. In the crown
is based on the use of the crown formazan dye, 7-nitro-2, ether method, a slight sodium-dependent positive bias in
12-dicarboxyl-16, 17-dihydro-5H, 15H-dibenzo (b, i) (1, lithium determination was evident at 157 mmol/L, but only
11, 4, 5, 7, 8) dioxatetraazacyclotetradecine (shown in clinically significant at physiologically extreme
Figure 1), which reacts rapidly with lithium in an alkaline concentrations (>188 mmol/L). Very high concentration of
mixture of water and dimethyl sulfoxide to form a hemoglobin (>325 mg/dL) in grossly hemolyzed
monovalent binary complex with increased absorbance at specimens were also found to cause a clinical significant
540 nm [11,13,14]. This mixture was found not to cause positive bias [10]. However, potassium, bilirubin, and
precipitation of serum proteins [13]. Similarly to the triglycerides did not cause detectable interference with
substituted porphyrin method, it determines lithium lithium analysis. In the substituted porphyrin method,
concentration by the difference of absorption at 540 and hemolysis, icterus, lipemia, and high and low sodium
700nm. Correlation between this method and the FAAS levels did not significantly interfere. In the crown
825
Lithium
formazan method, the crown formazan dye reacts with no completely absorbed in the gastrointestinal tract. Peak
other serum cations except for sodium because of its weak concentrations occur within 2 to 4 hours after
affinity. To minimize this problem, excess sodium ions administration of immediate-release formulations. Slow-
were added in reagents so that specimen-to-specimen release formulations are associated with later and lower
difference in sodium concentrations would have minimal peak concentrations. Lithium is neither metabolized nor
effects [13]. Interferences from other sources were bound to proteins. Approximately 95% of lithium is
minimal. excreted by the kidneys through two phases: the first phase
clears up to two thirds of an acute dose within 6 to 12
The exogenous interfering compounds for ISE and hours, followed by a second-phase elimination over the
colorimetric methods are the frequently prescribed drugs next 12 hours [24]. Lithium is also able to cross the
such as carbamazepine, n-acetylprocainamide, placenta and is excreted in breast milk. It should be used
procainamide, quinidine, lidocaine, and valproic acid [1]. with caution in pregnant and breastfeeding women [25].
However, these interferences are concentration dependent
and typically do not introduce clinically significant errors. Lithium has a low therapeutic index. Toxicity of lithium is
Moreover, one ISE method has been shown to be proportional to concentrations just above the therapeutic
positively biased by when the blood sample was collected range, although some patients may display signs of toxicity
in a red-top plastic Vacutainer Plus Tube (Becton at the upper end of it. Lithium intoxication most
Dickinson) containing a silica clot activator and silicone commonly occurs in the settings of hyponatremia, fluid-
surfactant [17]. Some ISEs introduce a positive bias in the volume depletion, and use of diuretics or angiotensin-
presence of high calcium concentrations (>8.9 mmol/L). converting enzyme inhibitors when renal excretion of
lithium is reduced [20]. The clinical manifestations of mild
Lithium Reference Intervals lithium intoxication include irregular coarse tremors,
Following the initiation of lithium therapy, dosage muscle weakness, nausea, vomiting, and diarrhea. Severe
adjustments should be based on serum concentrations lithium toxicity can lead to seizures, stupor, and coma.
determined on a biweekly or weekly basis. Once steady- Moreover, the complications of chronic lithium use
state concentrations and symptom remission are achieved, include renal insufficiency, nephrogenic diabetes insipidus,
the lithium concentration should be monitored every 1 to 3 leukocytosis, goiter, and hypothyroidism [2,24].
months [18]. Suggested lithium therapeutic concentrations Treatments for lithium intoxication include fluid repletion
are illustrated in Table 2. Serum lithium concentrations to restore sodium and water balance in hyponatremia, oral
between 1.5 and 2.5 mmol/L are considered mildly toxic, charcoal administration to decrease intestinal absorption
and between 2.5 and 3.5 mmol/L moderately toxic [2]. [26], and hemodialysis for severe lithium toxicity [27,28].
Urine lithium estimation is of extremely limited use except
to verify a patients compliance or monitor changes in Lithium Performance Goals
lithium renal-excretion patterns. Acceptable performance criteria for lithium by CLIA-88
are target value 0.3 mmol/L or 20%, whichever is
greater. Nevertheless, more stringent performance goals,
Lithium Reference Intervals such as less than 3% and 5% analytical imprecision, have
been reported elsewhere [1,29]. Currently, according to the
Prophylactic control of mania: 0.6-1.2 mmol/L [19] 2007 CAP survey, the imprecision for all the lithium
Treatment of acute mania: 0.8-1.5 mmol/L [20] measurements range from 3% to 10% for specimens with
Long-term control of manic-depressive illness: an average lithium concentration of 1.48 mmol/L.
0.4 to 0.8 mmol/L [21]
Mild toxicity: 1.5-2.5 mmol/L [2] References
Moderate toxicity: 2.5-3.5 mmol/L [2] 1 Linder MW, Keck PE Jr. Standards of laboratory
practice: antidepressant drug monitoring, National
Interpretation Academy of Clinical Biochemistry. Clin Chem.
Lithium has been effectively used for treatment of acute 1998;44:1073-1084.
mania and bipolar disorders for 3 decades, although its 2 Orsulak PJ. Agents for the treatment of bipolar
exact mechanism of action is unknown [1,2]. The effects disorder. In: Shaw LM, editor. The Clinical
of lithium on depolarization-provoked and calcium- Toxicology Laboratory: Contemporary Practice
dependent release of dopamine and norepinephrine from of Poisoning Evaluation. Washington, DC:
presynaptic nerve endings in the central nerve system, American Association for Clinical Chemistry;
neuronal secondary messenger pathways, and distribution 2001:237-245.
of Na+, Ca2+, and Mg2+ across neuronal membranes all 3 Kricka LJ. Optical techniques. In: Burtis CA,
have been suggested to contribute to its therapeutic effects Ashwood ER, Bruns DE, editors. Tietz Textbook
[22-24]. of Clinical Chemistry and Molecular Diagnostics.
St Louis: Mosby; 2006:61-91.
Lithium is available in both immediate and slow-release 4 Okorodudu AO, Burnett RW, McComb RB,
formulations. After ingestion, lithium ions are almost Bowers GN Jr. Evaluation of three first-
826
Lithium
generation ion-selective electrode analyzers for from clot activator in collection container. Clin
lithium: systematic errors, frequency of random Chem. 1997;43:675-679.
interferences, and recommendations based on 18 NIMH/NIH Consensus Development Panel.
comparison with flame atomic emission Consensus Development Conference Statement.
spectrometry. Clin Chem. 1990;36:104-110. Mood disorders: pharmacologic prevention of
5 Bertholf RL, Savory MG, Winborne KH, recurrences. Am J Psychiatry. 1985;142:469-476.
Hundley JC, Plummer GM, Savory J. Lithium 19 Roberts WL, McMillin GA, Burtis CA, Bruns
determined in serum with an ion-selective DE. Reference information for the clinical
electrode. Clin Chem. 1988;34:1500-1502. laboratory. In: Burtis CA, Ashwood ER, Bruns
6 Oesch U, Ammann D, Simon W. Ion-selective DE, editors: Tietz Textbook of Clinical Chemistry
membrane electrodes for clinical use. Clin Chem. and Molecular Diagnostics, Philadelphia:
1986;32:1448-1459. Saunders; 2006:2251-2318.
7 Kitazawa S, Kimura K, Yano H, Shono T. 20 Okusa MD, Crystal LJ. Clinical manifestations
Lithium-selective polymeric membrane electrodes and management of acute lithium intoxication.
based on dodecylmethyl-14-crown-4. Analyst. Am J Med. 1994;97:383-389.
1985;110:295-299. 21 Boton R, Gaviria M, Batlle DC. Prevalence,
8 Kimura K, Oishi H, Miura T, Shono T. Lithium pathogenesis, and treatment of renal dysfunction
ion selective electrodes based on crown ethers for associated with chronic lithium therapy. Am J
serum lithium assay. Anal Chem. 1987;59:2331- Kidney Dis. 1987;10:329-345.
2334. 22 Baldessarini RJ, Vogt M. Release of 3H-
9 Bastard G. Bastard replies. Phys Rev Lett. dopamine and analogous monoamines from rat
1988;60:2561. striatal tissue. Cell Mol Neurobiol. 1988;8:205-
10 Gorham JD, Walton KG, McClellan AC, Scott 216.
MG. Evaluation of a new colorimetric assay for 23 Chen G, Huang LD, Zeng WZ, H KM. Mood
serum lithium. Ther Drug Monit. 1994;16:277- stabilizers regulate cytoprotective and mRNA-
280. binding proteins in the brain: long-term effects on
11 Christenson RH, Mandichak JJ, Duh SH, cell survival and transcript stability. Int J
Augustyn JM, Thompson JC. Clinical Neuropsychopharmacol. 2001;4:47-64.
performance characteristics of a new photometric 24 Baldessarini RJ, Tarazi FI. Drugs and the
lithium assay: a multicenter study. Clin Chim treatment of psychiatric disorders: psychosis and
Acta. 2003;327:157-164. mania. In: Hardman JG, Limbird LE, editors.
12 Lyon AW, Whitley C, Eintracht SL. Analytic Goodman & Gilmans The Pharmacological Basis
evaluation and application of a novel of Therapeutics. New York: McGraw-Hill;
spectrophotometric serum lithium method to a 2001:485-520.
rapid response laboratory. Ther Drug Monit. 25 Schou M. Lithium treatment during pregnancy,
2004;26:98-101. delivery, and lactation: an update. J Clin
13 Thompson JC. Development of an automated Psychiatry. 1990;51:410-413.
photometric assay for serum lithium and use of 26 Linakis JG, Lacouture PG, Eisenberg MS, Maher
binding equilibrium expressions to optimize TJ, Lewander WJ, Driscoll JL et al.
results. Clin Chim Acta. 2003;327:149-156. Administration of activated charcoal or sodium
14 Gruson D, Lallali A, Furlan V, Taburet AM, polystyrene sulfonate (Kayexalate) as gastric
Legrand A, Conti M. Evaluation of a new lithium decontamination for lithium intoxication: an
colorimetric assay performed on the dade behring animal model. Pharmacol Toxicol. 1989;65:387-
dimension X-pand system. Clin Chem Lab Med. 389.
2004;42:1066-1068. 27 Jacobsen D, Aasen G, Frederichsen P, Eisenga B.
15 Pybus J, Bowers GN Jr. Serum lithium Lithium intoxication: pharmacokinetics during
determination by atomic absorption spectroscopy. and after terminated hemodialysis in acute
In: McDonald RP, editor. Standard Methods of intoxications. J Toxicol Clin Toxicol. 1987;25:81-
Clinical Chemistry. New York: Academic Press; 94.
1970:189-192. 28 Hauger RL, OConnor KA, Yudofsky S, Meltzer
16 Levy AL, Katz EM. A comparison of serum HL. Lithium toxicity: when is hemodialysis
lithium determinations using flame photometry necessary? Acta Psychiatr Scand. 1990;81:515-
and atomic absorption spectrophotometry. Clin 517.
Chem. 1969;15:787. 29. Fraser CG. Desirable standards of performance
17 Sampson M, Ruddel M, Albright S, Elin RJ. for therapeutic drug monitoring. Clin Chem.
Positive interference in lithium determinations 1987;33:387-389.
Table 1: Lithium Methods Summary

827
Lithium
Method 1: Flame atomic absorption spectrophotometry; quantitative

Principle of analysis: Absorption of light at 670.8 nm by Li0
Comments: Serum, plasma, urine, red blood cells
Method 2: Flame atomic emission spectrophotometry; quantitative

Principle of analysis: Emission of light at 670.8 nm by Li0
Comments: Serum, plasma
Method 3: Ion-selective electrodes; quantitative

Principle of analysis: Measurement of voltage difference generated by Li+ in contact with lithium ionophore
Comments: Serum, plasma, whole blood

Lithium + crown ether (400 nm) dye complex (600 nm)

Principle of analysis: Lithium bound to substituted porphyrin compound, resulting in decrease of absorbance of
visible light at alkaline pH
Comments: Serum, plasma; this is currently the most popular method in diagnostic laboratories.

Principle of analysis: Lithium bound to crown formazan dye, 7-nitro-2, 12-dicarboxyl-16, 17-dihydro-5H, 15H-
dibenzo [b, i] [1, 11, 4, 5, 7, 8] dioxatetraazacylotetradecine in an alkaline mixture of water and dimethyl sulfoxide to
form a monovalent binary complex with increased absorbance at 540 nm
Figure
Figure 1. Structure of crown formazan dye under conditions of the assay and with lithium bound within
the pocket of the 14-membered ring [13].
828
Luteinizing Hormone
Luteinizing Hormone
Gregory Hobbs
Name: Luteinizing hormone, LH

Clinical significance: Refer to Chapter 44, Pregnancy, and Chapter 50, The Gonads, in the 5th
Edition of Clinical Chemistry: Theory, Analysis, Correlation.
Half-life: 30 minutes
i anti- LHcoated microparticles, forming an antigen-

Luteinizing hormone (LH), a glycoprotein with a antibody complex. This mixture is transferred to a
molecular weight of approximately 30,000 daltons, is reaction cell, where the microparticles bind irreversibly
produced by the anterior pituitary. Like follicle- to a glass fiber matrix. After washing to remove
stimulating hormone (FSH), LH is also present as a unbound materials, an anti- LH antibody conjugated to
heterodimer with two different subunits, and The alkaline phosphatase is added, which binds to the
subunit (molecular mass 14000 D) is common to all antigen-antibody complex. After again washing the
gonadotropins and also to thyroid-stimulating hormone matrix cell to remove unbound materials, the substrate,
(TSH). The sub-unit (118 amino acids) is specific for methylumbelliferyl phosphate, is added and the
LH because of its amino-acid sequence and carbohydrate fluorescent product, methylumbelliferone, is measured.
content [1]. Thus immunoassays are specific for the
Roche Diagnostics produces an assay based on
subunit. Early immunoassays used radioisotopes.
electrochemiluminescence detection technology (ECL)
Although radioimmunoassays are available, current
[6,7]. This is a sandwich-based immunoassay that
immunoassays use enzyme-linked immunosorbent assay
involves two incubations. The first involves the sample,
(ELISA) [2], proprietary enzyme immunoassay (EIA)
a biotinylated monoclonal antibody specific for LH, and
methods, or chemiluminescent methods.
a second LH-specific monoclonal antibody labeled with
a ruthenium complex. An antibody-antigen-antibody
The Vitros immunodiagnostic method [3] uses an
sandwich forms during this incubation. In the second
immunometric assay technique. The LH in the sample is
incubation, streptavidin-coated microparticles are added,
captured by a biotinylated antibody on one binding site,
and the sandwich complex becomes bound to the
while a second antibody labeled with horseradish
microparticle by way of the biotin-streptavidin
peroxidase binds to a different site. The antibody
interaction. The reaction mixture is aspirated into a
complex is captured by streptavidin bound to plastic
measuring cell where the microparticles are captured
wells. The peroxidase bound to the solid phase is used to
magnetically on the surface of an electrode. Unbound
catalyze the oxidation of a luminol derivative, with the
substances are removed by a wash step. ECL works by
production of light.
the removal of the magnet and supply of voltage to the
electrode. The ruthenium in the complexes and
The Siemens (formerly DPC) procedure is a two-site
tripropylamine in the residual buffer are oxidized. The
sandwich chemiluminometric immunoassay [4]. LH in
two compounds react to produce an excited state of
the patient serum binds to an acridinium ester (AE)-
ruthenium. The spontaneous decay results in emission of
labeled mouse monoclonal antibody (Lite Reagent). A
a photon of light at 620 nm.
second antibody specific for LH attached to magnetic
particles captures the complex. The acridinium ester
Unfortunately, these assays are still not standardized [8],
remaining on the solid phase is reacted chemically to
and results are not comparable across methods.
release light. A direct relationship exists between the
Reference ranges also vary with method.
amount of LH in the patient sample and the amount of
relative light units (RLUs) detected by the Advia
Centaur system.
There is no established reference method.
Abbott Diagnostics produces an LH method based on its
It is preferred that any selected method be standardized
microparticle enzyme immunoassay (MEIA) technology
to a World Health Organization (WHO) reference
[5] for its AxSYM platform. In this method, LH binds to
material. The current practice is to use the second WHO
International Standard LH 80/552.
i
Luteinizing Hormone (LH)
Previous and current authors of this method: Sensitive LH methods are available from various
First edition: Not done manufacturers. As an example, we describe below the
Second edition: Not done assay available on the ACS 180 system, which is also
Third edition: Not done performed on the Bayer Centaur. Both systems are now
Fourth edition: Amadeo J. Pesce offered by Siemens Medical Solutions.
Fifth edition: Gregory Hobbs
829
Luteinizing Hormone
Specimen in moderate amounts, does not interfere with most

The choice of serum or plasma is dependent in part on immunoassay methods. Abbott notes that the sample
the manufacturer of the test system. For example, for the should be free of fibrin, which may cause erroneous
ACS 180 system, serum is the preferred specimen, while results [9]; for this reason, complete clot formation
for the Vitros or Abbott systems, serum, EDTA, or should take place.
heparinized plasma are all acceptable. However, some
systems show significantly increased results when Reference Interval
Interferences Selected reference interval values for the LH
Heterophile antibodies such as human anti-mouse concentration in serum using the WHO International
antibody can interfere in many immunoassays. The Standard 80/552 in milli-international units/mL (mIU/L)
presence of hemolysis, lipemia, or icterus, when present are:
Bioserve Abbott Roche

Patient Population ELISA Ortho Siemens AxSYM Elecsys
Females
Follicular phase 20 2.6-12 0.8-25.8 1-18 2.4-12.6
Midcycle peak 40-200 27-93 25-57.3 24-105 14-95.6
Luteal phase 20 0.8-15 0.8-27.1 0.4-20 1.0-11.4
Postmenopausal 20-100 13-85 39.7-103.5 15-62 7.7-58.5
Males 3-12 1.6-9.7 1.3-12.9 2-12 1.7-8.6
gel-separator tubes compared with specimens collected 3 Ortho Clinical Diagnostics. VITROS
in plain evacuated tubes. The sample should be Immunodiagnostic Products LH Reagent Pack.
refrigerated if not analyzed immediately. Samples may Instructions for Use. Publication Number
be refrigerated for up to 5 days (Abbott recommends 24 J03784. Rochester, NY: Ortho Diagnostics;
hours). Samples may be stored at 20C but not frozen 2002.
and thawed repeatedly. 4 Bayer HealthCare LLC. LH ADVIA Centaur
Assay Manual. 111736, Rev. J, 2005.
Interferences 5 Abbott Laboratories Diagnostic Division. LH in
Heterophile antibodies such as human anti-mouse Abbott AxSYM system, 34-2540/R8, Abbott
antibody can interfere in many immunoassays. The Park, IL 60064, 2003.
presence of hemolysis, lipemia, or icterus, when present 6 Roche Diagnostics. GmbH, D-68298. LH,
in moderate amounts, does not interfere with most Luteinizing Hormone, 11818252001v13.
immunoassay methods. Abbott notes that the sample Mannheim, Germany, 2005-06.
should be free of fibrin, which may cause erroneous 7 Roche Diagnostics website.
results [9]; for this reason, complete clot formation <http://us.labsystems.roche.com/products/ecl/ec
should take place. ltech.shtml>
8 Sturgeon, CM Ellis, AR. Standardization of
Interpretation FSH, LH, and hCG: current position and future
LH is made in the anterior lobe of the pituitary gland. prospects. Mol Cell Endo 2007;260-262:301-
LH production and release is under the control of the 309.
hypothalamus through gonadotropin-releasing hormone 9 Abbott Laboratories Diagnostics Division.
(GnRH). In women, LH and FSH are responsible for the Abbott AxSYM Sytem: LH. Abbott Park, IL:
physiological changes of the ovaries during the Abbott Diagnostics; 2003.
menstrual cycle. Under the influence of FSH, LH, and 10 Kicklighter EJ Norman RJ. The gonads. In:
estrogens, the follicles mature during the follicular Kaplan LA, Pesce AJ, eds. Clinical Chemistry:
phase. Theory, Analysis, and Correlation. St Louis:
In men, while FSH acts directly on the Sertoli cells in Mosby; 1989:650-663.
the spermatogenic tubule, LH influences spermato- 11 Liwnicz BH, Liwnicz RG. The
genesis indirectly by increasing testosterone synthesis in hypothalamopituitary system. In: Kaplan LA,
the adjacent Leydig cells. Pesce AJ, eds. Clinical Chemistry: Theory,
Analysis, Correlation. 3rd ed. St Louis: Mosby;
LH and the other pituitary gonadotropin, FSH, play a 1989:613-619.
critical role in maintaining the normal function of the 12 Scott MG, Ladenson JH, Green ED, Gast MJ.
male and female reproductive systems. Hormonal evaluation of female infertility and
reproductive disorders. Clin Chem
References 1989;35:620-629.
1 Niewoehner CB. Endocrine Pathophysiology. 13 Butt WR, Blunt SM. The role of the laboratory
Madison, CT: Fence Creek; 1998. in the investigation of infertility. Ann Clin
2 BioServ Diagnostics (Rostock, Germany) Biochem 1988;25:601-609.
website. <http://www.bioserv-diagnostics.com>
830
Luteinizing Hormone
Procedure: Siemens ACS:180 Assay Reagents

Assay Reagents
Editors note: the following procedure is used only Reagent Volume/Ingredients Storage/Stability
as an example of an automated immunoassay for LH Lite 5 mL/vial; polyclonal sheep anti-LH 2C8C; until
Reagent antibody (~1.344 mg/vial) labeled on the date vial
LH, and is not intended to promote this assay or acridinium ester in buffer, labeled with label or
any other assay. sodium azide, and microbicides cumulative 40
hours at room
temperature
PUBLISHERS NOTE: this material, derived from the
LH Solid 20 mL/vial; monoclonal mouse anti- 2C8C; until
Siemens package insert, cannot replace the Phase LH antibody (~1.066 mg/vial) expiration date
information provided by the manufacturer for the covalently coupled to pasramagnetic on the vial label
analysis of this analyte on their instrument. Please a articles in buffer, sodium azide, and or cumulative
refer all questions to the manufacturer. microbiocides 40 hours at room
s temperature
Luteinizing Hormone (LH) in Serum on the
Siemens Automated Chemiluminescence System
(ACS)-180 CAUTION: Sodium azide can react with copper and lead
plumbing to form explosive metal azides. On disposal,
Principle flush with a large volume of water to prevent the buildup
The Siemens ACS LH assay is a two-site of azides.
chemiluminometric (sandwich) immunoassay which
uses constant amounts of two antibodies that have BIOHAZARD: These reagents are intended for use with
specificity for the intact LH molecule. The first antibody human serum. Components from biological sources were
or Lite Reagent is a polyclonal sheep anti-LH antibody used in the manufacture of this product. It is
labeled with acridinium ester (AE). The second recommended that appropriate biosafety precautions be
antibody, or Solid Phase, is a monoclonal mouse anti-LH followed to prevent the transmission of infectious
antibody covalently coupled to paramagnetic particles. A agents.
direct relationship exists between the LH in the sample
and the RLUs detected by the ACS:180 systems. Preparing the Assay Reagents
1. For best results, thoroughly mix the Solid Phase by
This assay is intended for the quantitative determination inverting the vial before each use. Visually inspect
of luteinizing hormone (LH) in serum using the Siemens the bottom of the vial to ensure that all the particles
Automated Chemiluminescence System (ACS). Results are dispersed and suspended. If foaming occurs,
from this assay can be combined with other results to rim the top of the vial with applicator sticks to
assess the integrity of the pituitary-gonadal axis. remove the foam.
Specimen Collection and Handling 2. Place a septum cap on the Lite Reagent and on the
Serum is the recommended sample type for this assay. Solid Phase vials.
Cap and refrigerate specimens if testing is not done 3. Prewarming or bringing the reagents to room
immediately, or freeze specimens if testing is not done temperature before use is not required.
within 2 days. Freeze specimens only once, and mix
thoroughly after thawing. No further patient preparation Master Curve Calibration
is required. The ACS LH assay requires a Master Curve calibration
Assay Reagents when you use a new lot number of Lite Reagent and
For In-Vitro Diagnostic Use. Solid Phase. For each new lot of Lite Reagent and Solid
Caution: Phase, use the Master Curve card to enter the Master
1. Discard opened assay reagents that are at room Curve information at the System
temperature (20C to 30C) for a total of 40 hours. Do Management/Definitions/Master Curve screen. The
not use these reagents to calibrate the ACS:180 systems Siemens LH assay is standardized against the World
or assay samples. Health Organization (WHO) 2nd IRP 80/552 Reference
2. Do not use kit components beyond expiration Material.
date.
3. Do not mix different lot numbers of reagents. Two-Point Calibration Interval
The ACS LH assay requires a two-point calibration
Contents every 28 days. Also, perform a two-point calibration
Catalog Number Contents when Number
you change lot numbers of Lite Reagent and Solid
of Tests
672270 Six vials of LH Lite Reagent Phase300
or when your controls are repeatedly out of range.
Six vials of LH Solid Phase
One Master Curve card Quality Control
or For detailed information about entering quality-control
672271 One vial of LH Lite Reagent (QC) information, refer to Section 7, Calibration and
50 Quality Controls, in your ACS:180 Plus and ACS:180
One vial of LH Solid Phase Reference Manual. Enter quality-control information at
One Master Curve card the System Management/Definitions/Controls screen.
831
Luteinizing Hormone
additional volume required when performing

To monitor system performance and chart trends, duplicates or other tests on the same sample.
analyze three levels of QC materials for LH at least once 5. If automatic dilution is required, dispense Multi-
a day, usually at the beginning of the working day. Each Diluent 1 into a sample cup labeled with the
separate tray should have at least one level of QC pool to appropriate barcode label, and load the sample cup
detect between-tray drifts. This policy is in accordance on the sample tray.
with CLIA 88 requirements and Ciba-Cornings If manual dilution is required, refer to Procedural
recommendations that a minimum of three levels of Notes.
controls be run at least one time during a 24-hour period 6. Load the Lite Reagent and Solid Phase in adjacent
and when a two-point calibration is performed. positions on the reagent tray.
7. Press START. The ACS:180 system:
Please refer to the laboratorys QC manual for the a. dispenses 100 mL of sample into a cuvette
procedure for the reconstitution of these pools and the b. dispenses 100 mL of Lite Reagent and
interpretation of the QC results. All QC data must be incubates it for 5 min at 37C
entered into the LIS. c. dispenses 400 mL of Solid Phase and
incubates it for 2.5 min at 37C
Assay Procedure d. separates, aspirates, and washes the cuvette
For detailed information about operating the ACS:180 with reagent water
systems, refer to Section 6, Operating the ACS:180, in e. dispenses 300 mL each of Reagent 1 and
your ACS:180 Plus and ACS:180 Reference Manual. Reagent 2 to initiate the
NOTE: chemiluminescent reaction
If automatic tray and cup assignment is on, use the f. prints results according to the print option you
printed worklist to help you load calibrators, controls, selected, which is described in Section 5,
and patient samples into the correct tray and cup Defining System Parameters, in your ACS:180
positions. Plus and ACS:180 Reference Manual.
If automatic tray and cup assignment is off, and
your controls and patient samples are all barcoded, you Calculating Results
can load samples in any position. For detailed information about how the ACS:180
1. Schedule the requested tests or profiles for each calculates results, refer to Section 2, Understanding
calibrator, control, or patient sample. the ACS:180, in your ACS:180 Plus and ACS:180
2. Prepare and load ACS Calibrator B, if required. Reference Manual.
a. Reconstitute the low and high calibrators
according to the preparation instructions in the LH reference IntervalsSiemens
ACS Calibrator B product insert. As with all diagnostic tests, each laboratory should
b. Dispense the low and high calibrator into a establish its own reference intervals for the diagnostic
sample cup labeled with the appropriate evaluation of patient results.
barcode label.
c. Load the sample cups in any position on the To determine the Siemens ACS LH reference intervals,
sample tray. the data in the following table were generated.
Ensure that the low calibrator precedes the
high calibrator on the sample tray.
3. Prepare and load the controls.
a. Prepare the controls according to the
instructions in the control product insert.
b. Dispense the controls into a labeled sample
cup.
c. Load the sample cup in the appropriate
position on the sample tray.
4. Prepare the primary tubes or sample cups, and load
them on the sample tray.
This assay requires 100 mL of sample for a single
determination. This volume does not include the
unusable volume in the sample cup or the
832
Luteinizing Hormone
Expected Values
Mean Observed Range
Category N (mIU/mL) (mIU/mL)
Females
Normally menstruating
Follicular phase 63 5.0 0.825.8
Midcycle peak 8 43.0 25.057.3
Luteal phase 73 4.2 0.8-27.1
Pregnant 35 <0.1 <0.11.4
Postmenopausal
No hormone therapy 82 65.0 39.7103.5
Hormone therapy 12 25.5 12.047.8
Contraceptives 15 6.2 1.011.9
Ovarian failure 15 73.8 23.4123.0
Turners syndrome 5 70.5 64.375.7
Polycystic ovary disease 61 52.9 16.6114.0
Males
13-20 years 25 8.2 4.69.4
20-70 years 100 4.1 1.312.9
>70 years 25 30.6 11.356.4
Testicular failure 16 61.2 45.487.2
Klinefelters syndrome 8 52.5 44.267.7
Children 45 2.8 1.05.9
Procedural Notes endogenous LH values, whereas administration of

1. Dispose of hazardous and biologically estrogens or testosterone can result in decreased
contaminated materials according to your endogenous LH values. Administration of hMG to males
institutions practices. Discard all materials in a can result in elevated endogenous LH values, whereas
safe and acceptable manner and in compliance with administration of testosterone decreases endogenous LH
all federal, state, and local requirements. values.
2. Samples greater than 200 ng/mL must be diluted
and then repeated to obtain accurate results. Use Patient samples with high LH levels can cause a
the ACS Multi-Diluent 1 to dilute samples. paradoxical decrease in the RLUs (high-dose hook
3. Patient samples can be diluted automatically by the effect). In this assay, patient samples with LH levels as
ACS:180 system or manually. For automatic high as 1500 mIU/mL will assay greater than 200
dilutions, ensure that you load the ACS Multi- mIU/mL.
Diluent 1 and that you have defined values for
Dilution Setpoint and Dilution Ratio in System It is well known that heterophilic antibodies in human
Management. serum can react with reagent immunoglobulins,
4. For detailed information about setting system interfering with in-vitro immunoassays. Patients
parameters, refer to Section 5, Defining System routinely exposed to animals or to animal serum
Parameters, and Section 6, Operating the products can be prone to this interference, and
ACS:180, in your ACS:180 Plus and ACS:180 anomalous high values can be observed. Additional
Reference Manual. information may be required for diagnosis.
5. Manually dilute LH patient samples when patient
results exceed those values obtained by automatic Specimens have an insignificant
dilution or when your laboratory protocol requires that are effect on the ssay up to
manual dilution, not automatic dilution. hemolyzed 500 mg/dL of hemoglobin
6. Use ACS Multi-Diluent 1 to manually dilute lipemic 3000 mg/dL of triglycerides
patient samples, and then load the diluted sample icteric 20 mg/dL of bilirubin
onto the sample tray, replacing the neat sample.
7. Ensure that you mathematically correct the results Performance Characteristics
for the dilution. Specificity
8. To program automatic dilutions at the System The cross-reactivity of the ACS LH assay with TSH, -
Management/Definitions/Tests screen, Siemens LH, human chorionic gonadotropin (hCG), -hCG, FSH,
recommends you select a 1:2 or 1:5 dilution ratio prolactin, and human growth hormone (hGH) was
for optimal performance of the ACS LH assay. determined by adding these hormones to samples
containing LH. The level of LH in the samples was then
Limitations of Procedure determined.
Administration of gonadotropin-releasing hormone
(GnRH), clomiphene citrate, and human menopausal
gonadotropin (hMG) to females can result in elevated
833
Luteinizing Hormone
Performance Characteristics
LH Value Without LH Value With
Cross-Reactant Cross-Reactant (mIU/mL) Cross-Reactant (mIU/mL)
TSH, 1000 mIU/mL 4.8 5.0
10.2 11.6
71.5 70.7
-LH, 5 ng/mL 4.3 4.1
9.2 9.7
90.7 91.1
hCG, 200,000 mIU/mL 3.2 3.8
9.4 10.6
99.9 92.9
-hCG, 200,000 mIU/mL 2.9 2.9
10.5 10.5
69.3 65.6
FSH, 200 mIU/mL 2.7 2.6
9.4 9.4
77.2 79.0
Prolactin, 400 ng/mL 3.2 3.2
9.1 9.5
59.4 67.4
hGH, 100 ng/mL 3.1 3.4
9.2 10.7
57.8 60.2
Sensitivity LH is secreted by the anterior pituitary in response to
GnRH secreted by the hypothalamus [11]. In males, LH
Siemens reports that this assay has a minimum is also called interstitial cellstimulating hormone
detectable concentration of 0.09 mIU/mL. (ICSH) [10]. In both males and females, LH secretion is
regulated by a balance of positive and negative feedback
Reportable Range mechanisms involving the hypothalamic-pituitary axis,
The Siemens LH assay measures LH concentrations up the reproductive organs, and the pituitary and sex steroid
to 200 mIU/mL. hormones [10,12,13]. LH and the other pituitary
gonadotropin, FSH, play a critical role in maintaining
Interpretation the normal function of the male and female reproductive
LH is a glycoprotein hormone having two subunits. The systems. Table 1 shows the target tissues and actions of
subunit is similar to those of FSH, hCG, and TSH. The LH in females and males [10-13].
subunit is different from those of the other gycoprotein
hormones and confers its biochemical specificity [10].
Table 1. LH Target Tissues and Actions

Category Target Tissues Actions
Females Theca cells of the ovarian Stimulates production of androgens
follicles that FSH converts to estradiol during
the follicular phase
Graafian follicle Acts synergistically with FSH to cause
ovulation during midcycle peak
Corpus luteum Stimulates formation of the corpus luteum
after ovulation
Stimulates progesterone secretion during
the luteal phase
Males Leydig cells in the interstitial Stimulates testosterone secretion
tissue of the testes
Abnormal LH levels with corresponding increased or polycystic ovary disease in females and primary
decreased levels of FSH, estrogens, progesterone, and hypogonadism in males. Decreased LH levels are
testosterone are associated with a number of pathological associated with primary ovarian hyperfunction in
conditions [11]. Increased LH levels are associated with females and primary hypergonadism in males.
menopause, primary ovarian hypofunction, and
834
Lysozyme
Lysozyme
Arnold L. Schultz
Name: Lysozyme, muramidase, N-acetylmuramide glycanohydrolase,
N-acetylmuramyl hydrolase, globulin G1
Clinical interpretation:
Merck Index: 5457
the specific inactivation of human lysozymeT4-
bacteriophage conjugate has been used to detect
Lysozyme catalyzes the hydrolysis of the N-
enzymatically inactive forms of lysozyme in addition to
acetylmuramic acid (14) N-acetylglucosamine the biologically active forms[7] (method 6, Lysozyme
linkages in the polysaccharide component of the cell Methods Summary Table). Immunochemical
wall of Micrococcus lysodeikticus. The original method quantitation of lysozyme has been performed by
for the determination of lysozyme activity measured the immunoelectrophoresis; the lysozyme in serum or urine
time required for the sample to lyse part or all of a full- is electrophoresed into antibody-containing agarose gel,
grown culture plate of Micrococcus lysodeikticus and the immunoprecipitates are stained with Coomassie
(method 1, Lysozyme Methods Summary Table).[1] Brilliant Blue R-250[8] (method 7, Lysozyme Methods
Summary Table). Two radioimmunoassays (method 10,
The two most commonly used procedures for the assay Lysozyme Methods Summary Table) for lysozyme have
of lysozyme are based upon the lytic action of the been described.[9,10]
enzyme disrupting the bacterial cell wall resulting in the
dispersion of the cell contents and the clearing of the A method using living Micrococcus lysodeikticus cells
turbid bacterial suspension (method 2, Lysozyme
Methods Summary Table). The photometric method for loaded with trimethylphenylammonium ions (TMPA+)
the determination of lysozyme activity is based upon the as a marker has been reported.[11] This method utilized
decrease in turbidity with time when the sample is added a TMPA+ selective electrode to measure TMPA+
to a suspension of Micrococcus lysodeikticus.[2] released by lysis (method 8, Lysozyme Methods
Turbidity may be monitored at 540, 640, or 645 nm. In Summary Table). The initial rate of potential change is a
another variation of this technique, the cell lysis reaction function of the rate of lysis and is directly proportional
is monitored by nephelometry using white light.[3] to the lysozyme activity. A chromophoric substrate has
been evaluated as an alternative to the bacterial cell wall
The agarose plate or lysoplate technique uses a methods.[12] The substrate used was 3,4-dinitrophenyl
suspension of the bacteria in solidified agar contained in tetra-N-acetyl--chitotetraoside (method 9, Lysozyme
a Petri dish.[4] Methods Summary Table). Upon incubation with serum
Sample wells, which are cut into the agar, are filled, and samples in a sodium citrate buffer, the substrate is
the plate is allowed to stand at room temperature for 12 converted to 3,4-dinitrophenol. The increase in
to 18 h. During this time period a cleared zone of lysis absorbance at 410 nm is measured with respect to time.
develops in the initially translucent agarose gel (method
3, Lysozyme Methods Summary Table). The diameters Reference and Preferred Methods
of the cleared zones are related to the log of the Most of the methods described for the determination of
concentration of lysozyme in the unknown by means of a lysozyme activity in body fluids have significant
series of standards. limitations including substrates that are not available
commercially, lack of sensitivity, necessity for pure
Several other procedures for the determination of human standards that are not readily available, need for
lysozyme activity in body fluids have been investigated. specialized instrumentation, and laborious procedures.
Lysozyme activity has been measured with a
viscosimetric assay, employing glycol chitin as The two most widely accepted techniques for the
substrate[5] (method 4, Lysozyme Methods Summary determination of lysozyme activity are the turbidimetric
Table). Lysozyme action on the glycol chitin substrate method[13] and the lysoplate diffusion method.[14] Both
results in decreased viscosity of the substrate solution, of these methods are strongly affected by several factors
and the change in viscosity can be related to the that must be controlled to give reproducible results. The
lysozyme concentration. In addition to hydrolysis, lysoplate method is influenced by the pH and ionic
lysozyme catalyzes extensive transglycosylation. This strength of the gel medium. The turbidimetric method is
reaction has been used to relate the rate of release of affected by the technique used to prepare the bacterial
methylglycosamide from 14C-labeled chitobiose to cell suspension. Both methods are affected by the pH of
lysozyme concentration[6] (method 5, Lysozyme the reaction mixture, by the nature and concentrations of
Methods Summary Table). An immunoassay based on
C:\5th Edition Print Revised\Lysozyme_2009-06-24.doc
835
Lysozyme
the salts and ions present, and by the reaction (hypergranular progranulocytic leukemia), and M6
temperature.[15] (erythroleukemia); intermediate in M4 (myelomonocytic
leukemia); and high in M5 (monocytic histiocytic
Although both the turbidimetric and lysoplate methods leukemia).[28] Lysozyme, which is placed in
are relatively easy to perform and the results of the two circulation, is rapidly cleared by the kidney. Thus, in
methods correlate well,[16] the lysoplate method may those diseases in which large amounts of lysozyme are
offer an important clinical advantage.[14] Several released, both serum and urine measurements may be
investigators using the lysoplate procedure have reported made.
significantly elevated lysozyme activity in sera from
patients with active Crohns disease.[17-19] However, Lysozyme Performance Goals
when the turbidimetric method is used, the serum The between-run precision of the chromophoric method
lysozyme activity does not readily differentiate Crohns varies from 8% for high concentrations of lysozyme to
disease from other types of bowel inflammation.[20-23] 5% for values within the reference interval. The
sensitivity of the lysoplate method is 0.1 g/mL.
Specimen The sensitivity of the turbidometric method is at least 0.5
Lysozyme activity has been measured in serum, urine, g/mL, and its reproducibility is 18% at 0.5 g/mL and
stool, cerebrospinal fluid, gastric juice, bile, saliva,
4.4 at 25 g/mL.[29]
mothers milk, amniotic fluid, sperm, tears, transudates,
pus[24] and pleural fluid.[25]
References
EDTA plasma or serum may be used. Heparin interferes 1 Fleming A. On a remarkable bacteriolytic
with lysozyme activity through formation of heparin element found in tissues and secretions. Proc
lysozyme complexes. Serum should be separated from Roy Soc 1922;93:306-317.
the clot within 2 h of collection. Delayed separation of 2 Boasson EH. On the bacteriolysis by lysozyme,
the serum can result in spuriously high levels of J Immunol 1938;34:281-93.
lysozyme, presumably because of the breakdown of 3 Borgen J, Romslo I. Lysozyme determined in
leukocytes.[4] Serum lysozyme is stable for at least 1 serum and urine by a simple nephelometric
month at C if not repeatedly frozen and thawed.[25] method. Clin Chem 1977;23:1599-1601.
4 Osserma, EF, Lawlor DP. Serum and urinary
At pH 4.5 to 6.3, urinary lysozyme is stable for 4 days at lysozyme (muramidase) in monocytic and
room temperature. The higher the pH of the urine, the monomyelocytic leukemia. J Exp Med
more rapidly the enzyme activity decreases. Lysozyme 1966;124:921-51.
in urine is stable for 8 days at 4 C and for 18 days at 5 Lundbla G, Hultin E. Human serum lysozyme
20 C.[25] It has been reported that serum and urine (muramidase). I. Viscosimetric determination
samples that have been dried or pretreated on filter paper with glycol chitin and purification by selective
are stable for 13 weeks at room temperature.[26] adsorption. Scand J Clin Lab Invest
1966;18:201-8.
Tears are collected by having the patient stare at a strong 6 Dahlquist FW, Borders CL, Jacobson G,
light from a slitlamp. When tears have collected in the Raftery MA. The stereospecificity of human,
lower cul-de-sac, they are drawn into a micropipet by hen, and papaya lysozymes. Biochemistry
capillary action, with care being taken not to damage the 1969;8:694-700.
conjunctiva.[27] Because of their very high lysozyme 7 Maron E, Bonavida B. A sensitive
activity, tears are diluted 1:20 with the phosphate buffer immunoassay for human lysozyme in biological
before they are assayed.[24] fluids. Biochim Biophys Acta 1971;229:273-5.
8 Johansson BG, Malmquist J. Quantitative
Interferences immunochemical determination of lysozyme
There is a slight increase in lysozyme activity in icteric, (muramidase) in serum and urine. Scand. J Clin
lipemic, and hemolyzed serum when lysolyme is Lab Invest 1971;27:255-61.
measured by the lysoplate method. 9 Peeters TL, Depraeter YR. Vantrappen GR.
Radioimmunoassay for urinary lysozyme in
Lysozyme reference Intervals[24] human serum from leukemic patients Clin
1. Serum, 4 to 15.6 g/mL (0.28 to 1.1 mol/L) Chem 1978;24:2155-7.
10 Thomas MJ, Russo A, Craswell P, Ward M,
2. Urine, 0 to 1.4 g/mL (0 to 0.097 mol/L)
Steinhardt I. Radioimmunoassay for serum and
3. Tears, 1151 to 1383 g/mL (80 to 96 mol/L)
urinary lysozyme. Clin Chem 1981;27:1223-6.
Lysozyme activity is independent of age or sex.
11 DOrazio P, Meyerhoff ME, Rechnitz GA.
Membrane electrode measurement of lysozyme
Interpretation
using living bacterial cells. Anal Chem
In healthy persons serum and urine levels of lysozyme
1978;50:1531-4.
are usually insignificant. Serum and urine levels are
12 Turner GA, Ghneim H, Freeman J. Evaluation
higher in the Chdiak-Higashi syndrome and in some
of 3,4-dinitrophenyltetra-N-acetyl--
leukemias. In the leukemias they are low in M2
chitotetraoside as a substrate for the
(myeloblastic leukemia with maturation), M3
836
Lysozyme
measurement of lysozyme in normal and 21 Pruzanski W, Marcon N. Lysozyme in Crohns

pathological sera. Ann Clin Biochem disease. N Engl J Med 1975;293:611-2.
1979;16:51-3. 22 Peeters TL, Geboes K, Vantrappen GR. Serum
13 Litwack G. Photometric determination of lysozyme levels in Crohns disease (cont.). N
lysozyme activity. Proc Soc Exp Biol Med Engl J Med 1975;292:1349-50.
1955;89:401-3. 23 Pounder RE, Avella JR, McCallum H,
14 Peeters TL, Vantrappen GR. Factors Misiewicz JJ. Serum lysozyme in inflammatory
influencing lysozyme determinations by the bowel disease. Lancet 1975;2:228-9.
lysoplate method. Clin Chim Acta 1977;74:217- 24 Hankiewicz J, Swierczek E. Lysozyme in
55. human body fluids. Clin Chim Acta
15 Smolelis AN, Hartsell SE. Factors affecting the 1974;57:205-9.
lytic activity of lysozyme. J Bacteriol 25 Klockars M, Pettersson T, Riska H, Hellstrm
1952;63:665-74. PE, Norhagen A. Pleural fluid lysozyme in
16 Zucker S, Hanes DJ, Vogler WR, Eanes RZ. human disease Arch Intern Med 1979;139:73-7.
Plasma muramidase: a study of methods and 26 Maeda K, Ito K, Yamaguchi N. A simple
clinical applications. J Clin Lab Med lysoplate method of lysozyme determination
1970;75:83-92. with samples dried on filter paper. Clin Chim
17 Schussheim A, Josephson AS, Greenwald RA. Acta 1980;100:175-81.
Serum lysozyme in inflammatory bowel 27 Sen DK, Sarin GS, Mani K, Saha K.
disease. Gastroenterology 1976;70:632-4. Immunoglobulins in tears of normal Indian
18 Falchuk KR, Perrotto JL, Isselbacher KJ. Serum people. Br J Ophthalmol 1976;60:302-4.
lysozyme in Crohns disease, a useful index of 28 Miale JB. Laboratory medicine hematology. 6th
disease activity. Gastroenterology 1975;69:893- ed, St. Louis (MO): CV Mosby Co; 1982. p
6. 722-3.
19 Falchuk KR, Perroto JL, Isselbacher KJ. Serum 29 Daniels JC, Fukushima M, Larson DL, Abston
lysozyme in Crohns disease and ulcerative S, Ritzmann SE. Studies on muramidase
colitis. N Engl J Med 1975;292:395-7. (lysozyme). I. Serum and urine muramidase
20 Peeters TL, Vantrappen G, Geboes K. Serum activity in burned children. Tex Rep Biol Med
lysozyme levels in Crohns disease and 1971;29:13-39.
ulcerative colitis. Gut 1976;17:300-5.
837
Lysozyme
Tables
Methods of Lysozyme Analysis

Method 1: Micrococcus lysodeikticus; agar plate, qualitative, semiquantitative
Principle of analysis: Inhibition of growth; time of lysis of cells in plate proportional to concentration
Comments: Many body fluids; historical
Method 2: Micrococcus lysodeikticus; turbidimetric, photometric
Principle of analysis: Lysozyme dissolves suspended cells. Rate at which cell suspension clears is proportional
to concentration
Comments: Serum, urine, tears; linear to only 15 g/mL
Method 3: Micrococcus lysodeikticus; agarose diffusion
Principle of analysis: Cells immobilized in agarose. Diameter of zone cleared of bacterial suspension varies with
lysozyme concentration
Comments: Serum, urine, tears, many other body fluids; recommended procedure
Method 4: Glycol chitin; viscosimetric
Principle of analysis: Enzyme hydrolyses polysaccharide. Decrease in viscosity proportional to enzyme
concentration
Comments: Serum; not used in clinical studies
Method 5: Transglycosylation; liquid scintillation radioassay
Principle of analysis: 14C-labeled chitobiose forms methylglycosamide in presence of enzyme
Comments: Experimental; laborious; substrate not readily available; not used in clinical studies
Method 6: Bacteriophage conjugate; immunoassay
Principle of analysis: Inactivation of lysozymebacteriophage conjugate with antilysozyme; competition by
sample lysozyme
Comments: Serum, urine, tears, saliva; sensitive, laborious; substrate not readily available; not used in clinical
studies
Method 7: Electrophoresis; immunochemical
Principle of analysis: Electrophoresis of antigen into antibody-containing agarose gel to form precipitate
Comments: Serum, urine; laborious, lacks sensitivity
Method 8: Micrococcus lysodeikticus; electrochemical
Principle of analysis: Release of trimethylammonium ions from cells; measured with a selective electrode
Comments: Experimental; laborious; substrate and instrumentation not readily available; good sensitivity and
precision
Method 9: 3,4-Dinitrophenyl tetra-N-acetyl--chitotetraoside; photometric
Principle of analysis: Chromophoric substrate read at 410 nm
Comments: Serum; lacks sensitivity; substrate not readily available
Principle of analysis: Competitive assay using 125I-Labeled lysozyme
Comments: Serum, urine; reagents not readily available; good sensitivity and precision
Method 11: Nephelometry; immunochemical
Principle of analysis: Light scattering by antigenantibody complex
Comments: Serum, urine; reagents not readily available; good sensitivity and precision
838
Lysozyme
Figures
Lysozyme: Figure 1
Standard curve for agarose gel cell lysis assay. Cleared diameter versus lysozyme concentration. Notice that the x axis is a
three-cycle, logarithmic scale.
organism to 1 L of molten (60 to 70 C) 10 g/L agarose.

Procedure: Measurement of Lysozyme Activity by Discard after use.
Agarose Gel Cell Lysis Assay (Lysozyme Table: 6. Substrate plates. Pour freshly prepared buffer
Reaction conditions) substrate into a Petri dish, 85 mm in diameter, to a depth
of 4 mm. After the agarose solidifies, cut 16 sample
Principle wells 2 mm in diameter in four rows of four, 15 mm
Lysozyme diffuses through an agarose gel containing a
apart. The plates are stable for 2 weeks stored inverted at
suspension of Micrococcus lysodeikticus. A cleared zone
2 to 6 C in plastic bags.
ring of lysis develops in the initially translucent agarose
7. Stock lysozyme solution, 400 g/mL (27.8
gel.
Reagents mol/L). Dissolve 20 mg of grade I, 3crystallized,
1. Sodium phosphate, dibasic, 0.067 mol/L. chicken egg white lysozyme, 40,000 units/mg (Sigma
Dissolve 9.47 g of anhydrous dibasic sodium phosphate Chemical Co.) in 50 mL of 8.5 g/L sodium chloride in
in 1 L of deionized water. Stable for 6 months at 2 to 6 siliconized glassware. Stable at 2 to 6 C for 6 days.
8. Working lysozyme standards, 3, 15, 30, 60,
C.
2. Potassium phosphate, monobasic, 0.067 and 120 g/mL. Prepare a 120 g/mL standard by
mol/L. Dissolve 9.07 g of monobasic potassium diluting 3.0 mL of the stock lysozyme solution to 10.0
phosphate in 1 L of deionized water. Stable for 6 months mL with 8.5 g/L sodium chloride. The remainder of the
at 2 to 6 C. working standards are prepared by serially diluting the
3. Phosphate buffer, 0.067 mol/L, pH 6.3. Mix 120 g/mL standard with 8.5 g/L sodium chloride.
224 mL of 0.067 mol/L dibasic sodium phosphate and Prepare fresh daily.
776 mL of 0.067 mol/L monobasic potassium phosphate.
Check the pH with a pH meter. Stable for 2 months at 2 Assay
to 6 C. Equipment: Enlargerviewer to measure zones of
4. Agarose, 10 g/L. Dissolve 10 g of agarose in 1 clearance.
L of 0.067 mol/L phosphate buffer, pH 6.3, kept at a 1. Allow the necessary number of lysoplates to
temperature of 60 to 70 C, with stirring. Stable for 1 come to room temperature in their plastic bags.
month stored in a glass-stoppered bottle at 2 to 6 C. 2. Allow the samples to come to room
5. Buffered substrate. Prepare a uniform temperature.
suspension of Micrococcus lysodeikticus by mixing 500 3. Fill the sample wells with a volume of 25 L in
mg of dried Micrococcus lysodeikticus (Sigma Chemical as short a time interval as possible. Each filled
Co., St. Louis, MO) in 5 mL of phosphate buffer on a plate should contain the 5 working lysozyme
vortex mixer. Add the uniform suspension of the standards as well as the samples to be assayed.
Apply serum, plasma, and urine undiluted;
839
Lysozyme
dilute tears 1:20 with 0.067 mol/L phosphate Notes

buffer, pH 6.3. 1. A lysoplate test kit is commercially available
4. Carefully cover the plates tightly, and return from Kallestad Laboratories, Inc. (Austin, TX).
them to their plastic bags. 2. The standardization of the lysozyme activity
5. Incubate the plates at room temperature on a determination is a matter of concern. An unexplained
level surface for 12 to 18 h. The time of finding of a specific activity ratio of human
development of the lysoplates is not critical. lysozyme/hen white lysozyme activity between 8:1 and
12:1 for the lysoplate method has been observed[4] For
Calculation turbidimetric methods a ratio of 3:1 to 4:1 has been
1. At the end of the incubation period, measure the reported.[8] Methods for the isolation and purification
cleared-zone ring diameters that have of human lysozyme have been developed.[4,8]
developed to the nearest 0.1 mm with an However, the techniques used are beyond the scope of
enlarger-viewer (available from Kallestad many routine clinical chemistry laboratories. The
Laboratories, Inc., Austin, TX). All standard commercially available lysoplate test kit does contain
and sample ring diameters should be read in the reference materials prepared from pooled human urine.
same order and at the same time intervals that
were required to fill the sample wells.
2. For each lysoplate, plot the concentration of
each working standard on the logarithmic axis
against the corresponding cleared-zone ring
diameters on the linear axis of three-cycle
semilogarithmic graph paper. A typical
standard curve is presented in Lysozyme:
Figure 1.
3. For each sample, calculate the lysozyme
activity from the samples cleared ring diameter
and its intersection with the corresponding
value in micrograms per milliliter from the
reference curve.
4. For samples whose values are higher than 120
g/mL, dilutions in 8.5 g/L sodium chloride
must be made and the test repeated. The
resulting concentration is then multiplied by the
dilution factor. The dilution factor for the initial
determination of tears is 20.
840
Magnesium
Magnesium
Name: Magnesium
Atomic formula: Mg
Merck Index: 5469
Chemical class: Alkaline earth element
Forms: Free Mg2+, protein bound or in inorganic complexes
Principles of Analysis and Current Usage i widely used procedure [1]. Spectrophotometric methods
Magnesium, like calcium, has offered a challenge to for measurement of magnesium in biological fluids
analysts because of the ease of contamination and following deproteinization have been developed. Several
problems with binding to serum proteins. Serum of these methods have been automated and are in
magnesium accounts for only 0.3% of the bodys widespread use today.
magnesium. Approximately 30% to 35% of the total
magnesium ion concentration in serum is protein bound; Gindler and Heth [9] introduced the use of calmagite (1-[1-
the remainder is free, or ionized [1]. hydroxy-4-methyl-2-phenylazo]-2-naphthol-4-sulfonic
acid), a metallochromic dye, for direct determination of
The original methods used for magnesium analysis were magnesium without deproteinization (Table 1, Method 2).
precipitation techniques [2-4]. In these, the magnesium The alkaline, blue-colored working reagent forms a pink
was precipitated from solution in the form of salts such as magnesium-calmagite complex, and the intensity of the
MgNH4PO4. The amount of precipitate, measured color produced is measured at 532 nm. EGTA is used to
2+
gravimetrically or by analysis of phosphorus in the prevent Ca interference, and KCN is used to inhibit
precipitate, was directly related to the magnesium heavy-metal complexes. Polyvinylpyrrolidone and its 9-
concentration. These manually performed precipitation ethylene oxide adduct (Bion PVP, Bion Ne-9) have been
methods are no longer in use. employed to prevent serum proteins from shifting the
absorbance maxima of the Mg-calmagite complex [10].
Compleximetric techniques using EDTA [5] and other The reaction is quite rapid and completed in 60 seconds.
chelators [1,6] have been developed but are rarely used. This method has been automated.
Sensitive fluorometric assays employing calcein [1], o,o-
dihydroxyazobenzene [7], and 8-hydroxyquinoline are also Methylthymol blue, which has been used for calcium
available [1]. analysis [1], has also been employed for the analysis of
magnesium. Again, specific calcium chelators are used to
In the latter method (Table 1, Method 1), a chelated increase the specificity of the analysis. The color is
complex forms, has an excitation maximum at 420 nm, and quantitated bichromatically at 510 and 600 nm (Table 1,
fluoresces at 530 nm. Many fluorescence procedures Method 3).
employ chelating agents such as EGTA to inhibit
2+
interference from Ca . Another direct spectrophotometric method is based on the
reaction between magnesium and the dye Titan yellow
Magnesium can be measured by flame photometry, since (Clayton Yellow, Thiazol yellow, 2,2-[(diazoamino)-di-p-
its band emission peaks at 370 or 383 nm, but this is not a phenylene]bis[6-methyl-7-benzothiazolesulfonic acid]
disodium salt). Under the alkaline conditions of the
reaction, the colloid Mg(OH)2 forms [1,11].
iMagnesium
Previous and current authors of this method: The colloidal particles adsorb the reactive fraction of the
First edition: E. Christis Farrell Titan yellow dye to form a red complex (lake) measured at
540 nm (Table 1, Method 4). To prevent precipitation of
Methods edition: E. Christis Farrell
the dye-colloid complex, stabilizers such as polyvinyl
Second edition: E. Christis Farrell alcohol have been added to the reaction mixture. The
Third edition: Steven C. Kazmierczak stabilizers also increase the intensity of the color [1]. This
Fourth edition: Steven C. Kazmierczak method is rarely used.
Fifth edition: Steven C. Kazmierczak
841
Magnesium
proficiency studies, one must be aware of the use of four

Another method for analysis of magnesium in serum and different systems of units for the reporting of magnesium
urine utilizes the chelating agent chlorophosphonazo III concentrations:
(CPZ) (Table 1, Method 5) [12]. The reaction is initiated
by the introduction of sample to a dye reagent containing 0.5 mmol/L = 1.0 mEq/L = 1.22 mg/dL = 12.2 mg/L
CPZ and EGTA. The CPZ selectively complexes the
magnesium present in the sample while EGTA chelates Methods that relied on manual precipitation techniques
calcium. This first step results in an absorbance decrease at were relatively imprecise and difficult to automate. The
550 nm and an absorbance increase at 675 nm. In the fluorescent techniques, though sensitive, suffer from
second phase of the assay, reagent containing EDTA is problems of both background fluorescence and
added, which removes magnesium from the dye complex fluorescence quenching. Quenching effects in urine
with a resultant absorbance change. Less than 5% samples render fluorescence methods unsuitable for
interference has been reported from samples containing up analysis in urine.
to 200 mg/L calcium [12].
The direct spectrophotometric analyses, adapted to
The majority of magnesium determinations performed in automated techniques, are fairly accurate and precise. The
clinical laboratories measure the concentration of total Titan yellow dye method is adversely affected by the
magnesium present. However, the most important fraction impurity of the dye. Only a portion of the dye is reactive
is the free, or ionized, magnesium, because this is the with magnesium [1,18], and the degree of purity can affect
physiologically active fraction. Ionized magnesium is in the sensitivity and accuracy of the method. Titan yellow
equilibrium with protein bound and organic or inorganic methods appear to show a 10% to 15% positive bias when
complexed forms. The development of synthetic neutral compared to precipitation methods [19]. The precision (%
carriers for determination of ionized magnesium have coefficient of variation [CV]) of these assays for serum
enabled measurement of this form of magnesium. The magnesium values in the normal range is 8% to 14%.
selectivity of these ionophores allows for the measurement
of ionized magnesium even in the presence of The calmagite method has been adapted to many
pathophysiological concentrations of calcium. Intracellular automated analyzers. The reagents for the calmagite
determination of magnesium interference by calcium with reaction are more stable separately and should therefore be
the ionophore is generally ignored, since the ratio of added in two separate steps, or the reagent should be
magnesium to calcium within the cell is approximately mixed immediately before use. It may be advantageous to
1000:1 [13]. take an absorbance reading of the dye plus serum before
adding the alkaline cyanide reagent to develop the color.
Enzyme assays for magnesium have been described (Table Subtracting the blank reading from the final reaction color
1, Method 8) [14,15]. These assays are based on the gives a change of absorbance proportional to magnesium
principle that an Mg-ATP complex must be present to concentrations up to 40 mg/L. In addition, it has been
enable the phosphorylating enzyme to transfer the high- reported that the calmagite will precipitate if left in contact
energy phosphate from the ATP to another molecule. The with plastic lab ware for more than 15 to 20 minutes [20].
reaction rate is followed by monitoring the rate of NADPH
formation at 340 nm. In this assay, it is presumed that the On the basis of precision and accuracy, atomic absorption
is the preferred method.
Mg2+ binds to the ATP in solution and remains bound to
the ADP reaction product.
Specimen
Serum must be separated from the clot as soon as possible,
Reference and Preferred Methods or the level of magnesium will increase because of its
Measurement of magnesium with use of flame atomic elution from red blood cells [1]. Plasma is also an
absorption spectrometry (Table 1, Method 6) has been acceptable specimen. Serum or plasma separated from red
recognized as the traditional reference method for cells or stored in gel-barrier tubes are stable for 7 days at
magnesium [16], but ion chromatography has been room temperature or 4C and indefinitely at 20C [21].
considered as a candidate reference method [17]. Binding of magnesium to proteins and ligands in serum
Magnesium has a strong spectral emission or absorption and plasma is pH dependent. For measurement of ionized
line at 285.2 nm, which can be readily isolated and used to magnesium concentrations, some suggest simultaneous
measure magnesium concentration specifically in measurement of pH to allow adjustment of magnesium
biological fluids. The definitive method of magnesium concentration to a pH of 7.40 [22]. Urine should be
27
analysis is neutron activation with Mg (Table 1, Method acidified to pH 1 with concentrated HCl. If a precipitate
7). forms, shake, mix, acidify, and warm to 60C to redissolve
it [1].
When comparing methods or comparing the results of
842
Magnesium
Interferences loss of magnesium. Often hypomagnesemia is associated

2+
Erythrocytes contain about three times the Mg found in with hypocalcemia. Signs of magnesium deficiency
2+ include increased neuromuscular and cardiac excitability.
plasma, and the Mg is free within the interior of the cell,
Signs of magnesium excess include antagonism of nerve-
not bound to the membrane, as is calcium [1]. Thus in
impulse transmission, with consequent muscle weakness.
general, hemolyzed samples are unacceptable for analysis.
Some of the main causes of hypo- and hypermagnesemia
The presence of calcium gluconate in serum because of
are shown in Table 2.
intravenous administration can cause results of the Titan
yellow method to be 35% low [1]. Silicon, found in
Ionized magnesium represents approximately 70% of total
vacutainer tubes, does not interfere with measurements.
magnesium concentration. Although the proportion may
Blood anticoagulated with citrate, oxalate, or EDTA is
vary from subject to subject, it has been found to be
unacceptable because these compounds can chelate
remarkably consistent in sequential samples from a
magnesium ions. Specimens with increased bilirubin or
particular individual [24]. Ionized magnesium
those that are lipemic have been found not to interfere.
concentrations may show significant changes from normal,
despite no changes in total serum magnesium
Magnesium Reference Interval
concentrations.
Urinary excretion of magnesium is diet dependent, but
normally equals about one third of daily intake [1]. Serum
Magnesium Performance Goals
and cerebrospinal fluid magnesium concentrations are
A review of results of quality-control surveys and linear
shown below. Pregnancy typically results in a slow,
regression analyses of several adaptations of the calmagite
progressive decrease in serum magnesium to about 10%
reaction reveal a rather small, negative proportional bias (r
below nonpregnant values [23]. Aging has not been found
= 0.96) for the calmagite method compared to atomic
to have any appreciable effect on serum magnesium
absorption [25]. The methylthymol blue color reaction
concentrations. There is no appreciable variation in the
usually shows a closer correlation with atomic absorption.
patients magnesium levels throughout the day if exercise
The precision, %CV, of the methylthymol blue (MTB)
is avoided, and there is no difference between fasting and
method for serum magnesium concentrations within the
nonfasting magnesium levels. Newborns have essentially
interval is less than 5%. The coefficient of variation for the
the same serum concentrations of magnesium as adults.
calmagite methods for serum magnesium concentrations
within normal reference limits is less than 8%. Proposed
Magnesium Reference Intervals in different limits for measurement of magnesium suggest a limit for
biological fluids analytical error of 2.4% or less [17]. Clearly, improvement
Sample/Method Mg2+ in the precision of magnesium methods is indicated.
Serum and cerebrospinal fluid
Titan yellow [1] The acceptable Clinical Laboratory Improvement
for serum 18.2329.3 mg/L Amendments of 1988 (CLIA 88) performance criteria for
(1.52.4 mEq/L) magnesium require that laboratories obtain magnesium
Average value [1] values that are within 25% of peer-group laboratories.
for CSF 24.432.0 mg/L Precision goals based on biovariability data indicate a
(no reported sex difference) (2.02.7 mEq/L) maximum standard deviation of 0.02 mg/dL for serum
magnesium [26]. Intra-individual variation obtained in
Atomic absorption 15.825.5 mg/L healthy adults over a 4-week period found a %CV of 3.4
(1.32.6 mEq/L) [27]. The vast majority of magnesium methods used in the
clinical laboratory involve use of calmagite, methylthymol
Urine 7.3 12.2 mg/24 hrs blue, or Xylidyl Blue. Chlorophosphonazo III and
(6.0 10.0 mEq/24 hrs) arsenazol dye are used less frequently. A very small
(3.0 5.0 mmol/24 percent of labs (<1%) utilize an enzymatic method for
h ) magnesium.
References
Interpretation
Magnesium is a cofactor for many intracellular enzymes, 1 Henry RJ, Cannon DC, Winkelman JW. Clinical
including all those that use ATP as a substrate. It is present Chemistry Principles and Techniques. 2nd ed.
in all tissues and bone and is about equally distributed New York: Harper & Row; 1974
between soft tissue and bone. 2 Kramer B, Tisdall FF. A single technique for the
determination of calcium and magnesium in small
A major cause of hypermagnesemia is inappropriate amounts of serum. J Biol Chem. 1921;47:475-
dosage of therapeutic agents such as magnesium sulfate. 481.
Hypomagnesemia is caused by either decreased intake or 3 Briggs AP. A colorimetric method for the
843
Magnesium
determination of small amounts of magnesium. J A candidate reference method for the

Biol Chem. 1922;52:349-355. determination of magnesium in serum. Clin Chem
4 Denis W. The determination of magnesium in Clin Biochem. 1989;27:33-39.
blood, plasma and serum. J Biol Chem. 17 Thienpont LM, Van Nuwenborg JE, Stockl D. Ion
1922;52:411-415. chromatography as reference method for serum
5 Schwartzenbach G, Bidermann W, Banserter F. cations. J Chromatogr. 1997;789:557-568.
Komplex-one VI: neue einfache Titriermethoden 18 King HGC, Pruden G. Component of commercial
zur Bestimmung der Wasserhrte. Helv Chim Titan yellow most reactive towards magnesium:
Acta. 1946;29:811-818. its isolation and use in determining magnesium in
6 Thiers RE. Magnesium (fluorometric). In: Meites silicate minerals. Analyst. 1967;92:83.
S, ed. Standard Methods in Clinical Chemistry. 19 Simonsen DG, Westover LM, Wertman M. The
Vol 5. New York: Academic Press; 1965 determination of serum magnesium by the
7 Brien M, Marshall RT. An automated molybdivanadate method for phosphate. J Biol
fluorometric method for the determination of Chem. 1947;169:39-47.
magnesium in serum and urine using o,o- 20 Khayam-Bashi K, Liu TZ, Walter V.
dihydroxyazobenzene: studies on normal and Measurement of serum magnesium with a
uremic subjects. J Lab Clin Med. 1966;68:701- centrifugal analyzer. Clin Chem. 1977;23:289.
712. 21 Wilding P, Zilva JA, Wilde CE. Transport of
8 Schachter D. The fluorometric estimation of specimens for clinical chemistry analysis. Ann
magnesium in serum and urine. J Lab Clin Med. Clin Biochem. 1977;14:301-306.
1959;54:763-768. 22 Rayana MC, Burnett RW, Covington AK,
9 Gindler EM, Heth DA. Colorimetric DOrazio P, Fogh-Andersen N, Jacobs E et al.
determination with bound Calmagite of Guidelines for sampling, measuring and reporting
magnesium in human blood serum (abstract). Clin ionized magnesium in undiluted serum, plasma or
Chem. 1971;17:663. blood: Internation Federation of Clinical
10 Pierce Magnesium Stat Kit, Pierce Chemical Co., Chemistry and Laboratory Medicine (IFCC):
Rockford, IL; insert revised Feb. 1979. IFCC Scientific Division, Committee on Point of
11 Basinski DH. Magnesium (Titan yellow). In: Care Testing. Clin Chem Lab Med. 2005;43:564-
Meites S, ed. Standard Methods in Clinical 569.
Chemistry. Vol 5. New York: Academic Press; 23 Lind T. Clinical chemistry of pregnancy. Adv Clin
1965:137. Chem. 1980;21:1-24.
12 Dixon DJ, Denton J, Kaufman RA. New 24 Altura BT, Altura BM. Role of magnesium in
magnesium method for the Cobas Chemistry pathophysiological processes and the clinical
Systems using chlorophosphonazo III. Clin Chem. utility of magnesium ion selective electrodes.
1990;36:1068 Scand J Clin Lab Invest. 1996;56(suppl 224):211-
13 Elin RJ. Laboratory tests for the assessment of 221.
magnesium status in humans. Magnes Trace 25 Cohen SA, Daza JE. Calmagite method for
Elem. 1992;10:172-181. determination of serum magnesium modified
14 Tabata M, Kido T, Totani M, Marachi T. Direct [letter]. Clin Chem. 1977;23:289.
spectrophotometry of magnesium in serum after 26 Fraser CG, Peterson PH, Ricos C, Haeckel R.
reaction with hexokinase and glucose-6- Quality specifications. In: Haeckel R, ed.
phosphate dehydrogenase. Clin Chem. Evaluation Methods in Laboratory Medicine. R.
1985;31:703-705. New York: VCH Publishers; 1993:87-99.
15 Wimmer MC, Artiss JD, Zak B. A kinetic 27 Gonzlez-Revaldera J, Garca-Bermejo S,
colorimetric procedure for quantifying Menchn-Herreros A, Fernandez-Rodriguez E.
magnesium in serum. Clin Chem. 1986;32:629- Biological variations of Zn, Cu, and Mg in serum
632. of healthy subjects. Clin Chem. 1990;36:2140-
16 Kulpmann WR, Ruschke D, Buttner J, Paschen K. 2141.
844
Magnesium
Table 1: Methods for Magnesium Analysis

Method 1: Titration of fluorescent complex
a. Mg2+ + calcein Mg-calcein complex (fluorescent; ex, 420 nm;
em, 530 nm)
b. Mg2+ + o,o-dihydroxyazobenzene fluorescent complex
Comments:
a. Stat or small laboratories, difficulty with fluorescence background noise, as well as quenching
b. Rarely used
Method 2: Calmagite
Comments: Widely used in labs without Du Pont aca or atomic absorption; adapted to many automated analyzers;
slight negative bias; positive shift caused by lipemia
Method 3: Methylthymol blue
Principle of analysis: Mg2+ + methylthymol blue complex at 510 and 600 nm
Comments: Du Pont aca, most commonly reported method; good correlation with atomic absorption
Method 4: Titan yellow
_
Principle of analysis: Mg2+ + 2NaOH + Titan yellow PVP Mg(OH) Titan yellow (red lake; colloidal ppt)
2
Comments: Infrequently used; only one eighth sensitivity of calmagite
Method 5: Chlorophosphonazo III
_
Principle of analysis: Mg2+ + CPZ Mg-CPZ complex EDTAMg-EDTA + CPZ
Comments: Frequently used; minimal interference from Ca2+
Method 6: Atomic absorption
Comments: Used in a number of routine laboratories; excellent accuracy, reference method

Method 7: Neutron activation isotope dilution
Principle of analysis: Uses magnesium isotope 27Mg
Comments: Definitive method
Method 8: Enzyme colorimetric
Principle of analysis: Enzyme transfer of phosphate from MgATP to substrate is dependent on Mg2+ concentration
in serum; reaction monitored by coupling to indicator system such as NADPH or peroxidase
Comments: Newly developed; potential as new automated method
Table 2: Magnesium Changes in Disease

High magnesium Low magnesium
Uremia Mg2+-deficiency tetanus
Acute and chronic renal failure Loss of gastrointestinal secretions
Chronic glomerulonephritis Malabsorption syndrome
Addisons disease Diuretic therapy
Intensive antacid therapy Diabetic acidosis
Oxalate poisoning Treatment of diabetic coma
Lupus erythematosus Parenteral fluids (prolonged administration)
Multiple myeloma Hyperparathyroidism
Rickets
845
Magnesium
Procedure: Magnesium by Atomic Absorption temperature. Stable 3 months.

Spectrophotometry 4. Stock magnesium standard, 1000 mg/L (1
Adapted from Diagnotes, Lufkin Medical Laboratories, mg/mL, 82.2 mEq/L). Commercially available;
Minneapolis, MN, 55440. alternatively, dissolve 100 mg of magnesium metal
(commercially available in powder or ribbon form) in 2
Principle mL of 1 mol/L HCl in a 100 mL volumetric flask. After
The magnesium determinations by atomic absorption metal has completely dissolved, dilute to 100 mL with
spectroscopy are based on the fact that atoms of an distilled, deionized water. Store in a polyethylene bottle at
element in the ground, or unexcited, state absorb light of 4C. Stable for 12 months.
the same wavelength as that emitted by the element in the 5. Working standards
excited state. Each element has its own characteristic
absorption or resonance lines, and no two elements are Mg Stock Mg
known to have an identical resonance line. mg/L (mEq/L) (mL)
10 (0.82) 1
To avoid interference by phosphate ions, magnesium in 20 (1.64) 2
serum or urine is determined after sufficient dilution with 30 (2.47) 3
lanthanum oxide solution. 40 (3.28) 4
50 (4.11) 5
Reagents
Use special atomic-absorption glassware and water that are Assay
free of calcium and magnesium contamination.
1. Stock lanthanum, 50 g/L (314 mmol/L) in 4 Equipment: Absorption spectrophotometer with a calcium
mol/L HCl. Weigh 58.64 g of lanthanum oxide (La2O3), magnesium cathode lamp. Follow manufacturers
and wet with about 50 mL of deionized water. Add 250 instructions for proper analytical procedure and for
mL of concentrated HCl slowly and cautiously under hood. establishing a proper airacetylene fuel ratio.
Mix to dissolve powder. Dilute to 1 L with deionized A pipetter dilutor is used to prepare 1:50 dilutions
water. Store in dark at room temperature. Stable 6 months. of samples with the appropriate lanthanum solution.
2. Working lanthanum diluent for serum, 1 g/L Note
(6.28 mmol/L). Dilute 20 mL of stock lanthanum to 1 L The dilution (1:50) of serum or urine prevents protein
with deionized water. Store in dark bottle at room interference.
temperature. Stable 3 months.
3. Working lanthanum diluent for urine, 5 g/L
(31.4 mmol/L). Dilute 100 mL of stock lanthanum to 1 L
with deionized water. Store in dark bottle at room
846
Maternal-Fetal Screening
Peter OLeary and Barry Lewis
Clinical significance: Refer to Chapter 44, Pregnancy, in the 5th edition of Clinical Chemistry:
i 35 years and considered to be of advanced maternal age.
Principles of Analysis and Current Use
However in recent times, many women are choosing to
Screening Overview delay pregnancy, and the proportion of pregnant women
Screening is the systematic application of a test to older than 35 now approaches 20%.
identify asymptomatic subjects at sufficient risk of a
disorder who are likely to benefit from further diagnostic One in every 700 pregnancies has an NTD such as spina
investigation. Common screening practices include bifida or anencephaly, which occur when the fetal brain
looking for hypertension by measurement of blood or spine fail to develop normally during the first 4 weeks
pressure, atherosclerosis by measuring plasma of pregnancy. Spina bifida, which constitutes 50% of
cholesterol concentrations, and cervical cancer by the NTDs, is caused by an incomplete covering of the neural
Papanicolaou screen. The necessary parameters for a tube, leaving an open gap (open NTD) or skin-covered
good screening program [1] are: split (closed NTD) on the spinal cord. While a third of
babies with spina bifida live beyond 10 years, 20% die
(a) The condition is common and well defined in the first year, and the remainder are stillborn or die
(b) Simple tests detect with high specificity and soon after birth. Related defects include anencephaly
sensitivity, so there are few false positives and (small or absent neural tissue in the brain) and
few false negatives encephalocele (a malformation of the brain and skull
(c) The screening tests are cost effective allowing protrusion of brain tissue). Practically all
(d) Diagnostic confirmatory tests are available babies with anencephaly or encephalocele are stillborn
(e) Ethically acceptable options for intervention or or die soon after birth. The majority of NTDs can be
treatment prevented if the mother takes folic acid supplements (0.5
mg daily for 1 month before conception and up to 3
Screening for fetal genetic disorders such as months after conception) or consumes a diet rich in
aneuploidies and neural tube defects (NTD) fits within folate. In 2007, 54 countries had documented national
the screening criteria. We inherit half of our regulations for mandatory wheat flour, and this has been
chromosomes from each parent when a sperm fertilizes associated with a reduction in the prevalence of children
an ovum. The normal human chromosomal complement with spina bifida and other NTDs [2]. About 95% of all
is 46, or 23 pairs (22 pairs of autosomes and 2 sex NTDs occur in families with no history of these defects.
chromosomes), with each parent contributing one
chromosome per pair. The fertilized cell divides to form The birth prevalence [3], including number of live
an embryo and eventually the fetus. Cells with 46 births, stillbirths, and terminations of pregnancy per
chromosomes are called euploid. If the ovum or sperm 10,000 of selected fetal genetic disorders is as follows:
is missing a chromosome (45) or has an extra
chromosome (47), this is referred to as aneuploidy. The Spina bifida 3.39 (Atlanta, USA, 1999-2003)
most common aneuploidies are trisomy of chromosome Trisomy 21 17.44
21 (Down syndrome), trisomy 18, and trisomy 13, 45XO Trisomy 18 4.67
(Turners syndrome, affecting 1 in 2500 girls) and Trisomy 13 1.88
47XXY (Klinefelters syndrome, affecting about 1 in
600 to 800 boys). Most fetuses with any of these Analyte Description
chromosomal abnormalities are usually lost early in In the past decade, several biochemical and ultrasound
pregnancy. markers have been developed as part of a prenatal
screening algorithm for fetal chromosomal and structural
The risk of an adverse pregnancy outcome, such as a abnormalities. The American College of Obstetricians
chromosomal anomaly, increases with maternal age. In and Gynecologists has stated that first-trimester
the 1980s, just 5% of pregnant women were older than screening using measurements of both nuchal
translucency by ultrasound and biochemical markers
provide an effective screening test for Trisomy 21 in the
i
Maternal-Fetal Screening general population [4]. Nuchal translucency (NT) refers
Previous and current authors of this method: to the size of a fluid accumulation at the back of the fetal
First edition: Paul T. Russell neck. An increased measurement of NT is strongly
Methods edition: Not updated associated with an increased risk of Trisomy 21.
Second edition: Paul T. Russell Currently the recommended maternal serum markers for
Third edition: Steven C. Kazmierczak Trisomy 21 [5] in the first trimester include pregnancy-
Fourth edition: Jerzy Stanek, Lawrence A. Kaplan associated placental protein-A (PAPP-A) and free -
Fifth edition: Peter OLeary, Barry Lewis human chorionic gonadotropin (free -hCG). Second
trimester -fetoprotein (AFP), total hCG or free -hCG,
847
and unconjugated estriol (uE3) constitute the so-called

Triple screen. The inclusion of inhibin-A [6] in a four- Inhibin is a glycoprotein consisting of an subunit
test panel is called the Quadruple test. Screening for linked to either a A subunit (inhibin A) or to a B
NTDs by measuring maternal serum AFP can be offered subunit (inhibin B). Serum inhibin A concentrations
in the second trimester either as a stand-alone test or as begin to increase early in pregnancy, with the feto-
part of a Triple or Quadruple screen for fetal aneuploidy. placental unit being the major source. Very high inhibin
A concentrations are found at the end of pregnancy.
AFP ( globulin) is the major blood protein of early Several studies have established that measurements of
fetal life, later to be replaced by albumin. AFP is inhibin A in maternal serum are useful for the evaluation
produced in the yolk sac, gastrointestinal tract, and liver of fetuses with Trisomy 21 [13].
and is excreted into amniotic fluid by the immature fetal
kidneys. AFP passively enters into maternal circulation, PAPP-A measured during the first trimester has been
increasing the maternal serum AFP level. The AFP advocated as one of the best biochemical markers in
concentration peaks in fetal blood between 12 and 14 early screening to test for Trisomy 21. PAPP-A is one of
weeks and in maternal serum between 28 and 32 weeks several glycoprotein proteases produced by placental
of gestation. An abnormal increase in maternal serum trophoblast cells that interacts with insulin-like growth
AFP is the basis of the serum screening test for NTDs. factorbinding protein (IGFBPs), specifically IGFBP-4.
Leakage into the amniotic fluid is negligible in cases of Once IGF-I and IGF-II are released from their IGFBPs,
closed NTDs, and the maternal serum AFP is not they promote fetal growth and development through
increased. metabolic and differentiation pathways [14]. This
provides a biological rationale for PAPP-A influencing
The hormone hCG is a glycoprotein composed of two feto-placental growth and development, and particularly
non-covalently linked polypeptides: the alpha (-hCG) for an association between low PAPP-A and poor
and beta (-hCG) subunits. The individual subunits lack pregnancy outcome. In uncomplicated pregnancies,
biological activity but become active when linked to serum PAPP-A concentrations increase with gestational
form the intact complex. The subunit of hCG contains age until term. Low levels of maternal serum PAPP-A
92 amino acids and is identical to the subunit of the are associated with fetal Trisomy 21.
other glycoproteins, luteinizing hormone (LH), follicle
stimulating hormone (FSH), and thyroid-stimulating Other serum analytes that are currently under
hormone (TSH). The subunit is composed of 145 investigation include ADAM12 [15] and
amino acids, which distinguishes it from the other hyperglycosylated hCG (hCG-H) [16]. Human ADAM
12 (a disintegrin and metalloproteinase) is a placentally-
glycoproteins. In addition to intact hCG, free -hCG
derived glycoprotein that appears to be involved in
subunits, free subunits, and -core subunits are present
growth and differentiation. Similar to PAPP-A, ADAM
in the circulation. Since most immunoassays detect both
12 acts on insulin-like growth factorbinding proteins
intact and free -hCG, they are designated total hCG (IGFBP-3 and IGFBP-5) to regulate the bioavailability
assays. During pregnancy, hCG is produced by the and action of IGF-1 and II. Early in the first trimester (8
trophoblast layer of the placenta. Total hCG and free - to 10 weeks), maternal serum ADAM12 is decreased in
hCG concentrations increase in maternal blood to reach Trisomy 21 pregnancies, relative to unaffected
a peak at 8 to 10 weeks and then decline to 10% to 15% pregnancies, and then ADAM 12 increases relative to
of the peak concentrations and plateau from 20 weeks unaffected pregnancies by 16 weeks of gestation.
until term [7, 8]. The designation -hCG commonly Population modeling showed that a combination of
represents the total and free subunit of hCG. Assays ADAM 12 and PAPP-A measured at 8 to 9 weeks,
which measure either total or free -hCG have been combined with NT and free -hCG measured at 12
used, although free -hCG assays appear to provide weeks, could achieve a detection rate of 97% at a 5%
slightly improved predictive value in first- and second- false-positive rate.
trimester screening programs [9,10].
Estriol is produced by a unique biosynthetic pathway An essential component of successful embryo
during pregnancy, demonstrating the interdependency of implantation is hCG-H; it has been used as an early
fetus, placenta, and mother. Dehydroepiandrosterone marker of early abnormal pregnancy. In Trisomy 21, a
sulfate (DHAS), the major steroid produced by the fetal defect in cytotrophoblast differentiation into
adrenal gland, undergoes 16-hydroxylation (16OH- syncytiotrophoblast cells results in an accumulation of
DHAS) in the fetal liver and adrenal glands. When this cytotrophoblasts, the hCG-H producing cells. Variable
steroid reaches the placenta, the sulfate side chain is results have been reported when using hCG-H to predict
cleaved by placental sulfatase, and 16OH-DHA is Trisomy 21 in the first and second trimesters of
aromatized to form estriol, most of which is conjugated pregnancy. Assays with increased specificity for the
in the maternal liver for excretion through the urine. The hyperglycosylated isoform of hCG-H, with low sialic
concentration of plasma estriol rises 1000-fold during acid content, may achieve Trisomy 21 detection rates of
pregnancy [11]. The protein-bound, uE3, with a half-life 80% in both the first and second trimester [16].
of only 20 min, provides the most specific reflection of
the output of the fetoplacental unit at the time of Brief Description of Pathology
phlebotomy [12].
848
Most NTDs are malformations, although some of them Laboratories have a responsibility for monitoring
are disruptions (e.g., in amniotic band syndrome). They screening-program performance and should maintain
may be closed (e.g., spina bifida occulta) or open (e.g., data on analyte medians, detection rate, screen-positive
complete rachischisis or open encephalocoele). If not rate, maternal age distribution of the screened
lethal (all cephalad NTDs are lethal), they cause population, uptake of screening and prenatal diagnosis
irreversible disability. The maternal serum AFP test tests, and pregnancy outcome data. The level of posttest
detects approximately 80% of NTDs. Increased maternal counseling support needed may vary with the type of
serum AFP can occur for other reasons, such as multiple result and the resources of the referring practitioner to
pregnancy, omphalocele, gastroschisis, placenta accreta, deal with the issues surrounding an abnormal result.
and may also be associated with pregnancy-induced Women receiving an increased-risk test result, in
hypertension, miscarriage, preterm delivery, intrauterine particular, require adequate posttest counseling.
growth restriction, intrauterine fetal death,
oligohydramnios, and abruptio placentae. To increase Maternal screening programs for fetal anomalies
specificity for NTDs, maternal serum AFP is often evolved from the standard practice of offering diagnostic
adjusted for maternal weight, race, multiple gestation, amniocentesis or chorionic villous sampling (CVS) and
insulin-dependent diabetes mellitus, and prior family karyotyping to most women over the age of 35 years,
history of NTDs. based on the known association between advanced
maternal age and increased risk of delivering a baby
Most Trisomy 21 cases are formed spontaneously and with Trisomy 21. However, in 1984, this diagnostic
are noninheritable. They result from meiotic intervention detected only 30% of Downs-affected
nondisjunction. A minority occurs in offspring of pregnancies, because 70% of cases were born to the 95%
carriers of balanced translocations. The rate of the of women younger than 35 years of age. Using first-
nondysjunction forming trisomy 21 increases with trimester markers (NT, PAPP-A, and free -hCG) or
maternal age. Trisomy 21 is associated with fetal second-trimester maternal serum markers in algorithms
developmental defects, especially cardiac and that account for gestational age and maternal age, a risk
neurological abnormalities. for NTD or Trisomy 21 can be calculated for the fetus.
Several commercial software programs are available for
The Screen generating risk estimates. An additional level of
Screening Overview complexity results when several screening markers are
The maternal serum analysis serves only to screen for combined together (for example, maternal age, NT, and
diseasethat is, to identify those pregnancies at serum marker concentrations), each with a different unit
increased risk of the more common fetal anomalies such of measurement and contribution to the ultimate risk.
as Trisomy 21 and NTDs, which might benefit from Maternal age becomes increasingly influential over a
diagnostic testing through chorionic villous sampling or womens reproductive age. A practical way to combine
amniocentesis. Invasive diagnostic tests require several markers concurrently is to convert each
amniocytes or trophoblast tissue to undertake karyotype measurement into a risk estimate and then combine the
or other physicochemical measurements, and these risks into an individual risk for NTD or Trisomy 21. The
procedures are associated with a small risk (0.5 to 1%) program will usually be adjusted so that approximately
of spontaneous fetal loss. 5% of women screened will be identified as increased
risk and offered further diagnostic tests.
To facilitate informed decision making about the options
available, women and their partners should be provided First-trimester screens are performed between 10 and 13
with clear written and verbal information from a suitably weeks of gestation. PAPP-A provides a higher
qualified health professional. The information provided discrimination between affected and unaffected
should be available before pregnancy occurs or as early pregnancies at around 10 to 11 weeks of gestation,
as possible in pregnancy before women are faced with whereas NT is usually more straightforward when
making decisions. Counseling is a necessary prerequisite performed around 12 to 13 weeks of gestation. The
to ensure women have a thorough understanding of the second-trimester screens should be performed between
screening test and its accuracy. Women should be 14 and 22 weeks of gestation. However, because of the
informed of: possible need for repeat sampling and amniocentesis,
The practical aspects of screening, including the times earlier than 22 weeks are preferred. Most
conditions that are being screened laboratories recommend that second trimester screening
The type of tests, the timing of tests, and the be performed at 15 to 18 weeks.
approximate costs involved
The possibility that ultrasound scans may be To enable comparison of results between laboratories,
diagnostic of anomalies other than those for which reference intervals for normal pregnancies are calculated
the screening programs are designed around the medians for each gestational week for each
analyte. The observed results are then divided by the
The probabilistic nature of results, including
appropriate median and reported as the multiple of the
residual risk and the offer of a follow-up diagnostic
median, or MoM. The statistical use of medians, as
test if an increased risk (positive) result is
opposed to mean values, allows for simple manipulation
obtained
of non-Gaussiandistributed data. The median values are
849
calculated for each week of gestation. For a laboratory to which combines first-trimester serum PAPP-A and
have statistically reliable data, at least 100 screening second-trimester Triple or Quadruple screen into a risk-
results for each gestational week is recommended. The calculation algorithm. An alternative strategy is the
most difficult item to obtain is an accurate gestational contingent sequential screen, which uses the results of
age; therefore, a high-level ultrasound is recommended first-trimester screening (nuchal translucency, PAPP-A,
when screening results appear to be positive or and free -hCG) to categorize pregnancies into high,
borderline. moderate, and low risk. Those women at high risk
(greater than 1:30) are offered diagnostic chorionic
The accuracy of screening results is improved when the villus sampling, and those at lowest risk (risk less than
gestational age is calculated by ultrasonography or by 1:1500) are reassured and offered no further testing.
certain last menstrual period. A common cause of false- Those women at moderate risk (between 1:30 and
positive and false-negative results is misdating of the 1:1500) are offered second-trimester screening (AFP,
pregnancy. free -hCG or total hCG, uE3, and inhibin-A), with the
final risk calculated on all markers. The relative
First-Trimester Screening (913 Weeks) performance of the different strategies, each with a
First-trimester screening for Trisomy 21, based on screen-positive rate of 5%, is shown in Table 2.
maternal age, PAPP-A and free -hCG, and Estimates from the two most recent such studies,
measurement of nuchal translucency has been shown to SURUSS, conducted primarily in the United Kingdom,
detect around 85% of affected pregnancies with a false- and FASTER [17, 18], conducted in the United States,
positive rate of approximately 5% [17, 18]. are remarkably similar (Table 2).
Single Screen for NTDs (1420 Weeks) Reference and Preferred Methods
Measurement of only one analyte, maternal AFP, is used A number of analytical techniques for measurement of
as a screen to detect open NTDs, but other fetal AFP, hCG, uE3, and inhibin A have been described. The
pathologies can give raised AFP levels. The screening American College of Medical Genetics has published
cutoff for open NTDs is defined as 2.5 MoM for AFP, technical standards and guidelines for prenatal screening
with elevated values requiring further investigation by for Trisomy 21 [20] and open NTDs [21] covering pre-
ultrasound or other diagnostic tests. analytical, analytical, and results-reporting phases of the
screening program. When changing analytical methods,
Triple Screen for Chromosomal Defects (1420 Weeks) median values must be reestablished using the new
This screen measures maternal serum AFP, total or free method.
-hCG, and uE3, and correcting for the maternal age
(most important risk factor), weight, and race, an overall The U.S. Food and Drug Administration (FDA) has not
estimate of risk for Trisomy 21 can be calculated. Low licensed any kits for Trisomy 21 screening. While many
AFP and uE3 MoMs and high hCG MoMs are kits have Class II approval for other clinical
associated with an increased risk for Trisomy. A Triple applications, the manufacturers cannot make claims
screen can achieve a detection rate of 69% at a 5% false- about screening for Trisomy 21 [20]. On the other hand,
positive rate. In addition to Trisomy 21 detection, other the FDA licenses AFP kits as an aid in the diagnosis of
chromosomal abnormalities may be detected. Maternal open NTDs. They are classified as Class III devices and
serum uE3 is decreased both in fetal trisomy 18 and in approved for the measurement of AFP in maternal serum
Trisomy 21. However, maternal serum hCG is low in and amniotic fluid samples during the second trimester.
trisomy 18. Both immunometric and radioisotope assay formats are
Quadruple Screen for Chromosomal Defects (1420 available to measure AFP in the range 25 to 150 IU/mL
Weeks) [21].
The Quadruple screen is performed exactly like the
Triple screen but with the addition of inhibin A. The Conventionally, AFP assays are calibrated in either mass
multiple-marker combination with the highest screening (ng/mL) or International Units (IU/mL). Most hCG
performance currently available is AFP, uE3, hCG, and assays are calibrated as IU/mL against the 1st
inhibin A, together with maternal age (so-called quad International Reference Preparation (equivalent to the
marker screen). With this combination, a detection rate 3rd International Standard). Inhibin A assays are usually
of 80% at a 5% false-positive rate can be achieved. Like reported as pg/mL and free -hCG are calibrated either
hCG, inhibin A concentrations tend to be increased in as IU/mL or ng/mL.
Trisomy 21 pregnancies. Addition of inhibin A as the
fourth biochemical marker in screening has been Responses to Abnormal Screening Results
suggested to reduce the false-positive screening rate to There are many reasons for increased maternal serum
2% to 3% while increasing the detection rate. AFP results (false-positive values) that are not
associated with NTDs or chromosomal abnormalities.
Another screening approach that has been proposed is The most important are gestational age and fetomaternal
termed integrated screening [19], which uses first- and transfusion.
second-trimester markers to adjust a womans age-
related risk of having a child with Trisomy 21. A In the case where the maternal serum AFP is slightly
variation on this strategy is the serum integrated screen, elevated, the first step is to confirm the gestational age,
850
by ultrasound if possible, and then look for fetal Placental hemorrhage

structural anomalies such as an NTD. Intrauterine fetal demise
Fetuses < 1500 grams
If the maternal serum AFP is increased, an Fetal hydrops
amniocentesis can be considered. Clear AF with no Hemangiomas
blood contamination may be sent to the laboratory for Pregnancy complications (including intrauterine
AFP measurement and cell culture (for chromosomal growth retardation, hypertension,
analysis). If the amniotic fluid AFP concentration is oligohydramnios, placenta previa)
increased, some laboratories will undertake an assay for Fetal malformations such as gastroschisis,
acetylcholinesterase, although this measurement has gastrointestinal atresia, congenital nephrosis
been largely superseded by modern ultrasound Tumors: hepatocellular carcinoma and yolk-sac
techniques capable of detecting open NTDs. If the tumor
cholinesterase test is positive, an open NTD is most Nonneoplastic conditions: ataxia-telangiectasia
likely to be present, and ultrasonography and genetic syndrome, acute viral hepatitis, chronic active
counseling are recommended. hepatitis, cirrhosis of the liver.
A positive maternal serum screen for fetal chromosomal Because of the variety of other conditions that can lead
abnormalities is an indicator for referral to genetic to increased maternal serum AFP, further tests may be
counseling. Even if the estimated risk is considered low, required to exclude other causes of a positive screen for
there is still a chance that the pregnancy may not yield a NTDs [22]. These include testing for cocaine abuse that
normal outcome. The false-negative rate for Trisomy 21 increases risk for abruption, viral titers, and lupus
will be around 1:2000 to 1:3000, and the risk of an anticoagulant measurement to identify autoimmune
adverse outcome with an increased screening risk, even disease that might lead to placental thrombosis.
when Trisomy 21 is excluded by fetal chromosome
analysis, is still around 1:15. Potential adverse outcomes Maternal serum AFP results are adjusted for:
include spontaneous pregnancy loss, poor fetal growth, Maternal weight: obese women have spuriously lower
preterm birth, and neonatal death. median values due to a dilution effect [23]. However,
note that recent studies demonstrate that these women
Specimen have a two- to threefold increased risk of fetal NTD
Serum is the specimen of choice for the initial maternal [24].
screen. Tests for AFP, hCG, uE3, and inhibin A can be
readily performed on serum. Amniotic fluid may be Race: African Americans have higher median
required for positive AFP screen confirmation. AFP and, values than whites.
if necessary, acetylcholinesterase assays may be Twin gestation: Higher maternal serum median
performed on the amniotic fluid. The fluid must be free values are observed. However, amniotic fluid
from fetal blood contamination. For samples suspicious AFP values are not affected. An adjustment is
for blood contamination, a fetal hemoglobin assay may made to the maternal serum AFP by dividing
be performed to determine if this has occurred. Part of the value by 1.5.
the amniotic sample may be used for karyotyping, and Insulin-dependent diabetes mellitus: This
aseptic technique is essential for this type of specimen. population has a 20% lower median value of
Interferences AFP. Therefore, the patient-specific MoM
The usual interferences for each of the analytes as value for a woman with insulin-dependant
described in their respective chapters apply to the diabetes mellitus should be divided by 0.8.
determination of this possibility for each analyte.
However, the purpose of these tests is to obtain a risk Efficacy of Screening
estimate of the potential for a genetic defect. Because of Using a maternal serum cutoff of 2.5 MoM, the
the constantly changing concentrations of these analytes detection rate for anencephaly is expected to be 95% or
during pregnancy, the major error in the risk calculations greater, and 65% to 80% for open spina bifida [25]. In
is the estimate of gestational age [30]. Another possible twin pregnancy, the detection rate is approximately 50%
error is amniotic fluid contamination with fetal blood. In at the 5% false-positive rate (even after correcting for
this case, a fetal hemoglobin measurement may multiple gestation). In general about 1% to 2% of all
determine if this has occurred. women tested undergo amniocentesis. For every 15
women with unexplained increased serum AFP, only
Interpretation one will have a fetus with some form of NTD.
Screening for NTDs Confirmatory Laboratory Testing
There are differences in the upper limits of AFP MoM Amniotic fluid AFP and acetylcholinesterase are
values between whites and African Americans, and confirmatory laboratory tests. A qualitative test for
between patients with and without insulin-dependent acetylcholinesterase may be performed using gel
diabetes mellitus. Most commonly, increased maternal electrophoresis [26]. However, the polyacrylamide gel
serum AFP is not due to an NTD. Maternal serum AFP electrophoresis method is rarely performed because of
concentrations may also be observed in: its complexity and has been replaced by
Fetomaternal transfusion ultrasonography, which is now the definitive test for
851
open NTDs. laboratories can use the screen-positive rate as a

Screening for Chromosomal Defects reflection of the false-positive rate. The screen-positive
Historically, maternal age of 35 years or older at the rate is sensitive to precision and accuracy of the assay,
time of delivery has been used to identify women at the assay drift, and inappropriate median values for the
highest risk of having a child with Trisomy 21, and these unaffected population. According to the CLIA
women have been offered genetic counseling and guidelines, the screen-positive rate, established from 300
diagnostic amniocentesis or chorionic villous sampling. to 500 specimens, should fall within limits obtained by
However, the introduction of maternal serum screening large, experienced screening programs [17, 18]. The
and nuchal translucency measurements have screen-positive rate in the current data set can be
significantly improved the detection rate and decreased compared with the rate observed in data collected over
the screen-positive rate, with a decreased risk of the past 6 months. Any exceptions should be
spontaneous fetal loss. The American College of investigated for shifts in median values, or assay drifts.
Obstetricians and Gynecologists has recommended that
all women should be offered aneuploidy screening Calculation of the median MoM for each analyte should
before 20 weeks of gestation, regardless of maternal age be undertaken at least monthly to determine whether any
[4]. The following patterns occur in cases of significant deviation from 1.00 has occurred. The
chromosomal defects. median of the MoM values used in the risk calculation
may be adjusted for gestational age, maternal weight,
The risk for a woman having a baby with a diabetes, smoking status, and ethnicity and should lie
chromosomal defect is calculated from serum values, within 5% of the target value of 1that is, the median
gestational age, and maternal age. The risk can be adjusted MoM value should lie between 0.95 and 1.05.
expressed at the estimated time of deliver or at the date
the sample is drawn. Approximately 25% of fetuses with References
chromosome defects will spontaneously abort between 1. Wilson JMG, Jungner, G. Principios y metodos del
the date of screening and term [27]. Because the risks examen colectivo para identificar enfermeda.
will be different, the report must clearly state whether Boletin de la Oficina Sanitaria Panamericana
the risk is at term or at the date of screening. Corrections 1968;65:281-393.
of analyte values for weight, race, diabetes, and multiple 2. Morbidity and Mortality Weekly Report.
gestation are made in the same manner as the NTD risk 2008;57(01):8-10. Available at
assessment. For hCG, the effect of weight is many times <http://www.cdc.gov/mmwr/preview/mmwrhtml/m
that of race [23]. For twin gestation, the hCG MoM m5701a4.htm> Accessed June 2008.
value is divided by 2.0. For uE3, the MoM values are 3. International Clearinghouse for birth defects
corrected for weight, and a factor of 1.7 is used to surveillance and research MACDPAR.
correct for twin gestation. The results are expressed as a 2005;Chapter 5, p276. Available at
statistical risk for abnormality, which is based on an age- <http://www.icbdsr.org/filebank/documents/Report2
associated risk. This value is multiplied by a likelihood 005.pdf> Accessed June 2008.
ratio derived from the AFP, hCG and uE3 4. ACOG. Practice bulletin: clinical practice
measurements. It is clear then, that the use of multiple guidelines for obstetrician-gynecologists: No. 77.
analytes increases detection rates of aneuploidies [28]. Obstet Gynecol 2007;109:217-227.
Dual positivity at NTDs/Trisomy 21 screening may 5. Wald N, Kennard A, Hackshaw A, McGuire A.
anticipate numerous, frequently severe, pregnancy Antenatal screening for Downs syndrome. Health
complications [29]. Technol Assess UK 1998;2:1-112.
6. Spencer K, Wallace EM, Ritoe S. Second-trimester
Performance Goals dimeric inhibin-A in Downs syndrome screening.
The overall performance of a screening program is Prenat Diag 1996;16:1101-1110.
determined by the serum analyte assay characteristics, 7. Bateman BG. Vaginal sonography findings and
the risk cutoff level, whether dating was done by hCG dynamics of early intrauterine and tubal
ultrasound or dates, and the distribution of maternal ages pregnancies. Obstet Gynecol 1990;75:421-427.
in the population being screened. The CLIA guidelines 8. Braunstein GD, Rasor J, Adler D, Wade ME. Serum
[31] provide a framework for assessing satisfactory human chorionic gonadotropin levels throughout
performance that incorporate internal and accredited normal pregnancy. Am J Obstet Gynecol
external quality-control monitoring procedures to 1976;126:678-681.
demonstrate satisfactory analytical performance. 9. Palomaki G. A summary analysis of Down
syndrome markers in the late first trimester. Adv
Performance goals for detection rates (DR) and false- Clin Chem 2007;43:177-210.
positive rates (FPR) by screening programs that 10. Sancken U, Bahner D. Comparison of triple-risk
incorporate maternal age and gestational age estimated assessment of fetal trisomy 21, including total
by dates or ultrasound in the risk calculation are shown human choriogonadotropin (hCG) or its free beta-
in Table 5. subunit (free beta hCG). Fetal Diagn Ther
2003;18:122-127.
It is sometimes impractical to determine the detection 11. Klopper A. Criteria for the selection of steroid
rate by monitoring the outcome of all pregnancies, but assays in the assessment of fetoplacental function.
852
In: Klopper A, ed. Plasma Hormone Assays in the 23. Reynolds TM, Vrankren G, Van Neuten J. Weight
Evaluation of Fetal Well-Being. New York: correction of MoM values: which method? J Clin
Churchill Division of Longmens Press; 1976:20-35. Pathol 2006;59:753-758.
12. Distler W, Gabbe SG, Freeman RK, Mestman JH, 24. Werber MM, Louik C, Shapiro S, Mitchell AA. Pre-
Goebelsmann U. Estriol in pregnancy. V. pregnant weight in relation to risk of neural tube
Unconjugated and total plasma estriol in the defects. JAMA 1996;275:1089-1092.
management of pregnant diabetic patients. Am J 25. Knight GJ, Palomaki GE. Epidemiologic monitoring
Obstet Gynecol 1978;130:424-431. of prenatal screening for neural tube defects and
13. Canick JA, MacRae AR. Second-trimester serum Down syndrome. Clin Lab Med 2003;23:531-551.
markers. Seminar Perinatol 2005;29:203-208. 26. Goldfine C, Miller WA, Haddow JE. Amniotic fluid
14. Lawrence JB, Oxvig C, Overgaad MT, Sottrup- gel cholinesterase density ratios in fetal open
Jensen L, Gleich GJ, Hays LG, Yates JR, Conover defects of the neural tube and ventral wall. BJOG
CA. The insulin-like growth factor (IGF)-dependent 1983;90:238-240.
IGF binding protein-4 protease secreted by human 27. Morris JK, Mutton DE, Alberman E. Revised
fibroblasts is pregnancy associated plasma protein- estimates of the maternal-age-specific live birth
A. Proc Natl Acad Sci USA 1999;96:3149-3153. prevalence of Downs syndrome. J Med Screen
15. Spencer K. Aneuploidy screening in the first 2002;9:2-6.
trimester. Am J Med Genet 2007;145:18 -32. 28. Palomaki GE, Knight, GJ, McCarthy JE, Haddow
16. Cole LA. Hyperglycosylated hCG. Placenta JE, Donhowe JM. Maternal serum screening for
2007;28:977-986. Down syndrome in the United States: a 1995
17. Wald NJ, Rodeck C, Hackshaw AK, Walters J, survey. Am J Obstet Gynecol 1997;176:1046-1051.
Chitty L, Mackinson AM. First and second trimester 29. Zanini R, Tarantini M, Cerri V, Jacobello C, Bellotti
antenatal screening for Downs syndrome: the D, Lancetti S, Scalchi S, Groli C, Bianchi UA.
results of the Serum, Urine and Ultrasound Dual positivity for neural tube defects and Down
Screening Study (SURUSS). J Med Screen syndrome at maternal serum screening: gestational
2003;10:56-104. outcome. Fetal Diagn Ther 1998;13:106-110.
18. Malone FD, Canick JA, Ball RH, Nyberg DA, 30. Benn PA, Borgida A, Horne D, Briganti S, Collins
Comstock CH, Bukowski R, Berkowitz RL, Gross R, Rodis JF. Down syndrome and neural tube defect
SJ, Dugoff L, Craigo SD, Timor-Tritsch IE, Carr screening: the value of using gestational age by
SR, Wolfe HM, Dukes K, Bianchi DW, Rudnicka ultrasonography. Am J Obstet Gynecol
AR, Hackshaw AK, Lambert-Messerlian G, Wald 1997;176:1056-1061.
NJ, DAlton ME. First- and Second-Trimester 31. National Committee on Clinical Laboratory
Evaluation of Risk (FASTER) Research Standards. Maternal serum screening: approved
Consortium. First-trimester or second-trimester standard. NCCLS Document I/LA25-A. Wayne,
screening, or both, for Downs syndrome. N Engl J PA: NCCLS; 2004.
Med 2005;353:2001-2011. 32. Haddow JE, Palomaki GE, Knight GJ, Foster DL,
19. Wald NJ, Watt HC, Hackshaw AK. Integrated Neveux NL. Second-trimester screening for Downs
screening for Downs syndrome based on tests syndrome using maternal serum dimeric Inhibin A.
performed during the first and second trimesters. N J Med Screen 1998;5:115-119.
Engl J Med 1999;341:461-467. 33. Summers AM, Langlois S, Wyatt P, Wilson RD.
20. Palomaki GEBL, McDowell GA. Technical Prenatal screening for fetal aneuploidy. J Obstet
standards and guidelines: prenatal screening for Gynaecol Can 2007;29:146-161.
Down syndrome. Genet Med 2005;7:344-354.
21. Bradley LA, Palomaki GE, McDowell GA.
Technical standards and guidelines: prenatal
screening for open neural tube defects. Genet Med
2005;7:355-369.
22. Elias S, Simpson, JL. Maternal Serum Screening for
Fetal Genetic Disorders. New York: Churchill
Livingstone; 1992.
Table 1: Screening Tests, Analytes and Conditions in Maternal-Fetal Screening Programs.
Weeks Analyte Measurement Conditions
Detected
First-trimester 9-13 PAPP-A Correct for maternal weight, adjust for Trisomy 21
maternal Free -hCG or total hCG gestational age. Trisomy 18
serum Trisomy 13
markers Anencephaly
First-trimester 10-13 Nuchal translucency (NT) Determine gestational age by crown-rump Multiple
ultrasound length or biparietal diameter. pregnancy
measurements Some structural
defects
Second trimester 14-20 AFP Correct for maternal weight, adjust for NTDs
853
Free -hCG or total hCG gestational age. Trisomy 21

uE3 Trisomy 18
Inhibin A Trisomy 13
Table 2: Summary Data for First- and Second-Trimester Trisomy 21 Screens from the Prospective SURUSS
[17] and FASTER [18] trials.
Analytes Detection Rate (%) Optimal Time
for 5% (Gestation Weeks)
False-Positive Rate
First trimester NT only 64-70 12-13
Combined NT, PAPP- 82-87 11-13
A, free -hCG
Second trimester Triple screen (AFP, 69 15-17
hCG, uE3)
Quadruple screen 81 15-17
(AFP, hCG, uE3,
inhibin-A)
First and second Serum integrated 85-88 Serum PAPP-A at 11 weeks and
trimester (PAPP-A + Quadruple screen at 15-17 weeks
Quadruple screen)
Integrated (NT, PAPP- 94-96 All results withheld until the second
A + Quadruple trimester
screen)
Contingenta (NT, 88-94 First trimester screen for all with
PAPP-A + Quadruple screen only offered to
Quadruple screen) those with a moderate risk
Sequential 95 Step-wise testing, where 1.5% of
women screened in first trimester
identified at high risk are offered
invasive diagnostic testing; the
remaining women are offered second-
trimester Quadruple screening.
a
NT, PAPP-A and free -hCG are used to stratify a high-risk group (risk greater than 1:30) who are offered invasive
diagnostic testing, whereas women at low risk (risk less than 1:1500) are not offered further testing. Those at
intermediate risk are offered second-trimester Quadruple or Triple analyte screening
854
Table 3: Screening Program Framework for Fetal Anomalies, Including Trisomy 21 and Neural
Tube Defects.
Steps Protocols
Preanalytical Appropriate written patient information about:
Disorders detected by screening
Difference between screening and diagnostic tests
Program performance (detection and false-positive rates)
Risk and benefits of screening
Costs
Consent process for physician to discuss with patient
The next steps if the test is positive
Specimen Laboratory requirements for specimen collection, transport:
Patient demographics
Gestational age
Maternal weight, race, and age
Relevant clinical history of insulin-dependent diabetes, multiple pregnancy,
past history of NTDs or Trisomy 21 fetuses
AFP, hCG, Inhibin A and uE3 are stable in sera stored for several days at
4C to 8C and at 20C for several years; uE3 is not stable in whole
blood, and samples should be centrifuged and separated within 8 hours of
collection; free -hCG is unstable in specimens exposed to high daytime
temperatures or sunlight.
Analysis Result turnaround time
Clear arrangement for transmitting results to physician
Risk cutoffs explicitly defined in terms of detection rate and false-positive
rate
Quality Control Commercial control materials have the advantage of long-term stability for
monitoring assays over at least a year, and there is reduced risk of transmission
of bloodborne infectious agents. Standard within- and between-run assay
performance should be monitored.
Other factors that Gestational age dating method: calculation based on the first day of the last
modify the risk menstrual period, or (preferably) ultrasound measurement of the fetal
estimate crown-rump length (CRL) or biparietal diameter (BPD) are used.
Computer-based risk algorithms are available commercially or developed
in-house but must be validated carefully before introduced into routine
service.
Repeat testing is discouraged as part of screening unless the sample timing
or validity is suspected to be incorrect.
Risk algorithm Patient specific risks are generated from the analytical MoM values
adjusted for maternal weight, race, relevant past medical history. The risk
is calculated from the multiplying the likelihood ratios for each parameter
falling into an affected or unaffected Gaussian distribution and the a-priori
risk for Trisomy 21 based on the maternal age.
Results The patient-specific risk can be reported as risk at term or risk at the
time of screening, noting that approximately 25% of Trisomy 21 fetuses
will spontaneously abort between early second trimester and term.
855
Table 4. Common Maternal Serum Marker Patterns With Fetal Trisomies

Trisomy 21 Trisomy 18
Serum and 45, X and 45, X
marker with hydrops Trisomy 13 without hydrops
AFP low low low
hCG high normal low
uE3 low low low
Table 5: Current Screening Options and Screening Performance Goals.

