COr n e Al SECo n d Ed it io n

Topo aphy
A Gu e f Cl cal Appl ca
he Wavef Ea

Ed it o r : Mi Wa , MD, PhD
Co -Ed it o r : T ac y Sc h o d Swa tz , OD, MS, FAAO

SLACK Incorporated
 Editor:
Ming Wang, MD, PhD
Director, Wang Vision Cataract and LASIK Center
Clinical Associate Professor of O phthalmology, University of Tennessee
International President, Shanghai Aier Eye Hospital

Co-Editor:
Tracy Schroeder Swartz, OD, MS, FAAO
Center Director, VisionAmerica
Huntsville, Alabama
 www.slackbooks.com

ISBN: 978-1-55642-970-5

Copyright © 2012 by SLACK Incorporated

Cover images compliments of Ryan Vida, OD, Wang Vision Cataract and LASIK Center, Nashville, TN.

All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording or otherwise, without written permission from the publisher, except for
brief quotations embodied in critical articles and reviews.
The procedures and practices described in this publication should be implemented in a manner consistent with the profes-
sional standards set for the circumstances that apply in each specific situation. Every effort has been made to confirm the
accuracy of the information presented and to correctly relate generally accepted practices. The authors, editors, and publisher
cannot accept responsibility for errors or exclusions or for the outcome of the material presented herein. There is no expressed
or implied warranty of this book or information imparted by it. Care has been taken to ensure that drug selection and dosages
are in accordance with currently accepted/recommended practice. Off-label uses of drugs may be discussed. Due to continuing
research, changes in government policy and regulations, and various effects of drug reactions and interactions, it is recommended
that the reader carefully review all materials and literature provided for each drug, especially those that are new or not frequently
used. Some drugs or devices in this publication have clearance for use in a restricted research setting by the Food and Drug and
Administration or FDA. Each professional should determine the FDA status of any drug or device prior to use in their practice.
Any review or mention of specific companies or products is not intended as an endorsement by the author or publisher.
SLACK Incorporated uses a review process to evaluate submitted material. Prior to publication, educators or clinicians pro-
vide important feedback on the content that we publish. We welcome feedback on this work.

Published by: SLACK Incorporated

6900 Grove Road
Thorofare, NJ 08086 USA
Telephone: 856-848-1000
Fax: 856-853-5991
www.slackbooks.com

Contact SLACK Incorporated for more information about other books in this field or about the availability of our books from
distributors outside the United States.

Library of Congress Cataloging-in-Publication Data

Corneal topography : a guide for clinical application in the wavefront era / edited by Ming Wang ; co-editor Tracy S. Swartz.
-- 2nd ed.
p. ; cm.
Rev. ed. of: Corneal topography in the wavefront era / edited by Ming Wang ; coordinated by Tracy Swartz. c2006.
Includes bibliographical references and index.
ISBN 978-1-55642-970-5 (alk. paper)
1. Corneal topography. 2. Cornea--Pathophysiology. 3. Cornea--Surgery. I. Wang, Ming X., 1960- II. Swartz, Tracy
Schroeder. III. Wang, Ming X., 1960- Corneal topography in the wavefront era.
[DNLM: 1. Corneal Topography--instrumentation. 2. Cornea--anatomy & histology. 3. Cornea--physiology. 4. Corneal
Diseases. WW 220]
RE336.C685 2012
617.7’1907545--dc23
2011028147

For permission to reprint material in another publication, contact SLACK Incorporated. Authorization to photocopy items
for internal, personal, or academic use is granted by SLACK Incorporated provided that the appropriate fee is paid directly to
Copyright Clearance Center. Prior to photocopying items, please contact the Copyright Clearance Center at 222 Rosewood Drive,
Danvers, MA 01923 USA; phone: 978-750-8400; web site: www.copyright.com; email: [email protected]
 Ded icat ion
To: Professor Zhen-Ping Zhang

Professor Zhang, my dear friend and colleague, your untimely departure has left a void in the hearts of all of us who
care for you. We miss you, Lao Zhang.
 Con ten t s
Dedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
About the Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Contributing Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
Foreword by Renato Ambrósio Jr, MD, PhD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii

Section I Basic Topographic Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Chapter 1 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Tracy Schroeder Swartz, OD, MS, FAAO; Renzo Mattioli, PhD; Nancy K. Tripoli, MA;
Doug Horner, OD, PhD; and Ming Wang, MD, PhD
Chapter 2 Corneal Anatomy and Optics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Ashkan M. Abbey, MD and Sonia H. Yoo, MD
Chapter 3 Maps and Scales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Tracy Schroeder Swartz, OD, MS, FAAO; Ilan Cohen, MD; Ray-Ann Lin, MD; Megan Buliano, OD;
and Y. Ralph Chu, MD

Section II Topographic Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Chapter 4 Topographic Technologies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Tracy Schroeder Swartz, OD, MS, FAAO; Zuguo Liu, MD, PhD; Xiao Yang, MD; and Mei Zhang, MD
Chapter 5 Placido Disk-Based Topographic Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
ATLAS 9000 Corneal Topographer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Dianne Anderson, OD, FAAO
The Tomey TMS Corneal Topographer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Stephen D. Klyce, PhD and Bradford L. Tannen, MD
The Magellan Mapper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Michael J. Endl, MD and Claus M. Fichte, MD
The Keratron Corneal Topographers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Renzo Mattioli, PhD and Nancy K. Tripoli, MA
AstraMax Comprehensive Diagnostic Workstation With Polar Grid Topography . . . . . . . . . . . . . . . . . . 83
David D. Liu, PhD; Jia Qu, MD; Shihao Chen, MD, OD; Tracy Schroeder Swartz, OD, MS, FAAO;
and Ming Wang, MD, PhD
Topolyzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Mirko R. Jankov II, MD, PhD; Renato Ambrósio Jr, MD, PhD; Diogo L. Caldas, MD;
Ana Laura C. Canedo, MD; Leonardo N. Pimentel, MD; Bruno F. Valbon, MD; and Sissimos Lemonis
EyeSys 3000 and EyeSys Vista . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Sonya M. Dakin, COT and Joe S. Wakil, MD
Chapter 6 Tomographers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
The Bausch & Lomb Orbscan II/IIz Anterior Segment Analysis System . . . . . . . . . . . . . . . . . . . . . . . 103
Paul M. Karpecki, OD, FAAO
OCULUS Pentacam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Stephen S. Khachikian, MD and Michael W. Belin, MD, FACS
Precisio Surgical Tomographer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Aleksandar Stojanovic, MD and Xiangjun Chen, MD, MS
Visante Omni . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Amin Ashrafzadeh, MD
The Oculyzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Renato Ambrósio Jr, MD, PhD; Diogo L. Caldas, MD; Ana Laura C. Canedo, MD;
Leonardo N. Pimentel, MD; Bruno F. Valbon, MD; and Mirko R. Jankov II, MD, PhD
Optovue—Fourier Domain Anterior Segment Optical Coherence Tomography . . . . . . . . . . . . . . . . . . 143
Karolinne Maia Rocha, MD, PhD and J. Bradley Randleman, MD
viii Co n t e n t s

Artemis Epithelial Thickness Profile: A Surrogate for Stromal Surface Topography . . . . . . . . . . . . . . 148
Dan Z. Reinstein, MD, MA(Cantab), FRCSC, FRCOphth;
Timothy J. Archer, MA(Oxon), DipCompSci(Cantab); and Marine Gobbe, MST(Optom), PhD
Chapter 7 Topography and Wavefront Combined Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
The iTrace Combination Corneal Topography and Wavefront System . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Tracy Schroeder Swartz, OD, MS, FAAO and Joe S. Wakil, MD
The Nidek OPD Scan II: A Comprehensive Diagnostic and Planning Platform for Intraocular and
Refractive Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Phillip M. Buscemi, OD; Harkaran S. Bains, BSc; Murray McFadden, MD, FRCSC, DABO;
and Katherine E. Paton, MD, FRCSC, DABO
Topo-Aberrometry With Keratron Onda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Renzo Mattioli, PhD; Massimo Camellin, MD; and Nancy K. Tripoli, MA
Topcon KR-1W . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Naoyuki Maeda, MD

Section III Topographic Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

Chapter 8 Topography in Corneal Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Gaurav Prakash, MD and Amar Agarwal, MS, FRCS, FRCOphth
Chapter 9 Topography-Guided Contact Lens Fitting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Shawna Hill Vanderhoof, OD, FAAO; Randy Kojima, OD, FAAO, FOAA; Jason Jedlicka, OD, FAAO;
Matthew Lampa, OD, FAAO; and Tracy Schroeder Swartz, OD, MS, FAAO
Chapter 10 Pre-refractive Surgery Topographic Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
J. Bradley Randleman, MD
Chapter 11 Topography-Guided Custom Ablation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Tracy Schroeder Swartz, OD, MS, FAAO
LaserSight AstraPro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
David D. Liu, PhD; Zhengjun Fan, MD; Junhua Xiao, MD; and Bing Liu, MD
Topography-Guided Custom Ablation With iVis Suite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
Aleksandar Stojanovic, MD and Xiangjun Chen, MD, MS
Artemis VHF Digital Ultrasound Evaluation of Topography-Guided Repair . . . . . . . . . . . . . . . . . . . . 248
Dan Z. Reinstein, MD, MA(Cantab), FRCSC, FRCOphth;
Timothy J. Archer, MA(Oxon), DipCompSci(Cantab); and Marine Gobbe, MST(Optom), PhD
Scheimpflug-Based Topography-Guided Custom Ablation Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . 250
Renato Ambrósio Jr, MD, PhD; Frederico P. Guerra, MD; and Cristiana de Moraes Ramalho, MD
Chapter 12 Planning Keratorefractive Treatments Using Wavefront and Corneal Topography Data . . . . . . . . . . . . . . 253
Noel Alpins, FRANZCO, FRCOphth, FACS and George Stamatelatos, BScOptom
Chapter 13 Corneal Topography: What Will the Upcoming Decade Bring? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Lance J. Kugler, MD and Ming Wang, MD, PhD

Financial Disclosures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

 Ack n ow led gm en t s
I would like to express my sincere appreciation and gratitude toward the co-editor and project manager of this book,
Dr. Tracy S. Swartz, for her hard work and dedication. Without Tracy, this book would not have been possible.
I would like to also thank all the staff members of Wang Vision Cataract & LASIK Center: Dr. Helen Boerman,
Dora Sztipanovits, Dr. Kevin Jackson, Dr. Ryan Vida, Leona Walthorn, JJ Wang, Suzanne Gentry, Ashley Marlow, Terry
Hagans, and the entire team.
I have had the good fortune to have the chance to learn from some great teachers over my professional career, includ-
ing my PhD thesis advisor, Professor John Weiner; my MD (magna cum laude) thesis advisor and Nature paper co-author
Professor George Church at Harvard Medical School and Massachusetts Institute of Technology; my ophthalmology resi-
dency advisors Professors Larry Donoso and William Tasman at Wills Eye Hospital; my cornea and refractive surgery fel-
lowship mentors Professors Richard Forster, Scheffer Tseng, Eduardo Alfonso, Carol Karp, William Culbertson, and Lori
Ventura at Bascom Palmer Eye Institute; my Vanderbilt University colleagues, Professors Dennis O’Day, James Elliott, and
Donald Gass; and my colleagues at the University of Tennessee, Professors Barrett Haik, Natalie Kerr, and James Fleming.
I would like thank my national and international colleagues as well, including Professors Arun Gulani, Ilan Cohen,
David Chang, Ron Krueger, Aleksandar Stojanovic, Guiseppe D’lppolito, Francis Muier, Steve Klyce, Marguerite
McDonald, Dan Durrie, Steve Slade, George Waring, Terry Kim, Karl Stonecipher, Brian Boxer-Wachler, Terrence
O’Brien, Jay Pepose, Guy Guzarin, Noel Alpins, Tong Sun, Min-jun Zhang, Yong Tan, Li Li, Bang Chen, Bao-sung Liu,
Michael Zhou, Xiao-bing Wang, Ke-ming Yu, Zhi-yu Du, Qin-mei Wang, David Liu, Arthur Tan, Jin-feng Chai, Hui-yin
Liu, Ai-min Li, Ed Peterson, and Henry Zhou.
Often one learns as much from fellows that he trains, and I have been fortunate to have a great group of doctors who
have been my fellows over the years: Drs. Shin Kang, Ilan Cohen, Uyen Tran, Walid Haddad, Mouhab Aljaheh, Ke-ming
Yu, Yang-zi Jiang, Ray-Ann Lin, Lav Panchal, Lisa Marten, and Lance Kugler. I have learnt a great deal also from our
optometry residents over the years, Drs. Helen Boerman, David Coward, Shawna Hill, Tracy Winton, Dora Sztipanovits,
Kevin Jackson, Ryan Vida, Mark Raymond, and Michael George.
Finally, I want to thank my family for their unfailing support and love: my wife, Ye-jia “JJ” Wang; my father, Dr. Zhen-
sheng Wang; my mother, Dr. A-lian Xu; my brother, Dr. Ming-yu Wang; my godmother, June Rudolph; and my godfather,
Misha Bartnovsky.
 Ab ou t t h e Ed itor s
Ming Wang, MD, PhD is the Director of Wang Vision Cataract and LASIK Center in
Nashville, TN, Clinical Associate Professor of the University of Tennessee, Nashville, TN,
and International president of Shanghai Aier Eye Hospital, Shanghai, China.
Dr. Wang graduated from Harvard Medical School and Massachusetts Institute of
Technology (MD, magna cum laude) in Boston, MA, holds a doctorate degree in laser spec-
troscopy, and completed his residency at Wills Eye Hospital in Philadelphia, PA and his cor-
neal and refractive surgery fellowship at Bascom Palmer Eye Institute, Miami, FL. He is an
editorial board member of Cataract and Refractive Surgery Today and Refractive EyeCare.
A former panel consultant to the United States FDA Ophthalmic Device Panel and a
founding director of Vanderbilt Laser Sight Center, Nashville, TN, Dr. Wang published
a paper in the world-renowned journal Nature. This was followed by his editing 4 oph-
thalmic textbooks: Corneal Topography in the Wavefront Era: A Guide for Clinical
Application (SLACK Incorporated: Thorofare, NJ; 2006), Irregular Astigmatism: Diagnosis
and Treatment (SLACK Incorporated: Thorofare, NJ; 2008), Corneal Dystrophies and
Degenerations: A Molecular Genetics Approach (Oxford University Press: New York, NY;
2003), and Keratoconus and Keratoectasia: Prevention, Diagnosis and Treatment (SLACK Incorporated: Thorofare, NJ;
2010). Additionally, he has published over 120 papers and book chapters.
Dr. Wang has several US patents for his inventions of new biotechnologies to restore sight, including an amniotic mem-
brane contact lens, an adaptive infrared retinoscopic device for detecting ocular aberrations, and a digital eye bank for
virtual clinical trials. He is currently one of three investigators in the United States conducting an FDA-regulated clinical
trial to treat age-related loss of near vision (presbyopia). He introduced the femtosecond laser to China, and performed
China’s first LASIK procedure using this laser in 2005. He also performed the world’s first femtosecond laser-assisted
artificial cornea implantation (Alphacor), and the first Intacs procedure in the United States using a new version of Intacs
for keratoconus.
Dr. Wang was a recipient of the Academy of Ophthalmology Honor Award, as well as the Lifetime Achievement Award
of Association of Chinese American Physicians.
As the founding president of the Tennessee Chinese Chamber of Commerce, and co-owner and international president
of Shanghai Aier Eye Hospitals in Shanghai, China, which is the largest private eye hospital group in China today with
28 locations, Dr. Wang founded a 501c(3) nonprofit organization called the Wang Foundation for Christian Outreach to
China.
Dr. Wang specializes in refractive cataract surgery; keratorefractive surgery; corneal and external diseases, including
keratoconus; and anterior segment reconstructive surgeries with the amniotic membrane, stem cell, and keratoprosthesis.
He runs a busy international referral clinic for post-LASIK and postcataract surgery complications. He founded another
501c(3) nonprofit charity, the Wang Foundation for Sight Restoration, which has also helped patients from over 40 states
in the United States and 55 countries worldwide with all sight restoration surgeries performed free of charge.
Dr. Wang is a champion amateur ballroom dancer and a former finalist in the world ballroom dance championships in
the open pro-am international 10 dance. He plays the Chinese violin (er-hu) and accompanied country music legend Dolly
Parton on her CD Those Were the Days. Dr. Wang organized an annual classical ballroom dance sight charity event, the
EyeBall, which is now in its 6th year, and has drawn attendees from all over the United States and around the world.

Tracy Schroeder Swartz, OD, MS, FAAO currently serves as Center Director of
VisionAmerica in Hunstville, AL where she practices consultative optometry, specializing in
ocular surface disease and dry eye. Originally from Wisconsin, Dr. Swartz attended Indiana
University in Bloomington for her undergraduate education. During optometry school, she
served as an assistant instructor for the Biology department and became interested in pedi-
atrics and strabismus.
After completing her doctorate, she pursued a master’s degree in Physiological Optics,
specializing in pediatrics. She served as faculty at the IU School of Optometry for 4 years,
and earned the Indiana Chapter of the American Academy of Optometry Gordon Heath
Fellowship in 1996.
After completing her master’s, she relocated to Metro DC, where she specialized in refrac-
tive and corneal surgery, and earned her Fellowship in the American Academy of Optometry.
xii Ab o u t t h e Ed it o r s

She later joined Wang Vision Cataract and LASIK Center in Nashville, TN. Here she served as Director of Clinical
Operations, Residency Director for the Optometric Residency Program, and adjunct faculty to Indiana University School
of Optometry. While there, she edited 2 textbooks with Ming Wang, MD, PhD: Corneal Topography in the Wavefront
Era: A Guide for Clinical Application (SLACK Incorporated: Thorofare, NJ; 2006) and Irregular Astigmatism: Diagnosis
and Treatment (SLACK Incorporated: Thorofare, NJ; 2008), as well as authored numerous book chapters on refractive
surgery, topography, aberrometry, and anterior segment disease. She served as Co-Editor for the literature review column
for Cataract and Refractive Surgery Today from 2003 to 2008, and now serves on the editorial board for Optometry Times.
She left Nashville for Huntsville, AL where she became the Center Director for VisionAmerica. She edited 2 textbooks,
Keratoconus and Keratoectasia: Prevention, Diagnosis and Treatment (SLACK Incorporated: Thorofare, NJ; 2010)
followed by Cornea Handbook (SLACK Incorporated: Thorofare, NJ; 2010). She is a board member of the Optometric
Council for Refractive Technology, is a diplomat of the American Board of Optometery, serves as Adjunct Faculty for the
University of Waterloo School of Optometry in Waterloo, Ontario, Canada, and tries to be the best soccer mom she can.
 Con t r ib u t in g Au t h or s
Ashkan M. Abbey, MD (Chapter 2) Ilan Cohen, MD (Chapter 3)
Miami, Florida New York, New York

Amar Agarwal, MS, FRCS, FRCOphth (Chapter 8) Sonya M. Dakin, COT (Chapter 5)
Chennai, India Houston, Texas

Noel Alpins, FRANZCO, FRCOphth, FACS (Chapter 12) Michael J. Endl, MD (Chapter 5)
Cheltenham, Victoria, Australia Buffalo, New York
Amherst, New York
Renato Ambrósio Jr, MD, PhD (Chapters 5, 6, 11)
Tijuca, Rio de Janeiro Zhengjun Fan, MD (Chapter 11)
Beijing, China
Dianne Anderson, OD, FAAO (Chapter 5)
Naperville, Illinois Claus M. Fichte, MD (Chapter 5)
Amherst, New York
Timothy J. Archer, MA(Oxon), DipCompSci(Cantab)
(Chapters 6, 11) Marine Gobbe, MST(Optom), PhD (Chapters 6, 11)
London, United Kingdom London, United Kingdom

Amin Ashrafzadeh, MD (Chapter 6) Frederico P. Guerra, MD (Chapter 11)

Modesto, California Rio de Janeiro, Brazil

Harkaran S. Bains, BSc (Chapter 7) Doug Horner, OD, PhD (Chapter 1)

Fremont, California Bloomington, Indiana

Michael W. Belin, MD, FACS (Chapter 6) Mirko R. Jankov II, MD, PhD (Chapters 5, 6)
Nashville, Tennessee Belgrade, Serbia

Megan Buliano, OD (Chapter 3) Jason Jedlicka, OD, FAAO (Chapter 9)

Media, Pennsylvania Mankato, Minnesota

Phillip M. Buscemi, OD (Chapter 7) Paul M. Karpecki, OD, FAAO (Chapter 6)

Greensboro, North Carolina Lexington, Kentucky

Diogo L. Caldas, MD (Chapters 5, 6) Stephen S. Khachikian, MD (Chapter 6)

Tijuca, Rio de Janeiro Rapid City, South Dakota

Massimo Camellin, MD (Chapter 7) Stephen D. Klyce, PhD (Chapter 5)

Rovigo, Italy Port Washington, New York

Ana Laura C. Canedo, MD (Chapters 5, 6) Randy Kojima, OD, FAAO, FOAA (Chapter 9)
Tijuca, Rio de Janeiro Vancouver, British Columbia, Canada

Shihao Chen, MD, OD (Chapter 5) Lance J. Kugler, MD (Chapter 13)

Wenzhou, Zhejiang, China Omaha, Nebraska

Xiangjun Chen, MD, MS (Chapters 6, 11) Matthew Lampa, OD, FAAO (Chapter 9)
Oslo, Norway Silverton, Oregon

Y. Ralph Chu, MD (Chapter 3) Sissimos Lemonis (Chapter 5)

Bloomington, Minnesota Nürnberg Area, Germany
xiv Co n t r ib u t in g Au t h o r s

Ray-Ann Lin, MD (Chapter 3) Dan Z. Reinstein, MD, MA(Cantab), FRCSC, FRCOphth

Taipei, Taiwan, ROC (Chapters 6, 11)
New York, New York
Bing Liu, MD (Chapter 11) Paris, France
Beijing, China
Karolinne Maia Rocha, MD, PhD (Chapter 6)
David D. Liu, PhD (Chapters 5, 11) Cleveland, Ohio
Winter Park, Florida
George Stamatelatos, BScOptom (Chapter 12)
Zuguo Liu, MD, PhD (Chapter 4) Cheltenham, Victoria, Australia
Xiamen, China
Aleksandar Stojanovic, MD (Chapters 6, 11)
Naoyuki Maeda, MD (Chapter 7) Tromsoe, Norway
Yamadaoka, Suita, Japan
Bradford L. Tannen, MD (Chapter 5)
Renzo Mattioli, PhD (Chapters 1, 5, 7) New York, New York
Roma, Italy
Nancy K. Tripoli, MA (Chapters 1, 5, 7)
Murray McFadden, MD, FRCSC, DABO (Chapter 7) Durham, North Carolina
Langley, British Columbia, Canada
Bruno F. Valbon, MD (Chapters 5, 6)
Cristiana de Moraes Ramalho, MD (Chapter 11) Tijuca, Rio de Janeiro
Juiz de Fora, Minas Gerais, Brazil
Shawna Hill Vanderhoof, OD, FAAO (Chapter 9)
Katherine E. Paton, MD, FRCSC, DABO (Chapter 7) Denver, Colorado
Vancouver, British Columbia, Canada
Joe S. Wakil, MD (Chapters 5, 7)
Leonardo N. Pimentel, MD (Chapters 5, 6) Houston, Texas
Tijuca, Rio de Janeiro
Junhua Xiao, MD (Chapter 11)
Gaurav Prakash, MD (Chapter 8) Nanchang, China
Chennai, India
Xiao Yang, MD (Chapter 4)
Jia Qu, MD (Chapter 5) Guangzhou, China
Chashan Gaojiao Campus, Wenzhou, China
Sonia H. Yoo, MD (Chapter 2)
J. Bradley Randleman, MD (Chapters 6, 10) Miami, Florida
Atlanta, Georgia
Mei Zhang, MD (Chapter 4)
Guangzhou, China
 Preface
The first edition of the popular textbook, Corneal Topography in the Wavefront Era, became SLACK Incorporated’s
best seller at the American Society of Cataract and Refractive Surgery’s annual meeting in 2006 and has since become one
of their best-selling ophthalmic books. Since the publication of the first edition, a tremendous amount of advancement has
taken place in the field of corneal topography. Hence, we feel that a second edition of this textbook should be published at
this time in order to help our readers remain updated about the exciting developments in this field.
In 2011, there has been a resurgence of worldwide interest in the study of corneal topography for the following reasons:
• There has been a growing awareness that in order to improve visual quality of a primary keratorefractive treat-
ment, one does need to be concerned with how much we have altered the cornea.
• It is clear now that the best treatment strategy for a postkeratorefractive complex eye is a corneal topography-
driven treatment.
• The rising popularity of cataract surgery with the premium intraocular lens (IOL)—a new field in ophthalmology
called refractive cataract surgery—brings with it the need for a better understanding of the role played by the
cornea in an eye implanted with a presbyopic IOL.
• There has been an emergence of a myriad of new technologies to treat various ocular conditions and, therefore, it
is important to attain knowledge about the imaging of the cornea, Intacs and ultraviolet light cross-linking treat-
ments for keratoconus, and corneal intrastromal femtosecond laser technologies.
• New anterior imaging devices have been developed at an increasingly rapid pace, including new corneal tomog-
raphers such as Pentacam, optical coherent tomography, ultrasonic biomicroscopy, ocular response analyzer,
infrared devices, and ultrasonic devices.
The second edition of this corneal topography textbook has been produced with the goal of helping our colleagues to
become current in their knowledge of the tremendous advances in the corneal topography field and should serve as an
indispensable reference for all ophthalmologists and optometrists.
Below is my updated Form Fruste Keratoconus (FFKC) Criteria 2011:
2 D rule:
• >2 D difference in superior and inferior K readings outside the central 3 mm
• >2 D difference in the corresponding inferior corneal locations between 2 eyes
• Absolute value of K very high (over 50 D) in one eye
3-point touch:
• Coincidence of location of pathology of anterior and posterior elevation, pachymetry, and anterior curvature
• Displaced apex in all maps
Anterior & posterior elevation:
• Orbscan (best-fit sphere [BFS]):
¤ “Ominous purple” in the posterior surface
¤ Anterior greater than 15 to 20 µm
¤ Posterior greater than 20 to 25 µm (post-LASIK: 40 to 50 µm)
• Pentacam (relative to BFS in an 8-mm zone):
¤ Anterior elevation differences greater than +8 µm
¤ Posterior elevation differences greater than +16 µm
• Holladay Report (Pentacam, toric ellipsoid):
¤ Anterior or posterior elevation greater than 10 µm
Pachymetry:
• Bed 250 to 300 µm
• Normal: 535 µm, SD = 35 µm. No LASIK below 1 D (500 µm), no PRK below 2 D (465 µm)
• Keratoconus (KC): 430 µm, SD = 70 µm
• Thinnest area is more than 15 µm thinner than center
• The difference between thinnest areas between 2 eyes is greater than 15 to 20 µm
• Abrupt and more rapid “out-of-zone” pachymetry increase from thinnest point radially out
• Holladay Report (Pentacam): relative pachymetry that exceeds -5% (with respect to normal)
Irregular astigmatism orientation, amount, pattern:
• >3 D or more dioptic curvature change, in central 3-mm circle
• Asymmetric bowtie, bent or “squeezed,” bitemporal outward “hugging”
• Change of astigmatism orientation and amount
xvi Pr e fa c e

• Against-the-rule astigmatism, plus inferior steepening, the “C” pattern, suggesting peripheral marginal degenera-
tion
Topography-based FFKC detectors:
• Tomey: positive KC score with either the KC index or KC suspicion index
• EyeSys: I-S > 1.3
• Pentacam: red indicators
• Humphrey Atlas Pathfinder: red zone

Ming Wang, MD, PhD

Wang Vision Cataract & LASIK Center
Nashville, Tennessee
 Forew ord
It is with great honor that I have been given the privilege of writing the foreword for the Second Edition of Corneal
Topography: A Guide for Clinical Application in the Wavefront Era edited by Ming Wang, MD, PhD and Tracy Schroeder
Swartz, OD, MS, FAAO.
The introduction of wavefront aberrometry by the end of the last century determined a new era for refractive surgery.
An upheaval in understanding of optical quality beyond classic sphere and cylinder, and the advent of customized abla-
tions have made the popularity of wavefront sensing to rise significantly over the last decade. However, it is of critical
importance to understand the latest developments of corneal topography and its continuous role for corneal, refractive, and
anterior segment surgery.
Placido’s disc-based corneal topography was introduced in the mid 1980s and represented a true revolution for the
diagnosis and management of corneal disease, with critical applications for refractive surgery. In fact, the role of corneal
topography for screening, planning, and evaluating the results of refractive procedures was determinant for the acceptance,
efficacy, and safety of keratorefractive procedures. Interestingly, refractive surgery also determined the absolute need to
evolve the scientific knowledge on other areas related to corneal biology, such as biomechanics and wound healing. In addi-
tion, it is important to differentiate topography (from Greek words “to place” [topo] and “to write” [graphein] and tomogra-
phy (from the Greek words “to cut or section” [tomos] and “to write” [graphein]). Tomography enables the reconstruction
of the corneal architecture 3-dimensionaly, providing anterior and posterior elevation maps along with thickness maps.
Topography enables the characterization of the front corneal surface curvature. Corneal tomography has been accepted
to enhance the sensitivity and specificity for screening refractive candidates, as well as to provide objective parameters to
choose and plan the refractive procedure (increasing efficacy and safety) and to evaluate clinical results.
Corneal Topography: A Guide for Clinical Application in the Wavefront Era, Second Edition is a book that provides
clinicians with a current understanding of state-of-the-art corneal topographic instruments, including 3-dimensional
tomographic systems and combined topographers with whole eye aberrometers. The applications for planning topography-
guided custom ablations are also covered and illustrated. A last chapter on the future developments of corneal topography
is included, which also discusses corneal biomechanics importance and the integration of systems for calculating intra-
ocular lenses.
The editors are to be commended for the efforts to lead and edit the Second Edition of such an important book for
ophthalmology.

Renato Ambrósio Jr, MD, PhD

Instituto de Olhos Renato Ambrósio
Tijuca, Rio de Janeiro
 SECTIO N I

BASIC T O PO GRAPHIC
P RINCIPLES
Ch a p te r 1

Histo ry

Tracy Schroeder Swartz, OD, MS, FAAO ; Renzo Mattioli, PhD;

Nancy K. Tripoli, MA; Doug Horner, OD, PhD; and Ming Wang, MD, PhD

T
he interface of air with the tear layer on the human Two techniques, keratometry and keratoscopy, were used
cornea accounts for approximately two-thirds of during the past century before merging into the present-day
the eye’s refractive power. The typical cornea is computer-assisted video keratography or, as it is commonly
responsible for 43 D of the 60 D of the refractive power of called, corneal topography (CT). A comprehensive review
the eye. The average radius of curvature, 7.8 mm, generates can help us to understand the real meaning of keratometric
the majority of the refractive power of the cornea (about readings and topographic color maps.
+48.00 D). The posterior surface, with its concave shape
and stromal index similar to that of the aqueous, contrib-
utes about -5.00 D. In addition, the majority of astigmatism
originates from the corneal shape. It is not surprising that H ISTORY OF KERATOMETRY
great effort has been put forth to measure this surface of
the eye. In 1796, Ramsden created a telescope to magnify images
A normal cornea is not spherical but rather asphero- reflected off the cornea to measure corneal curvature.
torical, vaguely like a bell with flat sides that have been Magnification of the reflected images also magnifies the
slightly “squeezed” vertically. “With the rule” (wherein the normal instability of eye movements that becomes a serious
vertical axis is steeper and horizontal axis flatter) corneal issue in making accurate measurements. Ramsden is cred-
astigmatism from around 0.50 to 1.00 D is usually compen- ited with developing a doubling device that eliminates the
sated by the natural tilt of the crystalline lens. The central problems arising from normal instability of eye movements.
6- to 7-mm “apical cap” is ellipsoidal, with nearly constant Not until 1839 was another such instrument used, when
curvature only at its apex. The surrounding periphery is Kohlrausch used a telescope with adjustable mires. In 1881,
considerably flatter. Finally, in some corneas, a small steep Javal and Schiotz improved the instrument by using mires
junction to the limbus (like the bell’s edge) can be found. that were adjustable in size, and this model is still used
Corneal curvature measurement has been the subject clinically as the Haag-Streit ophthalmometer (Haag-Streit
of study since Father Christoph Scheiner’s works on the USA, Mason, OH). The modern configuration is shown in
human eye (1619).1 Ophthalmologists have tried to measure Figure 1-1. Since this and similar instruments measure the
corneal topographic characteristics for more than 150 years, curvature and astigmatism of the cornea, they were renamed
when A. Placido designed the keratoscopy target still in use “keratometers.” Bausch & Lomb (Rochester, NY) improved
today. on the keratometer in 1932 by adding a Scheiner’s disk

Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical

3 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 3 -10)
© 2012 SLACK Incorporated
4 Ch a p t e r 1

one image (h’), allowing calculation of r. Eye movements

are minimized because they affect both mires equally.
The radius-to-diopter conversion requires the following
equation:
(n’ – n)
F = ______
r
where F is the corneal surface power in D; n’ is the corneal
index of refraction; n is the refractive index surrounding
medium (for air, n = 1.0); and r is the corneal radius of
curvature in meters.
Javal’s index of 1.3375 has become the standard cor-
neal refractive index used. Note that the anterior tear film
surface index is closer to 1.376. Because the Javal ophthal-
mometer as well as Bausch & Lomb’s keratometer were
designed to measure the total corneal power, this reduced
index accounts for the negative power contribution of the
posterior corneal surface.
Figure 1-1. A modern keratometer. Keratometers acquire data from a central annular zone
of the cornea, which may vary in size depending on the
curvature and instrument. In the Reichert keratometer, the
to improve focusing mechanisms, circular mires, and the annulus is approximately 0.1-mm wide, and the diameter
ability to measure 2 meridians simultaneously, and this varies from 2.8 mm (in a 48.0-D cornea) to 3.5 mm (in a
model has essentially remained unchanged since that time 37.0-D cornea). It is important to remember the following
in the form of the Bausch & Lomb and Reichert (Depew, assumptions related to keratometry2 :
NY) models. Such instruments are designed to measure the • The formula used is based on spherical geometry. The
size of an image reflected off a convex surface using illumi- cornea, however, is not spherical but is a prolate (flat-
nated mires, a magnifying telescope, and a doubling prism. tened) ellipsoid. Thus, the central radius is slightly
Keratometry measurements are typically written in diop- steeper than actually measured.
ters (D). However, keratometers do not actually measure • Keratometry is based on 4 data points within the cen-
refractive power. They measure radius of curvature of the tral 3 mm of the cornea. It provides no insight into the
central 3 mm of the cornea. The formula for calculating area inside or outside of the 3-mm ring.
the corneal radius treats the cornea as a spherical reflecting • Keratometry theory assumes paraxial optics. While the
surface: approximation may be clinically acceptable for fitting
h’
__ -f
__
h
= x contacts or estimating corneal astigmatism, it may not
where h’ is the linear image size, h is the linear object size, be when measuring peripheral curvature.
f is the focal length, and x is the distance from mires to • Keratometers assume alignment of the corneal apex,
convex mirror focal plane. Note that the focal length, f, of line of sight, and instrument axis. However, this rarely
a spherical refracting surface is r/2, and the equation then occurs during actual measurement.
becomes: • The formula used to calculate the radius (r) approxi-
h’
__ = -r
__ mates the distance to the convex focal point, which,
h 2x in the case of the Reichert keratometer, may introduce
where r is the mirror radius of curvature. The distance from up to 0.12 D of error. This error may increase if the
the mires to the image approximates the distance from the instrument is not correctly focused or the operator
mires to the focal plane of the spherical convex mirror (x), accommodates during measurement.
resulting in the following formula: • Because the indices may differ between manufactur-
h’
__ -r
___ ers, one must be careful when comparing the readings
h
= -2d
in D between different instruments.
The distance, d, is fixed in most instruments. For exam-
• Reichert and similar keratometers are “one position”
ple, it is 75 mm for the Reichert keratometer. In Reichert and
instruments, able to measure 2 meridians 90 degrees
similar keratometers, the mire separation (h) is constant,
apart without moving the instrument position. While
and the size of the image (h’) is measured. Microscopic
this allows for quick measurement, it decreases the
eye movements make this measurement problematic, so a
clinician’s ability to measure irregular astigmatism.
doubling system is employed. A moveable prism is used to
The Haag-Streit ophthalmometer must be rotated
form a doubled image of the mires such that one image is
to measure each meridian, allowing the clinician to
displaced from the other and then manually aligned. The
better detect irregular astigmatism at the price of
distance required to align the images is equal to the size of
efficiency.
 H ist o r y 5

Figure 1-2. Topography timeline.

Modern corneal topographers calculate a simulated development of modern CT. These include Dekking,6 who
keratometry reading that incorporates the same paraxial in 1930 devised the first “cone,” and Bonnet,7 who in 1964
assumptions of traditional keratometry. These “simulated edited a book on CT, reporting elevation maps obtained
K” readings have been enriched by reporting curvature by stereo-photographic measurement of eyes sprayed with
along “semi-meridians” of the steepest and flattest merid- talcum, a technique more recently replaced by fluores-
ians at different zones (usually 3, 5, and 7 mm) to evaluate cein.8 Mandel published several works on corneal shape
irregular astigmatism in the peripheral cornea. models and contact lens fitting in the 1960s.9-11 In 1979,
Kuyama12 made 3-dimensional isometric computerized
maps, which were subsequently adopted by Sun/Nidek

KERATOSCOPY, C ORNEAL PKS-1000 (Fremont, CA).

A major step in the clinical utility of CT was the intro-
TOPOGRAPH Y, AND ITS duction of color-coded power maps. Several early maps
are shown in Figure 1-3. The first 2-dimensional, color-
C OLOR M APS coded map was published by Maguire and colleagues,13
and this is shown in Figure 1-3A. Following the intro-
duction of Maguire’s map, fully integrated commercial
In 1880, A. Placido devised his keratoscopy target, computer-assisted videokeratoscopes such as the Corneal
a flat disk with alternating black and white rings. Soon and Topographic Modeling Systems (CMS and TMS-1,
after in 1889, Javal attempted to give a quantitative mea- respectively; Computed Anatomy, New York, NY; Figure
sure of the corneal shape by placing Placido disks into 1-3B), the CT System (CTS; EyeSys Vision, Inc, Houston,
his ophthalmometer, behind the arc that carried the oph- TX; Figure 1-3C), and the VISIO (Visioptic, Inc, Houston,
thalmometric mires.3 Although the actual inventor of the TX; Figure 1-3D) introduced maps in which each point
first photokeratoscope is still a matter of discussion,4 it in the cornea was assigned a color according to its power
was Allvar Gullstrand who developed the algorithms for (Figure 1-4).
the first description of CT based on quantitative measure- As CT moved from a scientific curiosity to a clinical
ments. In a paper published in 1896, part of the work tool, disputes arose over what quantities were scientifically
that earned him a Nobel prize in 1911, Gullstrand5 illus- logical, most valuable for applications, most easily under-
trated how, following Javal’s suggestion, he developed a stood by clinicians, and ultimately most useful. A paral-
method of using a measuring microscope described as lel set of disputes concerned the terminology for types of
a “dividing engine” to determine the distance between instruments, measurements, maps, and algorithms. It is not
2 points on a keratoscopic photograph. Instantaneous radius surprising that eye professionals confused types of maps
of curvature was deduced from Gullstrand’s measurement with the algorithms that created them. False theoretical
through an arc-step algorithm, and the corneal meridian limits on some methodologies were assumed because of the
profiles that he was able to plot are surprisingly similar to limited computing power of earlier PCs. Much of the con-
those from modern computerized CT. fusion could have been avoided by an earlier understanding
It took about 90 years for photokeratoscopy to enter of the relationship between measuring corneal shape and
the computer age. A timeline is shown in Figure 1-2. measuring corneal optics and by recognizing the clinical
During those 90 years, several people contributed to the potential of keratoscopy.
6 Ch a p t e r 1

A B

C D

Figure 1-3. Early topographic maps.

Figure 1-5. Classical “bow-tie” image associated with astig-

matism.

the cone is exaggerated, and increasing the resolution of the

depiction would not increase the map detail.
The axial map has some utility as a simple descrip-
tor of corneal astigmatism, including cylinder, axis, and
irregularity. The example at the top-left in Figure 1-4 and in
Figure 1-5 shows the classical “bow-tie” or “hourglass.” The
vertical “yellow-orange” axis of symmetry of the hourglass
is the steeper meridian.
Despite its limitations, the axial map became known as
“the CT map” because of 2 factors. First, keratoscopes were
Figure 1-4. The colors of each point in an axial map represent thought of as merely Javal ophthalmometers or keratom-
the power associated with a sphere centered on the VK axis. eters newly capable of measuring a larger corneal area with
greater precision. In other words, they were used to assess
the need for simple optical correction of optics. They were
not intended to measure more sophisticated optical aberra-
Th e Ax ia l Map tions, much less the physical shape of the cornea. Second,
For many keratoscopes designed in the mid-1980s, axial the calculation of axial power from reflective data required
power was the primary quantity extracted from the raw only simple algorithms.16-18
data. As seen in Figure 1-4, the values and colors of each
point, P, in an axial map represent the power associated
to a sphere, Ra, having the same slant as the cornea and,
therefore, refracting a ray of light in the same way at each
THE “R EVOLUTION” AROUND
point, P. The assumption regarding the index used (1.3375)
is similar to that of keratometry. The axial map is more a
YEARS 1993 –1996: N EW
descriptor of corneal optics than shape.
The axial map is a traditional but poor descriptor of
ALGORITH MS AND M APS
corneal refraction because it does not take into account
spherical aberration,14,15 nor is the axial map a descriptor The calculus required for data analysis was the limit-
of shape. At each point, P, in Figure 1-4, the curvature and ing factor for several years following the introduction of
height of the centered sphere are not the same as the curva- computer-assisted CT. Because axial power was computed
ture and height of the cornea, and the discrepancy increases by algorithms16,18 that did not reconstruct the physical
toward the periphery. Therefore, the axial map distorts surface of the cornea, it was thought impossible to extract
shape. For example, in the map at the bottom-left, the wide accurate corneal height and instantaneous curvature from
red area indicates a keratoconic cone. However, the size of Placido disk images.19 There were authors who proposed
 H ist o r y 7

alternative algorithms. El Hage used a differential equation A

whose coefficients were fitted by polynomials.20 Doss and
colleagues proposed an algorithm that attempted to over-
come some of the spherical bias.21 Wang and colleagues
proposed the first arc-step method,22 van Saarloos and
Constable proposed an alternative arc-step,23 and Klein’s
more sophisticated algorithm14 used parabolas instead of
arcs. The 3 arc-step algorithms demonstrated that long-for-
gotten Gullstrand methods could not only approximate axial
power, but also measure corneal shape. Unfortunately, these
approaches could not be practically implemented without
further technical developments. The major challenge to
implementing an arc-step algorithm was to calculate “local”
quantities with high resolution while controlling the “noise”
that made maps inaccurate and nonreproducible and con-
fining the calculus complexity to stay within the limits of B
available computer power.22
In the early to the mid-1990s, the explosion of excimer
laser refractive surgery necessitated more accurate instru-
ments and more detailed representation of the corne-
al surface. The previous evaluation of keratoscopes on
spheres18,24-26 did not reflect their suitability for use in
surgical design. The first evaluations made on aspheric test
surfaces by Roberts,27 Cohen, Tripoli, and colleagues,28-31
and Carones and colleagues32 in 1993 through 1995 revealed
the inadequacy of spherically biased methods for measuring
height and curvature, especially in the periphery. It also
became evident that the use of an appropriate arc-step algo-
rithm on Placido disk images, rather than being impossible,
was quite accurate (less than or around 1 µm) for measuring
corneal height and curvature. Figure 1-6. (A) An instantaneous curvature map showing a
Once released from the over-simplicity of axial power, patient S/P myopic LASIK. (B) An axial map showing the same
instrument makers quickly developed new, useful quantities. in Figure 1-6A.
An instantaneous curvature map, calculated by an arc-step
algorithm, was introduced33-35 by the Keratron (Optikon
2000, Rome, Italy). An example is shown in Figure 1-6.
Curvature maps were soon adopted by most keratoscopes
under different names (eg, instantaneous, true, instantaneous
radius of curvature, local, tangential, or meridional).*
After the curvature map was introduced in keratoscopes,
it quickly revealed its role as an accurate descriptor of
corneal shape.33 Along a meridional section of the cornea
(Figure 1-7), as shown on the right, each zone has a center
of curvature that is not on the corneal axis, except at the
vertex. Without axial constraints, the curvature map can
track and display all transitions between flat and steep zones
and can reveal any local distortions, from the center to the
extreme periphery.**
The maps in Figure 1-7 illustrate the clinical value of Figure 1-7. Shape detail.
the curvature map’s depiction of detail. In both maps, the

* In their proposed standards, the ANSI committee has recently qualified the curvature map as “meridional curvature” to distinguish it from Gaussian and
other nondirectional measures. Unfortunately “meridional,” besides being an uncommon term, neglects a more important distinction between curvature
and axial power, both of which are measured along meridians. Therefore, while waiting for an eventual acceptance by CT manufacturers and by ISO, we
will refer only to “curvature” in this text.
** Although the precise boundaries and terminology of the “zones” are controversial among authors in the clinical literature, they are usually referred
to as a change in the corneal curvature.
8 Ch a p t e r 1

Figure 1-8. Progression of keratoconus shape in a number of

different patients.

Figure 1-9. The refractive map. Power is calculated according

green-blue transition (circle a, about 7 to 8 mm in size) to ray tracing.
reveals an abrupt change of curvature delimiting the “api-
cal zone”36 or “optical cap.” Bier’s primitive cornea model
described this zone in 195537 and defined the periphery
around it as “negative zone” in curvature. Actually, this
annular region may be very flat, but never goes to concav-
ity. Information about corneal shape that cannot be seen in
the axial map is revealed: (a) indicates the size of “corneal
cap,” (b) the size of keratoconus apex, (c) the surrounding
flat zone, and (d) a mild warpage induced by normal lid
pressure.
Curvature maps in Figure 1-8 show the degeneration of
the cap with increasing keratoconus.
A power or refractive map, where power is calculated
according to ray tracing (Figure 1-9), was proposed at the
same time by Roberts and was first introduced by EyeSys
in the Holladay diagnostic summary.38 In 1996, Barsky
and others proposed a Gaussian curvature map39,40 similar
to that shown in Figure 1-10. Its purpose was to overcome Figure 1-10. Gaussian map. (Reprinted with permission of Prof.
a limit of both axial and instantaneous curvature maps. B. Barsky.39,40 )
Because axial and instantaneous maps were based on a
center, they did not retain the same appearance on eccentric
corneas if the patient changed fixation. Unfortunately, the orange-red, and those below the sphere are green-blue. This
Gaussian map hides astigmatism, a clinically important universally accepted color convention in CT was inspired
feature. For this reason, Barsky proposed overlaid arrows to by geographic topography, in which warm colors represent
represent the amount and direction of astigmatism at each hills and mountains and blue shades are sea depth.
point (as seen in the bottom left map in Figure 1-9). Though
the Gaussian map is not common on commercial kerato-
scopes, it might yet find a useful application, for example,
in intraoperative situations when the patient cannot fixate.
C ORNEAL TOMOGRAPH Y
Corneal surface reconstruction algorithms also made
accurate measurement of corneal height feasible. Because a True elevation-based systems were developed to directly
representation of absolute height from a plane would have measure the shape of the cornea, and allowed evaluation
little meaning, the quantity is generally shown in “spheri- of the posterior surface for the first time. The first sys-
cal offset” maps. Construction is illustrated in Figure 1-11. tem, the Orbscan (Bausch & Lomb), used slit-scanning
The colors represent the height of the cornea with respect technology. This was followed by the use of Scheimpflug
to a reference sphere, measured in a direction parallel to imaging (Figure 1-12), and anterior segment ocular coher-
the axis. The points that are above the sphere appear in ence tomography (AS-OCT; Figure 1-13) Elevation-based
 H ist o r y 9

Figure 1-11. Construction of an elevation map.

Figure 1-13. Visante anterior segment optical coherence tomo-
grapher combined with the Atlas topographer enables mea-
surement of curvature, elevation, and pachymetry. (Reprinted
with permission of Carl Zeiss Meditec, Jena, Germany.)

it is the goal of this book to educate the reader to better

understand topography and its value in clinical decision
making. What follows is a review of corneal anatomy and
optical properties, a technical review of the technology,
map presentation and scaling, and in-depth information on
system-specific parameters, including using topography in
refractive surgery.

Figure 1-12. Scheimpflug image of a keratoconic cornea with

R EFERENCES
scarring.
1. Scheiner C. Oculus, Hoc Est. Fundamentum Opticum. Innsbruck,
Austria: Danielem Agricola (Oeniponti); 1619:12-17. http://gallica2.
bnf.fr/ark:/12148/bpt6k95041m.modeAffichageimage.langEN.f251.
systems capture 2-dimensional images, and create a vignettesnaviguer. Accessed July 11, 2011.
3-dimensional model; hence, the term tomographer rather 2. Horner DG, Salmon TO, Soni PS. Corneal topography. In: Benjamin
than topography is used to describe these systems. Some WJ, Borish, IM, eds. Borish’s Clinical Refraction. 2nd ed. St. Louis,
systems incorporate a Placido disc for curvature measure- MO: Butterworth Heinemann Elsevier; 2010:645-681.
ment, allowing direct measurement of both curvature and 3. Reynolds A. Introduction: history of corneal measurement. In:
Schanzlin D, Robin J, eds. Corneal Topography Measuring and
elevation. Pachymetry quickly became associated with Modifying the Cornea. New York, NY: Springer-Verlag; 1991:vii-x.
topography due to the capability of these systems to create 4. Levene JR. The true inventors of the keratoscope and photo-
maps of the corneal thickness. keratoscope. The British Journal for the History of Science, Vol. 2.
1965;8:324-342.
5. Gullstrand A. Photographic-ophthalmometric and clinical investi-
gations of corneal refraction. Am J Optom Arch Am Acad Optom.
C ONCLUSION 1966;43:143-214.
6. Dekking HM. Zur photographie der hornhautoberfläche. Albert Von
Graefe Archiv für Ophtalmologie. 1930;124(4):708-730.
CT has become an invaluable tool in ophthalmology, due 7. Bonnet R, Le Grand Y, Rapilly C. La Topographie Corneenne,
Paris, France: N. Desroches; 1964.
to advancements in technology and its application to refrac- 8. Warnicki JW, Rehkopf PG, Arrfa RC, Stuart JC. Corneal topogra-
tive surgery. Newest advances include indices for diagnosis phy using a projected grid. In: Schanzlin D, Robin J, eds. Corneal
of corneal diseases such as keratoconus, evaluation of Topography Measuring and Modifying the Cornea. New York, NY:
posterior cornea and other anterior segment structures, Springer-Verlag; 1991:25-32.
pachymetry measurement, and calculations for intraocular 9. Mandell RB. Keratometry and contact lens practice. The Optometric
Weekly. 1965;May 6:69-75.
lens (IOL) measurements. Interpretation of maps such as 10. Mandell RB. Corneal contour of the human infant. Arch Ophthalmol.
those reviewed briefly in this chapter is a learned skill, and 1967;77:345-348.
10 Ch a p t e r 1

11. Mandell RB. Mathematical model of the corneal contour. Brit J 26. Wilson SE, Verity SM, Conger DL. Accuracy and precision of
Physiol Opt. 1971;26(3):183-197. the corneal analysis system and the topographic modeling system.
12. Kuyama H, Sasamoto K, Maruyama S, Itoi M. A new photokeratom- Cornea. 1992;11:28-35.
eter for contact lens in clinic. J Jpn C L Soc. 1979;21(3):80-84. 27. Roberts C. Characterization of the inherent error in a spherically-
13. Maguire LJ, Singer DE, Klyce SD. Graphic presentation of biased corneal topography system in mapping a radially aspheric
computer-analyzed keratoscope photographs. Arch Ophthalmol. surface. J Refract Corneal Surg. 1994;10:103-116.
1987;105:223-230. 28. Cohen KL, Tripoli NK, Holmgren DE, Coggins JM. Assessment of
14. Klein SA. A corneal topography algorithm that produces continuous the height of radial aspheres reported by a computer-assisted kerato-
curvature. Optom Vis Sci. 1992;69:829-834. scope. Invest Ophthalmol Vis Sci (suppl). 1993;34:1217.
15. Roberts C. The accuracy of “power” maps to display curvature 29. Cohen KL, Tripoli NK, Holmgren DE, Coggins JM. Assessment
data in corneal topography systems. Invest Ophthalmol Vis Sci. of the power and height of radial aspheres reported by a computer-
1994;35:3524-3532. assisted keratoscope. Am J Ophthalmol. 1995;119:723-732.
16. Gersten M, Mammone RJ, Brunswick NJ, Larchmont NY. System 30. Tripoli NK, Cohen KL, Holmgren DE, Coggins JM. Assessment of
for topographical modeling of anatomical surfaces. US patent radial aspheres by the Keratron keratoscope using an arc-step algo-
4,863,260. September 5, 1989. rithm. Am J Ophthalmol. 1995;120:658-664.
17. Olsen T. On the calculation of power from curvature of the cornea. 31. Tripoli NK, Cohen KL, Obla P, Coggins JM, Holmgren DE. Height
Br J Ophthalmol. 1986;70:152-154. measurement of astigmatic test surfaces by a keratoscope that uses
18. Koch DD, Foulks GN, Moran CT, Wakil JS. The corneal EyeSys plane geometry reconstruction. Am J Ophthalmol. 1996;121;668-
system: accuracy analysis and reproducibility of first-generation 676.
prototype. J Refract Corneal Surg. 1989;6:423-429. 32. Carones F, Gobbi PG, Brancato R, Venturi E. Comparison between
19. Wilson SE, Wang JY, Klyce SD. Quantification and mathematical two computer-assisted keratoscopes in measuring aspheric surfaces.
analysis of photokeratoscopic images. In: Shanzlin DJ, Robin JB, Invest Ophtalmol Vis Sci (suppl). 1994:3748.
eds. Corneal Topography: Measuring and Modifying the Cornea. 33. Brancato R, Carones F. Topografia corneale computerizzata. Milan,
New York, NY: Springer-Verlag; 1991:1-9. Italy: Fogliazza; 1994.
20. El Hage SG. The computerized corneal topographer EH-270. In: 34. Mattioli R, Carones F, Cantera E. New algorithms to improve the
Shanzlin DJ, Robin JB, eds. Corneal Topography: Measuring and reconstruction of corneal geometry on the Keratron videokeratogra-
Modifying the Cornea. New York, NY: Springer-Verlag; 1991:11- pher. Invest Ophthalmol Vis Sci (suppl). 1995;36:1400.
24. 35. Mattioli R, Tripoli N. Corneal geometry reconstruction with the
21. Doss JD, Hutson RL, Rowsey JJ, Brown R. Method for calcula- Keratron Videokeratographer. Optom Vis Sci. 1997;74:881-894.
tion of corneal profile and power distribution. Arch Ophthalmol. 36. Kraff CR, Robin JB. Normal corneal topography. In: Shanzlin DJ,
1981;99:1261-1265. Robin JB, eds. Corneal Topography: Measuring and Modifying the
22. Wang J, Rice DA, Klyce SD. A new reconstruction algorithm for Cornea. New York, NY: Springer-Verlag; 1991:33-38.
improvement of corneal topographical analysis. J Corneal Refract 37. Bier N. A study of the cornea in relation to contact lens practice. Am
Surg. 1989;5:379-387. J Optom. 1956;33(6):291-304.
23. van Saarloos PP, Constable IJ. Improved method for calculation of 38. Holladay JT. Corneal topography using the Holladay diagnostic
corneal topography for any photokeratoscopic geometry. Optom Vis summary. J Cataract Refract Surg. 1997;23:209-221.
Sci. 1991;68:957-965. 39. Barsky BA, Klein SA, Garcia DD. Gaussian power, mean sphere,
24. Hannush SB, Crawford SL, Waring GO III, Gemmill MC, Lynn MJ, and cylinder representations for corneal maps with applica-
Nizam A. Accuracy and precision of keratometry, photokeratoscopy tions to the diagnosis of keratoconus. Invest Ophthalmol Vis Sci.
and corneal modeling on calibrated steel balls. Arch Ophthalmol. 1996;37(suppl):558.
1989;107:1235-1239. 40. Barsky B, Klein S, Garcia D. Gaussian power with cylinder vector
25. Maguire LJ, Wilson SE, Camp JJ, Verity SM. Evaluating the field topography maps. Optom Vis Sci. 1997;74:917-925.
reproducibility of topography systems on spherical surfaces. Arch
Ophthalmol. 1993;111:259-262.
Ch a p te r 2

Co rneal
Anato my and Optics

Ashkan M. Abbey, MD and Sonia H. Yoo, MD

T
he cornea is the transparent, avascular anterior por- significant refractive power so that light may be appropri-
tion of the eye that covers the iris, pupil, and ante- ately focused upon the retina. In fact, the anterior surface of
rior chamber. It comprises the external layer of the the cornea and its associated tear film account for 44 D of
eye along with the sclera, with which it is continuous. The the total 64 D of refractive power in the eye.4 With respect
transition area between the cornea and sclera is the limbus, to protection, the cornea provides both a barrier to fluid
a highly vascularized area of pluripotent stem cells. Fluid loss through tight junctions in apical epithelial cells and a
covers both its anterior (tear film) and posterior (aqueous) barrier to pathogens through membrane-spanning mucins
surfaces. The average horizontal diameter of the oval- at the apical surface. The cornea’s rigidity and ability to
shaped cornea is 11.7 mm, while the vertical diameter is quickly regenerate from mitotically active basal epithelial
10.6 mm, providing an overall area equivalent to one-sixth cells make it well-suited to act as a natural physical shield
of the circumference of the eyeball.1 The cornea is thinnest against ocular trauma.
at its center and undergoes progressive thickening toward
the periphery. On average, corneal thickness ranges from
0.5 mm centrally to 1 mm peripherally.2 The overall corneal
thickness tends to increase with age. ANATOMY AND P H YSIOLOGY
With respect to shape, the cornea is aspheric. The central
one-third (optic zone) of the cornea is almost spherical, and The normal cornea consists of 5 layers, listed from ante-
asymmetric flattening occurs with extension to the periph- rior to posterior: the epithelium, Bowman’s layer, stroma,
ery. The nasal superior portion of the cornea displays more Descemet’s membrane, and endothelium (Figure 2-1).
extensive flattening than the temporal inferior portion.
The cornea serves 2 primary functions in the eye: vision
and protection. With respect to vision, the cornea acts as a Ep it h eliu m
transparent tissue to allow light to be transmitted to the lens
and the retina. Its transparency is dependent on several fac- The corneal epithelium is composed of 5 to 7 layers
tors, including regularity and smoothness of the epithelium, of stratified, squamous epithelial cells and has an overall
its avascularity, and the size and arrangement of extracellu- thickness of 50 to 52 µm.5,6 It is a smooth, regular surface
lar and cellular components in the stroma, which are depen- to which the tear film adheres, and it plays an integral role
dent on the state of hydration, metabolism, and nutrition of as a barrier to pathogens and fluid and solute loss.
the stromal elements.3 Furthermore, the cornea provides

Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical

11 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 11-20)
© 2012 SLACK Incorporated
12 Ch a p t e r 2

Figure 2-2. Color-contrast polarized light microscopy at

the boundary of a keratoconus epikeratoplasty. The normal
(donor) human cornea is seen at the top, demonstrating the
lamellar structure. The keratoconic host cornea is at the bot-
tom (magnification 100×). Notice the disorganization of the
keratoconic lamellae.

regenerate after injury. It is attached anteriorly to the base-

ment membrane and blends into the more organized anterior
stroma. Its function remains unclear, but many have hypoth-
Figure 2-1. Normal histology of the cornea: epithe- esized that it plays a role in the maintenance of epithelial
lium, Bowman’s layer, stroma, Descemet’s mem- uniformity and structure and thereby helps to uphold the
brane, and endothelium. (Reprinted with permission
refractive power of the cornea.
from Wang M, ed. Keratoconus and Keratoectasia:
Prevention, Diagnosis, and Treatm ent. Thorofare,
NJ: SLACK Incorporated; 2010.)
St rom a
The corneal epithelium can be divided into 3 distinct The corneal stroma lies between the Bowman’s layer
sections based on cell type. Flattened superficial epithelial and Descemet’s membrane and composes approximately
cells (“squames”) compose the outer 3 to 4 layers of the 90% of the corneal thickness. It is primarily composed
epithelium. Below the superficial epithelial cells are 1 to of flattened bundles of Type I collagen fibrils arranged
3 layers of wing cells, aptly named for their thin wing-like in parallel, known as lamellae. These collagen fibrils are
extensions that project laterally from the cell body. The final embedded within a matrix of glycoproteins and proteogly-
layer is composed of basal cells, which are the mitotically cans that maintain adhesion between cells of the stroma and
active cells of the epithelium. These cells replicate and dif- allow for transport of nutrients and oxygen. Keratoconus
ferentiate to form all layers of the corneal epithelium. The produces compaction and a loss of the typical arrange-
basal cell layer is anchored to the basement membrane and ment of collagen fibrils in the anterior stroma (Figure 2-2).
stroma through an anchoring complex. In this complex, Proteoglycans also play a significant role in the regulation
a hemidesmosome located on the cytoplasmic side of the of spacing between collagen fibrils.
basal cell membrane adheres to the basement membrane The diameter of each collagen fibril ranges from 27 to
and is attached to the stroma through anchoring fibrils com- 35 nm, and the distance between each fibril is between
posed of Type VII collagen. 41.4 and 60 nm.8,9 The diameter of the collagen fibril is
The corneal basement membrane is an extracellular considered too small to result in significant scattering of
matrix that lies beneath the basal cell layer. Basal cells are light.10 It has been proposed that stromal transparency will
responsible for secreting its components, which include be maintained if the following 2 conditions are met: 1) col-
Type IV collagen, laminin, and other glycoproteins. The lagen fibrils maintain a fairly uniform small size and 2) col-
basement membrane separates the epithelium from the lagen fibrils are not spaced apart greater than approximately
stroma and provides a scaffold for the organization of the λ/2n, in which λ is the wavelength and n is the refractive
epithelium.7 index of the medium. Unlike the orientation of collagen
fibrils in other parts of the body, both the diameter of and
the distance between each collagen fibril in the corneal
Bow m a n ’s Layer lamellae are highly uniform. This regular arrangement is
thought to play a significant role in the establishment of the
Bowman’s layer is an acellular condensation of randomly shape, strength, and transparency of the cornea.
oriented collagen fibrils and proteoglycans that does not
 Co r n e a l An a t o my a n d O p t ic s 13

The primary cell type found within the corneal stroma is The average thickness of the tear film is approximately
the keratocyte, which is a type of fibroblast or specialized 7 µm, the majority of which is the aqueous layer.17 The tear
cell of connective tissue. The keratocyte is a flattened cell film serves the following functions: lubricates the anterior
that is interposed between lamellae, and its elongated pro- surface of the eye; acts as a smooth, transparent refractive
cesses often attach through gap junctions to those of other surface for the transmission of light; transfers nutrients and
keratocytes. Keratocytes secrete and maintain the collagen oxygen to the corneal epithelium; protects against infection;
and proteoglycans that form the majority of the corneal and removes desquamated epithelial cells from the surface
stroma. They also produce enzymes that are able to degrade of the eye.18 The lipid layer is produced by meibomian
older collagen fibrils in need of replacement. Furthermore, glands as a means by which to increase surface tension
keratocytes migrate to areas of stromal injury in order to and therefore prevent evaporation of the aqueous layer.
assist with scar formation and collagen repair.11 The aqueous layer is secreted by the main lacrimal glands
In normal corneal stroma, collagen fibrils are arranged and the accessory lacrimal glands of Wolfring and Krause
along the vertical and horizontal meridians, orthogonal and primarily serves to lubricate the corneal epithelium.
to one another.12-15 Furthermore, stromal collagen mass The mucin layer is manufactured by the goblet cells of the
progressively increases from the center of the cornea to conjunctiva and serves to reduce the surface tension of the
the periphery. Both of these characteristics of the corneal aqueous tear layer so that it may adsorb to the epithelium
stroma are integral to the maintenance of appropriate cor- and remain intact between blinks.
neal shape.

Cor n ea l In n er vat ion

Descem et ’s Mem b ra n e
The cornea has an extensive supply of sensory nerve
Descemet’s membrane is the basement membrane of the fibers, making it one of the most highly innervated tissues
corneal endothelium. It is synthesized by endothelial cells in the body. Corneal sensory nerves are derived from the
and is assembled at the basal surface of the cell layer. This ophthalmic division of the trigeminal nerve (cranial nerve
membrane is primarily composed of Type IV and VIII col- V). Nerve branches enter the anterior corneal stroma from
lagen, and it does not regenerate after injury. Its primary an annular nerve plexus near the limbus. At this entry point,
functions include filtering of solutes passing to and from myelination of the nerve axons is lost, which is essential for
epithelial cell layers and serving as a substrate for the the maintenance of corneal transparency.19 The nerve fibers
induction of polarity and differentiation of the overlying then proceed to run parallel to the epithelium, forming a
epithelium. sub-basal nerve plexus. From this plexus, axons without
Schwann cells penetrate Bowman’s layer and extend into
the epithelium. The corneal nerves serve 2 main functions.
En d ot h eliu m First, they provide protection by serving as the afferent por-
tion of an aversion reflex in response to any contact with
The corneal endothelium is a single layer of hexagonal the corneal surface. Second, corneal nerves secrete trophic
cells adherent to the posterior aspect of Descemet’s mem- factors for the preservation of corneal health. If corneal
brane. Their abundant organelles demonstrate the high nerves are damaged either by disease or trauma, the lack
level of metabolic and synthesizing activity of these cells. of adequate production of trophic factors could lead to
Endothelial cells in an adult are arrested in the G1 phase neurotrophic keratitis, which is characterized by epithelial
of the mitotic cell cycle, meaning that they do not replicate sloughing and impaired healing.
despite their capacity to do so.16 A major function of the
endothelium is the maintenance of corneal transparency
through the regulation of stromal hydration. The endo-
thelial cells provide a “leaky barrier” to aqueous humor C ORNEAL O PTICS
and contain numerous specialized ion transport systems to
transfer excess water out of the cornea. These 2 functions The cornea is the first structure that interacts with light
of the endothelium facilitate its regulation of the amount of as it is being transferred into the eye and contributes the
water within the corneal stroma. greatest amount of refractive power. The cornea can be
treated like an ophthalmic lens created from 3 different
interfaces: tears, corneal tissue, and the aqueous humor. To
Tea r Film calculate the power of the cornea, a thin lens formula can be
used 3 times, one for each interface. The thin lens formula
The tear film of the eye covers the anterior surface of the
is as follows:
cornea. It is composed of 3 separate layers: an anterior lipid (n-n1)
F= r_____
layer, a middle aqueous layer, and a posterior mucin layer.
14 Ch a p t e r 2

where F is the total refractive power at the surface of the

cornea; n is the refractive index of the air; and n1 is the
refractive index of the tear film. r is the radius of cur-
vature of the cornea in meters. For the average human
cornea, the central anterior corneal radius of curvature is
7.8 mm, and the posterior corneal radius of curvature is
6.5 mm. The indices of refraction are as follows: air, 1.00;
tears, 1.336; cornea, 1.376; aqueous humor, 1.336.
When using the above thin lens equation, the tear-air
interface power is 43.00. This first interface accounts for
the majority of corneal power because the change in density
from air to tears has the greatest differential. Using the thin
lens equation for the tear-cornea interface and the cornea-
aqueous humor interface, the resulting refractive powers
Figure 2-3. Effect of spherical aberration on a
are +5.1 D and -6.2 D, respectively. The total cornea power,
Snellen E.
calculated by combining these dioptric powers, is approxi-
mately +42.00 D for the average human cornea.20

Ph ysica l Con sid erat ion s Th at

Red u ce Visu a l Ab er rat ion s
The majority of a patient’s visual complaints arise from
lower-order aberrations, commonly known as refractive
errors. Defocus is considered absent when the punctum
remotum of the eye is at infinity in a situation where the
eye is not accommodating and a congruent image is per-
fectly focused on the retina.21 A hyperopic patient can use
accommodation to focus the image on the retina, while a
myopic patient has to move the object closer than infinity
Figure 2-4. Effect of coma on the Snellen E.
to focus the image.
Even when lower-order refractive errors are corrected
with spectacles or contact lenses, uncorrected higher-order
aberrations may decrease visual quality.22 Aberrations obliquely inducing astigmatism. This phenomenon explains
relevant to the cornea are spherical aberration, coma, and how a spherical cornea may register astigmatism when
oblique astigmatism. measured obliquely.
Spherical aberration is an expression of the difference in Corneal structure is such that aberrations are minimized.
refracting power for peripheral rays relative to central rays. This is accomplished by the prolate shape and small angle
This occurs when a point source of light is refracted by a Kappa. The cornea is not spherical, but rather aspheric. The
large-aperture optical system. Different zones of the aper- curve flattens toward the periphery as you move away from
ture have different focal lengths and, therefore, do not focus the center, and the shape is described as prolate. Natural
the point source of light to the same location. Normally, asphericity decreases the power of the peripheral parts of
spherical aberration is not important in the human eye the cornea, resulting in decreased spherical aberration and
because the pupil is sufficiently small. Figure 2-3 shows the coma.23 The asphericity of the human cornea is described
effect of spherical aberration on the Snellen E. by factor Q, which represents the difference in curvature
Coma occurs when an object is located off axis or if the from the center to the periphery of the cornea. The average
apex of the cornea is not properly aligned with the other value for a human’s cornea is -0.26. A perfect value for Q
optical elements of the eye. Similar to spherical aberration, would be -0.50, but the junction of the cornea and sclera at
it has little effect due to the pupil being relatively small. the limbus prohibits this in nature.
Coma can be thought of as a type of off-axis spherical Aberrations are further minimized by the location of the
aberration. It creates a series of images superimposed on corneal apex. If the apex of the cornea aligns with the pupil
each other with the brightest image being the smallest one. center, it would naturally intercept the visual axis. When
Figure 2-4 shows the effect of coma on the Snellen E. this does not occur, light rays get redirected and meet with
Oblique astigmatism is independent of pupil size and the retina at oblique angles. These redirected rays result in
cornea. It is induced when rays pass through the cornea, a lessened retinal response, producing what is known as
the “Stiles-Crawford Effect.” The pupil normally reduces
 Co r n e a l An a t o my a n d O p t ic s 15

Figure 2-5. Radial keratotomy

creates a central flattening, and
optic zones are often smaller than
the pupil size, resulting in halos
and night vision problems.

aberrant rays, which decreases the effect of oblique rays. The size of angle Kappa in reference to the ablation zone
Essentially, the same problem occurs if the pupil is not becomes important when this angle is large (Figure 2-8).
centered.20 If the visual axis is far from the geometric center of the
Pupil size is uniquely suited to limited aberrations, such cornea, the ablation will be effectively decentered. This is
as spherical aberration and diffraction. The pupil size that most often problematic with a hyperopic treatment where
is best for limiting higher-order aberrations and minimizing the central cornea is steeped around the visual axis.25 With
diffraction is 2.5 mm. The average pupil size is 5 mm, with a refractive surgery, pupil size related to the treatment zone
younger patients naturally having larger pupils.24 remains important. Patients with pupil sizes greater than
8 mm should be identified and appropriately educated about
visual side effects before undergoing keratorefractive surgi-
Cor n ea l Op t ics After cal correction.
Physiologically, the response of the cornea is com-
Keratorefract ive Su rger y plex following ablative procedures. The amount of tissue
removed can be predicted by Munnerlyn’s formula:
Keratorefractive surgery changes the natural shape of
S2D
____
the cornea, typically decreasing the natural safeguards T= 3
against aberrations. Keratorefractive surgeries are suc-
where S is the diameter of the treatment optical zone, D is
cessful because the optical properties of the eye can be
the refractive correction, and T is the maximum amount
manipulated by changing the shape of the cornea. With
of corneal tissue removed in microns. Clinically, for every
radial keratotomy, the central cornea is flattened due to
diopter of correction, approximately 12 µm of tissue are
relaxing incisions (Figure 2-5). When using excimer proce-
removed when using a 6-mm ablation.
dures, the prolate structure of the cornea is changed by the
Ablative, incisional, and INTACS all change keratomet-
removal of a convex positive meniscus for myopic ablations
ric readings. A 0.8-D change in K is associated with a 1-D
(Figure 2-6), a concave positive meniscus for hyperopic
change in refraction.23 This is important for surgical plan-
corrections (Figure 2-7), and a toric positive meniscus in
ning because creating a cornea that is too flat (less than
astigmatic corrections.
35 D) or too steep (more than 52 D) can result in a disabling
loss of visual quality.22
16 Ch a p t e r 2

A B

Figure 2-6. (A) Myopic LASIK

results in a flattened central C
zone as illustrated in this eleva-
tion map. Note the small cen-
tral zone. (B) Myopic LASIK
results in thinning in the center.
(C) Myopic LASIK creates posi-
tive spherical aberration, mani-
festing as halos. (A, B, and C
are reprinted with permission
from Trattler WB, Majmudar
PA, Luchs JI, Swartz T. Cornea
Handb ook. Thorofare, NJ:
SLACK Incorporated; 2010.)

Figure 2-7. (A) Hyperopic LASIK

results in a steepened central
zone. (Reprinted with permis-
sion from Trattler WB, Majmudar
PA, Luchs JI, Swartz T. Cornea A
Handbook. Thorofare, NJ: SLACK
Incorporated; 2010.) (continued)
 Co r n e a l An a t o my a n d O p t ic s 17

Figure 2-7 (continued). (B) Hyperopic LASIK creates negative spherical aberration. (Reprinted with per-
mission from Trattler WB, Majmudar PA, Luchs JI, Swartz T. Cornea Handbook. Thorofare, NJ: SLACK
Incorporated; 2010.)

REΔsp = MR Preop – MR Postop

A sample calculation is shown below:

OD OS
Ks Preop 45.00 (Average) 45.50 (Average)
MR Preop -6.00 -4.50
MR Postop -0.25 -0.50
Ks Postop 40.75 41.25

OD OS
REΔsp = MRpostop REΔsp = -0.25 – REΔsp = (-0.50) –
Figure 2-8. A large angle Kappa may complicate keratorefrac-
+ MRpreop (-6.00) = +5.75 D (-4.50) = +4.00
tive procedures.
Ksp = Kpreop 45.00 – 5.75 = 45.50 – 4.00 =
– REΔsp 39.25 D 41.50 D
Note the ratio of refractive change to keratometric
change is not 1:1. The mismatch is thought to result from To determine the change in corneal power based on the
the change in posterior corneal curvature following kera- refraction in the corneal plane, the change in refraction
torefractive surgery. In myopic ablative procedures, the must be converted using the following equation:
posterior corneal surface is thought to become more nega- REΔ sp
____________
REΔcp = 1-(0.014*REΔ )
tive at the same time the anterior corneal surface becomes sp
less positive. The natural power ratio between the 2 surfaces
This value can then be subtracted from the preoperative
is altered, and the assumptions used in keratometry create
keratometry value to determine the keratometry value for
significant errors in power measurements.
the corneal plane (Kcp).
The contact lens method uses the known values of power
and base curve combined with over-refraction to determine
In t rao cu la r Len s Ca lcu lat ion s the power of the cornea.26
After Refract ive Su rger y
OD OS
A considerable amount of research has been done to ORsp = MR with ORsp = -3.00 + ORsp = -1.50 +
improve the ability to calculate intraocular lens (IOL) CL+CL power (-3.00) = -6.00 (-1.50) = -3.00
powers after a refractive procedure. Several methods are
available to cataract surgeons to aid them in IOL power BC of CL = 337.5/7.4 = 45.61 337.5/7.5 = 45.00
calculation. The clinical history method, sometimes referred 337.5/BC (mm)
to as the refraction-derived method, involves subtracting ORcp = ORsp / -6.00/1+(0.014* -3.00/1+(0.014*
the change in refractive error induced by refractive surgery 1+(0.014* ORsp) -6.00) = -5.55 -3) = -2.13
from the average corneal power measured prior to surgery. KCL = BC of CL 45.61 + (-5.55) = 45.00 + (-2.13) =
The equation to determine the change in the corneal power + ORcp 40.06 D 42.87 D
from the known pre- and postoperative refractive and kera-
tometric values is
18 Ch a p t e r 2

Videokeratography measures the central corneal power profiles, significant treatments in less-than-ideal patients
inside the approximately 3-mm zone measured by keratom- may result in undesirable topographical abnormalities.
etry and may give a more accurate power to use in IOL cal-
culation formulas. For eyes that have had refractive surgery,
the corneal power derived from clinical history, contact
lens refraction, or videokeratography should be used in a R EFERENCES
third-generation theoretic formula, such as the Hoffer Q,
Holladay, or SRK/T, to calculate the IOL power used during 1. Duke-Elder S, Wybar KC. Cornea. In: Duke-Elder S, ed. System of
cataract surgery.27 Ophthalmology, Vol. 2: The Anatomy of the Visual System. London,
UK: Henry Kimpton Publishers; 1961:95-131.
Cheng and Lam28 investigated the K-value obtained
2. Gipson IK, Joyce NC, Zieske JD. The anatomy and cell biology
from the Gaussian optics formula (CalK) based on postop- of the human cornea, limbus, conjunctiva, and adnexa. In: Smolin
erative corneal topography by Orbscan II (Bausch & Lomb, G, Foster CS, Thoft RA, Azar DT, Dohlman CH, eds. Smolin and
Rochester, NY) and ultrasound pachymetry, comparing Thoft’s the Cornea: Scientific Foundations & Clinical Practice.
them to those obtained from the clinical history method Philadelphia, PA: Lippincott Williams & Wilkins; 2005:1-17.
3. McMenamin PG, Steele C, McGhee CNJ. Cornea: anatomy,
(estK). A high correlation was noted between the K-value
physiology, and healing. In: McGhee CNJ, Taylor HR, Gartry DS,
obtained by the clinical history method and the Gaussian Trokel SL, eds. Excimer Lasers in Ophthalmology: Principles and
optics formula (R = 0.97, P < 0.001). The mean difference Practice. London, UK: Martin Dunitz; 1997:41-45.
between the 2 methods was 0.13 D. However, Preussner and 4. Davson H. Physiology of the Eye. 5th ed. New York, NY: Pergamon
colleagues found that ray tracing is superior to Gaussian Press; 1990.
5. Maurice DM. The Cornea and the Sclera. London, UK: Academic
optics in calculating corneal power, the purpose of IOL
Press; 1985.
calculation.29 6. Warwick R. Eugene Wolff’s Anatomy of the Eye and Orbit. 8th ed.
Aramberri recommends adjusting the SRK/T formula London, UK: Chapman & Hall Medical; 1997:235.
using the double K method. He found using the preopera- 7. Vracko R, Benditt EP. Basal lamina: the scaffold for orderly cell
tive K value from keratometry or topography for the effec- replacement. Observations on regeneration of injured skeletal
muscle fibers and capillaries. J Cell Biol. 1972;55(2):406-419.
tive lens position calculation and using the clinical history
8. Hogan MJ, Alvarado JA, Wedell JE. Histology of the Human Eye.
method with the vergence formula improved the accuracy Philadelphia, PA: WB Saunders; 1971.
of IOL calculation S/P keratorefractive surgery.30 Gimbel 9. Krachmer JH, Mannis MJ, Holland EJ. Cornea, Vol I: Fundamentals,
and Sun31 evaluated target refractions based on measured Diagnosis and Management. 2nd ed. London, UK: Elsevier-Mosby;
and refraction-derived keratometric values by comparing 2005.
10 Farrell RA, McCally RL. Corneal transparency. In: Albert DM,
them with postoperative achieved refractions. Differences
Jakobiec FA, eds. Principles and Practice of Ophthalmology.
between target refractions were calculated using 5 IOL for- Philadelphia, PA: WB Saunders; 2000:629-643.
mulas and 2 A-constants. Refraction-derived keratometric 11. Arffa R. Grayson’s Diseases of the Cornea. 4th ed. St Louis, MO:
values led to more accurate IOL power calculations, along Mosby; 1997.
with using the Holladay 2 or Binkhorst 2 formula. 12. Meek KM, Boote C. The organization of collagen in the corneal
stroma. Exp Eye Res. 2004;78:503-512.
Chen and colleagues32 suggested using the flattest
13. Daxer A, Fratzel P. Collagen fibril orientation in the human corneal
keratometry value by calculation and aiming for -150 rather stroma and its implications in keratoconus. Inv Ophthal Vis Sci.
than plano in lens calculation to reduce the likelihood of 1997;38:121-129.
hyperopia after cataract surgery following keratorefractive 14. Meek KM, Tuft SJ, Huang Y, et al. Changes in collagen orienta-
surgery. Gimbel and colleagues33 also recommended using tion and distribution in keratoconus corneas. Inv Ophthal Vis Sci.
2005;46:1948-1956.
the smaller of the actual or refraction-derived keratometric
15. Hayes S, Boote C, Tift SL, Quantock AJ, Meek KM. A study of
values for calculating IOL power. As increasing numbers of corneal thickness, shape and collagen organization in keratoconus
baby boomers who are S/P keratorefractive surgery qualify using videokeratography and X-ray scattering techniques. Exp Eye
for cataract surgery, it is likely that specific parameters will Res. 2007;84:423-434.
develop to avoid hyperopic postoperative results. 16. Joyce NC, Meklir B, Joyce SJ, Zieske JD. Cell cycle protein
expression and proliferative status in human corneal cells. Invest
Ophthalmol Vis Sci. 1996;37:645-655.
17. Holly FJ, Lemp MA. Tear physiology and dry eyes. Surv Ophthalmol.