DR and FPR
Second trimester risk cutoff 1:300 at term [32] Dates Ultrasound
AFP, uE3, hCG 68/5.6 72/5.6
AFP, uE3, hCG, Inhibin-A 73/4.3 76/4.4
First trimester [33]
NT, PAPP-A, free-hCG 83/5.0
DR, Detection rate; FPR, False Positive Rate
Figure 1: Ultrasound image of a fetus showing

the nuchal translucency (NT) measurement
856
Metanephrines, Urine
METANEPHRINES (URINE)
Brett McWhinney
Name: Metanephrines: normetanephrine (NM), 3-o-methyl-norepinephrine;
metanephrine (M), 3-o-methylepinephrine
Clinical significance: Refer to Chapter 51, Adrenal Hormones and Hypertension, in the 5th edition of
Molecular formula: NM, C9H13NO3; M, C10H15NO3
Molecular mass: NM, 183.20 D; M, 197.23 D
Merck Index: NM, 6549; M, 5777
Chemical class: o-Methylated catechol derivative
i
Other methods have been developed in an attempt to
The metanephrines (metanephrine and normetanephrine) increase the analytical sensitivity and specificity of the
are present in very low amounts in normal urine, and as Pisano-based method. These include fluorometric [4-10],
a result, a number of sensitive measuring techniques gas chromatographic [11-14], and high-performance
have been developed. liquid chromatographic (HPLC) methods [15-20]. The
fluorescence methods (Table 1, Method 2) are both
Metanephrines are found in urine both in the free and based on the oxidation and tautomerization of M and
conjugated form, primarily as sulfates. The excretion NM to their fluorescent trihydroxyindole derivatives. It
rate of conjugated urinary normetanephrine and is assumed [9,10] that the oxidation products are the
metanephrine is about ten times that of the free same as those derived from the oxidation of epinephrine
metanephrines [1,2]. The conjugates are hydrolyzed by and norepinephrine in the trihydroxyindole method
being boiled in acid before sample extraction and adrenolutin and noradrenolutin. The fluorescence of the
included in the total metanephrines measured. trihydroxyindole products is measured at 510 to 520 nm
for NM and 520 to 531 nm for M, after excitation at 405
The most widely used method for the measurement of to 410 nm for NM and 415 to 422 nm for M. Many of
total metanephrines is that developed by Pisano in 1960 these methods employ a two-column extraction
[1]. It involves the separation of total metanephrines procedure (Table 1, Method 2a). In this procedure, the
from other urinary components by cation-exchange total hydrolyzed metanephrine fraction is extracted on a
chromatography, followed by oxidation with periodate to strong cation exchanger, and then M and NM are
yield vanillin, which is measured photometrically at 360 fractionated on another ion-exchange column. The
nm (Table 1, Method 1). This method does not separated fractions are oxidized by iodine or periodate to
distinguish between metanephrine (M) and their trihydroxy derivatives. Some procedures [8]
normetanephrine (NM). The same process using a separate the conjugated total metanephrine fraction from
liquid-liquid extraction rather than ion-exchange the free fraction and then hydrolyze before separating M
chromatography can be used to measure urinary 4- and NM in each fraction. This allows fractionation and
hydroxy-3-methoxymandelic acid (HMMA), also known quantitation of conjugated and free metanephrines.
as vanillylmandelic acid (VMA). The extent of Several methods employ only a single cation-exchange
adsorption and elution of metanephrines from cation- chromatography step (Table 1, Method 2b) to purify
exchange columns will vary. Gitlow [2] sought to total metanephrines partially. The amounts of M and NM
overcome this problem by employing two cation- in the eluate are estimated by oxidation of the eluate at
exchange columns for one urine specimen and pooling different pH values [7,9]. At pH 5 to 7, both NM and M
the eluates. will be oxidized, whereas at a lower pH (pH 1.5), only
M will be oxidized.
The Pisano method was modified by Gupta et al. [3] who
replaced cation-exchange chromatography with The gas-chromatography (GC) techniques (Table 1,
differential solvent extraction. An internal standard was Method 3) typically require either column
incorporated to correct for procedural losses. chromatography [11] or a liquid-liquid extraction [12] to
Interference from some related compounds was extract other catecholamines differentially from the
negligible. metanephrines. The metanephrines can then be
converted to vanillin by periodate oxidation. The vanillin
i is extracted and converted to its volatile silyl derivative
Urine Metanephrines
by N,o-bis(trimethylsilyl) acetamide [12]. The
derivatized compounds are separated on 3% OV-1
columns, and the vanillin derivative is detected by flame
Methods edition: Charles Sohr, Lawrence A. Kaplan
ionization. Alternatively, the trifluoroacetyl derivatives
of NM and M can be prepared, separated on 3% QF-
1:2% XF-1105 columns, and quantitated using an
Fourth edition: Charles Sohr
857
electron-capture detector [11]. A mass spectrometer has ensure minimum background fluorescence from
also been used as a detector [13,14]. detergents and other chemicals.
HPLC separation (Table 1, Method 4) and quantitation The gas chromatographic (GC) methods have the
of total metanephrines (conjugated and free) is usually capability for accurate analysis but require several
performed after acid hydrolysis to convert all the purification steps and derivatization. The GC-mass
metanephrines into the free form. Following this, the spectrometers are highly specific and sensitive but
metanephrines are partially purified by loading the require equipment that is not available in most
hydrolyzed urine on a cation-exchange column and, after laboratories.
washing to remove catecholamines, eluting with
ammonium hydroxide. This eluate is then transferred to The HPLC techniques are becoming more popular
an anion-exchange column for further cleanup. The because of their ability to separate M and NM quickly
metanephrines are then eluted with acetate buffer and (less than 15 and up to 30 min) with almost no
can be analyzed directly by HPLC. The mobile phase interference from any drug or natural biochemical. The
usually consists of a phosphate buffer, ion-pairing sensitivity of the HPLC methods with electrochemical
reagent octane sulfonic acid, and a small amount of 7.5% detection is in the nanogram range with between-run
acetonitrile. The most commonly used detector has been coefficients of variation reported to be less than 10%.
an electrochemical detector [17-20], but fluorescence Because the HPLC analysis provides highly specific
detection has also been employed [15,16]. measurements of both M and NM, and HPLC equipment
Standardization can be achieved by direct, external is rapidly becoming routine equipment in many clinical
standardization or by internal standardization. Internal chemistry laboratories, it is the recommended method
standards employed include 3-methoxy-4- for the analysis of urinary metanephrines.
hydroxybenzylamine [20] and 3-hydroxy-4-
methoxyphenethylamine [19]. Specimen
The recommended specimen is a complete 24-hour urine
Immunoassays have been developed for measurement of collection saved in a container containing 25 mL of
metanephrines (Table 1, Method 5). However, because concentrated HCl. Boric acid is not an acceptable
of their small molecular size, and because the preservative. The specimen should be refrigerated during
metanephrines lack sufficient immunogenic properties, collection. Properly collected urine (pH < 3) can be
the development of specific antibodies has been difficult stored at 4C to 8C for weeks and frozen at 20C for
[21,22]. One technique used to counteract this limitation many months.
is to use the N-acetylated derivatives of metanephrines.
By reaction with Bolton-Hunter reagent [3-(p- Interferences
hydroxyphenyl)propionic acid N-hydroxysuccinimide Interferences with the colorimetric procedure can result
ester], both metanephrines are readily converted to their from radiopaque x-ray contrast dye, the measurement of
N-acylhemisuccinates. These acetylated derivatives have other phenoxy acids, and certain drugs, such as
excellent immunogenic properties and allow for the propranolol and the triamterene diuretics. The
production of specific antibodies. fluorometric assay is subject to interference from -
methyldopa.
There is no accepted reference method for the Metanephrines Reference Interval
measurement of urinary metanephrines. Using the spectrophotometric assay, less than 1 mg of
total metanephrines are excreted in 24 hours.
The colorimetric assay lacks specificity, giving rise to
both false-positive and false-negative results [2,23]. In The ranges listed below indicate that significant
addition, the method is very tedious, requiring excellent differences may be obtained in laboratories, even when
control over the pH of the hydrolysis step and over the using similar methods. There are no reported sex or race
conditions for oxidizing M and NM to vanillin. The differences for the urinary excretion of metanephrines.
method also lacks sensitivity for metanephrine levels
found in a healthy population (<500 g/L). However, the Interpretation
method is technically fairly simple and can easily be The three catecholamines of physiological importance in
adopted by most laboratories. It may be useful in a the human are epinephrine (adrenaline), norepinephrine
population-screening situation. (noradrenaline), and dopamine (3-hydroxytyramine).
The important metabolites are 4-hydroxy-3-
The fluorometric methods are quite sensitive (0.25 to methoxymandelic acid (HMMA or VMA), homovanillic
0.35 g) but require an extensive number of acid (HVA), metanephrine, and normetanephrine
chromatographic steps to ensure specific and accurate
results. Excellent care must be taken with glassware to
858
Total Urinary Metanephrine Excretion
Age (years)
01 (HPLC) <0.87
Metanephrine <1.84
Normetanephrine
12 (HPLC) <1.04
(g/mg creatinine) <1.38creatinine)
(g/mg
24 (HPLC) <0.44 <0.81
47 (HPLC) <0.26 <0.67
7Adult (HPLC) <0.17 <0.41
g/24 hours g/24 hours
Adult (HPLC) [23,24] <395 <930
Adult (HPLC) [25] <300 <355
Adult (GC) [25] <260 <340
Adult (GC-MS) [26] <255 <620
Adult (ELISA) [22] <275 <320
concentration of 153 pmol/L measured by HPLC. For
normetanephrines at 671 pmol/L measured in plasma by
Interpretation HPLC, intra- and interassay coefficients of 4.7% and
The three catecholamines of physiological importance in 12%, respectively, have been reported [28]. The
the human are epinephrine (adrenaline), norepinephrine intraindividual variation of metanephrine and
(noradrenaline), and dopamine (3-hydroxytyramine). normetanephrine in urine of patients with
The important metabolites are 4-hydroxy-3- pheochromocytoma was found to be 15% (range 1% to
methoxymandelic acid (HMMA or VMA), homovanillic 34%) and 17% (range 0% to 35%) [29].
acid (HVA), metanephrine, and normetanephrine.
References
Testing for the catecholamines and their metabolites is 1 Pisano JJ. A simple analysis for
important in the diagnosis of the cause and extent of normetanephrine and metanephrine in urine.
certain neurological and endocrinological disorders such Clin Chim Acta 1960;5:406-414.
as orthostatic hypotension and pheochromocytoma. 2 Gitlow SE, Mendlowitz M, Bertani LM. The
biochemical technique for detecting and
Pheochromocytomas exhibit excessive storage and establishing the presence of
release of catecholamines. The majority of pheochromocytoma. Am J Cardiol
pheochromocytomas secrete norepinephrine, though 1970;26:270.
secretion of large amounts of epinephrine and dopamine 3 Gupta RN, Price D, Kearne, PM. Modified
is not uncommon. In some cases, epinephrine alone is Pisano method for estimating urinary
secreted; in these cases, the tumor is almost always metanephrines. Clin Chem 1973;19:611.
localized in the adrenal medulla. 4 Weil-Malherbe H, Smith ERB. The estimation
of metanephrine, normetanephrine, and 3,4-
The recommended screening tests for dihydroxymandelic acid in urine. Pharmacol
pheochromocytoma are urinary HMMA (VMA) and Rev 1966;18:331-341.
fractionated metanephrines. The positive predictive 5 Kahane Z, Vestergaard P. Fluorometric assay
value of elevated urinary metanephrines for for metanephrine and normetanephrine in urine.
pheochromocytoma is approximately 95%. When J Lab Clin Med 1967;70:333-342.
combined with an elevated VMA excretion rate, the 6 Kahane Z, Vestergaard P. Column
positive predictive value is approximately 100% (click chromatographic separation and fluorometric
here for VMA chapter). Because of paroxysmal estimation of metanephrine and
secretion of catecholamines and their metabolites from normetanephrine in urine. Clin Chim Acta
tumors, borderline results may be obtained. The analysis 1960;25:453-458.
should be repeated on a new specimen, and perhaps 7 Anton AH, Sayer DF. Distribution of
fractionated catecholamine analysis should also be metanephrine and normetanephrine in various
performed. animals and their analysis in biologic material. J
Pharmacol Exp Ther 1966;153:15-29.
Metanephrines Performance Goals 8 Taniguchi K, Kakimoto Y, Armstrong M.
Long-term precisions (% coefficient of variation [CV]) Quantitative determination of metanephrine and
for metanephrines and normetanephrines measured in normetanephrine in urine. J Lab Clin Med
urine by HPLC were found to be 5.4% at a concentration 1964;64:469-484.
of 1.01 mol/L and 3.8% at a concentration of 4.35 9 Bigelow LB, Weil-Malherbe H. A simplified
mol/L hour, respectively [27]. For metanephrines method for the differential estimation of
measured in plasma, intra- and interassay CVs of 7.0% metanephrine and normetanephrine in urine.
and 12%, respectively, have been reported at a Anal Biochem 1968;26:92-103.
859
10 Dalmaz Y, Peyrin L. Specific ion-exchange 20 Orsulak PJ, Kizuka P, Grab E, Schildkrant JJ.
chromatography and fluorometric assay for Determination of urinary normetanephrine and
urinary 3-o-methyldopamine. J Chromatogr metanephrine by radial-compression liquid
1976;116:379-394. chromatography and electrochemical detection.
11 Bertani LM, Dziedzic SW, Clarke, DD, Gitlow Clin Chem 1983;29:305-309.
SE. A gas chromatographic method for the 21 Wassell J, Reed P, Kane J, Weinkove C.
separation and quantitation of normetanephrine Freedom from drug interference in new
and metanephrine in human urine. Clin Chim immunoassays for urinary catecholamines and
Acta 1970;30:227-233. metanephrines. Clin Chem 1999;45:2216-2223.
12 Calseyde JF, van de Scholtis RJ, Schmidt NA, 22 Wolthers BG, Kema IP, Volmer M, Wesemann
Leyten CJ. Gas chromatography in the R, Westermann J, Manz B. Evaluation of
estimation of urinary metanephrines and VMA. urinary metanephrine and normetanephrine
Clin Chim Acta 1971;32:361-366. enzyme immunoassay (ELISA) kits by
13 Muskiet FA, Thomasson CG, Gerding AM, comparison with isotope dilution mass
Fremouw-Ottevangers DC, Nagel GT, Wolthers spectrophotometry. Clin Chem 1997;43:114-
BG. Determination of catecholamine and their 120.
3-o-methylated metabolites in urine by mass 23 Amery A, Conway J. A critical review of
fragmentography with use of deuterated internal diagnostic tests for pheochromocytoma. Am
standards. Clin Chem 1979;25:453-460. Heart J 1966;73:129-133.
14 Cenfell C, Binder SR, Khayam-Bashi H. 24 Graham PE, Smythe GA, Edwards GA, Lazarus
Quantitation of urinary normetanephrine and L. Laboratory diagnosis of pheochromocytoma:
metanephrine by reversed-phase extraction and which analytes should we measure? Ann Clin
mass-fragmentography analysis. Clin Chem Biochem 1993;30:129-134.
1982;28:25-28. 25 Rosano TG, Swift TA, Hayes LW. Advances in
15 Mell LO. Reverse-phase high pressure liquid catecholamine and metabolite measurements for
chromatography of urinary catecholamines and diagnosis of pheochromocytoma. Clin Chem
related acidic metabolites. In: Hank GL, ed. 1991;37:1854-1867.
Biological/Biomedical Applications of Liquid 26 Kema IP, Meiborg G, Nagel GT, Stob GJ,
Chromatography. Chromatography Science Muskiet FA. Isotope dilution ammonia-
Series 10. New York: Marcel Dekker; chemical ionization mass fragmentographic
1979:619-631. analysis of urinary 3-o-methylated
16 Freeman VS, Kiser EJ. Urinary total catecholamine metabolites. Rapid sample
metanephrines by high-pressure liquid cleanup by derivatisation and extraction of
chromatography with fluorometric detection lyophilised samples. J Chromatogr. B. Biomed
[abstract]. Clin Chem 1984;30:1045. Appl 1993;671:181-189.
17 Shoup RE, Kissinger PT. Determination of 27 Atef H, Prosser C, LeGatt D, Froment S,
urinary normetanephrine, metanephrine, and 3- Cembrowski GS. Medically allowable
methoxytyramine by liquid chromatography, imprecision in measuring urinary
with amperometric detection. Clin Chem metanephrines. Clin Biochem 2001;34:159-160.
1977;23:1268-1274. 28 Roden M, Raffesberg W, Raber W, Bernroider
18 Bertani-Dziedzic LM, Krstulovic AM, Dziedzic E, Niederle B, Waldhusl W, Gasic S.
SW, Gitlow SE, Cerqueira S. Analysis of Quantification of unconjugated metanephrines
urinary metanephrines by overset-phase high- in human plasma without interference by
performance liquid chromatography and acetaminophen. Clin Chem 2001;47:1061-1067.
electrochemical detection. Clin Chim Acta 29 Gerlo EAM, Stevens C. Urinary and plasma
1981;110:1-8. catecholamines and urine catecholamine
19 Binder SR, Sivorinovsky G. HPLC metabolites in pheochromocytoma: diagnostic
measurement of urinary metanephrines using value in 19 cases. Clin Chem 1994;40:250-256.
ion-moderated partition
chromatography[abstract]. Clin Chem
1984;30:1045.
Table 1: Methods of Analysis of Urinary Metanephrines

Method 1: Spectrophotometry
Principle of analysis: Metanephrines, separated by cation-exchange chromatography, are oxidized to vanillin,
which is measured at 360 nm. Measures total metanephrines.
Comments: Common; poor specificity and sensitivity; very laborious
Method 2: Fluorometry
Principle of analysis: Oxidation of metanephrines partially purified by cation-exchange chromatography to
trihydroxyindole derivatives; quantitation of fluorescence at 510 to 531 nm after excitation at 405 to 422 nm
Comments: Frequent; very good sensitivity and, with column separation, good specificity
Method 2a: Fluorometrytwo-column
Principle of analysis: Total metanephrines separated by cation-exchange chromatography are fractionated into
860
normetanephrine (NM) and metanephrine (M) before the oxidation reaction and fluorometric analysis.
Comments: Direct quantitation of NM and M; has also been used to estimate conjugated and free fractions
Method 2b: Fluorometrydifferential pH
Principle of analysis: Total metanephrines separated by cation-exchange chromatography are oxidized to vanillin
at pH 5 to 7 and pH 1.5. At high pH both NM and M are oxidized; at low pH only M is oxidized.
Comments: Same as for 2a
Method 3: Gas chromatography
Principle of analysis: Partially purified metanephrines are oxidized to vanillin, which is derivatized; derivatives
separated by gas chromatography and vanillin peak are detected by flame ionization. Measures total
metanephrines.
Comments: Rare; good sensitivity and specificity; fairly labor intensive
Method 4: High-performance liquid chromatography
Principle of analysis: Metanephrines are converted to free form by acid hydrolysis and partially purified by
cation-exchange/anion-exchange chromatography. Separation into NM and M fractions by reversed-phase HPLC,
with detection by fluorescence or electrochemical detectors.
Comments: Common; excellent sensitivity and very good specificity; reasonably rapid; preferred method
Method 5: Immunoassay (RIA or ELISA)
Principle of analysis: Antibodies produced against N-acetylated derivatives of metanephrine or normetanephrine
by coupling to bovine serum albumin as hapten
Comments: Requires overnight incubation
2. 70 mM Ammonium Pentaborate Buffer
Procedure: Fractionated Urinary Metanephrines by Ammonium Pentaborate. Store at room
High-Performance Liquid Chromatography (HPLC) temperature.
Ethylenediaminetetra acetic acid di-sodium
salt
Principle a) Dissolve 40 g ammonium
Reverse-phase isocratic high-performance liquid pentaborate and 1.0 g EDTA in 900 mL of
chromatography (HPLC) coupled with electrochemical
water.
detection (ECD) is utilized to determine the
b) Make up to 1000 mL.
concentration of urinary metadrenalines. Metadrenalines c) Store at room temperature.
are also referred to as metanephrines. Sample
purification is a two-step procedure: cation exchange and
3. 2 M Ammonium Hydroxide (Make fresh
anion-exchange extraction. Urinary metanephrines are weekly.)
all converted to the free, unconjugated form by acid Ammonia Solution
hydrolysis. The samples are then passed through Highly Corrosive. Use gloves when
disposable columns containing cation-exchange resin, handling in fume cupboard.
and the resin-bound metanephrines are then eluted onto a) Add 67.5 mL of ammonia solution to
anion-exchange columns. The columns are washed, and 400 mL water.
the metanephrines eluted with acetate buffer. A small b) Make up to 500 mL.
volume of this eluate is injected onto the isocratic HPLC c) Store at room temperature.
system. An Atlantis C18 reverse-phase column separates
the individual metanephrine peaks and the internal 4. 1 M Hydrochloric Acid.
standard peak (IS). Peak resolution is improved by the Concentrated HCl (10 M)
addition of an ion-pairing reagent, 1-octane sulfonic Highly Corrosive. Use gloves when
acid, to the mobile phase. The metanephrines (METS) handling.
are quantitated via ECD by being oxidized to their a) Add 100 mL concentrated HCl to
corresponding anthraquinones (i.e., the oxidation of the 800 mL of water.
phenolic group) at a fixed voltage potential, the resulting b) Make up to 1000 mL.
flow of electrons being measured as current. The electric c) Store at room temperature.
current produced is proportional to the molar
concentration of the METS over a wide linear range. 5. 1 M Acetic Acid.
Concentrated CH3COOH (17 M)
Reagents Highly Corrosive. Use gloves when
Note: All water used is deionized (18 megohm handling.
resistivity) a) Add 15 mL concentrated acetic acid
to 200 mL of water.
1. 0.05 M HCl b) Make up to 250 mL.
HCl (Assay 31.5-33.0%) (10M) MW 36.5 c) Store at room temperature.
Highly Corrosive. Use gloves when
handling. 6. 0.2 M Ammonium Acetate Buffer pH
a) Add 5 mL concentrated HCl to 950 6.0 (Urine Metanephrines Elution Buffer)
mL water. Make up to 1000 mL.
Ammonium Acetate. Store at 4C.
b) Store at room temperature.
861
a) Dissolve 15.4 g of ammonium Stock Internal Standard 500 M: MW 18906.

acetate in 950 mL of water. a) Weigh out 0.10 g of 4-hydroxy-3-
b) Adjust the pH to 6.00 with 1 M methoxybenzylamine.
acetic acid. b) Make up to 100 mL of 0.05 M HCl.
c) Make up to 1000 mL with water. c) Store frozen for up to 6 months.
d) Store at 4C. Working Internal Standard 2.5 M:
a) Dilute stock internal standard 1:200
7. Mobile Phase with 0.05 M HCl.
(A) Stock Mobile Phase Solution b) Store at 4C.
Sodium dihydrogen orthophosphate
dehydrate (AR Grade). MW 156.01. 2. Aqueous Metanephrine Standards:
Final concentration 0.8 M. Stock Standards (500 M solutions):
Ethylenediaminetetra acetic acid di-sodium Metanephrine MW 233.7; Sigma #M-8625
salt. Final concentration 134 M. i) Dissolve 0.1168 g in 1000 mL 0.05 M
1-Octanesulphonic acid sodium salt. MW HCl.
216.3. Final concentration 21 mM. ii) Store at 20C.
All reagents stored at room temperature.
Normetanephrine MW 219.7; Sigma #N-7127
a) Dissolve in 1500 mL of water: 248 g i) Dissolve 0.1098 g in 1000 mL 0.05 M
sodium dihydrogen orthophosphate (OSA) HCl.
b) 0.10 g EDTA 9.0 g OSA4C. ii) Store at 20C.
Expires in 6 months.
3-Methoxytyramine MW 203.7, Sigma #M4251
i) Dissolve 0.1019 g in 1000 mL 0.05 M
CALIBRATION STANDARDS
HCL.
LOW MEDIUM HIGH
ii) Store at 20C.
Vol. Conc. Vol. Conc. Vol. Conc.
(L) (M) (L) (M) (L) (M) 3. Calibration Standards:
Normetanephrine 400 0.8 2000 4.0 4000 8.0 a) Prepare three calibration
Metanephrine 250 0.5 1250 2.5 2500 5.0 standards from the Stock Standards as per the
3- 400 0.8 2000 4.0 4000 8.0 table below, making up each standard to 250 mL
Methoxytyramine with 0.05 M HCl.
b) Store 1.0-mL aliquots at -20C.
c) Make up to 2 liters with water.Store at
(B) Working Mobile Phase Quality Control
Biorad Lyphochek 1 and Lyphochek 2 are the
Sodium dihydrogen orthophosphate (AR controls used. Store at 4C.
Grade). Final concentration 0.08 M.
Acetonitrile HPLC grade. Final Prepare fresh weekly in 10 mL of 0.05 M HCl.
concentration 8.5% Place on a roller mixer for 10 min prior to use.
EDTA. Final concentration 13 M Store at 4C while in use. Discard after 7 days.
OSA. Final concentration 2.1 mM
Equipment
a) Add 200 mL of mobile-phase stock
solution and 170 mL of acetonitrile to 1.
a 2-L volumetric flask. a) pH Meter
Ross Combination pH electrode, 0-14 pH, Model
b) Add 1500 mL of water. 815600
c) Adjust pH to between 3.95 and 4.05 b) BioRad Cation-exchange resin

with phosphoric acid. columns; BioRad catalog number 1956017 (100
columns/kit).
d) Make up to 2 liters with water. Store at 4C.
e) Filter and de-gas under vacuum c) BioRad Anion-exchange resin columns;

through a 0.45-m filter. BioRad catalog number 1956018 (100 columns/kit).
Store at 4C.
f) Store at 4C. d) Waters 24-port extraction manifold attached
to a vacuum pump.
2. HPLC SYSTEM:
Standard Solutions
1. Internal Standard 4-Hydroxy-3- a) Pump
Methoxybenzylamine (HMBA):
862
Waters Model 2690 solvent delivery system. 8. Wash each column with 2 10 mL
Flow rate 1.5 mL/min. deionized water. Discard eluate.
b) Autosampler 9. Place numbered anion-exchange

Waters Model 2690 integrated autosampler. columns in plastic rack under the
appropriate cation-exchange column.
c) Column
Waters stainless steel Atlantis C18 10. Pipette 8.0 mL of 2 M ammonium
Cartridge Column 150 mm 4.6 mm hydroxide into each upper cation-
with 5 m packing. Part No. 186001344. exchange column, and allow to drain
completely through both columns. Apply
d) Data Manager a short vacuum to remove all the 2 M
Waters Empower Build No. 2154 ammonium hydroxide from the cation-
Chromatography Manager Project Metanephrines, Urine exchange column. Remove and save the
cation-exchange columns for washing.
e) Detector:
ESA Coulochem II 5200A Coulometric Detector 11. Attach the anion columns to the 24-port
(ECD) with a Model 5021 Conditioning Cell and a manifold in the positions previously
Model 5011 High Sensitivity Cell. occupied by its corresponding cation
Method 1Settings as shown in table below: column. Rinse the anion-exchange
N.B. These values must be determined for each columns with 2 10 mL deionized water.
cell and will vary depending on the age of cells AECD SETTINGS
and mobile phase. pGuard Cell
pCell potential +500 mV
l
PROCEDURE Channel 1
y
Cell potential +350 mV
Sample Extraction Ion Exchange Full scale range 500 nA
a
1. Aliquot 0.5 mL Standards Mets and 3 Filter 5 sec
MT Standards (3 levels) (stored in 2.25- sSignal output voltage 1 V
mL aliquots), QC (2 levels), and hBaseline offset 0%
patients urine into numbered 16 150 oChannel 2
mm glass tubes. rCell potential 250 mV
tFull scale range 500 nA
2. Add 100 L of internal standard (10 M Filter 5 sec
HMBA) and 0.25 mL of 1.0 M vSignal output voltage 1 V
hydrochloric acid to each tube, and cap aBaseline offset 0%
firmly. c
uum to remove all the water from the
3. Place the tubes in heater block set at anion column.
120C under a fume hood for 45 min.
12. Place appropriately numbered 5 LP tubes
4. Remove the tubes, and allow to cool. into the plastic rack, and add 400 L of 1
M acetic acid. Place the rack and tubes
5. Bio-Rad METS cation columns are under the corresponding columns. Add 5
removed from the fridge. The caps and mL of 0.2 M ammonium acetate buffer
tails are removed and the columns (metanephrine elution buffer) to each
placed on the 24-port vacuum extraction column, and allow metanephrines to
device to drain. elute. If a column does not drain, apply a
short vacuum to start the flow.
6. Transfer the samples to 25 mL LP, add
6.0 mL 70 mM ammonium pentaborate 13. Transfer the contents of each tube to
buffer, and mix. Adjust the pH to 6.5 numbered glass autosampler vials, and
0.10 using NaOH or HCl (5 M &1 M). inject 60 L.
7. Pour each sample onto separate labeled 5.3 System Settings Summary
cation columns which have been placed
onto the 24-port extraction manifold. Injection volume: 60 L
Allow to drain. If a column does not Flow rate: 1.5 mL/min
drain, apply a short vacuum to start the Run time: 25 mins
flow. Discard eluate. ECD Method Number: 1
Recycle mobile phase: yes
863
CALCULATIONS Where: PH = peak height of chromatographic

peak
Empower measures the peak heights, calculates
UNK = analyte of interest in unknown
the ratios of analytes to the IS, and constructs a
STD = analyte of interest in standard
standard curve of this ratio versus concentration
IS = internal standard
in mol/L, mol/day and Creat ratios. The
concentrations of the unknowns are then
Multiply the results obtained by the 24-hour urine
determined from this standard curve, using
volume to obtain the output as mol/24 hours.
their calculated peak height ratios. Check the
internal standard peak heights. Variations of
10% to 20 % are to be expected, but low Notes
recoveries of the internal standard suggest some 1. No interference has been seen in these HPLC
problem with the sample clean-up. systems, and no dietary restrictions are
necessary for persons to be tested.
For manually calculating the concentration of 2. Optimal working electrode potential can vary with
an analyte, use the following formula: time and should be checked at varying
intervals.
PH UNK PH ISSTD 3. Limit of Detection = 0.05 mol/L
Analyte concentration (mol/L)
PH STD PH ISUNK
864
IS
400.00
300.00
Normetadrenaline 3 Methoxy Tyramine
200.00
Metadrenaline
mV100.00
0.00
-100.00
-200.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00
Minutes
Figure 1: Ion-pair high-performance liquid chromatographic (HPLC) analysis by method described in text; electrochemical
detector (ECD) response versus elution time. A, Normetanephrine (6.15 minutes), metanephrine (8.55 minutes), internal
standard (12.5 minutes) and 3 methoxy tyramine (21 minutes).
865
Methotrexate
Methotrexate
Michael A. Pesce
Name: Methotrexate
Clinical significance: Refer to Chapter 53, Neoplasia, in the 5th edition of Clinical Chemistry:
Merck Index: 5861
Structure:
Methotrexate
NH2 CH3 CO2H
N CH2N CONHCH(CH2)2CO2H
N
N N
Chemical class: Pteridine derivative, a structural analog of folic acid
i
The fluorometric assay (Table 2, Method 1) involves
Methotrexate is given in both high and low doses as a protein precipitation with trichloroacetic acid and
cancer chemotherapeutic agent [1]. High-dose therapy is extraction into n-butanol. After re-extraction into an
often followed by leucovorin rescue (folinic acid, 5- aqueous phase, methotrexate is oxidized to 2,4-
formyl tetrahydrofolate). Monitoring is required to diaminopteridine-6-carboxylic acid by the reaction
establish the necessity and duration of the leucovorin scheme shown below.
rescue and to avoid renal damage [2-5]. In this latter
circumstance, concentrations of the order of 108 M
must be measured to establish that the drug is present at
a low and safe level. Methotrexate has a chemical
structure containing a pteridine ring with para-
aminobenzoic acid and glutamate moieties. Methotrexate
is metabolized to 7-hydroxy methotrexate, 2,4-diamino-
N10-methylpteroic acid (DAMPA) and methotrexate
polyglutamates. The pharmacological and physical
properties of methotrexate are summarized in Table 1,
and the ultraviolet absorbance spectrum is shown in
Figure 1.
Methotrexate may be measured in serum by

spectrofluorometric, spectrophotometric, high-
performance liquid chromatographic, and
immunochemical techniques.
i
Methotrexate
Methods edition: Michael A. Pesce
Fourth edition: Michael A. Pesce
Fifth edition: Michael A. Pesce
866
Methotrexate
Fluorescence is measured at an emission wavelength of methotrexate and use fluorescence detection for
450 nm after excitation at 370 nm [6]. quantitation of methotrexate [12,13]. Methotrexate is
oxidized by potassium permanganate (see above) to
The spectrophotometric assay is based on the binding of yield 2,4-diamino pteridine-6-carboxylic acid, which is
methotrexate to the enzyme dihydrofolate reductase detected at an excitation wavelength of 275 nm and an
(DHFR), which results in inhibition of the catalytic emission wavelength of 410 nm [12]. Methotrexate can
activity of DHFR (Table 2, Method 2) [7,8]. The affinity also be detected after reduction to 2,4-diamino-6-
of methotrexate for DHFR is much greater than the methylpteridine by the reaction shown at the top of the
natural substrate, dihydrofolic acid, for the binding sites following page.
on the enzyme. Methotrexate concentration is
determined by its ability to inhibit the reaction shown at Fluorescence of the 2,4-diamino-6-methylpteridine is
at the bottom of this page. detected at an emission wavelength of 463 nm after
excitation at 367 nm [13].
The change in absorbance of NADPH is measured at a
wavelength of 340 nm. The decrease in DHFR activity HPLC can also be used to measure methotrexate
attributable to inhibition by methotrexate is directly polyglutamates in erythrocytes. Two procedures are
related to the concentration of methotrexate in the used. The first approach measures methotrexate
sample. The methotrexate concentration is calculated by polyglutamates as a function of the total methotrexate
use of a calibration curve made with aqueous standards. concentration. Methotrexate polyglutamates are
converted to methotrexate by the enzyme gamma-
The general procedure for high-performance liquid
glutamyl hydrolase, followed by precipitation with
chromatography (HPLC) techniques involves protein
precipitation of the serum sample, followed by injection perchloric acid and injection of the extract into a C18
of an aliquot of the supernatant onto a reversed-phase column. The second approach measures the individual
C18 column (Table 2, Method 3). Methotrexate and its methotrexate polyglutamates by precipitation of the
packed cells with perchloric acid and injection of the
metabolites, 7-hydroxy methotrexate and DAMPA, are
detected by absorption of ultraviolet radiation at a supernatant into the C18 column. In both procedures,
wavelength of 308 nm [9]. Protein precipitation can be methotrexate is measured by photo-oxidation with
eliminated when the sample is passed through a ultraviolet irradiation in the presence of hydrogen
pretreatment column [10,11]. HPLC procedures have peroxide, followed by fluorometric detection of the
also been developed that either oxidize or reduce reaction products [14]. Capillary zone electrophoresis
867
Methotrexate
has also been used to detect methotrexate, 7- of the unbound enzyme-drug complex is determined by
hydroxymethotrexate, DAMPA, and methotrexate measurement of the amount of NADH formed by the
polyglutamates in blood [15]. conversion of glucose-6-phosphate to 6-
phosphogluconate, with the concurrent reduction of
The immunochemical assays are classified as NAD+ to NADH. The change in absorbance is measured
heterogeneous or homogeneous procedures. The at a wavelength of 340 nm and is proportionally related
heterogeneous methods can be divided into enzyme to the amount of methotrexate present in the sample
immunoassays and radiolabeled procedures. [22,23].
In the heterogeneous enzyme immunoassays (Table 2,
Method 4), methotrexate is labeled with an enzyme, such The FPIA assay involves competition between
as luciferase [16] or beta-galactosidase [17], and fluorescein-labeled methotrexate and drug in the sample
competes with methotrexate in the sample for binding for binding sites on an antibody (Table 2, Method 8).
sites on the antibody. One can separate the bound and The fluoresceinmethotrexate antibody complex is a
free labeled drug by using a second antibody, and large molecule and rotates slowly. When plane-polarized
determine the activity of the enzyme-bound fraction light strikes this molecule, the emitted fluorescent light
either by a bioluminescent reaction using luciferin and retains its polarization. In contrast, the unbound labeled
ATP as the substrate or spectrophotometrically using o- methotrexate rotates rapidly, and the fluorescent light is
nitrophenyl--D-galactopyranoside as the substrate and depolarized. The polarization is related to the amount of
measuring the production of o-nitrophenol at a fluorescein-bound complex and is inversely proportional
wavelength of 420 nm. to the concentration of methotrexate present in the
An enzyme immunoassay was developed in which sample [24].
methotrexate bound to polystyrene spheres competes
with methotrexate in the sample for an antibody labeled Reference and Preferred Methods
with the enzyme peroxidase [18]. The amount of There is no formal reference method for methotrexate,
peroxidase bound to the polystyrene spheres is but HPLC is the de facto reference method for
determined by measurement of the oxidation product of measurement because it separates methotrexate from its
o-phenylenediamine at a wavelength of 492 nm and is metabolites and can be used for both drug monitoring
related to the concentration of methotrexate in the and pharmacokinetic studies. When HPLC is used to
sample. measure methotrexate levels in serum with ultraviolet or
fluorescent detection, analytical sensitivity can be 10
The radiolabeled procedures are divided into competitive nmol/L [11]. The within-run coefficient of variation
protein binding (Table 2, Method 5) and (CV) at a methotrexate concentration of 5 mol/L, was
radioimmunoassays. The competitive protein-binding 9.7% [9]. HPLC can also be used to measure
procedure is based on a competition between 3H-labeled methotrexate polyglutamate concentrations in
erythrocytes. The inter-day CV of the method ranged
[19] or 125I-labeled [20] methotrexate and methotrexate
between 5.4% and 9.1% at a methotrexate polyglutamate
in the sample for binding sites on the enzyme DHFR.
concentration of 25 nmol/L. The limit of quantification
The unbound drug is separated by absorption to
for each of the methotrexate polyglutamates was 5
charcoal, and after centrifugation to pellet the charcoal,
nmol/L [14]. HPLC methods are not commonly used to
the supernatant containing the radiolabeled drug bound
measure methotrexate concentrations because they are
to DHFR is counted in a gamma-ray scintillation
time consuming and require expertise in the
counter. The amount of radioactivity in the supernatant
interpretation of the chromatograms and in
is related to the concentration of methotrexate in the
troubleshooting of the HPLC system.
sample.
A 2007 College of American Pathologists Participant
The radioimmunoassay methods are based on a
Summary Report showed that the FPIA procedure was
competition between 3H- or 125I-labeled methotrexate used by 96% of laboratories, with EMIT making up the
and methotrexate in the sample for binding sites on an remaining methods. FPIA is the preferred method for
antibody (Table 2, Method 6). The radiolabeled routine monitoring of methotrexate.
methotrexate bound to the antibody is separated from the
unbound radiolabeled drug by precipitation of the The FPIA method is a rapid and automated procedure
complex by the addition of a second antibody [21]. that can detect methotrexate at levels of 50 nmol/L. The
FPIA method uses 50 or 80 L of sample, and the
The homogeneous immunoassay procedures include an methotrexate results are obtained in about 18 minutes.
enzyme-multiplied immunoassay technique (EMIT) and The linearity of the method is up to 1 mol/L. A
a fluorescence polarization immunoassay (FPIA). significant advantage of this FPIA procedure is that the
Abbott TDx system automatically dilutes samples.
With the EMIT assay, methotrexate in the sample Methotrexate concentrations can be measured in
competes with methotrexate covalently linked to an undiluted samples and at dilutions of 10-, 100-, and
enzyme (glucose-6-phosphate dehydrogenase) for 1000-fold. The software program automatically
binding sites on an antibody (Table 2, Method 7). When calculates the methotrexate concentration that is in the
the methotrexate-enzyme complex is bound to the range of the calibration curve in the diluted sample. The
antibody, the enzyme activity is inhibited. The activity
868
Methotrexate
inter-imprecision CV of the method (n = 9) for with 7-hydroxymethotrexate was 4% and with DAMPA
methotrexate at levels of 0.08 and 0.79 mol/L were it was 20% [31].
9.1% and 3.0%, respectively [24].
The radioimmunoassay procedures are the most sensitive
The EMIT procedure is used by 4% of the laboratories to assays available and can measure less than nanomole
measure methotrexate. The EMIT method is a rapid and quantities of methotrexate. There is no cross-reactivity
automated procedure that can be adapted to most clinical from folate metabolites and trimethoprim. There is 5%
chemistry systems. The inter-day imprecision CV of the cross-reactivity with 7-hydroxymethotrexate and 100%
method was 7.9% at a methotrexate concentration of 0.8 cross-reactivity with DAMPA and methotrexate
mol/L [22]. The lower limit of detection with the EMIT diglutamate and triglutamate [29]. The competitive
assay is 0.3 mol/L, and the linearity of the method is up protein-binding and radioimmunoassays are time-
to 2 mol/L. consuming manual procedures and are not widely used
to measure methotrexate.
The following methods are not commonly used for
routine measurement of methotrexate. The Specimen
spectrofluorometric method is a nonspecific assay that is Serum, cerebrospinal fluid, plasma, or erythrocytes can
subject to interferences by folate and methotrexate be used for measurement of methotrexate. One report
metabolites because their chemical structures are similar suggests that in patients with very low methotrexate
to that of methotrexate. In addition, drugs not chemically concentrations, the drug may be detectable in capillary
related to pteridines (e.g., penicillin and ampicillin) will plasma but not in venous plasma [32]. Methotrexate is
also give a false-positive result [7]. The stable for 5 days in serum samples stored at 23C, 4C,
spectrofluorometric method has a lower limit of or 20C [22] and for 6 days in plasma stored at 4C or
detectability of 1000 nmol/L, is a long, tedious, manual at room temperature [33]. Methotrexate is stable in blood
assay and is not used for monitoring of methotrexate. for 3 days at room temperature or for 6 days if blood is
stored at 4C [33]. Methotrexate polyglutamates are
The spectrophotometric, enzyme-inhibition assay is rapid stable for 48 hours in blood that is in contact with a cold
and has been automated using discrete analyzers [22,25- pack [14].
27]. Choosing the right source of the enzyme DHFR is
essential for obtaining accurate results. When DHFR is Interferences
obtained from bacterial sources, such as Lactobacillus With the EMIT assay, there is 100% cross-reactivity
casei, the drug trimethoprim binds to the enzyme, giving with DAMPA. The amount of cross-reactivity obtained
false-positive results [28]. This does not occur when with 7-hydroxymethotrexate depends on when the
DHFR is obtained from bovine liver. The cross-reactivity sample is drawn after high dose methotrexate infusion.
of this assay with methotrexate metabolites is 100% with Samples drawn at 42 and 66 hours overestimate
methotrexate diglutamate and triglutamate, 10% with methotrexate concentrations by 5% and 31%,
DAMPA, and 1% with 7-hydroxymethotrexate [29]. There respectively. There is greater cross-reactivity in the 66-
is no cross-reactivity with folate metabolites. The method hour specimens because of the higher concentration of 7-
is not subject to interferences from bilirubin at hydroxymethotrexate [34]. There is no cross-reactivity
concentrations of 100 mg/L and hemoglobin at from leucovorin, 5-methyltetrahydrofolate, 5-
concentrations of 1000 mg/L [22]. The assay is sensitive fluorouracil, and 6-mercaptopurine [22]. The EMIT
to methotrexate at concentrations of 20 nmol/L. The assay shows no interference from bilirubin at
within-run CVs at methotrexate concentrations of 0.65 and concentrations of 100 mg/L, but there is a negative
1.90 mol/L were 6.12% and 4.74%, respectively [22]. interference from hemoglobin at concentrations greater
The spectrophotometric enzyme inhibition method than 1000 mg/L [22].
requires preparation of the reagents each time that
methotrexate is measured, which limits its use in the With the FPIA assay, there is 0.6% cross-reactivity with
clinical laboratory. 7-hydroxy methotrexate and 44% cross-reactivity with
DAMPA [24]. There is no cross-reactivity from
With the heterogeneous enzyme immunoassay leucovorin, trimethoprim, 5-methyltetrahydrofolate, 5-
procedures, methotrexate can be detected at nanomolar fluorouracil, or 6-mercaptopurine. There is no
concentrations. There is no cross-reactivity with 7- interference with the FPIA assay from bilirubin at
hydroxymethotrexate or folate metabolites, but there is concentrations of 200 mg/L, hemoglobin at
40% cross-reactivity with DAMPA [17]. These concentrations of 8000 mg/L, or triglycerides at
procedures are time consuming, require extensive concentrations of 6000 mg/L [24].
reagent synthesis by the laboratory, and are not useful
for routine monitoring of methotrexate. Methotrexate Therapeutic Interval
Serum concentrations after high-dose methotrexate
For the competitive protein-binding radiolabeled assay, infusion are typically monitored at 24, 48, and 72 hours
the enzyme DHFR must be obtained from bovine liver following therapy to help gauge the potential for
because of the binding of trimethoprim to DHFR toxicity. Concentrations over a 72-hour period following
obtained from Lactobacillus casei [30]. The lower limit a single-bolus infusion may differ by two orders of
of detectability is 10 nmol/L, and there is no cross- magnitude. Serum concentrations indicative of potential
reactivity with folate metabolites. The cross-reactivity toxicity are as follows:
869
Methotrexate
The slow clearance of 7-hydroxymethotrexate results in

Hours post dose Methotrexate concentration, mol/L plasma concentrations exceeding those of methotrexate
24 >5 at 24 and 48 hours after therapy. Therefore, falsely
48 >0.5 elevated methotrexate values will be obtained with
72 >0.1 analytical methods that exhibit substantial cross-
reactivity with 7-hydroxymethotrexate. The minor
Interpretation metabolite DAMPA is formed in the intestine by
High-dose MTX therapy is used in non-Hodgkins bacterial cleavage of glutamate. It is a weak inhibitor of
lymphoma, osteogenic sarcoma, breast carcinoma, acute DHFR and is usually not detected in plasma.
lymphocytic leukemia, and epidermoid cancer of the
head and neck. Methotrexate is used in low doses to treat The metabolism of methotrexate to 7-
rheumatoid arthritis and other chronic inflammatory hydroxymethotrexate and DAMPA represents
diseases. detoxification pathways, because they are not effective
inhibitors of the enzyme DHFR. Carboxypeptidase G2 is
Methotrexate acts as an antitumor agent by an enzyme that rapidly hydrolyzes methotrexate to the
competitively inhibiting the intracellular enzyme inactive metabolite DAMPA and can be used to rescue
dihydrofolate reductase (DHFR), which prevents the patients who exhibit delayed clearance of methotrexate
conversion of dihydrofolate to tetrahydrofolate. As a because of renal dysfunction. After treatment with
result, it blocks the production of reduced folates and carboxypeptidase G2, very high DAMPA concentrations
inhibits DNA, RNA, and purine synthesis. Leucovorin is are obtained. Since DAMPA significantly cross-reacts
given to rescue cells from the toxic effects of with the FPIA and EMIT assays, another method,
methotrexate because it replenishes the reduced folate preferably HPLC, should be used to monitor
pool by a pathway independent of the enzyme DHFR. methotrexate levels in patients receiving
carboxypeptidase G2 therapy [36].
Methotrexate is metabolized intracellularly to
polyglutamate derivatives by the enzyme Methotrexate is sparingly soluble in water at pH < 7.0,
folypolyglutamate synthase. Polyglutamate derivatives whereas 7-hydroxymethotrexate is about fourfold less
of methotrexate cannot be transported extracellularly soluble than methotrexate. Precipitation of methotrexate
unless they are hydrolyzed to the monoglutamate by the and 7-hydroxymethotrexate in the renal tubules at acid
enzyme gammaglutamyl hydrolase. Up to 7 pH causes renal dysfunction and delayed excretion of
polyglutamate forms of methotrexate can be measured in methotrexate, which increases the risk of toxicity.
erythrocytes. Within the cell, these methotrexate Therefore, in patients receiving high-dose therapy, urine
polyglutamates are potent inhibitors of the enzyme must be alkalinized so that drug toxicity is minimized.
DHFR. They also inhibit the enzyme thymidylate
synthase, resulting in a decrease in purine and DNA Methotrexate toxicity is related to the level and duration of
synthesis. These polyglutamate derivatives remain in the exposure to the drug. To identify patients at risk for toxicity,
cell in the absence of extracellular methotrexate and one should measure methotrexate at various times after
increase the possibility of cytotoxicity. Low-dose administration of the drug. Several studies have shown that
methotrexate therapy is routinely used to treat there is a high risk of toxicity if the drug concentration is
rheumatoid arthritis. The antiinflammatory properties of greater than 5 mol/L or 10 mol/L at 24 hours [1,37] or
methotrexate are probably due to the binding of greater than 0.5 mol/L [2], 1 mol/L [3], or 0.9 mol/L [4]
methotrexate polyglutamates to 5-aminoimidazole-4- at 48 hours after high-dose infusion. Therefore, monitoring
carboxamide ribonucleotide (A1CAR) transformylase, methotrexate levels identifies patients likely to develop
which leads to accumulation of adenosine in the blood. toxicity and allows the clinician to determine (1) the length of
Adenosine activates extracellular adenosine receptors time between doses and (2) the amount of leucovorin that
which inhibit the reproduction of tumor necrosis factor must be administered. It has been suggested that drug levels
alpha, IL6, and IL8 and increase the secretion of the be monitored and leucovorin doses adjusted until the
antiinflammatory cytokine IL10 [35]. Routine methotrexate concentrations are less than 50 nmol/L [5].
monitoring of erythrocyte methotrexate polyglutamate
levels has been proposed in rheumatoid arthritis patients Pharmacogenetics can play a role in understanding the
because of the correlation between methotrexate large interpatient variability in therapeutic responses to
polyglutamate concentrations and clinical efficacy. methotrexate. Methotrexate pharmacogenetics has
focused on the enzymes involved in folate metabolism
The major route of elimination of methotrexate and its and the transport of methotrexate into and out of cells.
metabolites is by the kidneys. Metabolites have been For example, the enzyme methylene tetrahydrofolate
detected in both the plasma and urine of patients reductase (MTHFR) catalyzes the conversion of 5-10
receiving high-dose therapy. The major metabolite, 7- methylene tetrahydrofolate to 5-methyl tetrahydrofolate.
hydroxymethotrexate, is formed in the liver by the A MTHFR 677CT polymorphism, which is an alanine to
enzyme aldehyde oxidase. It is an ineffective inhibitor of valine substitution, results in a variant MTHFR enzyme
DHFR and has a half-life two to three times longer than with decreased enzyme activity. As a result of this
that of methotrexate. polymorphism, patients treated with methotrexate have
an increased risk of hepatotoxicity [38].
870
Methotrexate
Pharmacogenetics can also be used to predict drug 7 Overdijk B, Van Der Kroef WMJ, Visser AAM,
resistance to methotrexate. Hooghwinkel GJM. The determination of
methotrexate in serum and urine. Clin Chim
Cellular resistance to methotrexate can be caused by a Acta 1975; 59: 177-182.
polymorphism in the DHFR gene, which significantly 8 Falk LC, Clark DR, Kalman SM, Long TF.
increases the enzyme activity of DHFR; by a variant Enzymatic assay for methotrexate in serum and
folypolyglutamate synthase enzyme, which decreases the cerebrospinal fluid. Clin Chem 1976; 22: 785-
polyglutamylation of methotrexate; or by a variant 788.
gamma glutamyl hydrolase enzyme, which increases the 9 Cairnes DA, Evans WE. High-performance
hydrolysis of the methotrexate polyglutamates to the liquid chromatographic assay of methotrexate,
monoglutamate. In the future, pharmacogenetic pre- 7-hydroxymethotrexate, 4-deoxy-4-amino-N10-
screening of methotrexate treated patients may be able to methylpteroic acid and sulfamethoxazole in
prevent drug toxicity and customize treatment to serum, urine and cerebrospinal fluid. J
optimize clinical efficacy. Chromatogr 1982; 231: 103-110.
10 Tong WP, Wisnicki JL, Horton J, Ludlum DB.
Methotrexate Performance Goals A direct analysis of methotrexate,
Survey data from the 2007 College of American dichloromethotrexate and their 7-hydroxy
Pathologists Participant Summary Report shows metabolites in plasma by high pressure liquid
imprecision values (% coefficient of variation) for chromatography. Clin Chim Acta 1980; 107:
methotrexate measurements performed by FPIA to be 67-72.
less than 8% in samples, with methotrexate 11 Breithaupt H, Kuenzlen E, Goebel G. Rapid
concentrations of approximately 3 and 29 mol/L. The high-pressure liquid chromatographic
CVs with the EMIT procedure are usually higher than determination of methotrexate and its
those obtained with FPIA. Acceptable performance metabolites 7-hydroxymethotrexate and 2,4-
criteria (CLIA-88) for measurement of methotrexate diamino-N10-methyl-pteroic acid in biological
require that laboratories be accurate to within 3SD or fluids. Anal Biochem 1982; 121: 103-113.
10% of the peer group mean, whichever is greater. 12 Nelson JA, Harris BA, Decker WJ, Farquhar D.
Analysis of methotrexate in human plasma by
The FPIA assay meets these goals and is the method of high-pressure liquid chromatography with
choice for measuring methotrexate. fluorescence detection. Cancer Res 1977; 37:
3970-3973.
References 13 Deen WM, Levy PF, Wei J, Partridge RD.
1 Isacoff WH, Townsend, CM, Eilber FR, Forster Assay for methotrexate in nanomolar
T, Morton DL, Block JB. High methotrexate concentrations with simultaneous direction of
therapy of solid tumors: observations relating to citrovorum factor and vincristine. Anal
clinical toxicity. Med Pediatr Oncol 1976; 2: Biochem 1981; 114: 355-361.
319-325. 14 Dervieux T, Orentas Lien D, Marcelletti J,
2 Isacoff WH, Morrison PF, Aroesty J, Willis, Pischel K, Smith K, Walsh M, Richerson R.
KL, Block JB, Lincoln TL. Pharmacokinetics HPLC determination of erythrocyte
of high-dose methotrexate with citrovorum methotrexate polyglutamates after low-dose
factor rescue. Cancer Treat Rep 1977; 61: 1665- methotrexate therapy in patients with
1674. rheumatoid arthritis. Clin Chem 2003; 49:
3 Nirenberg A, Mosende C, Mehta BM, Gisolfi 1632-1641.
AL, Rosen G. High-dose methotrexate with 15 Kuo CY, Wu HL, Kou H, Chiou SS, Wu DC,
citrovorum factor rescue: predictive value of Wu SM. Simultaneous determination of
serum methotrexate concentrations and methotrexate and its eight metabolites in human
corrective measures to avert toxicity. Cancer whole blood by capillary zone electrophoresis. J
Treat Rep 1977; 61: 779-783. Chromatogr 2003; A1014: 93-101.
4 Stoller RC, Hande KR, Jacobs SA, Rosenberg 16 Wannlund J, Azari J, Levine L, DeLuca M. A
SA, Chabner BA. Use of plasma bioluminescent immunoassay for methotrexate
pharmacokinetics to predict and prevent at the subicomole level. Biochem Biophys Res
methotrexate toxicity. N Engl J Med 1977; 297: Commun 1980; 96: 440-445.
630-634. 17 Al-Bassam MN, OSullivan MJ, Bridges JW,
5 Jolivet J, Cowan KH, Curt GA, Clendeninn NJ, Marks V. Improved double-antibody enzyme
Chabner BA. The pharmacology and clinical immunoassay for methotrexate. Clin Chem
use of methotrexate. N Engl J Med 1983; 309: 1979; 25: 1448-1452.
1094-1104. 18 Ferrua B, Milano G, Ly B, Guennec JY,
6 Kinkade JM, Vogler WR, Dayton PG. Plasma Masseyeff R. An enzyme immunoassay design
levels of methotrexate in cancer patients as using labelled antibodies for the determination
studied by an improved of haptens: application to methotrexate assay. J
spectrophotofluorometric method. Biochem Immunol Methods 1983; 60: 257-268.
Med 1974; 10: 337-350. 19 Arons E, Rothenberg SP, Costa M, Fischer C,
Igbal MP. A direct ligand-binding radioassay
871
Methotrexate
for the measurement of methotrexate in tissues 31 Buice RG, Evans WE, Karas J, Nicholas CA,
and biological fluids. Cancer Res 1975; 35: Sidhu P, Straughn AB et al. Evaluation of
2033-2038. enzyme immunoassay, radioassay, and
20 Methotrexate radioassay kit for the radioimmunoassay of serum methotrexate, as
determination of methotrexate in serum, compared with liquid chromatography. Clin
plasma, uring and cerebrospinal fluid. The Chem 1980; 26: 1902-1904.
Enzyme Center Inc, Malden, MA 02148. 32 Bomelburg TH, Ritter J, Schellong G. Analysis
21 Paxton JW, Ropwell FJ. A rapid, sensitive and of methotrexate concentrations in serum:
specific radioimmunoassay for methotrexate. comparison of capillary and venous blood
Clin Chim Acta 1977; 80: 563-572. samples. Klin Padiatr 1987; 199: 230-232.
22 Pesce MA, Bodourian SH. Enzyme 33 Limelette N, Ferry M, Branger S, Thuillier A,
immunoassay and enzyme inhibition assay of Fernandez, C. In-vitro stability study of
methotrexate with use of the centrifugal methotrexate in blood and plasma samples for
analyser. Clin Chem 1981; 27: 380-384. routine monitoring. Ther Drug Monit 2003; 25:
23 Finley PR, Williams RJ, Griffith F, Lichti DA. 81-87.
Adaptation of the enzyme-multiplies 34 Fotoohi K, Skarby T, Soderhall S, Peterson C,
immunoassay for methotrexate to the Albertioni. Interference of 7-
centrifugal analyser. Clin Chem 1980; 26: 341- hydroxymethotrexate with the determination of
343. methotrexate in plasma samples from children
24 Pesce MA, Bodourian SH. Evaluation of a with acute lymphoblastic leukemia employing
fluorescence polarization immunoassay routine clinical assays. J Chromatogr 2005; B
procedure for quantitation of methotrexate. 817: 139-144.
Ther Drug Monit 1986; 8: 115-121. 35 Grim J, Chladek,J, Martinkova J.
25 Brown LF, Johnson GF, Witte DL, Feld RD. Pharmacokinetics and pharmacodynamics of
Enzymatic inhibition assay for methotrexate methotrexate in non-neoplastic diseases. Clin
with a discrete analyser, the ABA-100. Clin Pharmacokinet 2003; 42: 139-151.
Chem 1980; 26: 335-338. 36 Buchen S, Ngampolo D, Melton RG, Hasan C,
26 Finley PR, Williams RJ. Methotrexate assay by Zoubek A, Henze G, Bode U, et al.
enzymatic inhibition, with use of the centrifugal Carboxypeptidase G2 rescue in patients with
analyser. Clin Chem 1977; 23: 2139-2141. methotrexate intoxication and renal failure. Br J
27 Imbert AM, Pignon T, Lena N. Enzymatic Cancer 2005; 92: 480-487.
assay for methotrexate with a centrifugal 37. Corn WR, Taylor RH, Pratt CB. Methotrexate:
analyzer (Cobas-Bio). Clin Chem 1983; 29: Therapeutic use and serum concentration
1317-1318. monitoring. In Taylor WJ and Finn AL (eds):
28 Bock JL, Pierce R. Trimethoprim interference Individualizing Drug Therapy: Practical
in methotrexate assays. Clin Chem 1980; 26: Applications of Drug Monitoring. New York,
1510-1511. Gross, Townsend, Frank, Inc. 1981;1:149-173.
29 Howell SK, Wang YM., Hosoya R, Sutow 38. Stamp L, Roberts R, Kennedy M, Barclay M,
WW. Plasma methotrexate as determined by ODonnel J, Chapman P. The use of low-dose
liquid chromatography, enzyme-inhibition methotrexate in rheumatoid arthritisare we
assay, and radioimmunoassay after high-dose entering a new era of therapeutic drug
infusion. Clin Chem 1980; 26: 734-737. monitoring and pharmacogenomics? Biomed
30 Hande K, Gober J, Fletcher R. Trimethoprim Pharmacother 2006; 60: 678-687.
interferes with serum methotrexate assay by the
competitive protein binding technique. Clin
Chem 1980; 26: 1617-1619.
872
Methotrexate
Table 1: Pharmacological and Physical Properties of Methotrexate

Structural formula:

Melting point: 185C to 204C
Solubility in water: At pH 5.0, 0.44 mg/mL; at pH 7.0, 8.90 mg/mL
pKas: 4.8, 5.5
Dose: 50 to 250 mg/kg of body weight or > 1 g/m2 for high-dose therapy
Protein binding: 30% to 70%
Volume of distribution: 0.87 L/kg of body weight
Major route of elimination: Renal
Half-life:
Biphasic t, 2.06 h; t, 10.4 h
Triphasic t, 0.75 h; t, 3.5 h; t, 10.4 h
Metabolites:
Serum Major: 7-hydroxymethotrexate
Minor: 2,4-diamino-N10-methyl pteroic acid (DAMPA)
Erythrocytes Methotrexate polyglutamates, n=2-7
873
Methotrexate
Table 2: Methods of Methotrexate Analysis

Method 1: Fluorometric; spectrofluorometric
Principle of analysis: Protein precipitation; extraction and oxidation with potassium permanganate
Comments: Not used; manual nonspecific assay; subject to interferences by folate and methotrexate metabolites
Method 2: Enzymatic inhibition; spectrophotometric
Principle of analysis: Inhibition of enzyme DHFR by methotrexate and measurement of residual enzyme activity
by addition of dihydrofolate and NADPH
Comments: Not commonly used; automated; sensitive to 20 nmol/L; cross-reactivity is 100% with methotrexate
diglutamate and methotrexate triglutamate and 10% with DAMPA; there is no interference from folate derivatives
or from 7-hydroxy methotrexate.
Method 3: High-performance liquid chromatography; HPLC.
Principle of analysis: Isolation by reversed-phase column chromatography
Comments: Used for pharmacokinetic studies; sensitive to 50 nmol/L; separates and detects methotrexate, 7-
hydroxymethotrexate, DAMPA, and intracellular methotrexate polyglutamates
Method 4: Enzyme immunoassay; spectrophotometric or bioluminescent
Principle of analysis: Binding of an enzyme to methotrexate; competition between enzyme-bound drug and free
drug for antibody; measures activity of antibody-bound enzyme
Comments: Not commonly used; sensitivity at nmol/L level; reagents must be prepared by laboratory; no
interference from folate derivatives or 7-hydroxymethotrexate; DAMPA cross-reacts
Method 5: Competitive protein binding; radiolabeled assay
Principle of analysis: Binding of 3H or 125I to methotrexate; competition between radiolabeled drug and free
drug for antibody; measures amount of radiolabeled drug bound to antibody
Comments: Not commonly used; sensitive to 10 nmol/L; see comments for enzyme inhibition assay
Method 6: Radioimmunoassay; radiolabeled assay
Principle of analysis: Binding of 3H or 125I to methotrexate; competition between radiolabeled drug and free
drug for antibody; measures amount of radiolabeled drug bound to antibody
Comments: Not commonly used; sensitive to 5 nmol/L; no interference from folate metabolites or 7-
hydroxymethotrexate; DAMPA, methotrexate diglutamate, and methotrexate triglutamate cross-react
Method 7: Enzyme-multiplied immunoassay; spectrophotometric
Principle of analysis: Binding of glucose-6-phosphate dehydrogenase to methotrexate; competition between
enzyme-bound drug and free drug for antibody; measures activity of enzymedrug complex not bound to antibody
Comments: An infrequently used automated procedure for monitoring methotrexate; sensitive to 300 nmol/L; no
interference from bilirubin at concentrations of 100 mg/L or folate metabolites; hemoglobin at concentrations of
greater than 1000 mg/L decreases methotrexate values; DAMPA and 7-hydroxy methotrexate cross-react
Method 8: Fluorescence immunoassay; fluorescence polarization
Principle of analysis: Binding of fluorescein to methotrexate; competition between fluorescein-bound drug and
free drug for antibody; measures polarization of bound antibody complex
Comments: Most commonly used method and automated procedure; sensitive to 50 nmol/L; no interference from
folate metabolites or 7-hydroxy methotrexate; DAMPA cross-reacts
DAMPA, 2,4-Diamino-N10-methylpteroic acid; DHFR, dihydrofolate reductase; NADPH, reduced nicotinamide adenine
dinucleotide phosphate.
874
Methotrexate
Figure
Figure 1: UV absorbance spectrum of methotrexate.
Absorbance spectrum of methotrexate in 0.1 M KOH (pH 13). At this pH, the molar absorptivities of
methotrexate are as follows: 370nm = 7100; 302nm = 22,000; and 257nm = 23,000 mol L1 cm1. (From
Seeger DR et al: J Am Chem Soc 71:1753-1759, 1949.)
875
Methylmalonic Acid
Methylmalonic Acid
Kevin Carpenter & Kathryn Green
Name: Methylmalonic acid (MMA)
Clinical interpretation: Refer to Chapter 52, Diseases of Genetic Origin, in the 5th edition of
Chemical class: Dicarboxylic acid
i comparison of the developed spot with an appropriate
standard (Table 1, Method 2). A simple and rapid gas
Methylmalonic acid (as its CoA ester) is an intermediate chromatographic method for the quantitative
in the propionate degradation pathway (Figure 1). determination of urinary methylmalonic acid, applicable
Deficient activity of the enzyme responsible for the on a routine clinical basis, has been described [3]. The
conversion of methylmalonyl CoA to succinyl CoA method is based on the isolation of a mono- to oligo-
(methylmalonyl CoA mutase) results in the organic carboxylic acid fraction from urine by a series of ether
aciduria known as methylmalonic aciduria, with a extractions and the subsequent gas chromatography
classical presentation of neonatal onset metabolic (GC) analysis of these acids as their trimethylsilyl
acidosis, hyperammonemia, and poor outcome if (TMS) esters and ethers.
untreated. The enzyme requires adenosylcobalamin as a
cofactor, so inborn errors in the cellular uptake and Since many organic acidurias present with a similar
processing of cobalamin will also result in increased clinical picture, it is now commonplace to perform urine
methylmalonic acid (MMA) in plasma and urine. In both organic-acid analysis by gas chromatography or gas
methylmalonyl CoA mutase and cobalamin defects, chromatography/mass spectrometry (GC/MS) in
concentrations of MMA in plasma and urine are symptomatic patients (Table 1, Method 3). These
significantly elevated. However, in recent times a mild methods may be qualitative, but if MMA is found to be
elevation in MMA has been shown to be a sensitive increased, it may be easily quantitated, especially when
marker for dietary cobalamin deficiency, and in this using GC/MS, since a stable isotope internal standard is
application, a sensitive specific assay for MMA in commercially available. Goodman and Markey [4]
plasma is required. described a standard organic-acid method that employs
ethyl acetate and ether extractions and TMS
The earliest methods for the quantitation of urinary derivatization. Although solvent extraction GC or
methylmalonic acid (MMA) were based on its reaction GC/MS methods are usually performed in urine, they
with diazotized p-nitroaniline. The emerald green color can be adapted to the much lower levels seen in plasma,
that developed after alkalinization with sodium especially where GC/MS is used in the selected ion-
hydroxide was measured at 620 nm [1] (Table 1, Method monitoring mode in conjunction with stable isotope
1). Various modifications have been published to internal standards [5]. Solid-phase extraction has also
improve specificity, but these procedures are of historic proved to be a suitable sample preparation technique for
interest only. plasma [6].
Separation of methylmalonic acid from interfering The other major technique in use is liquid
compounds was achieved by a variety of chromatography (LC) coupled to tandem mass
chromatographic techniques. Diethyl ether extractions of spectrometry (MS/MS) (Table 1, Method 5). A method
acidified urine were concentrated by evaporation and for the determination of MMA as the di-n-butyl ester
then applied to thin-layer chromatographic plates coated derivative in plasma and urine using LC-MS/MS has
with silica gel. An alternative method utilized an been described [7]. Further refinements of this procedure
aluminum-backed, cellulose thin-layer plate, developed [8], based on the extraction of dicarboxylic acids from
by ascending chromatography in n-butanolacetic acid samples with methyl-tert-butyl ether and derivatized
water [2]. MMA could be semiquantitated by with butanolic HCL to form dibutyl esters have been
found to be specific for MMA without interference from
succinic acid. With all dibutyl ester methods, care must
i
Methylmalonic Acid be taken to ensure the volatile product is not lost when
Previous and current authors of this method: removing the excess derivatizing agent [9].
Methods edition: Zulfikarali H. Verjee Reference and Preferred Methods
Second edition: Not updated No reference method is available. Although the GC/MS
Third edition: Not updated method is widely used, LC-MS/MS is becoming more
Fourth edition: Theodore Dashman, Zulfikarali H. commonplace in the clinical laboratory and has the
Verjee advantage of increased throughput over GC methods.
Fifth edition: Kevin Carpenter, Kathryn Green
876
Methylmalonic Acid
The thin-layer chromatography methods are qualitative Interpretation

or semiquantitative, cumbersome, and time consuming. Like many of the organic acidurias, methylmalonic
Compounds other than MMA also couple with the aciduria has a spectrum of clinical presentation and
stabilized diazotate of dianisidine (fast blue B). disease severity, depending on the residual enzyme
However, the solvent system used in TLC gives clear activity. In the most severe neonatal form, patients
and discrete separation of methylmalonic acid from these generally present acutely with a common set of clinical
compounds. and laboratory results. The signs and symptoms include
lethargy, failure to thrive, recurrent vomiting,
GC/MS is a sensitive and specific method, allowing dehydration, respiratory distress, and muscular
accurate quantitation of the low levels seen in normal hypotonia. If untreated, coma and death ensue. The most
plasma. GC with other detector systems may be used, common laboratory findings at onset of clinical illness
but selection of a suitable internal standard not present in are metabolic acidosis, ketosis, hyperammonemia,
normal body fluids may be troublesome. hypoglycemia or hyperglycemia, leukopenia,
thrombocytopenia, and anemia.
LC-MS/MS using isotope-labeled internal standards is
the method of choice if the instrumentation is available. Methylmalonic acid is a catabolic product of the amino
It offers excellent sensitivity, with low sample volumes acids valine, isoleucine, threonine, and methionine. Its
required, and can be used for both diagnosis of inborn excretion is greatly increased after a protein load in
errors of metabolism and micronutrient studies. patients who cannot metabolize MMA further. The
therapy for such patients, therefore, is the administration
Specimen of a protein-restricted diet. Carnitine and oral antibiotic
Traditionally, a random urine sample has been the therapy to reduce propionate derived from gut flora are
specimen of choice for detection of inborn errors of also used.
metabolism. Classical methylmalonic aciduria is likely
to be detected as part of an organic-acid profile, and Gross elevations of methylmalonic acid are seen in the
excretion is markedly elevated. Milder increases are urine of patients with congenital methylmalonic
more likely to be associated with cobalamin defects or acidemia (50 to 1000 times normal). The defective
deficiency, and although theoretically the higher values enzyme is mitochondrial L-methylmalonyl CoA mutase,
in urine should make it suitable for monitoring, many which catalyzes the conversion of L-methylmalonyl CoA
studies have used plasma or serum. In all sample types, to succinyl CoA. Isolated defects of adenosyl cobalamin
MMA is stable for many months at 20C. production lead to methylmalonic aciduria biochemically
indistinguishable from mutase defects. Methylmalonic
Interferences aciduria is accompanied by homocystinuria in early
The major interference in MMA estimation comes from processing defects in vitamin B12 (cobalamin)
the naturally occurring, structurally related isomer, metabolism. Absence of both the coenzymes
succinic acid, which is a product of MMA degradation methylcobalamin (MeCbl) and adenosylcobalamin
and usually present at concentrations greater than that of (AdoCbl) leads to a combined deficiency of the activity
MMA [8]. Interference from succinic acid is not usually of methionine synthetase and methylmalonyl CoA
an issue in GC/MS-based methods, since the two mutase, which leads to accumulation and excretion of
isomers are adequately separated chromatographically. homocysteine and methylmalonic acid.
Tandem MS methods have relied upon using different
transitions for MMA and succinate, although complete The milder end of the spectrum of inborn errors affecting
freedom from interference requires a short methylmalonyl CoA mutase may have a benign course,
chromatographic run, as well as careful selection of but patients are at risk of acute presentation during
transitions. intercurrent illness and catabolic stress.
Methylmalonic Acid Reference Interval Severe vitamin B12 deficiency can lead to marked
Urinary excretion of MMA in newborns can be up to 83 elevations of MMA in urine and plasma, although
mg/g creatinine (79.4 mol/mmol creatinine) (97.5 typically not to the levels seen in classical mutase
percentile, n = 57) [10]. Excretion falls with increasing defects. Mild elevations of MMA in plasma or serum are
age, so that above 10 years, the upper limit of normal is considered by many to be a better marker for subclinical
22.4 mg/g (21.6 mol/mmol creatinine). Upper limit of cobalamin deficiency than direct estimates of B12 levels.
normal excretion in adults is 3.8 mg/g (3.7 mol/mmol Other causes of mild elevations in plasma or serum
creatinine) [5]. Similarly, plasma or serum MMA levels are impaired renal function [12] and thyroid
concentrations are higher in infants younger than 6 disease [13].
months of age (0.12 to 1.57 mol/L) than in older
children and adults (0.06 to 0.34 mol/L) [9]. There is Methylmalonic Acid Performance Goals
generally a good consensus between studies with an To be able to detect vitamin B12 deficiency, methods
upper limit of normal between 0.26 and 0.35 mol/L. should be able to accurately measure concentrations of
Aging in males does not affect MMA concentrations in methylmalonic acid of approximately 1 mol/L in urine
serum, but females older than 60 years had slightly and 0.1 mol/L in serum. The GC method has good
higher values than younger women [11]. precision (coefficient of variation [CV] = 7.8% for
MMA concentration of 0.2 mol/L in plasma) [6]. A
877
Methylmalonic Acid
study of performance across 13 laboratories found

improvements were needed to reduce the analytical 8 Kushnir MM, Komaromy-Hiller G, Shushan B,
imprecision of most laboratories, and attention must be Urry FM, Roberts WM. Analysis of
focused on calibration issues [14]. dicarboxylic acids by tandem mass
spectrometry. High-throughput quantitative
References measurement of methylmalonic acid in serum,
1 Giorgio AJ, Plaut GWE. A method for the plasma, and urine. Clin Chem. 2001;47:1993-
colorimetric determination of urinary 2001.
methylmalonic acid in pernicious anaemia. J 9 Green AK. Unpublished.
Lab Clin Med. 1965;66:667-676. 10 Guneral F, Bachmann C. Age-related reference
2 Bhatt HR, Green A, Linnell JC. A sensitive values for urinary organic acids in a healthy
micromethod for the routine estimation of Turkish pediatric population. Clin Chem.
methylmalonic acid in body fluids and tissues 1994;40:862-868.
using thin layer chromatography. Clin Chim 11 Rasmussen K, Moller J, Lyngbak M, Pedersen
Acta. 1982;118:311-321. AM, Dybkjaer L. Age- and gender-specific
3 Gibbs BF, Itiaba K, Mamer OA, Crawhall JC, reference intervals for total homocysteine and
Cooper BA. A rapid method for the analysis of methylmalonic acid in plasma before and after
urinary methylmalonic acid. Clin Chim Acta. vitamin supplementation. Clin Chem.
1972;38:447-453. 1996;42:630-636.
4 Goodman S, Markey S. Diagnosis of organic 12 Lindgren A. Elevated serum methylmalonic
acidurias by GC-MS. Lab Res Methods Biol acid: How much comes from cobalamin
Med. 1981;6:105-114. deficiency, and how much comes from the
5 Marcell PD, Stabler SP, Podell ER, Allen RH. kidneys? Scand J Clin Lab Invest. 2002;62:15-
Quantitation of methylmalonic acid and other 19.
dicarboxylic-acids in normal serum and urine 13 Chong YY, Gupta MK, Jacobsen DW, Green R.
using capillary gas-chromatography mass- Serum homocysteine and methylmalonic acid
spectrometry. Anal Biochem. 1985;150:58-66. are not reliable indicators of cobalamin or folate
6 Kushnir MM, Komaromy-Hiller G. deficiency in patients with abnormal thyroid
Optimization and performance of a rapid gas function. Blood. 1993;82:94.
chromatography-mass spectrometry analysis for 14. Pfeiffer CM, Smith J, Miller D, Gunter EW.
methylmalonic acid. J Chromatogr B Biomed Comparison of serum and plasma
Sci Appl. 2000;741(2):231-241. methylmalonic acid measurements in 13
7 Magera MJ, Helgeson JK, Matern D, Rinaldo P. laboratories: an international study. Clin Chem.
Methylmalonic acid measurement in plasma 1999;45:2236-2242.
and urine by stable-isotope dilution and
electrospray tandem mass spectrophotometry.
Clin Chem. 2000;46:1804-1810.
878
Methylmalonic Acid
Table 1: MMA Methods Summary
Method 1: Colorimetric urine assay

Principle: Reaction of MMA with diazotized p-nitroaniline to generate emerald green color.
Usage: Of historical interest only.
Method 2: Thin-layer chromatography

Principle: Thin-layer chromatographic separation followed by localization with diazo reagent.
Usage: Largely superseded by GC or LC/MS methods.
Method 3: Gas chromatography/mass spectrometry of organic acids

Principle: Organic acids are extracted from urine using organic solvent (ethyl acetate/diethyl ether), converted to
TMS esters, and separated on capillary GC. Positive peak identification by mass selective detector.
Usage: Widely used as most common method for detecting all organic acidurias.
Comments: Method can be adapted to low levels seen in plasma by using stable isotope dilution quantitation and
selected ion monitoring.
Method 4: Liquid chromatography/tandem mass spectrometry

Principle: Methyl malonate is extracted from urine or plasma by solid-phase or solvent extraction, converted to
the dibutyl ester by reaction with butanolic HCl, and analyzed by tandem mass spectrometry, using stable isotope
internal standard. Interference from succinate is avoided by a short chromatographic separation.
Usage: Method of choice if instrumentation is available.
Comments: Excellent sensitivity and high throughput make this assay suitable for micronutrient studies.
Procedure: Gas Chromatography/Mass 9. Calibration standards are prepared

Spectrometry at 0.2, 0.5, 1.0 and 2.0 mol/L in
Principle dialyzed human plasma.
Based on methods published by Rasmussen
[11] and Kushnir [6], a rapid and sensitive GC/MS assay
for methylmalonic acid determination in serum and Assay
plasma is described. The assay utilizes an anion- Equipment: A suitable GC-MS system capable
exchange, solid-phase extraction, trimethylsilyl of operation in electron-impact ionization and selective
derivatization, and stable isotope internal standard ion-monitoring mode is required. The gas
quantitation. The method is sensitive enough to allow chromatograph is fitted with a 5% phenylmethyl-silicone
accurate quantitation of levels seen in normal capillary column, 25 m 0.20 mm, 0.33-m film
individuals. thickness. Solid-phase extraction is achieved using
Reagents strong anion-exchange columns (e.g., Bond Elut, SAX)
1. Methanol, A.R. 1. In a screw-capped glass tube, 100 L of the
2. Acetonitrile, A.R. internal standard d3MMA and 1 mL of
3. Methyl tert-butyl ether (MTBE), acetonitrile is added to each calibrator, control,
A.R. and test sample.
4. Formic acid, A.R. 2. The tubes are vortexed for 20 sec and
5. MTBSTFA (N,(tertiary centrifuged for 5 min at 2000 g.
butyldimethylsylyl)-N- 3. The contents of the tubes are transferred onto
methyltrifluoro-acetamide). the SPE columns, previously conditioned with 3
6. Methylmalonic acid stock standard mL methanol and 5 mL distilled water.
(MMA), 10 mmol/L in methanol. 4. The columns are washed by sequential addition
Place 118.1 mg of MMA in a 100-mL of 10 mL deionized water, 5 mL of methanol, 2
volumetric flask, add methanol to mL of MTBE, and dried after each wash with
dissolve, and bring to volume with vacuum for 3 min.
methanol. Store at 20C. 5. The analytes are eluted with 5 mL of elution
7. d3 Methylmalonic acid (Cambridge solvent consisting of 3% formic acid in MTBE.
Isotopes DLM-387) stock standard 6. The eluent is evaporated to dryness under a
(d3MMA), 1 mmol/L in methanol. stream of nitrogen at 35C.
Place12.1 mg of d3MMA in a 100-mL 7. The residue is reconstituted with 25 L of
volumetric flask, add methanol to MTBSTFA and 25 L acetonitrile and the tubes
dissolve, and bring to volume with incubated at 55 5C for 5 min.
methanol. Store at 20C. 8. GC/MS parameters:
8. MMA and d3MMMA working The mass selective detector is used in
standards at 10 and 15 mol/L, the electron-impact ionization mode
respectively, prepared in at 70 eV.
acetonitrile. Helium is utilized as a carrier gas.
879
Methylmalonic Acid
Injection port temperature 270C. Calculation

Initial column temperature 100C, Results are calculated by calculation of the ratio
ramped at 18C/min to 160C, of MMA to d3MMA signal peak areas in unknowns read
ramped at 50C/min to 300C, with a against a standard curve.
final hold of 2.5 min.
Interface temperature 300C. Notes:
Split injection, ratio of 1:20, and 1. MTBSTFA gives superior signal for the M-57
injection volume 1 L. ion compared to the corresponding M-15 ion
The ions monitored for the derivatives for TMS derivatives, but either may be used.
are: m/z 289 for MMA and m/z 292 2. The method adequately separates succinate
for d3MMA from MMA, allowing accurate quantitation.
Figure
Isoleucine
Valine
Methionine
Adenosylcobalamin
Propionyl CoA Methylmalonyl CoA Succinyl CoA
Cholesterol Methylmalonic acid

Odd Chain FA
Figure 1. Simplified pathway of propionate precursors and degradation. Conversion of methylmalonyl CoA to succinyl
CoA requires adenosylcobalamin as a cofactor.
880
Mycophenolic Acid
Mycophenolic Acid
Michal J Figurski, Magdalena Korecka, Leslie M Shawi
Name: Mycophenolic acid, MPA

Systematic (IUPAC) name: (E)-6-(4-hydroxy-6-methoxy-7-methyl-3-oxo-1,3-
dihydroisobenzofuran-5-yl)-4-methylhex-4-enoic acid
Clinical significance: Immunosuppressant for prevention of acute rejection of allograft in
transplant patients Refer to Chapter 54, Laboratory Evaluation of the
Transplant Recipient, in the 5th edition of Clinical Chemistry:
Theory, Analysis, Correlation
CAS number: 24280-93-1
Structure:
Principles of Analysis and Current Usage (CAP) survey [7], participating laboratories analyze MPA i
Mycophenolic acid is a fungal antibiotic, initially isolated concentrations either with HPLC/UV (60%) or mass
from a strain of Penicillium brevicompactum by the Syntex spectrometry (40%). The CAP survey shows no use of the
company (now in the Roche group) in the late 19th EMIT assay for MPA, since it is not approved by the U.S.
century. It is widely used to prevent allograft rejection in Food and Drug Administration (FDA) for use in the United
solid organ transplant patients. Currently, it is approved for States.
prevention of acute rejection in renal, heart, and hepatic
transplants and is used in various combinations with other Historically, the first method for determination of
immunosuppressants. There are ongoing clinical trials mycophenolic acid, published in 1968, was based upon
evaluating MPA in other administration protocols, as well separation of various antibiotics by paper chromatography
as for other organs, such as pancreas, intestine, lungs, and and then isolation of MPA using thin-layer
also bone marrow [1]. chromatography [8]. Several other methods for analysis of
MPA were developed in the following years, including
In addition, mycophenolic acid is under investigation as an turbidimetric analysis, thin-layer chromatography, and gas
immunomodulator in treatment of autoimmune diseases chromatography [9-11]. However, all these methods
such as lupus nephritis (Phase III study as of June 2007), involved a complex sample preparation step. In a report
pemphigus vulgaris (Phase III study as of June 2007), from a Syntex study published in 1990, a simple
myasthenia gravis (failed Phase III study) [2], immune HPLC/UV method for analysis of mycophenolic acid was
thrombocytopenic purpura, autoimmune hepatitis and IgA first mentioned, although no details were given [12]. This
nephropathy [3-5]. method was described later in 1994 [13]. High sensitivity
for MPA in this method was achieved using liquid-liquid
According to the International Proficiency Testing Scheme extraction from plasma and detection with ultraviolet light
for MPA, provided by Analytical Services International of 215 and 304 nm wavelength.
(ASI, London, UK), 45% of participating laboratories
analyze mycophenolic acid concentrations using EMIT, i
Mycophenolic acid
33% using HLPC/UV, and 20% using mass spectrometry; New method
2% of participants report use of other methods [6]. Fifth edition: Michal J Figurski, Magdalena Korecka,
According to the 2007 College of American Pathologists Leslie M Shaw
881
Mycophenolic Acid
pH can alter the protein binding of mycophenolic acid.
Later on, many other HPLC methods were developed for
analysis of mycophenolic acideither alone, with its The prepared sample is usually diluted prior to injection
metabolites, or as a part of whole panels of onto HPLC column. Larger injection volumes and less
immunosuppressive drugs. The methods utilizing dilution are commonly used in HPLC/UV methods,
ultraviolet detection are mainly applied to analysis of total whereas smaller volumes with larger dilutions are used for
mycophenolic acid and/or its glucuronide metabolite mass spectrometric methods. Application of liquid-liquid
(MPAG) concentration. For analysis of free MPA or the extraction (LLE)/SPE also allows for smaller injection
total concentration of other metabolites, such as volumes. A wide variety of mobile phases have been
mycophenolic acid acyl-glucuronide (AcMPAG), mass described for the analysis of MPA, although the primary
spectrometric detection is more suitable. Mass components of every mobile phase are water and an
spectrometric detection is applied exclusively in the organic solvent, either methanol or acetonitrile. Phosphoric
systems where multiple drugs are analyzed simultaneously. acid or phosphate buffer are often used in HPLC/UV
Starting from the year 2001, immunoassay methods for assays [13,14,17-21,23-25]. For LC/MS, the mobile phase
analysis of MPA were introduced and gradually gained is usually acidified with acetic or formic acid [15,26-31].
popularity because of their simplicity, low initial The majority of the recently published methods use
investment cost and lack of sample pretreatment step. gradients to decrease the runtime and increase selectivity.
Chromatographic Methods The most commonly used type of stationery phase for
Sample Preparation and Chromatography mycophenolic acid analysis is a reversed-phase C18
Sample preparation, the first step in chromatographic column, although reversed-phase C8 columns are also in
analysis of MPA, is the removal of proteins and other large use. Various column lengths and particle sizes are used
molecules that could compromise the performance of the across the published methods. Interested readers may find
chromatographic column. Several of the most commonly more information in a recent review [32]. The column
used methods are described below. length is usually between 150 and 250 mm, although
methods for shorter columns have also been published.
A simple and cost-effective method is protein precipitation
(PP), in which the sample is mixed with a precipitating An increasing number of methods utilize on-line sample
agent, centrifuged, and the supernatant is recovered. The clean-up techniques, performed using two columns, two
precipitating agent can be methanol [13], acetonitrile [14], pumps, and a switching valve. Most methods employ the
or sodium tungstate [15]. This process additionally frees cross-directional flow of the eluting phase; however, there
mycophenolic acid from albumin binding and allows for are also published methods that use the same flow
analysis of the total concentration of MPA. However, this direction of the washing and eluting phases. Diagrams of
method does not remove any small-molecule impurities, the switching valve connections for unidirectional and
which may interfere with the chromatographic analyses of cross-directional flow of eluting phase are shown in Figure
MPA. 1. In this set-up, the main column for chromatographic
separation of MPA is a regular, full-length, reversed-phase
Other methods of protein removal include solid-phase C18 or C8 column, and the washing column is a short
extraction (SPE) [16-19] or liquid-liquid phase extraction column of the same type, not exceeding 50 mm in length.
(LLE) [13, 20]. These methods allow for substantial The washing column is used as an adsorbent material,
sample clean-up and increase in concentration, which often retaining the MPA for analysis while allowing the solvents
results in increased sensitivity and a broader calibration and impurities to be removed by the quick flow of the
range for HPLC/UV methods. However, a substantial washing phase. The activation of the switching valve
drawback of these methods is their complexity and (switch from A to B position) ends the washing period and
required workload. In the case of SPE, an additional starts the eluting period [15,27].
drawback is the high cost of disposable cartridges.
Ultraviolet Detection
Plasma can also be injected directly onto HPLC without The ultraviolet (UV) spectrum of mycophenolic acid is
any sample pretreatment [21]. A system of two columns shown in Figure 2 in the range from 220 to 360 nm. Three
and a switching valve is employed for sample clean-up, major peaks of this spectrum are utilized in the published
with mobile phase composed of acetonitrile and phosphate methods: 215 nm (shown truncated), 254 nm, and 304 nm.
buffer (1:1 v/v). The latter peaks, despite their smaller intensity, provide the
advantage of less interference from other compounds of
Ultrafiltration (UF) is a necessary step in analysis of free human plasma.
MPA [22]. It removes proteins together with the fraction
of the drug that is bound to proteins. In this case, the Methods vary in terms of a compound used as an internal
micropartition devices or microfilter kits are used, with standard, but most of the published methods utilize
cutoff of around 30,000 D. It is essential that the sample is carboxybutoxy ether of mycophenolic acid (MPAC). There
not pretreated before this process, because the change of are also methods that do not use the internal standard at all.
882
Mycophenolic Acid
Using a standard sample clean-up procedure, such as lowering of that limit. A short summary of calibration
protein precipitation, mycophenolic acid can be detected at ranges and lower limit of quantification values (LLOQ)
concentrations around 0.2 to 0.3 mg/L, whereas some from several published methods are shown in Table 1.
sample clean-up techniques allow for considerable
Table 1: Method Information Summary for Selected HPLC/UV Methods

Reference Sample UV Peak Internal MPA* MPAG* AcMPAG*
(Chronologically) Prep [nm] Standard [mg/L] [mg/L] [mg/L]
13 LLE 215 & 304 Diazepam & 0.5-50 (no data)
indomethacin
21 Direct 215 None 0.1-20 (0.10)
injection
23 PP 215 Epilan D 0.25-15 (0.25) 3.4-218 (3.41) 0.20-15 (0.20)
14 PP 254 MPAC 0.5-20 (0.25) 5.0-200 (0.5) 2.5-100 (0.25)
17 SPE 306 Para-nitro-anilin 0.2-20 (0.2)
(PNA)
25 UF 304 MPAC 0.005-2.0 (0.005) free 1.0-150 (1.0) free
*Lower Limit of Quantitation (LLOQ) in brackets
Table 2: Method Information Summary for Selected HPLC/Fluorescence Methods

Reference Sample Prep Excitation & Emission [nm] Internal Standard MPA*
(Chronologically) [mg/L]
33 PP 310 & 430 None 0.2-20 (0.20)
34 PP 342 & 425 Naproxen 0.05-40 (0.05) total
0.005-1.0 (0.005) free
* Lower Limit of Quantitation (LLOQ) in brackets
Fluorescence Detection [27,36,37]. There are many chromatographic methods

As an alternative to ultraviolet detection, mycophenolic published for simultaneous determination of a whole panel
acid can also be analyzed using HPLC with fluorescence of immunosuppressants, where MPA and/or MPAG is just
detector [33,34]. Since fluorescence characteristics of one of multiple analytes [36,37]. Other methods allow for
chemical substances are generally more unique than simultaneous determination of MPA and both of its major
ultraviolet absorbance, the fluorescence methods are more metabolites, MPAG and AcMPAG [27,28]. Virtually all of
sensitive and allow for simpler sample pretreatment and the published methods utilize an electrospray ionization
shorter runtime. However, for the same reason, these source, either in negative or positive ionization mode.
methods do not allow for simultaneous analysis of MPA Positive ionization mode requires an ion-pairing agent, and
and its metabolites. The fluorescence of MPA is highly pH most methods utilize ammonium acetate to produce NH4+
dependent, thus this method of detection requires strict adducts. Na+ adducts are also used, although Na+ ions are
control of pH in the mobile phase, and the mobile phase actually a contamination of the sample from glassware that
itself is of complex composition. Validation information is used for sample preparation; therefore, this pairing ion is
for these methods is summarized in Table 2 above. less favorable. In the negative ionization mode, the
analytes minus a H+ ion are measured directly.
Mass Spectrometric Detection
Mycophenolic acid can be analyzed with much greater Table 3 summarizes MS parameters for several published
accuracy and precision using mass spectrometric (MS) methods, and Figure 3 shows the most common MS/MS
[29,35] or tandem mass spectrometric (MS/MS) assays fragments of the mycophenolic acid molecule.
883
Mycophenolic Acid
Table 3: Method Information Summary for Selected HPLC/MS & HPLC/MS/MS Methods
Reference Sample Method Ionization Mode MS Masses or MS/MS Transitions
(Chronologically) Preparation (m/z)
29 PP MS Positive, MPA: 343.4
Na+ adduct IS (N-hexadecyl--D-glucopyranoside) n/d
38 PP MS/MS Negative MPA: 319.0 191.1
IS (indomethacin): 356.0 312.2
MPAG: 495.5 319.4
26 SPE MS/MS Positive, NH4+ MPA: 337.7 207.2
adduct IS (MPAC): 437.6 207.2
MPAG: 513.6 207.2
27 PP + on-line MS/MS Positive, NH4+ MPA: 338.1 207.1
cleanup adduct IS (MPAC): 438.0 303.0
MPAG & AcMPAG: 514.2 303.0
Table 4: Summary of Method Validation Information for Selected Mass Spectrometric Methods
Reference Runtime Linearity Range LOD* LLOQ Accuracy Imprecision Recovery
(Chronologically) [min] [mg/L] [mg/L] [mg/L] [%] [%] [%]
29 14 MPA: 0.01-1.0 0.001 0.021 9.8 13.2 80-85
38 MPA: 0.1-30 0.05 0.1 12.28 90-91
6
MPAG: 1-300 0.5 1 13.37 73-77
26 MPA: 0.1-16 99
7 <15
MPAG: 1-200 99
27 MPA: 0.05-30 0.05 3.0-4.7 98-100
5 MPAG: 2-400 2 97.5-103 5.4-7.3 101-103
AcMPAG: 0.025-15 0.025 6.3-13 98-103
*LOD, Lower Limit of Detection; , LLOQ, Lower Limit Of Quantitation; , Calibration range reported using quadratic fit.
The mass spectrometric methods are generally viewed as complex labeled with MPA is added. The activity of the
requiring less chromatography because of the mass enzyme, directly proportional to the rate of formation of
selectivity of the detector instrument itself. However, in NADH, is estimated by measuring the change in
the case of analysis of mycophenolic acid, a phenomenon absorption at a wavelength of 340 nm at two points in
of in-source decomposition of MPAG back to MPA was time. After the first step, the antibodies present in the
observed, especially with elevated source temperatures solution that have not been bound to molecules of
[30]. An example of this phenomenon is shown in Figure mycophenolic acid bind to the MPA-labeled enzyme and
4. Table 4 above summarizes the available information thus decrease the enzymatic activity, which allows for
from method validation. calculation of MPA concentration.
Immunoassay Techniques The genetically engineered monoclonal antibody that is the

Enzyme-Multiplied Immunoassay Technique (EMIT) subject of the referenced patent [39] is said to be capable
An enzyme-multiplied immunoassay technique was of distinguishing MPA from mycophenolate mofetil, the
originally developed by Syva Corporation (Palo Alto, CA) ester pro-drug, and from the major metabolite, MPAG.
in 1972. The principle of the EMIT method is described in However, some cross-reactivity of this antibody was
full detail in Chapter 9, Principles for Competitive-Binding observed with AcMPAG. Since AcMPAG exhibits similar
Assays, of this book. An assay for mycophenolic acid was immunosuppressive potency to MPA in vitro [24], the
later developed by Dade-Behring company [39]. EMIT method may be perceived as allowing for estimation
of total immunosuppressive impact of MPA through
In general, the method involves several steps, shown measurement of all the active moieties of MPA
schematically in Figure 5. In the first step, the sample is simultaneously. This is sometimes presented as an
mixed with the antibody, the enzyme substrate, and advantage of this method in advertising materials.
nicotinamide adenine dinucleotide (NAD). In the second However, the EMIT results do not represent the true MPA
step, the enzyme glucose-6-phosphate-dehydrogenase concentrations, and the level of bias for each individual
884
Mycophenolic Acid
sample is unknown. A summary of several selected of patient results from one of the centers from this study is
publications comparing EMIT to either HPLC/UV or shown graphically in Figure 6. The coefficient of variation
MS/MS is shown in Table 5. (CV) of the within-run imprecision of this assay measured
on spiked plasma samples was 1.3% to 4.4% for samples
Table 5: Summary of Publications Comparing EMIT below 1 mg/L of MPA, and 0.4% to 4.7% for samples
to HPLC or MS Methods above 1 mg/L of MPA. Between-day imprecision was
1.1% to 9.4% CV and 0.0% to 6.7% CV for samples below
Reference Average Bias Average Bias and above 1 mg/L of MPA, respectively. The lower limit
Reference
Method of C0 [%] of AUC [%] of quantification for this method was established to be
[24] UV 151.2% approximately 0.3 mg/L of MPA. Recovery ranged from
81% to 120% depending upon the sample and laboratory,
[33] UV 11.1 15.2% with mean approximately 100% [43]. This assay is also
[40] UV 22% *
11.6% capable of measuring free MPA concentrations in the
range 0 to 200 g/L, with CV for 25, 75, and 150 g/L
[38] MS/MS 61.4 57.9%. control samples below 5% [42].
[31] MS/MS 18.7 26.8% 16.7% 22.5%
*
Cloned-Enzyme Donor Immunoassay (CEDIA)
Value not directly reported in the paper but derived Another enzymatic immunoassay for mycophenolic acid,
from the data. based on the cloned-enzyme donor immunoassay (CEDIA)
principle, was developed by Microgenics Corporation for
Total Mycophenolic Acid Assay on COBAS Integra use with regular chemistry analyzers. The CEDIA method
System utilizes enzyme fragment complementation of the
Recently a novel automated method for analysis of total genetically engineered Escherichia colis -galactosidase.
MPA was developed and approved by the U.S. Food and Two fragments of the enzyme, enzyme acceptor (EA) and
Drug Administration, using Roche Diagnostics Cobas enzyme donor (ED), are capable of combining to form the
Integra platform [41]. The method involves use of the active enzyme [44].
genetically engineered inosine-5-monophosphate
dehydrogenase (IMPDH) reagent, which is the direct target In the mycophenolic acid assay, the ED is conjugated with
enzyme of MPA. MPA. Similarly to the EMIT assay, the anti-MPA antibody
is introduced into the system and mixed with the sample.
The IMPDH catalyses the following reaction: IMP + NAD The mycophenolic acid and the conjugate of ED-MPA
+ H2O XMP + NADH + H+. First IMP binds to compete for the binding sites on the antibody. Whenever
IMPDH, which is followed by binding of NAD cofactor. the conjugate is bound to the antibody, it loses its ability to
Then reduced cofactor NADH is released, and the product combine with EA, and therefore no active enzyme is
xanthosine monophosphate (XMP) is released formed, resulting in decreased substrate cleavage activity
subsequently. This reaction is the rate-limiting step in in the sample. High enzyme activity results in a color
guanine nucleotide synthesis, a critical step in the change, which is directly proportional to the content of
proliferation of lymphocytes. Since mycophenolic acid MPA. This assay uses only two calibrators, 0 and 10 mg/L
inhibits the IMPDH directly, this method relies in fact on of MPA, and three QC samples, 1.1, 2.6, and 5.9 mg/L.
the measurement of direct immunosuppressive effect of
MPA. Therefore, there is no need for antibody or the The CEDIA assay has been used for assay of cyclosporin,
enzyme-labeled-MPA, as compared to the EMIT. tacrolimus, sirolimus, and everolimus for some time,
though it is a relatively new assay for mycophenolic acid.
Similarly to the EMIT method, the rate of formation of As of June 2007, only one paper on MPA analysis utilizing
NADH is measured by monitoring the change in this method has been published [45]. The within-run
absorption at a wavelength of 340 nm. Mycophenolic acid precision CV values did not exceed 5% for all QCs, while
binds to the active site of IMPDH uncompetitively and between-run precision was within 6%.
inhibits the above reaction, thus the concentration of MPA
in a sample is inversely proportional to the absorbance of Owing to the use of antibody, the CEDIA method is prone
NADH. The Cobas Integra assay for mycophenolic acid to the same deficiency as the EMIT method, that is, cross-
uses six calibrators, and the measuring range is 0.4 to 15 reactivity with AcMPAG, and possibly also other
mg/L of MPA. Comparison to the HPLC assay yielded the compounds in the sample. One of the studies found that the
following Passing/Bablok regression statistics for total degree of overestimation is dependent upon transplant
MPA: slope = 0.984, intercept = 0.096, r = 0.993 [42], or type. The bias for CEDIA versus HPLC/UV results
slope between 1.01 and 1.10, intercept between 0.12 and amounted to 44% for renal, 47% for lung, and 54% for
+0.068 mg/L and r > 0.99 from a multicenter trial [43]. heart transplant patients [45], which is comparable to the
reported EMIT bias.
This immunoassay method was validated in 4 independent Reference and Preferred Methods
laboratories in Europe and the United States. Comparison There is no formally established reference method of
885
Mycophenolic Acid
analysis of MPA, but the HPLC/UV method is commonly mycophenolate mofetil (MMF) by Roche, and as enteric-
perceived as a gold-standard method. It has served as a coated mycophenolate sodium salt (Myfortic) by Novartis.
widely used analytical method in the phase III pivotal trials After oral administration, MMF is readily hydrolyzed to
of mycophenolate mofetil in renal transplant patients, as MPA and rapidly absorbed, starting in the stomach.
well as in many other studies that included detailed Myfortic is mainly absorbed in the small intestine because
pharmacokinetic analysis of MPA. Despite the popularity of the enteric coating that reduces MPA absorption in the
of the EMIT method, certain limitations, including stomach. The median time to maximum MPA plasma
dynamic-range limitations and the mentioned positive bias concentration is between 0.25 and 1.25 hours for MMF
from MPA metabolites, have precluded this method from and 1.5 and 2.75 hours for Myfortic [32]. MPA is
being considered a gold-standard method. HPLC with extensively bound to plasma albumin (around 97% to
mass spectrometric detection is growing in popularity, 99%). Only the remaining free fraction of MPA of about
although it is not as widely used as the EMIT method 1% to 3% is pharmacologically active in in-vitro test
because of the high initial instrument cost. In addition, systems [22].
mass spectrometric methods offer little advantage over
ultraviolet detection when only total MPA and MPAG MPA is metabolized in the liver, gut, and kidneys by
measurements are of concern. uridine-diphosphate glucuronyl transferase enzymes,
mainly to the pharmacologically inactive glucuronide
Specimen metabolite, MPAG. At steady state, MPAG can be present
Mycophenolic acid is most commonly measured in plasma in blood plasma at concentrations as much as 100-fold
or serum. Some MS/MS methods for analysis of multiple higher than MPA. Two other metabolites, the acyl-
immunosuppressant drugs involve MPA analysis in whole glucuronide (AcMPAG) and the glucoside (MPAGc), are
blood; however, the standard of practice is to utilize also formed in smaller quantities. The AcMPAG
plasma or serum concentrations of MPA, and there is a metabolite was found in in-vitro test systems to be
lack of reference pharmacokinetic information for whole- pharmacologically active and of a potency similar to MPA
blood MPA concentrations. [47]. The metabolites are excreted mainly into urine and in
smaller quantities into bile. The excretion into bile allows
Samples should be collected and processed within 2 hours for the process of enterohepatic recycling of mycophenolic
following phlebotomy. Plasma is prepared from whole acid, whereby the MPAG that enters the gastrointestinal
blood by centrifugation. Plasma samples can be stored tract via bile is converted back to MPA by the hydrolytic
frozen without any loss in either total or free MPA and action of glucuronidase shed by enteric bacteria [48]. For
MPAG concentrations for at least one year at 24C. Due this reason, a second peak frequently appears in the MPA
to the instability of AcMPAG, it is recommended that the concentration versus time plot, usually within 4 to 10
samples are frozen at 80C and processed in the shortest hours after drug administration. This also accounts for a
time possible whenever this metabolite is of concern. great between- and within-patient variability of the area
Samples for which only the total level of analytes is of under the time-concentration curve (AUC) of MPA.
interest can be acidified prior to freezing, which stabilizes
AcMPAG concentration for at least 3 months of storage at Leukopenia is the main undesired effect of over-
24C [24]. However, when the free concentration and immunosuppression by mycophenolic acid, due to
free fraction of the analytes is of interest, acidification is bone-marrow suppression. Some other less severe but
not recommended, except for a separate aliquot of the more commonly occurring side effects of MPA include
primary sample, because it alters the binding between gastrointestinal disturbances, mainly diarrhea, and
proteins and mycophenolic acid. increased risk of viral infection, especially
cytomegalovirus [49].
Therapeutic Range and Pharmacokinetics
Mycophenolic acid acts by inhibiting inosine-5- The general variability in the pharmacokinetics of
monophosphate dehydrogenase (IMPDH)the enzyme mycophenolic acid and its relatively narrow therapeutic
essential in de novo guanosine nucleotide synthesis. T and window led to the recommendation of applying therapeutic
B lymphocytes critically depend on the de novo synthesis drug monitoring for assessment of the true exposure to this
of purines for their proliferative response, whereas other drug and dose modification. Typically the total content of
types of cells can use salvage pathways for purine MPA is used in monitoring; free MPA concentrations are
production. Thus MPA has a selective effect on the under investigation for their clinical relevance. The
proliferation of lymphocytes. In addition, mycophenolic therapeutic range of mycophenolic acid depends upon
acid suppresses antibody formation by B lymphocytes, whether there is concomitant use of other
resulting in a potent immunosuppressive effect [46]. immunosuppressants. Table 6 summarizes commonly used
MPA is currently marketed in two forms: as regimens and their respective MPA therapeutic ranges:
886
Mycophenolic Acid
Table 6: Therapeutic Ranges of MPA for Different Immunosuppressive Regimens [50]

Coadministered Immunosuppressant Target Range of MPA Trough Target Range of MPA AUC
[mg/L] [mgh/L]
Tacrolimus 1.9 4.0

30 60
Cyclosporine 1.0 3.5
The clinical utility of the 30 to 60 mg*h/L MPA AUC DC, Gerzon K, Gordee RS et al. Fermentation,
target range in the above table was derived from isolation, and biological properties of
investigations in renal transplant patients who received mycophenolic acid: antimicrobial Agents
concomitant cyclosporine and corticosteroid Chemother 1968;8:229-33.
immunosuppression [50]. The efficacy (freedom from [9] Noto T, Harada Y, Koyama K. A turbidimetric
acute early rejection) was confirmed in a recent bioassay method for determination of
prospective concentration-control investigation [51] mycophenolic acid. J Antibiot (Tokyo)
comparing a Bayesian dose adjustment targeting an 1970;23:96-8.
estimated MPA AUC of 45 mgh/L with empiric dosing of [10] Gainer FE, Hussey RL. Thin-layer
MMF in renal transplant patients who received MMF + chromatography of mycophenolic acid and
cyclosporine + corticosteroid and in a percentage of related compounds. J Chromatogr 1971;54:446-
patients, early exposure to monoclonal interleukin 2 8.
receptor antibody therapy. Two prospective therapeutic [11] Bopp RJ, Schirmer RE, Meyers DB.
drug monitoring trials that compare active monitoring of Determination of mycophenolic acid and its
MPA estimated AUC (FDCC trial) or active monitoring of glucuronide metabolite in plasma. J Pharm Sci
MPA trough concentration with a control arm with no 1972;61(11):1750-3.
active monitoring are under analysis and will hopefully, [12] Lee WA, Gu L, Miksztal AR, Chu N, Leung K,
together with the French Adaptation de Posologie du Nelson PH. Bioavailability improvement of
MMF en Greffe Rnale (APOMYGRE) study, provide the mycophenolic acid through amino ester
basis for further definition of therapeutic drug monitoring derivatization. Pharm Res 1990;7:161-6.
strategies in contemporary clinical practice [52]. [13] Sugioka N, Odani H, Ohta T, Kishimoto H,
Yasumura T, Takada K. Determination of a new
References immunosuppressant, mycophenolate mofetil,
[1] Shapiro R, Young JB, Milford EL, Trotter JF, and its active metabolite, mycophenolic acid, in
Bustami RT, Leichtman AB. rat and human body fluids by high-performance
Immunosuppression: evolution in practice and liquid chromatography. J Chromatogr B
trends, 1993-2003. Am J Transpl 2005;5: 874- 1994;654:249-56.
86. [14] Westley IS, Sallustio BC, Morris RG.
[2] Aspreva website. Validation of a high-performance liquid
<http://www.aspreva.com/clinical_trials.php> chromatography method for the measurement of
Accessed Jan 2008. mycophenolic acid and its glucuronide
[3] Fulton B, Markham A. Mycophenolate mofetil: metabolites in plasma. Clin Biochem
a review of its pharmacodynamic and 2005;38:824-29.
pharmacokinetic properties and clinical efficacy [15] Streit F, Shipkova M, Armstrong VW, Oellerich
in renal transplantation. Drugs 1996;51:278-98. M. Validation of a rapid and sensitive liquid
[4] Allison AC, Eugui EM. Mycophenolate mofetil chromatography-tandem mass spectrometry
and its mechanisms of action. method for free and total mycophenolic acid.
Immunopharmacology 2000;47:85-118. Clin Chem 2004;50:152-9.
[5] Ginzler EM, Dooley MA, Aranow C, Kim MY, [16] Zambonin CG, Aresta A, Palmisano F.
Buyon J, Merrill JT et al. Mycophenolate Determination of the immunosuppressant
mofetil or intravenous cyclophosphamide for mycophenolic acid in human serum by solid-
lupus nephritis. N Engl J Med 2005;353:2219- phase microextraction coupled to liquid
28. chromatography. J Chromatogr B 2004;806:89-
[6] ASI website. 93.
<http://www.bioanalytics.co.uk/html/latest_resu [17] Daurel-Receveur M, Titier K, Picard S, Ducint
lts.html, MPA report #42> Accessed Jan 2008. D, Moore N, Molimard M. Fully automated
[7] Klee G, ed. Participant Results, Surveys 2007. analytical method for mycophenolic acid
Washington, DC: College of American quantification in human plasma using on-line
Pathologists; 2007. solid-phase extraction and high-performance
[8] Williams BH, Boeck LD, Cline JC, DeLong liquid chromatography with diode array
887
Mycophenolic Acid
detection. Ther Drug Monit 2006;28:505-511. glucuronide metabolites. Clin Chem
[18] Patel ChG, Akhlaghi F. High-performance 2006;52:1962-1964.
liquid chromatography method for the [29] Platzer M, Jahn K, Wohlrab J, Neubert RHH.
determination of mycophenolic acid and its acyl Quantification of mycophenolate mofetil in
and phenol glucuronide metabolites in human human skin extracts using high-performance
plasma. Ther Drug Monit 2006;28:116-122. liquid chromatographyelectrospray mass
[19] Yin J, Wang S, Yang Gu, Yang Ge, Chen Y. spectrometry. J Chromatogr B 2001;755:355-
Molecularly imprinted solid-phase extraction 359.
for rapid screening of mycophenolic acid in [30] Vogeser M, Zachoval R, Spoehrer U, Jacob K.
human plasma. J Chromatog B 2006;844:142- Potential lack of specificity using electrospray
147. tandem-mass spectrometry for the analysis of
[20] Bahrami G, Mohammadi B. An isocratic high mycophenolic acid in serum. Ther Drug Monit
performance liquid chromatographic method for 2001;23:722-724.
quantification of mycophenolic acid and its [31] Premaud A, Rousseau A, Picard N, Marquet P.
glucuronide metabolite in human serum using Determination of mycophenolic acid plasma
liquid-liquid extraction: application to human levels in renal transplant recipients co-
pharmacokinetic studies. Clin Chim Acta administered sirolimus: comparison of an
2006;370:185-190. enzyme-multiplied immunoassay technique
[21] Teshima D, Kitagawa N, Otsubo K, Makino K, (EMIT) and liquid chromatographytandem
Itoh Y, Oishi R. Simple determination of mass spectrometry. Ther Drug Monit
mycophenolic acid in human serum by column- 2006;28:274-277.
switching high-performance liquid [32] Elbarbry FA, Shoker AS. Therapeutic drug
chromatography. J Chromatogr B 2002;780:21- measurement of mycophenolic acid derivatives
26. in transplant patients. Clin Biochem
[22] Nowak I, Shaw LM. Mycophenolic acid binding 2007;40:752-764.
to human serum albumin: characterization and [33] Hosotsubo H, Takahara S, Imamura R, Kyakuno
relation to pharmacodynamics. Clin Chem M, Tanaka T, Yazawa K et al. Analytic
1995;41:1011-1017. validation of the enzyme-multiplied
[23] Khoschsorur G, Erwa W. Liquid immunoassay technique for the determination of
chromatographic method for simultaneous mycophenolic acid in plasma from renal
determination of mycophenolic acid and its transplant recipients compared with a high-
phenol- and acylglucuronide metabolites in performance liquid chromatographic assay. Ther
plasma. J Chromatogr B 2004;799:355-360. Drug Monit 2001;23:669-674.
[24] Shipkova M, Schuetz E, Armstrong VW, [34] Shen J, Jiao Z, Yu YQ, Zhang M, Zhong MK.
Niedmann PD, Oellerich M, Wieland E. Quantification of total and free mycophenolic
Determination of the acyl glucuronide acid in human plasma by liquid chromatography
metabolite of mycophenolic acid in human with fluorescence detection. J Chromatogr B
plasma by HPLC and EMIT. Clin Chem Analyt Technol Biomed Life Sci 2005;817:207-
2000;46:365-372. 213.
[25] Wai-Ping Y, Vathsala A, Huei-Xin L, Shu Feng [35] Shaw LM, Korecka M, Van Breeman R, Nowak
Z, Eli C. Simple reversed-phase liquid I, Brayman KL. Analysis, pharmacokinetics and
chromatographic assay for simultaneous therapeutic drug monitoring of mycophenolic
quantification of free mycophenolic acid and its acid. Clin Biochem 1999;31:323-328.
glucuronide metabolite in human plasma. J [36] Bogusz MJ, Enazi EA, Hassan H, Abdel-Jawaad
Chromatogr B 2007;846:313-318. J, Ruwaily JA, Tufail MA. Simultaneous LC-
[26] Annesley TM, Clayton LT. Quantification of MS-MS determination of cyclosporine A,
mycophenolic acid and glucuronide metabolite tacrolimus, and sirolimus in whole blood as well
in human serum by HPLC tandem mass as mycophenolic acid in plasma using common
spectrometry. Clin Chem 2005;51:872-877. pretreatment procedure. J Chromatogr B Analyt
[27] Figurski MJ, Korecka M, Fields L, Waligrska Technol Biomed Life Sci 2007; 850(1-2):471-
T, Shaw LM. HPLC-MS/MS method for 480. Epub Jan 2007.
simultaneous quantification of total or free [37] Ceglarek U, Casetta B, Lembcke J, Baumann S,
fraction of mycophenolic acid and its Fiedler GM, Thiery J. Inclusion of MPA and in
glucuronide metabolites. Ther Drug Monit a rapid multi-drug LC-tandem mass
2007;29:511-512. spectrometric method for simultaneous
[28] Brandhorst G, Streit F, Goetze S, Oellerich M, determination of immunosuppressants. Clin
Armstrong VW. Quantification by liquid Chim Acta 2006;373(1-2):168-171.
chromatography tandem mass spectrometry of [38] Premaud A, Rousseau A, Le Meur Y, Lachatre
mycophenolic acid and its phenol and acyl G, Marquet P. Comparison of liquid
888
Mycophenolic Acid
chromatographytandem mass spectrometry mycophenolate mofetil. Ther Drug Monit 1995;17:681-
with a commercial enzyme-multiplied 684.
immunoassay for the determination of plasma [47] Shipkova M, Armstrong VW, Wieland E,
MPA in renal transplant recipients and Niedmann PD, Schtz E, Brenner-Weiss G et al.
consequences for therapeutic drug monitoring. Identification of glucoside and carboxyl-linked
Ther Drug Monit 2004;26:609-619. glucuronide conjugates of mycophenolic acid in
[39] US Patent 6225073. plasma of transplant recipients treated with
[40] Weber LT, Shipkova M, Armstrong VW, mycophenolate mofetil. Br J Pharmacol
Wagner N, Schuetz E, Mehls O et al. 1999;126:1075-1082.
Comparison of the emit immunoassay with [48] Bullingham R, Monroe S, Nicholls A, Hale M.
HPLC for therapeutic drug monitoring of Pharmacokinetics and bioavailability of
mycophenolic acid in pediatric renal transplant mycophenolate mofetil in healthy subjects after
recipients on mycophenolate Mofetil Therapy. single dose oral and intravenous administration.
Clin Chem 2002;48:517-525. J Clin Pharmacol 1996;36:315-324.
[41] US Patent 6107052. [49] Mourad M, Malaise J, Eddour DCh, De Meyer
[42] Dorn AR, Tsai J, Mountain L, Phillips M, M, Koenig J, Schepers R et al. Correlations of
Rugaber J, Zhou J et al. Assessment of total and mycophenolic acid pharmacokinetic parameters
free mycophenolic acid in transplant patient with side effects in kidney transplant patients
plasma by enzyme receptor assay. Ther Drug treated with mycophenolate mofetil. Clin Chem
Monit 2005;27:216-217. 2001;47:88-94.
[43] Luthe H, Figurski M, Korecka M, Marquet P, [50] Van Gelder T, Le Meur Y, Shaw LM, Oellerich
Liebisch G, Coffing MJ et al. Multicenter M, DeNofrio D, Holt C et al. Therapeutic drug
evaluation of the Roche total mycophenolic acid monitoring of mycophenolate mofetil in
assay on COBAS Integra Systems. Abstracts transplantation. Ther Drug Monit 2006;28:145-
of the AACC Annual Meeting, July 15-19, 154.
2007, San Diego, CA, USA. Clin Chem [51] Le Meur Y, Bchler M, Thierry A, Caillard S,
2007;53(6 Suppl):A2-232. Villemain F, Lavaud S et al. Individualized
[44] DA, Hubster EC, Kaufman MS, Ramon MK. mycophenolate mofetil dosing based on drug
Cloned-enzyme donor immunoassay (CEDIA) exposure significantly improves patient
for drugs-of-abuse screening. Clin Chem outcomes after renal transplantation. Am J
1995;941:92-98. Transplant 2007;7:2496-2503.
[45] Westley IS, Ray JE, Morris RG. CEDIA [52] Van Gelder T, Shaw LM. The rationale for and
mycophenolic acid assay compared with HPLC- limitations of therapeutic drug monitoring for
UV in specimens from transplant recipients. mycophenolate mofetil in transplantation.
Ther Drug Monit 2006;28:632-636. Transplantation 2005;80(2 Suppl):S244-53.
[46] Ransom JT. Mechanism of action of
Figures
PUMP B PUMP A PUMP B PUMP A PUMP B PUMP A PUMP B PUMP A
(ELUTING) (WASHING) (ELUTING) (WASHING) (ELUTING) (WASHING) (ELUTING) (WASHING)
AUTO- AUTO- AUTO- AUTO-

SAMPLER SAMPLER SAMPLER SAMPLER
1 1 1 1
10 2 10 2 10 2 10 2
9 9
COLUMN
COLUMN
3 3
WASH
3 3
WASH
9 9
Waste Waste
A Waste B A Waste B
8 4 8 4 8 4 8 4
WASH WASH
COLUMN COLUMN
7 5 7 5 7 5 7 5
6 6 6 6
THERMOSTAT: 50C THERMOSTAT: 50C THERMOSTAT: 50C THERMOSTAT: 50C

ANALYTICAL COLUMN ANALYTICAL COLUMN ANALYTICAL COLUMN ANALYTICAL COLUMN
MS/MS MS/MS MS/MS MS/MS
Figure 1. Switching valve layout for across-directional (left) and uni-directional (right)
flow of mobile phase through working column.
889
Mycophenolic Acid
14
12
10
mAU
6
220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
Wavelength [nm]
Figure 1: The UV spectrum of MPA, from [16]
275
193 207
Figure 3: Most common MS/MS
fragments of the mycophenolic acid
molecule.
NAD MPA PA S NAD

An M NADH
tib
od En
Antibody
zym
o dy
y
En
M Enzym
e
zy
PA e
m
Antib
e
Enzym
e
Antibody
Pr
D
S
NA
S
Enzym
e S
NAD
ody
NAD
Antib
dy
Antibody
e
bo E nzym NAD
MPA ti
An
Antibody
M
PA
NAD S
S
Figure 2: Schematic representation of EMIT principlebased on EMIT education materials from Dade-Behring.
890
Mycophenolic Acid
50000
45000 MPA
MPAG peak
MPAC Figure 4: Magnified portion of MS/MS
MPAG
40000
MPAC (IS) peak chromatogram of a free MPA sample, showing
35000
additional MPA peak, that originated from in-
source decomposition of MPAG.
30000 MPA peak
[counts]
25000
MPAG decomposed
in-source to MPA
20000
15000
AcMPAG peak
10000
5000
0
4 4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6
Time [min]
30
Y = 1.06 X 0.03; R2 = 0.96
25
Integra 400 [mg/L of MPA]
20
15
10
0
0 5 10 15 20 25 30
HPLC [mg/L of MPA]
Figure 5: Comparison of total MPA concentration

measurements by Integra 400 assay versus HPLC.
891
Mycophenolic Acid
List of Abbreviations Used in This Chapter

Acronym Explanation
AcMPAG Mycophenolic acid acyl-glucuronide
APOMYGRE Adaptation de Posologie du MMF en Greffe Rnale Study
AUC Area under the time-concentration curve
CAP College of American Pathologists
CAS Chemical Abstracts Service
CEDIA Cloned-enzyme donor immunoassay
CV Coefficient of variation
EA Enzyme acceptor
ED Enzyme donor
EMIT Enzyme-multiplied immunoassay technique
FDA U.S. Food and Drug Administration
FDCC Fixed Dose/Concentration Control study
HPLC High-performance liquid chromatography
IMP Inosine-5-monophosphate
IMPDH Inosine-5-monophosphate dehydrogenase (enzyme)
IUPAC International Union of Pure and Applied Chemistry
IS Internal standard
LC/MS Liquid chromatography/mass spectrometry
LLE Liquid-liquid extraction
LLOQ Lower limit of quantification
LOD Limit of detection
MMF Mycophenolate mofetil
MPA Mycophenolic acid
MPAG Mycophenolic acid glucuronide
MPAC Carboxybutoxy ether of mycophenolic acid
MS Mass spectrometer
MS/MS Tandem mass spectrometer
NAD Nicotinamide adenine dinucleotide
NADH Nicotinamide adenine dinucleotide - reduced
PP Protein precipitation
QC Quality control
SPE Solid-phase extraction
UF Ultrafiltration
UV Ultraviolet
XMP Xanthosine monophosphate
892
Myoglobin
Myoglobin
Alan H.B. Wu
Name: Myoglobin
Chemical class: Heme protein
Structure: See diagram below
i such as (1) spectrophotometry, (2) electrophoresis, (3)

chromatography, and (4) ultracentrifugation have been
Myoglobin is a heme protein that is reddish-brown. A
blood or urine sample turns this color when myoglobin is
present at a concentration exceeding 1 g/L. The visual developed to more specifically demonstrate the presence
examination of urine samples from patients with muscle of myoglobin in body fluids but are used only rarely.
disorders has been used in the past as a rapid bedside
test for the presence of myoglobin. However, most In urine, myoglobin can be detected using the pad for
patients with muscle disease have far less myoglobin blood on a urinalysis dipstick test (Table 1, Method 1).
present in the urine. Many physicochemical methods The heme group from free hemoglobin and myoglobin
catalyses the oxidation of tetramethylbenzidine, an
i indicator that turns the dipstick pad green. The pad is
Myoglobin
read visually against a color chart to determine the
semiquantitative degree of heme/myoglobulinuria (1+,
2+, etc). The urine sample can be pretreated with the
Methods edition: I-Wen Chen
addition of saturated ammonium sulfate, which will
salt-out the hemoglobin. After centrifugation, the
supernatant of the residual sample is retested with a
Fourth edition: I-Wen Chen
dipstick for blood, and a positive reaction is due to
Fifth edition: Alan H.B. Wu
893
Myoglobin
myoglobin. However, this procedure does not hemolytic complement by myoglobin-antimyoglobin

completely separate out hemoglobin from myoglobin, complexes. The amount of myoglobin present in the
therefore a false-positive result can be obtained in the sample is inversely proportional to the extent of
presence of significant hemoglobinuria [1]. Use of a hemolysis of the hemolysin-sensitized sheep cells used
microconcentrator membrane offers a more complete as the indicator system. Quantitative microcomplement
separation of myoglobin from hemoglobin and increased fixation is an extremely time-consuming and laborious
specificity for myoglobin detection [2]. technique. Few laboratories have mastered its intricacies,
and thus it is rarely used. In addition, its sensitivity of
Qualitative and semiquantitative immunological 300 mg/L is not adequate.
methods have also been used.
Counterimmunoelectrophoresis, with the simplicity of Currently, the determination of serum myoglobin for
the agar gel precipitin techniques and the speed of diagnosis of AMI and other muscle disorders is
electrophoresis, has been used to detect myoglobin in performed almost exclusively by immunoassay
both serum and urine (Table 1, Method 2) [3]. techniques because of their high analytical sensitivity,
Hemagglutination-inhibition assays use myoglobin- specificity, precision, and rapid turnaround time.
sensitized red blood cells (myoglobin chemically Radioimmunoassay (RIA) procedures have also been
attached to the membrane of the red blood cell). described for quantitative measurement of serum
Agglutination takes place when these red blood cells are myoglobin (Table 1, Method 6) [7]. However, in clinical
mixed with the antimyoglobin antibody (Table 1, laboratories, RIA has largely been replaced by
Method 3). When a sample containing myoglobin is automated 2-site non-isotopic immunoassays ( Table 1,
added, myoglobin competes with the red blood cell for Method 7). The capture antibody is linked to a bead or
myoglobin-binding sites on the antibody, and paramagnetic particle. The detecting antibody is linked
hemagglutination is inhibited [4]. to a tag that can produce a nephelometric, turbidimetric,
Counterimmunoelectrophoresis is a qualitative fluorometric, or chemiluminescence signal. Automated
procedure and has a sensitivity of only 2 mg/L. It is thus immunoassay analyzers perform a separation between
unsuitable for detection of acute myocardial infarction bound and free labels to produce a quantitative result.
(AMI).
Qualitative and quantitative methods for serum
A rapid, qualitative latex agglutination test for detection myoglobin have also been developed for point-of-care
of increased concentrations of myoglobin in serum has (POC) use (Table 1, Method 8) [8,9]. One POC method
been reported (Table 1, Method 4) [5]. In this test utilizes a self-calibrating immunoassay system whereby
procedure, any interfering rheumatoid factor present in whole blood is separated from plasma that reacts with
the sample is removed when 50 L of test serum is fluorescent antibody conjugates within a reaction
mixed with 5 L of rheumatoid factor absorption chamber [8]. Following an incubation step, the reaction
medium on a dark slide. About 50 L of a suspension of mixture flows down a detection lane by capillary action
latex particles with bound antimyoglobin antibodies is and the myoglobinfluorescent antibody complex is
then added and mixed with the serum by gentle tilting captured by a second antibody in a discrete measuring
for 5 min. Agglutination of the particles occurs in the zone. The intensity of the label (visual or fluorescence)
presence of myoglobin. A negative exponential is related to myoglobin concentration. The procedure
relationship has been found between the time of takes approximately 15 min to complete. Given that
agglutination and myoglobin concentrations. Myoglobin myoglobin measurements are most commonly used in
concentration was considered to be greater than 400 the context of diagnosis and exclusion of AMI, several
mg/L if agglutination occurred within 1 min, 150 to 400 POC devices have combined myoglobin with other
mg/L if agglutination time was between 1 and 2 min, 80 cardiac markers such as CK-MB and cardiac troponin
to 150 mg/L if agglutination time was between 2 and 3 into a single cartridge or slide [10,11]. International
min, and less than 80 mg/L (upper limit of reference guidelines such as those from the National Academy of
interval) if more than 3 min were required for Clinical Biochemistry have suggested that results of
agglutination. Red blood cell agglutination techniques cardiac markers should be reported within 1 hour of
are reported to be sensitive in the range of 50 to 100 blood collection [12]. Other groups such as the
mg/L [4]. A disadvantage of the technique is that some American Heart Association have suggested an optimal
sera containing a high concentration of rheumatoid turnaround time of less than 30 minutes [13]. These POC
factors gave positive agglutination reactions. The latex devices are designed for use at the patient bedside or
agglutination procedure is at best semiquantitative and is emergency department satellite laboratories to produce
somewhat subjective, requiring visual observation to results that can meet these groups recommended 30- to
establish an end point. The sensitivity of the latex test 60-minute turnaround time goals for reporting test
may not be adequate. The advantage of the latex test is results.
its simplicity.
A quantitative microcomplement fixation assay has been There are currently no reference methods or materials
developed to detect myoglobin in both the serum and for myoglobin in serum or urine. The preferred method
urine of patients with acute coronary syndrome or is a two-site sandwich immunoassay. Unfortunately,
myopathies [6] (Table 1, Method 5). This technique is these assays are not standardized to each other, and each
based on the consumption of a fixed amount of method has slightly different reference ranges.
894
Myoglobin
The chemical and physicochemical techniques to Interferences

measure myoglobin (e.g., spectrophotometry, The use of collection tubes with separator gels has been
electrophoresis, chromatography, and reported to both increase and decrease results [16].
ultracentrifugation) were not capable of measuring the Bilirubin has been found not to interfere. Rheumatoid
submicrogram range of myoglobin frequently found in factor (RF) has been found to affect results in some
certain muscle disorders, such as AMI or myositis, and assays, with a lack of concordance between
are thus not currently used. Immunoassays are more concentrations of RF and the magnitude of interference.
sensitive than the above techniques. However, a As with any 2-site immunoassays that use monoclonal
prerequisite of any of the immunochemical methods is antibodies, the presence of human anti-mouse antibodies
that the antisera be specific for myoglobin. Specific can potentially cause a false-positive result.
antisera used in such methods should not cross-react
appreciably with hemoglobin or other constituents of Myoglobin Reference Intervals
body fluids and should be capable of detecting Because myoglobin is the major protein in striated and
myoglobin to the submicrogram range. myocardial muscle and because muscle mass varies with
sex, age, and race, serum myoglobin levels may show
A major issue for myoglobin measurements is the lack of age, sex, and race dependence (Table 2). Results of
standardization among commercial manufacturers of measurements of serum myoglobin concentrations in
myoglobin assays. There has resulted in significant 292 ostensibly healthy subjects using a sensitive RIA
proportional biases in myoglobin results between revealed significantly higher myoglobin concentrations
manufacturers. The International Federation of Clinical in males than in females for both black and white
Chemistrys Committee on Standardization of Markers subjects [19]. Myoglobin concentrations in black males
of Cardiac Damage has begun a program to standardize were significantly higher than those in white males,
myoglobin measurements. Candidate secondary whereas no racial differences were observed in the
reference materials from various manufacturers have female population. Myoglobin concentrations were
been evaluated [14]. Calibrating 12 different automated higher in the older population in all groups except that of
testing platforms for myoglobin using a lyophilized black males, who showed no age difference. Reference
candidate material originating from human heart have intervals established with the use of various enzyme
reduced the bias for a human serum pool from 32% to immunoassay procedures showed that the 95% central
13%. However, until there is an accepted reference range of the normal distribution for males and females
standard that has been adopted by all manufacturers, varied significantly between different procedures. The
complete standardization of myoglobin will not be 97.5th percentile value ranged from 67 to 86 g/L for
possible. The lack of a reference method for myoglobin males and 50 to 63 g/L for females [16]. The age, sex,
further complicates this issue. As myoglobin is a small and race dependence of myoglobin concentrations can be
protein, it is likely that some mass-spectrometric method partially eliminated by correction of the myoglobin
will be needed. values for body mass using serum creatinine levels
(Table 2) [20].
Specimen
The preferred specimen is nonhemolyzed, nonlipemic Interpretation
serum. One study found no significant difference Myoglobin is a relatively small globular protein
between heparin plasma and serum, although EDTA composed of a single polypeptide chain of 153 amino
plasma samples were significantly lower than serum acids and an iron-containing heme prosthetic group
[15]. Another study found that different myoglobin identical to that of hemoglobin. It is found in skeletal
assays can show significant interference from different and cardiac muscle cells and is capable of reversibly
anticoagulants [16]. Each method should be investigated binding molecular oxygen and enhancing the rate of
with respect to the differences observed between serum oxygen diffusion through the muscle cell. Myoglobin is
and the use of various anticoagulants. an oxygen carrier in the cytoplasm and appears to be
responsible for the transport of oxygen from the muscle
Specimens are stable at 4C for 1 week and for up to 4 cell membrane to the mitochondrion. Myoglobins from
weeks at 20C [17]. Urine samples should be collected skeletal and cardiac muscles are immunologically
without preservative and assayed within 24 h of sample identical and thus cannot be distinguished from each
collection. If the test is not performed immediately after other by immunological methods.
sample collection, the urine sample should be kept at
2C to 8C and centrifuged to remove any debris before Myoglobin may leak from muscle tissue into the blood
the assay. Myoglobin in urine is stable under alkaline circulation as a result of damage to skeletal or cardiac
conditions (pH 9.0) for many days at 2C, 20C, and muscle and subsequently may appear in the urine
70C; sodium hydroxide can be added if long-term because of its relatively low molecular mass.
storage is needed [18]. The analyte is not stable under Myoglobinemia and myoglobinuria have been used in
acidic conditions (e.g., pH 5.5). Most commercial the diagnosis of myopathies and cardiomyopathies. A
methods for myoglobin are not cleared by the U.S. Food pronounced increase in serum myoglobin is an early
and Drug Administration for use in urine, and quantitative indicator of AMI. It has been reported that
laboratories must validate this specimen type prior to myoglobinemia is a more sensitive clinical indicator of
clinical use. AMI than increased activities or concentrations of serum
creatine kinase-MB (CK-MB) isoenzyme, especially
895
Myoglobin
when blood samples are drawn in the early stage of the serum CK and myoglobin. Patients with rhabdomyolysis
disease (e.g., within 3 hours after onset) [21]. However, and myoglobinemia are at risk for development of acute
because no assay can differentiate myocardial from renal failure [26]. Treatment of these patients with
skeletal muscle myoglobin, these assays are less specific. diuretics and alkalinization has been shown to minimize
As such, myoglobin assays for diagnosis of AMI have the risk of renal failure [27]. The aqueous stability of
not been widely used in the United States but have been myoglobin under basic pH conditions may facilitate its
used in some European countries [22]. Increased serum clearance through the kidneys. Calculation of myoglobin
myoglobin concentrations return to normal sooner than clearance has been suggested as a mechanism to predict
total CK and the CK-MB isoenzyme in AMI. The faster patients risk for renal failure with rhabdomyolysis
elimination of myoglobin from the blood may be [18,28]. Patients with a high serum myoglobin and low
advantageous in the assessment of reinfarctions urinary clearance are at greatest risk. Some researchers
occurring during the period of increased enzyme and clinicians have experimented with treating patients
activities. Recurrent increases of serum myoglobin are with rhabdomyolysis by removing myoglobin using
thus suggestive of an extension of the original infarction. veno-venous hemofiltration [29,30]. However, this has
not progressed beyond the experimental therapy stage.
Since myoglobin is also present in skeletal muscle and is
cleared almost exclusively through glomerular filtration, Myoglobin Performance Goals
increases of myoglobin will occur after acute muscle An assessment of the biological variation determines
trauma, in acute or chronic renal failure, severe analytical performance goals [31]. These studies are
congestive heart failure, prolonged shock, and in patients conducted in healthy subjects and involve serial blood
with myopathies of a variety of causes. Therefore, the collections. The biological variability for serum
use of serum myoglobin levels to furnish early, myoglobin assays has been determined by Panteghini et
quantitative documentation of AMI should be limited to al. [32]. The analytical, (CVA), intraindividual (CVI),
patients who do not have these underlying or associated and interindividual (CVG) variation were 6.9%, 11%,
problems and who are within the first 18 h of the clinical and 14%, respectively. Based on these figures, the
onset of infarction. In addition, serum myoglobin levels analytical goals for imprecision at 1/2CVI was 5.6%.
have been used in the evaluation of neuromuscular The number of values needed to estimate the
diseases such as muscular dystrophy, muscular atrophy, homeostatic set point of an individual to within 5% was
and polymyositis [21]. Percy et al. reported that the 25. The index of individuality (CVI/CVG) was 0.8. This
combination of CK, hemopexin, and myoglobin indicates that the use of population-based reference
measurements provided significantly better detection of intervals are appropriate for this analyte [31]. The
Duchenne muscular dystrophy carriers than when these critical difference required for determining statistically
tests were used alone or in the combination of CK and significant change in serial results (reference change
hemopexin without myoglobin [23]. values [RCV]) was 35%. However, the RCV in the
Panteghini study was calculated using blood collected
In studies conducted on myopathies in children [24], over several weeks, and it is probably higher than the
serum myoglobin concentrations for subjects 0.6 to 20 expected RCV values that are calculated over several
years were lower than adults, so it is necessary to use hours, as would be applicable for use in patients
assays with a high sensitivity of 2 g/L or less. Mean suspected of AMI. Studies have documented relatively
myoglobin concentrations of 17.1 and 12.5 g/L were high imprecision and marked disagreement between the
obtained from 37 males and 29 females, respectively, various commercial immunoassay procedures available
and were statistically different. Racial differences were for serum myoglobin [16]. Total CVs for specimens
not investigated. Investigations in children with muscle within the normal reference interval ranged from 3.4% to
diseases, however, have been rather disappointing [24]. 11%. Many of these assays do not meet goals established
Only 2 out of 14 patients studied had increased serum by biological variation.
myoglobin concentrations. The patients evaluated in this
study were relatively young, ranging from 0.3 to 16 Bombardieri et al. [33] studied circadian variation of
years of age, and were presumably in the relatively early serum myoglobin concentration in normal subjects and
stages of disease. This might have contributed to the low in patients with polymyositis and found the highest
rate of hypermyoglobinemia in this group of patients. levels at 9 am and the lowest between 6 pm and
Edwards et al. also found that measurement of midnight in both groups. The percent differences
myoglobin offered no advantage over CK for the between the highest and lowest values could be as high
investigation of any aspect of Duchenne muscular as 66%. Therefore it is important to take into account the
dystrophy [24]. However, serum myoglobin time of sample collection when myoglobin levels are to
measurements were found to be helpful in detection of be used to assess the degree of disease activity or to
Duchenne muscular dystrophy carriers when used with monitor the response to treatment of muscle diseases
CK and hemopexin [25]. such as polymyositis. In these cases, it may be desirable
to have samples drawn for myoglobin at approximately
Myoglobin measurements are also used for diagnosis of the same time each day.
patients with rhabdomyolysis. At one time, the urinalysis
blood dipstick test was widely used to detect myoglobin
in urine. However, a more effective indicator is the
presence of very high activities and concentrations of
896
Myoglobin
References characteristics and utilization of biochemical

1 Adams EC. Differentiation of myoglobin and markers in acute coronary syndromes.
hemoglobin in biological fluids. Ann Clin Lab Circulation 2007; 115: e356-75.
Sci 1971; 1: 208-21. 13 Braunwald E, Antman EM, Beasley JW, Califf
2 Kelner MJ, Alexander NM. Rapid separation RM, Cheitlin MD, Hochman JS et al.
and identification of myoglobin and ACC/AHA Guidelines for the Management of
hemoglobin in urine by centrifugation through a Patients With Unstable Angina and NonST-
microconcentrator membrane. Clin Chem 1985; Segment Elevation Myocardial Infarction:
31: 112-4. Executive Summary and Recommendations : A
3 Hibrawi H, Garrison FD, Smith HJ. A Report of the American College of
comparison of various agarose preparation in Cardiology/American Heart Association Task
counter-immunoelectrophoresis (CIE) system Force on Practice Guidelines. Circulation 2000;
for assaying urinary myoglobin. J Immunol 102: 1193-1209.
Methods 1977; 14: 59-63. 14 Panteghini M, Linsinger T, Wu AHB, Dati F,
4 Cloonan MJ, Donald TG, Neale C, Wilcken Apple FS, Christenson RH, et al.
DEL. The detection of myoglobin in urine and Standardization of immunoassays for
its application to the diagnosis of myocardial measurement of myoglobin in serum. Phase I:
infarction. Pathol 1976; 8: 313-20. Evaluation of candidate secondary reference
5 Ntorregaard-Hansen K, Hangaard J, materials. Clin Chim Acta 2004; 341: 65-72.
Ntorgaard-Pedersen B. A rapid latex 15 Pagani F, Bonetti G, Stefini F, Cuccia C,
agglutination test for detection of elevated Panteghini M. Serum and plasma samples for
levels of myoglobin in serum and its value in ACS:Systems cardiac markers. Clin Chem
the early diagnosis of acute myocardial 2000; 46: 1020-2.
infarction. Scand J Clin Lab Invest 1984; 44: 16 Zaninotto M, Pagani F, Altinier S, Amboni P,
99-103. Bonora R, Dolci A. Multicenter evaluation of
6 Kagen LJ, Scheidt S, Butt A. Serum myoglobin five assays for myoglobin determination. Clin
in myocardial infarction: the staccato Chem 2000; 46: 1631-7.
phenomenon. Is acute myocardial infarction in 17 Carraro P, Plebani M, Varagnolo MC, Rossetti
man an intermittent event? Am J Med 1977; 62: M, Burlina A. A new immunoassay for the
86-92. measurement of myoglobin in serum. J Clin
7 Stone MJ, Willerson JT, Gomez-Sanchez CE, Lab Anal 1994; 8: 70-5.
Weterman MR. Radioimmunoassay of 18 Wu AHB, Laios I, Green S, Gornet TG, Wong
myoglobin human serum: results in patients SS, Parmaley L, et al. Immunoassays for serum
with acute myocardial infarction. J Clin Invest and urine myoglobin: myoglobin clearance
1975; 56: 1334-9. assessed as a risk factor for acute renal failure.
8 Muller-Bardorff M, Sylven C, Rasmanis G, Clin Chem 1994; 40: 796-802.
Jorgensen B, Collinson PO, Waldenhofer U, et 19 Chen I-W, David R, Maxon HR, Sperling M,
al. Evaluation of a point-of-care system for Stein EA. Age-, sex-, and race-related
quantitative determination of troponin T and differences in myoglobin concentrations in the
myoglobin. Clin Chem Lab Med 2000; 38: 567- serum of healthy persons. Clin Chem 1980; 26:
74. 1864-80.
9 Beneteau-Burnat B, Baudin B, Vaubourdolle 20 Maxon HR, Chen IW, Gerson M, Sperling M,
M. Evaluation of Stratus CS stat fluorimetric Stein EA. Diagnostic specificity of serum
analyser for measurement of cardiac markers myoglobin radioimmunoassay for acute
Troponin I (cTnI), creatine kinase MB (CK- myocardial infarction is improved by using age-
MB), and myoglobin. J Clin Lab Anal 2001; , sex-, and race-specific reference intervals.
15: 314-8. Clin Chem 1982; 28: 2179.
10 Apple FS, Christenson RH, Valdes R Jr., 21 Askmark H, Osterman P, Roxin L, Venge P.
Andriak AJ, Berg A, Duh SH, et al. Radioimmunoassay of serum myoglobin in
Simultaneous rapid measurement of whole neuromuscular diseases. J Neurol Neurosurg
blood cardiac myoglobin, creatine kinase MB, Psychiatry 1981; 4: 68-72.
and cardiac troponin I by the Triage Cardiac 22 Roxin LE, Cullhed I, Groth T, Hallgren T,
Panel for detection of myocardial infarction. Venge P. The value of serum myoglobin
Clin Chem 1999; 45: 199-205. determinations in the early diagnosis of acute
11 Young GP, Murthi P, Levitt MA, Gawad Y. myocardial infarction. Acta Med Scand 1984;
Serial use of bedside CKMB/myoglobin device 215: 417-25.
to detect acute myocardial infarction in 23 Percy ME, Pichora GA, Chang LS, Manchester
emergency department chest pain patients. J KE, Andrews DF. Serum myoglobin in
Emerg Med 1999; 17: 769-75. Duchenne muscular dystrophy carrier detection:
12 Morrow DA, Cannon CP, Jesse RL, Newby a comparison with creatinine kinase and
LK, Ravkilde J, Storrow AB, et al. National hemopexin using logistic discrimination. Am J
Academy of Clinical Biochemistry Laboratory Med Genet 1984; 18: 279-87.
Medicine Practice Guidelines: clinical
897
Myoglobin
24 Chen IW, Sperling MI, Iannoccone S, Maxon 29 Nicolau D, Feng YS, Wu AHB, Bernstein SP,
H, Gelfand M. A sensitive radioimmunoassay Nightingale CH. Myoglobin clearance during
for detection of serum myoglobin in children continuous veno-venous hemofiltration with or
for evaluation of myopathies. J Clin Chem Clin without dialysis. Int J Artif Organs 1998; 21:
Biochem 1981; 19: 636. 205-9.
25 Edwards RJ, Rodeck CH, Watts DC. The 30 Amyot SL, Leblanc M, Thibeault Y, Geadah D,
diagnostic value of plasma myoglobin levels in Cardinal J. Myoglobin clearance and removal
the adult and fetus at risk for Duchenne during continuous venovenous hemofiltration.
muscular dystrophy, J Neurol Sci 1984; 63: Inten Care Med 1999; 25: 1169-72.
173-82. 31 Fraser CG. Biological variation: from principles
26 Hamilton RW, Hoplins MB, Shihabi ZK. to practice. Washington DC: American
Myoglobinuria, hematoglobinuria, and acute Association for Clinical Chemistry Press, 2001.
renal failure. Clin Chem 1989; 35: 1713-20. 32 Panteghini M, Pagani F. Biological variation of
27 Eneas JF, Schoefeld PY, Humphreys MH. The mlyoglobin in serum. Clin Chem 1997; 43:
effect of infusion of mannitol-sodium 2435.
bicarbonate on the clinical course of 33 Bombardieri S, Clerico A, Riente L, Del Chicca
myoglobinuria. Arch Intern Med 1979; 139: MG, Vitali C. Circadian variations of serum
801-5. myoglobin levels in normal subjects and
28 Laios I, Caruk R, Wu AHB. Myoglobin patients with polymyositis. Arthritis Rheum
clearance as an early indicator for 1982; 25: 1419-24.
rhabdomyolysis-induced acute renal failure.
Ann Clin Lab Sci 1995; 25: 179-84.
898
Myoglobin
Table 1: Methods of Myoglobin Analysis

Method 1: Urinalysis dipstick analysis; qualitative
Principle of analysis: Oxidation of a dye by the heme group to form a color
Comments: Positive results also occur with hemoglobin. Ammonium sulfate precipitation followed by
centrifugation can separate hemoglobin from myoglobin, but incomplete separation can occur at high urine
hemoglobin concentrations causing a false-positive result for myoglobin.
Method 2: Counterimmunoelectrophoresis; qualitative

Principle of analysis: Serum or urine is placed in electric field. Anodically migrating myoglobin and cathodically
migrating antibody move through agarose toward each other. A precipitin band forms where they meet.
Comments: Rare; lacks sensitivity; convenient technique; rarely used
Method 3: Hemagglutination inhibition; semiquantitative

Principle of analysis: Red blood cells are coated with myoglobin; addition of antibodies to myoglobin causes red
blood cells to agglutinate. Presence of myoglobin inhibits red blood cell agglutination. Read visually.
Comments: Rare; lacks sensitivity; difficult quality control
Method 4: Latex agglutination; semiquantitative

Principle of analysis: Latex beads coated with antibody to myoglobin. Presence of myoglobin causes beads to
agglutinate. Read visually.
Comments: Common; lacks sensitivity; semiquantitative; individual reading variability; easy to perform. Has
been replaced by immunoassays.
Method 5: Complement fixation; quantitative

Principle of analysis: Hemolysin-sensitized sheep red blood cells compete with myoglobinantimyoglobin
complex for complement. Extent of RBC lysis is inverse of myoglobin concentration
Comments: Rare; lacks sensitivity; very difficult to perform consistently. Has been replaced by immunoassays.
Method 6: Radioimmunoassay (RIA); quantitative

Principle of analysis: Myoglobin competes for limited antibody with 125I-labeled myoglobin
Comments: Infrequently used. Sensitive; requires RIA facility. Has been replaced by automated non-isotopic
immunoassays or point-of-care devices.
Method 7: Immunoassay, quantitative

Principle of analysis: Myoglobin bound to capture antibody. Second labeled antibody (enzyme, fluorophore,
chemiluminescent tag) forms sandwich and enables detection.
Comments: Most frequently used technique; allows rapid turnaround of results
Method 8: Lateral flow point-of-care immunoassay testing devices

Principle of analysis: Sandwich non-isotopic immunoassay using passive capillary flow to immobilized
antibodies to myoglobin
Comments: Point-of-care testing device immunoassays are conducted on whole blood and enable rapid
turnaround time for reporting results.
Table 2: Race-, Sex-, and Age-Specific Reference Intervals for Serum Myoglobin [20]
Groups Myoglobin (g/L)
Race Sex Age (years) Mean Range*
White Male 50 32 2052
50 38 2071
Female 50 28 1842
50 32 1660
Black Male All 42 2087
Female 50 28 2038
50 31 1853
*Range includes at least 95% of the population with 90% confidence (log normal distribution).
899
Opiates
Opiates
Dennis E. Coyle
Name: Opiates
Clinical significance: click here
Molecular structure: Morphine, 7,8-didehydro-4,5-epoxy-17-methylmorphinan-3,6-diol
Merck Index: 6129
Structure:
Molecular structure: Heroin, diacetylmorphine,
7,8-didehydro-4,5-epoxy-17-methylmorphinan-3,6-diol diacetate
Merck Index: 2932
Structure:
Molecular structure: dihydromorphinone, hydromorphone,
4,5-epoxy-3-hydroxy-17-methylmorphinan-6-one
C:\5th Edition Print Revised\Opiates_2009-07-02[1].doc

900
Opiates
Merck Index: 4714
Structure:
Principles of Analysis and Current Usage The screening for opiates is further complicated by the
Opioids are a class of drugs defined by their ability to fact that the hepatic metabolism of the opiates allows for
bind specifically to any of several subclasses of opioid the biotransformation of one opiate into another.[1]
receptors to produce an agonistic action. Opiates are Heroin serves as an example of this. Heroin is
commonly known as narcotic analgesics and derived metabolized into morphine and further into
from gum opium, such as heroin, morphine, and hydromorphone. Identification of the particular opiate
dihydromorphone. Synthetic compounds that have present in a sample becomes dependent on the time
opiate-like qualities include propoxyphene, methadone, between the last dose (ingestion or administration) and
meperidine, and fentanyl. Opiates as well as many other the collection of the sample. The opiates are metabolized
classes of drugs are subject to abuse. Because of this, to form an inactive 3-O-glucuronide, which is present in
testing for the presence of opiates and determining the the plasma and urine in much higher concentrations than
concentrations in overdose patients are major the parent drug. This major metabolic route must be
considerations of a toxic drug screen. This necessitates considered during sample preparation, since the
the need for rapid qualitative assay methods capable of glucuronide conjugate is not extracted under conditions
detecting the opiates in urine at overdose levels in the used for the parent opiate.
microgram-per-milliliter range.
Qualitative Tests
Analyses of opiates present at therapeutic levels, Thin-layer chromatography is performed on urine
however, must measure concentrations in the low adjusted to a range of pH 10 to 11 and extracted into
nanogram-per-milliliter range.[1] This range requires 25% ethanol in chloroform (v/v),[2] (method 1, Opiates
that the assay method used to measure therapeutic levels Methods Summary Table). The solvent is evaporated
have greater sensitivity than the methods used to and the residue taken up in a small amount of methanol
measure drug abuse. This requirement is of vital concern and applied to a thin-layer chromatography plate (silica).
because methods commonly used for drug screening, After development of the plate using chloroform
such as thin-layer chromatography (TLC), adequate for methanolammonia (90:10:1) as the mobile phase, the
detecting drug overdose, lack the sensitivity to detect position of the opiate is visualized by use of long- or
this class of drugs when the drugs are present at short-wave ultraviolet radiation, iodine vapor, and
therapeutic levels.[2] Screening methods of the iodineiodoplatinate solution or ninhydrin solution. Rf
immunoassay type are sensitive down to the picogram- values and brown color are specific criteria for the
per-milliliter range. However, they lack specificity (are identification of the drug.
cross-reactive) and therefore do not allow for
identification of the specific opiate involved.[3,4] One can modify the TLC method by employing
Amberlite, an XAD-2 resin.[4] This polymeric
absorbent resin allows differential elution of the opiates
i before application to the TLC plate. Separation is
Opiates
accomplished through adsorption (mainly through van
der Waals forces) of many water-soluble organic
compounds to the resin. This method also concentrates
Methods edition: Dennis Coyle
the sample, allowing detection of the drug at lower
concentrations than with untreated samples.
Fourth edition: Dennis Coyle
Fluorescence can be employed to detect nonspecifically
Fifth edition: Not updated
the presence of opiates[5,6] (method 2, Opiates Methods
901
Opiates
Summary Table). Urine is extracted at a range of pH 8.5 (1:1)[16] or methanol.[17] Solid-phase extraction has
to 9 into butanol/chloroform. The organic phase is then gained in popularity because it offers excellent
back extracted into acid, and oxidation of the opiates is recoveries and yields the best sample purification.[17]
accomplished with K3Fe(CN)6 (Opiates: Figure 1). A
fluorescent product is generated and is detected by a Immunoassays have also been employed for the
fluorometer. The excitation maximum is 250 nm, and the detection of opiates because of their high sensitivity and
emission maximum is 440 nm. ease of sample preparation. Most assays include a
competitive binding step involving competition between
Gas-liquid chromatography (GC) detection of the opiates the drug in the sample being analyzed and a labeled drug
is performed by extraction of urine, adjusted to a basic used for detection. The specificity (lack of cross-
pH, with ethyl acetate[7,8] (method 3, Opiates Methods reactivity) of the assay method is dependent on the
Summary Table). The organic phase is then separated antibody employed. Most antibody preparations exhibit a
and evaporated to dryness, and the residue is derivatized broad range of cross-reactivity between the different
with either trimethylsilylimidazole or heptafluorobutyric opiates.
anhydride. Separation and detection is accomplished
with an OV-series column with either a flame ionization The hemagglutination inhibition (HI) method utilizes
or electron-capture detector. The relative retention times formaldehyde-stabilized red blood cells complexed with
at specific column conditions allow presumptive a drugalbumin conjugate.[18] The red blood cells are
identification of the opiate. In some cases, an acid coated with the opiate. The test sample (urine) is added
hydrolysis step is employed before the extraction of the to the coated red blood cells along with antibody
urine.[2,7] This releases the opiate from the glucuronide directed against an opiate, usually morphine. The free
conjugate and therefore increases the amount of opiate, morphine, or a cross-reactive opiate in the
extractable opiate. Enzymatic hydrolysis of the sample, inhibits the agglutination (crosslinking) of the
glucuronide has also been used.[9] red blood cells by the antibody. The nonagglutinated red
blood cells form a diffuse pellet, whereas the
The GC method may be used with mass spectrometry agglutinated red cells settle and form a characteristic red
(GC/MS).[10,11] The method is the same as described ring (method 7, Opiates Methods Summary Table).
for GC; however, the mass spectrometer is used for
detection. This increases sensitivity and allows more The stable free radical-labeled assay technique (FRAT)
accurate identification of the opiate, using the mass- is a homogeneous competitive binding assay that uses a
fragmentation pattern (method 4, Opiates Methods spin-labeled drug (nitroxide radical attached to the
Summary Table).[12] drug)[16] (method 9, Opiates Methods Summary
Table). The spin label is detected by the use of electron
High-performance liquid chromatography (HPLC) is spin resonance spectrometry (ESR). The FRAT is able to
also useful in detection and identification of the measure the energy absorbed by the unpaired electrons
opiates[13,14] (method 5, Opiates Methods Summary (spin-label radical) when they change from a ground to
Table). Extraction of alkalinized urine is followed by an excited state. The spectrum is sharp when the drug is
chromatography on a reversed-phase silica column free, and it broadens and flattens when the drug binds to
(C18) with 0.1 M NaH2PO4 in 25% CH3CN in water, the antibody. The amount of spectrum sharpening
(increasing amplitude of the peaks) is proportional to the
pH 4.8. The separated opiates are detected by
amount of spin-labeled drug displaced or the amount of
spectrometric analysis at 254 nm. A modification of this
drug in the sample.
method employs the use of the fluorescence method
described earlier.[6,7] After extraction from urine, the
The enzyme-multiplied immunoassay technique (EMIT)
opiate is oxidized using K3Fe(CN)6 to form a
is basically the same assay as an RIA method except that
fluorescent dimer. The resulting product is then an enzyme-labeled drug replaces the radioactive
chromatographed on a silica column and detected using label[11,19] (method 10, Opiates Methods Summary
a fluorescence detector.[14] Table). An enzyme (lysozyme or malate dehydrogenase)
is covalently bound to an opiate (usually morphine). The
HPLC may also be used in combination with mass drug in the sample proportionately displaces the enzyme-
spectrometry (HPLC-MS).[15] The method is the same labeled drug. The enzyme is inhibited when the complex
as that used for HPLC with the mass spectrometer used is bound to the antibody and is active when unbound.
as the detector. The HPLC-MS coupling, however, does The free enzymedrug complex is able to convert a
suffer from a lack of sensitivity. The mass spectrometer substrate into products, causing a shift in the absorbance
in combination with the HPLC offers universality and of the solution (increase or decrease). The wavelength is
identification of the opiate (method 6, Opiates Methods 340 nm for the dehydrogenase reaction and 450 nm for
Summary Table). the lysozyme assay. The change in the absorbance is
proportional to the amount of drug in the sample. An
The use of solid-phase extraction procedures have been advantage of this assay over RIA methods is that the
used for GC, GC-MS, HPLC, and HPLC-MS methods. antibody-bound drug does not need to be removed
Samples are made alkaline and passed through a C18 (method 10, Opiates Methods Summary Table).
reversed-phase column.[16,17] The opiates are then
eluted from the column with dichloromethane-acetone
902
Opiates
ELISA (enzyme-linked immunosorbent assays) have methods require long periods of time for extraction and
become one of the most popular methods for assessment derivatization of the sample, they are not suited for the
of drug exposure (method 11, Opiates Methods emergency situation. When time is not a factor, an
Summary Table). Antibodies raised against an opiate additional step of hydrolysis of the glucuronide
are coated onto the surface of a polystyrene microtiter conjugate increases the sensitivity of detection.
plate. Blood or urine is added to the plate, along with an
opiateenzyme conjugate. Opiate in the sample and the The immunoassay techniques allow rapid detection of
opiateenzyme complex compete for antibody binding the presence of an opiate. Because of their high
sites on the surface of the well. Following an incubation, sensitivity and ability to be used without prior sample
unbound drug is removed by washing and a colorimetric treatment, these methods can be adapted for rapid and
reaction is used to measure how much drugenzyme easily automated analysis. Immunoassay methods most
conjugate is bound to the microtiter well. The measured frequently used include EMIT, ELISA, and CEDIA
absorbance is inversely proportional to the concentration technologies. The specificity of the assay is determined
of drug in the sample. by the nature of the antibody reagent used. For screening
purposes, a nonspecific antiserum is preferred, and one
Quantitative Assays that reacts with the glucuronide conjugate would have
Methods used for the quantitation of the amount of maximal sensitivity. Of the immunoassay methods
opiate present include GC, GC/MS, HPLC, and RIA discussed, EMIT is perhaps the best choice for a rapid
(methods 3 to 5 and 8, Opiates Methods Summary screening of samples. EMIT assays do not require
Table). separation of the antibody-bound drug from the unbound
drug and can produce results within 20 min. ELISAs
The chromatographic methods (GC, GC/MS, and HPLC) have also become extremely popular because of their
not only are able to quantitate the amount of drug but ease of use and adaptability for use with blood or urine
can also identify the specific drug present in the sample. specimens.[23] Unlike many of the other homogenous
The GC/MS technique is the only method in this immunoassays, ELISAs do not require pretreatment of
discussion that is able to identify unequivocably an whole blood samples.
individual opiate. This is possible because of the
methods ability to generate a unique mass- Confirmation of an opiate requires a different assay. The
fragmentation pattern for each individual opiate. These chromatographic methods (TLC, HPLC, GC, HPLC/MS,
methods therefore allow quantitation of mixtures of and GC/MS) are best suited for this type of analysis.
opiates resulting from the administration of more than Thin-layer chromatography, which is readily available,
one opiate or from the biotransformation (metabolism) is usually employed and is adequate for many overdose
of the opiate taken. cases. For those cases in which EMIT is positive and
TLC is negative, a specific extraction of the opiate
Radioimmunoassay (RIA) utilizes a radiolabeled drug followed by determination on a GC or GC/MS is usually
(3H, 14C, or 125I) in a heterogeneous competitive performed. The required instruments are generally
binding assay. The displacement of the radiolabeled drug available in toxicology laboratories. GC/MS is the
from the antibody by the unlabeled drug in the sample is method of choice because identification is verified by
the unique mass-fragmentation pattern of the separated
measured by either scintillation spectrometry (3H and
compound. This eliminates errors from any interfering
14C) or a gamma-ray scintillation counter (125I). The
compounds with the same retention time as a particular
free, labeled drug, not complexed to the antibody, is opiate present in the patient sample.
separated from the antibody-bound labeled drug by an
agent that precipitates the antibodydrug complex, such Quantitative
as a second antibody (an anti-antibody), charcoal Methods used for quantitation of the opiate are GC,
dextran,[20] or ammonium sulfate.[21] The amount of GC/MS, HPLC, HPLC/MS, and RIA. The
labeled, unbound drug is proportional to the amount of chromatographic procedures are the only suitable
drug in the sample (method 8, Opiates Methods procedures for the analysis of mixtures of opiates,
Summary Table). especially if a quantitation of all pharmacologically
active opiates present in the sample is desired. Usually
Reference and Preferred Methods for therapeutic monitoring only the opiate being
Qualitative administered is assayed. Of the chromatographic
The qualitative identification of an opiate in a toxic drug methods, GC/MS would be the method of choice, since
screen requires, first, the determination of the presence it has all of the advantages of the GC or HPLC methods
of an opiate and, second, its confirmation by a second along with the ability to confirm the identification of the
method. The confirmation may identify the specific separate opiates. The extraction method enables
opiate. TLC has sufficient sensitivity for detection of selection of either a total opiate level (hydrolysis before
some urinary opiates in the overdose patient. Practical extraction measures levels of both conjugated and
limits of detection of TLC are usually in the 1 to 2 unconjugated opiate) or an unconjugated opiate level
g/mL range.[22] Other chromatographic methods (that (extraction only, no hydrolysis). It must be noted that
is, GC, GC/MS, and HPLC) are able to detect nanogram acid hydrolysis will destroy some of the opiate (range of
quantities of opiates when the extraction and 4% to 15%). An enzymatic method (glucuronidase) does
derivatization procedures are optimized. However, if the not lead to a destruction of the opiate but requires more
903
Opiates
time, that is, 24 h for the enzymatic method compared to The specimen can be plasma, serum, urine, saliva, or
1 h for the acid hydrolysis.[10] tissue extract. The blood collection tube may be of any
type. All glassware, however, must be silanized to
When an individual opiate is being monitored for prevent adsorption of opiates to active sites. The samples
therapeutic use, immunological procedures are may be stored for 3 months at 20 C.
appropriate. In this circumstance EMIT and ELISA are
not quantitative. The RIA procedures have adequate Interferences
sensitivity. The type of antiserum used (antibody) must One must be aware that the EMIT assay does result in a
be considered for determination of the specificity of the positive test for codeine and dextromethorphan,
assay method.[3,4] The specificity of the antibody is compounds present in nonprescription cold
determined by how the opiate is presented for antibody remedies.[24,25] Cross reactivities for EMIT and
formation. For example, morphine is too small a ELISA procedures are presented in Opiates Table:
molecule to elicit antibody formation. The morphine Comparison of commonly used assay methods.
molecule must be covalently coupled to a larger
molecule such as albumin. The site of the chemical Consumption of foods that contain poppy seeds can
linkage of the morphine to the protein determines what result in excretion of significant quantities of morphine
part of the molecule of morphine elicits a response by and codeine in urine. Detection of 6-acetylmorphine in
the antibody-producing cell. Morphine conjugated to the urine can help rule out poppy seed ingestion as a cause
protein at the C-3 position (Opiates: Figure 1) elicits of increased opiate excretion.
antimorphine antibodies to the C-6 position and to the
nitrogen region between carbons 16 and 9. The antibody Opiates Therapeutic Range
therefore is sensitive to groups in this region. The 3-O- The therapeutic blood levels for the opiates have not
glucuronide derivative of morphine cross-reacts with the been established. Since heroin has not been approved by
antibody and is detected as free morphine. This analysis the FDA for clinical use, its therapeutic range has not
is valid if total morphine levels are to be measured. If, been investigated. Although hydromorphone is approved
however, pharmacologically active drug levels (amount for clinical use, its therapeutic range and basic
that would produce a response such as pain relief) are pharmacokinetic parameters have not yet been
needed, levels determined in this manner are unsuitable. established. The average peak plasma levels for
To remove interference by glucuronides, the sample can hydromorphone after a 2 mg intravenous or 4 mg oral
be extracted into an organic phase and the dried residue dose have been reported to be 31.6 ng/mL (5 min after
re-extracted into an aqueous solvent (saline or buffered the dose; n = 6) and 22 ng/mL (1 h after the dose; n = 6)
saline) before analysis. Another way to eliminate this respectively.[22] Morphine is the only opiate in which a
interference would be to use an antibody with specificity speculative therapeutic serum level has been reported. A
for the C-3 position. This has been accomplished by use serum concentration of 50 ng/mL is considered the
of a morphine molecule conjugated to the carrier protein minimum concentration of free morphine needed to
through its C-6 position. This allows detection of provide moderate analgesia.[26]
morphine without interference by the glucuronide and
without extraction of the sample. For a more detailed For purposes of screening for drugs of abuse, the two
examination of the effect of opiate structure on antibody most common cutoff concentrations for opiates are 300
specificity, refer to the paper by Findlay, Butz, and ng/mL and 2000 ng/mL. SAMHSA raised its cutoff from
Jones.[9] 300 to 2000 ng/mL in an effort to eliminate opiate
(morphine) positive results from poppy-seed-containing
A comparison of EMIT, ELISA, RIA, HI, GC, GC/MS, food items. Unfortunately, at the 2000 ng/mL cut-off
and HPLC detection levels is seen in Opiates Table: concentration, routine use of other opiate drugs such as
Comparison of commonly used assay methods. hydrocodone or oxycodone may not be detected.
GC or GC/MS would provide the best overall method for Interpretation

the determination of opiates. The method is the most Opiates are used as sedatives (premedicants), anesthetics
sensitive of the methods discussed and allows for both (narcotic nitrous oxidebalanced anesthetic technique),
identification and quantitation of the opiates. The and analgesics (for postoperative and chronic pain).
method also exhibits the lowest coefficient of variation Rapid determinations for opiates have recently been
(Opiates Table: Comparison of commonly used assay gaining more importance because of the advent of
methods). The method does not, however, lend itself narcotic antagonists (such as naloxone and nalorphine
well to automation and is not capable of rapid analysis hydrochloride), which allow quick reversal of the opioid
with large sample loads. action. The undesirable side effects (such as nausea,
convulsions, respiratory depression, and coma) can be
If separation is unnecessary and a specific drug is being quickly reversed by these antagonists.
analyzed, as in the case of therapeutic monitoring, the
RIA procedure is the method of choice. This method is Opiates have been demonstrated to affect levels of
able to handle large sample loads in a short period of pituitary hormones, though not to a clinically important
time (1 to 2 h). extent.[27] Because of the effects of morphine on
gastrointestinal motility, this drug has been used for
Specimen
904
Opiates
several diseases of the gastrointestinal tract such as reverse phase high performance liquid
diarrheal disorders and colitis. chromatography. Anal Chem 1977;49:359-63.
14 Jane I, Taylor JF. Characterization and
Opiate Performance Goals quantitation of morphine in urine using high
Acceptable performance criteria (CLIA-88) for pressure liquid chromatography with
qualitative measurement of opiates are based on 80% fluorescence detection. J Chromatogr
participant consensus 1975;109:37-42.
15 Pacifici R, Pichini S, Altieri I, Caionna A, Passa
.References AR, Zuccaro P. High-performance liquid
1 Brunk SF, Della M. Morphine metabolism in chromatographicelectroscopy mass
man. Clin Pharmacol Ther 1974;16:51-7. spectrometric determination of morphine and its
2 Mule SJ. Routine identification of drugs of 3- and 6-glucuronides: application to
abuse in human urine. I. Application of pharmacokinetic studies. J Chromatogr
fluorometry, thin-layer and gas-liquid 1995;B:664:326,.
chromatography. J Chromatogr 1971;55:255- 16 Fehn J, Megges G. Detection of 06-mono-
66. acetylmorphine in urine by GC/MS as evidence
3 Van Vunakis H, Wasserman E, Levine L. for heroin use. J Anal Toxicol 1985;9:134.
Specificities of antibodies to morphine. J 17 Bermejo AM, Ramos I, Fernandez P, Lopez-
Pharmacol Exp Ther 1972;180:514-21. Rivadulla M, Cruz ., Chiarotti M, Fucci N,
4 Bastos ML, Jukofsky D, Mule SJ. Routine Marsilli R. Morphine determination by gas
identification of drugs in human urine. III. chromatography/mass spectroscopy in human
Differential elution of the XAD-2 resin. J vitreous humor and comparison with
Chromatogr 1973;81:93-8. radioimmunoassay. J Anal Toxicol
5 Kupferberg H, Burkhalter A, Way EL. A 1992;16:372.
sensitive fluorometric assay for morphine in 18 Vanzetti G, Casani M, Vatente D. Detection of
plasma and brain. J Pharmacol Exp Ther morphine in urine by hemagglutination
1964;145:247-51,. inhibition with use of lyophilized reagents. Clin
6 Sansur M, Baccafurim A, Morgenstern S. Chem 1983;29:1376-9.
Automated fluorometric method for the 19 Rubenstein KE, Schneider RS, Ultman EF.
determination of morphine in urine.J Assoc Off Homogeneous enzyme immunoassay: a new
Anal Chem 1972;55:880-7 immunochemical technique. Biochem Biophys
7 Yeh SY, McQuinn RL. GLC determination of Res Commun 1972;47:846-51.
heroin and its metabolites in human urine. J 20 Honigberg TL, Stewart JT, Brown WJ, et al.:
Pharm Sci 1975;64:1237-9. Radioimmunoassay of hydromorphone. J Anal
8 Nicolau G, Vanhear G., Kaul B, Davidow B. Toxicol 1977;1:70.
Determination of morphine by electron capture 21 Honigberg TL, Stewart JT, Jackson, C.
gas-liquid chromatography. Clin Chem Radioimmunoassay procedure for
1977;23:1640-3. hydromorphone and hydrocodone in human
9 Findlay JA, Butz RF, Jones EC. Relationship plasma. J Pharm Sci 1980;69:1171-3.
between immunogen structure and antisera 22 Rowley GL, Rubenstein KE, Huisjen J, Ullman
specificity in the narcotic alkaloid series. Clin EF. Mechanism by which antibodies inhibit
Chem 1981;27:1524-37. hapten-malate dehydrogenase conjugates, an
10 Ebbighausen WO, Mowat J, Vestergaard P, enzyme immunoassay for morphine. J Biol
Kline NS: Stable isotope method for the assay Chem 1975;250:3759-66.
of codeine and morphine by gas 23 Kerrigan S, Phillips WH. Comparison of
chromatography-mass spectrometry: a ELISAs for opiates, methamphetamine, cocaine
feasibility study. Adv Biochem metabolite, benzodiazepines, phencyclidine,
Psychopharmacol 1973;7:135-46. and cannaboids in whole blood and urine Clin
11 Van der Slooten E.P, Van der Helm HJ. Chem 2001;47:540-7.
Comparison of the EMIT (enzyme multiplied 24 Mule SJ, Bastos ML, Jukofsky D. Evaluation of
immunoassay technique) opiate assay and a immunoassay methods for the detection of
gas-chromatographic-mass-spectrometric drugs of abuse in urine. Clin Chem
determination of morphine and codeine in urine. 1974;20:243-8.
Clin Chem 1976;22:1100-11. 25 Mule SJ, Whitlock E, Jukofsky D.
12 Wasels R, Bellvill F. Gas chromatographic- Radioimmunoassay of drugs subject to abuse:
mass spectrometric procedures used for the critical evaluation of urinary morphine
identification and determination of morphine, barbiturate, morphine, barbiturate and
codeine and 6-monoacetylmorphine. J amphetamine assays. Clin Chem 1975;21:81-6.
Chromatogr 1994;A:674:225. 26 Berkowitz BA, Ngai SH, Yang JC, Hempstead
13 Wu CY, Wittick JJ. Separation of five major J, Spector S. The disposition of morphine in
alkaloids in gum opium and quantitation of surgical patients. Clin Pharmacol Ther
morphine, codeine and thebaine by isocratic 1975;17:629-35.
905
Opiates
27 Schick R, Schusdziarra V. Physiological, of exogenous and endogenous opiates. Clin

pathophysiological and pharmacological aspects Physiol Bioche. 1985;3:43-60.
Tables
Methods for Opiate Analysis
Principle: Extract opiate; chromatograph; observe Rf factor (relative mobility) and color reaction presumptive
for opiate
Usage: Blood, plasma, serum, urine, tissue
Comments: Qualitative for high concentrations; can separate mixtures
Method 2: Fluorescence spectroscopy
Principle: Extract opiate; oxidize with K3Fe(CN)6; fluorescence measured
Usage: Blood, plasma, serum, urine
Comments: Nonspecific for opiates; qualitative
Principle: Extract opiate; derivatize (such as trimethylsilylimidazole); chromatograph; retention time identifies
opiate
Usage: Blood, plasma, serum, urine, tissue, saliva, hair, tissues, vitreous humor
Comments: Qualitative and quantitative; resolves mixtures; only a few drugs are known to interfere
Method 4: Gas chromatography-mass spectrometry (GC/MS)
Principle: Steps same as in 3; mass spectrometer identifies opiates and helps resolve complex mixtures
Usage: Blood, plasma, serum, urine, tissue, saliva, hair tissues, vitreous humor
Comments: Qualitative and quantitative; better identification of opiates than with GC (method 3)
Principle: Extract opiates; may or may not be derivatized (derivitization as in 2); retention time identifies opiate
Usage: Blood, plasma, serum, urine, tissue, saliva
Comments: Qualitative and quantitative; resolves mixtures; as efficient as GC but run time may be longer
Method 6: High-performance liquid chromatographymass spectrometry (HPLC/MS)
Principle: Steps as in Method 5; mass spectrometer ideitifies opiates and helps resolve complex mixtures
Usage: Blood, plasma, serum, urine
Comments: Qualitative and quantitative; resolves mixtures; run time may be longer than for GC/MS;
identification of opiates
Method 7: Hemagglutination inhibition (HI)
Principle: Opiate bound to fixed red blood cells competes with opiate bound in sample for reaction with
agglutinating antibodies; inhibition of agglutination indicates presence of opiates
Usage: Urine, serum, plasma
Comments: Very sensitive but nonspecific; used qualitatively; may be used quantitatively if opiate is known; no
extraction necessary
Principle: Radiolabeled opiate competes with opiate in sample for reaction with limited amount of antibody;
competitive binding assay
Usage: Urine, serum, plasma, tissue extract, saliva, breast milk
Comments: High sensitivity; qualitative and quantitative; may distinguish between glucuronide conjugate and
unconjugated opiate; no extraction necessary
Method 9: Free radical-labeled assay technique (FRAT)
Principle: Uses stable free-radical label (nitroxide) in competitive binding assay
Usage: Urine, serum, plasma, saliva
Comments: Qualitative and quantitative; commercially available
Method 10: Enzyme-multiplied immunoassay technique (EMIT)
Principle: Uses enzyme label in competitive binding assay
Usage: Urine
Comments: Qualitative; commonly used technique; no need to separate bound from unbound labeled drug
Method 11: Enzyme-linked immunosorbent assay (ELISA)
Principle: Drug in sample and drugenzyme conjugate compete for solid phase antibody
Usage: Serum, plasma, urine
Comments: Qualitative; frequently used; no need to separate bound from unbound labeled drug
906
Opiates
Comparison of Commonly Used Assay Methods for Preparation of Various Assay Tubes for
Serum Opiates Radioimmunoassay Procedure (Amounts are in
GC microliters)
Sample size (mL of plasma): 1 Method 1: Nonspecific binding (NSB)
Sensitivity: Electron capture: 0.1 ng/mL Blank serum: 100
Coefficient of variation: 2% Working buffer (WB): 400
Assay time: 1.5 to 2 h Labeled drug working solutions (LDWS):
reference: [8] 100
GC/MS Working antiserum solution (WAS):
Sample size (mL of plasma): 1 Standards:
Sensitivity: 0.5 ng/mL Sample:
Coefficient of variation: 5% Dextran-coated charcoal (DCC): 100
Assay time: 1.5 to 2 h Method 2: 100% binding
reference: [11] Blank serum: 100
HPLC Working buffer (WB): 300
Sample size (mL of plasma): 1 Labeled drug working solutions (LDWS):
Sensitivity: Fluorometric: 4 ng/mL 100
Coefficient of variation: 6% Working antiserum solution (WAS): 100
Assay time: 1.5 to 2 h Standards:
reference: [14] Sample:
HPLC/MS Dextran-coated charcoal (DCC): 100
Sample size (mL of plasma): 1 Method 3: Standards
Sensitivity: 10 ng/mL Blank serum:
Coefficient of variation: 6.9% Working buffer (WB): 300
Assay time: 1.5 to 2 h Labeled drug working solutions (LDWS):
reference: [14] 100
EMIT Working antiserum solution (WAS): 100
Sample size (mL of plasma): 0.05 Standards: 100
Sensitivity: 100 ng/mL Sample:
Coefficient of variation: 7% Dextran-coated charcoal (DCC): 100
Cross-reactivity: Other opiates, 3-O- Method 4: Test samples
glucuronide conjugate, dextromethorphan Blank serum:
Assay time: 20 min Working buffer (WB): 300
reference: [11] Labeled drug working solutions (LDWS):
ELISA 100
Sample size (mL of plasma): 0.010 Working antiserum solution (WAS): 100
Sensitivity: 100 ng/mL Standards:
Coefficient of variation: 5% (blood) and 8% Sample: 100
(urine) Dextran-coated charcoal (DCC): 100
Cross-reactivity: Other opiates
Assay time: 1 h
reference: [23]
RIA
Sample size (mL of plasma): 0.2
Sensitivity: 2.5 ng/mL
Coefficient of variation: 9%
Cross-reactivity: Other opiates,
dextromethorphan
Assay time: 2 h
reference: [21]
Hemagglutation assay
Sample size (mL of plasma): 0.1 (urine)
Coefficient of variation: Qualitative
Cross-reactivity: Other opiates, 3-O-
glucuronide conjugate, dextromethorphan
Assay time: 1 h
reference: [18]
907
Opiates
5. Working buffer (WB). 0.05 M phosphate

buffer, pH 7.4 with 0.1% gelatin and 0.6% NaCl. Place
Figures 1.0 g of gelatin in a 400 mL Erlenmeyer flask with 200
mL of distilled water. Heat the flask with stirring until
Opiates: Figure 1 the gelatin is completely dissolved. Allow the solution to
cool, and transfer to a 1 L volumetric flask. Add 1.31 g
of NaH2PO4, 10.87 g of Na2HPO4, 6.00 g of NaCl, and
700 mL of distilled water. Mix to dissolve salts, and
adjust pH to 7.4 with H3PO4. Bring volume to mark
with distilled water, and mix. Stable 1 week at 4 C.
6. Dextran-coated charcoal (DCC). Suspend 2.5
g of charcoal in 100 mL of distilled water. Add 0.25 g of
dextran to 100 mL of suspended charcoal. Mix overnight
Reaction for formation of pseudomorphine. (From at 4 C. Centrifuge the solution at 2000 g for 15 min.
Kupferberg, H., et al.: J. Pharmacol. Exp. Ther. 145:247- Remove the supernatant and suspend the dextran-coated
251, 1964.) charcoal in 400 mL of working buffer.
7. [7,8-3H(N)]-dihydromorphine. Stable at 20
C in the dark under N2 for 1 year. Check purity by TLC
Opiates: Figure 2 every 3 months according to manufacturers
recommended method.
8. [N-Methyl-3H]-morphine. (NEN) NET-653
(10 to 20 Ci/mmol). Prepare as described above for
dihydromorphine.
9. Diacetyl [I-3H(N)]-morphine (heroin).
10. Labeled-drug working solution (LDWS).
Dilute 20 L of dihydromorphine or heroin in 20 mL of
working buffer (0.25 Ci/mL). Dilute 16.7 L of
morphine in 20 mL of working buffer (0.25 Ci/mL).
11. Morphine, hydromorphone, and heroin.
Obtain pure salt form from manufacturer. Stable 1 year
if kept dry and under inert atmosphere (N2) at room
temperature.
12. Calibration standards. Dissolve 11 mg of
opiate in 50 mL of distilled water. This is solution A.
Dilute 100 L of solution A to 10 mL with drug-free
serum or plasma (solution 1). Dilute 50 L of solution A
to 100 mL with serum or plasma (solution 2). Store
Typical radioimmunoassay calibration curve for solutions 1 and 2 at 20 C in 1 mL aliquots. Stable for
hydromorphinone over concentration range of 2.5 to 25 about 3 months. Prepare the following solutions fresh
ng/mL. each day for a calibration curve:
Volume of Volume of Final standard

Solution solution working buffer concentration
Procedure: Radioimmunoassay no. (L) (L) (ng/mL)
The opiates contained in the sample are incubated with 2 250 9750 2.5
radiolabeled morphine, antibody, and phosphate buffer 1 25 9975 5.0
at room temperature. The unbound morphine is removed 1 75 9925 15.0
1 100 9900 20.0
by the addition of dextran-coated charcoal. An aliquot of 1 125 9875
the supernatant is then removed and placed in a 25.0
scintillation vial with scintillation fluid and counted by
use of a scintillation spectrometer. The percentage of 13. Antiserum. Obtain lyophilized antiserum to
labeled opiate bound to antibody is inversely morphine directed toward the 3-hydroxy and nitrogen
proportional to the logarithm of the drug concentration. moieties between positions C-9 and C-16. The
lyophilized serum is stable 1 year at 20C. Cross-
Reagents reactivity with hydromorphone, 78%; hydrocodone,
1. Gelatin, calf skin. 60 Bloom, Type IV. 59%; heroin, 100%; monoacetylmorphine; 100%; and
2. Charcoal. Stable indefinitely. morphine-3-glucuronide, 4.8%. Each vial contains
3. Dextran T 70. enough antiserum for 200 tubes.
4. Scintillation fluid. Cocktail designed for 14. Working antiserum solution (WAS).
aqueous samples (able to absorb 0.7 mL of water into 10 Reconstitute according to manufacturers instructions.
mL of cocktail). Dilute 0.6 mL of antiserum to 6 mL with working buffer
(60 samples). Stable for 1 week at 4 C.
908
Opiates
Assay
Equipment: Liquid scintillation counter, a Procedure: Opiates, Urine: Confirmation by SIM-
refrigerated centrifuge capable of generating 2000 g or MSD (GC/MS)
greater, 20 mL glass scintillation vials, and a liquid
scintillation counting cocktail designed for aqueous Principle
samples (able to absorb 0.7 mL of water into 10 mL of Opiates are first hydrolyzed and then are
cocktail). isolated from urine by liquidliquid extraction,
followed by acetylation derivatization and
1. To 10 75 mm labeled plastic or glass
analysis in the SIM mode by GC/MS to
disposable tubes, add 0.1 mL of the appropriate
generate qualitative findings.
standard or sample. See Opiates Table:
Preparation of various assay tubes for Specimen of Choice
radioimmunoassay procedure for addition of Urine (5 mL); spot aliquot. May be transported
sample and reagents to tubes. at room temperature. Store at 28 C prior to
testing.
2. To prepare a nonspecific binding (NSB) or
blank tube, add 0.1 mL of blank plasma, 0.1 mL Indications
of labeled drug solution, and 0.4 mL of working To confirm a presumptively positive "Opiates"
buffer. To prepare a 100% binding (100%B) immunoassay screen test result for the presence
tube, or zero-dose tube, add 0.1 mL of blank of codeine, morphine, and/or hydromorphone.
plasma, 0.1 mL of labeled drug solution, 0.1
mL of working antiserum solution, and 0.3 mL Reagents and Materials:
of working buffer. A. Toxi-A Extraction Tubes (Toxi-Lab, Inc.)
3. To each standard and sample tube, add 0.1 mL B. 16 100 mm (10 mL) glass tubes
of labeled drug working solution (55,550 cpm),
C. 5 ml volumetric glass pipet
0.1 mL of working antiserum solution, and 0.3
mL of working buffer. Incubate all tubes at D. 100 L Eppendorf pipet
room temperature for 50 min and at 4 C for 10 E. Glass Pasteur pipets
min. F. 5 mL glass "Reacti-vials" (Pierce Co.)
4. Add 0.1 mL of 4 C dextran-coated charcoal G. Ethyl acetate (Chromatoquality)
suspension to each tube. Incubate at 4 C for 10 H. Glusulase (Sigma). Store at 4 C. Stable
min. as per stated shelf life.
5. Centrifuge at 2000 g at 4 C for 20 min. Place I. Sodium acetate (1.1 M), pH 5.2
0.2 mL of supernatant into 20 mL scintillation
Dissolve 37.38 g sodium acetate in 5.5
vial. To each scintillation vial, add 0.5 mL of
mL of glacial acetic acid and dilute to 250
distilled water and 10 mL of counting cocktail.
mL with H2O. Store at room
Cap, and mix vials. Count for 10 min in
scintillation counter optimized for tritium temperature. Stable for 1 year.
counting. J. Pyridine (Chromatoquality)
K. Acetic anhydride (Chromatoquality)
Calculations L. Urine Positive Control
Calculate percent binding of standards and
(1) Reconstitute two lyophilized vials
samples by the following equation:
with 25 mL of deionized water each.
(2) Allow to stand 30 min, then transfer
Percent binding = Counts for sample - NSB counts
contents of both vials to a 500 mL
100
volumetric flask, rinsing the vials
Counts for 100% binding - NSB
several times with deionized water
and transferring washes to flask as
Plot log standard percent binding versus log
well.
concentration. Fit data using linear regression analysis
(Opiates: Figure 2). From slope (a) and intercept (b), (3) Dilute to mark with deionized
calculate concentration (X, in ng/mL) by the following water.
equation: (4) Aliquot 5 mL fractions into plastic
tubes and store at -20 C. Stable
for 1 year.
log X = (log y - b)/a or X = 10(log y - b)/a M. Negative Control
A urine pool previously testing below 10
For hydromorphone, as an example, the effective assay ng/mL by FPIA for opiates and negative
range is 2.5 to 20 ng/mL. The percentage of radiolabeled for opiates by GC/MS. 5 mL aliquots
drug bound to the antiserum in the absence of unlabeled stored at -20 C. Stable for 1 year.
drug is 30.0% + 5.4% (B0). The 50% displacement point
N. Stock Internal Standard (100 mg/mL)
(B/B0)50 occurs at 7.2 ng/mL. Deuterated D3 morphine. Store at 28 C.
Stable as per stated shelf life.
909
Opiates
O. Working Internal Standard (10 ng/l). To 4. Update the data file name, i.e.,
a 12 75 glass tube, add 300 mL of stock c:\HPCHEM\1\DATA\1118\POSCOC.
internal standard and dilute with 2.7 mL 5. Press Run Control.
of deionized water. Store at 4 C. Stable 6. Inject calibrator.
1 week. 7. At the end of the run, an
P. Calibration Standard (DASC Calibrator). uncalibrated report will print. To
"GC-MS Calibration Standard." calibrate, go to DATA ANALYSIS,
Main Panel, and load the calibration
Instrument and Parameters
data file specified above.
8. Choose Calibrate and Update
A. Hewlett-Packard 5972/5970 MSD GCMS
Levels.
and Chem Station.
9. Choose Recalibrate Level 1 Update
B. Heating block
Levels Update Qualifying ions
60 C update Response factors.
C. Centrifuge 10. Press OK, Exit.
Procedure 11. To print the report, press Quantitate
and Calculate Report.
With each run include a calibration standard 12. Make sure retention times,
and positive and negative controls, all calibration values and ion ratios
processed in singlecate. Process as with subject have been updated. If not, update
specimens, including hydrolysis. the levels again.
A. To a 10 mL glass test tube, add 5 mL 13. Subsequent injections, i.e., controls
urine, 1 mL of 1.1 M sodium acetate and patient material, will print
buffer, and 100 L of Glusulase. calibrated results.
B. Incubate at 60 C for 5 h.
B. Changing Retention Time Windows
C. To a 10 mL glass tube, add 5 mL of
hydrolysate and 100 L of working 1. Choose calibrate within the Data
internal standard. Vortex mix. analysis header.
D. Transfer the hydrolysate/internal standard 2. Choose Edit Compounts
mix to the fill line of a Toxi-A tube. 3. Highlight the compound to be
E. Extract for 5 min. Centrifuge for 2 min. updated and choose "View."
Using a disposable glass Pasteur pipet, 4. Update the retention time and OK
transfer the top layer to a Reacti-Vial. the selection.
F. Evaporate to dryness at 60 C under 5. Save the changes to the compound
nitrogen. and Exit.
G. Add 20 L of pyridine and 20 L of 6. Recalibrate, if necessary.
acetic anhydride. Vortex, cap, and heat at 7. Reprint the report.
60 C for 20 min.
H. Uncap and evaporate to dryness. C. Manual Calculation
I. Recap the vials. 1. List retention times of drug of
J. Reconstitute the sample in 50 L of ethyl interest for both control and
acetate (immediately before injecting) patient/subject sample.
vortex and inject 1 L into method 2. Calculate ion ratios for drugs of
"OPIATES." interest using the following:
NOTE: The following injection sequence
should be utilized: Codeine
Calibration standard Ratio in % =Individual 229, 282, or 341 ion area 100%
Positive control Sum of 229 + 282 + 341 ion areas
Negative control
Patient/subject
Hydromorphone
Calculations and Quality Control Ratio in % =Individual 228, 285 or 327 ion area 100%
NOTE: If using 5972 MSD refer to Addendum Sum of 228, 285, and 327 ion areas
III (Sequencing).
A. Calibration Morphine
1. From Top Level, choose method. Ratio in % = Individual 310, 327 or 369 ion area 100%
2. Press Load and Run Method. Sum of 310, 327 and 369 ion areas
3. Choose the Method of Interest.
910
Opiates
3. Calculate the concentration of the searched. The sample may

drug of interest using the following: then be considered negative
Codeine concentration in ng/mL = if these actions prove so.
Ratio of 327 ion area/330 ion area for Control or patient sample 500 Retention times may
Ratio of 327 ion area/330 ion area for calibration standard shift from day to
day, or even from
Hydromorphone Concentration in ng/ml = run to run.
Ratio of 327 ion area/330 ion area for Control or Patient sample 500
Interpretation and Notes
Ratio of 327 ion area/330 ion area for calibration standard
Morphine A. A positive finding indicates drug usage,