C ONCLUSION 1977;22:69-87.
18. Stein HA, Slatt BJ, Stein RM, Freeman MI. Fitting Guide for Rigid
and Soft Contact Lenses: A Practical Approach. Philadelphia, PA:
Mosby; 2002:3-12.
The understanding of natural corneal optics and how 19. Müller LJ, Marfurt CF, Kruse F, Tervo TM. Corneal nerves: struc-
they are changed with keratorefractive surgery is required ture, contents, and function. Exp Eye Res. 2003;76:521-542.
for all corneal surgeons. Surgical planning must incorporate 20. Robin JB, Rich LF, Elander RE. Principles and Practice of
the knowledge to safeguard the patient from visual quality Refractive Surgery. Philadelphia, PA: WB Saunders Co; 1997.
21. Fannin TE, Grosvenor T. Clinical Optics. 2nd ed. Boston, MA;
issues arising from drastic changes in corneal architecture.
Butterworth-Heinemann; 1996.
While night vision issues, glare, and distortion are mini- 22. Klyce S. Night vision after LASIK. Ophthalmology.
mized by increased ablation zones and improved ablation 2004;111(1):1,2.
 Co r n e a l An a t o my a n d O p t ic s 19

23. Munnerlyn C, Koons S, Marshall J. Photorefractive keratectomy: 29. Preussner PR, Wahl J, Lahdo H, Dick B, Findl O. Ray tracing for
a technique for laser refractive surgery. J Cataract Refract Surg. intraocular lens calculation. J Cataract Refract Surg. 2002;28:1412-
1988;14:46. 1419.
24. Baikoff G, Lutun E, Ferraz C, Wei J. Analysis of the eye’s anterior 30. Aramberri J. Intraocular lens power calculation after corneal
segment with optical coherence tomography: static and dynamic refractive surgery: double-K method. J Cataract Refract Surg.
Study. J Fr Ophtalmol. 2005;28(4):343-352. 2003;29:203-206.
25. Freedman KA, Brown SM, Mathews SM, Young RS. Pupil size and 31. Gimbel HV, Sun R. Accuracy and predictability of intraocular
the ablation zone in laser refractive surgery: considerations based on lens power calculation after laser in situ keratomileusis. J Cataract
geometric optics. J Cataract Refract Surg. 2003;29(10):1924-1931. Refract Surg. 2001;27:571-576.
26. Haigis W. Corneal power after refractive surgery for myopia: con- 32. Chen L, Mannis MJ, Salz JJ, Garcia-Ferrer FJ, Ge J. Analysis of
tact lens method. J Cataract Refract Surg. 2003;29:1397-1411. intraocular lens power calculation in post-radial keratotomy eyes. J
27. Hoffer KJ. Intraocular lens power calculation for eyes after refrac- Cataract Refract Surg. 2003;29:65-70.
tive keratotomy. J Refract Surg. 2004;20(6):783-789. 33. Gimbel HV, Sun R, Furlong MT, van Westenbrugge JA, Kassab
28. Cheng CK, Lam DS. Keratometry for intraocular lens power cal- J. Accuracy and predictability of intraocular lens power calcula-
culation using Orbscan II in eyes with laser in situ keratomileusis. tion after photorefractive keratectomy. J Cataract Refract Surg.
J Refract Surg. 2005;21(4):365-368. 2000;26:1147-1151.
Ch a p te r 3

Maps and Scale s

Tracy Schroeder Swartz, OD, MS, FAAO ; Ilan Cohen, MD;

Ray-Ann Lin, MD; Megan Buliano, OD; and Y. Ralph Chu, MD

C
orneal topography depicts the variations on the how it relates to scaling. The same map may appear differ-
surface of the cornea. Because the cornea is not ently depending on the scale used. Thus, determining which
perfectly spherical, nor does it perfectly coincide scale to use in a particular situation is the key to success.1
with any other geometrical abstraction, we need to consider Initially, a standardized absolute scale was proposed,2
the characteristics on each point along its surface in order to ranging from 9.0 to 101.5 D. The central portion of the range
understand its properties. was measured with 1.5-D step intervals and the extreme
The cornea has 2 surfaces responsible for the refraction limits of the range in 5.0-D step intervals. Although this
of light: the anterior and the posterior. The anterior surface range covered the entire power spectrum seen in corneal
is more important because approximately 90% of refraction practice, salient topographic features were occasionally lost
occurs there. The posterior surface contributes roughly 10% within the 5.0-D intervals, particularly at the low end of
of the total corneal power in a virgin eye. This assumption the scale. Hence, this was modified by the Klyce/Wilson
could lead to significant errors in eyes with keratoconus or Scale,3 which ranged from 28.0 to 65.5 D in equal 1.5-D
postkeratorefractive surgery. step intervals. It has still been argued that the 1.5-D interval
Numerous topographers enable measurement of the sur- is so wide that irregularities in corneal topography may be
face characteristics of the cornea. Each one uses slightly masked. An example of changes in topographic patterns
different formulas or techniques to derive the topographical associated with increased step size is shown in Figure 3-1.
map. To complicate matters, each topographer offers a wide Figure 3-1A shows more detailed irregularity compared to
range of plots to display the information obtained using its counterpart (Figure 3-1B) due to smaller step size.
similar but not identical scales, based on user preferences. However, it has been demonstrated that the 1.5-D scale
When evaluating a topographical map, one must note the can detect all the topographic characteristics identified
scale used and the type of map displayed to avoid misinter- by a more sensitive 1.0-D scale in a consecutive series
pretation. of patients that included contact lens-wearing corneas,
early to moderate and advanced keratoconus, penetrating
keratoplasties, extracapsular cataract extraction, excimer

SCALES laser photorefractive keratotomy, radial keratotomy, aphakic

epikeratophakia, and myopic epikeratophakia.4 In 1999,
the American National Standards Institute (ANSI) issued
To view a topographical map correctly, it is important to a report titled, Corneal Topography Systems—Standard
understand the general shape of the curvature patterns and Terminology, Requirements (ANSI Z-80.23-1999).5,6
Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical
21 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 21-32)
© 2012 SLACK Incorporated
22 Ch a p t e r 3

A B

Figure 3-1. (A) An axial curvature map using a step size of 0.5 D. (B) The same axial curvature map using a step size of 1.0 D.
Note the detail changes as the step size changes.

TABLE 3-1.

Differences Between the American National Standards Institute and Universal Standard Scale
SCALE NAME INTERVAL SIZE R ANGE OF POWERS (D) DISPLAYED POWERS (% OF 388 TOPOGRAPHIC
(D) MAPS OF 12 CORNEAL CONDITIONS)
ANSI 1.5 29 to 59 99.2
ANSI 1.0 34 to 54 96.6
ANSI 0.5 39 to 49 86.2
USS 1.5 30 to 67.5 99.9

This standard was anticipated to encompass the presenta- 1.5-D interval gives the best sensitivity, specificity, and
tion of information, the standardized scale, scale interval, range.7
and the representative color palette used for curvature and Table 3-2 describes the ability of topography scales to
elevation maps. It would allow users of corneal topogra- display corneal powers by relating interval size to dioptric
phers to directly compare the topography maps produced range. There is no standardization of topographic scales
by different manufacturers. However, Annex B of the ANSI between commercial companies. This makes it more diffi-
standard, which defines scale intervals, the scale center, and cult to compare examinations performed using different
a color convention, does not specify a single, well-defined systems. For example, in the absolute/standardized scale
color palette but rather suggests a variety of numeric and default mode, the Humphrey Atlas (Carl Zeiss Meditec,
color scale combinations.5 Jena, Germany) ranges from 38.5 to 50.0 D in 0.50-D inter-
In an effort to overcome this potential point of confusion, vals, while the out-of-the-box absolute default scale for the
an alternative color scale was proposed—the Universal Tomey TMS-4 (Tomey Corporation, Nagoya, Japan) ranges
Standard Scale (USS). The USS was able to overcome the from 9 to 101.5 D.
problems encountered with the ANSI standard by associat-
ing a single, well-defined numerical scale with a single,
well-defined color palette. Table 3-1 highlights the differ-
ences between the ANSI and USS scales. This method has I NTERPRETATION
been shown to produce maps that were consistent and could
be rapidly and correctly interpreted. The USS displays a When evaluating a topographic display either through
range of powers that encompass 99.9% of both naturally a printed report or on the instrument’s screen, one should
occurring and surgically induced corneal shapes. The USS study maps in a structured way to avoid mistakes in
is based on 1.5-D intervals and a well-defined color scale interpretation. The guidelines that follow will aid in suc-
that would be most relevant for average clinical uses.7 The cessful interpretation of the maps.8 Once a topographic
 Ma p s a n d Sca le s 23

TABLE 3-2. A

Topographical Maps Use

Colors to Denote Curvature
COLOR POWER (D)
Red 48.0 Steep
Orange/yellow 45.0
Yellow/green 43.5 Average
Green/light blue 42.0
Blue 39.0 Flat
B

image has been captured, it is scaled and graphically rep-

resented in a color-coded contour map. Depending on the
scale used for color conversion, identical maps may, in fact,
appear to be different. Scale ranges and color representation
vary among topographers, which makes direct comparisons
between topographers open to interpretation. Color-coded
contour maps of corneal surface power have been adopted
as a standard presentation scheme in corneal topography.
They markedly facilitate the viewer’s interpretation through
the association of power with color and recognition of
pathologic features with the patterns formed by the map Figure 3-2. (A) Mild dry eye causing irregular astigmatism with
contours. Color-coded contour maps were initially devel- the autoscale. (B) Mild dry eye appears less irregular when
oped for videokeratoscopy, and with the introduction of using the standard scale.
projection-based topography systems, a similar color-coded
system was applied to elevation maps. Warm colors, red
and orange, are used to represent relatively higher powers TABLE 3-3.
(steeper curvatures); green and yellow are used for pow-
ers associated with normal corneas; and cool color hues Comparison of Absolute and Relative Scales
of blue are used to denote relatively lower powers (flatter
curvatures). For elevation maps, high areas were depicted COLOR STANDARDIZED / NORMALIZED /
by warm colors, and low areas were depicted by cool colors. ABSOLUTE SCALES RELATIVE /AUTOSIZE
This concept, along with standard scale, provides an intui- Standardized Nonstandardized
tive basis for the interpretation of corneal topography.4 For
Good for comparison Comparison of maps is
example, in a case of keratoconus, the red area on a height
more difficult
map corresponds to the highest point, which is the apex of
the cone. In the same case, the red area on a curvature or Large step sizes Small steps
power map is the steepest area, which is usually located Low resolution High resolution
adjacent to the cone inferiorly.
Large range of powers Narrow range of powers
As you can see in Figure 3-2, the absolute/standard-
ized scale significantly expands the dioptric range used. Good for screening Subtle features present
Clinically significant irregularities may be masked when Good for gross pathology Good for detail
comparing eyes with widely disparate curvature readings.
When the range is standardized, the steepest curvatures
appear red, while the flattest curvatures appear blue.
General patterns remain the same, but intricate curvature Autosizing or normalized scales lend confusion. Because
changes disappear because the interval of measurement the range of powers for the cornea varies, physiologic
also increases from 0.25 to 0.50 D.1 asymmetry or astigmatism may be amplified by an adapt-
The use of a standardized and fixed color scale for able scale and may lead to misdiagnosis of pathology. An
routine clinical examination is important for consistent example of this is shown in Figure 3-2, a patient with mild
and correct evaluation of corneal topography (Table 3-3). dry eye. The irregular astigmatism is more evident using the
24 Ch a p t e r 3

auto scale. Alternatively, a cornea with substantial irregular

astigmatism can be made to look less abnormal with an STANDARD /ABSOLUTE SCALE
adjustable scale.4
The allocation of colors on an absolute/standardized The standard or absolute scale assigns the same color
scale is related to the distribution of corneal powers in the to a specific dioptric interval of curvature and forces that
normal population. Central corneal power has an approxi- cornea’s parameters to fit within that range. Although the
mate Gaussian distribution (represented by a bell-shaped range varies by instrument, each individual instrument’s
curve). The mean central corneal power is 43.50 D, which range is consistent. Therefore, for each map viewed, a direct
is depicted by a color from the middle of the spectrum. comparison of images from different eyes or from curva-
Approximately 66% of the population has a central corneal ture changes of one eye (preoperative and postoperative
power within one standard deviation of the mean (42 to refractive surgery patients) can be made quickly and accu-
45 D), and this is represented by the adjacent colors on the rately. Because there has been no standardization of scales
scale. Less than 3% of the population has a central corneal between commercial companies to date, this makes it more
power beyond ±3 SD, represented by red and dark blue. If difficult to directly compare examinations performed using
these colors are present on an absolute scale map, the cornea different systems.
is unlikely to be normal.7 A potential limitation of the standard scale is its inabil-
ity to highlight subtle corneal changes that are impor-
tant to evaluate when considering refractive surgery. Mis-
AUTOSIZE/N ORMALIZED interpretation can exist due to the large range in intervals
and loss of detail. Because the ranges and parameters are
SCALE standard, 2 different maps can be compared—unlike with
the autosize scale. This scale is most useful when compar-
ing maps over time and with varying interpreters. It gives
The normalized scale (relative, adaptive) or autoscale the best results for gross evaluations of the cornea.
automatically adjusts and subdivides a specific cornea into
several dioptric intervals based on the actual dioptric range
of the cornea measured. Based on color representation,
maps may appear to represent similar dioptric curvatures M AP D ISPLAYS
when using this scale, but depending on a patient’s specific
corneal curvature, these intervals may vary in range and Topographers capture data from points on the corneal
dioptric value. This may falsely give the appearance that all surface using various technologies. Placido imaging is the
maps represent equal curvatures. most commonly used system, but Scheimpflug imaging
A normalized scale uses a set number of colors that are and slit scanning are also employed. From these primary
automatically adjusted to fill the range of dioptric values data points, a wide range of maps can be displayed by the
for that single map. The mean power for that cornea is posi- computer. The maps are usually color coded for easier inter-
tioned in the center of the scale. The normalized scale has pretation by the clinician.
the advantage over the absolute scale by fusing nar rower
steps between the contours, which provides more detail.
Some systems limit how small the steps can be so that the
information generated is still clinically relevant. However,
Cu r vat u re Map s
as the scale may be different for almost every examination, Curvature or power maps include axial, tangential, and
it should be checked carefully before studying the map. refractive maps. The most commonly used curvature map
The automatic scale is a useful tool for examining a sin- is the axial map. This map depicts the anterior curvature
gle eye in greater detail, as the smaller range will highlight of the cornea at each point along its surface, usually up to
variation and subtle changes with the caveat that normal a diameter of 7 mm. To know the curvature of a point on a
variations may seem accentuated to the point of being spherical surface, we need to find the circle that best fits the
questionable surgical candidates. For example, the use of plane curve at the surface point of interest and determine
a normalized scale can produce a pair of maps of similar the local radius of curvature (K), where K = 1/R. The radius
appearance for a patient with advanced keratoconus in one of curvature (R) is measured in millimeters. Curvature is
eye (using a large step interval) and a subclinical cone in the inverse of the radius, measured in D The optical power
the other eye (using a small step interval).7 The automatic of the cornea at any given point is a function of the curva-
scale may also be beneficial in patients having undergone ture at that point but is not the same as the curvature. The
laser-assisted in situ keratomileusis (LASIK) with visual smaller the radius of curvature, the more curved the surface.
complaints where small irregularities may cause problems Curvature may be displayed using several types of maps:
with visual acuity. axial and tangential or meridional.
 Ma p s a n d Sca le s 25

A B

Figure 3-3. (A) An axial map. (B) A tangential map of the same eye. Note the focal changes seen with this map are more detailed
than those of the axial map.

Figure 3-4. Elevation and axial maps of

the same eye show that the inferior por-
tion is not actually elevated, but rather
depressed below that of the reference
sphere due to the sharp curve of the cor-
nea inferiorly.

Axial curvature maps are obtained by measuring the direction relative to the other points on the particular ring.
curvature of the cornea at each point relative to a specified For this reason, they may be referred to as “meridional
axis, usually the visual axis. The local radius for a corneal maps.” Tangential maps are also sometimes referred to as
surface point can be measured as the distance from the “local curvature” or “instantaneous curvature.” Because
point to the optic axis along the normal. Unfortunately, the axis of reference is different for each point, there is a
this requires the assumption that the center of curvature for higher degree of variability from point to point.9 There is
the specific surface point is located along the optic axis. also a smaller number of mathematical assumptions (ie, the
This radius is then called the axial radius of curvature. The sphericity of the cornea) used in this formula. The tangen-
center for all surface points is on the optic axis for spheri- tial map can recognize sharp power transitions more readily
cal surfaces only, and the cornea is an aspheric surface. than the axial map and eliminates the “smoothing” appear-
While this assumption was acceptable for keratometry, it ance that appears on the axial map for most topographers.
introduces significant error in the corneal periphery. This Because tangential maps are more likely to illustrate focal
effectively smooths small variations on the surface of the irregularities, they are useful in contact lens fitting.
cornea and provides a less irregular map that is easier to When reading a curvature map, it is important not to
read and understand.8 The axial map gives a global view confuse “steepness” (typically pictured by hotter colors)
of the corneal curvature as a whole. However, axial maps with height. Curvature simply means that the shape of the
tend to ignore minor variations in the local curvature. A cornea is changing, but it does not tell the direction of the
comparison between an axial map and a tangential map for change. For example, a curvature map of a keratoconic cor-
the same eye is shown in Figure 3-3. Figure 3-3A is an axial nea characteristically shows red inferiorly, corresponding to
map, which is less irregular than Figure 3-3B, a tangential the inferior distortion of the cornea inherent to the disease.
map. The elevation map of the same eye in Figure 3-4 shows
The tangential map uses calculations based on a different that the inferior portion is not actually elevated, but rather
mathematical approach to more accurately determine the is depressed below that of the reference sphere, due to the
peripheral corneal configuration. Tangential maps represent sharp curve of the cornea inferiorly.
the local curvature of the cornea at each point in meridional
26 Ch a p t e r 3

B
Figure 3-5. Many systems incorporate several maps into a
single display. This patient reported visual distortion and glare,
which was most easily explained by the refractive power map
(lower right).

The refractive map displays the calculated refractive

power of the cornea based on Snell’s law of refraction.
It correlates vision to corneal shape. It is not often used;
however, it may identify distortion in vision related to focal
irregularities in the cornea and explain vision loss not
apparent upon slit-lamp examination (Figure 3-5).

ELEVATION M APS Figure 3-6. O perator choices for alignment of anterior eleva-
tion map displays include (A) float alignment, where there are
no constraints assigned to the reference sphere and (B) cen-
A topographical map of a landscape depicts each point tered alignment, where the center of the sphere is constrained
to the viewing axis. (continued)
on the map in relationship to sea level. In corneal topog-
raphy, an elevation map needs to be depicted with relation
to a reference plane. This plane is often a sphere with a
diameter that most closely resembles the overall diameter of Changing the size, shape, or alignment of the reference
that specific cornea. Each point on the cornea that coincides sphere will have an impact on the topography map, just as
with the reference sphere is represented by the color green. changing the sea level would impact the heights of struc-
Warmer colors represent points that are higher than the ref- tures on the land surface. For practical applications, the
erence plane, and cooler colors represent points lower than reference plane is usually set to a sphere. The size of the
the reference plane. Remember that the refractive power of sphere is chosen such that it best fits the cornea in question.
the cornea is not represented by the elevation itself but by Choices for alignment vary, but often include float, cen-
changes in elevation. tered, pinned, or apex. In Figure 3-6, the same eye has been
In refractive surgery, elevation maps are extremely mapped using the 4 choices. When using float alignment,
important. Because removing tissue most directly affects there are no constraints assigned to the reference sphere.
the elevation of the cornea, we have to understand eleva- The diameter and the location of the sphere are chosen to
tion topography in order to grasp the effects that we are maximize the area of contact between the reference sphere
imparting on the cornea. The cornea is usually prolate, not and the cornea. This is the default alignment for most
perfectly spherical. Therefore, in order for a topographical topographers. When using centered alignment, the center
map to have a quantitative and qualitative meaning, we need of the sphere is constrained to the viewing axis with the
to know how the reference sphere correlates to the corneal ability to move along that axis. Pinned alignment imposes
surface. This attribute may be fixed in some topographers the pin (P) constraint, which forces the reference sphere and
and can be chosen by the user in others. the data surface to intersect on the viewing axis. The sphere
 Ma p s a n d Sca le s 27

Figure 3-7. An elevation and axial power map in an eye show-

Figure 3-6 (continued). O perator choices for alignment of ing with-the-rule astigmatism.
anterior elevation map displays include (C) pinned alignment,
which forces the reference sphere and the data surface to
intersect on the viewing axis and (D) apex alignment, which In refractive surgery, we often need to compare pre-
imposes both the center and pinned constraints. and postoperative topographical maps. The goal is to
determine where tissue was removed and how much tissue
was removed. We can then correlate this to the refractive
center may be off-axis in this scenario. Apex alignment effect that was achieved with the keratorefractive surgery.
imposes both center and pin (C + P) constraints. The sphere Elevation mapping directly illustrates tissue removal. Some
surface intersects the data surface on the viewing axis, and programs perform a pre- and postoperative analysis based
the center of the sphere lies on the viewing axis. on the anterior elevation and pachymetry maps to calculate
It is important to realize that the axial curvature map the exact amount of tissue that was removed at each point.
and anterior elevation map are related, but do not directly This is useful in detecting postsurgical irregularities such
correspond. Figure 3-7 shows an elevation map and an axial as decentered ablation or central islands. Curvature maps
power map in an eye that has with-the-rule astigmatism. should not be used to identify decentered ablations because
When we follow the vertical axis from a point superiorly they may exaggerate the decentration, exemplified in
toward the center, we see that elevation increases as we Figure 3-8.
approach the center and decreases as we move inferiorly. General understanding of the anterior elevation topo-
This means that the curvature is steep along this meridian, graphical map can be applied to the posterior elevation
as shown in the axial map. The steep meridian here is rep- topography. The posterior elevation map is created much
resented by the vertical bow-tie pattern. Note that the warm like the anterior elevation map in design, with the excep-
and cold colors are somewhat reversed when you compare tion that it represents the posterior surface of the cornea.
an elevation map to a curvature map because a change in The choices for alignment of the posterior surface elevation
elevation causes a change in curvature. maps are the same as those for the anterior surface: float,
centered, pinned, or apex. These are shown in Figure 3-9.
28 Ch a p t e r 3

A B

Figure 3-8. (A) Elevation maps rather than curvature maps should be used to identify decentered ablations. (B) Curvature maps
tend to exaggerate the decentered area and should not be used to identify decentered ablations.

A B

C D

Figure 3-9. O perator choices for alignment of posterior elevation map displays include (A) float alignment, where there are no
constraints assigned to the reference sphere; (B) centered alignment, where the center of the sphere is constrained to the view-
ing axis; (C) pinned alignment, which forces the reference sphere and the data surface to intersect on the viewing axis; and
(D) apex alignment, which imposes both the center and pinned constraints.
 Ma p s a n d Sca le s 29

Figure 3-11. Subtle topographic abnormalities in forme fruste

keratoconus. Note the slightly decentered posterior elevation,
and displacement of the thinnest point. There is mild asym-
metry on the curvature map.

Figure 3-10. (A) Pachymetry map from a healthy patient.

(B) Pachymetry map from a patient with keratoconus.

PACH YMETRY M APS

Slit scanning, Scheimpflug, and anterior segment optical
coherence tomography imaging technology yield pachym-
etry data. It gives us information not only about the central
or paracentral points of the cornea customarily obtained
by ultrasound pachymetry, but also about the distribution B
of the thickness along the surface. Figure 3-10 shows a
pachymetric map from a healthy patient (Figure 3-10A) and
a patient with keratoconus (Figure 3-10B).
Pachymetry information is useful in glaucoma assess-
ment, as well as in screening refractive surgery candidates
and estimating residual corneal bed thickness postopera-
tively. These data are invaluable when combined with the
other maps to rule out keratoconus or forme fruste kerato-
conus. Figures 3-11 and 3-12 illustrate the subtle differences
between the forme fruste keratoconus, true keratoconus,
and pellucid marginal degeneration. Corneal thinning cor-
responds to the apex of the cone in keratoconus (Figure
3-12A) and inferior to it in cases of PMD (Figure 3-12B).
Early topographical signs may only be found on the poste- Figure 3-12. (A) The ectasia in keratoconus tends to be central.
rior surface using tomography. (B) The ectasia in PMD tends to be inferior.
30 Ch a p t e r 3

Figure 3-13. When topography maps appear nonsensical, dry

eye may be the etiology, as in this case. With treatment, the
topography should become more regular.

Figure 3-14. Data loss resulting

from INTACS segments.

Top ograp h ic Map A corneal scar can interfere with the data acquisition
by a slit-scan topographer. The slit-scanning unit may
In ter p ret at ion misinterpret the opacity as a corneal surface, resulting in
errors in both pachymetry and posterior elevation maps.
When looking at a topography map, you need to consider A disrupted tear film can interfere with the reflection of
the clinical picture and interpret the map in that perspec- light from the cornea, create artificially steepened areas,
tive. Correct identification of the type of map is crucial—is or lead to miscalculations based on missing data points.
it a curvature map or elevation map? When comparing Irregularities resulting from a dry eye can be seen in
maps over time, it is necessary to examine the scale. Some Figure 3-13. INTACS corneal ring segments (Valley Eye
clinical entities that commonly interfere with the accuracy Associates, Appleton, WI) can also result in loss of data in
of topographic maps include corneal opacities, dry eye and certain systems, as seen in Figure 3-14.
insufficient tear film, eye movements, and contact lens
wear.
 Ma p s a n d Sca le s 31

Figure 3-15. Contact lens warpage

can result from overwear of both
soft and gas permeable lenses, and
lenses must be discontinued before
final diagnosis or IO L selection in
the case of patients pursuing cata-
ract surgery. It is easy to mistake
warpage for ectatic disease, and
pachymetry facilitates the differen-
tial diagnosis.

Minimal eye movements are required for data collection, surgeon who assesses the eligibility of a candidate for elec-
especially for the slit-scanning systems where data capture tive surgery to understand the characteristics of the different
requires more time. Some topographers may accomplish types of maps for proper interpretation and clinical utility.
this process faster than others, and this may be helpful in
patients who have difficulty opening their eyes or who can-
not maintain fixation for long.
Contact lens wear may distort the shape of the cornea. R EFERENCES
An example of contact lens warpage in a gas-permeable
lens wearer is shown in Figure 3-15. This is not uncommon 1. Lebow KA, Grohe RM. Differentiating contact lens induced warp-
in hard contact lens wearers and in prolonged soft contact age from true keratoconus using corneal topography. CLAO J.
1999;25(2):114-122.
lens wearers. These eyes may resemble early keratoconus
2. Maguire LJ, Singer DE, Klyce SD. Graphic presentation of com-
on topography. For this reason, it is recommended that puter-analyzed keratoscope photographs. Arch Ophthalmol.
contact lens wear be discontinued for a sufficient duration 1987;105:223-230.
to allow the cornea to return to its natural shape. A com- 3. Wilson SE, Klyce SD, Husseini ZM. Standardized color-coded
mon rule of thumb is 1 week plus 1 week for each decade of maps for corneal topography. Ophthalmology. 1993;100:1723-1727.
4. Oshike T, Klyce SD. Corneal topography: basic concepts. In:
gas-permeable lens wear. Tomographers facilitate this dif-
Brightbill FS, ed. Corneal Surgery—Theory, Technique, and
ferential diagnosis, but careful ultrasound in all 4 quadrants Tissue. 3rd ed. St. Louis, MO: CV Mosby; 1999.
can be used as well. 5. Smolek M, Klyce SD, Hovis JK. The universal standard scale: pro-
Epithelial surface irregularities such as superficial punc- posed improvements to the American National Standards Institute
tate keratitis, recurrent corneal erosions, and abrasions can (ANSI) scale for corneal topography. Ophthalmology. 2002;109:361-
369.
also interfere with topography. They typically cause focal
6. Secretariat, Optical Laboratories Association. American National
irregularities or loss of data (seen as white spots on maps). Standard for Ophthalmics—Corneal Topography Systems:
It is important to let such irregularities heal before impor- Standard Terminology, Requirements. Merrifield, VA: Optical
tant topography-based decisions are made. Laboratories Association. American National Standards Institute;
1999:ANSI Z80.23-1000.
7. Corbett MC, Rosen ES, O’Brart DPS. Presentation of topo-
graphic information. In: Corbett M, O’Brart D, Rosen E. Corneal
C ONCLUSION Topography: Principles and Applications. London, UK: BMJ
Books; 1999:31-59.
8. Applegate RA. Comment on characterization of the inherent error in
a spherically-biased corneal topography system in mapping a radi-
Topography has become an indispensable tool for the
ally aspheric surface. Refract Corneal Surg. 1994;10:113-114.
clinician in assessing the qualities of the corneal surface. 9. Roberts C. Characterization of the inherent error in a spherically-
The ability to interpret and understand various topographi- biased corneal topography system in mapping a radially aspheric
cal maps can be complex given the variety of technol- surface. Refract Corneal Surg. 1994;10:103-116.
ogy available. It is especially important for the refractive
 SECTIO N II

T O PO GRAPHIC
T ECHNO LO GIES
Ch a p te r 4

To p o grap hic
Te chno lo gie s

Tracy Schroeder Swartz, OD, MS, FAAO ; Zuguo Liu, MD, PhD;
Xiao Yang, MD; and Mei Zhang, MD

T
he principles of topography are based on the reflec- and a computer for image analysis. The number, position,
tions of a concentric ring of light upon the cor- color, and thickness of the rings vary between systems.
nea. Variations in curvature and astigmatism are Placido systems are typically divided into 2 types: near
represented as an asymmetry of the keratographic pat- (also called small targets) or distance (called large targets).
terns. Modern keratoscopes incorporate complex images Near target systems typically allow for imaging with lower
in the analysis of topographic anomalies. Refractive sur- illumination and enjoy greater corneal coverage. However,
gery and the quest for better understanding of corneal they are sensitive to focusing adjustments, and facial anato-
shape (versus curvature) resulted in new technologies with my may hinder measurement. Large target systems require
increased precision in evaluating complex corneal shapes. more illumination and are less sensitive to focusing error
Technologies include Placido disk imaging, PAR, slit scan- but cover less of the cornea.1
ning, Scheimpflug imaging, ultrasound, interferometric, Most systems project images of illuminated keratoscope
and optical coherence tomography (OCT) systems. This rings onto the corneal surface to produce a virtual image
chapter contains a brief overview of each technology. of the Placido disk about 4 mm behind the corneal vertex.2
An example is shown in Figure 4-1. They directly measure
the curvature of the cornea and calculate the elevation map

P LACIDO D ISK I MAGING using a coordinate system from the curvature data. This
requires assumptions about the corneal geometry, however.
Elevation is generated by fitting slope data to a predefined
Placido imaging is based on the overlay of concentric mathematical model, which may be spheric, aspheric, or a
mires on the cornea. Keratoscopes permitting the direct conical section. While this practice is reasonable in normal
observation of illuminated mires upon the cornea demon- corneas, it may result in serious error in diseased eyes or in
strate the Placido ring. The closer the mires, the steeper the eyes having undergone keratorefractive surgery.3
axis. The wider the rings, the flatter the axis. It was the first Studies regarding the accuracy of early Placido disk sys-
technology to be used to evaluate the shape of the cornea tems found acceptable levels of accuracy and reproducibil-
in conjunction with computer analysis. While systems may ity. However, most test objects were spheres.4 Systems tend
differ somewhat, all contain a transilluminated Placido tar- to be more accurate centrally than peripherally, and defocus
get in the shape of a cone or disk, an imaging system con- increases errors. Clear surfaces are required for clear mires.
taining an objective lens and camera, a video frame grabber, Reported accuracy of dioptric power varies from 0.1 to

Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical

35 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 35 -42)
© 2012 SLACK Incorporated
36 Ch a p t e r 4

Figure 4-2. AstraMax system’s polar grid. (Reprinted

with permission of Aleksandar Stojanovic, MD.)
Figure 4-1. A placido image. (Reprinted with permission of Joe S.
Wakil, MD.)

0.25 D and from approximately 0.018 to 0.045 mm in the

radius of curvature.5
Variants on Placido disk technology include a 3-dimen-
sional grid system, such as the AstraMax system (Lasersight
Technologies, Inc, Winter Park, FL) shown in Figure 4-2.
Rather than using a Placido disk, a polar grid provides both
radial and concentric data points, enabling measurement of
radial distance and rotational concentric changes in the cor-
nea. The AstraMax uses 3 cameras to obtain multi-angled
shots, measuring each data point independently of an adja-
cent point. A raw image is shown in Figure 4-3.

ELEVATION -BASED SYSTEMS

Several systems measure elevation directly: the PAR
Corneal Topography System (CTS; ParTech, Inc, New
Hartford, NY), Orbscan (Bausch & Lomb, Rochester, NY),
Scheimpflug imaging, and OCT. The first systems were the
PAR and the Orbscan slit-scanning systems, which use a Figure 4-3. AstraMax raw image of a keratoconic cornea.
triangulation method to measure the elevation, and curva-
ture data were calculated without the error of geometrical
assumption.6 can be obtained with epithelial irregularity or defects,
sutures, or stromal ulceration.
The PAR CTS was the first topography system to pro-
Ra ster stereograp h y duce an elevation map of the corneal surface using rasterste-
reography.8 It projected a grid onto the corneal surface and
a n d t h e PAR Cor n ea l computed elevation data based on the distortion of the grid.
The PAR system requires that a small amount of fluorescein
Top ograp h y System be placed in the tear film, and the images are collected using
Rasterstereography is a method of topographical evalu- standard fluorescence-based photography. Image acquisi-
ation that creates an elevation map. Unlike Placido disk tion is rapid and relatively insensitive to focusing. From the
systems, it does not depend on the reflectivity of the cor- known geometry of the grid projection and imaging system
neal surface and can provide information about the entire paths, rays can be intersected into 3-dimensional space to
corneal, limbal, and interpalpebral conjunctival surfaces.7 compute the X, Y, and Z coordinates of the surface.9
Because a smooth reflective surface is not required, images
 To p o gra p h ic Te c h n o lo gie s 37

The PAR CTS can provide elevation, curvature, and 45 degrees to the right and left of the video axis. Twenty
keratometry maps. Unlike Placido disk-based videokera- slits are projected from the left and 20 from the right.
toscopes, the PAR CTS produces a true elevation map and Proprietary software image registration attempts to mini-
requires neither a smooth reflective surface nor precise mize the influence of involuntary eye movements during
spatial alignment for accurate imaging.10 Based upon the data acquisition.
fact that the cornea is an asymmetric refractive surface, the The typical display used by the Orbscan incorporates
PAR corneal elevation map can be obtained by comparing 4 images: the anterior and posterior elevation maps, the cur-
corneal height with a spherical reference surface, which is vature (axial) map, and the pachymetry map. An example
the average cornea elevation or a certain preset standard is shown in Figure 4-4. When used for screening, Tanabe
height. The actual corneal surface is either above or below and colleagues recommended using 10- or 20-µm scales for
this reference surface measured at individual points. elevation maps, which best identified abnormal corneas.14
Unlike Placido disk-based videokeratoscopes, the PAR Modis and colleagues investigated the anterior and pos-
System requires neither a smooth reflective surface nor pre- terior corneal shape, curvature, and thickness of normal
cise spatial alignment for accurate imaging.11 The system human corneas using the Orbscan. Scanning slit topogra-
demonstrated the ability to image irregular, de-epithelial- phy seems to be a reliable technique for the evaluation of
ized, and keratectomized corneas. The PAR CTS can be normal corneas not only for anterior shape and curvature,
installed on slit-lamp microscopes, surgical microscopes, or but also for a real pachymetry gradient recording.15 This
automatic optometry instruments, allowing topographical was the first system to yield pachymetry data. However, the
examination intraoperatively. accuracy of these measurements remains controversial. It is
Currently, there is limited information on the normal generally accepted that Orbscan measurements of central
corneal topography with the PAR CTS, and it is rarely used corneal thickness are greater than ultrasonic pachymeter
in clinical practice. Naufal and colleagues12,13 investigated measurements in virgin eyes.16 The role of the Orbscan
corneal elevation maps in 100 normal eyes of 50 subjects pachymetry is limited by lack of repeatability for peripheral
using the PAR system. Five categories were identified in measurements, however, and is recommended for central
their study: unclassified, regular ridge, irregular ridge, CT measurements only.17
incomplete ridge, and island. They found the surface of the Kawana and colleagues compared central corneal thick-
normal cornea was not smooth and spherical, and surface ness measurements of 3 pachymetry devices in eyes after
irregularities ranged from small central islands of eleva- laser in situ keratomileusis (LASIK). They found in post-
tion to complete, elevated bands crossing the cornea. Priest LASIK eyes, Orbscan II scanning slit topography sig-
and colleagues13 investigated the accuracy and precision nificantly underestimated corneal thickness. Noncontact
of the elevation topography from the Tomey Topographic specular microscopy gave smaller thickness readings than
Modeling System (TMS-1) and the PAR CTS. Based on ultrasonic pachymetry, but these 2 units showed an excel-
quantitative analysis of elevation measurements, they con- lent linear correlation.18
cluded that the CTS represented surface topography more
accurately than the TMS-1.13
Sch eim p flu g Im agin g
Scheimpflug imaging is based on a principle named
TOMOGRAPH Y after and patented by Austrian Theodor Scheimpflug in
Vienna in 1904. To best appreciate the advantages of this
Tomography is the creation of a 3-dimensional model modality for ophthalmologic imaging, let us review basic
from 2-dimensional images and is accomplished using principles and limitations of an ordinary camera.
slit scanning, Scheimpflug imaging, OCT, and ultrasound A typical camera uses 3 imaginary surfaces, referred to
systems. as planes: the film plane, the lens plane, and the plane of
sharp focus. A camera’s film is fixed upon the film plane.
The lens plane passes through the optical center of the lens
Slit-Sca n n in g and is perpendicular to the lens axis. The lens will depict
any object that is positioned on the plane of sharp focus
Slit-scanning technology is currently used by a single crisply onto the film plane. For an ordinary camera, these
system, the Orbscan. The Orbscan uses a scanning slit- 3 planes are all parallel to each other, and, therefore, all
beam similar to the parallel pipes used in biomicroscopy 3 planes are perpendicular to the lens axis. As long as a
and direct stereotriangulation to measure the anterior cor- given application only calls for limited depth of focus,
neal surface. During the 1.5-second examination, 2 scan- this setup works well. The depth of focus is the range over
ning slit lamps project a series of 40 slit beams angled at which the film plane can be moved while maintaining an
image of acceptable sharpness.
38 Ch a p t e r 4

Figure 4 -4. Typical

O rbscan quad map of
a normal eye.

Ophthalmologic applications require an extended depth

of focus to effectively image the entire anterior segment.
In Scheimpflug imaging, the 3 planes are not parallel but
slanted so they intersect in a line or, simplified, in a point
of intersection. Put more elaborately, the Scheimpflug
principle states that if the lens is tilted in a way that the
resulting lens plane intersects the film plane, the plane of
sharp focus, due to its dependence upon the lens plane, must
also pass through that same line of intersection. This line is
sometimes referred to as the “Scheimpflug line.” The ben-
efit this setup brings is extended depth of focus. Extended
depth of focus has a trade-off: distortion of the image. The
Pentacam (OCULUS, Wetzlar, Germany) computes the
acquired picture to compensate for the distortion.
Scheimpflug imaging gives a complete picture from
the anterior surface of the cornea to the posterior sur- Figure 4-5. Clinical Scheimpflug image in a patient with kera-
toconus.
face of the lens as viewed through the pupil, as shown in
Figure 4-5. Three-dimensional models of the front surface
(Figure 4-6) and back surface (Figure 4-7), as well as are digitized and stored. Using the I-scan obtained by com-
pachymetry (Figure 4-8), are created. puting the analytic signal magnitude of the deconvolved
ultrasound signal, layer thickness measurements are made
with a precision of 2-µm standard deviation at 120-µm
Ult ra sou n d Digit a l intervals along each scan plane. The data are stored as an
Top ograp h y array, z (x,y), mapping thickness (z) onto horizontal and
vertical (x,y) spatial coordinates. Pachymetric maps are then
The Artemis system (ArcScan, Inc, Morrison, CO) uses constructed by plotting local thickness against measure-
high-frequency ultrasound scanning enhanced by digital ment point position. This technique provides the corneal
signal processing. Ultrasonic echo data from consecutive surgeon with a new tool for the topographic evaluation of
parallel B-scans of the cornea spaced at 250-µm intervals the thickness of anterior corneal layers in normal and
 To p o gra p h ic Te c h n o lo gie s 39

Figure 4-9. An ultrasound image of the cornea S/P LASIK.

S = Break in Bruch’s membrane. I = Interface. T = Flap demar-
cation. E = Epithelium. B = Bowman’s membrane, K = Flap.
P = Endothelium. H = O uter edge of flap. (Reprinted with per-
mission of Dan Z. Reinstein, MD.)

pathologic corneas with high precision. The resolution of

the Artemis is sufficient to distinguish individual corneal
layers such as the epithelium, stromal component of the
Figure 4-6. Pentacam anterior elevation map.
flap, and residual stromal bed. In addition, the technique
is not limited to optically transparent tissue.19 An example
of the images obtained with this technology is shown in
Figure 4-9. The common display used is the C12 array, illus-
trated in Figure 4-10. This array was created from scans of
the right cornea of a patient scanned prior to LASIK and
6 months postoperatively.
Reinstein and colleagues investigated precision, imaging
resolution, 3-dimensional thickness mapping, and clini-
cal utility of a new prototype—3-dimensional very high-
frequency (VHF; 50 MHz) digital ultrasound scanning
system for corneal epithelium, flap, and residual stromal
thickness after LASIK. They found VHF digital ultrasound
arc-B scanning provides high-resolution imaging and high-
precision 3-dimensional thickness mapping of corneal lay-
ers, enabling accurate anatomical evaluation of the changes
induced in the cornea by LASIK.20

Figure 4-7. Pentacam posterior elevation map.

I NTERFEROMETRIC SYSTEM
This technique uses laser holographic interferometry
fringe patterns to depict deviations of the corneal surface.
Interferometry is based on the principles of light wave inter-
ference. It records the interference pattern generated on the
corneal surface by 2 coherent wavefronts. High accuracy is
theoretically possible.1 Fringe patterns can be interpreted
as contour maps of surface elevations where the difference
between the 2 consecutive fringes is equal to an elevation
difference of half the light wavelength (about 0.5 µm).
The shape of the cornea can be calculated by adding this
amount to the reference sphere elevation. Unfortunately,
interferometric methods are sensitive to eye movements,
and a system is required to maintain head position. Despite
reported sensitivity of less than 0.1 D of curvature within
the 5-mm vertex5, the system is too complicated to gain a
Figure 4-8. Pentacam pachymetry map. foothold in the field.
40 Ch a p t e r 4

Figure 4-10. The C12 Artemis dis-

play. (Reprinted with permission
of Dan Z. Reinstein, MD.)

O PTICAL C OH ERENCE
TOMOGRAPH Y
This technology, which originally was used to evaluate
the posterior pole and retina, has been adapted for use on
the anterior segment. OCT uses cross-sectional scanning
based on reflection and scattering of light from the struc-
tures within the anterior segment. The light source is split
into measurement and reference beams. Ocular structures
reflect the measurement beam and interrelate with the Figure 4 -11. O CT yields a cross sectional image of the
reference light reflected from the reference mirror, creat- cornea.
ing interference. The coherent or positive interference, an
increased resulting signal, is measured by the interferom-
eter and allows determination of the reflecting structure’s
position.
A cross-sectional image of the anterior segment struc-
tures anterior to the lens is produced (Figure 4-11). It is
also more functional than confocal microscopy because
it can provide images of the entire anterior portion of the
eye quickly, including pachymetry maps (Figure 4-12).
Additional advantages include it being a noncontact test and
that it uses no coupling medium.
The use of light as an analysis medium also has disad-
vantages. The depth of penetration is limited to the anterior
segment, and image quality may be greatly reduced in cases
of scarring or pannus. Initial clinical use of anterior seg- Figure 4-12. O CT pachymetry mapping.
ment OCT was mainly focused on measurement of phakic
IOL placement, angle analysis in glaucoma, and flap detec-
tion after LASIK.21,22
 To p o gra p h ic Te c h n o lo gie s 41

C ONCLUSION 10. Belin MW, Cambier JL, Nabors JR, Ratliff CD. PAR Corneal
Topography System (PAR CTS): the clinical application of close-
range photogrammetry. Optom Vis Sci. 1995;72(11):828-837.
11. Belin MW, Zloty P. Accuracy of the PAR corneal topography sys-
Understanding topographic technology and how each tem with spatial misalignment. CLAO J. 1993;19(1):64-68.
system derives the maps is important for clinical interpreta- 12. Naufal SC, Hess JS, Freidlander MH¸Granet NS. Rasterstereography-
tion. Placido disk is the most widely used and understood by based classification of normal corneas. J Cataract Refract Surg.
1997;23:222-230.
clinicians. Tomographers are gaining popularity and offer 13. Priest D, Munger R. Comparative study of the elevation topography
views of both surfaces as well as optical pachymetry, attri- of complex shapes. J Cataract Refract Surg. 1998;24(6):741-750.
butes appreciated by both refractive and cataract surgeons. 14. Tanabe T, Oshika T, Yomidokor A, et al. Standardized color-coded
scales for anterior and posterior elevation mapping of scanning slit
corneal topography. Ophthalmology. 2002;107(7):1298-1302.
15. Modis L Jr, Langenbucher A, Seitz B. Evaluation of normal corneas
R EFERENCES using the scanning-slit topography/pachymetry system. Cornea.
2004;23(7):689-694.
16. Giraldez Fernandez MJ, Diaz Rey A, Cervino A, Yebra-Pimentel
1. Corneal topography. Ophthalmology. 1999;106(8):1628-1638. E. A comparison of two pachymetric systems: slit-scanning and
2. Binder PS. Videokeratography. CLAO J. 1995;21(2):133-144. ultrasonic. CLAO J. 2002;28(4):221-223.
3. Mandell RB. The enigma of the corneal contour. CLAO J. 17. Cho P, Cheung SW. Repeatability of corneal thickness measure-
1992;18:267-273. ments made by a scanning slit topography system. Ophthalmic
4. Koch DD, Foulks GN, Moran CT, Wakil JS. The Corneal EyeSys Physiol Opt. 2002;22(6):505-510.
System: accuracy analysis and reproducibility of the first generation 18. Kawana K, Tokunaga T, Miyata K, Okamoto F, Kiuchi T, Oshika T.
prototype. Refract Corneal Surg. 1989;5:424-429. Comparison of corneal thickness measurements using Orbscan II,
5. Mejia-Barbosa Y, Malacara-Hernandez D. A review of methods for non-contact specular microscopy, and ultrasonic pachymetry in eyes
measuring corneal topography. Opt Vis Sci. 2001;78:240-253. after laser in situ keratomileusis. Br J Ophthalmol. 2004;88(4):466-
6. Litoff D, Belin MW, Wynn SS, Smith RS. PAR technology cor- 468.
neal topography system. Invest Ophthalmol Vis Sci. 1991;32(4 19. Reinstein DZ, Silverman RH, Trokel SL, Coleman DJ. Corneal
suppl):922S. pachymetric topography. Ophthalmology. 1994;101(3):432-438.
7. Arffa RC, Warnicki JW, Rehkopf PG. Corneal topography using 20. Reinstein DZ, Silverman RH, Raevsky T, et al. Arc-scanning very
rasterstereography. Refract Corneal Surg. 1989;5(6):414-417. high-frequency digital ultrasound for 3D pachymetric mapping of
8. Nemeth J, Erdelyi B, Csakany B. Corneal topography changes the corneal epithelium and stroma in laser in situ keratomileusis.
after a 15 second pause in blinking. J Cataract Refract Surg. J Refract Surg. 2000;16(4):414-430.
2001;27:589-592. 21. Hirano K, Ito Y, Suzuki T, Kojima T, Kachi S, Miyake Y. Optical
9. Belin MW, Litoff D, Strods SJ, Winn SS, Smith RS. The PAR coherence tomography for the noninvasive evaluation of the cornea.
Technology Corneal Topography System. Refract Corneal Surg. Cornea. 2001;20(3):281-289.
1992;8(1):88-96. 22. Belin M, Holladay J, Michelson M, Woodhams JT, Ahmed I. The
Pentacam: precision, confidence, results and accurate “Ks.” Insert
to Cataract & Refractive Surgery Today. 2007;55:341-347.
Ch a p te r 5

Placid o
Disk-Base d To p o grap hic
Syste ms

ATLAS 90 0 0 C ORNEAL TOPOGRAPH ER

Dianne Anderson, OD, FAAO

T
he ATLAS 9000 Model (Carl Zeiss Meditec, Jena, photopic (with light) pupil images. With the exception of
Germany) corneal topographer (CT) is a large cone the keyboard and printer, the ATLAS system integrates all
Placido disk system designed to measure corneal hardware components in a single unit, including the image
curvature and produce color-coded maps for the detection acquisition optics, the system computer, and a flat panel
of corneal conditions and pathologies. This system also aids display screen.
in the screening and postoperative management of refrac-
tive procedures such as laser-assisted in situ keratomileusis
(LASIK), photorefractive keratectomy (PRK), and multifo- Im age Acq u isit ion
cal and toric intraocular lenses (IOLs) and in the selection
and fitting of rigid gas-permeable (RGP) contact lenses. The patented Cone-of-Focus technology used by the
The field of view of the camera system is large enough to ATLAS system is the most critical part of the system’s
capture the limbus boundaries of the eye, thus allowing alignment and focusing mechanism.1 The 22 rings of the
white-to-white (limbus-to-limbus) measurements, as well large cone are separated by a smaller Cone-of-Focus at
as the determination of the geometric center of the eye for the location of the 9th ring. This smaller cone protrudes
better contact lens fitting and eye registration. outward and reflects on the eye at a fixed, known distance
The ATLAS 9000 Placido disk-based system projects a from the camera. This reflection appears on the live video
series of 22 concentric infrared (950 nm) light rings onto image as a larger separation between the rings surrounding
the cornea of the eye. An image of the reflected rings is the small cone. As the cone is brought closer to the cornea,
captured with a digital camera. The ATLAS system ana- the cone ring moves closer to the inner ring. As the cone
lyzes thousands of data points in the image to measure the is brought further from the cornea, the cone ring moves
distances between rings and their relationships with each closer to the outer ring. The properly focused image results
other in order to reconstruct the corneal surface into a color- in even separation between the rings surrounding the cone.
coded topography map. The ATLAS system can display the ATLAS’ Cone-of-Focus uses a triangulation method to find
image of the cornea in a variety of ways, including curva- the exact location and distance of the rings. A “triangle”
ture, elevation, and aberrations. The ATLAS system also is formed between the cone, the corneal surface, and the
has the ability to capture both scotopic (without light) and Placido rings.
Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical
43 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 43-102)
© 2012 SLACK Incorporated
44 Ch a p t e r 5

During an examination, the patient’s chin is positioned in

the chin rest assembly in front of the projection head. The
clinician administering the exam aligns the projection head
and, using a joystick, focuses on the patient’s eye and acti-
vates the system. The user presses the joystick or spacebar
to stop data acquisition. The ATLAS SmartCapture Image
Analysis technology captures images at a rate of 15 per sec-
ond before the joystick button/spacebar is pressed,2 which
stops the collection of images. When using this option,
SmartCapture analyzes these images and chooses the 4 best
images for processing on the “exam preview screen,” or the
best acceptable single image on the “exam results screen.”
After you capture an image, the “exam preview screen”
is the default display setting for the ATLAS 9000. Both the
“exam preview screen” and the “exam results screen” allow Figure 5-1. ATLAS Axial Curvature View may be displayed in
you to evaluate image quality, recapture or save images, mm or D.
acquire additional images, edit pupil and limbus contours,
select the doctor and operator for the exam, and add remarks.
While the “exam preview screen” displays the last 4 images
(displayed in chronological order, left to right) in the image
frame buffer (or 4 best images if the SmartCapture option is
used), the “exam results screen” displays only the last image
(or best image if the SmartCapture option is on).

Im age An a lysis: View s

ATLAS supports 13 different types of views of the exam
image: Axial Curvature, Tangential Curvature, Elevation,
Irregularity, Rings Image, Keratometry, Refractive Power,
Mean Curvature, Corneal Wavefront, Image Simulation,
Point Spread Function (PSF), Modulation Transfer Function
(MTF), and PathFinder II Corneal Analysis. Masterfit II Figure 5-2. The Tangential View gives more detail about the
Contact Lens Software is an optional program for RGP topographical characteristics of the cornea compared to the
designing. axial map.

Axial Cu rvat u re View

Tan gen t ial Cu rvat u re View
The Axial Curvature View displays the axial curvature
of the cornea using 23 colors to represent dioptric power or The Tangential Curvature View (Figure 5-2) displays the
radius of curvature in mm (Figure 5-1). The Axial View is cornea as a topographical view but bases the calculation on
actually a calculation of curvature rather than “true” refrac- the local curvature of a given meridian, resulting in more
tive power when displayed as keratometric diopters (D). The detail than the Axial View. The Tangential View bases its
color-coded display visually corresponds to the “flat” and calculation on a different mathematical approach and is also
“steep” axes, well-known to keratometry users. known as instantaneous or local curvature. This algorithm
The Axial View treats every single data point as if the is more accurate at calculating the periphery of the cornea.
whole cornea would be a sphere that matches the local This view may also recognize sharp transitions in power
slope and position of every data point. The distance from more easily and eliminates the “smoothing” appearance of
the point of interest on the corneal surface to the optical the topographical view as seen in the axial or sagittal view.
axis defines the axial radius of curvature. The Axial View This is often useful in following healing trends after surgery
has been established in the field, but it does not represent or in viewing pathologic eyes. The Tangential View is help-
the true local curvature. Axial Views naturally incur more ful for locating the size and position of specific features on
smoothing when measuring the corneal surface toward the the corneal surface, such as the apex of corneal steepening
periphery and may not reflect subtle changes. Because it in keratoconus, boundaries of a treatment zone after corneal
shows a “perfect sphere” as flat, the Axial View is unable to refractive surgery, pterygia, and corneal scarring.3
take into account corneal spherical aberration (SA).
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 45

Figure 5-3. Elevation View displays the difference in height of Figure 5-4. Irregularity View uses nonspherical surfaces as the
the cornea in micrometers (µm) from a reference sphere. reference object rather than the reference sphere used in the
elevation map.

The Tangential View does not force the center of cur- Irregu larit y View
vature of the best-fit sphere to lie along the optical axis.
The location of the center of curvature for each point is The Irregularity View takes the best-fit surface meth-
calculated by a best-fit sphere to the local neighborhood at odology of the Elevation View one step further by using
the point of interest on the corneal surface. The distance nonspherical surfaces as the reference object (Figure 5-4).
from the point of interest to the center of the locally best-fit A best-fit toric ellipsoid surface is subtracted from the
sphere represents the tangential (or instantaneous) curva- corneal elevation data to yield the irregularity that cannot
ture. The Tangential View may be a better representation of be described by a reference ellipsoid that accounts for both
true local curvature. the asphericity and toricity of the cornea. The Irregularity
View shows how different the cornea is from a normal eye,
Elevat ion View representing the elevation difference between the “normal”
best-fit ellipsoid surface and the actual measured corneal
The Elevation View displays the difference in height surface.5
of the cornea in microns (µm) from a reference sphere
(Figure 5-3). ATLAS superimposes a reference sphere onto Rin gs Im age View
the corneal surface, measuring the difference between the
reference and the real corneal surface. Elevation is mea- The Rings Image View or photokeratoscopic image dis-
sured as differences in height from the reference sphere as plays the actual captured image of the eye with the reflected
positive and negative µm. The corneal surface may be above rings (Figure 5-5). In addition to the Rings Image, Scotopic
the reference sphere, resulting in a positive difference and (no light, large pupil) and Photopic (with light, small pupil)
appear red in the Standard Colors, or below the reference images may also be displayed if captured at image acquisi-
sphere, resulting in a negative difference and appear blue in tion. Overlay options include Zoom Rings Image to display
the Standard Colors.4 a zoomed-in rings image to 10 mm; Limbus Contour to
The reference sphere may be automatically computed or display green markings at the location of the limbus contour
created using parameters entered by the user. It is important and to measure the horizontal visible iris diameter (HVID);
to keep in mind that the elevation is measured in µm, not Geometric Center to display a crosshair enclosed by a
D, and that Axial and Elevation Views cannot be directly circle designating the point on the cornea corresponding to
compared. The value of measuring the corneal surface in the center of the limbus contour; and Rings to outline ring
µm is that small elevations or depressions on the corneal edges. Data options in the display include pupil diameter,
surface created by surgery or pathology can be carefully HVID, and photopic and scotopic pupil diameter.
documented, giving a level of detail not possible using the
curvature views. Because the shape of the normal cornea is Keratom et ry View
aspheric (ellipsoidal) rather than spherical, it is challenging The Keratometry View simulates the data obtained
to locate subtle deviations in elevation maps even in normal using a standard keratometer (Figure 5-6). This view pro-
eyes, as well as those with nonpathologic corneal shape vides readings for each semi-meridian in 3 zones: central
changes, such as astigmatism. (0 to 3 mm), midperiphery (3 to 6 mm), and periphery
(6 to 9 mm). The steepest semi-meridians in each zone are
displayed in red D values, and the flattest meridians are
46 Ch a p t e r 5

Figure 5-5. The Photokeratoscoptic Image displays the actual Figure 5-7. Refractive Power View shows the power of the eye
captured image of the eye with its reflected rings. Scotopic (no in its refractive state, measured only in D.
light, large pupil) and Photopic (with light, small pupil) images
may also be displayed. O verlays such as limbus contour may
be used to determine the horizonatal visible iris diameter.

Figure 5-8. The Mean Curvature View illustrates the average

curvature of the 2 principal curvatures for any given point on
the cornea, and contains curvature information along 3 dimen-
Figure 5-6. Keratometry View simulates the data obtained sions of every single data point.
using a standard keratometer. Values for the 3 zones are dis-
played as well.
power and radius for the same exam. Significant differences
can be seen in the periphery and in eyes with unusual or
displayed in blue D values. The data are listed by semi-
abnormal shapes due to the effect of the SA.
meridian and zone on the left of the screen. The simulated
keratometer values at the top left of the display are com-
puted in a similar way to the computations performed by Mean Cu rvat u re View
keratometers commonly used to measure the central 3 mm, The Mean Curvature View is the average curvature of
assuming the principal axes have a 90-degree shift. the 2 principal curvatures for any given point on the cor-
Eccentricity values are also displayed for meridians at nea (Figure 5-8). As the Axial and Tangential Views are
0, 10, 15, 20, 25, and 30 degrees. defined along individual meridians (or slices of the cor-
nea), the Mean View contains curvature information along
Refract ive Power View 3 dimensions of every single data point. The mean curvature
is the average curvature in all directions around the data
The Refractive Power View shows the power of the eye
point, while axial and tangential curvature is limited to
in its refractive state, measured only in D (Figure 5-7).
the semi-meridian. As a result, the Mean View suppresses
Sometimes called a Snell’s Law Map, this view measures
corneal astigmatism and is insensitive to patient fixation.
the refraction of light rays using Snell’s Law as they pass
The advantage of the Mean View is that it may better detect
through the corneal surface. Taking into account SA, the
subtle abnormalities because it has the ability to suppress
Refractive Power View more closely illustrates how light
nonpathological features, such as corneal astigmatism,
rays behave as they strike the cornea. Refractive Power
while preserving pathological features such as keratoconus.
Views may differ from Axial Views in their readings of
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 47

Figure 5-9. Corneal wavefront represents aberrations based Figure 5-10. The Zernike table view displays the names and
upon topography, and described using Zernike Polynomials values of the Zernike coefficients (up to order 4), along with
like total eye aberrometry. horizontal bar graphs of the Zernike coefficients values.

Corn eal Wavefron t View

The Corneal Wavefront View displays corneal wave-
front aberrations (Figure 5-9). The corneal wavefront is
calculated from corneal topography using an algorithm
that incorporates the shape of the cornea using ray tracing,
technology that traces light rays from the fovea out of the
eye. Wavefront aberration is measured as the optical path
difference, the difference between the corneal wavefront
and an ideal wavefront at the entrance pupil.
Total aberration of the eye as measured by wavefront
aberrometers depends on pupil size and accommoda-
tion and generates the refractive error of the eye. Corneal
wavefront aberrations are relatively stable over time, are
not affected by pupil size or accommodation, and do not Figure 5-11. Image Simulation View displays a corneal focal
plane image simulation based upon the corneal PSF.
provide information about the refractive error of the eye.
Using the Wavefront View Editor, corneal wavefront can
be displayed in color, number, profile, surface, or Zernike
table format. The Zernike table view displays the names The PSF shows the distribution of light on the focal plane of
and values of the Zernike coefficients (up to order 4), the anterior cornea, not including the effects of diffraction.
along with horizontal bar graphs of the Zernike coefficient The PSF View can be displayed as an Image (default view),
values (Figure 5-10). Various coefficients can be selected Surface, or Profile view.
or removed. You can also enable or disable compensation
of Stiles-Crawford weighting (directional sensitivity of the Modu lat ion Tran sfer Fu n ct ion View
photoreceptors) in the calculation of the aberrations. The MTF View is a monochrome display representing
the MTF arising from the Corneal Wavefront (Figure 5-13).
Im age Sim u lat ion View The MTF View can be displayed as a Profile (default view),
The Image Simulation View displays a corneal focal Surface, or Image view.
plane image simulation based upon the corneal PSF
(Figure 5-11). Images available for use in the simulation Pat h Fin der II Corn eal An alysis
include a 20/100 “E” image, an eye chart (20/50 to 20/10), PathFinder II is a unique software feature for the ATLAS
and a beach scene image. 9000 System, designed to assist in identifying abnormal and
pathologic corneas based solely on anterior corneal topog-
Poin t Spread Fu n ct ion View raphy.6 It uses statistical parameters derived from anterior
The PSF View is a monochrome display representing corneal topography for its analysis (Figure 5-14). Clinical
the PSF arising from the Corneal Wavefront (Figure 5-12). data make it possible to compare findings in an individual
48 Ch a p t e r 5

(SVM) algorithm, PathFinder II displays the probability

(from 0% to 99%) that the corneal topography matches
each of the categories established in the PathFinder II clini-
cal study database.7 Of the 12 corneal parameters analyzed,
3 evaluate the corneal irregularity measurement (CIM), or
mean apical curvature/toric keratometric mean (TKM),
and shape of the cornea (Shape Factor). The remaining
9 parameters are derived from the Mean Curvature View.
An SVM algorithm performs classification by construct-
ing multidimensional hyperplanes that optimally separate
the clinical (or training) data into specific categories
using the analyzed parameters. SVMs are considered to
be one of the most successful classification algorithms in
the data processing field.8 For each topography exam, the
Figure 5-12. PSF View is a monochrome display representing PathFinder II SVM algorithm assigns a percent probability
the PSF arising from the Corneal Wavefront. of the exam to match the trained categories.
Each corneal parameter used by PathFinder II is col-
or-coded (green = normal, yellow = borderline, red =
abnormal) to indicate the degree to which it matches the
values established for the normal population, based on a
multicenter clinical study. The multiclinical sample data
set covers 95% of the normal eye population with a 95%
confidence. The statistical parameters (CIM, TKM, and
Shape Factor) are displayed in bar plot format with color-
coding shown in the bar plot itself. The 9 other parameters,
based on the Mean Curvature View, are displayed in the
Mean Curvature View Data Box, and the values themselves
are color coded.
PathFinder II analyzes 12 different corneal parameters
for a particular exam and compares them to a comprehen-
sive clinical database that contains parameter reference val-
Figure 5-13. MTF View is a monochrome display representing ues for 5 categories of conditions and pathologies: normal,
the MTF based on the Corneal Wavefront. keratoconus pattern, myopic laser vision correction (LVC),
hyperopic LVC, and other.
The normal cornea is typically aspherical in shape with
TKM, CIM, and Shape Factor in the normal range. No
abnormal or asymmetric topography patterns, such as infe-
rior steepening, abnormal elevation, or curvature, can be
seen in the Mean Curvature View. PathFinder II’s clinical
database for the normal category includes normal corneas
with no corneal pathology and no history of refractive sur-
gery or other ocular surgery. These corneas were screened
with other modalities, such as corneal pachymetry, and
were determined to be good candidates for LVC.
PathFinder II clinical database for the keratoconus pat-
tern category includes established keratoconus, suspect
keratoconus (subclinical/forme fruste keratoconus), and
pellucid marginal degeneration. Keratoconus is defined as
the true pathological condition, which causes thinning and
Figure 5-14. PathFinder II Analysis assists in the identifica- wrinkling of the cornea, along with a cone-like protrusion
tion of abnormal and pathologic corneas based on statistical of the cornea in its later stages. Keratoconus can also pro-
parameters derived from anterior corneal topography.
duce high regular astigmatism, irregular astigmatism, and
topographical patterns with inferior steepening.
PathFinder II’s clinical database for the myopic LVC
patient to expected findings from different populations of
category includes corneas that have undergone myopic LVC
individuals with eyes that have various known conditions
(LASIK, PRK, or laser-assisted subepithelial keratectomy
or pathologies. Using a trained Support Vector Machine
[LASEK]).
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 49

Figure 5-15. The Trend with Time Display

shows the currently active exam and up
to 3 previous exams for the same eye in
chronological order.

PathFinder II’s clinical database for the hyperopic LVC • The 2-Difference Display shows 2 exams for the
category include corneas that have undergone hyperopic same eye and illustrates their difference. The top
LVC (LASIK, PRK, or LASEK). exam (Exam 1) is subtracted from the bottom exam
PathFinder II’s clinical database for the “other” category (Exam 2).
include corneal diseases, pathologies, or surgeries that can • The 3-Difference 5-Map Display provides informa-
potentially affect the corneal surface, such as postoperative tion regarding changes between multiple exams, help-
corneal transplant, intracorneal ring segments, radial kera- ful in monitoring healing following corneal surgery.
totomy, post-LASIK ectasia, corneal scars, etc. It is also useful to follow contact lens-induced change
The PathFinder II display incorporates 4 views. The over time to determine the effect that the lens is hav-
Probability View displays the previous 5 categories with ing on the cornea. Three exams for the same eye are
blue color bars representing the 2 highest probabilities. The selected, and 2 differences are calculated. Exams 1,
Parameters View consists of bar graphs of the 3 statisti- 2, and 3 are shown, respectively, left to right on the
cal data parameters: CIM, TKM, and Shape Factor. The top of the display. Exam 1 will be subtracted from
User-Selectable View default is the Axial Curvature View Exam 2, and a difference map will be calculated and
and is customizable. The Mean Curvature View, a special displayed on the lower left. Exam 2 will be subtracted
mean curvature view with 9 additional parameters, cannot from Exam 3, and a difference map will be calculated
be customized. and displayed on the lower right.
• The Trend with Time Display shows the currently
active exam and up to 3 previous exams for the same
Im age An a lysis: Disp lays eye (Figure 5-15). The exams are always displayed in
The ATLAS software lets you select 8 different display chronological order (left to right, top to bottom).
types for the exam data: • The Trend Analysis Display shows up to 4 exams
• The Single View Display shows a single view of a for the same eye, along with a table of data analysis
single exam for the selected patient. parameters (Figure 5-16). By default, only the current
day’s exams for the same eye are displayed, if pos-
• The Overview Display shows 4 different views of the
sible. However, you can select any 4 exams to include
same exam for the selected patient.
in the Trend Analysis Display. The exams are always
• The OD/OS Compare Display shows 2 different displayed in chronological order (left to right). The
views of left and right eye exams for the same patient Trend Analysis Data Parameters Table displays api-
on the same display. Two views on the left side of cal radius (Ro), sagittal height (for a specified chord
the display are for the right eye (OD) exam, and the length); eccentricity along either the flat, steep, or
same views for the left eye (OS) are on the right side. custom meridian; flat and steep Sim Ks; HVID; and
ATLAS automatically selects the exam of the other pupil diameter values for each image. The mean and
eye based on the date of exam, but manipulation of standard deviation values for these parameters are
the exam is easily performed. also displayed for the images.
50 Ch a p t e r 5

Figure 5-16. The Trend

Analysis Display shows
up to 4 exams for the
same eye, along with
descriptive parameters.

• The Custom Display can show 4 different views

for the selected patient. Any combination of exams
and views can be selected. This differs from the
Overview Display, where all views are for the same
exam. You can select the exams and views for the
Custom Display. ATLAS has 6 default parameter sets:
Pathology, Cataracts, Contact Lens Fitting, Refractive
Surgery, Screening, and Screening (Expanded).
Default displays can be customized within each of
these parameter sets.

Ma sterFit II
Con t act Len s Soft w a re
The optional MasterFit II Contact Lens Software enables
the user to design RGP contact lenses and to simulate the Figure 5-17. MasterFit II Fitting Rules are set to these default
resulting fluorescein pattern on the cornea. MasterFit II values. Users can reset these values to personal preferences.
features both topographic- and keratometric-based lens fit-
ting methods.9 Based on customizable fitting preferences,
MasterFit II will recommend the most appropriate lens
type, including spherical, aspheric, front toric, back toric,
and bitoric designs, and designs a custom lens that best
achieves the desired fitting preferences (Figure 5-17).
MasterFit II also features region-specific databases of
stock (finished) lenses that may be used as an alternative to
creating custom lens designs.
The Fluorescein Map is a green color map of simulated
fluorescein tear film layer thickness (between the back
surface of the contact lens and the cornea) for the current
selected lens design shown over the ring image of the eye
(Figure 5-18). The shades of green correspond to the values Figure 5-18. MasterFit II Simulated NaFl Pattern Display shows
the recommended RGP lens based on the chosen fitting rules.
(µm) shown on the scale at the left of the screen—darker
Lens design thumbnails on the left allow the user to add
green areas have small tear film thicknesses, while brighter and edit up to 5 lens designs to the recommended design.
greens areas have larger tear film thicknesses. The software Horizontal and vertical tear film thickness graphs appear
will initially place the lens at the geometric center of the below and to the right of the simulated image, respectively.
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 51

cornea. You can manually move the lens by clicking inside • There should be midperipheral clearance of approxi-
it and dragging up-down and/or left-right. mately 40 µm, allowing for unobstructed movement
MasterFit II’s topography-based fitting optimizes the along the vertical meridian (Sim K axis closest to
tear film clearance. The topographic fitting algorithm 90 degrees).
attempts to create a contact lens design that will provide The user may create multiple lens designs per topography
the selected tear lens thickness using the methodology exam to simulate the effect of adjusting the available lens
described in Figure 5-17: parameters on the fluorescein pattern. After finalizing the
• The center of the lens should clear the central cornea lens design, the user may print the lens design using the order
by approximately 20 µm. form templates provided, or the order form may be saved as a
• There should be a midperipheral clearance of approxi- PDF and sent by e-mail to the preferred laboratory.
mately 8 µm along the horizontal meridian (Sim K
axis closest to 180 degrees).

TH E TOMEY TMS C ORNEAL TOPOGRAPH ER

Stephen D. Klyce, PhD and Bradford L. Tannen, MD

T
he analysis of corneal topography is now the stan- in the measured surface become much greater than a
dard of care in anterior segment practices. The few wavelengths of the measuring light wavelength, the
first CT commercially available was the Computed fringes produced will merge and become indistinguish-
Anatomy, Inc (New York, NY) Corneal Modeling System able. In essence, interferometry is too sensitive a mea-
(CMS) in 1987.10,11 It incorporated many of the Louisiana surement technique to apply to the imperfect optics of
State University Eye Center research laboratory findings in the eye. Profilometry is the technique used by slit-based
its implementation.12 Of greatest impact was the incorpora- topographers and fluorescein-stained tear film techniques
tion of the color-coded display of corneal surface curva- (rasterstereography). The early version Orbscan I used
ture.13 This device had limited commercial success owing anterior corneal surface data obtained with a scanning
to its cost but was remarkable in that it measured both ante- slit to estimate corneal topography, while the Pentacam
rior corneal curvature as well as corneal thickness profiles, (OCULUS, Wetzlar, DE) uses a rotating slit to obtain these
capabilities subsequently available in the Bausch & Lomb data. However, direct measurement of the corneal profile
(Rochester, NY) model Orbscan II. Computed Anatomy does not provide sufficient resolution for accurate depiction
was acquired by the Tomey Corporation (Nagoya, Japan) in of corneal topography. The Placido reflective technology
the early 1990s following the introduction of the Computed appears to have the requisite sensitivity of measurement
Anatomy Topography Modeling System (TMS-1). This and is employed by all of the commercially successful CTs,
model became the earliest “work horse” CT, supplanted including the upgraded Orbscan IIz. Developing technolo-
eventually by subsequent Tomey models. At this writing, gies like high-frequency ultrasound and high-speed anterior
the Tomey TMS-4a (“TMS-4”) is commercially available segment optical coherence tomography continue to evolve,
with software version 5.2D. This section will review the but at present have limited utility in clinical practice in
features of the TMS-4 as well as the capabilities of the measuring corneal topography.14,15
software. It should be noted that not every capability of the There are 2 basic approaches used with the Placido
TMS-4 software can be covered here; priority is given to disks. Some use a large faceplate with a large working dis-
those of most frequent clinical use. tance to project images of mires onto the corneal surface.
This approach has the advantage of less critical focus and
the disadvantage of corneal shadows created by the brow
Gen era l Ch a racter ist ics and the nose of the patient, which can obscure portions of
the peripheral mire images. Other CTs use a small cone-
Several methodologies have been used to measure cor- shaped Placido disk target with a short working distance.
neal curvature, including interferometry, profilometry, and The advantage of this arrangement is a greater potential
Placido disk reflective methods. While interferometry has coverage of the corneal surface, with the disadvantage of
the greatest potential sensitivity of measurement, interfer- being more sensitive to error in focus. Both types have seen
ence produces fringes according to the deviation of a mea- commercial success. All of the Tomey models have used the
sured surface from a reference surface. If the distortions small cone-type Placido disk.
52 Ch a p t e r 5

Top ograp h ic Disp lays

CTs project a Placido pattern onto the corneal surface,
capture this with a digital camera, determine the precise
location of each of the mires, and convert these positions
into curvature at each measurement point. The curvature
data are expressed in a number of ways; mm are often used
for contact lens applications, while the units of D are most
often used for clinical diagnostics. The curvature data avail-
able from CTs are generally conditioned with the kerato-
metric index (1.3375), such that the values presented include
the dioptric power of the tear film/air interface (about 48 D)
and the dioptric power of the endothelial/aqueous humor
interface (about -5 D). This makes dioptric power values
equivalent to total corneal power and conveniently corre- Figure 5-19. The Tomey TMS-4 single standard power map
using the Klyce /Wilson scale. This is a cornea from a normal
spondent to keratometry.
eye. This map conveys all the characteristics of a normal cornea:
The scales available on the TMS-4 include the absolute uniform central powers tapering toward the limbus, relatively
scale, the Klyce/Wilson scale, the Maguire/Waring scale, smooth contours, and K readings near 43 D. An outline of the
a user-adjustable scale, and a normalized scale. Normal pupil is shown with its location and diameter noted in the lower
clinical use of the TMS-4 involves use of one of the left-hand corner. The central cross indicates the position of the
1.5-D contour interval scales (absolute or Klyce/Wilson), vertex normal at the alignment axis of the TMS-4. The central
which show the relevant topographic features without over- small square represents the position of the pupil center. Corneal
emphasizing details.16 The scales are implemented with a statistics are displayed in the lower panel. Sim K1 is labeled Ks
color palette that consists of contrasting colors that make and Sim K2 is labeled Kf in this version of the software.
the contour boundaries obvious so that corneal topographic
irregularities are not masked. This strategy of a combina-
tion of a fixed range, a fixed interval, and a well-selected
color palette is essential for proper interpretation of corneal
topography. Further emphasis on the importance of proper
topographic scale choice is presented elsewhere in this
book.
Curvature data are commonly calculated with 3 differ-
ent approaches. The standard map with the TMS-4 displays
corneal surface dioptric power with a spherical approxima-
tion method commonly referred to as axial power (Figures
5-19 and 5-20A). This is the baseline map that is ordinar-
ily used in routine clinical screening. The strength of this
method for calculating and displaying corneal topography
is that it removes the corneal positive SA component from Figure 5-20. The TMS-4 multimap display is used here to illus-
trate the appearance of a single exam (cornea after PRK) with
the power calculation. The remaining curvature displayed
the 3 different power maps and the height map. (A) The stan-
retains more of the shape characteristics seen in the normal
dard power display. (B) The refractive power map. Note the
cornea: steeper (higher power) in the central region and central powers are the same as A, but allowing for SA steepens
flatter (lower power) in the periphery. the peripheral powers. (C) The instantaneous map shows the
A second display of corneal curvature available on the “red ring” in the transition zone, which modern algorithms
TMS-4 is the refractive power map (Figure 5-20B). This have overcome for the most part. Note the additional noise
method for calculating corneal power uses Snell’s law, compared to A. (D) Without amplification, the true height map
which is necessary for accurate ray tracing in order to is unable to show useful detail of the shape changes respon-
evaluate the corneal contribution to the aberrations mea- sible for the power maps.
sured with wavefront sensors. Note that the negative SA of
the natural lens (as well as the modern aspheric IOLs) com-
pensates to some extent for the positive SA of the cornea. instantaneous map contains significant noise artifact that
A third display of corneal curvature available on the may mislead the user to assume that a cornea has an abnor-
TMS-4 is the instantaneous map (Figure 5-20C). This mally high amount of higher-order aberrations (HOAs).
label is shortened from instantaneous radius of curvature, However, the instantaneous map increases the detail of
which offers the most detailed look at local changes in the curvature changes, particularly in the corneal periphery,
corneal curvature. Because of the method of calculation, the and may be the only means to evaluate the peripheral
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 53

optics of the cornea after myopic refractive surgery (MRS).

Generally, amplifying corneal power distribution beyond
that shown with the standard axial power map shows details
that are not of consequence to vision or to diagnostics.
However, the instantaneous map clearly shows the transi-
tion zone characteristics after refractive surgery, and these
may be responsible for night vision complaints after refrac-
tive surgery (Figure 5-20C).
The final map type available on the TMS-4 is the height
map (Figure 5-20D). The raw height map is of itself not
useful diagnostically except in extremely aberrated cor-
neas. The main component of the height data is a sphere,
and this masks the small curvature changes that are clearly
shown with the power maps. However, the TMS-4 offers
2 additional displays that allow inspection of distortions in Figure 5-21. The TMS-4 power difference map is useful for
true cornea shape. The enhanced height map is created by examining the actual effect of a refractive surgical procedure
on the corneal topography. (A) The preoperative cornea.
subtracting an average sphere that fits the raw height data.
(B) Postoperative result. (C) The difference between pre- and
The enhanced height map produces an aberration structure,
postoperative examinations.
which bears a similarity to wavefront displays as both are
in units of µm. A second routine, the height difference map,
displays the difference in height data between 2 corneal
examinations. This allows the change in shape induced by
refractive surgery to be displayed, which can be useful to
compare planned versus achieved shape changes.
A difference map is also available to show changes in
corneal power with time or condition. As with the height
change map, the power difference map permits the evalu-
ation of changes in corneal power induced by refractive
surgery, contact lens wear, or other condition (Figure 5-21).
Each of the maps available on the TMS-4 embodies
a number of user-selectable features. The cursor can be
moved around on a map to view the radius and power at any
measured point. The same routine can be used to estimate
the distance and difference in power between any 2 points. Figure 5-22. TMS-4 statistics display. (A) Raw image of the
This same ruler function can be used to obtain a calibrated eye with the mires overlaid with the green rings found by the
“white-to-white” measurement. The TMS-4 measures the computer is useful for verifying the accuracy of the processing.
margins of the pupil in the video image it captures, and (B) The standard map display of corneal topography. The right
the pupil outline can be superimposed on any of the maps, half of the figure displays the statistics calculated for the exami-
nation. Indices that are green in color indicate values within
as noted in Figure 5-22. Astigmatism is calculated in a
±2 standard deviations of the value found in normals. Those
number of ways: orthogonal, instantaneous, and zonal.
yellow in color indicate suspect values that lie between 2 and
Orthogonal astigmatism is similar to standard keratometry. 3 standard deviations from the normal. Those red in color rep-
Instantaneous astigmatism gives an indication of corneal resent abnormal values more than 3 standard deviations from
irregularity, while zonal astigmatism is useful for plan- the normal. The cutoff values are indicted for every statistic in
ning astigmatic surgery and for contact lens applications. the scale at the bottom of the panel. These change appropri-
Other features that are user selectable include the presence ately for each statistic by moving the mouse cursor over one
or absence of the eye image, a routine for verifying the of the statistics. The definition for each statistic is obtained in
accuracy of the mire tracking, a slide-making utility, exam the same manner.
backup, import and export, and several grid overlays for
measurement.
By itself, the color-coded contour map is useful for automatic discrimination software described below. Each
classifying corneal topography, but it does not lend itself of the indices was evaluated for a group of normal corneal
directly to quantitative evaluation of specific values of topographies to determine the mean and standard deviation
clinical interest. A number of corneal topographic indices for the normal population. To assist the user in distinguish-
are available on the TMS-4; some, such as the simulated ing among normal, potentially abnormal (suspect), and
keratometry data, were designed for direct clinical use. clearly abnormal corneal topography, each index is color
Other indices were designed primarily for use in the coded. A green-colored index indicates a value that is
54 Ch a p t e r 5

within 2 standard deviations from the normal average. A overall global shape factor. The newest software version
yellow-colored index indicates a caution (ie, that an index also permits the user to obtain eccentricity values for each
is 2 to 3 standard deviations from the normal average). A meridian on the corneal surface. A positive (normal) value
red-colored index indicates a value that is more than 3 stan- is obtained for a prolate surface, a nil value for a sphere,
dard deviations away from the normal average. Some of the and a negative value is used to indicate an oblate surface.
statistical indices can be displayed on the standard maps as Higher-than-normal values are found with keratoconus, and
in Figure 5-19. More extensive statistics are found on the negative values are often found with symptomatic contact
statistics display (see Figures 5-20 through 5-22). lens wear and MRS corrections.
The first indices enumerated for the TMS-4 and now The standard deviation of corneal power (SDP) is
standard on all CTs are those associated with simulated calculated from the distribution of all corneal powers in
keratometry (SimK).17,18 SimK1 gives the dioptric power an examination.21 The coefficient of variation of corneal
and associated angle of the principal meridian, SimK2 power (CVP) is calculated from SDP divided by the grand
gives the dioptric power and associated angle of the merid- average of corneal powers. This fundamental statistic is
ian orthogonal to the principal meridian, and MinK gives high when there is a broad range of powers in the corneal
the dioptric power and associated angle of the meridian surface and has been found to be a good measure of corneal
with the lowest overall dioptric power. The simulated kera- multifocality. High values of CVP are found in moderate to
tometric cylinder of the corneal surface (Cyl) is obtained severe keratoconus corneas, as well as during corneal trans-
from the SimK readings. plants in the early postoperative period. Manifest refraction
The surface asymmetry index (SAI) measures the of an eye with high CVP will be difficult to achieve, but
difference in corneal powers at every ring (180 degrees attention to refraction is important in such a patient to attain
apart) over the entire corneal surface.19,20 The SAI is often spectacle tolerance. The CVP value given has been scaled
higher than normal in keratoconus, PKP, decentered MRS up by a factor of 1000.
procedures, trauma, and contact lens warpage. Adequate The irregular astigmatism index (IAI) is an area-com-
spectacle correction is often not achieved when SAI is high. pensated average summation of inter-ring power variations
The surface regularity index (SRI) is a correlate to potential along every meridian for the entire corneal surface ana-
visual acuity (PVA) and is a measure of local fluctuations lyzed.22 The IAI increases as local irregular astigmatism
in central corneal power.17,18 When SRI is elevated, the in the corneal surface increases. IAI is high in corneal
corneal surface ahead of the entrance pupil will be irregu- transplants shortly after surgery; persistence often heralds
lar, leading to a reduction in best spectacle-corrected visual suboptimal best spectacle-corrected vision. The analyzed
acuity. High SRI values are found with dry eyes, contact area (AA) gives the fraction of the corneal area covered by
lens wear, trauma, and PKP. the mires that could be processed by the TMS-4 software.
The PVA is derived from a clinical correlation18 of SRI AA is lower than normal for corneas with gross, irregular
versus the best spectacle-corrected visual acuity. The PVA astigmatism, which causes the mires to break up and not
is given as the range of best spectacle-corrected Snellen be resolved. A lower-than-normal AA is found with early
visual acuity that might be expected from a functionally postoperative corneal transplants, advanced keratoconus,
normal eye with the topographical characteristics of the and trauma. AA can also be artificially low during a squint
analyzed cornea. Diagnostic evaluation should consider or when the eyes are not opened wide.
the fact that tear film breakup can greatly influence PVA The elevation/depression diameter (EDD) is the equiva-
(and SRI). Prolonged gazing at a fixation target by a patient lent diameter of the area found to contain powers within
without blinking can produce tear film breakup, transiently the pupil 1 D or more from the mode. It is calculated from
reduced vision, and abnormal values of PVA and SRI. twice the square root of this area divided by pi. The units
With proper blinking, abnormal values of PVA are associ- are millimeters. The elevation/depression power (EDP)
ated with true irregular corneal astigmatism as is often calculates the average power of apparent islands (or pen-
observed with keratoconjunctivitis sicca, contact lens warp- insulas) and valleys for those areas of the cornea that are
age, lamellar keratoplasty, and herpes keratitis. SRI and within the demarcated pupil. If the pupil is not available
PVA are extremely valuable diagnostically to differentiate for a given exam, EDP is calculated from an area 4 mm
visual deficit; if an eye exhibits good corneal potential but in diameter centered on the CT axis. Together with EDD,
suffers visual loss when best-corrected, the deficit will be EDP can be used to estimate the size of so-called central
associated with compromised internal optics, retinal dis- islands after excimer laser sculpting. Any power within the
ease, or neural deficit. pupil that is 1 D or more beyond the mode (most frequently
The average corneal power (ACP) is an area-cor- occurring power) is multiplied by the cornea local area it
rected average of the corneal power ahead of the entrance represents (area compensation); this total is then divided
pupil.19,20 This value is generally a more accurate mea- by the total area of the summed powers. The units are D.
surement to use for IOL calculations than keratometry Normal corneas with high cylinder, corneal grafts, and
values, particularly in postsurgical corneas. The corneal clinical keratoconus will also exhibit degrees of abnormal
eccentricity index is a measure of corneal eccentricity, an EDP and EDD.
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 55

Modern corneal topography algorithms and devices offer

reliable and repeatable measurements of surface curvature.
In addition to the basic features described above, the TMS-
4 offers several advanced features that take advantage of the
consistent examinations that can be obtained.