Same as for hydromorphone using but cannot be used to interpret
morphine areas. impairment. All positive results are noted
4. Compare control and as "positive," with no quantitative ng/mL
patient/subject's retention time, ion report.
ratios, and concentration to that of B. Sensitivity300 ng/mL; negatives
the calibration standard. A positive reported as "none detected."
sample will include: C. SpecificityNo known interfering
A retention time match 1% of standard compounds.
Three ion ratios matching within 20% D. Recovery80%.
of standard
Addendum I: GC-MS Calibration Standard
A calculated concentration equal to or
above the threshold cutoff of 300 ng/mL.
1. Prepare the following stock standards (a) through
D. Result Acceptance/Rejection
(d).
Acceptance:
NOTE: Prepare fresh for each new lot.
Retention times match within 1%.
(a) Phencyclidine (PCP)
Results in ng/mL units must exceed documented cutoff
To a 10 mL volumetric flask add 11.4 mg of
levels of 300 ng/mL.
PCPHCl. Dissolve and dilute to mark with
Ion ratios must match within 20%. methanol.
If multiple drug peaks are being analyzed the retention
(b) 9-Carboxy-11-nor--9-THC (cannabinoid
times must be unique for each of
metabolite) (THC).
the different drugs. Example:
Make sure that the target software (c) Benzoylecgonine (BE)
has properly assigned different To a 10 mL volumetric flask, add 12.5 mg of
retention times to the different benzoylecgonine tetrahydrate. Dissolve and
"opiates". Codeine, dilute to mark with methanol.
Hydromorphone, and Morphine (d) Into a 10 mL volumetric flask add:
must have unique retention times on
the calibration run. Drug Amounts (mg)
Rejection and Corrective Action: Amphetamine sulfate 13.7
Results below documented cutoffs Methamphetamine HCl 12.3
which meet all other acceptance Desmethyldiazepam 10
criteria, should be reported as
Temazepam 10
negative or unconfirmed.
Codeine 10
When ion ratios do not match the
range stated on the report, several Morphine sulfate 11.8
options are available: Amobarbital 10
(a) Reinject the patient/subject Butabital 10
sample. Secobarbital 10
(b) Reinject the calibration Pentobarbital 10
standard in the cal mode and Phenobarbital 10
reassess the patient ion Phentermine HCl 12.35
ratios.
Hydromorphone HCl 11.2
(c) Calculate the ion ratios for
the calibrator and the patient
Dilute to the mark with methanol.
in question manually.
(d) If these steps do not rectify
the problem, the sample
may be injected into a scan
program and library
911
Opiates
2. Pool Spiking (b) Transfer aliquots into 16 100 polyethylene

(a) To a 2 L volumetric flask, add 200 mL of tubes
negative urine pool. (c) Cap and freeze at -20C in racks marked
(b) To the flask add: with the respective lot number
200 L (0.2 mL) of standard 1(a)
2000 L (2.0 mL) of standard 1(b)* ADDENDUM III Sequencing
600 L (0.6 mL) of standard 1(c)
1000 L l (1.0 mL) of standard 1(d) 1. Click on the "Sequence" header and choose
*For standard 1(b), THC, transfer the "Load."
contents from two commercial vials, rinsing 2. Choose the proper sequence for the mode of
each vial three times with deionized water, operation required, i.e., choose the THC
and transferring the washings to the flask. sequence for running THC assays.
(c) Mix flask with constant Note: The sequence tables are simply
swirling while diluting to mark with deionized water. templates and can be modified as you
(d) Transfer contents of flask to see fit. Remember that multiple
wide-mouth flask for aliquoting. methods can be run within a
sequence. (If THC sequence is
3. Target concentrations
chosen, you may run any method
THC, DRUGS1, COCAINE, or any
Drug Concentration (ng/mL) combination.)
3. Choose "Edit Sample Log Table." The Sample
Amobarbital 500 Log Table appears.
Butalbital 500 4. Fill out the Sample Log Table. Use one line per
Pentobarbital 500 injection. The injections will be made
Secobarbital 500 sequentially in the order they are listed in the
table. To set up multiple injections from one vial,
Phenobarbital 500
make a separate line in the Sample Log Table for
THC metabolite 100 each injection .
BE metabolite 300 5. For each injection, select the sample type
Morphine 500 (Sample, Blank, or Calibration)
Codeine 500 Note: When a calibration is specified the
Hydromorphone 500 calibration LEVEL identifier must
Phencyclidine 100 be specified. For our purpose type
Amphetamine 500 in a 1. Choose REPLACE for Upd
Methamphetamine 500 RF, Upd RT, and Upd Ql.
Phentermine 500 6. Enter the position of the vial in the autosampler
tray.
4. Lot testing and verification 7. Assign a data file name (sample number or
description).
(a) Assign sequential lot number to the pool.
8. Type in the name of the method to use to analyze
(b) Submit pool for testing/verification prior to
the sample or type a question mark (?) in the
use, including:
field to display a list of available methods.
TDx FPIA analysis for each of the
9. Type in an identifier for the sample injection.
individual drugs (PCP, THC, BE,
(Optional)
amphetamines, opiates,
benzodiazepines, and barbiturates) 10. Once the Sample Log Table is complete, it may
be printed by choosing "Print Brief Format"
EMIT analyses for the individual
within SEQUENCE.
drugs
11. Save the Sample Log Table.
Specific GC-MS quantitative
analyses run in parallel with current 12. Choose "Load and Run Sequence."
lot of DASC calibrator a. Click on "Full Method" under Method
(c) Acceptable analyses include: Sections to Run.
TDx FPIA positive results for all drug classes b. Update the Data File Directory to reflect
the date of the run, i.e.,
EMIT positive results for all drug classes
C:\hpchem\1\data\0715\ .
GCMS quantitation within 10% of target value
c. Click on "Run Sequence" to start the run.
Acceptable calculated GC-MS control data
5. Aliquoting and storage
(a) Aliquot approximately two-thirds of the total
pool in 5 mL fractions and one third of the
pool in 10 mL fractions
912
Organic Acid Screening

Kevin Carpenter i
Clinical Interpretation Chapter 52, Diseases of Genetic Origin, in the 5th edition of Clinical

Organic acid analysis has become a key component in Following removal of the solvent by evaporation, the
the diagnosis of inherited metabolic disease. The term concentrated residue containing the organic acids needs
organic acids covers a wide range of low-molecular- to be derivatized to allow separation by gas
weight carboxylic acids, including those derived from chromatography. Since the technique was first described,
intermediary metabolism of amino acids, carbohydrates, the majority of workers have used trimethylsylyl (TMS)
lipids, and biogenic amines. Although methods such as esters, and almost all published data for retention times
capillary electrophoresis [1] and tandem mass and mass spectra are for TMS esters. The derivatization
spectrometry [2] are used, the commonest methods in is easily and rapidly performed using N,O-
use are based on gas chromatography [3]. bis(trimethylsilyl)trifluoroacetamide with 1% trimethyl
chlorosilane (BSTFA/1% TMCS). Care must be taken to
The principle steps of organic acid analysis are (1) ensure no water remains in the extract, because this will
isolation of the organic acids from physiological fluids, impair derivatization. Problems of formation of multiple
(2) formation of volatile derivatives, (3)chromatographic TMS esters for compounds with more than one site of
separation using capillary gas chromatography, and (4) derivatization may make quantitation difficult but in
detection of the compounds, now most commonly practice does not cause problems with diagnosis.
achieved using benchtop mass selective detectors (mass
spectrometers). Separation of the derivatized organic acids requires
capillary chromatography with temperature
Before extraction, urine samples may be treated with an programming. Typically the oven temperature will start
oximating agent such as ethoxylamine hydrochloride in the 80C to 100C range, and temperature gradients of
that preserves ketoacids. Without oximation, a 4C to 6C per minute are employed to selectively elute
significant proportion of ketoacids are converted to their the compounds. The rate chosen will depend upon the
corresponding 2-hydroxyacid. In practice, organic acid throughput requirements of the laboratory, but there is
analysis with or without oximation still allows little to be gained from very slow gradients, especially
identification of abnormal patterns and appropriate when mass selective detectors can distinguish co-eluting
interpretation of the findings. A recent review of peaks. The column used will typically be 30 meters in
analytical methods used to screen urine organic acids by length with an internal diameter of 0.2 to 0.5 mm.
participants in the European Research Network for the Stationary phases are a matter of personal choice, and
evaluation and improvement of screening, Diagnosis, retention characteristics are published for many
and treatment of Inherited Metabolic disorders external compounds of interest on OV1, OV5, and OV17 (or
quality assessment schemes (ERNDIM-QAP) found just equivalent) columns. To allow for variations in
over half used oximation [4]. temperature programming, column length, and carrier-
gas flow rates, retention data is quoted compared to
Extraction of organic acids from urine is most alkanes of different chain lengths run under the same
commonly performed using solvent extraction from conditions. For example, a compound eluting midway
acidified, salt-saturated urine. It is most convenient to between the C14 and C16 alkanes would have a
dilute the urine to a fixed concentration of creatinine retention index of 15, and this would be fixed for any
prior to extraction to control for concentration of the given stationary phase.
sample. Addition of one or more internal standards will
serve to control the extraction process and allow for a Detection and data acquisition is best performed using a
reproducible qualitative scoring of abnormal compounds mass selective detector (mass spectrometer). Although
(although true quantitation requires calibration and flame ionization detection can be used, mass
ideally a stable, isotope-labeled internal standard for spectrometry allows positive identification of the peaks,
each compound). Solvents used are typically ethyl identification of co-eluting compounds, and the ability to
acetate, sometimes in combination with diethyl ether. detect very small amounts of pathognomonic compounds
Solid phase extraction using silicic acid [5] or ion hidden in the baseline. Data acquisition is not started
exchange [6] columns have been used but do not retain until the solvent peak has returned to baseline, after
some of the neutral compounds which may be seen using which the mass spectrometer is operated in full scan
solvent extraction and are of diagnostic value (e.g., mode over a mass range of 50-500m/z, with a scan cycle
glycerol). time of around 0.5 seconds, allowing reliable peak
i
Organic acid screening
New method
Fifth edition: Kevin Carpenter
913
reconstruction in Total Ion Current mode and the ability Abnormal results require a detailed interpretation,
to search for key compounds by using extracted ion including significance, differential diagnosis,
searches. Comparison of mass spectra obtained with recommendations for additional biochemical testing,
library entries, in conjunction with retention time, is and in vitro confirmatory studies.
used to identify peaks. Several commercial libraries are
available for purchase, but most centers measuring Specimen
urinary organic acids build their own in-house libraries Organic acids are well cleared by the kidney, and urine
based upon experience and availability of samples from is the near-universal sample of choice for analysis.
patients with organic acidurias. Most modern software Samples should be stored and transported at or below
systems allow the construction of a quantitation 20C. Under such conditions, organic acids are stable
database which allows the software to identify and label for many years. The volume required will be dependent
peaks based on their retention time and the presence of a on the concentration of the urine (see procedure below).
number of ions in the appropriate proportions.
An important precursor to the analytical process is
Reference and Preferred Methods obtaining a specimen most likely to be informative for
There is no formal reference method for organic acid diagnosis of inherited metabolic disease. Many
screening. conditions may present in an episodic manner, and
diagnostic metabolites may only be detectable during
The American College of Medical Genetics publishes clinical crises. For this reason, samples should always be
Standards and Guidelines for Clinical Genetics sought during an acute exacerbation of the condition.
Laboratories [7], the 2006 edition of which states the The absence of pathological findings in samples
requirements for organic acid analysis under the collected at times other than during acute exacerbations
following headings: cannot be taken to exclude these conditions.
Extraction: Interferences
Internal standards added to sample of fixed Since the organic acid method is designed to identify a
creatinine concentration. broad range of excreted compounds, many exogenous
Solvent extraction. substances (or their metabolites) will also be detected.
Oximation performed either routinely or under This includes many drugs and food additives, but in
specific circumstances. practice, the technique requires experienced
interpretation, part of which is the identification of such
Derivatization: compounds and knowledge of their origin. A small
TMS esters formed using BSTFA/TCMS. number of urine samples appear to contain unknown
substances that inhibit effective extraction of some
GC/MS Analysis: compounds. Careful monitoring of internal standard
Capillary column separation responses should allow identification of potentially poor
Detection by quadrupole or ion trap mass extraction, and a request for a repeat sample should be
spectrometer. made.
Computer library of mass spectra for confirmation
of peak identities. Interpretation
The potential number of disorders detectable by organic
Calibration: acid analysis stands at over 30. Interpretation of the
Reporting of quantitative data without the profile obtained is a highly skilled job requiring
availability of a reference standard should be experience of what to expect in normal and disease
reduced to a minimum. states. For this reason, organic acid analysis is usually
performed in specialist centers where the required
Isotope-dilution, selected ion mass spectrometry
experience is concentrated. An excellent source of
should be used for more accurate analyte
information on interpretation of findings has been
quantification.
published by Kumps et al. [8].
Chromatogram Analysis:
Organic Acid Performance Goals
Identification of peaks relies on mass spectra.
The primary performance goal of organic acid analysis is
the correct identification of metabolites and
Quality Control: interpretation of their significance. For the majority of
Tuning of mass spectrometer should be checked on disorders, the abnormalities are gross, and identification
daily basis. is not a problem. However, several key marker
The internal standards in each specimen serve as a metabolites are excreted in trace amounts, and it is
QC check for each specimen. useful to know the sensitivity of the method on a day-to-
day basis. Regular analysis of a spiked urine with
Reporting: addition of hexanoylglycine, octanedioylglycine, 4-
Diagnostic specificity of acute versus asymptomatic hydroxybutyrate, and orotic acid at levels typical of mild
conditions. disease (5, 10, 100, and 15 mol/mmol creatinine,
Identification of all relevant organic acids listed. respectively) will ensure adequate sensitivity is being
914
achieved and flag when column change or MS ion Reagents

source cleaning is required. 1. BSTFA/1% TMCS (N,O-
bis(trimethylsilyl)trifluoroacetamide with 1%
Appropriate interpretation of the profiles obtained is trimethyl chlorosilane)
critical and as mentioned previously requires experience 2. Heptadecanoic acid 20 mmol/L in ethanol
and skill normally only available in specialist centers. (internal standard). Weigh out 0.54g of
Participation in an external quality assurance program heptadecanoic acid, and dissolve in 100 mL of
which incorporates interpretation (e.g., ERDIM-QAP; ethanol.
http:// www.erndimqa.nl is an essential component of 3. 1,3 15N orotic acid 1 mmol/L (internal
ensuring performance goals are met. standard). Weigh out 15.8 mg of 1,315N orotic
acid, and dissolve in 500 L 1M NaOH.
References Transfer to a 100 mL volumetric flask with
1 Garcia A, Barbas C, Aguilar R, Castro M. multiple washings and make up to volume with
Capillary electrophoresis for rapid profiling of deionized water.
organic acidurias. Clin Chem 1998; 44: 1905- 4. Hydrochloric acid 6M.
1911. 5. Sodium chloride. Analar
2 Pitt JJ, Eggington M, Kahler SG. 6. Ethyl acetate. Analar.
Comprehensive screening of urine samples for
inborn errors of metabolism by electrospray Assay
tandem mass spectrometry. Clin Chem 2002; Equipment:
48: 1970-1980. A suitable GC-MS system capable of
3 Tanaka K, West-Dull A, Hine DG, Lynn TB, operation in electron impact ionization;
Lowe T. Gas-chromatographic method of full scan mode is required. The gas
analysis for urinary organic acids. II. chromatograph is fitted with a 5%
Description of the procedure, and its application phenylmethyl-silicone capillary column,
to diagnosis of patients with organic acidurias. 25 mm 0.20 mm, 0.33 m film
Clin Chem 1980; 26: 1847-1853. thickness.
4 Bonham JR. Personal Communication. 2007- Glass tubes, 16 100 mm with PTFE
07-17. lined screw cap.
5 Hoffmann G, Aramaki S, Blum-Hoffmann E, Fortuna Optifix dispenser containing ethyl
Nyhan WL, Sweetman L. Quantitative analysis acetate and set at 2 mL.
for organic acids in biological samples: batch Pasteur capillary pipettes, glass,
isolation followed by gas chromatographic- disposable.
mass spectrometric analysis. Clin Chem 1989; Supply of dry air.
35:5 87-595. Heating block set at +60C.
6 Thompson JA, Markey SP. Quantitative
Pearce Reacti-therm heating block with
metabolic profiling of urinary organic acids by
Reacti-vap head set to +37C.
gas chromatography-mass spectrometry:
Centrifuge.
comparison of isolation methods. Anal Chem
Vortex mixer.
1975; 47: 1313-1321.
Estimate urine creatinine concentration, and dilute urine
7 American College of Medical Genetics.
to creatinine of 1.0 mmol/L with deionized water. For
Standards and Guidelines for Clinical Genetics
samples with initial creatinine less than 1.0 mmol/L,
Laboratories 2006 Edition. Internet page 2006
dilute to 0.5 mmol/L and use 2 mL of sample and 4 mL
[cited 2007 Aug 9];Available from: URL:
of ethyl acetate to extract (see below).
http://www.acmg.net/Pages/ACMG_Activities/s
tds-2002/f.htm
1. In a screw-capped glass tube containing
8 Kumps A, Duez P, Mardens Y. Metabolic,
approximately 0.5 g of sodium chloride, add 10
Nutritional, Iatrogenic, and Artifactual Sources
L of each of the two internal standards,
of Urinary Organic Acids: A Comprehensive
heptadecanoate and 1,3 15N orotic acid, and 1
Table. Clin Chem 2002; 48: 708-717.
mL of diluted urine.
2. Acidify with 50 L of 6M HCl and add 2 mL of
Procedure: Gas Chromatography/Mass ethyl acetate.
Spectrometry 3. The tubes are vortexed for 2 min and
Principle centrifuged for 5 min at 2000 g.
Using ethyl acetate, organic acids are extracted from 4. Transfer the upper solvent layer to a second
acidified, salt-saturated urine which has been diluted to a tube, being careful not to carry any aqueous
fixed creatinine concentration. The solvent is removed phase across.
by evaporation and the sample derivatized to form 5. The solvent is evaporated to dryness under a
trimethylsylyl esters. Derivatized organic acids are stream of air at 37C.
separated by capillary gas chromatography and detected 6. The residue is reconstituted with 50 L of
by mass selective detector. BSTFA/TCMS, and the tubes incubated at 60C
for 30 min.
915
7. GC/MS parameters:
The mass selective detector is used in Full scan data is obtained and peaks identified on the
the electron impact ionization mode basis of retention time and mass spectra and qualitatively
at 70 eV. scored by comparison of total ion response to the
Helium is utilized as a carrier gas. internal standards. Key marker compounds such as
The injection port temperature 240C. acylglycines, 4-hydroxybutyrate, orotate, and 3-
Initial column temperature 80C, held hydroxyglutarate, which may be critical in small
for 1 minute then ramped at 6C/min quantities, may be sought using extracted ion
to 250C, ramped at 20C/min to chromatograms over narrow time windows looking for
270C, with a final hold of 2.5 min. the presence of unique ions present in the mass spectra
Interface temperature 280C. of the compound of interest.
Split injection ratio of 1:15 and
injection volume 2 L.
916
Osmolality
Osmolality
Goce Dimeski
Name: Osmolality
Clinical significance: Refer to Chapter 30, Renal Function, in the 5th edition of Clinical Chemistry:
i
An infrequently measured parameter is vapor-pressure
Osmolality is a colligative property of solutions that depression. The vapor-pressure osmometer actually
depends on the number of dissolved particles present in measures the dew-point depression of the vapor, that is,
the solution. The three types of solutes most often the vapor in equilibrium with the solution being
encountered in biological fluids are electrolytes, organic measured. The more dissolved particles present
molecules, and colloids. In healthy persons, the principle (increased osmolality), the lower the vapor pressure of
contributors to the measured osmolality are Na+, Cl, and the aqueous component of the solution. An important
bicarbonate. In persons with diabetes or renal failure, exception to this is found when the solute itself is a
glucose and urea may also contribute significantly to the volatile substance.
measured osmolality. As the number of dissolved
particles increases, the freezing point and vapor pressure Formulae for Calculating Osmolality
of a solution are decreased, and the osmotic pressure and A stat procedure frequently used to estimate the
boiling point are increased [1]. In dilute solutions, there osmolality of a fluid is the calculated osmolality. To
is a linear change in the colligative properties as the calculate serum osmolality, one sums up the molar
solute concentration increases. It is important to realize concentrations of those substances which contribute
that it is not the mass concentration but the molal most to the measured osmolalitynamely, sodium,
concentration (that is, moles per kilogram of solvent) glucose, and urea. The contribution of chloride and
that is the basis of colligative properties. For example, in bicarbonate is handled by multiplying the sodium
biological fluids the concentration of salts and low- concentration by a variable factor as discussed below
molecular-mass organic compounds (such as glucose [2].
and urea) affect osmolality far more than albumin, which
is present in a large mass amount but because of its large Although there are many equations used to calculate
molecular mass is present at low molal concentrations. osmolality, the simplest and most frequently used is:
The molar concentration of plasma albumin, when
present at 40 g/L, is approximately 0.58 mmol/L. In Eq. 1 [3]
contrast, the molar concentration of plasma NaCl, when
present at 5.8 g/L, is approximately 150 mmol/L. Calculated mOsm/L H2O = 2 [Na+] + glucose (mmol/L)
Normal plasma is approximately 93% water, with the + urea (mmol/L)
remaining 7% being solids (proteins, lipids, etc). The or
electrolytes are dissolved in the water component. Since Calculated mOsm/L H2O = 2 [Na+] +
there are 2 osmotically active components in NaCl (Na+
and Cl), the osmotic contribution is 0.93 2, or 1.86 glucose (mg/dL) + BUN (mg/dL
times the concentration of Na+ (2 or 1.86 accounting for 18 2.8
Na+ and Cl, with Cl being the major accompanying
anion). where sodium concentration is expressed in
milliequivalents per liter (mEq/L) or mmol/L, and the
The most frequently measured colligative property used concentrations of glucose and urea or blood urea
to estimate fluid osmolality is freezing-point depression. nitrogen (BUN) are divided by their respective formula
When 1 mole of solute is added to 1 kg of water, the equivalent weights, 180 and 28, to convert these mass
freezing point is depressed by 1.858C. The laboratory concentrations into millimoles per liter (mmol/L) [4].
osmometer actually employs a sensitive heat thermistor This formula is proposed to balance simplicity and ease
to measure the heat released by the freezing fluid and of calculation with the goal of obtaining results that
relates it to the freezing point of the fluid. closely compare to a measured osmolality. Other more
accurate formulas that have been suggested include the
i
following two:
Osmolality
Previous and current authors of this method: Eq. 2 [5]
First edition: Lawrence A. Kaplan
Calculated mOsm/L H2O = 1.86 [Na+] + glucose
Methods edition: Lawrence A. Kaplan
Second edition: Lawrence A. Kaplan (mmol/L) + urea (mmol/L) + 9
Third edition: Steven C. Kazmierczak, Lawrence A. or
Kaplan Calculated mOsm/L H2O = 1.86 [Na+] +
Fourth edition: Lawrence A. Kaplan glucose (mg/dL) + BUN (mg/dL) + 9
Fifth edition: Goce Dimeski 18 2.8
917
Osmolality
Eq. 3 [6] respond to volatile solutes that may be clinically

Calculated mOsm/L H2O = encountered [15,16]. The contribution to serum
osmolality of several commonly seen alcohols (e.g.,
1.86 [Na+] + 1.38 [K+] + glucose (mmol/L) + urea
ethanol, methanol, and isopropanol) is directly
(mmol/L) + 7.45
proportional to their molar concentrations when
or
osmolality is measured by freezing-point depression
Calculated mOsm/L H2O =
osmometry. However, vapor-pressure osmometers do
1.89 [Na+] + 1.38 [K+] + 1.08[glucose (mg/dL)] + not detect these volatile substances, because volatile
18 solutes contribute to the total vapor pressure present
1.03[BUN (mg/dL)] + 7.45 above the solution, though they decrease the vapor
2.8 pressure of the water portion of the solution [17].
Therefore, the decreased vapor pressure of water is
Rasouli and Kalantari [7] have suggested formula 3 for counterbalanced by the increase in vapor pressure caused
computer calculations on automated analyzers. There are by the presence of a volatile substance. As the
many other formulas suggested for calculation of concentration of the volatile substance increases in
osmolality [2,6,8,9,10,11]. It is important to note that aqueous solutions, the vapor pressure above the solution
although osmolality should be expressed as milliosmoles actually increases, giving falsely low osmolality results.
per kilogram (mOsm/kg) of water, it is most frequently
calculated and reported as milliosmoles per liter Apart from the problem encountered with volatile
(mOsm/L) of water. In dilute solution like plasma at solutes, the osmolality of most specimens measured by
normal temperature and pressure, the osmolality and freezing-point depression correlates well with the
osmolarity are effectively the same. The slight error osmolality measured by vapor-pressure depression. In
introduced due to the difference between these two practice, vapor-pressure osmometers are seldom used.
expressions has little clinical significance [5]. The 9 or Because of its greater sturdiness and precision, most
7.45 mOsm/kg is added to represent the contribution of laboratories use the freezing-point osmometer.
other osmotically active substances in plasma such as
K+, Ca2+, Mg2+, and their unmeasured anions and The calculated osmolality is not meant to provide a
proteins. definitive result but only one sufficient for most acute
emergency service work. Used as such, the calculated
The calculation of urine osmolality is less accurate than osmolality can provide clinicians with valuable
plasma [12]. The commonest formula is Urine information with no extra analysis time. The error
osmolality = 2 (Na+ + K+) + Urea (mmol/L). Other introduced by calculation of the serum osmolality as
formulae may include in the calculation the glucose that mOsm/L instead of mOsm/kg can be clinically
may be present in some urine samples, namely from significant only if the percentage of water in the sample
diabetic patients. The presence of glucose, ammonium, (approximately 93%) changes significantly. Thus highly
or foreign substances such as mannitol will result in lipemic samples or samples with large amounts of
underestimation of the calculated urine osmolality [13]. protein (as from multiple-myeloma patients) can have an
inaccurate calculated osmolality. The laboratory might
Fecal osmolality is infrequently requested in not wish to report a calculated osmolality on such
investigations of chronic idiopathic diarrhea. Calculated specimens or alternatively might affix a comment to the
fecal osmolality should be used in preference to results warning the physician of a possible error. The
measured osmolality, which increases after defecation calculated osmolality will also give falsely low results in
due to continued bacterial fermentation [14]. The Na+ the presence of volatile solutes. Additionally, the
and K+ are measured in the fecal water, and their sum is inclusion of K+, Ca2+, or Mg2+ concentrations in
doubled to account for anions: Calculated fecal formulas has not resulted in a more accurate calculated
osmolality = 2 (Na+ + K+). osmolality.
A 2007 College of American Pathologists (CAP) quality Specimen

assurance survey report indicated that the vast majority Blood drawn without the use of any anticoagulant is the
(97.9%) of laboratories used a freezing-point depression preferred sample. The serum should be removed from
osmometer. It should be pointed out that nearly all the clotted blood cells as rapidly as possible. Well-
automated analyzers now determine a calculated serum centrifuged urine or other body fluids are also acceptable
osmolality, or it can be calculated by laboratory samples. Random urine samples can vary considerably
information systems. Thus the most frequently reported in the concentration of analytes and are rarely useful
serum osmolality is a calculated serum osmolality. clinically. Therefore an aliquot of a 24-h urine collection
is the preferred specimen for the measurement of urine
Reference and Preferred Methods osmolality.
Although there is no reference method for osmometry,
the freezing-point depression method is considered the Any fluid left uncovered will have an increased
de-facto reference method. osmolality because of evaporation and concentration of
sample solutes. Serum osmolality has been reported [18]
It has been pointed out that the two types, vapor-pressure to be stable for 3 days at room temperature, 14 days at
and freezing-point osmometers, differ in their ability to 7C, 14 days at 21C, and as long as 56 days at
918
Osmolality
78C. Urine osmolality may be stable at 4C for up to an emergency room situation. It must be pointed out that
24 h [19]. Urine samples should be centrifuged to the osmolal gap may depend less on the equation
remove particulate matter. employed than on the population tested. For example, in
patients with increased unmeasured solutes, such as
Fecal specimens requested for osmolality estimation are patients in circulatory shock, renal failure, in very low
useful in diagnosis of watery diarrhea. Samples require birth weight neonates, hyperosmolal nonketotic coma,
centrifugation, and the supernatant should be used for alcoholic ketoacidosis, diabetic acidosis (due to increase
the Na+ and K+ measurement. Dilution of solid fecal in ketone concentration), or in patients with multi-organ
samples in water is not recommended [14]. failure (due to increased cellular permeability with
leakage of amino acids and other cellular components),
Interferences there will be a larger difference between calculated and
As previously mentioned above, volatile substances such measured osmolalities [11]. An osmolal gap > 10
as alcohols cause a significant bias in osmolality mOsm/kg has been applied clinically as indicating the
measured using vapor-pressure depression osmometers. presence of other osmotically active compounds, but
Use of citrate as an anticoagulant has been found to because this is dependent on the laboratory equipment
significantly increase measured osmolality [20]. and the equation chosen, institutions should calculate
Hemolysis, even when present in gross amounts, has their own reference range for the osmolal gap.
been found not to interfere with osmolality measured by
use of freezing-point depression [21]. The osmolal gap has been suggested as a rapid test for
the presence of ingested volatiles, especially alcohols
Any particulate matter, microclots in serum/plasma, [6,11,25,26,27,28]. Alcohols have a significant effect on
particulate matter in urine, feces, or other fluid types are the serum osmolality because of their low molecular
capable of altering the freezing or vaporizing process, weights. There is a very good correlation between the
thus leading to inaccuracies. Therefore, all samples alcohol concentration measured enzymatically and that
should be free of such particulate matter and may require determined by calculation [29]. The most common cause
centrifugation prior to analysis. of coexisting hyperosmolality and coma is ethanol
ingestion [29]. If an increased osmolal gap is present, a
Osmolality Reference Interval screen for blood ethanol should be performed
The reference interval for serum osmolality in a healthy immediately. If the screen is negative, a screen for other
person, as measured using a freezing-point depression toxic substances (e.g., ethanol, methanol, isopropanol,
osmometer, is 280 to 295 mOsm/kg [22]. The acetone, ethylene glycol) that can increase serum
approximate contribution from osmotically active osmolality should be run. Additionally, if indirect ISE
substances is as follows: Na+ 140; Cl 105; HCO3 27; methods are used for measuring Na+ and Cl in samples
K+ 4; Glucose 5; Urea 5; Ca2+ 2; Mg2+ 1; HPO42 1; with hyperlipidemia or hyperproteinemia, Na+ and Cl
protein 1 and organic compounds ~5. Individuals living results will be decreased (pseudohyponatremia and
at high altitude (i.e., > 3800 m) have been found to have pseudohypochloremia); the reverse occurs with
significantly lower serum osmolality compared to hypoproteinemia [30,31], which will translate as an
individuals living at sea level [23]. inaccurate osmolal gap. Low serum osmolality can be
seen in cases of overhydration, hyponatremia, and cases
Urine osmolality can range from 50 to 1000 mOsm/kg of inappropriate secretion of antidiuretic hormone. The
[13,24]. osmolal gap should be used with caution and must be
used in conjunction with clinical information.
Interpretation Urine osmolality varies in response to plasma osmolality
Osmolality measurements are most frequently used to and sodium regulation. The appropriate physiological
help determine whether a patient is in a hyperosmolal response is that urine osmolality should be high when
state. Such states can be seen in renal failure plasma osmolality is high and vice versa. Urine
(hyperuremia), diabetes (hyperglycemia), and disease osmolality measurements are performed as a test of the
states in which there is water loss in excess of solute kidneys concentration ability. After fluid restriction for
loss. In the latter clinical states, the degree of 12 to 14 h, a patient with normal renal function should
hyperosmolality parallels the degree of hypernatremia. be able to concentrate urine to approximately 800
Hyperosmolality due to urea is far less dangerous than mOsm/kg. Loss of this function is seen early in renal
when due to glucose. Urea freely diffuses across cell failure, and a concentration ability of only 400 mOsm/kg
membranes, resulting in concentration equilibrium. would indicate severe renal dysfunction [32,33].
Glucose does not diffuse across cell membranes and
causes water movement out of cells. Hyperosmolality Urine osmolal gap has been proposed to serve as a
due to hyperglycemia is often associated with acute surrogate for urine ammonium in cases of
illness, shock, and renal failure, including deficit in body hyperchloremic acidosis. The urine osmolal gap formula
potassium. suggested and used by Kirschbaum et al. (Osmolal gap=
0.5 x( urine osmolality 2 x[Na+ + K+] + Urea
The use of the osmolal gap (the difference between [mmol/L]) obtained the following correlation against
calculated and measured osmolality) to screen for the measured ammonium: y = 2.17x + 10, and r2 of 0.6. Any
presence of exogenous substances can be very useful in gap between the measured and calculated largely
919
Osmolality
represents ammonium. In summary, the derivation of the 7 Rasouli M, Kalantari KR. Comparison of
urine osmolal gap is very cumbersome in that many methods for calculating serum osmolality:
analytes are required to be measured, yet it is far easier multivariate linear regression analysis. Clin
on modern analyzers to directly measure urinary Chem Lab Med 2005; 43: 635-640.
ammonium. 8 Weisberg HF. Osmolality. Lab Med 1981; 12:
81-85.
9 American Association of Clinical Pathologists,
The calculated fecal osmolalities subtracted from the
Commission on Continuing Education:
measured plasma osmolality (or if plasma osmolality is
Osmolality Clinical Chemistry Check Sample
not measured, use 290 mOsm/kg for approximation) is No. CC-71, Chicago, 1971, p. 27.
used to work out the osmotic gap. A value of < 50 10 Khajuria A, Krahn J. Osmolality revisited -
mOsm/kg is proposed as a suitable cut off to deriving and validating the best formula for
differentiate osmotic from secretory diarrhea. In osmotic calculated osmolality. Clin Biochem 2005; 38:
diarrhea, the osmotic gap is > 50 mOsm/kg, and this may 514-519.
be due to ingestion of magnesium (e.g., magnesium- 11 Purssell RA, Lynd LD, Koga Y. The use of the
containing laxative abuse) or malabsorbed osmole gap as a screening test for the presence
carbohydrates. In secretory diarrhea, the value is < of exogenous substances. Toxicol Rev 2004;
50mOsm/kg, and this is associated with colonic 23: 189-202.
stimulant laxatives, hormonal causes (e.g., tumors 12 Roscoe MH. A comparison of calculated and
producing vasoactive intestinal peptide), and some estimated osmolarities of urine. J Clin Pathol
gastrointestinal infections (e.g., cholera) [14]. 1960; 13: 514-517.
13 Penney MD, Walters F. Are osmolality
measurements clinically useful? Ann Clin
Osmolality Performance Goals
Biochem 1987; 24: 566-71.
Based on intra- and interindividual variation of serum
14 Duncan A. Screening for surreptitious laxative
osmolality, as measured by freezing depression in
abuse. Ann Clin Biochem 2000; 37: 1-8.
healthy adults, the coefficient of variation was found to
15 Rocco RM. Volatiles and osmometry. Clin
be less than 1% [33]. However, review of the 2007 CAP
Chem 1976; 22: 399.
data shows serum osmolality, as measured by freezing-
16 Barlow WK. Volatile and osmometry II. Clin
point depression osmometers, has an interlaboratory
Chem 1976; 22: 1230-1232.
coefficient of variation of 1.1% to 1.5% for osmolality
17 Mercier DE, Feld RD, Wittell DL. Comparison
within the normal reference interval. Vapor-pressure
of dewpoint and freezing point osmolality. Am
osmometers have worse precision at all levels of
J Med Technol 1978; 44: 1066-1069.
osmolality2.3% to 2.8% coefficient of variation within
18 Seifarth CC, Miertschischk J, Hahn EG, Hensen
the normal reference interval. It is clear that osmolality
J. Measurement of serum and plasma
measurements by either method do not meet the goal of
osmolality in healthy young humans influence
achieving the desirable CV of < 1%.
of time and storage conditions. Clin Chem Lab
Med 2004; 42: 927-932.
References
19 Weissman N, Pilegg VJ. Inorganic ions. In
1 Stevens SC, Neumayer F, Gutch CF. Serum
Henry R.J, Cannon DC, and Winkleman JW,
osmolality as a routine test. Nebraska Med J
editors: Clinical chemistry principles and
1960; 45: 447.
techniques, New York, 1974, Harper & Row,
2 Weisberg HF. Osmolalitycalculated delta
Publishers, Inc., p. 739.
and more formulas. Clin Chem 1975; 21: 1182-
20 Smith JC, Lewis S, Holbrook J, Seidel K, Rose
1184.
A. Effect of heparin and citrate in measured
3 Worthley LI, Guerin M, Pain RW. For
concentrations of various analyses in plasma.
calculating osmolality, the simplest formula is
Clin Chem 1987; 33: 814-816.
the best. Anesth Intensive Care 1987; 15: 199-
21 Rocks BF, Sherwood RA, Cook JGH. Whole
202.
blood osmolality. Ann Clin Biochem 1986; 23:
4 Holmes JH. Measurement of osmolality in
106-108.
serum urine and other biologic fluids by the
22 Grasbeck R, Alstrom T. Reference values in
freezing point determination. In preworkshop
Laboratory Medicine: the current state of the
manual on Urinalysis of Renal Function
art. New York, NY Wiley, 1981.
Studies, Chicago, 1962.
23 Zeballos J, Galdos B, Quintanilla A. Plasma
5 Dorwart WV, Chalmers L. Comparison of
osmolality in subjects acclimatized at high
methods for calculating serum osmolality from
altitude. Lancet 1973; 1: 230-231.
chemical concentrations, and the prognostic
24 Kirschbaum B, Sica D, Anderson FP. Urine
value of such calculations. Clin Chem 1975; 21:
electrolytes and the urine anion and osmolar
190-194.
gaps. J Lab Clin Med 1999; 133: 597-604.
6 Bhagat CI, Garcia-Webb P, Fletcher E, Beilby
25 Redetzki HM, Koerner TA, Hughes JR, Smith
JP. Calculated vs. measured osmolalities
AC. Osmometry in the evaluation of alcohol
revisited. Clin Chem 1984; 30: 1703-1705.
intoxication. Clin Toxicol 1972; 5: 343-363.
920
Osmolality
26 Pappas AA, Gadsden RH Jr, Gadsden RH, Procedure: Osmolality by Freezing-Point Depression
Groves WE. Computerized calculation of
osmolality and its automatic comparison with Principle
observed serum ethanol concentration. , Am J The sample or standard is supercooled below its freezing
Clin Pathol 1982; 77: 449-451. point and then agitated by a rapidly vibrating probe. This
27 Coakley JC, Tobgui S, Dennis PM. Screening induces rapid crystallization of the sample, and it begins
for alcohol intoxication by the osmolar gap. to freeze. Heat of fusion is released by the crystallization
Pathology 1983; 15: 321-323. and is accurately measured by a thermistor probe present
28 Krahn J, Khajuria A. Osmolality gaps: in the solution. The released heat is related to the
Diagnostic accuracy and long-term variability. freezing-point depression (1 mOsm of solute depresses
Clin Chem 2006; 52: 737-739. the freezing point by 0.001858C), which is converted to
29 Loeb JN. The hyperosmolar state. N Engl J osmolality.
Med 1974; 290: 1184-1187.
30 Dimeski G, Mollee P, Carter A. Effects of Reagent
hyperlipidaemia on plasma sodium, potassium Osmolality standards can be purchased from instrument
manufacturers or other commercial sources. Calibrators
and chloride measurements by an indirect ion
are used to standardize the instrument for either serum or
selective (ISE) measuring system. Clin Chem
2006; 52: 155-156. urine osmolalities.
31 Dimeski G, Barnett RJ. Effects of total plasma
A relatively simple procedure for preparing osmolality
protein concentration on plasma sodium,
potassium, and chloride measurements by an standards has been described [34]. Reagent-grade NaCl
is heated to 200C overnight to drive off any water
indirect ion selective electrode measuring system.
present in the crystals. After it cools, the desired amount
Crit Care and Resus 2005; 7: 12-15.
32 Ricos C, Alvarez V, Cava F, Garcia-Lario JV, of NaCl is weighed out (see below) and added to 1 kg of
class I distilled water. The most convenient way to do
this is to fill a 1-L class A volumetric flask with water at
databases on biologic variation: pros, cons, and
progress. Scand J Clin Lab Invest 1999; 59: 20C to the mark. Add exactly 1.8 mL of additional
water, and then add the salt to dissolve.
491-500.
33 Fraser CG, Cummings ST, Wilkinson SP,
Neville RG, Knox JD, Ho O, MacWalter RS. Desired osmolality Grams of NaCl per kilograms
Biological variability of 26 clinical chemistry (mOsm/kg) of H2O
analytes in elderly people. Clin Chem 1989; 35: 100 3.094
783-786. 300 9.476
34 Johnson RB, Hoch H. Osmolality. In Meites S, 500 15.93
editor: Standard methods of clinical chemistry, 1000 32.12
vol. 5, New York, 1965, Academic Press, Inc.
The standard should be stored in clean polyethylene or
borosilicate glass bottles. Always pour the standards into
clean test tubes before use. When stored properly in
closed containers, the standards will be stable for 3 to 6
months at room temperature.
921
Oximetry
Oximetry
Goce Dimeski i
Name: Oximetry, hemoximetry, CO-oximetry

Clinical significance: Refer to Chapter 40, Hemoglobin; Chapter 29, Acid-Base Control and Acid-
Base Disorders; and Chapter 55, Toxicology, in the 5th edition of Clinical Chemistry: Theory,
Principles of Analysis and Current Usage < 10% of total hemoglobin. The oxygen dissociationi
Oximeters (or hemoximeters) are specialized curve with Hb F is shifted to the left when compared to
spectrophotometers used to determine the oxygen HbA.
saturation and hence the amount of oxygen bound to
hemoglobin. Depending on the type of oximeter, the Oxyhemoglobin and Deoxyhemoglobin
analytes that can be measured are total hemoglobin, Oxyhemoglobin is the main Hb fraction, composing
hemoglobin fractions or derivatives, and the oxygen 94% to 98% of total Hb in arterial blood. It is the
saturation. The hemoglobin fractions are of clinical oxygenated fraction of hemoglobin, and fully saturated it
interest in assessment of oxygenation. The different can bind four oxygen molecules (1 gm of fully saturated
hemoglobin fractions absorb light at different and Hb (O2Hb) binds 1.39 mL of O2).
specific wavelengths in the visible region (520 to 622
nm, Table 1) and thus permit use of spectrophotometry Hemoglobin without oxygen is referred to as
to determine their concentrations. CO-oximetry is a term deoxyhemoglobin. Other names that have been used are
derived from the technical ability to measure deoxyHb, desaturated Hb or reduced Hb.
simultaneously the concentration of the hemoglobin Deoxyhemoglobins main role is as a buffer for
fractions: carboxyhemoglobin (COHb, FCOHb), hydrogen ions. The functional fractions (O2Hb and HHb)
oxyhemoglobin (O2Hb, FO2Hb), deoxyhemoglobin are also used to determine the oxygen saturation (sO2).
(HHb, FHHb), methemoglobin (metHb or Hi, FMetHb),
and sulfhemoglobin (SHb). In turn, the sum of all Carboxyhemoglobin (Carbon Monoxide)
fractions is the total hemoglobin concentration (ctHb). Carboxyhemoglobin is formed when carbon monoxide
The oxygen saturation (sO2, SaO2) is determined from (CO) binds to hemoglobin. CO gas can easily displace
the O2Hb and HHb. The term CO-oximeter originates O2 from Hb, owing to its greater (~220 times greater)
from the first commercially available oximeter, the IL affinity for Hb. CO binds even more tightly to Hb F,
282 CO-oximeter produced by Instrumentation making infants particularly vulnerable to its effects [2].
Laboratory (IL) in 1979. There are two sources of CO: exogenous and
endogenous. The exogenous sources include tobacco
The Hb fractions are divided into functional fractions smoke, malfunctioning furnaces and gas dryers,
(O2Hb, HHb), which are capable of reversibly binding combustion of motor fuels (68%), burning buildings, and
O2, and dysfunctional fractions (COHb, MetHb, SHb), industrial processes (12%). The endogenous (human)
which are incapable of reversibly combining with O2 source is a by-product of heme catabolism, resulting in a
under physiological conditions. Increased concentrations background carboxyhemoglobin saturation of 0.4 to 0.7
of dysfunction hemoglobins result in a decreased % in healthy subjects [3].
capacity to carry oxygen and deliver it to the tissues,
increasing the likelihood of cellular hypoxia. Persons who have inspired carbon monoxide in chronic
Measurement of the Hb fractions is important because or acute doses usually are seen as medical emergencies.
they may be direct contributors to the clinical picture. Thus a means to measure the extent of carbon monoxide
exposure rapidly, either through measurement of carbon
This section presents methods for determination of the monoxide concentration in blood or through
hemoglobin derivatives: O2Hb, HHb, COHb and CO, measurement of the fraction of hemoglobin that is
metHb, SHb, and oxygen saturation (sO2) in blood. converted to carboxyhemoglobin, is needed.
Hemoglobin These measurements help the clinician ascertain (1)

Hemoglobin is a tetrameric protein, and the usual form whether COHb is present in clinically significant or
in adults comprises two and two chains and is called toxic amounts and (2) what treatment modality can be
hemoglobin A (Hb A). Hemoglobin in the fetus and used to treat the patient effectively. CO or COHb can
newborn is mainly fetal Hb (Hb F) (50% to 100 %) [1], lead to tissue hypoxia, acidosis, or effects on the central
and it has a slightly different absorbance spectrum to Hb
A (Figure 2). After birth, Hb A gradually replaces Hb F, i
and at around 3 months of age, the Hb F will usually be Oximetry
New method:
Fifth edition: Goce Dimeski
922
Oximetry
nervous system but does not produce cyanosis. the skin is picked up to determine the sO2. Pulse
Treatment of the CO-poisoned patient begins with oximeters are capable of providing a noninvasive
supplemental oxygen for the hypoxia and to accelerate estimate of tissue oxygen saturation via the measurement
elimination of CO from the body. The half-life of COHb of the differential absorption at two different light
is 5 to 6 hours [4]. wavelengths (660 and 940 nm) by oxygenated and
deoxygenated hemoglobin [9,10,11]. Deoxyhemoglobin
Methemoglobin absorbs more energy at 660 nm, whereas O2Hb absorbs
MetHb is hemoglobin that has the ferrous ion (Fe2+) more energy at 940 nm. Pulse oximeter probes consist of
oxidized to the ferric (Fe3+) state, hence it is unable to a photodetector and two light-emitting diodes. One diode
bind oxygen. MetHb is formed by either hereditary or emits at 660 nm (red band of the spectrum) and the
acquired means. Some metHb forms naturally in RBCs second at 940 nm (infrared band of the spectrum). The
and can amount to 1% to 2% of total Hb [5]. In normal photodiodes are continuously switched on and off
persons, the Fe3+ is reconverted to Fe2+ by the action of several hundred times per second to record the
NADH metHb (cytochrome b5) reductase. The absorption of O2Hb and HHb during pulsatile and
hereditary form is due to mutation leading to NADH nonpulsatile flow.
metHb reductase deficiency, or to amino acid
substitution in the alpha or beta globin chain, resulting in It is the ratio of the absorbance at the two wavelengths
hemoglobin M (Hb M). Life expectancy with hereditary from which the oxygen saturation is determined. These
metHb is significantly shortened in most individuals. instruments are calibrated from data on healthy
The acquired methemoglobinemia has a fairly acute individuals who had simultaneous CO-oximetry
onset and is a result of exposure to agents such as measurement of arterial oxygen saturation ranging from
nitrites, nitrates, sulfonamides, aniline dyes, acetanilid, 75% to 100%. Pulse oximeters are not reliable when the
phenacetin, local anesthetics (benzocaines, lignocaine, saturation is < 75% [9]. A ratio of absorbance (660/940
etc), dapsone etc [4,5]. The other acquired form is after nm) of 0.43 corresponds to 100% oxygen saturation, and
prolonged exposure of blood to air, with the O2Hb being a ratio of 3.4 corresponds to 0% oxygen saturation. In
oxidized to form metHb. the absence of dysfunctional hemoglobins, a ratio of 1.0
corresponds to an oxygen saturation of 85%. Pulse
Persons who develop methemoglobinemia classically oximeters do not measure the dysfunctional hemoglobins
present with hypoxia and cyanosis. The measurement of (COHb and MetHb). Elevated levels of dysfunctional
metHb is important for prompt medical intervention hemoglobins will result in false-normal sO2 [10]. Pulse
either by removing the causative agent or commencing oximeters are not affected by the difference in the
treatment with metHb reducing agents, namely absorption spectra of Hb A and Hb F [12]. A major
methylene blue or, less effectively, with ascorbic acid. technological advance in pulse oximetry is the new pulse
Patients with cytochrome b5 reductase deficiency oximeter, Masimo Rad-57 (Masimo Inc), capable of
respond to methylene blue therapy but not those with Hb measuring MetHb and COHb using 8 wavelengths [11].
M disease. The additional therapies for such patients
may include blood transfusions and hemodialysis [6]. 2. CO-oximeters
These patients do not respond well to oxygen therapy. Laboratories that measure hemoglobin fractions perform
The half-life of metHb is ~ 55 minutes [4]. this as a stat procedure using benchtop blood-gas
analyzers equipped with CO-oximeter modules. CO-
Sulfhemoglobin oximeters are divided into discrete wavelengths or
Sulfhemoglobin is formed by the linkage of sulfur to continuous spectrum, which may be with or without
hemoglobin. Sulfhemoglobin appears in some (slide technology) hemolysis.
individuals rather than metHb after exposure to
sulfonamides, phenacetin, acetanilid, and trinitrotoluene. CO-oximeters offer major advantages in that they
Other drugs or compounds that have been implicated in require an extremely short time of analysis (~1 min) and
causing elevated SHb are dapsone, metoclopramide, and little or no sample preparation, measure the total Hb and
H2S gas [7]. The sulfur binding to hemoglobin is stable the fractions simultaneously, and require minimal
and irreversible. Because sulfhemoglobin stays in the technical skills.
circulation until the red blood cells complete their life
cycle; it causes significant and prolonged hypoxia and a. Discrete Wavelengths CO-oximeters
cyanosis. There is no known treatment for elevated SHb The earliest CO-oximeter, IL 282, utilizes an aspirated
levels other than discontinuation of the implicated agent. arterial whole-blood sample (125 L) to analyze the
Transfusion may be necessary in severe cases. various fractions of hemoglobin. It consists of a
hemolyzer unit (e.g., ultrasound device) to hemolyze the
Oximeters cells, a lamp (e.g., Tl-Ne hollow-cathode lamp) to
Oximeters belong to one of the two major classes: pulse provide the light source, a lens system to focus the light
oximeters or CO-oximeters. onto four interference filters to provide monochromatic
light of specific wavelengths (535.0, 585.2, 594.5, and
1. Pulse Oximeters 626.6 nm), a photodetector to detect the amount of light
The first pulse oximeter was developed in 1974 by a absorbed by the different Hb fractions, and a readout
Japanese firm and was based on the work of Takuo device. The sum of the absorbances of the fractions is
Aoyagi [8]. The spectrum of light transmitted through used to compute the total hemoglobin, and the fractions
923
Oximetry
are expressed as a percentage of total hemoglobin. readings from 450 to 700 nm to construct its continuous
Calculations are performed automatically by a built-in spectrum [18]. An example of the non-hemolyzing type
microprocessor from absorbance measurements and is the Siemens (Bayer) Rapidlab 1265 analyzer, which
molar absorptivity constants. The concentrations of each utilizes a slide cell technique. It measures the
fraction are summated to yield a total hemoglobin absorbance readings at 47 wavelengths on whole blood
concentration. The proportions of each hemoglobin to construct its continuous spectrum. The sample
speciesO2Hb, metHb, and COHbare displayed. chamber has a sliding cell design that opens and closes
Calibration of the output of the instrument is (narrowing the pathway to create a narrower sample
accomplished daily using a dye solution. SHb, measured path) to allow for sample absorbance measurement. The
in the Evelyn and Molloy method [13], can be detected sample from the slide cell continues to be used for
but not quantitated in the IL 282 CO-oximeter. When gases and electrolyte measurements [19].
SHb is present, the instrument will display a positive
result for metHb and a negative result for COHb. Oxygen Saturation (sO2)
Presence of Hb F and SHb may interfere with The sO2 provides an immediate estimate of whether
quantitative measurements. A dye (amaranth) is used as additional oxygen can be carried by Hb. Oxygen
a calibrator. saturation can be accurately calculated by the use of
Equation 1 [20], directly determined from the measured
The next phase in the development of the discrete O2Hb and HHb fractions (Equation 1), estimated
wavelength CO-oximeters was in the mid-1980s, with (Equations 2,3,4), or determined by pulse oximetry.
the major change being an increase in the number of
wavelengths (6 or 7 wavelengths in general; but some, Equation 1:
such as the AVL 912, used as many as 17 wavelengths),
including a correction for lipid interference (wavelength
reading at ~ 670nm), and some even included correction
for bilirubin interference (wavelength reading ~ 488
nm). Although the Hb fractions have specific
wavelength peaks, the wavelength selection between the
different analyzers is variable. Even though wavelengths It is often referred to as the fractional sO2 (SaO2) because
used are provided, manufacturers do not explain in detail it is determined from the two hemoglobin fractions. This
the exact mathematical calculations involved in deriving is the most accurate method for sO2 estimation, since it
the results [14]. CO-oximeters are either (1) hemolyzing is a direct measure of the parameters that affect sO2. CO-
types, with samples being hemolyzed either oximetry is considered the gold standard for oxygen
ultrasonically (e.g., AVL Omni 6, CC270, IL682, saturation measurement.
OSM3) or chemically (e.g., IL482), or (2) non-
hemolyzing types (e.g., AVOXimeters 1000 and 4000, Estimated oxygen saturation does not account for
IL Synthesis 35) to make measurement on unhemolyzed variations in 2,3 DPG levels, nor the presence of
blood [15]. dysfunctional hemoglobins (COHb, MetHb). These
equations assume no presence of dysfunctional
Early CO-oximeters lacked accuracy at low hemoglobins. If any of these conditions change, the
concentrations (<5%) for COHb when compared to a GC estimated sO2 will be inaccurate. Oxygen saturation can
method and also differed in the degree of interference be calculated from the measured parameters PO2 and pH,
from Hb F, temperature, bilirubin, SHb, and medical on the basis of standard oxygen-dissociation curves [21].
dyes (e.g., methylene blue with absorbance peak at 550
to 570 nm) [16]. The other feature differing among the Equation 2: sO2 = [(PO2 + 150 x PO2)-1 x 23 400] +1-1
CO-oximeters was that not all were capable of providing
quantitated SHb. Some only showed its presence when it The Severinghaus equation assumes the following
was above certain levels such as > 1.5% (e.g., parameters are normal: pH = 7.40, PCO2 = 40 mm Hg
Radiometer ABL 700). and 2,3-diphosphoglycerate = 5 mmol/L and no presence
of dysfunctional Hbs. If any of these conditions change,
b. Continuous Spectrum CO-oximeters the sO2 will be inaccurate.
The most recent CO-oximeters measure a continuous
spectrum on a sample. Multiple readings are obtained, a Equation 3: The Radiometer ABL uses the following
spectrum is constructed, and it is compared to a model Hill equation to determine oxygen saturation from the
or typical data set. By pattern recognition, it can measured pH and O2:
compensate for any interference from turbidity,
therapeutic dyes, Hb F, or other factors. The usual
spectral range used is between 450 and 700 nm. The
available CO-oximeters are either hemolyzing or non-
hemolyzing types. Examples of systems using sample
hemolysis are the Radiometer 700 & 800 series
analyzers, which measure 128 different readings in which
between 478 and 672 nm to construct the continuous
spectrum [17]. The GEM (IL) 4000 analyzer takes 1400
924
Oximetry
Found in the equation is a Bohr factor of 0.48 and a CO-oximeters

standardized P50 of 26.6 mm Hg. Spuriously high COHb values may be encountered in the
presence of HBF or with increased oxygen saturation of
Equation 4: The Siemens (formerly Bayer) analyzers use fetal blood when older CO-oximeter models are used
the relationship described by Kelman [22] and Thomas [29]. In addition, the presence of other hemoglobins such
[23]. as SHb complicates these methods.
Discrete wavelength CO-oximeters were prone to

interferences from HbF, SHb, bilirubin, and medical
dyes. For example, the discrete CO-oximeter on the
Radiometer ABL 625 was unable to measure Hb M
N = pO2 10[0.48(pH(37) -7.4) 0.0013BE(B)] methemoglobin, unlike the continuous type used on the
735 [4]. The IL-482 CO-oximeter was able to detect
Pulse oximetry is a third method for sO2 estimation. The SHb indirectly by indicating a spuriously high metHb
saturation calculated from pulse oximetry is a functional value and a negative value for COHb [15,30]. The likely
saturation, SpO2. It is the percentage of HbO2 compared reason for SHb interference is that early CO-oximeters
with the sum of HbO2 and HHb only. Because these measured metHb absorbance at one wavelength, 630 nm.
systems do not measure COHb and metHb, the SpO2 is SHb has an absorbance peak at ~ 620 nm, which
falsely elevated in the presence of dysfunctional overlaps to 630 nm, which would have been reported as
hemoglobins. In the presence of normal SpO2 with metHb [10].
cyanosis, CO-oximetry must be used to determine the
saturation status and presence of any dysfunctional Significant bilirubin interference (1.2% COHb per 496
hemoglobin. mol/L bilirubin) was found with the CCD250 CO-
oximeter [31].
The spectrophotometric cyanmethemoglobin method is Perflubron emulsion (Oxident), a non-hemoglobin,
the reference method for total hemoglobin estimation in oxygen-carrying blood substitute, was shown to interfere
whole blood [24]. There are no stated reference methods with hemolyzing, discrete-type CO-oximeters but not
for hemoglobin fractions or oxygen saturation with non-hemolyzing types. The effects were variable,
estimation, but currently available co-oximeters, either but in general, it decreased the O2Hb, increased COHb
stand-alone or part of blood-gas analyzers, provide and MetHb, or produced turbidity errors [15].
acceptable methods for their estimation.
In one such system (an earlier model Chiron [Bayer] 800
Specimen that used 10 wavelengths to calculate the various
CO-oximetry has become the method of choice for fractions) the presence of HbF produced falsely elevated
measurement of the different hemoglobin fractions. metHb values in the clinically significant range, > 5%
Samples are whole blood anticoagulated with heparin [32]. This was overcome in the later version of the same
(reduced heparin syringes contain ~ 7 IU/mL, balanced model, with the introduction of 40 wavelengths [33].
heparin syringes contain ~ 35 or more IU/mL) or less
frequently, EDTA (1.5 mg/mL). With the ease of use Pulse Oximeters
and rapid analysis of samples on blood gas analyzers Pulse oximetry cannot distinguish the different types of
equipped with CO-oximetry or stand-alone CO- hemoglobins such as COHb and metHb [34]. With
oximeters, prolonged sample storage requirements are increasing COHb or metHb, the reported oxygen
diminished. Samples collected in this manner are stable saturation remains > 90%. MetHb absorbance at 660 nm
at least 5 days at ambient temperatures or at 4C [21]. is like that of HHb, and its absorbance at 940 nm is
Although significant loss of CO can occur during markedly greater than that of HHb or O2Hb. As a result,
storage, freezing of the sample can help maintain CO metHb will contribute to the perceived absorbances of
concentrations [25]. Some reports suggest that samples both HHb and O2Hb in a pulse oximeter. In fact, in the
can be stabilized by the addition of sodium dithionite presence of 100% metHb, the ratio is 1% or 85% oxygen
[26]. For maintaining optimal stability, anticoagulated saturation. At ~ 30% metHb, the ratio reaches a plateau,
blood should be sealed in vials with a minimum of air and the apparent oxygen saturation becomes stable at ~
space and stored frozen or at least at 3C prior to 85% [10]. COHb absorbs approximately the same
analysis [27]. COHb has been found to be stable even in amount of 660 nm light as does O2Hb [35].
postmortem blood samples stored in Vacutainer tubes for
up to 2 years at 3C with or without preservative [28]. Other factors that can lead to limitations with pulse
oximeters are sensitivity to motion, decreased peripheral
Interferences perfusion, improper probe placement, bright ambient
Spectrophotometric methods light, tattoos, and dark nail polish [36,37,38,39]. Dark
Turbidity due to proteins, lipids, and cellular matter is a skin has been found to cause overestimation in SpO2
potential problem with spectrophotometric methods. during hypoxia in dark-skinned individuals [40].
925
Oximetry
Reference Intervals bind more tightly to hemoglobin than oxygen does, it

Oxyhemoglobin and Deoxyhemoglobin also interferes with the release of the oxygen already
The O2Hb concentration in an arterial sample is 94% to bound to hemoglobin, because it causes a leftward shift
98%, and 1% to 2% for HHb. Cyanosis develops when of the oxyhemoglobin dissociation curve.
the level of HHb reaches 5g/dL (~30 %) [41].
Chronic and acute exposures are judged safe or unsafe
Carboxyhemoglobin (COHb) by the resulting fraction of COHb that results. The
Nonsmokers not otherwise exposed to external sources current standard for maximum exposure in the United
of CO have 1% to 2% COHb because of endogenous States allows a maximum concentration of 50 ppm over
production. Smokers have been found to have as little as 8 hours [44]. Concentrations of 200 ppm in inhaled air
2.1% and as much as 10% COHb, depending on are considered dangerous, but the lower limit of a safe
individual rates of tobacco consumption. Reference exposure is difficult to determine; in some persons, a 3%
intervals have been discussed by Allen [42] in more blood concentration of COHb results in impaired
detail. alertness and psychomotor performance.
Reference Intervals for COHb in Blood [41] The concentration of COHb should be considered only a
guide to diagnosis and does not always correlate closely
Population Range Mean S.D.
with clinical findings. Correlation of blood COHb levels
(n) (%) (%) (%)
with patient outcome is weak. This is because CO
Navy hospital patients persists in tissues long after COHb levels have returned
Nonsmokers to normal.
34 0.421.50 0.85 0.26
Smokers Therapy for carbon monoxide poisoning involves
34 1.538.47 3.97 1.88 reoxygenation of the blood with oxygen therapy. The
Navy divers half-life of CO in blood in a person breathing room air is
Nonsmokers 5 hours, 20 min. With a tight-fitting oxygen mask and
17 Not given 0.88 0.21 100% O2, the half-life is reduced to 1 hour, 20 min.
Smokers
19 Not given 4.66 1.68 Methemoglobin
In normal individuals, a small amount of the hemoglobin
Levels of 10% to 20% generally cause headache and in erythrocytes is oxidized to methemoglobin, and
fatigue, whereas values from 20% to 30% are typically methemoglobin concentration remains < 2% [6]. The
associated with severe headache, nausea, vomiting, majority of cases of methemoglobinemia are acquired
dizziness, blurred vision, and fainting. Coma and and mild, and therapy is focused on removal of the
convulsions can occur when COHb levels get above causative toxin. Healthy individuals with levels up to
40%. Respiratory failure and death are typically seen 15% tend to be asymptomatic. Levels between 15% and
when levels exceed 70%. 30% lead to varying symptoms such as hypoxia and
cyanosis. Levels of 30% to 50% cause dyspnea,
Methemoglobin headache, fatigue and dizziness. Breathlessness,
Levels of 1% to 2% can occur naturally in RBCs [5]. metabolic acidosis, CNS depression, confusion, and
coma are quite likely to follow levels > 50%, whereas
Sulfhemoglobin levels > 70% are fatal [4].
Concentrations > 1% should be regarded as abnormal.
Sulfhemoglobin
Interpretation SHb is an uncommon cause of cyanosis. Unlike metHb,
Carbon monoxide is a combustible, nonirritating gas that SHb has few adverse clinical consequences and can
is odorless and tasteless. Because it is a by-product of resolve spontaneously with the physiological turnover of
incomplete combustion, it is usually produced in areas red blood cells. SHb concentration of > 5 g/L (>3.0%)
where humans can be exposed to it. When carbon can produce detectable cyanosis [45]. However, it has
monoxide is inhaled, it is readily absorbed into the blood been observed that levels as low as 2g/L have produced
and binds avidly to hemoglobin. Affinity constants for dark complexion and discolored blood [46].
the COHb complex are several hundred times greater Concentrations > 1% should be regarded as abnormal. It
(220 to 250) than those for oxyhemoglobin [12,43]. The is not clear, but it has been reported that levels of 20% to
increased affinity constant is the result of an increase in 60% may be benign in some patients [7].
the absolute affinity of CO for Hb. Thus in the following
reaction: Performance Goals
k1 A review of the 2007 blood oximetry proficiency
CO + Hb HbCO surveys of the College of American Pathologists shows
k2 similar precisions for the vast majority of analyzers
included in the survey. The O2Hb coefficients of
k1 is one tenth the k1 for the binding of O2 to Hb, but k2 variation (CVs) are higher at the low end (O2Hb ~ 41 %)
is 2400-fold less than that for the dissociation of O2 < 2.5% and decrease down to ~ 1% with values at the
from oxyhemoglobin. Not only does carbon monoxide high end (O2Hb ~ 95 %). The COHb CVs are higher at
926
Oximetry
the low end (COHB ~ 5 %), where ~ 85 % of analyzers oximeters that use overdetermined systems.
have CVs < 15 %, and with the remaining analyzers, the Clin Chem 1997;43:189-190.
CVs are as high as ~ 27%. For the higher end (COHB ~ 15 Shephard RJ. Carbon Monoxide, the Silent
37 %), the CVs for all the analyzers are < 4 %. The Killer. Springfield, IL: Charles C Thomas;
proficiency samples do not contain suitable levels of 1983:44-67.
MetHb to properly test the analyzers capability. The 16 Mahoney JJ, Vremen HJ, Stevenson DK, Van
concentration of MetHb in all the samples is < 1 %. Kessel AL. Measurement of
carboxyhemoglobin and total hemoglobin by
References five specialized spectrophotometers (CO-
1 Speakman ED, Boyd JC, Bruns DE. oximeters) in comparison with reference
Measurement of methemoglobin in neonatal methods. Clin Chem1993;39:1693-1700.
samples containing fetal hemoglobin. Clin 17 Radiometer ABL800 reference manual.
Chem 1995;41:458-61. 18 IL GEM 4000 reference manual.
2 Vreman HJ, Mahoney JJ, Stevenson DK. 19 Siemens (Bayer) Rapidlab 1200 reference
Carbon monoxide and carboxyhemoglobin. manual.
Adv Pediatr 1995;42:303-34. 20 Severinghaus JW. Simple, accurate equations
3 Widdop B. Analysis of carbon monoxide. Ann for human blood O2 dissociation computation. J
Clin Biochem 2002;39:378-391. Appl Physiol 1979;46:599-602.
4 Haymond S, Cariappa R, Eby CS, Scott MG. 21 Rem J, Siggaard-Andersen O, Norgaard-
Assessment of oxygenation in Pedersen B, Storensen S. Hemoglobin
methemoglobinemia. Clin Chem 2005;51:434- pigments: photometer for oxygen saturation,
444. carboxyhemoglobin, and methemoglobin in
5 Modarai B, Kapadia YK, Kerins M, Terris J. capillary blood. Clin Chim Acta 1972;42:101-8.
Methylene blue: a treatment for severe 22 Kelman GR. Digital computer subroutine for
methaemoglobinaemia secondary to misuse of the conversion of oxygen tension into
amyl nitrite. Emerg Med J 2002;19:270-1. saturation. J Appl Physiol 1966;21:1375-6.
6 Baraka AS, Ayoub CM, Yazbeck-Karam V, 23 Thomas LJ. Algorithm for selected blood acid-
Kaddoum RN, Gerges FJ, Hadi UM, Dagher base and blood-gas calculations. J Appl Physiol
CM. Prophylactic methylene blue in a patient 1972;33:154-8.
with congenital methemoglobin. Can J Anesth 24 Recommendation for reference method for
2005;52:258-61. haemoglobinometry in human blood
7 Wu C, Kenny MA. A case of sulfhemoglobin (International Committee for the
and emergency measurement of sulfhemoglobin Standardisation in Haematology ICSH 1995)
with OSM3 CO-oximeter. Clin Chem and specifications for international
1997;43:162-6. haemoglobincyanide standard (4th edition). J
8 Severinghaus JW, Astrup PB. History of blood Clin Pathol 1996;49:271-4.
gas analysis. VI. Oximetry. J Clin Monit 25 Dominguez AM, Halstead JR, Domanski TJ.
1986;2:270-288. The effect of postmortem changes on
9 Carter BG, Carlin JB, Tibballs J, Mead H, carboxyhemoglobin results. J Forensic Sci
Hochmann M, Osborne A. Accuracy of two 1964;9:330-341.
pulse oximeters at low arterial hemoglobin- 26 Ainsworth CA, Schloegel EL, Domanski TJ,
oxygen saturation. Crit Care Med Goldbaum LR. A gas chromatographic
1998;26:1128-33. procedure for the determination of
10 Wright RO, Lewander WJ, Woolf AD. carboxyhemoglobin in postmortem blood
Methemoglobin: etiology, pharmacology, and samples. J Forensic Sci 1967;12:529-537.
clinical management. An Em Med 27 Chance DH, Goldbaum LR, Lappas NT.
1999;34:646-656. Factors affecting the loss of carbon monoxide
11 Barker SJ, Curry J, Redford D, Morgan S. from stored blood samples. J Anal Toxicol
Measurement of carboxyhemoglobin and 1986;10:181-189.
methemoglobin by pulse oximetry: a human 28 Kunsman GW, Presses CL, Rodriguez P.
volunteer study. Anesthesiology 2006;105:892- Carbon monoxide stability in stored
7. postmortem blood samples. J Anal Toxicol
12 Zijlstra WG, Buursma A, Meeuwsen-van der 2000;24:572-78.
Roest WP. Absorption spectra of human fetal 29 Vremen HJ, Ronquillo RB, Arlagno RL,
and adult oxyhemoglobin, de-oxyhemoglobin, Schwartz HC, Stevenson DK. Interference of
carboxyhemoglobin, and methemoglobin. Clin fetal hemoglobin with spectrophotometric
Chem 1991;37:1633-8. measurement of carboxyhemoglobin. Clin
13 Evelyn KA, Malloy HT. Microdetermination of Chem 1988;34:975-7.
oxyhemoglobin, methemoglobin, and 30 Rausch-Madison S, Mohsenifar Z.
sulfhemoglobin in a sample of blood. J Biol Methodological problems encountered with co-
Chem 1938;126:655-662. oximetry in methemoglobinemia. Am J Med
14 Yakuwa N, Suzuoka T, Forrest ARW, Osawa Sci 1997;341:203-6.
M, Takeichi S. Data processing in CO-
927
Oximetry
31 Vreman HJ, Stevenson DK. 48 Van Slyke DD, Hiller A, Weisiger JR, Cruz
Carboxyhemoglobin determined in neonatal WO, Plazin J. Determination of carbon
blood with a CO-oximeter unaffected by fetal monoxide in blood and of total and active
hemoglobin. Clin Chem 1994;40:1522-1527. hemoglobin and methemoglobin contents of
32 Lynch PL, Bruns SE, Boyd JC, Savory J. normal human blood. J Biol Chem
Chiron 800 system CO-oximeter module 1946;166:121-148.
overestimates methemoglobin concentration in 49 Conway EJ. Microdiffusion Analyses and
neonatal samples containing fetal hemoglobin. Volumetric Error. 4th ed. London: Crosby,
Clin Chem 1998;44:1569-70. Lockwood & Sons; 1957:7-13, 326-329.
33 Keijzer M H. Comment on the overestimation 50 Collinson HA, Rodkey FL, ONeal JD.
of methemoglobin concentrations in neonatal Determination of carbon monoxide in blood by
samples with the Chiron 800 system CO- gas chromatography. Clin Chem 1968;14:162-
oximeter module. Clin Chem 1999;45:1313-4. 171.
34 Mack E. Focus on diagnosis: Co-oximetry. 51 Ostrander CR, Cohen RS, Hopper AO, Cowan
Pediatr Rev 2007;28:73-74. BE, Stevens GD, Stevenson DK. Paired
35 Hampson NB. Pulse oximetry in severe carbon determinations of blood carboxyhemoglobin
monoxide poisoning. Chest 1998;114:1036-41. concentration and carbon monoxide excretion
36 Costarino AT, Davis DA, Koen TP. Falsely rate in term and preterm infants. J Lab Clin
normal saturation reading with pulse oximeters. Med 1982;100:745-755.
Anesthesiology 1987;67:830-831. 52 Vreman HJ, Kwang LK, Stevenson DK. Carbon
37 Clayton D, Webb RK, Ralston AC, Duthie D, monoxide in blood: an improved microliter
Runciman WB. A comparison of the blood-sample collection system, with rapid
performance of 20 pulse oximeters under analysis by gas chromatography. Clin Chem
conditions of poor perfusion. Anesthesia 1984;30:1382-1386.
1991;46:3-10. 53 Zwart A, Buursma A, van Kampen EJ,
38 Hinkelbein J, Genzwuerker HV, Sogl R, Fiedler Oeseburg B, van der Ploeg PH, Zijlstra WG. A
F. Effect of nail polish on oxygen saturation multi-wavelength spectrophotometric method
determined by pulse oximetry in critically ill for the simultaneous determination of five
patients. Resuscitation 2007;72:82-91. haemoglobin derivatives. J Clin Chem Clin
39 Severinhaus JW, Naifeh KH, Koh SO. Errors in Biochem 1981;19:457-463.
14 pulse oximeters during profound hypoxemia. 54 Zwart A, Buursma A, Oeseburg B, Zijlstra WG.
J Clin Monit 1989;5:72-81. Determination of hemoglobin derivatives with
40 Bickler PE, Feiner JR, Severinghaus JW. the IL 282 CO-oximeter as compared with a
Effects of skin pigmentation on pulse oximeter manual spectrophotometric five-wavelength
accuracy at low saturation. Anesthesiology method. Clin Chem 1981;27:1903-1907.
2005;102:715-9. 55 Zijlstra WG, Buursma A, Meeuwsen-van der
41 Grace RF. Pulse oximetry: gold standard or Roest WP. Absorption spectra of human fetal
false sense of security? Med J Aust and adult oxyhemoglobin, de-oxyhemoglobin,
1994;160:638-44. carboxyhemoglobin, and methemoglobin. Clin
42 Allen LC. Determination of carbon monoxide Chem 1991;37:1633-8.
in blood. Therapeutic Drug Monitoring and 56 Dennis RC, Valeri CR. Measuring percent
Toxicology. Washington, DC: American saturation of hemoglobin, percent
Association for Clinical Chemistry; 1984:1-10. carboxyhemoglobin and methemoglobin and
43 Goldsmith JR, Landau SW. Carbon monoxide concentrations of total hemoglobin and oxygen
and human health. Science 1968;162:1352- in blood of man, dog, and baboon. Clin Chem
1359. 1980;26:1304-1308.
44 Norman P, Wherrett JR. Carbon monoxide 57 Zwart A, Buursma A, van Kampen EJ, Zijlstra
poisoning. Therapeutic Drug Monitoring and WG. Multicomponent analysis of hemoglobin
Toxicology. Washington, DC: American derivatives with reversed-optics
Association for Clinical Chemistry; 1984:11- spectrophotometer. Clin Chem 1984;30:373-9.
16. 58 Zwart A, Buursma A, van Kampen EJ, Zijlstra
45 Tfelt-Hansen P, De Vries P, Saxena PR. WG. Results of routine determination of
Triptans in migraine: a comparative review of clinically significant hemoglobin derivatives by
pharmacology, pharmacokinetics and efficacy. multicomponent analysis. Clin Chem
Drugs 2000;60:1259-87. 1986;32:972-8.
46 Flexman AM, Del Vicario G, Schwartz SKW. 59 Blanchette E, Tagilaferro D, Hotaling T,
Dark green blood in the operating theatre. Uretsky LS. Improvements in robustness of the
Lancet 2007;369:1972. Evelyn and Molloy
47 Van Slyke DD, Neill JM. The determination of methemoglobin/sulfhemoglobin manual assay.
gases in blood and other solutions by vacuum Clin Chem 2006;52(Suppl A):191.
extraction and manometric measurement. J Biol 60 Biochemical aspects of hematology. Chapter
Chem 1924;61:523-573. 46. In: Tietz Textbook of Clinical Chemistry.
3rd ed. Philadelphia: Saunders; 1987:1677.
928
Oximetry
61 Rodkey FL, Hill TA, Pitts LL, Robertson RF. 63 Rand RN. Practical spectrophotometric
Spectrophotometric measurement of standards. Clin Chem 1969;15:839-863.
carboxyhemoglobin and methemoglobin in 64 Frings CS, Broussard LA. Calibration and
blood. Clin Chem 1979;25:1388-1393. monitoring of spectrometers and
62 Beutler E, West C. Simplified determination of spectrophotometers. Clin Chem 1979;25:1013-
carboxyhemoglobin. Clin Chem 1984;30:871- 1017.
874.
Table 1: Absorption Peaks of Hemoglobin Fractions

Fraction Wavelength (nm)
O2Hb 542, 577
HHb 431, 556
COHb 538, 568
MetHb 500, 630
SHb 620
Table 2: Matrix of Millimolar Absorptivities

Millimolar Absorptivities Expressed in cm-1mmol-11 (10-2m2mol-1; Lineic Absorbances)
(nm) HHb O2Hb COHb metHb S Hb

500 4.09 5.05 5.35 9.04 7.20
569 11.27 11.27 14.27 4.10 8.10
577 9.40 15.37 10.00 4.10 8.10
620 1.23 0.24 0.33 3.35 20.80
760 0.43 0.13 0.03 0.24 1.04
Reproduced with permission from Zwart A et al. Determination of hemoglobin derivatives with the IL 282 CO-oximeter as
compared with a manual spectrophotometric five-wavelength method. Clin Chem 1981;27:1903.
Figure 1: Absorption spectra of Hemoglobin Fractions

Absorption spectra of reduced hemoglobin (HHb), carboxyhemoglobin (COHb), oxyhemoglobin (O2Hb), methemoglobin
(MetHb), sulfhemoglobin (SHb), and lipid.
Reproduced with permission from Widdop B. Analysis of carbon monoxide. Ann Clin Biochem 2002;39:378-391.
929
Oximetry
Figure 2: Absorption spectra of adult and fetal hemoglobin

Light absorption spectra of the common derivatives of HbA and HbF in the visible range. Left panel: O2Hb (1) and HHb
(2). Right panel: MetHb (1) and COHb (20. The absorptivity is expressed in Lmmol-1 cm-1.
Reproduced with permission from Zijlstra WG, Buursma A, Meeuwsen-van der Roest WP. Absorption spectra of human
fetal and adult oxyhemoglobin, de-oxyhemoglobin, carboxyhemoglobin and methemoglobin. Clin Chem 1991;37:1633-8.
Historical Methods lactic, or hydrochloric, which acts to release CO (and

CO2) to the atmosphere inside the closed dish. In the
inner chamber of the dish is a solution of palladium
1. Carbon Monoxide (CO) chloride. The PdCl2 is reduced by the CO to metallic
palladium as the CO diffuses through the atmosphere in
a. Comparative Color Test
the dish and settles into the inner section of the dish.
Individuals with CO poisoning often have blood that is
After allowing time for CO to react as completely as
cherry red in appearance. However, lack of a cherry-red
possible, an aliquot of inner fluid is removed, and the
appearance does not exclude poisoning. One
modification of this simple visual inspection test is to remaining palladium chloride is titrated with I, or
compare 0.5 mL of blood diluted with 10 mL of 10 measured photometrically at 278 nm. The amount of
mmol/L ammonia with that of normal blood treated in palladium chloride remaining is inversely proportional to
the same manner. A pink coloring indicates the presence the CO content of the sample. The method is most useful
of COHb. However, blood from patients with cyanide when the CO concentrations are at levels of 10% or
poisoning may also cause a pink coloration. more.
Gas chromatography can also be used to measure CO

that has been liberated in a closed system. The column
b. Carbon Monoxide (CO) Measurements packing frequently used is Molecular Sieve 5A, which is
CO is present in blood both as dissolved CO and as used to separate and analyze many gas mixtures. In
COHb. Direct analysis procedures measure total CO by many cases, a thermal conductivity detector is used to
releasing the CO from COHb and from solution and then monitor the column effluent. Calibration can be
determining the amount of CO by one of several accomplished by use of a gas containing a known
measurement techniques: manometry, microdiffusion, amount of carbon monoxide, or use of blood that has
gas chromatography, or infrared analysis. been tonometered to contain a known percentage of
carboxyhemoglobin [50]. In the gas chromatographic
Manometric techniques use the Van Slyke apparatus method of Collinson et al. [50] as modified by Ostrander
[47,48]. Other gases present, such as oxygen, nitrogen, et al. [51] and Vreman et al. [52], blood is collected in
and carbon dioxide, are chemically removed, and then capillary tubes that have been coated internally with
the carbon monoxide is liberated by the addition of heparin and saponin. The anticoagulated blood is mixed
acidified ferricyanide, and the volume percent of CO is with potassium ferricyanide (K3Fe[CN]6, 100 g/L) in a
measured. vial, and hemoglobin-bound carbon monoxide is
liberated into the headspace of the vial. The vials air
A Conway diffusion method [49] employs a dish space is sampled with a needle through the septum cap
comprising two concentric chambers. In the outer and injected into the gas chromatograph. Air, the carrier
chamber, sample is mixed with an acid, usually sulfuric,
930
Oximetry
gas (50 mL/min), must be purified by passage through a over time. Simply put a couple of blood drops on white
catalytic converter and a moisture trap, which uses filter paper and look for color change. Unlike the
magnesium perchlorate prior to usage. Although the brightening of the original dark red HHb color with
method is very precise and accurate, it is too slow and exposure to air, the chocolate brown color of metHb
labor intensive to be used in emergent situations. does not change [4]. The only problem is that this does
not distinguish metHb from SHb, which if present, also
c. CO Reference and Preferred Methods gives blood a dark brown appearance.
The modified GC method [51,52] of Collinson et al [50]
can detect as little as 0.1 nL of CO in 2 L of blood. b. Photometric Methods
Reproducibility of the method at 1.80 mL of CO/L is Measurements of the proportion of hemoglobin fractions
0.04 mL of CO/L. Linearity of the method extends to in blood are usually made photometrically, employing
9.36 mL of CO. various wavelengths at which the hemoglobin species in
blood absorb light. Because the absorption spectra
Gas chromatography can be used to measure the total overlap, it is usually necessary to make two or more
CO content in blood regardless of binding state; in measurements and then solve simultaneous equations.
addition, gas chromatography requires only a small Spectra of various hemoglobin species are shown in
sample volume. Because it measures total CO, pre- Figure 1. Binary mixtures can be analyzed with
measurement steps are simplified, and CO measurement measurements at two wavelengths. Since one normally
in other tissues is possible. The disadvantage of gas encounters oxyhemoglobin (O2Hb), deoxyhemoglobin
chromatography as a routine tool for CO analysis in the (Hb), and COHb in blood to be analyzed for COHb,
clinical chemistry laboratory is that it is a relatively methods have been developed to convert these samples
complex and slow procedure compared to into two-component systems. This is accomplished by
spectrophotometry. However, the GC method is highly the addition of sodium dithionite (Na2S2O4) to blood,
accurate and can serve as a reference method.
which converts O2Hb to HHb. The HHb and COHb are
The spectrophotometric method of Zwart et al. [53,54] then measured at 420 and 432 nm, respectively. The
compares favorably with the four-wavelength method presence of other hemoglobins, such as Hb F or SHb,
when SHb is not present. There are three major sources complicates these methods. To cope with the presence of
of error in this technique. Even the introduction of a these hemoglobins, one can increase the number of
small amount of oxygen in the detergent solution can measurements, and the resulting simultaneous equations
cause significant error in the analysis. Use of literature can be solved to quantitate these other Hb species.
values of molar absorptivities in an uncalibrated Availability of microprocessors or computers simplifies
spectrophotometer can bias the results. In addition, the solution of the equations, but one is also required to
presence of other hemoglobin moieties may make the standardize the multicomponent systems to determine
method inaccurate. the constants used in the equations. Some of the tedium
of performing these tasks can be avoided by purchase of
In blood samples containing fetal hemoglobin, dedicated systems that are programmed to perform the
spuriously high COHb values were encountered by calculations and have been standardized at the factory.
Zwart et al. [55] using co-oximetry techniques. These These topics have been previously reviewed [42,44].
values increased with the increased oxygen saturation of
fetal blood. However, in the absence of fetal blood, the Zwart and co-workers [53] have developed a
instrument compared well with the reference method spectrophotometric method that allows the measurement
cited here. Zwart and co-workers [53] obtained the of hemoglobin derivatives in a simple hemolysate in the
regression equation for the IL-282 CO-oximeter as presence of all hemoglobins: O2Hb, metHb, HHb, and
follows: SHb. Absorbances of the hemolysate must be measured
at each of five wavelengths, since a five-component
% CO Hb (IL-282) = 1.03 (%CO Hb [ref. mixture requires five simultaneous equations of the
method]) 0.05 following general type to be constructed from Beers
law:
for n = 20. The range of %Hb CO was 10 to 80. Similar
correlation was obtained with gas chromatography [56]. Anm = a1(nm)bC1 + a2(nm)bC2 ...
Precision of measurements was determined to be 0.4%, + a5(nm)bC5
expressed as the maximum range of differences of
triplicate measurements. Precision of the total where Anm is the absorbance at wavelength nm, a is the
hemoglobin measurement of a single sample expressed
molar absorptivity of the numbered species at
as the standard deviation was 0.8 g/L at a mean of 141.6
wavelength nm, b is the pathlength of the cuvette used,
g/L, n = 20.
and C is the concentration of the numbered species.
Zwart [53] used wavelengths that allow measurement of

2. Hemoglobin Fractions
the absorbances at a wavelength maximum wherever
a. Blood Color Test
possible: Hb, 760 nm; O2Hb, 577 nm; COHb, 569 nm;
In a cyanosed patient, the differentiation of HHb from
MetHb can be done by observing the blood color change metHb, 500 nm; SHb, 620 nm. Molar absorptivities at
these wavelengths were derived from standards and are
931
Oximetry
given in Table 2: Millimolar Absorptivities. Because of 1. Label a cuvette C1, and add to it 1.5 mL of
the large differences in a, two pathlengths are required to phosphate buffer and 1.5 mL of water. Zero the
keep observed absorbances within the range of most spectrophotometer at 620 and 630 nm.
instruments without the need for dilution. At 620 and 2. Obtain a fresh whole-blood sample (EDTA or
760 nm, b = 0.200 cm, but at 500, 569, and 577 nm, b = lithium heparin).
0.007 cm. Aliquot 0.2 mL of blood into a tube containing
7 mL of distilled water. Mix by gentle
The hemolysate is prepared when one mixes heparinized inversion. Wait 5 to 10 minutes. Add 4.0 mL of
whole blood in a syringe with a 5% solution of a K2HPO4 buffer. Mix the specimen, and
nonionic detergent, nonoxynol (Sterox). One filters the centrifuge at 4000 rpm for 10 min to remove
cellular debris from the hemolysate by expelling the cell stroma and unhemolyzed cells.
syringe contents through a cotton fiber filter. 3. Transfer 3.0 mL of clear supernatant from step
Absorbances are measured in a narrow bandpass 2 into a cuvette labeled C2,
spectrophotometer (<1 nm) against a water blank. Blood measure the absorbance at 630 nm, and
samples with various proportions of hemoglobin species designate this value A1.
are prepared and used for calibration. 4. Transfer 1.0 mL of clear supernatant from step
2 into a cuvette labeled C3, and add 1.0 mL of
Zwart et al. [57,58] next used a multiwavelength method K2HPO4 buffer and 1.0 mL of water. Mix by
for multicomponent analysis of Hb derivatives. The gentle inversion.
lithium heparin whole-blood sample was hemolyzed 5. Add 0.1 mL of KCN to cuvettes C2 and C3.
using100 mL/L solution of Nonidet-P40. The absorption Cover with Parafilm, and invert several times to
spectrum was determined between 480 and 650 nm and mix thoroughly. Allow to stand for 5 minutes.
absorbance measurements taken in increments of 2 nm 6. Measure the absorbance of cuvette C2 at 630
(86 absorbance readings) with a diode array nm (A2) and 620 nm (A3) against a blank
spectrophotometer. The concentrations of the individual cuvette C1.
components are calculated using the spectrophotometer 7. Add 0.1 mL of K3F(CN)6 solution to cuvettes
computer. C1 and C3. Cover with Parafilm, and invert
several times to mix thoroughly. Allow to stand
1. Methemoglobin and Sulfhemoglobin Method for 30 minutes.
Measurement is based on the original Evelyn and 8. Set spectrophotometer to 540 nm and adjust to
Molloy [12] reference method with modifications as zero absorbance with cuvette C1. Measure
suggested by Blanchette et al. [59] of the Teitz [60] absorbance of C3 (A4).
textbook method.
Calculations
Principle: Total Hb (g/dL) = F1 A4
MetHb has an absorbance peak at 530 to 535 nm. The MetHb (g/dL) = F2 (A1-A2)
addition of cyanide converts metHb to cyanmetHb. The SHb g/dL) = 80 F3 (A3-F4(A1-A2)-3 F5 A4)
decrease in absorbance is proportional to the metHb MetHb % = [(MetHb (g/dL)/total Hb g/dL)] 100
concentration. SHb has an absorbance peak at 618 to 622 SHb % = [(SHb (g/dL)/total Hb g/dL)] 100
nm. SHb is not converted by cyanide to cyanmetHb so
there will be no change in absorbance. Where F1=35.1, F2=33.6, F3=0.64 and F4=0.300 and
F5=0.022
Reagents
1. Potassium phosphate buffer, 0.15 mol/L, pH 6.6: 2. Carboxyhemoglobin Manual Spectrophotometry
Dissolve 17.1 g of K2HPO43H2O (or 13.2 g anhydrous) Method
in 500 mL of distilled water and 10.2 g of KH2PO4 in In the selection of a general method for
500 mL of distilled water. Transfer 100 mL of each the carboxyhemoglobin measurements, however,
two solutions to a new beaker. Adjust the pH to 6.6 considerations of accuracy and simplicity predominated
using the K2HPO4 solution. over versatility. The preferred method presented below
is based on Rodkey et al. [61] as modified by Beutler
2. KCN solution: Dissolve 500 mg of KCN in ~ 7 mL and West [62].
of water, and dilute to 10 mL.
Principle
3. K3F(CN)6 solution: Dissolve 2.0 g of in ~ 7.0 mL Hemoglobin in blood is mixed with sodium hydrosulfite
distilled water, and dilute to 10 mL. (Na2S2O4), which reacts with oxyhemoglobin and
methemoglobin to produce reduced hemoglobin, thereby
Assay converting a multicomponent system into a two-
Equipment: Spectrophotometer, narrow bandpass (<1 component system: carboxyhemoglobin and reduced
nm), that has been calibrated and checked regularly for hemoglobin. Absorbance measurements are made at two
wavelength accuracy, photometric accuracy, and wavelengths, 420 and 432 nm, using a narrow bandpass
linearity and for the presence of stray radiant energy. spectrophotometer, which if properly calibrated, allows
one to employ literature values for molar absorptivities
932
Oximetry
of reduced Hb and Hb CO in the calculations using reagent-grade water. Bring volume to 1 L, and mix well.
Beers law. Prepare pH 6.85 buffer by mixing equal amounts of each
stock buffer. Check pH of solution and adjust to pH 6.85
At each wavelength used in the measurement step, with 0.1 M HCl or NaOH. This solution, reagent 1, is
Beers law gives the absorbance as follows: stable for 2 months at 4C to 8C. Check weekly for
bacterial growth.
A420 = (aHb420[1 x] + aHb CO420[x])bC 2. Hemolyzing solution. Buffer, reagent 1,
1 diluted one part to 9 parts of distilled water. Prepare and
A432 = (aHb432[1 x] + aHb CO432[x])bC use as needed from buffer stock.
3. Hb CO diluting solution. To 25 mg of sodium
2
hydrosulfite, add 20 mL of buffer, reagent 1 (see above).
Prepare this solution with minimum introduction of air.
where a is the molar absorptivity of each species at each
In a procedure described by Rodkey [61], a test tube is
wavelength, b is the cuvette pathlength in cm, C is the
filled to the top with buffer containing glass beads for
molar concentration of hemoglobin iron in solution, and
agitation. Solid sodium hydrosulfite is added to the full
x is the fraction of the total amount of hemoglobin that is
tube, and the tube is covered and inverted to dissolve the
present as carboxyhemoglobin. Note that (1 x)
sodium hydrosulfite. This reagent must be prepared
represents the fraction of the hemoglobin present as
immediately before use.
reduced hemoglobin.
Assay
Combining equations 1 and 2 and solving for x
Equipment: Spectrophotometer, narrow
yields the following:
bandpass (<1 nm), that has been calibrated and checked
regularly for wavelength accuracy, photometric
accuracy, and linearity and for the presence of stray
3
radiant energy. Procedures and tools for these quality
x = ___________(A432)(aHb420)
assurance measures are given by Rand [63] and Frings
(A420)(aHb432)___________ [64].
(A420[aHb CO432 aHb432]) 1. Add 25 L of whole anticoagulated blood to 3
(A432[aHb CO420 aHb420]) mL of hemolyzing solution in a small test tube.
Mix two or three times by inversion. Let stand
about 5 min to ensure complete lysis of cells.
Rodkey [61] has published carefully derived molar
2. In a 1-cm glass cuvette, mix 0.2 mL of the
absorptivity values (a) required to solve the above
hemolysate from step 1 and 2.3 mL of the Hb
equation. The values given below are derived from
CO diluting solution. Cover with Parafilm, and
human hemoglobins:
invert to mix. Let stand 10 to 60 min, no longer.
3. Read the absorbances of the sample at 420 and
Wavelength aHb aHb CO
432 nm against a matched 1-cm glass cuvette
420 0.988 1.970
containing Hb CO diluting solution. Record
432 1.317 1.237
A420 and A432.
The user must verify the literature values of molar Calculations
absorptivities on the spectrophotometer used for analysis 1. Calculate Ar = A420/A432 for each unknown.
and should regularly check wavelength calibration, 2. Using equation 4, calculate the fraction of Hb
photometric accuracy and linearity, absence of stray CO, using literature values of the constants as
radiant energy, and cuvette pathlengths. follows: [61]
Equation 3 has been simplified by Beutler [62] to the F1 = 1.3330, F2 = 0.4787, F3 = 1.9939,
following: after confirming that they are valid to use in
%x = __(1 [Ar][F1])100%__ your laboratory procedure. Verify as often as
Ar(F2 F1) F3 + 1 the spectrophotometer is subjected to source-
lamp changes or electronic repair.
where Ar = A420/A432 Calibration and Controls
1. Frequent confirmation of the reference intervals
F1 = aHb432/aHb420
for smokers and nonsmokers is helpful to
F2 = aHb CO432/aHb420 ensure accuracy.
F3 = aHb CO420/aHb420 2. Determination of molar absorptivity ratios can
%x = percentage of Hb CO. be accomplished as described by Beutler
[62].
Reagents a. Prepare a hemolysate using 6 mL of
1. KH2PO4/K2HPO4 buffer, 0.1 mol/L, pH the hemolysate reagent and 50 L of
whole blood from a nonsmoking
6.85. Weigh 13.6 g of KH2PO4 (anhydrous), and
person with no known recent exposure
dissolve in approximately 800 mL of reagent-grade to other sources of carbon monoxide.
water. Bring volume to 1 L and mix well. Weigh 17.4 g b. Using a class A volumetric pipet, add
of K2HPO4, and dissolve in approximately 800 mL of 2 mL of the hemolysate to each of two
933
Oximetry
class A 25-mL volumetric flasks. F2 = AHb CO432/AHb420;

Accuracy in pipetting is crucial to the F3 = AHb CO420/AHb420.
accuracy of the determination of these
molar absorptivities. Calibration of the
3. Measurement of Carboxyhemoglobin in Breath
pipets and flasks is recommended.
Sensors have been developed that enable measurement
c. In a well-ventilated hood and with
of carbon monoxide in air or expired breath samples.
frequent swirling and stirring, allow
With this technique, CO is oxidized at the anode, with
carbon monoxide gas to pass through
resultant generation of an electrical signal as follows:
one flask while compressed air is
passed through the other. Use a bright CO + H2O CO2 + 2H+ + 2 e-
light to dissociate any traces of carbon
monoxide from the hemoglobin in the Oxygen in the air is reduced at the cathode:
air flask.
d. Bring each flask to volume with Hb 2O + 8H+ + 8e- 4H O
2 2
CO diluting solution, and mix well.
Immediately measure the absorbance The current that is generated in this system can be
(A) at 432 and 420 nm for each diluted displayed as parts per million (ppm) or as a percentage
hemolysate, using matched 1-cm glass of COHb saturation if measured in breath. This
cuvettes and blanking against Hb CO technique offers the advantages of being noninvasive
diluting solution. and relatively inexpensive, and it requires little training
e. Calculate molar absorptivity factors for use [3].
(F) from the ratio of observed
absorbance values as follows:
F1 = AHb432/AHb420;
934
Parathyroid Hormone (PTH, Parathyrin)

Susan Vickery and Edmund J Lamb
Name: Parathyroid hormone, parathyrin, parathormone, PTH

Molecular mass: 9500 D
Merck Index: 6898
Chemical class: Polypeptide
i Different antibodies were available that bound to
epitopes either in the mid-region or the C-terminus of
Parathyroid hormone (PTH) is a single polypeptide PTH. Highly pure isotopically labeled intact PTH would
consisting of 84 amino acids. Intact PTH (1-84) is then competitively displace bound sample PTH [6]. Such
secreted by the parathyroid glands in response to low assays were insensitive, owing to the poor cross-
circulating (ionized) calcium concentration. The most reactivity between sample PTH and the non-human
important role of PTH is to increase blood calcium antibody [7,8].
concentration by increasing bone resorption, increasing
the tubular reabsorption of calcium in the kidney, and It then became apparent that CPTH fragments secreted
increasing the production of 1,25- by the parathyroid glands and produced during the
dihydroxycholecalciferol (vitamin D), which metabolism of PTH were being measured in addition to
subsequently increases absorption of calcium from the intact PTH (1-84) [9]. CPTH fragments, are found in
gut [1]. PTH determination is important in the diagnosis varying amounts in all patients but accumulate in
and management of primary hyperparathyroidism patients with renal disease. Circulating concentrations of
(oversecretion of PTH from the parathyroid glands), CPTH fragments accounted for up to 95% of the
secondary hyperparathyroidism (e.g., due to kidney measured PTH in patients with end-stage renal disease
disease), vitamin D deficiency, and other hyper- and (ESRD), arising from increased glandular secretion of
hypocalcemic states. CPTH fragments, increased metabolism of intact PTH,
and reduced glomerular filtration [8,10]. For patients
The complete amino acid sequence of human PTH has with renal disease, PTH concentrations were
been determined [2]. Parathyroid cells synthesize a pre- overestimated and unreliable. Attempts were made to
pro-hormone, consisting of 115 amino acids, which is use an antibody that could recognize an epitope in the N-
cleaved immediately after its synthesis to give the 90 terminal region of PTH, which the CPTH fragments lack
amino acid compound pro-PTH [3]. The removal of six [11]. However, the poor sensitivity of these assays
amino acids from the carboxyl terminus of the molecule limited their clinical use.
yields the final glandular product PTH with residues 1-
84. Classical biological activity, mediated via interaction First-generation RIAs have been replaced with
with the PTH receptor PTH1R, resides in the N-terminus immunometric assays that detect PTH using two
of the molecule. C-terminal (CPTH) fragments are also different antibodies targeted at two distinct epitopes in
directly secreted by the parathyroid gland. Within the the PTH molecule. Typically, the first antibody,
liver and kidney, PTH (1-84) is cleaved into N-terminal immobilized to a solid phase, binds to a region near the
and CPTH fragments. Circulating PTH is a mixture of C-terminus and removes PTH from the serum. A second
PTH (1-84) and CPTH fragments [4]; this raises labeled antibody then binds to the N-terminus of the
problems when immunological techniques are employed immobilized PTH [12]. Only full-length PTH (1-84) is
for the measurement of this hormone. capable of binding both antibodies, so CPTH fragments
are not measured. These assays have been termed intact
Berson et al. described the first radioimmunoassay (RIA) PTH assays. They are highly sensitive, reasonably
for the determination of PTH in 1963 [5]. The assay specific, readily automated, and can be used with a
utilized a single antibody obtained from polyclonal variety of labels: radiolabeled isotopes,
antisera raised against partially pure non-human PTH. chemiluminescent, biotinylated, or enzyme-linked labels
[13]. Such second-generation PTH assays have been in
routine clinical use for more than a decade.
i
Parathyroid hormone (PTH)
Previous and current authors of this method: In the mid-1990s, a Canadian research group discovered
First edition: Not done that another molecular form of PTH was being detected
Methods book: Helmut W. Minne, Reinhart Ziegler by the second-generation assays [14,16]. In high-
Second edition: Not done performance liquid chromatography, this molecule co-
Third edition: Not done chromatographs with synthetic PTH (7-84). The
Fourth edition: Helmut W. Minne, Reinhard Ziegler discovery that the second-generation intact PTH assays
Fifth edition: Susan Vickery, Edmund J. Lamb measure other PTH molecules has lead to the
935
development of specific PTH (1-84) assays, referred to Specimen

here as third-generation assays. Third-generation assays EDTA plasma is the preferred sample; PTH in EDTA
are defined by their ability to measure whole, intact PTH whole blood is stable for up to 48 h at room temperature
(1-84) without cross-reacting with PTH (7-84), now prior to separation (centrifugation >1000g for 15 to 20
thought to be a major C-terminal circulatory form that min) [23]. This is of benefit in clinical situations where
cross-reacts with second-generation assays. Exploitation rapid separation of samples may be hard to achieve.
of the difference between the two N-terminal regions has Once separated, EDTA plasma should be frozen at
provided an antibody that exclusively detects PTH (1- 20C (for up to 3 months) prior to analysis. Recurrent
84); polyclonal antibodies are targeted to epitopes within freeze/thaw cycles are not recommended.
the first four amino acids from the N-terminal [17].
Third-generation assays at present are expensive, and Serum specimens collected into plain tubes can also be
very few are automated [18]. used, providing samples are transported rapidly to the
laboratory, allowed to clot at room temperature, and then
With the introduction of third-generation PTH assays, separated and frozen without delay (1). To facilitate
numerous studies have been undertaken to compare interpretation, serum calcium, albumin (to allow
second- and third-generation assays. Recent studies have correction of calcium concentration), and renal function
shown that third-generation assays are more sensitive (i.e., creatinine and eGFR) should always be measured at
(>23%) for detecting increased plasma PTH in primary the same time as PTH concentration.
hyperparathyroidism than first- or second-generation
assays [19,20]. In contrast, a study comparing a third- Interferences
generation Bio-PTH (1-84) assay with a second- Unless a third-generation assay solely measuring PTH
generation intact PTH assay concluded a good (1-84) is used, CPTH fragments will be detected in
correlation between the two assays, with both providing addition to PTH (1-84); this is particularly applicable to
similar clinical information [21]. renal disease patients. The instability of PTH in serum
can result in spuriously decreased values in samples that
Reference and Preferred Methods are not analyzed immediately following collection or
A reference preparation of purified human PTH for PTH immediately frozen. One manufacturer has reported that
immunoassays (WHO 79/500) was prepared in 1981 samples from patients routinely receiving high dose
[22]. However, its content is formally defined in biotin therapy may show falsely decreased results. As
International Units (0.1 IU/ampoule), whereas PTH with all other immunoassay procedures, the presence of
measurements tend to be expressed in mass or molar heterophilic antibodies could have the potential to cause
units. Some manufacturers do claim traceability to interference in PTH measurements. Following good
79/500, but its nominal mass concentration is probably laboratory practice, grossly hemolyzed, icteric, or
inaccurate and does not provide a sound basis for lipemic samples should not be used.
conversion to the use of mass units for reporting PTH.
Further, stocks of 79/500 are becoming depleted. A PTH Reference Interval
newer preparation based on recombinant PTH and The upper limit of normal for PTH in healthy individuals
defined in mass units has been prepared by the National is dependent on the assay used, for example:
Institute of Biological Standards and Controls (NIBSC
95/646; http://www.nibsc.ac.uk/documents/ifu/95- First-generation PTH RIA:
646.pdf) but has not been widely adopted to date. 10 to 65 ng/L, Nichols Advantage Chemiluminescence
Intact Parathyroid Hormone
There is currently no reference method for PTH
measurement. Second-generation PTH immunometric assays:
11 to 79 ng/L, Intact PTH assay; Siemens Medical
Second-generation immunometric intact PTH assays are Solutions Diagnostics
currently used in most clinical laboratories. The
distinguishing feature between these assays is their mode Third-generation PTH immunometric assay:
of detection; there is a clear decline in the use of 5 to 39 ng/L, Whole PTH (1-84) Specific Scantibodies
radiolabeled isotopes, where chemiluminescence (using Laboratory, Inc.
acridinium esters) is now the preferred choice [7]. Most
commercial assays are now fully automated. Of the 257 To convert ng/L to pmol/L, multiply by 0.106.
participants in the United Kingdom National External
Quality Assessment Scheme (UK NEQAS) for plasma The lack of international standardization makes direct
PTH in March 2007, all of which used second- comparisons between assay systems difficult.
generation assays, the major reagent manufacturers were Laboratories must use a reference range appropriate for
Roche Diagnostics Ltd (39%, Modular and Elecsys their assay. A further difficulty in defining reference
platforms), Diagnostic Product Corporation (30%), intervals for PTH is the high prevalence of (a) occult
Siemens Medical Solutions Diagnostics (18%, Advia primary hyperparathyroidism and (b) vitamin D
Centaur platform), and Abbott Diagnostics (5%). deficiency in the reference populations studied.
No significant gender- or race-related differences have

been noted. Plasma concentrations during the first few
936
days of life are lower when compared with those in older Siemens Medical Solutions Diagnostics (5.2%), Roche
children [24]. For older children, however, PTH Diagnostics Ltd (5.7%), Abbott Diagnostics (6.4%), and
concentrations are lower and in a narrower range when Diagnostics Product Corporation (10.6%) were
compared with those in adults [25]. In adults, a acceptable. Reported intraindividual variation for PTH is
significant increase in concentration has been noted with 25.9% in healthy individuals [32]. Assuming a typical
aging. One study found that compared with women performance goal of 50% of the biological variation, this
under 40 years of age, women in their fifth through is achieved by most modern PTH assays. Within-
seventh decades of life had PTH concentrations that laboratory CVs for the Siemens Medical Solutions
were 24%, 36%, and 47% higher, respectively [26]. A Diagnostics Intact PTH assay are typically 10.2%, 5.9%,
diurnal variation, with highest values occurring in the and 4.3% for plasma PTH concentrations of 36, 232 and
early to mid-midmorning hours, has been described by 855 ng/L, respectively (Department of Clinical
some but not all investigators. Biochemistry, East Kent Hospitals NHS Trust, personal
communication).
Interpretation
PTH measurement is used in the differential diagnosis of References
hypercalcemia. 1. Ashby JP, Newman DJ, Gow SM. Clinical
application of intact parathyroid hormone
As second-generation PTH immunometric assays are assays. Ann Clin Biochem 1997;34:588-98.
routinely used, the question of whether CPTH fragments 2. Brewer HB, Jr., Fairwell T, Ronan R, Sizemore
(primarily PTH [7-84]) detected by these assays possess GW, Arnaud CD. Human parathyroid hormone:
biological activity needs to be addressed. It has been amino-acid sequence of the amino-terminal
proposed that PTH (7-84) is an antagonist to PTH (1-84) residues 1-34. Proc Natl Acad Sci USA
in inhibiting the turnover of bone and preventing an 1972;69:3585-8.
increase in blood calcium concentration [27]. In primary 3. Niall HD, Sauer RT, Jacobs JW, Keutmann HT,
hyperparathyroidism and vitamin D deficiency, PTH (7- Segre GV, O'Riordan JL, et al.. The amino-acid
84) accounts for less than 10% of intact PTH. However, sequence of the amino-terminal 37 residues of
accumulation of PTH (7-84) has been shown in patients human parathyroid hormone. Proc Natl Acad
with chronic kidney disease [28]. In this scenario, an Sci USA 1974;71:384-8.
increased PTH concentration measured by a second- 4. Murray TM, Rao LG, Divieti P, Bringhurst FR.
generation assay may be indicative of bone disease, but Parathyroid hormone secretion and action:
if a significant amount of PTH (7-84) is present evidence for discrete receptors for the carboxyl-
inhibiting PTH (1-84)s action, bone disease is less terminal region and related biological actions of
likely. It can therefore be argued that specifically carboxyl- terminal ligands. Endocr Rev
measuring PTH (1-84) is of merit. 2005;26:78-113.
5. Berson SA, Yalow RS, Aurbach GD.
Like first-generation assays, second-generation assays Immunoassay of bovine and human parathyroid
overestimate PTH (1-84) concentrations, and may be hormone. Proc Natl Acad Sci USA
providing a worse picture of bone turnover for patients 1963;49:613-7.
with renal disease than is the case [27]. The Nichols, 6. Goodman WG, Salusky IB, Juppner H. New
Diagnostic System Laboratories, and Instars second- lessons from old assays: parathyroid its
generation PTH assays have all been found to receptors, and the potential biological hormone
overestimate PTH (1-84) concentrations [16]. However, (PTH), relevance of PTH fragments. Nephrol
the clinical implications of third-generation assays are Dial Transplant 2002;17:1731-6.
controversial. A number of studies have concluded that 7. Carter AB, Howanitz PJ. Intraoperative testing
no additional diagnostic information is obtained from for parathyroid hormone - A comprehensive
third-generation assaysfor example, Scantibodies review of the use of the assay and the relevant
Whole PTH (1-84) Specific assay [29,30]. Perhaps of literature. Arch Pathol Lab Med
greater clinical relevance is determining both PTH (1- 2003;127:1424-42.
84) and (7-84) concentrations and their ratio in those 8. Segre GV. Advances in Techniques For
patients with known renal bone disease [18]. Measurement of Parathyroid-Hormone -
Current Applications in Clinical Medicine and
PTH Performance Goals Directions For Future-Research. Trends
Second-generation PTH assays do not agree with each Endocrinol Metab 1990;1:243-7.
other, and this is confirmed by external quality 9. Brossard JH, Whittom S, Lepage R. Carboxyl-
assessment data. Such differences can be attributed to terminal fragments of parathyroid hormone are
the lack of standardization, variation in recovery of PTH not secreted preferentially in primary
(1-84), and differences in the amount of cross-reactivity hyperparathyroidism as they are in other
with PTH (7-84). Data from UK NEQAS has estimated hypercalcemic conditions. J Clin Endocrinol
between 69% and 102% cross-reactivity with PTH (7- Metab 1993;77:413-9.
84) for all commercially available second-generation 10. Nguyen-Yamamoto L, Rousseau L, Brossard
assays [31]. The between-all method coefficient of JH, Lepage R, Gao P, Cantor T, et al. Origin of
variation (CV) was 34% in the March 2007 UK NEQAS parathyroid hormone (PTH) fragments detected
for plasma PTH. However, within-method CVs for
937
by intact-PTH assays. Eur J Endocrinol clinical characterization of the Bio-PTH (1-84)

2002;147:123-31. immunochemiluminometric assay and
11. Martin KJ, Hruska K, Freitag JJ. Clinical utility comparison with a second- generation assay for
of radioimmunoassays for parathyroid parathyroid hormone. Clin Chem 2004;50:385-
hormone. Min Electrolyte Metab 1980;3:283- 90.
90. 22. Zanelli JM, Gaines-Das RE. The first
12. Nussbaum SR, Zahradnik RJ, Lavigne JR, international reference preparation of human
Brennan GL, Nozawa-Ung K, Kim LY, et al. parathyroid hormone for immunoassay:
Highly sensitive two-site immunoradiometric characterization and calibration by international
assay of parathyrin, and its clinical utility in collaborative study. J Clin Endocrinol Metab
evaluating patients with hypercalcemia. Clin 1983;57:462-9.
Chem 1987;33:1364-7. 23. Teal TK, Reed M, Stevens PE, Lamb EJ.
13. Martin KJ, Akhtar I, Gonzalez EA. Parathyroid Stability of parathyroid hormone ex vivo in
hormone: New assays, new receptors. Semin haemodialysis patients. Ann Clin Biochem
Nephrol 2004;24:3-9. 2003;40:191-3.
14. Brossard JH, Lepage R, Cardinal H, Roy L, 24. Mayne PD, Kovar IZ. Calcium and phosphorus
Rousseau L, Dorais C, et al.. Influence of metabolism in the premature infant. Ann Clin
glomerular filtration rate on non-(1-84) Biochem 1991;28 :131-42.
parathyroid hormone (PTH) detected by intact 25. Cioffi M, Corradino M, Gazzerro P, Vietri MT,
PTH assays. Clin Chem 2000;46:697-703. Di Macchia C, Contursi A, et al.. Serum
15. D'Amour P, Brossard JH, Rousseau L, Nguyen- concentrations of intact parathyroid hormone in
Yamamoto L, Nassif E, Lazure C, et al. healthy children. Clin Chem 2000;46:863-4.
Structure of non-(1-84) PTH fragments secreted 26. Kotowicz MA, Melton LJ, 3rd, Cedel SL,
by parathyroid glands in primary and secondary O'Fallon WM, Riggs BL. Effect of age on
hyperparathyroidism. Kidney Int 2005;68:998- variables relating to calcium and phosphorus
1007. metabolism in women. J Bone Miner Res
16. Lepage R, Roy L, Brossard JH, Rousseau L, 1990;5:345-52.
Dorais C, Lazure C, et al. A non-(I-84) 27. Malluche HH, Mawad H, Trueba D, Monier-
circulating parathyroid hormone (PTH) Faugere MC. Parathyroid hormone assays -
fragment interferes significantly with intact evolution and revolutions in the care of dialysis
PTH commercial assay measurements in uremic patients. Clin Nephrol 2003;59:313-8.
samples. Clin Chem 1998;44:805-9. 28. Brossard JH, Cloutier M, Roy L, Lepage R,
17. Gao P, Scheibel S, D'Amour P, John MR, Rao Gascon-Barre M, D'Amour P. Accumulation of
SD, Schmidt-Gayk H, et al. Development of a a non-(1-84) molecular form of parathyroid
novel immunoradiometric assay exclusively for hormone (PTH) detected by intact PTH assay in
biologically active whole parathyroid hormone renal failure: importance in the interpretation of
1-84: Implications for improvement of accurate PTH values. J Clin Endocrinol Metab
assessment of parathyroid function. J Bone 1996;81:3923-9.
Miner Res 2001;16:605-14. 29. Godber IM, Parker CR, Lawson N, Hitch T,
18. Blumsohn A, Al Hadari A. Parathyroid Porter CJ, Roe SD, et al. Comparison of intact
hormone: what are we measuring and does it and whole molecule parathyroid hormone
matter? Ann Clin Biochem 2002;39:169-72. assays in patients with histologically confirmed
19. Silverberg SJ, Gao P, Brown I, Logerfo P, post-renal transplant osteodystrophy. Ann Clin
Cantor TL, Bilezikian JP. Clinical utility of an Biochem 2002;39:314-7.
immunoradiometric assay for parathyroid 30. Goodman WG. New assays for parathyroid
hormone (1-84) in primary hormone (PTH) and the relevance of PTH
hyperparathyroidism. J Clin Endocrinol Metab fragments in renal failure. Kidney Int
2003;88:4725-30. 2003;64:S120-S4.
20. Yamashita H, Gao P, Noguchi S, Uchino S, 31. Lamb EJ, Vickery S, Ellis AR. Parathyroid
Watanabe S, Ogawa T, et al.. Comparison of hormone, kidney disease, evidence and
intact and whole parathyroid hormone levels guidelines. Ann Clin Biochem 2007;44:1-4.
after parathyroidectomy between patients with 32. Ankrah-Tetteh T, Wijeratne S, Swaminathan R.
primary and secondary hyperparathyroidism. J Intra individual variation in serum thyroid
Bone Miner Res 2003;18:S100-S. hormones, parathyroid hormone, and insulin
21. Inaba M, Nakatsuka K, Imanishi Y, Watanabe like growth factor-1 (IGF-1). Ann Clin Biochem
M, Mamiya Y, Ishimura E, et al.. Technical and 2007;in press.
938
Table 1
Method 1: First Generation PTH Radioimmunoassay
Principles of analysis:
PTH (both intact PTH (1-84) and fragments) competes with 125I-labelled fragments for binding to
antibodies to specific fragments or the intact PTH molecule. Antibodies used will bind either to the C-
terminal, mid-molecule or to the N-terminal region of PTH. Separation of bound and free ligand is
achieved by use of dextran-coated charcoal or a second antibody.
Method 2: Second Generation Intact PTH Immunometric Assays

A signal (e.g. chemiluminescent) polyclonal antibody targeted to the N-terminal region of PTH and a
second (e.g. biotinylated) polyclonal antibody targeted to the C-terminal region of PTH detects PTH (1-
84) and some truncated PTH fragments. The second polyclonal or capture antibody is typically coated
to latex particles or bound to the wells of microtiter plates. Unbound protein is removed by a series of
washing steps. The concentration of antibody-bound PTH is directly proportional to the signal
produced (e.g. relative light units (RLUs)).
Method 3: Third Generation Bioactive/Whole PTH Immunometric assay

A signal (e.g. chemiluminescent) polyclonal antibody targeted to amino acids 1-4 of the N-terminal of
PTH and a second (e.g. biotinylated) polyclonal antibody to the C-terminal of PTH ensures that only
intact PTH (1-84) is detected. The second polyclonal or capture antibody is typically coated to latex
particles or bound to the wells of microtiter plates. Unbound protein is removed by a series of washing
steps. The concentration of antibody-bound PTH is directly proportional to the signal produced (e.g.
RLUs).
939
Phenylalanine
Phenylalanine
Theodore Dashman, Helen K. Berry
Name: Phenylalanine, -amino--phenylpropionic acid

Clinical significance:
Merck Index: 7150
Chemical class: amino acid
Structure:

i tyrosine uses the N(O)-trifluoroacetyl methyl esters
The determination of phenylalanine in serum is most prepared by reaction of the amino acid mixture with
frequently used to confirm the diagnosis of acetonitrile and trifluoroacetic anhydride, followed by
hyperphenylalaninaemias, that is, phenyl-ketonuria and methylation using diazomethane [5]. One can use a
disorders of tetrahydrobiopterin metabolism, and to variety of column packings. Conditions of gas flow and
follow changes in phenylalanine blood levels in persons column and oven temperatures are adjusted for the
undergoing therapy for phenylketonuria. One method for particular derivative. Detection is by flame ionization or
measurement of phenylalanine is automatic ion- mass spectrometry with selective ion monitoring.
exchange chromatography using an amino acid analyzer
(Table 1, method 1). The single-column method of Piez Reversed-phase high performance liquid
and Morris forms the basis for operation of most amino chromatography has been developed for rapid analysis of
acid analyzers [1]. The ion-exchange resins are highly plasma amino acids (Table 1, method 3). The most
cross-linked sulfonated styrene copoly-mers that have an common practice is to derivatize the amino acids with
affinity for both the ionic and nonionic portion of the one of several reagents, such as, dabsy lchloride, phenyl-
amino acid molecule [2]. Amino acids are eluted from isothiocyanate, or 9-fluorenylmethyl chloro-formate, but
the resin when the column temperature is increased and the most popular compound is o-phthala-adehyde [6].
buffers of different pH and ionic strength are used. The The advantages of these methods are high precision,
column effluent is reacted with ninhydrin and the optimal identification of primary and secondary amino
produced blue reaction product is quantified acids, and economic mobile phase [7].
spectrophotometrically at 570 nm.
The Guthrie microbiological inhibition assay is used for
Gas-chromatographic procedures are used for separation semiquantitative measurement of phenylalanine in
of amino acids, including phenylalanine, in blood, particularly in newborn screening programs [8].
physiological mixtures (Table 1, method 2). A number The test is based on the inhibition of growth of Bacillus
of derivatizing agents have been used for the gas subtilis by -2-thienylalanine and the ability of
chromatography of amino acids. Trimethylsilyl phenylalanine to overcome this inhibition (Table 1,
derivatives are frequently used [3], as are derivatives method 4).
prepared with N,N-bis(trimethyl-silyl)trifluoroacetamide
(BSTFA) [4]. A method specific for phenylalanine and The fluorometric method of McCaman and Robins [9-
13] is preferred for routine use in the clinical laboratory
from the standpoint of simplicity, sensitivity, and cost.
i
Phenylalanine The procedure is based on the enhanced fluorescence
Previous and current authors of this method: produced when phenylalanine condenses with ninhydrin
First edition: Helen K. Berry in the presence of the dipeptide L-leucyl-L-alanine
Methods edition: Helen K. Berry (Table 1, method 5). The reaction is carried out at pH
Second edition: Not updated 5.9. The fluorescent compound is stabilized by addition
Third edition: Not updated of an alkaline copper tartrate reagent. The fluorescence
Fourth edition: Theodore Dashman, Helen K. Berry produced by excitation at 365 to 390 nm is measured at
Fifth edition: Not updated 489 to 515 nm.
940
Phenylalanine
Techniques for quantitating both phenylalanine and fluorometric method with a blank correction is rapid and
tyrosine by tandem mass spectrometry (MS/MS) have accurate and thus is the recommended procedure.
also been described [14]. Studies have shown that
calculation of the molar ratio of phenylalanine to Use of MS/MS can diagnose phenylketonuria with much
tyrosine can allow detection of affected newborns under lower false positive rates (approximately 1/100)
24 h of age without increasing the rate of false positive compared with fluorometry. Also, in addition to other
test results. In addition to phenylketonuria, MS/MS can aminoacidopathies, MS/MS can also screen for organic
also detect other aminoacidopathies, including maple acid disorders and disorders of fat metabolism. Because
syrup urine disease and homocystinuria. The of the low false-positive rates, this method is an efficient
disadvantages to MS/MS include the initial high cost of way to screen for large numbers of disorders while
equipment and the need for sample derivatization. One minimizing the cost of follow-up.
way around this is employ stable isotope dilution
reversed-phase LS-MS [15]. Specimen
Either serum or plasma may be used. The nature of the
Reference and Preferred Methods anticoagulant does not affect the measurement. The
The amino acid analyzer yields accurate and precise specimens are stable at 4 to 8 C for about a week.
measurements of phenylalanine. From derivatization to Blood dried on filter paper is used for regional screening
producing a report takes 2 to 3 h and is expensive in programs and is stable for mailing and handling at room
terms of both cost of equipment and cost of reagents. temperature. Ideally, specimens should not be collected
until 48 h following birth. Approximately 2 days worth
Gas chromatography of amino acids from physiological of protein intake is usually required to result in the
fluids requires a preliminary separation of amino acids accumulation in an abnormal concentration of amino
from other substances that may react with the acids in affected infants.
derivatizing agents. Although gas chromatography has
excellent sensitivity and a short analysis time and the Interferences
apparatus is less expensive than an automatic amino acid The most important limitation of the ion-exchange
analyzer, the problems involved in preparation of the chromatography method is the presence in the sample of
specimen and preparation of the derivative make it compounds with primary amino acid moieties. These
undesirable for routine work. compounds react with ninhydrin and may coelute with a
plasma amino acid. These compounds include a variety
Although the fluorometric method yields less accurate of drugs, such as ampicillin. For specimens analyzed
results than the amino acid analyzer, it has the advantage using the Guthrie procedure, antibiotics or other
of being able to analyze large numbers of samples substances that inhibit bacterial growth must not be
rapidly with a fair degree of precision. However, present. Otherwise, false negative results may occur.
reaction of amino acids other than phenylalanine may Premature infants may show false positive results
yield a nonspecific fluorescence that results in because immature hepatic function may cause increased
overestimation of phenylalanine concentrations in amino acid concentrations in serum.
specimens from normal persons.
Phenylalanine Reference Interval
In a cooperative study, values obtained by several Serum phenylalanine values measured in several studies
laboratories using the fluorometric procedure were of normal persons were, in milligrams per liter: 15 + 3
compared with values measured in the same specimens [13], 15.5 + 3.4 [17] and 21 + 5 [18]. For comparison,
by the amino acid analyzer [16]. At the lowest mean phenylalanine concentration in 330 specimens
phenylalanine level tested, 16 mg/L, the mean values from infants and children measured by the amino acid
from the fluorometric procedure were significantly analyzer was 12 mg/L, with values ranging from 8 to 16
higher than control values measured on the analyzer. At mg/L [19]. One recent study that compared the
the phenylalanine concentration of 36 mg/L, higher fluorometric assay with MS/MS found a cutoff of 4.3
values were obtained by the fluorometric procedure, mg/dL for the flouorometric procedure compared with
except in instances in which a serum blank (ninhydrin 3.0 mg/dL for MS/MS [14]. The difference in cutoff
without peptide) was used to correct for nonspecific values between fluorometry and MS/MS is probably due
fluorescence. At concentrations of phenylalanine above to fluorescent interference which raises the concentration
50 mg/L, none of the mean values from fluorometric of phenylalanine.
procedures was significantly different from the values
measured by the amino acid analyzer. Dilution of the
specimen when one is measuring phenylalanine in
patients with phenylketonuria reduces the error from
nonspecific fluores-cence. However, when
phenylalanine concentrations are very low in a
nonphenylketonuric patient, the fluorometric procedure
may cause overestimation of the serum phenylalanine
content, and a phenylalanine deficiency may be falsely
undiagnosed. For general screening purposes, the
941
Phenylalanine
Phenylalanine Concentration in Some Genetic Defects

Blood phenylalanine
Defect concentration (mg/L) mmol/L
Classical phenylketonuria (phenylalanine hydroxylase deficiency)
Untreated 300 1.81
Treated, good control 3080 0.1820.484
Variant phenylketonuria (partial phenylalanine
hydroxylase deficiency) untreated 40200 0.2421.21
Hyperphenylalaninemia 40300 0.2421.816
Dihydropteridine reductase deficiency[20] 160530 0.969-3.208
Biopterin synthetase deficiency[20] 210490 1.2712.966
Normal 9 0.055
942
Phenylalanine
Interpretation 13 McCaman MW, Robins E: Fluorimetric method

Phenylalanine concentrations greater than 40 mg/L are for the determination of phenylalanine in serum.
rarely seen except in patients who have a genetic defect J Lab Clin Med 1962;59:885-90
in phenylalanine metabolism. Ranges reported in such 14 Chace DH, Sherwin JE, Hillman SL, Lorey F,
patients are shown in the preceding [20,21]. In a few Cunningham GC. Use of phenylalanine-to-
rare disorders, galactosemia, hereditary fructose tyrosine ratio deter-mined by tandem mass
intolerance, hereditary tyrosinemia, and transient spectrometry to improve newborn screening for
tyrosinemia of the newborn, abnormal tyrosine phenylketonuria of early discharge specimens
metabolism may cause a secondary elevation of collected in the first 24 hours. Clin Chem
phenylalanine [21]. 1998;44:2405-9
15 Tuchman M, McCann MT. Phenyl-alanine and
References tyrosine quantification by stable isotope dilution
1 Piez KA, Morris L. An automatic procedure for liquid chromato-graphy-mass spectrometry
the automatic analysis of amino acids. Anal from filter paper blood spots. Clin Chem 1999;
Biochem 1960; 1:187-201 45:571-3
2 Benson JV, Patterson JA. Accelerated 16 Ambrose JA. Report on a cooperative study of
chromatographic analysis of amino acids various fluorometric procedures and the Guthrie
commonly found in physiological fluids on a bacterial inhibition assay in the determination
spherical resin of specific design, Anal of hyperphenylalaninemia, Centers for Disease
Biochem 1965;13:265-80 Control, Health Services and Mental Health
3 Blau K. Biomedical applications of gas Administration, Public Health Service, Atlanta,
chromatography 1968 vol. 2, New York Plenum 1972, U.S. Government Printing Office.
Press. 17 Wong PWK, OFlynn ME, Inouye T.
4 Gehrke CW, Leimer K. Trimethylsilylation of Micromethods for measuring phenylalanine and
amino acids derivatization and chromatography. tyrosine in serum. Clin Chem 1964; 10:1098-
J Chromatogr 1971;57:219-38 104
5 Zagalak MJ, Curtius HC, Leimbacher W, 18 Hsia DY, Berman JL, Slatis HM. Screening
Redweik U. Quantitation of deuterated and non- newborn infants for phenyl-ketonuria. JAMA
deuterated phenylalanine and tyrosine in human 1964; 188:203-6
plasma using the selective ion monitoring 19 Berry HK, Porter LJ. Newborn screening for
method with combined gas chromatography phenylketonuria. Pediatrics 1982;70:505-6
mass spectrometry: application to the in vivo
measurement of phenylalanine-4- 20 Danks DM, Bartholome K, Clayton BE, Curtius
monooxygenase activity. J Chromatogr 1977; H, Grbe H, Kaufman S, Leeming R, et al.
142:523-31 Malignant hyper-phenylalaninemiacurrent
6 Teerlink T, van Leeuwen PAM, Houlijk A, status (June 1977). J Inherited Metab Dis 1978;
Guthrie R, Susi A. Plasma amino acid 1:49-53
determined by liquid chromatography within 17 21 Berry HK. The diagnosis of phenylketonuria.
minutes. Clin Chem 1994; 40:245-9 Am J Dis Child 1981; 135:211-3
7 Worthen HG, Lui H. Automatic pre-column
derivitization and reversed phase high
performance liquid chromatography of primary
and secondary amino acids in plasma with
photo-diode array and fluorescence detection. J.
Liq Chromatogro 1992; 15:3323-41
8 Guthrie R, Susi A. A simple phenyl-alanine
method for detecting phenyl-ketonuria in large
populations of newborn infants. Pediatrics 1963
;32:338-43
9 Bambach K, Burkey R. Micro-determination of
lead by dithizone. Indust Eng Chem Anal Ed
1942 ;14:904-7
10 Cholak J, Hubbard D, Burkey, R.
Microdetermination of lead in biological
material with dithizone at high pH. Anal Chem
1948; 20:671-2
11 Cholak J. Analytical methods for deter-
mination of lead. Arch Environ Health
1964;8:222-31
12 Kopito L, Schwachman H: Measurement of
lead in blood, urine and scalp hair by atomic
absorption spectrometry. Stand Methods Clin
Chem. 1972;7:151-62
943
Phenylalanine
Tables Method 3: High performance liquid chromatography

Principle of analysis: Precolumn
Table 1: Phenylalanine Methods Summary derivitization of amino acids, separation by
Method 1: Automatic column chromatography reversed-phase chromatography; fluorescence
Principle of analysis: Cation-exchange detection
chromatography with detection of amino acids Reaction: Formation of an amino acid O-
by ninhydrin phthala-dehyde derivative in the presence of a
Reaction: Triketohydrindene hydrate sulhydryl reagent
(ninhydrin) + RCH(NH2)CO2 yields Usage: Plasma
Comments: Derivatization may take up to 1 h
diketohydrindylidenediketohydrindamine
and chromatography up to 2 h
(Reuhlmans purple) Method 4: Guthrie microbiological assay
Comments: Serum, plasma, urine, or other Principle of analysis: Inhibition of growth of
physiological fluids; most accurate and Bacillus subtilis by -2-thienylalanine;
preferred inhibition overcome by phenylalanine
Method 2: Gas chromatography Comments: Dried blood spots collected on
Principle of analysis: Separation of filter paper; routinely used for newborn infant
derivatized amino acids on glass columns; screening; semiquantitative
packing selected depends on derivative; Method 5: Fluorometric
detection by flame ionization Principle of analysis: Conversion of
Reaction: Formation of N(O)- phenylalanine to fluorescent derivative
trimethylsilylamino acid esters or of amino acid Reaction: Mechanism of reaction of
alkyl ester followed by acylation; alkyl groups phenylalanine with ninhydrin to yield
methyl through pentyl have been used fluorescent compound is not known
Comments: Serum or plasma; requires Comments: Serum, plasma, blood eluted from
removal of interfering substances; not widely filter paper; recommended for routine use; has
used for routine analysis; most valuable when been automated
using stable iosotopes
Reaction Conditions for Fluorometric Analysis of Phenylalanine

Condition Requirements
Temperature 80 C
pH 5.9
Final concentration of reagents Succinate buffer: 352 mmol/L
(before dilution with copper reagent)* Ninhydrin: 7.0 mmol/L
L-Leucyl-L-alanine: 0.59 mmol/L
Fraction of sample volume (protein- 0.0625
free supernatant)
Sample volume 100 L
Sample Protein-free supernatant of serum/plasma or protein-
free supernatant of diluted serum/plasma
methanol eluate of dried blood spot may be used
Time of reaction 25 min
Wavelengths Excitation: 365390 nm
Emission: 489515 nm
Linearity With most fluorometers the reaction is linear between
5 and 80 mg/L; the range can be extended up to 800
mg/L using the dilutions specified
Precision: mean (& CV) mg/L mg/L mg/L
16 (11.6%) 19 (12.8%)
55 (11.8%) 35 (7.1%)
110 (6.3%) 140 (10.5%) 167 (4.7%)
*Volume of copper reagent can vary from 3.0 to 4.0 mL.
**From Berry, H.K., and Porter, L.J.: Newborn screening for phenylketonuria, Pediatrics 70:505-506, 1982.
From Sigma Chemical Co., St. Louis, Mo., Technical Bulletin No 60F.
From:H.K. Berry's laboratory, Children's Hospital; Cincinnati,OH.
944
Phenylalanine
Example of Phenylalanine Concentration Calculation*

Sample g per assay mL of serum per = g per mL
size (mL) (from curve) assay serum
0.010 0.133 0.000833 160
0.020 0.264 0.00167 158
0.040 0.551 0.00333 165
*Average = 161 g/mL 1000 mL/L = 161 mg/L

1000 g/mg
6. Trichloroacetic acid (TCA), 0.6 mol/L (98

Procedure: Fluorometric Determination of g/L). Prepare volumetrically; 0.3 mol/L, 1:1 dilution of
Phenylalanine 0.6 mol/L. Stable 6 months at refrigerated temperature.
(Reaction Conditions for Phenylalanine Analysis Table) 7. Phenylalanine standard solution
Stock solution, 200 mg/L (1.21 mmol/L). Dissolve 20
Principle mg of L-phenylalanine in water and dilute to a final
The chemical basis of the method is the specifically
volume of 100 mL. Store frozen. Stable for 1 to 2
enhanced fluorescence of phenylalanine after reaction
months.
with ninhydrin in the presence of a peptide, L-leucyl-L-
Working standard solution, 20 mg/L (0.121 mmol/L).
alanine. A reagent blank is included to correct for
Combine 1 mL of stock solution, 4 mL of water, and 5
nonspecific fluorescence. This procedure is available as
mL of TCA 0.6 mol/L to prepare phenylalanine solution
a kit from Sigma Chemical Co., St. Louis, MO.
containing 20 mg/L in TCA 0.3 mol/L. Store frozen.
Reagents Stable for 2 to 4 weeks.
All reagents are prepared in deionized water unless
Assay
otherwise stated.
Equipment: Heating bath set at 80 C (ethylene glycol
1. Succinate buffer, 0.6 mol/L, pH 5.9. Dissolve
produces more uniform temperature than water), a
16.2 g of Na2C4H4O46H2O (disodium succinate
centrifuge, and a fluorometer. One of the following is
hexahydrate) in 90 mL of water. Adjust pH with 1 M acceptable: Turner Model 110 or 111 Fluorometer with
HCl. Dilute to 100 mL. Store at 0 to 5 C. Stable for 2 General Purpose Lamp; filters: Primary 7-60 (Turner
to 4 months. 110-611); Secondary 2A (Turner 110-816) or 8 (Turner
2. Ninhydrin, 30 mmol/L (5.34 g/L). Dissolve 110-817) + 65A (Turner 110-825); Farrand Model A
534 mg of ninhydrin in approximately 80 mL of water Fluorometer, filters as above; Aminco Bowman
and bring to a final volume of 100 mL. Store in a brown Spectrophotofluorometer, activation 365 nm and
bottle at 0 to 5 C. Stable for 7 to 10 days. emission 515 nm.
3. L-Leucyl L-alanine, 5 mmol/L (1095 1. Preparation of protein-free supernatant
mg/L). Prepare volumetrically by dissolving 109.5 mg
For analysis of specimens from normal
in water and dilute to a final volume of 100 mL. Store
persons: Combine 1 volume (usually 0.100 mL)
frozen at 20 C in aliquots of 1 mL. Stable for 4 to 6
of serum or plasma with 1 volume of TCA 0.6
months.
mol/L.
4. Ninhydrinpeptide reagent. Prepare daily by
combining 5 volumes of succinate buffer reagent with 2 For analysis of specimens from untreated
volumes of ninhydrin reagent and 1 volume of peptide patients with phenylketonuria: Dilute 1 volume
reagent. of serum with 9 volumes of water; combine 1
5. Copper reagent volume of diluted serum with 1 volume of TCA
Stock solution. Na2CO3, 151 mmol/L; CuSO4, 2.40 0.6 mol/L.
For analysis of specimens from phenylketonuric
mmol/L; sodium potassium tartrate, 2.51 mmol/L.
patients under treatment: Dilute 1 volume of
1. 16.0 g of Na2CO3 (anhydrous sodium
serum with 2 or 5 volumes of water (depending
carbonate) on expected degree of phenylalaninemia);
2. 0.650 g of NaKC4H4O66H2O (sodium combine 1 volume of diluted serum with 1
potassium tartrate hexahydrate) volume of TCA 0.6 mol/L.
3. 0.600 g of CuSO45H2O (copper sulfate The serum-TCA solutions are mixed well and
pentahydrate) allowed to stand at room temperature for 10
Dissolve each in about 300 mL of water; mix together in minutes. The tubes are centrifuged at 3000 rpm
order listed; dilute to 1 L. Store at room temperature. (approximately 1000 g) for 10 minutes. The
Stable for 6 to 8 months. protein-free supernatant is used for analysis.
Working reagents. For daily use dilute 1 2. Standard curve. Pipet 0.010, 0.020, 0.030, and
volume of stock with 9 volumes of water. 0.040 mL of phenylalanine working standard,
945
Phenylalanine
corresponding to 0.2, 0.4, 0.6, and 0.8 g per Example

assay, into small test tubes. Adjust total volume Serum diluted 1:6 for preparation of protein-free
to 0.040 mL by addition of TCA 0.3 mol/L. supernatant:
3. Sample. Prepare three tubes for each specimen; 0.020 mL of serum + 0.100 mL of water +
use 0.010, 0.020, and 0.040 mL of protein-free 0.120 mL of TCA (0.6 mol/L) =
supernatant; bring to final volume to 0.04 mL 0.240 mL total volume
with TCA 0.3 mol/L.
4. Prepare a reagent blank by pipetting 0.040 mL Calculated volume of serum per assay:
of TCA 0.3 mol/L (in duplicate). __0.020 mL of serum__ 0.010 mL (sample
size) = 0.000833 mL
5. Add 0.60 mL of ninhydrinpeptide mixture to
0.240 mL (total volume)
each standard, sample, and blank tube. Mix
thoroughly.
One calculates the concentration of phenylalanine by
6. Place in heating bath at 80 C for exactly 25 dividing the micrograms of phenylalanine in each tube
min; remove; cool in tap water for 5 minutes. (from the standard curve) by the volume of serum per
7. Add 4 mL of dilute copper reagent, mix, and assay as shown in Phenylalanine Table: Example of
allow to stand for 10 min. Read in fluorometer. phenylalanine concentration calculation.
Calculation
The fluorescence of each standard minus the Phenylalanine Reference Interval
fluorescence of the reagent blank is plotted versus the
respective phenylalanine concentration. The
concentration of phenylalanine in the unknown specimen
is obtained from interpolation of the standard curve.
Average the results of three determinations for each
unknown. Dilute if the fluorescence is greater than the
highest value on the standard curve.
946
Phenytoin
Phenytoin
Gus Koerbin
Name: Phenytoin
Clinical significance: Refer to Chapter 47, Nervous System, in the 5th edition of Clinical Chemistry:
Merck Index: 7130
Chemical class: Hydantoin
Structure:
A second method for measuring phenytoin involved

i
Principles of Analysis and Current Usage extraction and separation by thin-layer chromatography
[3,4]. Quantitation was achieved by either ultraviolet
Phenytoin is an anticonvulsant drug which can be useful scanning of the plate or by elution of the drug from the
in the treatment of epilepsy. The primary site of action
appears to be the motor cortex, where spread of seizure plate and recording of its ultraviolet absorption (Table 1,
activity is inhibited by possibly promoting sodium efflux Method 2).
from neurons. Phenytoin tends to stabilize the threshold
against hyperexcitability caused by excessive stimulation Gas chromatographic techniques (GC) were for many
or environmental changes capable of reducing years the primary technique used to analyze phenytoin
membrane sodium gradient. Phenytoin reduces the and other antiepileptic drugs in biological fluids (Table
maximal activity of brainstem centers responsible for the 1, Method 3). In a review by Rambec and Meijer [5],
tonic phase of tonic-clonic (grand mal) seizures. flame ionization detection (FID) or the more sensitive
nitrogen-selective detection (NSD) systems were the
Most anticonvulsant drugs have specific ultraviolet most widely used. NSD may have considerable
spectral characteristics (Figure 1) that were exploited in advantage over FID in terms of specificity and
the initial attempts to monitor the therapeutic levels of sensitivity, since almost all antiepileptics except valproic
these drugs [1,2]. Phenytoin was extracted into an acid contain nitrogen. However, NSD requires more
organic solvent and reextracted into aqueous solution, frequent and careful maintenance. Approximately half of
and the absorbance of the solution was recorded (Table the gas chromatographic methods described by Rambec
1, Method 1). By use of the appropriate solvents, and Meijer do not prepare derivatives of the
phenytoin could be extracted with relative specificity. anticonvulsant drugs [6-9]. However, of those
By recording the difference spectra of the extract at two procedures that use derivatization, methylation is most
pH values, one could also increase the specificity. The commonly used [5]. In the past, column support
difference spectra were obtained by use of a dual-beam materials most frequently used were acid-washed
spectrophotometer and placement of one solution into silanized materials (e.g., Gas Chrom-Q, Chromosorb-W,
the reference compartment and the other into the usual and Supelcoport). Nowadays fused-silica capillary
test compartment. columns are preferred because of their high temperature
stability and inertness, which not only removes the need
for derivatization but offers high sensitivity, selectivity,
and speed [10]. In the method described by Volmut et al.
[11], six antiepileptic drugs may be analyzed
i
Phenytoin simultaneously by capillary gas chromatography.
Previous and current authors of this method: High-performance liquid chromatography assays
First edition: Steven J. Soldin (HPLC) require some form of sample preparation before
Methods edition: Steven J. Soldin separation on a column (Table 1, Method 4). Sample
Second edition: Steven J. Soldin preparation usually involves either organic solvent
Third edition: Paul Salm, Paul J. Taylor, Julia M. extraction of the drugs or protein precipitation. Nearly
Potter all methods use reverse-phase chromatography with
Fourth edition: Not updated ultraviolet absorption spectrometry for detection [12-20]
Fifth edition: Gus Koerbin
947
Phenytoin
Immunological procedures (Table 1, Methods 5a-5j) antibody, and a patient sample containing the drug, as
developed for the measurement of phenytoin include follows:
radioimmunoassay (RIA), enzyme immunoassay (EIA),
fluorometric enzyme immunoassay (FEIA), fluorescence 1. FDR + antibody + anticonvulsant
immunoassay (FIA), fluorescence polarization 2. AntibodyFDR + antibodyanticonvulsant + FDR
immunoassay (FPIA), nephelometric inhibition + anticonvulsant
immunoassay (NIA), cloned-enzyme donor 3. FDR -galactosidase umbelliferone + galactose
immunoassay (CEDIA), particle-enhanced turbidimetric
immunoassay (PETINIA) and dry-film multilayer
immunoassay (DFMI). The drug in the sample competes with the FDR for the
limited number of antibody-binding sites. The FDR not
The RIA procedures (Table 1, Method 5a) involve the bound to antibody is hydrolyzed by -galactosidase to
standard competitive-binding procedure in which there is produce the fluorescent product. Therefore, the
competition between the labeled drug with that from the
fluorescence produced is proportional to the
sample for a limited amount of antibody [21]. concentration of the drug in the sample.
The introduction of the enzyme-multiplied immunoassay
technique (EMIT) into the clinical chemistry laboratory
Fluorescence polarization immunoassay (FPIA) uses the
in the 1970s made possible the routine monitoring of not principle of competitive binding assay and measures
only phenytoin but all the routinely prescribed binding of tracer directly by fluorescence polarization
anticonvulsant drugs (Table 1, Method 5b). This
(Table 1, Method 5e). The fluorescent tracer consists of
homogeneous enzyme immunoassay is based on the fluorescein covalently attached to the drug being
competitive protein-binding technique, with an enzyme analyzed. When excited with polarized light, this tracer
as the label and an antibody as the binding protein.
will emit light with a degree of polarization that is
When the enzyme-labeled drug becomes bound to an inversely proportional to its rate of rotation; the more
antibody to the drug, the activity of the enzyme is slowly the tracer molecule rotates, the greater the
reduced. Free drug in the sample competes with the
polarization signal of the emitted light. When a specific
enzyme-labeled drug for binding to the antibody, antibody binds to the tracer, its rotation is slowed, and
decreasing the amount of antibody-bound labeled drug, the emitted light increases in polarization. It is a
and thereby decreases the antibody-induced inactivation
competitive-binding assay in which a specific
of the enzyme. The enzyme activity then correlates with fluorescein-labeled drug competes with endogenous drug
the concentration of the free drug. from a patient sample for a limited amount of antibody.
The amount of bound, labeled drug is inversely
Fluorometric enzyme immunoassay (FEIA) is a proportional to the drug in the patient sample. The
competitive-binding method that is based on the radial
intensity of the polarized fluorescence is proportional to
partition immunoassay principle that combines solid-
the amount of antibody-bound, labeled anticonvulsant.
phase immunological techniques and radial The greater the polarized fluorescence, the smaller the
chromatography (Table 1, Method 5c) [22]. In FEIA,
amount of drug in the patient sample.
phenytoin in the sample is premixed with enzyme-
labeled conjugate and spotted onto a glass-fiber filter
paper which contains immobilized antibody at the Another method for the measurement of phenytoin
analysis site. These two constituents then compete for employs a labeled prosthetic group. In this assay, a
antibody binding sites. After a suitable incubation prosthetic group necessary for the activity of the enzyme
period, unbound label is removed from the center of the is covalently linked to the drug to be measured: for
reaction zone through radial elution by applying a wash example, phenytoin linked to flavine adenine
solution. This solution contains the substrate for the dinucleotide (FAD) (Table 1, Method 5f). The FAD is
enzyme that initiates enzymatic activity simultaneously necessary for the enzyme apoglucose oxidase to function
with the wash. The enzymatic rate of the bound fraction [25]. If a limited amount of antibody to the drug is
at the center of the reaction zone is quantified by front- present, the drug will bind to the antibody. In the
surface fluorescence and is inversely proportional to the presence of drug in the patient sample, less FAD-labeled
concentration of drug present in the sample [23]. drug will bind to the antibody, making more FAD
available to bind to the apoprotein. The enzyme is then
Competitive protein binding has also been used to activated, and the amount of activity is proportional to
measure phenytoin in serum or plasma [24] (Table 1, the amount of drug in the patient sample. The enzyme
Method 5d). With this method, the drug is labeled with a converts glucose and O2 to gluconolactone and H2O2.
derivative of the fluorogenic enzyme substrate The color reaction is generated by use of peroxidase and
umbelliferyl--D-galactoside. This fluorogenic drug a dye substrate, 3,3, 5,5-tetramethylbenzidine, which
reagent (FDR) remains nonfluorescent under the when oxidized, forms a blue-colored product. In this
conditions of the assay until hydrolysis catalyzed by - assay, the patient serum or plasma sample is first diluted
galactosidase yields the fluorescent product. When with distilled water, and a specific volume of the
antibody to the anticonvulsant drug reacts with the FDR, prepared specimen is placed on the reagent area of the
the FDR is prevented from acting as a substrate for - strip. The strip is then placed in the instrument, a solid-
galactoside. Competitive binding reactions are set up phase reflectance photometer. The strip color is at 740
with a constant amount of FDR, a limiting amount of
948
Phenytoin
nm, and the results are calculated from a calibration hapten drug), as well as acting as a filter to prevent
curve. protein interference with the immunoreaction. The
sample travels from the topcoat layer through an iron-
A rate nephelometric inhibition immunoassay is also oxide screen to the signal layer. The iron oxide serves to
commonly used (Table 1, Method 5g). In this assay, block excitation of the label outside the signal layer. The
phenytoin is covalently linked to a protein. Drug from signal layer contains a monoclonal antibody specific for
the patients sample competes with this protein-bound the drug being assayed, conjugated with a fluorescent-
drug for binding to an antibody. The rate of change of labeled hapten in an agarose matrix coated onto a clear,
light scatter is measured by the instrument [26]. polyester-film base. The sample antigen displaces the
labeled antigen from the binding sites of the antibodies.
A homogeneous immunoassay technique, cloned- The unbound, labeled antigen then diffuses into the
enzyme donor immunoassay (CEDIA), is a unique upper layers where it cannot be measured. Equilibrium
method that utilizes genetically engineered - occurs within 3 to 5 min, after which the fluorescent
galactosidase enzyme fragments (Table 1, Method 5h). signal from the remaining bound conjugate is measured
The CEDIA principle is based on the use of two inactive from the bottom of the test module by front-surface
-galactosidase fragments which have been constructed fluorimetry. The fluorescent intensity measured is
using recombinant DNA techniques. One of these is inversely proportional to the antigen concentration in the
designated the enzyme donor (ED), a small amino- serum sample.
terminal peptide, and the other is the enzyme acceptor
(EA), a larger protein that has residues deleted near the
There is no reference method for phenytoin. The ideal
amino terminus [27]. The EA and ED each on their own
analytical procedure for anticonvulsant drugs including
show no enzymatic activity; however, active -
phenytoin should be as follows:
galactosidase forms when the fragments recombine
1. Both accurate and precise, affording the reliable
spontaneously. The formation of this active enzyme can
quantitation of the antiepileptic drugs.
be measured spectrophotometrically by the hydrolysis of
2. Technically easy, so that little effort is required in
a chromogenic substrate such as o-nitrophenyl--D-
training to perform the task.
galactopyranoside. A ligand molecule such as a drug or
3. Rapid, to minimize delay in reporting results to
hormone can be covalently linked to the ED peptide,
clinician. Many requests for drug quantitation
forming an ED-ligand conjugate, such that the enzyme
arise as a result of problems in patient
complementation reaction is not affected. Adding a
management.
ligand-specific antibody to the system will block or
4. Good sensitivity and specificity
inhibit spontaneous assembly of the enzyme by binding
Unfortunately, no method meets all these requirements,
to the ED-ligand conjugate and therefore decrease the
and the decision on which procedure to adopt depends
amount of enzyme formed. Active enzyme formation is
largely on the laboratory conditions and on personal
influenced by the concentration of antibody available for
preferences of those involved.
binding to the ED-ligand conjugate. In turn, the
concentration of available antibody is dependent on the
concentration of the analyte in the sample that can Spectrophotometric methods were initially used for the
competitively bind to the antibody [28]. The amount of measurement of phenytoin concentrations in body fluids
enzyme created and the rate of substrate hydrolysis is [1,2]. Although differential extraction procedures were
directly proportional to the sample analyte concentration. developed to overcome interference by other substances,
particularly drugs, the specificity of these and other
Another homogeneous immunoassay is the particle- wet-chemistry procedures is questionable, and the
enhanced turbidimetric immunoassay (PETINIA), where possibility of drug interferences still exists. For these
small latex particles are coated with antibody, antigen, or reasons, alternative modes of analysis were developed,
hapten (Table 1, Method 5i). This phenytoin assay is a employing either chromatographic or immunological
non-isotopic competitive-binding immunoassay based on procedures.
the inhibition of agglutination of latex particles coated
with covalently-bound phenytoin by a monoclonal Thin-layer chromatography procedures have been
antibody against phenytoin. The degree of inhibition of described that semiquantitate or quantitate many
the latex phenytoin particle agglutination is proportional anticonvulsant drugs [3,4]. This procedure, because of
to the amount of drug in the patients sample. The rate of both its lack of specificity and semiquantitative nature, is
agglutination is monitored at 340 nm by measuring the not a very popular mode of analysis.
change in turbidity after the particle-bound phenytoin,
antibody, and patients sample are mixed. Gas chromatography (GC) affords reliable analysis for
phenytoin analysis on microscale samples. The
Another procedure is based on a combination of procedure is easily automated and also allows
multilayer film technology and competitive simultaneous determination of other anticonvulsive
immunoassay (Table 1, Method 5j). [29]. A coated drugs. This is an obvious advantage, since many patients
multilayer film chip is encased in a plastic test module. with epilepsy are on multiple drug regimens. Although
The serum sample is applied to the topcoat layer in the steadily losing ground to both high-performance liquid
test module. This layer contains buffer components, chromatography (HPLC) and some of the competitive
surfactants, and other reagents (i.e., antibody and labeled binding assays, GC is useful in laboratories with a large
949
Phenytoin
workload. Disadvantages of GC include the used. The saliva concentration provides an estimate of
requirements of specialized equipment and considerable the free drug concentration for phenytoin.
expertise.
Interferences
The adoption of HPLC for therapeutic drug monitoring Icteric or hemolyzed specimens have less effect on the
has been one of the most significant advances in the chromatographic methods than on immunoassays.
clinical laboratory. The early spectrophotometric
methods, thin-layer chromatography, and gas Therapeutic Reference Interval
chromatography have been largely replaced by HPLC. The therapeutic reference interval for phenytoin is 10 to
20 mg/L (40 to 80 mol/L).
Some advantages of HPLC over earlier
techniques: Interpretation
1. Analyte volatility and thermal stability, so Therapeutic drug monitoring of anticonvulsants is the
essential for gas chromatography, are not best established of any medical therapy. The prevention
required for this type of chromatography. of any single seizure is important because each episode
2. Relatively little sample workup is required is potentially a life-threatening event. Epilepsy is a
before analysis. general term for a group of disorders in which there are
3. Characteristically, methods involving HPLC recurrent seizures. The causes of epilepsy are numerous.
require only a short analysis time. In many patients, a particular pathological brain lesion is
4. Derivatization of the analyte is not required. not identified (i.e., the seizures are idiopathic). The
5. For laboratories with a large workload, the cost major known causes of epilepsy include head injury,
per analysis is very low because reagent costs cerebrovascular accidents (strokes), tumors (both
are low. primary and secondary), infections (meningitis and
6. HPLC affords good sensitivity. encephalitis), electrolyte disturbances, and alcohol and
7. HPLC methods are readily automated. Routine other drugs. The description of a seizure determines the
analyses therefore require a minimum of category or type of epilepsy (Table 2). Such definition is
technician time, a major factor in arriving at the important; different forms of epilepsy respond to
low cost per analysis. different anticonvulsant drugs. It should not be
8. HPLC makes possible the simultaneous overlooked that the treatment of some secondary
analysis of most anticonvulsant drugs in a seizures (e.g., hypoglycemia, electrolyte disturbances,
microsample when required. fever) is primarily the treatment and correction of the
underlying problem.
Some limitations of HPLC in drug analysis
are as follows: The major drugs currently used in the treatment of
1. Spectrophotometric, fluorometric, and epilepsy can be divided into three groups with regard to
amperometric detectors are commonly used. their mechanisms of action [11,12,30]. All of them alter
Therefore the analyte in question must either neuronal conductance of ions, particularly sodium and
absorb light, fluoresce, or be electrochemically calcium. Phenytoin inhibits high-frequency, repetitive
active. neuronal firing, as does carbamazepine. Phenobarbital
2. Equipment is expensive. Investment of capital and valproic acid enhance the synaptic inhibition
is only warranted if the laboratory has a associated with gamma-aminobutyric acid (GABA), as
substantial workload. do the benzodiazepines (e.g., diazepam and
clonazepam). Ethosuximide alters calcium currents in
Phenytoin can be determined by radioimmunoassay [21]. the thalamus. In the past few years, the practice of
Although extremely sensitive, these techniques have the monotherapy with anticonvulsant drugs has become
disadvantages of a relatively long analysis time. RIA increasingly accepted as desirable, and a concerted
requires the availability of a liquid or gamma-ray attempt should be made to control an individual patients
scintillation counter and availability and stability of seizures with a single drug rather than multiple
suitable antisera. anticonvulsants. Patients on monotherapy tend to have a
lower incidence of side effects and drug interactions, as
FPIA and immunoturbidimetric immunoassay have in well as better compliance overall. However, if it
the main become the assays of choice for the clinical becomes clear that in an individual patient, control is not
laboratory. The 2007 College of American Pathologists obtained with a single agent, then combination therapy is
(CAP) Participant Summary Report showed that more utilized. The combination will comprise drugs from
than 80% of participant laboratories used one of these groups with different mechanisms of action.
methods.
Specimen The wide acceptance of the role of therapeutic drug
Plasma or serum may be used for the gas monitoring of anticonvulsants is due not only to the
chromatographic or high-performance liquid importance of good clinical control of epilepsy, but to
chromatographic methods. Serum is the preferred the widely varying pharmacokinetics of the drugs
specimen for the immunoassays. Sera can be stored at between individuals. The major pharmacokinetic
room temperature for several hours. If frozen at 20C, parameters of phenytoin are summarized in Table 3. The
samples are stable for at least 1 year. Saliva may also be timing of the sample to be assayed in relation to the dose
950
Phenytoin
is important. The most appropriate sample for routine drugs: a systematic review. Ther Drug Monit
monitoring is the trough concentration, immediately 1980;2:385-96.
prior to the next dose of the drug. When introducing a 6. Meijer JWA. Simultaneous quantitative
drug or adjusting the dose, it is not generally useful to determination of anti-epileptic drugs, including
measure serum concentrations until the patient has carbamazepine, in body fluids. Epilepsia
reached steady state. From a practical point of view, the 1971;12:341-52.
steady state may be considered as being approached at a 7. Gardner-Thorpe C, Parsonage MJH, Smethurst
time interval equivalent to five times the elimination PF, Toothill C. A comprehensive gas
half-life; for phenytoin, this is more than 5 days. chromatographic scheme for the estimation of
antiepileptic drugs. Clin Chim Acta
In most patients maintained at a steady dosage, stable 1972;36:223-30.
serum phenytoin levels are achieved, but there may be 8. Toseland PA, Albani M, Gauchel FD. Organic
wide inter-patient variability in levels with equivalent nitrogen selective detector used in gas-
dosages. Patients with unusually low levels may be chromatographic determination of some
noncompliant or hypermetabolizers of phenytoin. anticonvulsant and barbiturate drugs in plasma
Unusually high levels may result from liver disease, and tissues. Clin Chem 1975;21:98-103.
congenital enzyme deficiency or drug interactions which 9. Rambeck B, Meijer JW. Comprehensive
result in metabolic interference. method for the determination of anti-epileptic
drugs using a novel extraction technique and a
The plasma therapeutic reference intervals shown are fully automated gas chromatograph.
total concentrations. If a drug is highly protein bound, a Arzneimittelforschung 1979;29:99-103.
decrease in binding protein may potentially increase the 10. Plotczyk LL. Application of fused-silica
unbound moiety of drug. If in addition, there is rate- capillary gas chromatography to the analysis of
limited clearance of that drug, then the unbound moiety underivatized drugs. J Chromatogr
will increase, and there will be a high likelihood of 1982;240:349-60.
clinical toxicity. Such is the case with phenytoin. At the 11. Volmut J, Matisova E, Pham Thi H.
same time, the total plasma concentration may remain Simultaneous determination of six antiepileptic
within the reference interval. In these circumstances, drugs by capillary gas chromatography. J
some practitioners would also measure the free Chromatogr 1990;527:428-35.
phenytoin concentration, utilizing equilibrium dialysis or 12. Meatherall R, Ford D. Isocratic liquid
ultracentrifugation. chromatographic determination of theophylline,
acetaminophen, chloramphenicol, caffeine,
Anticonvulsant Drugs Performance Goals anticonvulsants, and barbiturates in serum. Ther
Survey data from the 2007 CAP Participant Summary Drug Monit 1988;10:101-15.
Report show imprecision values (% coefficient of 13. Riedmann M, Rambeck B, Meijer JW.
variation) for phenytoin to range from 3.6% to 8.5% in Quantitative simultaneous determination of
samples with drug concentrations within the therapeutic eight common antiepileptic drugs and
range. Acceptable Clinical Laboratory Improvement metabolites by liquid chromatography. Ther
Amendments performance criteria (CLIA 88) for Drug Monit 1981;3:397-413.
measurement of phenytoin require that laboratories be 14. Soldin SJ, Hill JG. Rapid micromethod for
accurate to within 25% of the peer-group mean. measuring anticonvulsant drugs in serum by
Imprecision for free phenytoin ranges between 10% and high-performance liquid chromatography. Clin
15% at values of 1.85 mg/L with the majority of assays Chem 1976;22:856-9.
undertaken by FPIA. 15. Christofides JA, Fry DE. Measurement of
anticonvulsants in serum by reverse-phase ion-
References pair liquid chromatography. Clin Chem
1. Plaa GL, Hine CH. A method for the 1980;26:499-501.
simultaneous determination of phenobarbital 16. Kabra PM, Koo HY, Marton LJ. Simultaneous
and diphenylhydantoin in blood. J Lab Clin liquid-chromatographic determination of 12
Med 1956;47:649-57. common sedatives and hypnotics in serum. Clin
2. Svensmark, O, Kristensen, P. Determination of Chem 1978;24:657-62.
diphenylhydantoin and phenobarbital in small 17. Kabra PM, Stafford BE, Marton, LJ.
amounts of serum. J Lab Clin Med Simultaneous measurement of phenobarbital,
1963;61:501-7. phenytoin, primidone, ethosuximide, and
3. Pippenger CE. Scott JE, Gillen HW. Thin-layer carbamazepine in serum by high-pressure liquid
chromatography of anticonvulsant drugs. Clin chromatography. Clin Chem 1977;23:1284-8.
Chem 1969;15:255-60. 18. Ou CN, Rognerud CL. Simultaneous
4. Huisman JW. The estimation of some important measurement of ethosuximide, primidone,
anticonvulsant drugs in serum. Clin Chim Acta phenobarbital, phenytoin, carbamazepine, and
1966;13:323-8. their bioactive metabolites by liquid
5. Rambeck B, Meijer JW. Gas chromatographic chromatography. Clin Chem 1984;30:1667-70.
methods for the determination of antiepileptic 19. Adams RF, Vandemark FL. Simultaneous high-
pressure liquid-chromatographic determination
951
Phenytoin
of some anticonvulsants in serum. Clin Chem 25. Tybach R. Adaptation of prosthetic-group,

1976;22:25-31. labeled, homogeneous immunoassay to reagent
20. Szabo GK, Browne TR. Improved isocratic strip format. Clin Chem 1981;27:1499-504.
liquid-chromatographic simultaneous 26. Nishikawa T, Kubo H, Saito M. Competitive
measurement of phenytoin, phenobarbital, nephelometric immunoassay method for
primidone, carbamazepine, ethosuximide, and antiepileptic drugs in patient blood. J Immunol
N-desmethylmethsuximide in serum. Clin Methods 1979;29:85-9.
Chem 1982;28:100-4. 27. Henderson DR, Friedman SB, Harris JD,
21. Cook CE, Christensen HD, Amerson EW. Manning WB, Zoccoli MA. CEDIA, a new
Radioimmunoassay of anticonvulsant drugs: homogeneous immunoassay system. Clin Chem
phenytoin, phenobarbital, and primidone. In: 1986;32:1637-41.
Kellaway P, Petersen IS, eds. Quantitative 28. Engel WD, Khanna PL. CEDIA in-vitro
Analytical Studies in Epilepsy. New York: diagnostics with a novel homogeneous
Raven; 1976. immunoassay technique. J Immunol Methods
22. Price CP, Newman DJ. Radial partition 1992;150:99-102.
immunoassay. In: Price CP, Newman DJ, eds. 29. Jandreski M.A, Shah JC, Gardincius J, Bermes
Principles and Practice of Immunoassay. New EW. Clinical evaluation of five therapeutic
York: Stockton; 1991. drugs using dry-film multilayer technology on
23. Giegel JL, Brotherton MM, Cronin P, the OPUS immunoassay system. J Clin Lab
DAquino M, Evans S, Heller ZH, Knight WS, Anal 1991;5:415-21.
Krishnan K, Sheiman M. Radial partition 30. Albers GA, Peroutka SJ. Neurologic disorders.
immunoassay. Clin Chem 1982;28:1894-8. In: Melmon KL, Morrelli HF, Hoffman BB,
24. Wong RC, Burd JF, Carrico RJ, Buckler RT, Nierenberg DW, eds. Melmon and Morrellis
Thoma J, Boguslaski RC. Substrate-labeled Clinical Pharmacology: Basic Principles in
fluorescent immunoassay for phenytoin in Therapeutics. 3rd ed. New York: McGraw-Hill;
human serum. Clin Chem 1979;25:686-91. 1992.
Tables
Table 1: Phenytoin Methods Summary
Method 1: Extraction and ultraviolet spectroscopy
Principle of analysis: Anticonvulsant extracted free of other ultraviolet compounds; ultraviolet spectrum or the
ultraviolet difference spectra are specific for the drug.
Comments: Interference with drugs and endogenous compounds
Principle of analysis: Extracted drug separated by TLC and its Rf noted; drug quantitatively eluted and
quantified by ultraviolet spectroscopy
Comments: Technically difficult to achieve reproducibility
Principle of analysis: C18 solid phase extraction; capillary column; flame ionization detection
Comments: Can resolve phenytoin and other anticonvulsants simultaneously; requires simple extraction
Principle of analysis: Organic solvent extraction of drugs or protein precipitation of sample; anticonvulsants are
separated by reversed-phase chromatography and monitored by ultraviolet spectroscopy
Comments: Can resolve phenytoin and other anticonvulsants except valproate simultaneously; commonly used
Method 5: Competitive-binding assays
a. Radioimmunoassay (RIA)
Principle of analysis: Competitive binding of radioactive ligand; radio-labeled hapten competes with unknown
for antibody binding site
Comments: Problems of radioactive usage; slower than other immunoassays
b. Enzyme-multiplied immunoassay technique (EMIT)
Principle of analysis: Competitive binding with drug attached to enzyme
Comments: In common use
c. Fluorometric enzyme immunoassay (FEIA)
Principle of analysis: Combination of competitive binding with enzyme-labeled conjugate for antibody binding
site (on a solid-phase matrix) and radial chromatography
d. Substrate-labeled fluorescence immunoassay (SLFIA)
Principle of analysis: Competitive-binding substrate label; substrate competes with unknown antigen
for antibody binding site
Comments: Fluorescence interferes; not widely used
e. Fluorescence polarization immunoassay (FPIA)
952
Phenytoin
Table 1: Phenytoin Methods Summary (Cont)

Principle of analysis: Competitive binding with fluorescein-labeled drug; fluorescent drug competes with unknown for
antibody site
Comments: In common use (Abbott systems)
f. Ames test strip
Principle of analysis: Antibody binding to drug labeled with prosthetic group prevents glucose oxidase activity;
competition with endogenous drug allows prosthetic group to activate enzyme; glucose oxidase is coupled to
colorimetric reaction
Comments: Designed for inexpensive instrument; reflectance spectrophotometer; rapid test
g. Rate nephelometric inhibition immunoassay
Principle of analysis: Competitive binding for hapten; hapten-protein competes with unknown for antibody
binding site
Comments: Requires specialized instrument
h. Cloned-enzyme donor immunoassay (CEDIA)
Principle of analysis: Competitive binding with enzyme donor (ED)ligand conjugate for Ab binding site;
endogenous drug modulates enzymatic activity by influencing free ED-ligand conjugate available for
complementation
Comments: Can use routine clinical chemistry analyzers (e.g., Roche systems)
i. Particle-enhanced turbidimetric immunoassay (PETINIA)
Principle of analysis: A non-isotopic competitive-binding immunoassay based on the inhibition of agglutination
of latex particles. Antigen is coupled to latex particles, and the antibody remains in the solution phase; inhibition
of agglutination formation
Comments: Can use routine clinical chemistry analyzers (e.g., Siemens Dimension family of analyzers)
j. Dry-film multilayer immunoassay (OPUS)
Principle of analysis: Combination of competitive binding with a fluorescent-labeled hapten in an agarose
matrix and multilayer film technology; labeled antigen competes with sample antigen for antibody binding site
Comments: Requires specialized instrument; not widely used.
Table 2: Classification of Epileptic Seizures*

Type and Characteristics
I. Partial (focal, local). Electroencephalography (EEG) may reveal localized abnormality over seizure focus.
A. Simple partial: consciousness is not impaired. Includes Jacksonian motor epilepsy (confined to a single
muscle group or limb).
B. Complex partial: consciousness is impaired. Most commonly arise in the temporal lobe. Include many
different symptoms and behavior patterns.
C. Secondarily generalized: both (A) and (B) may spread to involve both hemispheres, with convulsions of
all limbs and loss of consciousness.
II. Generalized seizures. Simultaneous initiation of epileptic activity in both hemispheres.
Absence: Brief impairment of consciousness. Motor activity may not occur.
A. Myoclonic: isolated muscle jerks
B. Clonic: rhythmic contractions, loss of consciousness
C. Tonic-clonic (grand mal): major convulsions, loss of consciousness,
autonomic activity, and prolonged depression of central nervous system
function
* Modified from Commission on Classification and Terminology of the Internation League against Epilepsy. Epilepsia
1981;22:489-501.
Table 3: Pharmacokinetic properties.

Anticonvulsant: Phenytoin
Volume of distribution (L/kg): 0.40.8
Elimination half-life (hr): 726
Comments: Dose-dependent kinetics with saturable hepatic metabolism; induction as for phenobarbital; protein
binding important: NB hypoalbuminemia (e.g., renal disease)
953
Phenytoin
Figures
Figure 1
Anticonvulsant Drugs. Ultraviolet absorption spectra of 5 drugs in acetonitrile. Solid line, carbamazepine (10 g/mL); thick
dashed line, ethosuximide (100 g/mL); dashed and dotted line, phenytoin (100 g/mL); regular dashed line,
phenobarbital (100 g/mL); dotted line, primidone, (100 g/mL)
Figure 2
Chromatogram of a serum sample spiked 10 g/mL phenytoin (PT), with internal standard (CA) at 20 g/mL. The serum
and column were acidified prior to the extraction with 0.5 M HCl. Injected volume, 2 L of extract.
peak height relative to an internal standard. Tolybarb is

Procedure: High-Performance Liquid used as an internal standard.
Chromatography Analysis of Anticonvulsant Drugs
Specimen
Principle Serum or plasma.
Phenytoin is separated by reversed-phase liquid
chromatography and quantitated by measurement of Reagents and Materials
954
Phenytoin
1. HPLC-grade acetonitrile, methanol, samples are removed from the air stream on
isopropanol and chloroform can be purchased reaching dryness.
from Fisher Scientific (Winnipeg, Manitoba, e. Reconstitute the residue with 100 L of
Canada). mobile phase.
2. Standards. The pure drug can be purchased 3. Chromatographic parameters.
from Sigma Chemical Co., St. Louis, MO a. Flow rate 2.0 mL/min.
(phenytoin). b. Wavelength 204 nm.
3. Phosphate buffer. c. Sensitivity 0.08 AUFS(absorbance units at
a. 10 mmol/L potassium phosphate full scale).
buffer (pH 6.3) is prepared by d. Column temperature ambient.
dissolving 2.72 g of KH2PO4 in 1900
e. Injection volume 4 L.
mL of distilled water. The pH is
f. Run time 7 min.
adjusted to 6.3 with about 2 mL of 1 M
NaOH, then the volume is brought to 2
A typical chromatogram is shown in the Figure 2:
L.
b. 1 mmol/L potassium phosphate Calculations
buffer (pH 6.8) is prepared by In each sample, peak heights (PH) of each drug
dissolving 13.6 g of KH2PO4 in 90 are compared to the internal standard from which peak
mL of distilled water. The pH is height ratios (PHR) are calculated. Each unknown
adjusted to 6.8 with about 3 mL of 10 sample PHR is compared to a calibrator PHR of known
M NaOH, then the volume is brought concentration. The calculation of drug concentration is
to 100 mL. as follows.
4. Stock internal standards are prepared by [DRUG] = [CAL] PH of sample/ PH of internal
dissolving 30 mg of tolybarb in 20 mL of standard
methanol. PH of calibrator/ PH of internal
5. Solvent extraction solution is prepared by standard
combining 500 L of the stock internal standard
where [DRUG] = drug concentration (mg/L)
solution with 500 mL of 5% isopropanol in
[CAL] = calibrator concentration (mg/L)
chloroform.
6. Mobile phase is prepared by combining 1300
mL of phosphate buffer (pH 6.3) with 350 mL Procedure: Antiepileptic Drugs Including Valproate
of methanol and 350 mL of acetonitrile. The by Capillary Gas Chromatography
mixture is degassed by stirring in a magnetic
mixer for 30 min, then filtered by suction Principle
through a 0.45-m pore filter (Millipore Corp., Phenytoin is extracted from acidified serum
Bedford, MA). The apparent pH of the mobile samples by a simple solid-phase (C18) extraction
phase is 7.0. process. Prior to evaporation, basic methanol is added to
the eluate from extraction column to convert the acid
Assay drug to its potassium salts in order to reduce the loss of
1. Equipment. The liquid chromatographic volatile drug. The residue is then redissolved in acidic
system consists of a Series 2 pump, a model methanol, and 2 L is injected into the gas chromatograph
LC-75 variable wavelength detector, an R-100 equipped with a split capillary inlet system and a flame
strip chart recorder (all from PerkinElmer ionization detector.
Corp., Norwalk, CT), and a model 7105 injector
(Rheodyne Inc., Cotati, CA). A 5-m particle Specimen
size, 250 4.6 mm Supelcosil LC-1 analytical Serum.
column and a 5-m particle size, 20 4.6 mm
Supelguard column (Supelco, Bellefonte, PA) Reagents and Materials
are used to effect the drug separations. 1. Phenytoin (SPOFA, Prague, Czechoslovakia
and VEB Arzneimittelwerk, Dresden,
2. Sample preparation.
Germany).
a. Add 100 L of 1 mmol/L phosphate buffer 2. Stock standard solution of the drug is
(pH 6.8) and 1.5 mL of solvent extracting prepared in analytical reagent-grade methanol
solution (containing internal standard) to 100 to a concentration of 1 g/L.
L of serum or plasma.
3. Stock internal standard is prepared by
b. Vortex mix the sample for 30 sec and dissolving 40 mg of MPPH [5-(4-
centrifuge briefly to separate the two phases. methylphenyl)-5-phenylhydantoin] in 20 mL of
c. Aspirate the aqueous phase (top layer) to methanol.
waste, and transfer the organic phase (lower 4. 0.5 M HCl solution in water. Dilute 4.2 mL of
layer) to a clean tube. concentrated HCl to 100 mL with H2O.
d. The organic layer is evaporated at room
temperature with a dried air stream. The
955
Phenytoin
5. 0.05 M KOH in methanol. Add 0.281 g of polarization immunoassay. After a filtration step, a
KOH to 80 mL of methanol, and bring to 100 clear supernatant containing the drug is analyzed.
mL with methanol. Unlabeled drug in the sample competes with the
6. 0.05 M HCl in methanol. Dilute 42 L of tracer labeled with fluorescein for limited antibody
concentrated HCl to 100 mL of methanol. sites. The tracer, when excited by linearly
polarized light, emits fluorescence with a degree of
Assay polarization inversely related to its rate of rotation.
Equipment: A HP-5790A gas chromatograph If the patient sample contains high concentrations
(Hewlett-Packard, Avondale, PA) equipped with a split of the drug, the polarization value is low. If the
injection port (ratio 1:20) using an empty glass liner and patient sample contains a low concentration of the
flame ionization detection. A fused-silica capillary drug, the polarization value is high.
column packed with cross-linked 5%
phenylmethylsilicone gum phase HP-5 (Hewlett- II. Specimens of Choice
Packard), 25 m 0.20 mm ID, 0.33 m film thickness is Serum (2 mL) drawn in a red-top Vacutainer tube
used to separate the drug of interest. without separator gel. (NOTE: Gel tubes are not
Sample preparation recommended because of their potential to
decrease drug concentration during storage. This
1. Add 100 L of 0.5 M HCl and 10 L of internal
is negated somewhat by always aliquoting serum
standard solution (MPPH) to 500 L of serum,
into a separate pour-off tube immediately.)
and thoroughly vortex mix.
Recommended drawing time is the trough level,
2. Precondition the extraction column (a 2-mL immediately prior to the next scheduled dose of
polyethylene syringe packed with 200 mg of drug, or during the acute phase of toxicity.
Silipor C18) with a 1-mL wash of methanol
Specimens may be transported at RT. Separate
followed by a 1-mL wash of 0.5 M HCl.
serum, and store at RT for up to 24 hr until time of
3. Pour mixture (from step 1) onto the column, analysis or prepare ultrafiltrate (Section VI.A),
and allow the mixture to flow through. Then then freeze at 20C until time of testing.
rinse the column with two 1-mL volumes of
Grossly hemolyzed, lipemic, or icteric specimens
H2O, and dry by applying a vacuum.
are unacceptable samples for routine analysis.
4. The drug is then eluted from the column with 1
mL of methanol. A 50-L aliquot of 0.05 M III. Indications
KOH is added to the methanol solution. The For the monitoring of phenytoin for therapeutic
solvent is then evaporated under nitrogen at drug monitoring (TDM) purposes, as an aid in
40C in a heating block. The residue is setting the dosing regimen, and to assess toxicity
reconstituted in 50 L of 0.05 M HCl in when total phenytoin drug analysis does not
methanol. provide adequate information.
5. Chromatographic parameters:
a. Injection temperature 240C IV. Reagents and Material
b. Detector temperature 300C A. Bio-Rad Control Materials (Level 1 and 2)
c. Dual ramp column temperature Store at 4C. Stable as labeled
program (60C to 250C)
B. 100-L and 1-mL Eppendorf pipets
d. Nitrogen inlet pressure 100 kPa
e. Injection volume 2 L C. Amicon CentriFree Micropartition Filters
(Amicon)
Calculation D. Sample Cartridges (Abbott)
1. A single-point calibration: E. Sample Cuvettes (Abbott)
Peak Height Ratio (PHR) = Peak Height of Drug F. TDx Reagent Pack Adapter for FLx
Peak Height of IS
G. Calibration and Routine Carousels (Abbott)
[Drug] = PHR of Drug [Standard] H. Buffer (Abbott); store at room temperature
PHR of Standard The following Abbott Laboratories Phenytoin
Assay Kit Materials:
where: IS = Internal Standard I. Phenytoin Reagent Pack* (Batch Assay)
[Drug] = concentration of drug (mg/L) J. Free Phenytoin Calibrators*
[Standard] = concentration of standard (mg/L) *Store at 4C. Stable for 1 month after opening.
Free Phenytoin V. Instrument and parameters.

Abbott TDx/FLx (Abbott Laboratories TDx and Abbott TDx/FLx; however, any of the automated drug
TDx/FLx Operations Manual.) analyzers which have a linear range to 0.5 g/mL are
I. Principle acceptable
The concentration of free, or unbound, phenytoin
in samples can be determined using fluorescence
956
Phenytoin
VI. Procedure 1. Performing the calibration:

Each routine run will generally consist of QC, Assays should be calibrated whenever:
along with patients processed in singlicate. A new assay is introduced
Controls and patients require Amicon filtering; An assay activation (new reagent pool) is
calibrators do not. issued
A. Pretreatment (for control and patient The memory circuit board (Board #2) is
specimens) replaced
When conducting ultrafiltration prior to Assay control values fall outside the acceptable
measurement of free phenytoin, check that
range specified in the specific assay section and
the room temperature at that time does not
exceed 25C. If the temperature is > 25C, the control is not suspect.
contact the area supervisor. If the sample is Assays may require recalibration (when indicated
ultrafiltered at room temperature of 30C to by unacceptable control data) whenever:
37C, protein binding will be affected, and a A new lot number of reagent is used
different reference range may need to be A new lot of buffer is used
applied: Any dispense component is replaced
12 g/mL at 25C Any instrument calibration procedure is performed
1.53 g/mL at 37C A wide variation in room temperature is
1. Free phenytoin calibrators, controls, experienced
and patient specimens should be mixed
by gentle inversion. Avoid foaming; if The TDx/FLx System stores a calibration curve for
this should occur, allow sample to sit each assay. The random-access calibration curves
until foam dissipates. are separate and distinct from the calibration
2. Number an Amicon filter for each curves stored for the batch assays. Therefore,
specimen to be tested, and place in a prior to running an assay, the instrument must be
suitable rack. A maximum of 20 calibrated for that assay in the mode of operation
specimens can be run in one batch. being run.
3. Pipet 1 mL of control or patient C. Sensitivity: 0.5 g/mL; report negatives as < 0.5
sample to be assayed into its g/mL
corresponding Amicon filter.
4. Cap the Amicon filter sample Notes and Interpretations
reservoir.
A. Optimal therapeutic concentrations of free
5. Place filter into a fixed-angle phenytoin will generally be between 1 and 2
centrifuge, and centrifuge samples for
g/mL. Frequency and severity of toxic
25 min.
effects due to drug therapy will increase as
6. After centrifugation, transfer 100 L the free phenytoin level exceeds 3.0 g/mL.
of ultrafiltrate into appropriate sample
Results may be reported up to 10 g/mL.
wells, and place in a sample carousel.
B. Specificity: no known interfering substances
B. Calibration (no Amicon filtering required)
957
Phosphorus and Phosphate

Name: Phosphorus, inorganic phosphorus, phosphate
Molecular mass: Phosphate 97.00, 96.00 D (phosphorus at 30.98)
Chemical class: Inorganic anions (P is nonmetallic group 5 element)
Forms: H2PO4, HPO42
Principles of Analysis and Current Usage hydrazine sulfate [2] (Table 1, Method 2). This also
Elemental phosphorus does not exist to any appreciable permitted rapid color development at room temperature.
extent in the body, and so methods have been directed Methods were developed for automated analyzers based on
toward analysis of the two phosphate anions, which the work of Hurst and Kraml [3], in which the
interchange rapidly depending on the pH. i concentration of stannous chloride is doubled to increase
The monovalent and divalent anion forms are present in linearity.
serum at about equal concentrations in acidosis, in a ratio
of 1:9 in alkalosis, in a ratio of 1:4 at pH 7.4, and in a ratio Taussky and Shorr [4] used ferrous ammonium sulfate as a
of 100:1 in a pH 4.5 urine. Thus it is impossible to say reducing agent (Table 1, Method 3). The merits and
with any certainty what the molecular mass of inorganic drawbacks of the use of these different reducing reactions
phosphate is. Therefore, the units traditionally chosen have been reviewed by Martinek [5].
have been milligrams or millimoles of phosphorus (in a
volume) but never milliequivalents of phosphate, since that Simonsen et al. [6] showed that phosphate can be
would change rapidly with the charge. Because measured using the 340-nm absorbance of the unreduced
phosphorus exists in the body as phosphate, and because phosphorusmolybdate complex (Table 1, Methods 4 and
that is the form measured, the two terms are used 5). This made the assay faster and simpler and improved
interchangeably. reagent stability. It also produced three to four times more
absorbance change than the molybdenum blue reductions.
The oldest and still most commonly used methods are Close to 80% of laboratories use this technique today (see
based on the reaction of phosphate ions with molybdate to Table 2).
form complex structures such as ammonium
phosphomolybdate; (NH4)3[P(Mo3O10)4]. One can Enzymatic methods have been developed to measure
serum inorganic phosphate (Table 1, Method 6). A cited
measure phosphomolybdate complexes directly or convert advantage of this technique is the stability of organic
them to molybdenum blue using a wide variety of reducing phosphate compounds in the neutral range, where most
agents (Table 1, Methods 1 to 6). Some of the earlier enzymes function. A number of enzyme-based methods
methods required heating or produced colored complexes have been described. One early assay proposed by Schultz
that were unstable. Molybdenum blue is a complex et al. [7] used a coupled enzyme system of glycogen
heteropolymer of unknown structure whose absorbance is phosphorylase, phosphoglucomutase, and glucose-6-
usually measured at 660 nm (Figure 1A). Approximately phosphate dehydrogenase. The amount of phosphate
20% of laboratories reporting in College of American present was determined by measurement of the NADPH
Pathologists (CAP) proficiency surveys use a formed. A coupled enzymatic method has also been
phosphomolybdate reduction assay. marketed [8]. The steps of the enzymatic reaction involve
formation of ribose-1-phosphate and hypoxanthine from
Kutter and Cohen used stannous chloride stock reagent to inorganic phosphate and inosine, oxidation of
produce a more intense color with phosphomolybdic acid hypoxanthine to uric acid and H2O2, and formation of a
(Table 1, Method 1) [1]. Hurst stabilized the dilute
stannous chloride working reagent by combining it with chromogen-using peroxidase and a chromogenic substrate.
Ghetta et al. [9] recently described the use of
polynucleotide phosphorylase, an enzyme that catalyzes
i First Edition phosphorolysis of polynucleotides, with release of
E. Christis Farrell nucleotide diphosphates. In the presence of polyadenylic
Methods E. Christis Farrell acid, phosphate is converted into ADP by this enzyme.
Second Edition E. Christis Farrell ADP then reacts with phosphoenolpyruvate in a pyruvate
Third Edition Steven C. Kazmierczak kinasecatalyzed reaction, giving rise to adenosine 5-
Fourth Edition Steven C. Kazmierczak triphosphate and pyruvate. Finally, pyruvate oxidizes
958
NADH through the action of L-lactate dehydrogenase, with accounts for this difference. However, others have found
a concomitant decrease in absorbance at 340 nm. Only no significant difference between serum and plasma [17].
approximately 2% of laboratories utilize enzymatic No significant difference has been seen in concentrations
phosphate methods. from specimens collected into plastic and those collected
into glass serum separator tubes [18]. Storage of whole-
Reference and Preferred Methods blood specimens at room or refrigerated temperatures for
There is no reference method for serum phosphorus. prolonged periods will result in increased phosphate
concentrations. Prompt separation of serum or plasma
The majority of laboratories today perform analysis of following collection is necessary. Separated serum or
phosphate by employing direct analysis of the phosphorus- plasma is stable for 4 days at room temperature, 7 days at
molybdate complex. There are two factors that complicate 4C and indefinitely at 20C [19].
what seems to be a simple, straightforward adaptation of
this reaction to an automated system. The first factor is that Serum values will be lower after meals, and it is best that
aqueous standards react considerably more slowly than the patient be fasting before the sample is drawn.
protein-based standards and may not attain the end-point at Phosphate will also be lower during the menstrual period.
the same time as serum unknowns [10]. Albumin can be A diurnal variation has been found, with a minimum
added to the standards to increase the reaction rate, but the occurring at 8:00 am and highest around 2:00 am [20].
albumin may also contain phosphate. The second factor is Intravenous glucose and fructose also physiologically
the unusual influence of pH; acidity is necessary, but as the lower phosphate.
pH decreases to 1.0, the reaction is substantially slowed,
and as it is increased beyond 2.0, spontaneous reduction of Urine may contain larger quantities of organic phosphates,
molybdate takes place [10,11]. which can decompose on exposure to elevated
temperatures. When acidified with HCl, urine phosphate is
In the automated methods, a serum blank is needed stable for more than 6 months [12]. Urine samples are
because of the presence of a great number of 340-nm usually diluted 1:10 with normal saline before analysis.
absorbing compounds and drugs. The blank should also
contain surfactant, to clear serum turbidity, and sulfuric Interferences
acid, to correct for the effect of pH on the ultraviolet Compounds such as citrate, fluoride, oxalate, tartrate, and
absorption of many compounds. EDTA can form complexes with molybdate and decrease
color development. Heparin is the preferred anticoagulant.
Additional problems encountered with the Leakage of phosphate from cells falsely increases serum or
phosphomolybdate-reduction type of assay include plasma concentrations. Whereas losses from erythrocytes
instability of the reducing reagent. This is especially true produce a positive influence, hemoglobin will produce a
for the stannous chloride reagent, which also shows poor strong negative photometric influence. For these reasons,
adherence to Beers law [12]. Stabilization of SnCl2 with the preferred methods of phosphate analysis require correct
hydrazine sulfate has been used to overcome this blanking of the individual sera. Use of 340- and 380-nm
difficulty. filters and 340-nm monitoring with blanking is optimum.
Crouch and Malmstadt studied the kinetics of the overall Icteric sera should be properly blanked. Blanking with use
molybdenum blue reaction and found that the rate was first of bichromatic analysis with 340- and 380-nm wavelengths
order with respect to phosphate but that the velocity was is helpful. Lipemic sera, whose absorbance is higher at the
highly sensitive to changes in pH and molybdate shorter wavelengths, can produce a positive bias with some
concentration [13]. It has also been noted using the ANS methods.
procedure that at higher pH readings of 2 or 3, nonspecific
reduction of the molybdate reagent takes place [14]. Phosphate Reference Intervals
Reference intervals for serum phosphate are age dependent
The reaction conditions of these reduction assays are and also method dependent, as illustrated below. Mean
strong enough to hydrolyze glucose phosphate, creatine concentrations in neonates are approximately 2.0 mmol/L
phosphate, and other organic phosphates, creating a higher than adults [21]. Concentrations of phosphate show
positive error. Experience with the enzymatic methods a gradual decline with age. Healthy elderly individuals
shows these to be less affected by organic phosphate show values approximately 0.2 mmol/L lower when
compounds [15]. compared with individuals < 60 years of age [22]. Age-
related changes are more pronounced in older males as
Specimen compared with females. Decreases in phosphate in older
Serum is preferred by some because it has been suggested men may be related to decreased tubular reabsorption due
that phosphate concentration in plasma samples are to higher parathyroid hormone (PTH) concentrations [23].
approximately 0.2 to 0.3 mg/dL lower than in serum [16].
The release of intracellular phosphate during clotting Serum phosphate concentrations show a decrease
959
following food intake, as a result of phosphorylation of

glucose and metabolism [24]. Factors that may cause an
increase in serum phosphate include exercise and
pregnancy.
Serum Phosphorus Urine

Method mg/dL (mmol/L) g/24 hr
Molybdenum blue methods
Henry TCA Filtrate 12
In children 4.07.0 (1.292.26) 0.50.8
In adults 2.54.8 (0.811.55) 0.31.3
Unreduced phosphomolybdate
methods
Gilford 340 nm rate, adults 2.54.8 (0.811.55)
Abbott 340 nm kinetic, adults 2.04.6 (0.651.49)
Urinary phosphate excretion is influenced by a variety of Phosphorus Performance Goals

Acceptable performance criteria for measurement of
factors, including muscle mass, renal function, patient age,
phosphate requires that laboratories be accurate to within
and time of day. A diurnal variation in urine phosphate
the greater of 0.3 mg/dL or 10.7% of the peer-group
excretion has been noted, with the highest output occurring
mean (Clinical Laboratory Improvement Amendments of
in the afternoon. No significant gender-related differences
1988 [CLIA 88]). Precision goals based on biovariability
have been reported.
data indicate a maximum coefficient of variation (CV) of
4.3% for serum phosphate measurements and 13.2% for
Interpretation
urine [25]. Data from proficiency testing surveys indicate
Phosphorus and calcium metabolism are intertwined. In
laboratories using the phosphomolybdate method for
healthy persons, as serum calcium levels rise, those of
phosphorus meet this requirement for precision.
phosphorus fall, under the influence of PTH. Control of
Intraindividual variation for serum phosphate measured
phosphorus levels is in part accomplished by regulation of
over periods greater than 1 month has been reported as
renal excretion. A more detailed description of the levels
ranging from approximately 2% to 7% [26].
of phosphorus in disorders of bone disease is presented in
Chapter 33 of Kaplan and Pesces Clinical Chemistry:
Theory, Analysis, Correlation, 5th edition. In addition to References
1 Kuttner T, Cohen HR. Microcolorimetric studies.
being present in plasma, phosphorus is a major
I. A molybdic acid stannous chloride reagent: the
intracellular anion. A significant portion of both plasma
microestimation of phosphate and calcium in pus,
and intracellular phosphorus is in the form of
plasma, and spinal fluid. J Biol Chem
organophosphate compounds such as glucose-6-phosphate
1927;75:517-531.
and 2,3-diphosphoglycerate. Methods of measurement of
serum phosphate usually attempt to measure only the 2 Hurst RO. The determination of nucleotide
inorganic phosphate. However, fairly rapid fluctuations in phosphorus with stannous chloride-hydrazine
serum inorganic phosphate can occur because the serum sulphate reagent. Can J Biochem 1964;42:287-
inorganic phosphate concentration is influenced by 292.
carbohydrate metabolism. Part of the metabolic processing 3 Kraml M. A semi-automated determination of
of carbohydrates is the formation of organic phosphate 3- phospholipids. Clin Chim Acta 1966;13:442-448.
and 6-carbon intermediates in the cell cytoplasm. This 4 Taussky HH, Shorr J. Microcolorimetric method
causes a phosphate shift from plasma into the cell. for determination of inorganic phosphorus. J Biol
Chem 1953;202:675-685.
In diabetes, severe loss of phosphate is possible, since 5 Martinek RG. Review of methods for determining
carbohydrate metabolism is deranged, and phosphate tends inorganic phosphorus in biological fluid. Am J
to pass from the cell into extracellular fluid and then into Med Technol 1970;32:337-345.
plasma. It is then extracted and excreted by the kidney.
6 Simonsen DG, Wertman M, Westover LM, Mehl
Another example of the interrelationship between
JW. The determination of serum phosphate by the
phosphorus and carbohydrates can be found in the use of
insulin therapy to treat diabetic ketoacidosis. There can be molybdivanadate method. J Biol Chem
a large migration of phosphorus from extracellular to 1946;166:747-755.
intracellular space accompanying the passage of glucose 7 Schultz DW, Passanneau JV, Lowry OH. An
into cells because of the formation of 3- and 6-carbon enzymatic method for the measurement of
phosphate intermediates. inorganic phosphate. Anal Biochem 1967;19;300-
960
314. 18 Hill BN, Laessig RH, Koch DD, Hassemer DJ.