Screen in g Soft w a re
The presence of keratoconus or suspect keratoconus pat-
tern in corneal topography is a contraindication for standard
refractive surgery as it puts the cornea at risk for developing
keratectasia. This is a particular concern for LASIK, where
keratectasia can occur 6 to 18 months or more after surgery Figure 5-23. The TMS-4 keratoconus screening tool analyzing
in corneas exhibiting suspect keratoconus topography23 as a cornea with moderate keratoconus. The statistical indices
well as for PRK, where keratectasia can occur 4 to 5 years presented are fully described in the text.
after surgery in such corneas (personal communication, R.
Zaldivar, 2005). Corneal topography is the most sensitive
method for detecting suspect keratoconus24,25 and, because
of this, often there are no other signs, although careful reti-
noscopy may reveal a slight scissoring of the light reflex in
these cases. Suspect keratoconus topographic patterns often
present as a subtle asymmetry in corneal power. Because
all corneas exhibit some degree of asymmetry, in order to
differentiate between these normal variations and changes
consistent with suspect keratoconus, a quantitative analysis
is indicated.24,25 The TMS-4 has 2 keratoconus screening
programs (Figure 5-23). The Klyce/Maeda method22 uses
discriminant analysis and a decision tree to report the simi-
larity in percent of a corneal topography to clinical kerato-
conus. The Smolek/Klyce26 method extends this approach
Figure 5-24. The TMS-4 keratoconus screening tool analyz-
using neural networks to assign a similarity value to corneas ing a cornea that is suspect keratoconus. Note that the Klyce /
with clinical keratoconus and gives a severity index of the Maeda method reports the absence of clinical keratoconus,
pathology. In addition, the latter method was also trained on while the Smolek/Klyce method reports the presence of sus-
suspect keratoconus topographies (Figure 5-24). A number pect keratoconus. This is not a contradiction: the Klyce /Maeda
of complementary corneal topographic indices were devel- method only recognizes clinical keratoconus; suspect kerato-
oped to aid in the discrimination of keratoconus from other conus was not part of its training.
corneal pathologies in order to increase the specificity of
the tests.22 The opposite sector index represents the maxi- report the absence of keratoconus, with keratoconus indices
mum difference between average corneal powers between colored green. Therefore, it is important that these utilities
any 2 opposite sectors. The differential sector index rep- not be used as general screening programs for refractive
resents the maximum difference between average corneal surgery because abnormal corneal conditions other than
powers between any 2 sectors. The center/surround index is keratoconus are also contraindications for conventional
the difference in average corneal power between the central refractive surgical techniques.
3-mm diameter of an analyzed area and an annulus sur- Hjortdal and co-workers27 were the first to show the util-
rounding this central area from an inner radius of 1.5 mm to ity of Fourier decomposition in corneal topography analy-
an outer radius of 3 mm. This index is used to capture cen- sis. They demonstrated that the average power, cylinder, and
tral keratoconus. These and other indices already described irregular astigmatism could be extracted from each mire.
are used in the calculation of the keratoconus index (KCI) The TMS-4 implements a Fourier decomposition display
and the keratoconus severity index. (Figure 5-25) along the lines suggested by these authors.
Here is one caveat in the use of these screening utilities: With this implementation, the power values along each of
although the training sets of topography examinations con- the mires are fit sequentially with a 1-dimensional Fourier
tained a number of different types of corneal topographic Series analysis, which provides a radial micro-zonal analy-
anomalies in addition to normal corneas and those with sis. This routine decomposes corneal power data into zonal
keratoconus, they were trained to detect only keratoconus. spherical equivalent, zonal cylinder, zonal asymmetry, and
Hence, presented with a clear example of topographic pellu- zonal higher-order irregularity. In addition, each of these
cid marginal degeneration, the screening utilities will each variables is also given for 3- and 6-mm zones. These data
56 Ch a p t e r 5

Figure 5-26. The TMS-4 has an extensive contact lens program

with a large built-in library of commercial contact lenses as well
Figure 5-25. The TMS-4 Fourier decomposition display. Here,
as customizable one-off routines. The map simulates the clini-
a keratoconus cornea is decomposed into spherical, cylindri-
cal fluorescein test and is useful for inspecting the lens clear-
cal, asymmetrical, and higher-order irregularity components.
ance. It updates if base curve or other parameters are adjusted.
Because the analysis is done on a ring-by-ring basis, the cylin-
der component is not radially linear.

and its descendants, including scales and color palettes that

can be useful for the analysis of the quality of refractive promote rapid and accurate interpretation, display types
outcomes for different approaches. The approach has been that allow the clinician to examine topography in greater
used to evaluate corneal topographic effects of cataract sur- detail and to appreciate changes over time, topographic
gery,28 irregular astigmatism and contrast sensitivity after statistical indices that provide quantitative assessment
PRK,29 and keratoconus progression30 ; to establish norma- of corneal shape factors and corneal optical quality, and
tive data for normal corneas versus those with pathology screening programs that differentiate suspect keratoconus
or surgery31; to examine the effects of overnight wear of and clinical keratoconus from normal individual variations
contact lenses worn for orthokeratology32 ; and to evaluate in corneal shape. Corneal topography analysis has made
visual acuity after PKP.33 huge strides from the early 1980s, when keratometry and
The TMS-4 has contact lens fitting software that has keratoscopy were state of the art. Without this essential
evolved from the TMS-1 program, one of the first such tool, which has become the standard of care in anterior
utilities available on a CT. Using elevation and position data segment practice, there is no doubt that refractive surgery
allows a 3-dimensional representation of the cornea to be would have remained underdeveloped and would not have
analyzed for a contact lens fit using several rules similar to achieved the enormous success it has today. It was the CT
those used by contact lens fitters. Eccentricity and K read- that taught us that small functional optical zones, central
ings are available from the corneal statistics, and these are islands and peninsulas, and decentrations are clear detrac-
used to select an initial contact lens fitted to the corneal tors of visual quality. It was the CT that detected suspect
surface from those lens characteristics that are stored on the and early keratoconus to classify these patients as ineligible
TMS-4. The contact lens is then mathematically lowered for standard refractive surgical procedures. These are just a
onto the corneal surface by properly controlling the lens tip few of the many advancements that define the critical role
and tilt. This allows the clearance between the contact lens the TMS CT has played in the development of corneal shape
and the corneal surface to be calculated, and the amount of analysis, a science that has provided both the clinician and
clearance is indicated using a simulated fluorescein exam the refractive surgery patient with better information, safer
(Figure 5-26). If necessary or desired, the operator can then surgery, and ultimately clearer vision for the past 20 years
make adjustments to the lens design, such as base curve and and will continue to do so in the years to come.
diameter while observing the effect on the fit. This program
has been used successfully to obtain good fits with RGP
contact lenses for both normal and irregular corneas.34,35 Ack n ow led gm en t s
This work was supported in part by an unrestricted grant
Con clu sion to the Department of Ophthalmology, Mount Sinai School
of Medicine from Research to Prevent Blindness Inc, New
The TMS-4 is the latest Placido disk-based Tomey CT, York, NY. The authors are indebted to Marguerite B.
whose predecessors included the very first clinical device McDonald, MD, Ophthalmic Consultants of Long Island,
capable of producing a data-rich map of corneal shape, the New York, for providing the patient examinations used to
Computed Anatomy CMS. Nearly all of the advances in illustrate this chapter.
corneal topography were made with the CMS topographer
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 57

TH E M AGELLAN M APPER
Michael J. Endl, MD and Claus M. Fichte, MD

O
ver the past decade, the number of corneal refrac-
tive procedures has reached an all-time high.
Increasingly, eye-care providers have become aware
of potential postoperative complications. One of the most
feared outcomes following subtle laser procedures is the
development of progressive corneal ectasia. If the physician
is to fulfill his or her primary goal of “do no harm,” then
prevention is the key.
Although various authors have published risk criteria as
a guide to practitioners,36 preoperative screening for corne-
al ectasia presently has no standardized detection methods.
Clinicians currently rely on keratometric, slit lamp,
and standard topographic “red flags” when evaluating the
candidacy of a keratorefractive patient. Some CTs—like
the Magellan Mapper from Nidek (Fremont, CA)—provide
refractive surgeons with a helpful tool to avoid unwanted
complications.
The Magellan features new software that includes a neu-
ral network application capable of predicting various cor-
neal diseases and postsurgical outcomes. Based on corneal
statistics derived from topographic data, the software clas-
sifies and predicts the probability of several categories in an
easy-to-read bar graph below the traditional axial map. This
is the first application of an artificial intelligence system
that uses a previously trained set of logic rules “learned”
from sets of ectatic and normal patient topographies.36
Unlike previous keratoconus screening programs, the Figure 5-27. This topographical map has the characteristics of
Magellan is able to differentiate between astigmatism, a normal cornea (99.0%).
keratoconus suspects (KCS), true keratoconus, and pellucid
marginal degeneration. In addition, the mapper’s neural net-
work is able to assign a percentage of probability, or grade, values. In addition, an easy-to-interpret bar graph with clas-
to these disease states. This will potentially allow users to sification categories is displayed that includes a percentage
better document and follow their patients for progression of that category’s probability. Placing the computer mouse
over time. Lastly, the software can also detect the probabil- over any of the abbreviated indices or classifications reveals
ity of previous myopic and hyperopic refractive surgery, as a full explanation of the title and its significance.
well as PKP. Refractive surgery is no longer just in the realm of the
Another feature of the Magellan software is the high LASIK surgeon, as cataract/IOL surgery now requires the
resolution of the map displays, made possible through same precise results. The dramatic technology advances
improvements in the dual-edge ring-finding algorithms. As in small-incision phacoemulsification, aspheric multifocal
a cone-based Placido system, the mapper enjoys a smaller IOLs, and diagnostic instrumentation have led our patients
working distance than traditional Placido disk systems; this to demand the same “freedom from glasses” that our
improves corneal coverage. The cone features a streamlined LASIK patients have come to expect. The new paradigms
30-ring projector. However, the system can locate both require results that are within ±0.25 D sphere with ≤0.50 D
edges of each ring, which allows the Magellan’s new algo- cylinder as well as good contrast sensitivity and functional
rithm to detect 60 rings of data. This provides an astound- reading vision. Expectations often exceed what current
ing 21,600 data points. As these are evenly distributed IOLs can provide unless we are “right on the money.” To
across the cornea, that translates to twice the resolution of this end, the authors have found a second application of the
traditional topographers. Magellan Mapper extremely useful.
Figure 5-27 illustrates a typical Magellan printout with Aspheric IOLs have been shown to effectively reduce
the traditional axial map, a grouping of indices, and their SA in implanted eyes, with improved optical quality over
58 Ch a p t e r 5

previous spherical lenses.37 Studies have shown that cus-

tomized selection of aspheric IOLs based on corneal wave-
front measurements is possible and produces better visual
outcomes compared to unselected patients implanted with
aspheric IOLs.
As opposed to wavefront aberrometers that provide the
sum total HOAs of the entire visual system, the Magellan
provides corneal HOA calculations that can be used in
everyday IOL selection by the cataract surgeon. This is
something the authors have been routinely doing since 2005
by attempting the “best match” selection from “aberration-
free” (eg, Bausch & Lomb AO platforms) or negative SA
IOLs in the marketplace (such as Tecnis [Abbott Medical
Optics [AMO], Santa Ana, CA] and Wavefront [Alcon, Fort
Worth, TX] platforms).38,39

Clin ica l Ap p licat ion s

Corn eal Refract ive Su rgery
Any information that can help the refractive surgeon
avoid the dreaded postoperative complication of progressive
corneal thinning is beneficial. Keratoconus prevalence has
been reported in the range of 50 per 100,000.40 However, as
these patients are many times dissatisfied with their glasses
Figure 5-28. This topographical map has the characteristics
or contact lenses, the incidence is significantly higher for associated with clinical keratoconus (68.5%) with a severity
the refractive surgeon. index of 10.7% . This map also contains features characterized
There remains no substitute for a complete medical his- as unclassified variations (31.5%).
tory and full slit-lamp examination to best evaluate a preop-
erative patient’s potential risk for ectasia following excimer
laser surgery. Although pachymetry and preoperative kera- It should be noted that any removal of corneal tissue could
tometry readings are an essential part of every refractive theoretically accelerate the progression of early ectatic
surgery evaluation, corneal topography has become the disease. Most experts agree that surface ablation poses a
“standard of care” when it comes to identifying irregular lower risk than the creation of a corneal flap (whether 90 or
corneal astigmatism and disease states. 160 µm) plus ablation depth. Moreover, recent studies of
Despite advances in instrumentation and graphics, most up to 12 years have demonstrated the long-term safety and
modern kerato-mapping systems lack the ability or pro- refractive stability of PRK.41
gramming to help identify early progressive disorders, such Figure 5-29 shows a preoperative map that registers
as keratoconus or pellucid marginal degeneration. Most properties associated with KCS in a patient who underwent
practitioners are comfortable recognizing advanced thin- uncomplicated surface ablation. The same patient’s pre-
ning and irregular or asymmetric maps. However, detecting operative Orbscan, shown in Figure 5-30, did not display
early changes or possible forme fruste keratoconus remains any of the classic “red flags” associated with early ectasia.
a clinical dilemma. Such red flags include pachymetry readings with a thinnest
The advanced resolution and user-friendly bar graphs point less than 470 µm or more than a 100-µm difference
of the Magellan Mapper provide a much needed screen- from the thinnest point to the 7-mm zone; posterior eleva-
ing tool for the identification of abnormal corneal shape tion (compared to best fitting sphere) greater than 40 µm;
preoperatively. Using this statistical information may help high irregularity indices at 3- and 5-mm zones; and (per-
prevent undesirable postoperative outcomes by recommend- haps most important) the overall correlation of the highest/
ing against surgery for patients with evidence of progressive thinnest point coinciding on the anterior, posterior, and
ectatic disease. Figure 5-28 shows a Magellan printout of a pachymetry maps.42
patient having properties consistent with keratoconus. Also Figure 5-31 shows the postoperative stable Magellan
note that the indices listed in red are abnormal. classification graph consistent with characteristics of MRS.
Alternatively, patients classified with mild character- This case may represent a patient who was spared the
istics that are consistent with KCS on the bar graph may increased risk of postoperative progressive irregular thin-
be best treated by surface ablation or PRK (if at all). ning brought on by the creation of a microkeratome cut.
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 59

Figure 5-31. The Magellan analyzes the characteristics of the

cornea and determines if the characteristics are associated
with various conditions. In this case, the map is characteristic
of MRS (98.8%).

Figure 5-29. The topographical image shows characteristics

associated with a cornea with 1.52 D of cylinder (99.0%) and
KCS (32.5%).

Figure 5-32. This map shows characteristics associated with

Figure 5-30. The preoperative O rbscan map of the same eye unclassified variations (88.5% likelihood) and keratoconus
shown in Figure 5-29. Note there is no indication of ectasia. suspect (11.5%).

It should be noted that the previously mentioned char- from the best-fitting sphere, and the difference between the
acteristics for Orbscan screening compliment the outlined thinnest region and the 7-mm zone on corneal pachymetry
Magellan statistical classification. This is illustrated in is more than 100 µm. Again, these characteristics are con-
Figures 5-32 and 5-33. Both show cases where the Magellan sistent with early keratoconus-like changes, and this is a
topographical maps are consistent with mild KCS, while patient who perhaps should be followed without any refrac-
the Orbscan maps appear unassuming at first glance. Upon tive surgery until further information is gathered.
further review, the “hottest” region on the Orbscan poste- As refractive technology continues to evolve, wavefront-
rior float (upper right of quad map) is greater than 40 µm based excimer ablations are emerging as the procedure of
60 Ch a p t e r 5

Figure 5-33. O rbscan map of the eye shown in Figure 5-32.

Note the elevation on the posterior float (upper right of quad
map).

choice for optimum visual quality. In order for these types

of customized treatments to work, a match must be made
between the wavefront data and the pattern to be traced on
the patient’s cornea.
Because the preoperative wavefront information is the
entire basis for the ablation profile, the capture of a patient’s
true corneal shape has become more important than ever.
One cannot expect a perfect outcome if the data used to cre- Figure 5-34. This axial map has the characteristics associated
with unclassified variations (OTH = 39.5% likelihood), 0.85 D
ate the treatment plans are less than optimal. The Magellan
of cylinder (AST = 38.9%), and normal topography (21.6%).
Mapper’s statistical corneal analysis is designed to differen-
tiate between classic early ectasia signs and other abnormal
shapes that may be more likely to appear with extended
contact lens wear and corneal warpage (personal communi-
cation, Drs. S. Klyce and M. Smolek, November 5, 2004).
Most refractive surgeons recommend discontinuation
of soft lenses at least 2 weeks prior to wavefront measure-
ment and hard lenses or RGP lenses a minimum of 1 month
before refractive evaluation. As contact lens warpage
from RGP lens use has been noted to produce an irregular
corneal shape for up to 6 months,43 any axial maps that
appear asymmetrical on preoperative evaluation should be
followed and correlated with clinical history before surgery
is considered.
Figure 5-34 shows an axial map of a refractive surgery
candidate that on initial exam displayed some mild irregu-
lar astigmatism. The Magellan classifier registers in the
“other” category. Note that no ectatic characteristics were
triggered. Upon further history, this patient had been out of
her extended-wear soft contact lenses for just 1 week.
Figure 5-35 shows her follow-up exam after being out of
lenses for over 1 month. We now see a more regular axial
map with the reclassification of only “normal” and “astig-
matism.”
The previous case is a perfect example of subtle con-
tact lens warpage detected by the Magellan that might
otherwise be overlooked. These findings allow the refrac-
tive surgeon to collect more accurate wavefront information Figure 5-35. This map shows characteristics associated with
and thus produce better postoperative results. a normal cornea (75.6%), as well as features of a cornea with
0.82 D of cylinder (AST 10.3%).
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 61

Figure 5-36. The aberrometry display includes

the selected topographical map and wavefront
maps related to typical aberrations (SA, coma,
trefoil, 2-degree astigmatism, irregularities, and
other HOAs) with corresponding PSFs.

Len s-Based Refract ive Su rgery minor decentration or tilt in the capsular bag. The previous
generation of IOLs from all manufacturers induces various
The bouquet of IOL choices are ever expanding so that amounts of positive SA.
the concept of “one size fits all” has become a historical Patients with previous MRS (including radial kera-
artifact. At the present time, IOL surgeons must choose totomy, lamellar procedures, and excimer laser ablations)
from lenses that are aberration neutral or correct for nega- typically have a significantly increased value to their posi-
tive or positive SA. tive corneal SA above +0.28 µm. Conversely, those patients
In general, HOAs degrade optical quality. Although with a prior history of hyperopic refractive surgery tend to
there may be some benefits, such as a natural defense present with corneas measuring overall negative SA.
against ocular chromatic aberrations44 or a decrease of As each patient’s corneal shape and wavefront profile is
image deterioration by myopic defocus,45 most clinicians unique, our practice has found the corneal-HOA calcula-
agree that, for HOAs, “the lower the better.” tion feature of the Magellan Mapper extremely valuable
The cornea, with its high refractive power, is a primary in choosing the best IOL to match the patient’s anatomy.
contributor to HOAs. Based on a typical 6-mm papillary Figure 5-36 is an example of the corneal aberrometry page
diameter, the cornea produces about + 0.28 µm of SA.46 available for the surgeon’s review. The corneal aberrometers
The AMO Tecnis family of lenses (ie, ZA00 multifo- module adopts a model based on the Zernike decomposition
cal; ZB00 monofocal) induces -0.275 µm of SA to attempt to derive aberrometric terms from corneal surface wavefront
to neutralize the “average” positive SA produced by the analysis. All of the aberrometry work was part of a larger
cornea. Dr. Holladay has suggested that a small amount of software program created with Dr. Stephen Klyce during
residual negative SA may be beneficial in presbyopes as it his long-term scientific consultancy for Nidek Technologies
can increase the cornea’s central power at near with pupil- (personal communication, Cesare Tanassi, April 10, 2010).
lary constriction.47 Starting from the topographic map, the software com-
The Alcon family of lenses uses different negative SA putes the most significant Zernike terms (up to the 30th
factors based on research attempting to approximate the order). Thanks to a technique of wavefront approximation
residual SA measured in the emmetropic youthful eye. with Zernike polynomials, typical wavefront aberrations
The ReStor (3.0) SN6AD1 corrects -0.10 µm of SA and the (such as SA, coma, trefoil, 2-degree astigmatism, and other
aspheric wavefront IQ monofocal and toric (SN6 series) HOAs) can be expressed as weighted sums of Zernike
lenses adjust for -0.20 µm. The intent here is to produce a terms. Wavefront deformation maps are built and plotted in
small residual amount of positive SA as seen in young adult coded colors. For each aberration, the chromatic PSF and
eyes of athletes and US Navy pilots at their peak visual the simulated eye-chart are calculated and plotted.
performance.48,49 As illustrated in Figure 5-36, the aberrometry page
The Bausch & Lomb AO platforms (Crystalens, Akreos) displays several components: the selected topography,
are aberration “neutral” or without any positive or negative wavefront deformation maps related to typical aberrations
aberration induced. Thus, these lenses are less sensitive to (SA, coma, trefoil, 2-degree astigmatism, irregularities,
62 Ch a p t e r 5

Figure 5-37. Simulated Snellen eye chart. This

can be a very useful tool when attempting to
explain the effects of corneal shape and previ-
ous refractive surgeries on visual function to
patients.

Figure 5-38. Measurement of SA can be

beneficial when considering implantation
of an aspheric IO L with a lower amount
of asphericity. This value can be found
beneath the spherical aberrometry map
(second row, left).

and other HOAs), the residuals map, PSF, and simulated IOLs available were calculated based on the cornea’s posi-
eye charts. For each aberrometric term, the following are tive SA at a 6-mm pupil, we recommend that the maps also
illustrated: wavefront deformation map, color-coded RMS be set at 6 mm to attempt to best use the wavefront data
value (bad, suspect, normal depicted as red/yellow/green), collected.
PSF, and simulated eye chart. Following are examples of how we have used our
The surgeon can review the PSFs and simulated eye Magellan in deciding what IOLs to implant in our cataract
charts (Figure 5-37) in either photopic or scotopic mode. patients. Figure 5-38 of patient PR shows the right eye has
For each aberration term, the simulated eye chart will be RMS = +0.177 µm corneal SA and very low other HOAs,
calculated and displayed below the corresponding wave- a “good match” for one of the Alcon platforms. The OS
front deformation. (Figure 5-39) has RMS = + 0.30 µm and low other HOAs, a
The effect of pupil size can be varied from 3 to 9 mm. It “good match” for one of the Tecnis platforms. The surgeon
is important to measure and correctly set the patient pupil can now decide whether to “mix and match” or just choose
parameters as HOAs become significant (>4 mm) and large one of these platforms.
(increase ≥6 mm). Moreover, as the negatively aberrated
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 63

Figure 5-39. Higher amounts of SA are better

addressed with an IO L with a larger amount of
asphericity.

Figure 5-40. Negative SA found in this eye com-

bined with an IO L with negative SA correction
would suffer increased HOAs. An aberration-free
IO L is a better choice.

Figure 5-40 of patient VK’s OD shows negative SA RMS other factors? Her OS “virgin” cornea shows rms of only
of -0.021 and high amounts of other HOAs. This patient +0.185 SA with medium other HOAs. The OS is theoreti-
had no previous eye surgery, and if an “average” positive cally a candidate for one of the Alcon wavefront platforms
corneal SA had been assumed, the surgeon may have actu- (set for -0.20 SA). Having this HOA information and
ally induced more HOAs by implanting a negative SA lens. evaluating the effects of SA as well as of the other HOAs
An aberration-free (Akreos) platform was chosen here to can be helpful to making a better IOL decision and makes
attempt to avoid further changes to the low measured SA it less likely that you will be surprised by an unanticipated
and not exacerbate the higher degrees of coma if the lens outcome in vision quality.
were to tilt or decenter. Figure 5-43 of patient JW’s OD shows the classi-
Figures 5-41 and 5-42 are of patient JH who had “old” cal topographical pattern post-RK surgery. Observe the
LASIK in her OD and left her OS untreated for monovision. 4 flatter regions over the old corresponding RK cuts. Note
Her OD shows high rms of +0.85 SA and very high rms for the extremely high HOAs in all categories. The patient ada-
other HOAs. A decision now has to be made. Does one want mantly wanted multifocal visual rehabilitation. After exten-
to go SA neutral with an aberration-free platform or choose sive counseling, the best option for him was the aberration-
a negatively aberrated lens (such as Tecnis set at -0.275 SA), free Crystalens AO.
reducing the +SA somewhat but risking an increase in the
64 Ch a p t e r 5

Figure 5-41. A map from an eye S/P LASIK with

increased SA (+0.853 µm). This patient may benefit
from a stronger aberration-correcting IO L.

Figure 5-42. The fellow eye of the patient in Figure

5-41 was left untouched for monovision and thus had
far less SA.

A B

Figure 5-43. (A) Classical topographical pattern following radial keratotomy. Note the 4 flatter regions corresponding to the
RK incisions and extremely high HOAs in all categories. The patient adamantly wanted multifocal visual rehabilitation. (B) The
Magellan uses the indices to categorize the patient and correctly identified this as an eye with a history of MRS. The patient
wanted to undergo implantation of a presbyopia-correcting lens, but multifocal I classical topographical pattern post-RK surgery.
O bserve the 4 flatter regions over the old corresponding RK cuts. Note the extremely high HOAs in all categories. The patient
adamantly wanted multifocal visual rehabilitation. IO Ls are contraindicated in eyes with this amount of irregular astigmatism.
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 65

A B

Figure 5-44. (A) This patient was referred for surgery O D due to a PSC cataract. Refraction and manual Ks were “spherical,”
but the Magellan found topography with 99% artificial intelligence-based association with keratoconus with a severity index of
2.9% . Note the inferior steepening, but the central mires are quite regular, consistent with his refraction and Ks (both manual
and IO L Master, Carl Zeiss, Jena, Germany). (B) The navigator view shows very high HOA RMS in almost all categories. An
aberration-free IO L platform was the most reasonable choice. Multifocal IO L implantation should be avoided.

A B

Figure 5-45. (A) This patient presented for cataract surgery with a history of hyperopic LASIK. Note the well-centered excimer
ablation on the axial map with the neural network indices correctly identifying the previous hyperopic refractive surgery (96.6%
HRS). (B) The aberration map shows a high degree of negative corneal SA (-0.311). An aberration-free IO L or perhaps an earlier-
generation spherical IO L that will induce some positive SA would be beneficial.

Figure 5-44 is of patient MK, a 54-year-old man Figure 5-45 illustrates patient CB presenting for cataract
referred for OD surgery for a PSC cataract. His refraction surgery with a history of hyperopic LASIK. Note the well-
and manual Ks were close to “spherical.” His past ocular centered excimer ablation on the axial map with the neural
history was not contributory. The Magellan aberrometer network indices correctly identifying the previous hyper-
view shows topography with 99% artificial intelligence- opic refractive surgery (96.6% HRS). Here, the aberration
based association with keratoconus with a severity index map (Figure 5-45B) shows a high degree of negative corne-
of 2.9%. Note the inferior steepening, but the central mires al SA (-0.311). The surgeon can consider an aberration-free
are quite regular, consistent with his refraction and Ks IOL or perhaps in this case an earlier-generation spherical
(both manual and IOL Master, Carl Zeiss Meditec). The IOL that will induce some positive SA to best match this
navigator view shows very high HOA RMS in almost all wavefront.
categories. In our view, the aberration-free Akreos AO
platform was the most reasonable choice.
66 Ch a p t e r 5

Ot h er Refract ive Applicat ion s Con clu sion

In addition to the Magellan’s advanced screening fea-
The Magellan Mapper and its novel classification system
tures and corneal wavefront profiles, the mapper has
can provide clinicians with accurate, easy-to-read topogra-
advanced contact lens-fitting software with simulated fluo-
phies as well as color-coded corneal wavefront aberration
rescein test images and a library of available contact lens
maps that complement any refractive evaluation. The high-
parameters. Last, the topographer has an available ortho-
resolution images and expanded range of indices make this
keratology module. Detailed axial scans, dual and quad
topographer an invaluable tool for pathology detection and
multimaps, and difference mapping provide the clinician
management.
with precise fitting and therapeutic diagnostics.

TH E KERATRON C ORNEAL TOPOGRAPH ERS

Renzo Mattioli, PhD and Nancy K. Tripoli, MA

T Th e Qu est for Accu rate

he Keratron family of CTs are Placido-based, non-
spherically-biased, and computer-assisted (Figure
5-46). The original Keratron, introduced in 1993 Cor n ea l Top ograp h y
by Optikon 2000 (Rome, Italy) was designed to satisfy the
demand for more precise and intuitive corneal measure- In the past 25 years, the color-coded corneal maps
ment by overcoming limitations in an earlier generation of produced by CTs have become an indispensable tool for
CTs. In 1999, Optikon recognized the need for a portable, clinical understanding of corneal shape. The challenge of
modular topographer and introduced the Keratron Scout, videokeratoscopy has been to present clinically relevant
a CT that can be handheld, fitted on a generic slit lamp corneal shape information in pictorials that are well under-
with a slide adaptor, or positioned intraoperatively with a stood to discourage overinterpretation. In addition to maps
weight-balanced arm and sterile disposable covers. In 2002, of corneal power, CTs have also mapped corneal height
an external unit, the Keratron Bridge, was added to update and used the height unit to fit contact lenses and predict
the Keratron providing image grabbing, power supply, and or evaluate refractive surgical techniques, such as PRK,
USB interface with any standard PC, either desktop or lap- LASIK, LASEK, and intrastromal rings.
top. The Keratron Piccolo was introduced in 2006 to take Since the early 1990s, a new generation of lathes has
advantage of USB2 general-purpose computer technology. allowed fabrication of custom contact lenses with far more
A version of the Keratron Scout, the Keratron Piccolo, fits complex shapes that can conform to the large variety
on a slit-lamp carriage and uses the display of a PC rather of keratoconic corneas. In the same period, ophthalmic
than an integral display of its own. The Keratron Onda is excimer laser technology has been greatly improved by the
a new-generation instrument that acquires corneal topogra- introduction of flying spot technology and high-speed eye-
phy and the ocular wavefront (OW) contemporarily and at trackers, which allow custom modelling and custom retreat-
various accommodation states of the eye. Its PC is embed- ments. Both of these applications could be linked to CTs
ded, and it can be connected via LAN cable to an external and guided by corneal topography measurements. However,
PC. All currently available Keratrons are equipped with the applications needed data with higher accuracy and local
corneal wavefront analysis, IR pupillometry (except the spatial resolution than could be achieved by spherically-
Piccolo), and a number of new accessories. biased methods.
Scout is a full 32-bit software package that is compat- In the past, Placido-based CTs were criticized as theo-
ible with the Windows XP, Windows Vista, and Windows 7 retically incapable of producing accurate corneal height
(Microsoft Corporation, Redmond, WA) operating systems. information.50-52 It was argued that the corneal surface
It operates all Keratron topographers, even those dating description could not be recovered from Placido images
back to 1993-1994, and can be upgraded from Optikon’s without irrevocable limits on accuracy. The Keratron was
Web site at www.optikon.com. designed to overcome such criticism. The problems to be
Before describing these and other Keratron family solved were (1) accurately measuring the cornea so that the
software features and clinical applications, it is important maps and other applications are trustworthy; (2) providing
to understand the Keratron’s unique method of corneal a color-coded map unit that shows corneal shape detail;
reconstruction that reveals corneal shape details that are not (3) supporting software processing with precision hard-
commonly evident with most other topographers. ware; (4) rigorously testing the accuracy produced by the
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 67

Figure 5-46. Scout software

processes corneal topography
and wavefront maps from all
the Keratrons. (A) Keratron with
Keratron Bridge. (B) Keratron
Scout, handheld. (C) Keratron
Scout, intraoperative. (D) Keratron
Piccolo. (E) Keratron O nda.

+ (dz/dx) 2) 3/2
(1____________
hardware/software configuration; and (5) using the con- Ri =
(d2x/dx2)
figuration to demonstrate as much information as possible
where z is sagittal height and x is the distance from the
to the clinician. This section describes the challenges of
axis. The sphere used to measure axial power does not have
Placido-based CTs and the Keratron’s design implementa-
the same instantaneous curvature as the cornea at the same
tions to meet those challenges.
reflective position unless the cornea is spherical. Unlike the
sphere, the center of curvature does not necessarily lie on
Clin ical In form at ion in a Corn eal the CT axis. Axial power at any point has been shown to be
Top ograph y Color-Coded Map the average of the instantaneous curvature from the center
Prior to the introduction of the Keratron, the leading to that point.53
commercial CTs showed maps of “corneal power” as mea-
sured along a number of radials. The unit shown was axial Trad it io n a l “Sp h er ica lly Bia sed ”
power whose mathematical definition is the power of a Met h od s to Deduct Axial Power From
sphere, centered on the CT axis that has the same tangent Placido Reflect ion s
as the cornea at each measurement position on a corneal
profile. Axial power is inversely proportional to the radius Prior to the Keratron, the leading commercial CTs
337.5 derived axial power by (1) measuring the power of the
of the sphere, using the “thin lens law” (D = _____
Ra
), and that
radius is the distance from the surface position to the CT’s cornea to magnify the image of a mire54 or (2) matching
axis measured along a surface normal. Mathematically, it the size of a reflected mire to the sizes of mires reflected
is defined as from a series of spheres.55,56 Figure 5-47, left, shows the
(1 + (dz/dx) 2)1/2
x_____________ problems of these methods. The diagram shows rays from
Ra = (dz/dx) a mire reflecting from a cornea (solid line) and a sphere
where z is sagittal height and x is the distance from the (dotted line). A point on the sphere (A) and a point on the
axis. Axial power represents the “tilting” of the surface cornea (B) will reflect the same ray to the lens and form
downward from apex to periphery. Its historical relevance the same image in the reflected mire pattern. However, the
lies in its use in the ophthalmometer, and its scientific rel- sphere and the cornea have different axial radii, heights,
evance is derived from the fact that the tangent to a surface and instantaneous curvatures. Traditional axial power mea-
determines refraction. surement records the radius of the sphere as the radius of
Because the corneal surface is smooth, its height changes curvature of the cornea. This results in mild error for axial
gradually from position to position. The surface’s rate of radii and more extreme error for height and instantaneous
change, the tangent, reveals more detail about corneal shape curvature.
than does height. However, the “bending” of the cornea The false assumption that a sphere and a cornea, which
reveals even more details. The geometric unit that describes reflect the same ray to the lens, have the same powers and
the bending of the cornea in a given direction through a heights is called “spherical equivalency” or “spherical
given point is the local, also called tangential, meridional bias.”55 When axial power is measured by mire magnifi-
or instantaneous curvature, which is the inverse of the cation or mire size, corneal height cannot be determined
radius of a circle whose profile matches the profile of the because, for a selected reflection position, an infinite fam-
corneal surface at each measurement position. The radius of ily of combinations of tangents and heights can produce
the sphere (Ri) is defined as the same axial power. It has been definitively demonstrated
68 Ch a p t e r 5

Figure 5-47. (Left) Spherically

biased algorithms identify the ray
reflected from the cornea as equiv-
alent to a ray from the same mire
(M) reflected on a sphere. (Right)
The Keratron arc-step instead
reconstructs the whole corneal
profile as a continuous sequence
of arcs, from vertex to periphery,
resulting in higher accuracy of both
heights and curvatures.

that this assumption leads to incorrect height measure-

ment.54,55 The reliance of early CTs on spherically biased
measurement techniques resulted in the false assertion that
all reflective CTs must be inaccurate for measuring height
and instantaneous curvature.50,51

Corn eal Su rface Recon st ruct ion Usin g

t h e Kerat ron Arc-Step Algorit h m
An arc-step algorithm such as that used by the Keratron
is a surface reconstruction method that is not spherically
biased and is not completely new. It was introduced by
Allvar Gullstrand in 189657 and resurrected in different
formats by Doss in 198158 and by other authors between
1989 and 1992.59-61 Nevertheless, the arc-step was usu-
ally assumed to be difficult to implement in a commercial
CT without smoothing data with polynomial fitting or the
like.62
The Keratron identifies a continuous sequence of arcs,
starting at the vertex, that accurately reconstruct a corneal
profile along each of 256 radials, as shown in the right
image of Figure 5-47. A cornea reflects rays from each mire
through the lens to produce a unique reflected mire pattern
for that cornea. The radii of the arcs in the Keratron’s recon-
Figure 5-48. Representations of a normal, astigmatic cornea
structed profile are adjusted until they reflect rays from (left) and a keratoconic cornea (right) include the axial power
the appropriate mires to the lens and produce the reflected (top) and curvature maps (middle) and color-coded local cur-
pattern. The generic arc-step reconstruction method can vature along the 9 o’clock corneal profile (bottom).
describe aspheric corneal profiles because the second and
more peripheral arcs need not be centered on the CT axis.
There is one and only one sequence of arcs that can satisfy and instantaneous curvature for shape visualization. The
a unique sequence of mire reflections along a radial. detail that is missing from an axial power map, but is
An unsmoothed arc-step algorithm was implemented on depicted by an instantaneous curvature map, is clinically
the Keratron to provide accurate height measurements and significant.66,67 For example, on an axial map of a normal
to produce meaningful maps of instantaneous curvature.63 astigmatic cornea such as that in Figure 5-48 (top, left),
When the profile is reconstructed, it yields the instanta- the toricity of regular astigmatism seems to extend to the
neous curvature, height, tangent, and axial power at each periphery. When the same cornea is depicted using instan-
intersection of 2 arcs. This obviates the need for calculating taneous curvature as shown in the same figure, middle left,
one corneal descriptor from another because curve-fitting it now appears that the astigmatism is confined within an
and mathematical derivations necessarily erode measure- approximately circular region of curvature above 37 D, the
ment accuracy and/or spatial resolution.64,65 “corneal cap” (green and yellow). That region is surrounded
by a flat peripheral region (blue), which becomes even flat-
Kerat ron Map s of Axial an d Cu rvat u re ter toward the periphery, reaching a maximum flatness of
15 D (23 mm) or more. The existence of the corneal cap,
Un its which appears on most normal corneas, has implications
Maps that depict unbiased units demonstrate the clini- for contact lens fitting and also may prove valuable in early
cal utility of axial power for traditional reference to optics diagnosis of disease. The diagram in Figure 5-48 (bottom,
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 69

left) represents a corneal profile from 9 o’clock to 3 o’clock.

The relatively uniform curvature across the corneal cap in
this direction, shown in green and corresponding to the
green in the curvature map, changes in the periphery, which
is relatively flat, shown in blue corresponding to the blue in
the map.
Viewing an axial power map of keratoconus, as shown in
Figure 5-48 (top, right), can lead to a misinterpretation of
shape because a large corneal region seems to be affected
by the cone. The instantaneous curvature map (Figure 5-48,
middle, right) shows that the most highly curved region of Figure 5-49. The instantaneous curvature map (right) of a
the cone is only 2 mm in diameter, and curvature decreases post-PRK cornea clearly shows shape features, like the steep
rapidly to a large corneal region that is unaffected by the bending over the edge of the ablation and central islands, that
cone. The diagram (Figure 5-48, bottom, right) shows that are not clearly visible in the axial power map (left) of the same
the cone is characterized by a highly curved central region cornea.
that involves the corneal vertex, surrounded by a somewhat
flatter annular region. Much of the corneal cap is unaf-
fected by the cone.
The importance of depicting corneal shape details is
well illustrated by maps of a cornea after PRK, such as that
in Figure 5-49. On this axial map (left), the steepness of the
edge appears to involve a large annular region surrounding
a generally flat center. On the curvature map (right), the
steep bending of the cornea over the ablation’s edge can be
appreciated. More importantly, some central island anoma-
lies, which are critical to vision following PRK, are not Figure 5-50. The instantaneous curvature map (left) and height
visible on the axial map but are seen on the curvature map. or “spherical offset” map (right) depict the same cornea with
contact lens warpage.
Detecting and understanding corneal warpage caused
by contact lenses is also aided by an instantaneous curva-
ture map, as shown in Figure 5-50. On the curvature map
(left), the distinctive pattern includes a flat annular region positions where the sphere’s surface and the cornea’s surface
(blue) surrounded by alternating steep (red) and flat (blue) coincide. This is illustrated in Figure 5-50, right. Positions
regions. The map on the right side is obtained by subtract- where the cornea rises above the sphere are colored in
ing corneal height from the height of a sphere. Although the shades of red, and positions where the cornea dips below the
map contains less detail than either axial or curvature maps, sphere are in shades of green-blue. With this convention, the
it shows that a highly curved corneal region, indicated by map is analogous to a topographic map of the earth.
the large cross, differs from the surrounding region by only The positioning of the sphere radically alters the appear-
5 µm. To reveal this detail requires an operator’s option ance of the map, and users must understand and become
for positioning the reference sphere and also precise height accustomed to the alternative views of height. When we
measurement so that small color steps describe localized first introduced “spherical offset” height maps, users were
corneal shape features. frequently disappointed by the smooth, general appearance
of this kind of map and the lack of detail, even on diseased
corneas. To allow them to enhance heights in the region of
Th e Sph erical Offset Map
interest (eg, the apex of a keratoconus, the optical zone of
A map of absolute sagittal height (ie, depth with respect a postrefractive treatment), the Keratron allows the sphere
to the corneal plane) would always appear as a series of to be fit tangent to the apex (as in Figure 5-51) or to any
concentric colored doughnuts because the decreases in position selected by the user, either tangent to a given point
height from center to periphery are much greater than any or passing through 3 reference points, which is especially
circumferential variations in height. Therefore, another useful to align pre-post ablation height differences as in
surface similar to the cornea’s must be subtracted and the Figure 5-52.
differences in heights mapped. Simulations of normal cor-
neas have been proposed as the basis for subtraction, but
the normative shape of the cornea is disputed, complex, Kerat ron Ba sic Feat u res
and highly variable, making the use of a single “normal”
reference of little value. Therefore, the Keratron subtracts The accuracy and cost-effectiveness of a measuring
a sphere, a shape that can be easily visualized by users. instrument depend on the appropriateness of each of its
Difference values near zero are colored yellow and represent components to the overall objective. To implement the
70 Ch a p t e r 5

Figure 5-51. The height scale resolution of the spherical offset

map can be adjusted to measure detail like this 1.9-µm ectatic
scar.

Figure 5-53. Four model surfaces were fabricated to simulate

the 4 clinical conditions shown on the left. Accuracy was
always within ±1.5 µm.

85 µm.54,74 Another frequent criticism of CTs is their

disregard for the circumferential rise and fall of an astig-
matic surface, which theoretically deflects mire reflections.
Evidence that this real optical effect produces inconsequen-
tial errors in the reconstruction on surfaces with up to 12 D
of regular astigmatism has been published elsewhere.75,76
Attempting to measure more complex surfaces encoun-
tered in clinical and surgical practice can lead to even larger
height and curvature errors with spherically biased meth-
ods.72,73 The multiple transitions from flat to steep to flat
Figure 5-52. The pre-post height difference (A minus B) after
areas present a real challenge for topographic algorithms.
an application of the Schwind ESIRIS laser treatment to an
Optikon conducted in-house testing on 4 surfaces that
aspheric PMMA surface (top row) and to a living eye (bottom
row), which shows the accuracy of the Keratron Scout for mimicked clinical cases and presented the results at the
measuring corneal heights. 1996 Association of Research in Vision and Ophthalmology
annual meeting (Figure 5-53).77 In spite of the greater
complexity of such shapes, the error remained within less
arc-step algorithm on a commercial instrument required than 1.5 µm. The testing has been repeated on the Keratron
special attention to inherent problems, for example numeri- Scout with results similar to the Keratron. The examples in
cal instability and extremely sensitive ring tracing.51,59 The Figures 5-48 through 5-53 demonstrate the flexibility of the
subpixeling and other proprietary numerical methods and arc-step for tracking complex profiles without smoothing
design solutions adopted for the Keratron are described details.
in more detail elsewhere.68 The following is a list of the In addition to other applications like custom contact
Keratron’s main features and the resulting unique benefits. lenses and laser links, the Keratron’s height accuracy is very
useful for verifying the specified custom laser ablation pat-
Heigh t Accu racy tern as implemented, including file transfer, ablation design
algorithm, and “fluence.” Figure 5-52 shows an example of
Earlier CTs had proven successful in measuring
such a custom ablation pattern (on the left) and the post-
spheres,56,69-71 but were highly inaccurate for measuring
minus pretreatment height measurement by a Keratron
aspheres.54,55 Since 1993-1994, the Keratron has been
Scout (on the right). In the top row, the ablation was applied
tested by several independent researchers on nonspherical
on curved plastic surfaces, and the yellow annular zone
profiles.72 Tripoli and colleagues73 measured, on a set of
(difference = 0 µm) corresponds to the untreated zone.
normal prolate, rotationally symmetric surfaces, an error
The bottom row shows the effect of the same specified
within 0.25 µm within the central 3-mm zone and an error
ablation on a living eye. This example depicts a case by
within 1 µm for entire surfaces. Under similar conditions,
Dr. Camellin, treated with the ESIRIS laser by Schwind
a TMS-1 (Computed Anatomy) height error was within
Eye-Tech solutions (Kleinhostheim, Germany) and the
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 71

Figure 5-54. The Keratron “Dekking” type cone (left) achieves

greater corneal coverage than large “Placido disk” topogra-
phers (right) when imaging the same cornea.
Figure 5-56. The Keratron software includes a repeatability
check, shown here assessing the stability of acquisition by a
surgeon through 7 intraoperative topographic images of the
same eye.

Rep eatabilit y Ch eck

In spite of EPCS automated capture enabling, harsh
movements and other occurrences (eg, blinking, tear film
disruption) can still create artifacts that affect overall accu-
racy, especially in hand-held mode. To screen out unaccept-
Figure 5-55. The Keratrons’ Eye Position Control System able images, a repeatability check feedback function was
(EPCS) includes an infrared beam that intercepts the corneal introduced in the Scout software. The repeatability check
vertex and permits image acquisition only when the cornea is window, seen in Figure 5-56, can be recalled at any time, or
within a narrow range of Z distance. preset to automatic whenever 3 or more images of the same
eye are acquired and processed. Normal eye repeatability is
Scout-ORKw (Optimized Refractive Keratectomy—wave- considered “good” with a deviation of the best fit sphere of
front) corneal wavefront link.78 processed maps within 0.24 D and “very good” if less than
0.12 D. Using this feature for self-training, it is not difficult
C orn ea l C overage, Fo cu sin g, an d to obtain repeatability well within 0.1 D, thus reducing the
acquisition root mean square (RMS) height error to within
Posit ion in g less than 0.4 µm over a 6-mm zone.
The use of a cone containing lighted mires is not new.
It was introduced by Dekking in 1930 to place the mire Ran ge of Measu rem en t
arrangement close to the cornea in order to measure a larger Even normal eyes frequently have an instantaneous cur-
area.79 The Keratron mire reflections cover about 80% to vature range between 20 and 50 D. In keratoconus, scars,
90% of a normal cornea in a well-opened eye, illustrated and postsurgery, corneal curvature often shows a much
in Figure 5-54. However, the cone can be very sensitive greater range of power. To achieve their full potential, CTs
to focal distance. To overcome this theoretical limitation, must also be able to cover such extreme cases without losing
the Keratrons are equipped with a patented Eye Position too much local information or introducing artifacts. Figure
Control System (EPCS), pictured in Figure 5-55, in which 5-57 shows a case in which the local curvature ranges from
an infrared beam intercepts the corneal vertex. With this less than 10 D (dark blue) in some areas to above 120 D
system, a picture can be taken only when the focusing, (the measured point = 127.6 D) in others. Mires that are too
the “Z” positioning, is within a range as tight as 0.05 mm, close together limit an instrument’s capability for measur-
which can guarantee a repeatability of ± 0.1 D or better (see ing abrupt changes across the corneal surface. On the other
next paragraph and Figure 5-56). Lateral (X, Y) decenter- hand, if the mires are too sparse, height reconstruction and
ing is not critical because of a software method called Eye detection of local distortion are impaired. With these oppo-
Misplacement Control System (EMCS), which recalculates site needs in mind, the Keratron’s mire arrangement was
the spatial mire position in the measured direction.68 optimized both in number and size of steps (28 mires uni-
Cantera and associates have verified that a misalignment in formly spaced about 0.166 mm on a 43-D sphere), making
any direction, up to 1 mm, induces less than 0.1 D of change use of both black-to-white and white-to-black mire borders.
in the resulting curvature map.80
72 Ch a p t e r 5

Figure 5-57. The Keratron’s mire arrangement and subpixel

edging algorithms allow the capture of detail in high resolution Figure 5-58. The Keratron can track the pupil edge in photopic
and over a large range of curvature. In this cornea, curvatures and mesopic conditions and report it in topographic maps.
as steep as 2.64 mm (128 D) and as flat as 37.5 mm (9 D) are
measured.

instruments. Furthermore, a measure of individual natural

If mires were tracked at the middle of the rings as in most pupil dilation at night, or under mesopic-scotopic visual
classic “cone type” videokeratoscopes, twice the number of conditions, is often required in clinical preoperative prac-
rings would be needed for the same measurement distribu- tice. A dual infrared + visible illumination system has been
tion. This would limit the instrument’s curvature measure- implemented in both the Keratron Scout’s (Figure 5-58,
ment range to a maximum of 60 to 70 D.68 Detecting the left side), and the Keratron’s “lamp board.” An assembly
borders of alternating black and white mires instead ensures of software, firmware, and electronics allows acquisition
accurate tracking of mires even with complex, tangled pat- of the eye image and tracking of the pupil edges in both
terns. In Figure 5-57, only some small edematous areas are mesopic and standard photopic light conditions.
unprocessed because the data were automatically discarded The reconstruction of the pupil by the Keratron is per-
by tracking algorithms as unsuitable. Occasionally, special formed on an internal reconstruction of a “clean picture”
processing functions such as “dots editing” may be needed from the keratoscope image.68 The mires must have been
to recover mires that were automatically tracked incorrectly. tracked correctly without induction of artifacts. In some
cases, a “pupil editor” function can be used for hand cor-
Spat ial Resolut ion rection. The pupil position at the 2 extreme illumination
conditions is measured and represented with respect to
The Keratron’s mire arrangement samples an average corneal vertex. On average, in a normal population, the
normal cornea at 166-µm spacing steps. This requires pupil center does not change significantly from photopic
special edging techniques to prevent smoothing and los- to mesopic condition.81,82 Nevertheless, in some cases,
ing important information about localized distortion such the pupil center can be substantially displaced (0.25 mm
as scars. Precise mires tracking is accomplished by the in the case of Figure 5-58, right). In such a case, although
Keratrons due to their subpixeling technique. Rather than the laser treatments are tracked on the photopic pupil, the
just searching for black-to-white and white-to-black edges ablated zone could be more properly centered on the largest
thresholds, the technique uses the entire gray scale infor- naturally dilated area.
mation from all the pixels of each profile sector captured
by the CCD camera. The resulting equivalent accuracy in Th e “Move Axis”
detection of a ring position is less than 1/20 of a pixel.68 The
small ectasia in Figure 5-51, whose height is less than 2 µm, To measure curvature relative to the virtual pupil, surface
is detectable because of special tracking and arc-stepping measurements must be recalculated relative to a new axis.
algorithms that take full advantage of the crisp mire edge As with recalculating off-center images, a lateral transla-
reflections. It could not be detected by nonreflective meth- tion will result in errors. Instead, the Keratron performs a
ods because the confusion in surface-scattered projected 3-dimensional rotation of the corneal model. The “move
lines or slit light edges is typically much larger than the axis” feature can also shift the corneal representation to
height of the feature. The mire reflections instead naturally any point desired by the user. This is useful to visualize
describe even the smallest corneal scar (see Figure 5-51, the symmetry surrounding points of interest such as the
left side). keratoconic cone apex shown in Figure 5-59. This example
also shows how the flat annular area surrounding the apex
Pupil Trackin g (A) creates a high astigmatism, typically against the rule
in patients having an inferior keratoconus, because the
The use of links between the corneal wavefront and the meridional cut across the keratoconus and the apex is the
excimer laser in which a complex ablation pattern must be local flattest meridian, and the steepest meridian is perpen-
precisely aligned with pupil eye-trackers requires accurate dicular. Moving the axis to the apex (B) shows the actual
measurement of the pupil center and edge, a feature that symmetry of the shape of keratoconus.
was less important when CTs were primarily diagnostic
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 73

Figure 5-59. Keratron move-axis feature. Different views of the same

keratoconus (A) before and (B) after moving the keratoscope axis to the
cone center show that the keratoconus is symmetric.

Figure 5-60. Using the Keratron move-axis feature (C) produces the same result as asking the patient to change fixation (B).

When an image is acquired while a subject gazes at the Process Ed it in g an d “Dots Ed it in g”

fixation light, as is shown in Figure 5-60, and the opera-
tor moves the axis, the resulting map looks remarkably When CT users acquire a corneal image and are shown
similar to a map produced by requesting the subject to only a colored map, they are unaware of the many choices
change his or her gaze. Thus, moving the axis produces a that have been automatically made by the instrument. All
virtual rotation of the eye.83 In Figure 5-60, normal fixa- measuring systems can occasionally produce artifacts or
tion produces a typical postablation “knee,” a red ring (A). skip over regions. The ability to change the parameters that
If the patient is asked to fixate to the side, the vertex moves control Keratron processing can benefit the clinician who
toward the periphery, and a small “hourglass” indicates the understands his or her objective. The Keratron offers the
high astigmatism of the “knee zone” (B). The same image user the options to improve ring reconstruction in difficult
can be created using the Keratron’s move-axis feature (C), cases, for example, by altering the sensitivity of borders,
showing that the feature mathematically rotates the cornea. the incremental ring limit, the maximum number of merid-
The move-axis (C) preserves the entrance pupil position, ians that can be interpolated, the sensitivity for finding the
whereas this important reference is lost when the patient processing center, or the sensitivity for excluding suspected
changes fixation (B). This feature is particularly useful to extraneous points. A versatile software feature called “Dots
measure the actual astigmatism when patients cannot keep Editing” includes a number of tools to allow the user to
fixation, such as during suture adjustment after PKP with reassign every black/white (green “dots”) or white/black
the intraoperative Keratron Scout.84 threshold (red “dots”) to its proper ring. A complete remov-
al of artifacts with this tool can be especially important
before custom contact lens (CL) height-fitting or custom
74 Ch a p t e r 5

Figure 5-61. CLMI identifies a highly probable keratoco-

nus (PPK = 98.5%) shown with 4 maps: curvature (top
left), axial (top right), Gaussian (bottom left), and spheri-
cal offset with the sphere positioned tangent to the apex
indicated by Cc (bottom right). The CLMI numerical
values are shown in the screening panel (A). When PPK
exceeds criterion by 20% , description data can be seen
by clicking on yellow (>20%) or red (>45%) buttons.
The description data include the 8-mm search area (B),
the steepest 2-mm circle (C), as well as the average (Cc/
Ca /Cg) in the circle and its location (D) from the vertex
in polar coordinates.

laser link. The user is also permitted to clean a ring pattern CLMI was evaluated on 132 normal and 112 keratoconic
by erasing artifactual areas but is not allowed to add rings eyes in 2005 by Twa87 who applied a receiver operator
or segments of rings arbitrarily. characteristic curve analysis. He found that CLMI accu-
racy could be 90% at second place after his proposed C4.5
In t raop erat ive Fu n ct ion s decision tree, with a specificity as high as 99% and a 79%
sensitivity. This statistical analysis of screening had optimal
The Keratron Scout can be fit into an optional weight-
results with axial maps.87
balanced trolley (see Figure 5-46C), and sterile accessories
The CLMI were implemented in Scout software in
allow the surgeon to take intraoperative corneal topography
2005.89 In scout.exe terminology, axial magnitude (Ma) is
(ie, before and after a treatment). While the Keratron Scout
used for screening, and curvature (Cc) is used for descrip-
is sterile sleeved, the surgeon can position and use upside-
tion (Figure 5-61). The screening panel includes Ma and
down buttons to start acquisition for either eye, capture an
a statistic, percentage probability of keratoconus (PPK).
image, and even browse in the database and save the topog-
If PPK is lower than a criterion, then a green circle indi-
raphy without needing to touch the PC keyboard. The intra-
cates that the eye is not keratoconic. If PPK is higher than
operative Keratron Scout has proved very useful in adjust-
a criterion, either a yellow (for suspect) or red (for highly
ing sutures after PKP and LKP,84 during PTK treatments,
probable) button allows the user to get a description panel
and in measuring preoperative and/or de-epithelialized
on the map. The description includes the “C1” circle (2-mm
corneas before and after LASEK topo-link treatments.85
steepest zone), the circle’s location relative to the vertex
in polar coordinates, and the average D in that circle, Cc,
Ca, Cg for tangential, axial, or Gaussian curvature maps,
Con e Lo cat ion a n d Magn it u d e respectively (top left, top right, and bottom left in Figure
In d ices, Gau ssia n Map , a n d 5-61). The main advantage of CLMI over other indices is
that it locates the keratoconus and shows the clinician exact-
Follow -Up Feat u res ly where CLMI located the presumed cone apex, rather than
using a hidden process. The clinician can look at the map
Th e Con e Lo cat ion an d Magn it u d e to decide if the Ma or PPK could be due to an artifact or
In d ices some clinical or surgical reason other than a keratoconus.
Furthermore, the Cc parameter, which is the average diop-
The Cone Location and Magnitude Indices (CLMI), tric power in the steepest circle C1 on curvature maps, is an
introduced by Drs. C. Roberts and A. Mahmoud in 2001,86 especially valuable quantifier to follow the keratoconus, as
were created to automatically distinguish keratoconic from described in a later section.
normal corneas. CLMI is calculated by an algorithm that
locates the steepest 2-mm diameter circular area of a diop- Th e Gau ssian Map
tric map (axial, tangential, or Gaussian curvature). The cen-
tral keratoconus index (KCI) is the difference between the The Gaussian is the product of the steepest and flattest
average value within the steep circle and the average value curvature in a plane perpendicular to a point of a surface.
of the remaining corneal area within 8 mm. The peripheral According to the Theorema Egregium by Carl Friderich
KCI measures the asymmetry of the cornea87 as the differ- Gauss (1828), this product is invariant for a flexible non-
ence between the average within the steep circle and the elastic surface. Gaussian curvature is therefore an intrinsic,
average within a circle on the opposite side of the cornea. invariant property of surfaces.90 To illustrate the prop-
A recently introduced calculation88 allows continuous mea- erty, the Gaussian curvature of a flat sheet of paper is 0.
surement from central to peripheral keratoconus. When the paper is bent to any curvature in one direction,
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 75

Figure 5-62. Gaussian curvature is an intrinsic property

of any flexible, inelastic surface. Squeezing such a surface
induces astigmatism on the (tangential) curvature map, but
the product of the principal curvatures, C1 x C1, is invariant,
and thus the Gaussian map is unchanged.

Figure 5-63. A comparison between curvature (upper

row) and Gaussian maps (lower row) of 1 astigmatic
(left) and 2 keratoconic (center and right) corneas
shows that Gaussian maps do not depict astigmatism
unless it is shown as a superimposed vector field.
Gaussian maps do clearly reveal ectasia, whose loca-
tions are calculated independent from their distance
from the corneal vertex.

(C1 x C2)
_______
it will remain straight (0 curvature) in the perpendicular M= 2
direction. The steep times flat product will be 0, so the
It can be easily demonstrated that the square of the
Gaussian curvature will remain 0. Squeezing a spherical
2 measures differs by the square of astigmatism:
surface, as shown in Figure 5-62, will make it steeper in one
direction and proportionally flatter orthogonally. Surgeons G2 = M2 + (C1 – C2) 2
who fit sutures on a donor cornea after PKP know that
“Mean curvature” maps may appear quite similar to
nonuniform tension will create astigmatism in the axial
Gaussian if astigmatism is not high, but they are not
and (tangential) curvature maps. But, unless the wound is
invariant for the surface because they do not fulfill Gauss’
stretched to introduce cornea elasticity or permanent inelas-
Theorema Egregium.
tic mechanical distortions, the Gaussian map will remain
Figure 5-63 shows differences between Gaussian maps
constant.
(lower row) and curvature maps (upper row). Note the pair
Unlike axial and tangential curvature calculations, the
of maps of an astigmatic cornea on the left. On the Gaussian
Gaussian curvature calculation is independent of the dis-
curvature map (bottom, left), the regular astigmatism
tance from the corneal vertex. Therefore, Gaussian maps
cannot be seen, and the map appears no different from a
can unequivocally show the apex of a keratoconus. This
rotationally symmetric cornea. Similarly, neither the blue
application was first proposed in 1997 by Barsky and col-
sides nor the astigmatism in the apex of the keratoconus are
leagues.91
visible in the middle and bottom right maps. Gaussian maps
Gaussian curvature as implemented in the Scout soft-
with vectors at points around the cornea allow visualization
ware in 200790 is not actually the product, but the geomet-
of astigmatism and its pattern (eg: Figure 5-61, bottom left).
ric average in D, the square root of the product C1 x C2:
However, a curvature unit that hides astigmatism, as well as
G = (C1 x C2)1/2 the flat sides that surround keratoconus apices, is not ideal
for a complete clinical evaluation. Gaussian maps should
where C1 and C2 are the 2 principal curvatures in that point.
always be viewed in conjunction with maps that depict other
Some authors have proposed a “Mean curvature” unit,92
units and should never replace them.
which is the arithmetic average of the same curvatures:
76 Ch a p t e r 5

Figure 5-64. Follow-up of a case of keratoconus

is shown in curvature maps and a plot of CLMI Cc
from 6 tests over a period of almost 2 years. The
first 3 tests were done before, and the fourth test
after CXL treatment. The final 2 tests were done
after an ICR implant. The parameter Cc proves to
be ideal to show the natural progression of kera-
toconus and the stabilization after treatments.

Figure 5-65. Follow-up of the same case as in

Figure 5-64 is shown in a plot of HOAs of the
corneal wavefront and corneal O PD maps. The
higher-order RMS parameter quantifies visual
improvement, rather than a shape factor, and is
confirmed by patients’ self-report.

In the new Scout software release, the CLMI applies if a Gaussian map had been computed at least once. Moving
Gaussian measurement to better locate the keratoconus the axis will affect all of the maps on the screen and the
apex as Cg, the CLMI circle C1 on the Gaussian map, and Sim-K readings but will not affect CLMI nor the aberra-
the unbiased location of the ectasia. There is a move axis to tions. Corneal wavefront maps can also be shown with pupil
Gaussian apex option that places the map center of any type size and aberrations chosen from the Zernike pyramid.
of map (curvature, axial, and spherical offset) on the center In Figure 5-64, we see the history of a keratoconic eye
of circle C1. This option can be applied automatically to a over almost 2 years and after 2 kinds of treatments. In the
number of maps, taken at different dates, in order to better first 3 topographies, the progression of the cone steepness
evaluate Sim-K and other shape factors within the follow- is evident by the increase of Cc (ie, the average curvature in
up feature. D within the circle at the apex). After CXL treatment (test
#8-3), Cc regressed by more than 2 D. After implantation of
Th e Scout Follow-Up Feat u re ICRs, the apical steepness dropped even more dramatically,
by more than 6 D. Figure 5-65 shows the same patient’s
When a patient has undergone a series of 3 or more tests,
tests, for which the clinician has selected the Corneal WA
rather than comparing them in pairs (eg, with “pre-post”
(bottom tab, red circle, high orders maps) and to plot their
differences like those in Figure 5-52), the clinician can
RMS value (upper tab, red circle).
look at all the tests in a single window in order to evaluate
This figure gives us a different perspective than Figure
over time the changes of the various maps and parameters
5-64. Rather than changes in the corneal shape, it shows the
including CLMI, Sim-K and other indices, or individual
changes in the quality of vision. It explains why, in this par-
corneal and ocular aberrations.
ticular case, the patient could not appreciate the regression
Figures 5-64 through 5-67 show some examples of the
of cone steepness after CXL, but recognized the greater
Scout follow-up window. In this window, the clinician can
improvement after ICR, which dramatically reduced HOAs,
choose 1) the patient’s eye OD/OS (tabs on the left side);
especially coma.
2) the map type (circled in red, bottom left); and 3) the
Critical to the decision of whether a keratoconic patient
parameter (circled in red, top-left) to be plotted. The maps
should have a CXL treatment or not, in addition to corneal
are updated in real-time and are shown in the middle of the
thickness, is whether his or her ectasia is stable or progress-
window, just below the plot of the selected measurement
ing. This assessment is not an easy task when changes are
unit. Their axes can be moved to the entrance pupil (button
slower or smaller than in the case shown in Figures 5-64
at the bottom left) or to the “Gaussian apex” (Cg), but only
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 77

Figure 5-66. An unoperated keratoconic eye

is followed over 8 years (2002 through 2010).
Average Sim-K is plotted, and axial maps are
shown.

Figure 5-67. The same case as in Figure 5-66 is

shown after (1) moving the axis of the axial maps
to the “Gaussian apex,” Cg, as calculated by
CLMI, and (2) recalculating Sim K. The recalcu-
lated Sim K shows the likely progression during
the 4-year gap between tests.

and 5-66. Most researchers who do not have a topographer

with CLMI and the “move axis” feature determine whether
Wavefron t An a lysis
the keratoconic ectasia is progressing by examining average Wavefront representations have been implemented in
Sim-K values, which are calculated on the corneal vertex of the Keratron Scout software since 2001.93 These include
axial maps. This measurement can be reliable when the ker- measuring the Corneal Wavefront (CW)* from Keratron
atoconus apex is central. But in peripheral keratoconus, the corneal topography, importing the OW from most com-
disease progression can reduce the flat Sim-K reading rath- mercial aberrometers through files containing Zernike
er than increasing the steep Sim-K reading, making Sim-K terms and pupil data, and calculating the Internal Wavefront
a very poor index as shown in Figure 5-66. Four years after (IW) as the linear subtraction of Zernike terms (IW =
the first 2 tests, even axial maps show an increase in the OW – CW). Other wavefront representations include refrac-
cone magnitude. Average Sim-K has increased by about tion maps from selected Zernike terms and simulating
0.5 D, but in the subsequent maps, this value looks very visual function (PSF, MTF) and performance (letter chart,
fuzzy. If the axes of the same maps are moved to the visual acuity, night vision). A more detailed description of
Gaussian apex (see Figure 5-67), the etiology of average these functions,94,95 as well as wavefront analysis96 and
Sim-K is clearer and is consistent with CLMI Cc and Cg. Zernike polynomials can be found elsewhere,97 and in the
After 4 years, the actual progression was almost 2.5 D, and Scout “on-line-help.” Here below, we summarize some
in the following years, another 0.5 to 0.7 D. By referring to basic topics of the new features that allow CTs to “talk the
such a plot, the clinician can more objectively decide, and same language” as total wavefront analyzers.
involve the patient in the decision, whether to do a CXL or
perhaps wait for another 6 months of follow-up. * The terms corneal or ocular wavefront are used here as generic descrip-
It has to be pointed out that plotting changes of only a tions of either wavefront error/wavefront aberration or optical path differ-
ence (OPD). According to the OSA standard, both units can be used but
few tenths of a D requires a well-calibrated Keratron and a
the reader must be aware that they are opposite in sign. OPD maps appear
good operator who makes use of the “repeatability check” similar in color convention to height maps, while wavefront error maps have
(see Figure 5-56) to get precise measurements. opposite colors.
78 Ch a p t e r 5

Figure 5-68. The right (top) and left eye (bottom) of

one of the authors are shown as Keratron topography
(left), ocular O PD (imported from a S/H aberrometer),
and Keratron corneal O PD (right column).

Th e Corn eal Wavefron t: A New Way to

Represen t Corn eal Opt ics
The purpose of wavefront analysis (WA) is to measure
wavefront aberrations, and the purpose of CT is to mea-
sure corneal shape. However, more than 70% of the eye’s
refraction is due to the corneal first surface. Corneal distor-
tion always affects total OW aberration more significantly
than any other interface. The CW, the component of the
OW due to the cornea, can be calculated from CT height
data and fit with a Zernike decomposition just as the OW
can be measured by an aberrometer and fit with Zernikes.
Placing the 2 Zernike representations side-by-side allows
inspection of the OW and the contribution of the CW.
Figure 5-68 shows maps of the OW of one of the author’s
Figure 5-69. A cornea with an inferior keratoconus accelerates
right and left eyes, measured with a Shack Hartmann aber-
the rays in the superior and retards the rays in the inferior as
rometer, and of the CW, derived from corneal topography.
compared with the ideal cornea. These physical distortions
Only the high-order Zernike polynomials from 3rd to 5th create a reversed aberrated wavefront whose height is scaled
radial order are shown. As usual, there is similarity between about 1:3 to the corneal distortion.
the OW and CW maps that shows how high-order aberra-
tions of the cornea affect total ocular aberrations. However,
when total ocular aberrations are very low, there is much space.”99,107 If a cornea had an ideal shape and minimal
less similarity between OW and CW representations of the internal aberrations, then rays traced from the fovea would
same eye, and internal aberrations IW become the predomi- exit parallel from the cornea and form a flat wavefront (see
nant component of the OW. Figure 5-69, green lines and yellow map on the top right)
and the aberrations would be zero. If the cornea is distorted
From Corn eal Geom et ry to Corn eal from this ideal shape (example Figure 5-69, orange lines),
Wavefron t Maps the rays in one section (the superior) enter the air first where
they accelerate. Other rays (the inferior) remain in the cor-
Since 1994, several authors have conducted studies of the
nea for a longer path and are retarded. In the time it takes
CW from CT data, including Howland and colleagues98;
for a ray to travel about 3 µm in the cornea, it would travel
Hemenger and colleagues99; and, in 1997, Schwiegerling
about 4 µm in the air (ie, rays travel [n-1] times faster in air
and Greivenkamp.100 A comprehensive collection of CW
than in cornea, where n is the assumed refractive index).
and IW studies under different patient conditions are
For example, in Figure 5-69, a corneal distortion (b) of
among the work of Marcos and others101,102 and Artal
3 µm will produce an OPD (a) 1 µm different from the ideal.
and colleagues.103-106 Figure 5-69 illustrates a model of a
As a result, we can adopt a simple “rule of 3” from corneal
simplified procedure based on the optical path difference
elevations:
(OPD; ie, the “least time” Fermat principle) in the “object
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 79

Every 3 µm of distortion from the ideal shape of the cor-

nea will produce about a +1-µm difference in the OPD map
and a -1-µm difference in the wavefront error map.
This simple rule can help estimate the thickness of a soft
contact lens to add to the cornea or the amount of corneal
tissue to remove from the cornea in order to produce a flat
CW with no aberration, although some precautions should
be followed for different refractive indices. The CW OPD
maps (see Figure 5-68) closely resemble topographic height
maps, but there are conceptual differences between the
“spherical offset,” which shows a cause, and OPD data,
which show the effect. The “ideal cornea” is in fact an ellip-
soid centered on the corneal vertex. Its definition depends
on refractive indices and pupil position and requires Figure 5-70. Pupil diameter and Zernike data selected in a
matching the manifest patient refraction with the corneal Scout wavefront window will affect the corneal wavefront
map, which further affects the simulation of the patient’s
curvature at a preset “refractive zone.” In fact, the Zernike
vision.
term, defocus, Z(2, 0), which depends on the axial length,
is the only Zernike term that cannot be measured from the
cornea alone and must be input from the manifest spherical
equivalent refraction.

Zern ike Decom p osit ion of Corn eal

Wavefron t
In the wavefront module of the Scout software, the wave-
front surface is calculated from corneal height, and Zernike
polynomials up to the 7th order are least-squares best-
fit.108 For an accurate result, all corneal positions within
the selected pupil, weighted for their subtended area, are
involved in the fitting process. Piston and Tilt components
(Zernike radial orders 0 and 1), although calculated, are not
usually displayed.
Figure 5-71. A number of alternative Zernike term formats are
Zern ike Select io n a n d Grap h ica l shown by the Keratron software.
Represen tat ion s of t h e Aberrat ion s
The user can select the pupil size, from 3 to 9 mm, and (“Histogram,” “Color boxes,” “RMS skyline,” “Zernike
any combination of Zernike terms to be represented by the list,” or “Aberration summary”).
wavefront map by clicking on specific buttons, arranged as A clinically relevant representation is the “Aberration
suggested by the OSA standard pyramid.109 For example, summary.” Here, the aberrations at the opposite sides of the
the user can select all orders from 2 to 7, the high orders OSA pyramid (“sine” + “cosine,” or positive and negative F
only, the high orders and astigmatism, only radial orders index) are presented not as individual Zernike terms, but as
from 3 to 5, or any combination of individual orders. When their RMS sum. This sum is expressed in µm as well as in
the selection is confirmed, the relevant weighted sum of “equivalent defocus” D and axis. For example, in the case
Zernikes is calculated and reported in the map. The ray in Figure 5-71, the amount of primary coma is 0.13 D at
tracing, the PSF, MTF, or visual-simulation selected in the 75 degrees, or 0.097 µm RMS (or the combination of sine
display panel at the top right in Figure 5-70 are updated and cosine terms) at the selected pupil size. The equiva-
to show the effect of using only the selected Zernike lent defocus proposed by Dr. L. Thibos110 expresses the
terms. Such a decomposition of the wavefront into selected wavefront variance in familiar dioptric terms. Reporting
Zernike “ingredients” can support the clinician’s under- the Zernike terms in defocus equivalent D has the further
standing of how individual or combined aberrations affect advantage that D are fairly independent of pupil size, as
vision. In the “Zernike” display panel (Figure 5-71), a shown in a study by Indiana University.111 A green-yellow-
spectrum of the selected Zernike polynomials can be pic- red pattern in Figure 5-71 suggests a likely normality of
torially illustrated by selecting from the 5 different formats individual aberrations.
80 Ch a p t e r 5

Figure 5-73. Example of a simulated PSF from a corneal wave-

Figure 5-72. From wavefront O PD maps, PSF, MTF, and visual
front map (O PD) compared with a drawing of the equivalent
simulations are calculated using FFTs.
manifest PSF by the subject in scotopic conditions. (Reprinted
with permission of Renzo Mattioli, PhD.)

From t h e Wavefron t to Poin t Spread

Fu n ction , Modu lation Tran sfer Fu n ction , and retinal problems may cause these simulations to differ
an d Visu al Sim u lat ion s from reality. However, in an example (Figure 5-73), we can
appreciate how well the corneal PSF matches the patient’s
The calculation of the PSF, the MTF, the visual simu- drawing of a small spot of light. Both the CW and the draw-
lation of 2 different letter charts (from 20/500 to 20/100 ing were made under the same lighting, and therefore pupil,
and from 20/100 to 20/20), and pictures of night vision conditions. This example shows that the CW simulations,
are not done by simple ray-tracing, but by complex math- which directly derive from the corneal PSF, can be espe-
ematics using Fourier optics on the pupil size area and the cially realistic in the presence of high corneal aberrations,
Fast Fourier Transform (FFT), as summarized in Figure which are difficult to measure directly with an aberrometer.
5-72. Only this process produces faithful simulations112 by
including the effect of both the wave propagation and the
diffraction created by the pupil edge. Simplified patterns Lin k s W it h Excim er La ser s
or retinal spot-diagrams should not be confused with the
real PSF. Keratron height data can be output to files for any
“third-party” software application. Specific links have been
Th e Clin ical Ut ilit y of C o rn ea l developed in agreement with Schwind Eye-Tech Solutions
Wavefron t Measu rem en t (Kleinostheim, Germany) and Nidek (Gamagori, Japan).
These include a “topo-link” with Schwind, the Optimized
CW measurement is not intended to replace WA. This Refractive Keratectomy—topography (ORK-t) software
new representation has special clinical utility because it (now obsolete) and a versatile link with Nidek “FinalFit.”
evaluates corneal optics using the same “language” as Optikon pioneered with Schwind the adaptation of CW
wavefront analyzers and can be decomposed by Zernike features to link a Keratron with an excimer laser system.
fitting into representations analogous to WA, including The Scout software’s CW was linked to the Optimized
simulated visual acuity, PSF, MTF, and other quantities that Refractive Keratectomy—wavefront (ORKw), software that
assess quality of vision. In this respect, the CW is a more had been previously developed by Schwind for OW links
sensible representation of corneal optics than either axial or with ESIRIS and AMARIS. The first case of a photo-
refractive maps. Unlike WA, CT data allow investigators to ablative treatment using a corneal wavefront link was made
assess the CW under varying conditions such as pupil size by Dr. M. Camellin on February 2002.78 Other medi-
and position. cal researchers extended this experience in the next few
CW analysis can help clinicians understand patient com- years, including Bonci,113 Barraquer (Bogotà, Colombia),
fort or discomfort despite apparently bad or good corneal Arbelaez,114 Rummelt,115 Siganos,116 and others. (For more
conditions, may aid in the design of surgical interventions, complete and updated documentation, contact Schwind’s
and may predict the efficiency of strategies that concentrate APM department and the Web site at www.eye-tech-solu-
on removing only extremely disturbing aberrations. Visual tions.com.)
simulations of the CW are not to be regarded as a curiosity In principle, the following are advantages of a CW link
but as new, very powerful tools for clinicians. In addition with respect to topography, OW, or both:
to internal aberrations, light scattering, media opacity, • Optimization of the ablation pattern volume
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 81

• Simpler comparison of results with the target

• Overcoming some limits of Munnerlyn formulas117,118
• The possibility of aberration management
• The possibility of interaction with aberrometers
A more detailed description of these arguments can be
found in quoted papers and in a book on LASEK and ASA
edited by Camellin.119

Fit t in g Con t act Len ses W it h

t h e Kerat ron ’s Soft w a re Figure 5-74. The Scout software module CLE is used to fit to
a keratoconic cornea using a “virtual trial set” and a custom
designed triconic lens.
The Keratron’s accuracy in tracking both small local
and peripheral corneal distortion produces sensitive height-
difference fluorescein pattern algorithms to guide a user
through easy design and simulation of contact lenses fitting.
The user interface is ergonomic for easy familiarization,
even for old-fashioned practitioners. Classic methods allow
users to select or fit lenses according to central K-read-
ings and corneal diameter. However, the Scout software’s
Contact Lens Editor (CLE) introduces a number of radi-
cally new height-fitting methods and protocols.
The CLE is a software module that is accessed by
clicking any map or processed image from the database.
Although maps usually cover 80% to 90% of the cornea, a
preliminary extrapolation, placing 3 points around the lim-
Figure 5-75. In this example of the CLE’s use to fit a young
bal edge, is required to fill the gaps and to measure corneal patient with an ortho-K contact lens, the protocol projects
diameter at the same time. a 10.1-mm lens to get 15 µm of clearance at the apex and a
Some specially automated “lens-family protocols” are parallel fit in the midperiphery, with a “compression factor”
provided to guide the user through a large number of virtual of -1.75 D.
trial-sets including the first, made for Horus (Verona, Italy)
contact lenses120 and similar sets including TS (Milano,
Italy) and Soleko (Pontecorvo, Italy). the map is sufficient to select the meridian and see
how the relevant clearance section changes.
Th e Scout Con tact Len s Ed itor • Moving the lens by drag-and-drop to any point on the
The Keratron’s CLE windows and menu offer several cornea simulates real lens movement on the eye and
choices: is used to find the point of maximum symmetry, a
recommended first step.
• Select a virtual trial set from among a wide number
of CL manufacturers (registered users, see Scout CD- • Tilting the lens along the flattest and steepest merid-
ROM or Optikon Web site) to simulate the fitting of ians, using 4 buttons on the map, simulates the effect
a lens. of blinking, and gravity.
• Fit a custom multi-curve contact lens according to • Indenting the lens (ie, pushing in or pulling out) simu-
classical criteria “Contour” and “OC+2.” lates the effect of indentation and compares this with
the pattern of a tight lens.
• Or, fit a custom lens using a “height-fitting” method
that includes multi-curve (up to 5 zones), multi-conic • Programming the “lens selection protocol” (ie, the
(Figure 5-74), toric (up to 4 zones), or Ortho-K lenses way the lens is first chosen within any trial-set). The
(5 zones, inverse geometry, Figure 5-75). protocol can be either classical curvature (ie, accord-
A number of tools allow the user to evaluate and opti- ing to K-readings and toricity) or the default “height-
mize fitting and design: fitting” protocol, which leaves an “apical clearance”
of 10 µm.
• A clearance panel (eg, bottom right in Figure 5-74B
and C) plots the cross-sectional vertical distance • Evaluating the residual cylinder of a front spherical
between the lens and the cornea in µm. Clicking on lens or designing the ideal front toric with a lens
power input and calculator panel.
82 Ch a p t e r 5

Figure 5-74 shows contact lens fitting to a keratoconic

cornea (A) using the Scout CLE and 2 methods: B) a lens
selected from a “virtual trial set” and C) a custom-designed
triconic lens created automatically with the “height-fitting”
(aspheric) feature. The “clearance panel” shows how well
the contact lens conforms to the flat corneal meridian
when the lens is properly placed on the cone’s apex. Figure
5-75 shows an example of Ortho-K contact lens fitting.
The user can create or select from among the available
“height-fitting” protocols to modify the data. The protocol
in this example projects a 10.1-mm lens to get 15 µm of
clearance at the apex and a parallel fit in the midperiphery,
with a “compression factor” of -1.75 D. The operator must
then input only the desired correction in the panel (A), and
the software will automatically calculate the parameters
of a 5-curve CL reported in panel (C). Note the similar-
ity between the desired (B) and achieved clearance plots Figure 5-76. CALCO windows include the tear-layer screen,
(D), despite natural corneal asymmetries. It is unique to which contains data on how a lens fits a corneal semi-merid-
ian according to the parameter of clearance between the
the Scout that all geometric components of the lens are not
cornea and lens.
empirically extrapolated from Sim-K, Ro, and e values,
but are calculated on actual corneal heights, measured
by Keratron’s along the flat meridian. This optimizes the to fitting parameters rather than complex geometric ones.
design, even with complex and asymmetric corneas. In fact, the first step in lens design is to define the desired
Lens designs can be saved and attached to the relevant tear layer shape (Figure 5-76). The “tear layer” screen
corneal topography, up to 8 different designs per image. contains data on how a lens fits a corneal semi-meridian.
The CLE fluorescein simulations have advantages over The lens design is controlled by the clearance between
real fluorescein testing at the slit-lamp because the topog- cornea and lens. Diameters and lifts are chosen in order to
raphers’ “virtual reality” fitting is static and does not stress optimize the lens with respect to power, thickness, optical
the cornea. A physical test can hide how much the cornea zone, correction of lens SAs, and multifocal optic options.
is fitting itself to the lens as the lens shapes, possibly creat- This approach allows full control of the design because the
ing patient discomfort and warpages like those shown in parameter of clearance between the cornea and lens is much
Figure 5-50. closer to fitting philosophy than the parameter of eccentric-
As a general result, many optometrists using a Keratron ity. Moreover, working with clearance frees the fitter from
have improved their RGP first lens success fitting rate from the overhead of managing complex geometrical parameters.
the traditional 30% to 40% and up to 75% and above.121 The use of Calco Lens Design has dramatically improved
first fit success rate and patient comfort in a number of
Calco Len ses an d Ot h er Th ird -Part y complex situations.
Lin ks Figure 5-77 illustrates some examples:
• Early stage keratoconus: The design is chosen as a
Not including Berkeley UC’s OPTICAL research proj-
conform fit, moderately flat. The slit-lamp image
ect,122 Eikon’s (Firenze, Italy) Calco was the first com-
confirms comfortable wear.
mercial third-party software provider to manufacture fully
custom contact lenses based on Keratron data beginning • Post-hyperopic PRK treatment: The design leaves a
in 1996.123-125 In Latin, “Calco” means “cast.” The basic small central clearance; The slit-lamp image shows
idea of Calco Lens Designs is a strict integration between lens stability and good physiological response.
corneal topography and contact lens manufacture. The • A rare superior keratoconus: An axial symmetry
earliest contact lens manufacturing lathes were designed design, confirmed by the patient and the slit-lamp
to cut sphere-based geometries. New-generation lathes (eg, image as having good positioning, high visual acuity,
the Rank-Pneumo Optoform, by Taylor Hobson, Leicester, and comfort.
UK, and later Precitech Keene, NH) overcame these limi- Following Eikon’s Calco, other fully custom methods
tations by designing complex aspherical profiles made of designed to link the Scout with third-party software have
a large number of small steps with a resolution down to been developed by companies all over the world. Dr.
10 nm. Such geometries can significantly improve comfort Edward’s WAVE126,127 (Figure 5-78) is very popular, espe-
and global performance of an RGP lens. A natural evolu- cially in the United States, as well as Europe’s ProCornea’s
tion was to manufacture an optimized contact lens based EyeLite and Ing. Liffredo’s Focal Points.128 These pro-
on accurate elevation data from the Scout. This gives the grams import complete data from the Keratron’s Scout
fitter the capability of designing an aspheric lens according software to specialized labs (eg, Eikon, VerkerOptik [now
 Pla c id o D isk-Ba se d To p o gr a p h ic Syst e m s 83

Figure 5-77. Examples of CALCO lens designs including early Figure 5-78. The popular WAVE program
stage keratoconus, post-hyperopic PRK treatment, and a rare (www.wavecontactlenses.com) is shown in a
superior keratoconus. screen shot. (Reprinted with permission from
Renzo Mattioli, PhD.)