8 Adam A, Boulanger J, Azzouzi M, Ers P. Comparison of plastic versus glass serum
Colorimetric versus enzymatic determination of separator (SST) blood-drawing tubes for common
serum phosphorus. Clin Chem 1984;30:1724- clinical chemistry determinations. Clin Chem
1725. 1992;38:1474-1478.
9 Ghetta A, Matus-Ortega M, Garcia-Mena J, Deho 19 Winsten S. Collection and preservation of
G, Tortora P, Regonesi ME. Polynucleotide specimens. Stand Methods Clin Chem 1965;5:1-
phosphorylase-based photometric assay for 17.
inorganic phosphate. Anal Biochem 20 Calvo MS, Eastell R, Offord KP. Circadian
2004;327:209-214. variation in ionized calcium and intact parathyroid
10 Michelsen OB. Photometric determination of hormone: evidence for gender differences in
phosphorus as molybdovanadophosphoric acid. calcium homeostasis. J Clin Endocrinol Metab
Anal Chem 1957;29:60-62. 1991;72:69-76.
11 Daly JA, Ertinghausen G. Direct method for 21 Evans SE, Durbin GM. Aspects of the
determining inorganic phosphorus in serum with physiological and pathological background to
Centrifichem. Clin Chem 1972;18:263-265. neonatal clinical chemistry. Ann Clin Biochem
12 Henry RJ. Clinical Chemistry: Principles and 1983;20193-207.
Technics. 2nd ed. New York: Harper & Row; 22 Thompson SP, While DA, Hosking DJ, Wilton
1974:726. TJ, Pawley E. Changes in osteocalcin after
13 Crouch SR, Malmstadt HV. A mechanistic femoral neck fracture. Ann Clin Biochem
investigation of molybdenum blue method for the 1989;26:487-491.
determination of phosphate. Anal Chem 23 Tietz NW, Shuey DF, Wekstein DR. Laboratory
1967;39:1084-1089. values in fit aging individualssexagenarians
14 Tietz N. Fundamentals of Clinical Chemistry. 2nd through centenarians. Clin Chem 1992;38:1167-
ed. Philadelphia: Saunders; 1976:915. 1185.
15 Zoppi F, Giachetti M. More on enzymatic assay 24 Garb S. Clinical Guide to Undesirable Drug
of inorganic phosphorus [letter]. Clin Chem Interactions and Drug Interferences. New York:
1985;31:1242. Springer; 1972.
16 Carothers JE, Kurtz NM, Lemann J. Error 25 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
introduced by specimen handling before Hernandez A, Jimenez CV et al. Current
determination of inorganic phosphate databases on biological variation. Pros, cons, and
concentrations in plasma and serum. Clin Chem progress. Scand J Cin Lab Invest 1999;59:491-
1976;22:1909-1912. 500.
17 Miles RR, Roberts RF, Putnam AR, Roberts WL. 26 Young DS. Effects of Preanalytical Variables on
Comparison of serum and heparinized plasma Clinical Laboratory Tests. Washington, DC:
samples for measurement of chemistry analytes AACC Press; 1997.
[letter]. Clin Chem 2004;50:1704-1706.
961
Tables
Table 1: Methods of Phosphorus Analysis
Method 1: SnCl2*
Principle of analysis: Reduction of phosphomolybdate to molybdenum blue
Comments: Early continuous flow; replaced by improved methods because of reagent instability, poor linearity, and
lack of precision
Method 2: SnCl2 + Hydrazine*
Comments: Continuous flow; good accuracy, limited linearity, better reagent stability than method 4
Method 3: NH4FeSO4*
Comments: Discrete analyzers; not so sensitive as ultraviolet methods
Method 4: No reduction*
Principle of analysis: 340 nm, monitoring
Comments: Discrete and continuous flow; good accuracy, precision, sensitivity
Method 5: No reduction*
Principle of analysis: Bichromatic
Comments: Some discrete analyzers; biochromatic (340 and 380 nm) about one-half sensitivity of 340 nm; blanking
problems with hemolysis and lipemia
Method 6: Enzymatic
_PNP
HPO42 + inosine hypoxanthine + ribose-1-phosphate
_
Hypoxanthine + 2H2O + 2O2 XOD uric acid + 2H2O2
_
H2O2 + chromogenic substrate peroxidase reddish purple complex
(Amaximum, 555 nm)
Comments: Discrete analyzers; new assay
*General principle: PO4 + H2SO4 + (NH4)6Mo7O24 4H2O Mo-PO4 complex
Mo/PO4 complex + reducing agent heteropolymeric blue complex
PNP, Purine nucleoside phosphorylase (EC 2.4.2.1); XOD, xanthine oxidase (EC 1.2.3.2); chromogenic substrate is
N-ethyl-N-(3-methylphenyl)-N-acetylethylenediamine.
962
Table 2: Reaction Conditions for Serum Phosphate Analysis

Condition: Sample volume
Nonreduction at 340 nm*: 3 L
Reduction method: 10 L
Condition: Fraction of sample volume
Nonreduction at 340 nm*: 0.013
Reduction method: 0.014
Condition: Final concentration of reagents
Nonreduction at 340 nm*:
Ammonium molybdate: 2.25 mmol/L
Sulfuric acid: 750 mmol/L, pH 1.55
Tween 80: 0.45 mL/L
Reduction method:
Sodium acetate: 55.7 mmol/L
Acetic acid: 328 mmol/L
Sodium metabisulfate: 37.2 mmol/L
p-Methylammonium phenol sulfate: 11.0 mmol/L
Ammonium paramolybdate 7 H2O: 587 mol/L
H2SO4: 16.2 mmol/L
CU2+: as catalyst
Nonreduction at 340 nm*: 37C
Reduction method: 30C
Condition: Wavelength
Nonreduction at 340 nm*: 340 nm
Reduction method: 620 nm
Condition: Reaction
Nonreduction at 340 nm*: End point with serum blank
Reduction method: Two-point kinetic
Condition: Time
Nonreduction at 340 nm*: 200 sec
Reduction method: 300 sec
Condition: Precision (X, %CV)
Nonreduction at 340 nm*: 25 mg/L, 9.2% CV; 78.7 mg/L, 5.3% CV
Reduction method: 24.9 mg/L, 7.2% CV; 77.3 mg/L, 4.8% CV
Nonreduction at 340 nm*: 80 mg/L
Reduction method: 150 mg/L
X, Mean; CV, coefficient of variation.
*As analyzed on COBAS analyzer (Roche Inc., Nutley, NJ), using reagent obtained from Gilford Diagnostics,
Oberlin, Ohio.
As analyzed on CentrifiChem Analyzer (Baker Instruments, Dallas, Texas), using reagent from Smith-Kline-
Beckman (Sunnyvale, CA).
From College of American Pathologists quality assurance survey data for nonreduction and reduction method
groups (overall).
963
Figure
Figure 1
A, Absorption spectra of reduced phosphomolybdate complex. Dashed line, Reagent versus H2O blank; dotted-
dashed line, serum and reagent versus H2O blank; solid line, serum and reagent versus reagent blank. Reagent used
metabisulfite as reducing agent. Serum standard contained 50 mg/L of inorganic phosphate.
B, Absorption spectra of unreduced phosphomolybdate complex. Curve a, Reagent versus H2O blank; Curve b,
hemolyzed serum versus reagent blank; Curve c, lipemic serum versus reagent blank; Curve d, icteric serum versus
reagent blank; Curve e, 80 mg/L of standard versus reagent blank. In b to e, 20 L of serum and 2.0 mL of reagent
were mixed. The final reaction was performed at pH 1.55. In this determination, hemolysis was defined as deep red
color, lipemia as creamy serum, and icteric as a bilirubin content of 24 mg/L.
964
Metanephrines, Plasma Free

Brett McWhinney i
Name: Metanephrines: normetanephrine (NM), 3-O-methyl-norepinephrine;

metanephrine (M), 3-O-methylepinephrine
Clinical significance: Refer to Chapter 51, Adrenal Hormones and Hypertension, in the 5th edition
of Clinical Chemistry: Theory, Analysis, and Correlation
Molecular formula: NM, C9H13NO3; M, C10H15NO3
Molecular mass: NM, 183.20 D; M, 197.23 D
Merck Index: NM, 6549; M, 5777
Chemical class: O-Methylated catechol derivative
Structure:
Principles of Analysis and Current Usage hydroxide/methanol [1:3]) to remove any impurities that
Recently, Eisenhofer and colleagues have proposed that might otherwise contaminate the final elution. Columns
plasma free metanephrines (metanephrine [M] and are then activated by washing with a solution of 1%
normetanephrine [NM]) may be a superior analytical KOH in methanol to bring up the negative charges on
tool for the investigation of the patient with a diagnosis the column matrix. Residual 1% KOH in methanol is
of pheochromocytoma [1,2]. washed off the column with water, after which the
columns are loaded with sample. Cationic species in
There are theoretical reasons why plasma metanephrines samples of plasma are retained on the column, while the
may be superior diagnostically, including that tumor bulk of the sample is eluted to waste. Addition of 1 mL
tissue has a high activity of COMT, an enzyme that of water prior to loading of the columns with samples
converts catecholamines to metanephrines, thus causing helps to retard positively charged metanephrines on the
selective synthesis of metanephrines in tumor tissue column, increasing their recovery from less than 30% to
[3,4,5]. typically more than 70%. Subsequent washing of
columns with acetic acid followed by a slightly basic
Plasma concentrations of total metanephrines (i.e., free solution of ammonium phosphate, water, and methanol
plus conjugated metanephrines) are 20- to 30-fold higher serves to further purify the finally eluted sample.
than free metanephrines. They therefore reflect largely
the concentration of conjugated metanephrines. Glass tubes are placed under the columns, and a solution
Considerable amounts of free metanephrines are of 10% ammonium hydroxide/methanol solution (1:3) is
produced in adrenal medullary or pheochromocytoma passed through the columns to disrupt ion bonding of
tumor tissue. Sulfate-conjugated metanephrines are analyte to column matrix and the elute collected and
formed predominately in gastrointestinal tissue, owing to concentrated using nitrogen evaporation at 80C. The
the high concentrations of monoamine-preferring metanephrines, unlike catecholamines, are stable in the
sulfotransferase [6]. highly basic ammonium hydroxide/methanol solution,
Plasma free metanephrines are present in pg/mL which is fully evaporated away under nitrogen. The
concentrations, and as a result, many of the urinary residue is resuspended in a minimum quantity of mobile
metanephrine assays do not have the sensitivity required. phase ready for injection onto the HPLC system.
In the early 1990s, Eisenhofer developed an HPLC assay
using coulometric detection to measure plasma free The HPLC-EC method relies on separation of the
metanephrines. The plasma metanephrines were analytes of interest on an HPLC column followed by
extracted from plasma and their concentration their ordered elution off the column and post-column
determined by reverse-phase isocratic high-performance detection by way of the current generated from the
liquid chromatography coupled with electrochemical oxidation or reduction of the eluting analytes. The triple
detection (HPLC-ED). electrode system allows the removal of acetaminophen at
the first reducing electrode. The second measuring
The extraction procedure serves to isolate and electrode is therefore free from interference. This is an
concentrate the low pg/mL concentrations of free improvement on the original Eisenhofer method, where
metanephrines in plasma into a purified low-volume acetaminophen interfered with NM quantitation.
form appropriate for injection onto the HPLC apparatus.
Extractions are performed by ion-exchange Enzymatic immunoassays based on microtiter plate
chromatography, using AccuCAT silica solid-phase technology have recently been developed for the
cation exchange columns. Extraction columns are first determination of plasma free NM and M. However,
washed with the eluting solution (10% ammonium compared to HPLC assays, they are very susceptible to
i
Plasma Free Metanephrines
New method
965
cross-reactivity and artifacts caused by non-specific The heparinized blood sample should be centrifuged and
binding. There is also very poor agreement between separated within 2 hours of blood collection. Each
results obtained from different immunoassay aliquot should be at least 2 mL, placed into plastic tubes
manufacturers. These assays also lack an internal clearly marked with the date and patient ID and frozen
standard to monitor recovery through the extraction immediately (e.g., on dry ice). The sample should be
process. stored at 20C or colder [8].
Recently, liquid chromatography with tandem mass Interferences

spectrometry (LC-MS/MS) methods have been The method of Eisenhofer suffers from interference by
developed. Sample preparation still involves off-line acetaminophen [9,10], a frequently prescribed analgesic
solid-phase cleanup prior to injection on to the LC- drug. Modifications to the original assay with changes to
MS/MS. Further on-line cleanup occurs using reverse- the extraction procedure and detector settings have now
phase columns and switching valves prior to the elution removed this problem. Medication-associated false-
into the MS/MS [7]. positive results can occur with phenoxybenzamine and
tricyclic antidepressants, producing levels 1.9- to 2.6-
Reference and Preferred Methods fold higher results compared with patients on no
There is no reference method for either plasma M or medications [11].
NM. The HPLC technique is the most popular because it
can separate M and NM quickly (less than 20 min) with Metanephrines Reference Interval
almost no interference from any drug or natural Using the original Eisenhofer method, the following
biochemical. The sensitivity of the HPLC methods with concentrations of NM and M were found in 178
electrochemical detection is in the picomole range, with normotensive individuals, including 100 men and 78
between-run coefficients of variation (CVs) reported to women, ages ranging from 18 to 72 yrs [12].
be less than 10%. Because the HPLC analysis provides
highly specific measurements of both M and NM, and Reference Ranges (Adjusted for Gender or Age)
HPLC equipment is rapidly becoming routine equipment Reference Ranges (Not Adjusted for Gender or Age)
in many clinical chemistry laboratories, it is the Analyte Geometric Upper
Geometric Upper Lower
Lower
recommended method for the analysis of plasma free Analyte Mean
Mean Limit Limit
Limit Limit
metanephrines. Normetanephrine (NMN)
(pg/mL)
Normetanephrine (NMN)45 112 18
Mobile phase eluting from the outlet of the analytical (nmol/L)
age < 38 yr (pg/mL) 0.25
40 0.61
95 0.10
17
column and containing the analytes of interest passes Metanephrine (MN)
age < 38 yr (nmol/L) 0.22 0.52 0.09
through a triple electrode system, consisting of a series (pg/mL)
age > 38 yr (pg/mL 2751 61
125 1221
Model 5021 conditioning cell, followed by a Model (nmol/L)
age > 38 yr (nmol/L)0.140.28 0.31
0.68 0.060.11
5011 analytical cell. Both cells are connected to an ESA Metanephrine (MN)
Model 5100A Coulochem detector. The first electrode in Male (pg/mL) 29 62 21
the Model 5021 conditioning cell is set to a potential of Male (nmol/L) 0.1 0.31 0.11
about +0.41 V; the second and third electrodes are set to Female (pg/mL) 24 55 11
potentials of +0.01 and about 0.35V respectively. Female (nmol/L) 0.12 0.28 0.06
Because of variations between cells, the potential of a
new cell may require adjustment according to Reference ranges above are adjusted for a highly
voltamograms that should be generated when cells are significant influence (P<0.001) of gender on plasma
first installed. Coulometric oxidation at the first concentrations of metanephrine and total metanephrine
electrode serves to screen out irreversibly oxidized and a highly significant influence of age on plasma NM
contaminating species. The second electrode reduces and total NM.
compounds that may otherwise interfere with subsequent
detection by a reducing potential at the third analytical The development of newer assays has also resulted in the
electrode. The hydroxyl group on metanephrines is establishment of other reference ranges. The upper limits
normally in the reduced state, but the compounds are of the reference ranges were 165 pg/mL for NM and 100
reversibly oxidized and reduced so that measurement by pg/mL for M [6,7,13,14].
reduction is possible after the compounds are oxidized at
the first electrode in the conditioning cell. These values are higher than those established by
Eisenhofer. It has been proposed that the differences in
the upper limit may be attributable to variations in the
Specimen specimen collection procedures [15].
It is preferable that the sample be obtained in the
morning after an overnight fast. The sample should be Slight to moderate elevation of NMN (112 to 400
collected into an evacuated glass collection tube pg/mL) or MN (61 to 236 pg/mL) can be seen in false-
containing heparin (in any form) as an anticoagulant positive patients, and this maybe due to hypertension or
(e.g., a green-top Vacutainer tube). It is preferable, but medications. Additional biochemical tests can provide
not essential, to draw the sample without a tourniquet further information for firmly establishing a diagnosis.
[7]. Conclusive diagnosis of pheochromocytoma in most
patients can be achieved by coupling the clonidine
966
suppression test with measurements of plasma NMN the best test to diagnose pheochromocytoma. The
[11]. reasons include:
1. Plasma free metanephrines are produced
Interpretation continuously by metabolism of catecholamines
Testing for the catecholamines and their metabolites is within pheochromocytoma cells; this is in
important in the diagnosis of the cause and extent of contrast to the episodic secretion of
certain neurological and endocrinological disorders such catecholamines.
as orthostatic hypotension and pheochromocytoma. 2. Stress causes large increases in catecholamine
release, whereas plasma free metanephrines
Pheochromocytoma is an uncommon but important remain relatively unaffected.
cause of hypertension. Its secretory products 3. Urine markers such as HMMA and
catecholamines and their metabolic productsare metanephrines are different metabolites from
powerful pressor agents, and severe hypertension and free metanephrines and are produced in
stroke may be the sequelae if the condition is not different parts of the body by metabolic
identified and treated [16,17]. processes not directly related to the tumor [6].
There is substantial literature which suggests that many 4 The conjugated urine metanephrines are
cases of pheochromocytoma are not picked up during produced outside of tumor tissue, and HMMA
life [18]. (VMA) is produced predominantly in the liver.
To date, diagnostic strategies for pheochromocytoma The clonidine suppression test was first introduced to
have depended primarily on the measurement of distinguish patients with pheochromocytoma from those
catecholamines and their metabolites in a 24-hour timed with falsely positive results. Clonidine suppresses
urine collection. These collections are inconvenient for norepinephrine release by sympathetic nerves.
patients, and inaccurate collections are not uncommon. Therefore, a decrease in plasma norepinephrine after
There is disagreement as to the testing protocol that clonidine administration suggests the elevated result was
should be followed, because each urine catecholamine due to sympathetic activation, and a lack of decrease
fraction will occasionally be normal in a person with suggests a pheochromocytoma. Compelling evidence has
significant clinical disease [19]. The secretory products shown that the coupling of the clonidine suppression test
of pheochromocytoma tissue may also be produced to with measurements of plasma NM and M can give a
excess or accumulate in conditions such as stress, renal conclusive diagnosis of pheochromocytoma [25].
failure, and certain drug therapies [20]. This is
potentially a particular problem in the acute hospital Metanephrines Performance Goals
setting with seriously ill, stressed patients [3,21]. For M measured in plasma, intra- and interassay CVs of
7.0% and 12%, respectively, have been reported at a
Plasma catecholamines have been advocated as a concentration of 153 pmol/L measured by HPLC. For
diagnostic tool, but this has not proved popular, firstly NM at 671 pmol/L measured in plasma by HPLC, intra-
because of false negatives resulting from overlap and interassay CVs of 4.7% and 12%, respectively, have
between some pheochromocytomas and the upper limit been reported [10]. The HPLC method for the analysis
of the normal range, and secondly because of frequent of plasma free NM and M has been found to be highly
false-positive results due to sympathoadrenal precise in the low normal and high abnormal values
overactivity in other pathophysiological states and relevant for the screening and diagnosis of
inadequately prepared patients [11,22]. pheochromocytoma.
Eisenhofer has presented compelling evidence that References

plasma metanephrines may be superior to both urine 1 Raber W, Raffesberg W, Bischof M, Scheuba
catecholamine fractions and plasma catecholamines C, Niederle B, Gasic S et al. Diagnostic efficacy
because of tumor-specific synthesis of metanephrines of unconjugated plasma metanephrines for
[5,23]. These theoretical grounds appear to be borne out detection of pheochromocytoma. Arch Intern
by clinical application [1,11,24]. Therefore, a negative Med 2000;160:2957-2963.
test for plasma free metanephrines means that a 2 Eisenhofer G. Biochemical diagnosis of
pheochromocytoma is highly unlikely, so no other tests pheochromocytoma: is it time to switch to
are necessary [6]. plasma free metanephrines? J Clin Endocrinol
Metab 2003;88:550-552.
Plasma free metanephrines provide a very good initial 3 Goldstein DS, Eisenhofer G, Kopin IJ. Sources
diagnostic test for exclusion of pheochromocytoma. and significance of plasma levels of catechols
Interindividual variations of plasma free metanephrines and their metabolites in humans. J Pharm Exp
are lower, and plasma half-lives are longer than those of Ther 2003;305:800-811.
catecholamines [1]. 4 Eisenhofer G, Walther MM, Huynh T, Sheng-
Tinget Li, Bornstein SR, Vortmeyer A et al.
Comparison of biochemical tests or combinations of Pheochromocytomas in von Hippel-Lindau
tests has shown that plasma free metanephrines provide syndrome and multiple endocrine neoplasia
type 2 display distinct biochemical and clinical
967
phenotypes. J Clin Endocrinol Metab 15 Grouzmann E, Fathi M, Gillet M, de Torrente

2001;86:1999-2008. A, Cavadas C, Brunner H et al. Disappearance
5 Eisenhofer G, Keiser H, Friberg P, Mezey E, rate of catecholamines, total metanephrines, and
Huynh TT, Hiremagalur B et al. Plasma neuropeptide Y from the plasma of patients
metanephrines are markers of after resection of pheochromocytoma. Clin
pheochromocytoma produced by catechol-O- Chem 2001;47:1075-1082.
methyltransferase within tumours. J Clin 16 Lenders JWM, Eisenhofer, Mannelli M, Pacak
Endocrinol Metab 1998;83:2175-2185. K. Phaeochromocytoma. Lancet 2005;366:665-
6 Eisenhofer G. Free or total metanephrines for 675.
diagnosis of phaeochromocytoma: what is the 17 Eisenhofer G, Goldstein DS, Sullivan P, Csako
difference? Clin Chem 2001;47:988-989. G, Brouwers FM, Lai EW et al. Biochemical
7 Lagerstedt SA, OKane DJ, Singh RJ. and clinical manifestations of dopamine-
Measurement of plasma free metanephrine and producing paragangliomas: utility of plasma
normetanephrine by liquid chromatography- methoxytyramine. J Clin Endocrinol Metab
tandem mass spectrometry for diagnosis of 2005;90:2068-2075.
phaeochromocytoma. Clin Chem 2004;50:603- 18 McNeil AR, Blok BH, Koelmeyer TD, Burke
611. MP, Hilton JM. Phaeochromocytoma
8 Willemsen JJ, Sweep CJG, Lenders JWM, Alec discovered during coronial autopsies in Sydney,
Ross H. Stability of plasma free metanephrines Melbourne, and Auckland. Aust N Z J Med
during collection and storage as assessed by an 2000;30:648-652.
optimized HPLC method with electrochemical 19 Lenders JWM, Pacal K, Walther MM, Lineham
detection. Clin Chem 2003;49:1951-1953. WM, Mannelli M, Friberg P et al. Biochemical
9 Roden M, Raffesberg W, Raber W, Niederle B, diagnosis of pheochromocytoma. JAMA
Waldhausl W, Gasic SL. Quantification of 2002;287:1427-1434.
unconjugated metanephrines in human plasma 20 Eisenhofer G, Huysmans F, Pacak K,
without interference by acetaminophen. Clin MacClellan MW, Sweep F, Lenders JWM.
Chem 2001;47:1061-1067. Plasma metanephrines in renal failure. Kidney
10 Lenders JW, Eisenhofer G, Armando I, Keiser Int 2005;67:668-677.
HR, Goldstein DS, Kopin IJ. Determination of 21 Raber W, Raffesberg W, Waldhausl W, Gasic
metanephrines in plasma by liquid S, Roden M. Exercise induces excessive
chromatography with electrochemical detection. normetanephrine responses in hypertensive
Clin Chem 1993;39:97-103. diabetic patients. Eur J Clin Invest
11 Eisenhofer G, Goldstein DS, Walther MM, 2003;33:480-487.
Friberg P, Lenders JWM, Keiser HR. 22 Pacak K, Ilias I, Adams KT, Eisenhofer G.
Biochemical diagnosis of pheochromocytoma: Biochemical diagnosis, localization and
how to distinguish true- from false-positive test management of pheochromocytoma: focus on
results. J Clin Endocrinol Metab 2003;88:2656- multiple endocrine neoplasia type 2 in relation
2666. to other hereditary and sporadic forms of the
12 Lenders JWM, Pacak K, Eisenhofer G. New tumour. J Int Med 2005;257:60-68.
advances in the biochemical diagnosis of 23 Eisenhofer G, Rundquist B, Aneman A, Froberg
pheochromocytoma, moving beyond P, Dakak N, Kopin IJ et al. Regional release
catecholamines. Ann NY Acad Sci and removal of catecholamines and
2002;970:27-40. extraneuronal metabolism to metanephrines. J
13 Heider EC, Davis BG, Frank EL. Clin Endocrinol Metab 1995;80:3009-3017.
Nonparametric determination of reference 24 Eisenhofer G, Lenders JWM, Goldstein DS,
intervals for plasma metanephrine and Mannelli M, Csako G, McClellan MW et al.
normetanephrine. Clin Chem 2004;50:2381- Pheochromocytoma catecholamine phenotypes
2384. and prediction of tumor size and location by use
14 Sawka AM, Jaeschke, Singh RJ, Young WF. A of plasma free metanephrines. Clin Chem
Comparison of biochemical tests for 2005;51:735-744.
pheochromocytoma: measurement of 25 Goldstein DS, Eisenhofer G, Flynn JA, Wand
fractionated plasma metanephrines compared G, Pacak K. Diagnosis and localization of
with the combination of 24-hour urinary pheochromocytoma. Hypertension
metanephrines and catecholamines. J Clin 2004;43:907-910.
Endocrinol Metab 2003;88:553-558.
968
Methods of Analysis of Plasma Free Metanephrines

Method 1: High-performance liquid chromatography mass spectrometry
Principle of analysis: Metanephrines are extracted off-line using solid-phase extraction followed by
on-line sample cleanup.
Comments: Expensive capital equipment; rare; good sensitivity and specificity; fairly labor intensive
Method 2: High-performance liquid chromatography
Principle of analysis: Free metanephrines are purified by mixed-mode ion exchange solid-phase
extraction chromatography. Separation into NM and M fractions by reversed-phase HPLC with
detection by electrochemical detectors.
Comments: Common; excellent sensitivity and very good specificity; reasonably rapid; preferred
method
Method 3: Immunoassay (RIA or ELISA)
Principle of analysis: Antibodies produced against N-acetylated derivatives of metanephrine or
normetanephrine by coupling to bovine serum albumin as hapten
Comments: Requires overnight incubation; lack of internal standard
phosphoric acid. Bring volume to 1000 mL with

Procedure: Plasma Free Metanephrines by deionized water.
High-Performance Liquid Chromatography
(HPLC) 2. 10 mM Ammonium Phosphate, pH 8.5
Dissolve 2.3 g monobasic ammonium phosphate
Brett C. McWhinney (MW = 115.03) in 800 mL deionized water.
Adjust the pH to 8.5 with 5% ammonium
Principle hydroxide (5 mL ammonium hydroxide in 95 mL
The plasma free metanephrines are extracted from deionized water). Bring volume to 1000 mL with
plasma and their concentration determined by deionized water, and add 0.1 g EDTA. Store
reverse-phase isocratic high-performance liquid refrigerated for up to 6 months. This is a 20 mM
chromatography coupled with electrochemical ammonium phosphate solution. Prepare 10 mM
detection. solution by mixing 100 mL deionized water with
100 mL of the 20 mM solution.
The extraction procedure isolates and concentrates 3. 5 % Ammonium Hydroxide/95%
the low picomolar concentrations of free Methanol Solution
metanephrines in plasma into a purified low- Add 5 mL concentrated ammonium hydroxide
volume form appropriate for injection onto the solution and 95 mL HPLC-grade methanol.
HPLC apparatus. Extractions are performed by Prepare this solution fresh immediately before
reverse-phase ion-exchange chromatography using extraction. Sufficient for 24 samples. Store in a
silica solid-phase octyl + benzyl sulfonic acid flask and cover to minimize contact with air
cation-exchange columns. Hydrophobic/cationic while in use. Expiry: 3 hours.
species in samples of plasma are retained on the 5. 0.1 M Acetic Acid
column, while the bulk of the sample is eluted to Dilute 6 mL of glacial acid with deionized water
waste. Subsequent washings of the columns to a total volume of 1 L. Store at room
removes compounds that were only retained by temperature for up to 1 year.
either hydrophobic or ion-exchange interaction. 6. Mobile Phase
(a) Mobile Phase Stock Solution
The metanephrines are eluted and concentrated Add 78 g of sodium dihydrogen orthophosphate
using nitrogen evaporation at 80C. The (NaH2PO4.2H2O), 1.28 g of 1-octane sulfonic acid,
metanephrines, unlike catecholamines, are stable in and 100 mg EDTA to a clean 1-L flask. Make up
the highly basic elution reagent. The residue is to 1 L with deionized water, and stir thoroughly to
resuspended in a small volume of mobile phase and dissolve.
injected onto the HPLC system. Store at 4C for up to 6 months.
(b) Mobile Phase for HPLC-EC Assay of
Metanephrines
REAGENTS AND EQUIPMENT Add 200 mL of mobile-phase stock solution and
Reagents 150 mL of HPLC-grade acetonitrile to a clean 2-
L flask. Make up to 1950 mL with deionized
1. 100 mM Sodium Phosphate, pH 6.0 water, and adjust the pH with HPLC grade
Dissolve 14.2 g disodium phosphate in 800 mL phosphoric acid, with stirring, to pH 3.25. Make
deionized water. Adjust the pH to 6.0 with up to a final volume of 2 L with deionized water,
and filter through a 0.45 m type HV (Millipore)
969
filter under vacuum. Decant to a clean bottle, and

pump through HPLC system. (D) External Metanephrine Standard
Store at 4C for up to 2 months Working Standard For HPLC-EC
(depending on background current). Calibration
The standard mix (2400 pmol/L) is
NOTE: During equilibration of a new column with prepared by mixing together 300 L of the
mobile phase, always run mobile phase at 1 Calibration Standard (Plasma
mL/min for at least 12 hours while recycling and Metanephrine STD 80nmol/L), 500 L of
before the injection of standards. Adjust pH and Plasma Metanephrine Internal Standard
acetonitrile concentration as required. (50 nmol/L HMBA in acetic acid) and
400 L of Mobile Phase. Inject 100 L
7. Standard Solutions on HPLC column.
(A) Plasma Metadrenaline Internal NOTE: When calibrating HPLC-EC,
Standard: 4-Hydroxy-3- always compare peak heights of the new
Methoxybenzylamine (HMBA) working standard solution with those of
that used for the previous assay. If peaks
Plasma Metanephrine Stock Internal areas differ by more than 5%, make up
Standard (500 M HMBA) new working standard solution, and use
Weigh out 0.10 g of 4-hydroxy-3- solution that compares most favorably
methoxybenzylamine. Dissolve in 100 mL of 0.05 with the other or the standard from the
M HCl. Store frozen for up to 6 months. previous HPLC run. Note also, however,
that new detector cells or mobile-phase
Plasma Metanephrine Intermediate Internal conditions or extended time between
Standard (10 M HMBA) assays may give rise to different detector
Dilute plasma metanephrine stock internal responses to injection of identical
standard 1:50 with 0.05 M HCl. Store at 4C. standards.
Plasma Metanephrine Internal Standard Equipment

(40 nM HMBA)
Dilute plasma metanephrine intermediate 1. LABORATORY APPARATUS
internal standard 1:250 with 0.2 M acetic acid.
Store at 4C. a. Solid-phase extraction (SPE)
cartridges
(B) Aqueous Metanephrine Standards Clean screen columns 10 cc/200 mg
Stock Standards (500 M) UCT Catalog no. ZSDAU020 (50
(a) Metanephrine Stock Standard. columns/box)
Dissolve 116.8 mg of metanephrine in 1000 b. Waters 24-port extraction
mL 0.05 M HCl. Store at 20C. manifold attached to a vacuum pump
(b) Normetanephrine Stock
Standard. 2. HPLC SYSTEM
Dissolve 109.8 mg of
Normetanephrine in 1000 mL 0.05 M a. Pump
Waters pump, Model 616.
HCl. Store at 20C.
Flow rate usually 1.0 mL/minute
b. Autosampler
Plasma Metanephrine Intermediate
Waters 717 plus autosampler
Standards (10 M)
with heater/cooler module
Dilute Metanephrine and
attached
Normetanephrine stock standards
c. Column
1:50 with 0.1 M acetic acid. Store at
Phenomenex stainless steel
4C. Luna C18 (2) columns, 4.6 mm
150 mm with 5-m packing.
(C) Calibration Standard (Plasma Metanephrine
Part No. 00F-4252-E0.
Standard 80 nmol/L) Immediately preceding the
(a) Combine Metanephrine and Normetanephrine analytical column is an
into one calibration standard (80 nmol/L). integrated Phenomenex
Add 2 mL of each intermediate standard, Security Guard holder
and make up to 250 mL with 0.1 M acetic containing a C18 insert. Part
acid No. AJ0-4287.
(b) Store in 2.5-mL aliquots at 20C. d. Data Manager
970
Waters Empower Build No. 3. Activate columns with 1 mL 100

2154, Chromatography mM sodium phosphate buffer, pH
Manager 6.0, in methanol under gravity. A
e. Detector light vacuum (1 to 5 mmHg) may be
ESA Coulochem II 5200A used to start the flow, but once it has
Coulometric Detector (ECD) commenced, release any vacuum
with conditioning and present. Again, do not allow columns
analytical cells dedicated to to dry out.
Plasma Metanephrine Assay.
Settings as shown in table Sample Extraction Procedure
below:
1. Remove standards, QC, and patient
ECD SETTINGS plasma samples from 20C freezer,
Guard Cell and thaw at room temperature. Spin
Cell Potential +450 mV thawed plasma at 3000 g for 10 min
Channel 1 to pellet fibrin and other particulate
Cell Potential +10 mV matter.
Full Scale Range 100 nA 2. To the standard tube, add 30 L of
Filter 5 sec Plasma Metanephrine Standard (80
nmol/L) and 2 mL of water.
Signal Output Voltage 1V
Baseline Offset 0% 3. To the appropriate 10-mL conical
Channel 2 plastic tubes, add 1 mL of QC and 1
mL of patient plasma (or 1 mL of
Cell Potential -350 mV
water for the blank).
Full Scale Range 100 nA
Filter 5 sec
4. Add 1 mL of deionized water to all
tubes, followed by 50 L of Plasma
Signal Output Voltage 1V
Metanephrine Internal Standard (50
Baseline Offset 0% nM HMBA).
N.B. These values must be determined for
5. Cap and vortex the tubes for 15 sec,
and centrifuge at 3000 rpm for
each cell and will vary depending on the age
10min. (It is essential that samples
of cells and mobile phase.
are free of particulate matter for their
efficient passage through extraction
columns).
Standards, controls, and specimens are 6. Add the samples to their appropriate
treated similarly. The extracts are stable extraction column. Pass the whole
when stored at 4C for up to 4 days. solution slowly through the column
under gravity. (Elution time should
Once step 1 of the Column Preparation be between 5 and 10 min.)
Procedure has started, begin step 1 of the 7. Wash columns with 3 mL water.
Sample Extraction Procedure so that the 8. Wash columns with 1 mL 100 mM
samples are ready for extraction when acetic acid under gravity. When flow
the last step of the Column Preparation is finished, dry the column out by
procedure is finished. increasing the vacuum to 20 mmHg
for 3 minutes.
Column Preparation Procedure
9. Wash columns with 2 mL 10 mM
1. Wash extraction columns with 3 mL ammonium phosphate buffer pH 8.5
of methanol under gravity. A light under gravity. A light vacuum (1 to 5
vacuum (1 to 5 mmHg) may be used mmHg) may be used to start the
to start the flow and draw through flow, but once it has commenced,
approximately 1 mL, then release release any vacuum present.
any vacuum present. Do not allow 10. Wash columns with 2 mL deionized
columns to dry out. water under gravity. A light vacuum
2. Wash columns with 2 mL deionized (1 to 5 mmHg) may be used to start
water under gravity. A light vacuum the flow, but once it has commenced,
(1 to 5 mmHg) may be used to start release any vacuum present. Elute all
the flow, but once it has commenced, water.
release any vacuum present. Again, 11. Wash columns with 3.0 mL of 100%
do not allow columns to dry out. methanol. A light vacuum (1 to 5
971
mmHg) may be used to start the Recycle mobile phase: yes

flow, but once it has commenced,
release any vacuum present. Allow Retention time for internal standard (IS)
methanol to drip through the must be 17 to 19 minutes. Total run time is
columns. Apply full vacuum to ~40 minutes
remove any residual methanol.
12. Place labeled glass culture tubes (13 CALCULATIONS
100 mm) into a rack, and position The Empower Chromatography Data
under the needles. Manager measures the peak heights,
13. Elute samples with 3 mL of 5% calculates the ratios of the analytes of
NH4OH/methanol under gravity. A interest to the IS, and constructs a
light vacuum (1 to 5 mmHg) may be standard curve of this ratio versus
used to start the flow, but once it has concentration in pmol/L. The
commenced, release any vacuum concentrations of the unknowns are then
present. Finally, use a vacuum to determined from this standard curve,
elute all the NH4OH/methanol. using their calculated peak height ratios.
14. Take glass tubes to the heating block Check the internal standard peak heights.
set at 80C, and dry the sample down Variations of up to 20 % are to be
completely using a strong stream of expected, but low recoveries of the
nitrogen (approximately 20 min). internal standard suggest some problem
15. Dissolve the residue in 120 L of with the sample clean-up, and the sample
mobile phase, vortex for 30 sec, and should be re-assayed.
centrifuge the tubes at 2000 rpm for
5 min. For manually calculating the concentration
16. Transfer the samples to limited- of an analyte, use the following formula:
volume vials, and inject 100 L onto PH UNK PH ISSTD
the HPLC. Concentration pmoles / L Std CONC
PH STD PH IS UNK
After extraction, samples are stable at

4C for at least 4 days. PH=peak height of UNK or STD
UNK = analyte of interest in unknown
System Settings Summary STD = analyte of interest in standard
IS = internal standard
Injection volume: 100 L
Flow rate: 1.0 mL/min
Run time: 40 min
972
450.00
IS
400.00
350.00
300.00
250.00 NM
200.00 M
mV 150.00
100.00
50.00
0.00
-50.00
-100.00
-150.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00

Minutes
Figure 1: Ion-pair high-performance liquid chromatography (HPLC) analysis of a normal plasma sample by method
described in text: electrochemical detector (ECD) response versus elution time. Normetanephrine (NM) 15 minutes,
metanephrine (M) 20.5 minutes, and internal standard (IS) 28 minutes.
973
A
1000.00
900.00
800.00
700.00
600.00
mV 500.00
NM
400.00 M IS
300.00
200.00
100.00
0.00
-100.00
-200.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00
Minutes
Figure 2: Ion-pair high-performance liquid chromatography (HPLC) analysis of a normal plasma sample spiked with 100
mg/L acetaminophen by method described in text: electrochemical detector (ECD) response versus elution time.
Acetaminophen (A) 12minutes, normetanephrine (NM) 14 minutes, metanephrine (M) 18 minutes and internal standard
(IS) 25 minutes.
974
Porphobilinogen Screening and Quantitation
Porphobilinogen (PBG) Screening and Quantitation

Enrico Rossi
Name: Porphobilinogen, PBG, 5-(aminomethyl)-4-(carboxymethyl)-1H-

pyrrole-3-propionic acid
Clinical significance: see Interpretation page 4, line 6, left column.
Merck Index: 7469
Chemical class: Pyrrole
Structure:
Refer to Chapter 39, Iron, Porphyrin, and Bilirubin Metabolism, in the 5th edition of Clinical Chemistry: Theory, Analysis,
Correlation.
i extracted into the chloroform layer. Color that remains in

Porphobilinogen (PBG) is a water-soluble monopyrrole the supernatant aqueous phase is usually caused by PBG;
precursor of heme found in high concentrations in the this can be confirmed following further extraction with
urine of persons suffering from one of the acute butanol [4].
porphyrias. PBG levels are highest during the acute
phases of acute intermittent porphyria (AIP), hereditary Problems With the Watson-Schwartz Screening Test
coproporphyria, or variegate porphyria. The Watson-Schwartz screening test has several
shortcomings and is no longer recommended. The chief
Production of this abnormal metabolite was the most drawback is lack of adequate sensitivity, a problem
notable feature of the first recorded case of drug-induced especially evident in dark-colored urine samples [5].
acute porphyria. It occurred in 1889 in a woman who, Pigments usually present in urine, such as urobilinogen,
after ingestion of the drug sulfonmethane (Sulfonal), indole, and indican, not only interfere with the color
excreted red urine and died [1]. In 1931, Sachs reaction, they can also lead to spurious color
discovered that although the metabolite produced a color development, resulting in lack of specificity [6]. Even a
reaction with Ehrlichs aldehyde (or diazo) reagent, it PBG concentration of 50 to 100 mol/L (11 to 22 mg/L),
was different from urobilinogens [2]. well above the reference range of < 9 mol/L (<2 mg/L),
may not be detected [7].
For many years, the standard test for clinical detection of
PBG was the test introduced by Watson and Schwartz in Samples from acute porphyria patients with current
1941 [3]. The method is qualitative and indicates the symptoms usually have PBG concentrations >100
presence or absence of PBG. The basis of the procedure mol/L (<22 mg/L), but there is a risk that a patient with
is the reaction between PBG and Ehrlichs aldehyde porphyria may be missed. It is well known that diagnosis
reagent (p-dimethylaminobenzaldehyde in strong acid) should not be made on clinical grounds alone, and the
to form a red condensation compound. Sodium acetate is laboratory carries the responsibility for successful
added to the reagent to increase the pH. This maximizes detection of PBG.
the color produced by urobilinogen and renders it
preferentially soluble in chloroform. Thus if color is A further difficulty is an apparent decrease in sensitivity
produced during the first step of the reaction, a of the Watson-Schwartz screening test with aging of the
chloroform extraction is carried out. The colored urine sample. This has been observed even in urine
products of urobilinogen and other compounds are then samples tested on the same day of collection. The
reasons are not well described, but it has been suggested
i that indole compounds which inhibit the condensation
PBG Screening and Quantitation reaction with Ehrlichs aldehyde reagent form on
Previous and current authors of this method: standing [8].
Methods edition: Michael D.D. McNeely, Robert S. A simpler method is the Hoesch test, which uses
Franco, Tony Kraft, and William E. Ehrlichs aldehyde reagent without sodium acetate. No
Schreiber extractions are required. However, this test is also not
Second edition: Not updated recommended; it has the same shortcomings as the
Third edition: Not updated Watson-Schwartz test. Positive results have been
Fourth edition: Michael D.D. McNeely observed with indoles, indole-3-acetic acid, alpha-
Fifth edition: Enrico Rossi
975
methyldopa, phenazopyridine hydrochloride, and end- eluate is reacted with Ehrlichs reagent. At very high
stage alcoholic malnutrition [9]. A popular urinary concentrations of PBG, diluted urine can be
misconception is that strip tests for urinary urobilinogen analyzed directly by use of Ehrlichs reagent, with
that employ Ehrlichs aldehyde reagent will routinely minimal interference from interfering compounds.
change color in the presence of PBG. Thus it has been
suggested that a negative strip test will rule out acute A more rapid procedure [14] employs batch adsorption
intermittent porphyria (AIP). This is inappropriate, since of PBG onto Dowex 2 resin rather than employing
PBG will not reliably change the test strips color. column chromatography (Table 1, Method 1b). By
adjustment of the diluted urine pH to about 10 to 11,
Recommended Porphobilinogen Screening Method maximal adsorption of PBG to the resin is achieved.
More recently, a sensitive and specific screening test has After several water washings of the resin and
been described which uses a syringe prepacked with an centrifugation, the PBG is quantitatively eluted by
ion-exchange resin. As originally described by Buttery et multiple extractions with 1 N acetic acid.
al., [10] the test used anion-exchange resin in test tubes;
however, it is now available commercially with the resin High-performance liquid chromatography (HPLC) can
supplied in more convenient, disposable plastic syringes also be used to separate and quantitate PBG (Table 1,
and has been formally evaluated (Thermo Electron Method 2) [15]. The difficulty that arises is maintaining
Corporation, Louisville, CO) [11]. The urine sample is HPLC equipment and reagents in readiness for a
adsorbed onto the resin, washed and eluted through a relatively infrequent test which is usually not batched,
series of procedures involving the addition and removal because the result is required urgently. Either a
of a 5-micron filter to the syringe and a final color combination of a strong anion-exchange stationary-phase
formation with Ehrlichs reagent. Visual assessment of column and a strong cation-exchange stationary-phase
color formation is accomplished with use of a color chart column or a single C18 reversed-phase column is used to
or a dye color calibrator. separate PBG from -aminolevulinic acid. Analysis time
is under 15 min, and no extraction of urine is necessary
Quantitative Porphobilinogen Assays with analysis being performed directly on diluted urine.
The earliest quantitative assays for PBG depended on
reaction with Ehrlichs reagent (p- Reference and Preferred Methods
dimethylaminobenzaldehyde [DMAB]) to form a red The reference method for PBG in urine is stable-isotope
compound (Table 1, Method 1). The reaction has been dilution liquid chromatography-tandem mass
described as occurring in two phases [12]. In the first, spectrometry (LC-MS/MS), first reported in 2001 [16].
the rapidly occurring step, porphobilinogen (or any The procedure is briefly described as Method 3 in Table
pyrrole with an unsubstituted alpha hydrogen) reacts 1. It requires 1 mL of urine and uses 5-(aminoethyl)-4-
under strongly acidic conditions to form a red
(carboxymethyl)-1H-2,4-[13C]PBG (2.75 g) as the
condensation product whose absorbance maximum is at
internal standard. As expected from a reference method,
555 nm (PBG Quantitation: Figure 1 and Reaction 1). In
this technique shows greater analytical specificity and
the second, the slower step, the red compound reacts
improved clinical sensitivity and specificity compared to
with an additional pyrrole molecule to form a colorless
other available methods.
dipyrrylphenylmethane product (Porphobilinogen
Quantitation: Reaction 2).
For the first time, mild elevations (13 to 66 mol/L, 3-15
mg/L) of PBG could be reliably detected and the true
The rate and extent of each reaction is dependent on the
reference range for PBG established. The quantitative
reaction conditions. Using Ehrlichs reagent prepared
Mauzerall and Granick method obtained a PBG
with HCl, there is an increase in the rate of the second
reference range of < 15 mol/day (<3 mg/day), whereas
reaction, especially at high concentrations of pyrrole,
LC-MS/MS gives a much lower and probably more
resulting in decreased linearity of the reaction [13].
correct range of < 2.2 mol/day (<0.5 mg/day).
Replacing the HCl by perchloric acid (HClO4, 2-4 N),
Mauzerall and Granick increased the linearity of the Porphobilinogen Screening Method
reaction and stabilized the red condensation product so The preferred screening method uses an anion-exchange
that the molar absorptivity of the reaction product was resin, as first described by Mauzerall and Granick [13]
almost twice that of the reaction product formed when for their quantitative technique. Adaptation of the
employing HCl [13]. quantitative resin method into a rapid method suitable
for screening took place in stages. The originally
The methods employing Ehrlichs reagent are described method required the addition of resin to a test
distinguished by the procedure used to separate PBG tube, and the subsequent washing and elution steps
from the interfering compounds that are present in urine caused delays because of the need for several
(Table 1). centrifugation steps to remove supernatant fluids from
the mobile resin bed in the bottom of the tube [10].
In the quantitative method of Mauzerall and Granick Blake et al. [17] described several useful modifications:
(Table 1, Method 1a), PBG is adsorbed onto Dowex 2 The anion-exchange resin is purchased from
resin, and the resin is washed to remove other urinary Bio-Rad in the required acetate form and
constituents. The PBG is then eluted from the column, immobilized in a syringe.
with essentially quantitative recovery [13], and the
976
A range of color standards based on methyl red Specimen

solutions allow the result to be reported semi- For screening, a freshly voided urine specimen collected
quantitatively as not detected, positive +, during or immediately after an attack of acute abdominal
positive ++ and positive +++. pain is recommended. The assay should be conducted
Using a syringe with an attached filter within several hours to prevent the degeneration of PBG
decreases the time required to approximately 10 or the formation of an inhibitor to the reaction with
minutes. Ehrlichs reagent [19]. After several days at room
temperature, and especially if exposed to light, colorless
These modifications have been incorporated into a PBG polymerizes to form porphyrins and other pigments
commercial urine PBG kit available from Thermo that darken the urine. At alkaline pH (pH 9), PBG is
Electron Corporation, Louisville, Colorado stable at 4C for 1 to 2 weeks. If the assay is to be
(www.thermo.com/clinicalchemistry). This test is more delayed for a longer time, an aliquot of a well-mixed
sensitive and specific than the Watson-Schwartz method, urine collection may be frozen at 20C for several
and an evaluation of the commercial kit has been months [20].
published [11].
Twenty-four-hour urine specimens for quantitation of
The College of American Pathologists (CAP) external PBG are not recommended. They delay diagnosis and
quality control scheme does not include porphyrins or increase the risk of losses during the collection period. A
PBG. However, the scheme administered by the Royal random urine reference range obtained with the
College of Pathologists of Australasia (RCPA) in Mauzerall and Granick quantitative method of < 1.5
Australia has circulated urine samples for PBG mol/mmol creatinine has been determined [17].
screening. Over a 4-year period from 2003 to 2006, the
proportion of laboratories persisting with the outmoded Interferences
Watson-Schwartz or Hoesch screening methods declined Inappropriate specimen preservation and storage is the
from 23% to 10%. During the same period, enrolled most important aspect affecting accurate measurement of
laboratories using the Watson-Schwartz or Hoesch PBG.
screening methods obtained a sensitivity of 92%
compared to 98% for those using the recommended Stored acidified and exposed to light at room
anion-exchange resin screening method. The specificity temperature, urines containing PBG would be expected
for all methods was 100%. to rapidly lose it, usually leading to significant darkening
of the urine sample and interference with both the
Porphobilinogen Quantitation Method Watson-Schwartz and Hoesch screening methods.
The preferred method for quantitation of PBG is one of Pigments normally present in urine, such as
those that utilizes an ion-exchange resin for isolation of urobilinogen, indole, and indican, are also thought to
PBG with subsequent spectrophotometric quantitation cause increased interference to these screening methods
using DMAB. The Mauzerall and Granick method [13] with increasing storage time [8]. The preferred anion-
has been widely used in clinical laboratories and has exchange resin screening method has a decisive
stood the test of time to become a literature citation advantage in removing these interfering pigments prior
classic. The use of column chromatography to isolate to the color reaction [11].
PBG should not be a problem in most clinical
laboratories. Porphobilinogen Reference Interval
The reference interval for a random urine is < 2 mg/L
Routine analysis of urines should employ either the (<9 mol/L), but a creatinine-corrected result is
column or batch chromatographic method to isolate PBG preferred. The recommended reference interval of < 1.5
from interfering compounds such as urea and ascorbic mol/mmol creatinine is based on the 95th centile for
acid [18]. If a sample is known to contain very high 290 random urines [17].
concentrations of PBG (tenfold higher than normal),
direct analysis of diluted urine can be performed The older literature quotes reference intervals based on
satisfactorily; results will be only 5% too high [14]. Both 24-hour urine collections, which are not recommended,
methods report quantitative recovery of PBG from the since they delay diagnosis and increase the risk of losses.
ion-exchange resin and an assay sensitivity of 0.44 mol Moore and Labb [14] reported a reference interval of
(0.1 mg) of PBG (4.4 mol/mL or 1 mg/mL of urine). 5.7 3.5 mol/day (1.3 0.8 mg/day). Determination of
reference intervals with the LC-MS/MS method gives a
The batch method presented below also incorporates lower PBG excretion of the order of 0 to 2.2 mol/day (0
measurement of absorbance at 2 wavelengths to allow to 0.5 mg/day). Reference intervals obtained by the
estimation of the contribution of interfering compounds established colorimetric methods [13] probably
that are not removed by the ion-exchange procedure overestimated PBG excretion because of incomplete
[14]. Using the batch method, it is possible to process removal of interfering substances from the sample before
several samples simultaneously in about 1 h. This is complexation with Ehrlich reagent [16]. PBG excretion
particularly useful when several members of a family are in children is equivalent to that of adults when the values
examined at the same time. are corrected for body weight.
977
Interpretation methods for urinary porphobilinogen. Ann Clin

Urinary PBG concentrations are greatly increased in Biochem. 1990;27:165-6.
acute attacks of acute intermittent porphyria (AIP), 7 Buttery JE, Carrera AM, Pannall PR. Reliability
hereditary coproporphyria, and variegate porphyria. In of the porphobilinogen screening assay.
AIP, levels up to 440 mol/L (100 mg/L) of PBG may Pathology 1990;22:197-8.
be found. AIP, the most commonly seen acute porphyria 8 Watson CJ, Taddeini L, Bossenmaier I. Present
in the United States and northern Europe, most Status of the Ehrlich Aldehyde Reaction For
frequently is accompanied by acute abdominal pain, Urinary Porphobilinogen. JAMA. 1964;190:501-
constipation, and vomiting. AIP is almost exclusively 4.
described only after puberty, most likely resulting from 9 Pierach CA, Cardinal R, Bossenmaier I, Watson
the increase in sex hormones, which are known to CJ. Comparison of the Hoesch and the Watson-
exacerbate symptoms of porphyria. Acute attacks Schwartz tests for urinary porphobilinogen. Clin
become progressively less likely with advancing age, Chem 1977;23:1666-1668.
especially after menopause in women. Any positive 10 Buttery JE, Chamberlain BR, Beng CG. A
screening test result should be followed up by a sensitive method of screening for urinary
quantitative PBG and urinary porphyrin assay. porphobilinogen. Clin Chem 1989;35:2311-2.
11 Deacon AC, Peters TJ. Identification of acute
During acute porphyric attacks, urinary PBG porphyria: evaluation of a commercial screening
concentrations will be very high (>133 mol/day [30 test for urinary porphobilinogen. Ann Clin
mg/day]) and can be detected by most qualitative Biochem 1998;35:726-732.
screening tests for PBG [9]. Between acute episodes, 12 Treibs A, Herrmann E. Pyrrole dyes IV: the
urinary PBG excretion for some patients will be too low Ehrlich reaction of pyrroles and indoles. Hoppe-
to detect with the screening procedure, even though Seylers Z Physiol Chem 1955;299:168-185.
urinary PBG is above normal. Other patients may have 13 Mauzerall D, Granick S. The occurrence and
completely normal PBG excretion between acute determination of delta-aminolevulinic acid and
episodes, although the more sensitive LC-MS method porphobilinogen in urine. J Biol Chem
often detects elevated PBG in cases where the screening 1956;219:435-446.
procedures return a negative result. 14 Moore DJ, Labb RF. A quantitative assay for
urinary porphobilinogen. Clin Chem
In acute intermittent porphyria, there is a decrease in the 1964;10:1105.
enzyme uroporphyrinogen I synthase (URO-S), also 15 Lim CK, Rideout JM, Samson DM.
known as hydroxymethylbilane synthase. This enzyme is Determination of delta-aminolaevulinic acid and
responsible for converting PBG to uroporphyrinogen I in porphobilinogen by high-performance liquid
the synthesis of heme. The assay of this enzyme allows chromatography. J Chromatogr 1979;185:605-
the diagnosis of acute intermittent porphyria to be made 611.
on a portion of the subset of patients with normal PBG 16 Ford RE, Magera MJ, Kloke KM, Chezick PA,
levels during the latent period of the disease, although Fauq A, McConnell JP. Quantitative
there is an area of overlap between the lower end of the measurement of porphobilinogen in urine by
normal range and the upper end of ranges for AIP stable-isotope dilution liquid chromatography-
patients [21]. The enzyme assay is most useful in family tandem mass spectrometry. Clin. Chem
studies, where the enzyme activity of the proband is 2001;47:1627-1632.
known [22]. 17 Blake D, Poulos V and Rossi E. Diagnosis of
porphyria: recommended methods for peripheral
References laboratories. Clin Biochem Revs 1992;13[suppl
1 Magnussen CR. Acute intermittent porphyria: 1]:S1-24.
clinical and selected research aspects, Ann Int 18 Pounty FTG. Acute porphyria: some properties of
Med 1975;83:851-864. porphobilinogen. Biochem J 1945;39:446-451.
2 Sachs P. Ein Fall von akuter Porphyrie mit 19 Watson CJ, Bossenmaier I, Cardinal R. Acute
hochgradiger Muskelatrophie. Klin Wochenschr intermittent porphyria. JAMA 1961;175:1087-
1931;10:1123-1125. 1091.
3 Watson CJ, Schwartz S. Simple test for urinary 20 Bossenmaier I, Cardinal R. Stability of -
porphobilinogen. Proc Soc Exp Biol Med aminolevulinic acid and porphobilinogen in urine
1941;47:393-394. under varying conditions. Clin Chem
4 Schwartz S, Keprios M, Schmid R. Experimental 1968;14:610-614.
porphyria: type produced by lead, 21 Bottomley SS, Bonkowsky HL, Kreimer-
phenylhydrazine and light. Proc Soc Exp Biol Birnbaum M. The diagnosis of acute intermittent
Med 1952;79:463-468. porphyria: usefulness and limitations of the
5 Schreiber WE, Jamani A, Pudek MR. Screening erythrocyte uroporphyrinogen I synthase assay.
tests for porphobilinogen are insensitive. The Am J Clin Pathol 1981;76:133-139.
problem and its solution. Am J Clin Pathol 22 Astrup EG. Family studies on the activity of
1989;92:644-9. uroporphyrinogen I synthase in diagnosis of acute
6 Buttery JE, Carrera AM, Pannall PR. Analytical intermittent porphyria. Clin Sci Mol Med
sensitivity and specificity of two screening 1978;54:251-256.
978
Tables
Table 1: Methods for Urinary Porphobilinogen (PBG) Quantitation
Method 1: Ion-exchange
a. Column
b. Batch
Principle of analysis: Partially purified PBG quantitated after reaction with Ehrlichs reagent
a. PBG separated from interfering compounds by column chromatography
b. PBG separated from interfering compounds by batch ion-exchange chromatography
Comments:
a. Common; standard method
b. Common; may be more suitable for clinical laboratory

Principle of analysis: PBG separated by HPLC and quantitated spectrophotometrically at 240 nm
Comments: Rare; requires special equipment, but should be very specific
Method 3: Liquid chromatographytandem mass spectrometry

Principle of analysis: Following solid-phase extraction of PBG, LC-MS/MS analysis performed. PBG and
internal standard monitored through their own precursor and product-ion settings
Comments: Method exhibits greater analytical specificity and improved clinical sensitivity and specificity as
compared with other methods. Requires solid-phase extraction of urine to remove urine components that suppress
the ion signal.
Reaction Conditions for Porphobilinogen (PBG) Quantitation by Reaction with Ehrlichs Reagent
Sample: Ion-exchange chromatography extract of urine
Temperature: Ambient temperature
Sample volume: 1 mL urine; ion-exchange chromatography column eluate, 2 mL
Fraction of sample volume: 0.5 (Ehrlichs reaction)
Final concentration of reagents:
HCl: 1.46 mol/L
Acetic acid: 6.6 mol/L
p-Dimethylaminobenzaldehyde (DMBA): 67 mmol/L
Time of reaction: 6 to 8 min
Linearity: Up to at least 100 g/mL
Sensitivity: 0.1 g (1 g/mL of urine)
Figures
Figure 1: Porphobilinogen Quantitation.

Absorbance spectrum of red condensation product produced by reaction of PBG and Ehrlichs reagent. Dashed line,
Ehrlichs reagent versus water; dotted-dashed line, Ehrlichs reagent and PBG versus water; solid line, Ehrlichs reagent
and PBG versus Ehrlichs reagent.
979
Porphobilinogen Quantitation: Reaction 1
Porphobilinogen Quantitation: Reaction 2
Screening Procedure: Resin Assay for mL aliquots into 2 mL syringes and cap with a rubber
Porphobilinogen plug. Store ready for use.
3. Acetic acid (1 mol/L). To 17 volumes of
Principle distilled water, add 1 volume of glacial acetic acid.
This screening test is more sensitive and specific than 4. Ammonia solution. To 3 volumes of distilled
the Watson-Schwartz test it replaces [17]. The anion- water add 1 volume of concentrated (33%) ammonia
exchange resin is prepacked in a syringe. The urine solution.
sample is adsorbed onto the resin, washed and eluted 5. 5-Micron filters (Argyle, St Louis, MO).
through a series of procedures involving the addition and 6. Methyl red color standards. For the stock
removal of a 5-micron filter and a final color formation solution dissolve 33.3 mg of methyl red sodium salt in
with Ehrlichs reagent. Visual assessment of color 3.3 mL concentrated hydrochloric acid and 25 mL of
formation is accomplished with use of a dye color distilled water, then make up to 200 mL with distilled
calibrator for the in-house method given below or a color water. For the working solution, dilute the stock solution
chart for the commercial version available from Thermo with 0.1 mol/L hydrochloric acid as shown below.
Electron Corporation, Louisville, Colorado; Standard Dilution Absorbance
www.thermo.com/clinicalchemistry. Range
25 mol/L 1:200 0.10-0.15
Reagents 50 mol/L 1:100 0.25-0.30
1. Ehrlichs aldehyde reagent (18.8 mmol/L in 100 mol/L 1:50 0.50-0.60
7.2 M HCl). Combine 700 mg of reagent-grade p-
dimethylaminobenzaldehyde, 150 mL of concentrated 5. Control urine. A fresh, otherwise normal,
HCl, and 100 mL of deionized water. The solution urine is run. Commercially available urine controls
should be colorless or very light yellow. Brown or usually contain no detectable PBG and can be used as
brownish-red material should not be used. Store in negative controls. A positive control can be prepared by
amber bottle at room temperature. It will remain stable addition of PBG (Sigma Chemical Co., St. Louis, MO;
for months. Discard if discolored. no. P1134) to a final concentration of 75 mol/L, 16.97
2. Anion-exchange resin (AG1-X2 200-400 mesh, mg/L. The spiked pool should then be frozen at 20C in
acetate form BioRad, Richmond CA). Mix 100 g resin 2 mL aliquots. It is stable for up to 6 months.
with 100 mL distilled water. While mixing, draw up 2
980
Assay add approximately 900 mL of distilled water to dissolve

1. Check urine pH is > 6; if not, add one drop of salt. Add distilled water to mark, mix well, and store at
ammonia solution. room temperature. Stable for 6 months.
2. Remove the syringe cap and fit filter securely to 3. Dowex 2-X8 resin (200 to 400 mesh) acetate
the syringe. Expel the water in the syringe, form. The resin may be prepared in large batches (100 to
remove the filter, and draw up 1 mL of urine 200 g), since it is stable for at least several months.
using the graduations on the syringe. To help Suspend the resin in 4 volumes of sodium acetate
with mixing, introduce an air bubble into the solution to convert it to the acetate form. Mix well, and
syringe. Replace the filter, and mix the resin allow all but the very fine particles to settle. Decant and
and urine for 10 seconds. discard the supernatant. Wash the resin about eight times
3. Expel the urine to waste. Draw up 1 mL of with 4 volumes of distilled water. These washes can be
water and an air bubble and mix for 10 seconds. facilitated by brief centrifugation to pack the resin. A
4. Expel the water to waste. Draw up 1 mL of 1 stock resin suspension is prepared by the addition of 1
mol/L acetic acid and an air bubble and mix for volume of distilled water to 1 volume of washed, packed
10 seconds. resin. The resin will settle quickly and should be
5. Expel the acetic acid into a clear tube completely suspended before use.
containing 1 mL of Ehrlichs reagent and mix. 4. 1 M acetic acid (1 mol/L). Take 5.7 mL of
Wait 3 minutes and then compare the color of glacial acetic acid, and dilute with distilled water to 100
the solution to the methyl red solutions against mL in a volumetric flask. Mix thoroughly, and store at
a white background. room temperature. Stable for 6 to 12 months.
6. Report as follows: 5. Ammonium hydroxide, concentrated (28%
Not detected NH3).
PBG concentration <25 mol/L +
PBG concentration 25-50 mol/L ++ Assay
PBG concentration 50-100 mol/L +++ Equipment: Narrow band-width (10 nm)
PBG concentration >100 mol/L ++++ spectrophotometer, general-purpose centrifuge,
12 mL graduated centrifuge tube.
Notes 1. Place 2 mL of thoroughly resuspended resin
1. This resin-based method overcomes the major into a 12 mL centrifuge tube, and centrifuge
drawback of the Watson-Schwartz screening method, sufficiently to pack the resin (1 min at several
namely lack of adequate sensitivity, a problem especially hundred g). Remove the supernatant, and add 1
evident in dark-colored urine samples. Pigments usually mL of urine onto the packed resin followed by
present in urine such as urobilinogen, indole, and indican 0.1 mL of concentrated ammonium hydroxide.
not only interfere with the color reaction, they can also Resuspend thoroughly, centrifuge, and discard
lead to spurious color development, resulting in lack of supernatant. The PBG is bound to the resin.
specificity. 2. Wash the resin four times with 5 mL aliquots of
distilled water. The resin is packed after each
Procedure: Quantitative Porphobilinogen Assay wash by centrifugation. Remove the final
Using Ehrlichs Reaction [14] supernatant.
3. Add 2 mL of 1 M acetic acid to the packed
Principle resin, and resuspend thoroughly. Pack the resin
PBG Quantitation Table: Reaction conditions with by centrifugation, and save the supernatant,
Ehrlichs reagent. which contains PBG. Elute three more times
PBG and other pyrroles condense in an acid medium with 2 mL aliquots of 1 M acetic acid, and
with p-dimethylaminobenzaldehyde (DMAB) to form a combine all supernatants together.
colored product. There is evidence [13,14] that further 4. Adjust the total volume of pooled supernatants
reactions occur, resulting in fading of the color with to 10 mL with 1 M acetic acid, and mix well. If
time. For this reason, all measurements should be made necessary, centrifuge to remove any resin that
at a consistent time, between 6 and 8 min after the contaminates the supernatants.
addition of DMAB. The DMAB is contained in a 5. Pipet 2 mL of the pooled eluates into a clean
modified Ehrlichs reagent that contains acetic acid to tube, and add 2 mL of Ehrlichs reagent to each
increase the color intensity. tube at 2-min intervals. Allow the color of each
tube to develop for 6 mins at ambient
Reagents temperature, and read the absorbance.
1. Ehrlichs reagent (modified): (p- 6. Read the absorbance (A) (1 cm cuvette) at 525
dimethylaminobenzaldehyde [DMAB] 134 mmol/L; nm and 555 nm using a blank consisting of 2
HCl, 2.92 mol/L; acetic acid, 13.2 mol/L). Dissolve 2 g mL of 1 M acetic acid and 2 mL of Ehrlichs
of DMAB in 25 mL of concentrated HCl, and add 75 mL reagent.
of glacial acetic acid. Store refrigerated in the dark up to
several months [1]. If the clear reagent becomes yellow, Calculation
it should be discarded.
2. Sodium acetate, 100 g/L (1.2 mol/L). Place
100 g of sodium acetate in a liter volumetric flask, and
981
where A555 = absorbance at 555 nm of unknown; 10 = 2. The red condensation product that is measured
dilution factor for urine; in this assay has a strong absorption peak at 555
0.114 = A555 of 1 g/mL of PBG when reacted nm and a weaker peak at 525 nm. Most
pigments that cause interference absorb more
with modified Ehrlichs reagent strongly at the lower wavelength. Therefore the
24-h excretion (mg/24 h) =g of PBG/mL Total urine ratio A525/A555 should be close to 0.83. Urines
volume (mL) (1 mg/1000 g)
giving a value > 1 for this ratio should not be
reported as containing elevated levels of PBG.
Notes
This is frequently the case when the A555 is
1. An extinction coefficient of 6.2 104 Lmol-
only slightly elevated. Under these
1cm-1 has been described for the reaction
circumstances, the assay should be repeated
product [14]. The factor of 0.114 was derived with a fresh specimen.
from a molar absorptivity of approximately 2.6 3. It is essential that the water washes of the resin
104 Lmol-1cm-1, but it should be verified after addition of urine be performed four times.
for each spectrophotometer [14]. For this Fewer washes will result in unacceptable
purpose, preweighed vials of PBG are carryover of interfering substances. Likewise, it
commercially available (Sigma Chemical Co., is essential that at least four elutions be
St. Louis, MO). The PBG should be dissolved performed with acetic acid to ensure adequate
in 1 M acetic acid and assayed in the reaction recovery of PBG.
described above at several concentrations. One
can verify the entire procedure by adding a
known, relatively high (for example, 20
mg/mL) amount of PBG to normal urine and
performing the assay.
982
Procainamide and N-Acetyl Procainamide
Procainamide and N-Acetylprocainamide

Matthew T. Olson and William Clarke
Name: Procainamide N-acetylprocainamide
Clinical Vaughan Williams class Ia antiarrhythmics with multiple indications
significance: Refer to Chapter 14, Therapeutic Drug Monitoring, in the 5th Edition of
Molecular C13H21N3O C15H22N3O2
formula:
Molecular 235.54 D 277.37 D
weight:
Merck Index: 7553 7845
Chemical Amines Amines
class:
Structure:
i polarization immunoassay (FPIA), enzyme-multiplied

immunoassay technique (EMIT), particle-enhanced
Procainamide (PA) is a Vaughan Williams class Ia turbidimetric inhibition immunoassay (PETINIA), or
antiarrhythmic drug that clinicians prescribe acutely for cloned-enzyme donor immunoassay (CEDIA). The
pharmacological conversion of atrial and ventricular frequency of these assays in use is given in Table 1. This
arrhythmias and for chronic suppression of these chapter will focus on these four methods. Readers
arrhythmias and others when they are refractory to more interested in the historical use of colorimetry of a
common treatment modalities [1,2]. Hepatic arylamine N- derivative [9], gas chromatography [10], or high-
acetyltransferase 2 (NAT2) acetylates PA to yield an performance liquid chromatography [11-13] may consult
equivalently therapeutic active metabolite, N- this chapter in the previous edition. Those interested in the
acetylprocainamide (NAPA), which is at least equally developmental use of tandem mass spectrometry for TDM
responsible for the adverse cardiac effects associated with of antiarrhythmics may find recent investigations a useful
PA therapy [3,4]. The production of NAPA by acetylation starting point [14, 15].
on NAT2 occurs at a genetically determined rate that
varies widely among patients [5, 6], so NAPA Methods involving FPIA [16] are based on the fact that a
concentrations cannot be predicted from those of PA stationary fluorophore fluoresces in the same plane of light
without genetic information about the NAT2 that excites it. If the fluorophore moves after excitation
polymorphisms in the patient [7]. For these reasons, as the result of rotational relaxationthe emitted light will
therapeutic drug monitoring (TDM) of PA necessitates not be polarized, because the electronic field has been
monitoring of NAPA. changed to new planes. Small molecules have faster
rotational relaxation than large molecules, so fluorophores
According to the 2007 College of American Pathologists that are bound to antibodies preserve the plane of polarized
(CAP) survey [8], participating laboratories analyze PA excitation light more than those that are not bound to
and NAPA concentrations with either fluorescence antibodies. Figure 1 provides a schematic for FPIA. In
FPIA for PA and NAPA [17-19], a known amount of
fluorescein-tagged drug and anti-drug antibody is added to
i
Procainamide and N-Acetyl Procainamide the cuvette containing an aliquot of patient serum or
Previous and current authors of this method: plasma. Tagged and untagged drug compete to bind the
First edition: Not done antibody. When more drug is present, less tagged drug
Methods edition: John E. Sherwin binds the antibody, so a smaller proportion of the
Second edition: Not updated fluorescence will be polarized. The instrument irradiates
Third edition: Not updated the cuvette with polarized excitation light, and the intensity
Fourth edition: John E. Sherwin, Curt Grote of fluoresced light on the planes parallel, Iv, and
Fifth edition: Matthew T. Olson, William Clarke
983
orthogonal, Ih, to the excitational plane is detected. These and EMIT require anti-drug antibody, CEDIA has the
values are converted into a value for polarization, P, by advantage in that the only other drug-specific entity it
equation 1(1): requires is a very small drug-labeled ED rather than whole
enzyme. The direct relationship between enzyme rate and
P = (Iv Ih)/(Iv + Ih) antigen is demonstrated in Figure 2A, EMIT is shown in
Figure 4, and CEDIA is shown in Figure 5.
Drug concentration and P are inversely proportional at a
slope defined by the experimental conditions. Linear Reference and Preferred Methods
interpolation of P onto a concentration-versus-P graph Traditionally, TDM of PA and NAPA has been performed
yields the drug concentration. Drug concentration and by GC and HPLC, but the immunoassays described above
polarization are inversely related, as in Figure 2A. have largely replaced these methods as preferred methods
because of their speed and the fact that they obviate the
Instruments based on PETINIA [20, 21] make extraction sample preparation required in chromatographic
turbidimetric measurements of light scattering in the methods. There is no formally defined reference method
cuvette. When drug-coated microparticles bind anti-drug for PA and NAPA; however, the HPLC method functions
antibody, agglutination occurs, and more light is scattered. as a de facto reference method because it is not subject to
Drug from the patient competes with the particle-bound the immunochemical interferences inherent in the modern
drug for antigen-binding sites and so prevents methods [27]. Among the immunoassays, FPIA is the most
agglutination in a concentration-dependent fashion. widely used method among respondents to the 2007 CAP
Because agglutinated particles scatter light, the light survey [8]. Several studies have evaluated intermethod
scattering is inversely related to the amount of drug variability [19] of the immunoassays and compared
present. The inverse plot is shown in Figure 2B, and the immunoassay to HPLC for this analyte [17, 18, 26]. These
method is shown in Figure 3. PETINIA offers the results have shown good agreement. All the methods meet
advantage of a higher dynamic range than other the acceptable performance standards outlined by the
immunoassays in exchange for more susceptibility to Clinical Laboratory Improvement Amendments of 1988
interference [22]. (CLIA-88).
Both EMIT [23] and CEDIA [24] involve Specimen

spectrophotometric monitoring of enzymatic reactions by Because procainamide and NAPA clear the body rapidly
drug-labeled enzymes which are disabled by binding to under normal conditions, the trough concentration is most
anti-drug antibody. Higher concentrations of drug bind clinically useful. Blood should be drawn 1 hour or less
more antibody, so the enzymatic reaction rate increases in before the next scheduled dose. Serum or plasma can be
a linear fashion. The rate of reaction is linearly related to drawn, and either EDTA or oxalate are acceptable
functional enzyme. The two methods differ on the basis of anticoagulants. The sample can be kept at 2C to 8C for
how the enzyme is labeled with drug and how antibody up to 24 hours and at 20C for up to 2 weeks [28].
binding affects enzyme kinetics. As shown in Figure 1A, Heating the sample to 60C immediately prior to analysis
EMIT methods use whole enzymes, such as glucose-6- (in protocols that desire virus denaturation) is shown not to
phosphate dehydrogenase (G6PD), which have the drug of be associated with a decrease in drug concentrations [29].
interest covalently attached. Free anti-drug antibodies bind Excessive free heme, bilirubin, and lipids all interfere with
the drug on the enzyme and inactivate the enzyme by steric CEDIA, EMIT, and PETINIA. As discussed elsewhere,
hindrance. At higher drug concentrations in the blood, NAPA must accompany PA measurement, and the
more patient drug binds the free anti-drug antibodies, less concentrations should be reported separately, not as a sum
G6PD is inactivated, so more glucose-6-phosphate is [28, 30].
converted to glucose-6-phosphogluconolactone. This
reaction can be followed via the NADP+/NADPH Procainamide and N-Acetylprocainamide Therapeutic
transition, which results in a change in absorbance at 340 Range
nm. Linear interpolation into a positively sloped graph of Procainamide [28]
drug concentration versus spectrophotometric absorbance Therapeutic: 4.0-8.0 mg/L
yields the drug concentration [25,26]. In contrast to EMIT Toxic: >10 mg/L
methods, which use whole enzymes, CEDIA methods, N-Acetylprocainamide [28]
depicted in Figure 1B, use recombinant fragments of an Therapeutic: 10-20 mg/L
enzyme, usually -galactosidase. The method involves two Toxic: >40 mg/L
kinds of fragments: large enzyme acceptors (EA) and
small drug-labeled enzyme donors (ED). The ED consists Interpretation
of an essential peptide fragment to make the EA Procainamide causes more clinical side effects than can be
functional. Anti-drug antibody binding of the drug moiety discussed in detail here. Effectiveness [31] and toxicity
on an ED prevents EA-ED association, formation of active [32] are the central reasons for TDM of class Ia
enzyme, and the degradation of a galactopyranoside which antiarrhythmics. Procainamide typically achieves
is absorbent at 420 and 570 nm [24]. Whereas both CEDIA therapeutic effect at the defined ranges, but patients with
984
extensive chronic ischemia require higher concentrations References

of drug [31]. However, PA and NAPA commonly cause 1 Koch-Weser I, Klein S. Procainamide dosage
QTc prolongation, ventricular extrasystoles, and schedules, plasma concentrations, and clinical
tachycardia. These side effects result from the mechanism effects. JAMA. 1971;215:1454-1460.
of therapeutic action: inhibition of fast sodium channels in 2 Atkinson AJ, Ruo T, Piergies A. Comparison of
the myocardium and His-Purkinje conduction system. In the pharmacokinetic and pharmacodynamic
addition to TDM, patient monitoring also involves regular properties of procainamide and N-
electrocardiogram monitoring and assessment of renal acetylprocainamide. Angiology. 1988;39:655-657.
function by laboratory analysis of blood urea nitrogen and 3 Funck-Bretano C, Light R, Lineberry M, Wright
plasma creatinine. Regular electrocardiograms are G, Roden D, Woosley R. Pharmacokinetic and
essential in monitoring, because QTc prolongation cannot pharmacodynamic interaction of N-acetyl
be predicted from drug concentrations alone in many cases procainamide and procainamide in humans. J
[3,4] and can lead to the fatal torsades-de-pointes Cardiovasc Pharmacol. 1989;14:364-373.
arrhythmia. 4 Miller J, Zipes D. Therapy for cardiac
arrhythmias. In: Zipes D, Libby P, Bonow R,
The half-lives of PA and NAPA are 3.5 0.8 and 8.1 1.3 Braunwald E, eds. A Textbook of Cardiovascular
hours, respectively [28, 33, 34]. Organic cation Medicine. 7th ed. Phladelphia: Saunders; 2005.
transporters eliminate both PA and NAPA by active 5 Okumura K, Kita T, Chikazawa S, Komada F,
secretion in the proximal tubule of the nephron. Because Iwakawa S, Tanigawara Y. Genotyping of N-
the same mechanism eliminates many other drugs, such as acetylation polymorphism and correlation with
fluoroquinolones [35] and central acetylcholinesterase procainamide metabolism. Clin Pharmacol Ther.
inhibitors [36], co-administration of these drugs can result 1997;61:509-517.
in potentially dangerous fluctuations of drug levels and 6 Deguchi T, Mashimo M, Suzuki T. Correlation
warrants TDM, especially since PA and NAPA already between acetylator phenotypes and genotypes of
compete for the same transporters and mutually prolong polymorphic arylamine N-acetyltransferase in
the drug effects [3,34]. The organic anion transporters do human liver. J Biol Chem. 1990;265:12757-
not appear to be involved to a clinically relevant extent 12760.
[37]. Concentrations of PA and NAPA are also followed in 7 Grant D, Blum M, Beer M, Meyer U.
patients with chronic renal disease, in cardiogenic shock, Monomorphic and polymorphic human arylamine
and in edematous states [38-42]. Hemodialysis removes N-acetyltransferases: a comparison of liver
NAPA but not PA [30,43], so it introduces an additional isoenzymes and expressed products of two gloned
need for TDM. Finally, PA ranks among the top drug Genes. Mol Pharmacol. 1990;39:184-191.
choices for treating supraventricular arrhythmias in 8 Z-B Therapeutic Drug Monitoring General
neonates, and standard of care includes intensive TDM of Survey: College of American Pathologists (CAP);
PA and NAPA [44], because the clinical situation involves 2007.
rapidly changing conditions and immature renal function. 9 Sitar D, Graham D, Rango R, Dusfresne L,
The relationship between drug levels and the occurrence Ogilvie R. Modified colorimetric method for
and severity of most common non-urgent systemic side procainamide in plasma. Clin Chem.
effects associated with PA and NAPA has not been fully 1976;22:379-80.
explored. However, contemporary interest in the drug- 10 Atkinson AJ, Parker M, Strong J. Rapid gas
induced lupus often associated with PA therapy has lead to chromatographic measurement of plasma
several investigations [45-47]. Chronic PA use is shown to procainamide concentration. Clin Chem.
cause dose-dependent T-cell disturbances [45], leading 1972;18:643-647.
occasionally to drug-induced lupus [46]; this is a clinical 11 Kessler K, Ho-Tung, Steele B, Silver J, Pickoff
diagnosis supported by positive anti-chromatin antibody A, Narayanan S et al. Simultaneous quantitation
tests [47]. There is no evidence that drug monitoring plays of quinidine, procainamide, and N-
any role in predicting or diagnosing these adverse effects. acetylprocainamide in serum by gas-liquid
chromatography with a nitrogen-phosphorous
Procainamide and N-Acetylprocainamide Performance selective detector. Clin Chem. 1982;28:1187-
Goals 1190.
All the modern immunoassays show less than an 8% 12 Stearns F. Determination of procainamide and N-
coefficient of variation (CV) for PA. The performance acetylprocainamide by high-performance liquid
criteria outlined by CLIA-88 require no more than 25% chromatography. Clin Chem. 1981;27:2064-2067.
difference from the peer-group mean. These standards 13 Dutcher J, Strong J. Determination of plasma and
apply to the minimal therapeutic concentration of 4.0 N-acetylprocainamide concentration by high-
mg/L. The same performance criteria and observed CV pressure liquid chromatography. Clin Chem.
apply for NAPA, except that the clinical decision 1977;23:1318-1320.
concentration is 10.0 mg/L. Currently available assays 14 Li S, Liu G, Jia J, Liu Y, Pan C, Yu C et al.
meet these necessary criteria. Simultaneous determination of ten antiarrhythmic
985
drugs and a metabolite in human plasma by liquid therapeutic drugs. Ther Drug Monit.
chromatography tandem mass spectrometry. J 1994;16:612-615.
Chromatogr B Analyt Technol Biomed Life Sci. 30 Campbell T, Williams K. Therapeutic drug
2006;Nov17:174-181. monitoring: antiarrhythmic drugs. Br J Clin
15 Dhananjeyan M, Bykowski C, Trendel J, Sarver J, Pharmacol. 1998;46(4):307-319.
Ando H, Erhardt P. Simultaneous determination 31 Myerburg R, Kessler K, Kiem I, Pefkaros K,
of procaine and para-aminobenzoic acid by LC- Conde C, Cooper D et al. Relationship between
MS/MS method. J Chromatogr B Analyt Technol plasma levels of procainamide, suppression of
Biomed Life Sci. 2007;847(2):224-230. premature ventricular complexes and prevention
16 Gutierrez M, Gomez-Hens A, Perez-Bendito D. of recurrent ventricular tachycardia. Circulation.
Immunoassay methods based on fluorescence 1981;64:280-290.
polarization. Talanta. 1989;36:1187-1201. 32 Takada M, Goto T, Kotake T, Saito M, Kawato
17 Haver V, Audino N, Burris S, Nelson M. Four N, Nakai M et al. Appropriate dosing of
fluorescence polarization immunoassays for antiarrhythmic drugs in Japan requires therapeutic
therapeutic drug monitoring evaluation. Clin drug monitoring. J Clin Pharm Ther. 2005;30:5-
Chem. 1989;35:138-140. 12.
18 Laurino J, McKay J, Fischberg E, Schwenzer K, 33 Jamali F, Alballa R, Mehvar R, Lemko C. Longer
Kaufman R. Comparison of three immunoassays plasma half-life for procainamide, utilizing a very
for the determination of N-acetylprocainamide: sensitive high-performance liquid
determination of false elevations by a polyclonal chromatography assay. Ther Drug Monit.
antibody-based method. Ther Drug Monit. 1988;10:91-96.
1994;16:312-315. 34 Roden D, Beele S, Higgins S, Wilkinson G,
19 Tisdale J, Padhi I, Ware J, Svensson C. Smith R, Oates J et al. Antiarrhythmic efficacy,
Comparison of FPIA with LC for quantification pharmacokinetics, and safety of N-
of PA and NAPA in urine. Ther Drug Monit. acetylprocainamide in human subjects:
1996;18:693-697. comparison with procainamide. Am J Cardiol.
20 Cambiaso C, Riccomi H, Masson P, Heremans J. 1980;46:463-468.
Automated nephelometric immunoassay II: its 35 Bauer L, Black D, Lill J, Garrison J, Raisys V,
application to the determination of haptens. J Hooton T. Levofloxacin and ciprofloxacin
Immunol Methods. 1974;5:293-302. decrease procainamide and N-acetylprocainamide
21 Grange J, Roch A, Quash G. Nephelometric assay renal clearances. Antimicrob Agents Chemother.
of antigens and antibodies with latex particles. J 2005;49:1649-1651.
Immunol Methods. 1977;18:365-375. 36 Cummings J, Frank J, Cherry D, Kohatsu N,
22 Ritchie R. The foundations of immunochemistry. Kemp B, Hewett L et al. Guidelines for managing
In: Wild D, ed. The Immunoassay Handbook. Alzheimers disease: part II, treatment. Am Fam
Oxford: Elsevier; 2005. Physician. 2002;65:2525-2534.
23 Rubenstein K, Schneider R, Ullman E. 37 Lam Y, Boyd R, Chin S, Chang D, Giacomini K.
Homogenous enzyme immunoassay: a new Effect of probenecid on the pharmacokinetics and
immunochemical technique. Biochem Biophys pharmacodynamics of procainamide. J Clin
Res Commun. 1972;47:846-851. Pharmacol. 1991;31:429-432.
24 Henderson D, Friedman S, Harris J, Manning W, 38 Jurgens G, Graudal N, Kampmann J. Therapeutic
Zoccoli M. CEDIA, a new homogenous drug monitoring of antiarrhythmic drugs. Clin
immunoassay system. Clin Chem. 1986;32:1637- Pharmacokinet. 2003;42:647-663.
1641. 39 Dresser M, Leabman M, Giacomini K.
25 Henry P, Dhruv R. More sensitive EMIT for PA Transporters involved in the elimination of drugs
and NAPA in plasma, serum and urine. Clin in the kidney: organic anion transporters and
Chem. 1988;34:957-960. organic cation transporters. J Pharm Sci.
26 Walberg C, Wan S. Clinical evaluation of the 2001;90(4):397-421.
EMIT PA and NAPA assay. Ther Drug Monit. 40 Lee W, Kim R. Transporters and renal drug
1979;1:47-56. elimination. Ann Rev Pharmacol Toxicol.
27 Mehta A. The how and why of therapeutic drug 2004;44:137-166.
monitoring. Anal Proc. 1995;32:347-351. 41 Mizuno N, Niwa T, Yoshihisa Y, Sugiyama Y.
28 Valdes R, Jortani S, Gheorghiade M. Standards of Impact of drug transporter studies on drug
laboratory practice: cardiac drug monitoring. Clin discovery and development. Pharmacol Rev.
Chem. 1998;44:1096-1109. 2003;55:425-461.
29 Dasgupta A, Bard D. Effect of heating sera under 42 Raebel M, Carroll N, Andrade S, Chester E,
conditions necessary for deactivation of human Lafata J, Feldstein A et al. Monitoring drugs with
immunodeficiency virus on commonly monitored a narrow therapeutic range in ambulatory care.
Am J Manag Care. 2006;12:268-274.
986
43 Braden G, Fitzgibbons J, Germain M, Ledewitz 45 Richardson B. DNA methylation and autoimmune

H. Hemoperfusion for treatment for N- disease. Clin Immunol. 2003;109:72-79.
acetylprocainamide intoxication. Ann Intern Med. 46 Rubin R. Drug-induced lupus. Toxicology.
1986;105:64-65. 2005;209:135-147.
44 Moffett B, Cannon B, Friedman R, Kertesz N. 47 Schett G, Rubin R, Steiner G, Hiesberger H,
Therapeutic levels of intravenous procainamide in Muller S, Smolen J. Comparative analysis of
neonates: a retrospective assessment. antichromatin antibody specificity in lupus
Pharmacotherapy. 2006;26:1687-1693. erythematosus cell-positive and cell-negative
sera. Arthritis Rheum. 2000;43:420-428.
987
Table 1: The methods participants in the 2007 CAP survey use for the analysis of PA and NAPA are given with a
breakdown according to popularity in terms of raw numbers and percentages. These data are adapted from citation
[8].
Method N labs % labs

FPIA 278 70.6
EMIT 54 13.7
PETINIA 41 10.4
CEDIA 21 5.3
Total 394
CEDIA, Cloned-enzyme donor immunoassay; EMIT, enzyme-multiplied immunoassay technique; FPIA,
fluorescence polarization immunoassay; PETINIA, particle-enhanced turbidimetric inhibition immunoassay.
Figure 1: Schematic for FPIA. When no drug is present, as in the top reaction, the polarity is conserved because the
fluorescein-labeled drug is bound to antibody so moves little. More drug prevents the labeled drug from binding, and
the polarity of the sample decreases.
Fluoroscein labeled Drug
+
YY Polarity conserved
YY Anti-drug Antibody Polarized Light
+
YY Polarity dispersed
Patient Drug
988
Figure 2: Sample plots demonstrating the relationship between antigen concentration and observed parameters in
competitive (A) and non-competitive (B) relationships. In methods such as EMIT and CEDIA, the observed
parameter is the reaction rate of the enzyme, which correlates directly with antigen concentration, as in (B). In
methods such FPIA, which measures polarization, and PETINIA, which measures scattered light, more antigen
decreases the magnitude of these measurements, leading to an inverse relationship, as in (A).
(A) (B)
Observed parameter
Observed parameter
Antigen concentration Antigen concentration
Figure 3: PETINIA. In the presence of drug, anti-drug antibody binds drug rather than the particle-bound drug.
Therefore, less agglutination occurs, and less light scatters.
Y
Drug-coated
Latex Beads
Y Y
Light scatters
orthogonal to
+ incident
Incident Light
YYY Anti-drug
Source
+ antibody
Patient drug
YYY Light passes
parallel to
incident
989
Figure 4: EMIT. Drug-labeled G6PD binds antibody in the absence of patient drug. The binding inactivates the
enzyme, so no reaction occurs. In the presence of patient drug, which occupies the antibody binding sites, the free
G6PD is able to convert G6P at a higher rate.
No reaction
+
Drug labeled
G6PD
YY
YY
Anti-drug antibody substrate
+
YY
Patient drug product
Figure 5: CEDIA. The top scenario involves the absence of patient blood. The enzyme acceptors are inactive in the
absence of enzyme donors, so no reaction takes place. At higher levels of patient drug, the antibody binding sites are
occupied with patient drug, and the reaction rate increases.
No reaction
Enzyme
acceptors
+
+ Enzyme donors
YY
Y+ Y
substrate
Anti-drug antibody
product
Patient drug
YY
990
Table 2: Methods for Procainamide and N-Acetylprocainamide Analysis

Method 1: Fluorescence polarization immunoassay (FPIA)
Principle of analysis: Patient drug displaces drug-labeled fluorophores from fluorescent
plane polarity, conserving antibody complexes.
Comments: Most popular method in use for PA and NAPA TDM
Method 2: Enzyme-multiplied immunoassay technique (EMIT)
Principle of analysis: Patient drug displaces drug-labeled whole G6PD from a disabling
complexation with anti-drug antibody.
Comments: Commonly used
Method 3: Cloned-enzyme donor immunoassay (CEDIA)
Principle of analysis: Patient drug frees a drug-labeled recombinant peptide which
associates with incomplete -galactosidase to form an active enzyme.
Comments: Only requires a small drug-labeled peptide. Reaction rate monitored by
conversion of a galactopyranoside.
Method 4: Particle-enhanced turbidimetric inhibition assay (PETINIA)
Principle of analysis: Patient drug dissipates the light-scattering agglutination of drug-
labeled micro- or nanoparticles with anti-drug antibody.
Comments: Commonly used
991
Progesterone
Progesterone
John Galligan
Name: Progesterone
Clinical significance: Refer to Chapter 48, General Endocrinology, in the 5th edition of Clinical
Merck Index: 57-83-0, (C A Registry Number)
Structure:
i before GC analysis employing flame ionization detection

[8].
Progesterone is synthesized in the ovary, the adrenal
cortex, and the placenta from the biosynthetic precursor The application of a competitive protein-binding (CPB)
technique [9] (Table 1, Method 5) led to the
5-pregnenolone. The major excretory metabolite
development of a series of methods that employed
present in urine is pregnanediol.
various extraction solvents, different sources of binding
The original methods for assessing progesterone activity
proteins, different tritiated steroids, and a variety of
were bioassays. The increase in mouse uterine weight
techniques for separating the free and protein-bound
was correlated with progesterone content in the infusate
steroid fractions. Some of these methods purified the
(Table 1, Method 1) [1]. Subsequently, quantitative
steroid before CPB analysis. The radioactive steroid
chemical methods were described (Table 1, Method 2).
used was either tritiated corticosterone [9-15] or
These included the 2,4-dinitrophenylhydrazine
progesterone [16-18]. The corticosteroid-binding protein
derivative method [2] and methods employing
used in these competitive binding methods was obtained
spectrophotometric measurements at 240 nm [3], sulfuric
from various sources. Solvents used to extract the
acid chromogen [4], fluorescence [5,6], and
progesterone were petroleum ether [9-11,17,18], ethyl
thiosemicarbazide derivative [7].
acetate [13], hexane [14] and diethyl ether [15,16]. In
some publications, chromatographic purification of the
With the advent of isotopically-labeled steroids and
plasma extract before the protein-binding assay was
derivatization compounds, the technique of double-
reported to be essential for the removal of interfering
isotope derivatization was applied to the measurement of
steroids. Chromatographic systems employing thin-layer
plasma progesterone levels (Table 1, Method 3).
silica gel [9,11,16], paper [13], or Celite columns [15]
have been reported. Separation of the free and protein-
Gas chromatography (GC) has been used to improve the
bound fractions was performed with either Florisol
accuracy and rapidity of progesterone analysis (Table 1,
(activated magnesium silicate 60-200 mesh) [9-15,17] or
Method 4), with a series of solvent extractions and paper
Sephadex [16,18] as the absorbing medium.
chromatography being used to purify the plasma extract
The use of radioimmunoassay for the determination of
plasma progesterone concentration was first reported by
i
Progesterone Abraham et al. [19] and Furuyama and Nugent [20]
Previous and current authors of this method: (Table 1, Method 6). These reports were followed by a
First edition: Not done series of publications that reported the use of different
Methods edition: James A. Demitriou extraction solvents, chromatographic purification
Second edition: Not updated systems, radioactive steroids, antibodies, and methods
Third edition: Not updated for separating the free and antibody-bound fractions [21-
Fourth edition: James A. Demitriou 31].
Fifth edition: John Galligan
992
Progesterone
Non-isotopic immunoassays are now available, and they Interferences

have become the predominant method as various forms The presence of hemoglobin, lipemia, and icterus have
of these assays have been adapted for use on automated no significant effect on the described assays, but this
immunoassay analyzers [32-34]. Such non-isotopic should be confirmed for automated immunoassay
assays measure progesterone directly and do not require systems.
its extraction from the sample. Reagents containing 8-
anilino-1-naphthalene sulfonic acid (ANS), salicylate, Heterophilic antibodies present in some patient samples
danazol, dexamethasone, cortisol, or low pH have been can react with reagent immunoglobulins and interfere
used to displace progesterone from endogenous binding with immunoassays. Patients who have been regularly
proteins. A range of chemiluminescent, fluorescent, exposed to animals or have received immunotherapy or
luminescent, and photometric substrates have been diagnostic procedures utilizing immunoglobulins may
utilized to monitor the activity of the assay. Separation elicit such antibodies as human anti-mouse antibodies
of the free and antibody-bound progesterone is achieved (HAMA). Other heterophilic antibodies such as anti-
by incorporating antibody-coated microparticles, sheep, anti-rabbit or anti-goat may also be present in
polystyrene beads, or paramagnetic particles. patient samples. Although assay reagents now include
Immunoassays for progesterone incorporate antibodies blocking agents to minimize the possibility of
generated against progesterone conjugated to protein, interference, interaction between patients sera
such as bovine serum albumin through links at positions containing heterophilic antibodies and assay reagents
C-3, C-4, C-6, C-7, C-11, or C-20. All antibodies so still occurs [41], the possible outcome being anomalous
generated have the potential to cross-react with other C- results and subsequent inappropriate treatment.
21 corticosteroids present in the sample.
Progesterone Reference Intervals
Isotope dilution gas chromatography/mass spectrometry Males, adult <0.4 ng/mL (<1.3 nmol/L)
is the accepted reference method used to determine Females, adults
target values for progesterone. Thienpont et al. [35] have Follicular 0.1 to 1.5 ng/mL (0.3 to 4.8 nmol/L)
reported on five years experience in two laboratories, Luteal 2.5 to 28 ng/mL (7.9 to 89 nmol/L)
determining reference-method values in serum-based Midluteal 5.7 to 28 ng/mL (18 to 89 nmol/L)
materials. For both laboratories, the mean overall Postmenopausal
coefficient of variation (CV) was < 2%, the method bias 0.1 to 0.3 ng/mL (0.3 to 0.9 nmol/L)
was < 1%, and the maximum total analytical error in all
cases was < 3%. More recently, developments of highly
sensitive liquid chromatography/tandem mass The concentration of progesterone varies with time
spectrometry (LC/MS/MS) instrumentation have before and after ovulation. During pregnancy, values
resulted in precise, accurate, and sensitive methods for may increase by 10 to 100 times over those seen in
the determination of low-concentration analytes, nonpregnant individuals. Marked increases are observed
including progesterone. As a result, a candidate after the 20th week of pregnancy [42].
reference measurement procedure for the estimation of Given the significant bias reported [37] for some
progesterone has now been evaluated [36]. immunoassay analyzers, the use of laboratory-validated,
method-specific reference ranges would seem
In the majority of clinical laboratories, automated non- mandatory.
isotopic immunoassays are used to measure
progesterone. However, a recently published report [37] Interpretation
which focused on the performance of six automated Under the influence of follicle-stimulating and
immunoassay analyzers documented significant bias in luteinizing hormones, progesterone secretion by the
some methods and considerable variation in results for ovary starts to rise during the ovulatory phase of the
the same sample assayed in different laboratories. This menstrual cycle and increases to a maximum 6 to 8 days
report highlights the need for assays to be stringently after ovulation. The corpus luteum formed after
validated to ensure they are fit-for-purpose. ovulation is the major source of progesterone secretion
during the luteal phase of the menstrual cycle. If
Specimen fertilization of the ovum does not occur, the corpus
Serum or plasma collected with heparin or EDTA as the luteum atrophies, the progesterone concentration falls,
anticoagulant is the preferred specimen; however, the and menstruation occurs. However, with fertilization of
specimen requirements for particular immunoassay the ovum, the corpus luteum is stimulated by the
systems should be verified. Saliva has also been increasing levels of chorionic gonadotropin from the
suggested as a possible fluid [38]. A progressive fetoplacental unit, which continues to secrete the
decrease in progesterone has been observed in specimens hormone. Progesterone concentrations continue to rise
collected into gel-barrier tubes when compared with throughout pregnancy. During pregnancy, progesterone
samples collected into tubes without gel [39]. No is synthesized in the placenta from maternal cholesterol
significant differences have been noted between samples as the substrate.
collected in glass or plastic tubes [40]. Samples are
stable for 7 days at 4C or for up to 3 months at 20C. The major use of progesterone assays is the detection of
ovulation. Midluteal progesterone concentrations of
greater than 10 ng/mL of plasma indicate luteinization of
993
Progesterone
the follicle. However, there is evidence of corpus luteum 2 Zander J, Simmer H. Die chemische
formation without ovulation (the luteinized, unruptured Bestimmung von Progesterone in organischen
follicle syndrome) in some infertile women [43]. In this Substraten. Klin Wochenschr 1954;32:529-540.
case, it is recommended that three or more samples be 3 Short RV. The chemical estimation of
obtained so that they will allow bracketing of the progesterone in peripheral blood. J Endocrinol
midluteal peak (such as days 19, 21, and 33). It must be 1958;16:415-425.
stressed that the interpretation is dependent on accurate 4 Oertel GW, Weiss SP, Eik-Nes KB.
determination of the start of the last menstrual period Determination of progesterone in human blood
[38]. Luteal-phase insufficiency or short luteal phase has plasma. J Clin Endocrinol Metab 1959;19:213-
also been characterized in cases of infertility. Multiple 218.
progesterone assays during the luteal phase are required 5 Short RV, Levett I. The fluorometric
to identify this syndrome, which is characterized by low determination of progesterone in human plasma
progesterone concentrations throughout the cycle [44]. during pregnancy and the menstrual cycle. J
Endocrinol 1962;25:239-244.
The measurement of progesterone in early pregnancy, as 6 Heap RB. A fluorescence assay for
an indicator of potential threatened abortion, has not progesterone. J Endocrinol 1964;30:293-305.
been found to be of clinical use and is not justified. 7 Sommerville IF, Pickett MT, Collins WP,
However, the measurement of progesterone (and hCG) Denyer DC. A modified method for the
concentration may be a useful adjunct to quantitative determination of progesterone in
ultrasonography in the investigation of patients human plasma. Acta Endocrinol 1963;43:101-
presenting to emergency departments with first-trimester 109.
bleeding or pain [45]. It was also reported that in such 8 Yannone MG, McComas DB, Goldfien A. The
circumstances, patients with progesterone concentration assay of plasma progesterone. J Gas
5 ng/mL will almost certainly have a nonviable Chromatogr 1964;2:30-33.
pregnancy. In stable patients with a progesterone 9 Neill JD, Johansson EDB, Datta JK, Knobil E.
concentration > 22 ng/mL, there is a strong likelihood of Relationship between plasma levels and
a viable intrauterine pregnancy. luteinizing hormone and progesterone during
the menstrual cycle. J Clin Endocrinol Metab
Estimation of progesterone concentration has been used 1967;27:1167-1173.
to determine the probability of pregnancy in women 10 Johansson EDB. Progesterone levels in
undergoing ovarian stimulation with GNRH and peripheral plasma during the luteal phase of the
gonadotropins for in-vitro fertilization. However, a normal menstrual cycle measured by a rapid
recent systematic review and meta-analysis does not competitive binding technique. Acta Endocrinol
support this association [46]. 1969;61:592-606.
11 DeSouza MLA, Williamson HO, Moody LO,
The level of progesterone in saliva samples has also been Diczfalusy E. Further assessment of the
shown to provide a noninvasive method for determining reliability of progesterone assays by
profiles of menstrual cycles [47]. competitive protein binding. Acta Endocrinol
1970;64:(suppl 147)171-183.
Progesterone Performance Goals 12 Lurie AO, Patterson RJ. Progesterone in non-
Data was obtained from the College of American pregnancy: an assay for the clinical chemistry
Pathologists (CAP) 2007 Y-A and Y-B Ligand (Special) laboratory. Clin Chem 1970;16:856-860.
Participant Summary. The performance of 17 13 Martin BT, Cooke BA, Black WP. Evaluation
immunoassay analyzers (~1136 participants) in of a rapid method for the measurement of
measuring progesterone over the range ~1.6 to 31.5 plasma progesterone by competitive protein
ng/mL was documented. Of the 17 analyzer groups, 1 binding. J Endocrinol 1970;46:369-377.
group had 5/6 samples with CVs > 10%, and 3 other 14 Demetriou JA, Austin FG. A rapid competitive
groups had 3/6 samples with > 10% CV. The evaluation protein binding assay for plasma progesterone.
criteria for progesterone in this Survey is peer-group Clin Chim Acta 1971;33:21-32.
target value 3 SD. Under this criteria, 16/17 analyzer 15 Stone S, Nakamura RM, Mishell DR Jr,
groups had acceptable performance. Thorneycroft IH. A modified technique for the
assay of progesterone in blood using Celite
When reviewing the all results medians, 1 analyzer column chromatography. Steroids 1971;17:411-
groups medians differed by > 15% in 6/6 samples, 422.
another group differed in 4 samples, and a further 3 16 Yoshimi T, Lipsett MB. The measurement of
analyzer groups had 3 samples > 15% from the all plasma progesterone. Steroids 1968;11:527-
results medians [48]. 540.
17 Horth CE, Palmer RF. Measurement of
References progesterone in human plasma. Clin Endocrinol
1 Hooker CW, Forbes TR. A bioassay for minute 1972;1:199-207.
amounts of progesterone. Endocrinology 18 Pichon MF, Milgrom E. Competitive protein
1947;41:158-169. binding assay of progesterone without
chromatography. Steroids 1973;21:335-346.
994
Progesterone
19 Abraham GE, Swerdloff R, Tulchinsky D, reference method values by isotope dilution gas
Odell WD. Radioimmunoassay of progesterone. chromatography/mass spectrometry: a five-year
J Clin Endocrinol Metab 1971;32:619-624. experience of two European reference
20 Furuyama S, Nugent CA. A radioimmunoassay laboratories. Eur J Clin Chem Clin Biochem
for progesterone. Steroids 1971;17:663-674. 1996;34:853-860.
21 Morgan CA, Cooke ID. A comparison of the 36 Tai SS-C, Xu B, Welch MJ. Development and
competitive protein-binding assay and evaluation of a candidate reference
radioimmunoassay for plasma progesterone measurement procedure for the determination of
during the normal menstrual cycle. J Endocrinol progesterone in human serum using isotope
1972;54:445-456. dilution liquid chromatography/tandem mass
22 Kutas M, Chung A, Bartos D, Castro A. A spectrometry. Anal Chem 2006;78:6628-6633.
simple progesterone radioimmunoassay without 37 Coucke W, Devleeschouwer N, Libeer J-C,
column chromatography. Steroids 1972;20:697- Schiettecatte J, Martin M, Smitz J. Accuracy
716. and reproducibility of automated estradiol-17
23 Youssefnejian E, Florensa E, Collins WP, and progesterone assays using native serum
Sommerville IF. Radioimmunoassay of plasma samples: results obtained in the Belgian
progesterone. J Steroid Biochem 1972;3:893- external assessment scheme. Human
901. Reproduction 2007;22:3204-3209.
24 Thorneycroft IH, Stone SC. Radioimmunoassay 38 Wood P, Groom G, Moore A, Ratcliffe W,
of serum progesterone in women receiving oral Selby C. Progesterone assays: guidelines for the
contraceptive steroids. Contraception provision of a clinical biochemistry service.
1972;5:129-146. Ann Clin Biochem 1985;22:1-24.
25 DeVilla GO, Roberts K, Wiest WG, Mikhail G, 39 Hilborn S, Krahn J. Effect of time of exposure
Flickinger G. A specific radioimmunoassay of of serum to gel-barrier tubes on results for
plasma progesterone. J Clin Endocrinol Metab progesterone and some other endocrine tests.
1972;35:458-460. Clin Chem 1987;33:203-204.
26 Cameron EDH, Scarisbrick JJ. 40 Reinartz JJ, Ramey ML, Fowler MC. Plastic vs
Radioimmunoassay of progesterone. Clin Chem glass SST evaluated serum-separator blood-
1973;19:1403-1408. drawing tubes for endocrinologic assays. Clin
27 Calabresi E, Pazzagli M, Fiorelli G, Forti G, Chem 1993;39:2535-2536.
Serio M. Determination of progesterone in 41 Kricka LJ. Interferences in immunoassay: still a
human plasma by radioimmunoassay. J Nucl threat. Clin Chem 2000;46;1037-1038.
Biol Med 1974;18:30-37. 42 Edelstam G, Karlsson C, Westgren M, Lowbeer
28 Tea NT, Castainer M, Roger M, Scholler R. C, Swahn M-L. Human chorionic gonadotropin
Simultaneous radioimmunoassay of plasma (hCG) during third-trimester pregnancy. Scand
progesterone and 17-hydroxyprogesterone J Clin Lab Invest 2007;67:519-522.
normal values in children, in men and women 43 Abdulla U, Diver MJ, Hipkin LJ, Davis JC.
throughout the menstrual cycle and in early Plasma progesterone as an index of ovulation.
pregnancy. J Steroid Biochem 1975;6:1509- Br J Obstet Gynaecol 1983;90:543-548.
1516. 44 Hinney B, Henze C, Kuhn W, Wutke W. The
29 Aso T, Guerrero R, Cekan A, Diczfalusy E. A corpus luteum insufficiency: a multifactorial
rapid 5-hour radioimmunoassay of progesterone disease. J Clin Endocrinol Metab 1996;81:565-
and oestradiol in human plasma. Clin 570.
Endocrinol 1975;4:173-182. 45 Murray H, Baakdah H, Bardell T, Tulandi T.
30 Lebel M, Grosse JH. A rapid and precise Diagnosis and treatment of ectopic pregnancy.
method for measurement of physiological CMAJ 2005;173;905-912.
variations of human plasma progesterone. J 46 Venetis CA, Kolibianakis EM, Papanikolaou E,
Steroid Biochem 1978;9:989-993. Bontis J, Devroey P, Tarlatzis BC. Is
31 Scarisbrick JJ, Cameron EHD. progesterone elevation on the day of human
Radioimmunoassay of progesterone: chorionic gonadotrophin administration
comparison of [1,2,6,7-3H]-progesterone and associated with the probability of pregnancy in
progesterone-[125I]-iodohistamine in-vitro fertilization? A systematic review and
radioligands. J Steroid Biochem 1975;6:51-56. meta-analysis. Human Reproduction Update
32 Product Insert. Architect System. Abbott 2007;13:343-355.
Laboratories 2004;69-6610/R5:2. 47 Gandara BK, Leresche L, Mancl L. Patterns of
33 Product Insert. Access Immunoassay System. salivary estradiol and progesterone across the
Beckman Coulter 2005;33550:1. menstrual cycle. Ann N Y Acad Sci
34 Product Insert. Advia Centaur. Siemens 2007;1098:446-450.
Medical Solutions Diagnostics 2004;118696:2. 48 College of American Pathologists. Participant
35 Thienpont LM, Van Nieuwenhove B, Stockl D, summary surveys 2007: Y-A, Y-B, ligand
Reinauer H, De Leenheer A. Determination of (special). Washington, DC: CAP; 2007.
995
Progesterone
Table 1: Methods for Measurement of Progesterone