Optiek Verkerk, Amersfoort, The Netherlands], ProCornea

[Eerbeek, The Netherlands], Essilor [Wiener Neudorf, their clinical practice and studies. Since then, the Optikon
Austria], No7 [East Sussex, UK], Eyequip [Jacksonville, Keratron team has improved the software and hardware
FL], etc) to manufacture fully custom lenses. of this family of CTs. All Keratron owners are invited to
register at www.optikon.com in order to download software
updates and be notified by e-mail. We hope that all current
Con clu sion and future Keratron users will continue to appreciate our
commitment to maintain our state-of-the-art topographic
Thousands of users and researchers from around the systems. I invite them to continue to report to us their feed-
world have been using the Keratron CTs since 1993 in back and suggestions.

ASTRAM AX C OMPREH ENSIVE D IAGNOSTIC W ORKSTATION

W ITH P OLAR G RID TOPOGRAPH Y
David D. Liu, PhD; Jia Q u, MD; Shihao Chen, MD, OD;
Tracy Schroeder Swartz, OD, MS, FAAO ; and Ming Wang, MD, PhD

I
nnovation and technology advancement has afford- corneal elevation with a range from 0.00 to 2.60 mm and
ed ophthalmologists and optometrists with diagnostic accuracy to ±1 µm. AstraMax acquires images in less than
equipment of unprecedented sophistication. As such, cli- 0.2 seconds, eliminating the need for software or straps to
nicians expect more measurements and diagnostic capabili- compensate for eye movements. AstraMax also generates
ties from modern-day topographers. LaserSight’s AstraMax anterior and posterior axial, instantaneous curvature maps
(Winter Park, FL) multi-camera topographer is designed for and different comparative options of the displays, process-
this purpose with hardware and software integration that ing data within 10 to 15 seconds. Information can also be
gives the system a variety of corneal and pupil diagnostic displayed as point spread and modulation transfer functions
functions.129 One of AstraMax’s most critical measure- (PSF and MTF), Zernike coefficients, plus a variety of
ments for custom ablation is the anterior and posterior physician-adjusted preferences to customize the software.
84 Ch a p t e r 5

Figure 5-80. The Holladay Diagnostic Summary displays criti-

cal clinical information.

posterior corneas. Single map, dual maps, difference maps,

Figure 5-79. A quick sequence of exposure to targets using
and quad maps are also available for one or more corneal
controlled illumination is captured within 0.2 seconds.
exams displays. Both 2- and 3-dimensional color maps and
3-dimensional contour displays are available. The Holladay
Diagnostic Summary displays critical clinical information
Mu lt ip le Pa ra m eter (Figure 5-80).
Mea su rem en t s in a Sin gle Corn eal Opt ical An alysis Tools
Acq u isit ion AstraMax offers advanced optical analysis capabilities
AstraMax uses synchronized, multi-camera imaging using Zernike analysis, ray-tracing modeling, and image
systems to capture multiple images for various measure- simulation. It incorporates corneal wavefront analysis, cor-
ment purposes without the mechanical scanning of the neal MTF, and corneal PSF. All optical analysis functions
acquisition head. This eliminates the source of error asso- are displayed 3-dimensionally. These analysis tools can
ciated with the movement and the extended discomfort help the clinicians understand the optical property of the
associated with scanning technologies. Anterior topogra- cornea and communicate the diagnosis to the patient. For
phy, whole corneal pachymetry, and posterior topography example, the corneal PSF illustrates the superior resolv-
can be acquired in a single measurement. Dynamic pupil ing capability of a prolate-shaped cornea compared to the
size and location relative to the visual axis center under oblate shape (Figure 5-81).
various illumination conditions, cornea limbus diameter, AstraMax also provides corneal Zernike analysis capa-
and geometric center can be acquired as well. All of these bilities to analyze the corneal aberration. Figure 5-82
parameters are particularly critical for a custom ablation analyzes SA with respect to the asphericity (Q value) of
algorithm when using laser refractive surgery. the corneal models, including spherical, prolate and oblate
As shown in Figure 5-79, a quick sequence of exposure ellipsoid, toroid, and astigmatic oblate and prolate models.
to targets using controlled illumination is captured within
0.2 seconds. This simultaneous acquisition eliminates the Polar Grid an d Con cen tric Rin g Pattern s
error caused by mechanical movement of scanning mea- The pattern used in AstraMax is a combined concentric
surement techniques. ring and polar grid pattern130 as illustrated in Figure 5-83.
Standard corneal topography uses a set of concentric rings
Com preh en sive Diagn ost ics Displayed as a reflective target to measure the axial aberration of the
in a Sin gle Screen corneal surface. The radius of each ring at each angular
direction (the radial direction) represents the surface aber-
AstraMax offers standard data displays, including ante-
ration of the point along the radial direction only. Along
rior and posterior elevation and curvature, pupil and
each ring, there is no information along the circumferen-
limbus measurement, pachymetry, and optical analysis.
tial direction. With a polar grid pattern as represented in
Elevation, axial curvature, instantaneous curvature, and
Figure 5-83, information along radial and circumferential
refractive power maps are available for both anterior and
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 85

A B

Figure 5-81. The corneal PSF illustrates the superior resolving capability of a prolate-shaped cornea.

Figure 5-82. The corneal PSF illustrates

the reduced resolving capability associ-
ated with oblate shape.

A B

Figure 5-83. SA analysis with respect to the asphericity (Q value) of the corneal models, including spherical,
prolate and oblate ellipsoid, toroid, and astigmatic oblate and prolate models.

directions are captured and processed. For concentric asymmetrical objects, critical information is missing along
rings and axial symmetrical objects, measurement and the circumferential direction with concentric ring-only
analysis are performed only along radial directions. For Placido.
86 Ch a p t e r 5

A B

Figure 5-84. (A) The pattern used in AstraMax is a combined concentric ring and polar grid pattern. (B) The target used to cre-
ate reflections on the cornea.

A B

Figure 5-85. (A) Image of the anterior Placido reflection image of the patented grids for a sphere or axial symmetrically aspheric
surface. Reflected image of a cornea with 4 D of corneal astigmatism. (B) Axial image of a cornea with 4 D of corneal astigma-
tism.

O rt h ogon al Polar Grid an d Ax ial A sphere is an axial symmetrical surface where con-
centric rings alone are adequate to measure the anterior
Asym m et rical Objects surface. An astigmatic surface, however, best illustrates the
For axial asymmetrical objects, AstraMax performs effectiveness of the polar grid to address torsional twist.
analysis using concentric rings as well as spokes. For radial Figure 5-84 is the image of the anterior Placido reflection
direction, concentric rings are used. For circumferential image of the patented grids for a sphere or an axial symmet-
direction, a new measurement parameter, the twist angle rical aspheric surface. The spokes are roughly 10 degrees
map, is obtained to further illustrate the complex nature apart. Figures 5-85 and 5-86 are the reflected images of 4 D
of optically axial asymmetrical objects. Measurement and astigmatic and 8 D astigmatic surfaces of the polar grid pat-
analysis are performed along both axial and circumferential terns. The 10-degree spokes are “twisted” toward the flat
directions. Image is stretched differently in all directions, axis (horizontal axis). The amount of twist varies according
and image is twisted according to the twist angle. AstraMax to the magnitude of the astigmatism and the angular region
captures the reflective twist, and, likewise, refractive twist of the surface. Figure 5-87 is a summary of the twist angle
measurement can be obtained using optical ray tracing of for various astigmatic surfaces with respect to a no-twist
the anterior twist measurements. spherical or axial symmetrical aspheric surface. It can be
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 87

A B

Figure 5-86. (A) Reflected image of a cornea with 8 D of corneal astigmatism. Note the 10-degree spokes are “twisted” toward
the flat axis (horizontal axis). The amount of twist varies according to the magnitude of the astigmatism and the angular region
of the surface. (B) Axial image of a cornea with 8 D of corneal astigmatism.

Figure 5-88. Image twist occurs where the twist bends toward
the flat axis. The minimum twist occurs at the steep and flat
axis, while the maximum twist occurs in the middle of the
steep and flat axis.

Figure 5-87. A summary of the twist angle for various additional image-forming complexity to the imaging
astigmatic surfaces with respect to a no-twist spherical optical system (in this case, the astigmatic reflective
or axial symmetrical aspheric surface. or refractive surfaces).

Accu m u lated Tw ist An gle an d Tw ist

observed that the twist has several characteristics with
regard to the optical property of the cornea: Ch an ge Rate
• The object no longer reflects rays in the same plane Image twist occurs according to Figure 5-88, where the
as the incident rays. Instead, it may reflect rays in a twist is bending toward the flat axis. The minimum twist
different plane, causing the object image to twist. occurs at the steep and flat axis, while the maximum twist
• By applying Snell’s law, this twist ray characteristic occurs in the middle of them. In this figure, between 0 and
will also be in the refractive domain, causing the 10 degrees, the image is compressed along the circum-
image formed to be twisted. ferential direction. Likewise, between 80 and 90 degrees,
the image is stretched. In the middle between 40 and
• The twist also causes objects to reflect in a way
50 degrees, there was very little compression, or stretching,
that may enhance (compress) or weaken (stretch)
even though the maximum accumulated twist occurs in this
the images of different parts of the object, causing
middle region.
88 Ch a p t e r 5

Figure 5-89. The accumulated twist angle within 0 - Figure 5-91. By taking the derivative of the accumulated twist
and 90-degree region where the horizontal axis is angular angle, we can get the twist rate of change of this 4-D astig-
direction theta and the vertical axis is the accumulated twist matic surface, where the range of the twist rate is between
angle. For this 8 D astigmatic surface, the maximum twist approximately 1 to -1 degree for the angular range of 0 and
angle is 4.8 degrees at 45-degree location. Note that the 90 degrees.
accumulated twist angle fits a second-order polynomial quite
well, and its coefficients are proportional to the magnitude of
the astigmatism as can be seen from this figure, in this case,
a factor of 2.

Figure 5-92. By taking the derivative of the accumulated twist

angle, we can get the twist rate of change of this 8-D astig-
matic surface, where the range of the twist rate is between
approximately 2 to -2 degrees for the angular range of 0 and
Figure 5-90. The accumulated twist is clockwise, toward the 90 degrees. Note that a linear relationship fits the data quite
flat axis. There were no twists at 0 and 90 degrees. For this well, and slope and offset are proportional to the magnitude of
4 D astigmatic surface, the maximum twist angle is 2.4 degrees the astigmatism, in this case a factor of 2 compared to Figure
at 45-degree location. 5-92.

The accumulated twist angle within the 0- and 90-degree the illumination angle allows for detailed capture of the
region is documented in Figures 5-89 and 5-90, where the pupil and iris and iris recognition. This important infor-
horizontal axis is angular direction theta and the vertical mation, when combined with the simultaneously acquired
axis is the accumulated twist angle. The accumulated twist topography data, is crucial to the alignment and registration
is clockwise, toward the flat axis. There was no twist at of the eye during laser refractive surgery.131-133
0 and 90 degrees. The maximum twist happens at about AstraMax is capable of using controlled illumination
45 degrees. Please note that the accumulated (whole body) light sources and capturing the pupil size and location with
twist is bending toward the flat axis (0 degrees), and the respect to the visual axis under various conditions. Figure
maximum bending occurs at about 45 degrees and slowly 5-93 is the captured scotopic pupil data. In Figure 5-94, the
returns to zero twist at 90 degrees. Therefore, the rate of same eye captured the pupil data under photopic conditions.
twist (twist angle per angular section) has to change sign Both the pupil size and location are dynamic, changing
where the maximum twist occurs (see Figures 5-89 and with time and illumination condition. Proper assessment is
5-90). By taking the derivative of the accumulated twist important to determine the nominal pupil offset under the
angle, we can get the twist rate of change as shown in proper conditions when data are to be used in laser refrac-
Figures 5-91 and 5-92. tive surgery.
Due to AstraMax’s unique design of illumination and
Dyn am ic Pupil Measu rem en t an d Iris imaging optics, iris patterns are distinctively revealed,
recorded, and analyzed under programmable illumination
Recogn it ion intensity as shown in Figure 5-95. Iris registration and
AstraMax uses programmable infrared illumination alignment capabilities are important, particularly when
to allow the image capture and measurement of the pupil treating patients with significant astigmatism.133,134
parameter of diameter and location. The unique design of
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 89

Figure 5-93. Captured scotopic pupil data showing the Figure 5-95. Iris patterns are distinctively revealed, with
pupil size (7.18 mm). pupil diameter, which is used to align the patient eye
during surgery.

Figure 5-94. The same eye as in Figure 5-93 is cap-

tured under photopic condition, showing pupil size of Figure 5-96. The 3-camera system uses 8 radial acquisi-
4.57 mm. tion meridians to optically measure the corneal thick-
ness along normal direction. (A) This is the view from
the middle camera. (continued)
W h ole Eye In st a n t a n eo u s an d
Sim u ltan eou s Opt ical Pach ym et ry
The advantage of the technique is the instantaneous and
AstraMax uses 3 gen-locked cameras arranged at the simultaneous acquisition of the whole data points by all
proper angle to capture an optically bent focal plane that 3 cameras at the same time. The disadvantage is the some-
matches the contour of the cornea through an optical pro- what loose data density. Future improvement includes addi-
jector of a starburst pattern onto the cornea. Three stereo tional meridians with dense data points and more advanced
cameras capture the starburst pattern, and advanced image- image processing to obtain higher spatial resolution.
processing algorithms digitally measure the whole eye cor- Other than whole-eye pachymetry, additional anterior
neal thickness. As can be seen in Figure 5-96, the current segment information such as anterior chamber depth and
design offers 8 radial acquisition meridians to optically IOL thickness and lens opacity can also be obtained (see
measure the corneal thickness along the normal direction. Figure 5-96). Posterior topography is obtained by the
90 Ch a p t e r 5

B C

Figure 5-96 (continued). The 3-camera system uses 8 radial acquisition meridians to optically measure the
corneal thickness along normal direction. (B) This is the view from the side camera. (C) This is the view from
the other side camera.

combination of anterior topography and whole-eye optical function of astigmatism, and the distribution is highly
pachymetry. correlated to the optical transfer function of the cornea.
Simultaneous and dynamic pupillometry, iris recognition,
and corneal topography are critical to the overall assess-
Con clu sion ment of the patient’s eye and, in particular, to the clinicians
involved with laser surgery when alignment of the eye and
An instantaneous and simultaneous comprehensive diag- visual axis is of significance. A comprehensive diagnostic
nostic instrument is clinically important, particularly in workstation with instantaneous and simultaneous measure-
laser refractive surgery. Polar grid, combined with concen- ment capabilities provides effective tools in topo-guided
tric rings, offers additional information not available from laser refractive ablation for particularly problematic and
a concentric ring-only Placido system. Twist angle is a complex corneas.131

TOPOLYZER
Mirko R. Jankov II, MD, PhD; Renato Ambrósio Jr, MD, PhD; Diogo L. Caldas, MD;
Ana Laura C. Canedo, MD; Leonardo N. Pimentel, MD; Bruno F. Valbon, MD;
and Sissimos Lemonis

T
he Allegretto Topolyzer (Alcon, Fort Worth, TX; The Topolyzer has 11 rings of the Placido disks that
Figure 5-97) is a Placido disk corneal topography are projected on the cornea (Figure 5-98), so that up to
system with an integrated autokeratometer, which 22 reflection edges are recorded by a CCD camera and
facilitates best focusing on the reflected mires on the cor- up to 22,000 measuring points are detected by the soft-
nea and allows for an automatic acquisition of the Placido’s ware (Figure 5-99). Basically, the reflected image of the
image. This provides quick, precise, and very repeatable Placido’s disk is captured, and the slope (distance) of the
data with excellent image centration. The instrument has rings is computed for calculating the topographical maps.
a software link with the Alcon WaveLight excimer laser The distance between the reflected rings will be smaller in
platform for custom topographic-guided treatments.135-137 steeper corneas and larger in flatter surfaces. The integrated
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 91

Figure 5-97. WaveLight Allegretto Topolyzer system.

Figure 5-99. Raw photokeratoscopic photo of the Placido
disk’s reflection from the cornea, together with the kera-
tometer marks. Note how the nose (lower left) and the eye-
lashes together with the eyebrows (superiorly) form a shadow,
decreasing the number of detected ring edges, thus the amount
of measured data.

Figure 5-98. Red and blue lines represent the edges

of the rings as detected by software. Smaller distance
between the lines represents steeper areas of the cor-
nea, while larger distance represents flatter surfaces.

keratometer enables the Topolyzer to have an accuracy

of 0.1 D for keratometric readings (15 µm). The range of
measurements is between 9 and 99 D (3- to 38-mm radius). Figure 5-100. Automatic release comprises 3 view windows:
There is a calibration routine using a model eye with a left window for rough x, y, and z centration and middle
radius of 8.00 mm, which should be repeated periodically and right window for automatic release, together with the
according to the manufacturer’s recommendations. alignment instructions in the form of an arrow for correct
focusing.

Mea su rem en t Tech n iq u e

If the case is considered for customized topography-
During the examination, the patient should be comfort- guided ablation treatment (T-CAT), at least 4 (and up to 8)
ably positioned at the chin and forehead support, with eyes measurements should be taken sequentially (Figure 5-101).
aligned in the direction of the side marking line. The patient In this case, only the measurements taken automatically
is asked to blink normally, open his or her eyes, and look at would be eligible for export and further analysis for the
the fixation target in the center of the rings. The alignment topography-guided customized treatments on the treatment
and focusing are facilitated by the integrated autokeratom- planner portal software of the WaveLight Allegretto Wave
etry, which guides for perfect positioning, so that it auto- excimer laser.
matically takes the image (Figure 5-100). In cases of poor It is important to note some important parameters of
quality of the measured surface where automatic imaging is quality to proceed with T-CAT. The pupil should be identi-
not possible, one can take the measurement with the manual fied by a black and white circle, ensuring that the software
activation arbitrary to the examiner. As with any Placido did not erroneously recognize the real edge of the pupil
disk-based topographer, Topolyzer measurements are sensi- as one of the inner circles (switch off the visualization
tive to tear film deficiency and irregularities; therefore, a of the corneal rings and only check for the pupil margin
strict measurement protocol and regular blinking is impor- detection).
tant for correct measurements.
92 Ch a p t e r 5

Figure 5-101. Repeatability demonstration “in vivo.”

Figure 5-102. Photokeratoscopic photo of the Placido

disks’ reflection from the cornea, together with the
software edge detection analysis: pupil (center and
diameter), limbus (iris center and diameter), and ring
edges of the Placido disks’ reflections.

Another parameter is the percentage of the imaged image. The corneal apex is defined in cartesian polar coor-
or measured area compared to the full available image dinates as x = 0 and y = 0. This is the center of the Placido’s
(abbreviation used by the software is analyzed area [AA]). reflex, and it is very important to have the patient adequate-
Depending on the settings, the software either extrapolates ly fixating to the fixation light inside the central inner ring.
the area that is not really measured, based on the surround- The pupil is identified, and its center is positioned accord-
ing measured ones, marking that area with black points, or ingly to the apex. The limbus edge (or iris edge, as defined
it will simply be omitted and left as a white area without by the software) is also identified, and the best circle is fit
any information. such that its center is also characterized (see Figure 5-102).
Only the area with uninterrupted Placido rings (Figure The overview display provides most of the relevant
5-102) and without the black dots will be considered for clinical information (Figure 5-103). This display contains
T-CAT. The measured area should be at least 60%. As this the photokeratoscopic image with the edge detection layer
figure strongly depends on preoperative Ks, as well as the (upper left) along with the sagittal or tangential curvature
surface quality, one should aim to have at least enough data map (upper right). The real (not simulated) central kerato-
to cover the area of the intended ablation optical zone, typi- metric readings from a built-in keratometer for horizontal
cally 6.0, or 0.5-mm smaller in cases of the patients with and vertical axis (or steeper and flatter, depending on the
smaller pupils. As an additional safety measure, the treat- software settings), together with the astigmatism and its
ment planner portal software will automatically exclude axis, are also displayed (lower left) and color coded: any
measurements not providing adequate amounts of data for data related to the flatter axis are displayed in blue, while
the desired optical zone. data showing the steeper axis are marked in red. Next is
the white-to-white measurement of the limbus (or corneal
diameter, as defined by the software), and finally the cor-
Disp lay Mo d es neal eccentricity (ecc), another conic constant to describe
asphericity (Q). The formula for converting ecc to Q is as
The Topolyzer exam provides relevant clinical informa- follows:
tion, starting from the evaluation of the photokeratoscopic
Q = -ecc2
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 93

Figure 5-103. O verview display—an example of pure myopic Figure 5-105. O verview display—an example of the patient
astigmatism, sagittal map with relative O CULUS color scale in Figure 5-104 3 days after the implantation of 2 intracorneal
and 0.25-D interval. ring segments of 200 and 250 µm (nasally and temporally,
respectively). Tangential map shows absolute O CULUS color
scale and 0.50-D interval. Note the remarkable decrease of the
corneal curvatures and astigmatism, as well as formation of a
relatively regular optical zone of approximately 4.5 mm on the
keratometric profile map (lower left).

Figure 5-104. O verview display—an example of keratoco-

nus, tangential map with absolute O CULUS color scale, and
Figure 5-106. O verview display—an example of myopic LASIK
0.50-D interval. Note the distortion of the rings on the raw
of -5.00 D. The tangential map shows absolute O CULUS color
image (upper left).
scale and 0.50-D interval. Note a regular optical zone of
approximately 5.5 mm represented as a straight line on the
keratometric profile map (lower left).
The keratometric profiles along the major orthogonal
axis calculated along all of the measured corneal diameter
are also provided (lower left corner), together with kerato- used for correct ablation zone centration planning in case
metric readings at 3-, 5-, and 7-mm zones (lower right cor- of excimer laser treatment.
ner). The Topolyzer software performs a Fourier analysis, a
Topolyzer is capable of recognizing quite complex cor- mathematical process that decomposes a periodic function
neal shapes, such as keratoconus (Figure 5-104), cornea in a number of sine and cosine waves. The topographic
with intracorneal ring implanted for KC of the same patient map generated by the Topolyzer can be divided into indi-
(Figure 5-105), cornea after myopic LASIK (Figure 5-106), vidual components by Fourier analysis, allowing the study
and cornea after hyperopic LASIK (Figure 5-107). Cases of spherical equivalent, decentralization, regular astig-
of elevated hyperopic astigmatism with a very high kappa matism, and irregularities from a single captured image
angle are adequately detected (Figure 5-108) and can be (Figure 5-109).
94 Ch a p t e r 5

Figure 5-107. O verview display—an example of myopic Figure 5-109. Fourier display—topography of a significant
LASIK of +4.0 D, tangential map shows absolute O CULUS superior decentration of a myopic LASIK treatment. (Map on
color scale and 0.50-D interval. Note a regular optical zone the left with the absolute O CULUS color scale and 0.50-D
of approximately 4.5 mm represented as a straight line on the interval next to it.) The same curvature map is decomposed
keratometric profile map (lower left). Deflection circle of a into 4 different components: spherical (upper middle), decen-
hyperopic transition zone can be seen as a circular area of tration (upper right), regular astigmatism (lower middle), and
the missing rings on the raw image (upper left). That area is irregularities (lower right). Note a significant decentration
erroneously interpreted by the software as extremely flat and, component, while other components are quite low.
therefore, color-coded with light blue on the curvature map
(upper right).

Figure 5-110. Zernike display—topography of a significant

superior decentration of a myopic LASIK treatment (same
Figure 5-108. O verview display—an example of compound
patient seen in Figure 5-110). The curvature map is decom-
hyperopic astigmatism of Sph: +1.0 D and Cyl: +4.0 D @ 175.
posed into Zernike coefficients expressed in mm (vector mode
Tangential map shows absolute O CULUS color scale and
and 3-dimensional view chosen on this picture). Note a gross
0.50-D interval. Note a significant kappa angle, represented by
decentration described by significant tilt and coma coeffi-
a line between the center of the pupil (cross) and the corneal
cients of 8.22 and 1.99 µm, respectively, and some additional
apex (the central dot) on the raw image (upper left).
irregularities of 3-foil, 4-foil, and 5-foil, while other coefficients
remained relatively low.

The Topolyzer exam also provides an analysis of cor-

neal wavefront. Zernike polynomial decomposition of the Keratometric indices are calculated to provide artificial
anterior surface of the cornea provides lower-order aber- intelligence to detect and stage keratoconus (Figure 5-111).
rations and HOAs. The understanding of corneal SAs may A contact lens-fitting software module with a big database
be used for customizing aspheric IOLs in cataract surgery. for different RGP and soft contact lenses is also available
Also, coma and other HOA terms may be useful for diag- (Figure 5-112).
nosing keratoconus, irregular ablation, and other irregular
conditions of the cornea (Figure 5-110).
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 95

Figure 5-111. Keratoconus indices display—topography of a Figure 5-113. Height map display—topography (upper right)
mild case; together with the curvature map (upper right with of a significant superior decentration of a myopic LASIK treat-
the absolute O CULUS color scale and 0.50-D interval next ment (the same as in Figure 5-109); software automatically
to it), a map of the decentration component from the Fourier chooses the best-fit reference toric ellipsoid for a “pinned”
analysis is displayed (upper left). In the upper middle, between fit, where the reference shape touches the corneal map in its
the maps, a series of indices with their respective values is center (an asphere with a radius of 0.27 mm and eccentricity
displayed (red being outside normal limits), as well as the clas- of 0.26 in this example).
sification of the keratoconus level (KC stage 1-2 in this case).

The Topolyzer is a topography system efficient in gener-

ating data and useful maps for diagnosis of corneal diseases,
but one of the most important applications of this equipment
is the ability to plan and export data to the Alcon WaveLight
laser systems.135,138,139 For this reason, expression of data
as height maps with the pupil center data is essential (Figure
5-113). The main indications for customized topography-
guided ablations using the Topolyzer are primary cases of
high astigmatism, especially those with high kappa angle,
irregular astigmatism due to corneal scars and retreatment
of a small optical zone, decentered ablation and postradial
keratotomy, and penetrating or lamellar keratoplasty. Data
validation for quality, repeatability, and proper centration
prior to exporting, as described earlier (measurement tech-
nique), is essential for predictable outcomes.
The Topolyzer is a topography system that is a power-
Figure 5-112. Contact lens display—fluorescein trial simula- ful diagnostic tool, with good accuracy and repeatability,
tion of a fit for a mild keratoconus (the same as in Figure 5-111) which are very important for correct and successful mea-
with a RGP CL (r = 7.70, eccentricity of 0.45 and the outer surement of various natural and iatrogenic corneal shapes.
diameter of 9.80 mm). Moreover, one of the most important applications of the
Topolyzer is planning customized ablations for LVCs with
the Alcon WaveLight laser systems and T-CAT treatment
planner portal software.
96 Ch a p t e r 5

EYESYS 30 0 0 AND EYESYS VISTA

Sonya M. Dakin, COT and Joe S. Wakil, MD

T EyeSys Design Ph ilo sop h y

he EyeSys 3000 system manufactured by EyeSys
Vision Inc (Houston, TX) is a patented multiresolu-
tion Placido technology along with high-resolution The EyeSys 3000 uses a multiresolution Placido technol-
4-camera optics that provides unparalleled corneal mea- ogy to ensure the most reliable and reproducible informa-
surement reliability and precision.140 The multiple-resolu- tion possible. The Vista CT is designed to be detached and
tion corneal reflection pattern encodes reflection features used with a laptop computer as a stand-alone, portable,
in such a way as to automatically adjust the measurement hand-held corneal topography unit with full software ben-
resolution depending upon the curvature of the surface. The efits. The EyeSys system was developed in 1986 as an
downfall of most CTs on the market today is the fact that industry leader, and with its most recent release of the
they use a standard number of rings to identify and quantify EyeSys 3000 and Vista, it is still an industry leader. The
all corneal abnormalities. In certain cases, subtle changes EyeSys Vista CT is designed for convenient handheld use
are overlooked because they are lost in the “red sea” due to and has special electronic features such as auto-capture and
extremely steep keratometric readings.141 In high curvature tilt sensing along with auto-correction to ensure accurate
areas, high-frequency patterns can merge together so that and precise corneal topography and keratometric data on
they cannot be reliably resolved while low-frequency pat- par with any tabletop unit. This includes using the system in
terns stay far enough apart so that a measurement can be a vertical mode on supine patients, such as during surgery
made. In low curvature areas, patterns spread out so that (ie, to document flap orientation and condition at the end of
high-resolution patterns are required to maintain high spa- standard LASIK procedures).
tial sampling. The EyeSys Placido is noted for using edge detection of
The EyeSys 3000’s technology offers the customization the Placido mires to provide precise determination of ring
based on the overall keratometric condition of the cornea. location for highly accurate corneal topography measure-
For instance, for a relatively high curvature corneal area ments. The Placido design of the Vista is the same as that of
(keratoconus), the EyeSys needs only 14 rings to accurately the larger EyeSys 3000 system, covering the central cornea
identify and display the keratometric characteristics. When at around 0.9 mm to over 10 mm in the normal peripheral
evaluating a moderate-curvature cornea (normal eyes), the cornea. This larger Placido design provides for a longer
EyeSys increases the number of rings to 79, and, lastly, working distance than smaller cylindrical Placido cones.
the very low-curvature corneas increase the rings to the The longer 12.5-mm working distance has the advantage
maximum number, which are 131 rings. By adapting the of being less sensitive to positional errors in focusing and
resolution of these areas, the EyeSys 3000 can increase alignment that significantly improves the reproducibility
measurement reliability and precision using techniques and accuracy of the system. The EyeSys Vista hardware has
that are not possible with single-resolution reflection pat- this advantage working in its favor under both conditions—
terns.142 The EyeSys 3000 also has the ability to locate the mounted on the slit-lamp base and when used in a handheld
pupil’s centroid and measures its size by using a dedicated manner. The reproducibility results are impressive at less
IR pupil camera. than ±0.10 D in the hands of an experienced user.
Along with the EyeSys 3000, EyeSys Vision also produc- It is important to always remember that, with corneal
es an EyeSys Vista unit, which brings new freedom to cor- topography measurements, one is measuring the anterior
neal topography while retaining the quality and excellence tear film surface of the cornea—truly the “lens of the eye.”
that are the hallmark of EyeSys products. The Vista offers This is the refractive surface that provides the vast major-
eye-care practitioners a simple affordable and dependable ity (>75%) of the eye’s refractive power. Please note that
tool for precise corneal topography. The portability and although some topography systems emphasize that the cor-
unparalleled versatility of the EyeSys Vista enables it to nea has anterior and posterior surfaces, this can be mislead-
perform topography in all clinical environments. Patients ing because, optically, with tear film on the living cornea,
confined to a bed or wheelchair can easily be examined, there are actually 3 refractive interfaces: air/tear film, tear
and supine patients can be mapped in the operating room film/corneal stroma, and corneal stroma/aqueous humor.
under a surgical microscope. Effectively, nature has provided for these last 2 refractive
interfaces of the cornea to practically cancel each other
out, leaving the anterior tear film as the effective refractive
power interface of the cornea.
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 97

Figure 5-114. Main topography display. Figure 5-115. Holladay display. Note the corneal parameters
listed at the bottom: Sim Ks, radial power values, pupil mea-
surements, asphericity, CU index, I-S value, and PC acuity.
To ensure ideal vision correction, matching the available
surgical procedures or corrective lens choices in addressing
each patient’s visual deficits depends heavily upon perform- corneal indices to comprehensively analyze the cornea and
ing the correct diagnosis of the aberration sources in each simplify corneal diagnosis (Figure 5-115). These indices
patient’s eyes. The EyeSys is designed to meet this goal for include the inferior-superior (I-S) index, a corneal unifor-
every patient in an easy-to-use and comprehensive format, mity index (CUI), an aspheric Q factor of the cornea, and
creating a new standard of visual function analysis well the unique potential corneal acuity (PCA) measurement that
beyond the antiquated Snellen refraction. provides a direct objective measure of mire quality related
to the tear film of the eye, in effect helping measure poten-
tial dry eye problems.
Disp lays Beyond the numeric data of K readings, refractive
power readings, and other corneal indices, color maps
The corneal topography calculations and displays are are provided to depict the unique features of each cornea.
generated from the EyeSys 3000 and Vista Placido image of With the 2-dimensional images, there are a number of
the cornea and are fully customizable within the software. algorithms available, providing complete corneal analysis.
Using advanced edge detection software, the Placido image These include the Axial Map, Local Radius of Curvature
of the cornea is analyzed across all of the ring edges from Map (Tangential), Refractive Map, Z Elevation Map, High-
center to periphery. Along with patient name and identifi- Order Aberration, Contact Lens Map, and Wavefront Map
cation number, OD/OS label, and time of exam, standard (Figures 5-116 and 5-117).
keratometric readings are generated at the 3-mm zone of The Vista hardware is used to capture the Placido image
the cornea to accurately measure its central curvature and automatically upon centering and focusing of the video
simulate keratometer measurements for routine use. These image as the patient fixates upon a coaxial fixation point.
K readings are provided with every corneal topography This automatic capture occurs through triangulation of a
display, as shown in Figure 5-114. low-power laser beam reflected off of the corneal vertex at
Additionally, a refractive power reading of the cornea a calibrated distance from the Placido. Once captured, the
is calculated for the 3-mm zone in similar fashion to the Placido image is displayed to the user with the ring edge
K readings, but based on Snell’s law of refraction to more detection highlighted, so that the user can easily confirm
accurately describe the refractive power of the central cor- correct image processing. This helps avoid artifacts and
nea. A single effective refractive power reading is given errors in processing that can sometimes be problematic
for the entire central 3-mm zone of the cornea to be used in very irregular corneas, such as those following corneal
primarily for IOL calculations. This number has been transplantation. Figure 5-118 shows a typical Placido image
popularized by Jack Holladay, MD, as an improvement in capture verification screen with the edge detection clearly
understanding central corneal refractive power contribution shown, confirming correct analysis to the user. Also, the
in more accurately calculating the necessary IOL power user can obtain the K readings and refractive power read-
for eyes postrefractive surgery. The Holladay Diagnostic ings from this immediate display following image capture.
Summary available with the compatible EyeSys software The Corneal Topography Summary Displays can be cus-
package to the Vista hardware has become an industry tomized to allow for any of the color maps available to be
standard—a single-page display that provides additional displayed with up to 4 maps at once.143 The EyeSys Vista
98 Ch a p t e r 5

Figure 5-116. Comprehensive mapping display including Figure 5-118. Verification display. The clinician is able to
tangential, refractive, elevation, and corneal wavefront aber- check the rings to ensure the capture if well done.
rations maps.

Figure 5-119. 1.5 D of with-the-rule astigmatism.

Figure 5-117. Comprehensive mapping display showing infe-
rior steepening and coma on the wavefront map. The I-S value
is 1.4, and this is suspicious for ectasia.
A Comparison Display Map is also available on each of
the different color map displays available with both EyeSys
systems. These maps are informative in revealing the differ-
as well as the EyeSys 3000 is generally configured to show
ences that occur over time (ie, pre- and postsurgery), as well
the Axial Map, Local Radius of Curvature Map, Refractive
as over periods of healing or treatment. Figure 5-120 shows
Map, and Z Elevation Map.144 The Corneal Wavefront Map
corneal stabilization after contact lens removal in a patient
can be added as desired and positioned in any order, so there
planning for excimer laser surgery. The refractive power
is full flexibility. The example of the corneal topography
difference between these 2 exams is much less than 0.25 D
summary display in Figure 5-119 shows almost 4 D of with-
over most of the corneal surface, therefore indicating that
the-rule astigmatism, where the steep corneal curvature is
the cornea has changed insignificantly between the 2 days
oriented vertically. As with all color maps, it is important
examined. The difference map is overall green, indicating
to pay attention to scaling. With the EyeSys Vista as well as
little or no change in corneal topography.
the EyeSys 3000, there is a choice to allow for auto-scaling
The Holladay Diagnostic Summary shown in Figure
features or user-defined fixed scaling units. Typically,
5-121 demonstrates the power of corneal topography in
0.5-D step sizes are used to fulfill most of the needs to
diagnosing an early keratoconus case. This unique display
provide good corneal detail while also providing adequate
generated by EyeSys software has been widely used for
range to most corneas. User choice is available to generate
more than a decade in providing a single-page overview of
practically all scaling requirements.
the cornea with a number of valuable corneal indices. The
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 99

Figure 5-120. The comparison map compares topography over Figure 5-121. Holladay diagnostic summary in a patient with
time, such as in this case of corneal stabilization after contact early keratoconus.
lens removal in a patient planning for excimer laser surgery.

keratoconus as tear film break-up and irregularity over the

cornea displayed shows the classic pattern of inferior steep- ectasia is typical. The corneal indices at the bottom of the
ening in the Local Radius of Curvature Map in the upper Holladay Diagnostic Summary are consistent with kera-
right map and also a correlating change in the Asphericity toconus having an I-S index greater than 1.0 (in this case
Difference Map in the lower left demonstrating that the 4.79) and a CUI of less than 60% (in this case 20%) with a
area of the keratoconus significantly alters the normal rate degraded PCA of 20/100. With the data presented, a clini-
of flattening of the cornea from center to periphery. The cian only needs to complete the patient history to rule-out
Potential Corneal Acuity Map in the lower right reveals contact lens warpage in making the keratoconus diagnosis
a pattern of inferior mire distortion expected in classic with a high degree of certainty.

R EFERENCES
1. McKay T. A Clinical Guide to the Humphrey Corneal Topography 10. Gormley DJ, Gersten M, Koplin RS, Lubkin V. Corneal modeling.
System. Dublin, CA: Humphrey; 1998; Section 2:4-6. Cornea. 1988;7:30-35.
2. ATLAS Corneal Topography System: Model 9000 Instrument and 11. Mammone RJ, Gersten M, Gormley DJ, Koplin RS, Lubkin VL. 3-
ATLAS Review Software [computer program]. Version 3.0. Jena, D corneal modeling system. IEEE Trans Biomed Eng. 1990;37:66-
Germany: Carl Zeiss Meditec; 2007: Section 3:9. 72.
3. Anderson D, Kojima R. Topography: a clinical pearl. Optom 12. Klyce SD. Computer-assisted corneal topography: high resolu-
Management. 2007;42(2):35. tion graphical presentation and analysis of keratoscopy. Invest
4. Roberts C. A practical guide to the interpretation of corneal topog- Ophthalmol Vis Sci. 1984;25:1426-1435.
raphy. CL Spectrum. 1998;13(3):25-33. 13. Maguire LJ, Singer DE, Klyce SD. Graphic presentation of com-
5. Navarro R, Gonzalez L, Hernandez JL. Optics of the average nor- puter-analyzed keratoscope photographs. Arch Ophthalmol.
mal cornea from general and canonical representations of its surface 1987;105:223-230.
topography. J Opt Soc Am A Opt Image Sci Vis. 2006;23(2):219- 14. Swartz T, Marten L, Wang M. Measuring the cornea: the latest
232. developments in corneal topography. Curr Opin Ophthalmol.
6. ATLAS Corneal Topography System: Model 9000 Instrument and 2007;18:325-333.
ATLAS Review Software [computer program]. Version 3.0. Jena, 15. Konstantopoulos A, Hossain P, Anderson DF. Recent advances in
Germany: Carl Zeiss Meditec; 2007: Section A:1-18. ophthalmic anterior segment imaging: a new era for ophthalmic
7. Bagherinia H, Chen X, Flachenecker C, et al. Support vector diagnosis? Br J Ophthalmol. 2007;91:551-557.
machine (SVM)-based classification of corneal topography. Poster 16. Wilson SE, Klyce SD, Husseini ZM. Standardized color-coded
presented at: Association for Research in Vision and Ophthalmology maps for corneal topography. Ophthalmology. 1993;100:1723-1727.
(ARVO) Annual Meeting; April 27-May 1, 2008; Ft. Lauderdale, 17. Dingeldein SA, Klyce SD, Wilson SE. Quantitative descriptors of
FL: 1023/D879. corneal shape derived from computer-assisted analysis of photoker-
8. Noble WS. What is a support vector machine? Nature Biotech. atographs. Refractive & Corneal Surgery. 1989;5:372-378.
2006;24(12):1565-1567. 18. Wilson SE, Klyce SD. Quantitative descriptors of corneal topogra-
9. ATLAS Corneal Topography System: Model 9000 Instrument and phy: a clinical study. Arch Ophthalmol. 1991;109:349-353.
ATLAS Review Software [computer program]. Version 3.0. Jena, 19. Alimisi S, Miltsakakis D, Klyce S. Corneal topography for intra-
Germany: Carl Zeiss Meditec; 2007: Section B:1-30. ocular lens power calculations. J Refract Surg. 1996;12:S309-S311.
100 Ch a p t e r 5

20. Maeda N, Klyce SD, Smolek MK, McDonald MB. Disparity of 41. Rajan MS, Jaycock P, O’Brart D, et al. A long term study of
keratometry readings and corneal power within the pupil after photorefractive keratectomy; 12 year followup. Ophthalmology.
refractive surgery for myopia. Cornea. 1997;16:517-524. 2004;111:1813-1824.
21. Mafra CH, Dave A, Wilson SE, Klyce SD. Extracapsular cataract 42. Orbscan Keratoconus Screening Indices: Operators Manual and
surgery: corneal topographic alterations. Invest Ophthalmol Vis Sci. Procedural Recommendations. Rochester, NY: Bausch & Lomb;
1993;34:1248. July 2004.
22. Maeda N, Klyce SD, Smolek MK, Thompson HW. Automated 43. Wilson SE, Lin DTC, Klyce SD, et al. Topographic changes in con-
keratoconus screening with corneal topography analysis. Invest tact lens-induced corneal warpage. Ophthalmology. 1990;97:734.
Ophthalmol Vis Sci. 1994;35:2749-2757. 44. McLellan JS, Marco S, Prieto PM, Burns SA. Imperfect optics may
23. Randleman JB, Russell B, Ward MA, Thompson KP, Stulting be the eye’s defense against chromatic blur. Nature. 2002;417:174-
RD. Risk factors and prognosis for corneal ectasia after LASIK. 176.
Ophthalmology. 2003;110:267-275. 45. Marcos S, Barbero S, Jimenez-Alfaro I. Optical quality and depth
24. Holladay JT. Keratoconus detection using corneal topography. of field of eyes implanted with spherical and aspheric intraocular
J Refract Surg. 2009;25:S958-S962. lenses. J Refract Surg. 2005;21:223-235.
25. Klyce SD. Chasing the suspect: keratoconus. Br J Ophthalmol. 46. Wang L, Dai E, Koch DD, Nathoo A. Optical aberrations of the
2009;93:845-847. human anterior cornea. J Cataract and Refract Surg. 2003;29:1514-
26. Smolek MK, Klyce SD. Current keratoconus detection methods 1521.
compared with a neural network approach. Invest Ophthalmol Vis 47. Price FW Jr. How important is it to reduce or eliminate spheri-
Sci. 1997;38:2290-2299. cal aberration and is there an advantage to having some present?
27. Hjortdal JO, Erdmann L, Bek T. Fourier analysis of video-kerato- In: Chang DF, ed. Curbside Consultation in Cataract Surgery:
graphic data. A tool for separation of spherical, regular astigmatic 49 Clinical Questions. Thorofare, NJ: SLACK Incorporated;
and irregular astigmatic corneal power components. Ophthalmic 2007:37-42.
Physiol Opt. 1995;15:171-185. 48. Levy Y, Segal O, Avni I, Zadok D. Ocular higher order aberrations
28. Olsen T, Dam-Johansen M, Bek T, Hjortdal JO. Evaluating surgi- in eyes with supranormal vision. Am J Ophthalmol. 2005;139:225-
cally induced astigmatism by Fourier analysis of corneal topogra- 228.
phy data. J Cataract Refract Surg. 1996;22:318-323. 49. Grimson JM, Schallhorn SC. Contrast sensitivity: establishing
29. Tomidokoro A, Soya K, Miyata K, et al. Corneal irregular astig- normative data for use in screening prospective naval pilots. Aviat
matism and contrast sensitivity after photorefractive keratectomy. Space Environ Med. 2002;73:28-35.
Ophthalmology. 2001;108:2209-2212. 50. Wilson SE, Wang JY, Klyce SD. Quantification and mathematical
30. Oshika T, Tanabe T, Tomidokoro A, Amano S. Progression of analysis of photokeratoscopic images. In: Shanzlin DJ, Robin JB,
keratoconus assessed by Fourier analysis of videokeratography data. eds. Corneal Topography: Measuring and Modifying the Cornea.
Ophthalmology. 2002;109:339-342. New York, NY: Springer-Verlag; 1991:1-9.
31. Tanabe T, Tomidokoro A, Samejima T, et al. Corneal regular and 51. Klyce SD, Wang JY. Considerations in corneal surface reconstruc-
irregular astigmatism assessed by Fourier analysis of videoker- tion from keratoscope images. In: Masters BR, ed. Noninvasive
atography data in normal and pathologic eyes. Ophthalmology. Diagnostic Techniques in Ophthalmology. New York, NY: Springer-
2004;111:752-757. Verlag; 1990:76.
32. Hiraoka T, Furuya A, Matsumoto Y, et al. Quantitative evaluation 52. Merlin U. I cheratoscopi: caratteristiche e attendibilita. In: Buratto
of regular and irregular corneal astigmatism in patients having L, Cantera E, Dal Fiume E, Genisi C, Merlin U, eds. Topografia
overnight orthokeratology. J Cataract Refract Surg. 2004;30:1425- Corneale. Milano, Italy: CAMO; 1995:43-56.
1429. 53. Klein SA, Mandell RB. Axial and instantaneous power conversion
33. Miyai T, Miyata K, Nejima R, et al. Visual acuity measurements in corneal topography. Invest Ophthalmol Vis Sci. 1995;36:2155-
using Fourier series harmonic analysis of videokeratography data in 2159.
eyes after penetrating keratoplasty. Ophthalmology. 2005;112:420- 54. Cohen KL, Tripoli NK, Holmgren DE, Coggins JM. Assessment
424. of the power and height of radial aspheres reported by a computer-
34. Szczotka LB, Reinhart W. Computerized videokeratoscopy contact assisted keratoscope. Am J Ophthalmol. 1995;119:723-732.
lens software for RGP fitting in a bilateral postkeratoplasty patient: 55. Roberts C. Characterization of the inherent error in a spherically-
a clinical case report. CLAO J. 1995;21:52-56. biased corneal topography system in mapping a radially aspheric
35. Bufidis T, Konstas AG, Mamtziou E. The role of computerized cor- surface. J Refract Corneal Surg. 1994;10:103-116.
neal topography in rigid gas permeable contact lens fitting. CLAO 56. Koch DD, Foulks GN, Moran CT, Wakil JS. The corneal EyeSys
J. 1998;24:206-209. system: accuracy analysis and reproducibility of first-generation
36. Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assess- prototype. J Refract Corneal Surg. 1989;6:423-429.
ment for ectasia after corneal refractive surgery. Ophthalmology. 57. Gullstrand A. Photographic-ophthalmometric and clinical
2008;115(1):37-50. Investigations of corneal refraction. Am J Optom Arch Am Acad
37. Bellucci R, Morselli S. Optimizing higher-order aberrations with Optom. 1966;43:143-214.
intraocular lens technology. Curr Opin Ophthalmol. 2007;18(1):67- 58. Doss JD, Hutson RL, Rowsey JJ, Brown R. Method for calcula-
73. tion of corneal profile and power distribution. Arch Ophthalmol.
38. Endl MJ. Contrast sensitivity & patient satisfaction: Softport AO 1981;99:1261-1265.
Aberration-Free IOL vs. a wavefront-designed IOL. Paper presented 59. Wang J, Rice DA, Klyce SD. A new reconstruction algorithm for
at: American Society of Cataract and Refractive Surgery Annual improvement of corneal topographical analysis. J Corneal Refract
Meeting; March 17-22, 2006; San Francisco, CA. Surg. 1989;5:379-387.
39. Fichte CM. A one year follow-up comparing two higher order 60. van Saarloos PP, Constable IJ. Improved method for calculation of
aberration IOLs. In: Current Concepts in Ophthalmology. Paper corneal topography for any photokeratoscopic geometry. Optom and
presented at: Johns Hopkins Ophthalmology Meeting; March 10-12, Vis Sci. 1991;68:957-965.
2007; Vail, CO. 61. Klein SA. A corneal topography algorithm that produces continuous
40. American Academy of Ophthalmology. Basic and Clinical Science curvature. Optom Vis Sci. 1002;69:829-834.
Course, Section 8: External Disease and Cornea. San Francisco, 62. Campbell C. Reconstruction of the corneal shape with the MasterVue
CA: Author; 2001-2002:305. corneal topography system. Optom Vis Sci. 1997;74:901.
 Pla c id o D isk-Ba se d To p o gra p h ic Syst e m s 101

63. Mattioli R, Carones F, Cantera E. New algorithms to improve the 85. Vinciguerra P, Camesasca FI. Smoothing in Excimer refractive
reconstruction of corneal geometry on the Keratron videokeratog- surgery. In: Vinciguerra P, Camasasca FI, eds. Refractive Surface
rapher. Invest Ophthalmol Vis Sci (suppl). 1995;36:1400. Ablation. Thorofare, NJ; SLACK Incorporated; 2007:3-16.
64. El Hage SG. The computerized corneal topographer EH-270. In: 86. Mahmoud AM, Roberts C, Herderick EE, Lembach RG, Markakis G.
Shanzlin DJ, Robin JB, eds. Corneal Topography: Measuring and The cone location and magnitude index (CLMI). Invest Ophthalmol
Modifying the Cornea. New York, NY: Springer-Verlag; 1991:11-24. Vis Sci. 2001;42:S898.
65. El-Hage SG. Suggested new methods for photokeratoscopy: a com- 87. Twa MD, Parthasarathy S, Roberts C, Mahmoud AM, Raasch TW,
parison of their validities. I. Am J Optom Arch Am Acad Optom. Bullimore MA. Automated decision tree classification of corneal
1971;48:897-912. shape. Optom Vis Sci. 2005;82(12):1038-1046.
66. Brancato R, Carones F. Topografia Corneale Computerizzata. 88. Mahmoud AM, Roberts CJ, Lembach RG, Twa MD, Herderick EE,
Milan, Italy: Fogliazza; 1994:2.12-17.3. McMahon TT; the CLEK Study Group. CLMI: The cone location
67. Chan WK, Carones F, Maloney RK. Corneal topographic maps: and magnitude index. Cornea. 2008:27(4):480-487.
a clinical comparison. In: International Society of Refractive 89. Mattioli R. CLMI (Cone Location + Magnitude) Indices in
Keratoplasty—Abstract Book. Paper presented at: Pre-Academy the Keratron Scout. Milano, Italy: AIM Congress srl - AIM
meeting of the International Society of Refractive Keratoplasty; Group. September 2005. http://www.refractiveonline.com/2005/
October 28-November 3, 1994; San Francisco, CA. eng/38mattioli/player.html. Accessed August 8, 2011.
68. Mattioli R, Tripoli N. Corneal geometry reconstruction with the 90. Mattioli R. The Gaussian curvature map. Presented at
Keratron videokeratographer. Optom Vis Sci. 1997;74:881-894. Refractiveonline, Milano, Italy: September 2007. www.refracti-
69. Hannush SB, Crawford SL, Waring GO III, Gemmill MC, Lynn MJ, veonline.com/2007/eng/04mattioli/player.html. Accessed August 8,
Nizam A. Accuracy and precision of keratometry, photokeratoscopy 2011.
and corneal modeling on calibrated steel balls. Arch Ophthalmol. 91. Barsky BA, Klein S, Garcia DD. Gaussian power, mean sphere
1989;107:1235-1239. and cylinder representations for corneal maps with applica-
70. Maguire LJ, Wilson SE, Camp JJ, Verity S. Evaluating the repro- tions to the diagnosis of keratoconus. Invest Ophthalmol Vis Sci.
ducibility of topography systems on spherical surfaces. Arch 1996;37(suppl):558.
Ophthalmol. 1993;111:259-262. 92. Tang M, Shekhar R, Miranda D, Huang D. Characteristics of kerato-
71. Wilson SE, Verity SM, Conger DL. Accuracy and precision of the conus and pellucid marginal degeneration in mean curvature maps.
Corneal Analysis System and the Topographic Modeling System. Am J Ophthalmol. 2005;140:993-1000.
Cornea. 1992;11:28-35. 93. Mattioli R. Representation of corneal wavefront from corneal
72. Carones F, Gobbi PG, Brancato R, Venturi E. Comparison between topography. Milano, Italy: AIM Congress srl - AIM Group.
two computer-assisted keratoscopes in measuring aspheric surfaces. September 2001. http://www.refractiveonline.com/eng/pages/edi-
Invest Ophtalmol Vis Sci (suppl). 1994:3748. tion.aspx?anno=2001. Accessed date August 8, 2011.
73. Tripoli NK, Cohen KL, Holmgren DE, Coggins JM. Assessment 94. Mattioli R. Corneal topography and wavefront analysis. In: Caimi
of radial aspheres by the Keratron keratoscope using an arc-step R, Brancato R, eds. The Aberrometers. Milan, Italy: Fabiano;
algorithm. Am J Ophthalmol. 1995;120:658-664. 2003:113-132.
74. Cohen KL, Tripoli NK, Holmgren DE, Coggins JM. Assessment of 95. Mattioli R, Frondizi S, Rubeo A, Saldutti M. The Keratron
the height of radial aspheres reported by a computer-assisted kera- Scout and wavefront analysis. In: Caimi F, Brancato R, eds. The
toscope. Invest Ophthalmol Vis Sci (suppl). 1993;34:1217. Aberrometers. Milan, Italy: Fabiano; 2003:243-257.
75. Tripoli NK, Cohen KL, Obla P, Coggins JM, Holmgren DE. Height 96. Marcos S. Aberrations and optical quality of the eye: what the clini-
measurement of astigmatic test surfaces by a keratoscope that uses cian needs to know. In: Caimi F, Brancato R, eds. The Aberrometers.
plane geometry reconstruction. Am J Ophthalmol. 1996;121:668- Milan, Italy: Fabiano; 2003:15-26.
676. 97. Tripoli NK. The Zernike polynomials. In: Caimi F, Brancato R, eds.
76. Wang JY, Rice DA, Klyce SD. Analysis of the effects of astigmatism The Aberrometers. Milan, Italy: Fabiano; 2003:51-64.
and misalignment on corneal surface reconstruction from photok- 98. Howland H, Buettner J, Applegate R. Computation of the shapes
eratoscopic data. Refract Corneal Surg. 1991;7(2):129-140. of normal corneas and their monochromatic aberrations from vid-
77. Mattioli R, Carones F. How accurately can corneal profile heights be eokeratometric measurements. Vision science and its applications.
measured by Placido-based videokeratography? Invest Ophthalmol Technical Digest Series. 1994;2:54-57.
Vis Sci (suppl). 1996;37:4273. 99. Hemenger RP, Tomlinson A, Oliver K. Corneal optics from videok-
78. Camellin M, Federici R, Federici S. Clinical applications of the eratographs. Ophthalmic Physiol. 1995;15:63-68.
aberrometric functions of the Keratron Scout. In: Caimi F, Brancato 100. Schwiegerling J, Greivenkamp JE. Using corneal height maps and
R, eds. The Aberrometers. Canelli, Italy: Fabiano; 2003:259-268. polynomial decomposition to determine corneal aberrations. Optom
79. Dekking HM. Zur photographie der hornhautoberfl-eche. Graefes Vis Sci. 1997;74:906-916.
Arch Ophtalmol. 1930;124:708-730. 101. Marcos S. Aberrations and visual performance following standard
80. Cantera E, Carones F, Brancato R, Cantera I, Neuschuler R. laser vision correction. J Refract Surg. 2001;17:596-601.
Evaluation of a new autofocus device for computer-assisted corneal 102. Marcos S, Barbero B, Llorente L, Merayo-Lloves J. Optical
topography. Invest Ophtalmol Vis Sci. 1994;35(suppl):2063. response to LASIK for myopia from total and corneal aberrations.
81. Walsh G. The effect of mydriasis on the pupillary centration of the Invest Ophthalmol Vis Sci. 2001;42:3349-3356.
human eye. Ophthalmic Physiol Opt. 1988;8:178-182. 103. Artal P, Guirao A, Berrio E. Compensation of corneal aberrations
82. Wyatt HJ. The form of the human pupil. Vision Res. 1995;35:2021- by the internal optics in the human eye. J Vision. 2001;1:1-8.
2036. 104. Artal P, Berrio E, Guirao A, Piers P. Contribution of the cornea and
83. Mattioli R, Carones F. An algorithm to recalculate topographic internal surfaces to the change of ocular aberrations with age. J Opt
maps from any axis for computer assisted videokeratography. In: Soc Am A. 2002;19:137-143.
International Society of Refractive Surgery—Abstract Book. Paper 105. Artal P, Guirao A, Contributions of the cornea and the lens to the
presented at: International Society of Refractive Surgery Meeting; aberrations of the human eye. Optics Letters. 1998;23:1713-1715.
October 29-November 3, 1995; Atlanta, GA. 106. Guirao A, Gonzalez C, Redondo M, Geraghty E, Norrby S, Artal P.
84. Mattioli R, Federici R. Intra-operatory corneal topography—Tools Average optical performance of the human eye as a function of age
and applications. Acts from the congress “Refractive on Line,” ICH- in a normal population. Inv Ophth Vis Sci. 1999;40:203-213.
Milano, Sept. 2001. http://www.refractiveonline.com/eng/pages/
edition.aspx?anno=2001. Accessed August 8, 2011.
102 Ch a p t e r 5

107. Guirao A, Artal P. Corneal wave aberration from videokeratogra- 126. Maller K. Using CAD/CAM lenses for orthokeratology. Review of
phy: accuracy and limitations of the procedure. J Opt Soc Am A. Contact Lenses. 2004:16-20.
2000;17:955-965. 127. Linton G. Software increased one OD’s RGP fit. Contact Lens
108. Malacara D. Optical Shop Testing. 2nd ed. New York, NY: John Spectrum. 2000:60.
Wiley & Sons; 1992:472-488. 128. Brown K. Focal points: the power of computing in GP lens
109. Thibos LN, Applegate RA, Schwiegerling JT, Webb RH, VSIA design and manufacturing. Boston Update North American Edition.
Standards Taskforce Members. Standards for reporting the optical 2005;16:4-9.
aberrations of eyes, vision science and its applications. OSA Trends 129. Stojanovic A. LaserSight alternatives in topography-based custom
in Optics & Photonics. 2000;35:110-130. ablation. In: Wang M, ed. Corneal Topography in the Wavefront
110. Thibos LN. Wavefront data reporting and terminology. J Refract Era: A Guide for Clinical Application. Thorofare, NJ: SLACK
Surg. 2001;17:S578-S583. Incorporated; 2006:146-154.
111. Thibos LN, Hong X, Bradley A, Cheng X. Statistical variation of 130. D’souza HM, Sarver EJ, Wakil YS, inventors; Eyesys Technologies
aberration structure and image quality in a normal population of Inc, assignee. Method of corneal analysis using a checkered Placido
healthy eyes. J Opt Soc Am A. 2002;19:2329-2348. apparatus, US patent 5841511; November 24, 1998.
112. Smith S. Digital Signal Processing, 2nd ed. San Diego, CA: 131. Liu DD, Santiago D, Wu L, et al. Irregular astigmatism: LaserSight
California Technical Publishing; 1999. ellipsoid model and topography-driven aspheric treatment. In: Wang
113. Bonci P. Decentramenti in chirurgia rifrattiva. In: Buratto L, M, ed. Irregular Astigmatism Diagnosis and Treatment. Thorofare,
Picardo V, eds. Focal Points su Chirurgia Refrattiva. Canelli, Italy: NJ: SLACK Incorporated; 2007:195-209.
Fabiano; 2003:11-25. 132. Bueeler M, Mrochen M, Seiler T. Maximum permissible lateral
114. Arbelaez MC. Early work shows safety and efficacy. Ophthalmology decentration in aberration sensing and wavefront guided corneal
Management. 2003:56-58. ablation. J Cataract Refract Surg. 2003;29:257-263.
115. Rummelt V. Selective corneal wavefront diagnosis and excimer laser 133. Stojanovic A, Suput D. Strategic planning in topography-guided
treatment for vision improvement. Highlights of Ophthalmology. ablation of irregular astigmatism after laser refractive surgery.
2003;31(5):27-30. J Refract Surg. 2005;21(4):369-376.
116. Siganos D. Quality of vision improved with ORK-W system. Euro 134. Bueeler M, Mrochen M, Seiler T. Maximum permissible torsional
Times. May 16, 2003. http://www.escrs.org/eurotimes/May2003/ misalignment in aberration sensing and wavefront guided corneal
quality.asp. Accessed August 8, 2011. ablation. J Cataract Refract Surg. 2004;30:17-25.
117. Munnerlyn CR, Koons J. Photorefractive keratectomy: a technique 135. Falavarjani KG, Hashemi M, Modarres M, Sanjari MS, Darvish
for laser refractive surgery. J Cataract Refract Surg. 1988;14:46-52. N, Gordiz A. Topography-guided vs wavefront-optimized surface
118. Gatinel D, Hoang-Xuan T, Azar DT. Determination of corneal ablation for myopia using the WaveLight platform: a contralateral
asphericity after myopia surgery with the excimer laser: a math- eye study. J Refract Surg. 2011;27(1):13-17.
ematical model. Invest Ophthalmol & Vis Sci. 2001;42:1736. 136. Cummings AB, Mascharka N. Outcomes after topography-based
119. Mattioli R, Camellin M. La aberrometria corneale ed il link topo LASIK and LASEK with the WaveLight Oculyzer and Topolyzer
aberrometrico. In: Camellin M, ed. LASEK and ASA. Canelli, Italy: Platforms. J Refract Surg. 2009:1-8.
Fabiano; 2004:231-253. 137. Jankov MR, Panagopoulou SI, Aslanides IM, Hajitanasis GI,
120. Risoldi U, Venturini S, De Nicolò N. RGP contat lenses Horus e Pallikaris GI. Topography-guided treatments with WaveLight
topografo corneale, una soluzione integrata, moderna ed efficace. ALLEGRETTO WAVE for irregular astigmatism. J Refract Surg.
Professional Optometry. 1996;8:32-41. 2006;22(4):335-344.
121. No7 Contact Lenses. No7 Contact Lenses: Why Topography? 138. Cummings AB, Mascharka N. Outcomes after topography-based
Video. http://www.no7contactlenses.com/index.php?option=com_ LASIK and LASEK with the WaveLight Oculyzer and Topolyzer
hdflvplayer&Itemid=241&title=No7-Contact-Lenses:-Why-Topogr Platforms. J Refract Surg. 2009;Sep 2:1-8.
aphy?&compid=1&id=2&page=1. Accessed August 15, 2011. 139. Kanellopoulos AJ, Pe LH. Wavefront-guided enhancements using
122. Garcia D, Barsky B, Bolles B. The OPTICAL Project at UC Berkeley: the WaveLight excimer laser in symptomatic eyes previously treated
Computer Aided Cornea Modeling and Visualization. 1996. with LASIK. J Refract Surg. 2006;22(4):345-349.
http://www.eecs.berkeley.edu/~ddgarcia/optical/SIGGRAPH96_ 140. González Pérez J, Cerviño A, Giraldez MJ, Parafita M, Yebra-
ET.html. Pimentel E. Accuracy and precision of EyeSys and Orbscan
123. Manganotti A, Pedrotti E, Sbabo A. RGP contact lenses custom systems on calibrated spherical test surfaces. Eye Contact Lens.
made links with topographical apex saving in keratoconus. Our 2004;30(2):74-78.
experience and comparison with traditional lenses. EuVision. 141. McMahon TT, Anderson RJ, Joslin CE, Rosas GA; Collaborative
2004:6-13. Longitudinal Evaluation of Keratoconus (CLEK) Study Topography
124. Manfredini M. L’evoluzione applicativa delle lenti a contatto Analysis Group. Precision of three topography instruments in kera-
RGP: dalle lenti a contatto su misura tradizionali, alle LAC toconus subjects. Optom Vis Sci. 2001;78(8):599-604.
Custom Made realizzate mediante link informatico con topografo 142. McCarey BE, Zurawski CA, O’Shea DS. Practical aspects of a cor-
corneale EuVision. 2003:8-14. http://www.manganottioculista.it/ neal topography system. CLAO J. 1992;18(4):248-254.
Area_Professionisti/1%20APPLICAZIONE%20LAC%20e%20 143 Wang L, Koch DD. Corneal topography and its integration into
LENTI%20SPECIALI/3%20LAC%20CUSTOM-MADE%20 refractive surgery: application in refractive surgery. Medscape
da%20topografia%20corneale.pdf. Accessed August 8, 2011. Reference Web site. http://www.medscape.com/viewarticle/504198_
125. Manfredini M. Corneal topography in contact lens practice. In: 2. Accessed February 25, 2011.
Soper JW, ed. A Comprehensive Study and Reference Guide. Vol I. 144. Fung MW, Raja D, Fedor P, Kaufman SC. Corneal topography and
Herndon, VA; Contact Lens Society of America (CLSA); 2003:397- imaging. Medscape Reference Web site. http://emedicine.medscape.
408. com/article/1196836-overview. Accessed February 25, 2011.
Ch a p te r 6

To mo grap he rs

TH E BAUSCH & LOMB O RBSCAN II/II Z ANTERIOR

SEGMENT ANALYSIS SYSTEM
Paul M. Karpecki, OD, FAAO

T
he Bausch & Lomb Orbscan II/IIz Anterior Segment are used to construct mathematical representations of the
Analysis system (Rochester, NY; Figure 6-1) per- true topographic surfaces of the anterior segment, including
forms a complete anatomical analysis of the anterior maps of elevation (z) versus horizontal and vertical (x and
segment of the eye. The system combines 2 technologies: a y) coordinates. The mathematical surface representations,
calibrated video and scanning slit-beam system that mea- which typically have continuous second-order derivatives,
sures anterior segment geometry and an advanced Placido are used to calculate slope and curvature at any point and
disk system that measures the curvature of the anterior in every direction. A tracking system measures involuntary
surface of the cornea. eye movement and is used to accurately assemble the math-
During an examination, the patient fixates on a blinking ematical surface representation from the 40 slit images. The
light source, which is coaxial with a calibrated video imag- resulting calculations are used to describe 4 elements of the
ing system. The video imaging system performs 40 scans anterior segment: anterior and posterior corneal elevations,
through light slits projected at a 45-degree angle. In two corneal power, and corneal thickness. Keratometric power
0.75-second periods, 20 slits are projected sequentially on is calculated using a standard keratometric index.
the eye from the left and right sides of the video axis. The The Orbscan system acquires, analyzes, and presents
5-mm central zone of the cornea is sampled twice, once in the calculated data. Displays use a color scale to show rela-
each direction, through these overlapping slits, as illustrated tive elevations, providing a 3-dimensional view of surface
in Figure 6-2. Before the slit scans, an additional image is topography. In all elevation maps, green is the reference
captured using the Placido rings. surface, or zero level. Red is high, positive, and anterior
The Orbscan performs noninvasive measurements of to the reference surface, while blue is low, negative, and
thousands of points on 4 surfaces of the anterior segment posterior to the reference surface. A commonly used view
of the eye: the anterior cornea, posterior cornea, ante- of surface topography is the quad map, which presents
rior iris, and anterior lens. An example of a raw image is anterior elevation, posterior elevation, corneal power, and
shown in Figure 6-3. The system measures 9600 points pachymetry maps in one view. Figure 6-4 is a quad map
(240 from each of the 40 slits). These point measurements showing typical with-the-rule astigmatism. In this figure,
Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical
103 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 103 -166)
© 2012 SLACK Incorporated
104 Ch a p t e r 6

Figure 6 -1. O rbscan II/IIz diagnostic systems. (Reprinted with

permission of Bausch & Lomb.) Figure 6 -2. Slit scans used for data acquisition.

Figure 6 -3. Structures of the anterior seg-

ment analyzed by O rbscan include the ante-
rior cornea, posterior cornea, anterior lens,
and anterior iris. (Reprinted with permission
of Bausch & Lomb.)

Figure 6-4. O rbscan quad map showing

typical with-the-rule astigmatism. (Reprinted
with permission of John Vukich, MD.)
 To m o gra p h e r s 10 5

the anterior corneal float (elevation) appears in the upper

left corner, and the posterior float appears in the upper right
corner. Keratometric topography is shown in the lower left,
and corneal thickness is shown in the lower right.

Th eor y of Mea su rem en t

The Orbscan system uses hybrid measurement technol-
ogy. Specular reflection from a Placido disk is used to cal-
culate the surface curvature of the cornea by measuring the
specularly reflected image of a set of concentric mires. Slit
scanning is used to measure the anterior segment geometry;
its basic measurement is the absolute elevation of optical
surfaces. These curvature and elevation measurements are
combined to develop accurate maps of absolute anterior and Figure 6 -5. O rbscan raytrace triangulation. (Reprinted with
posterior surface elevation and pachymetry. Curvature is permission of Bausch & Lomb.)
displayed as a keratometric power.

Diffu se Reflect ion s required to accurately locate internal surface points when
they lie behind an optical interface that refracts the slit
When a slit beam intercepts an optically smooth surface,
beams and the conjugate image rays. The planar slit beam,
it is split into a specular reflection and a refracted beam that
diffusely reflected from the convex corneal shell, appears
penetrates the surface and is volume scattered by internal
as an annular arc in the video image. The outer and inner
scattering centers. Like surface diffuse reflection, volume
edges of this arc correspond to the anterior and poste-
scattering is omnidirectional. This important property
rior surfaces of the diffusely reflected volume, shown in
allows surface points to be independently observed and
Figure 6-5.
triangulated and gives Orbscan the capability to measure
To locate a point on the anterior surface, an outer edge
arbitrary surface shapes—convex or concave, aspheric or
point is first detected to sub-pixel accuracy. From the video
irregular. Volume (or diffuse) scattering is typically negli-
calibration, the detected edge point is then translated into its
gible from liquids such as the tear film and aqueous humor
conjugate 3-space ray. An example is shown in Figure 6-6.
because the constituent molecules are very small compared
This ray is represented mathematically as:
to the illuminating wavelength. In contrast, volume scatter-
ing is significant from the lens, the iris, and the cornea. For y = U + Vr
that reason, Orbscan sees through the tear film and captures
where the (x, y, z) vector U defines the ray origin at the prin-
the image of the diffusely scattered light from the corneal
ciple/nodal point of the camera optics; the vector V defines
volume that is directly illuminated by the slit beam.
its direction; and the scalar r specifies the distance of a ray
Because typical internal scatters are generally smaller
point from the origin.
than the wavelength of visible light, the magnitude of
Direct triangulation is used to locate points on the
scattering is inversely proportional to the third or fourth
external surface of the cornea. In direct triangulation, a
power of the optical wavelength (Rayleigh scattering: λ 4 for
ray is intersected with the calibrated outer surface of the
spherical particles, λ3 for cylindrical fibers). As a conse-
illuminating slit beam. This surface can be mathematically
quence, the diffusely scattered return consists of the short-
represented as:
est wavelengths found in the interrogating beam (the rea-
sons that Tyndall reflections from the cornea appear blue). S(x) = 0
Calculations of the beam and ray refraction depend on the
where the vector x represents the (x, y, z) coordinates of any
physiologic refractive indices of the various ocular tissues
valid point on the outer beam surface. Direct triangulation
and humors. Orbscan assumes that the standard physiologic
finds r such that S(U + Vr) = 0.
refractive index of air is 1.000; aqueous is 1.336; and cornea
Surfaces are triangulated one at a time, from front
is 1.376.
to back. Thus, all of the refracting surfaces in front of a
desired surface point are known a priori and can be used to
Trian gu lat ion of Com plex Su rfaces calculate all the necessary refractions.
Triangulation is required to map complex surfaces,
such as those of the anterior eye. Raytrace triangulation is
106 Ch a p t e r 6

the fixation light reflex with the instrument axis. As slit-

scan technology is relatively insensitive to misalignment
of the eye, some misalignment is expected. Following data
acquisition, the system determines the properly aligned
fixation-reflex axis and places it at the center of the map.
This fixation-reflex alignment ensures that the map center
is always a point of stationary elevation with zero surface
slope; for convex surfaces, this point is always a local maxi-
mum. Thus, the local surface normal at the map center is
aligned with the map axis. This form of alignment is crucial
for determining many relative properties that are axis-
based, such as axial and tangential curvature. Because the
fixation-reflex axis is very sensitive to surface inclination,
Figure 6 -6. O rbscan triangulation map of a Lincoln-head refractive surgery will almost always alter this axis; for this
penny (area in yellow circle is mapped). (Reprinted with per-
reason, standard alignment of a postoperative eye is unlike-
mission of Bausch & Lomb.)
ly to coincide exactly with the preoperative alignment.

Su rface Rotat ion

Fou n d at ion for An a lysis
Surfaces measured by Orbscan can be rotated to any
Th e Measu rem en t of Top ograph y other point of view by selecting a view center, which
becomes the surface point that is rotated to the map center.
The Orbscan system’s analysis produces a set of data The view axis is automatically chosen to be coincident with
describing the true topographic surfaces of the anterior and the local surface normal. Because 3-dimensional rotations
posterior cornea, the anterior iris, and the anterior lens. All are not commutative, the meridional orientation of a rotated
other measures and displayed maps are derived from these view generally depends on all the rotations that preceded
true 3-space surfaces. This approach was taken because it it. To eliminate the ambiguity that order dependency might
avoids the pitfalls of high- and low-frequency noise gener- create and to ensure that the rotated meridians are as close
ated by numerical transformation. as possible to the standard alignment meridians, all surface
All numerical transformations generate noise. Mathe- rotations are computed as simple single-axis rotations from
matic extraction of curvature from elevation (differentia- the standard alignment position, which is always centered
tion) generates high-frequency noise, while the transforma- on the fixation-reflex axis.
tion required to deduce elevation from curvature or slope
(integration) generates low-frequency noise. Although high- Relat ive Versu s Absolute Prop ert ies
frequency noise is more apparent, low-frequency noise is
more difficult to deal with; in addition, the most important Relative properties depend on the alignment, size, and
optical aberrations are low-frequency. Consequently, direct shape of a reference object; when the reference object
measurement of elevation and computation of curvature changes, so does a relative measure. In contrast, absolute
were the methods selected to extract the geometric informa- properties are intrinsic properties of the surface and are
tion of optical surfaces in Orbscan. not alignment dependent; they are measured directly from
Another advantage of directly measured topography is the surface. Only absolute measurements can identify true
that it facilitates surface analysis and display from other ocular landmarks, such as optical axes. Although absolute
points of view. Like other corneal measurement systems, properties are superior, most properties are relative by their
Orbscan aligns with the fixation-reflex axis of the eye, nature. In Orbscan, both relative and absolute properties are
which is appropriate when examining optical properties mapped, as described in Table 6-1.
in the visual zone. However, in contact lens fitting and
other procedures in which the apical shape of the cornea is
important, the viewpoint should be aligned with the axis of Geom et r ic An a lysis
best symmetry of the anterior surface. This alignment can
be easily and accurately achieved by rotating the directly Elevat ion
measured topographic surfaces in physical space. Looking at an irregular corneal surface is like looking
at the mountains and valleys of the earth. When drawn to
Fixat ion -Reflex Align m en t scale, the mean curvature of the earth overwhelms even
During data acquisition with Orbscan, the eye is aligned the most significant topographic features. Hills and valleys
by having the patient fixate on a blinking light source that are only apparent if elevation is shown with respect to the
is coaxial with the video system while the operator aligns mean sea level. Similarly, corneal irregularities can only be
 To m o gra p h e r s 10 7

TABLE 6-1.

Preview of Relative and Absolute Properties

GEOMETRIC OR OPTICAL PROPERTY TYPE R EFERENCE OBJECT (PARAMETERS)
Surface topography Complete
Normal elevation (floating alignment) Relative Surface (size and shape)
Normal elevation (axis alignment) Relative Axis + surface (size and shape)
Axial elevation Relative Axis + surface (size and shape)
Slope Relative Axis
Mean and astigmatic curvatures Absolute
Irregular curvature Absolute (Aperture size)
Axial pseudo-curvature Relative Axis
Tangential and sagittal curvatures Relative Axis
Axial thickness and depth Relative Axis
Normal thickness and depth Absolute
Optical power Relative Axis
Normal power Absolute

Figure 6 -7. True elevation, also known as topographic elevation, is the perpendicular distance Z of a point on the
cornea from the system reference plane. (Reprinted with permission of Bausch & Lomb.)

seen after a reference surface is mathematically removed. fit error with no additional constraints. Axial and pinned
Changes in the reference surface, like changes in sea level, alignments each add one additional constraint (indicated
can dramatically affect the perceived topography of the below by the X in Figure 6-8). Axial alignment forces the
corneal landscape, while its true topography (z as a func- sphere center to lie on the view axis, while pinned align-
tion of x and y) remains unchanged. Figure 6-7 illustrates ment forces the sphere surface to include the view center.
true elevation, also known as topographic elevation. It is Axial pinned alignment employs both constraints.
the perpendicular distance Z of a point on the cornea from The advantage of displaying elevation with respect to a
the system reference plane. True elevation data are used to best-fitting sphere is that a sphere is rotationally symmetric,
determine pachymetry. and thus it is completely described by its radius and center.
A reference sphere may be oriented with a surface in Unfortunately, corneal surfaces are not spherical. To view
3 different ways: floating, axial, or pinned, demonstrated elevation asymmetries with respect to the axisymmetric
in Figure 6-8. Floating alignment minimizes the surface surface that fits the cornea no matter what its shape, it is
108 Ch a p t e r 6

Figure 6 -8. Reference sphere orien-

tation. (Reprinted with permission of
Bausch & Lomb.)

Figure 6-9. Close-fitting reference sur-

faces. (Reprinted with permission of
Bausch & Lomb.)

necessary to select a rotor reference surface (Figure 6-9). An terior Segm en t an d Ch am ber Dept h s
The name rotor is derived from its method of construction,
which is to find a surface of revolution by spinning the data The anterior chamber depth (ACD) from the posterior
surface around the view axis. A rotor (without modifiers) is cornea to the anterior lens or iris is calculated differently
the mean surface of revolution. The high rotor (or low rotor) as the straight-line axial distance between anterior chamber
is the surface of revolution lying just above (or below) the surfaces. Anterior chamber volume is easily determined by
data surface. Orbscan setup options and parameters that integrating this distance across the cornea. In addition, the
affect the relative elevation of a single topographic surface rate of change of this distance in radial directions can be
include surface rotation from the instrument axis; reference used to estimate the anterior angle.
surface type (plane, sphere, cone, or rotor); reference sur-
face size, shape, and alignment (floating, axial, or pinned); Su rface Slop e an d Su rface Cu rvat u re
and elevation direction (normal to the reference surface or Surface slope measures the rate of change of surface ele-
axially directed). vation in a particular direction. Determination of the radi-
ally directed slope of the corneal surfaces is important in
Pach ym et ry contact lens fitting and in the implantation of intracorneal
Corneal thickness is calculated as the distance from the rings for refractive correction. Surface curvature measures
anterior to the posterior surface, in the direction perpendicu- the bending (or rate of change of slope, or second deriva-
lar to the anterior surface. Orbscan pachymetry calculations tive) intrinsic to a curve or a surface. Because curvature
have been correlated with manual ultrasound pachymetry; is inversely related to radius of curvature, a small radius
in general, Orbscan determinations of corneal thickness sphere has a large curvature.
are several percent thicker than ultrasound measurements Surface curvature can be described as cutting the sur-
of the same corneas. Orbscan pachymetry measurements face of an object with a plane and then fitting a circle to
can be automatically converted to their acoustic equivalent the plane intersection. The inverse radius of the circle
values; the correlation factor can be set individually for gives the surface curvature in the direction of the plane.
each Orbscan system. The refractive surgeon can overlay Obviously, the value measured is highly dependent on the
the intended radial and astigmatic keratotomy (RK and AK) orientation of the cutting plane. Every surface point has an
cuts on any Orbscan anterior corneal map. The values of infinite number of surface normal planes, each containing
minimum corneal thickness are shown on each incision. the local surface normal, but each cutting the surface in a
 To m o gra p h e r s 10 9

TABLE 6-2.

Assumed Refractive Indices and Typical Spherical Curvatures

OCULAR SURFACE INDEX TYPE POSTERIOR INDEX CURVATURE (GEOMETRIC) CURVATURE (DIOPTERS)
Keratometric cornea Standard 1.3375 128 1/m 43.2 D
keratometric
Anterior cornea Physiologic 1.376 128 1/m 48.1 D
Posterior cornea Physiologic 1.336 149 1/m – 6.0 D
Anterior lens Physiologic 1.425 98 1/m 8.7 D

different direction. Thus, every point on a smooth sur- to understanding Orbscan maps is that curvature expressed
face has a direction-dependent curvature, whose complete in D is proportional to an assumed refractive index differ-
description is captured by a mathematical object known as ence. The assumed index difference may be physiologic
a tensor. Orbscan calculates the complete curvature tensor (the real value averaged over the population), or it may be
field from the derivatives of the fitted topographic surfaces. invented (eg, the standard keratometric index). Table 6-2
From this tensor field, the curvature at any surface point lists the assumed posterior indices (air = 1) for each optical
and in any direction can be directly calculated. surface in the anterior segment, together with their typical
A theorem of differential geometry states that every curvatures, expressed in both geometric and D units. Note
point on a smooth surface has a minimum and a maximum that the curvature (in D) of the posterior cornea is nega-
curvature (called principle curvatures), which lie in perpen- tive, not because the geometric curvature is negative, but
dicular surface normal planes. Thus, the curvature at any because the interfacial index difference is negative (Δn =
smooth surface point can be completely specified by 3 inde- -0.040 = 1.336 – 1.376). The standard keratometric index is
pendent quantities: the minimum and maximum (principle) applicable only to the anterior cornea. When so applied, the
curvatures and their direction. Two curvatures at a point in anterior cornea is referred to as the keratometric cornea.
nonprinciple directions do not contain sufficient informa-
tion to construct the complete curvature tensor. Mean an d Ast igm at ic Cu rvat u re
Curvature of an optical surface is directly related to the
Mean curvature is a measure of absolute local sphericity.
focusing power of a normally incident bundle of light rays.
A local sphere is the one that best fits a point and its surface
Because of this property, curvature is often expressed in
derivatives. The inverse radius of this sphere is an absolute
diopters (D; the unit of optical power) and is often referred
measure of the local spherical component of curvature.
to as power, which is a potential cause of confusion.
Because it is absolute, mean curvature is intrinsic to the sur-
Orbscan emphasizes the distinction between curvature and
face and is independent of surface alignment. Absoluteness
power by exclusively reserving the name power for opti-
is important in the diagnosis of certain corneal diseases like
cal properties. Curvature, whether expressed in standard
keratoconus, as it ensures that any geometric abnormality
geometric units (inverse meters, 1/m) or radius of curvature
will appear as it exists. Hence, keratoconus appears as a
units (millimeters, mm), or scaled by D, is always called
symmetric local maximum because the mean curvature of
curvature by Orbscan.
a cone increases toward its apex.
Three concepts are important in understanding Orbscan’s
Maps of mean curvature display the variation of local
mapping of surface curvature. First, curvature applies only
sphericity. Thus, the mean curvature of a normal cornea
to a single surface. In contrast, optical power is generally
is typically very uniform, even when its astigmatism is
calculated for a sequence of surfaces (always beginning
significant. Mean curvature typically filters out global
with the anterior cornea). Second, surface curvature is not
astigmatism in favor of the local spherical component. As
single-valued, but direction-dependent. A complete specifi-
a rule, any residual astigmatism seen in a mean curvature
cation of curvature requires 3 values: the 2 principle curva-
map is greatly reduced and rotated 90 degrees from the
tures and their orientation. As a color contour map (without
real astigmatism. To see the local axes of astigmatism, the
overlays) can only show the variation of one value, many
principle directions overlay is used.
different curvature maps are used to display combinations
Astigmatic curvature is a measure of absolute local cyl-
of the 3 principle values in useful ways. These maps are
inder. Astigmatic curvature typically filters out the global
grouped into 2 families: absolute local curvatures (mean,
spherical component in favor of astigmatism or local cylin-
astigmatic, and irregular) and relative axis-based curvatures
der. As maps of astigmatic curvature display the variation
(axial, tangential, and sagittal). The third concept important
110 Ch a p t e r 6

of local cylinder, an eye with regular astigmatism will have

a fairly uniform astigmatic map. The bow-tie pattern typi-
cally associated with astigmatism and seen in axis-based
maps (eg, axial, tangential, and sagittal curvatures) is not
physical, but is really an artifact of the measurement.

Irregu lar Cu rvat u re

A normal but ametropic eye with regular astigmatism
is correctable with spherocylindrical spectacles. Surface
irregularity includes the curvature variation of an optical
surface that cannot be corrected with a spherocylindrical
lens, thereby producing an uncorrectable loss in visual
acuity.
As the mean and astigmatic curvature maps of a normal Figure 6 -10. The axial power map in O rbscan, similar to the
eye are fairly uniform, the variation of these curvatures is sagittal map from Placido systems, produces its image using
a measure of surface irregularity. Irregular curvature is only the Placido rings. (Reprinted with permission of Bausch
the statistical combination of the standard deviations of & Lomb.)
the mean and astigmatic curvatures, measured over a local
aperture (typically, 1 mm in diameter). Surface irregularity
caused by incisional keratotomy (such as the RK artifact) is left panel). Light rays, originally parallel to an arbitrary
recognizable in maps of irregular curvature. power axis, are refracted by both surfaces of the cornea
and are brought to focus on the power axis, which is also
Axial, Tan gen tial, an d Sagittal Cu rvatu res refracted. The refractive index is included in the definition
so that power measured in either direction is the same.
Axial, tangential, and sagittal curvatures are relative
Optical performance can be calculated from topographic
properties measured with respect to the arbitrary view
surface data when the material refractive indices are known
axis. Because the measurement directions radiate from or
(see Table 6-2).
encircle this axis, these axis-based maps always contain a
Peripheral rays are brought to a shorter focus than cen-
conspicuous axial artifact, the familiar bow-tie pattern seen
tral paraxial rays. This is illustrated in Figure 6-11. This
in Figure 6-10. The direction of this artifact effectively
spherical aberration is apparent in optical power maps of
locates the meridians of astigmatism, but only with respect
the cornea, which show increasing power peripherally.
to the view axis.
When astigmatism is present, the familiar bow-tie pattern
Unlike absolute curvature, which is a property of the
will also be seen. This pattern arises from the interaction of
surface and requires no reference object, axis-based cur-
light, collimated along the line of sight, with a smooth opti-
vature maps change dramatically when their axis is reposi-
cal surface or sequence of surfaces. Optically, the bow-tie
tioned. Axis-based maps also distort corneal abnormalities,
pattern is real, while geometrically it is an artifact.
like keratoconus, making it impossible to locate the true
conical apex.
The curvature of an aligned sphere is correctly deter-
Norm al Power
mined by any of the axis-based curvatures and the mean Normal power is a measure of the normally incident
curvature (which gives the correct value even when the mean focusing power of a sequence of surfaces, beginning
sphere is off center). Axial curvature, however, is not a true with the anterior cornea (see Figure 6-11, right panel).
measure of curvature but is a spherical equivalent not appli- Unlike optical power, which is a relative property defined
cable to aspheres or to asymmetric surfaces. Tangential and with respect to some arbitrary power axis, normal power is
sagittal maps, although true measures of curvature, do not an attempt to define an optically relevant absolute measure.
contain sufficient information to construct the complete Normal power can be calculated locally for any point
curvature tensor, except for the special case of aligned axi- on the anterior corneal surface by orienting a ray pencil (a
symmetric surfaces. tight bundle of paraxial rays) initially perpendicular to the
point of interest. The ray pencil is mathematically propa-
gated through the specified surface sequence and the mean
Th eor y of Op t ica l An a lysis paraxial focus calculated. The normal power is calculated
from the mean focal length.
Opt ical Power Because the normal power of an anterior surface point
is independently calculated, neither a common focal point
Optical power is defined as the posterior refractive index nor power axis exists. However, when the optical power
n2
divided by the posterior focal length: P = __
f
(Figure 6-11, axis is aligned to the local surface normal, normal power
2
 To m o gra p h e r s 111

Figure 6-11. O ptical

power and normal
power. (Reprinted
with permission of
Bausch & Lomb.)

and paraxial optical power are equivalent. Hence, normal surfaces that rotate with the eye and any reference objects,
power is the paraxial portion of optical power as seen from whereas movable centers are dependent on the rotational
any point on the cornea. point of view (eg, entrance pupil center). Nine centers used
by Orbscan are defined in Table 6-3. The tabulated accu-
Opt ical Axes racy reflects the theoretical precision with which these cen-
ters can be calculated from a set of topographic surfaces.
An optical axis can be defined for any 2 surfaces by
the alignment of a point source and its 2 catoptric images.
Thus, an optical axis is the refracted light ray oriented
perpendicular to both surfaces. In tilted optical systems,
Op t ica l Su r face Ir regu la r it y
like the human eye, the optical axis is not a straight line, Irregularity is a statistical term that describes the vari-
but is bent or curved by refraction occurring between the ation of values in a data set. Optical surface irregularity
2 defining surfaces. is proportional to the standard deviation of surface cur-
Normal pachymetry and normal anterior segment depth vature. As applied to the Orbscan, it is a spatial variation
are both distances measured along the refracted light ray in curvature, quantified as the statistical combination of
that is initially perpendicular to the anterior corneal sur- the standard deviations of the mean and toric curvatures.
face. Maps of these quantities each show a local extreme, Irregularity is calculated for a specific area using the fol-
which locates a 2-surface optical axis. The local minimum lowing formula:
in corneal thickness is the origin of the corneal optical axis,
Irregularity = I = [(σ (κ)) 2 + (σ (Δκ)) 2] ½
as defined by its anterior and posterior surfaces. Similarly,
where
the local maximum in anterior segment depth is the origin
σ = standard deviation
of the anterior optical axis, as defined by the anterior cor- κ1 + κ 2
nea and lens. κ = mean curvature = ______ 2
Both of these optical axes lie close to the traditional opti- Δκ = toric curvature = |κ1 – κ 2|
cal axis of the eye, which is defined by the confluence of κ1 and κ 2 represent the principle curvatures
the 4 Purkinje images. However, because 4 images cannot
The Orbscan system calculates irregularity using the
generally be brought into alignment, the traditional axis is
following method. First, initial parameters are set using
undefined for many eyes. When the traditional axis exists,
default values that can be changed in an initialization file.
it is coincident with both the corneal and anterior segment
The beginning default pupil diameter is 3 mm, and the
optical axes, which always exist.
increment for increasing the diameter is 2 mm. With these
settings, there are 4 possible irregularity zones on the pupil
(3, 5, 7, and 9 mm). Next, the algorithm defines the sam-
Ocu la r La n d m a rk s pling rules for each zone. Nine concentric rings are defined
Used b y Orb sca n in the 3-mm zone; each ring is divided into segments using
a formula to ensure uniform sampling density (eg, the cen-
In general, a surface center is any well-defined point on ter ring has 3 segments, the second ring has 9 segments, the
a surface—usually the anterior cornea. A landmark center third has 15 segments, etc.) The center point of each seg-
is any uniquely defined physical or optical point of the eye, ment is defined. In the third step, the algorithm calculates
whereas a reference center is defined by the arbitrary refer- all of the map zone statistics at each center point, including
ence object employed. Fixed centers are points on ocular the principle and mean, toric curvatures, etc. Then, the
112 Ch a p t e r 6

TABLE 6-3.

Ocular Landmarks (Surface Centers) Used by Orbscan

SURFACE DEFINITION TYPE ACCURACY MAP MARK
CENTER (TYPICALLY
WHITE)

Fixation reflex Corneal reflex point of a fixating patient, measured Fixed High Oblique cross,
by a coaxial optical system and corrected for acqui- landmark x
sition misalignment
Pachymetry Anterior corneal point with minimum normal thick- Fixed Low C
minimum ness; defines the 2-surface optical axis of the cornea landmark
Anterior segment Anterior corneal point with maximum normal ante- Fixed Low S
maximum rior segment depth; defines the 2-surface optical landmark
axis of the anterior cornea and lens combination
Anterior corneal Geometric center of the cornea, or the location Fixed Low Triangle
apex where the axis of best anterior symmetry intersects landmark
the anterior surface
Entrance pupil Physical pupil of the eye imaged through the cor- Movable Medium Dot
nea. Its center is taken to be the geometric centroid landmark
of the pupil image
Sphere center (of View-axis projection of its apical center of curvature Movable Exact Circle
an axisymmetric reference
reference object)
View center Point at which the view axis pierces the surface; Movable Exact Black cross
placed at the map center in standard alignment reference (map center)
Summit Highest surface point measured with respect to the Movable High
current view axis; placed on the view center (and reference
therefore at map center) in standard alignment of a
convex surface
Instrument or Point at which the instrument axis (defined by the Fixed High
system center video camera) pierces the data surface; if located on reference
the unrotated anterior cornea, is a measure of acqui-
sition misalignment

algorithm calculates the standard deviation of these values difference, which is 0.3375 for the keratometric surface.
for the first zone and applies the formula for calculating Thus, the D equivalent of irregularity is about one-third the
irregularity. These calculations are repeated for each zone. curvature measure in reciprocal meters.
Optical surface irregularity is proportional to the stan-
dard deviation of surface curvature. Consequently, only Clin ical Im plicat ion s of Irregu larit y
axis-independent surface curvatures are used in the calcula- Calcu lat ion s
tion, as only they are true surface properties. These include
the mean curvature, which is a measure of local surface Optical surface irregularity is important to the refractive
sphericity, and astigmatic curvature, which is a measure of surgeon because it often represents a loss in best-corrected
local cylinder. As both curvature variations are important, visual acuity (BCVA). Regular astigmatism has a low
the standard deviations of the mean and astigmatic curva- irregularity, while irregular astigmatism has a high degree
ture are statistically combined (their variances are added) of irregularity. Higher-order aberrations (HOAs) may also
to yield the irregularity in standard curvature units (recip- have a high degree of irregularity. Measures of irregular-
rocal meters). Curvature in reciprocal meters can be con- ity may also be useful as a supporting indicator (but not
verted to D by multiplying it by the surface refractive index the sole indicator) in evaluating keratoconus. A 3-mm
 To m o gra p h e r s 113

Figure 6-12. This O rbscan quad map

shows a posterior float of approximately
0.063 D, a strong red flag for forme
fruste keratoconus or pellucid marginal
degeneration (PMD). (Reprinted with
permission of John Vukich, MD.)

irregularity greater than 1.5 or a 5-mm irregularity greater

than 2.0 may be an early sign of keratoconus, while a 3-mm
irregularity greater than 2.0 or a 5-mm irregularity greater
than 2.5 may indicate full keratoconus. Irregularity cannot
be corrected with spherocylindrical lenses or with current
surgical techniques.

Clin ica l Ex a m p les

Risk of Ectasia
One of the most frequent uses of Orbscan is in the screen-
ing of patients for LASIK surgery. The risk of ectasia can be
accurately assessed using 5 indices: the number of abnor-
mal maps using the normal band scale, a variance of more
than 1.00 D in astigmatism between the eyes, keratometric
or corneal steepness on the mean power map, a posterior
surface float greater than 0.05 D (the difference between
the highest and lowest spots; Figure 6-12), 3-mm and
5-mm irregularity (Figure 6-13), and a minimum peripheral
corneal thickness that is not at least 20 µm greater than the
thickness of the central cornea (Figure 6-14).
Figure 6-12 shows a posterior float of approximately
0.063 D, indicated by the dark reddish color near the Figure 6-13. This O rbscan quad map shows
center of the posterior elevation map (upper right). A pos- surface irregularity of 2.1 D in the 3-mm zone
terior float greater than 0.050 raises a strong red flag for and of 2.2 D in the 5-mm zone, suggesting
forme fruste keratoconus or pellucid marginal degeneration the clinician look for other signs, such as the
(PMD) and confirms the diagnosis if supported by other number of abnormal maps or a posterior float
maps and measurements. Note that when the posterior float greater than 0.050 mm (50 µm).
is greater than 0.050, posterior elevation is rarely the only
abnormal map. In this example, the pachymetry map shows
a significant irregularity in corneal thickness, with a central Figure 6-13 shows surface irregularity of 2.1 D in the
area whose thickness is only approximately 0.050 mm— 3-mm zone and of 2.2 D in the 5-mm zone, indicated by the
2 indices suggesting that LASIK may not be appropriate for circled scores in the gray area in the center of the screen. It
this patient. does not indicate forme fruste keratoconus in and of itself
114 Ch a p t e r 6

Figure 6 -14. In the pachymetry map at

the lower right corner of this quad map,
the thinnest area is 30 µm thinner than
the central cornea.

but raises suspicion to look for other signs, such as the An terior Ch am ber Map
number of abnormal maps or a posterior float greater than
0.050 mm (50 µm). Irregularity at 3.00 and 5.00 mm may The anterior chamber map, which shows the depth of
simply indicate the presence of HOAs without pathology, the anterior chamber, can be useful in surgical planning for
so this index should only be used in conjunction with other phakic intraocular lenses (IOLs). The endothelial anterior
findings to diagnose keratoconus. map (Figure 6-15) shows the true available space. This map
In the pachymetry map, shown in Figure 6-14, the lower is likely to grow in its usefulness in the future as the use of
left corner of this quad map, the thinnest area is more than phakic IOLs increases.
30 µm thinner than the central cornea (marked in red).
Peripheral pachymetry measurements in which the periph- Post-LASIK Follow-Up
eral cornea is not at least 20 µm thicker than the central Approximately 1 month following LASIK, the Orbscan
cornea may be a sign of keratoconus, which should be cor- system may be used to assess problems such as dry eye
roborated with other observations. Another potential source and corneal edema and to map the change in corneal shape
of concern is a cornea in which the thinnest point is outside as a reference for an enhancement procedure. Earlier than
the central 5 mm of the cornea.1 1 month, reflections from the recovering tear film or corne-
One abnormal map on the Orbscan normal band scale al edema may produce an Orbscan reading that incorrectly
does not usually indicate forme fruste keratoconus, but suggests ectasia.
requires patient education or having the patient return for
monitoring of changes in 6 to 12 months. Two abnormal Dry Eye an d Corn eal Edem a
maps may indicate early keratoconus; or, if the posterior
float is abnormal with a slightly thinner cornea (less than Dry eye that continues for more than 1 month after
500 µm), 2 abnormal maps may still indicate keratoconus LASIK can be evaluated in a distinctive Orbscan map,
depending on other variables described in this chapter. If shown in Figure 6-16. Note the irregularity on the kerato-
a patient has 2 abnormal maps but no indication of forme metric mean power map as well as the significant missing
fruste keratoconus, surface ablation would likely be a bet- data points. These anomalies confirm that the maps are
ter procedure than LASIK for this patient. Three or more not reliable. After artificial tears were instilled, the images
abnormal maps is a contraindication to corneal surgery and were normal. An Orbscan map with irregularity on the
often indicates a high risk of post-LASIK ectasia.2 anterior elevation map and thinning on pachymetry are
A difference of more than 1.00-D cylinder between the indicators of dry eye.5
eyes or increasing cylinder over time is also a risk factor Corneal edema also produces a distinctive Orbscan map
for keratoconus that can be detected with the Orbscan.3 (Figure 6-17). The specular reflection from corneal edema
Because keratoconus is known to be an asymmetric condi- might be interpreted as Descemet’s layer, and, therefore, the
tion, one eye usually progresses faster than the other.4 A edema will often look similar to a case of ectasia.
higher risk of keratoectasia is also suggested by a K reading
of more than 46.00 D at the steepest point on the Orbscan
keratometric mean power map.
 To m o gra p h e r s 115

Figure 6 -15. The endothelial anterior

chamber map shows the depth of the
anterior chamber.

Figure 6 -17. Post-LASIK corneal edema. (Reprinted with per-

mission of Dan Durrie, MD.)
Figure 6 -16. Post-LASIK dry eye.

Post-LASIK En h an cem en t Plan n in g

LASIK surgery results in significant changes to the
corneal surface geometry (Figure 6-18). The blue areas
in the center and at the edges of the Orbscan map show
the changes in the cornea that result from LASIK, includ-
ing a flattening of the center and sides of the cornea with
respect to the reference sphere. Although the map appears
Figure 6 -18. Corneal topography S/P myopic LASIK. (Reprinted
to show that the cornea is concave, this is not the case; the with permission of John Vukich, MD.)
depression in the center of the cornea is lower than the ref-
erence sphere, but not lower than the outer corneal edges. A
3-dimensional Orbscan map of the post-LASIK cornea pro-
disorder. PMD typically presents with topography similar
vides the surgeon with a means to visualize the cornea and
to Figure 6-19. Note the classic kissing birds appearance
determine an enhancement strategy when an enhancement
in the bottom left keratometric axial map, as well as the
procedure is desired.
extreme peripheral elevation on the anterior and posterior
floats (upper right and left maps). Note also that the steepest
Pellucid Margin al Degen erat ion part of the cornea is located more inferiorly in PMD, and
PMD is a rare form of corneal ectasia, with an arcuate the axial topography shows a bending of the bow-tie pat-
band of corneal thinning in the inferior cornea. The Orbscan tern. The keratometric mean map shows a localized inferior
system can be used in the diagnosis of patients with this corneal steepening, especially in the peripheral areas.
116 Ch a p t e r 6

Figure 6-19. PMD. (Reprinted with permis-

sion of Tim Cavanaugh, MD.)

Figure 6 -20. Keratoconus.

Keratocon u s Rayt race An alysis as a Diagn ost ic Tool

Diagnosis of keratoconus is facilitated by the use of Raytrace analysis can calculate optical properties even
Orbscan maps, which show the magnitude and location of more informative than power. The retinal point spread
corneal thinning and protrusion (Figure 6-20). Orbscan function (PSF) is the retinal image of a point of light. It
maps can also be used to detect early keratoconus (which contains all the information needed to reconstruct what
always appears initially on the posterior surface) and the patient sees. The PSF describes how an object point is
document the progression of the disorder. Note the normal aberrated (spread out) in the retinal image. PSF size and
pachymetry map and the irregularity in the 3- and 5-mm shape indicate the ocular aberration. In an emmetropic,
zones and the deeper cone on the posterior surface. The unaberrated eye, the image is concentrated at one point. In
mean power map shows Ks in the inferior cornea greater an astigmatic eye, the image takes on an elongated shape,
than 54 D. Indices for the detection of keratoconus are while in monocular diplopia, the image is double peaked.
described above, in the discussion of the risk of post-LASIK Figure 6-21 shows an original picture and a simulated reti-
ectasia. nal image, produced with the retinal PSF, in a patient with
monocular diplopia.
 To m o gra p h e r s 117

optimal results based on those data. Taking into consider-

ation variables such as pachymetry and the new placement
of 2- to 1-mm laser flying spot placement are factors that
allow the Zyoptix system to be one of the few systems that
is tissue paring when moving from conventional to wave-
front. Furthermore, it is becoming apparent that wavefront
data can also enhance topography diagnostic capabilities.
In a study reported by Pepose,6 posterior float signs of
ectasia noted on the Orbscan Diagnostic System were dif-
ferentiated from PMD and keratoconus based on wavefront
Figure 6 -21. Reconstruction of patient’s view of an object data. Patients with PMD tended to have higher degrees
using retinal PSF in raytrace analysis. (Reprinted with permis- of peripheral aberrations on the Zywave system including
sion of Bausch & Lomb.) trefoil, whereas patients with keratoconus tended to have
higher degrees of coma. Combining the Wavefront data
with the Orbscan data will enhance diagnostic capabilities
The patient complained of double vision, halo, and and wavefront data, potentially improving outcomes.
night driving difficulties, but only for the left eye. Raytrace
analysis through the anterior corneal surface did not reveal
any visual impairment. When both corneal surfaces were
raytraced, the left eye showed several ghost images, which
Con clu sion
are associated with the posterior corneal surface. An analy- This section has described the Bausch & Lomb Orbscan
sis of posterior elevation maps showed that each eye had a Anterior Segment Analysis System, focusing on the prin-
posterior cone, but the off-axis eccentricity of the OS cone ciples underlying its measurement and analysis technology.
resulted in the optical aberrations reported by the patient. The Orbscan system provides unique tools for the refractive
Retreatment was not undertaken because of concern that surgeon to use in the evaluation and diagnosis of visual
it would increase the posterior ectasia and the associated disorders of the anterior segment.
visual impairment.

O rbscan Diagn ost ics an d Wavefron t Ack n ow led gm en t

Tech n ology
The author thanks Barry Eagan, Director, Optical
A unique aspect of the Bausch & Lomb Zyoptix Systems Design & Numerical Analysis, Bausch & Lomb,
Diagnostic Workstation is that it combines wavefront aber- Rochester, NY for his assistance in the development of this
rometry information with data from the Orbscan system. section.
This allows for potential greater data capture and potential

OCULUS P ENTACAM
Stephen S. Khachikian, MD and Michael W. Belin, MD, FACS

W
ith the increased popularity of refractive sur- system from curvature data by making geometric assump-
gery and the high success rates for corneal tions about the cornea and its position relative to the Placido
transplantation, there has been an increasing disk.7
need to understand corneal contour. Computerized corneal There are significant limitations in trying to describe the
topography has advanced our knowledge of the complex cornea with a Placido-based curvature map (Figure 6-22).
corneal surface beyond what was previously possible with The area of corneal coverage is limited to about 60% of the
earlier examination equipment. Traditional computerized corneal surface, eliminating important data for many periph-
corneal analysis relied on the processing of a Placido-type eral or paracentral pathologies. Placido maps provide no
reflected image. This precludes obtaining data from the information about the posterior corneal surface. Without mea-
posterior corneal surface. True topography implies shape surement of the anterior and posterior surfaces, pachymetric
and requires the generation of an x, y, and z coordinate maps depicting the distribution of corneal thickness cannot
system. Placido-based systems create such a coordinate be made. Additionally, there are limitations in attempting to
118 Ch a p t e r 6

Figure 6 -22. Placido-based curvature map demonstrat-

ing limited corneal coverage.
Figure 6-24. Pentacam-generated elevation-based curva-
ture map on the same eye as shown in Figure 6-23.

Figure 6 -25. Pentacam

Anterior Segment Analysis
System.

Figure 6 -23. Standard Placido-based curvature display.

reconstruct the corneal surface based on curvature measure-

ments.8
The standard topographic curvature (axial or sagittal)
is a reference-based measurement (Figure 6-23). It is not
The Pentacam Anterior Segment Analysis System
a unique property of the cornea. The same shape can have
(OCULUS, Wetzlar, Germany) is an elevation-based
many different “curvatures” depending on which axis is
topographer that represents a significant advancement in
used to make the measurement. Most of us visualize the eye
corneal and anterior segment imaging (Figure 6-25). The
as a Gullstrand-reduced eye, assuming that the eye is sym-
Pentacam uses 2 cameras to obtain imaging of the anterior
metric, with the line of sight, visual axis, center of pupil, and
segment. The central camera is used for fixation monitoring
corneal apex all crossing at a common point. This, however,
and pupil measurement, and the second Scheimpflug cam-
is not the case.9,10 Furthermore, we assume that the mea-
era captures optical cross-sections of the anterior segment.14
surement axis of the Placido system also coincides. Most
The Scheimpflug principle, the photographic technique
people do not look through the center of the cornea. The line
on which the camera is based, was first described in 1875
of sight and the measurement axis of the videokeratoscope
by Theodore Scheimpflug as a way to minimize distor-
are not the same.11,12 In contrast to the reflection-based
tion in aerial photographs. Scheimpflug photography is an
Placido measurement, each elevation system uses a direct
alternative method of capturing images by adding an angle
triangulation technique to measure the anterior corneal sur-
between the lens and the film (Figure 6-26). Scheimpflug
face. Curvature data are directly calculated from the eleva-
cameras capture images with a better spatial accuracy than
tion data using second-order derivatives (Figure 6-24).13
a traditional camera containing a coaxial lens and film
 To m o gra p h e r s 119

Figure 6-27. Pentacam-generated corneal cross-section with

edge detection lines.

Figure 6 -26. Diagram of Scheimpflug-generated corneal cross-

section, showing plane of focus and lens position.

optical system. By rotating around the point of fixation, the

Scheimpflug system reduces the artifact created by small
movements during image acquisition.15,16 The Scheimpflug
camera rotates 360 degrees around a single point of fixa-
tion as the patient focuses on a central light source. The
camera completes several rotations before coming to a
steady speed. It then obtains images over a 2-second period
without vibrations and image artifact that might occur dur-
ing acceleration or deceleration. The slit images are photo-
Figure 6 -28. PAR CTS grid projected onto fluorescein-
graphed on an angle from 0 to 180 degrees to avoid shadows
stained tear film.
from the nose. Every picture is a complete image through
the cornea in the specific angle, so a real 360-degree image
of the anterior segment of the eye is acquired. The camera fluorescein onto the tear film for the grid to be viewed and
provides 25 or 50 images during a scan of less than 2 sec- relied on the distortion of the grid pattern to determine ele-
onds duration. Each image typically contains approximately vation by a technique similar to triangulation called raster-
500 elevation points, which are then analyzed to yield photogrammetry. Because it was a projection-based system,
12,500 to 25,000 total data points. The images generated by measurement of the posterior corneal surface or anterior
the rotating Scheimpflug camera are used to locate the ante- segment was not possible. Optical cross-sectioning was first
rior and posterior corneal surfaces as well as the iris and introduced commercially in the mid 1990s. The Orbscan
anterior lens surface. By taking a series of cross-sectional used parallel segmental cross-sectioning (no shared points)
images around a common central point, a 3-dimensional and relied on a Placido image to augment the anterior sur-
reconstruction of the anterior segment can be created. face measurements and reportedly to assist in image reg-
While the Pentacam is primarily a topography instrument, istration. The rotating system used by the Pentacam offers
the cross-sectional Scheimpflug photographs obtained also certain advantages. Because each image shares a common
have clinical importance (Figure 6-27). point (center of rotation), image registration should be more
The OCULUS Pentacam was neither the first elevation- accurate (Figure 6-29).
based corneal analysis system nor the first to use optical Precise image registration is a prerequisite for accurate
cross-sectioning. Elevation-based systems were first report- topographic data. Because the distance travelled during
ed to have benefits over Placido-based systems in 1991.13 rotation is less in the center, the point density is great-
The PAR CTS used a projection grid system to measure est in the central cornea and decreases in the periphery.
anterior corneal elevation (Figure 6-28). It had advantages Additionally, the edge detection software on the Pentacam
over Placido-based systems in that accurate measurements appears less susceptible to corneal haze than the Orbscan,
were possible on distorted corneas that could not be fully and the system does not suffer from the inaccuracy in locat-
analyzed by reflective systems and offered complete cor- ing the posterior corneal surface common to the Orbscan in
neal coverage. The PAR CTS required the instillation of the acute postoperative cornea.17
120 Ch a p t e r 6

Figure 6 -29. Diagram showing the overlapping radial imaging

of the Pentacam (left) and parallel imaging of the O rbscan
(right).

Figure 6 -31. Four-view refractive display on the Pentacam.

For routine use, it is recommended to use the manual

setting with the BFS set to sphere, float, and the diameter
set to 8.5 mm. We also suggest that, for routine screen-
ing, the overall area of the map displayed on the screen be
limited to the central 9.0-mm zone. This does not eliminate
the peripheral coverage; it just masks it from the screening
Figure 6 -30. Pentacam-generated Zernike polynomials. view and makes the inspection of the maps easier. When
viewing the maps limited to the central 9.0-mm area, a valid
map would have no or very minimal extrapolated data.
While primarily designed as a topographic unit, the In the authors’ experience, in all but markedly abnormal
Pentacam also provides cross-sectional visualization of the corneas, it is possible to obtain maps free of extrapolated
cornea and the anterior chamber, as well as an objective data. For refractive screening, we are dealing with pre-
lens densitometry assessment. It calculates the pachymetry sumed normal corneas, and quality images should almost
of the cornea from limbus to limbus with an accuracy of always be obtained. These settings are chosen to maximize
±5 µm. True elevation measurements allow for computa- sensitivity and allow for a quick visual screening based on
tions of anterior and posterior sagittal (axial) and tangential color and pattern recognition. It should be understood that
curvature, anterior and posterior elevation maps, ACD when screening for refractive surgery, the goal is to identify
measurement, angle approximation, corneal and anterior patients who are not “normal.” The recommended settings
chamber volumes, and anterior surface-derived Zernike are designed for that purpose.
polynomials (Figure 6-30). It is important to realize that what we call an “elevation”
map does not really display the actual raw elevation data,
but shows the elevation data against some reference sur-
Clin ica l Ap p licat ion s face. The most commonly used reference surface is a BFS.
The shape, scales, and colors should be chosen to simplify
Refractive surgical screening is the most common clini- screening. Different scales, shapes, and color bars may be
cal application of the Pentacam. Screening should be rapid, appropriate for other uses. Others have proposed using a
technician friendly, and accurate. The vast majority of toric-ellipsoid as the base reference surface for screening
patients can be quickly and safely screened with the 4-map patients under the premise that a toric-ellipsoid more closely
refractive display (Figure 6-31). The recommended maps in follows the normal prolate shape of the cornea. While this
the composite display include anterior and posterior corneal is true (ie, toric-ellipsoid more closely mimics the shape of
elevation using a best-fit sphere (BFS), pachymetric distri- the cornea), this is exactly opposite of what a functional ref-
bution, and anterior sagittal curvature. Screening requires erence surface should do when screening normal patients.
consistent scales, colors, and settings to allow the practitio- The purpose of the reference surface is to highlight or mag-
ner to rapidly evaluate the preoperative map. nify surface features to allow the clinician to identify things
 To m o gra p h e r s 121

Figure 6 -32. Four-view refractive display of a cornea with iso- Figure 6 -33. Four-view refractive display of a cornea showing
lated posterior elevation. an abnormal displacement of the thinnest point.

such as astigmatism or conical protrusion (ectasia or kera- ultrasound pachymetry and Placido-based topography. The
toconus). A reference surface, such as a toric-ellipsoid, will sagittal curvature (lower right), the map most commonly
more closely fit the astigmatic cornea or the conical cornea displayed on a Placido system, is completely normal. The
and will effectively mask the pathology. Keratoconus, for curvature map is symmetric and shows minimal astigma-
example, is a progressive disorder, the hallmarks of which tism. The simulated keratometry readings are in the mid
are stromal thinning, conical protrusion, corneal scarring, 40s. Central ultrasound pachymetry readings were 520
decreased spectacle-corrected visual acuity, Fleischer ring, µm. All of these values would be considered well within
and, in advanced cases, Vogt’s striae.18 The composite the normal range. This is, however, an incomplete picture.
screening display outlined previously (see Figure 6-31) has The pachymetric distribution is distinctly abnormal with a
significant advantages for identifying ectatic change. marked inferior displacement of the thinnest region. The
Figure 6-32 is a 4-view composite map of an asymptom- thinnest area of the cornea is 499 µm, significantly thin-
atic patient presenting for refractive surgical evaluation. ner than the central reading. The posterior elevation map
The map demonstrates the importance of both the posterior shows a well-circumscribed, clearly demarcated, island of
corneal surface and the pachymetric distribution map. The positive elevation representing the area of ectatic change.
patient had 20/20 best spectacle-corrected vision, normal The area of posterior ectasia and the thinnest corneal region
Placido topography, and central ultrasound pachymetry coincide. This patient has subclinical keratoconus in spite of
of 540 µm. The Pentacam composite display reveals the normal Placido topography and normal ultrasound pachym-
limitations of relying on anterior curvature and central cor- etry. The Pentacam clearly reveals the shortcomings of the
neal thickness measurements alone. The anterior elevation original incomplete corneal analysis.
(upper left) and anterior sagittal curvature (lower right) are
all normal, both maps displaying a small amount of astig-
matism. The pachymetric map (lower left) shows a normal Refract ive Screen in g
central reading of 539 µm. The pachymetric distribution is
significantly displaced, however, where the thinnest region While the basis for the development of ectasia may ulti-
(501 µm) is inferotemporal to the central cornea. The dif- mately be a structural, genetic, or biochemical abnormality
ference between the thinnest reading and the apical reading of the cornea, measurable structural components, such as
is 38 µm. The posterior elevation (upper right) is distinctly elevation and pachymetry, should be thoroughly evaluated.
abnormal. The posterior map shows a well-demarcated, With subclinical disease, curvature alone may not provide
paracentral island of positive elevation (>34 µm) off the enough information to detect early corneal abnormality.
BFS representing an ectatic change. The pachymetric dis- The goal of the Belin/Ambrósio Enhanced Ectasia
tribution also corresponds to the area of posterior elevation, display (Figure 6-34) is to combine elevation-based and
further increasing its significance. pachymetric corneal evaluation in an all-inclusive display.
Similarly, the patient depicted in Figure 6-33 would This gives the clinician a global view of the structure of the
have a completely normal exam if one relied solely on cornea and allows the physician to quickly and effectively
122 Ch a p t e r 6

Figure 6-34. The Belin /Ambrósio

Enhanced Ectasia screening dis-
play.

screen patients for ectatic disease. The elevation maps and ectasia in patients reportedly with normal preoperative
pachymetric data are placed side-by-side in a comprehen- exams.23 How many of these so-called “normal” exams are
sive display. By evaluating these measurements from dif- truly normal and how many had undiagnosed changes on
ferent perspectives, the ability to identify abnormalities is their posterior corneal surface is unknown. Additionally,
increased. The elevation and pachymetric components of reported residual bed computations based on central ultra-
the display are designed to be complimentary. sound readings may significantly overestimate the actual
The creation of a pachymetric map allows for the identi- residual bed. Like flap thickness unpredictability, pachy-
fication of the true thinnest point and can contrast the thin- metric distribution inconsistency may represent a previ-
nest point with the geometric center of the cornea. In 12% ously unrecognized confounding variable.
of normal patients, the pachymetry difference between
the thinnest point and the geometric center of the cornea
is more than 10 µm.19 The distance between the thin- Cor n ea l Mor p h olog y
nest point and the geometric central point is significantly
higher in keratoconus patients. A reliable pachymetric map The Pentacam’s complete corneal analysis is also
is therefore essential for determining the localization and advantageous in characterizing keratoconus morphology.
value of the cornea’s thinnest point. Along with the thin- Traditionally, the morphology of the cone was categorized
nest point evaluation, the data from a full pachymetric map by descriptions based solely on anterior curvature analysis.
enable characterization of the thickness profile of the cor- Sagittal curvature, however, is a very poor indicator of cone
nea. Because the absolute central (or thinnest) pachymetry location and morphology. Figure 6-35 reveals the limita-
value varies significantly among a normal population, the tions of curvature analysis in the assessment of cone loca-
relationship between the central and peripheral cornea can tion. Sagittal curvature maps, whether generated by Placido
be an indicator of pathologic thinning. The normal cornea systems or by elevation systems, do not depict shape nor do
is thinner at the center, with a gradual increase in thick- they accurately locate the cone. One can see that the sagit-
ness toward the periphery.20-22 These concepts and the tal curvature map (lower right) would incorrectly place the
ability to accurately measure the corneal thickness limbus- apex of the cone below the limits of the 9.0-mm display.
to-limbus add significantly to our diagnostic capability. The true apex of the cone, however, is slightly below the
These data can only be obtained with an elevation-based pupillary margin as shown in both the pachymetry map and
topographer and accurate measurement of the posterior the anterior and posterior elevation maps. Accurately iden-
corneal surface. tifying cone location and size are crucial in planning Intacs
The Belin/Ambrósio display allows the clinician to view surgery and corneal transplantation. Elevation maps are
the pachymetric profile and view the corneal elevation as it inherently more accurate than sagittal curvature displays in
appears with an optimized reference surface. The literature depicting the morphology of the cone and should be used to
is replete with numerous articles of iatrogenic post-LASIK classify keratoconus.
 To m o gra p h e r s 12 3

Figure 6-36. Corneal indices cal-

culated by the Pentacam.

Figure 6 -35. Four-view refractive display revealing proper

cone localization on the elevation and pachymetric maps.
TABLE 6-4.

In d ividu al Displays
Pentacam Corneal Indices
Single map displays are available for corneal thickness, Steep (K2/Rs): Simulated keratometry value for the
anterior and posterior elevations, and anterior and poste- steepest meridian on both the anterior and posterior
rior sagittal and tangential curvatures. They are selected by surfaces computed at the central 3-mm zone.
choosing the “1 Large Color Map” option and then select- Flat (K1/Rf): Simulated keratometry value for the
ing the map from the drop-down list. All of the single maps flatest meridian on both the anterior and posterior
share a common presentation of corneal indices, which is surfaces computed at the central 3-mm zone.
shown on the left side of the display (Figure 6-36).
Rf/K1: Flat central radii in the 3-mm zone.
Corn eal In d ices Rs/K2: Steep central radius in the 3-mm zone.
Rm/Km: Mean central radius, arithmetic average of
The steep (K2/Rs) and the flat (K1/Rf) simulated kera-
Rf and Rs.
tometry values are shown for both the anterior and posterior
surfaces. These are computed at the central 3-mm zone and QS: Stands for “quality statement” and is a check on
are always orthogonal (ie, 90 degrees apart). The orienta- the image acquisition quality. Here “OK” means the
tion of the major and minor meridians is shown in the small acquired image was of sufficient quality.
diagram on the left side of the field. It should be noted that, Axis: Axis of corneal astigmatism (red for steep,
even in highly irregular corneas, the simulated K values blue for flat and user selectable).
will be reported as orthogonal and that this does not suggest
Astig.: Corneal astigmatism in the central 3 mm.
that the cornea is regular. Other values depicted are listed
in Table 6-4. Q-val.: Corneal shape factor of the cornea in
30 degrees.
Com pare Exam s Rper : The mean radii of the zone between the 7- and
9-mm ring.
These displays allow the user to show different exams
from the same patient to compare and to observe the differ- Rmin: Minimum radii of the cornea.
ences between exams such as pre- and post-LASIK (Figure For the conversion of the curvature readings into refractive power on
6-37). Any of the available single maps may be selected, the anterior surface, a refractive index of n = 1.3375 is used and for
and their differences will be computed. Additionally, the the posterior surface the refractive index n = 1.376 for cornea and
user may left-click on any of the maps, and the correspond- 1.336 for aqueous is used.
ing individual point value will be displayed on all the maps.
This display is particularly useful for showing surgical
effect or following a progressive disease.
124 Ch a p t e r 6

TABLE 6-5.

Curvature Indices
Index of Surface Variance (ISV): Gives the deviation
of individual corneal radii from the mean value. This
index is elevated in all types of irregularity of the cor-
neal surface (scars, astigmatism, deformities caused by
contact lenses, keratoconus, etc).
Index of Vertical Asymmetry (IVA): Gives the degree of
symmetry of the corneal radii with respect to the horizontal
meridian as axis of reflection. Elevated in cases of oblique
axes of astigmatism, in keratoconus, or in limbal ecstasies.
Keratoconus Index (KI): Elevated especially in kera-
Figure 6 -37. Comparison of 2 examinations using the Pentacam. toconus.
Center Keratoconus Index (CKI): Elevated especially
in central keratoconus.
Index of Height Asymmetry (IHA): Gives the degree
of symmetry of height data with respect to the horizon-
tal meridian as axis of reflection. Analogous to IVA,
although sometimes more sensitive.
Index of Height Decentration (IHD): This index is
calculated from Fourier analysis of height and gives the
degree of vertical decentration. Steeper in keratoconus.
Radii minimum (RMin): Gives the smallest radius of
curvature in the entire field of measurement. Elevated in
keratoconus.
Topographical Keratoconus Classification (TKC):
Based on anterior corneal data the keratoconus is classi-
fied based on the Amsler/Muckenhirn staging.
Note: This classi cation ignores the degree of thinning and ectatic
changes on the posterior surface. This classi cation would classify as
normal eyes with signi cant posterior changes associated with thinning
if the anterior surface was uninvolved and is also prone to false positives
Figure 6-38. Topometric display on the Pentacam. in corneas with a displaced apex.

Top om et ric Display

Pach y m et r ic Disp lay
The Topometric display (Figure 6-38) is designed to ana-
lyze the cornea based on anterior and posterior curvature The individual pachymetric display (Figure 6-39)
data. This display presents a number of anterior curvature provides the color map of the distribution of corneal thick-
values commonly found on Placido-based systems. The ness, the thickness of the cornea at the apex, the pupil-
display shows anterior and posterior sagittal curvature lary center, and the thinnest corneal point. It provides
with their associated simulated K values, pachymetry, true the corneal volume calculated at the 10-mm optical zone
net power, and asphericity values. Additionally, it displays centered on the apex, the chamber volume at the 12-mm
the 8 curvature indices listed in Table 6-5. Note that these zone centered at the apex, and the ACD. The 2 large graphs
indices, which were derived from Placido-based data only, (Corneal Thickness Spatial Profile [CTSP] and Percentage
consider the anterior corneal surface. Their predictive and Thickness Increase [PTI]) and the progression index show
descriptive abilities are limited by not evaluating either the how the corneal thickness changes relative to data from a
posterior surface or the pachymetric distribution. normal population. The graphs show the “average” normal
values and ±2 SD (95% confidence interval). Pachymetric
progression data are very useful in screening for ectatic
disease.
 To m o gra p h e r s 12 5

Figure 6 -39. Pentacam Individual Pachymetric display. Figure 6 -40. The Holladay Equivalent Keratometry Readings
detailed report.

Hollad ay “Eq u iva len t K” Ph a k ic In t rao cu la r Len s

Det a il Rep or t Sim u lat ion Progra m
As more and more of the early refractive surgery
The 3-dimensional simulation program for phakic IOLs
patients develop cataracts, the need for accurate postre-
was designed to assist surgeons in the preoperative plan-
fractive IOL calculations increases. Unfortunately, many
ning for the implantation of phakic IOLs into the anterior
of these patients do not have their prerefractive corneal
chamber and to determine whether adequate clearances are
measurements available, making the historical method of
present for safe implantation (Figure 6-41).
IOL calculation impossible. Standard keratometry assumes
The specific phakic IOL type and power can be chosen
normal spherocylindrical optics and assumes a constant
from the database. The selected lens is located on the iris
relationship between the anterior and posterior corneal sur-
automatically. The minimum distances between phakic IOL
faces. Standard keratometry predicts overall corneal power
and endothelium and natural crystalline lens are calculated
based solely on anterior curvature measurements. These
in all dimensions and automatically displayed. The phakic
assumptions are usually accurate in “normal” eyes, and this
IOL alignment can be modified by the user. The phakic
explains the excellent refractive results typically seen with
IOL simulation software offers an automatic calculation of
modern cataract surgery. After refractive surgery, however,
the refractive power of the phakic IOL with reference to the
these assumptions no longer hold true, and using standard
subjective refraction of the patient.
“K” values in IOL formulas may lead to a significant devia-
Additionally, due to the normal growth of the lens with
tion from the desired refractive result.
age and subsequent shallowing of the anterior chamber, an
Because the relationship between the anterior and poste-
aging prediction module was incorporated. The age of the
rior corneal surfaces deviates from typical after refractive
patient can be changed while clicking on the arrow keys or
surgery, the optical approximation of the corneal power
typing in a specific age. The software assumes a growing
requires data from both anterior and posterior surfaces as
of 18 µm per year for the crystalline lens. The outcome of
well as the index of refraction of the cornea and aqueous.
this is an anterior movement of the iris toward the cornea.
Additionally, the power change from the corneal center
Depending on the predicted age, the clearances in the
to the periphery may change dramatically after refractive
Scheimpflug image change. The minimum clearances in
surgery, and the overall power may be more dependent
the anterior chamber are shown on the lower right-hand side
(postoperatively) on the pupillary size. The Equivalent
of the display.
Keratometer Reading (EKR) uses all of the above informa-
tion to produce a graphic and tabular representation of the
“adjusted” postsurgical “K” readings at varying pupil sizes
(Figure 6-40). In the absence of historical data, the use of
Con t act Len s Fit t in g Disp lay
the EKR (at an appropriate pupillary size) in standard IOL The complex interaction between a rigid contact lens and
formulas may allow for more precise postrefractive IOL the corneal surface can only be simulated if you know the
calculation. true corneal shape (elevation) and if the system is capable
126 Ch a p t e r 6

Figure 6 -41. Pentacam phakic IO L simulation program. Figure 6 -42. Contact Lens Fitting display on the Pentacam.

of imaging more of the corneal surface that is typical of a designs. The lens selection can then be modified based on
reflective Placido-based system. If the total corneal shape the simulated lens/cornea alignment and predicted fluores-
is known, rigid gas-permeable lenses can be “fit” to the cein pattern.
corneal surface and the lens/corneal relationship can be The development of elevation-based imaging systems
predicted. The lenses can be adjusted both in overall diam- that allow assessment of both the anterior and posterior cor-
eter, base curve, and peripheral curves, and a simulated neal surfaces marks an evolutionary change in computer-
fluorescein pattern can be produced. ized corneal analysis. Accurate assessment of both corneal
The Contact Lens Fitting Display (Figure 6-42) allows surfaces and complete pachymetric distribution significant-
the practitioner to either design a custom lens or to select ly enhances our knowledge and should allow for better and
from a large number of commercially available rigid lens more complete corneal and refractive evaluations.

P RECISIO SURGICAL TOMOGRAPH ER

Aleksandar Stojanovic, MD and Xiangjun Chen, MD, MS

P
recisio (Figure 6-43) is a modern corneal topogra- and the focus camera—and is further converted to a digital
pher/tomographer that measures limbus-to-limbus format for analysis. The main camera is mounted on a rota-
elevation of the anterior and posterior corneal sur- tional wheel, perpendicular to the projection of the slit, to
faces, corneal pachymetry, as well as the ACD, using the capture the images of the anterior and posterior surfaces of
Scheimpflug principle. It is designed as a custom surgical the cornea, the anterior surface of the iris, and the crystal-
data acquisition device to be used as a part of the iVIS-Suite line lens, while the focus camera, mounted on the rotational
laser system. Its resolution, sensitivity, and specificity are axis, detects the positional data of the examined eye and
fine-tuned for detecting the details of corneal morphology monitors the fixation (Figure 6-44). For each examination
with accuracy and repeatability suited for use in custom lasting 1 second, the main camera acquires 50 images dur-
ablation treatments. ing its 360-degree rotation, generating 39,000 data points
During the examination, Precisio projects a rotating on both the anterior and posterior surfaces. The elevation
slit onto the anterior segment of the eye at an angle of data captured by the main camera are paired with the posi-
20 degrees from its rotational axis, which coincides with tional data from the focus camera. The focus camera also
the visual axis of the examined eye. The slit image is captures the iris and limbal vessel architecture with the help
captured by 2 integrated CCD cameras—the main camera of a secondary illumination system consisting of 4 infrared
 To m o gra p h e r s 127

Figure 6 -43. Precisio topo /tomographer.

Figure 6 -45. Yellow shaded concentric area from 8 to 14 mm

from the pupil center showing iris and limbal vessel architec-
ture used for eye registration and cyclotorsional tracking.

Figure 6-44. The main camera captures a perpendicularly

projected slit during a 360-degree rotation, while the focus
camera, located on the rotational axis, detects the positional
data.

LEDs and 4 white LEDs for illumination of a concentric

area from 8 to 14 mm from the pupil center (Figure 6-45).
The image of the iris and limbal vessel architecture is used
Figure 6-46. Precisio (top) uses a higher optical magnification
for eye registration and for cyclotorsional tracking during compared to Pentacam (bottom) and consequently covers a
the surgery with the iVIS-Suite laser system. smaller space.
As a surgical data acquisition device, Precisio shows
some clear differences compared to the other Scheimpflug
topo/tomographers, designed primarily for diagnostic pur- • A fully automated acquisition process based on
poses (eg, Pentacam). To maximize the accuracy of the 4 software-aided motors, which allows an examina-
data from the cornea, the main camera of Precisio uses a tion without introduction of any operator’s vibration
higher optical magnification compared to the Pentacam to the system
(Figure 6-46), thus covering only the space between the
cornea and the anterior surface of the lens. Consequently, • An eye tracker, which compensates for x, y micro
a higher resolution image is obtained, sacrificing, however, movements and cyclotorsion of the eye during the
information on the structure behind the surface of the crys- examination
talline lens. In order to maximize the repeatability of the Finally, Precisio provides quality control of the surgical
data, the following unique features are implemented: data by not allowing export to the custom ablation planning
• A heavy and rigid mechanical construction to mini- software until the requirement for repeatability of ±3 µm
mize vibration induced by the rotating structure dur- for the anterior elevation examinations within the central
ing the acquisition 6 mm is reached.
128 Ch a p t e r 6

Figure 6-47. Precisio’s default maps—anterior elevation (left), Figure 6 -49. Posterior curvature maps—axial (left), tangential
posterior elevation (right superior), and pachymetry (right infe- (right superior), and total power (right inferior).
rior) in a case of preclinical keratoconus.

Figure 6 -50. ACD maps.

Figure 6-48. Anterior curvature maps—axial (left), tangential
(right superior), and total power (right inferior).
Placido-based systems’ detection of localized increases in
curvature.24,25 Although such measurements are reproduc-
Precisio’s default display presents anterior elevation, pos- ible in normal corneas, this is not the case in extremely
terior elevation, and pachymetry maps in one view (Figure irregular corneas, such as in keratoconus, where a deviation
6-47). The curvature maps (axial, tangential, and mean in the tear film thickness will significantly affect the mea-
power) of the anterior and posterior corneal surfaces are surements. A thin tear film will cause nonexisting surface
also available (Figures 6-48 and 6-49), as well as ACD map irregularities to show as artifacts, whereas a thick tear film
(Figure 6-50) and a 3-dimensional view of the cornea. may prevent some real irregularities from showing. Even
Clinically, Precisio can be used as an instrument for more importantly, the Placido-based systems do not mea-
diagnostic purposes, for topography-guided refractive and sure the posterior surface of the cornea, which as a rule
therapeutic treatments, as well as for analysis of the effec- shows keratoconus changes before the anterior surface. The
tiveness of the refractive treatments. early keratoconus changes do not show on the anterior sur-
face due to epithelial remodeling, which causes a masking
effect by initially smoothing the keratoconic changes of the
Diagn o st ics anterior corneal surface.26,27 The Precisio’s high-resolution
Scheimpflug images provide precise anterior and posterior
Probably the most important task of preoperative evalu- elevation and pachymetry maps of the cornea and represent
ation in refractive surgery is the early detection of kera- a solid base for identification of preclinical keratoconus (see
toconus. Before the development of scanning slit systems, Figure 6-47).
the diagnosis of keratoconus was mainly dependent on the
 To m o gra p h e r s 12 9

Figure 6 -52. “Regularity index” represents an indication of

Figure 6-51. “Clinical follow-up” display: Difference map (left) irregularity of the corneal surface. Normal virgin cornea (left)
between anterior elevation maps before (right superior) and and a cornea after a decentered LASIK (right).
after (right inferior) CXL treatment in a keratoconus eye.

described by Alessio.28 Basically, to program a refractive

To monitor progression of corneal shape changes, surgery, pupillometry and the patient’s refraction are also
Precisio’s “clinical follow-up” mode provides comparison fed to the CIPTA software, which then compiles a custom
of 2 examinations of the same eye performed at different ablation plan with the aim of transforming the actual cor-
times (Figure 6-51). neal shape into a regular aspheric shape of desired curva-
Precisio also provides so-called “Regularity index,” ture, within the treatment zone suggested by pupillometry.
which represents an indication of irregularity of the corneal The volume of the ablation is defined by the intersection
surface compared to its best aconic surface within the cen- between the corneal anterior surface as detected by Precisio
tral 6-mm area (Figure 6-52). and the targeted aspheric surface determined by CIPTA.
The curvature of the targeted surface is determined by sub-
tracting the amount of desired dioptric change from the pre-
Su rgica l Design operative curvature. Excimer laser-based lamellar corneal
transplantation programmed by Corneal Lamellar Ablation
Any excimer laser treatment is based on an ablation plan for Transplantation (CLAT) software is also built around
expressed in µm of corneal tissue to be removed at any the corneal topography/tomography information imported
particular coordinate of its surface. This implies that, inde- from Precisio. CLAT first defines an “Ideal Corneal Bed”
pendent of the type of the information used for the custom with uniform thickness of residual healthy corneal tissue,
ablation planning (wavefront aberrometry, corneal curva- based on the thinnest point of the pachymetry map provided
ture, corneal elevation topography, or corneal wavefront), by Precisio. CLAT ablation of the host cornea is performed
a final calculation must be performed to find the depth of to achieve this uniform bed, which then takes on membrane
ablation expressed in µm on the corneal level. Placido-based properties and assumes a regular shape. Finally, a trephine
topography measures corneal curvature in D. To obtain is used to cut the donor’s cornea to the diameter of the
the necessary corneal elevation data, a mathematical con- receiving bed, while the laser adjusts the thickness of donor
version from D to µm is needed, resulting in “secondary cornea to suit the residual thickness of the receiving bed by
data” prone to cumulative error. Precisio, however (along performing an even thickness ablation from the endothelial
with other Scheimpflug-based topographers), measures the surface of the donor cornea. Figure 6-53 shows the principle
corneal shape directly in µm of corneal elevation, providing of the CLAT, while Figure 6-54 shows an example of topo-
high-quality “primary data” for topography-guided custom graphic outcome after CLAT transplantation.
ablation planning. Precisio’s “Surgical follow-up mode—iVerify” evaluates
Precisio is designed to provide data for the iVIS-Suite the effect of the laser ablation by comparing the delivered
custom ablation system. In order to accomplish this, corneal (real) ablation map with the planned (expected) ablation
anterior elevation and pachymetry data are fed to Corneal map. The real ablation map is calculated by subtracting
Interactive Program Topographic Ablation (CIPTA) plan- the postoperative elevation map from the preoperative
ning software, while the iris and limbal vessel architecture elevation map, while the expected ablation map is imported
data are transferred to the laser and are used for eye regis- from CIPTA. Thanks to the Precisio’s registration informa-
tration and cyclotorsional eye tracking during the surgery. tion, any cyclotorsional or x, y misalignment between the
The details of the working principle of CIPTA have been preoperative and the postoperative maps is accounted for.
130 Ch a p t e r 6

Figure 6 -55. Ablation difference map (left) shows the differ-

Figure 6-53. (A) CLAT performed in a keratoconus eye. (B) An ence between the real (right superior) and the expected (simu-
“Ideal Corneal Bed” with uniform thickness of healthy cor- lated) ablation (right inferior), in an eye with excellent results
neal tissue is obtained by pachymetry-guided laser ablation. after treatment for low myopia.
(C) A uniform bed assumes a regular shape after the ablation.
(D) The thickness of donor cornea is then reduced to suit the
residual thickness of the receiving bed by a laser ablation from
the endothelial surface. (E) A trephine is used to cut the donor’s
cornea to the diameter of the receiving bed. (F) The button is
turned around with the endothelial side facing the host bed
and (G) sutured in place.

Figure 6 -56. Ablation difference map (left) shows the differ-

ence between the real (right superior) and the expected (simu-
lated) ablation (right inferior), in an overcorrected eye after
treatment for low myopic astigmatism.

Figure 6 -54. Anterior elevation map in an eye with keratoco-

nus, before (left) and after (right) the CLAT treatment.
Con clu sion
Precisio is a unique topographer/tomographer, especially
The ablation difference map shows the differences in the designed to provide precise surgical data for topography-
maximum ablation depth, the volume, and the achieved guided custom ablation and laser-assisted corneal lamellar
change in spherical equivalent and cylinder between the transplantation, to analyze the results of refractive surgery
real and the expected ablation (Figures 6-55 and 6-56). and to provide reliable diagnostic information for corneal
These differences are the result of the efficiency of the evaluation and presurgical screening.
laser and the corneal biological and biomechanical factors.
Thus, iVerify is meant to be used as a tool for sophisticated
nomogram building.
 To m o gra p h e r s 131

VISANTE O MNI
Amin Ashrafzadeh, MD

A
new era was ushered in with the advent of optical
TABLE 6-6.
coherence tomography (OCT). Initially, the OCT
was recognized for the diagnostic capabilities in
the treatment of retinal disorders. The Stratus OCT has an
Maps Included in the Predefined Themes
820-nm diode that was optimized to reach the retina and STANDARD ALTERNATE HOLLADAY
reduce light scatter. The 820-nm light, however, is not opti-
mal for anterior segment. The 1310-nm light has increased Anterior axial Anterior mean Anterior axial
water absorption, allowing the Visante OCT to have a curvature curvature curvature
20-fold increase in light intensity with decreased retinal (ATLAS) (ATLAS) (ATLAS)
exposure. This allows for faster scanning at the same signal- Pachymetry Pachymetry Pachymetry
to-noise ratio, reducing motion artifacts. Additionally, the (Visante OCT) (Visante OCT) (Visante OCT)
1310-nm light has reduced scattering of the light, allowing Anterior eleva- Anterior eleva- Anterior eleva-
for better penetration into turbid and more opacified tis- tion sphere tion toric ellip- tion toric ellip-
sue. This increased signal-to-noise ratio allows for a faster (ATLAS) soid (ATLAS) soid (ATLAS)
processing with increased detail of structures, such as the
cornea, sclera, and the anterior chamber angle.29 Anterior mean Posterior mean Anterior mean
curvature curvature (Both) curvature
(ATLAS) (ATLAS)
Tech n ica l In for m at ion Relative Relative Relative
pachymetry pachymetry pachymetry
Visan te An terior Segm en t OCT (Visante OCT) (Visante OCT) (Visante OCT)

The Visante Anterior Segment OCT (Carl Zeiss Meditec, Posterior eleva- Posterior eleva- Posterior eleva-
Dublin, CA) is a real-time, low-energy infrared laser, time tion sphere tion toric ellip- tion toric ellip-
domain OCT employing a 1310-nm light. Images may be (Both) soid (Both) soid (Both)
acquired using 3 modes. The Anterior Segment Mode,
referred to as “Ant Seg,” is a 16-mm by 6-mm image with
256 optical “A Scans” in a span of 0.125 seconds. The High
Resolution Mode, referred to as the “Hi Res” mode, is a The pachymetry scans are performed in either the
10-mm by 3-mm image with 512 optical “A Scans,” in a standard pachymetry mode, which performs 8 sequential
span of 0.25 seconds. The Pachymetry Mode is a modified images in rotating fashion spaced every 22.5 degrees apart,
Hi Res Mode, 10-mm by 3-mm image, however, with only or a Global Pachymetric Map, where 16 sequential pachy-
128 optical “A Scans” in a span of 0.0625 seconds. The metric images are acquired spaced every 11.25 degrees
axial (depth) resolution of the Visante OCT is 18 µm, and apart. The Pachymetric Map is acquired in 0.5 seconds,
the lateral (transverse) resolution is 60 µm. This information and the Global Pachymetric Map is acquired in 1.0 seconds.
is presented in Table 6-6. The repeatability of the pachymetric map in the standard
The Visante is capable of rotating the OCT drum mode was noted to be 7 µm standard deviation in the cen-
360 degrees. The images can then be acquired using ter and 14 µm standard deviation in the periphery.30 The
2 modes: Ant Seg or Hi Res Scans may also be per- central corneal 10-mm circular pachymetric map provides
formed in the “Enhanced” mode where 4 sequential images 2048 data points.
are acquired and then compressed into a single image. The relative pachymetric map compares the patient’s
Additionally, the Ant Seg or the Hi Res modes can also be corneal pachymetric map against a standard normal cornea
performed in a dual- or quad-scan mode. The dual-scan with a central pachymetry of 550. This relative pachymetric
mode is performed at the 20- to 200-degree axis and at map produces a percent deviation compared to the norm.
the 160- to 340-degree axis. This dual-scan mode is also This map provides a rapid visual analysis of abnormality.
capable of being rotated, but the degree separation remains This information is directly calculated by the Visante and
constant. In the quad scan mode, the 4 scans are in equal is presented in the middle, lower position.
45-degree separations. The Visante 3.0 software also allows The Visante is also equipped with an optometer that
for customization of the scan in terms of the number of ranges from +20 to -35 D. The optometer is used to bring the
scans and also the degrees and the type of scans. focusing reticle into view for patients with refractive error.
132 Ch a p t e r 6

TABLE 6-7.

Scan Modes
ANT SEG HI R ES PACHY
Scan Size 16 mm by 6 mm 10 mm by 3 mm 10 mm by 3 mm
# of A-Scans 256 512 128
Acquisition 0.125 seconds 0.25 seconds 0.0625 seconds per B-scan; 0.5 seconds for regular
time pachymetry, 1.0 second for global pachymetry
Modes Single, double, Single, double, quad, 8 scan regular pachymetry or 16 scan global
quad, single single enhanced pachymetry
enhanced

Additionally, it can also be used to induce accommodation. ATLAS Corn eal Top ograph er
The ability to induce accommodation allows for studying
the movement of the anterior surface of the crystalline lens The ATLAS corneal topographer is a Placido disk-based
or the accommodating IOLs. The capacity to evaluate the system. Models 993, 995, and 9000 are capable of link-
anterior surface lens information in the setting of phakic ing with the Visante 3.0. The ATLAS topographers have a
IOLs is important as the vault distance may dramatically 22-ring Placido disk, 18 superior, and 22 inferior to the point
change with accommodation. of fixation. The latest model, the 9000, uses a 950-nm light
The Visante is also equipped with several software source, which is not visible to the patient, is well-tolerated
modules. The refractive module allows for placement of by dilated or light-sensitive patients, and does not induce
“Residual Stromal Bed” line in the Hi Res modes using any alteration in pupil size.
standard, enhanced, or combined scans that can be adjusted Because the information obtained by the ATLAS topog-
to the desired amount as a set distance from the endothe- rapher is from the reflection of light from the Placido disk,
lial surface. This module may be either used for surgical all data pertain to the anterior surface. The Visante 3.0 ver-
planning, patient education, or postoperative evaluation of sion is upgraded with a tracking device to ensure corneal
outcomes. pachymetric mapping occurs with sequential scans centered
In the Anterior Segment (Ant Seg) modes, the “Anterior to the same location. Additionally, the vertex of the cornea
Chamber Rainbow” provides standardized lines of 0.5, is recognized as the point with greatest signal in a verti-
1.0, and 1.5 mm distances into the anterior chamber from cal scan. The vertex of the ATLAS scans and the Visante
the endothelial surface. This tool is very useful for the pachymetric scans are matched to represent the same loca-
evaluation of the anterior chamber when considering phakic tion. By using the corneal thickness map (pachymetric
IOLs. data), with known anterior surface curvature data from the
The “Phakic IOL” tool has a list of Artisan lenses ATLAS, the posterior corneal curvature data are calculated.
(Ophthec, Groningen, Netherlands) and models including
the 202 (5-mm Verisyse) and 204 (6-mm Verisyse) that are
approved by the US FDA. The “Safety Distance” tools mea- Disp lays
sure the distance from the endothelial surface to the edge of
To ease clinical use, various displays may be chosen.
either the model or the real postoperative anterior chamber
The ATLAS supplies the axial, tangential, mean curvature,
implant. Additionally, the “Vault Distance” tools are used
anterior elevation, corneal refractive, and corneal wavefront
to measure the distance from the posterior surface of the
maps. The Visante supplies pachymetry data and enables
implant to the anterior surface of the crystalline lens.
calculation of posterior surface topography and power.
The “Irido-Corneal” tools allow for measurement of the
Table 6-7 presents preset formats of presentation; however,
angles with more precise tools. The tool is manually placed
they can also be personalized.
at the scleral spur. Once placed, the Visante will auto-
matically calculate the distance along the internal corneal
surface at 500 and 750 µm. Then, the tool will continue to Th e Hollad ay Rep ort
calculate the appropriate angle-opening distance (AOD) at In cooperation with Jack T. Holladay, MD, MSEE,
500- and 750-µm marks along with the trabeculo-iris sur- Houston, TX, the Holladay Report has been created to
face area (TISA) at the 500- and 750-µm areas. produce a single page of information that provides all per-
tinent information in evaluating candidacy for refractive
 To m o gra p h e r s 13 3

Figure 6 -57. A normal, spherical cornea. Figure 6 -58. A normal toric cornea.

procedure. The axial curvature map, placed in the top left surface is then represented as points of elevation above (red)
corner, is imported directly from the ATLAS. The mean or below (blue) such theoretical best-fit surface. Sphere and
curvature map, placed in the bottom left corner, is also toric ellipsoid are 2 such surfaces, and they provide vari-
imported directly from the ATLAS. The anterior elevation able data. Figure 6-58 illustrates the variations between the
map is directly imported from the ATLAS and is presented BFS and the best-fit toric ellipsoid for both the anterior and
in the top right corner. The elevation data, combined with the posterior surfaces. The BFS model over such astigmatic
the pachymetric data, allow for calculation of the posterior cornea has areas that fall below (blue) such sphere surface
surface, and the posterior elevation map is presented in the on the vertical meridian and areas that are above (red) such
right lower corner. The pachymetry map is in the middle. sphere surface on the horizontal meridian. By taking such
reference surface and altering it to a best-fit toric ellipsoid,
Refract ive Evalu at ion the curvatures of the astigmatic cornea are now compared
to a more similar geometric surface shape. This example
The evaluation of the LASIK patient both preoperatively
for this patient points out the lack of ease in “user friendli-
and postoperatively is enhanced with the Visante Omni.
ness” of BFS data. The best-fit toric ellipsoid represents a
Accurate posterior surface information may detect early
modified model that takes into account a regular astigmatic
ectasia in candidates as well as those who are postkeratore-
cornea.
fractive surgery.
Figure 6-59 demonstrates forme fruste keratoconus.
A normal, spherical cornea is presented in Figure 6-57.
A 36-year-old Hispanic woman presented for consider-
The Holladay Report of a 25-year-old White woman with
ation of refractive surgery. Her manifest refraction was
plano refraction and 20/15 visual acuity in both eyes is
-0.50 -3.25 x 018 in her right eye (20/20) and -1.25 -2.75 x
shown. It is to be noted that because her corneal pachym-
175 in her left eye (20/25). Her ultrasound corneal pachym-
etry is 506 µm by the Visante, her relative corneal pachy-
etry values were 531 µm OD and 521 µm OS. Despite a rela-
metric map uniformly shows an 11% deviation from the
tively normal axial map, the Visante Omni demonstrated an
normal 550-µm cornea. Clearly, her corneas are normal,
elevation of greater than 10 µm above the reference best fit
but the percent deviation data are used to spot not only eyes
toric ellipsoid on posterior elevation maps. The threshold
with uniform deviation, but more importantly the eyes with
for concern on the posterior elevation map is 10 µm. Note
abnormal pattern deviation.
the correlation between the slightly heavier inferior bow tie
A normal toric cornea is presented in Figure 6-58. A
on the axial map, highlighted inferior temporal area on the
33-year-old White woman presented with a history of stable
mean curvature map, the inferotemporal displacement of
refraction. Her manifest refraction resulted in 20/15 in each
the thinnest point, and the elevation inferotemporally on the
eye with -0.50 -4.00 x 003 OD and -0.50 -4.00 x 180 OS.
anterior and posterior elevation maps. The plus sign on all
Corneal ultrasound pachymetry was 550 µm OD and
6 maps indicates the same anatomical location. Typically,
549 µm OS. The Visante Omni presented a regular bow-tie
3 indicators of ectasia contraindicate LASIK, and this
astigmatism with a regular pachymetry map. Mean curva-
patient has 6. The patient was diagnosed with forme fruste
ture and the anterior elevation maps were unremarkable.
keratoconus.
The calculated posterior elevation presented the highest
Figure 6-60 demonstrates keratoconus. A 37-year-old
elevation point at 4.32 µm in the best-fit toric ellipsoid map.
White man presented for consideration of refractive sur-
The purpose of the elevation map is to compare a
gery. His manifest refractions were -1.00 -1.50 x 025 OD
patient’s cornea to some theoretically perfect geometric
(20/25) and -1.50 -1.25 x 152 OS (20/30). Ultrasound
surface that should nearly match. Deviation from such
corneal pachymetry found 430 µm OD and 427 µm OS.
134 Ch a p t e r 6

Figure 6 -59. Forme fruste keratoconus. Figure 6-61. PMD.

Figure 6 -60. Keratoconus.

B

Note the 19 µm of anterior elevation on the anterior best fit

toric ellipsoid surface and 45.49 µm of protrusion above the
posterior surface. This disparity is noted as a consequence
of the corneal epithelial thinning noted at the tip of the
cone, which masks the stromal protrusion, yet there are no
compensatory mechanisms on the endothelial surface.
Figure 6-61 demonstrates PMD. A 54-year-old Black
woman had a refraction of +2.50 -3.75 x 093 (20/30) and
+1.50 -4.00 x 093 (20/40) with central ultrasound cor-
neal pachymetry of 515 µm OD and 503 µm OS. Note the
Holladay Report OD shows a 54.55-µm anterior corneal
surface elevation above the best fit toric ellipsoid and a
79.20-µm anterior elevation above the posterior surface Figure 6-62. (A) A patient with a history of myopic LASIK.
O riginal prescription was -6.00 -1.25 x 178. The posterior sur-
elevation.
face appears normal using the Visante O mni. (B) In the same
Figure 6-62 demonstrates a patient with a history of patient, the Pentacam suggested several areas were elevated.
myopic LASIK. A 23-year-old White woman presented
following treatment for -6.00 -1.25 x 178 due to residual
refractive error. The Visante revealed a cornea with normal
Figure 6-63 demonstrates a patient 9 years post-LASIK,
posterior surface. In comparison, the Pentacam found sev-
now suffering from post-LASIK ectasia. Her left eye was
eral elevated areas (more than 10 µm) on the posterior cor-
evaluated using the Visante Omni and the Pentacam. Over
neal surface best-fit toric ellipsoid. Over a period of 2 years
the course of 1 year, the Visante presented a consistent,
with repeated exams, the refraction as well as the Visante
slowly progressive ectasia OS. The Pentacam results were
Omni exams were stable.
less suggestive.
 To m o gra p h e r s 135

Figure 6 -64. LASIK flap in patient who presented for enhance-

ment. The flap was measured to be thin, and the patient chose
to undergo PRK rather than relift the flap.

Figure 6 -63. Ectasia post-LASIK.

Figure 6-64 demonstrates the Visante Omni’s ability to

evaluate LASIK flap morphology. Microkeratome blades
create variable thickness flaps. Thin flaps are difficult to
lift and may tear. In rare cases, the flap is too thick, with
insufficient residual stromal bed for further enhancement.
In these cases, PRK may be more appropriate. This patient
was evaluated using the Visante, was found to have a thin
Figure 6 -65. A model for Verisyse lens placement. The
LASIK flap, and underwent PRK enhancement.
-10.00-D implant decreased the endothelial clearance to
1.5 mm, so a -7.00-D implant was used.
Ph akic In t raocu lar Len ses
At the time of this writing, there are only 2 types of
phakic intraocular implants that are approved by the Figure 6-65 illustrates the Visante Omni’s ability to
US FDA: the Verisyse/Artisan and the Visian lens. The model the implantation and check for safety issues prior
Verisyse/Artisan-type implants are anterior chamber, iris to surgery. A 42-year-old woman with the prescription of
fixed lenses. The Visian lens is implanted into the posterior -11.00 -2.75 x 021 (20/30) and central corneal ultrasound
chamber. One of the key issues with the Verisyse/Artisan- pachymetry of 556 was evaluated for the Verisyse implant
type implants is the endothelial clearance of the implant. in conjunction with an IntraLase LASIK for residual refrac-
The current standard is to maintain a minimum of 1.5 mm tion. VeriCalc suggested the endothelial clearance of a
from the edge of the implant to the endothelium. -12 D, 6-mm implant would be 2.00 mm. The Visante, how-
The rise of the crystalline lens above a straight line from ever, showed less than 1.5 mm of clearance. This required
the deepest portion of the angle is a concept that Dr. George a change in planning.
Baikoff described as “crystalline lens rise” (CLR).31 He Although the -10.0 D, 6-mm implant had just 1.5 mm of
noted that a high CLR is a contraindication for the phakic clearance, one has to allow for a slightly decentered implant
IOLs due to increased risk of complications, such as pig- surgically, as the pupil is a dynamic structure. I opted for a
mentary dispersion. With an increasing lens rise of 18 µm -7.0 D, 6-mm implant that had a 1.60-mm endothelial clear-
per year, the age of the patient must be considered. Thus, ance on the model. Postoperatively, the Visante Omni mea-
implantation of Verisyse/Artisan implants for patients with sured a 1.60-mm endothelial clearance. The residual refrac-
more than 300 µm of CLR is contraindicated (personal tive error was corrected using iLASIK (Abbott Medical
communication, G. Baikoff, September 21, 2010). Optics, Mellpitas, CA).
The Visante has a set of models of the Verisyse/Artisan Figure 6-66 demonstrates postoperative analysis for
implants. After the Ant Seg view is acquired, the endothe- Intacs for keratoconus. Intrastromal corneal ring segments,
lial clearance Rainbow Tool allows for a quick visualiza- such as Intacs, were originally devised as a reversible alter-
tion and a guide to center the Verisyse/Artisan implant. native to LASIK surgery. Currently, Intacs are used for the
The implant models 202 or 204 with 5- and 6-mm optics, treatment of ectasia. Visante measures peripheral pachym-
respectively, need to be placed with the flat portion of the etry, guiding the IntraLase channel placement. Once the
implant slightly anterior to the posterior pigmented iris implant is inserted intrastromally, the Visante can confirm
line. Once the model is in place, safety tools can be placed postoperative results. Using the pachymetric maps or the
to measure the endothelial clearance, and vault tools allow Holladay Report in such patients requires special attention
measurement of the distance from the phakic IOL to the to the “surface fit lines.” The Visante follows the normal
anterior crystalline lens surface. contour of the cornea to set the surface fit lines (in blue) to
136 Ch a p t e r 6

Figure 6-67. Evaluation of patients with complex cataracts

such as a posterior polar cataract with adhesion to the poste-
rior capsule can reduce surgical complications.

Figure 6 -66. Evaluation of INTACS placed for keratoconus.

calculate the pachymetric data. A sudden change as seen

with Intacs may not be recognized automatically and may
need manual adjustment of such surfaces.