Method 1: Biological; bioassay
Principle of analysis: Increase in mouse uterine weight is measured.
Comments: Historical interest
Method 2: Chemical; spectrometric or fluorometric measurement
Principle of analysis: Conversion of steroid to chromogen or fluorogen; absorbance or fluorescence is measured.
Comments: Many interferences, historical interest
Method 3: Isotopic; double-isotope derivatization
Principle of analysis: Isotope ratio of added steroid and radiolabeled endogenous steroid is determined.
Comments: Labor intensive; large quantity of isotope required, historical interest
Method 4: Vapor-phase chromatography; gas-liquid chromatography
Principle of analysis: Chromatographic separation and detection by flame ionization, electron capture, or
nitrogen sensors.
Comments: Labor intensive; limited in number of specimens that can be tested, historical interest
Method 5: Protein binding; competitive protein binding
Principle of analysis: Ratio of free to bound radioactive steroid determined after binding by steroid-specific
binding globulin.
Comments: Lacks specificity, but relatively large numbers of specimens can be processed
Method 6: Immunoassay
a. Radioisotopic/steroid-specific antibody
b. Enzyme/fluorescence/steroid-specific antibody
a. Ratio of free to bound radioactive steroid determined after binding by steroid-specific antibody
b. Bound chemiluminescent, enzyme or fluorescent-labeled ligand determined after binding by steroid-
specific antibody
Comments:
a. Specific, simple, and relatively large numbers of specimens can be processed
b. Simple, readily automated, rapid, capable of processing greater numbers than other assays with no
radioactive waste generated
Method 7: Isotope dilution gas chromatography/mass spectrometry
Principle of analysis: Internal standard, ionization, sorting and separation of ions according to mass and charge,
detection and quantification
Comments: Complex instrumentation, high sensitivity, high specificity, medium throughput, reference method,
requires skilled analyst
Method 8: Liquid chromatography/tandem mass spectrometry
Principle of analysis: HPLC, ionization, sorting and separation of ions according to mass and charge, further
separation of compound fragments, detection, and quantification
Comments: Complex instrumentation, high sensitivity, high specificity, medium throughput, proposed reference
method, requires skilled analyst
996
Prolactin
Prolactin
Sheila Dawling
Name: Prolactin
Clinical interpretation: Refer to Chapter 48, General Endocrinology, in the 5th edition of Clinical
Conversion factor: ng/mL 21.2 mIU/L [WHO 3rd IS 84/500]
Merck Index: 7683
i enzymatic activity of the eluate was assayed by use of a
Human prolactin is a single-chain polypeptide hormone fluorogenic substrate, 4-methylumbelliferyl--
of 198 amino acid residues with a molecular mass of galactoside. Although 48-hour incubation was required
approximately 22,000 D [1]. It is secreted by lactotrophs to optimize the sensitivity to 2.5 ng/mL, the results
of the anterior pituitary gland under the control of correlated well with a radioimmunoassay.
hypothalamic factors. The high structural homology of
prolactin with growth hormone (16%), their similar The non-isotopic immunoassays available today are
cross-reactivity in bioassays, and the 20- to 100-fold rapid, sensitive, reliable, and easily automated. The most
enrichment of growth hormone over prolactin in the common methods in use today are immunometric assays
pituitary gland hampered early attempts at separation, employing two antibodies raised towards different
and it was only in 1970 that prolactin was definitively epitopes of the prolactin molecule (Table 1, Method 4).
identified as a discrete hormone [2]. Prolactin is assayed One antibody, designated the capture antibody, is
as an aid in the clinical diagnosis and management of usually immobilized in proprietary fashion to the
amenorrhea, galactorrhea, and hypothalamic-pituitary reaction vessel wall, to particle(s). The second
disorders. Magnetic resonance imaging (MRI) and other (detection) antibody forms a sandwich with the captured
scanning techniques play an important complementary prolactin and carries a label which is then used to
role. Assay methods have evolved from bioassays, determine the extent of the binding (Figure 1a). Labels
through radio-immunoassays and immuno-radiometric can be enzymes, fluorophores, or chemiluminescent
assays, to modern non-isotopic immunometric assays probes, and to a large extent the nature of the detection
with an assortment of detection methods (see Table 1). label determines the sensitivity of the method. All assays
are highly specific for prolactin with respect to growth
Bioassays for prolactin (Table 1, Method 1) are now hormone and pituitary and reproductive hormones and
only used in research settings, particularly to are described individually in the Reference and Preferred
characterize variant molecular weight prolactins [3,4]. Methods heading.
Early radio-iodine immunoassays were improved by the
use of purified human prolactin (Table 2, Method 2) [5]. Prolactin circulates in human plasma in various forms.
Both bioactivity and immuno-reactivity results Studies using gel-filtration chromatography [8,9] have
correlated well in normal persons and from patients with identified three distinct forms: approximately 60% to
hyperprolactinemia under both basal and stimulated 90% is a 23kDa prolactin monomer; 15% to 30% is a 40
conditions [6]. The manufacture of a monoclonal to 60 kDa big prolactin; and 0% to 10% is a >100 kDa
prolactin antibody in the early 1970s and its big, big prolactin, now commonly referred to as
incorporation into immuno-radiometric kits using macroprolactin. This normal distribution pattern is
radioisotope-labeled antibody as the tracer marked the preserved in the majority of patients with elevated
first clinically useful prolactin assays (Table 1, Method prolactin irrespective of the physiological/pathological
3). The first of these assays was a competitive binding cause. The three forms react differently: in radio-
enzyme assay [7] in which serum was incubated with receptor assays, 77% of the receptor activity was found
rabbit anti-prolactin antibody and -galactosidase- with monomer prolactin; only 12% and 11%,
labeled prolactin. After retention of the antibody-bound respectively, were found with big and macroprolactin. In
prolactin on a column of anti-rabbit IgG-sepharose, the contrast, monomer prolactin represents 58% of the total
immunoreactivity, big prolactin 18%, and
i
macroprolactin 24% [10].
Prolactin Macroprolactin
Previous and current authors of this method: Little is known of the clinical significance of big
First edition: Not done prolactin, and it has not received much attention to date.
Methods edition: C. Dennis Ashby Macroprolactin, however, has been studied extensively.
Second edition: Not updated It is formed in the circulation and has a longer half-life
Third edition: Not updated than monomeric prolactin [11]. Although active in vitro,
Fourth edition: C. Dennis Ashby its lack of activity in vivo can be explained, since its
Fifth edition: Sheila Dawling
997
Prolactin
large size impedes transfer out of the capillaries to target (PEG), Protein A, Protein G or anti-human IgG) have
receptors [12,13]. However, macroprolactin is not a been used.
homogeneous entity. The molecular mass of the
common fraction is in the range 150 to 170 kDa and is GFC is the oldest method (Table 2, Method 1) and is
frequently [14] though not always [15] associated with considered the gold standard in that it allows
IgG. There is evidence that it behaves as an anti- simultaneous quantitation of all three molecular forms of
prolactin autoantibody [16] and is commonly found in prolactin. It is, however, time consuming and expensive,
patients with idiopathic hyperprolactinemia [17]. Recent adding significantly to the cost of the analysis, and
thought is that circulating acidic prolactin isoforms, requires equipment unlikely to be available in non-
found only in these patients, may act as the antigenic specialist laboratories. Precipitation with PEG (Table 2,
stimulus for continued macroprolactin production [18]. Method 2) is relatively nonspecific but easy and
Lower (120 kDa) and higher (200 to 699 kDa) mass inexpensive to perform. PEG is readily obtained and
forms, together with some heavily glycosylated forms highly stable. PEG polymers sequester water from the
and those containing IgA and IgM, are less commonly specimen into their matrix, causing macromolecules to
described [19,20]. The overall prevalence of exceed their solubility and precipitate. The main
macroprolactinemia in the general population is thought drawback is that some immunometric methods are
to be about 1% to 2% (0.5% in healthy people [21], and affected by the presence of the PEG in the supernate.
in patients who have macroprolactinemia, the amount Precipitation of macroprolactinimmunoglobulin G
present correlates with increases or decreases in complexes has been described using a number of
monomeric prolactin over time [22-25]. relatively inexpensive reagents (Table 2, Method 3).
Ultrafiltration (Table 2, Method 4) employs a standard
The reactivity of macroprolactin in immunoassays membrane filter (available from Amicon in a variety of
causes problems when interpreting an increased serum pore sizes) with a molecular weight cut-off of around
prolactin concentration. It is therefore of extreme 100 kDa. These devices are freely porous to 45kDa and
importance that laboratories recognize this potential completely retain above 100 kDa, with variable
source of error and take steps to minimize the reporting penetrance in between. The clear filtrate is analyzed for
of misleading results. Identification of macroprolactin, prolactin content by immunoassay. Since there are no
which has minimal biological activity but can be the additives to interfere with immunoassays, it can be used
cause of high prolactin concentrations in serum by where precipitating techniques are not able to be applied.
immunoassay, can help resolve the diagnostic confusion Comparative studies show good agreement between
and avoid expensive investigations and inappropriate ultrafiltration, PEG precipitation, and GFC, most notably
treatment. The additional expense incurred by the when macroprolactin is the predominant species. The
laboratory is offset many-fold by these avoidable costs. additional sample processing and dilution steps for
Before an analysis for macroprolactin is undertaken, the macroprolactin elimination necessarily introduce
specimen must first be shown to have an increased total imprecision into the method, with assay coefficient of
prolactin concentration. A review of clinical studies [26] variation (CV) ranging from 3% to16%, depending on
reveals a cut-off of approximately 40 ng/mL (800 the methodology and choice of specimen [29].
mIU/L), most commonly used to initiate a
macroprolactin assay, but this should be adapted to the Reference and Preferred Methods
population under study. There is no designated reference method for prolactin,
but the overwhelming preference is for an automated
One common approach has been to use two different immunoassay for clinical applications. These offer the
immunoassaysone with a broad spectrum of reactivity advantage of simplicity and make use of reagents that
(DELFIA or Elecsys) and the other with a low cross- have been optimized, validated, and quality controlled
reactivity (BC Access or Bayer Centaur)and to by the manufacturer. Distribution of analytical methods
highlight discrepant results as possibly due to the for prolactin among the 1500 laboratories returning
presence of macroprolactin. However, the case study by results for immunometric assays to the College of
Ismail shows the fallacy of this approach and highlights American Pathologists (CAP) in 2006-7 is shown in
the need for a more rigorous scientific base [27]. A Table 3. Although 1500 laboratories participated in the
sounder approach is to re-run the prolactin analysis using spot challenges, only one third as many took part in the
the same immunological method after removal of any calibration and linearity verification surveys.
macroprolactin present. These methods are referred to as Comparative data are shown for the 312 laboratories
macroprolactin assays and have recently been compared participating in the 2006 United Kingdom National
and reviewed by Kavanagh et al. [28] (see Table 2 and External Quality Assurance Scheme (UKNEQAS).
Appendix). Two general strategies are available for the Performance in these surveys is discussed later in the
analysis of macroprolactin. The first is to separate the section under Prolactin Performance Goals.
prolactin fractions by gel-filtration chromatography For an objective comparison of available prolactin
(GFC) and determine the immuno-reactivity of each methodologies, optimal characteristics and performance
fraction. The second and more readily available and criteria should be defined. Suggested criteria are
cost-effective strategy is to remove any macroprolactin presented here. Immunoassay reagents (i.e., immunogen,
present and then determine the residual label, and calibration standards) should be of human
immunoreactivity due to monomeric prolactin. Both origin. The procedure should be calibrated against a
ultrafiltration and precipitation (with polyethylene glycol reference standard with known international unit
998
Prolactin
potency, such as WHO 84/500. An analytical or repeat analysis of the same sample produce markedly
measurement range from 0.5 to 200 ng/mL is desirable, different results, or when recovery on dilution is not
preferably with an option for automatic dilution of linear. Interference from heterophile antibodies is
samples with high concentrations of prolactin. The discussed in the section on Interferences below.
challenge is to extend the measurement range without
sacrificing precision at low concentrations. The ability of While some analytical problems encountered in prolactin
the procedure to allow accurate and precise quantitation assays are shared with many other hormones that are
of levels in the 20 to 40 ng/mL range is crucial, because measured by immunometric assays, in the case of
this range represents the discrimination between normal prolactin, it is differences in cross-reactivity with
and elevated results. The antibodies should be tested for macroprolactin that are responsible for most of the
cross-reactivity and be specific enough not to show observed variation and inconsistencies between different
responses at less than 0.1% at pathological methods[34]. Methods are classed as being highly
concentrations with growth hormone (GH), luteinizing reacting (DELFIA, Elecsys I), low reacting (Beckman
hormone (LH), follicle-stimulating hormone (FSH), Access, Bayer Centaur), and intermediate reacting
human placental lactogen (PL), and thyroid stimulating (Immulite, Vitros ECi). None of the technical inserts for
hormone (TSH). Cross-reactivity with human chorionic the methods listed contain specific cross-reactivity with
gonadotrophin (HCG) should be less than 0.01%. Apart macroprolactin, although the Roche insert does describe
from growth hormone, human prolactin has no a procedure for its precipitation with PEG.
significant homology with other polypeptide hormones,
and thus no other interferences should be expected. For the identification and measurement of
macroprolactin, the specimen is first analyzed by
The features of the most commonly used immunometric immunoassay to determine the total prolactin present. If
prolactin assays methods are compared in detail in Table hyperprolactinemia is demonstrated, yet symptoms are
4, and in Figures 2 through 4. The Wallac DELFIA not supportive, or if there is over-recovery on dilution, a
method is also included as this method, although not treatment step is performed and the prolactin analysis
approved for clinical use in the United States, is used repeated on the treated products. Two sample treatments
extensively in research because it shows almost equal are described in more detail in the appendix. Gel-
cross-reactivity with all molecular forms of prolactin. filtration chromatography (GFC) is the gold standard but
is labor intensive and expensive, whereas PEG
All methods should be tested for the presence of a hook precipitation is simpler and far less expensive.
effect. The hook effect occurs in immunometric assays Macroprolactin assays have recently been compared and
when the excess antigen in the patient sample saturates reviewed by Kavanagh et al. [28], and a more thorough
the binding sites on both the capture and detection account of the significance of this entity can be found in
antibodies, thus preventing the formation of the the appendix.
sandwich and the immobilization of the detection
antibody. This is most likely to happen when both Specimen
capture and detection antibodies are incubated with the Serum is the sample of choice for prolactin, but EDTA
patient specimen in a single step (see Table 4). In a two- and heparin plasma may be acceptable, depending on the
step assay, patient specimen is reacted with the capture methodology used (not for methods with active metal
antibody prior to the addition of the capture antibody (in ions). Gel separators do not affect results. However,
some assays excess unbound prolactin is washed from there may be instances where a specific formulation of
the reaction system prior to the addition of the detection tube additives can affect one or more methods adversely
antibody). In this latter case, a maximal (though (see comments in Interferences Section below).
underestimated) response is obtained, whereas in the
former case, a falsely low or even subnormal result is No special handling procedures are necessary, though
obtained [30,31]. With the prevalence of MRI and other multiple freeze/thaw cycles should be avoided. Serum
imaging techniques, the diagnosis of a pituitary adenoma can be stored up to 2 weeks at 2C to 8C and for longer
is unlikely to missed, although it may be classified periods at or below 10C [35].
incorrectly as a nonfunctioning pituitary adenoma rather
than a prolactinoma and be treated surgically rather than Physical stress has been reported to both increase and
with drugs [32]. In one series of patients, nonfunctioning decrease serum prolactin concentrations, whereas
adenomas produced more pituitary damage (visual emotional stress may elevate blood levels of prolactin
impairment, deficiency of ACTH, FSH, LH, and TSH) [36]. Patients should be under resting basal conditions
than did prolactinomas secreting sufficient hormone for before sampling [37]. Diurnal variation in prolactin has
a hook effect [33] and will therefore carry a high degree been described with mean concentrations greater
of clinical suspicion. The hook effect can be between 04:00 and 08:00 hours, and between 20:00 and
demonstrated by serial dilution of the specimen either in 22:00 hours [38,39].
isotonic saline, zero-prolactin calibrator, or system
diluent, as specified in the manufacturers instructions. Interferences
Example results are shown in Table 5. The presence of hemolysis (below 0.5 g/dL
hemoglobin), hyperbilirubinemia (below 20 mg/dL), or
Interference by heterophile antibodies should be lipemia (below 1000 mg/dL triglycerides) generally do
suspected when serial specimens from the same patient not interfere with the reported immunometric assay
999
Prolactin
methods in Table 4. Cross-reactivity with other

reproductive and pituitary hormones is negligible and The most recent type of interference discovered was
was addressed earlier. from a blood collection tube additive, the organosilicone
surfactant Silwet L-720, present in certain lots of BD
For the assays that have biotin-streptavidin components, Vacutainer SST tubes. The presence of this compound
specimens from patients receiving high-dose biotin was responsible for desorbing capture antibodies from
supplements (>5mg/day) may be problematic. It is the solid phase polystyrene beads used in the Immulite
recommended that a minimum of 8 hours elapse between methodology [45]. Although the problem was first noted
biotin ingestion and specimen collection for prolactin with the total T3 assay, resulting in falsely increased
analysis. Isolated cases of high titer antibodies against results, it is common to the methodology, and other
streptavidin are also mentioned in the literature, but competitive binding assays were similarly affected. In
supporting data are difficult to locate. The presence of contrast, immunometric assays were negatively affected,
antibodies to the cross-linker used for attachment of the although to a lesser degree [45]. Such findings have led
ruthenium ion in the Roche assay (Method 4f) has been to heightened awareness of potential interference in
documented in a few patients with the thyroid assays but immunoassays and have called for decanting of
is equally applicable to all assays with this structure calibrators and controls into the same tube type as used
[40]. for patient specimens [46].
Interference by heterophile antibodies should be Prolactin Reference Interval

suspected when serial specimens from the same patient, A variety of physiological, pharmacological, and
or repeat analysis of the same sample, produce markedly pathological conditions can affect the circulating
different results or when recovery on dilution is not concentrations of prolactin, and these conditions should
linear. Heterophile antibodies are polyclonal be considered when results are being interpreted. Greatly
autoantibodies, either IgG or IgM, which bind across a increased concentrations of serum prolactin are present
broad spectrum of different animal antibodies used in kit during the first 6 weeks or so of life but rapidly decrease
manufacture, probably through the Fc region. thereafter to approximately 5 ng/mL, where it remains
Heterophile antibodies thus form sandwiches between throughout most of childhood. For both males and
the capture and detection antibodies, and so a positive females, prolactin concentrations increase progressively
signal is produced even in the absence of prolactin from prepuberty to puberty and to adulthood levels of up
(Figure 1b). In rare cases (10%) the heterophile antibody to approximately 20 ng/mL for males and 25 ng/mL for
binds to just one of the reagent antibodies and prevents females [47,48]. Although adult female prolactin levels
formation of the sandwich, giving a false-negative are usually higher than those in adult males, there is a
signal. Assays employing enzyme labels are rather more considerable overlap. Levels of prolactin are increased
prone to this type of interference because of the degree over baseline in females who are taking oral
of modification to the antibody during the conjugation contraceptives and up to 10-fold during the later stages
process. Human anti-mouse antibodies (HAMA) are one of pregnancy, declining to basal levels by some 90 days
specific example that binds mouse immunoglobulins. postpartum. Breast feeding stimulates prolactin secretion
Heterophile antibodies are more likely to be seen in and slows the decline postpartum. In postmenopausal
people who are exposed to animals, such as farm women, serum concentrations of prolactin decrease to
workers, veterinarians, laboratory technicians, or those below those seen in cycling females [37].
exposed repeatedly to blood products or medications
based on blood products. Many manufacturers formulate Interpretation
their reagents to minimize this possibility of interference A variety of pharmacological agents are known to affect
from HAMA, but such interferences have been the the circulating concentration of prolactin.
subject of several recent reviews [41,42]. Analysis by an Hyperprolactinemia can result from dopamine receptor
alternative method is usually sufficient to align the antagonists, such as phenothiazines, butyrophenone
measured result with the clinical picture, but in some antipsychotics, risperidone, verapamil, and
cases it may be necessary to prove that these antibodies metoclopramide, and from dopamine-depleting agents,
exist. In this case, a broad spectrum passive precipitant such as methyldopa and reserpine. Tricyclic
such as PEG can be used, but it is preferable to use an antidepressants, cocaine, opiates, and estrogens can also
active blocking agent such as The Scantibodies increase prolactin concentrations, but these elevations
Heterophilic Blocking Reagent Tubes (HBRT). The are usually modest (rarely exceeding 150 ng/mL) [49].
reagent is a proprietary mix of lyophilized mouse anti- Administration of dopamine substitutes or receptor
human IgM with high affinity for human anti-animal agonists such as l-dopa, apomorphine, and ergot
antibodies commonly used in reagent manufacture derivatives suppress prolactin secretion [37].
(mouse, goat, sheep, and rabbit) and to rheumatoid
factor. Each tube contains sufficient reagent to inactivate Psychological stress is a common cause of modest
500 L of patient sera and is incubated for 1 hour at increases in serum prolactin concentration, but the most
room temperature. In comparison to PEG, the specimen common pathological cause of an elevated serum
is not diluted, and the immunoassay signal is only prolactin is a pituitary adenoma. These are usually small,
minimally affected [43,44]. Although FDA-approved for slow growing, and benign. Prolactinomas are classified
this purpose, the value obtained from a sample treated by their size, and the serum prolactin concentration
with HBR should not be used as a reportable result. parallels tumor size quite closely. Microprolactinomas
1000
Prolactin
(<10 mm in diameter) are typically associated with specifications for analytical imprecision derived from
prolactin concentrations of 40 to 100 ng/mL. studies of biological variation [52] indicate a desired
Macroprolactinomas (>10 mm diameter) are typically assay imprecision of 3.5%. Most assays have similar
associated with concentrations > 250 ng/mL, and which imprecision values (% coefficient of variation) of 5% to
do occasionally exceed 1000 ng/mL. It is important to 8%, although there is much larger variation across the
distinguish between a macroprolactinoma and the entire group (typically 20%). This is most likely
compound macroprolactin; the two are quite distinct attributed to the presence of macroprolactin. UKNEQAS
entities. The discrepancy between a large pituitary specifically targets macroprolactin in its survey sets.
adenoma and only a mildly elevated prolactin (<100 Over the last 12 years, they have circulated some 10
ng/mL) should be a signal for further investigation for challenges containing macroprolactins and recently
either compression of the pituitary stalk by a non- introduced an interpretative challenge. This circulation
prolactin-secreting adenoma or to a heavily secreting shows 83% of labs in the scheme screen routinely for
macroprolactinoma producing a hook effect on macroprolactin above 700 mIU/L (34 ng/mL) by PEG
immunoassay. This distinction is especially important, precipitation [25].
since drug treatment is particularly effective for
prolactinomas, reducing both size and prolactin Recent literature stresses the importance of adequately
secretion, whereas a non-secreting adenoma probably identifying the presence of macroprolactin and
requires surgical intervention. Bromocriptine and appreciating its significance. In one retrospective study
cabergoline are the primary drug therapies for of patients previously diagnosed with prolactinoma,
prolactinomas [50]. A combined diagnostic and macroprolactin has been proven to account entirely for
treatment pathway is provided by Schlechte [32]. (Figure the hyperprolactinemia in 42% of patients, a third of
5) whom presented with signs and symptoms of
hyperprolactinemia [53]. This degree of misdiagnosis is
Prolactin is occasionally produced ectopically by oat cell no longer acceptable. To this end, the onus has been
carcinomas of the lung and renal carcinomas. Circulating placed on kit manufacturers to characterize their
concentrations of prolactin increase progressively with antibodies with respect to cross-reactivity with
the severity of chronic renal failure and as expected are macroprolactin, with a view to reducing it [54]. In
unaffected by hemodialysis [51]. Prolactin concentration addition, the Pituitary Society and other leading groups
can be moderately increased with herpes zoster of endocrinologists are calling for automatic reflexing of
infection, especially when prominent in dermatomes T4- a PEG precipitation test to exclude the presence of
T5, in polycystic ovary syndrome, and with excessive macroprolactin in all specimens with moderate
exercise or stress. Hyperprolactinemia is also observed hyperprolactinemia (>35 ng/mL). Drafting of
in patients with primary hypothyroidism, probably due appropriate guidelines are underway [55-57].
to the stimulation of pituitary lactotrophs by elevated
TSH and/or TRH concentrations. For this reason, References
workup for hyperprolactinemia should include 1 Shome B, Parlow AF. Human pituitary
assessment of thyroid status [32]. prolactin: the entire linear amino acid sequence.
J Clin Endocrinol Metab 1977; 45: 1112-5.
The most common symptoms of hyperprolactinemia in 2 Frantz AG, Kleinberg DL. Prolactin: evidence
premenopausal women are amenorrhea and infertility. that it is separate from growth hormone in
Galactorrhea occurs in about 80% of these patients, and human blood. Science 1970; 170: 745-7.
some have menstrual irregularities. The majority of 3 Nicoll CS. Bioassay of prolactin: analysis of the
prolactinomas are small at diagnosis in women, and pigeon crop sac response to local prolactin
headaches and neurological deficits are rare. By contrast, injection by an objective and quantitative
prolactinomas in men typically tend to be large at method. Endocrinol 1967; 80: 641-655.
diagnosis and nay cause cranial nerve dysfunction, 4 Whitaker MD, Klee GG, Kao PC, Randall RV,
visual-field loss, and panhypopituitarism, with Heser DW. Demonstration of biological activity
associated impotence, infertility, and loss of libido [32]. of prolactin molecular weight variants in human
sera. J Clin Endocrinol Metab 1983; 58: 826-
Prolactin Performance Goals 30.
Acceptable performance criteria (CLIA-88) for 5 Sinha VN, Selby FW, Lewis VJ, Vanderlaan
measurement of serum prolactin requires that WP. A homologous radioimmunoassay for
laboratories be accurate to within 3 SD of the peer human prolactin. J Clin Endocrinol Metab
group mean. Precision data from the CAP 2006-7 1973; 36: 509-16.
Participant Summary Reports are shown in Table 3 for 6 Tanaka T, Shishiba Y, Gout PW, Beer CT,
1500 laboratories returning results for immunometric Noble RL, Friesen HG. Radioimmunoassay and
assays to CAP in 2006-7. Only a third as many bioassay of human growth hormone and human
participated in the calibration and linearity verification prolactin. J Clin Endocrinol Metab. 1983; 56:
surveys, and the spread of these data shows the bias 18-20.
between methods. Comparative data are shown for the 7 Ishikawa K, Narita O, Noguchi H, Kato K.
312 laboratories participating in the 2006 United Practical enzyme immunoassay for prolactin in
Kingdom National External Quality Assurance Scheme human serum. Clin Chim Acta 1982; 121: 181-
(UKNEQAS), showing imprecision and bias. Desirable 7.
1001
Prolactin
8 Suh HK, Frantz AG. Size heterogeneity of 20 Schiettecatte J, van Opdenbosch A, Anckaert E,
human prolactin in plasma and pituitary De Schepper J, Poppe C, Velkeniers B, Smitz J.
extracts. J Clin Endocrinol Metab. 1974; 39: Immunoprecipitation for rapid detection of
928-35. macroprolactin in the form of prolactin-
9 Guyda HJ. Heterogeneity of human growth immunoglobulin complexes. Clin Chem 2005;
hormone and prolactin secreted in vitro: 51: 1746-8.
immunoassay and radioreceptor assay 21 Ichihara K, Miyai K. Detection of
correlations. J Clin Endocrinol Metab. 1975; asymptomatic prolactinoma by mass screening
41: 953-67. program. Jap J Clin Pathol 1990; 38: 667-74.
10 Farkouh NH, Packer MG, Frantz AG. Large 22 Hattori N, Ikekubo K, Ishihara T, Moridera K,
molecular size prolactin with reduced receptor Hino M, Kurahachi H. Correlation of antibody
activity in human serum: high proportion in titres with serum prolactin levels and their
basal state and reduction after thyrotropin- clinical course in patients with anti-prolactin
releasing hormone. J Clin Endocrinol Metab. autoantibody. Eur J Clin Endocrinol 1994; 130:
1979; 48: 1026-32. 438-45.
11 Hattori N, Inagabi C. Anti-prolactin (PRL) 23 Fahie-Wilson MN, John R, Ellis AR.
autoantibodies cause asymptomatic Macroprolactin; high molecular forms of
macroprolactinaemia; bioassay and clearance circulating prolactin. Ann Clin Biochem 2005;
studies of PRL immunoglobulin G complex. J 42: 175-92.
Endocrinol Metab. 1997; 82: 3107-10. 24 Carlson HE, Markoff E, Lee DW. On the nature
12 Andersen AN, Pedersen H, Djursing H, of serum prolactin in two patients with
Andersen BN, Friesen HG. Bioactivity of macroprolactinaemia. Fertil Steril 1992; 58: 78-
prolactin in a woman with an excess of large 87.
molecular size prolactin, persistent 25 Ellis AR, Fahie-Wilson M, Axcell M, Hill RP.
hyperprolactinemia and spontaneous Macroprolactin(s): composition and reactivity
conception. Fertil Steril. 1982; 38: 625-8. in immunoassays and laboratory interpretation
13 Leite V, Cosby H, Sobrinho LG, Fresnoza A, of results of an unusual patient serum. Ann
Amparo Santos M, Friesen HG. Clin Biochem 2006; 43: 57-62.
Characterisation of big, big prolactin in patients 26 Gibney J, Smith TP, McKenna. Clinical
with hyperprolactinemia. Clin Endocrinol 1992; relevance of macroprolactin. Clin Endocrinol
37: 365-72. 2005; 62: 633-43.
14 Cavaco B, Leite V, Amparo Santos M, 27 Ismail AAA, Walker PL, Fahie-Wilson MN,
Arranhado E, Sobrinho LG. Some forms of big Jassam N, Barth JH. Prolactin and
big prolactin behave as a complex of macroprolactin: a case report of
monomeric prolactin and immunoglobulin G in hyperprolactinemia highlighting the
patients with macroprolactinaemia or interpretation of discrepant results. Ann Clin
prolactinoma. J Endocrinol Metab 1995; 80: Biochem 2003; 40: 298-300.
2342-6. 28 Kavanagh L, McKenna TJ, Fahie-Wilson MN,
15 Hattori N. The frequency of Gibney J, Smith TP. Specificity and clinical
macroprolactinaemia in pregnant women and utility of methods for the detection of
the heterogeneity of its etiologies. J Clin macroprolactin. Clin Chem 2006; 52: 1366-72.
Endocrinol Metab. 1996; 81: 586-90. 29 Prazeres S, Amparo Santos M, Ferreira HG,
16 Hattori N, Ikekubo T, Ikekubo K, Moridera K, Sobrinho LG. A practical method for the
Hino M, Kurahachi H. Autoantibody to human detection of macroprolactinaemia using
prolactin in patients with idiopathic ultrafiltration. Clin Endocrinol 2003; 58: 686-
hyperprolactinemia. J Clin Endocrinol. Metab. 90.
1992; 75: 1226-9. 30 Comtois R, Robert F, Hardy J.
17 Jackson RD, Wortsman J, Malarkey WB. Immunoradiometric assays may miss high
Characterisation of a large molecular weight prolactin levels. Ann Intern Med 1993; 119:
prolactin in women with idiopathic 173-4.
hyperprolactinemia and normal menses. J 31 Unnikrishnan AG, Rajaratnam S, Seshadri MS,
Endocrinol Metab. 1985; 61:258-64. Kanagasapabathy AS, Stephen DC. The hook
18 Hattori N, Ikekubo K, Nakaya Y, Kitagawa K, effect on serum prolactin estimation in a
Inagaki C. Immunoglobulin G subclasses and patient with prolactinemia. Neurol India 2001;
prolactin (PRL) isoforms in 49: 78-80.
macroprolactinaemia due to anti-PRL 32 Schlechte JA. Prolactinoma. N Engl J Med
autoantibodies. J Clin Endocrinol Metab 2005; 2003; 349:2035-2041.
90: 3036-44. 33 St-Jean E, Blain F, Comtois R. High prolactin
19 Diver MJ, Ewins D, Worth RC, Bowles S, may be missed by radiometric assay in patients
Ahlquist JAO, Fahie-Wilson MN. Pregnancy in with macroprolactinomas. Clin Endocrinol
a patient with markedly raised concentrations of 1996; 44:305-9.
circulatory big-big (macro) prolactin a case 34 Gilson G, Schmit P, Thix J, Hoffman J-P,
report. Clin Chem. 2001; 47: 346-8. Humbel RL. Prolactin results for samples
1002
Prolactin
containing macroprolactin are method and Lombardi G. Prolactinomas resistant to

sample dependent. Clin Chem 2001; 47: 331-3. standard dopamine agonists respond to chronic
35 Dale JC, Pruett SK. Phlebotomyminimalist cabergoline treatment. J Clin Endocrinol Metab
approach. Mayo Clin Proc 1993; 68:249-55. 1997; 82: 876-83.
36 Gann DS, Lilly MP. The endocrine response to 51 Nagel TC, Freinkel N, Bell RH, Friesen H,
injury. Prog Crit Care Med 1984; 1: 15-47. Wilber JF, Metzger BE. Gynecomastia,
37 Frantz, AG. Physiology in medicine: prolactin. prolactin and other peptide hormones in
N Engl J Med 1978; 298: 201-7. patients undergoing chronic hemodialysis. J
38 Oestreicher M, Oestreicher S, Schmidt- Clin Endocrinol Metab 1973; 36: 428-32.
Gollwitzer M. Circadian rhythms of hormones 52 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
in pregnancy. Z Geburtschilfe Perinatol. 1985; Hernandez A, Jimenez CV. Current databases
189: 140-3. on biologic variation: pros, cons and progress.
39 Logue, FC, Fraser WD, Christie J, et al. The Scand J Clin Lab Invest 1999; 59: 491-500.
circadian rhythm of PTH (1-84) and 53 Donadio F, Barbieri A, Angioni R, Montovani
neprhogenous cyclic AMP in normal men and G, Beck-Peccoz P, Spada A, Lania AG. Patients
hyperparathyroid patients: temporal relationship with macroprolactinaemia: clinical and
with prolactin secretion. Proc ACB Natl Meet. radiological features. Eur J Clin Invest 2007;
p 102, 1990. 37: 552-7.
40 Ando T, Yasui J, Inokuchi N, Usa T, Ashizawa 54 John R, McDowell IFW, Scanlon MF, Ellis AR.
K, Kamihara S, Eguchi K. Non-specific Macroprolactin reactivities in prolactin assays:
activities against ruthenium crosslinker as a an issue for clinical laboratories and equipment
new cause of assay interference in an manufacturers. Clin Chem 2000; 46:884-5.
electrochemiluminescent immunoassay. Jap J 55 Casanueva FF, Molitch ME, Schlechte JA, Abs
Intern Med 2007; 46: 1225-9. R, Bonert V, Bronstein MD et al. Guidelines of
41 Kricka LJ. Human anti-animal antibody the Pituitary Society for the diagnosis and
interferences in immunological assays. Clin management of prolactinomas. Clin Endocrinol
Chem. 1999; 45: 942-56. 2006; 65: 265-73.
42 Levinson SS, Miller JJ. Towards a better 56 Suliman AW, Smith TP, Gibney J, McKenna
understanding of heterophile (and the like) TJ. Frequent misdiagnosis and management of
antibody interference with modern hyperprolinactinemic patients before the
immunoassays. Clin Chim Acta. 2002; 325: 1- introduction of macroprolactin screening:
15. application of a new strict laboratory definition
43 Ellis MJ, Livesey JH. Techniques for of macroprolactinemia. Clin Chem 2003; 49:
identifying heterophile antibody interference 1504-9.
are assay specific: study of seven analytes on 57 Gibney J, Smith TP, McKenna TJ. The impact
two automated immunoassay analysers. Clin on clinical practice of routine screening for
Chem 2005; 51: 639-41. macroprolactin. J Clin Endocrinol Metab 2003;
44 Preissner CM, OKane DJ, Singh RJ, Morris 90: 3927-32.
JC, Grebe SKG. Phantoms in the assay tube: 58 Smith TP, Suliman AM, Fahie-Wilson MN,
heterophile antibody interferences in serum McKenna TJ. Gross variability in the detection
thyroglobulin assays. J Clin Endocrinol Metab of prolactin in sera containing big big prolactin
2003; 88: 3069-74. (Macroprolactin) by commercial
45 Bowen RAR, Chan Y, Ruddel ME, Hortin GL, immunoassays. J Clin Endocrinol Metab 2002;
Csako G, Demsoky SJ, Remaley AT. 87: 5410-5.
Immunoassay interference by a commonly used 59 Olukoga AO, Kane JW. Macroprolactinemia:
blood collection tube additive, the validation and application of the polyethylene
organosilicone surfactant Silwet L-720. Clin glycol precipitation test and clinical
Chem 2005; 51: 1874-82. characterization of the condition. Clin
46 Kricka LJ, Park JY, Senior MB, Fontanilla R. Endocrinol 1999; 51: 119-126.
Processing controls in blood collection tubes 60 Sapin R, Kertesz G. Macroprolactin detection
reveals interference. Clin Chem 2005; 51: by precipitation with protein A-Sepharose: a
2422-3. rapid screening method compared to
47 Guyda, HJ, Friesen HG. Serum prolactin levels polyethylene glycol precipitation. Clin Chem
in humans from birth to adult life. Pediatr Res 2003; 49: 502-5.
1973; 7: 534-40. 61 Vierra JGH, Tachibana TT, Obara LH, Maciel
48 Schmidt-Gollwitzer, M., and Saxena, B.B. RMB. Extensive experience and validation of
Radioimmunoassay of human prolactin, Acta polyethylene glycol preparation as a screening
Endocrinol 1975; 80: 262-74. method for macroprolactinemia. Clin Chem
49 Molitch ME. Medication-induced 1988; 44: 1758-9.
hyperprolactinemia. Mayo Clin Proc 2005; 80: 62 Sustarsic LW, Fahie-Wilson M, Vanrieken L,
1050-7. Walker K, Del Rosario I, El Shami AS. Matrix
50 Colao A, Di Sarno, Sarnacchiaro F, Ferone D, effect of PEG precipitation on detection of
Di Renzo G, Merola B, Annunziato L, macroprolactin in IMMULITE and IMMULITE
1003
Prolactin
2000 prolactin assays. Clin Chem 2002; 48 IgG binding reagent. J Immunol 1984; 133:
(S6): A62. 969-74.
63 Schiettecatte J, DeSchepper J, Velkeniers B, 65 Lindmark R, Thoren-Tolling K, Sjoquist J.
Smitz J, van Steirteghem A. Rapid detection of Binding of immunoglobulins to protein A and
macroprolactin in the form of prolactin- immunoglobulin levels in mammalian sera. J
immunoglobulin G complexes by Immunol Methods 1983; 62: 1-13.
immunoprecipitation with anti-human IgG-
agarose. Clin Chem Lab Med 2001; 39: 1244-8. Appendix Data at
64 Bjorck L, Kronvall G. Purification and some www.acb.org.uk/annclinbiochem/Webwise/Elli
properties of streptococcal protein G, a novel sAppendix1.pdf
Table 1: Methods of Analysis for Prolactin

Method 1: Bioassay
Principle of analysis: Cell replication stimulated by prolactin
Comments: Historical; useful for establishing bioreactivity; insensitive for clinical samples
Principle of analysis: Competitive binding assay; isotope-labeled prolactin competes with endogenous hormone
for limited amount of antibody.
Comments: Rare and mainly historical; applicable to most laboratory samples; lacks specificity
Method 3: Immuno-radiometric assay
Principle of analysis: Antibody linked to solid phase reacts with prolactin in test sample. Second prolactin
antibody radioactively labeled reacts in proportion to amount of prolactin bound to solid phase.
Comments: Rare; more specific than method 1; radioactivity a disadvantage
Method 4: Non-isotopic immunometric assays
Principle of analysis: Prolactin forms sandwich complex between a capture antibody attached to a solid phase
support and second labeled antibody. Label may be an enzyme, fluorophore, or chemiluminescent tag.
Comments: Most common methods in use; easily automated; good sensitivity and specificity; two-step assays are
less prone to hook effect. Take care to exclude macroprolactin.
(See table 4 for comparison of most commonly used commercially available methods.)
Table 2: Methods of Analysis for Macroprolactin

Method 1:
Principle of analysis: Gel-filtration chromatography (GFC)
Comments: Simultaneous quantitation of PRL in each molecular fraction identified; reference method;
expensive, time consuming; not readily available
Method 2:
Principle of analysis: PEG precipitation based on molecular size
Comments: Readily available and inexpensive; nonspecific for Ig class; most common method in use; correlates
well with GFC; may be incompatible with some immunoassays (Abbott Axsym)
Method 3:
Principle of analysis: Specific precipitation of immunoglobulin G and IgG-macroprolactins using anti-hIgG-
agarose, Protein A-agarose, Protein G-agarose
Comments: Easy to use and inexpensive; fractions containing other Ig classes are not precipitated.
Method 4:
Principle of analysis: Ultrafiltration by centrifugation
Comments: Readily available; inexpensive and easily performed; macroprolactins may dissociate; dilution of
specimen is not required; compatible with all immunoassays
1004
Prolactin
Table 3: Methods Used for Prolactin Measurement and Their Performance as Reported by CAP
and UKNEQAS During 2006-7
CAP 2006 - 7 2006

Average of 7 spot Linearity / Calibration UKNEQAS
challenges Verification
Method % of Labs mean ng/mL cv % low ng/mL high ng/mL % of labs cv % % Bias
4a + 4b 12.1 24.5 5.7 1.9 160.9 17.0 5.7 15.0

4c 29.5 19.0 5.9 1.8 132.8 28.2 5.5 -10.0
4d 22.7 19.6 5.6 1.7 129.6 8.3 7.0 3.0
4e 16.3 19.8 5.7 1.8 123.5 16.3 7.5 -15.0
4f 10.5 25.0 6.0 2.0 145.6 24.4 3.5 10.0
4g 7.2 19.7 5.3 6.0 124.7 <1 --- 10.0
Others 1.7 5
All Results 100.0 21.0 13.4 100.0 20.0
a Abbott Axsym
b Abbott Architect
c Bayer Centaur
d Beckman-Coulter Access / DxI
e Diagnostic Products Immulite Series
g Vitros ECi
NEQAS data kindly supplied by Andy Ellis, UK NEQAS for Peptide Hormones, Edinburgh.
Table 4: Comparison of the Most Commonly Used Immunometric Assays for Prolactin
ULOQ ng/mL
Ab Hook Effect
Captue Ab*/ Detection Ab* / Conjugation Separation LLOQ ULOQ Extended by Calibration Type /
Method Addition Detection System Absent to:
solid phase Conjugate System method ng/mL ng/mL Automatic Frequency
Steps ng/mL
Dilution
mc mouse / mc rabbit / particles trap 4-Methylumbelliferyl-phosphate Wet
4a --- 1 0.6 200 no 10,000
microparticles ALP in fiber Fluorescent product Full curve 28 day
mc mouse / 2 with Acridinium (N-sulfonyl) carboxamide Electronic for lot#

mc rabbit /
4b paramagnetic --- interim magnet Chemiluminescence with 0.6 200 2,000 10,000 2 wet ajustors each
Acridinium ester
particles wash Peroxide and NaOH trigger kit and 28 day
mc mouse / Acridinium ester Electronic for lot#
pc goat /
4c paramagnetic Biotin - Streptavidin 2 magnet Chemiluminescence with 0.3 200 1,000 30,000 2 wet ajustors each
Acridinium ester
particles Acid and NaOH trigger kit and 28 day
mc mouse /
pc goat / pc goat anti-mouse Chemiluminescence with Lumi-Phos Wet
4d paramagnetic 1 magnet 0.25 200 no not stated
ALP IgG 530 substrate Full curve 28 day
particles
Electronic for lot#
mc mouse / mc goat / centrifugal Adamantyl-1,2-dioxetane phosphate
4e --- 1 0.16 150 3,000 20,500 2 wet ajustors each
plastic bead ALP wash substrate Fluorescent product
kit and 28 day
mc mouse / mc mouse / Electrochemiluminescence with Ru Electronic for lot#
4f paramagnetic Ruthenium Biotin - Streptavidin 2 magnet Electron amplification with 0.05 470 4,700 12,690 2 wet ajustors each
particles complex Tripropylamine kit and 28 day
Luminol derivative substrate
mc sheep / mc mouse / Peracid salt peroxide generator Wet
4g Biotin - Streptavidin 1 wash 1.4 329 no 20,680
well wall HRP Electron amplification with 3- Full curve 28 day
Chloro,4-hydroxy acetanilide
mc mouse / Acidic chelator dissociates Eu3+ Wet
mc mouse /
4h microtiter plate Biotin - Streptavidin 1 wash Micellar enhancer 0.04 250 no 2,500 Full curve each
Europium
well wall Time resolved fluorimetry at 613 nm assay
a Abbott Axsym ALP Alkaline phosphatase

b Abbott Architect HRP Horseradish Peroxidase
c Bayer Centaur * mc = monoclonal; pc = polyclonal
d Beckman-Coulter Access / DxI
e Diagnostic Products Immulite Series
g Vitros ECi
h Wallac DELFIA
1005
Prolactin
Table 5: Demonstration of Hook Effect with Prolactin Assay
Sample Measured Final "Correct" 1600

Dilution Prolactin Result Response 1400
Measured PRL ng/mL

ng/mL ng/mL ng/mL 1200
Neat 12 12 392,000
1000
x2 15 30 196,000
800
x10 36 360 39,200
x50 133 6,650 7,840 600
x250 1504 376,000 1,568 400

x500 784 392,000 784 200
0
0 200 400 600
Dilution Made
Table 6: Incidence of Hyperprolactinemia Explained by Macroprolactin by Immunoassay

Prevalence
Immunoassay Method
%
Method used:
Data abstracted from Fahie- Abbott Architect 16 - 17
Wilson et al. [23] Bayer Centaur 5
Bayer Immuno-1 26
Beckman-Coulter Access / DxI <1
Diagnostic Products Immulite Series 10
Roche Elecsys Systems I 16 - 17
Roche Elecsys Systems II 5
Tosoh AIA 24
Wallac DELFIA 15
Figure 1:
1a: Typical sandwich immunoassay format with antibodies directed to different epitopes of the prolactin molecule. The
capture antibody may be attached to a surface such as a well wall or bead. The second antibody is labeled with a marker
that facilitates quantitation of the amount of sandwich formation, hence determination of prolactin concentration in the
specimen.
1b: Interference from heterophile antibody. The capture and detection antibodies are linked together by a nonspecific
antibody, producing a signal indistinguishable from that produced by prolactin.
1006
Prolactin
Figure 2
Prolactin in
patient sample Wash ALP
Y
off
YY
Y ALP
Substrate
YY
Detection Ab ALP
Y
Prolactin Capture Ab
coated Microparticles
Capture on Filter
Fluorescent
Product
CH3
CH3
Alkaline
O + PO4
Phosphatase
OH P O O O
O O
OH
4-Methylumbelliferyl phosphate 4-Methylumbelliferone
Microparticle-enhanced immunoassay (MEIA) see Table 4, Method 4a. Sample, microparticles (capture molecules) and
enzyme-labeled (ALP) detection antibodies are transferred to the reaction cell. During the incubation, analytes bind to both
antibodies, creating an immune complex with the microparticles. An aliquot is transferred to the matrix cell, where the
immune complex binds irreversibly to the glass fiber matrix. After washing the matrix to remove unbound materials, an
alkaline phosphatase substrate, 4-methylumbelliferyl phosphate (MUP), is added to the matrix. The bound ALP-conjugate
catalyzes the hydrolysis of 4-methylumbelliferyl phosphate (MUP) to the fluorescent product 4-methylumbelliferyl (MU).
The MEIA optics measures the rate at which MU is generated on the glass fiber matrix.
Figure 3
Biotinylated Capture Ab- Sandwich

Capture Ab Acridinium label Formation
YY
Incubate
Y
Prolactin in
patient sample
Streptavidin
coated magnetic
Y
micro-particles
1007
Prolactin
Magnetic
Capture
Pre-trigger Y
Y Addition Y
RLU
Y Trigger
Addition
Measurement
Y Y
Two-step chemiluminescent immunometric assay Abbott Chemiflex Technology (Table 4, Method 4b).
In the first step, patient sample and paramagnetic microparticles coated with capture antibody are incubated. After washing
to remove unbound components (and reduce possibility of a hook effect from excess prolactin), detection antibody labeled
with an acridinium ester is added and incubated. Pre-trigger (peroxide) solution releases the complexed prolactin-labeled
detection antibody from the microparticles, which are drawn towards the magnet. Trigger solution (sodium hydroxide)
optimizes the pH for the release of chemiluminescence, which is detected by the liminometer and quantified in relative
light units (RLU). The acridinium ester used in this assay has been specially formulated with a sulfopropyl group to
improve its aqueous solubility and a sulfonamide-leaving group to improve its stability. Method 4c (Bayer Centaur)
follows a similar pathway.
Figure 4
Ru(tpy)2 2+
Ruthenium bis(2,2:6,2-
Biotinylated Capture Wash
Y
terpyridine)
Capture Ab Ab-Ru label off
YY
Y Photons
e- TPA
H+
Y Ru2+
base state
3+
RuTPA
TPA+
TPA
Y
Y
Prolactin in Electron donor

e-
patient sample Magnetic (Tripropylamine, TPA)
Capture + Platinum Electrode + ensures that the reaction
Streptavidin
Magnet continues to cycle,
coated magnetic
multiplying the signal
micro-particles
Two-step electrochemiluminescent immunometric assay Roche Diagnostics (Table 4, Method 4f).

In the first step, patient sample is incubated with biotinylated capture antibody. Next, streptavidin-coated paramagnetic
microparticles and detection antibody labeled with a ruthenium compound are incubated. A sandwich complex is formed,
which becomes bound to the magnetic particles by the interaction of streptavidin and biotin. After washing to remove
unbound components, the magnetic particles are attracted out of solution by a magnet onto the electrode. Application of a
voltage to the electrode then induces chemiluminescent emission, which is measured by the photomultiplier. The addition
of an electron donor (tripropylamine) ensures that the reaction continues to cycle, thus improving the sensitivity.
1008
Prolactin
Figure 5
A combined diagnostic and treatment pathway for prolactinoma.

Reproduced from Schlechte JA [32]: Prolactinoma, N Engl J Med 2003; 349:2035-41, with permission pending.
Figure 6
An 18-laboratory survey of 10 samples known to contain macroprolactin (defined as hyperprolactinemia that corrected on
treatment with PEG). Results are shown for nine different commercially-available immunometric assays.
Reproduced from Smith TP, Suliman AM, Fahie-Wilson MN, McKenna TJ [58]: Gross variability in the detection of
prolactin in sera containing big big prolactin (macroprolactin) by commercial immunoassays, J Clin Endocrinol Metab
87:5410-5, 2002, with permission pending.
1009
Prolactin
Figure 7
Calibration curve (o) for gel-filtration chromatography and the prolactin () curve of a serum containing a preponderance
of macroprolactin (bbPRL) and some monomeric prolactin (mPRL). Molecular weight markers are indicated as follows: A,
dextran blue; B, thyroglobulin; C, gamma globulin; D, ovalbumin; E, myoglobin; F, cyanocobalamin (vitamin B12).
Reproduced from Olukoga AO, Kane JW [59]. Macroprolactinemia: validation and application of the polyethylene glycol
precipitation test and clinical characterization of the condition, Clin Endocrinol 51:119-126, 1999, with permission.
Appendix
Identification & Measurement of Macroprolactin Two general strategies are available for the analysis of
macroprolactin. The first is to separate the prolactin
Different commercially-available assays have varied fractions by gel-filtration chromatography (GFC) and
responses to macroprolactins. This can be explained in determine the immuno-reactivity of each fraction. The
part by the orientation of the prolactin peptide within the second and more readily-available and cost-effective
macromolecule relative to the epitopes to which the strategy is to remove any macroprolactin present and
antibodies were raised, and to a lesser extent by the then determine the residual immuno-reactivity due to
steric hindrance offered by the detection label. It was monomeric prolactin. Ultrafiltration, and precipitation
initially assumed that the relative responses of with polyethylene glycol (PEG), Protein A, Protein G, or
commercial assays would be stable for the majority of anti-human IgG have all been used for this purpose.
specimens tested. Thus those labs using assays that react Both methods are described in more detail.
only weakly might be tempted to assume that an elevated
prolactin is unlikely to be caused by macroprolactin. Gel-Filtration Chromatography (GFC) (see Table 2,
However, the pattern does vary with some samples Method 1)
[25,34], and hyperprolactinemia due to macroprolactin This is considered the gold standard because it allows
can be observed in all assays tested to date. Figure 6 simultaneous quantitation of all three molecular forms of
shows results of an 18-laboratory survey of ten samples prolactin. It is, however, time consuming and expensive,
known to contain macroprolactin (defined as adding almost $300 to the cost of the analysis, and
hyperprolactinemia that corrected on treatment with requires equipment unlikely to be available in non-
PEG) [58]. Results are shown for nine different specialist laboratories. Many systems haven been used,
commercially-available immunometric assays, and there of which the description below is one example [59].
is a 2.3- to 7.8-fold variance in results obtained for
individual specimens (since these data were generated, Principle
Roche has reformulated the antibodies and reduced their GFC separates molecules based on their size or
cross-reactivity with macroprolactin fourfold). The molecular weight. The gel (stationary phase) is
results suggest that depending on the assay used, packed into a column equilibrated with elution
macroprolactin may account for up to 25% of reported buffer. The column is calibrated with molecular
hyperprolactinemias. Table 6 shows the estimated size markers either prior to or concurrently with
prevalence of hyperprolactinemia due to macroprolactin the mixture to be separated, which is applied to
in samples with elevated total serum prolactin by the top of the column. Crevices or pores within
individual assay method [23]. A meta-analysis of the gel particles can accommodate molecules
reported incidence in clinical studies shows good up to a certain size (the cut-off), while larger
agreement [26]. molecules (above the cut-off) are excluded from
the pores and thus pass through the column
1010
Prolactin
quickly in the void volume. Molecules of

smaller size penetrate the pores and retain on Calculation
the column; over time they dissociate from the 1. The area under the total prolactin
pores and elute from the column in order of curve is determined by integration.
decreasing molecular mass. The range of pore 2. The regions of interest are identified
sizes in the gel preparation determines the against the molecular size markers,
speed and resolution of the analysis. The monomeric (23 kDa), big (40 to 60
column filtrate is collected for fixed time kDa) and macro (>100 kDa) prolactin
intervals by use of an automatic fraction fractions.
collector. A UV detector signal provides 3. The area of the macroprolactin peak is
monitoring of the separation of the molecular calculated as a percentage of the total
weight protein markers. Alternate fractions are prolactin activity in the specimen.
then subjected to prolactin analysis by
immunometric assay, and this is usually a Interpretation
widely cross-reacting method such as the Samples are considered macroprolactin positive
DELFIA assay. when the area under the prolactin curve 30%
of the total area of prolactin elution.
Materials required The normal pattern of activity is 60% to 90%
Peristaltic pump prolactin monomer; 15% to 30% big prolactin;
Automated fraction collector and 0% to 10% macroprolactin
UV spectrophotometer set at 280 nm
Sephacryl S-300 HR Comment
Isotonic buffer Although considered the gold standard, there
Tris Buffer, 10 mmol/L, pH 7.4, are four main disadvantages:
containing 140 mmol/L sodium In sera with a low affinity complex, there is the
chloride, 1.25 mol/L calcium chloride, potential for dissociation of the prolactin from
0.5 mmol/L magnesium chloride and the antibody complexes during the lengthy GF
0.2% sodium azide run. There is little evidence that this often
Molecular weight markers, for example: happens in practice. There is inherent
Dextran blue (MW 2.5 106) cumulative imprecision (and cost) when
Bovine thyroglobulin (MW 6.7 105) measuring prolactin in 30 to 40 different
Bovine gamma globulin (MW 1.58 105) fractions. There may be disproportionate loss of
Chicken ovalbumin (MW 4.4 104) prolactin from some fractions by absorption
Horse myoglobin (MW 1.7 104) onto the gel during the run, which will skew the
Cyanocobalamin (MW 1350) percentage distribution pattern for prolactin. It
Prolactin standard 9000 mIU/L is labor intensive and costly because multiple
immunoassays are required, which limits its
Immunometric prolactin assay system application to research institutions. Although
Method longer columns give better molecular size
1. A 1.6 70 cm column is packed to a bed height separations, they take longer and produce more
of 55 to 60 cm and equilibrated at 4C with fractions. When using the S-300 gel, the big
isotonic buffer. prolactin peak may be shouldering on the
2. The column is calibrated with the molecular macroprolactin, or if only small in size may be
weight markers. completely incorporated into it. It is, however,
3. 1 mL of filtered serum is applied to the column useful for investigating specimens with unusual
and run at a flow rate of 30 mL/h 110 1.0 mL. macroprolactins. S-100 gel gives better
4. Fractions collected over the next 3.7 hr. separation within the middle masses and can be
used to expand the area between macro and
Protein in each fraction is measured by UV monomeric prolactin, but macroprolactin elutes
spectrophotometry at 280 nm to determine the location close to the void volume[28]. S-200 gel may be
of the molecular weight markers. Within the molecular a good compromise for routine use.
weight areas of interest, alternate fractions are analyzed
for prolactin by immunoassay PEG Precipitation Method
(Table 2, Method 2) Precipitation with PEG is easy and
Results inexpensive to perform, requiring only about 10 minutes
The amount of prolactin determined in each preparation time and adding about $11 to the cost of the
fraction analyzed is plotted against the fraction test. PEG is readily obtained and highly stable. It is,
number. however, relatively nonspecific.
Separation and location of the molecular weight
markers on this system is shown in Figure 7, Principle
with the DELFIA prolactin immunoassay The specimen is first analyzed by immunoassay
concentrations overlaid in a patient with to determine the total prolactin present. If
macroprolactinemia. hyperprolactinemia is demonstrated, a PEG
1011
Prolactin
precipitation is performed and the supernate Interpretation

reanalyzed using the same immunoassay The difference between the two results indicates the
method. Patient specimen is mixed with an contribution of macroprolactin to the original prolactin
equal volume of 25% PEG solution, vortex determination. If the corrected prolactin result falls
mixed, and incubated for several minutes. PEG within the reference range for the patient, they are
polymers sequester water from the specimen described as having macroprolactinemia. In the absence
into their matrix, causing macromolecules to of correction for dilution and recovery with controls or
exceed their solubility and precipitate, leaving calibrators, a drop of > 60% is considered definitive for
smaller molecules in solution. The size of the macroprolactinemia, while a drop of < 40% indicates
PEG will determine the molecular weight cut- presence of prolactin monomer. A grey zone exists
off. After centrifugation, the supernate is between 40% and 60%.
analyzed by immunoassay.
Comment
Materials Required Imprecision is 3% to 10%, and because of the small
Polyethylene glycol (PEG) 6000 or 8000 dilution made, it is applicable to specimens with only
available from most standard chemical moderate hyperprolactinemia [59,61]. Although there is
manufacturers under-recovery of monomeric prolactin (only
Balance approximately 75% is recovered), when this is corrected
DI or distilled water by concomitant analysis of calibrators or patient pools, it
Magnetic stir plate correlates well (r = 0.80) with GFC and is much less
Vortex or rotating shaker expensive and more easily performed. The main
Small centrifuge drawback is that some immunometric methods are
General laboratory supplies affected by the presence of the PEG in the supernate.
These include Abbott Axsym and DPC Immulite [62].
Working Reagent Preparation However, this does not preclude use of PEG with these
1. 250 g/L (25% w/v) polyethylene glycol solution assays, particularly if the purpose is to identify the
in water presence of macroprolactin rather than accurately
2. Add 25 g PEG to approx 60 mL deionized determine the monomeric prolactin concentration.
water at room temperature. Stir till dissolved on
a magnetic plate. QS to 100 mL with deionized Other Methods
water. Precipitation of macroprolactin-immunoglobulin G
3. Working reagent can be stored at room complexes has been described using a number of
temperature for 7 days. reagents (Table 2, Method 3). They are all relatively
inexpensive, adding an additional $25 to the cost of the
Method assay. The first assay used goat anti-human IgG-agarose
Combine sufficient patient serum with an equal volume with a binding capacity of 1,5 mg hIgG/mL of resin.
of working-strength PEG solution in a small test tube This assay had a 2-hour incubation and a dilution factor
(typically 150 L). of 20-fold, and therefore was not sufficiently sensitive to
Vortex mix for 30 seconds, or mix on a rotator for 10 detect mild macroprolactinemia [63]. A later
min. modification uses an IgG-agarose resin with a binding
Centrifuge for 5 min between 1500 g and 10,000 g capacity of 20 mg/mL of resin. This protein G
within 5 to 30 min of preparation. polypeptide binds to the Fc region of all subclasses of
Analyze the supernate by the same immunoassay method the human IgG molecule. Protein G-agarose is available
used with the native specimen. in ready-to-use suspension (Roche) [64]. 150 L patient
serum is mixed and incubated with 300 L Protein G-
Calculations agarose suspension, giving a threefold dilution. After
The specimen has been diluted by a factor of 2 during centrifugation for 2 min at 9500 g at 20C the supernate
the preparation, and this must be accounted for by is analyzed by immunoassay. The imprecision is < 2%,
multiplication of the instrument result by the dilution and results agree well with PEG precipitation when
factor. macroprolactin is the predominant form. Two sera which
Approximately 14 % (range 0% to 40%) of monomeric normalized on PEG precipitation but did not normalize
prolactin is co-precipitated with PEG [60]. The best with protein G-agarose were shown to react with anti-
practice is to work up a calibrator and QC material, IgA- and anti-IgMsepharose, demonstrating the
processed similarly, to determine the assay recovery. specificity of this method and the conclusive presence of
These should be factored into the calculation for the IgA- and IgM-associated macroprolactins [20].
specimen result.
Example Calculation: Protein A is another polypeptide that binds the Fc region
Monomeric PRL = 2(PEG Patient Instrument Reading) of human immunoglobulin molecules, especially hIgG1,
Set point value for calibrator or control hIgG2, and hIgG4, but only a fraction of IgG3 [65]. 300
Divided by L of a Protein A-sepharose suspension (Protein A CL-
2(PEG Control or Calibrator Instrument Reading) 4B in 20 mmol/L borate buffer) is added to 150 L
Macro PRL = Original untreated PRL result patient serum. After incubation for 15 min at room
monomeric PRL treated result temperature, the sample is centrifuged at 2000 g for 5
1012
Prolactin
min at 20C. Prolactin is determined in the supernate by minutes. Spin times of up to 5 hours have been used at
immunoassay. This assay uses a shorter incubation time 450 g, but it is likely that the filter clogs on prolonged
and produces only a threefold dilution of sample so is contact and results become more variable. The clear
more applicable when there is only mild filtrate is analyzed for prolactin content by
hyperprolactinemia. Total imprecision is better than immunoassay. Unlike some other methods, the specimen
12%. The assay appears to specifically target IgG is not diluted so is ideal for patients with only moderate
complexes, while other forms of macroprolactins are not hyperprolactinemia. Additionally, since there are no
removed [60]. additives to interfere with immunoassays, it can be used
where precipitating techniques are not able to be applied.
Ultrafiltration (Table 2, Method 4) employs a standard Comparative studies show good agreement with PEG
membrane filter (available from Amicon in a variety of precipitation and GFC, most notably when
pore sizes) with a molecular weight cut-off of around macroprolactin is the predominant species. Some
100 kDa. These devices are freely porous to 45kDa and comparative studies demonstrate over-recovery of
completely retain above 100 kDa, with variable monomeric prolactin, which is most likely caused by
penetrance in between. Specimen (typically 0.5 to 1 mL) dissociation of macroprolactin during centrifugation or
is loaded into a reservoir above the filter, and a small because of dilution in PBS, or because high dilution
tube is attached to the base of the device to collect the factor (20) used in some studies introduces
filtrate. Some studies use a dilution of 25 L specimen in imprecision. Values for imprecision range from a CV of
500 L PBS. The entire apparatus is loaded into a fixed- 3% to 9% to 16%, depending on the methodology and
angle rotor centrifuge and spun at 1000 g for 30 to 45 choice of specimen [29]. Additional cost are
approximately $16.
1013
Prostate-Specific Antigen (PSA)

Hassan M.E. Azzazy
Name: Serine protease, PSA
Clinical significance: Refer to Chapter 53, Neoplasia, in the 5th edition of Clinical
Molecular mass: 28.43 kDa
Subunit: Monomer (single chain)
Isoelectric pH: Range from 6.8 to 7.5
Coding gene location and size: Chromosome 19, 6 kb
i many of the available methods are difficult to compare
because their results are dependent on the standard that
Human prostate-specific antigen (PSA) is produced is used, the specificity of the antibodies, and the purity
exclusively by the epithelial cells lining the prostatic of the antigen employed in the generation of antibodies.
acini and ducts of prostatic tissue. PSA is a single-chain
glycoprotein that contains 237 amino acids and N-linked PSA occurs in serum in different molecular forms.
oligosaccharide [1]. Functionally, PSA is a kallikrein- Approximately 70% is bound to protease inhibitors,
like serine protease with chymotrypsin- and trypsin-like primarily 1-antichymotrypsin and 2-macroglobulin,
enzymatic activity [2]. Because of its high specificity for and a small amount is complexed with 1-antitrypsin and
prostate tissue, PSA is the preferred serum marker for protein C. Free PSA, defined as the portion of PSA not
prostate cancer [3,4]. PSA is present in seminal fluid in bound to serum proteins, accounts for ~30% of total
high concentrations and cleaves the high-molecular-mass PSA [5,6]. Antibodies can react with different epitopes,
seminal vesicle protein, semenogelin, the major and the different molecular forms of PSA can influence
structural protein of the seminal coagulum (to increase antibody-binding characteristics.
sperm motility). Semenogelin is formed at ejaculation
and appears to be the physiological substrate for PSA Some manufacturers advocate the use of the free PSA
[4]. molecule as the standard, whereas others recommend the
PSAanti-chymotrypsin complex. The former is better
In serum, PSA can be joined to other proteins characterized, but the latter is the form that predominates
(complexed PSA [cPSA]), or it can exist as free PSA in plasma. Clearly, some effort is needed by the
(fPSA). Several assays are available that measure free, manufacturers of the kits to promote intermethod
complexed, and/or total PSA. Assays for free or cPSA agreement [5].
are used when the total PSA result is marginally raised,
but not when total PSA result is highly elevated. Free Early methods used to measure PSA included isoelectric
PSA measurements can be used to improve the focusing, immunodiffusion, immunoelectrophoresis, and
specificity of PSA for prostate cancer, especially when rocket immunoelectrophoresis [7]. Subsequent tests for
total PSA values are between 4 and 10 ng/mL (gray PSA incorporated standard enzyme-linked
zone). Free PSA is expressed as a ratio of total PSA immunosorbent assays and solid-phase immunometric
[%fPSA = fPSA/total PSA 100]. Men with prostate assays with polystyrene beads.
cancer may have a smaller proportion of fPSA and more
cPSA when compared with men with benign prostatic Zundel et al. [8] developed an EIA for PSA that uses
hyperplasia (BPH) or prostatitis. microtiter wells coated with monoclonal antibody to
PSA. After the specimen is added to the wells, a second
monoclonal antibody to PSA, labeled with horseradish
Methods for PSA Analysis peroxidase, is added along with the substrate
Immunoassays for PSA are widely available and tetramethylbenzene.
typically use non-isotopic labels such as enzymes,
fluorescent labels, or chemiluminescence. Unfortunately, A solid-phase, two-site, immunoradiometric assay
(IRMA) that uses two murine monoclonal antibodies for
different epitopes on the PSA molecule has been
i
Prostate-Specific Antigen (PSA) described [9]. A high-dose hook effect has been
Previous and current authors of this method: reported for this assay, with artificially low measured
First edition: Not done PSA values found with serum concentrations above 5000
Methods edition: Not done ng/mL [10]. An EIA version of the IRMA PSA assay has
Second edition: Not done also been developed. The principle is the same as the
Third edition: William R. Jackson, John A. Lott IRMA; however, the label is alkaline phosphatase rather
Fourth edition: William R. Jackson, John A. Lott than 125I, and the substrate is p-nitrophenyl phosphate.
1014
Another FIA for PSA is available that uses monoclonal clinical outcomes. Sotelo et al. (2007) have compared
antibodies to PSA coated on magnetic beads. The the free and total PSA values obtained using two
resulting complex binds with an anti-PSAalkaline commercial PSA assays. Using a free/total PSA ratio of
phosphatase conjugate; 4-MUP is the substrate. 20% as the threshold for biopsy, the free/total PSA
ratio was found to be discordant between the assays,
PSA can be measured using a microparticle-capture EIA leading to a discrepancy in biopsy recommendations and
system. PSA binds to inert microparticles that hold cancer detection rates [16].
monoclonal antibody to PSA. A second antibody to
PSA, labeled with ALP, binds to the PSA on the Reference and Preferred Method
microparticles. A 4-MUP substrate is used. Currently, there is no reference method for PSA. Several
sensitive PSA immunoassays are commercially available
A chemiluminescent immunoassay is also available for and generate acceptable results.
PSA. Polyclonal antibodies to PSA labeled with
acridinium ester are used, plus a second antibody to Specimen
PSA. The labeled PSAanti-PSA complex is separated PSA is stable in serum for up to 24 hours at 2C to 8C.
magnetically, then chemiluminescence is generated and For longer and extended storage periods, serum should
measured with a luminometer. A summary of the be stored frozen at 20C and at 70C, respectively.
analytical principles of the commercially available PSA For complexed PSA (cPSA), serum is stable for up to 48
assays is shown in Table 1. hours at 2C to 8C and at 70C for extended storage.
For PSA measurements, in particular for free PSA, blood
Circulating levels of cPSA can be determined either should be centrifuged within 3 hours of collection to
directly by cPSA specific assays [11] or indirectly by isolate serum or plasma. Free PSA is more susceptible to
subtracting fPSA from total PSA using assays decay than cPSA; decay is slower in plasma than in
standardized against one another [12]. An assay that serum [17].
measures cPSA has been developed. The assay measures
PSA complexes with 1-antichymotrypsin and other The serum half-life of PSA is 2.2 to 3.2 days [18]. Small
minor PSA complexes. It employs an antibody to block but significant increases in PSA may occur following
and inactivate fPSA in the sample [13]. Other cPSA prostatic manipulation, with greater increases occurring
assays are available on automated platforms and in in men with enlarged prostates [19]. In addition,
ELISA format. diagnostic intervention such as transrectal ultrasound or
biopsy can also result in increased PSA concentrations.
Long-term monitoring may represent a challenge PSA is occasionally found in urine in men and is an ideal
because of between-method variations caused by forensic marker in rape cases. It is a component of
calibration using different standard and differences in seminal plasma, even in vasectomized males. PSA is
method design. Additionally, different commercial PSA detectable in semen stains for as long as 1 year and is
immunoassays employ monoclonal-monoclonal or stable in the vagina for approximately 2 days.
monoclonal-polyclonal antibody combinations which Matsuzawa et al. [20] described a latex particle
may differ in their recognition of the free and complexed agglutination test for PSA that is based on the use of an
PSA forms. Two international standards representing the antibody to PSA.
main immunoreactive PSA components in serum, free
PSA and PSA 90:10 (ratio of bound-to-free PSA), are It is preferred to collect blood prior to prostate biopsy or
now available for calibration of PSA immunoassays.[14] cystoscopy. Although DRE does not elevate PSA serum
Although these reference materials may help to levels, it is prudent to wait several days prior to blood
minimize between-method differences, studies designed collection for PSA [21]. Blood drawing for PSA should
to characterize the molecular epitope structure and be delayed for several weeks following prostate biopsy,
specificities of diagnostically relevant antibodies specific surgery, or resolution of prostatitis [22].
for PSA and its derivatives could improve comparability
of PSA immunoassay results [15]. Interferences
Serum PSA values may be decreased by 50% in men
Although several commercial assays for PSA and its using finasteride (Proscar) for over 6 months or after
derivatives in serum are available, physicians making surgical or medical castration. Total and free PSA serum
clinical decisions should be alerted to the potential levels may increase upon transurethral resection of the
standardization bias among PSA assays. For example, prostate (TURP), needle biopsy, ejaculation, or exercise;
AxSYM total and free PSA are standardized using the cPSA levels are less affected. Antiandrogens, LHRH
World Health Organization (WHO) reference material agonists and antagonists and finasteride decrease cPSA
and the Stanford 90:10 reference material, respectively. levels. Digital rectal exam (DRE) may not cause
IMMULITE total and free PSA assays are referenced clinically significant increases in PSA [23]. The
against the WHO National Institute for Biological presence of heterophilic antibodies in serum may
Standards and Control First International Standard interfere with results generated using immunoassays
(NIBSC 1st IS) 96/670 and WHO NIBSC 1st IS 96/688, employing murine antibodies.
respectively [16]. Measurements for PSA and its
variations obtained by commercial assays are not
necessarily interchangeable and may lead to different
1015
Reference Intervals test for cancer rises to 40% to 50% or higher when
Reference intervals for PSA depend on the assay used. serum levels exceed 10 ng/mL [29].
Reference intervals for PSA in males are listed as 0 to 4
ng/mL when measured using immunoassays. Age- PSA-level increases with age and day-to-day
specific reference intervals recommended using intraindividual variation have been reported. According
chemiluminescent immunoassay are as follows: 40 to 49 to the American Society for Therapeutic Radiation,
years, 2.7 ng/mL; 50 to 59 years, 3.7 ng/mL; 60 to recurrence after radiation therapy is defined as three
69 years, 5.1 ng/mL; and 70 to 79 years; 7.0 ng/mL. consecutive increases in PSA above nadir (<0.5 ng/mL)
PSA is found in women in low concentrations, with [30].
values typically less than 0.15 ng/mL [24].
Serum PSA concentrations correlate directly with the
The standard cutoff for biopsy consideration, PSA value various clinical stages of prostate cancer, but they cannot
of 4.0 ng/mL, is equivalent to a cPSA concentration of be relied upon to determine the clinical stage of the
3.2 ng/mL. The lower threshold of 2.5 ng/mL is disease. PSA concentrations also correlate with the
equivalent to cPSA value of 2.2 ng/mL [23,25]. weight of prostatic tissue resected during a transurethral
resection of the prostate and with the volume estimated
Interpretation by transrectal ultrasonography. The ratio of PSA
An ideal tumor marker is specific for a tissue, tumor, or concentration to the volume of prostatic tissue is useful
both. It is released from the tumor into the blood or in differentiating patients with benign and malignant
urine, has a short half-life, and indicates the presence of diseases of the prostate [31,32]. Increasing PSA values
tumor before clinical symptoms; also, its concentrations are associated with advancing pathological stages of
fluctuate in proportion to the tumor mass. Although a prostate cancer and correlate positively with increasing
perfect tumor marker has not yet been discovered, PSA tumor volume [33]. PSA concentrations are helpful in
exemplifies many qualities of an ideal tumor marker. determining the necessity of radionuclide bone scan in
PSA is useful for staging, monitoring, and screening for newly diagnosed, untreated prostate cancer.
prostate cancer, as well as detecting recurrent disease.
Unfortunately, PSA is specific to prostate tissue not to PSA values should decrease to undetectable
prostate cancer. It is also found in abnormal concentrations following radical prostatectomy. Since
concentrations in normal and benign changes of the the serum half-life of PSA is relatively long (3.2 days),
prostate such as benign prostatic hypertrophy (BPH) assessment of PSA measurements should be delayed
[26]. Osterling et al. [27] reported that not only cancer until 2 or 3 weeks after surgery. Positive PSA values of
causes an elevation in the serum PSA; it can be elevated 0.2 to 0.6 ng/mL after radical prostatectomy are
from digital rectal examinations, cystoscopic indicative of residual disease. PSA is also useful in
examinations, and prostatic biopsies and trauma. The monitoring prostatic cancer patients who have
prevalence of foci of malignant cells is an incredible undergone radiotherapy and anti-androgen therapy
40% in men over age 50. Because the incidence of [33,34].
clinical prostate cancer in the United States was 165,000
cases in 1993, there is clearly a great deal of latent The use of PSA as a screening test for prostate cancer is
disease. Although the normal range of PSA is controversial. Although it demonstrates good sensitivity,
conventionally considered between 0.0 and 4.0 ng/mL, PSA lacks clinical specificity. Also, the sensitivities and
there is no cutoff PSA value at which prostate cancer is specificities of serum PSA assays change when different
not present, but rather there is a continuum of prostate cutoff points are used. Armbruster [35] illustrated a
cancer risk at all values of PSA [28]. hypothetical use of PSA for screening the general male
population for prostate cancer.
The usefulness of PSA as an early detector of prostate
cancer by itself is questionable, owing to the overlap in Measurement of PSA cannot be the sole factor used for
PSA values seen in patients with BPH and in those with the diagnosis of disease but must be used together with
organ-confined prostate cancer (see 27a). PSA values clinical findings. PSA must always be used in
combined with digital rectal examination findings and/or conjunction with a direct rectal examination (DRE) for
transrectal ultrasonography may lead to early detection the initial screening of patients for prostate cancer. The
of prostate cancers. The sensitivity and specificity of a American Cancer Society recommends that the PSA test
screening test for prostatic cancer is markedly increased and DRE should be offered annually to men of age 50
when these test combinations are used. years with life expectancy of 10 years or more. Men at
higher risk, of sub-Saharan African descent, and men
PSA diffuses into the systemic circulation in increased with a first-degree relative diagnosed with prostate
amounts only after significant changes have occurred in cancer at < 65 years should begin testing at age 45 years
the architecture of the prostate gland. Approximately [29].
38% to 48% of patients with organ-confined prostate
cancer have PSA values in the reference interval [27]. The fPSA and cPSA tests are frequently ordered when
Only 30% to 40% of men who present with moderate total PSA values are between 4 and 10 ng/mL. Tests
PSA elevations (4 to 10 ng/mL) actually have prostate from the same manufacturer should be used when
cancer confirmed by biopsy. The specificity of the PSA measuring total and fPSA values for the same patient.
The relative proportions of fPSA and cPSA vary
1016
between men with prostate cancer and those with other 4 Lilja H. A kallikrein-like serine protease in
benign prostatic diseases. Compared with men with prostate fluid cleaves the predominant seminal
prostatic cancer, patients with prostate hypertrophy or vesicle protein. J Clin Invest 1985;76:1899-903.
prostatitis have a higher proportion of free PSA and less 5 Lilja H, Christensson A, Dahln U, Matikainen
cPSA. Assays for cPSA are approved by the U.S. Food MT, Nilsson O, Pettersson K, Lvgren T.
and Drug Administration (FDA) for detection of prostate Prostate-specific antigen in serum occurs
cancer in men > 50 years in conjunction with DRE, as predominantly in complex with 1-anti-
well as for monitoring therapy in prostate cancer chymotrypsin. Clin Chem 1991;37:1618-25.
patients. 6 Jung K, Brux B, Lein M, Rudolph B,
Kristiansen G, Hauptmann S et al. Molecular
PSA kinetics provides a dynamic picture of prostate forms of prostate-specific antigen in malignant
cancer activity, and PSA doubling time (PSADT) has and benign prostatic tissue: biochemical and
been proposed as a determinant of the presence and diagnostic implications. Clin Chem 2000;46:47-
pattern of disease recurrence after definitive therapy 54.
[36]. Although definitive thresholds are yet to be 7 Kuriyama M, Wang MC, Papsidero LD, Killian
established, patients with a shorter PSADT should be CS, Shimano T, Valenzuela L et al.
treated aggressively, and those with a longer PSADT Quantitation of prostate-specific antigen in
may represent a slow-growing process for which delayed serum by a sensitive enzyme immunoassay.
or local therapy may be more appropriate. Table 2 Cancer Res 1980;40:4658-62.
presents variations of PSA measurements and their 8 Zundel D, Jarry H, Kestler D, Holzapfel G,
proposed clinical applications. Bartels H, Scheit KH, Wuttke W. Development
and evaluation of an enzyme-linked
Performance Goals immunoassay for the prostate-specific antigen
Survey data from the 2007 College of American utilizing two monoclonal antibodies. Urol Res
Pathologists participant summary report show 1990;18:327-30.
imprecision values (% coefficient of variation [CV]) for 9 Lindstedt G, Jacobsson A, Lundberg PA,
total prostate-specific antigen procedures ranging from Hedelin H, Pettersson S, Unsgaard B.
3.7% to 9.7% at a mean concentration of 10 ng/mL. Determination of prostate-specific antigen in
Imprecision values for free PSA assays range from 3.2% serum by immunoradiometric assay. Clin Chem
to 7.0% at a mean concentration of 2 ng/mL; imprecision 1990;36:53-8.
values for complexed PSA assays range from 4.5% to 10 Ooi DS, Escares EA. High-dose hook effect
5.1% at a mean concentration of 6.5 ng/mL; and those in IRMA-count PSA assay of prostate-specific
for PSA ratio range from 4.3% to 8.7% at a mean ratio antigen. Clin Chem 1991;37:771-2.
of 0.2 [37]. 11 Brawer MK, Meyer GE, Letran JL, Bankson
DD, Morris DL, Yeung KK, Allard WJ.
Acceptable Clinical Laboratory Improvement Measurement of complexed PSA improves
Amendments performance criteria (CLIA 88) for specificity for early detection of prostate cancer.
measurement of total, complexed, and free prostate- Urology 1998;52:372-8.
specific antigen require that laboratories be accurate to 12 Okihara K, Fritsche HA, Ayala A, Johnston
within the greater of 3 SD or 0.2 ng/mL of the peer- DA, Allard WJ, Babaian RJ. Can complexed
group mean. PSA ratio is currently not graded. Within- prostate-specific antigen and prostatic volume
subject biological variation has been determined to be enhance prostate cancer detection in men with
approximately 14% [38]. Desirable specifications for total prostate-specific antigen between 2.5 and
analytical imprecision derived from studies of biological 4.0 ng/mL? J Urol 2001;165:1930-6.
variation indicate an assay imprecision of no greater than 13 Parsons JK, Brawer MK, Cheli CD, Partin AW,
7.0% [39]. Djavan R. Complexed prostate-specific antigen
(PSA) reduces unnecessary prostate biopsies in
References the 2.6-4.0 ng/mL range of total PSA. BJU Int
1 Blanger A, van Halbeek H, Graves HC, 2004;94:47-50.
Grandbois K, Stamey TA, Huang L et al. 14 Rafferty B, Rigsby P, Rose M, Stamey T,
Molecular mass and carbohydrate structure of Gaines Das R. Reference reagents for prostate-
prostate-specific antigen: studies for specific antigen (PSA):establishment of the first
establishment of an international PSA standard. international standards for free PSA and PSA
Prostate 1995;27:187-97. (90:10). Clin Chem 2000;46:1310-7.
2 Akiyama K, Nakamura T, Iwanaga S, Hara M. 15 Berger P, Sturgeon C, Bidart JM, Paus E, Gerth
The chymotrypsin-like activity of human R, Niang M et al. The ISOBM TD-7 Workshop
prostate-specific antigen, gamma- on hCG and related molecules. Towards user-
seminoprotein. FEBS Lett 1988;42:154-160. oriented standardization of pregnancy and
3 Osterling JE. Prostate-specific antigen: a critical tumor diagnosis: assignment of epitopes to the
assessment of the most useful tumor marker for three-dimensional structure of diagnostically
adenocarcinoma of the prostate. J Urol and commercially relevant monoclonal
1991;145:907-23. antibodies directed against human chorionic
1017
gonadotropin and derivatives. Tumour Biol Schrder FH, Hugosson J, Roobol MJ, et al.
2002;23:1-38. Screening and prostate-cancer mortality in a
16 Sotelo RJ, Mora KE, Prez LH, Novoa J, randomized European study. New England
Carmona O, De Andrade R et al. Assay Journal of Medicine 2009; 360(13):13201328.
standardization bias: different prostate cancer
detection rates and clinical outcomes resulting Andriole G, Crawford E, Grubb R, et al. Seven
from different assays for free and total prostate- year mortality results and related findings from
specific antigen. Urology 2007;69:1143-6. the prostate
17 Piironen T, Pettersson K, Suonp M, Stenman component of the PLCO randomized cancer
UH, Oesterling JE, Lvgren T, Lilja H. In vitro screening trial. New England Journal of
stability of free prostate-specific antigen (PSA) Medicine. Published online ahead of print
and prostate-specific antigen (PSA) complexed March 18, 2009.
to alpha 1-antichymotrypsin in blood samples.
Urology 1996;48:81-7. 28 Thompson IM, Ankerst DP, Chi C, Lucia MS,
18 Pontes JE, Jabalameli P, Montie J, Foemmel R, Goodman PJ, Crowley JJ et al. Operating
Howard PD, Boyett J. Prognostic implications characteristics of prostate-specific antigen in
of disappearance rate of biologic markers men with an initial PSA level of 3.0 ng/mL or
following prostatectomy. Urology 1990;36:415- lower. JAMA 2005;294:66-70.
9. 29 Smith RA, Cokkinides V, Eyre HJ. Cancer
19 The Internal Medicine Clinic Research screening in the United States, 2007: a review
Consortium. Effect of digital rectal examination of current guidelines, practices, and prospects.
on serum prostate-specific antigen in a primary CA Cancer J Clin 2007;57:90-104.
care setting. Arch Intern Med 1995;155(4):389- 30 Freedland SJ, Sutter ME, Dorey F, Aronson
92. WJ. Defining the ideal cutpoint for determining
20 Matsuzawa S, Itoh Y, Miyauchi C, Hara M, PSA recurrence after radical prostatectomy.
Inoue T, Tsuda R. Immunologic determination Prostate-specific antigen. Urology 2003;61:365-
of seminal secretions by a latex microtiter 9.
technique. J Forensic Sci 1982;27:848-54. 31 Vesey SG, Goble M, Ferro MA, Stower MJ,
21 Lilja H, Haese A, Bjrk T, Friedrich MG, Hammonds JC, Smith PJ. Quantification of
Piironen T, Pettersson K et al. Significance and prostatic cancer metastatic disease using
metabolism of complexed and noncomplexed prostate-specific antigen. Urology 1990;35:483-
prostate-specific antigen forms and human 6.
glandular kallikrein 2 in clinically localized 32 Veneziano S, Pavlica P, Querz R, Nanni G,
prostate cancer before and after radical Lalanne MG, Vecchi F. Correlation between
prostatectomy. J Urol 1999;162:2029-34. prostate-specific antigen and prostate volume,
22 Oberpenning F, Schmid HP, Fuchs-Surdel W, evaluated by transrectal ultrasonography:
Hertle L, Semjonow A. The impact of usefulness in diagnosis of prostate cancer. Eur
intraoperative manipulation of the prostate on Urol 1990;18:112-6.
total and free prostate-specific antigen. Int J 33 Stamey TA, Kabalin JN, McNeal JE, Johnstone
Biol Markers 2002;17:154-60. IM, Freiha F, Redwine EA, Yang N. Prostate-
23 Price CP, Allard J, Davies G, Dawnay A, Duffy specific antigen in the diagnosis and treatment
MJ, France M et al. Pre- and post-analytical of adenocarcinoma of the prostate. II. Radical
factors that may influence use of serum prostatectomy treated patients. J Urol
prostate-specific antigen and its isoforms in a 1989;141:1076-83.
screening programme for prostate cancer. Ann 34 Stamey TA, Kabalin JN, Ferrari M, Yang N.
Clin Biochem 2001;38:188-216. Prostate-specific antigen and the diagnosis and
24 Mayo Clinic Test Catalog, 2007. treatment of adenocarcinoma of the prostate.
25 Polascik TJ, Oesterling JE, Partin AW. IV. Anti-androgen treated patients. J Urol
Prostate-specific antigen: a decade of discovery: 1989;141:1088-90.
what we have learned and where we are going. 35 Armbruster DA. Prostate-specific antigen:
J Urol 1999;162:293-306. biochemistry, analytical methods, and clinical
26 Duffy MJ. New cancer markers. Ann Clin application. Clin Chem 1993;39:181-95.
Biochem 1989;26:379-87. 36 Ramrez ML, Nelson EC, Devere White RW,
27 Oesterling JE, Chan DW, Epstein JI, Kimball Lara PN Jr, Evans CP. Current applications for
AW Jr, Bruzek DJ, Rock RC et al. Prostate- prostate-specific antigen doubling time. Eur
specific antigen in the preoperative and Urol 2008;54:291-302.
postoperative evaluation of localized prostatic 37 College of American Pathologists. 2007
cancer treated with radical prostatectomy. J Participant Survey. Northfield, IL: CAP; 2007.
Urol 1988;139:766-72. 38 Panteghini M, Pagani F, Bonora R.
27a. Preanalytical and biological variability of
http://www.cancer.gov/cancertopics/factsheet/d prostatic acid phosphatase and prostate-specific
etection/PSA antigen in serum from patients with prostatic
1018
pathology. Eur J Clin Chem Clin Biochem 43 Brawer MK, Cheli CD, Neaman IE, Goldblatt J,
1992;30:135-9. Smith C, Schwartz MK et al. Complexed
39 Rics C, Alvarez V, Cava F, Garca-Lario JV, prostate-specific antigen provides significant
Hernndez A, Jimnez CV et al. Current enhancement of specificity compared with total
databases on biologic variation: pros, cons and prostate-specific antigen for detecting prostate
progress. Scand J Clin Lab Invest 1999;59:491- cancer. J Urol 2000;163:1476-80.
500. 44 Benson MC, Whang IS, Pantuck A, Ring K,
40 Loeb S, Catalona WJ. Prostate-specific antigen Kaplan SA, Olsson CA, Cooner WH. Prostate-
in clinical practice. Cancer Lett 2007;249:30-9. specific antigen density: a means of
41 Catalona WJ, Smith DS, Ratliff TL, Dodds KM, distinguishing benign prostatic hypertrophy and
Coplen DE, Yuan JJ et al. Measurement of prostate cancer. J Urol 1992;147:815-6.
prostate-specific antigen in serum as a 45 DAmico AV, Renshaw AA, Sussman B, Chen
screening test for prostate cancer. N Engl J Med MH. Pretreatment PSA velocity and risk of
1991;324:1156-61. death from prostate cancer following external
42 Uzzo RG, Pinover WH, Horwitz EM, Parlanti beam radiation therapy. JAMA 2005;294:440-7.
A, Mazzoni S, Raysor S, et al. Free prostate- 46 Oudard S, Banu E, Scotte F, Banu A, Medioni
specific antigen improves prostate cancer J, Beuzeboc P et al. Prostate-specific antigen
detection in a high-risk population of men with doubling time before onset of chemotherapy as
a normal total PSA and digital rectal a predictor of survival for hormone-refractory
examination. Urology 2003;61:754-9. prostate cancer patients. Ann Oncol
2007;18:1828-33.
Table 1
Table 1: Analytical Principles of Typical Commercially Available PSA Assays
Method 1:
Principle: RIA (polyclonal Abs)
Signal: 125I radiation
Formula:
Ab + PSA + PSA - 125I Ab - PSA + Ab - PSA - 125I + PSA + PSA - 125I
Separation step
Count bound radiation
Method 2:
Principle: IRMA (monoclonal Abs)
Signal: 125I radiation
Formula:
Plastic bead- Ab + PSA + Ab - 125I
Plastic bead- Ab - PSA - Ab - 125I + PSA + Ab - 125I
+ PSA - Ab - 125I
Separation step
Count bound radiation
Method 3:
Principle: EIA (monoclonal Abs)
Signal: PNP absorbance
Formula:
Plastic bead- Ab + PSA + Ab - ALP
Plastic bead- Ab - PSA - Ab - ALP + PSA +
Ab-ALP + PSA - Ab-ALP
1. Wash step
2. Add PNPP substrate
3. Quench
4. Measure PNP product absorbance at 405 and 450 nm
Method 4:
Principle: Sandwich ELISA (monoclonal/polyclonal Abs)
Formula:
Plate Ab + PSA + Ab HRP Plate Ab PSA Ab HRP + PSA + PSA Ab HRP + Ab HRP
1. Wash step
2. Add TMB (3, 3, 5, 5 tetramethyl-benzidine) substrate
3. Quench (sulfuric acid)
4. Measure color change absorbance at 450 nm
1019
Method 5:
Principle: MEIA (monoclonal Abs)
Signal: 4-MU fluorescence
Formula:
Microparticle- Ab + PSA + Ab - ALP
Microparticle- Ab - PSA - Ab-ALP + PSA +
Ab-ALP + PSA - Ab - ALP
1. Wash step
2. Add 4-MU substrate
3. Quench
4. Measure 4-MU product fluorescence at 448 nm
Method 6:
Principle: FIA (polyclonal Abs)
Signal: 4-MU fluorescence
Formula:
Glass-fiber paper tab Ab + PSA + Ab - ALP
Glass-fiber paper tabAb - PSA - Ab - ALP + PSA +
Ab - ALP + PSA - Ab - ALP
1. Wash step
2. Add 4-MU substrate
3. 360 nm excitation light
4. Measure 4-MU product fluorescence at 450 nm
Method 7:
Principle: CIA (monoclonal/polyclonal Abs)
Signal: Acridinium ester chemiluminescence
Formula:
Paramagnetic particle- Ab + PSA + Ab - AE
Paramagnetic particle- Ab - PSA - Ab - AE + PSA +
Ab-AE + PSA - Ab-AE
1. Wash/separation
2. Add H2O2
3. OH-
4. Measure chemiluminescence of acridinium ester at 400 nm
See text for further details. Some of these assays are available for experimental use only, pending FDA approval. Some
assays are available for routine use in Europe.
Ab, Antibody; CIA, chemiluminescence immunoassay.[35]
Table 2: Variations of PSA Measurements [40]

PSA Measurement Definition Application
Total PSA [41] Includes free and bound PSA in Most commonly reported value
serum FDA approved for detection of prostate cancer (upper
limit of normal 4.0 ng/mL)
Useful in management and monitoring recurrence after
treatment
fPSA [42] Unbound PSA Higher values are observed in benign conditions
Improves prostate cancer detection in high-risk
population of men with a normal total PSA and DRE
cPSA [43] PSA complexed with other Higher values are observed in malignancies
proteins such as 1-
antichymotrypsin
PSA density [44] Serum PSA divided by the Facilitates interpretation of PSA values by reference to
prostate volume (as estimated gland size
by ultrasound)
PSA velocity [45] Measurement of longitudinal May be used to predict presence and aggressiveness of
change in PSA over a specific prostate cancer prior to intervention
time interval
PSA doubling time [46] Measurement of time for PSA Dependent upon initial PSA measurement
levels to double Useful for detecting relapse after therapy
1020
Pyruvic Acid
Pyruvic Acid
Name: Pyruvic acid, pyruvate
Clinical significance: see Interpretation page 2, line 94, right column.
Merck Index: 7925
Chemical class: -Ketocarboxylic acid
Structure:
Refer to Chapter 29, Acid-Base Control and Acid-Base Disorders, in the 5th edition of Clinical

i Many reports describe the use of the enzymatic method
using protein-free filtrates formed by acid precipitation of
The earliest method for measuring blood pyruvate was serum proteins. At pH 7.4, the reaction strongly favors the
based on the fact that pyruvate is a ketone. Ketones react quantitative conversion of pyruvate to lactate. The reaction
with 2,4-dinitrophenylhydrazine (DNPH) to produce a is easily monitored by following the consumption of
colored hydrazone (Table 1, Method 1). NADH, either spectrophotometrically at 340 nm [3-5] or
This reaction was usually performed on a protein-free fluorometrically [6,7]. This method is easily adapted to
filtrate after treatment of serum with a strong acid many automated instruments that monitor the reaction by
(trichloroacetic acid [TCA], perchloric acid [PCA], or measuring either absorbance [3] or fluorescence [6,7].
metaphosphoric acid [MPA]). Since other ketones react
with DNPH, most methods were an attempt to extract the A biosensor has also been developed for the measurement
acid hydrazones [1] with an organic solvent such as of pyruvate in whole blood or serum (Table 1, Method 3)
ethyl acetate, xylene, toluene, or benzene. The extract was [8]. The pyruvate sensor is a hydrogen peroxide sensor
then back extracted with an alkaline solution (10% covered with a membrane containing immobilized
Na2CO3) to separate the acid hydrazones. After a wash pyruvate oxidase and protected with a cellulose acetate
with the organic solvent, the aqueous solution was made membrane. The reaction of pyruvate in the presence of
more alkaline with NaOH, resulting in a red color (Amax pyruvate oxidase is as follows:
about 430 nm). Attempts to purify the pyruvic acid pyruvate

hydrazone used chromatography on alumina or silica-gel Pyruvate + HPO42 + O2 + H2O
columns and paper chromatography [2].
acetyl phosphate + CO2 + H2O2
Current methods for the analysis of pyruvic acid employ a
reaction catalyzed by muscle lactate dehydrogenase (LD)
(Table 1, Method 2): The H2O2 is measured amperometrically and related to
pyruvate concentrations present in the sample. Several
LD, pH 7.4 cofactors, including thiamin pyrophosphate, phosphate,
Pyruvate + NADH + H+ lactate + NAD+ and calcium chloride, are required for the reaction
catalyzed by pyruvate oxidase [8].
A rapid HPLC method has also been described that

measures branched-chain keto acids and pyruvate in blood
i
Pyruvic Acid [9]. The pyruvate and other keto acids are derivatized with
Previous and current authors of this method: o-phenylenediamine to give fluorescent derivatives which
First edition: Nancy Gau are then separated chromatographically on a reversed-
phase column using a binary gradient.
Methods edition: Nancy Gau
Second edition: Not updated A method for measuring pyruvic acid by capillary
Third edition: Steven C. Kazmierczak electrophoresis with amperometric detection has recently
Fourth edition: Nancy Gau been described [10]. In this technique, pyruvic acid reacts
Fifth edition: Steven C. Kazmierczak with NH2OH to form oxime as the final product. Pyruvic
1021
Pyruvic Acid
acid is quantified by capillary electrophoretic separation,

followed by amperometric detection of remaining NH2OH. Interferences
Hemolysis can cause an increase in pyruvate. One recent
Reference and Preferred Methods study found that samples with plasma hemoglobin
There is no reference method for the measurement of concentrations of 0.9 g/dL showed increased measured
pyruvate. pyruvate concentrations by approximately 30% [14].
Decreases in measured pyruvate have been described for
The hydrazone method was a manual, laborious procedure enzymatic methods where the reagent LD is contaminated
that lacked specificity. In the method of Friedemann and with pyruvate kinase [15].
Haugen [1], it was estimated that 70% to 96% of the color
extracted was produced by the pyruvate hydrazone [11- Pyruvic Acid Reference Interval
13]. Other endogenous ketones interfered, including Pyruvate concentrations in fasting venous and arterial
acetoacetate, levulinic acid, and -ketoglutaric acid. blood are < 0.10 mmol/L. Pyruvate measured in capillary
Chromatographic purification of the pyruvate hydrazone blood of 141 healthy newborns ranged from 0.01 to 0.14
only made the analysis more laborious, and the analysis mmol/L, with a mean of 0.03 mmol/L [16]. Measurement
was further complicated by the presence of two hydrazone of pyruvate, in addition to lactate, in cerebrospinal fluid is
isomers for each ketone. important in the assessment of numerous primary and
acquired disorders affecting the central nervous system.
The enzymatic method is the method of choice. The The upper 90th percentile limit for cerebrospinal fluid
reagents are reasonably stable, and the reaction is easily pyruvate from birth to 15 years of age showed a linear
automated. Recovery of added pyruvate is close to 100% decrease from 0.148 to 0.139 mmol/L [17].
[3-5], and the precision of 2.5% (percent coefficient of
variation) at 0.08 mmol/L is acceptable [4]. The analysis Interpretation
can be performed in an end-point or kinetic mode. Pyruvate is a product of the Embden-Meyerhof pathway,
which oxidizes glucose to three-carbon compounds. When
A major concern in all methods is the instability of pyruvic peripheral cells are sufficiently oxygenated, most of the
acid at room temperature. To preserve the sample, many pyruvic acid produced is further oxidized in mitochondria
workers recommend preparing a protein-free supernatant to CO2 and water by the citric acid cycle. However, under
of serum or plasma as rapidly as possible. Although PCA conditions of tissue anoxia, when the cellular PO2 is low
and TCA are widely used, Marbach and Weil recommend
and the NAD/NADH ratio is low, pyruvate is reduced to
the use of metaphosphoric acid [4]. They found that
lactate. This raises the NAD/NADH ratio and allows
NADH was much more stable when MPA was used as the
continued anaerobic metabolism of glucose by the
precipitating agent.
Embden-Meyerhof pathway. Thus the ratio of serum
lactate to pyruvate is a measure of the oxidative-reductive
Specimen
state of the body. Mild exercise will cause an increase in
Blood should be drawn in an iodoacetate (gray-top)
both lactate and pyruvate but will not substantially change
phlebotomy tube to prevent metabolism of glucose to
the ratio. The determination of the lactate/pyruvate ratio
pyruvate by blood cells. Pyruvate is very unstable in blood,
has also been suggested as an aid to estimate the severity
and proper technique is essential. The blood should be
of circulatory failure, since the ratio is considered to be a
drawn from a fasting patient. A tourniquet can be used
marker of cytosolic redox status [18,19]. The higher the
during phlebotomy, since blood stasis for up to 2 min does
ratio, the more severe the tissue anoxia is presumed to be.
not produce any change in pyruvate concentration [5].
Similarly, this ratio can be useful in allowing
Clenching and unclenching of a fist also does not affect
determination of the extent of an alcohol-induced ketosis.
pyruvate levels. Mild exercise will increase the blood
pyruvate concentration.
Measurement of lactate and pyruvate in cerebrospinal fluid
can help in the diagnosis of inborn errors of metabolism
The serum should be separated immediately at refrigerated
that can affect the central nervous system, such as pyruvate
temperatures and kept at 4C to 8C until analysis.
dehydrogenase deficiency [20] and respiratory chain
Analysis should be performed as soon as possible. Delays
disorders [21].
of 1 hour or more before deproteinization have been found
unacceptable [8]. Preservation of the sample by
Pyruvic Acid Performance Goals
preparation of a protein-free supernatant of serum or
Pyruvic acid is not currently regulated for proficiency
plasma is recommended. The sample can be stored frozen,
testing. Inter- and intraindividual biological variation
but for long-term stability, it is best to treat the sample
measured in healthy individuals over a 5-day period was
with metaphosphoric acid (to a final concentration of 40
determined to be 13.0% and 15.2%, respectively [22].
g/L) and separate the protein-free supernatant by
Desirable specifications for analytical imprecision derived
centrifugation. This supernatant is stable for 6 days at
from studies of biological variation indicate an assay
room temperature, 8 days at refrigerated temperatures, and
imprecision of no greater than 7.6% [23].
for up to 42 days when frozen [4].
1022
Pyruvic Acid
acids. Scand J Clin Lab Invest 1952;4:189-191.