C at a ract Evalu at ion an d Su rgica l Figure 6-68. A patient with a CrystaLens implant, seen here
Plan n in g where there is no movement of the optics when provoked with
the optometer to induce “accommodation.”
Cataract is the clouding of the crystalline lens. It is
the most commonly performed refractive procedure with
more than 2 million surgeries per year in the United States. however, are all 820- to 840-nm light technology with some
Although cataract is not the main central focus of this sec- limitations in what structures they can observe. With the
tion, it is important to note a few cases. 820- to 840-nm technology, the ability to view intraocular
In postcorneal refractive surgery patients, the ATLAS implants, the healed LASIK flaps, and healed endothelial
corneal topography can provide data for the 0-, 1-, 2-, and keratoplasty interfaces are significantly reduced compared
3-mm zones, which can be used on the postrefractive sur- to the 1310-nm light technology.
gery calculators such as the one on the American Society The other limitation of the 820- to 840-nm technologies
of Cataract and Refractive Surgery Web site (www.ascrs. is the size of the acquired image. For example, the Cirrus
org).32 OCT is, at the time of the writing, limited to 2- by 4-mm
The Visante can also be used to evaluate the images. That image size is far too limiting to fully evaluate
postkeratorefractive surgery patients, as well as those the cornea. The Spectralis (Heidelberg Engineering, Vista,
with complex cataracts, such as a posterior polar cata- CA), although not FDA approved for anterior segment OCT
ract with adhesion to the posterior capsule (Figure 6-67). at the time of this writing, is providing claims of ability to
Additionally, new technologies can be evaluated. A patient provide 16- by 3-mm images. Such images become invari-
with a CrystaLens (Bausch & Lomb) implant is noted here, ably difficult in acquiring as the narrow window requires
and there is no movement of the optics that can be noted exquisite patient cooperation. With the capacity to provide
even when provoked with the optometer to induce “accom- the 16- by 6-mm images, the Visante still remains the most
modation” (Figure 6-68). comprehensive imaging device.
The side viewing screen and the small footprint of the
Fut u re Developm en ts instrument also make it a practical device. The ability to
Visante anterior segment OCT gained FDA approval scan the patient as the family watches and to analyze the
in 2005. Its limitations are accentuated by the rapidly images along with the patient creates a very sophisticated,
growing field of technological advances in the OCT field. elegant real-time patient evaluation and education tool.
The Visante is a time domain technology capable of Visante’s capacity to link with the ATLAS topographer
2048 A-scans in 1 second. Competing technologies such as along with refractive and iridocorneal software capabilities
the spectral domain instruments provide 10 times or more makes this instrument truly unique and the most compre-
scans, resulting in an exquisitely more-detailed image. The hensive anterior segment imaging device.
current commercially available spectral domain OCTs,
 To m o gra p h e r s 137

TH E O CULYZER
Renato Ambrósio Jr, MD, PhD; Diogo L. Caldas, MD; Ana Laura C. Canedo, MD;
Leonardo N. Pimentel, MD; Bruno F. Valbon, MD; and Mirko R. Jankov II, MD, PhD

T
he Oculyzer (Alcon-WaveLight, Fort Worth, TX;
Figure 6-69) is a comprehensive cornea and anterior
segment analyzer based on the same hardware plat-
form as the Pentacam, with customized software capability
for planning customized corneal photoablations based on
corneal front elevation data. The system uses a rotating
Scheimpflug camera for a 3-dimensional scanning of the
cornea and anterior segment, which is aligned to a second
frontal pupil camera.
The Scheimpflug image is based on the principle named
after the Austrian Captain Theodor Scheimpflug (1865-
1911). In 1904, Cap. Scheimpflug described and patented
a geometric rule that describes an innovative orientation of
the plane of focus of a camera, in which the lens plane is not
parallel to the image plane. The technique uses the 3 imagi-
nary planes: the film plane, the lens plane, and the plane of
sharp focus in nonparallel manner, so that they intersect in a
line (Scheimpflug intersection), with a virtual point of inter-
section (Figure 6-70A). Thereby, in a Scheimpflug camera,
the lens is tilted in a way that the resulting lens plane
intersects the film plane and the plane of focus in a form
of a line. This setup extends the depth of focus, providing
more sharpness to points of the image located at different
planes, with the cost of having minor distortion of the image
(Figure 6-70B), which is, in the case of the Pentacam and Figure 6 -69. The O culyzer Instrument (Alcon-
Oculyzer exams, compensated by the software calculations. WaveLight).
In a normal or ordinary camera, the film plane, the lens
plane, and the plane of sharp focus are all parallel to each
other, and, therefore, all 3 planes are perpendicular to the
lens axis (Figure 6-71). This generates a limited depth of
focus when compared to the Scheimpflug image (Figure
6-72).
The Oculyzer is a corneal tomography system, from
the Greek words “to cut or section” (tomos) and “to write”
(graphein), because it provides a 3-dimensional reconstruc-
tion of the corneal picture. It is capable of capturing 25 (in Figure 6-70A. (A) Scheimpflug’s principle. (B) Image distortion
1 second) or 50 (in 2 seconds) Scheimpflug images during by the Scheimpflug system.
1 scan for a 3-dimensional tomographic reconstruction.
It is also capable of acquiring single Scheimpflug images
for documentation (Figure 6-73). Up to 500 points are The images are obtained in slit thickness of 80 µm and
measured per single image during a scan, which generates the range of 14 mm, allowing a complete image through
up to 25,000 true elevation points that are measured and the cornea (Figure 6-74). To avoid the shadows of the nose,
analyzed by the Oculyzer software. The system integrates a the slit images are photographed from the temporal side
second frontal pupil camera, which serves the purposes of at angles from 0 to 180 degrees (Figure 6-75), in order to
controlling fixation and alignment compensation, as well as reconstruct a 3-dimensional model of the anterior segment
the correction of eye movements. The frontal pupil camera of the eye (Figure 6-76). More measurements are per-
is also important for the acquisition mode and enables the formed around the corneal center because this is the point
detection of the size and orientation of the pupil image. of intersection of the images. The range for corneal radius
138 Ch a p t e r 6

Figure 6 -71. O rdinary camera schematics.

(Reprinted with permission of Thomas Auer.)

Figure 6 -72. Comparison of

Scheimpflug and ordinary
images. Note the expanded
depth of field.

Figure 6 -73. Acquiring scan display with

options for exams on the O culyzer software.
 To m o gra p h e r s 13 9

Figure 6 -74. Slit projection on the cornea. (Reprinted with

permission of Thomas Auer.)

Figure 6 -76. Three-dimensional tomographic reconstruction

of the cornea and anterior segment picture from Scheimpflug
images.

measurement is from 3 to 38 mm, and its refractive power

measurement range is from 9 to 99 D. The repeatability of
the thickness measurements is 3 µm, while for the curvature
measurements, it is 0.2 D.
During the examination, the patient should be comfort-
ably positioned at the instrument with adjustment of height
and proper placement on the chin rest and forehead strap.
The patient is asked to blink normally and to open both eyes
and stare at the fixation target. After alignment is close to
perfect, the patient is asked to blink a couple of times, and
B the final alignment is done for the automatic release mode
in order to start the scan. We advise using the 25 single
Scheimpflug images 3-dimensional scan as a routine.
Image quality should be checked right after the exam is
taken, and only images with an acceptable quality should
be considered. The quality specification shows several
parameters separately for the front and back surfaces of the
cornea: the area of coverage of the measurements (analyzed
area), blinking (lost segments), alignment (x, y, and z axis),
and fixation of the eye (Figure 6-77). If the case is consid-
ered for customized topography-guided ablation treatment
(T-CAT), at least 4 (and up to 8) scans should be taken
sequentially. The exam can be taken in the manual release
mode, but such scans are not accepted by T-CAT software
for treatment planning.
The exams are sensitive to any drops previously applied
to the cornea, mainly ocular surface dyes such as fluo-
rescein. These would lead to artifacts of the slit images
and result in false calculations. Pupil dilation is also not
Figure 6 -75. Slit images are photographed at an angle of 0 to advised because it may also affect the centration of T-CAT
180 degrees. (Reprinted with permission of Thomas Auer.)
140 Ch a p t e r 6

Figure 6 -77. Q uality assessment of the scan box. (Reprinted

with permission of Thomas Auer.)

treatments, unless the T-CAT treatment is planned imme- Figure 6 -78. (A) Total internal reflection of light at the iris-
diately afterward, presumably with the same pupil size. corneal angle. (B) Extrapolation of edge detection from the
Dilation of the pupil is necessary only for diagnostic pur- software for angle measurements.
poses in the case with interest in examining IOLs or the
crystalline lens.
The 3-dimensional Scheimpflug image scanning pro-
vides data from the anterior and posterior surfaces of the
cornea, the anterior iris, and the lens (see Figure 6-73).
Because the system uses the visible blue light (custom-
designed 475-nm, UV-free), it is sensitive to any corneal
opacity rendering hyper-reflective images and inaccurate
contour analyses by the Oculyzer software. This also makes
the images optically sensitive to total internal reflection at
the peripheral cornea, thus not enabling visualization of the
anterior chamber angle, as well as anything posterior to the
iris (Figure 6-78A). Although the exam does not provide a
direct visualization of the iris-corneal angle, the sophisti-
cated extrapolation software is capable of providing a value
for angle estimation with high accuracy. Very importantly,
the clinician should re-evaluate the extrapolation calcula-
tion on the Scheimpflug image for ensuring proper clinical
evaluation of the anterior chamber angle (Figure 6-78B).
However, this approach has been successfully used in clini- Figure 6 -79. (Top) Narrow angle in a patient with glaucoma
cal practice to screen narrow angle cases and to document prior to cataract surgery. (Bottom) The same eye post-cataract
surgery. Note the widening of the angle in the absence of the
the improvement after cataract surgery (Figure 6-79).
crystalline lens.
 To m o gra p h e r s 141

Figure 6-80. Diagram scheme for (A) local/tangential

A B (or instantaneous) curvature and (B) axial/sagittal cur-
vature calculations.

Figure 6 -81. Sagittal (superior) and

tangential (inferior) maps from both
eyes of a refractive candidate with
very asymmetric keratoconus (not
unilateral). Scale used was American
Scale with 61 colors, absolute (nor-
mal). Note the tangential maps high-
lighting the cone are surrounded by
a flatter zone. Both corneas were
prolate with average asphericity (Q )
for 30 degrees of -0.23 in O D and
-0.26 in O S. Interestingly, the topo-
metric indices for the 8-mm front
curvature zone did not detect a kera-
toconic pattern in O D and detected
KC stage 2 in O S.

The Oculyzer is a powerful diagnostic tool that also and served as the basis for classic screening parameters
serves therapeutic purposes. The exam provides detailed such as the Rabinowitz-McDonnell,33,34 which are based
corneal tomographic information, along with anterior on the steepness of the cornea, superior-inferior asymmetry,
chamber maps and Scheimpflug images for optical scatter- and between-eyes asymmetries.
ing densitometry of the cornea and crystalline lens. The clinical value for the curvature maps from the pos-
The Oculyzer is a corneal tomography system, and terior corneal surface relies on the more accurate corneal
the Scheimpflug scanning provides considerably larger power measurements, mainly for assessing IOL calcula-
area measurement of the cornea compared with Placido’s tions. There are, however, different approaches for calculat-
topography systems. It is also not sensitive to irregularities ing the relationship between the radii of curvature of the
on the corneal reflex and interference from the tear film. anterior and posterior cornea. The BESST formula,35 the
Having measured the true elevation raw data of the front true net power, the keratometric power deviations, and other
and back corneal surfaces, this exam generates curvature methods36,37 have been described. However, it still remains
maps using a sagittal (axial) and tangential (instantaneous) controversial if K values from Scheimpflug photography do
algorithm. In the sagittal map, the curvature of the cornea is improve accuracy for routine IOL power calculations.38
determined at each measured point at a normal (90-degree) The Oculyzer provides a corneal thickness map with
angle to its surface referenced to the mid-line (axis), while accuracy and repeatability within less than 3 µm.39 Central
tangential maps evaluate the local radius at each measured thickness and thinnest pachymetry data are provided. In
point of data (Figure 6-80). This highlights differences addition, the CTSP and the PTI graphs are calculated from
among the measured points and results in a more “noisy” the pachymetry map (Figure 6-82). Starting from the thin-
color pattern for the tangential maps, which facilitates diag- nest point outward, the CTSP describes the rate of increase
nosis of keratoconus (Figure 6-81). Sagittal curvature maps of corneal thickness by noting the averages of pachymetric
have been more popular for screening refractive candidates values within imaginary annular rings concentric to the
142 Ch a p t e r 6

Figure 6 -82. Pachymetric maps

and CTSP and PTI graphs from
the same case as in Figure 6-81.

thinnest point with increasing diameters in 0.1-mm steps. The diameter of the cornea used for calculating the best fit
The PTI takes the percentage of increase from the thinnest impacts the calculation. We advise setting a fixed zone of
point for the average at each ring.22,40,41 9 mm or preferably 8 mm. Guidelines for interpretation of
A pachymetric progression index (PPI) is also cal- elevation maps have been well-described by Belin.46,47
culated for every 1-degree meridian along the complete The Oculyzer exam also provides an analysis of corneal
360 degrees, starting at the thinnest point. The average of wavefront. Zernike polynomial decomposition of the ante-
all meridians and the one with lower and higher values are rior and posterior surfaces of the cornea provides lower-
presented. In a normal population, the averages and SD of order aberrations and HOAs. The understanding of corneal
PPI of the minimal, maximal, and average of all meridians spherical aberrations may be used for customizing aspheric
(PPI-Ave, PPI-Max, PPI-Min) are 0.58 ± 0.3, 0.85 ± 0.18, IOLs in cataract surgery.48 The impact of the posterior cor-
and 0.13 ± 0.33.22,40 The pachymetric index will be higher neal wavefront is still not well understood. Also, coma and
if the cornea gets thicker more abruptly from the thinnest other HOA terms may be useful for diagnosing keratoconus
point out to the periphery (PTI and CTSP graphs falling and related conditions.
down). The graphs enable the diameter zone where the From the 3-dimensional model created, the Oculyzer
increase in thickness is higher than the mean of a normal allows a detailed assessment of anterior chamber angle,
population. height, volume, and depth,49 becoming a useful tool to
The best parameters developed for diagnosing kerato- complement the study of glaucoma and for planning the
conus are the “relational” thinnest, which is the thinnest implantation of anterior chamber lenses.50-52
pachymetric value divided by the pachymetric progression. The Scheimpflug images also enable the documentation
The “Ambrósio Relational Thinnest” (ART) is thereby cal- and quantification of corneal and lens opacity and other
culated for the minimal (ART-Min), average (ART-Mid), abnormalities. For example, the increased reflectivity in
and maximal (ART-Max). The “ART” concept combines the posterior surface of the cornea may be an indication of
thinnest with the pachymetric distribution, which facilitates cornea guttata (“sign of the camel” in the densitogram).53
the identification of an abnormal cornea despite its thin- Scheimpflug documentation of cataract has also been very
nest value. The ART-Mid and ART-Max have AUROC of useful for supporting and also planning refractive lens sur-
0.98 and 0.99, with cut-offs of 426 and 339 µm, respec- gery indications.54
tively.42,43 For practical reasons, we have opted not to The Oculyzer is a very powerful diagnostic tool, but
perform LASIK if ART-Max is lower than 400 µm. This one of the most important applications of the Oculyzer is
tomographic approach for pachymetric characterization planning customized ablations for laser vision corrections
has been useful for increasing the sensitivity to detect very with the Alcon WaveLight laser system. Although the exam
subtle abnormalities in refractive candidates who may be at provides a 3-dimensional model of the cornea, the link for
higher risk for ectasia.44,45 customization currently only considers front elevation and
The anterior and posterior elevation maps are a gener- corneal apex-pupil centroid relation data. We have used this
alized form of comparing the true corneal shape with a approach for all therapeutic cases as well as for most hyper-
reference shape, usually a (best fit) sphere or toric ellipsoid. opic procedures with very successful outcomes.55
 To m o gra p h e r s 14 3

O PTOVUE—FOURIER D OMAIN ANTERIOR SEGMENT

O PTICAL C OH ERENCE TOMOGRAPH Y
Karolinne Maia Rocha, MD, PhD and J. Bradley Randleman, MD

O
ptical coherence tomography (OCT) is a real-
time, high-resolution, high-speed, noninvasive, and
noncontact optical device that provides 2- and
3-dimensional visualization of anterior segment morphol-
ogy. This technique generates unique imaging capabilities
for the cornea and anterior segment and provides valuable
information for both keratorefractive and lenticular surgery,
including flap architecture analysis, residual stromal bed
thickness determination, keratoconus screening, and IOL
power calculations.

Pr in cip les of Op t ica l

Figure 6-83. Fourier domain O CT CAM-L raster module scan
Coh eren ce Tom ograp h y of a normal cornea. All anatomical layers are visible.

OCT images are acquired by measuring the intensity and

time delay of wave lights diffracted from anatomical struc- Module (CAM) for the Optovue RTVue-100 OCT was
tures passing through an established reference path. Low- approved in 2007, and it provides full-range, artifact-free
coherence light from a super-luminescent diode source is high-pixel density cross-sectional imaging of the anterior
reflected off ocular structures at different axial depths and segment.64,65
is processed to create an axial scan.56 OCT systems provide The higher-resolution anterior segment images obtained
cross-sectional scans of anterior segment structures such as with Fourier domain OCT systems result from a scan rate
the cornea, iris, anterior chamber angle, and lens as well of 26,000 axial scans per second, axial resolution of 5 µm,
as the retina and optic nerve. However, the pigmentation and transverse resolution of 15 µm compared with a scan
on the posterior side of the iris blocks the high-resolution rate up to 2048 axial scans per seconds, axial resolution
visualization of the ciliary body.57-60 of 18 µm, and transverse resolution of 60 µm of the time-
Initial time-domain OCT technologies used detection domain OCT systems. Furthermore, the increased axial
of backscattered light from the tissues by mechanically resolution of spectral-domain OCT allows visualization of
sweeping a mirror in a known reference position. A com- anatomical details that are not observed with time-domain
mercially available time domain OCT (Visante OCT, Carl OCT, including discreet epithelial thickness profiles and
Zeiss Meditec, Jena, Germany) operates at a scan rate of Bowman’s layer, among others (Figure 6-83).64,66
2048 axial scans per seconds, axial resolution of 18 µm, and
transverse resolution of 60 µm. Anterior segment OCT sys-
tem operates at 1310 nm to attenuate the signal from highly Im agin g a n d
scattering structures (eg, sclera),61 contrary to the retinal
imaging system that operates at 830 nm.62 Ex a m in at ion Tech n iq u es
With the advent of a spectral-domain OCT (Fourier
domain) in 2002, the image acquisition speed has consis- The add-on Cornea/Anterior Segment Module lens (CAM
tently increased. The spectral-domain OCT systems use a mode) should be adjusted prior to anterior segment high-
charge-coupled device camera to register the diffraction resolution scan acquisition. Two anterior segment models
grating of wave lights returning from the eye structures. The are available: a wide-angle lens (long lens or CAM-L) and a
intensity and time delay of the wave lights are calculated high-magnification lens (short lens or CAM-S) for anterior
using a mathematical formula, a Fourier transformation. segment imaging. Select either CAM-L or CAM-S mode
The images in Fourier domain OCT systems (Optovue to assess the tab contents. The wide-angle lens (CAM-L,
RTVue-100, Optovue Inc, Fremont, CA) are acquired up 6.0 to 2.0 mm) is the most commonly used because it bal-
to 100 times more quickly when compared to the standard ances depth of field with transverse image capture. The
time-domain OCT devices.63 A Cornea/Anterior Segment currently available anterior segment maps for the Optovue
144 Ch a p t e r 6

Figure 6 -84. Fourier

domain O CT of thin
LASIK flap for myopia,
revealing a uniform flap
thickness.

are pachymetry map, cornea power validation, cross-sec- edge of the LASIK flap is determined by direct visualiza-
tional scan line (CL-line), cross line (CL-cross line), ante- tion of the area of increased reflectivity corresponding
rior chamber angle (CL-angle), raster scan (CL-raster), and with the flap-stroma interface.67-69 The flap tool of the
3-dimensional view of all corneal layers (CL-3D Cornea). high-resolution cross-line scan may be used with the add-on
Users can program a series of scan types for imaging strate- lens to measure flap thickness. The anterior surface of the
gies such as clinical trials, specific corneal diseases, and cornea and the flap-stroma interface are used as anatomical
postoperative follow-up. landmarks for this purpose. In addition to the central flap
Image acquisition is obtained while the patient sits thickness, the overall flap architecture can be evaluated
in front of the scanning device with the forehead and (Figure 6-84). Using the OCT, a recent study demonstrated
chin stabilized by a head rest. External illumination and that some modern mechanical microkeratomes make essen-
scan optimization can be adjusted by clicking “Auto P,” tially planar flaps, similar in architecture to flaps created
which automatically adjusts for differences in polarization. with the femtosecond laser.69
Patients are asked to fixate on the target light source, and Previous studies have suggested actual residual stromal
consecutive scans are performed of the operated eye using bed (RSB) thickness may vary significantly from predicted
the high-resolution scanning mode. values due to variation in flap thicknesses.70-73 Further, a
The “Analyze” menu activates the view and tools for nonuniform meniscus-shaped mechanical microkeratome
manual measurements of the captured images. Editing flap can lead to deeper disruption of collagen lamellae
tools include removal of tracing lines, area of interest in the corneal periphery, resulting in peripheral corneal
selection, zoom, and “Undo/Redo.” Measurement tools steepening and central corneal flattening.74 The OCT can
include distance tool, area tool, point line, text annotation, provide important information, such as the overall RSB,
zoom, OCT noise level, snapshot, and video. Diagnostic and may prove especially useful when determining if there
tools include progression, symmetry, and comparison is sufficient RSB for LASIK retreatments. Some cases of
analysis. ectasia after LASIK may be due to excessive, irregular flap
thickness (Figure 6-85).
Although many topographic indices have been described
Clin ica l Ap p licat ion s for the diagnosis and progression of keratoconus,24,75-78 one
of the most important findings in keratoconus detection is
Refract ive Su rgery focal corneal thinning.34,79,80 The OCT pachymetry maps
have shown high reproducibility in detecting eccentric and
Fourier domain OCT is a useful tool when evaluating asymmetric corneal thinning.81 Li and coworkers have dem-
LASIK flaps, stromal bed thickness prior to enhancement, onstrated higher repeatability in the 0- to 5-mm central area
and preoperative evaluation for keratoectasia. The posterior of the OCT pachymetry map and have suggested keratoconus
 To m o gra p h e r s 14 5

Figure 6-85. Fourier domain O CT

of an ectatic cornea post-LASIK.
Note the highly variable residual
stromal bed thickness, flap thick-
ness, and the meniscus shape of
the flap.

analysis of the central 5-mm diameter zone.81,82 The Optovue

RTVue-CAM pachymetry map is divided into 8 zones and
2-, 5-, and 6-mm annular rings. The map is obtained in
0.32 seconds and comprises 8 meridional scans. The average
of 6-mm paracentral pachymetry measurements of superior
(S), inferior (I), temporal (T), nasal (N), superotemporal (ST),
superonasal (SN), inferotemporal (IT), and inferonasal (IN)
zones are displayed. The central measurement corresponds to
the average pachymetry of the central 2 mm. The pachymetry
map also includes minimum corneal thickness and location;
anterior and posterior corneal power and curvature radius;
and total corneal power (Figure 6-86).
RTvue CAM-L mode (CL-line) scans provide high-
resolution imaging of the corneal architecture and the
epithelial thickness profile of keratoconic eyes. Specific
areas of reactive epithelial hyperplasia and residual
stromal bed can be precisely assessed. Reinstein and co-
authors have demonstrated that the epithelium is thinner
in the apex of the cone and thicker surrounding the apex
of the stromal cone (creating a “doughnut” shape) using Figure 6 -86. O ptovue RTVue-CAM O CT pachymetry map.
the Artemis very high-frequency (VHF) digital ultra-
sound.27,83 The authors also suggested that the epithelial
compensation of areas surrounding the cone can interfere topography measurements of the cornea power (K) after
with the interpretation of anterior corneal surface topog- refractive surgery tend to overestimate the K readings fol-
raphy. An advantage of the Fourier domain OCT is cross- lowing laser corrections for myopia and underestimate for
sectional high-resolution scans that demonstrate specific hyperopia, frequently resulting in significant hyperopia in
areas of epithelial changes overlying the corneal stroma. eyes with previous myopic surgery and myopic error in eyes
The epithelium can be significantly thinner in the apex of with previous hyperopic surgery.84-92
the cone with reactive thickening compensation immedi- Very high-speed OCT can directly measure both anterior
ately inferior to the conical apex in severe keratoconus as and posterior corneal power and should increase the accu-
demonstrated in Figure 6-87. racy of the corneal power calculation. OCT-based formulas
Len s Su rgery for calculating IOL powers are in active clinical trials.
With the OCT, true corneal power can also be assessed
for IOL calculation. Contrary to other devices that only
Advan ced Diagn ost ic Im agin g
analyze the anterior corneal surface or curvature, this The high resolution of the Fourier domain OCT and the
program compares both the anterior and posterior corneal ability to visualize all cornea layers is extremely useful for
surfaces. Errors in measuring true corneal power affect IOL diagnosis, management, and follow-up of complex corneal
calculation outcomes, especially in cataract patients who cases. Figure 6-88 shows a patient who presented with
had previous refractive surgery. Standard keratometry and significant visual symptoms and irregular astigmatism
146 Ch a p t e r 6

Figure 6 -87. Fourier

domain O CT of a normal
cornea (top) and severe
keratoconus (bottom).

Figure 6 -88. Fourier domain O CT of LASIK

interface fluid causing topographic irregular
astigmatism as demonstrated by the Pentacam
(bottom right).

7 months post-LASIK despite a normal clinical appear- Optovue RTVue-CAM can also produce high-resolution
ance. The diagnosis of interface fluid was made with images of intrastromal corneal rings, allowing for accurate
OCT. Figure 6-89A illustrates a case of a patient who depth and position of the intrastromal ring. Superficial
developed progressive irregular astigmatism after phakic implant of corneal ring segments may cause epithelial-stro-
IOL implantation. Fourier domain high-resolution CL-line mal breaks, fibrosis, and extrusion, while very deep implants
scan demonstrated irregular wound healing. After wound can result in anterior chamber perforation. Figure 6-90
revision, the patient’s symptoms resolved. The apposition illustrates a case where the corneal ring segment was placed
of the surgical wound 1-week postwound revision is dem- too deep (bottom) and one example of stromal and epithelial
onstrated in Figure 6-89B. irregularities surround the implant (top).
 To m o gra p h e r s 14 7

Figure 6-89. (A) Fourier domain O CT of

A apposition of the surgical wound after
phakic IO L implant; Pentacam (bot-
tom right) shows irregular astigmatism.
(B) O ne week after wound revision.

Con clu sion

Fourier domain OCT provides accurate assessment of
the anterior segment with applications including corneal
diagnostic procedures, treatment planning, and monitoring
outcomes. Certainly, high-resolution anterior segment OCT
can help anterior segment surgeons to improve their clinical
evaluation and surgical techniques.

Figure 6 -90. Fourier domain O CT demonstrating

depth and position of the corneal intrastromal rings.
148 Ch a p t e r 6

ARTEMIS EPITH ELIAL TH ICKNESS P ROFILE: A SURROGATE

FOR STROMAL SURFACE TOPOGRAPH Y
Dan Z. Reinstein, MD, MA(Cantab), FRCSC, FRCO phth; Timothy J. Archer,
MA(O xon), DipCompSci(Cantab); and Marine Gobbe, MST(O ptom), PhD

T
he corneal epithelium is a highly active, self-renew- Currently, the only validated method capable of measur-
ing layer; a complete turnover occurs in approxi- ing epithelial thickness continuously over a 10-mm area is
mately 5 to 7 days.93 Despite this high turnover VHF digital ultrasound.115,116 The first confirmed measure-
rate, the epithelium must maintain the same thickness ment of the epithelium of the cornea in vivo was reported
profile over time to maintain corneal power and, hence, by our group in 1993 using VHF digital ultrasound, dem-
ocular refraction. The epithelium accounts for an average of onstrating that acoustic interfaces that were being detected
1.03 D of corneal power at the central 2-mm diameter were indeed located spatially at the epithelial surface and
zone.94 The corneal epithelium is known to have the ability at the interface between epithelial cells and the surface
to alter its thickness profile to try and re-establish a smooth, of Bowman’s layer.117 In 1994, we reported the first high-
symmetrical optical surface to compensate for changes to precision 3-dimensional thickness mapping of the corneal
the stromal surface.95,96 For example, central epithelial epithelium for the central 3-mm diameter.118 The area of
thickening after myopic excimer laser ablation has been acquisition was increased to 10 mm using the Artemis VHF
widely reported,97-101 while epithelial breakdown over an digital ultrasound (ArcScan Inc, Morrison, CO) in 2000.115
advanced cone in keratoconus demonstrates that the epithe- The Artemis has since been used to measure the epithe-
lium must have thinned over the cone.26,102-104 Epithelial lial thickness in a wide variety of situations, the majority of
changes such as these will have an impact on the ocular which are summarized in this section, and the effect these
refraction; however, the biggest clinical impact of epithelial changes have on ocular refraction and the implications
changes is to corneal front surface topography. Because for misdiagnosis by corneal front surface topography are
the epithelium compensates for stromal irregularities, the discussed. This section will focus on the diagnosis of early
presence of an irregular stromal surface is either partially keratoconus, epithelial thickness changes after refractive
or totally masked from corneal front surface topography. surgery and Intacs, and establishing a true diagnosis after
Therefore, corneal front surface topography does not always complicated LASIK.
tell the whole story and in some cases does not provide the
necessary information to establish a correct diagnosis.
Ideally, we would like a method of directly measuring Ar tem is Tech n olog y
stromal surface topography. Some have suggested measur-
ing topography after epithelial removal and have shown All the epithelial thickness data set out in this section
an increase in irregularity.105 However, this is not a clini- were obtained using the Artemis 1 VHF digital ultrasound
cally feasible test. Alternatively, the anterior stromal surface arc-scanner. Artemis VHF digital ultrasound is carried
shape can be appreciated by an examination of its proxy— out using an ultrasonic standoff medium, and so provides
the epithelial thickness profile. Based on an understanding the advantages of immersion scanning (eg, the tear film
of the pattern of the normal epithelial thickness profile, any is not incorporated into the corneal or epithelial thickness
localized abnormal epithelial changes can be relied upon as measurement, and there is no physical contact of the trans-
a mirror of a relative localized stromal surface irregularity. ducer with the cornea). The patient sits and positions the
Therefore, the epithelial thickness profile can be thought of chin and forehead into a headrest while placing the eye in a
as a form of stromal surface topography. soft-rimmed eye-cup. Warm sterile normal saline (33°C) is
Different methods have been used to measure corneal filled into the darkened scanning chamber. The patient fix-
epithelial thickness: OCT,106,107 slit-lamp-adapted OCT,108 ates on a narrowly focused aiming beam, which is coaxial
confocal microscopy,109,110 optical pachymetry,111 and with the infrared camera, the corneal vertex, and the center
focusing confocal microscopy.112 However, these techniques of rotation of the scanning system. The technician adjusts
only allow single-point measurement of epithelial thickness, the center of rotation of the system until it is coaxial with
so they cannot be used to investigate the epithelial thick- the corneal vertex. In this manner, the position of each scan
ness profile. Most studies have reported central epithelial plane is maintained about a single point on the cornea, and
thickness measurements, and a few studies also provided corneal mapping is therefore centered on the corneal vertex.
epithelial thickness in the peripheral cornea, but the number A speculum is not required as patients find it comfortable
of points measured in the periphery is limited.111,113,114 to open the eye without blinking in the warm saline bath
 To m o gra p h e r s 14 9

and voluntary elevation of the upper lid produces exposure

of the central 10 mm of cornea in virtually all patients.
Performing a 3-dimensional scan set with the Artemis 1
takes approximately 2 to 3 minutes per eye.
A broadband 50-MHz VHF ultrasound transducer (band-
width approximately 10 to 60 MHz) is swept by a reverse
arc high-precision mechanism to acquire B-scans as arcs
that follow the surface contour of anterior or posterior seg-
ment structures of interest. The Artemis possesses a unique
scan-arc adjustment mechanism to enable maximum per-
pendicularity (and signal-to-noise ratio) to be obtained for
scanning any of the different curvatures within the globe
(ie, cornea, iris plane, and retina). The resolution of the sys-
tem is 21 µm, meaning that distinct echo-peaks can be seen
on the I-scan for a corneal layer thicker than 21 µm.
For 3-dimensional scan sets, the scan sequence consists
of 4 meridional B-scans at 45-degree intervals. Each scan
sweep takes about 0.25 seconds and consists of 128 scan
lines or pulse echo vectors. During the acquisition of each
scan, data are converted (in near real-time) to a B-scan dis-
played on the computer screen.
Ultrasound data are digitized and stored. The digi-
tized ultrasound data are then transformed using patented Figure 6 -91. Artemis VHF digital ultrasound horizontal B-scan
Cornell University (Ithaca, NY) digital signal process- of a cornea 6 months after LASIK with a flap created using the
ing technology I-scan. A linear polar/radial interpolation VisuMax femtosecond laser. (Top) A geometrically corrected
B-scan after readjusting the data to account for the arc-scan
function is used to interpolate between scan meridians
motion of the transducer. (Middle) Digital signal processing is
to produce a Cartesian matrix over a 10-mm diameter in performed on the B-scan signal, and layer thickness measure-
0.1-mm steps. The interpolation function also includes ments are obtained by a computer algorithm on the I-scan,
auto-correlation of back surface curvatures to center and resulting in the red line image of the interfaces. (Bottom)
align the meridional scans. This is our standard scanning Nongeometrically corrected B-scan, which magnifies the
protocol as it provides sufficiently high-density information image to make the interfaces easier to identify. The nongeo-
in the central cornea with lower density of information in metrically corrected B-scan is the initial output and appears
the periphery where it is less needed. relatively flat because the arc-scan motion of the transducer
Interfaces between tissues are detected at the location of keeps it perpendicular to the cornea.
the maximum change in acoustic impedance (the product
of the density and the speed of sound). It was first demon-
strated in 1993 that acoustic interfaces being detected in • A nongeometrically corrected B-scan, which magni-
the cornea were located spatially at the epithelial surface fies the image to make the interfaces easier to iden-
and at the interface between epithelial cells and the anterior tify. The nongeometrically corrected B-scan is the
surface of Bowman’s layer.117 The posterior boundary of the initial output and appears relatively flat because the
stroma with VHF digital ultrasound is located at the inter- arc-scan motion of the transducer keeps it perpen-
face between the endothelium and the aqueous as this is the dicular to the cornea.
location of the maximum change in acoustic impedance. We have previously published the 3-dimensional repeat-
Therefore, stromal thickness with VHF digital ultrasound is ability of thickness measurements using the Artemis VHF
measured from the front surface of Bowman’s layer to the digital arc-scanner and demonstrated very high repeat-
back surface of the endothelium. ability of 3-dimensional wide-area pachymetry for the epi-
Figure 6-91 shows an example B-scan of a patient thelium, stroma, cornea, flap, and residual stromal bed.116
6 months after LASIK, including the following: The repeatability at the corneal vertex was 0.58 µm for
• A geometrically corrected B-scan after readjusting epithelium, 1.78 µm for stroma, 1.68 µm for cornea, 1.68 µm
the data to account for the arc-scan motion of the for flap, and 2.27 µm for residual stromal bed. The region-
transducer repeatability within the central 1-mm radius was 1.01 µm
• A redline image that represents the result after digital for epithelium, 3.44 µm for stroma, 3.35 µm for cornea,
signal processing by a computer algorithm on the 2.81 µm for flap, and 3.97 µm for residual stromal bed.
I-scan
150 Ch a p t e r 6

Therefore, the Artemis is capable of reliably detecting

changes in epithelial thickness of less than 1 µm. This com-
pares favorably with the other methods of single-point epi-
thelial thickness measurement; the repeatability has been
reported to be 3.24 µm using SL-OCT,108 3.33 µm using
Humphrey Zeiss OCT,112 and 3.41 µm using the OptoVue
OCT (unpublished data, DZ Reinstein). Therefore, these
other instruments are only capable of reliably measuring a
change of more than 6 µm.

Ep it h elia l Th ick n ess

Profile in Nor m a l Eyes
Before we look at changes to the epithelium, it is
important to obtain a baseline measurement in a normal
population so that epithelial changes can be confirmed and
quantified. We have previously published a study that char-
acterized the in vivo epithelial thickness profile in a popu-
lation of normal eyes with no ocular pathology other than
refractive error.119 In this study, the epithelial thickness
profile was measured across the central 10-mm diameter of
the cornea for 110 normal eyes of 56 patients, and these data
were averaged. Epithelial thickness values for left eyes were
reflected in the vertical axis and were superimposed onto
the right eye values so that nasal/temporal characteristics
could be combined.
The average epithelial thickness map (Figure 6-92)
revealed that the epithelium was not a layer of homogeneous
thickness as had previously been thought, but followed
a very distinct pattern; on average, the epithelium was
5.7-µm thicker inferiorly than superiorly and 1.2-µm thicker
temporally than nasally. The pattern of thicker epithelium
inferiorly than superiorly and thicker epithelium nasally Figure 6 -92. Artemis nongeometrically cor-
rected B-scan and epithelial thickness profile
than temporally was consistent across a majority of eyes
for 1) a population of 110 normal untreated
in the population sampled. The average central epithelial eyes, 2) a population of 54 keratoconic eyes,
thickness was 53.4 µm, and the standard deviation was only 3) a population of 24 eyes after myopic
4.6 µm (range: 43.5 to 63.6 µm). This indicated that there LASIK, 4) a population of 14 eyes after radial
was little variation in central epithelial thickness in the keratotomy, 5) a population of 65 eyes after
population. The thinnest epithelial point within the central hyperopic LASIK, 6) a case example of an eye
5 mm of the cornea was displaced 0.33 mm (±1.08) tempo- after 2 weeks of wear of an orthokeratology
rally and 0.90 mm (±0.96) superiorly with reference to the lens, and 7) a case of post-LASIK ectasia.
corneal vertex.
Figure 6-92 also shows an example of a B-scan of a
tarsus provides a template for the outer shape of the epi-
normal cornea. The epithelium appears regular in thick-
thelial surface. During blinking, which occurs on average
ness. The finding that the epithelium was thinner superiorly,
between 300 to 1500 times per hour,120 the vertical traverse
rather than being a uniform thickness throughout, was a
of the upper lid is much greater than that of the lower lid.
surprising result; however, we have recently suggested that
Doane121 studied the dynamics of eyelid anatomy during
eyelid mechanics and blinking might be responsible for
blinking and found that during a blink, the descent of the
the nonuniform epithelial thickness profile seen in normal
upper eyelid reaches its maximum speed at about the time it
corneas. In 1994, we originally suggested that eyelid blink-
crosses the visual axis. As a consequence, the eyelid might
ing and friction onto the corneal surface may regulate the
effectively be chafing the surface epithelium during blink-
epithelial thickness profile.117 We postulated that the eyelid
ing, with greater forces applied on the superior cornea than
might effectively be chafing the surface epithelium during
on the inferior cornea. This could explain why the epithe-
blinking and that the posterior surface of the semi-rigid
lium was found to be thinner superiorly.
 To m o gra p h e r s 151

Ep it h elia l Th ick n ess correlation between the thickness of the thinnest epithelium
and the difference in thickness between the thinnest and the
Profile in Kerato con ic Eyes thickest epithelium. This indicated that, as the epithelium
thinned, there was an increase in the irregularity of the
In keratoconus, the epithelium is known to thin in the epithelial thickness profile. These factors demonstrate that
area overlying the cone, and in advanced keratoconus, there the degree of potential epithelial compensation is dependent
may be excessive epithelial thinning leading to a breakdown on the severity of the keratoconus and that the epithelium
in the epithelium. Epithelial thinning over the cone has changes with the progression of the disease.
been demonstrated using histopathologic analysis of kera- Figure 6-92 also shows a B-scan for a keratoconic cor-
toconic corneas,102 while Haque and colleagues103 found nea, which demonstrates the lack of homogeneity in epi-
that central epithelial thickness was thinner in keratoconus thelial thickness as well as central corneal thinning. There
than in normal corneas using OCT. Transmission electron is epithelial thinning over the cone and relative epithelial
microscopy (TEM) has demonstrated that the epithelium thickening adjacent to the stromal surface cone.
was irregular in thickness in keratoconus.104 As with the epithelial thickness profile seen in a normal
In 2009, we published a study that characterized the population, the reason for the presence of an epithelial
in vivo epithelial thickness profile in a population of doughnut pattern in keratoconus could also be explained
54 keratoconic eyes.26 The subjects included for study had by eyelid mechanics and blinking. In keratoconus, because
previously been diagnosed with keratoconus, and the diag- the cone is protruding, the apex would be the first point
nosis was confirmed by clinical signs such as microscopic of contact with the eyelid, resulting in increased chafing
signs at the slit-lamp, corneal topographic changes, high and therefore thinning of the epithelium at the apex of the
refractive astigmatism, reduced best-corrected visual acu- cone. As the cone becomes more severe and the protrusion
ity and contrast sensitivity, and significant level of HOAs, becomes more pronounced, it makes sense that the epithe-
in particular vertical coma. The epithelial thickness profile lium at the apex of the cone becomes thinner. This theory
was measured across the central 10-mm diameter of the is supported by the finding that the thinnest epithelium
cornea for 54 keratoconic eyes of 30 patients, and the data in the central zone of the epithelial doughnut pattern was
were averaged. Epithelial thickness values for left eyes were thinner for eyes with steeper keratometry (more advanced
reflected in the vertical axis and were superimposed onto keratoconus). As the cone becomes more severe, we also
the right eye values so that nasal/temporal characteristics found that the surrounding annulus of epithelium in the epi-
could be combined. thelial doughnut pattern is thicker. The presence of annular
The average epithelial thickness profile in keratoconus epithelial thickening suggests that the eyelid is applying less
(see Figure 6-92) revealed that the epithelium was signifi- force during blinking around the sides of the cone, allowing
cantly more irregular in thickness compared to the normal the epithelium to grow up to the inner surface template of
population. The epithelium was thinnest at the apex of the the eyelid; in a sense, the eyelid inner surface is tented over
cone, and this thin epithelial zone was surrounded by an the apex of a bulging cone, producing lateral valleys that
annulus of thickened epithelium. The location of the thin- are then filled with thickened epithelium. The epithelial
nest epithelium within the central 5 mm of the cornea was reaction can be thought of in terms of draping a cloth over
displaced 0.48 mm (±0.66 mm) temporally and 0.32 mm an irregular surface with the edges pulled taut. A taut cloth
(±0.67 mm) inferiorly with reference to the corneal vertex. would not lie evenly across the surface; rather, it would be
The mean epithelial thickness for all eyes was 45.7 ± 5.9 µm in contact with the peaks of the irregular surface and tent-
(range: 33.1 to 56.3 µm) at the corneal vertex, 38.2 ± 5.8 µm shaped above the surface on either side of the peaks before
at the thinnest point (range: 29.6 to 52.4 µm), and 66.8 ± gradually returning to the surface in areas without irregu-
7.2 µm (range: 54.1 to 94.4 µm) at the thickest location. The larities. The anterior surface of the cornea can be seen to be
epithelial thickness was outside the range observed in the similar to such a tent shape in the B-scan in Figure 6-92.
normal population in both the thinnest and thickest regions,
demonstrating the extent of the change in epithelial thick- Diagn osin g Early Kerato con u s Usin g
ness in keratoconus.
While all eyes exhibited the same epithelial doughnut Epit h elial Th ickn ess Profiles
pattern, characterized by a localized central zone of thin- As we have just shown in the previous 2 sections, there
ning surrounded by an annulus of thick epithelium, the is a distinct difference in the epithelial thickness profile in
thickness values of the thinnest point and the thickest point keratoconus compared to that of normal corneas. We have
as well as the difference in thickness between the thinnest also shown that the epithelial thickness profile changes with
and the thickest epithelium varied greatly between eyes. the progression of the disease. Therefore, we can assume
There was a significant correlation between the thinnest that the epithelial doughnut pattern would be observed in
epithelium and the steepest keratometry (D), indicating that early keratoconus, where the degree of epithelial thick-
as the cornea became steeper, the epithelial thickness mini- ness changes would be expected to be relatively small, and
mum became thinner. In addition, there was a significant epithelial thickness profiles can be used as a screening tool
152 Ch a p t e r 6

for keratoconus. We have recently proposed this method

of using epithelial thickness profile maps as a new adjunc-
tive diagnostic tool, with the aim to provide both higher
specificity and higher sensitivity to diagnose early cases of
keratoconus when topography is equivocal.27
In keratoconus, the cone is often represented by a high,
eccentric apex on both anterior and posterior elevation
BFS.122,123 Our hypothesis is that front and back corneal
surfaces are yoked, meaning that any back surface ectatic
change will therefore be accompanied by a front stromal
surface ectatic change. The epithelial doughnut pattern that
we have observed in keratoconus will act to minimize the
extent of the cone on anterior elevation BFS. Therefore,
epithelial changes could potentially fully compensate the
stromal surface irregularity for small amounts of stromal
front surface ectasia and render a completely normal ante-
rior elevation BFS, while the ectasia would still be apparent
on the posterior surface elevation BFS.
Therefore, mild keratoconus might be picked up earlier
on posterior surface elevation BFS than anterior surface
elevation BFS; something that early adopters of Orbscan
became convinced of in the mid- to late 1990s and was first
presented by our group in 2003.124 However, this does not
necessarily mean that posterior elevation BFS changes are
the first anatomical morphological changes in keratoconus,
just that these are the first detectable changes. Interestingly,
it was not until 2008 that a paper suggesting this concept
actually came to press; Schlegel and colleagues125 sug-
gested that early posterior surface changes could be occur-
ring in patients with no symptoms and with anterior surface
modifications that are so small that they would not be
detected. The authors concurred with our previously pre-
sented hypothesis that the corneal epithelium could remodel
itself and mask or minimize some early anterior surface
topographical changes. However, not all posterior eleva-
tion BFS changes will be due to keratoconus, which is why
there is a need for a diagnostic tool to confirm or exclude a
diagnosis of keratoconus in eyes with an eccentric posterior
elevation BFS. The key to detecting keratoconus by geomet- Figure 6 -93. Central keratometry, ATLAS front corneal surface
ric anatomical changes appears to be the ability to examine topography, and PathFinder corneal analysis (indicated below
the topography map), O rbscan anterior and posterior elevation
the anterior and posterior stromal surfaces.
BFS, and Artemis epithelial thickness profile for 3 example
eyes where the diagnosis of keratoconus might be misleading
from topography. The final diagnosis based on the epithelial
Ca se Ex a m p les thickness profile is shown at the bottom of each example.
O D = right eye, O S = left eye.
Figure 6-93 shows 3 selected examples where epithelial
thickness profiles helped to interpret and diagnose anterior
and posterior elevation BFS abnormalities. In each case, the 45.25/43.25 D x 76, and PathFinder corneal analysis classi-
epithelial thickness profile appears to be able to differenti- fied the topography as normal. Orbscan II posterior eleva-
ate cases where the diagnosis of keratoconus is uncertain tion BFS showed that the posterior elevation BFS apex was
from normal.27 decentered inferotemporally. Corneal pachymetry mini-
Case 1 (OS) represents a 25-year-old man with a mum by handheld ultrasound was 479 µm. Contrast sensi-
manifest refraction of -1.00 -0.50 x 150 and a best spec- tivity was slightly below the normal range measured using
tacle-corrected visual acuity of 20/16. ATLAS corneal the CSV-1000 (Vector Vision Inc, Greenville, OH). There
topography demonstrated inferior steepening, which would was -0.30 µm (OSA notation) of vertical coma on WASCA
traditionally indicate keratoconus. The keratometry was aberrometry. Corneal hysteresis was 7.5 mm Hg, and
 To m o gra p h e r s 15 3

corneal resistance factor was 7.1 mm Hg, which are low, but as well as the level of vertical coma (coma = 0.066 µm).
these could be affected by the low corneal thickness. The Corneal hysteresis was 8.9 mm Hg, and corneal resistance
combination of inferior steepening, an eccentric posterior factor was 8.8 mm Hg, both within normal limits. In this
elevation BFS apex, and thin cornea raised the suspicion case, only the slightly eccentric posterior elevation BFS
of keratoconus, although there was no suggestion of kera- apex and the low-normal corneal thickness were suspicious
toconus by refraction, keratometry, or PathFinder corneal for keratoconus, while all other screening methods gave no
analysis. Artemis epithelial thickness profile showed a pat- indication of keratoconus. However, the epithelial thickness
tern typical of keratoconus with an epithelial doughnut profile showed an epithelial doughnut pattern characterized
shape characterized by a localized zone of epithelial thin- by localized epithelial thinning surrounded by an annulus
ning displaced inferotemporally over the eccentric posterior of thick epithelium, coincident with the eccentric posterior
elevation BFS apex, surrounded by an annulus of thick elevation BFS apex. Epithelial thinning with surrounding
epithelium. The coincidence of an area of epithelial thin- annular thickening over the eccentric posterior elevation
ning with the apex of the posterior elevation BFS, as well as BFS apex indicated the presence of probable subsurface
the increased irregularity of the epithelium, confirmed the keratoconus. In this case, it seems that the epithelium had
diagnosis of early keratoconus. fully compensated for the stromal surface irregularity so
Case 2 (OD) represents a 31-year-old woman with a that the anterior surface topography of the cornea appeared
manifest refraction of -2.25 -0.50 x 88 and a best spectacle- perfectly regular. Given the regularity of the front surface
corrected visual acuity of 20/16. Atlas corneal topography topography and the normality of nearly all other screening
demonstrated a very similar pattern to case 1 of inferior parameters, it is feasible that this patient could have been
steepening, therefore suggesting that the eye could also be deemed suitable for corneal refractive surgery and subse-
keratoconic. The keratometry was 44.12/44.75 D x 148, and quently developed ectasia. As we were able to also consider
PathFinder corneal analysis classified the topography as sus- the epithelial thickness profile, this patient was rejected for
pect subclinical keratoconus. Orbscan II posterior elevation corneal refractive surgery. This kind of case may explain
BFS showed that the apex was slightly decentered nasally. some reported cases of ectasia “without a cause.”27
Corneal pachymetry minimum by handheld ultrasound Knowledge of the differences in epithelial thickness
was 538 µm. Contrast sensitivity was in the normal range. profile between the normal population and the keratoconic
There was 0.32 µm (OSA notation) of vertical coma on population allowed us to identify several features of the
WASCA aberrometry. Corneal hysteresis was 10.1 mm Hg, epithelial thickness profile that might help to discrimi-
and corneal resistance factor was 9.8 mm Hg, which are nate between normal eyes and keratoconus-suspect eyes.
well within normal range. The combination of inferior Mapping of the epithelial thickness profile may increase
steepening, against-the-rule astigmatism, and high degree sensitivity and specificity of screening for keratoconus
of vertical coma raised the suspicion of keratoconus, which compared to current conventional corneal topographic
was also noted by PathFinder corneal analysis. Artemis screening alone and may be useful in clinical practice in
epithelial thickness profile showed a typical normal pattern 2 very important ways.
with thicker epithelium inferiorly and thinner epithelium First, epithelial thickness mapping can exclude the
superiorly. Thicker epithelium inferiorly over the suspected appropriate patients by detecting keratoconus earlier or
cone (inferior steepening on topography) was inconsistent confirming keratoconus in cases where topographic chang-
with an underlying stromal surface cone, and therefore es may be clinically judged as being “within normal limits.”
the diagnosis of keratoconus was excluded. This patient Epithelial information allows an earlier diagnosis of kera-
would have been rejected for surgery given a documented toconus as epithelial changes will occur before changes on
PathFinder corneal analysis warning of suspect subclinical the front surface of the cornea become apparent. Epithelial
keratoconus, but given the epithelial thickness profile, this thinning coincident with an eccentric posterior elevation
patient was deemed a suitable candidate for LASIK. BFS apex, and in particular if surrounded by an annu-
The anterior corneal topography in case 3 (OD) bears no lus of thicker epithelium, is consistent with keratoconus.
features related to keratoconus. The patient is a 35-year-old Excluding early keratoconic patients from laser refractive
woman with a manifest refraction of -4.25 -0.50 x 4 and a surgery will reduce and potentially eliminate the risk of
best spectacle-corrected visual acuity of 20/16. The refrac- iatrogenic ectasia of this etiology and therefore increase the
tion had been stable for at least 10 years, and the contrast safety of laser refractive surgery. From our data, 136 eyes
sensitivity was within normal limits. The keratometry was out of 1532 consecutive myopic eyes screened for refrac-
43.62/42.62 D x 74, and PathFinder analysis classified the tive surgery demonstrated abnormal topography suspect
topography as normal. Orbscan II posterior elevation BFS of keratoconus. All 136 eyes were screened with Artemis
showed that the apex was slightly decentered inferotempo- VHF digital ultrasound arc-scanning, and individual epithe-
rally, but the anterior elevation BFS apex was well-centered. lial thickness profiles were mapped. Out of 136 eyes with
Corneal pachymetry minimum by handheld ultrasound suspect keratoconus, only 22 eyes (16%) were confirmed as
was 484 µm. Pentacam keratoconus screening indices were keratoconic.126
normal. WASCA ocular HOAs were low (RMS = 0.19 µm)
154 Ch a p t e r 6

Second, epithelial thickness profiles may be useful -3.40±1.63 D (range: -0.75 to -6.00 D). The average post-
in excluding a diagnosis of keratoconus despite suspect LASIK epithelial pachymetric map in the population (see
topography. Epithelial thickening over an area of topo- Figure 6-92) showed that the epithelium was thicker in a
graphic steepening implies that the steepening is not due central, circular region approximately 6 mm in diameter.
to an underlying ectatic surface. In such cases, excluding The central epithelial thickening can also be seen on the
keratoconus using epithelial thickness profiles appears B-scan. The central zone of epithelial thickening was sur-
to allow patients who otherwise would have been denied rounded by an annulus of epithelial thinning between the
treatment due to suspect topography to be deemed suitable 6- and 8-mm diameter. The mean thickest epithelial thick-
for surgery. From our data, out of the 136 eyes with suspect ness for all eyes was 64.3±2.7 µm (range: 58.7 to 69.1 µm).
keratoconus screened with Artemis VHF digital ultrasound The thickest epithelial point was displaced 0.45±0.68 mm
arc-scanning, 114 eyes (84%) showed normal epithelial temporally and 0.66±0.70 mm inferiorly on average with
thickness profile and were diagnosed as nonkeratoconic reference to the corneal vertex. Measurement of epithe-
and were deemed suitable for corneal refractive surgery. lial thickness profile provided evidence that the epithelium
One year post-LASIK, follow-up data126 and preliminary appears to behave as to reverse the change in stromal cur-
2-year follow-up data127 on these demonstrated equal sta- vature and return the cornea to its preoperative curvature
bility and refractive outcomes as matched control eyes. by remodeling itself to mirror the volume of stromal tissue
Randleman et al, in their paper assessing risk factors for removed.
ectasia, reported that ectasia might still occur after uncom- In a separate study, we investigated changes in epithe-
plicated surgery in appropriately screened candidates.128 lial thickness profile as a function of myopia treated in a
Mapping of epithelial thickness profiles might provide an group of 37 eyes after myopic LASIK. The eyes were split
explanation for these cases; it could be that a stromal sur- into low myopia (-1.00 D to -4.00 D), moderate myopia
face cone was masked by epithelial compensation and the (-4.25 to -6.00 D), and high myopia (-6.25 to -13.50 D).132
front surface topography appeared normal. We demonstrated that the amount of epithelial thickening
was dependent on the level of myopia. There was more
Ep it h elial Ch an ges After Refract ive epithelial thickening for high myopia than for low myopia
Su rgery with maximum thickening centrally and progressively less
thickening centrifugally in the low myopia group, whereas
The importance of epithelial changes in corneal refrac- a more homogenous thickening in the moderate myopia
tive surgery has probably been underestimated. Significant and high myopia groups within the 5-mm diameter was
changes in epithelial thickness profiles in both PRK97,98 noted. Knowledge of the epithelial thickness profile change
and LASIK99-101 have been demonstrated and implicated between the pre- and postoperative epithelium is of interest
in regression as well as the inaccuracy of topographi- as it will affect the refractive power of the epithelium. The
cally guided excimer laser ablation.129 The curvature of increased epithelial thickening at the corneal vertex could
Bowman’s layer in the center of the normal cornea is on partly explain the myopic shift described in the early period
average greater than that of the epithelial surface.130 As the after myopic LASIK. Interestingly, we demonstrated that
refractive index of epithelium and stroma are sufficiently the myopic refractive shift due to epithelial profile changes
different (1.401 versus 1.377),131 the epithelial-stromal was more significant for low than high myopic ablations
interface constitutes an important refractive interface with- because the gradient of epithelial thickening from center to
in the cornea, with a mean power contribution estimated at periphery was steeper for low myopia than moderate and
approximately -3.60 D.130 Thus, unpredicted changes in the high myopia, although the epithelium thickened progres-
epithelial lenticule after surgery will result in unplanned sively with increasing central ablation depth.
refractive shifts. This is one of the reasons why current These studies show that consideration of epithelial thick-
ablation depths and profiles (“nomograms”) differ from ness changes might help improve the accuracy of myopic
theoretical ablation profiles—they incorporate the average LASIK as epithelial healing and thickness changes play a
change of epithelial power for a given level of stromal sur- role in the final refractive result.
face flattening (level of myopia treated). Thus, the under-
standing of epithelial dynamics and their patterns begins to Lo n git u d in a l Ep it h elia l Th ick n ess
unfold,100,101 and these factors may potentially be used to
improve the accuracy of corneal refractive outcomes.
Ch an ges After Myopic LASIK
The epithelial thickness profile was measured in a popu-
Ep it h elial Th ickn ess Profile After lation of 24 eyes at 1 day, 1 month, 3 months, 6 months,
Myopic LASIK and 12 months after myopic LASIK. The mean preopera-
tive spherical equivalent was -3.40±1.63 D (range: -0.75 to
We have characterized the epithelial thickness pro- -6.00 D). The average epithelial thickness map at each
file in a population of 24 eyes 12 months after myopic time point (Figure 6-94) demonstrated a large change in
LASIK. The mean preoperative spherical equivalent was epithelial thickness overnight, characterized by a central
 To m o gra p h e r s 15 5

Figure 6 -94. Top row: Maps of the average epithelial thickness before and 1 day, 1 month, 3 months, 6 months, and 12 months
after myopic LASIK. Middle row: Maps of the change in epithelial thickness between each pair of time points. Bottom row: Maps
of the t-test p -value comparing the change between each pair of time points; the areas in blue indicate a statistically significant
change (p <0.05). The difference maps show that there was a change overnight with central epithelial thickening and peripheral
epithelial thinning, followed by epithelial thickening until 3 months, after which the epithelial thickness stabilized.

zone of thickening of approximately 1 to 2 µm within the equivalent was -5.45±3.0 D (range: -2.00 to -9.50 D),
5-mm diameter, surrounded by an annulus of epithelial although preoperative data were not available for 4 eyes.
thinning of approximately 4 to 6 µm between the 5- and The average epithelial pachymetric map in the popula-
8-mm diameters. This demonstrates the speed at which the tion (see Figure 6-92) showed that the epithelium became
epithelium responds to stromal surface changes. Between thicker in a central, circular region approximately 4 mm in
1 day and 1 month, the epithelium had thickened across diameter. The central zone of epithelial thickening was sur-
the 7-mm diameter zone by up to 5 µm, with more pro- rounded by an annulus of epithelial thinning. Interestingly,
nounced thickening within the central 4 mm. Between the epithelial thickness profile was very similar to that
1 and 3 months, the epithelium had continued to thicken of eyes after myopic LASIK ablation, although the surgi-
in the central 7-mm diameter zone by approximately an cal techniques were completely different; the epithelium
additional 1 µm. The t-test p-value maps shown in Figure responded to changes in curvature alone after RK without
6-94 highlight the regions where the changes were statisti- tissue removal. The exaggerated change in curvature after
cally significant between each pair of time points. There RK can be visualized in the B-scan shown in Figure 6-92,
were no further changes in epithelial thickness at the 6- and where the cornea appears like a tabletop. The central zone
12-month visits, supported by the t-test difference map that of epithelial thickening was slightly smaller for eyes after
showed no regions of statistically significant difference RK than for eyes after LASIK, indicating that the size of
between 3 and 6 months and between 6 and 12 months. the optical zone was smaller for eyes after RK than for eyes
This demonstrated that the epithelium had become stable after LASIK.
3 months after LASIK. These epithelial changes partially
explain the regression seen after myopic LASIK in the first Ep it h elial Th ickn ess Profile After
3 months and agree with the common finding that refractive Hyp eropic LASIK
stability is attained after 3 months.133
We have previously published a study that character-
Ep it h elial Th ickn ess Ch an ges After ized the epithelial thickness profile in a population of
65 eyes at least 3 months after hyperopic LASIK using a
Rad ial Keratotom y 7-mm ablation zone with the MEL80 excimer laser (Carl
The epithelial thickness profile was measured in a popu- Zeiss Meditec).134 The average epithelial pachymetric
lation of 14 eyes after radial keratotomy for myopia. The map in the population (see Figure 6-92) showed that the
mean time of measurement after surgery was 17±6 years epithelium was thinner in a central, circular region approxi-
(range: 1.6 to 26 years). The mean preoperative spherical mately 4 mm in diameter with the thinnest epithelium in the
156 Ch a p t e r 6

superior half of this region. The mean thinnest epithe- retreatments might be performed without risk of apical
lial thickness for all eyes was 39.7±5.6 µm (range: 26.9 to syndrome while also allowing some patients to have retreat-
52.7 µm). The thinnest epithelial point was displaced ment who would otherwise have been rejected for further
-0.19±0.64 mm temporally and 0.93±0.82 mm superi- surgery due to high keratometry postoperatively.
orly on average with reference to the corneal vertex. The
central, thin epithelium was surrounded by an annulus of Ort h okeratology
thicker epithelium, with the thickest epithelium temporally.
We previously described epithelial, stromal, and corneal
The mean thickest epithelial thickness for all eyes was
thickness changes in a patient before and during ortho-
89.3±14.6 µm (range: 63.6 to 124.6 µm). The thickest epi-
keratology treatment and were able to demonstrate that
thelial point was displaced -2.49±1.04 mm temporally and
refractive changes were mainly due to epithelial thickness
-0.73±1.42 mm inferiorly on average with reference to the
changes.136 The epithelial thickness changes appeared to
corneal vertex. The mean radial distance of the location
reveal the pattern of compression produced by orthokeratol-
of the thickest epithelium from the corneal vertex was
ogy lenses. In both eyes, there was a zone of central epithe-
3.12±0.25 mm (range: 2.3 to 3.6 mm). The diameter of the
lial thinning within the central 1.5-mm radius, surrounded
annular zone of maximum epithelial thickening was 6.8 mm
by a midperipheral annulus of epithelial thickening at the
in the horizontal meridian and 6.7 mm in the vertical merid-
2-mm radius. The Artemis epithelial thickness profile and
ian, which matches the programmed laser ablation optical
B-scan of the right eye is shown in Figure 6-92. There also
zone diameter of 7 mm where the maximum ablation depth
appeared to be a degree of central stromal thickening and
was at the 7-mm diameter. This demonstrates that the
midperipheral stromal thinning, although the magnitude of
epithelium was compensating for the paracentral stromal
the stromal changes was less than one-third of the magni-
tissue removal due to the hyperopic ablation. Knowledge
tude of the epithelial changes. Using epithelial thickness
of the epithelial thickness profile after hyperopic LASIK
mapping, we were able to differentiate epithelial thickness
may partly explain hyperopic regression and could be used
changes from stromal thickness changes to understand the
to improve the accuracy of hyperopic ablation, particularly
mechanism of refractive changes and demonstrate that the
with respect to induced cylinder.
majority of the refractive correction in orthokeratology is
In this study, we also suggested that central epithelial
caused by changes to the epithelial thickness profile, which
thickness may be a more useful indicator than keratometry
explains the temporary nature of the effect.
to determine whether further treatment may be attempted.
The epithelial thickness changes observed add more
There was a statistically significant and strong correlation
weight to the theory that the epithelium remodels to fit
between the attempted spherical equivalent refraction and
the template in front of the cornea. In orthokeratology,
both the thinnest (R2 = 0.350, p<0.001) and thickest epithe-
the template normally provided by the posterior surface
lium (R2 = 0.765, p<0.001). This indicated that the mini-
of the semi-rigid tarsus is replaced by a contact lens that
mum epithelial thickness was thinner and the maximum
is designed to fit tightly to the center of the cornea and
epithelial thickness was thicker in eyes after higher refrac-
loosely paracentrally. Therefore, the epithelium is chafed
tive ablations. It is currently assumed that hyperopic LASIK
and squashed by the lens centrally while the epithelium is
should be limited according to postoperative curvature as
free to thicken paracentrally where the lens is not so tightly
too much steepening can result in epitheliopathy or apical
fitted.
syndrome; it is generally accepted that the postoperative
curvature should not exceed 49.00 to 50.00 D.135 However,
the results from the present study suggest that epithelial Ectasia Diagn osis
thickness might be a better indicator. Whereas the thinnest Ectasia is one of the most devastating potential conse-
epithelium is correlated with the postoperative curvature quences of LASIK, and it behooves us to prevent it from
(R2 = 0.210, p<0.001), the postoperative curvature alone happening in every possible way. In the first section, we
can be misleading. In an example case from the study,134 described the use of epithelial thickness mapping for the
the maximum simulated keratometry of 50.80 D would most early diagnosis of keratoconus as an aid in the determi-
likely prevent the surgeon from treating further hyperopia; nation of candidacy for LASIK with the aim of avoiding
however, the central epithelial thickness of 41.7 µm would ectasia given that the majority of post-LASIK ectasia has
suggest that the cornea could be steepened further without been caused by undetected keratoconus.128,137-140 Epithelial
resulting in epithelial breakdown. On the other hand, anoth- thickness maps can also be used to confirm and monitor the
er case from the present study demonstrates that the epithe- diagnosis of post-LASIK ectasia. The example in Figure
lial thickness can be thin (33.7 µm) although the cornea was 6-92 shows the epithelial thickness profile of an eye with
still relatively flat postoperatively (46.40 D). The curvature post-LASIK ectasia. The epithelial thickness profile dem-
limit would allow further hyperopic ablation, whereas the onstrates similar features to that of a keratoconic eye, char-
thin, central epithelium would indicate that further steepen- acterized by localized central thinning coincident with the
ing might increase the risk of apical syndrome. Therefore, back surface cone on topography, surrounded by an annulus
using epithelial thickness measurements, hyperopic of thick epithelium.
 To m o gra p h e r s 15 7

We have also hypothesized that epithelial thickness and LASIK worldwide, it is becoming increasingly evident
profile maps could be a useful adjunct to topography in that there is a distinct need for a method of determining the
monitoring patients after collagen corneal cross-linking layered anatomy of the changes induced. Without an accu-
(CXL) treatment. We have used Artemis to measure epi- rate anatomical diagnosis, topography or wavefront-guided
thelial thickness profile in vivo across the central 10-mm treatments may lead to a suboptimal treatment plan.
corneal diameter of a patient before CXL for post-LASIK Here, we show some examples where the epithelium
corneal ectasia and at intervals up to 2 years after CXL. masked true topographic and/or wavefront error to be cor-
The epithelial thickness profile was altered, with a slight rected, and the Artemis provided essential information for
reduction of the area of epithelial thinning and decreased treatment planning.
peripheral thickening. This resulted in minimizing the dif-
C ase 1: Ph otot h erap eu t ic Keratectom y
ference between the thinnest and thickest epithelium and
Treat m en t fo r Ir reg u la r Ep it h elia l
might indicate an improvement of the condition. However,
little change was seen on front corneal surface topography, Th ickn ess Profile
which suggests that the epithelium remodeling masked We have previously published a case of a 60-year-old
the stromal surface changes. This suggests that epithelial male NASA employee who was referred to our clinic with
thickness profiles might be a more sensitive tool to monitor severe visual difficulties in his dominant left eye.95 The
changes in ectasia than front corneal surface topography patient complained that the vision in the left eye was not
and keratometry alone. compatible with that of the right eye and that he felt as
if “everything is coming in from the side.” The patient
In t racorn eal Rin g Segm en ts also complained of reduced contrast sensitivity. These
symptoms meant that the patient preferred to keep his left
We have also previously published a case report show-
eye closed, sometimes wearing an eye patch. His original
ing the epithelial thickness changes that occur after the
manifest refraction in his left eye was -6.25 -0.50 × 180.
insertion of intracorneal ring segments.141 The Artemis
He underwent automated lamellar keratoplasty (ALK),
scan showed that the epithelium was thinner directly over
which resulted in a residual astigmatism of -1.50 D. He
the intracorneal ring segment and thicker to the side of the
was later retreated by arcuate keratotomy (AK) proce-
ridge caused by the intracorneal ring segment. The epitheli-
dures. He then underwent LASIK with a newly created
al reaction is another example of the eyelid tenting effect of
flap to treat residual compound hyperopic astigmatism of
a ridge, which allows the epithelium to grow into the space
+2.50 -1.75 × 95 and a second LASIK procedure with a fur-
created on either side of the ridge, while also thinning on
ther flap cut to treat -1.00 -0.75 × 75. Finally, he underwent a
top of the ridge due to the localized eyelid contact.
LASIK enhancement by relifting a flap to treat a refraction
of +1.00 -1.00 × 65.
Epit h elial Th ickn ess Mappin g for True His UCVA in the left eye was 20/32, improving to 20/25
Diagn osis of Irregu lar Ast igm at ism with -0.50 -0.50 × 80. Topography was irregular with a
While LASIK and PRK are already relatively safe pro- central flattened optical treatment zone of approximately
cedures today, we are constantly striving to make them even 4 mm in diameter, slightly displaced inferiorly (Figure
safer. We need to prevent complications, and when these do 6-95A). Within this central zone, there was an outer ring of
occur, we need methods for correcting them and restoring extreme flattening to a power of about 37.5 D surrounding
visual function. In keeping with basic principles of surgery, a central area with a power of about 39 D at a diameter of
accurate imaging and biometry will be the cornerstone of approximately 1.5 mm. Higher-order aberrations were sig-
these goals because accurate diagnosis enables optimal nificantly raised. The Artemis epithelial profile (see Figure
treatment planning. 6-95A) revealed a central area of thin epithelium (44 µm)
Surface topography has been the mainstay of diagnostic covering a diameter of approximately 1 mm surrounded by
testing in complicated LASIK. Recently, the introduction concentric rings of thick (up to 75 µm) and thin (down to
of aberrometry has greatly enhanced our diagnostic capa- 32 µm) epithelium. The horizontal cross-sectional B-scan
bilities in being able to understand in a quantitative way (Figure 6-95B) shows the undulations of the surface of
how irregular astigmatism and other shape irregularities Bowman’s layer accompanied by partial epithelial com-
produce visual complaints. However, neither the under- pensation within the troughs. It was found that each ring of
standing of the optical defect nor the surface shape of the thickened epithelium coincided with the rings of flattening
cornea will necessarily provide a diagnosis for the cause of on topography. This is an example of the known phenomena
the problem.129 This is due to the fact that internal corneal of the epithelium remodeling itself to try to regularize the
refractive interfaces (such as epithelial-stromal interface) front surface of the cornea142 ; it had become thicker to fill
are not being measured independently. The anatomical in troughs in the stromal surface and thinner over peaks
cause of a surface abnormality may only be understood at in the stromal surface. However, the irregular topography
an internal corneal level (eg, irregularities in the flap versus shows that the epithelium had not been able to completely
the stromal bed). With burgeoning surgical rates of PRK compensate for the stromal irregularities.
158 Ch a p t e r 6

A B

Figure 6 -95. (A) TMS-3 front corneal surface topography before and after Artemis-assisted transepithelial PTK plotted on the
same scale. The topography is significantly more regular postoperatively, with a relatively normal optical zone. The map on the
right shows the change in the topography following the PTK treatment. Bottom row: Artemis epithelial thickness profile before
and after Artemis assisted transepithelial PTK plotted on the same scale. The concentric rings of thin and thick epithelium
have been significantly regularized. The map on the right shows the change in the epithelium following the PTK treatment.
(B) Non-geometrically corrected horizontal B-scans through the visual axis of a cornea before and 3 months after Artemis assist-
ed transepithelial PTK. There is a large z-axis zoom in scale (2.1 mm represented horizontally on the image) relative to lateral
distance (10 mm represented vertically on the image). The concentric rings of epithelial thinning and epithelial thickening are
marked on the pre-PTK B-scan. The epithelium has become significantly more regular as a consequence of the Artemis-assisted
trans-epithelial PTK smoothing of the stromal surface irregularities.

In this case, topography and wavefront analysis did not superiorly of the pupil margins with the aim of improving
provide information on the etiology of the surface irregu- the near function, similar to a presbyopic correction. Six
larities. Mapping of the epithelial thickness map allowed months after the second procedure, the surgeon recom-
the diagnosis of irregular stromal surface to be made. The mended a lamellar graft, which the patient refused. Since
visual symptoms were most probably due to micro-optical then, the patient sought advice from several surgeons to
scattering within the cornea as well as HOAs. Therefore, improve his quality of vision in the left eye.
the treatment plan was split into 2 parts: (1) perform a VHF On presentation to the London Vision Clinic in February
digital ultrasound-assisted transepithelial phototherapeutic 2009, the patient’s refraction was +6.50 -8.00 x 110 achiev-
keratectomy (PTK) using the epithelium as a mask to focus ing CDVA of 20/50. ATLAS front corneal surface topogra-
the laser ablation on the areas of raised stroma in order to phy was irregular showing an asymmetric bow-tie pattern
regularize the stromal surface and (2) remove any remain- decentered superiorly (Figure 6-96A). Simulated central
ing epithelium and perform a wavefront-guided ablation to keratometry was 42.12 x 4/36.75 x 94. The bow-tie pat-
attempt to correct the higher-order aberrations also present. tern was surrounded inferiorly by a crescent-shaped zone
The treatment successfully regularized the stromal surface, of corneal steepening, extending nasally and temporally.
confirmed by the regularization of the epithelial thickness Higher-order aberrations were raised. The Artemis epithe-
profile (see Figure 6-95A), which can also be seen on the lial thickness profile (see Figure 6-96A) was very irregular
postoperative B-scan (see Figure 6-95B). The HOAs were in thickness with up to 35 µm variation within the central
dramatically reduced, the contrast sensitivity was improved 4-mm corneal diameter. The epithelial thickness map dem-
from below normal to high normal, and the BSCVA was onstrated 2 small zones of thin epithelium (40 µm), approx-
improved to 20/20. imately 1 mm in diameter each, at a 2-mm radius inferiorly
and superiorly from the corneal vertex; these regions of thin
C ase 2: Ph otot h erap eu t ic Keratectom y
epithelium were coincident with the trapezoidal incisions.
Treat m en t fo r Ir reg u la r Ep it h elia l
The epithelium was thicker (up to 75 µm) along the horizon-
Th ickn ess Profile tal meridian centrally, extending nasally, and at the 3-mm
A 50-year-old man was referred to our clinic complain- radius inferonasally. In this case, the epithelial thickness
ing of poor vision in his left eye following several refractive profile was masking a significant proportion of the stromal
procedures. His original manifest refraction was approxi- irregularity from front corneal surface topography, mean-
mately -5.50 D. He underwent a radial keratotomy proce- ing that this proportion of the stromal irregularity would
dure in June 1982, followed by another surgery 2 months not be taken into account by a topography-guided ablation
later consisting of trapezoid stromal incisions inferiorly and algorithm. Therefore, the optimal treatment plan was to
 To m o gra p h e r s 15 9

Figure 6 -96. (A) Atlas front corneal surface topography before and after Artemis-assisted transepi-
thelial PTK plotted on the same scale. The superior astigmatic flattening has been significantly
reduced postoperatively, highlighted by the bow-tie pattern seen on the difference map showing
up to 3.50 D of steepening. Bottom row: Artemis epithelial thickness profile before and after
Artemis-assisted transepithelial PTK plotted on the same scale. The superior and inferior regions
of thin epithelium have become thicker postoperatively, and the epithelium has also become
thinner centrally, resulting in a significantly smoother epithelial thickness profile. The map on
the right shows the change in the epithelium following the PTK treatment that highlights the
10 µm to 15 µm of thickening in the regions where the epithelium was thinnest before surgery.
(B) Artemis digital subtraction pachymetry simulation of the pattern of remaining epithelium after
45 µm of epithelial tissue removal by PTK. The white areas indicate where the stromal surface
would be exposed (ie, the areas where stroma would be ablated). The stromal surface would first
be exposed in these areas because the epithelium was thinner to compensate for the raised areas
on the irregular stromal surface. Middle: Intraoperative photograph showing the epithelial pattern
after 45 µm of PTK ablation. O n inspection, the pattern of exposed stroma closely resembles
the Artemis-predicted pattern. Right: Predicted stromal ablation profile calculated based on the
Artemis epithelial thickness data. The stromal ablation profile shows the greatest ablation of
25 µm was in the superior and inferior regions where the epithelium was thinnest, coincident
with the location of the extra trapezoidal incisions performed during the second RK procedure.

perform an Artemis-assisted transepithelial PTK proce- shows the simulated epithelial pattern next to the actual
dure to target the component of the stromal irregularity intraoperative photograph of the cornea after 45 µm of
compensated for by the epithelium. Once the irregularity is ablation. These images show that the pattern of remaining
reduced, the epithelium should become sufficiently smooth epithelium matched that predicted by the Artemis epithe-
for a topography-guided ablation to be used to correct the lial maps, with stroma exposed in the 2 regions where the
remaining irregularity. epithelium was thinnest, 2 mm superiorly and inferiorly of
The epithelial thickness profile was used to generate a the corneal vertex. The concentration of stromal ablation
simulation of the pattern of epithelium after regular inter- superiorly and inferiorly in the vertical meridian meant that
vals of ablation, which were used to monitor the progress the stromal ablation pattern (see Figure 6-96B) was similar
of the PTK ablation during the procedure. Figure 6-96B to a hyperopic astigmatic ablation.
160 Ch a p t e r 6

Nine months after the procedure, the astigmatism had also on the Orbscan BFS surface shape map. Beneath
been halved so that the manifest refraction was +4.50 -4.50 the raised (R) area, the epithelial thickness is seen to be
x 101 and the CDVA had improved to 20/20. The post- reduced, due to invagination by the underlying Bowman’s
operative Artemis epithelial thickness profile (see Figure layer (B). Bowman’s (B) is highly irregular, showing
6-96A) showed that the epithelium was much more regu- 3 major ultrasonic discontinuities (*) representing either
lar in thickness. The epithelial thickness difference map cracks or microfolds in the flap surface. Pachymetric maps
(see Figure 6-96A) demonstrated that the epithelium had of the epithelium, residual stroma, and stromal component
become thicker in the superior and inferior regions where of the flap are shown in Figure 6-97A. The epithelial thick-
the maximum ablation was performed and had become ness profile is seen to vary continuously, filling in and
thinner centrally where the epithelium was thickest before smoothing out the surface of Bowman’s layer. The thin-
the procedure. The change could also be seen on ATLAS nest point within the residual stromal bed, determined by
front corneal surface topography (see Figure 6-96A) where 3-dimensional thickness mapping, was 223 µm. The resid-
the difference map showed a significant astigmatic change. ual stromal layer thickness profile appears slightly asym-
The PTK procedure had succeeded in reducing the stromal metric or decentered in the nasal direction. Inspection of
irregularity, and the presence of a smoother epithelium the stromal component of the flap map showed the reason
meant that the impact of epithelial masking on front corneal for this—the stromal component of the flap was thicker
surface topography was also reduced. Therefore, the patient temporally than nasally.
was then suitable for treatment using a topography-guided A diagnosis was made of flap malposition and possible
ablation as the majority of the remaining irregularity was asymmetric biomechanical shift. In addition, the residual
apparent on front corneal surface topography. stromal thickness was noted to be too thin for further
This case demonstrates the influence the epithelium can under-the-flap ablation, despite the fact that the preopera-
exert on the manifest refraction; in this case, a refraction tive parameters would have implied that there was room for
change of +2.24 -3.97 x 120 was achieved with a trans- further treatment.
epithelial PTK ablation alone. This example shows that the This case clearly illustrates the importance of anatomi-
pattern of irregular epithelium and the associated refrac- cal diagnosis, in contrast to a topographical description,
tive effect must be taken into account when planning PTK in planning the management of LASIK complications. By
and/or custom ablation (topography guided or wavefront topography alone, this case may well have been diagnosed
guided). as a decentration. The eye may well have then undergone a
topographically guided treatment under the flap. Given the
C a se 3: To p o grap h ic D iagn o sis o f
low residual stromal thickness, it is conceivable that further
D ecen t rat io n : Is It Rea lly a La ser
tissue removal would have led to further mechanical shifts
Decen t rat ion ? and an unpredictable result, with a high possibility of induc-
Decentration is a diagnosis made postoperatively by ing progressive ectasia.143
inspection of topography. Decentration denotes off-center
ablation. We have found that what appears to be decentra-
tion by topography is not always due to off-center ablation. Con clu sion
In the following example, a patient presented to us com-
plaining of monocular double vision after LASIK. The ini- In this section, we have demonstrated that knowledge
tial refraction was -6.50 D. Treatment was carried out with of the epithelial thickness profile is a useful tool in under-
the Moria LSK-One microkeratome and the Nidek EC5000. standing and planning corneal refractive procedures. We
Preoperative corneal thickness by Orbscan was measured have shown that the epithelium always remodels itself to fit
as 516 µm. With an ablation depth of 90 µm, the predicted a template (normally provided by the eyelid) and that this
postoperative residual stromal thickness was 266 µm. On mechanism results in compensation for stromal irregular-
examination, UDVA was 20/70, and manifest refraction ity, which then impacts the validity of front corneal surface
was +3.00 -3.75 x 96, yielding a CDVA of 20/40+2. Slit- topography. Therefore, the epithelial thickness profile needs
lamp examination showed a clear cornea, with an unre- to be considered alongside front corneal surface topography
markable flap possessing a few very faint, faded shallow- to build a picture of true stromal irregularity.
appearing vertical microfolds. Orbscan anterior elevation We have described the major areas where the epithelium
BFS map (Figure 6-97A) provided a differential diagnosis can make a significant difference. Mapping of the epithelial
of decentration of the ablation zone or ectasia. Aberrometry thickness profile may increase sensitivity and specificity of
demonstrated coma-like HOAs. screening for keratoconus compared to conventional front
Horizontal 3D Artemis VHF digital ultrasound B-scan corneal surface topography. This should help reduce cases
cross-section of the cornea revealed anatomical features of post-LASIK ectasia and therefore improve the safety of
that provided further diagnostic information. Figure 6-97B LASIK. Knowledge of epithelial thickness profile changes
shows the B-scan demonstrating a flatter (F) nasal side of after myopic and hyperopic LASIK may partly explain
the cornea, with a raised (R) surface temporally as found refractive regression and could be used to improve the
 To m o gra p h e r s 161

B Figure 6-97. (A) O rbscan ante-

rior BFS (default 10-mm zone fit)
plot of the cornea in a patient
presenting with monocular diplo-
pia and a topographic diagnosis
of “decentered ablation,” proved
incorrect by B-scan imaging in
the plane represented by the hori-
zontal black line. Flatter (F) and
raised (R) areas are correlated to
the ultrasound B-scan in Figure
6-97B. Right: Artemis pachymetric
maps of the epithelium, residual
stroma, and stromal component of
the flap. Inspection of the epithe-
lial thickness profile demonstrates
the error introduced by epithelial
compensation if one were to attempt topography-guided or wavefront-guided ablation to correct the optical defect. The residual
stromal layer thickness profile appears slightly asymmetric or decentered in the nasal direction. Inspection of the stromal compo-
nent of the flap map showed the reason for this—the stromal component of the flap was thicker temporally than nasally. B-scan
imaging (Figure 6-97B) confirms that laser ablation would be a less optimal management strategy in this case, in which there is
extreme flap bunching due to inadequate distension. (B) Horizontal VHF digital ultrasound corneal B-scan through the visual
axis of the right cornea of the patient in Figure 6-97A. The upper image (1) shows the geometrically corrected image, while the
lower image (2) shows the raw ultrasound data with axial zoom to better appreciate the interfaces. The surface of epithelium
(E), Bowman’s (B), and the keratectomy interface (I) are labeled. It is clearly noted that Bowman’s surface is highly irregular,
with numerous true microfolds (*), which were only very faintly visible on slit-lamp examination, due to the impressive epithelial
compensation producing excellent smoothing of the corneal surface. The diagnosis of “decentered ablation” is clearly less likely
than that of an inadequately distended flap, producing surface asymmetry. Appropriate management would most likely involve
flap distension and repositioning, not further laser ablation.

accuracy of corneal ablations. In cases of complicated cor- pachymetry will be skewed by the epithelial changes that
neal refractive surgery, knowledge of epithelial thickness will have occurred after tissue removal (eg, achieved depth
profile is crucial in establishing the correct diagnosis and after a myopic ablation will be underestimated due to the
planning further treatment. epithelial thickening that will have occurred).148 Epithelial
There are numerous other situations where the epi- changes should also be considered in flap thickness mea-
thelium should be taken into account that have not been surement after excimer laser ablation. A postoperative mea-
covered in this chapter. We have also published numerous surement of flap thickness will include epithelial changes;
other reports demonstrating epithelial compensation for therefore, a more accurate method of measuring flap thick-
stromal irregularities in cases of asymmetric resection in ness at the time of flap creation is to combine the stromal
automated lamellar keratoplasty,129 asymmetric LASIK component of the flap (measured postoperatively) with
flaps,129,144,145 microfolds,144,145 flap malposition,144-146 the preoperative epithelium.149,150 This method effectively
short flap,144-146 and free cap malrotation.147 This weight resets the epithelium to the thickness profile at the time of
of evidence for epithelial compensation proves that it always surgery.
occurs and that the epithelium must be considered in rela- Orthopedic surgery was practiced without pre- and post-
tion to any corneal interaction. operative anatomical imaging until the discovery of X-ray
Consideration of epithelial changes also applies to imaging in 1895, by Wilhelm Konrad Roentgen. Perhaps,
corneal measurements. For example, an estimation of epithelial imaging will have a similar impact on corneal
achieved ablation depth derived from the change in corneal refractive surgery.
162 Ch a p t e r 6

R EFERENCES
1. Pflugfelder SC, Liu Z, Feuer W, Verm A. Corneal thickness indices 26. Reinstein DZ, Gobbe M, Archer TJ, Silverman RH, Coleman DJ.
discriminate between keratoconus and contact lens-induced corneal Epithelial, stromal, and total corneal thickness in keratoconus:
thinning. Ophthalmology. 2002;109(12):2336-2341. three-dimensional display with Artemis very-high frequency digital
2. Tanabe T, Oshika T, Tomodokoro A, et al. Standardized color-coded ultrasound. J Refract Surg. 2010;26:259-271.
scales for anterior and posterior elevation maps of scanning slit 27. Reinstein DZ, Archer TJ, Gobbe M. Corneal epithelial thick-
corneal topography. Ophthalmology. 2002;109(7):1298-1302. ness profile in the diagnosis of keratoconus. J Refract Surg.
3. Davis LJ, Schechtman KB, Begley CG, Shin JA, Zadnik K. 2009;25(7):604-610.
Repeatability of refraction and corrected visual acuity in kera- 28. Alessio G, Boscia F, La Tegola MG, Sborgia C. Topography-
toconus. The CLEK Study Group. Collaborative Longitudinal driven photorefractive keratectomy: results of corneal interac-
Evaluation of Keratoconus. Optom Vis Sci. 1998;75(12):887-896. tive programmed topographic ablation software. Ophthalmology.
4. Zadnik K, Steger-May MA, Fink BA, et al. Between-eye asymmetry 2000;107(8):1578-1587.
in keratoconus. Cornea. 2002;21:671-679. 29. Huang D, Izatt JA. Physics and fundamentals of anterior segment
5. Liu Z, Pflugfelder SC. Corneal thickness is reduced in dry eye. optical coherence tomography. In: Steinert RF, Huang D, eds,
Cornea. 1999;18(4):403-407. Anterior Segment Optical Coherence Tomography. Thorofare, NJ:
6. Pepose JS, Applegate RA. Making sense out of wavefront sensing. SLACK Incorporated; 2008:3-10.
Am J Ophthalmol. 2005;139(2):335-343. 30. Visante OCT User Manual System Software [computer program].
7. Belin MW. Intraoperative raster photogrammetry—the PAR Corneal Version 3.0. Jena, Germany: Carl Zeiss Meditec; 2009.
Topography System. J Cataract Refract Surg. 1993;19:188-192. 31. Baı¨koff G. Anterior segment OCT and phakic intraocular lenses: a
8. Belin MW, Zloty P. Accuracy of the PAR Corneal Topography perspective. J Cataract Refract Surg. 2006;32:1827-1835.
System with spatial misalignment. CLAO J. 1993;19:64-68. 32. Hill W, Wang L, Koch DD. IOL power calculation in eyes that have
9. Arffa RC, Klyce SD, Busin M. Keratometry in refractive surgery. undergone LASIK/PRK/RK. http://iol.ascrs.org. Accessed July 18,
J Refract Surg. 1986;2:6. 2011.
10. Rubin ML. Optics for Clinicians. Gainesville, FL: Triad Publishing 33. Rabinowitz YS, McDonnell PJ. Computer-assisted corneal topogra-
Company; 1993. phy in keratoconus. Refract Corneal Surg. 1989;5(6):400-408.
11. Mandell RB. The enigma of the corneal contour. CLAO J. 34. Rabinowitz YS. Keratoconus. Surv Ophthalmol. 1998;42(4):297-
1992;18:267-273. 319.
12. Arffa RC, Warnicki JW, Rehkopf PG. Corneal topography using 35. Borasio E, Stevens J, Smith GT. Estimation of true corneal power
rasterstereography. Refract Corneal Surg. 1989;5:414-417. after keratorefractive surgery in eyes requiring cataract surgery:
13. Litoff D, Belin MW, Winn SS, et al. PAR Technology Corneal BESSt formula. J Cataract Refract Surg. 2006;32(12):2004-2014.
Topography System. Inv Ophthalmol Vis Sci. 1991;32:922. 36. Holladay JT, Hill WE, Steinmueller A. Corneal power measure-
14. Swartz T, Marten L, Wang M. Measuring the cornea: the latest ments using scheimpflug imaging in eyes with prior corneal refrac-
developments in corneal topography. Curr Opin Ophthalmol. tive surgery. J Refract Surg. 2009;25(10):862-868.
2007;18:325-333. 37. Jin H, Rabsilber T, Ehmer A, et al. Comparison of ray-tracing
15. Wegener A, Laser H. Image analysis and Scheimpflug photog- method and thin-lens formula in intraocular lens power calculations.
raphy of anterior segment of the eye--a review. Klin Monatsbl J Cataract Refract Surg. 2009;35(4):650-662.
Augenheilkd. 2001;218:67-77. 38. Shammas HJ, Hoffer KJ, Shammas MC. Scheimpflug photography
16. Müller-Breitenkamp U, Hockwin O. Scheimpflug photography in keratometry readings for routine intraocular lens power calculation.
clinical ophthalmology. A review. Ophthalmic Res. 1992;24:47-54. J Cataract Refract Surg. 2009;35(2):330-334.
17. Ciolino J, Belin MW. Changes to the posterior cornea after LASIK 39. Amano S, Honda N, Amano Y, et al. Comparison of central corneal
and PRK. J Cataract & Refract Surg. 2006;32:1426-1431. thickness measurements by rotating Scheimpflug camera, ultrasonic
18. Miller D, Greiner JV. Corneal measurements and tests. In: Albert pachymetry, and scanning-slit corneal topography. Ophthalmology.
DM, Jakobiec FA, eds. Principles and Practice of Ophthalmology. 2006;113(6):937-941.
Philadelphia, PA: WB Saunders; 1994:7. 40. Ambrosio R Jr. Percentage thickness increase and absolute differ-
19. Ambrósio R Jr, Klyce SD, Wilson SE. Corneal topographic and ence from thinnest to describe thickness profile. J Refract Surg.
pachymetric screening of keratorefractive patients. J Refract Surg. 2010;26(2):84-86; author reply 86-87.
2003;19:24-29. 41. Luz A, Ursulio M, Castaneda D, Ambrosio R Jr. [Corneal thickness
20. Jonsson M, Behndig A. Pachymetric evaluation prior to laser in situ progression from the thinnest point to the limbus: study based on
keratomileusis. J Cataract Refract Surg. 2005;31:701-706. a normal and a keratoconus population to create reference values].
21. Ciolino JB, Khachikian SS, Belin MW. Comparison of corneal Arq Bras Oftalmol. 2006;69(4):579-583.
thickness measurements by ultrasound and Scheimpflug photogra- 42. Ambrósio R Jr, Nogueira LP, Caldas DL, et al. Evaluation of cor-
phy in eyes that have undergone laser in situ keratomileusis. Am J neal shape and biomechanics before LASIK. Int Ophthalmol Clin.
Ophthalmol. 2008;145:75-80. 2011;51(2):11-38.
22. Ambrósio R, Alonso RS, Luz A, Coca Velarde LG. Corneal- 43. Ambrósio Jr R, Caiado ALC, Guerra FP, et al. Novel pachymetric
thickness spatial profile and corneal-volume distribution: tomo- parameters based on corneal tomography for diagnosing keratoco-
graphic indices to detect keratoconus. J Cataract Refract Surg. nus. J Refract Surg. 2011; in press
2006;32(11):1851-1859. 44. Ambrosio R Jr, Dawson DG, Salomao M, Guerra FP, Caiado AL,
23. Klein SR, Epstein RJ, Randleman JB, Stulting RD. Corneal ectasia Belin MW. Corneal ectasia after LASIK despite low preoperative
after laser in situ keratomileusis in patients without apparent preop- risk: tomographic and biomechanical findings in the unoperated,
erative risk factors. Cornea. 2006;25:388-403. stable, fellow eye. J Refract Surg. 2010:1-6.
24. Rabinowitz YS. Videokeratographic indices to aid in screening for 45. Belin MW, Ambrosio R Jr. Corneal ectasia risk score: statistical
keratoconus. J Refract Surg. 1995;11:371-379. validity and clinical relevance. J Refract Surg. 2010;26(4):238-240.
25. Binder PS, Lindstrom RL, Stulting RD, et al. Keratoconus and cor- 46. Belin MW, Khachikian SS. An introduction to understanding eleva-
neal ectasia after LASIK. J Refract Surg. 2005;21:749-752. tion-based topography: how elevation data are displayed—a review.
Clin Experiment Ophthalmol. 2009;37(1):14-29.
 To m o gra p h e r s 16 3

47. Ciolino JB, Belin MW. Changes in the posterior cornea after laser 68. Stahl JE, Durrie DS, Schwendeman FJ, Boghossian AJ. Anterior
in situ keratomileusis and photorefractive keratectomy. J Cataract segment OCT analysis of thin IntraLase femtosecond flaps.
Refract Surg. 2006;32(9):1426-1431. J Refract Surg. 2007;23:555-558.
48. Packer M, Fine IH, Hoffman RS. Aspheric intraocular lens selection 69. Rocha KM, Randleman JB, Stulting RD. Analysis of thin micro-
based on corneal wavefront. J Refract Surg. 2009;25(1):12-20. keratome flap architecture using Fourier-domain optical coherence
49. Rabsilber TM, Khoramnia R, Auffarth GU. Anterior chamber mea- tomography. Paper presented at: American Society of Cataract and
surements using Pentacam rotating Scheimpflug camera. J Cataract Refractive Surgery (ASCRS) meeting; April 9-14, 2010; Boston,
Refract Surg. 2006;32(3):456-459. MA.
50. Tehrani M, Dick HB. Scheimpflug biometry of the anterior segment 70. Das S, Sullivan LJ. Comparison of residual stromal bed and flap
after implantation of foldable iris-fixated lenses. J Refract Surg. thickness in primary and repeat laser in situ keratomileusis in myo-
2006;22(3):243-246. pic patients. J Cataract Refract Surg. 2006;32(12):2080-2084.
51. Lovisolo CF, Reinstein DZ. Phakic intraocular lenses. Surv 71. Reinstein DZ, Srivannaboon S, Archer TJ, Silverman RH, Sutton H,
Ophthalmol. 2005;50(6):549-587. Coleman DJ. Probability model of the inaccuracy of residual stro-
52. de Souza RF, Allemann N, Forseto A, Barros PS, Chamon W, Nose mal thickness prediction to reduce the risk of ectasia after LASIK
W. Ultrasound biomicroscopy and Scheimpflug photography of part I: quantifying individual risk. J Refract Surg. 2006;22(9):851-
angle-supported phakic intraocular lens for high myopia. J Cataract 860.
Refract Surg. 2003;29(6):1159-1166. 72. Randleman JB, Hewitt SM, Lynn MJ, Stulting RD. A comparison
53. Kwon RO, Price MO, Price FW Jr, Ambrosio R Jr, Belin MW. of 2 methods for estimating residual stromal bed thickness before
Pentacam characterization of corneas with Fuchs dystrophy treated repeat LASIK. Ophthalmology. 2005;112(1):98-103.
with descemet membrane endothelial keratoplasty. J Refract Surg. 73. Prandi B, Baviera J, Morcillo M. Influence of flap thickness on
2010:1-8. results of laser in situ keratomileusis for myopia. J Refract Surg.
54. Nixon DR. Preoperative cataract grading by Scheimpflug imaging 2004;20(6):790-796.
and effect on operative fluidics and phacoemulsification energy. 74. Dupps WJ Jr, Roberts C. Effect of acute biomechanical changes
J Cataract Refract Surg. 2010;36(2):242-246. on corneal curvature after photokeratectomy. J Refract Surg.
55. Cummings AB, Mascharka N. Outcomes after topography-based 2001;17(6):658-669.
LASIK and LASEK with the WaveLight Oculyzer and Topolyzer 75. Maeda N, Klyce SD, Smolek MK, Thompson HW. Automated
Platforms. J Refract Surg. 2009:1-8. keratoconus screening with corneal topography analysis. Invest
56. Huang D, Swanson EA, Lin CP, et al. Optical coherence tomogra- Ophthalmol Vis Sci. 1994;35:2749-2757.
phy. Science. 1991;254(5035):1178-1181. 76. Schwiegerling J, Greivenkamp JE. Keratoconus detection based on
57. Radhakrishnan S, Goldsmith J, Huang D, et al. Comparison of videokeratoscopic height data. Optom Vis Sci. 1996;73:721-728.
optical coherence tomography and ultrasound biomicroscopy for 77. Rabinowitz YS, Rasheed K. KISA% index: a quantitative videoker-
detection of narrow anterior chamber angles. Arch Ophthalmol. atography algorithm embodying minimal topographic criteria for
2005;123(8):1053-1059. diagnosing keratoconus. J Cataract Refract Surg. 1999;25:1327-
58. Müller M, Dahmen G, Pörksen E, et al. Anterior chamber angle 1335.
measurement with optical coherence tomography: intraobserver and 78. Tang M, Shekhar R, Miranda D, Huang D. Characteristics of kerato-
interobserver variability. J Cataract Refract Surg. 2006;32(11):1803- conus and pellucid marginal degeneration in mean curvature maps.
1808. Am J Ophthalmol. 2005;140:993-1001.
59. Dada T, Sihota R, Gadia R, Aggarwal A, Mandal S, Gupta V. 79. Krachmer JH, Feder RS, Belin MW. Keratoconus and related
Comparison of anterior segment optical coherence tomography and noninflammatory corneal thinning disorders. Surv Ophthalmol.
ultrasound biomicroscopy for assessment of the anterior segment. 1984;28:293-322.
J Cataract Refract Surg. 2007;33(5):837-840. 80. Vinciguerra P, Camesasca FI. Prevention of corneal ectasia in laser
60. Wong HT, Lim MC, Sakata LM, et al. High-definition optical in situ keratomileusis. J Refract Surg. 2001;17:S187-S189.
coherence tomography imaging of the iridocorneal angle of the eye. 81. Li Y, Meisler DM, Tang M, et al.Keratoconus diagnosis with opti-
Arch Ophthalmol. 2009;127:256-260. cal coherence tomography pachymetry mapping. Ophthalmology.
61. Radhakrishnan S, Rollins AM, Roth JE, et al. Real-time optical 2008;115(12):2159-2166.
coherence tomography of the anterior segment at 1310 nm. Arch 82. Li Y, Shekhar R, Huang D. Corneal pachymetry mapping with
Ophthalmol. 2001;119(8):1179-1185. high-speed optical coherence tomography. Ophthalmology.
62. Hee MR, Izatt JA, Swanson EA, et al. Optical coherence tomography 2006;113:792-799.
of the human retina. Arch Ophthalmol. 1995;113(3):325-332. 83. Reinstein DZ, Archer TJ, Gobbe M. Stability of LASIK in topo-
63. Keane PA, Bhatti RA, Brubaker JW, Liakopoulos S, Sadda SR, graphically suspect keratoconus confirmed non-keratoconic by
Walsh AC. Comparison of clinically relevant findings from high- Artemis VHF digital ultrasound epithelial thickness mapping:
speed Fourier-domain and conventional time-domain optical coher- 1-year follow-up. J Refract Surg. 2009;25(7):569-577.
ence tomography. Am J Ophthalmol. 2009;148(2):242-248.e1. 84. Randleman JB, Loupe DN, Song CD, Waring GO III, Stulting RD.
64. Sarunic MV, Asrani S, Izatt JA. Imaging the ocular anterior seg- Intraocular lens power calculations after laser in situ keratomileusis.
ment with real-time, full-range Fourier-domain optical coherence Cornea. 2002;21:751-755.
tomography. Arch Ophthalmol. 2008;126:537-542. 85. Koch DD. New options for IOL calculations after refractive surgery.
65. Wojtkowski M, Bajraszewski T, Targowski P. A realtime in vivo J Cataract Refract Surg. 2006;32:371-372.
imaging by high-speed spectral optical coherence tomography. Opt 86. Chokshi AR, Latkany RA, Speaker MG. IOL calculations after
Lett. 2003;28:1745-1747. refractive surgery. J Cataract Refract Surg. 2006;32:1984.
66. Wylegała E, Teper S, Nowińska AK, Milka M, Dobrowolski 87. Seitz B, Langenbucher A. Intraocular lens calculations status after
D. Anterior segment imaging: Fourier-domain optical coherence corneal refractive surgery. Curr Opin Ophthalmol. 2000;11:35-46.
tomography versus time-domain optical coherence tomography. 88. Gimbel HV, Sun R. Accuracy and predictability of intraocular
J Cataract Refract Surg. 2009;35(8):1410-1414. lens power calculation after laser in situ keratomileusis. J Cataract
67. Kim JH, Lee D, Rhee KI. Flap thickness reproducibility in laser Refract Surg. 2001;27:571-576.
in situ keratomileusis with a femtosecond laser: optical coherence 89. Aramberri J. Intraocular lens power calculation after corneal
tomography measurement. J Cataract Refract Surg. 2008;34:132- refractive surgery: double-K method. J Cataract Refract Surg.
136. 2003;29:2063-2068.
164 Ch a p t e r 6

90. Speicher L. Intra-ocular lens calculation status after corneal refrac- 111. Perez JG, Meijome JM, Jalbert I, Sweeney DF, Erickson P. Corneal
tive surgery. Curr Opin Ophthalmol. 2001;12:17-29. epithelial thinning profile induced by long-term wear of hydrogel
91. Odenthal MT, Eggink CA, Melles G, Pameyer JH, Geerards AJ, lenses. Cornea. 2003;22:304-307.
Beekhuis WH. Clinical and theoretical results of intraocular lens 112. Sin S, Simpson TL. The repeatability of corneal and corneal epithe-
power calculation for cataract surgery after photorefractive keratec- lial thickness measurements using optical coherence tomography.
tomy for myopia. Arch Ophthalmol. 2002;120:431-438. Optom Vis Sci. 2006;83:360-365.
92. Arce CG, Soriano ES, Weisenthal RW, et al. Calculation of intraoc- 113. Li HF, Petroll WM, Moller-Pedersen T, Maurer JK, Cavanagh HD,
ular lens power using Orbscan II quantitative area topography after Jester JV. Epithelial and corneal thickness measurements by in vivo
corneal refractive surgery. J Refract Surg. 2009;25(12):1061-1074. confocal microscopy through focusing (CMTF). Curr Eye Res.
93. Hanna C, O’Brien JE. Cell production and migration in the epithe- 1997;16:214-221.
lial layer of the cornea. Arch Ophthalmol. 1960;64:536-539. 114. Patel SV, McLaren JW, Hodge DO, Bourne WM. Confocal micros-
94. Simon G, Ren Q, Kervick GN, Parel JM. Optics of the corneal epi- copy in vivo in corneas of long-term contact lens wearers. Invest
thelium. Refract Corneal Surg. 1993;9:42-50. Ophthalmol Vis Sci. 2002;43:995-1003.
95. Reinstein DZ, Archer T. Combined Artemis very high-frequency 115. Reinstein DZ, Silverman RH, Raevsky T, et al. Arc-scanning very
digital ultrasound-assisted transepithelial phototherapeutic keratec- high-frequency digital ultrasound for 3D pachymetric mapping of
tomy and wavefront-guided treatment following multiple corneal the corneal epithelium and stroma in laser in situ keratomileusis.
refractive procedures. J Cataract Refract Surg. 2006;32:1870-1876. J Refract Surg. 2000;16:414-430.
96. Vogt A. Textbook and Atlas of Slit Lamp Microscopy of the Living 116. Reinstein DZ, Archer TJ, Gobbe M, Silverman RH, Coleman DJ.
Eye. Bonn: Wayenborgh Editions; 1981. Repeatability of layered corneal pachymetry with the Artemis
97. Gauthier CA, Holden BA, Epstein D, Tengroth B, Fagerholm very high-frequency digital ultrasound arc-scanner. J Refract Surg.
P, Hamberg-Nystrom H. Factors affecting epithelial hyperpla- 2010;26(9):646-659.
sia after photorefractive keratectomy. J Cataract Refract Surg. 117. Reinstein DZ, Silverman RH, Coleman DJ. High-frequency ultra-
1997;23:1042-1050. sound measurement of the thickness of the corneal epithelium.
98. Lohmann CP, Reischl U, Marshall J. Regression and epithelial Refract Corneal Surg. 1993;9:385-387.
hyperplasia after myopic photorefractive keratectomy in a human 118. Reinstein DZ, Silverman RH, Trokel SL, Coleman DJ. Corneal
cornea. J Cataract Refract Surg. 1999;25:712-715. pachymetric topography. Ophthalmology. 1994;101:432-438.
99. Srivannaboon S, Reinstein DZ, Sutton HFS, Silverman RH. Effect 119. Reinstein DZ, Archer TJ, Gobbe M, Silverman RH, Coleman DJ.
of epithelial changes on refractive outcome in LASIK. Invest Epithelial thickness in the normal cornea: three-dimensional dis-
Ophthalmol Vis Sci. 1999;40:S896. play with Artemis very high-frequency digital ultrasound. J Refract
100. Reinstein DZ, Srivannaboon S, Silverman RH, Coleman DJ. Limits Surg. 2008;24:571-581.
of wavefront customized ablation: biomechanical and epithelial fac- 120. Bentivoglio AR, Bressman SB, Cassetta E, Carretta D, Tonali P,
tors. Invest Ophthalmol Vis Sci. 2002;43:E-Abstract 3942. Albanese A. Analysis of blink rate patterns in normal subjects. Mov
101. Reinstein DZ, Srivannaboon S, Silverman RH, Coleman DJ. The Disord. 1997;12:1028-1034.
accuracy of routine LASIK; isolation of biomechanical and epithe- 121. Doane MG. Interactions of eyelids and tears in corneal wetting and
lial factors. Invest Ophthalmol Vis Sci. 2000;41(Suppl):S318. the dynamics of the normal human eyeblink. Am J Ophthalmol.
102. Scroggs MW, Proia AD. Histopathological variation in keratoconus. 1980;89:507-516.
Cornea. 1992;11:553-559. 122. Rao SN, Raviv T, Majmudar PA, Epstein RJ. Role of Orbscan II in
103. Haque S, Simpson T, Jones L. Corneal and epithelial thickness in screening keratoconus suspects before refractive corneal surgery.
keratoconus: a comparison of ultrasonic pachymetry, Orbscan II, Ophthalmology. 2002;109:1642-1646.
and optical coherence tomography. J Refract Surg. 2006;22:486- 123. Lim L, Wei RH, Chan WK, Tan DT. Evaluation of keratoconus in
493. Asians: role of Orbscan II and Tomey TMS-2 corneal topography.
104. Aktekin M, Sargon MF, Cakar P, Celik HH, Firat E. Ultrastructure Am J Ophthalmol. 2007;143:390-400.
of the cornea epithelium in keratoconus. Okajimas Folia Anat Jpn. 124. Reinstein DZ. High-frequency ultrasound versus optical coher-
1998;75:45-53. ence tomography. Paper presented at: American Academy of
105. Gatinel D, Racine L, Hoang-Xuan T. Contribution of the cor- Ophthalmology 2003 Annual Meeting; November 15-18, 2003;
neal epithelium to anterior corneal topography in patients having Anaheim, CA.
myopic photorefractive keratectomy. J Cataract Refract Surg. 125. Schlegel Z, Hoang-Xuan T, Gatinel D. Comparison of and cor-
2007;33:1860-1865. relation between anterior and posterior corneal elevation maps in
106. Wang J, Fonn D, Simpson TL, Sorbara L, Kort R, Jones L. normal eyes and keratoconus-suspect eyes. J Cataract Refract Surg.
Topographical thickness of the epithelium and total cornea after 2008;34:789-795.
overnight wear of reverse-geometry rigid contact lenses for myopia 126. Reinstein DZ, Archer TJ, Gobbe M. Stability of LASIK in corneas
reduction. Invest Ophthalmol Vis Sci. 2003;44:4742-4746. with topographic suspect keratoconus, with keratoconus excluded
107. Feng Y, Simpson TL. Comparison of human central cornea and by epithelial thickness mapping. J Refract Surg. 2009;25:569-577.
limbus in vivo using optical coherence tomography. Optom Vis Sci. 127. Reinstein DZ, Archer TJ, Gobbe M. Stability of LASIK in corneas
2005;82:416-419. with topographic suspect keratoconus confirmed non-keratoconic
108. Wirbelauer C, Pham DT. Monitoring corneal structures with by epithelial thickness mapping: 2-years follow-up. Paper presented
slitlamp-adapted optical coherence tomography in laser in situ ker- at: American Academy of Ophthalmology 2009 Annual Meeting;
atomileusis. J Cataract Refract Surg. 2004;30:1851-1860. October 24-27, 2009; San Francisco, CA.
109. Ladage PM, Yamamoto K, Ren DH, et al. Effects of rigid and 128. Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assess-
soft contact lens daily wear on corneal epithelium, tear lactate ment for ectasia after corneal refractive surgery. Ophthalmology.
dehydrogenase, and bacterial binding to exfoliated epithelial cells. 2008;115:37-50.
Ophthalmology. 2001;108:1279-1288. 129. Reinstein DZ, Silverman RH, Sutton HF, Coleman DJ. Very high-
110. Moller-Pedersen T, Vogel M, Li HF, Petroll WM, Cavanagh HD, frequency ultrasound corneal analysis identifies anatomic correlates
Jester JV. Quantification of stromal thinning, epithelial thickness, of optical complications of lamellar refractive surgery: anatomic
and corneal haze after photorefractive keratectomy using in vivo diagnosis in lamellar surgery. Ophthalmology. 1999;106:474-482.
confocal microscopy. Ophthalmology. 1997;104:360-368.
 To m o gra p h e r s 16 5

130. Patel S, Reinstein DZ, Silverman RH, Coleman DJ. The shape of 142. Lazzaro DR, Aslanides IM, Belmont SC, et al. High frequency
Bowman’s layer in the human cornea. J Refract Surg. 1998;14:636- ultrasound evaluation of radial keratotomy incisions. J Cataract
640. Refract Surg. 1995;21:398-401.
131. Patel S, Marshall J, Fitzke FW. Refractive index of the human cor- 143. Reinstein DZ, Srivannaboon S, Sutton HFS, Silverman RH, Shaikh
neal epithelium and stroma. J Refract Surg. 1995;11:100-105. A, Coleman DJ. Risk of ectasia in LASIK: revised safety criteria.
132. Reinstein DZ, Srivannaboon S, Gobbe M, et al. Epithelial thick- Invest Ophthalmol Vis Sci. 1999;40(Suppl):S403.
ness profile changes induced by myopic LASIK as measured by 144. Reinstein DZ, Silverman RH. Artemis VHF digital ultrasound tech-
Artemis very high-frequency digital ultrasound. J Refract Surg. nology. In: Wang M, ed. Corneal Topography in the Wavefront Era.
2009;25:444-450. Thorofare, NJ: SLACK Incorporated; 2006:207-226.
133. Reinstein DZ, Archer TJ, Gobbe M. LASIK for myopic astigma- 145. Reinstein DZ, Silverman RH, Archer TJ. VHF digital ultrasound:
tism and presbyopia using non-linear aspheric micro-monovision Artemis 2 scanning in corneal refractive surgery. In: Vinciguerra
with the Carl Zeiss Meditec MEL80 platform. J Refract Surg. P, Camesasca FI, eds, Refractive Surface Ablation: PRK, LASEK,
2011;27(1):23-37. Epi-LASIK, Custom, PTK and Retreatment. Thorofare, NJ: SLACK
134. Reinstein DZ, Archer TJ, Gobbe M, Silverman RH, Coleman DJ. Incorporated; 2006:315-330.
Epithelial thickness after hyperopic LASIK: three-dimensional dis- 146. Reinstein DZ, Archer TJ. Evaluation of irregular astigmatism with
play with Artemis very high-frequency digital ultrasound. J Refract Artemis very high-frequency digital ultrasound scanning. In: Wang
Surg. 2010;26(8):555-564. M, ed. Irregular Astigmatism: Diagnosis and Treatment. Thorofare,
135. Varley GA, Huang D, Rapuano CJ, Schallhorn S, Boxer Wachler BS, NJ: SLACK Incorporated; 2007:29-42.
Sugar A. LASIK for hyperopia, hyperopic astigmatism, and mixed 147. Reinstein DZ, Rothman RC, Couch DG, Archer TJ. Artemis very
astigmatism: a report by the American Academy of Ophthalmology. high-frequency digital ultrasound-guided repositioning of a free
Ophthalmology. 2004;111:1604-1617. cap after laser in situ keratomileusis. J Cataract Refract Surg.
136. Reinstein DZ, Gobbe M, Archer TJ, Couch D, Bloom B. Epithelial, 2006;32:1877-1882.
stromal, and corneal pachymetry changes during orthokeratology. 148. Reinstein DZ, Archer TJ, Gobbe M. Corneal ablation depth readout
Optom Vis Sci. 2009;86:E1006-1014. of the MEL80 excimer laser compared to Artemis three-dimension-
137. Barraquer JI. Queratomileusis y queratofakia. Bogotá, Colombia: al very high-frequency digital ultrasound stromal measurements.
Instituto Barraquer de America; 1980:342. J Refract Surg. 2010;26(12):949-959.
138. Wellish KL, Glasgow BJ, Beltran F, Maloney RK. Corneal ectasia 149. Reinstein DZ, Sutton HF, Srivannaboon S, Silverman RH, Archer
as a complication of repeated keratotomy surgery. J Refract Corneal TJ, Coleman DJ. Evaluating microkeratome efficacy by 3D corneal
Surg. 1994;10:360-364. lamellar flap thickness accuracy and reproducibility using Artemis
139. Seiler T, Koufala K, Richter G. Iatrogenic keratectasia after laser in VHF digital ultrasound arc-scanning. J Refract Surg. 2006;22:431-
situ keratomileusis. J Refract Surg. 1998;14:312-317. 440.
140. Randleman JB, Trattler WB, Stulting RD. Validation of the Ectasia 150. Reinstein DZ, Archer TJ, Gobbe M, Johnson NF. Accuracy and
Risk Score System for preoperative laser in situ keratomileusis reproducibility of Artemis central flap thickness and visual out-
screening. Am J Ophthalmol. 2008;145:813-818. comes of LASIK with the Carl Zeiss Meditec VisuMax femtosec-
141. Reinstein DZ, Srivannaboon S, Holland SP. Epithelial and stromal ond laser and MEL 80 excimer laser platforms. J Refract Surg.
changes induced by Intacs examined by three-dimensional very 2010;26(2):107-119.
high-frequency digital ultrasound. J Refract Surg. 2001;17:310-318.
Ch a p te r 7

To p o grap hy
and Wave fro nt Co mb ine d
Syste ms

TH E ITRACE C OMBINATION C ORNEAL TOPOGRAPH Y AND

W AVEFRONT SYSTEM
Tracy Schroeder Swartz, OD, MS, FAAO and Joe S. Wakil, MD

T
he iTrace system manufactured by Tracey Soviet military applications and was developed by scientist
Technologies Corp (Houston, TX) is uniquely Vasyl Molebny, PhD,1 and his team of engineers in Kiev,
designed to combine full-surface Placido corneal Ukraine in collaboration with ophthalmic surgeon Ioannis
topography with the advantages of ray-tracing aberrometry, Pallikaris, MD, and his colleagues in Crete, Greece. The
a robust wavefront technology to measure quality of vision. combination of this aberrometry information with corneal
The iTrace incorporates the Placido components manufac- topography in a compact, cost-efficient package provides
tured by EyeSys Vision, Inc, (Houston, TX) with integrated valuable clinical information in elucidating the ocular
software application produced by Tracey Technologies to sources of aberrations, primarily as to whether they are
deliver corneal topography and total eye aberrometry data corneal or lenticular for added diagnostic power.
acquisition and analysis (Figure 7-1).
At the core of this instrument is the unique technology
providing optical ray tracing of the eye, as is evidenced by iTrace En gin eer in g
the product name, the iTrace. A sequential series of thin,
infrared beams or rays of light on the order of 100 µm each Ad va n t ages
is projected into the entire entrance pupil of the eye using a
Important in any system design is the philosophy behind
software-programmable, scanning pattern measuring hun-
it and the utility goals for its practice and use. The iTrace
dreds of points within milliseconds. Each of these points
features a dual-purpose system with full corneal topography
represents the entrance of parallel, sampling light rays into
and wavefront capabilities along with integrated software
the eye, which become refracted by the eye’s optical power
analysis on a single computer platform. This device dem-
eventually focusing on the retina. By locating the spot on
onstrates the far greater power of the combination of these
the retina where each thin beam of light is focused, a direct
2 technologies.
aberration measurement is made leading to calculations
In addition, the slim engineering allows the eye under
for a complete aberration profile and optical performance
measurement to fixate directly through the instrument’s
examination of the eye. This technology stems from former
transparent optics at a distance target. For example, the
Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical
167 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 167-198)
© 2012 SLACK Incorporated
168 Ch a p t e r 7

Figure 7-2. Default topographic display upon data capture.

Note the distortion of the mires secondary to a corneal graft.

Figure 7-1. The iTrace unit’s slim profile enables true distance
target measurement binocularly for strong control of accom-
modation. (Reprinted with permission of Joe S. Wakil, MD.)

instrument may be placed such that the patient looks

through an office window at a tree across the street while
the fellow eye also has an unobstructed view of the same
target. This helps provide an important and proprietary
feature in allowing for binocular open field fixation of a Figure 7-3. Binocular topographic display includes 2 maps
(customized) and various indices valuable to refractive and
real object at a truly far distance much greater than 20 feet
cataract surgeons.
during wavefront measurement. Natural binocular vision
overcomes instrument myopia, a phenomenon where mon-
ocular fixation on an optically simulated “far” target results
K readings, but based on Snell’s Law of Refraction to
in excessive myopia. This phenomenon plagues all conven-
more accurately describe the refractive power of the cen-
tional autorefraction systems in measuring, particularly in
tral cornea. A single effective refractive power reading is
younger patients.
given for the entire central 3-mm zone of the cornea to be
used primarily for IOL calculations. This number has been
popularized by Jack Holladay, MD, as an improvement in
Cor n ea l Top ograp h y understanding central corneal refractive power contribution
Feat u res of t h e iTrace System in more accurately calculating the necessary IOL power for
eyes postrefractive surgery.2
The corneal topography displays are generated from Beyond the numeric data of K readings, refractive power
Placido images of the cornea and are fully customizable readings, and other corneal indices, color maps are pro-
within the iTrace software. Using advanced edge detection vided to depict the unique features of each cornea. For both
software, the Placido image of the cornea is analyzed across 2- and 3-dimensional color maps, a number of algorithms
all the ring edges from center to periphery (Figure 7-2). are available providing complete corneal analysis. These
The binocular display is shown in Figure 7-3. This display include the Axial Map, Local Radius of Curvature Map,
is often preferred by cataract surgeons due to its content. Refractive Map, Z Elevation Map, and Wavefront Map.
Simulated keratometry values, spherical aberrations, and The standard Axial map (Figure 7-4) is based on the
corneal astigmatism (delta value) are easily located on the keratometric formula wherein the curvature of the cornea is
printout and facilitate intraocular lens (IOL) selection. presented with respect to the optical axis of the instrument.
Additionally, a refractive power reading of the cornea This color map has been commonly used over the years as
is calculated for the 3-mm zone in similar fashion to the an extension of the keratometer to the entire corneal surface.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 169

Figure 7-4. Axial map in a patient with keratoconus. Figure 7-5. Local radius of curvature map, otherwise known as
the tangential or instantaneous radius of curvature map, in a
patient with keratoconus.
However, in applying the simple keratometric formula, this
map only provides accurate curvature data and corneal
refractive power data in the central corneal region. The data
outside of the 3-mm zone is simply not accurate in terms
of either the true curvature of the cornea or the refractive
power generated in the peripheral cornea.
The Local Radius of Curvature Map, otherwise known
as the Tangential Map or Instantaneous Radius of Curvature
Map (Figure 7-5), is used to accurately depict the true cor-
neal shape in terms of millimeter radius of curvature from
center to periphery. This map is a mathematical derivative
of the axial color map and can be most useful in appreciat-
ing the detail of corneal surface features both centrally and
peripherally. Contact lens fitters will derive more accurate
curvature measurements in the periphery to better fit con-
tact lenses, while refractive surgeons will better appreciate Figure 7-6. Refractive map in the same patient with kerato-
the sharpness of transition zone edges from excimer laser conus.
ablations that commonly generate halos and glare to post-
surgical eyes with small effective optical zones using this
map. Also, for the critical diagnosis of keratoconus, the The Z Elevation Map (Figure 7-7) illustrates the physi-
Local Radius of Curvature Map will emphasize the true cal nature of the cornea in terms of its height difference in
size and nature of an ectasia as observed in the slit-lamp microns from a reference sphere. The algorithms used to
microscope. This is very important to note as it is necessary choose a reference sphere are important, as one can gener-
to know the actual apex location and severity of curvature ate a number of different elevation measurements based
generated by the keratoconus to treat appropriately. The solely upon the particular reference sphere. iTrace uses an
Local Radius of Curvature Map is typically the best source algorithm that uses a reference sphere based on the best
of information for such corneal shape information. fit to the central cornea. The Z Elevation Map illustrates
The Refractive Map (Figure 7-6) accurately calculates depressions and elevations from a best-fit sphere. The user
the refractive power across the entire corneal surface as can modify the reference sphere to obtain more detail in
it applies Snell’s Law of Refraction instead of the basic some cases. Without providing a reference sphere, the abil-
keratometric formula, which once again is only correct for ity to discern the very subtle height differences across the
the limited central corneal region. The Refractive Map is cornea would be difficult. Because the cornea is generally
always presented in diopter units. The Refractive Map pro- aspheric, there will always be some differences in the height
vides the clinician with appreciation for the greater refrac- from a sphere. Elevation mapping is critical for planning
tive power of the normal cornea in the periphery than in the excimer laser enhancement surgeries, including surgical
center despite the fact that the normal cornea does flatten correction of decentered ablations, and in studying corneal
in the periphery. This map is an accurate source of data to disease progression.
truly understand corneal optics and their contribution to the The corneal wavefront (CW) map (Figure 7-8) is an
total ocular power. extension of wavefront analysis of the eye applied strictly
170 Ch a p t e r 7

Figure 7-7. Z elevation map in the same patient with kerato- Figure 7-9. Verification display including simulated keratom-
conus. etry, spherical aberration values, and corneal astigmatism
(delta) value to aid IO L selection.

capture occurs through triangulation of a low-power laser

beam reflected off of the corneal apex at a calibrated dis-
tance from the Placido. Once captured, the Placido image
is displayed with the ring edge detection highlighted to
confirm correct image processing. This helps avoid arti-
facts and errors in processing that can sometimes be prob-
lematic in very irregular corneas, such as those following
corneal transplantation. Figure 7-9 shows a typical Placido
image capture verification screen with the edge detection
clearly shown confirming correct analysis. Keratometry
and refractive power readings are also displayed following
image capture.
Figure 7-8. CW map showing coma in the same patient with
keratoconus.
The Corneal Topography Summary Display can be cus-
tomized to allow for up to 4 of any of the available color
maps to be displayed. The iTrace is generally configured to
show the Axial, Local Radius of Curvature, Refractive, and
to the cornea. It is a Zernike calculation of the wavefront
Z Elevation. The example in Figure 7-10 shows significant
errors generated by the corneal surface. This is useful
keratoconus.
in understanding the cornea’s contribution to the overall
Within the displays, scaling is also customizable.
aberrations in the eye. Subtracting the CW contributions
Autoscaling features or user-defined fixed scaling units
from the entire ocular wavefront (OW) measurement to
may be used. Typically, 0.5-D step sizes are used to provide
elucidate aberrations generated in the internal eye struc-
good corneal detail while also providing adequate range.
tures is a powerful diagnostic tool. This Wavefront Map is
User choice is available to generate practically all scaling
described in micron units of wavefront error and must not
requirements.
be confused with corneal height determination. There is a
An example of the Z Elevation Map in 3 dimensions is
direct correlation of the CW Map, which only applies to the
shown in Figure 7-11. This effect can be applied to all maps
area of the cornea within the entrance pupil, to that of the
and is included in the summary display.
Refractive Map in the same corneal area.
A Comparison Display reveals differences that occur
over time. This may be useful in patients with contact lens
distortion preoperatively (Figure 7-12) or to monitor treat-
Cor n ea l Top ograp h y ment (Figure 7-13). The difference map is overall green,
Ver ificat ion a n d An a lysis indicating little or no change in corneal topography.
The caliper tool can be used to determine the axis of
Disp lays toric IOL when marked prior to surgery and scanned. The
landmarks of the eye can be used to indicate the proper
The iTrace system captures the Placido image automati-
placement (Figure 7-14).
cally upon centering and focusing of the video image as the
patient fixates upon a coaxial fixation point. This automatic
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 171

Figure 7-10. CT summary display included for maps and Figure 7-13. Corneal topography comparison display of higher
commonly used indices. This is an example of a patient with order aberrations in a LASIK patient prior to (top) and following
keratoconus. (bottom) dry eye treatment. The improvement in the inferior
portion of the cornea was significant enough to bring the
uncorrected acuity from 20/30 to 20 /20.

Figure 7-11. Three-dimensional elevation map. All maps may

be viewed in 2 or 3 dimensions.
Figure 7-14. Use of the caliper tool. This may be used in sur-
gery in patients with high astigmatism by marking the cornea,
capturing the image, and locating the proper axis relative to
the horizontal and vertical meridian. Alternatively, conjunc-
tional landmarks are used.

Ab errom et ry/ Wavefron t W it h

t h e iTrace Ray Tracin g Prin cip le
Historically, the optical property of the eye has been
described in terms of a single group as a simple refraction.
Refraction can be measured with a variety of instruments:
retinoscopy, refractometry, and autorefraction. However,
these measurements represent only spherocylindrical mea-
Figure 7-12. Comparison display for a keratoconic patient
surements of the aberrations in the eye. To measure the
who had underworn his gas permeable lenses. Immediately
after removal, the cone was larger. After cessation of lenses,
spatially resolved optical properties of the eye, we must
the vision improved to 20 /20 best corrected with improved use another terminology to describe these properties. One
topography. method is by characterizing the wavefront or the total aber-
rations of the eye’s optical system.
172 Ch a p t e r 7

Figure 7-17. Representation of a defocused wavefront from

Figure 7-15. Plane and spherical wavefronts. (Reprinted with the ideal.
permission of Joe S. Wakil, MD.)

A
Figure 7-16. Wavefront for myopic optical system. (Reprinted
with permission of Joe S. Wakil, MD.)

A wavefront is an imaginary surface joining all points

in space that are reached at the same time by a lightwave
propagating through a medium. The solid, vertical red line
in Figure 7-15 depicts such a wavefront. Parallel light enter-
ing the eye would have a flat or plane wavefront. Each point
in the lightwave arrives at the imaginary surface in front of
the eye at the same time.
For the human emmetropic eye, the wavefront is curved.
When light travels different distances through variable B
media at different speeds, the imaginary surface that Figure 7-18. (A) Irregular wavefront results from the human eye
represents the points reached at the same time is curved. because it is not purely spherical or aspheric. (B) Variations in
For a myopic eye of identical length, the wavefront would refractive power at each point over the entrance pupil can also
be more curved secondary to the greater refractive power, be demonstrated as an irregular wavefront of the eye in both
as drawn in Figure 7-16. 2- and 3-dimensional maps.
The difference between an ideal wavefront and the actu-
al wavefront over the entire surface of the optical system
is calculated in terms of microns of deviation. When the The eye, however, is rarely described by a purely spheri-
actual wavefront is more advanced than the ideal wavefront, cal or aspheric surface. The resulting wavefront error is
the error is described as a positive deviation in microns. If irregular in deviation pattern as depicted in cross-section
the actual wavefront were retarded, the deviation would in Figure 7-18. Variations in refractive power at each point
have a negative value. This spatially resolved deviation of over the entrance pupil can also be demonstrated as an
the measured wavefront from the ideal as a 2-dimensional irregular wavefront of the eye in both 2- and 3-dimensional
color map and a 3-dimensional display is shown for the maps.
defocused wavefront in Figure 7-17.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 17 3

Figure 7-19. Visual rep-

resentation of a Zernike
polynomial. (Reprinted
with permission of
Sonya Dakin.)

The deviations between the actual wavefront and ideal

wavefront are referred to as aberrations. In 1934, Fritz
Zernike published a paper describing a set of polynomials
that could be used to expand the aberration function. Each
polynomial represents a particular type of optical aberration.
Each wavefront can be described by mathematical coef-
ficients that, when taken as a whole, reconstruct the wave-
front map as shown in Figure 7-19, which shows the Zernike
modes for an expansion through 6th order polynomials.
There are 4 technologies commercially available to
measure aberrations in the eye. These include: Hartmann-
Shack, Tscherning, differential skiascopy, and ray tracing.
Figure 7-20. Schematic layout of ray tracing aberrometer.
Tracey Technologies uses the ray tracing principle that
(Reprinted with permission of Joe S. Wakil, MD.)
projects a sequential series of thin laser beams through
the entrance pupil parallel to the eye’s line of sight. The
location of each beam of light focused onto the retina is location in the entrance pupil and the reflected location
measured by capturing the exiting reflected light and focus- from the retina. This means that highly aberrated eyes can
ing it onto position-sensing detectors. Ray tracing measures be easily measured with ray tracing. Second, because the
the forward aberration of light passing into the eye. The pattern of laser spots projected through the entrance pupil
Hartmann-Shack principle assumes an ideal point source of is rapidly controlled with software, the system can track the
light is generated from the retina, therefore measuring the pupil size and project all 256 points into a pupil as small
reverse aberrations of light as it passes from the retina out as 2 mm or as large as 8 mm. Third, because each point
the exit pupil of the eye. Measuring forward aberrations is is measured separately, the software’s task of locating the
more physiologic in analyzing vision as it follows the natu- center of each spot is much easier; therefore, processing
ral path of light into the eye. is very fast, requiring only basic computer power, which
Ray tracing has additional advantages over other tech- helps allow for a cost-efficient system. Measurement has
nologies. First, the rapid sequential capture of data means successfully been performed in children with amblyopia,
that there is no confusion in the analysis between the origin
174 Ch a p t e r 7

Figure 7-21. Shift of retinal spot

location based on local aberrations
of the eye. Myopia is described on
the left; hyperopia on the right.
(Reprinted with permission of Joe S.
Wakil, MD.)

where higher order aberration (HOA) quantification aids

diagnosis.3
Figure 7-20 is a schematic layout of the ray tracing tech-
nique developed by Tracey Technologies. Once the position
of point 1 is determined, the laser beam is moved to a new
position, and the location of the next point on the retina is
then determined. This process continues until 256 separate
points have been projected through the entrance pupil,
which occurs in approximately 100 msec. If the eye were
emmetropic, then all 256 points would fall on one spot in
the center of the macula as shown in Figure 7-19. Generally,
local aberrations at the beam’s entry point on the cornea or
the lens cause a shift in the location on the retina. Figure
7-21 illustrates a single measurement point for a myopic and
hyperopic eye. After a series of points are projected sequen-
tially through the entrance pupil, the retinal spot pattern is
created (Figure 7-22).

Ad va n t ages of Com b in ed Figure 7-22. Schematic eye and retinal spot pattern illustrating
the resulting spot pattern (256 points). (Reprinted with permis-
Wavefron t / Cor n ea l sion of Joe S. Wakil, MD.)
Top ograp h y Disp lay
The iTrace system measures total ocular aberrations formula across the cornea’s surface. By subtracting the
of the eye (aberrometry) using ray tracing and calculates corneal aberrations from the total ocular aberrations of the
corneal anterior surface aberrations from the corneal topog- eye, the aberrations of the internal optics (primarily the
raphy data acquired through the Placido image. This cal- lens) are generated. This is a good first approximation of
culation involves simply performing a Zernike polynomial aberrations attributed to the internal optics of the eye. This
computation on the data generated from the refractive map also provides clinical insight into the optical relationship
of the cornea using the classic Snell’s Law of Refraction between the crystalline lens and cornea.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 175

Figure 7-23. Case example of a keratoconic patient with irregu- B

lar astigmatism. Note the wavefront pattern for the entire eye
(upper right) closely resembles the pattern of the CW (lower
left), suggesting the majority of the aberrations are due to the
cornea.

Locating the sources of astigmatism is a prime example

of the power of the combined Wavefront and Corneal
Topography Display. In Figure 7-23, a keratoconic eye
is shown. The upper right map provides the whole eye
wavefront and clearly shows irregularity, which closely
resembles the CW map (lower left). This indicates that the
majority of the visual disturbance is found on the cornea.
Figure 7-24 depicts an eye with an iris cyst superiorly Figure 7-24. Case example of a patient with iris cysts causing a
tilting the crystalline lens and inducing astigmatism. Here, tilt of the crystalline lens, and refractive astigmatism. The cyst
the CW is green, indicating a regular cornea while the inte- is easily seen on the eye image. The combination topography
rior optics reveal coma secondary to lens tilt. and wavefront display suggests the astigmatism is not on the
Figure 7-25 is an example from a patient who underwent cornea but rather due to the tilting of the lens secondary to the
cyst. Note the green CW, indicating few aberrations generated
implantation of a ReStor presbyopia-correcting lens (Alcon
on the cornea.
Laboratories, Fort Worth, TX) and suffered from glare and
mild vision loss. The clinical picture at the slit lamp did not
indicate a problem, and the surgeon was perplexed as to the
etiology of the patient’s complaint. The iTrace identified
coma resulting from tilting of the ReStor lens.
Figure 7-26 illustrates the advantage of the iTrace with
regard to refractive surgery candidacy. The combination
of wavefront and topography identifies those with early
ectasia who should not undergo LASIK. Coma on CW may
indicate early keratoconus or, in this case, pellucid marginal
degeneration. Clinical exam and pachymetry are used to
rule out lens warpage and dry eye, which also manifest as
inferior steepening.
The iTrace uniquely demonstrates changes in the lens
during accommodation.4,5 Using the open field window, the
patient fixates on a distance target binocularly or monocu- Figure 7-25. Lenticular coma resulting from a tilted ReStor lens
larly. The measurement is then repeated while looking at a (Alcon, Fort Worth, TX). (Reprinted with permission of Joe S.
near point target, such as a reading card using an attached Wakil, MD.)
near-point reading rod. The difference map represents the
accommodative change of the eye. Accommodation tradi-
tionally calls for a spherical change in the lens power across to assist the patient’s visual processing to enhance near
the pupil, but in many cases the iTrace demonstrates that vision. New studies are being performed to understand
cylinder as well as HOAs are induced by the lens, which pro- these mechanisms in natural accommodation and accom-
vide an increased depth of field (pseudo-accommodation) modating IOL technology. Figure 7-27 shows the refractive
176 Ch a p t e r 7

Figure 7-26. Moderate coma in a patient presenting for LASIK.

Coma on CW may indicate early ectasia, and contact lens
overwear dry eye should be treated prior to proceeding with
LASIK. Should the asymmetry fail to resolve without treatment,
LASIK may be ill advised.

Figure 7-28. A nuance of the system is the ability to measure

aberrometry through spectacles. This allows an objective eval-
uation of the glasses and is extremely helpful when patients
with early cataracts complain about their spectacles being
incorrect. This display may be used to educate the patient to
the location of the visual problem.

The ability to reveal corneal aberrations separately from

B those of the lens combined with the ability to measure
a patient’s ocular aberrations at true far- and near-point
fixation is providing a new level of diagnostic information
on how the eye accommodates and how well we are able
to return such function through either accommodating or
multifocal implants.
An additional nuance unique to the iTrace is the ability
to perform objective evaluation of the spectacle correc-
tion. This is valuable when patients present reporting dis-
satisfaction with the glasses. The Visual Function Display
(Figure 7-28) can be used to educate patients that they are
dissatisfied with their glasses due to lenticular changes that
cannot be corrected with spectacles. In this example, the
Figure 7-27. (A) The iTrace unit with a reading rod attached. refractive error is negligible while the internal aberrations
(Reprinted with permission of Joe S. Wakil.) (B) Difference are significant. This patient was 20/20 in each eye, but
map of near and far measurements showing accommodative
mild cataracts reduced the vision subjectively. The patient
response.
presented upset about the quality of vision, but left with an
understanding that only cataract surgery would improve the
change from far (A) and near point (B) measurements. Note vision. Measurements can be obtained over gas-permeable
the change in lenticular aberrations provides almost 2 D of lenses as well to demonstrate the benefit of this modality
accommodation for this hyperopic patient. While we expect upon vision.6
changes in the lens aberrations during accommodation, the The iTrace can also be used to fit contact lenses.
classic model of accommodation does not explain the dra- Wavetouch hydrogel lenses are fitted, and then the patient
matic shift of 90 degrees in the lens astigmatism observed is scanned to determine the over-aberrometry measure-
in this patient during accommodation. ments (Figure 7-29). These are sent to the Wavetouch
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 17 7

manufacturer, and quarterly replacement lenses are created.

These can be used to correct healthy corneas with 20/20
vision as well as those with reduced vision secondary to
irregular astigmatism.

Con clu sion

The power of the iTrace to measure the refraction of the
total eye through ray tracing aberrometry and integrating it
with the corneal aberrations derived directly from the cor-
neal topography measurement provides tremendous diag-
nostic information to enhance vision correction and disease
diagnosis. The iTrace represents a new platform of refrac-
Figure 7-29. WaveTouch hydrogel lenses can be fit to correct
aberrations measured over the lens and incorporated into the
tion diagnosis to understand a patient’s quality of vision and
lens prescription. Lenses are traditionally fitted, and then the to pinpoint the source of aberrations in the eye.
patient is measured using the iTrace to determine the appropri-
ate prescription using aberrometry. These parameters are used
to build a quarterly replacement lens.

TH E N IDEK OPD SCAN II: A C OMPREH ENSIVE D IAGNOSTIC AND

P LANNING P LATFORM FOR I NTRAOCULAR AND R EFRACTIVE SURGERY
Phillip M. Buscemi, OD; Harkaran S. Bains, BSc;
Murray McFadden, MD, FRCSC, DABO ; and Katherine E. Paton, MD, FRCSC, DABO

C
orneal topography was the first technology that severely aberrated eyes, and refractive wavefront maps that
objectively allowed ophthalmologists to perform separate corneal from lenticular aberrations.7
pre- and postoperative assessment of the cornea, Additional software programs are available for use with
which contributes the greatest magnitude of HOAs to the the OPD Scan family of instruments that allow enhanced
eye. While this technique was an indirect assessment, clini- diagnostic ability, visual quality simulation, and IOL
cal wavefront aberrometry enabled direct measurement of power calculations. For example, corneal disease screening
the optical effects of intraocular surgery and corneal sur- software using the Corneal Navigator detects keratoconus,
gery. Combining these 2 techniques allows greater analysis keratoconus suspect, pellucid marginal degeneration, and
for preoperative planning and postoperative assessment of postsurgical corneas. Modulation transfer functions (MTFs)
vision. For example, the separation of internal aberrations of the ocular, corneal, and internal and HOAs allow the
(mainly lenticular) from corneal aberrations allows the evaluation of visual performance before and after surgery.
surgeon to determine whether lenticular or surface aberra- The IOL Station software has recently been introduced for
tions are the source of visual blur. Isolation of the source IOL power calculations and for selection of spherical and
of the aberrations enhances surgical planning and provides aspheric IOLs for virgin corneas and postrefractive surgery
greater diagnostic insight. cases.
The Nidek OPD Scan II and the new OPD Scan III The combined use of the OPD SCAN II and OPD Station
combine aberrometry and corneal topography to allow the will be described here with attention to clinical examples
separation of corneal front surface aberrations and internal that are commonly encountered. Explanation of the full
aberrations. Corneal topography is measured using Placido clinical utility of the data and exam sets available would
disk technology. Measurement of ocular (whole eye) aber- occupy several chapters. For example, the OPD Scan II
rations is performed with spatial dynamic skiascopy, a plots 20 different maps; the OPD Station plots an addi-
technique that allows a wide dynamic range of measure- tional 30 maps. We have therefore limited the examples to
ment of spherical refractive error ranging from -20 to those that are clinically pertinent to a general ophthalmic
+22 D and ±12 D of astigmatism, the ability to measure practice.
178 Ch a p t e r 7

TABLE 7-1.

Specifications of the OPD-SCAN II

Spherical power -20 to +22 D
Cylinder power ±12 D
Axis 0 to 180 degrees
Wavefront data points 1440
Zernike order 8th order (selectable)
Aberrometry maps Distribution of refractive power due to wavefront (D)
Wavefront distortion based on Zernike functions (µm)
Pupil size 6 mm
Corneal topography data points >6800 points
Topography maps Axial, instantaneous, refractive, elevation, difference
maps
Accommodation control Autofog using simulated scenery
Pupillometry Photopic and mesopic diameters
Infrared light
Eye centration landmarks Visual axis (coaxially sighted corneal light reflex)
Photopic and mesopic line of sight
Photopic angle kappa
Mesopic angle kappa

Feat u res The OPD Scan III offers higher-resolution aberrometry

and other improvements compared to the OPD Scan II.
The OPD Scan II is a combination aberrometer, corneal The OPD Scan III features higher-resolution topography
topographer, autorefractor, autokeratometer, and pupillom- with 33 rings and 11,880 data points versus 19 rings and
eter. All measurements are performed on the same optical 6840 data points currently available on the OPD Scan II. In
axis without moving the patient. This allows direct cor- addition, the mire illumination has been changed to blue,
relation of ocular, corneal, and internal aberrations of the which dramatically reduces pupil artifacts due to constric-
eye. The technical specifications are presented in Table 7-1. tion during measurement. The aberrometry functions have
Important features include the small footprint similar to a also been enhanced in the OPD Scan III. The wavefront
conventional autorefractor; a large cone Placido disk that measurements are currently performed for a 6-mm pupil
is less sensitive to eye movement and has a larger range diameter, which will be increased to 9 mm (user selectable)
of focus than small cone Placidos; an internal eyetracker with the OPD Scan III. Additionally, the number of data
and autoshot function to reduce the effect of saccadic eye points used for measuring and plotting wavefront maps has
movements during measurement; infrared aberrometry increased from 1440 currently to 2520 data points along
measurements that reduce the need for dilation; and approx- with 20% faster acquisition times with the OPD Scan III.
imately 5 times more wavefront data points compared to Other new features in the OPD Scan III specifically for
most Hartmann Shack aberrometers. The measurement cataract surgery include retroillumination image capture
modes can be selected by the user to perform only one for lens opacity and average pupillary power (APP) for
specific measurement, such as autorefraction, autokeratom- IOL calculations. Average pupillary power is the average
etry, or corneal topography based on the clinical workup. power within the 4-mm central cornea calculated using
Physiologic dark-adapted pupil measurements are pre- 5 to 7 Placido mires that comprise approximately 2000 data
ferred, and if pupil dilation is required, we recommend the points. Average pupillary power may serve as a more accu-
use of 2.5% phenylephrine to avoid any cycloplegic artifact rate reflection of corneal power for IOL calculations com-
in measuring wavefront data. pared to averaging the 2 principle meridians of the cornea.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 17 9

A host of other improvements including database manage-

ment and ease of service have also been implemented on the
OPD Scan III. As the OPD Scan III is a very new machine,
the patient examples in this chapter have been acquired with
the OPD Scan II.

Refract ive Wavefron t Map s

The OPD Scan II plots refractive wavefront maps that are
unique to this instrument and therefore require a detailed
explanation of the interpretation and use in clinical prac-
tice. The raw data from the spatial dynamic skiascopy are
converted to refractive wavefront maps. The OPD, Internal
OPD, and OPD higher-order maps are refractive wavefront
maps that plot the distribution of refractive power across
the pupil created by ocular aberrations. Interpretation of
the maps involves evaluation of the uniformity of color,
the symmetry of the patterns, metrics such as root mean
square, and refractive values. The difference in power from
the warmest color to the coolest color (or vice versa) on the
Figure 7-30. O PD SCAN II refractive wavefront maps of (A) the
map determines the maximal difference in refractive power
entire eye, (B) the internal (lenticular) aberrations, and (C) the
across the pupil. HOAs of the entire eye.
A number of options are available for the selection of
scale step size and color palette. Our preferred color palette
and scale for refractive wavefront maps is the “adjustable aberration can occur due to increasing cataractous changes
common scale” setting with the autostep function on and or excimer laser corneal ablation. Increasing cylinder values
the International Standards Organization (ISO) or the indicate high astigmatism or coma aberration. Postoperative
Smolek/Klyce color palette. While it is arguable that the causes of high astigmatism include decentered IOLs, tilted
normalized scale will provide the highest detail for any IOLs, IOLs placed at the incorrect axis, or a decentered
given exam (wavefront or corneal topography), when com- refractive ablation. Multicolor OPD maps, with greater
paring changes from visit to visit, the adjustable common asymmetry and root mean square (RMS) values over
scale allows direct comparison due to the consistent color 0.50 D, indicate that HOAs are likely playing a significant
palette and step size between serial exams. Based on our role in the distribution of refractive power across the pupil.
clinical experience, both color palettes are adequate for Figure 7-30B plots the internal OPD map. The inter-
clinical practice, although the Smolek/Klyce colors tend nal OPD map subtracts the effects of the corneal surface
to provide greater contrast between scale steps, enhanc- and plots the remaining refractive wavefront. As only the
ing subtle changes in refractive power within the pupil. corneal surface data are removed, the internal OPD map
The scale for these maps is plotted in diopters where light incorporates the effects of all structures behind the corneal
green indicates no refractive power, warmer colors indicate tear film and epithelium. In eyes with posterior ectasia
myopic power, and cooler colors indicate hyperopic power. or severely distorted corneas, such as in severe pellucid
The OPD map plots the distribution of refractive power marginal degeneration or postpenetrating keratoplasty, the
across the pupil, incorporating the effects of all lower- and aberrations of the back corneal surface will register and
higher-order aberrations (Figure 7-30). In Figure 7-30A, the may predominate on this map. However, in normal eyes,
OPD map has a red asymmetric bow-tie pattern, indicat- the internal OPD map can be considered an accurate repre-
ing the combined effects of lower and higher aberrations sentation of the refractive wavefront due to the physiologic,
on refraction. The difference in power across the pupil is pathologic, or pseudophakic lens. Figure 7-30B presents
approximately 1.26 D with the warmer (red) colors indi- an example with a symmetric bowtie indicating lenticular
cating myopic powers comprising the central pupil. The astigmatism. In this case, corneal astigmatism (1.76 D,
autorefraction (measured for a 2.30-mm pupil diameter), dk value on the bottom row of Figure 7-30) and lenticu-
3-mm wavefront refraction, and 5-mm wavefront refraction lar astigmatism were compensating each other, indicated
provide the effect on refractive power as the pupil dilates. by the minimal refractive astigmatism on the OPD map
Increasing hyperopia peripherally indicates increasing neg- (autorefraction cylinder value +0.25 D, Figure 7-30A).
ative ocular spherical aberration, whereas increasing myo- Figure 7-30C presents the OPD HO map. The OPD HO
pia indicates positive spherical aberration. Positive spherical map removes the effects of the lower-order aberrations such
180 Ch a p t e r 7

as sphere and cylinder, plotting the effects of the HOAs on

refractive power distribution (see Figure 7-30C). The color
changes across the map indicate that approximately 1 D
of refractive change is caused by HOAs. Comparison with
the OPD map determines whether HOAs are the source of
the refractive difference within the pupil. In this case of a
mild myope, the majority of refractive difference is due to
HOAs (1 D due to HO aberrations versus 1.26 D total dif-
ference). This finding indicates that visual complaints from
this patient cannot be addressed by reducing refractive error
alone and will require reduction or elimination of wavefront
error.
Traditional Zernike-based higher-order maps are plotted
by conversion of data from the OPD maps. Convolution
of the Zernike maps allows simulation of the MTF, point
spread function (PSF), and visual acuity of the entire eye
(ocular) or corneal and internal aberrations. Simulations
based on each individual aberration within the Zernike
table or combinations of aberrations are also possible.
Figure 7-31. Corneal navigator analysis of a keratoconus
suspect (KCS). Note the classification system (lower left).
Clin ica l Use NRM denotes a normal cornea, AST denotes astigmatism, KC
denotes keratoconus, PMD denotes pellucid marginal degen-
Peer-review publications have reported the use of OPD eration, PKP denotes penetrating keratoplasty, MRS denotes
Scan II in refractive surgery,8 corneal disease screening,9,10 myopic refractive surgery, HRS denotes hyperopic refractive
alignment of toric IOL and implantable collamer lens,11,12 surgery, OTH denotes a classification other that the ones listed
and the study of aberration compensation after refractive above.
surgery.13 Comparison of OPD Scan II refractions to mani-
fest refraction have shown differences less than 0.35 D in
sphere and 0.25 D difference in cylinder.14 Repeatability Although the incidence of keratoconus is extremely low
and reproducibility of the total, corneal higher-order and in the general population, keratoconus suspects and patients
internal wavefront (IW) measurements easily surpass the with frank keratoconus present at refractive surgery cen-
tolerances required for clinical practice.15 Bland-Altmann ters at a much higher rate for alleviation of their refractive
plots indicate that the dark-adapted mesopic pupillometry error. Advanced cases of keratoconus and pellucid marginal
measurements with the OPD Scan II are consistent with degeneration have distinct corneal topographic features that
the scotopic pupil diameter measured with the Procyon enable quick diagnosis. However, cases with suspicious cor-
and Colvard pupillometers.16 The data from these studies neal topography not indicative of advanced corneal disease
clearly indicate that the OPD Scan II can be used for a vari- remain a clinical conundrum.
ety of diagnostic purposes with acceptable accuracy and Figure 7-31 outlines a case of a patient who presented
reproducibility for routine clinical practice. for a refractive evaluation with corneal features that were
consistent with a keratoconus suspect classification. The
Corn eal Disease Screen in g features of this topography indicate that the left cornea
is classified as a keratoconus suspect. The right eye of
Corneal disease screening represents the major appli- this patient was classified as normal. The lack of mirror
cation of corneal topography. The OPD Scan II includes image symmetry between corneas at presentation, clini-
the axial, instantaneous, refractive, and elevation corneal cal examination, serial corneal pachymetry, and long-term
topography maps. Additionally, the Corneal Navigator cor- follow-up did confirm progression to keratoconus. The
neal disease screening module in the OPD Station software Corneal Navigator uses a color scheme where green is nor-
employs a neural network for comparison of corneal charac- mal, yellow values warrant further investigation, and red
teristics to a database of normal, diseased, and postsurgical indicates an abnormal value of the various indices based on
corneal topographies. Based on the comparison, the cornea neural network analysis. In this example of a keratoconus
under measurement can be classified as normal, keratocon- suspect, the surface asymmetry index (SAI) is high, which
ic, a keratoconus suspect, pellucid marginal degeneration, is often associated with keratoconus (see Figure 7-31).9 The
and postmyopic refractive surgery among other classifica- interested reader is directed to a comprehensive review by
tions. Klyce and colleagues9 on the clinical interpretation of all
the indices presented with the Corneal Navigator.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 181

In t rao cu la r Su rger y
Preop erat ive Plan n in g
Refractive surgery is rapidly incorporating IOL surgery
(phakic and toric lens implantation). This shift in attitude
is due to patient expectations, the increased accuracy of
IOLs, and the introduction of wavefront IOLs. For example,
aspheric IOLs are designed to compensate for corneal
spherical aberration that addresses the loss in visual quality
after spherical IOL implantation. Aspheric accommodating
IOLs incorporate wavefront principles to allow better visual
performance than standard accommodating lenses. Hence,
the need for wavefront assessments is rapidly becoming a
clinical necessity for many cataract surgeons and general
ophthalmologists.
Aspheric IOLs are implanted based on the compensation
of corneal and aspheric IOL spherical aberration. However, Figure 7-32. Internal O PD maps of the normal (top left) and
the selection of the correct aspheric IOL requires measure- cataractous eyes.
ment of the corneal spherical aberration. The distribution
of corneal spherical aberration across a population is
widespread, following a bell-shaped curve.17 Additionally, present with emmetropia or mild myopia centrally sur-
differences in HOAs across ethnic groups has been report- rounded by hyperopia (Figure 7-32). Cataractous eyes
ed.17 Hence, implanting one type of aspheric IOL in all present with increasing myopia anywhere on the internal
patients is unrealistic. For example, significantly increasing map along with flecks of power randomly distributed on
negative ocular spherical aberration in a patient with pre- the internal OPD map (see Figure 7-32). Progression of
existing negative corneal spherical aberration may result the cataract can be followed with serial measurements and
in a diminution of visual quality akin to keratoconic eyes comparison of the internal OPD maps.
or posthyperopic ablation, both of which can cause high The axial corneal topography map, keratometry, and
negative spherical aberration. The aspheric IOL needs to be autorefraction from the OPD Scan II will assist the surgeon
tailored to the individual’s preoperative corneal spherical in planning the placement of the phaco incision, limbal
aberration. The OPD Scan II provides the corneal spherical relaxing incisions, and toric IOLs and toric phakic IOLs.
aberration values for aspheric IOL selection calculated from
the corneal topography. Postop erat ive Evalu at ion
Baby boomers are the largest subset of the population Postoperatively, the OPD Scan II can be used to evaluate
to undergo laser in situ keratomileusis (LASIK) or pho- toric IOL alignment, tilt, and torque. Serial corneal topog-
torefractive keratectomy (PRK) worldwide. As this demo- raphy will allow the assessment of corneal wound healing
graphic ages, the need for accurate postrefractive surgery and of the effect on corneal astigmatism over time.
IOL calculations will increase, along with the need for a In cases of refractive surprise or dysphotopsia post-
selection of specific aspheric monofocal IOLs or presbyopic implantation, the effect of an incorrectly placed limbal
IOLs. Hyperopic LASIK tends to induce negative spherical relaxing incision, phaco-incision, or IOL can be determined
aberration, and myopic ablations tend to induce positive by comparing corneal topography, the OPD map, and
spherical aberration. The magnitude of induction varies the internal OPD. This enables the surgeon to determine
based on aspheric or conventional ablation algorithms.18 surface versus internal sources and plan the appropriate
The distribution of residual postoperative corneal spheri- retreatment. Figure 7-33 presents a case of a patient who
cal aberration in refractive surgery patients will likely be complained of shadowing and ghosting after IOL implan-
abnormal, leading to the introduction of more aspheric tation. Slit-lamp examination indicated a well-centered
IOLs. Corneal topography to measure corneal spherical IOL in the capsular bag. OPD Scan II measurement shows
aberration and postoperative keratometry will be instru- the corneal topography was normal, yet the internal OPD
mental in patients who have undergone LASIK and PRK. maps show myopic and hyperopic areas bisecting the pupil,
Additionally, postrefractive surgery IOL calculations are creating 2 focal planes on the retina resulting in ghosting.
available on the Nidek IOL Station for such cases with the The 2 refractive planes were due to a subtle tilt of the IOL,
Camellin-Callosi formula. resulting in coma, a common cause of ghosting and diplopia
Cataracts cause typical patterns on the internal OPD (see Figure 7-33).
map. For example, the phakic eye of a young adult will
182 Ch a p t e r 7

Figure 7-33. Measurement of a patient with a tilted IO L.

Figure 7-34 presents a 1-page summary recommended

for cataract surgery by one of the authors (MM). This
88-year-old male patient presented with 20/40 pinhole
acuity and a 2+ nuclear sclerotic cataract. The cornea was
classified as “normal” by the corneal navigator screening
software. The axial topography was unremarkable with a
difference in keratometry (dk value) of 0.34 D, indicating
mild corneal astigmatism (see Figure 7-34). The preop-
erative dk astigmatic axis is useful intraoperatively as it is
frequently easy to shift the incisional axis to reduce corneal
astigmatism. Pre- and postoperative keratometry values are
valuable during implantation of toric lenses and evaluation
of both the keratometric and real total visual outcomes (see
Figure 7-34).
The Zernike/OPD graph plots the magnitude of each
ocular lower- and higher-order Zernike aberration in µm
Figure 7-34. Preoperative O PD Station assessment of an
(see Figure 7-34). This value theoretically represents the 88-year-old patient with a cataract. Map on the top left is
deviation from an individual with perfect vision. We have an axial corneal topography map. The figure on the top right
elected to present this example with a real-world mesopic lists Zernike wavefront aberration values for the whole eye.
pupil diameter of 3.70 mm as pharmacologically dilated Aberration number 12 is ocular spherical aberration. The O PD
pupils are not representative of the conditions this patient map (middle row, left side) is a refractive wavefront map plot-
functions under for daily living activities (see Figure 7-34). ting the refractive distribution due to the lower- and higher-
Furthermore, miotic pupils are often encountered in this order aberrations of the whole eye. The internal O PD map
segment of the population; hence, peripheral aberrations (middle row, right side) is a refractive wavefront map plotting
the refractive distribution due to the lower- and higher-order
from an artificially dilated pupil are clinically irrelevant.
aberrations excluding the front corneal surface. In this case,
The OPD map is flagged “High RMS,” indicating sig- aberrations are due to a cataract. The map on the bottom left
nificant visual problems likely due to optical aberrations, is the PSF, which plots the retinal image quality of the ocular
with the differences in color indicating greater than 6 D HOAs. The figure at the lower right lists Zernike values for
difference across the pupil (see Figure 7-34). The internal corneal aberrations. Aberration number 12 is corneal spheri-
OPD map shows a pattern similar to the OPD map and not cal aberration. All Zernike aberrations are reported for the
the corneal topography, indicating an internal (lenticular) patient’s mesopic pupil diameter of 3.7 mm out to the 6th
cause of the optical aberrations (see Figure 7-34). The high Zernike order. Spherical Aberration correcting IO Ls are calcu-
central myopic area and the high internal cylinder values lated at a 6 mm lens diameter and the corneal mires are not
compromised by a reduced pupil diameter.
(4.34 D) are commensurate with cataract formation. These
data confirm that the primary visual problem is cataractous
in origin. The internal OPD RMS is 5.15 D, indicating Figure 7-35 presents the postoperative OPD document-
significant internal aberrations, which not surprisingly are ing the remarkable optical transformation from elective
affecting vision. The PSF simulates the retinal image qual- cataract surgery with implantation of a Tecnis ZCBOO
ity with best-corrected vision. The corneal Zernike values aspheric IOL (Abbott Medical Optics, Santa Ana, CA). The
indicate +0.427 µm of corneal spherical aberration, which axial topography map indicates the corneal astigmatism
aids in the selection of the best aspheric IOL for this patient remained unchanged, despite a 2.75-mm temporal incision.
(see Figure 7-34). The ocular Zernike graph plots a marked diminution of all
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 18 3

Figure 7-36. MTF/O PD/Tot difference plot. The area ratios on

the graph represent the ratio between the curves; A/B is the
ratio between the average curve and the best MTF curve; A/D
indicates the relation of the average curve and the diffraction
limit; H /B indicates the ratio of the HOAs and the best MTF
curve; H /D indicates the ratio between HOAs curve and dif-
fraction limit.

0.201 µm compared to a preoperative ocular higher-order

wavefront error of 0.944 µm (see Figure 7-35).
Figure 7-36 plots the preoperative and postoperative
MFT graphs, which allow an assessment of visual perfor-
Figure 7-35. Postoperative O PD Station assessment of an eye
that underwent aspheric IO L implantation. Map on the top mance after IOL implantation. The H/B ratio indicates the
left is an axial corneal topography map. The map on the top area under the best-corrected curve of the patient compared
right is a Zernike graph of the ocular (whole eye) aberrations. to a subset of emmetropes with excellent visual quality (see
Aberration number 12 is ocular spherical aberration. The O PD Figure 7-36). In this case, the H/B ratio has increased from
map (middle row, left side) is a refractive wavefront map plot- 17.9% preoperatively to 66.8% postoperatively, indicating
ting the refractive distribution due to the lower- and higher- the significant increase in visual performance after IOL
order aberrations of the whole eye. The internal O PD map implantation (see Figure 7-36). The patient’s postoperative
(middle row, right side) is a refractive wavefront map plotting uncorrected Snellen acuity remains 20/25+ to date, and he
the refractive distribution due to the lower- and higher-order
is satisfied with the outcome of surgery.
aberrations excluding the front corneal surface. In this case,
the visual complaint is due to the IO L. The map on the bot- The combination of corneal topography and aberrometry
tom left is the PSF, which plots the retinal image quality of the in the OPD Scan II allows thorough assessment, diagnosis,
ocular HOAs. The map on the lower right is a Zernike graph and management of elective lens surgery. Additionally, the
of the corneal aberrations. outcomes of intraocular surgery can be superbly document-
ed. It is anticipated that surgeons will be able to use the pre-
operative measurement of corneal spherical aberration and
lower and higher optical aberrations including a change postoperative measurements of total aberration to develop
in spherical aberration from -0.063 µm preoperatively to individual nomograms that will allow them to deliver more
+0.02 µm postoperatively (see Figure 7-35). The OPD map accurate outcomes to patients.
documents a postoperative autorefraction of 0.00 -0.50 x
71 and an RMS value of 0.46 D, indicating a significant
reduction in aberrations across the pupil. The internal OPD Refract ive Su rger y Screen in g
indicates the optical power of the implanted IOL with this
eye’s new optics and an RMS deviation of 1.27 D versus Wavefront measurements are commonplace during
5.15 D, preoperatively (see Figure 7-35). The PSF indicates LASIK and PRK consultations. The OPD Scan II mea-
the excellent best-corrected retinal image quality postop- sures refraction, the pre-existing corneal and lenticular
eratively and a final ocular higher-order wavefront error of aberrations, and the preoperative retinal image quality and
184 Ch a p t e r 7

simulates potential visual performance after treatment of

HOAs. Recent findings with different excimer platforms
indicate patients with lower magnitudes of preoperative
(0.30 µm or less) ocular HOAs tend to have greater induc-
tion of HOAs postoperatively after wavefront-guided abla-
tion.19,20 However, the interaction of aberrations is also
a determinant of visual quality. Based on these findings,
a prolate ablation or wavefront ablation can be planned
based on the magnitude of HOAs preoperatively. Using
the OPD Scan II, patients can be screened for corneal or
lenticular contraindications. Additionally, selection of the
correct laser algorithm is possible for candidates undergo-
ing excimer surgery.
Figure 7-37. Pupillography with the O PD SCAN II presenting
Wavefront-guided ablation has reignited the controversy a hyperopic patient with a large angle kappa. The pink circle
regarding centration of the laser ablation. Initially con- denotes the photopic pupil shape, PDist denotes the distance
sidered a problem only in hyperopic patients due to angle between photopic pupil center and the optical axis, MDist
kappa, the benefits of centration closer to the visual axis denotes the distance between the mesopic pupil center and
in myopes have been documented by 2 recent studies.21,22 the optical axis. The pink and blue crosses represent the phot-
Better optical quality by centering on the “visual axis ” in opic and mesopic pupil centers.
myopes has been reported.21,22 Currently, the OPD Scan II
is the only unit with an automated, repeatable method of
measuring the difference between the line of sight (pupil over the entire mesopic pupil. The corneal surface data are
center) and the “visual axis” (coaxially sighted corneal used to create a compensation matrix that addresses the
light reflex) and transferring these data to an excimer laser loss of excimer energy as the ablation moves peripherally.
(Figure 7-37). The measurement of difference in pupil cen- The ocular spherical aberration is used to create a target
ter (either mesopic or photopic) and visual axis is possible spherical aberration value postoperatively while maintain-
with this unit due to the combined corneal topography and ing physiologic corneal asphericity. Without topographic
aberrometry measurements on the same optical axis. data, an accurate asphericity target cannot be calculated. In
Custom ablation treatment of normal and postsurgical a prospective, contralateral, randomized study of 32 myopes
eyes requires meticulous measurement in order to negate who underwent LASIK with OPA in one eye and conven-
spurious data that may be incorporated into treatment tional ablation in the fellow eye, the refractive outcomes
planning. One method to mitigate measurement errors is were similar between groups. However, the best-corrected
to repeat measurements and average all maps. The OPD visual acuity, postoperative HOAs, spherical aberrations,
Station incorporates an averaging function that averages up and contrast sensitivity favored the OPA group.23 The
to 10 measurements to yield a composite map. The option authors attribute these outcomes to the postoperative prolate
to exclude maps of lower measurement quality is provided corneas and the inclusion of corneal topography and OW
in order to reduce the effects of inordinately high or low data.23
values. In conclusion, the combined use of corneal topography
The field of excimer laser surgery continues to advance and wavefront data aids surgeons to achieve excellent
from treating solely refraction a decade ago to incorporat- refractive outcomes following cataract surgery, phakic IOL
ing the treatment of HOAs of the cornea or the entire eye. implantation, and LASIK. The OPD Scan II is a unique
Corneal topography data and ocular spherical aberration combination corneal topographer and aberrometer that aids
from the OPD Scan II are used in optimized prolate ablation the clinical assessment, surgical planning, and postopera-
algorithms (OPA) to create a physiologically prolate cornea tive assessment of surface and intraocular surgery.