References 13 Drewes PA. Carbohydrate derivatives and
1 Friedmann TE, Haugen GE. Pyruvic acid. II. The metabolites. In: Henry RJ, Cannon DC,
determination of keto acids in blood and urine. J Winkelman JW, eds. Clinical Chemistry:
Biol Chem 1943;147:415-442. Principles and Technics. 2nd ed. Hagerstown,
2 Seligson, D, Shapiro B. Alpha-keto acids in blood MD: Harper & Row; 1974:1335-1340.
and urine studied by paper chromatography. Anal 14 Chuang CK, Wang TJ, Yeung CY, Hsieh WS,
Chem 1952;24:754-755. Lin DS, Ho SC et al. Interference and blood
3 Hansen JL, Freier EF. Direct assays of lactate, sample preparation for a pyruvate enzymatic
pyruvate, beta-hydroxybutyrate and acetoacetate assay. Clin Biochem 2006;39:74-77.
with a centrifugal analyzer. Clin Chem 15 Henderson AR. A source of error in blood
1978;24:475-479. pyruvate determination. J Clin Pathol
4 Marbach EP, Weil MH. Rapid enzymatic 1971;24:475.
measurement of blood lactate and pyruvate. Clin 16 Nielsen J, Ytrebo LM, Borud O. Lactate and
Chem 1967;13:314-325 pyruvate concentrations in capillary blood from
5 Segal S, Blair AE, Wyngaarden JB. An enzymatic newborns. Acta Paediatr 1994;83:920-922.
spectrophotometric method for the determination 17 Benoist JF, Alberti C, Leclercq S, Rigal O, Jean-
of pyruvic acid in blood. J Lab Clin Med Louis R, Ogier de Baulny H et al. Cerebrospinal
1956;48:137-143. fluid lactate and pyruvate concentrations and their
6 Antonius A, Clark M, Pilkington JRE. A ratio in children: age-related reference intervals.
semiautomated fluorometric method for the Clin Chem 2003;49:487-494.
enzymatic determination of pyruvate, lactate, 18 Huckabee WE. Relationships of pyruvate and
acetoacetate and beta-hydroxybutyrate levels in lactate during anaerobic metabolism. III. Effect of
plasma. J Lab Clin Med 1966;68:340-356. breathing low-oxygen gases. J Clin Invest
7 Cramp DG. Automated enzymatic fluorometric 1958;37:264-271.
method for the determination of pyruvic and 19 Medina JM, Tabernero A, Martin-Barrientos J.
lactic acids in blood. J Clin Pathol 1968;21:171- Metabolic fuel utilization and pyruvate oxidation
174. during the postnatal period [review]. J inherit
8 Chariot P, Tariney R, Ami-Said M, Herigault R, Metab Dis 1996;19:432-442.
Adnot S, Gherardi R. Optimal handling of blood 20 Matsuda J, Ito M, Naito E, Yokota I, Kuroda Y.
samples for routine measurement of lactate and DNA diagnosis of pyruvate dehydrogenase
pyruvate. Arch Path Lab Med 1994;118:695-697. deficiency in female patients with congenital
9 Pailla K, Blonde-Cynober, F, Aussel C, DeBandt lactic academia. J Inherit Metab Dis 1995;18:534-
JP, Cynober L. Branched-chain keto-acids and 546.
pyruvate in blood: measurement by HPLC with 21 Jackson MJ, Schaefer JA, Johnson MA, Morris
fluorometric detection and changes in older AA, Turnbull DM, Bindoff LA. Presentation and
subjects. Clin Chem 2000;46:848-853. clinical investigation of mitochondrial respiratory
10 Lu X, Huang WH, Ai F, Wang ZL, Cheng JK. chain disease. A study of 51 patients. Brain
Indirect determination of pyruvic acid by 1995;118:339-357.
capillary electrophoresis with amperometric 22 Panteghini M, Pagani F. Biological variation of
determination. J Chromatogr B Analyt Technol lactate and pyruvate in blood. Clin Chem
Biomed Life Sci 2007;857(2):347-351. 1993;39:908.
11 McArdle B. The quantitative estimation of 23 Ricos C, Alvarez V, Cava F, Garcia-Lario JV,
pyruvic and (L-oxyglutaric acids by paper Hernndez A, Jimnez CV et al. Current database
chromatography in blood, urine and cerebrospinal on biologic variation: pros, cons and progress.
fluid. Biochem J 1957;66:144-148. Scand J Clin Lab Invest 1999;59:491-500.
12 Kulonen E, Carpen E, Ruokolainen T. Experience
with paper chromatography in the study of -keto
1023
Pyruvic Acid
Table 1: Pyruvate Methods

Method 1: Dinitrophenylhydrazone
Principle of analysis: Pyruvate and other ketones react with 2,4-dinitrophenylhydrazine to form the hydrazone
derivative; acid hydrazones are extracted and alkalinized to produce a red color.
Comments: Plasma, serum; nonspecific, of historical interest
Method 2: Enzymatic
Principle of analysis: Pyruvate is quantitatively converted to lactate at pH 7.4 by lactate dehydrogenase; the reaction
is monitored at 340 nm by following the concomitant conversion of NADH to NAD+ or by fluorescence using
excitation at 360 nm and emission at 470 nm.
LD
Pyruvate + NADH + H lactate + NAD
Comments: Plasma, serum; highly specific, automated, preferred method
Method 3: Amperometric
Principle of analysis: Pyruvate reacts with pyruvate oxidase which has been immobilized on an artificial membrane.
H2O2 produced measured amperometrically.
Pyruvate + HPO42 + O2 + H2O acetyl phosphate + CO2 + H2O2
Pyruvate oxidase
4. Lactic dehydrogenase (LD). Muscle LD can be

Procedure: Enzymatic Analysis of Pyruvic Acid purchased from Boehringer Mannheim Diagnostics
(including reaction conditions for enzymatic measurement
of pyruvic acid) (Indianapolis, IN; No. 127230) as an ammonium sulfate
suspension. Prepare working reagent by diluting 50 mL of
Principle the stock suspension with 450 mL of distilled water. Use
Muscle lactate dehydrogenase is used to immediately.
catalyze the reaction between pyruvic acid and 5. Working reagent. Add 40 mL of diluted LD
solution and 400 mL of NADH solution to 10 mL with
NADH to form lactate and NAD+. The reaction working buffer. Stable 24 hr at refrigerated temperatures.
is monitored by measurement of the decreasing 6. Stock pyruvic acid, 25 mmol/L. Place 2.2 g of
absorbance at 340 nm. This procedure can be sodium pyruvate in a 1-L flask. Dissolve in 200 mL of 0.1
readily adapted to most automated analyzers. M HCl, and bring to volume with additional 0.1 M HCl.
Stable for 3 months at refrigerated temperatures.
7. Working pyruvic acid standard, 0.5 mmol/L.
Reagents Dilute 1 mL of stock pyruvic acid to 50 mL with distilled
1. Stock buffers
water in a volumetric flask. Prepare fresh daily.
0.1 M Na2HPO4. Dissolve 1.42 g of Na2HPO4
salt in 80 mL of distilled water in a 100-mL volumetric Assay
flask, and bring to volume with additional distilled water. Equipment: ABA-100 or any automated
Stable for 1 year at refrigerated temperatures. spectrophotometric analyzer capable of reading at 340 nm.
0.1 M KH2PO4. Dissolve 1.36 g of KH2PO4 in
80 mL of distilled water in a 100-mL volumetric flask, and
bring to volume with additional distilled water. Stable for Instrument Parameters
1 year at refrigerated temperatures. Temperature 37C
2. Working buffer, 0.1 mol/L phosphate buffer. Mode Kinetic
Mix 20 mL of 0.1 M KH2PO4 and 80 mL of Na2HPO4 Direction Down
buffers. Check that the pH is 7.4 0.05; adjust if necessary Time 15 sec
with 0.1 M HCI or NaOH. Stable for 2 months at Revolutions Infinity
refrigerated temperatures. Check pH periodically, and Filter 340/380 nm
examine for the presence of bacterial growth or a
precipitate. Syringe plate 1:11
3. NADH. Weigh 20 mg of pure NADH. Dissolve Decimal 0.000
in 1 mL of distilled water. Prepare fresh, and use within 1 Calibration factor 0.500
h. Zero setting 0.000
1024
Pyruvic Acid
Calculation
1. Prime working reagent through syringe plate. Subtract the first reading of position 1 (blank)
2. Set above parameters, and arrange sample tray in from the first reading of other positions to get change of
following manner: absorbances (A); then:
a. H2O
b. 0.5 mmol/L pyruvate standard
c. 0.5 mmol/L pyruvate standard
d. Patient sample
e. Patient sample, diluted fourfold with
saline solution
3. Rotate to position 31, go to test mode, and set Notes
zero. Push calibrate button, and set scale to read 0.500 with 1. This method is linear up to 1 mmol/L.
scaling knob. Push start button. As the carousel rotates, it
2. High lactate (to 40 mmol/L), bilirubin (to 200
will begin printing the change of absorbance readings
mg/L), and gross hemolysis (2 g of
every 15 sec for the first sample. It will print a total of
hemoglobin/L) will not interfere with the
three readings for each sample and then rotate to the next
analysis.
sample.
3. Grossly lipemic sera will not affect the analysis.
4. After the third absorbance reading on the last
sample is completed, stop instrument. Carousel will not
stop automatically.
Table 2: Reaction Conditions for Enzymatic Measurement of Pyruvic Acid

Temperature: 37C
pH: 7.4
Sample volume: 25 L
Fraction of sample volume: 0.0909
Final concentration of reagents:
Phosphate buffer: 90.9 mmol/L
NADH: 1.1 mmol/L
LD: varies with lot
Linearity: 0 to 1 mmol/L
1025
Renin
Renin
Greg Ward i
Name: Renin (PRA, plasma renin activity, direct renin concentration, DRC)
Clinical significance: Refer to Chapter 28, Physiology of Body Water and Electrolytes, in the
5th Edition of Clinical Chemistry: Theory, Analysis, Correlation.
Molecular mass: ~ 40,000 D
Chemical class: Glycoprotein, enzyme EC 3.4.23.15
Principles of Analysis and Current Usage indirect or enzyme kinetic assays, whereas whole-
molecule assays have been referred to as direct renin
Renin (EC 3.4.23.15; formerly EC 3.4.4.15 and EC concentration (DRC) or active renin assays or total
3.4.99.19) is a 340-amino-acid aspartyl protease renin assays for those assays which measure both active
glycoprotein (MW ~ 40000 daltons) which catalyses the renin and prorenin. Currently, prorenin assays or total
cleavage of angiotensin I from angiotensinogen [1,2]. renin assays do not have routine clinical application and
The juxtaglomerular cells of the kidney synthesize the will not be further discussed. In general, the terminology
inactive precursor prorenin that is cleaved into the active PRA and DRC will be used, consistent with the 2008
proteolytic enzyme renin, which is stored and released in Endocrine Society Clinical Practice Guideline [7].
response to stimuli from secretory granules.
Physiological factors important for renin release include Plasma Renin Activity (Table 1, Method 1)
sodium depletion, hypotension, hypovolemia, and - Plasma renin activity is measured by the determination
adrenergic stimulation [1,2,3]. of the amount of angiotensin I generated after incubation
of plasma at 37C, buffered to pH 6.0 for optimal
Active renin released into the circulation cleaves the activity. The substrate angiotensinogen is not added, but
decapeptide angiotensin I from its substrate serum rather endogenous angiotensinogen normally present in
angiotensinogen, an 2-globulin synthesized in liver and excess amounts is utilized. The amount of endogenous
kidney, to angiotensin I, which is converted by angiotensinogen is not at the concentration required for
angiotensin-converting enzyme to the potent Vmax and is sometimes closer to the Km. Research
vasoconstrictor, angiotensin II. Angiotensin II, together methods where addition of exogenous angiotensinogen
with ACTH and potassium, are responsible for synthesis or synthetic substrates are used have been described and
and secretion of aldosterone. The renin-angiotensin- are termed assays for plasma renin concentration (PRC).
aldosterone system is critical to both sodium The PRA is dependent on both the amount of active
homeostasis and maintenance of blood volume and renin and angiotensinogen for generation of angiotensin
blood pressure. This system is controlled by renin I, and since this is the rate-limiting step, it is considered
activity, which is the rate-limiting step [1,2,3]. to reflect the ability to generate angiotensin II in vivo
[1,2,5,6].
Prorenin is secreted constitutively from the kidney and
other organs in contrast to the tight regulation of active Angiotensin I generated is protected from degradation or
renin release. Prorenin concentrations in normal serum modification by the addition to reaction buffer of phenyl
are approximately 10-fold greater than those of active methyl sulfonyl fluoride (PMSF) and EDTA, which
renin, and conversion to active renin appears not to respectively inhibit the enzymes angiotensinase and
occur in plasma in vivo. In-vitro irreversible angiotensin-converting enzyme. The angiotensin I
cryoactivation of prorenin to active renin occurs in liquid generated during the incubation is quantitated by
plasma at refrigerated temperatures below 25C, competitive radioimmunoassay using125I-angiotensin I as
increasing with decreasing temperature [1,4,5]. Plasma tracer and separation of antibody-bound tracer from free
active renin concentrations are very low, on the order of achieved by PEG separation or antibody-coated tubes.
approximately 0.2 to 0.8 pmol/L (8 to 32 ng/L or 12 to Although a blank tube is recommended in some
48 mU/L, depending on the conversion factor for the procedures, it has been demonstrated that a blank is not
assay), which corresponds to a plasma renin activity of necessary, provided the amount of angiotensin I
approximately 1 to 4 ng/mL/hr. generated is 10 times greater than the blank signal. This
is achieved by extending the incubation times for
Assays for renin have been classified as those which samples with low PRA. The standard incubation time for
measure enzyme activity (plasma renin activity [PRA]) the generation of angiotensin I is 90 to 180 min. Samples
and immunoassays for the renin molecule [4,5,6,7]. with PRA < 1 ng/mL/hr should be incubated for 18
Enzyme activity assays have also been referred to as hours, which allows for lower activity to be accurately
i
Renin
New method
Fifth edition: Greg Ward
1026
Renin
and precisely quantitated. The limit of detection for active conformation) was eliminated in both systems by
incubation times of 90 min, 3 hours, and 18 hours is, using an incubation time of 30 minutes at 37C.
respectively, 0.65 ng/mL/hr, 0.33 ng/mL/hr, and 0.06
ng/mL/hr. The PMSF inhibition of angiotensinase cannot Reference and Preferred Methods
be maintained during the 18-hour incubation unless pH There is no reference method for renin.
6.0 is maintained. Further renin is unstable at pH > 8.0,
and at pH 3.5, conversion of prorenin to renin occurs According to the 2007 General Ligand Survey by the
[1,2,5]. American College of Pathologists, more than 90% of 40
participating laboratories used PRA methods with
Angiotensin I formation in PRA assays has also been incubation at pH 6.0 and RIA of angiotensin I. The
determined by EIA using -galactosidase-labeled remainder (<5 labs) measured DRC with no data
angiotensin I [8] or liquid chromatography/mass available.
spectrometry [9]. Both DRC and PRA assays are suitable for measurement
of normal or low amounts of renin. In the case of low
Direct Renin Concentration renin levels, PRA with an 18-hour incubation is
All two-site immunometric (sandwich) assays for DRC preferred over DRC [5,7,14].
utilize capture antibodies which immuno-extract both
active renin and prorenin. Specificity is achieved by Specimen
using signal antibodies towards renin in the active (open) Critical to the collection and preparation of samples for
conformation and short incubation times at 37C. renin analysis are procedures which prevent
cryoactivation of renin resulting in falsely elevated
Two-site immunoradiometric assays (IRMA) with 125I- levels. EDTA is an inhibitor of angiotensin-converting
labeled signal antibodies have used capture antibodies, enzyme and is the preferred anticoagulant for PRA and
which are either coupled to magnetic beads [10] or can also be used for DRC [1,2,5]. In the case of DRC
biotinylated and react with avidin-coated beads [11] to assays, all anticoagulants tested were observed to be
create a solid phase for separation of the sandwich suitable [13].
complex from unreacted signal antibody (Table 1,
Method 2). Deinum and co-workers [12] demonstrated It is recommended that the sample be collected mid-
that overestimation of renin in samples with low morning, after the patient has been ambulant for a
concentrations was not due to cross-reactivity with minimum of 2 hours but seated for 5 to 15 minutes prior
inactive prorenin, but rather the conformation of to collection [7]. The sample should be collected at room
prorenin had changed to that of the active form under the temperature (not placed on ice, because the lower
conditions of incubation, resulting in formation of temperature will cause cryoactivation). The sample
additional measurable sandwich complexes. Thus these should be centrifuged at room temperature and the
assays are specific for both active renin and activated plasma snap frozen.
prorenin. The conformational change and overestimation
was eliminated by reducing the incubation time from 24 Renin is stable in whole EDTA blood and plasma at
to 6 hours and increasing the incubation temperature room temperature for 24 hours. Samples stored at 20C
from 22C to 37C. The functional sensitivity (interassay are stable for up to 15 months [2,5].
coefficient of variation [CV] < 20%) for the latter assay
which eliminated prorenin interference was 4 mU/L. To avoid cryoactivation, frozen samples should be
thawed rapidly prior to analysis. Fans and lukewarm
water have been variously used. Frozen samples should
Automation of DRC assays has been achieved with not be thawed in a refrigerator, and frozen samples
immunoluminometric assays (ILMA) which utilize which thaw during transportation should be rejected for
chemiluminescence labels (Table 1, Method 3). In the analysis [1,2,5].
Nichols Advantage System, the signal antibody was
labeled with an acridinium ester and the capture Interferences
antibody biotinylated. Separation of the sandwich No analytical interferences in renin assays have been
complex was achieved by the use of paramagnetic reported. Heterophilic antibody interference is possible
particles coated with streptavidin. The functional for the two-site immunometric DRC assays, but this has
sensitivity of this method was 2.65 mU/L [13]. In routine not yet been reported.
practice, this system could not reliably measure low
renin levels [5]. The Nichols Advantage System is no Many medications can cause plasma renin levels to
longer commercially available. The DiaSorin Liaison either decrease (-adrenergic blockers, clonidine, -
automated analyzer has a DRC method (currently not methyldopa, NSAIDS) or increase (potassium wasting or
available in the United States) where the capture sparing diuretics, ACE inhibitors, dihydropyridines,
antibody is coupled to paramagnetic particles, and the angiotensin IItype I receptor blockers). This can result
signal antibody is labeled with amino-butyl-ethyl- in false-positive or false-negative aldosterone/renin
isoluminol. The manufacturers quoted functional ratios, and recommendations are made in clinical
sensitivity is < 2 mU/L practice guidelines for medication use during
(http://www.diasorin.com/upload/prodottiesistemi/Direct investigation of hypertension [1,2,7].
%20Renin.pdf). Prorenin interference (conversion to
1027
Renin
pseudohypoaldosteronism [2,14]. PRA is best used for

Renin Reference Interval measurement of low renin levels; both PRA and DRC
As with all diagnostic tests, it is recommended that are suitable for normal and high renin levels.
reference intervals be determined by each laboratory to
conform to the characteristics of the population being In most clinical situations, renin levels are measured
tested. together with aldosterone and the aldosterone-to-renin
ratio (ARR). A recent assessment of current literature
Renin levels are highest in newborn children, with a demonstrated ARR to be superior to measurement of
rapid fall after 3 months, whereby they slowly decrease aldosterone or renin in isolation for patients with PA [7].
to adult levels. In adults, renin levels decrease with Renin levels are low in PA. The use of PRA and DRC in
aging, presumably due to decreased renal function PA is controversial [4, 5, 16-19]. In general, two-site
[2,15]. immunometric assays are unable to reliably measure low
renin levels [4, 5, 15, 16]. Early two-site IRMAs
Renin has a diurnal variation, with the highest level measured activated prorenin in addition to active renin
observed on awakening and values then progressively [4,5,12]. In normal patients, the prorenin concentration is
falling through the day. approximately 10 times the renin concentration, whereas
patients with PA have up to 100 times more prorenin
Renin levels are influenced by posture, with the change than active renin. Thus at low DRC, even small
in blood pressure on standing resulting in an increased conversion of prorenin can affect the accuracy and
renin level. clinical reliability of the assay. The problem of prorenin
activation was overcome by short incubation times at
Renin levels are lower in blacks than whites. 37C [12]. Initial studies on a two-site ICMA
demonstrated that this assay could measure DRC in
Dietary sodium intake influences the renin level, with patients with low PRA [5, 20-22]. In routine use, the
salt restriction or high-salt diets respectively stimulating two-site ICMA could not reliably measure low DRC.
or suppressing renin levels. This has been attributed to a variety of factors, including
insufficient range of low renin levels used in the
In general, morning reference intervals for PRA and standard curve, variation between reagent lots, false high
DRC are 1 to 4 ng/mL/hr and 8 to 35 mU/L, levels (this could be unproven heterophilic antibody
respectively. interference), and machine maintenance [5]. It has been
recommended that the best test for measurement of low
Reference intervals in use are shown below. renin levels is PRA, with an 18-hour incubation followed
by two-site immunometric assay [5, 14].
Plasma Renin Activity Reference Intervals
(ARUP, Renin levels are low in adrenal disease where there is
http://www.aruplab.com/guides/ug/tests/00 excess production of deoxycorticosterone (DOC),
70105.jsp) including: 11--hydroxlase deficiency; 17--hydroxylase
deficiency; primary glucocorticoid resistance; DOC
Adults, normal sodium diet: producing tumors; and sometimes ectopic ACTH
Supine: 0.2-1.6 ng/mL/hr syndrome [2, 14].
Upright: 0.5-4.0 ng/mL/hr
Newborn (1-7 days): 2.0-35.0 ng/mL/hr Renin levels are elevated in the following adrenal
Cord blood: 4.0-32.0 ng/mL/hr diseases: primary adrenal insufficiency; 21-hydroxylase
Children, normal sodium diet, supine: deficiency; aldosterone synthase deficiency; 3--
1-12 months: 2.4-37.0 ng/mL/hr hydroxysteroid dehydrogenase deficiency; side-chain
13 months-3 years: 1.7-11.2 ng/mL/hr cleavage enzyme deficiency; and steroid acute
4-5 years: 1.0-6.5 ng/mL/hr regulatory protein deficiency [2, 14].
6-10 years: 0.5-5.9 ng/mL/hr
11-15 years: 0.5-3.3 ng/mL/h Renin levels are normal in secondary adrenal
insufficiency (hypopituitarism).
Direct Renin Concentration Reference Intervals [2]
Adults: Renin levels can be either normal or elevated in renal
Supine: 3-21 ng/L (5-34 mU/L) artery stenosis (RAS) or reninoma [2, 14]. In RAS, the
Upright: 5-30 ng/L (8-48 mU/L) clinical sensitivity and specificity is improved by using
Children: captopril renography rather than captopril-stimulated
Mean DRC, neonatal period 73.1 ng/L PRA or peripheral PRA [14]. In renography, renin is
Mean DRC at 3 months 30.4 ng/L measured in the renal veins both before and after
administration of captopril, with a renal-vein ratio
Interpretation greater than 1.5 used to demonstrate lateralization of
Renin levels are used in the investigation of patients with renin excess. Reninoma is rare, and lateralized renin
secondary hypertension, most commonly primary secretion in the absence of RAS is suggestive; further
hyperaldosteronism (PA) and renal artery stenosis but radiological studies are required to search for a small
also reninoma, congenital adrenal hyperplasia, and kidney mass. Renin secretion in these patients varies
1028
Renin
over time, and a single renin measurement does not immunoassay specific for active renin. Clin
exclude reninoma [14]. Chem 1992;38:1959-1962.
11 Derkx FHM, de Bruin RJA, van Gool JMG, van
Renin Performance Goals den Hoek MJ, Beerendonk CCM, Rosmalen F
Survey data from the 2007 CAP participant summary et al. Clinical validation of renin monoclonal
report show imprecision values (CV) of PRA to be antibody-based sandwich assays of renin and
approximately 20% for specimens with mean activities prorenin, and use of renin inhibitor to enhance
ranging from approximately 1 to 12 ng/mL/hr. prorenin immunoreactivity. Clin Chem
Acceptable Clinical Laboratory Improvement 1996;42:1051-1063.
Amendments performance criteria (CLIA 88) for 12 Deinum J, Derkx FHM, Schalekamp MADH.
measurement of renin require that laboratories be Improved immunoradiometric assay for plasma
accurate to within 3 SD of the peer-group mean. The renin. Clin Chem 1999;45:847-854.
ranges of renin in the CAP survey are higher than those 13 de Bruin RA, Bouhuizen A, Diederich S,
seen in PA and may not reflect the ability of laboratories Perschel FH, Boomsma F, Deinum J.
to measure low PRA. Validation of a new automated renin assay. Clin
Chem 2004;50:2111-2116.
Most renin measurements are used for screening of PA 14 Rossi GP, Seccia TM, Pessina AC. Clinical use
where low renin levels are observed. Currently, only 18- of laboratory tests for the identification of
hour incubation PRA can consistently measure these low secondary forms of arterial hypertension. Crit
renin levels. DRC assays using two-site ICMA have the Rev Clin Lab Sci 2007;44:1-85.
potential to measure these low concentrations but so far 15 Gordon RD. The challenge of more robust and
have been inconsistent when used routinely. reproducible methodology in screening for
primary aldosteronism. J Hypertens
References 2004;22:251-255.
1 Sealey JE. Plasma renin activity and plasma 16 Sealey JE, Trenkwalder P, Gahnem F,
prorenin assays. Clin Chem 1991;37:1811- Catanzaro D, Laragh JH. Plasma renin
1819. methodology: inadequate sensitivity and
2 Cartledge S, Lawson N. Aldosterone and renin accuracy of direct renin assay for clinical
measurements. Ann Clin Biochem applications compared with traditional
2000;37:262-278. enzymatic plasma renin activity assay
3 Lavoie JL, Sigmund CD. Minireview: overview [commentary]. J Hypertens 1995;13:27-30.
of the renin-angiotensin system, an endocrine 17 Morganti A, Pelizzola D, Mantero F, Gazzano
and paracrine system. Endocrinology G, Opocher G, Piffanelli A.
2003;144:2179-2183. Immunoradiometric versus enzymatic renin
4 Sealey JE, Laragh JH. Renin and prorenin: assay: results of the Italian Multicenter
advances and declines in methodology. Clin Comparative Study. Authors reply. J
Chem 1996;42:993-994. Hypertens 1995;13:31.
5 Sealey JE, Gordon RD, Mantero F. Plasma 18 Mnard J, Guyene T-T. Renin assays: a debate
renin and aldosterone measurements in low for clinicians, not only for specialists
renin hypertensive states. Trends Endocrinol [commentary]. J Hypertens 1995;13:367-369.
Metab 2005;16:86-91. 19 Morganti A, Pelizzola D, Mantero F, Gazzano
6 Schalekamp MADH, Derkx FHM, Deinum J, G, Opocher G, Piffanelli A.
Danser AHJ. Newly developed renin and Immunoradiometric versus enzymatic renin
prorenin assays and the clinical evaluation of assay: results of the Italian Multicenter
renin inhibitors. J Hypertens 2008;26:928-937. Comparative Study. J Hypertens 1995;13:19-
7 Funder JW, Carey RM, Fardella C, Gomez- 26.
Sanchez CE, Mantero F, Stowasser M, Young 20 Hartman D, Sagnella GA, Chesters CA,
WF Jr, Montori VM. Case detection, diagnosis MacGregor GA. Direct renin assay and plasma
and treatment of patients with primary renin activity compared. Clin Chem
aldosteronism: an Endocrine Society Clinical 2004;50:2159-2161.
Practice Guideline. J Clin Endocrinol Metab 21 Perschel FH, Schemer R, Seiler L, Reincke M,
2008;93:3266-3281. Deinum J, Maser-Gluth C et al. Rapid screening
8 Sharpe S, Verkerk R, Sasmito E, Theauws M. test for primary hyperaldosteronism: ratio of
Enzyme immunoassay of angiotensin and plasma aldosterone to renin concentration
rennin. Clin Chem 1987;33:1774-1777. determined by fully automated
9 Fredline VF, Kovacs EM, Taylor PJ, Johnson chemiluminescence immunoassays. Clin Chem
AG. Measurement of plasma renin activity with 2004;50:1650-1655.
use of HPLC-electrospray-tandem mass 22 Unger N, Schmidt IL, Pitt C, Walz MK, Philipp
spectrometry. Clin Chem 1999;45:659-664. T, Mann K, Petersenn S. Comparison of active
10 Simon D, Hartmann DJ, Badouallie G, Caillot renin concentration and plasma renin activity
G, Guyenne T-T, Corvol P et al. Two-site direct for the diagnosis of primary hyperaldosteronism
in patients with an adrenal mass. Eur J
Endocrinol 2004;150:517-523.
1029
Renin
Table 1: Methods of Analysis for Renin

Method 1: Plasma renin activity (PRA) radioimmunoassay (RIA) for angiotensin I
Principle of analysis: Competitive binding of angiotensin I with 125I-labeled angiotensin I for limited amount of
anti-angiotensin I
Marker: 125I
Signal: Radioactivity
Comments: Manual; 18-hour incubation required for low renin levels. Uses endogenous angiotensinogen as
substrate.
Method 2: Immunoradiometric assay (IRMA) for direct renin concentration (DRC)
Principle of analysis: Noncompetitive binding of active renin to excess 125I-labeled antibody
Marker: 125I
Signal: Radioactivity
Comments: Not commonly used now; manual
Method 3: Immunoluminometric assay (ILMA) for DRC
Principle of analysis: Noncompetitive binding of active renin to excess acridinium derivativelabeled antibody
Marker: Acridinium derivatives
Signal: Luminescence
Comments: Fully automated; functional sensitivity indicates that low renin can be measured, but not observed so
far in routine practice. Assay has 30-minute incubation at 37C to eliminate conversion of prorenin to activated
prorenin.
1030
Rheumatoid Factor
Rheumatoid Factor
Terry Pry i
Name: Rheumatoid factor

Clinical significance: Refer to Chapter 7, Immunological Reactions, and Chapter 8,
Immunochemical Techniques, in the 5th edition of Clinical Chemistry: Theory, Analysis, and
Correlation.
Principles of Analysis and Current Usage Reference and Preferred Methods

Rheumatoid factors (RF) are immunoglobulins of any There are no reference methods for RF measurement that
isotype with antibody activity directed against antigenic have been approved at this time. Preferred methods for
sites on the Fc region of human or animal detecting RF are the following:
immunoglobulin G (IgG). After the discovery by Waaler
[1] that the sera from patients with rheumatoid arthritis Latex Agglutination
agglutinated sheep erythrocytes coated with rabbit anti- The earliest tests, which are still widely used clinically,
sheep erythrocyte antibody, it was determined that the rely on the agglutinating properties of the IgM class of
serum factor responsible for the agglutination was a RF. IgG, usually human or rabbit, is bound to a
high-molecular immunoglobulin of the IgM class. IgM- particulate carrier, and the presence of RF is then
RF is the main isotype identified by clinically available detected by agglutination or flocculation of the
diagnostic assays for RF detection. Assays for RF are the respective indicator system. Carrier particles frequently
most widely used serological tests as an aid for the used include latex and erythrocytes. In one application of
diagnosis of rheumatoid arthritis (RA). this method, the procedure is performed by mixing IgG-
coated latex particles with serum on a Teflon-coated
Traditionally, RF was detected by a latex flocculation card (Table 1, Method 1). Latex particles function as
test and the sensitized sheep-cell agglutination (Waaler- inert spacers in the antigen-antibody lattice, increasing
Rose) test [1,2]. Conventional qualitative methods for the size of the complexes.
the measurement of RF-IgM utilize visual detection and
have depended upon the agglutination of particles (e.g., The test is considered positive if agglutination of the
latex, erythrocytes) coated with human or animal IgG. latex particles is observed following the mixing period.
The latex-IgG agglutination technique of Singer and Advantages of the latex agglutination procedure include
Plotz has become the most popular method for detection a rapid assay time and no need for specialized or
of RF [3]. Qualitative (positive/negative) and expensive equipment. Either qualitative or semi-
semiquantitative (titer unit) assessment of RF are quantitative analysis is provided. The RF antibody
provided. Inherent problems accompany these methods, content of a serum involves doubling dilutions of the
such as subjectivity, difficulty in standardization, and serum and determination of an end point (the last
imprecision [3,4]. doubling dilution at which agglutination can be
visualized). The reciprocal of this dilution is known as
Measurement of RF was greatly improved through antibody titer. It is not possible to quantify RF by these
adaptation of automated quantitative serological tests methods except by titer, which is recognized to be
such as enzyme immunoassay (EIA), nephelometry, and inaccurate [4,5].
turbidimetry that utilize the advantage of objective
instrument measurement on a single sample dilution. The latex agglutination test is sensitive, but it can result
These tests are superior in that they are easy to use, in a fairly high number of false positives [6].
provide rapid results, have good precision, and have the Nonspecific agglutination of latex particles by sera from
ability to measure RF in absolute terms. normal individuals is not uncommon [7].
The principle of commonly used methods for Immunoturbidimetric

demonstrating RF involves use of an indicator system An immunoturbidimetric RF assay involves an antigen-
such as latex or erythrocytes, to which human IgG is antibody reaction between RF in the sample and
attached. The presence of RF is recognized by denatured human IgG, which has been adsorbed to latex
aggregation, flocculation, or precipitation of various particles (Table 1, Method 2) [4]. Immunoturbidimetric
indicator systems upon addition of specimen. Several methods measure the resulting agglutination as an
assay methods are available for RF determination and absorbance change, with the magnitude of the change
are listed in Table 1. being proportional to the quantity of RF in the sample.
The actual concentration is then determined by
interpolation from a calibration curve prepared from
calibrators of known concentration.
i
Rheumatoid Factor
New method
Fifth edition: Terry Pry
1031
Rheumatoid Factor
qualitative agglutination methods, including false

Nephelometry positives from bacterial contamination, lipemia, and
Similar to immunoturbidimetric methods, nephelometry drying effect due to extended reaction times.
involves mixing RF and antibody under antibody-excess
conditions. The concentration of resulting antigen- RF Reference Interval
antibody complexes can be determined by light Reference intervals are method dependent but are
dispersion (Table 1, Method 3) [7,8]. When a beam of typically < 30 IU/mL [12]. It is recommended that each
light is passed through tubes containing a fixed amount laboratory determine its own reference range based upon
of antibody and variable concentrations of RF, the its particular locale and population characteristics.
concentration of immune complexes formed in the tube Results expressed in KU/L are equivalent to IU/mL. A
will determine the extent of light scattering. In contrast WHO International Standard Preparation of Rheumatoid
to the immunoturbidimetric method, which measures Arthritis Serum (NIBSC code: 64/2) is available to
transmitted light in the reaction solution, nephelometry achieve traceability of results.
measures the amount of light scattered at angles varying
from 0 to 90 degrees. Since antibody concentration Interpretation
remains constant, the light scattered is proportional to Rheumatoid arthritis is an autoimmune disorder of
the concentration of RF in the mixture. uncertain etiology, but there is evidence that an interplay
of genetic predisposition (HLA-DR shared-epitope
Enzyme-Linked Immunosorbent Assay genes) and environmental factors (e.g., smoking) are
Enzyme-linked immunosorbent assay (ELISA) is used involved [13]. RA is characterized by symmetrical and
for the qualitative, quantitative, or semi-quantitative erosive synovitis that typically affects hands, wrists, and
determination of IgM RF in human serum. Purified RF feet initially and later may involve any joint lined by
antigen (human IgG) is attached to the wells of a synovial membrane. Chronic fluctuating disease course,
polystyrene microtiter plate (Table 1, Method 4) [9]. if left untreated, results in progressive joint destruction,
Diluted patient specimens, controls, and calibrators are deformity, disability, and premature death. RA is the
incubated at room temperature in the microwells. Any most common autoimmune disease, affecting 1% to 2%
RF-IgM antibody present binds to the immobilized of the adult population, with twice as many women as
human IgG to form antigen-antibody complexes. men [14]. The age of onset varies, peaking between 20
Unbound antibody is washed from the wells, and and 40 years for women, while often occurring at age 60
enzyme-conjugated antihuman IgM is added. The to 80 in men. Since structural joint damage is
enzyme conjugate binds to the antigen-antibody irreversible, early recognition and treatment are currently
complex. Excess conjugate is washed away and a being emphasized, with the goal of halting the
specific substrate added. Bound enzyme conjugate progression of the disease.
begins a hydrolytic reaction, causing color development.
After a specific time, the reaction is stopped. Autoimmunity is one of the hallmarks of RA that
distinguishes this disease from other inflammatory and
The intensity of the generated color is proportional to the degenerative joint diseases such as psoriatic arthritis or
amount of RF-specific IgM antibody bound to the wells. osteoarthritis. The first autoimmune response described
The results are read on a spectrophotometer (ELISA in RA was RF. This observation formed the basis for a
reader). The net absorbance is calculated by subtracting pathogenetic concept that considers RA as one of the
the absorbance value for the specimen blank from the systemic autoimmune diseases closely related to
value for the antigen-coated microwell. A calibration systemic lupus erythematosus, scleroderma, or Sjgrens
standard that is assayed with each plate is then used to syndrome.
calculate the RF-IgM activity in IU/mL from the net
absorbance value. Although RF has proven to be a useful diagnostic and
prognostic test, the etiology of RFs and the precise role
Specimen they play in the pathogenesis of RA is still unresolved.
Serum is the preferred specimen for RF testing. IgM-RF synthesis can be induced by immune complexes
Acceptable storage of specimen has been suggested for and polyclonal B-cell activators. Transient synthesis of
20C to 25C (1 day), 2C to 8C (3 days), and 20C (1 IgM-RF accompanies secondary immune responses and
month) [10]. is part of the immunoregulatory process. RF is produced
during bacterial and viral infections, probably in
Interferences response to immune complexes containing microbial
All fluid-phase immunochemical assays may be antigens. Polyclonal stimulation induces low-affinity
susceptible to nondetection of antigen excess and should polyreactive IgM-RF, which can also be found in healthy
be carefully assessed. Pharmaceuticals may affect RF individuals. There is evidence of a preclinical or
concentration as follows: (1) interferon alfa-2a and symptomatic phase of the disease.
methotrexate may decrease serum RF levels; (2)
methyldopa, oral contraceptives, and oxyphenisatin may The diagnosis of RA is made on the basis of history and
increase serum RF levels; and (3) nonsteroidal physical examination findings together with laboratory
antiinflammatory drugs may decrease or have no testing and imaging of joints. The diagnosis may be
significant effect on serum RF levels [11]. Several difficult in early stages of the disease. The
potential interferences have been described for internationally accepted 1987 classification criteria of
1032
Rheumatoid Factor
the American College of Rheumatology, although and enzyme immunoassay (1% of laboratories, n = 13).
primarily developed as a basis for recruitment of patients Imprecision values (% coefficient of variation) for
into clinical trials, are often also used as a diagnostic tool quantitative measurement of RF were acceptable and
for individual patients [15]. For the latter purpose, generally less than 10%. Criteria for measurement of RF
especially in early diagnosis, performance is far from require that laboratories achieve 80% participant
optimal. These guidelines include a set of seven common consensus of qualitative results (negative or positive) or
criteria that should be considered: (1) morning stiffness within 3 SD of peer group mean for quantitative
> 1 hour, (2) arthritis of > 3 of 14 joints/joint groups, (3) results. Acceptable performance was achieved for
arthritis of hand joints, (4) symmetrical swelling of joint, qualitative reporting, with > 98.5% of laboratories
(5) rheumatoid nodules in specific locations, (6) positive reporting the consensus test result.
RF, and (7) radiographic changes suggestive of joint
erosion. The presence of four of seven criteria is References
required to establish the presence of RA for study 1 Whaler E. On the occurrence of a factor in
purposes. In addition, the first four criteria must be human serum activating the specific
present for 6 weeks. The European League Against agglutination of sheep blood corpuscles. Acta
Rheumatism (EULAR) recently recommended that the Pathol Microbiol Scand. 1940;17:172-178.
following factors predicting persistent and erosive 2 Rose HM, Ragan C, Pearce E, Lipmann MO.
disease should be measured: number of swollen and Differential agglutination of normal and
tender joints, ESR or CRP, levels of RF and anti- sensitized sheep erythrocytes by sera of patients
citrullinated protein (anti-CCP) antibodies, and with rheumatoid arthritis. Proc Soc Exp Biol
radiographic damage [16]. Med. 1948;68:1-11.
3 Singer J, Plotz C. The latex fixation test.
The presence of RF is still considered an important Application to the serologic diagnosis of
feature of RA. RF can be found in 70% to 90% of RA rheumatoid arthritis. Am J Med. 1956;21:888-
patients, though they are far from specific for this 92.
disease. The autoantibodies most frequently found in 4 Melamies L, Ruutsalo H, Nissila H. Evaluation
patients with RA are antibodies against IgG (IgM of a quantitative immunoturbidimetric assay for
rheumatoid factor) and antibodies against citrullinated Rheumatoid factors. Clin Chem. 1986;32:1890-
proteins (anti-CCP). Meta-analysis of clinical studies 1894.
involving anti-CCP and RF conclude that anti-CCP 5 Finley P, Hicks J, Williams J, Hinlicky J, Lichtl
antibodies are more specific than RF for diagnosing RA D. Rate nephelometric measurement of
and may better predict erosive disease [17,18]. rheumatoid factor in serum. Clin Chem.
1979;25:1909-1914.
As a diagnostic test, the use of RF is limited because a 6 National Committee for Clinical Laboratory
positive RF can also be detected in other rheumatic Standards. User Comparison of Quantitative
disorders, infections, and in apparently healthy Clinical Laboratory Methods Using Patient
individuals (Table 2) [19]. The prevalence of RF Samples: Proposed Guideline. Vol 5. Wayne,
positivity in the normal population increases with age; PA, 1985, NCCLS.
by age 70, up to 10% of the population may be positive. 7 Ash K. Reference intervals (normal ranges): a
Thus the presence of a positive serum RF is not challenge to laboratories. Am J Med Tech.
sufficient in itself to make a diagnosis of RA. In 1980;46:504-11.
addition, about 15% of patients meet diagnostic criteria 8 Roberts-Thomson P, McEvoy R, Langhans T,
for RA and do not have a positive test for RF. Bradley J. Routine quantification of rheumatoid
Importantly, combining the presence of RF with anti- factor by rate nephelometry. Ann Rheum Dis.
CCP can further increase the specificity for RA. The 1985;44:379-383.
combined use of RF and anti-CCP is currently 9 Highton J, Hessian P, Small B, Palmer D.
recommended [16]. A negative result in both assays Clinical advantages from measurement of IgM-
makes RA unlikely, although it does not exclude the rheumatoid factor by enzyme immunoassay.
disease entirely. A positive result in both tests raises the Rheumatology. 1986;25:20-25.
probability of RA to approximately 90% to 100%. The 10 Guder WG, Narayanan S, Wisser H, Zawta B.
prevalence of RA is markedly increased in individuals Samples: From the Patient to the Laboratory:
with RF antibodies of more than one isotype, most The Impact of Preanalytical Variables on the
frequently a combination of IgM and IgA. Quality of Laboratory Results. Darmstadt,
Germany: Git Verlag; 1996.
Rheumatoid Factor Performance Goals 11 Young DS. Effects of Drugs on Clinical
Survey data from the 2007 College of American Laboratory Tests. 4th ed. Washington, DC:
Pathologists Participant Summary Report summarize AACC Press; 1995:3-527.
both qualitative and quantitative assay methods. A 12 Tietz NW. Clinical Guide to Laboratory Tests.
majority of laboratories reporting quantitative RF results 3rd ed. Philadelphia: Saunders; 1995:544-5.
utilized the immunoturbidimetric method. Quantitative 13 Klareskog L, Padyukov L, Lorentzen J,
RF test results were reported using turbidimetric Alfredsson L. Mechanisms of disease: genetic
immunoassay methods (62% of laboratories; n = 526), susceptibility and environmental triggers in the
nephelometric methods (37% of laboratories; n = 315),
1033
Rheumatoid Factor
development of rheumatoid arthritis. 17 Nishimura K, Sugiyam D, Kogat Y, Tsuji G,

Rheumatology. 2006;2:425-433. Nakazawa T, Kawano S et al. Meta-analysis:
14 Lee DM, Weinblatt ME. Rheumatoid arthritis. diagnostic accuracy of anti-cyclic citrullinated
Lancet. 2001;358:903-11. peptide antibody and rheumatoid factor of
15 Arnett FC, Edworthy SM, Bloch DA, McShane rheumatoid arthritis. Ann Intern Med.
DJ, Fries JF, Cooper NS et al. The American 2007;146:797-808.
Rheumatism Association 1987 revised criteria 18 Avouac J, Gossec L, Dougados M. Diagnostic
for the classification of rheumatoid arthritis. and predictive value of anti-cyclic citrullinated
Arthritis Rheum. 1988;31:315-24. protein antibodies in rheumatoid arthritis: a
16 Combe B, Landewe R, Lukas C, Bolosiu H, systematic literature review. Ann Rheum Dis.
Breedveld F, Dougados M et al. EULAR 2006;65:845-851.
recommendations for the management of early 19 Mierau R, Genth E. Diagnosis and prognosis of
arthritis: report of a task force of the European early rheumatoid arthritis, with special
Standing Committee for International Clinical emphasis on laboratory analysis. Clin Chem
Studies Including Therapeutics (ESCIST). Ann Lab Med. 2006;44:138-143.
Rheum Dis. 2007;66:34-45.
1034
Rheumatoid Factor
Table 1: Methods for Rheumatoid Factor

Principle: Latex particles coated with IgG (usually human or rabbit) are mixed with serum; presence of RF in
test serum results in agglutination of latex particles. Provides qualitative or semiquantitative results.
Comments: Visual detection method; large inter-observer variation when Ab titers are near assay sensitivity.
Method 2: Turbidimetry
Principle: Agglutination of latex particles coated with IgG antigen. Increasing levels of RF cause increasing
agglutination of latex particles, resulting in increased turbidity of the solution. The level of turbidity is measured
by absorbance spectrophotometry and is proportional to the RF level in the sample.
Usage: Frequently used (62% of laboratories reporting quantitative results)
Comments: Provides rapid (typically 10 minutes) quantitative as well as semi-quantitative or qualitative
results. Readily applied to automated general chemistry analyzers utilized to process routine chemistry tests on
patient specimens.
Method 3: Nephelometry
Principle: Agglutination of latex particles coated with IgG antigen. Increasing levels of RF cause increasing
agglutination of latex particles, resulting in increased turbidity of the solution. Measuring light reflected at various
angles using a nephelometer assesses turbidity.
Usage: Frequently used (37% of laboratories reporting quantitative results)
Comments: Provides quantitative as well as semi-quantitative or qualitative results. Requires specialized
instrumentation (nephelometer).
Method 4: Enzyme-Linked Immunosorbent Assay
Principle: ELISA technology in which human IgG attached to wells of a polystyrene micro-titer plate. Addition
of specimen containing RF binds to immobilized IgG, resulting in antigen-antibody complex. Subsequent addition
of enzyme-conjugated anti-human IgM binds the RF immune complex. The level of RF is proportional to the
level of enzyme activity.
Usage: Rarely used (<1% of laboratories reporting quantitative results)
Comments: Provides quantitative as well as semi-quantitative or qualitative results.
Table 2: Causes of Rheumatoid Factor Positivity
Age older than 60 years
Rheumatoid diseases
Rheumatoid arthritis
Sjgrens syndrome
Systemic lupus erythematosus
Viral infections
Hepatitis C
Hepatitis B
HIV infection
Parvovirus
Influenza
Bacterial infections
Endocarditis
Osteomyelitis
Tuberculosis
Chronic inflammatory conditions

Inflammatory bowel disease
Sarcoidosis
Pulmonary fibrosis
Primary biliary cirrhosis
Malignancy
1035
Rubella
Rubella
Terry Pry and Greg Maine i
Name: Rubella
Clinical significance: Refer to Chapter 7, Immunological Reactions, and Chapter 8,
Immunochemical Techniques, in the 5th edition of Clinical Chemistry: Theory, Analysis, and
Correlation.
Principles of Analysis and Current Usage epidemiological reasons. As such, in many countries,
Rubella, also known as German measles, is a common determining a patients immune status to rubella IgG is
and normally benign viral disease that can occur in part of the routine antenatal screening panel.
unimmunized individuals. It has been largely eradicated
in countries with well-established vaccination programs, The laboratory diagnosis of rubella infection became
but outbreaks still occur, and the disease is still possible following the isolation of the virus in cell
widespread in developing countries. It is a highly cultures in 1962 [2]. Rubella virus, a Togavirus, is an
contagious viral disease, usually transmitted from person enveloped virus with a single-stranded RNA genome.
to person via the respiratory route. Following respiratory The structural proteins consist of capsid and
transmission of rubella virus, replication of the virus is glycoproteins E1 and E2. Since viremia is usually
thought to occur in the nasopharynx and regional lymph detectable only in the last week before the onset of rash,
nodes. Rubella spreads via the lymph nodes into the virus isolation techniques are not suited for routine
blood, where it induces an immune response resulting in diagnosis because of the short period of viral excretion
lasting immunity. Viremia occurs 5 to 7 days after into the blood stream.
exposure, with spread of the virus throughout the body
and the chance of transplacental infection of the fetus. The diagnostic laboratory tests developed for the
The virus can be detected in secretions from the assessment of rubella infection are based on a number of
nasopharynx at approximately 4 days before to 4 days serological methods discussed below. Figure 1 shows the
following the onset of the typical rash associated with a typical relation between the virological and
rubella infection. immunological features of infection. The appearance of
IgM antibody to rubella virus is seen very early
Although relatively mild in most instances, rubella can following the onset of the illness. IgM typically reaches
lead to serious complications for pregnant women, peak concentrations within 7 to 10 days after infection
especially during the first 4 months of gestation. and then gradually disappears within 4 to 5 weeks. The
Intrauterine transmission of a primary rubella infection measurement of rubella IgM is primarily used for the
can cause severe fetal damage, leading to miscarriage or assessment of primary rubella infection in pregnant
severe birth defects in the newborn. Up to 20% of infants women. In contrast to IgM, IgG antibody continues to
born to mothers infected during the first half of increase for several months following the onset of rash,
pregnancy develop congenital rubella syndrome (CRS). and elevated levels of serum antirubella IgG may persist
The most common congenital defects are cataracts, heart for decades. The avidity of rubella IgG antibody is low
disease, deafness, and mental retardation. following primary antigenic stimulus but matures rapidly
to high avidity within 2 to 3 months. Hence, the clinical
Clinical diagnosis of rubella infection is difficult; only utility of the rubella IgG avidity assay is limited and
half of illnesses diagnosed clinically as acute rubella must be used in the context of other rubella serological
infection are actually found to be caused by the rubella testing described below to distinguish between a primary
virus. Since rubella-like illnesses can be induced by and non-primary rubella infection.
other viruses that have no teratogenic potential,
laboratory differentiation of rubella from other rash- Neutralization Antibody Test
causing infections (e.g., measles, parvovirus B19, human The isolation of the rubella virus gave rise to tests based
herpes virus 6, enteroviruses in developed countries, and on the ability of viral-specific antibodies to neutralize
various endemic arbovirus) is essential [1]. Accurate the infectivity of the virus (Table 1, Method 1).
laboratory diagnosis of past or recent rubella infection is Neutralization is used to identify both virus isolates and
vital for counseling a pregnant patient, as well as for immune response to the virus. Neutralizing antibodies,
i
Rubella
Methods edition: Not done
Third edition: Not done
Fifth edition: Terry Pry, Greg Maine
1036
Rubella
which may be of the IgG or IgM class, appear in serum rubella virus antigen (Table 1, Method 4). This test is
approximately 1 to 2 days following the onset of rash most useful for detecting serological evidence of past
and reach peak titers usually within 3 to 4 weeks. rubella infection (IgG antibodies) and is thus suited for
Rubella IgM antibody is rarely detectable for longer than screening of immunity to rubella.
4 to 5 weeks following an acute illness. This test is
rarely used today because it is too time consuming to Hemolysis-in-Gel Test
perform and is a labor-intensive procedure. The single radial hemolysis, or the hemolysis-in-gel test,
has been used as a simple and sensitive method for
Complement Fixation Test rubella screening (Table 1, Method 5). The test is based
Like the neutralization procedure, the complement on the diffusion of antibodies through an agarose gel
fixation test is rarely used today. This test is based upon plate containing erythrocytes coated with rubella
the fixation of complement added to antigen-antibody antigen. Complement is included within the gel or may
complexes (Table 1, Method 2). Visualization of the be layered on top of the gel. Patient serum is added to
complement fixation reaction is mediated through the wells punched into the gel. Complement-mediated lysis
use of sensitized erythrocytes. If a specimen contains of erythrocytes occurs if the serum diffusing through the
rubella-specific antibodies, complement will attach to gel contains antibodies to rubella. Following an
the antigen-antibody complexes and be unavailable for overnight incubation, the diameter of the hemolytic
lysis of the sensitized erythrocytes. Rubella-specific zones is measured. The area of the hemolyzed zone is
complement-fixation antibodies generally appear in proportional to the concentration of antibody within the
serum within 2 to 3 weeks following infection and may sample.
remain detectable for 2 to 5 years after the infection.
Although this test is simple to perform and requires no
The complement fixation test is relatively insensitive and special equipment, the shelf life of reagents is limited to
cannot be used as a single test for the diagnosis of a a few weeks. Some sera have been found to give atypical
rubella infection. Test results are dependent upon the hemolysis zones. The presence of a nonhemolytic area in
quality of the antigen used in the procedure, particularly the center of the zone, called a halo, has been attributed
the proportions of soluble and virion components [3]. to rheumatoid factor or some other factors [6]. In
Both the neutralization and complement fixation addition, some acute-phase sera contain a blocking
procedures have been replaced by the hemagglutination factor which, when present in high concentration,
inhibition (HI) test (see below). completely blocks hemolysis.
Hemagglutination Inhibition (HI) Test Enzyme Immunoassays

The HI test was the first technique that was widely Rubella-IgG Assay: Most assay formats are
available for the detection of antibodies to rubella. This indirect, in which rubella antigen (peptide, recombinant,
test is based on the ability of rubella virus to agglutinate or whole-virus antigen) is attached to the solid phase
avian erythrocytes (Table 1, Method 3) [4]. Inhibition of (Table 1, Method 6). The sensitivity and specificity of
agglutination by the binding of specific antirubella the assay have been shown to be affected by the source
antibodies to the agglutinin is utilized in this test. of antigen. The typical test sample is maternal or fetal
Following infection with rubella, hemagglutination- blood. The limitation with this assay is that it cannot
inhibition antibodies increase very rapidly, with distinguish between active infection and immunity.
maximum titers occurring approximately 8 days Ideally, rubella IgG serology should be performed in
following the onset of rash. The hemagglutination women prior to pregnancy to verify rubella immunity. If
inhibition test detects all classes of antibody and the patient is nonimmune, the rubella vaccine should be
therefore can be used for the selective detection of IgM- administered and the test repeated to verify the
class specific antibodies. development of humoral immunity to the rubella virus.
The HI test demonstrates excellent correlation with the Rubella-IgM Assay: The development of
neutralization procedure but is simpler to perform and sensitive and specific enzyme-linked immunoassay
more sensitive. In addition, the HI procedure is able to (ELISA) techniques for IgM, which is present in serum
detect all classes of antibody and does not require live for a limited period following infection, has been a
virus, instead utilizing antigen that can be stored major advancement in the diagnosis of rubella in early
lyophilized. A titer of 1:8 is considered negative, and a pregnancy (Table 1, Method 7). ELISA tests are simple
titer of > 1:32 suggests an earlier infection or successful to perform, and many have been automated. Two assay
vaccination [5]. The test, however, does have drawbacks. formats are described: (1) indirect ELISA and (2) IgM-
The presence of nonspecific inhibitors, particularly low- capture ELISA. In indirect ELISA, viral antigen is
density beta-lipoproteins and agglutinins, can interfere bound to the solid phase, and a labeled antihuman IgM is
with the assay, and these must be removed prior to used as the conjugate. These assays now generally use a
analysis. Kaolin, heparinmanganous chloride, dextran, pretreatment step that includes an absorbent reagent to
and red blood cells have been utilized for this purpose. remove rubella IgG from the sample. In IgM-capture
ELISA, antihuman IgM antibodies are attached to the
Passive Hemagglutination Test solid phase, and a conjugate of rubella virus antigen and
The passive hemagglutination test is based on the labeled antirubella virus antibody is used for detection.
agglutination of erythrocytes that have been coated with Although, IgM capture ELISA assays generally
1037
Rubella
demonstrate reduced non-specific assay interference, patients serum rubella IgG levels should be above 29 IU
interference is still observed and must be considered [7]. to provide a reliable AI result [13].
The increased sensitivity of recent rubella IgM assays Latex Agglutination

can lead to the detection of nonspecific IgM. Since the The latex agglutination procedure for antibodies to
incidence of rubella has diminished in countries with rubella is often used for the semiquantitative
high rubella vaccination coverage, the proportion of determination of immune status. The procedure is
nonspecific rubella IgM reactivity among all samples performed by mixing latex particles, previously coated
with rubella IgM reactivity has increased. Nonspecific with rubella virus antigen, with untreated serum on a
false positives are possible due to (1) other IgM Teflon-coated card (Table 1, Method 9). Once added, the
antibodies which cross-react arising in patients infected mixture is mechanically rotated for a short period, and
with Epstein-Barr virus, cytomegalovirus, and human the results read macroscopically. The test is considered
parvovirus B19, and (2) rheumatoid factor resulting from positive if agglutination of the latex particles is observed
patients with autoimmune diseases such as rheumatoid following the mixing period. Advantages of the latex
arthritis and systemic lupus erythematous [5]. Therefore, agglutination procedure include a rapid assay time, no
rheumatoid factor, parvovirus IgM, and heterophile tests need for specialized or expensive equipment, and no
should be done if a false-positive result is suspected. In need for pretreatment of serum prior to analysis.
addition, low levels of IgM may not be indicative of
primary infection when sensitive tests are used, because PCR
IgM antibody has been shown to persist for much greater The use of reverse transcription followed by PCR
than the typical 8 to 12 weeks [8]. A rubella IgM- amplification has enabled the detection of rubella virus
positive result alone does not accurately predict the risk RNA in tissue (Table 1, Method 10) [14]. This method is
of fetal infection and should be followed by other rubella typically used as a reflex to test for fetal infection
serological tests described below prior to prenatal following detection of a maternal primary infection to
diagnosis. increase the reliability of the diagnosis [15]. Studies
show detection of rubella RNA by PCR has a sensitivity
Rubella-IgG Avidity Assay: The rubella IgG avidity of 83% to 100% and 100% specificity [16,17]. A rubella
assay measures the functional binding affinity (Avidity IgM EIA recombinant assay uses a recombinant rubella
Index) of the IgG class of antibody in response to virus-like particle and is based upon the IgM capture
infection (Table 1, Method 8). Avidity is a measure of principle.
the overall strength of binding of an antigen with many
antigenic determinants and multivalent antibodies, Reference and Preferred Methods
whereas affinity refers to the strength of binding between The hemagglutination inhibition test is generally
a single antigenic determinant and an individual considered the gold-standard test against which other
antibody combining site. The rubella IgG avidity test is rubella tests are measured. The use of ELISA has
ordered when both IgG and IgM rubella antibodies are become the predominant method of analysis for both
detected in the mother. Following primary infection with IgG and IgM antibody to rubella, because they are
rubella virus, low-avidity IgG is produced first. As the easier, more rapid, and more sensitive than earlier
infection progresses, the affinity of the rubella IgG neutralization and HI tests. A number of compounds
increases with time, and the presence of high-avidity have been employed as substrates, including those
rubella IgG can be used in conjunction with other whose products are measured spectrophotometrically, by
serological tests as a marker of past or recurrent fluorescence, or by chemiluminescence. This technique
infection. The rubella-IgG avidity assay differentiates is the most common in use for the determination of
between antibodies with high or low avidity to the recent infection (IgM) and immune status (IgG).
antigen and therefore helps to differentiate between
primary and non-primary infection. The assay measures Specimen
the functional binding affinity of the IgG class of Serum or plasma for rubella serological testing should
antibody in response to infection. ideally be collected at the onset of disease and 2 to 3
weeks later in order to determine acute and convalescent
In avidity assays, the patients IgG is allowed to bind to titers. Serum can be stored at refrigerator temperatures
its antigen, followed by elution with or without a protein (4C to 8C) for 5 to 7 days and should be frozen if
denaturant, such as urea or diethylamine. The protein testing is to be delayed. There is no significant
denaturant dissociates weakly bound antibodies from difference in antibody levels from blood collected by
their specific antigens, but strongly bound antigens fingerstick versus venipuncture [18]. Adaptation of
remain intact. Thus the proportion of IgG remaining rubella IgM immunoassay using dried venous blood
antigen-bound indicates the relative amount of high- spots has also been described [19]. Saliva also proved
affinity IgG. The avidity index (AI) is defined as the useful as an alternative, less invasive sample [20].
ratio of (high avidity rubella IgG present)/(total rubella
IgG titer) 100. The cutoff point differentiates the low Interferences
avidity index and the high avidity index. Different values The potential for assay interference varies by assay
of the low avidity index cutoff points have been method. As described earlier, potential interferents
reported, ranging from < 30% to 55% [9-12]. The include rheumatoid factor, human anti-mouse antibodies,
1038
Rubella
and cross-reacting IgM antibodies (Epstein-Barr, CMV, infection may not contain any detectable amounts of
parvovirus B19). rubella IgG antibodies or may have antibody levels less
than 10 IU/mL. Moreover, antibody levels after
Rubella Reference Interval vaccination may drop under the threshold of 10 IU/mL.
IgM antibodies in response to a primary rubella infection
can be detected in the patient by a sensitive EIA within 5 Rubella Performance Goals
to 10 days after the onset of the rash; they disappear Survey data from the 2007 College of American
rapidly, usually within 7 to 10 weeks. In some Pathologists Participant Summary Report indicate that
individuals, low levels of IgM may last for an extended approximately 67% of laboratories (n = 626) reporting
time, up to 1 year. IgM antibodies in response to quantitative results used a chemiluminescence
vaccination can be detected in the vaccinee between 3 immunoassay, and 24% used an enzyme immunoassay
and 12 weeks post-vaccination. to measure rubella antibody. Imprecision values (%
coefficient of variation) for quantitative measurement of
Interpretation rubella antibody was generally less than 15%. Criteria
Rubella was first discovered in the early 18th century for measurement of rubella antibody require that
and described as a relatively mild, self-limiting disease laboratories achieve 90% participant consensus of
with few complications. However, following a rubella qualitative results (negative or positive), and acceptable
epidemic in 1941, an increased incidence of neonatal performance was achieved across all assay methods.
cataract cases associated with deafness, cardiac
anomalies, and microcephaly was observed [21]. From References
these initial observations, the medical community 1 Banatvala J, Best J, OShea S, Dudgeon J.
recognized the teratogenic effects of the rubella virus. Persistence of rubella antibodies after
These and other defects (including mental retardation vaccination: detection after experimental
and low birth weight) are now collectively known as the challenge. Rev Infect Dis. 1985;7(suppl.1):S86-
congenital rubella syndrome (CRS) [8]. 90.
2 Weller TH, Neva FA. Propagation in tissue
Immunization programs in most developed countries culture of cytopathic agents from patients with
have essentially eradicated rubella infections. However, rubella-like illness. Proc Soc Exp Biol Med.
rubella may reemerge if vaccination programs cannot be 1962;111:215-225.
adequately sustained. In undeveloped nations, outbreaks 3 Cradock-Watson JE. Laboratory diagnosis of
among individuals not vaccinated against the virus are rubella: past, present and future. Epidemiol
still common. Between immunization and natural Infect. 1991;107:1-15.
infections, > 85% of people carry antibodies to the 4 Stewart G, Parkman P, Hopps H, Douglas R,
rubella virus, and > 90% of these will have lifelong Hamilton J, Meyer H. Rubella-virus
immunity to the disease [22]. Primary rubella virus hemagglutination-inhibition tests. N Engl J
infection occurring after birth is typically a mild, self- Med. 1967;276:554-7.
limiting disease characterized by a maculopapular rash, 5 Mendelson E, Aboudy Y, Smetana Z,
pyrexia (fever), tiredness, and lymphadenopathy. Tepperberg M, Grossman Z. Laboratory
Maternal infection acquired within the first trimester of assessment and diagnosis of congenital viral
pregnancy can be a serious threat to the developing infections: Rubella, cytomegalovirus (CMV),
fetus. If a pregnant woman is infected with rubella varicella-zoster virus (VZV), herpes simplex
during the first trimester, there is an approximately 90% virus (HSV), parvovirus B19 and human
chance that the fetus will be infected, and the majority of immunodeficiency virus (HIV). Reprod
these will have congenital defects. If the fetus is exposed Toxicol. 2006;21:350-382.
to rubella between the 12th and 20th weeks of gestation, 6 Hedman K, Salonen EM, Keski-Oja J, and
the risk of fetal infection declines to approximately 50%, Rih K. Single-serum radial hemolysis to
with about 30% of infants suffering complications. detect recent rubella virus infection. J Infect
Rubella infections that occur after 20 weeks of gestation Dis. 1986;154:1018-1023.
are not normally associated with any complications. 7 Tipples G, Hamkar R, Mohktari-Azid T, Gray
M, Ball J, Head C et al. Evaluation of rubella
An algorithm for assessment of rubella infection in IgM enzyme immunoassays. J Clin Virol.
pregnancy is shown in Figure 2. Screening pregnant 2004;30:233-238.
women for the presence of antirubella IgG antibodies 8 Banatvala J, Brown D. Rubella. Lancet.
typically is best performed when a women plans to 2004;363:1127-37.
become pregnant, or at the first prenatal visit to her 9 Bottinger B, Jensen P. Maturation of rubella
physician. If the antirubella IgG is detected, the woman IgG avidity over time after acute rubella
has immune status. If there are no detectable antibodies, infection. Clin Diagn Virol. 1997;8:105-111.
a nonpregnant woman is usually offered vaccination but 10 Thomas H, Charlett A, Cubie H. Specific IgG1
should avoid pregnancy for 3 months. avidity maturation after rubella vaccination; a
comparison with avidity maturation after
Protection and immunity is considered with antibody primary infection with wild rubella virus.
levels of 10 to 15 international units (IU) of IgG per mL Serodiag Immunother Infect Dis. 1995;7:75-80.
[23]. Specimens taken very early in the acute phase of
1039
Rubella
11 Guttierrez J, Rodriguez M, Ory F, Piedrola G, 17 Revello M, Baldanti F, Sarasini A, Zavattoni

Maroto M. Reliability of low-avidity IgG and M, Torsellini M, Gerna G. Prenatal diagnosis of
of IgA in the diagnosis of primary infection by rubella virus by direct detection and
rubella virus with adaptation of a commercial semiquantitation of viral RNA in clinical
test. J Clin Lab Anal. 1999;13:1-4. samples by reverse transcription PCR. J Clin
12 Mubareka S, Richards H, Gray M, Tipples G. Microbiol. 1997;35:708-13.
Evaluation of commercial rubella 18 Pinsky N, Loepfe T, Jacobson R, Vierkant R,
immunoglobin G avidity assays. J Clin Poland G. Comparison of fingerstick versus
Microbiol. 2007;45:231-233. venipuncture for antibody testing of measles
13 Enders G, Knotek F, Pacher U. Comparison of and rubella. Scand J Infect Dis. 2003;35:107-
various serological methods and diagnostic kits 109.
for the detection of acute, recent, and previous 19 Karapangiotidis T, Riddell M, Kelly H.
rubella infection, vaccination, and congenital Detection of rubella immunoglobin M from
infections. J Med Virol. 1985;16:219-232. dried venous blood spots using a commercial
14 Grangeot-Keros L, Enders G. Evaluation of a enzyme immunoassay. Diagn Microbiol Infect
new enzyme immunoassay based on Dis. 2005;53:107-111.
recombinant rubella virus-like particles for 20 Vijaylakshmi P, Muthukkaruppan V,
detection of immunoglobulin M antibodies to Rajassundari A, Korukluoglu G, Nigatu W,
rubella virus. J Clin Microbiol. 1997;35:398- Warrener L et al. Evaluation of a commercial
401. rubella IgM assay for use on oral fluid samples
15 Tang J, Aarons E, Hesketh L, Strobel S, for diagnosis and surveillance of congenital
Schalasta G, Jauniauz E. Prenatal diagnosis of rubella syndrome and postnatal rubella. J Clin
congenital rubella infection in the second Virol. 2006;17:265-268.
trimester of pregnancy. Prenat Diagn. 2003; 21 Gregg NM. Congenital cataract following
23:509-512. German measles in the mother. Trans
16 Mace M, Cointe D, Six C, Levy-Bruhl D, Opthalmol Soc Aust. 1941;3:35-46.
Parent du Chatelet I, Ingrand D et al. 22 Llewellyn-Jones D. Obstetrics and Gynecology.
Diagnostic value of reverse transcription-PCR St Louis: Mosby; 1994:135.
of amniotic fluid for prenatal diagnosis of 23 Skendzel L. Rubella immunity: defining the
congenital rubella infection in pregnant women level of protective antibody. Am J Clin Pathol.
with confirmed primary rubella infection. J Clin 1996;106:170-74.
Microbiol. 2004;42:4818-20.
1040
Rubella
Table 1: Methods for Rubella
Method 1: Neutralization test

Principle: Antibodies neutralize the infectivity of virus
Comments: Detects IgG and IgM antibodies; laborious; performed only in reference labs
Method 2: Complement fixation test
Principle: Presence of rubella antibodies results in attachment of complement to AgAb complexes; complement
not available for lysis of hemolysin-sensitized cells
Comments: Detects IgG-class antibodies only
Method 3: Hemagglutination inhibition test
Principle: Based on inhibition of erythrocyte agglutination by binding of antirubella antibodies to an agglutinin
Comments: Detects all classes of antibody; interference by nonspecific inhibitors
Method 4: Passive hemagglutination test
Principle: Agglutination of erythrocytes coated with rubella virus antigen
Comments: Detects evidence of past infection; useful for screening of rubella immunity
Method 5: Hemolysis-in-gel test
Principle: Antibody diffusion through agarose gel containing erythrocytes coated with rubella antigen;
complement-mediated cell lysis occurs if sera contains antibodies to rubella
Comments: Limited reagent shelf life; some interferences noted
Method 6: Rubella IgG
Principle: ELISA: Detection of virus-specific IgG Ab bound to a solid phase by a labeled secondary anti-IgG
Ab.
Usage: Marker of past infection
Comments: Limitation of result: not conclusive of latent or active infection. Comparisons between
commercial IgG assays correlate well with each other, however; no international standard yet exists. Fast and
sensitive commercialized assays available. Detection typically via chemiluminescent or fluorescent substrate
Method 7: Rubella IgM
Principle: ELISA: Detection of virus-specific IgM Ab bound to a solid phase by a labeled secondary anti-IgM
Ab. Detection typically via chemiluminescent or fluorescent substrate.
Usage: Marker of active or recent rubella infection
Comments: IgM antibodies may persist for months or years; false positives a limitation; fast and sensitive
commercialized assays available
Method 8: Rubella IgG Avidity
Principle: ELISA: Two rubella IgG ELISA tests are performed on the same sample. After primary antibody
incubation, the solid phase of each assay is washed with either buffer (control) or a chaotropic reagent (urea,
diethylamine) to remove low-avidity antibodies. After another wash, the conjugate is added, and the signal is
measured. The avidity index is the ratio of (rubella IgG titer with chaotropic wash step/rubella IgG titer with
buffer wash step) 100, that is, the ratio of rubella high-avidity IgG over the total rubella IgG titer.
Usage: Differentiates between primary infection (low avidity IgG) and recurrent or persistent infection (high
avidity IgG).
Comments: Avidity testing should be performed early in gestation (within the first trimester). Fully
automated high-throughput analytical systems available with integrated avidity assays.
Principle: Latex particles coated with virus antigen mixed with serum; presence of rubella antibody in test serum
results in agglutination of latex particles
Comments: Visual detection method; large inter-observer variation when Ab titers are near assay sensitivity
Method 10: PCR
Principle: RNA extraction, reverse transcription, amplification, and detection of amplified sequences using
fluorescent probes or dye in a specialized analyzer.
Usage: Detection of viral RNA in clinical samples
Comments: Not prone to contamination. Interpretation of very low result questionable. Samples can be
frozen for subsequent analysis by other labs.
1041
Rubella
Figure 1:
IgG / IgG Avidity

1042
Rubella
Figure 2: Algorithm for assessment of rubella infection in pregnancy:

1043
Salicylates
Salicylates
Gus Koerbin and Julia M. Potter
Name: Salicylate, acetylsalicylic acid, aspirin
Acetylsalicylic acid Salicylic acid
Molecular formula: C9H8O4 C7H6O3
Molecular mass: 180.15 D 138.12 D
Chemical class: Benzoic acid derivative Hydroxybenzoic acid
Structure:
Acetylsalicylic acid Salicylic acid

i plasma by extraction into acidified ethylene dichloride,
followed by back extraction into an aqueous phase as the
Acetylsalicylic acid (aspirin) is absorbed from the ferric complex, which is then measured at 540nm. In
stomach and rapidly hydrolyzed to salicylic acid 1948, Keller published a modification of this method [2]
(salicylate) and acetic acid in the blood. Aspirin which which eliminated the extraction steps (Table 1, Method
passes into the small intestine is hydrolyzed prior to 1b), and total salicylate was measured as the sum of
absorption or within the mucosa. Salicylate is rapidly acetylsalicylate and its metabolites. In 1950, Trinder [3]
absorbed into the blood. There are a variety of methods published the well-known protein precipitation method
for determination of salicylate in serum or plasma. The using mercuric chloride and hydrochloric acid (Table 1,
most widely used methods in the clinical laboratory Method 1c). Color is produced when salicylate is
employ either colorimetry or fluorescence polarization complexed with ferric ion supplied as ferric nitrate.
immunoassay, but other methods include high-
performance liquid chromatography (HPLC), gas A second type of colorimetric method involves reducing
chromatography (GC), ultraviolet spectroscopy, and salicylate with a mixture of phosphotungstic and
fluorescence spectrophotometry. phosphomolybdic acids (Folin-Ciocalteu reagent) (Table
1, Method 2) [4]. Weichselbaum and Shapiro [5]
Historically, the earliest method for measurement of published a method in which serum was mixed with this
salicylates is the colorimetric method, which measures reagent, and the resulting blue color was read at 660 nm.
the absorbance of the complex formed between In 1950, Smith and Talbot [6] published a similar
salicylates and ferric ions (Table 1, Method 1). The first method but shortened the color development step from
widely accepted method using this technique was 40 min to 3 min.
described by Brodie et al. [1] (Table 1, Method 1a) in
1944. In this procedure, salicylate is separated from Ultraviolet spectrophotometry has also been employed
as an analytical technique to measure acetylsalicylate
plus salicylate (Table 1, Method 3). Routh et al. [7]
i
Salicylates extracted the acetylsalicylate using the Brodie procedure
Previous and current authors of this method: and quantified the drugs using differential
First edition: Joseph Svirbely spectrophotometry by measuring the differences in
Methods edition: Joseph Svirbely absorbance at pH 9 and pH 13.5.
Second edition: Joseph Svirbely
Third edition: Ross L.G. Norris, Donald Davis, Julia Spectrofluorometry has also been used for the
M. Potter measurement of salicylates (Table 1, Method 4). In
Fourth edition: Not updated 1948, Saltzman [8] published a spectrofluorometric
Fifth edition: Gus Koerbin, Julia M. Potter method in which the salicylate was separated from
1044
Salicylates
protein by precipitation with dilute tungstic acid. Strong have described a method which enables the simultaneous
alkali was added to convert the salicylate into a more determination of acetylsalicylic acid and its metabolites
fluorescent form. The ion was excited at 370 nm, and on 0.2 mL of plasma or serum. Sample preparation is a
fluorescent emission was monitored at 460 nm. Potter relatively simple protein precipitation step using
and Guy [9] have published a micromethod for plasma perchloric acid. Chromatography utilizes a reversed-
that allows determination of protein-bound salicylate, phase column with a mobile phase of methanol in
free salicylate, acetylsalicylate, and total salicylate in aqueous buffer. The eluent is monitored with a UV
serum. Sephadex gel columns were used to separate the detector at 235 nm. Limit of detection was quoted as
protein-bound salicylate from the free forms. Salicylate being 0.16 mg/dL, which is adequate for most clinical
was measured by spectrofluorometry, with excitation at laboratory applications. The HPLC method of Mays et
305 nm and emission at 405 nm. al. [16] has an improved limit of detection, which may
be an advantage in pharmacokinetic studies where low-
In another fluorometric assay, Lever and Powell [10] in dose salicylates are used. However, the more complex
1973 described an enhancement and alteration of the sample preparation requiring 1 mL of plasma offsets this
salicylate fluorescence by the addition of magnesium advantage for most clinical laboratory applications. The
acetate tetrahydrate and sodium barbitone to the serum. method of Davis [17] utilizes a simple trichloroacetic
Fluorescence is measured with excitation at 340 nm and acid precipitation of proteins from 0.2 mL of plasma,
emission at 390 nm. and reverse-phase chromatography and use of
fluorescence detection avoids potential theophylline
The fluorescence polarization immunoassay (FPIA) interference as reported by Miceli et al. [18].
(Table 1, Method 5) assay for salicylates is one of the
most frequently used methods. A change in the Reference and Preferred Methods
polarization of fluorescent radiation emitted by the There is no reference method for salicylate
fluorescein-labeled tracer is measured. measurement. The simplest method for measuring
You and Bittikofer [11] reported an enzymatic assay that salicylates on a routine basis is the spectrophotometric
has good precision at low salicylate levels (less than 10 method of Keller, which yields a measure of total
mg/dL). The assay utilizes salicylate hydroxylase salicylate and its metabolites. The method of Brodie et
derived from Pseudomonas cepacia (Table 1, Method 7). al. requires time-consuming extraction steps and
Salicylate hydroxylase catalyzes the stoichiometric, involves the use of a toxic solvent, ethylene dichloride.
unidirectional conversion of salicylate and NAD(P)H to Recovery with the Brodie method is less than 90%
catechol and NAD(P) in the presence of molecular because of the double extraction. Likewise, Trinders
oxygen. The change in absorbance at 340 nm is method offers no advantages over Kellers method and
monitored and compared to a standard. A more specific also makes use of toxic mercuric ion.
modification of this method, which is claimed to be
adaptable to most discrete analyzers, was reported by Ultraviolet spectrophotometry measures salicylate and
Morris [12]. The improved specificity is gained by acetylsalicylate and has a sensitivity of 0.5 mg/dL.
monitoring the production of catechol rather than the However, because of its lack of specificity, it offers no
consumption of NADH. Several metabolites and analogs advantages over colorimetric procedures. The main
of salicylate which interfere in the former assay because advantage of spectrofluorometric methods is that they
they also consume NADH do not interfere in this have better sensitivity than the colorimetric methods.
modification in that they do not produce catechol. These methods do not measure acetylsalicylate directly,
since acetylsalicylate does not fluoresce. Instead they
Gas chromatography has been used for measuring derive the value indirectly by subtracting free salicylate
salicylates, since acetylsalicylate and its metabolites can from the total salicylate measured after the sample is
be separated and quantitated independently (Table 1, hydrolyzed.
Method 6). GC procedures [13,14] incorporate a
sialylation step after extraction of the salicylates from HPLC is the method of choice for separating and
the biological sample. In the method of Thomas et al. quantitating acetylsalicylate and its major metabolites.
[13], plasma is extracted into ether along with p-toluic Its use is recommended in studies requiring
acid (the internal standard). Following evaporation of the discrimination between these compounds. HPLC, in
ether, silylation is achieved by heating in the presence of common with GC, has good sensitivity and specificity,
bis (trimethylsilyl) trifluoroacetamide (BSTFA). but because no derivatization is required, it has the
Separation on the GC is achieved with a 5% OV-17 advantage of more rapid and simple sample preparation.
column run isothermally at a temperature of 150C, and Another benefit of both GC and HPLC is that large
detection is with a flame ionization detector (FID). numbers of samples can be processed simultaneously,
Acetylsalicylate and salicylate are both quantitated provided that the instrument is equipped with an
separately. Rance et al. [14] used a similar procedure, automatic injector. Because of economies of scale,
but by using temperature programming, they were able substantial cost benefits can be attained using automated
to measure salicylamide in addition to salicylate and HPLC or GC where large numbers of samples are to be
acetylsalicylate. assayed.
HPLC has been applied successfully to the quantitation Spectrophotometric assays remain the overall methods
of salicylate (Table 1, Method 8). OKruk et al. [15] of choice because they are relatively free from
1045
Salicylates
interference, have adequate accuracy and precision, and The pharmacokinetics of salicylate are dose dependent
are inexpensive and readily available to all laboratories. (Table 2), and hepatic clearance is saturable. The major
However, where equipment and expertise is available, metabolites are the glucuronide conjugate and salicyluric
HPLC is to be preferred. and gentisic acids. Further, the drug is protein bound,
and both plasma protein binding and clearance alter over
Clearly the final decision on the most appropriate the range of standard therapeutic doses. With saturation
method needs to be tailored for the individual laboratory. kinetics and changing protein binding, accumulation of
Factors to be considered can be broken down into salicylate is not uncommon, and the syndrome of
technical and nontechnical factors. Technical factors salicylism occurs (including tinnitus, nausea,
include accuracy, specificity, sensitivity and precision. hyperventilation, and confusion). With increasing
These need to be considered within the anticipated range concentrations and in overdose (either chronic or acute),
of concentrations to be assayed. Nontechnical factors there will also be respiratory alkalosis, metabolic
include cost, numbers of samples to be processed, and acidosis, vomiting, hyperthermia, coma, or seizures,
availability of expertise (i.e., if more technically among other signs and symptoms.
demanding assays such as HPLC are to be used).
Although at usual therapeutic doses and urinary pH only
Specimen a small percentage of salicylate (<5%) is cleared by the
Either serum or plasma are suitable samples for analysis. kidney, renal excretion can be used to advantage in the
There is no known interference by commonly used treatment of overdose. Salicylate is a weak acid (pKa 3),
anticoagulants. Most methods can be readily adapted for and if the pH of urine is made alkaline, the excretion of
urine specimens. salicylic acid can be increased significantly by
preventing the back-diffusion of ionized salicylate into
Interferences peritubular capillaries. In addition to alkalinization of
A major shortcoming of methods using the Folin- urine in this circumstance, gastric lavage should be
Ciocalteu reagent is the false-positive color produced by carried out and activated charcoal used. The further
reaction with tyrosine, tryptophan, and uric acid. A therapy of forced diuresis is not recommended or
report by Duffens et al. [19] suggests that diflunisal, a considered necessary, since the additional fluid load
difluorophenyl analog of salicylic acid, interferes in both carries the risk of pulmonary edema and further
the colorimetric and FPIA assays. They review other complicates the already complex systemic acid-base and
reports, showing cross-reactivities of greater than 200% disturbed electrolyte balances. In cases of severe
with the FPIA assay and greater than 50% for the toxicity, such as that with signs of cerebral involvement
Trinder method. As diflunisal fluoresces, it would be (seizure, coma, etc.) and/or hyperpyrexia, hemodialysis
expected to interfere in the fluorometric assays also. should be considered. This allows redress of the acid-
Sarma et al. [20] have shown that HPLC is not subject to base and electrolyte problems, as well as removal of
diflunisal interference. salicylate.
Salicylates Therapeutic Range

Analgesia Low dose 210 mg/dL (0.14 - 0.72mmol/L)
Anti-inflammatory High dose 530 mg/dL (0.36 2.16mmol/L)
Salicylism >20 mg/dL (>1.44mmol/L)
Marked toxicity >30 mg/dL (>2.16mmol/L)
Interpretation
Aspirin and nonacetylated salicylates (e.g., sodium Salicylates Performance Goals
salicylate) are available in numerous pharmaceutical Survey data from the 2007 College of American
formulations, many of which have been developed to Pathologists Participant Summary Report shows
limit gastric irritation, microulceration, and the imprecision values (% coefficient of variation) for
gastrointestinal intolerance and blood loss which may measurement of salicylates range from 1.8% to 7.0% at
ensue. In general, the formulations (enteric-coated and concentrations of approximately 60 mg/dL for assays
slow-release) are designed to encourage passage of the with > 20 participants. Acceptable performance criteria
drug from the stomach into the small intestine, from (Clinical Laboratory Improvement Amendments 88) for
where absorption then occurs. These changes alter the measurement of salicylate requires that laboratories be
pharmacokinetic profiles, particularly of the absorptive accurate to within 3.0 SD or 10% of the peer-group
phase. Nonionized aspirin is absorbed from the stomach mean, whichever is greater.
and then undergoes rapid hydrolysis within plasma
(Table 2). Hydrolysis is also rapid within the intestine References
and its mucosa. The compound benorilate is an ester of 1 Brodie BB, Udenfriend S, Coburn AF. The
salicylic acid and acetaminophen (paracetamol) which determination of salicylic acid in plasma. J
undergoes hydrolysis following absorption, releasing Pharmacol 1944;80:114-7.
both pharmacologically active moieties. 2 Keller W. A rapid method for the determination
of salicylates in serum or plasma. Am J Clin
Pathol 1947;17:415-7.
1046
Salicylates
3 Trinder P. Rapid determination of salicylate in in biological fluids by gas-liquid

biological fluid. Biochem J 1954;57:301-3. chromatography. J Pharm Pharmacol
4 Folin O, Ciocalteu V. On tyrosine and 1973;25:201-4.
tryptophan determinations in proteins. J Biol 14 Rance ML, Jordan BI, Nichols JD. A
Chem 1927;73:627-50. simultaneous determination of acetylsalicylic
5 Weichselbaum TE, Shapiro I. A rapid and acid, salicylic acid and salicylamide in plasma
simple method for the determination of salicylic by gas-liquid chromatography. J Pharm
acid in small amounts in blood plasma. Am J Pharmacol 1975;27:425-9.
Clin Pathol 1945;9:42-4. 15 OKruk RJ, Adams MA, Philp RB. Rapid and
6 Smith MJ, Talbot JM. The estimation of plasma sensitive determination of acetylsalicylic acid
salicylate levels. Br J Exp Pathol 1950;31:65- and its metabolites using reversed-phase high-
69. performance liquid chromatography. J
7 Routh JI, Shane NA, Arredondo EG, Paul WD. Chromatogr 1984;310:343-52.
Method for the determination of acetylsalicylic 16 Mays DC, Sharp DE, Beach CA, Kershaw RA,
acid in the blood. Clin Chem 1967;13:734-43. Bianchine JR, Gerber N. Improved method for
8 Saltzmann . Fluorophotometric method for the the determination of aspirin and its metabolites
estimation of salicylate in blood. J Biol Chem in biological fluids by high-performance liquid
1948;174:399-404. chromatography: application to human and
9 Potter GD, Guy J. A micromethod for the animal studies. J Chromatogr 1984;311:301-9.
analysis of plasma salicylate. Proc Soc Exp Biol 17 Davis D. Method of salicylate quantitation
Med 1964;116:658-60. utilizing trichloroacetic acid and reverse-phase
10 Lever M, Powell JC. Simplified fluorimetric chromatography. Unpublished data.
determination of salicylate. Biochem Med 18 Miceli JN, Aravind MK, Cohen SN, Done AK.
1973;7:203-7. Simultaneous measurements of acetaminophen
11 You K, Bittikofer JA. Quantification of and salicylate in plasma by liquid
salicylate in serum by use of salicylate chromatography. Clin Chem 1979;25:1002-4.
hydroxylase. Clin Chem 1984;30:1549-51. 19 Duffens KR, Smilkstein MJ, Bessen HA,
12 Morris HC, Overton PD, Ramsey JR, Campbell Rumack BH. Falsely elevated salicylate levels
RS, Hammond PM, Atkinson T, Price CP. due to diflunisal overdose. J Emerg Med
Development and validation of an automated, 1987;5:499-503.
enzyme-mediated colorimetric assay of 20 Sarma L, Wong SH, Dellafera S. Diflunisal
salicylate in serum. Clin Chem 1990;36:131-5. significantly interferes with salicylate
13 Thomas BJ, Solomonraj G, Coldwell BB. The measurements by FPIA-TDx and UV-VIS aca
estimation of acetylsalicylic acid and salicylate methods. Clin Chem 1985;31:1922-3.
1047
Salicylates
Tables
Table 1: Salicylates Methods Summary Table
Method 1: Ferric nitrate complex; spectrophotometric
Principle of analysis: Salicylates + Fe3+ Ferric complex (Amax 540 nm)
a. Extracts salicylate into ethylene dichloride
b. Eliminates extraction step
c. Deproteinates with mercuric chloride
Comments: Most commonly used clinical salicylate procedure; sensitive to 20 mg/dL; linear to 100
mg/dL; only interference is salicylate metabolites
a. Method 1a uses toxic ethylene dichloride
b. Method 1b less complicated but specific enough for clinical needs
c. Method 1c uses toxic mercuric chloride
Method 2: Folin-Ciocalteu reagent reduction; spectrophotometric
Principle of analysis: Phosphotungstic acid + phosphomolybdic acid reduce salicylates to form blue
complex (Amax 660 nm)
Comments: Not commonly used because of interferences; tryptophan, tyrosine, and uric acid react
with reagent to give falsely high readings
Method 3: Spectrophotometry; differential ultraviolet spectrophotometric
Principle of analysis: Measures salicylates + acetylsalicylate by differences in absorbance at pH 9.0
and pH 13.5
Comments: Not commonly used; has advantage of measuring acetylsalicylate; requires narrow
bandpass ultraviolet spectrophotometer; sensitive to 0.5 mg/dL
Method 4: Fluorometry; spectrofluorometric
Principle of analysis: Protein can be removed by precipitation; alkali intensifies fluorescence of
salicylate
Comments: Not commonly used; has advantage of greater sensitivity; requires smaller sample than
spectrophotometry
Method 5: Fluorescence polarization immunoassay; polarization immunoassay
Principle of analysis: Drug in sample competes with fluorescent-labeled drug for antibody sites;
change in fluorescent radiation is monitored
Comments: Widely used; commercially available and convenient; fewer interferences than
colorimetric methods but diflunisal a problem
Method 6: Gas chromatography; chromatographic separation
Principle of analysis: Acetylsalicylate and major metabolites extracted with ether; derivation with
BSTFA (see text); analysis with gas chromatograph using temperature programming and OV-17
column monitored by flame ionization detector
Comments: Used in research; reference method; sensitive to 0.1 mg/dL; linear to 200 mg/dL; no
interfering compounds
Method 7: Enzymatic; spectrophotometric
Principle of analysis: Salicylate hydroxylate converts salicylate to catechol with the oxidation of
NADPH to NADP; reaction rate, monitored by measurement of decrease in A340, is proportional to
salicylate concentration
Comments: Increasing in use since commercially available; highly specific
Method 8: High-performance liquid chromatography; chromatographic separation
Principle of analysis: Plasma or serum cleaned up by a protein precipitation; salicylate and/or
metabolites separated on reversed-phase column and quantitated by UV or fluorescence detection
Comments: Research; reference method some clinical labs use; sensitive to 0.1 mg/dL; linear to > 500
mg/dL; no interfering compounds; simpler sample preparation of GC
1048
Salicylates
Table 2: Pharmacokinetic Parameters of Aspirin and Salicylates

Compound: Aspirin
Dose:
Protein binding (%): 80 to 90
Half-life: 10 to 15 min
Comments: Protein binding < salicylate
Compound: Salicylate
Dose: 600 mg bolus
Half-life: 2 to 4 hr
Comments: Protein binding decreases as salicylate concentration increases; also altered in uremia; NB
hypoalbuminemia
Compound: Salicylate
Dose: 4 g/day
Half-life: 20 hr
*Whether given as aspirin or salicylate or molar equivalent dose of benorilate.
Figures
Figure 1: Absorbance characteristics of colorimetric salicylate procedure.

1, Ferric nitrate reagent versus water; 2, acetylsalicylate standard in ferric nitrate reagent
versus water; 3, acetylsalicylate standard in ferric nitrate reagent versus ferric nitrate reagent
blank.
1049
Salicylates
Figure 2: Example of standard curve for ferric nitrate assay for salicylic acid.
Absorbance at 540 nm versus concentration of salicylate (g/mL).
Figure 3: Examples of HPLC chromatograms for salicylate assay.

Chromatogram A is of blank plasma, and chromatogram B is of the same plasma spiked with
250 mg/L salicylate. Compound I is the internal standard, and compound II is salicylate.
1050
Salicylates
Figure 4: Typical standard curve for the assay of salicylate by HPLC.

This demonstrates linearity up to at least 600 mg/L. Regression analysis produced the
following equation: y = 0.040x + 0.0040, with a correlation coefficient of 0.999.
Procedure: Spectrophotometric Method

Assay
Principle Equipment: 10 nm bandpass spectrophotometer.
The following spectrophotometric method is based 1. Label six test tubes: standard blank,
on salicylates complexing with ferric ion, which is control, control blank unknown(s), and unknown(s)
supplied by ferric nitrate in dilute nitric acid. The blank.
resulting complex is water soluble, absorbs at 540 2. To the standard, standard blank, control,
nm (Figure 1), and is read against a serum blank and unknown tubes, add 3.6 mL of ferric nitrate
pipetted into dilute nitric acid. solution.
3. To unknown and control blank tubes, add
Specimen 3.6 mL of dilute nitric acid.
4. Measure 0.3 mL of standard into standard
tube. Mix.
1. Ferric nitrate solution (6.7 g/L, 24.8
5. Measure 0.3 mL of control into control
mmol/L). Dissolve 2.5 g of Fe(NO3)39H2O in
and control blank tubes. Mix.
1.0 mL of concentrated nitric acid, and dilute to 6. Measure 0.3 mL of unknown into
250 mL with distilled water. Filter through unknown and unknown blank tubes. Mix.
Whatman No. 1 filter paper into a glass bottle. 7. Read the standard, control, and unknown
Stable for 1 year at room temperature. against their respective blanks in the
2. Dilute nitric acid (62 mmol/L). Dilute spectrophotometer at a wavelength of 540 nm.
1.0 mL of concentrated nitric acid to 250 mL with
distilled water. This is stable for 2 years at room Calculation
temperature. [Absorbanceunknown/Absorbancestandard] x 10 mg/dL
3. Stock salicylate standard (1 g/L, 7.24 = salicylate (mg/dL)
mmol/L). Dissolve 100 mg of salicylic acid in 100
mL of water. This is stable for 1 year at 4C to 8C. Notes
4. Working salicylate standard (10 mg/dL, 1. Because of the presence of iron in the
0.724 mmol/L). In a 10-mL volumetric flask, unknown serum or plasma sample, a low
dilute 1.0 mL of stock standard to 10 mL with absorbance reading corresponding to
distilled water. This is stable for 3 months at 4C to approximately 2 mg/dL or less should not be
8C. considered significant.
5. Control. A serum sample containing 2. The colorimetric reaction is linear from 0
known amounts of salicylate. to 100 mg/dL (Salicylates: Figure 2).
1051
Salicylates
3. Analysis of urine. This procedure can be Procedure: To respective 1-mL centrifuge tubes,
used for a urine specimen after a heating step to add 200 L of serum or plasma. Add 200 L of
remove volatile interferences (especially ketones). internal standard solution, and vortex. Centrifuge at
Place a test tube containing 5 mL of centrifuged high speed for 5 sec in a Microfuge (Beckman
urine in a 100C heating block or boiling water for Instruments). Decant the supernatant into clean
15 min. After cooling, use as a specimen in the tubes, and inject 50-L volumes onto the HPLC
ferric nitrate procedure. Phenothiazine, column.
homogentisic acid, and thiocyanates will interfere
and cause false-positive results. Salicylate cannot Equipment: An HPLC system equipped with a
be confirmed in urine in the presence of these fluorometric detector and a radial compression
compounds. module for the column (Millipore-Waters).
Procedure: High-Performance Liquid Column: Millipore Waters Radial-PAK cartridge

Chromatography (8 NV C 18 4 (NOVA-PAK) part #86342). Flow
rate: 2.0 mL/min. Excitation wavelength: 300 nm.
Principle Emission wavelength: 370 nm.
In the HPLC procedure, the native fluorescence of
salicylate is used for quantitation after preliminary Calculation
sample cleanup by protein precipitation and The retention time for salicylate is approximately
separation on a reversed-phase column. 3.3 min for the internal standard and 5.0 min for
salicylate. Sample chromatograms are presented as
Specimen Salicylates: Figure 3.
1. Deionized water. 18 MOhm quality or Compute peak height ratio of salicylate versus
better. internal standard for the standard and unknowns:
2. Glacial acetic acid. Analytical reagent Peak height ratio = Peak height of compound
(AR) grade. Peak height internal standard
3. Acetonitrile. HPLC grade.
4. Mobile phase. Mix 700 mL deionized Calculate a standard curve by using the ratio for the
water, 50 mL glacial acetic acid, and 250 mL single standard and assuming a straight line passing
acetonitrile. Filter through a 0.22-mm filter prior to through the origin. The concentrations of unknown
use. May be recycled (with mixing) for up to 1 samples can then be determined by interpolation
week or 200 injections. from this line.
5. Stock salicylate standard (1.5 g/L, 10.9 Notes
mmol/L). Prepare a stock standard solution 1. Both between- and within-day coefficients
containing 1.500 g salicylic acid in 1 L by of variation for this method were less than
weighing 1.74 g of sodium salicylate and 1.2% when assessed using spiked plasma
transferring to a 1-L volumetric flask. Use at concentrations of 0.75, 15, and 30
deionized water to dissolve the sodium salicylate, mg/dL salicylate.
and dilute to the mark.
2. Limit of sensitivity is 0.1 mg/dL (0.007
6. Serum salicylate standard (30 mg/dL,
mmol/L), with linearity up to at least 60
2.2 mmol/L). Prepare a 30 mg/dL serum standard
mg/dL (4.32 mmol/L). Linearity is
by mixing 10 mL of stock standard with 50 mL of
demonstrated in Salicylate: Figure 4.
drug-free serum. This can be stored in 0.5-mL
aliquots at 18C for at least 12 months. 3. There are no known interfering
7. o-Methoxybenzoic acid internal substances.
standard (200 mg/L, 1.31 mmol/L). Dissolve 20 4. It is essential that internal controls at both
mg o-methoxybenzoic acid in 100 mL of 5% the high and low ends of the reporting
trichloroacetic acid. AR-grade trichloroacetic acid range be included in each batch of
from the Ajax Company produces the least samples assayed when single-point
chromatographic baseline noise. This solution is calibration is used as recommended above.
stable at 4C for at

Methods Kaplan

Uploaded by

Copyright:

Available Formats

Methods Kaplan

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Methods Kaplan

Uploaded by

Copyright:

Available Formats

1

Kaplan and Pesces : Clinical Chemistry:

Gus Koerbin, BAppSci, AFAIM

A work of 144 Methods of Analysis describing current methodology.

*Published by Mosby, an affiliate of Elsevier; 2009

Kaplan and Pesces : Clinical Chemistry:

Pesce Kaplan Publishers

Previous copyrighted printed editions: 1987, electronic editions 1996, 2003

Library of Congress Cataloging-in-Publication Data

Bound version and electronic version.

Printed in the United States of America

Contributors to Methods of Analysis

Marion Black, BSc(Hons), Dip. Ed, MAACB Joris R. Delanghe

John R. Burnett, MB ChB, MD, PhD, Goce Dimeski, BSc

Atholl Johnston, BSc, MSc, PhD,

Edmund Lamb, PhD, FRCPath Danni L. Meany, PhD

Denise A. McKeown, MSci, AMRSC

Peter OLeary, BSc, MAACB, AFACHSE, Terry Pry, PhD

Ravinder Jit Singh, PhD John G. Toffaletti, PhD

Jillian R. Tate, BSc(Hons), MSc Odette Youdell, BAppSci(Hons), MAACB

Preface for the 2009 edition of Methods in Clinical Chemistry

47. Copper The following analytes are in Volume

92. Luteinizing Hormone 115. Pyruvic Acid

137. Triglycerides 141. Urine Porphyrin Quantitation

and LC-MS/MS methods for 25-OH-D have been

Table 1: Methods for 25-OH-Vitamin D Assay

Table 2: Performance of 25-OH-Vitamin D Methods in Various Laboratories as

Sample Method n Method Mean SD CV

Automated IDS EIA 13 44.7 10.7 23.9

Chromatographic ligand-binding assay 1 48 0 0

DiaSorin liaison 37 40.7 6.4 15.8

DiaSorin liaison, total 8 41.2 5.7 13.9

DiaSorin RIA 50 44.1 6.4 14.5

HPLC 10 48.5 16.6 34.3

IDS EIA 56 49.3 9.4 19

IDS RIA 31 45.2 4.5 10

LC-MS 20 50.8 9.3 18.2

All Methods 228 45.4 7.5 16.6

n, Number of laboratories using a particular method.

Table 3: Vitamin D Metabolite Concentrations in Disorders of Calcium Homeostasis

Figure 3: Representative HPLC-UV chromatograms.

Figure 7: Comparison of the proposed HPLC method with LC-MS/MS. [19]

Procedure: 25-OH-Vitamin D-Commercial Assays.

Names: 1-Antitrypsin, 1-protease inhibitor

summary of an analysis of published genetic 1-Antitrypsin: Methods Summary Table

1-Antitrypsin: Methods Summary Table (cont)

Name: Acetaminophen, N-acetyl-p-aminophenol, paracetamol

Molecular formula: C8H9NO2

Quantitative Analysis determination of acetaminophen in serum (Table 1,

of methanol/1% acetic acid, containing pyridine and derivatization to the trimethylsilyl

Following an overdose, when it is likely that

Table 1: Acetaminophen Methods Summary

Chromatogram of HPLC analysis for acetaminophen:

Procedure: High-Performance Liquid Chromatography

limit of detection of 1 mg/L. In each sample, the peak height of acetaminophen is

[Acetaminophen] = [CAL] peak height unknown/peak height of internal standard

where [Acetaminophen] = drug concentration (mg/L),

Adrenocorticotropic Hormone (ACTH)

Adrenocorticotropic Hormone (ACTH)

Name: Adrenocorticotropic hormone (ACTH)