TOPO -ABERROMETRY W ITH KERATRON O NDA

Renzo Mattioli, PhD; Massimo Camellin, MD; and Nancy K. Tripoli, MA

T
he Keratron Onda (Figure 7-38) is a combined instru- (CT) and aberrations of the total OW contemporarily. The
ment, developed and produced by Optikon 2000 instrument also tracks and measures the pupil in photopic
SpA (Rome, Italy) to measure corneal topography and scotopic or mesopic conditions and performs dynamic
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 18 5

Microlens Array (MLA) designed by Optikon, with a spatial

resolution at the cornea of 128 µm. It can measure wave-
front over a pupil area up to 7 x 7 mm, within a sensor area
about 7.9 x 10.5 mm. The light source for the retinal reflex
is a low-coherence, low-power collimated beam (<30uW at
λ = 840 nm), well within safety limits.
The instrument embeds a PC of ETX type with WinXPe
and is equipped with a 7” wide screen color display with
touch-screen user interface. It can process both CT and OW
sensor images immediately following acquisition and/or
transfer them to the Keratron Scout software. This allows
tests to be printed, saved, edited, and analyzed using all the
features previously developed for Keratrons. For example,
tests can be compared over time, test repeatability checked,
Figure 7-38. The Keratron O nda. (Reprinted with permission of images edited, contact lenses designed and fit, and lasers
Renzo Mattioli, PhD.) linked. All information about corneal topography imple-
mented on previously developed Keratron models is avail-
able, including the Maloney and CLMI indices.
pupillometry. It is able to evaluate patients’ residual capa-
bility to accommodate by measuring the OW in different
conditions of accommodation.24 The name “Onda” means Cor n ea l Top ograp h y
“wave” in both Italian and Spanish.
The Onda is the latest in the successful line of Keratron a n d Wavefron t Ab er rom et r y
instruments. The Keratron has been the first reflection-type
corneal topographer that is not spherically biased to intro-
in a Sin gle In st r u m en t :
duce color-coded maps of the instantaneous (ie, local or Tech n ica l Solu t ion s
tangential) curvature (1993) 25,26 “spherical offset” height
maps and the “move axis” feature (1994) 27 and CW analysis Ret ro -Mires
(2001) 28,29 —all of which are rigorously underpinned by
an arc-step algorithm that guarantees submicron accuracy To combine topography and aberrometry in a single
measurements30,31 and high spatial resolution maps even instrument presents several challenges. First, there must be
of irregular corneas (see Chapter 5, “The Keratron Corneal a compromise between corneal coverage and the range of
Topographers” section). measurable aberrations. A good arc-step analysis of reflec-
The IW, produced by refraction through internal ocular tive topography requires that mires are equally spaced. The
structures, can be calculated as the difference between the central hole in the Dekking cone or Placido disk must be
OW, captured by aberrometry, and CW, derived from cor- very small and distant so that its reflection does not leave a
neal topography. All of these wavefronts can be represented gap in the center. In Keratrons, this hole diameter is around
by all of the analysis tools available in the Keratron Scout 6 mm at 70 mm distance from the cornea. However, aber-
software, including Zernike terms (RMS, skyline, histo- rometry requires a visual field larger than the maximum
gram, and aberrations summary) and visual simulations pupil and optics closer to the eye so that all rays of the wave-
(Visus, PSF, MTF; see Chapter 5, “The Keratron Corneal front coming back from the pupil are collected by the mea-
Topographers” section). suring optics. For this reason, topo-aberrometers usually
sacrifice central coverage in their corneal topography, as in
the pattern on the bottom left of Figure 7-39. Furthermore,
Ba sic Feat u res CT requires visible light (λ = 550 ÷ 650 nm) to optimize the
reflected mire’s contrast with the iris background, whereas
As a corneal topographer, the Keratron Onda shares aberrometry requires near infrared (λ = 780 ÷ 900 nm) so
similar features with other Keratrons: accuracy, corneal as not to induce pupil constriction, which restricts the size
coverage up to 80% to 90%, and a wide range of curvature of the acquired image.
measurement (from 0 to 130 D).32,33 Optikon handled this problem by adopting a proprietary
The aberrometry acquisition system includes a high-res- system of “retro-mires” as shown in Figure 7-39. The retro-
olution motorized system to compensate for defocus from mires (204, Figure 7-39) are positioned behind a mirror
higher than +6 D to lower than -11 D. This allows measure- (207, Figure 7-39) that splits the optical path by their differ-
ment of myopic and hyperopic patients from -20 to +10 D (at ent wavelengths, so they can fill the CT pattern in the center
vertex distance of 14 mm), although there is a proportional (Figure 7-39, bottom right). Thus, the Keratron Onda’s mire
reduction of the maximum measurable pupil at extreme val- pattern is the same as in the other Keratrons, with a uniform
ues. The OW is measured in the image space. The sensor is a step beginning in the cone center, without compromise.34
186 Ch a p t e r 7

Figure 7-40. Keratron O nda measurements on a sphere at

several Z distances, with and without ADC compensation.
(Reprinted with permission of Renzo Mattioli, PhD.)

Figure 7-39. O ptikon’s proprietary retro-mire method allows

combining Keratron topography with other measures of the
eye that can be made at the same time without compromising
the efficacy of the Keratron arc-step reconstruction. (Reprinted
with permission of Renzo Mattioli, PhD.)

Following the path behind the mirror, the OW can thus be

measured through a central hole much wider and closer
(Ø>12 mm, at about 40 mm from the cornea) than in a
standard Keratron.
Figure 7-41. The short “bivalent cone” of the Keratron O nda
Eye Position Con trol System With An alog- allows the operator to choose quickly between “near” and
to-Digital Con version Com pen sation “far,” to get either greater corneal coverage or a less invasive
approach for deep-set eyes, without needing to remove the
All Keratrons use a proprietary technique involving cone and recalibrate.
2 photocells called the Eye Position Control system (EPCS;
see Chapter 5) 35 that allows acquisition only at a preset dis-
tance. In the Keratron Onda, the acquisition of topography, distances as selected by the operator from the touch-screen
wavefront, and also pupillometry imaging may require a (Figure 7-42). One distance allows greater corneal cover-
longer time and may induce patient movement. Therefore, age, while the other is a less invasive approach that is espe-
the Keratron Onda has larger cells to widen the “OK range” cially useful when imaging deep-set eyes.
during which an image can be acquired by a factor of 6.
An analog-to-digital conversion (ADC) measurement in Aberrom etry Measu rem en t: Frequen cy
the “Z axis” (Figure 7-40) has been added. The resulting
Dem odu lation Meth od Versus Cen troids
configuration compensates for eye movements within about
0.55 mm, resulting in an error less than ±0.1 D when imag- As in all OW measurements in “object space,” a light
ing a standard 43.5-D sphere. This is illustrated by Figure source beam (λ = 840 nm) is projected into the eye to pro-
7-41, in which the violet curve shows ADC compensation, duce a retinal point reflection, which is then projected back
compared to the yellow curve, which shows no compensa- to a lenslet array. In the classic Shack Hartman method,
tion. The Z displacement measured by the ADC is recorded the displacement (Δx, Δy) of the centroids on the sensor
within each image, and if the patients move more than the from their ideal position yields a measure of the X and Y
allowable 0.55 mm, the operator is warned to discard the gradients, and, from the gradients, one can reconstruct the
test. wavefront aberration map (Figure 7-43).
The limit of this method is that the centroids are mea-
Sh ort Bivalen t Con e surable only while they stay within their cells as in Figure
7-43A, and a more highly distorted pattern might not be
Because of the retro-mire solution described above, measurable, as in Figure 7-43B.
the Keratron Onda cone is one-third shorter, and its The Keratron Onda employs another method, using
hole is twice as wide as other Keratrons. Also, unlike frequency demodulation in the FFT domain (Figure 7-44).
other Keratrons, which offer 2 cones that can be physi- Methods of this type have been devised by several authors
cally exchanged, the Keratron Onda has 2 pairs of pho- for use with different sensors,36-39 one of which was
tocells in its EPCS that allow measurement at 2 different
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 18 7

Figure 7-44. Keratron O nda’s frequency demodulation method

to get gradients from the wavefront sensor image.

Figure 7-42. The new Keratron O nda cone with ADC com-
pared to a Keratron traditional cone.

Figure 7-45. In this case, due to a bad sensor image, the recon-
struction of gradients and the OW map are affected by “phase
jump” artifacts.

Figure 7-43. Classic method of measuring x,y gradients from

the centroids displacement and then extracting the wavefront cells, and the Keratron Onda can measure wider aberrations,
from the gradients (or from the Zernike gradients best-fit to robustness in calculations, and fidelity in reconstruction
the wavefront gradients). (Reprinted with permission of Renzo of the aberrations.42 The limits of the method are that cal-
Mattioli, PhD.)
culations are somewhat slower and some “phase jumps” are
not corrected by phase unwrapping as discussed below.
well described by Dr. Edwin Sarver at the International
Wavefront Congress in 2006.40 In the FFT transform of Ph ase Ju m p Art ifacts an d Th eir Detect ion
the sensor image, the 2 small areas circled in red in Figure Due to an opacity in the subject’s crystalline lens and
7-44 contain all the information we need about the X and cornea reflections in a poor sensor image, phase unwrap-
Y displacement of a regular orthogonal lenslet projection. ping created a “phase jump,” making the map in Figure
Then, by anti-transforming the FFT, centered and masked 7-45 unreliable. Such a situation is analogous to a “ring
at these points, we get maps of the phases (Figure 7-44, jump” in Placido corneal topography. In this situation,
bottom left), and then a so-called “phase unwrapping” pro- double-clicking on the sensor image, as in Figure 7-46, will
cedure41 reconstructs the gradients X and Y (Figure 7-44, show maps of the X and Y gradients. The user can accept
bottom right). The abrupt phase jump from black to white the acquisition and fix the artifact later or redo the acquisi-
(Figure 7-44, bottom left) shows that, in the sensor image, tion at once.
the centroids have shifted to an adjacent “cell.” The next In Figure 7-47, the same eye has been acquired again
phase jump from white to black indicates that they have with better centering. The reflections are reduced, and
shifted “2 cells away,” and so on. we can see correct reconstruction of the gradients and the
In practice, this method offers the advantages of a larger resulting correct OW map, in spite of some crystalline lens
dynamic because the centroids are not constrained in their opacity.
188 Ch a p t e r 7

Figure 7-46. The operator can double-click on the wavefront

sensor image to detect phase-jump artifacts on a topo-aber-
rometry test just after acquisition. Figure 7-48. Mesopic-scotopic (IR=infrared) and photopic
(VL=visible light) pupil images and sizes are shown just after
topo-aberrometry acquisition.

Figure 7-47. The operator acquired the same case as in Figure Figure 7-49. Dynamic pupillometry can be shown as a slow-
7-45 again. This time, in spite of opacities in the crystalline motion movie. Here, one of the author’s pupils changes from
lens, the reconstruction of gradients and the O W map are scotopic to photopic conditions over 800 ms. The changes
correct. have been transferred to the Scout software. Pupil size versus
time response to visible light is plotted on the right.

Pupillom et ry
In addition to corneal topography and wavefront aber-
rometry, the Keratron Onda acquires and measures the
patient’s pupil in 2 extreme conditions: “scotopic,” which
is taken in infrared light under mesopic conditions, and
“photopic,” taken about 1 second after turning on the LEDs
(Figure 7-48). Actually, the instrument captures an entire
movie sequence of images under these changing conditions,
an example of which is shown in Figure 7-49. After turning
the LEDs on, we can see that it takes about 800 millisec-
onds for the pupil to go into miosis. This sequence can be
recorded at up to 50 fps and saved in the PC for further
Figure 7-50. Acquisition of a topo-aberrometry test. The
analyses. operator can align the patient’s eye and view: the live image
of the topography CCD (1) and of the wavefront sensor (3),
Acqu irin g Top o -Aberrom et ry the position of the light beam on the entrance pupil (2), and
the Z distance of the eye according to the EPCS (4). He can
When “Topo-Aberrometry” (corneal topography + select the near/far distance cone detectors (5) and the defocus
wavefront aberrometry) is selected with the Keratron Onda compensation (7) and can activate or deactivate “Auto-RX” (6)
display touch screen, the “Acquisition” screen appears and “fogging” (7) functions.
(Figure 7-50).
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 18 9

Figure 7-51. The corneal and OWs are shown side by side. Figure 7-52. Curvature or axial maps and Sim-K can be shown
The Rx refractive data at the bottom-center can be calculated on the Keratron O nda display before transferring the topo-
at either a selected, photopic (VL), or scotopic (IR) pupil, with aberrometry to the external PC with the Scout software.
a single click.

The patient’s refraction taken from the OW is shown as

The operator can choose the type of cone (see Figure “Rx” (see Figure 7-52). This refraction can vary according
7-50, 1). Wavefront optics compensation for defocus can to the selected vertex distance (VD), cylinder type (positive,
be selected manually by sliding a cursor from +6 to -11 D negative, or automatic), and dioptric resolution (0.01, 0.12,
(see Figure 7-50, 7) or automatically by activating the “Auto or 0.25 D). Most important, the refraction can be greatly
Rx” button (see Figure 7-50, 6). Activating the “Fogging” affected by pupil size, which can be manually selected
button (see Figure 7-50, 8), presets the acquisition to show between as small as 3 mm and up to a maximum of the
a fogging target in 3 or more steps, from virtual “close” size measured by the wavefront sensor. Alternatively, the
(-1 D) to “far” distance (+1 D beyond the max defocus that operator can click either of the 2 little round “Rx” buttons
he or she can objectively accommodate). Note that fogging to calculate pupil size at photopic (VL) or scotopic (IR)
is obviously unnecessary with presbyopic or pseudophakic conditions as measured by the Keratron Onda’s topographic
patients. pupillometry feature. Eventually, it will be possible to
The operator can monitor the acquisition while simul- input the refraction (Rx) measured by topo-aberrometry as
taneously centering the keratoscopic image of the eye (see a “manifest refraction” in the test database for any Scout
Figure 7-50, 1) and paying attention to the wavefront sen- software applications (CW, CL, and laser links).
sor image (see Figure 7-50, 3). A yellow circle (see Figure
7-50, 2) shows where the beam is projected into the eye. The
beam should fall within the pupil in order to have a valid Clin ica l Ca ses
retinal reflection in the screen (see Figure 7-50, 3) but at a
distance from the corneal vertex to avoid corneal parasitic Case 1: A 51-Year-Old Pat ien t W h o
reflections (as also described in Figure 7-44). The EPCS Un derwen t Rad ial Keratotom y in 1989
visual feedback (see Figure 7-50, 4) and its characteristic
varying intensity sound tells the operator when he or she is The topographic map (Figure 7-53B) shows a small
at the proper Z distance from the eye. optical zone, which is usual after radial keratotomy (RK)
Pressing the button at the distance of “OK” starts the treatment of moderate myopia, and the irregularities in
“Auto RX” and/or the patient’s “Fogging” phases, depend- the midperiphery that resulted from the 6 incisions. The
ing on which buttons were previously activated. The images OW map (Figure 7-53C) has been set to exclude the low
then freeze if the EPCS detects an acceptable Z distance. orders (astigmatism and defocus) because it is important to
Then, the visible LEDs turn on and, after a minimum of understand the influence of the incisions and the resulting
800 ms, an additional image is acquired for pupillometry small optical zone on spherical aberration. Coma is also
with visible light. Finally, the “Preview” screen appears frequent in RK patients because the asymmetrical bulging
(see Figure 7-48). of the radials is usually enhanced in the inferior sector. In
At this point, the operator may decide whether to dis- this case, the amount of spherical aberration (0.42 µm) and
card the image and proceed with another acquisition, send coma (0.69 µm) are greater than normal, considering that
the topo-aberrometric test to an external PC for further this analysis was made at a pupil size of 4.75 mm. Most of
processing and storage, or process the test immediately the aberration in this case originates in the anterior cornea
(Figures 7-51 and 7-52). Immediate processing allows the (Figure 7-53D), but the IW (Figure 7-53E) is affected by
operator to examine the maps and data such as Sim-K (see the posterior corneal surface, whose distortions match the
Figure 7-52) and refraction without necessarily saving the incisions. The precise position and number of the incisions
entire examination. is clearly visible in Figure 7-53A.
190 Ch a p t e r 7

Figure 7-54. Wavefront analysis of a pseudophakic patient.

An HOA due to fibrosis of the anterior capsule can be seen
in the superior sector. (Reprinted with permission of Massimo
Camellin, MD.)

Figure 7-53. Wavefront analysis of an RK surgical case. Some

internal aberrations are a result of posterior corneal surface
deformation that corresponds to the radial incisions. (Reprinted
with permission of Massimo Camellin, MD.)

Case 2: A 70 -Year-Old Pseudoph akic

Patien t W h o Com plain ed of Nigh t Glare
Clinically, the implanted IOL was spherical and appears
to be well-centered with respect to the pupil. The cor-
neal topography (Figure 7-54B) is unremarkable, but the
wavefront sensor image (Figure 7-54A) shows an anterior
capsular phimosis with reactive fibrosis in the upper left. Figure 7-55. Wavefront analysis of a penetrating keratoplasty
While the total ocular component of coma (0.038 µm) and for a keratoconus. Some internal aberrations are due to the
deep folds inside the transplant. (Reprinted with permission of
spherical aberration (0.24 µm) indicate perfect centering
Massimo Camellin, MD.)
of the lens, some higher-order irregularities in the superior
sector (Figure 7-54C) upon further analysis appear to be
due to internal aberration (Figure 7-54E). refractive index of the air/stroma interface is much greater
Thus, the patient’s complaint of night glare is supported than that of the stroma/water interface, then obviously
objectively, and one can attribute responsibility to the supe- the posterior corneal distortions contribute less aberration.
rior sector of the anterior capsule. However, their contribution is still evident in Figure 7-55E.
Figure 7-55A shows the wavefront sensor grid on which the
Case 3: A 43-Year-Old Pat ien t W h o vertical striae are due to folds of the donor cornea.
Un derwen t Pen et rat in g Keratoplast y for
Keratocon u s in His Left Eye
Con clu sion
The patient has high residual astigmatism that results in
poor visual acuity (VA = 5/10) with spectacle correction (OS The Keratron Onda measures corneal topography and
-4 sph +6.25 cyl at 65 degrees). Topographic analysis (Figure aberrations of the total OW contemporarily; tracks and
7-55B) shows substantial regular astigmatism (Figure 7-55C). measures the pupil in photopic and scotopic or mesopic
Ocular aberrometry is set to show higher orders, except conditions; and performs dynamic pupillometry. Combining
coma and spherical aberration, which are deselected (Figure these functions in a single instrument required redesign-
7-55C-E). The remaining components have highly disturb- ing the cone architecture and developing an alternative
ing microirregularities. The OW (Figure 7-55C) is quite for centroid displacement, as well as other adjustments for
similar to the CW (Figure 7-55D). Few components of the increased acquisition time. The Scout software that runs all
IW have opposite signs from the CW, so the internal aber- Keratron operations coordinates the multiple functions of the
rations clearly stem from the posterior surface of the cornea. Keratron Onda while offering the operator great flexibility.
Therefore, the patient has stromal traction forces that distort Examples of preliminary clinical applications demonstrated
both the front and the rear surface of the cornea. Because the the remarkable advantage of the combined instrument.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 191

TOPCON KR-1W
Naoyuki Maeda, MD

W
ith the advancements in refractive surgery, cata-
ract surgery, and contact lenses, there is a
huge demand to evaluate the quality of vision
objectively. Corneal topographic analysis is very useful to
show the optical quality of the corneas following refractive
surgery or with corneal diseases such as keratoconus.43-45
Interpretation of color-coded maps reveals the corneal
irregular astigmatism qualitatively.46 Topographic indi-
ces such as SRI47 or the results of Fourier analysis48 and
Zernike analysis45 can show the effects of corneal irregular
astigmatism on vision quantitatively.
Recently, the compensational association between cor-
nea and internal optics became obvious not only for regular
astigmatism but also for irregular astigmatism or HOAs.49
For this reason, instruments that measure both corneal and
ocular HOAs may be useful for the comprehensive evalua-
tion of the optical system of the eye.
Figure 7-56. The Topcon unit provides 4 functions: autokera-
tometry, autorefractometry, videokeratography, and wavefront
sensing.
Ma in Feat u res
The Topcon KR-1W (Figure 7-56) provides 4 functions:
autokeratometry, autorefractometry, videokeratography, Top ograph y
and wavefront sensing, sequentially in a session. Table 7-2 This machine uses Placido rings for corneal topography.
indicates the specifications of the machine. Many original Infrared illumination is used to project the Placido rings,
articles and reviews have already been published with improving the patient’s fixation during measurements by
the previous machine (KR-9000PW) or prototype of this reducing the luminance and minimizing miosis of the pupil
instrument.50-77 The machine is capable of fully-automated during the wavefront sensing.
measurement including autofocus and X-Y alignment, Output display for corneal topography consists of a mire
enabling measurements for both eyes by just touching the image, corneal power map, and corneal HOA map (Figure
center of the pupil on the instrument’s touch screen. 7-57). The axial power and instantaneous power map with
the Smolek/Klyce scale79 are set as the default. An absolute
Keratom et ry an d Refractom et ry scale with 1.5-D steps best illustrates clinically significant
Keratometry and refractometry are performed like con- abnormality in corneal shape, while screening for clinically
ventional autorefractometers or autokeratometers. In addi- insignificant differences in topography.80 Conventional
tion to this conventional autorefractometry, this machine videokeratoscope users easily perform visual inspection of
may provide better objective refractions than conventional topographic maps using a standard scale.
ones with the aid of wavefront technology, especially fol- In addition to the conventional display of corneal power
lowing refractive surgery or corneal surgeries.78 Although distribution, the KR-1W enables assessment of corneal
conventional autorefraction obtains sphere and cylinder HOAs both qualitatively and quantitatively. Figure 7-57
values based on paracentral values (from 1.4- to 3.0-mm illustrates the color-coded map of corneal HOAs in addi-
diameter), this wavefront sensor can calculate sphere and tion to the corneal indices and corneal HOAs for 4 mm,
cylinder using the whole data inside the defined pupil: from 6 mm, and actual pupil size. The absolute scale is used for
inside 2 mm (minimum) to 8 mm (maximum) diameter the display of the corneal HOA map to show the clinically
and default 4 mm for day vision and 6 mm for night vision. significant HOAs or irregular corneal astigmatism with the
The difference between these 2 methods may be small in rapid pattern recognition. Different from the conventional
normal eyes but increases dramatically in cases following power maps, irregular astigmatism is easily differentiated
refractive surgery with optical zone abnormality or in post- from sphere and cylinder with the corneal HOA map. By
keratoplasty cases. comparing the corneal HOA map and the ocular HOA map,
the origin of irregular astigmatism can be identified.
192 Ch a p t e r 7

TABLE 7-2.

Specifications of KR-1W
Auto-refractometry Spherical refractive power: -25 to +22 D
Cylindrical refractive power: 0 to 10 D
Measured minimum pupil diameter: 2 mm
Auto-keratometry Corneal curvature radius: 5.00 to 10.00 mm
Corneal refractive power: 67.50 to 33.75 D
Corneal astigmatic power: 0 to 10 D
Ocular HOAs Hartmann-Shack wavefront aberrometer
HOA: Zernike polynomial up to 10th or 4th order
Aberration display: Total aberration, HOA
Measuring range: -25 to +22 D
Measuring area: 8.0 mm
Corneal topography/corneal HOAs Numbers of Placido rings: 19
Corneal curvature radius: 5.00 to 10.00 mm
Corneal refractive power: 67.50 to 33.75 D
Cornea shape measurement area: 0.8 mm to 9.2 mm
Axial power map, Instantaneous power map
HOA: Zernike polynomial up to 10th or 4th order
Range of pupil distance measurement 20 to 85 mm
External connection terminal USB (input/output, output), RS-232C (output),
LAN (input/output)

and wavefront aberrations are calculated by expanding the

wavefront into sets of Zernike polynomials.
The total aberration map represents the aberrations of
the eye including defocus (sphere), regular astigmatism
(cylinder), and irregular astigmatism (Figure 7-58; lower
center). The ocular total aberration map relates to the
uncorrected visual acuity (UCVA), allowing estimation of
UCVA using the total aberration map.
The ocular HOA map (Figure 7-58; lower right) eliminates
lower-order aberrations and isolates only the HOAs, high-
lighting irregular astigmatism. As HOAs relate to the best
spectacle-corrected visual acuity (BSCVA), BSCVA may be
estimated using the HOA map. In addition to the ocular total
and HOA maps, PSF, MTF, and simulated retinal images of
Landolt rings may also be displayed (Figure 7-59).
Figure 7-57. The color-coded map of corneal HOAs in addi-
tion to the corneal indices and corneal HOAs for 4-mm and
actual pupil size (4.47-mm). Rep resen t at ive Ca ses
The result from a typical emmetropic eye is displayed in
Aberrom et ry Figure 7-60. Mire image (upper left) consists of a smooth
concentric circle, and axial power map (upper center)
The purpose of aberrometry is to evaluate the optical indicates no astigmatism. Corneal HOA map (upper right)
quality of the eye by analyzing the shape of its wave- shows the relatively flat wavefront without clinically signif-
front.50,51 This machine uses Hartmann-Shack sensors, icant irregular astigmatism at the anterior corneal surface.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 19 3

Figure 7-58. Comparison of the topography with corneal and Figure 7-60. A typical emmetropic eye is displayed. Mire
whole eye aberrations aids in diagnosis. The ocular HOA map image (upper left) consists of a smooth concentric circle, and
(lower right) eliminates lower-order aberrations and isolates axial power map (upper center) indicates no astigmatism.
HOAs, highlighting irregular astigmatism. BSCVA may be esti- Corneal HOA map (upper right) shows the relatively flat
mated using the HOA map. wavefront without clinically significant irregular astigmatism at
the anterior corneal surface. The Hartmann image (lower left)
shows the regular grid pattern. A uniform pattern is seen in the
ocular total aberration map (lower center) indicating emmetro-
pia, with minimal irregularity in the HOA map (lower right).

Figure 7-59. PSF, MTF, and simulated retinal images of Landolt

rings. Point spread function (PSF, lower left), 2-dimensional
modulation transfer function (MTF, lower center), MTF graph
for X and Y axis (lower right), and simulated retinal images of
Landolt rings (right) are shown.
Figure 7-61. Myopic astigmatism. Vertical bowtie indicating
regular astigmatism with prolate asphericity is seen on the
corneal axial power map. O cular total aberration shows a slow
The Hartmann image (lower left) shows the regular grid oval wavefront in the center characteristic of severe myopia
pattern. A uniform pattern is seen in the ocular total aberra- with regular astigmatism. The flat wavefront in both corneal
tion map (lower center) indicating emmetropia, with mini- and ocular HOA maps indicates mild irregularity.
mal irregularity in the HOA map (lower right).
Figure 7-61 illustrates myopic astigmatism. Vertical
bow-tie pattern that indicates regular astigmatism with Distorted mires and localized abnormal steepening
prolate asphericity is seen in the corneal axial power map. are typical of a keratoconic eye (see Figure 7-58). Ocular
Ocular total aberration shows a slow oval wavefront in the total aberration map illustrates myopic astigmatism, while
center characteristic of severe myopia with regular astigma- the corneal HOA map reveals coma—the combination of
tism. The flat wavefront in both corneal and ocular HOA advanced wavefront superiorly and delayed wavefront infe-
maps indicate lack of irregularity. riorly.68 The corneal HOA map reveals that the irregular
astigmatism is attributed to the cornea. The simulated reti-
nal images are characteristic of coma aberration.
194 Ch a p t e r 7

Figure 7-63. For each pair of the standard Zernike terms, a sin-
gle magnitude and axis value are calculated using Campbell’s
simplified Zernike functions

Figure 7-62. An eye after conventional myopic LASIK. The flat

area seen in the center of the axial power map (top left) and
flat wavefront seen in the ocular total aberration map (lower
left) are the result of myopic correction. In the corneal and
ocular HOA maps, advanced wavefront in the center indicates
mild increase of spherical aberration.

An example of an eye following conventional myopic

LASIK is shown in Figure 7-62. The flat area seen in the
center of the axial power map and flat wavefront seen in
the ocular total aberration map are the result of myopic
correction. In the corneal and ocular HOA maps, advanced
wavefront in the center indicates mild increase of spherical
aberration.

Figure 7-64. In a myopic patient, the map for total HOA (top)
Zer n ike Vector Map shows mild but complex patterns. Zernike vector maps (bot-
tom row) clearly show that the pattern is due to the combina-
HOA maps are a combination of several Zernike terms. tion of a small amount of coma and spherical aberration.
It is sometimes difficult to perform visual inspection of
wavefront maps. Pairs in the Zernike terms (Figure 7-63)
such as in trefoil and coma make it difficult to analyze the
characteristics of ocular HOAs quantitatively. Describing
the Zernike terms as vector components with a magnitude
and orientation similar to the cylinder power and axis may
be helpful. For each pair of the standard Zernike terms,
a single magnitude and axis value are calculated using
Campbell’s simplified Zernike functions81 as shown in
Figure 7-63.
In the Zernike vector map, the magnitudes and orienta-
tions of coma, trefoil, secondary astigmatism, and tetrafoil
in addition to the spherical aberration were shown. At a
glance of the map, the character of the HOA in each clinical
case can be easily understood. In a myopic patient (Figure
7-64), the map for total HOA (top, left) shows mild but com-
plex patterns. On the other hand, Zernike vector maps (bot-
tom) clearly show that the pattern is due to the combination Figure 7-65. Zernike vector analysis in a patient with kerato-
of a small amount of coma and a small amount of spherical conus. Zernike vector maps indicate magnitudes and angles
aberration. Figure 7-65 reveals the Zernike vector analysis for each term. Coma is the dominant aberration, followed by
in a patient with keratoconus. The output of Zernike vector trefoil and secondary astigmatism.
maps indicates the magnitudes and angles for each term.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 19 5

Figure 7-66. The IO L selection map can be used as a screening Figure 7-67. Sequential HOA measurements were obtained as
tool for premium IO L candidates. It highlights corneal irregular subjects were asked to inhibit blink for 10 seconds and maps
astigmatism, regular astigmatism, and spherical aberration. were captured every second.

It is obvious that coma aberration is dominant followed by Last, evaluate the amount of corneal cylinder (Figure
trefoil and secondary astigmatism in this eye.68 7-66; bottom right). Surgical correction of regular astigma-
tism using toric IOL, touch-up with excimer laser, or limbal
relaxing incisions should be considered for the cases with
In t rao cu la r Len s Select ion high corneal cylinder.

With the development of new technology, aspheric, toric,

and multifocal IOLs are currently available. To ensure Ser ia l Mea su rem en t s of
successful implantation of these newer IOLs, the optical
properties of the cornea must be evaluated prior to cataract Ocu la r Wavefron t Ab er rat ion
surgery.
It is well-known that the ocular refraction is dynamic
The IOL selection map (Figure 7-66) can be used as a
due to the changes in accommodation, pupil size, and tear
screening tool for candidates of new-technology IOLs. In
film. On the other hand, most of the autorefractometers and
this program, the screening process consists of 4 steps.
aberrometers measure the refraction with static fashion.
First, evaluate the amount of corneal irregular astigmatism
KR-1W enables us to measure HOA sequentially for 10
as the total corneal HOA (Figure 7-67; bottom left). If
seconds.
there is a clinically significant amount of corneal irregular
During the measurements of sequential HOA, subjects
astigmatism, it will be important to exclude such a candi-
were asked to inhibit blink for 10 seconds to obtain maps
date from multifocal IOLs. Also, surgeons can educate the
every second (Figure 7-67). Serial measurements of HOAs
patient about the possibility of a limited outcome postopera-
in healthy eyes69 showed variations even in clinically nor-
tively due to irregular astigmatism when total corneal HOA
mal subjects (classified into 4 groups by pattern: stable,
is beyond the clinically significant range.
small fluctuation, saw tooth, and others). The sequential
The second step involves the IOL power calculation. By
changes of HOAs in dry eye70,71 soft contact lens wear,72
comparing the corneal power between center and paracen-
or after punctal plug insertion73 were shown using the pro-
tral areas, surgeons estimate the need for special IOL power
totype machine.
calculation. This will be especially important for cataract
following LASIK or keratoplasty.
The third step is to assess corneal spherical aberration
when considering aspherical IOLs. There is a wide range of Con clu sion
variation of corneal spherical aberration even in the normal Using the combination of Placido disk-based corneal
population, and aspherical IOLs for the eyes with negative topographer and Hartmann-Shack wavefront sensor, the
corneal spherical aberration or spherical IOLs for the eyes detailed information about corneal and ocular HOAs can be
with extremely positive corneal spherical aberration should obtained in the clinic.
be avoided.
196 Ch a p t e r 7

R EFERENCES
1. Molebny VV, Panagopoulou SI, Molebny SV, Wakil JS, Pallikaris 21. Okamoto S, Kimura K, Funakura M, Ikeda N, Hiramatsu H,
IG. Principles of ray tracing aberrometry. J Refract Surg. 2000;16: Bains HS. Comparison of myopic LASIK centered on the coaxi-
S570-S575. ally sighted corneal light reflex or line of sight. J Refract Surg.
2. Holladay JT, Lynn MJ, Waring GO, et al. The relationship of visual 2009;25:S944-S950.
acuity, refractive error, and pupil size after radial keratotomy. Arch 22. Wu L, Zhou X, Chu R, Wang Q. Photoablation centration on the
Ophthalmol. 1991;109:70-76. corneal optical center in myopic LASIK using AOV excimer laser.
3. Kirwan C, O’Keefe M. Higher order aberrations in children with Eur J Ophthalmol. 2009;19(6):923-929.
amblyopia. J Pediatr Ophthalmol Strabismus. 2008;45(2):92-96. 23. El Danasoury AM. NIDEK optimized prolate ablation for the
4. Iida Y, Shimizu K, Ito M, Suzuki M. Influence of age on ocular treatment of myopia with and without astigmatism. J Refract Surg.
wavefront aberration changes with accommodation. J Refract Surg. 2009;25:S136-S141.
2008;24(7):696-701. 24. Mattioli R. Topo-aberrometry with Keratron Onda. Congress:
5. Win-Hall DM, Glasser A. Objective accommodation measurements “Refractive on line,” ICH-Milano, Italy, Sept. 2009. http://www.
in pseudophakic subjects using an autorefractor and an aberrometer. refractiveonline.com/2009/eng/52mattioli/player.html.
J Cataract Refract Surg. 2009;35(2):282-290. 25. Carones F, Gobbi PG, Brancato R, Venturi E. Comparison between
6. Gemoules G, Morris KM. Rigid gas-permeable contact lenses two computer-assisted keratoscopes in measuring aspheric surfaces.
and severe higher-order aberrations in postsurgical corneas. Eye Invest Ophtalmol Vis Sci (suppl). 1994:3748.
Contact Lens. 2007;33(6 Pt 1):304-307. 26. Brancato R, Carones F. Topografia corneale computerizzata.
7. MacRae S, Fujieda M. Slit skiascopic-guided ablation using the Fogliazza, Milano (Italy), 1994.
Nidek laser. J Refract Surg. 2000;16:S576-S580. 27. Mattioli R, Carones F, Cantera E. New algorithms to improve the
8. McAlinden C, Moore JE. Comparison of higher order aberrations reconstruction of corneal geometry on the Keratrontm videoker-
after LASIK and LASEK for myopia. J Refract Surg. 2010;26:45- atographer. Invest Ophthalmol Vis Sci. 1995;36(suppl):1400.
51. 28. Mattioli R. Rappresentazione della wavefront corneale della
9. Klyce SD, Karon MD, Smolek MK. Screening patients with the topografia. Congress: “Refractive on Line,” ICH-Milano, Italy,
corneal navigator. J Refract Surg. 2005;21:S617-S622. Sept. 2001. http://www.refractiveonline.com/eng/pages/edition.
10. Holladay JT. Keratoconus detection using corneal topography. aspx?anno=2001.
J Refract Surg. 2009;25:S958-S962. 29. Mattioli R. Corneal topography and wavefront analysis. In: Caimi
11. Gatinel D, Hoang-Xuan T. Objective assessment of the quality of F, Brancato R, eds. The Aberrometers. Canelli, Italy: Fabiano;
vision before and after repositioning of a dislocated iris-fixated 2003:113-132.
aphakic anterior chamber lens. J Refract Surg. 2007;23:S1005- 30. Tripoli NK, Cohen KL, Holmgren DE, Coggins JM. Assessment
S1010. of radial aspheres by the Keratron keratoscope using an arc-step
12. Carey PJ, Leccisotti A, McGilligan VE, Goodall EA, Moore CB. algorithm. Am J Ophthalmol. 1995;120:658-664.
Assessment of toric intraocular lens alignment by a refractive 31. Cohen KL, Tripoli NK, Holmgren DE, Coggins JM. Assessment
power/corneal analyzer system and slitlamp observation. J Cataract of the power and height of radial aspheres reported by a computer-
Refract Surg. 2010;36:222-229. assisted keratoscope. Am J Ophthalmol. 1995;119:723-732.
13. Gatinel D, Adam PA, Chaabouni S, et al. Comparison of corneal and 32. Mattioli R, Tripoli NK. Corneal geometry reconstruction with the
total ocular aberrations before and after myopic LASIK. J Refract Keratron videokeratographer. Optom Vis Sci. 1997;74:881-894.
Surg. 2010;26(5):333-340. 33. Mattioli R, Frondizi S, Rubeo A, Saldutti M. The Keratron
14. Nissman SA, Tractenberg RE, Saba CM, Douglas JC, Lustbader JM. Scout and wavefront analysis. In: Caimi F, Brancato R, eds. The
Accuracy, repeatability, and clinical application of spherocylindri- Aberrometers. Canelli, Italy: Fabiano; 2003:243-257.
cal automated refraction using time-based wavefront aberrometry 34. Mattioli R, Frondizi S, inventors; Optikon 2000 S.p.A., assignee.
measurements. Ophthalmology. 2006;113:577. Multi-purpose ophtalmological apparatus, US patent 7878654;
15. Barreto J Jr, Netto MV, Cigna A, Bechara S, Kara-José N. Precision February 1, 2011.
of higher order aberration repeatability with NIDEK OPD-SCAN 35. Mattioli R, inventor; Optikon Oftalmologia S.p.A., assignee. An
retinoscopic aberrometry. J Refract Surg. 2006;22:S1037-S1040. apparatus for topographical analysis of anatomical surfaces in par-
16. McDonnell C, Rolincova M, Venter J. Comparison of measurement ticular of cornea, US patent 5,526,073; March 30, 1994.
of pupil sizes among the colvard pupillometer, procyon pupillom- 36. Takeda M, Ina H, Kobayashi S. Fourier-transform method of fringe-
eter, and NIDEK OPD-SCAN. J Refract Surg. 2006;22:S1027- pattern analysis for computer-based topography and interferometry.
S1030. JOSA. 1982;72(1):156-160.
17. Lim KL, Fam HB. Ethnic differences in higher-order aberrations: 37. Carmon Y, Ribak EN. Phase retrieval by demodulation of a
spherical aberration in the South East Asian Chinese eye. J Cataract Hartmann-Shack sensor. Optics Comm. 2003;215:285-288.
Refract Surg. 2009;35:2144-2148. 38. Ribak EN, Carmon Y, Talmi A, Glazer O, Srour O, Zon N. Full
18. Hori-Komai Y, Toda I, Asano-Kato N, Ito M, Yamamoto T, Tsubota wave front reconstruction in the Fourier domain. Proc. SPIE.
K. Comparison of LASIK using the NIDEK EC-5000 optimized 2006;6272(54):1-9.
aspheric transition zone (OATz) and conventional ablation profile. 39. Horwitz LS. Wavefront sensor using near-field diffraction effects.
J Refract Surg. 2006;22:546-555. 5th International Congress on Wavefront Sensing and Optimized
19. Venter J. Wavefront-guided custom ablation for myopia using the Refractive Correction; February 21-23, 2004; Whistler, Canada.
NIDEK NAVEX laser system. J Refract Surg. 2008;24:487-493. 40. Sarver EJ. Extracting wavefront error from S/H images using spatial
20. Subbaram MV, MacRae SM. Customized LASIK treatment for demodulation. 7th International Congress on Wavefront Sensing
myopia based on preoperative manifest refraction and higher and Optimized Refraction Correction; January 26-29 2006; Nassau,
order aberrometry: the Rochester nomogram. J Refract Surg. Bahamas.
2007;23:435-441.
 To p o gr a p hy a n d Wa ve fr o n t Co m b in e d Syst e m s 197

41. Ghiglia DC, Pritt MD. Two-Dimensional Phase Unwrapping: 62. Fujikado T, Kuroda T, Maeda N, et al. Light scattering and optical
Theory, Algorithms and Software. New York, NY: Wiley; 1998:1- aberrations as objective parameters to predict visual deterioration in
57. eyes with cataracts. J Cataract Refract Surg. 2004;30:1198-1208.
42. Ribak EN, Canovas C. Comparison of Hartmann analysis methods. 63. Amano S, Amano Y, Yamagami S, et al. Age-related changes
Applied Optics. 2007;46(10):1830-1835. in corneal and ocular higher-order wavefront aberrations. Am J
43. Klyce SD, Maeda N, Byrd TJ. Corneal topography. In: Kaufman Ophthalmol. 2004;137:988-992.
HE, Barron BA, McDonald MB, eds. The Cornea. 2nd ed. Boston, 64. Tanabe T, Miyata K, Samejima T, Hirohara Y, Mihashi T, Oshika
MA: Butterworth-Heinemann; 1998:1055-1075. T. Influence of wavefront aberration and corneal subepithelial haze
44. Applegate RA, Hilmantel G, Howland HC, et al. Corneal first on low-contrast visual acuity after photorefractive keratectomy. Am
surface optical aberrations and visual performance. J Refract Surg. J Ophthalmol. 2004;138:620-624.
2000;16:507-514. 65. Yamane N, Miyata K, Samejima T, et al. Ocular higher-order aberra-
45. Naoyuki M. Evaluation of optical quality of corneas using corneal tions and contrast sensitivity after conventional laser in situ keratomi-
topographers. Cornea. 2002;21:S75-S78. leusis. Invest Ophthalmol Vis Sci. 2004;45:3986-3990.
46. Maguire LJ, Singer DE, Klyce SD. Graphic presentation of com- 66. Fujikado T, Kuroda T, Ninomiya S, et al. Age-related changes in
puter-analyzed keratoscope photographs. Arch Ophthalmol. ocular and corneal aberrations. Am J Ophthalmol. 2004;138:143-
1987;105:223-230. 146.
47. Wilson SE, Klyce SD. Quantitative descriptors of corneal topogra- 67. Bellucci R, Morselli S, Piers P. Comparison of wavefront aber-
phy. A clinical study. Arch Ophthalmol. 1991;109:349-353. rations and optical quality of eyes implanted with five different
48. Oshika T, Tomidokoro A, Maruo K, Tokunaga T, Miyata N. intraocular lenses. J Refract Surg. 2004;20:297-306.
Quantitative evaluation of irregular astigmatism by Fourier series 68. Kosaki R, Maeda N, Bessho K, et al. Magnitude and orientation of
harmonic analysis of videokeratography data. Invest Ophthalmol Zernike terms in patients with keratoconus. Invest Ophthalmol Vis
Vis Sci. 1998;39:705-709. Sci. 2007;48:3062-3068.
49. Artal P, Guirao A, Berrio E, Williams DR. Compensation of corneal 69. Koh S, Maeda N, Hirohara Y, et al. Serial measurements of
aberrations by the internal optics in the human eye. J Vis. 2001;1:1- higher-order aberrations after blinking in normal subjects. Invest
8. Ophthalmol Vis Sci. 2006;47:3318-3324.
50. Maeda N. Wavefront technology in ophthalmology. Current Opin 70. Koh S, Maeda N, Hirohara Y, et al. Serial measurements of higher-
Ophthalmol. 2001;12:294-299. order aberrations after blinking in patients with dry eye. Invest
51. Maeda N. Wavefront technology and LASIK application. In: Azar Ophthalmol Vis Sci. 2008;49:133-138.
DT, Koch DD, eds. LASIK: Fundamentals, Surgical Techniques, 71. Koh S, Maeda N, Hori Y, et al. Effects of suppression of blinking on
and Complications. New York, NY: Marcel Dekker, Inc; 2002:139- quality of vision in borderline cases of evaporating dry eye. Cornea.
151. 2008;27:275-278.
52. Kuroda T, Fujikado T, Maeda N, Oshika T, Hirohara Y, Himashi T. 72. Koh S, Maeda N, Hamano T, et al. Effects of internal lubricating
Wavefront analysis in eyes with nuclear or cortical cataract. Am J agents of disposable soft contact lenses on higher-order aberrations
Ophthalmol. 2002;134:1-9. after blinking. Eye & Contact Lens. 2008;34:100-105.
53. Koh S, Maeda N, Kuroda T, et al. Effect of tear film break-up on 73. Koh S, Maeda N, Ninomiya S, et al. Paradoxical increase of visual
higher-order aberrations measured with wavefront sensor. Am J impairment with punctal occlusion in a patient with mild dry eye.
Ophthalmol. 2002;134:115-117. J Cataract Refract Surg. 2006;32:689-691.
54. Kuroda T, Fujikado T, Maeda N, Oshika T, Hirohara Y, Mihashi 74. Fujikado T, Shimojyo S, Hosohata J, et al. Wavefront analysis of eye
T. Wavefront analysis of higher-order aberrations in patients with with monocular diplopia and cortical cataract. Am J Ophthalmol.
cataract. J Cataract Refract Surg. 2002;28:438-444. 2006;141:1138-1140.
55. Kuroda T, Fujikado T, Ninomiya S, Maeda N, Hirohara Y, Mihashi 75. Mihashi T, Hirohara Y, Koh S, Ninomiya S, Maeda N, Fijikado T.
T. Effect of aging on ocular light scatter and higher order aberra- Tear film break-up time evaluated by real-time Hartmann-Shack
tions. J Refract Surg. 2002;18:S598-S602. wavefront sensing. Jpn J Ophthalmol. 2006;50:85-89.
56. Ninomiya S, Fujikado T, Kuroda T, et al. Changes of ocular aberra- 76. Oie Y, Maeda N, Kosaki R, et al. Characteristics of ocular higher-
tion with accommodation. Am J Ophthalmol. 2002;134:924-926. order aberrations in patients with pellucid marginal corneal degen-
57. Ninomiya S, Maeda N, Kuroda T, et al. Evaluation of lenticular eration. J Cataract Refract Surg. 2008;34:1928-1934.
irregular astigmatism using wavefront analysis in patients with 77. Maeda N. Clinical application of wavefront aberrometry—A review.
lenticonus. Arch Ophthalmol. 2002;120:1388-1393. Clinical & Experimental Ophthalmology. 2009;37:118-129.
58. Maeda N, Fujikado T, Kuroda T, et al. Wavefront aberrations mea- 78. Thibos LN, Hong X, Bradley A, Applegate RA. Accuracy and
sured with Hartmann-Shack sensor in patients with keratoconus. precision of objective refraction from wavefront aberrations. J Vis.
Ophthalmology. 2002;109:1996-2003. 2004;4:329-351.
59. Ninomiya S, Fujikado T, Kuroda T, et al. Wavefront analysis in 79. Smolek MK, Klyce SD, Hovis JK. The Universal Standard Scale:
eyes with accommodative spasm. Am J Ophthalmol. 2003;136:1161- proposed improvements to the American National Standards
1163. Institute (ANSI) scale for corneal topography. Ophthalmology.
60. Kelly JE, Mihashi T, Howland HC. Compensation of corneal hori- 2002;109:361-369.
zontal/vertical astigmatism, lateral coma, and spherical aberration 80. Wilson SE, Klyce SD, Husseini ZM. Standardized color-coded
by internal optics of the eye. J Vis. 2004;4:262-271. maps for corneal topography. Ophthalmology. 1993;100:1723-1727.
61. Fujikado T, Kuroda T, Maeda N, et al. Wavefront analysis of an eye 81. Campbell CE. A new method for describing the aberrations of the
with monocular triplopia and nuclear cataract. Am J Ophthalmol. eye using Zernike polynomials. Optom Vis Sci. 2003;80:79-83.
2004;137:361-363.
 SECTIO N III

T O PO GRAPHIC
APPLICATIO NS
Ch a p te r 8

To p o grap hy
in Co rneal Disease s

Gaurav Prakash, MD and Amar Agarwal, MS, FRCS, FRCO phth

C KERATOCONUS
orneal topography is a vital tool in the assess-
ment of the diseased cornea. Many disorders of
the cornea, such as keratoconus, are identified by
subtle changes in the topography before there are visual Keratoconus is a bilateral, noninflammatory disorder
symptoms or clinically visible signs in slit-lamp biomi- of the cornea that, despite years of study, is not com-
croscopy. As topography looks at the central as well as pletely understood. Early researchers focused on the opti-
the peripheral cornea, many disorders that can be missed cal and clinical signs, morphology, and rehabilitation by
on central evaluation by the keratometry can instead the means of glasses, contact lenses, and keratoplasty.15,16
be diagnosed by an astute clinician using a topography With increasing application of corneal topography, multiple
device.1-7 indices have been described to diagnose cases with early
However, measurement technique, diurnal variation of or forme fruste keratoconus, ranging from simple grading
the corneal hydration, calibration of the device, position- scores to advanced algorithms to monitor progression.17-25 It
ing of the patient’s head, post-LASIK status, and use of is important to be comfortable with a grading system and use
anesthetic drops can influence topographic evaluation, it in sequential follow-ups in the patients. The indications of
and therefore it is essential to reduce these factors to a topography are not limited to diagnosis and monitoring pre-
minimum.8-12 In the case of subclinical disease, there is a intervention progression. With the advent of intrastromal
narrow zone of high sensitivity and low specificity neces- ring implantation and collagen cross-linking, postoperative
sitating consistent and precise measurements. Recent topog- follow-up with topography can provide vital information in
raphy devices, such as the Pentacam, have been found to be collecting more information on the disease process.
more accurate than Orbscan when measuring corneas with Keratoconus is generally associated with adolescents and
certain conditions.13-15 The most classical corneal disorders young adults and has been historically known to stabilize
identified by topography are corneal ectasias: keratoconus, in older age groups, probably because of increased cross-
pellucid marginal degeneration, or post-LASIK ectasia. In linkage in the cornea with growing age.26 It can occur both
addition, the irregular astigmatism resulting from corneal sporadically or inherited as a genetic disorder. Keratoconus
dystrophies, degenerations, scarring, inflammatory condi- may be associated with atopy, corneal dystrophies, cata-
tions, pterygium, or dry eye can also be quantified with the racts, Lebers’, as well as systemic disorders such as Pierre
help of topography. Robins’ sequence, congenital adrenal hyperplasia, Turner’s
syndrome, and Noonen syndrome.27-47 Syndromic associa-
tions, although rare, should be ruled out in all cases.

Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical

201 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 201-214)
© 2012 SLACK Incorporated
202 Ch a p t e r 8

Figure 8-1. The earliest suspicion of keratoconus is an inferior B

or central steepening, with or without a difference between
the keratometry of the fellow eye. (Reprinted with permission
of Tracy Schroeder Swartz, O D, MS, FAAO.)

Recently, interest in biomechanics and long-term out-

comes in the fellow eye in keratoconus has increased.
Bilaterality of keratoconus remains a debatable topic.
Certain large studies concluded that unilateral keratoconus
is a rare entity and up to 50% of fellow eyes of a keratoconic
eye eventually develop keratoconus, with the greatest risk of
development in the first 6 years of diagnosis.48-54
The earliest suspicion of keratoconus is an inferior or Figure 8-2. (A) A case of forme fruste keratoconus. Note the
central steepening, with or without a difference between the steepest keratometric value of 47.7 and 47.5, the astigmatism of
keratometry of fellow eyes (Figure 8-1). There may be mild 3.4 D, and the asymmetric bow-tie. The patient has a normal
astigmatism, with or without any corneal thinning. Classic corneal thickness and posterior elevation as of now. (B) A case
signs of keratoconus, such as Munson’s sign, Rizzuti’s with steep corneal curvature, low astigmatism, normal poste-
rior float, and low pachymetry. This patient had a fellow-eye
sign, and Fleischer ring, may not be present. According to
forme fruste keratoconus.
a classification proposed by Rabinowitz, this presentation
is known as keratoconus suspect.55 Several indices have
been used to differentiate these patients from normal eyes,
particularly in patients presenting for elective refractive
procedures. However, “keratoconus suspect” is a diagnosis
with high sensitivity and therefore holds a high rate of false-
positives. Some of these patients will not develop keratoco-
nus and are simply on the tail end of the normal population
distribution.
Forme fruste keratoconus (FFKC) presents with topo-
graphic changes not causing any visual symptoms and no
ocular morbidity other than mild spherocylinder refractive
error (Figure 8-2A). These patients are at a higher risk of
developing keratoconus. The fellow eyes in cases of FFKC
(Figure 8-2B) may have normal topography or a mildly
steep cornea, and are also at a higher-than-normal risk for Figure 8-3. A case of suspected early keratoconus, with kera-
tometry values <50 D. Even though the pachymetry seems rel-
keratoconus. These eyes are also at risk of developing post-
atively normal and the astigmatism is low, this patient should
LASIK ectasia, so excimer laser ablation should be avoided. not undergo LASIK.
Early keratoconus is another term used to describe patients
who have no slit-lamp findings but have subtle signs like
early scissoring on retinoscopy in addition to topographic islands (Figures 8-3 through 8-5). An island of dioptric
deviations seen in “forme fruste” keratoconus.55 power results from increased corneal curvature surround-
As keratoconus advances, the topographic features asso- ed by concentric annuluses of decreasing power.7,16,24,56
ciated with keratoconus manifest. These include refractive This island may be in either of the 2 inferior quadrants
 To p o gr a p hy in Co r n e a l D ise a se s 203

Figure 8-4. A case of moderate keratoconus. Note the asym- Figure 8-6. A case of atypical keratoconus with superior
metric bowtie, high posterior elevations, and low central meridians being steeper than the inferior. Note that other signs
corneal thickness. The extremely low pachymetry readings in of keratoconus, such as asymmetric bowtie, thin cornea, and
the superior cornea are artifactual, due to a superior corneal high posterior elevation, are present.
nebulomacular haze.

Figure 8-7. A case of keratoglobus. Note the generalized high

Figure 8-5. A case of advanced keratoconus. Note the very keratometry (>60 D) with astigmatism, low pachymetry values
high astigmatism (15 D), corneal thinning, and high posterior extending from limbus to limbus, and very high posterior and
as well as anterior elevation. anterior elevations.

(in primary keratoconus) or central (in secondary kerato- ring implantation and collagen cross-linking topography
conus); however, the steepening involves more than one can be used to monitor treatment efficacy, stability, and
quadrant as the disease progresses. A difference in the pow- improvement in the keratometric power of the anterior cor-
ers of the interior and superior cornea is called inferosupe- neal surface.61-64
rior power asymmetry. Skewing of the steepest radial axes
above and below the horizontal meridian may also occur.
Increased posterior elevation may be noted on tomography,
where the area of maximum steepening corresponds with
KERATOGLOBUS
the area of reduced corneal pachymetry. In pellucid mar-
ginal degeneration, the maximal thinning is just below the Unlike keratoconus, keratoglobus characterizes a more
area of maximum dioptric power. An asymmetric bow-tie, diffuse involvement of the cornea.65,66 Generalized thin-
or “kissing bird” appearance, may be seen in some patients, ning is seen extending from limbus to limbus (Figure 8-7).
while others will demonstrate a globus type of cone or, Features suggestive of keratoconus, keratoglobus, or pellu-
rarely, superior keratoconus where steeper values may be cid marginal degeneration can be found in the fellow eyes,
found superiorly rather than inferiorly (Figure 8-6). suggesting a similar etiology for the 3.67,68 Clinical find-
In addition to topographic signs, higher-order aberra- ings in keratoglobus include the lack of a specific bowtie
tion (HOA) profiles in early keratoconus are being inves- or similar pattern but rather a diffusely steep cornea with
tigated.57-60 Alio and Shabayek have suggested grading irregular keratometry values, a high posterior elevation,
criteria for the same.60 Topography following intrastromal and reduced pachymetry. Keratoglobus has been associated
204 Ch a p t e r 8

B
Figure 8-8. Mooren’s ulcer must be differentiated from pel-
lucid marginal degeneration.

with other disorders like posterior polymorphous dystrophy,

Lebers congenital amaurosis, subcapsular cataracts, and
Rubinstein-Taybi syndrome.68-73 There are sporadic reports
of hydrops in a case of keratoglobus, which is managed
using hyperosmotics with or without intracameral gas injec-
tions.74-76

Figure 8-9. (A) Early pellucid marginal degeneration. Note the

P ELLUCID M ARGINAL “crab claw sign,” also called the “bow-tie sign.” The thinnest
point does not coincide with the steepest one, and, more
D EGENERATION markedly, the point with the highest posterior elevation is dis-
placed inferiorly. (B) An advanced case of pellucid marginal
degeneration. Note that the area of steepening (the claw area
A cousin to keratoconus is pellucid marginal degenera- on the topography map) lies above the thinnest area on the
tion (PMD), which should be differentiated from Mooren’s pachymetry map. The points of highest posterior and anterior
elevation are displaced inferiorly. The thinnest point has a
ulcer (Figure 8-8). The term pellucid means “clear” and is
relatively normal keratometry.
used to differentiate PMD from Terrien’s marginal degen-
eration. PMD demonstrates peripheral thinning, which is
opacified or translucent. PMD can present as progressive
astigmatism with difficulty in contact lens wear, frequent
change of glasses, or an incidental fellow eye of a “normal
KERATOECTASIA
eye” with 20/20 acuity.
PMD differs from keratoconus because the area of thin- Keratoectasia, or ectasia following corneal refractive
ning most often occurs inferiorly to the area of increased surgery, is one of the most dreaded and challenging compli-
dioptric power (Figure 8-9). The midperipheral cornea may cations for refractive surgeons. It is important to understand
decrease in power above the inferior oblique meridians. that all subtractive methods where the excimer laser is used
Literature describes flattening of the central cornea along a (LASIK, epi-LASIK, LASEK, and PRK) and incisional
vertical axis and steepening of the inferior corneal periph- methods (radial keratotomy or astigmatic keratotomy)
ery, which may extend into midperipheral inferior oblique predispose the cornea to varying levels of biomechanical
corneal meridians.77 The “claw sign,” which was previously weakening. The common presenting sign in most of these
thought to be strongly associated with PMD, has recently patients is increasing astigmatism months to years after
been shown to be nonspecific and may be associated with refractive surgery. Disturbance of strong peripheral and
keratoconus.78 The HOA profiles and Gaussian fitting pat- anterior corneal collagen fibers following RK may induce
terns for corneal topographic maps have been found to be ectasia, but because excimer is presently the predominant
different for keratoconus and PMD, perhaps because of the method of elective surgical correction, the subsequent dis-
difference in the location of the apex and resulting power cussion is based on the biomechanics of excimer procedures.
distribution.79,80
 To p o gr a p hy in Co r n e a l D ise a se s 205

Figure 8-10. A topographically normal LASIK candidate with Figure 8-11. Note the low amount of astigmatism and slightly
moderate astigmatism, normal keratometry values, good decentered ablation in this post-LASIK patient. The clinical
corneal thickness, and low posterior elevation. The refrac- picture is stable for 2 years without progression of astigmatism,
tive error was stable at -4 D sphere and -1.75 D cylinder for change in uncorrected or best-corrected vision, increase in
3 years. Based on topography and refraction, he is a good posterior elevation, or reduction in pachymetry.
candidate for LASIK.

A limit of 250 µm of unablated cornea or residual bed

thickness (RBT) was assumed adequate for corneal bio-
mechanical strength.81,82 However, as cases with ectasia
occurred in patients with RBT greater than 250 µm, this
numerical limit was found to be inadequate. As understand-
ing of corneal hysteresis, elasticity, and other biomechanical
properties increased, it was realized that a simple numerical
limit may not be safe.83
Multiple parameters and ectasia scoring schemes have
been used in the literature.84-106 Randleman’s Ectasia Risk
Score System is one such recent index based on topography,
age, preoperative corneal thickness, RBT, and spherical
equivalent.85,92,94 This scoring system was subsequently Figure 8-12. Post-LASIK ectasia. Note the high keratometry,
validated in a follow-up study.94 However, another study astigmatism, and elevated posterior float. There has been a
progressive deterioration over a period of the last 7 months,
found using the score would have put the patients at a higher
after an initial stability of 1 year in this patient.
risk warning as these patients eventually did not develop
ectasia on follow-up.95 Conversely, cases with low risk
scores using the Randleman criteria and cases with RBT
more than 300 µm have also been demonstrated to develop
post-LASIK ectasias.96,97 This suggests that other biome-
chanical evaluations and accurate predictions of variable
tissue behavior as a function of time may be required to
provide an improved risk assessment for keratoectasia.
When evaluating patients for elective surgical correc-
tion, it is important to select cases with low topographic
risk for ectasia using topography (Figure 8-10). It is normal
to have a mild amount of topographic astigmatism post-
LASIK with other parameters being stable (Figure 8-11).
Progressive astigmatism, increased irregularity, increasing
posterior float, and a progressive decrease in pachym-
etry are highly suggestive of keratoectasia (Figures 8-12
through 8-14). Cases with even mild interface or surface Figure 8-13. Suspicious case of post-LASIK ectasia: Note the
asymmetric bowtie pattern, high posterior elevation, and low
haze may result in a falsely low value upon optical pachym-
pachymetry readings in this 8-months-postoperative patient.
etry measurement using slit-scanning technology, and The astigmatism and keratometry are still numerically normal
diagnosis must not be made on this isolated criterion.98,99 and thus were not denoted as “red” on the color scale.
206 Ch a p t e r 8

Figure 8-14. Post-radial keratotomy ectasia. Note the periph-

eral data loss, high irregular astigmatism, multiple areas of
high posterior elevation, and a trifocal distribution of corneal
thinning. These cases can be rehabilitated using contact lenses
and /or keratoplasty.

B
In such cases, ultrasound pachymetry validates measure-
ment. Furthermore, all cases of post-LASIK astigmatism
should not be automatically linked to keratoectasia. Flap
contraction, surface irregularity, and severe dry eyes may
result in topographic findings that incorrectly suggest kera-
toectasias.

D RY EYES
Dry eye disease comprises a constellation of symptoms
and signs resulting from primary or secondary reduction in
the amount or quality of the tear film or problems with the
ocular surface. Inferior steepening in the absence of thin-
ning pachymetry is often seen (Figure 8-15). A review study
noted the prevalence ranged from 7% in the United States
to 33% in Taiwan and Japan and summarized risk factors
as advanced age, female gender, smoking, extreme heat or Figure 8-15. (A) Dry eye disease. Inferior steepening with
normal pachymetry is a common sign. (Reprinted with permis-
cold weather conditions, low relative humidity, use of video
sion of Tracy Schroeder Swartz, O D, MS, FAAO.) (B) In severe
display terminals, history of refractive surgery, contact lens cases of dry eye, melts, scarring, and neovascularization occur.
wear, and certain medications.100 The features of dry eye
range from nonspecific difficulty in reading and near work
and tired eyes to paradoxically increased tearing.
to change because of fluctuation of tear film stability has
Topography using Placido imaging is a noninvasive
previously been used.103-105
method to assess the tear break-up time (TBUT), a test that
requires a smooth anterior corneal-tear film-air complex.
Cases of unstable tear film can be screened using topogra-
phy. In a study comparing noninvasive and invasive TBUT,
short average tear break-up values are associated with small
C ONTACT LENS W ARPAGE
differences between these 2 methods, whereas long average
tear break-up values are associated with large differences The phenomenon where habitual contact lens wear alters
between the methods.101 High-speed videokeratoscopy has corneal topography is called “corneal warpage.” Warpage
been used as an investigational tool in the evaluation of dry has been reported with all lens types: PMMA, gas perme-
eyes.102 A new tear film stability analysis system (TSAS) able, and hydrogel. Clinical signs include reduced vision,
based on the phenomenon that topographic maps are subject irregular keratometry mires, and irregular topography.
 To p o gr a p hy in Co r n e a l D ise a se s 207

Figure 8-16. Contact lens warpage is often seen in

patients with a history of gas-permeable lens wear
but can occur with hydrogel lens wear as well.
(Reprinted with permission of Tracy Schroeder
Swartz, O D, MS, FAAO.)

Often, there is an absence of slit-lamp findings. Topographic

patterns may resemble keratoconus or PMD, or they may be
irregular with no particular pattern (Figure 8-16). CL warp-
age is often identified in patients seeking elective vision
correction. When preparing for keratorefractive surgery
or intraocular lens implantation, resolution of topographic
irregularities is required. Resolution of contact lens warp-
age should be closely monitored using refractive error and
corneal curvature. Irregularity typically improves after lens
wear is discontinued, but, in some cases, does not com-
pletely resolve. The amount of time required for resolution
varies widely. Mobillia and Kenyon105 reported that, when
chronic lens wearers discontinue contact lens wear, the
refractive error changes quickly, and astigmatism increases.
The change reaches a maximum approximately 3 days
after lens removal. These corneal contour and physiologic
changes require time to resolve.
While long-term gas-permeable wear is commonly
thought to cause more warpage, Wang et al106 found that Figure 8-17. Irregular astigmatism secondary to severe EBMD.
soft extended wearers required 11.6 weeks, more than (Reprinted with permission of Tracy Schroeder Swartz, O D,
the 8.8 weeks required by gas-permeable lens wearers. MS, FAAO.)
Resolution of topographic abnormality averaged nearly
8 weeks with an incidence of 12%.

C ORNEAL D EGENERATIONS
C ORNEAL D YSTROPH IES Several corneal degenerations may alter corneal topog-
raphy. Most commonly known to affect topography are
Anterior corneal dystrophies affecting the epithelium, Mooren’s ulcer, Terrien’s ulcer, band keratopathy, and
Bowman’s membrane, and anterior stroma may alter the Slazmann’s nodules. Mooren’s ulcer, a peripheral ulcerative
cornea’s topographical landscape. The 3 main epithelial condition typically presenting in patients older than 40, is
dystrophies include epithelial basement membrane dystro- often rapidly progressive. Terrien’s ulcer, another peripheral
phy (EBMD), Reiss-Buckler dystrophy, and Meesman’s ulcerative condition, occurs more in men and at a younger
dystrophy, with EBMD being the most common (Figure age than Mooren’s ulceration.
8-17). Stromal dystrophies may have more severe effects on Band keratopathy results from calcium salts in Bowman’s
vision due to loss of transparency, but may also cause RCE layer causing opacification of the anterior cornea. It can
and dry eye. These include granular, lattice, Schnyder’s, and cause severely irregular topography and poor tear film
macular dystrophies. distribution.
208 Ch a p t e r 8

Figure 8-20. Herpetic keratitis resulted in scarring and irregular

astigmatism. The effect on vision is illustrated in the refractive
map, and vision was limited to 20 /30 in spectacles.
Figure 8-18. Salzmann’s nodules may cause severe corneal
irregularity and vision loss. (Reprinted with permission of Tracy
Schroeder Swartz, O D, MS, FAAO.) becomes the only alternative in such cases, providing the
eye is quiet and anatomy is normal.

I MMUNE-M EDIATED
D ISORDERS
Immune-mediated disorders often cause corneal pathol-
ogy and irregular astigmatism. Diseases range from the
Figure 8-19. Corneal scarring, in this case from a corneal very mild to devastatingly severe and include atopic kerato-
ulcer corresponding to the flat blue area, may cause irregular conjunctivitis, marginal corneal infiltrates associated with
astigmatism. (Reprinted with permission of Tracy Schroeder blepharoconjunctivitis (Figure 8-21), nonulcerative keratitis,
Swartz, O D, MS, FAAO.)
ocular mucous membrane pemphigoid, peripheral keratitis,
and peripheral ulcerative keratitis associated with systemic
immune-mediated diseases, including rheumatoid arthritis,
Salzmann’s nodular degeneration may cause irregular Wegener granulomatosis, systemic lupus erythematosus,
astigmatism (Figure 8-18) and scarring and may require and inflammatory bowel disease. Treatment includes topical
surgical removal. steroids and systemic immunosuppressive agents, but clini-
cal findings may progress despite aggressive treatment.

C ORNEAL SCARRING
P TERYGIUM
Opacification of the cornea may result from infection,
trauma, chronic inflammation, and surgery. Infectious kera- Pterygium (Figure 8-22) is an elastotic disorder of the
titis can cause severe corneal scarring with severe corneal cornea, occurring more frequently in the nasal area of the
irregularity (Figure 8-19). Particularly devastating infec- bulbar conjunctiva than the temporal. UV light may induce
tions include Acanthamoeba keratitis, bacterial keratitis, corneal invasion of matrix metalloproteinases, express-
fungal keratitis, and sclerokeratitis. ing altered limbal epithelial basal cells.107,108 Without the
Herpes keratitis, caused by the herpes simplex virus advent of topography, the indications for pterygium removal
(HSV), is a leading cause of corneal blindness in indus- included cosmetic complaint, encroachment on the pupil-
trialized nations. A single severe infection, or repeated lary axis, or increasing cylindrical power. With topogra-
infections, may result in scarring and irregular astigmatism phy, early changes such as increasing irregularity within
(Figure 8-20). Prior to surgical correction, irregular astig- the central 5-mm zone prior to it affecting the refraction
matism should be addressed cautiously with a gas-permeable indicate surgical intervention. Correction can occur earlier,
contact lens and frequent lubrication. In some cases, corneal preventing a dissection affecting the pupillary axis. Both
opacification limits visual prognosis, and corneal thinning topographic profile and manifest astigmatism have been
may limit laser treatment options. Penetrating keratoplasty shown to improve after removal of the pterygium.109,110
 To p o gr a p hy in Co r n e a l D ise a se s 209

A B

Figure 8-21. (A) Marginal keratitis causing superior data loss and irregular astigmatism. (B) Slit-lamp view of the superior cornea.

Figure 8-22. (A) (a) Early pterygium with

B minimal corneal involvement. (b) Thick
pterygium encroaching the cornea with
scarring. (B) Topographic map showing a
pterygium nasally O S, and the steepening
adjacent to the lesion (nasally). (Reprinted
with permission of Tracy Schroeder Swartz,
O D, MS, FAAO.)

P OSTKERATOPLASTY in improvement of uncorrected visual acuity by reducing

the postoperative astigmatism.111,112 Intraoperative topog-
raphy devices have been shown to reduce postoperative
Topography plays a vital role in the postoperative man- cylinder.113 In cases of postkeratoplasty astigmatism per-
agement and optical rehabilitation of cases that have under- sistent after suture removal, arcuate keratotomy or excimer
gone keratoplasty (Figures 8-23 through 8-25). Selective laser ablation may be performed based on the topography
suture removal based on the topography may be very useful findings.114-117
210 Ch a p t e r 8

Figure 8-23. Post-penetrating keratoplasty with a glued IO L.

Figure 8-26. Small central zones are often noted in patients

with a history of radial keratotomy). (Reprinted with permission
of Tracy Schroeder Swartz, O D, MS, FAAO.) The small irregu-
lar zone seen on the axial map O D (top row) corresponds to
the easily noted irregularity seen on the Placido image. The left
eye (bottom row) was 20 /20 and asymptomatic.

R ADIAL KERATOTOMY
Incisional procedures alter corneal curvature resulting
in a change in the refractive error. Unfortunately, it is not
uncommon for astigmatism to increase in magnitude and
Figure 8-24. Intralase endothelial keratoplasty with a glued irregularity. Patients with a history of RK often demonstrate
IO L—1-year postoperatively. excessively small optical zones and severe central flattening
with a loss of best-corrected vision (Figure 8-26).

KERATOREFRACTIVE
SURGERY
Excimer procedures ablate tissue, altering corneal shape
to reduce the refractive error and allow the patient lens
independence upon refractive correction. Resulting topo-
graphical patterns should be easily recognizable. Normal
myopic treatments ablate centrally, resulting in central
thinning corresponding to the ablation area and a central
plateau. This creates positive spherical aberration (Figure
8-27). Hyperopic treatments result in central steepening
Figure 8-25. Postkeratoplasty astigmatism. Note the high
irregular astigmatism due to tight sutures and the abrupt shift and negative spherical aberration (Figure 8-28). Decentered
in the pachymetry due to graft-host size disparity. myopic and hyperopic ablations, when significant, result in
polyplopia (Figure 8-29).
 To p o gra p hy in Co r n e a l D ise a se s 211

A
A

Figure 8-28. (A) Hyperopic excimer treatments cause central

steepening. (B) Hyperopic excimer treatments cause negative
spherical aberration. (Reprinted with permission from Trattler
WB, Majmudar PA, Luchs JI, Swartz TS. Corneal Handbook.
Thorofare, NJ: SLACK Incorporated; 2009.)

Figure 8-29. Decentered myopic ablation. Note the coma on

the wavefront map corresponds to the asymmetry noted on the
Figure 8-27. (A) Myopic excimer treatments cause central flat- axial and elevation maps. This can be differentiated from ecta-
tening seen on the elevation map. (B) Myopic excimer treat- sia by questioning if the visual disturbance was noted immedi-
ments cause central thinning. (C) Myopic excimer treatments ately after surgery (suggesting decentration) or developed over
cause a more oblate cornea and positive spherical aberration. time (suggesting ectasia). (Reprinted with permission of Tracy
(Reprinted with permission from Trattler WB, Majmudar PA, Schroeder Swartz, O D, MS, FAAO.)
Luchs JI, Swartz TS. Corneal Handbook. Thorofare, NJ: SLACK
Incorporated; 2009.)
212 Ch a p t e r 8

C ONCLUSION 16. Rabinowitz YS. Keratoconus. Surv Ophthalmol. 1998;42(4):297-

319.
17. Sherwin T, Brookes NH. Morphological changes in keratoco-
nus: pathology or pathogenesis. Clin Experiment Ophthalmol.
To conclude, judicious use of corneal topography can be 2004;32(2):211-217.
useful for the clinician in cases that are likely to be missed 18. Parker J, Ko WW, Pavlopoulos G, Wolfe PJ, Rabinowitz YS,
on clinical examination only. There is a unique role of Feldman ST. Videokeratography of keratoconus in monozygotic
twins. J Refract Surg. 1996;12(1):180-183.
topography in cases with corneal ectasias and other patholo- 19. Rabinowitz YS. Videokeratographic indices to aid in screening for
gy, which can be well-supplemented by the use of wavefront keratoconus. J Refract Surg. 1995;11(5):371-379.
aberrometry in these cases. As combined platforms will 20. Maeda N, Klyce SD, Smolek MK. Comparison of methods for
evolve and topography devices will become more sensitive detecting keratoconus using videokeratography. Arch Ophthalmol.
and specific, accurate predictions on cases like post-LASIK 1995;113(7):870-874.
21. Rabinowitz YS, McDonnell PJ. Computer-assisted corneal topogra-
ectasia or keratoconus suspects will improve. phy in keratoconus. Refract Corneal Surg. 1989;5(6):400-408.
22. Twa MD, Parthasarathy S, Roberts C, Mahmoud AM, Raasch TW,
Bullimore MA. Automated decision tree classification of corneal

R EFERENCES shape. Optom Vis Sci. 2005;82(12):1038-1046.

23. Li X, Yang H, Rabinowitz YS. Longitudinal study of keratoconus
progression. Exp Eye Res. 2007;85(4):502-507.
24. Li X, Yang H, Rabinowitz YS. Keratoconus: classification scheme
1. Swartz T, Marten L, Wang M. Measuring the cornea: the lat- based on videokeratography and clinical signs. J Cataract Refract
est developments in corneal topography. Curr Opin Ophthalmol. Surg. 2009;35(9):1597-1603.
2007;18(4):325-333. 25. Karimian F, Aramesh S, Rabei HM, Javadi MA, Rafati N.
2. Wilson SE, Ambrosio R. Computerized corneal topography and its Topographic evaluation of relatives of patients with keratoconus.
importance to wavefront technology. Cornea. 2001;20(5):441-454. Cornea. 2008;27(8):874-878.
3. Naseri A, McLeod SD, Lietman T. Evaluating the human optical 26. Li X, Rabinowitz YS, Rasheed K, Yang H. Longitudinal study of
system: corneal topography and wavefront analysis. Ophthalmol the normal eyes in unilateral keratoconus patients. Ophthalmology.
Clin North Am. 2001;14(2):269-273, vii. 2004;111(3):440-446.
4. Mejía-Barbosa Y, Malacara-Hernández D. A review of methods 27. Rabinowitz YS. Tangential vs sagittal videokeratographs in the
for measuring corneal topography. Optom Vis Sci. 2001;78(4):240- “early” detection of keratoconus. Am J Ophthalmol. 1996;122(6):887-
253. 889.
5. Herrmann C, Ludwig U, Duncker G. Corneal topography. Analysis 28. Rabinowitz YS, Yang H, Brickman Y, et al. Videokeratography data-
of the corneal surface. Ophthalmologe. 2008;105(2):193-204. base of normal human corneas. Br J Ophthalmol. 1996;80(7):610-
6. Cairns G, McGhee CN. Orbscan computerized topography: attri- 616.
butes, applications, and limitations. J Cataract Refract Surg. 29. Bron AJ, Rabinowitz YS. Corneal dystrophies and keratoconus.
2005;31(1):205-220. Curr Opin Ophthalmol. 1996;7(4):71-82.
7. Rao SN, Raviv T, Majmudar PA, Epstein RJ. Role of Orbscan II in 30. Lin ML, Jhanji V, Vajpayee RB. Keratoconus associated with the
screening keratoconus suspects before refractive corneal surgery. Pierre Robin sequence. Cont Lens Anterior Eye. 2010;33(1):41-42.
Ophthalmology. 2002;109(9):1642-1646. 31. Cremona FA, Ghosheh FR, Rapuano CJ, et al. Keratoconus associ-
8. Giráldez-Fernández MJ, Díaz-Rey A, García-Resua C, Yebra- ated with other corneal dystrophies. Cornea. 2009;28(2):127-135.
Pimentel-Vilar E. Diurnal variations of central and paracen- 32. Sherwin JC, Hewitt AW, Ruddle JB, Mackey DA. Genetic isolates in
tral corneal thickness and curvature. Arch Soc Esp Oftalmol. ophthalmic diseases. Ophthalmic Genet. 2008;29(4):149-161.
2008;83(3):183-191. 33. Simon GJ, Grinbaum A. Anhidrotic ectodermal dysplasia with
9. Martin R, de Juan V, Rodríguez G, Cuadrado R, Fernandez I. spontaneous corneal perforation and keratoconus. Ophthalmic Surg
Measurement of corneal swelling variations without removal of Lasers Imaging. 2004;35(6):495-498.
the contact lens during extended wear. Invest Ophthalmol Vis Sci. 34. Georgiou T, Funnell CL, Cassels-Brown A, O’Conor R. Influence of
2007;48(7):3043-3050. ethnic origin on the incidence of keratoconus and associated atopic
10. Asano-Kato N, Toda I, Sakai C, et al. Pupil decentration and iris disease in Asians and white patients. Eye (Lond). 2004;18(4):379-
tilting detected by Orbscan: anatomic variations among healthy 383.
subjects and influence on outcomes of laser refractive surgeries. 35. Kim J, Netto MV. Keratoconus associated with hyperimmunoglobu-
J Cataract Refract Surg. 2005;31(10):1938-1942. lin E syndrome. Cornea. 2004;23(1):93-96.
11. Asensio I, Rahhal SM, Alonso L, Palanca-Sanfrancisco JM, 36. Hughes AE, Dash DP, Jackson AJ, Frazer DG, Silvestri G. Familial
Sanchis-Gimeno JA. Corneal thickness values before and after oxy- keratoconus with cataract: linkage to the long arm of chromosome
buprocaine 0.4% eye drops. Cornea. 2003;22(6):527-532. 15 and exclusion of andidate genes. Invest Ophthalmol Vis Sci.
12. Maldonado MJ, López-Miguel A, Nieto JC, Cano-Parra J, Calvo 2003;44(12):5063-5066.
B, Alió JL. Reliability of noncontact pachymetry after laser in situ 37. Iqbal A, Jan S, Babar TF, Khan MD. Corneal complications of ver-
keratomileusis. Invest Ophthalmol Vis Sci. 2009;50(9):4135-4141. nal catarrh. J Coll Physicians Surg Pak. 2003;13(7):394-397.
13. Read SA, Collins MJ. Diurnal variation of corneal shape and thick- 38. Incorvaia C, Parmeggiani F, Costagliola C, Perri P, Tittoni M,
ness. Optom Vis Sci. 2009;86(3):170-180. Sebastiani A. Congenital adrenal hyperplasia due to 21-hydroxylase
14. Nilforoushan MR, Speaker M, Marmor M, et al. Comparative deficiency associated with bilateral keratoconus. Am J Ophthalmol.
evaluation of refractive surgery candidates with Placido topography, 2003;135(4):557-559.
Orbscan II, Pentacam, and wavefront analysis. J Cataract Refract 39. Igarashi S, Makita Y, Hikichi T, Mori F, Hanada K, Yoshida A.
Surg. 2008;34(4):623-631. Association of keratoconus and Avellino corneal dystrophy. Br J
15. Krachmer JH, Feder RS, Belin MW. Keratoconus and related Ophthalmol. 2003;87(3):367-368.
noninflammatory corneal thinning disorders. Surv Ophthalmol. 40. Gusek-Schneider GC, Langenbucher A, Seitz B. Association of kera-
1984;28(4):293-322. toconus and Möbius’ syndrome. J Pediatr Ophthalmol Strabismus.
2001;38(1):47-48.
 To p o gra p hy in Co r n e a l D ise a se s 213

41. Lichter H, Loya N, Sagie A, et al. Keratoconus and mitral valve 64. Ertan A, Karacal H, Kamburoğlu G. Refractive and topographic
prolapse. Am J Ophthalmol. 2000;129(5):667-668. results of transepithelial cross-linking treatment in eyes with intacs.
42. Koçak Altintas AG, Gül U, Duman S. Bilateral keratoconus associ- Cornea. 2009;28(7):719-723.
ated with Hashimoto’s disease, alopecia areata and atopic kerato- 65. Baillif S, Garweg JG, Grange JD, Burillon C, Kodjikian L.
conjunctivitis. Eur J Ophthalmol. 1999;9(2):130-133. Keratoglobus: review of the literature. J Fr Ophtalmol.
43. Macsai M, Maguen E, Nucci P. Keratoconus and Turner’s syndrome. 2005;28(10):1145-1149.
Cornea. 1997;16(5):534-536. 66. Cameron JA. Keratoglobus. Cornea. 1993;12(2):124-130.
44. Wilhelmus KR. Keratoconus and progressive cone dystrophy. 67. Karabatsas CH, Cook SD. Topographic analysis in pellucid margin-
Ophthalmologica. 1995;209(5):278-279. al corneal degeneration and keratoglobus. Eye (Lond). 1996;10(Pt
45. Elder MJ. Leber congenital amaurosis and its association with 4):451-455.
keratoconus and keratoglobus. J Pediatr Ophthalmol Strabismus. 68. Rumelt S, Rehany U. Surgically induced keratoglobus in pellucid
1994;31(1):38-40. marginal degeneration. Eye (Lond). 1998;12(Pt 1):156-158.
46. Ascaso FJ, Del Buey MA, Huerva V, Latre B, Palomar A. Noonan’s 69. Patel SP, Sajnani MM, Pineda R. Posterior polymorphous dystro-
syndrome with keratoconus and optic disc coloboma. Eur J phy and keratoglobus in a child. J Pediatr Ophthalmol Strabismus.
Ophthalmol. 1993;3(2):101-103. 2010;8:1-3.
47. Stokes DW, Parrish CM. Axenfeld’s anomaly associated with 70. Harissi-Dagher M, Dana MR, Jurkunas UV. Keratoglobus in
Down’s syndrome. Cornea. 1992;11(2):163-164. association with posterior polymorphous dystrophy. Cornea.
48. Fontes BM, Ambrósio R Jr, Salomão M, Velarde GC, Nosé W. 2007;26(10):1288-1291.
Biomechanical and tomographic analysis of unilateral keratoconus. 71. Ku JY, Grupcheva CN, Fisk MJ, McGhee CN. Keratoglobus and
J Refract Surg. 2009;24:1-5. posterior subcapsular cataract: surgical considerations and in vivo
49. Schweitzer C, Roberts CJ, Mahmoud AM, Colin J, Maurice-Tison S, microstructural analysis. J Cataract Refract Surg. 2004;30(1):237-
Kerautret J. Screening of forme fruste keratoconus with the ocular 242.
response analyzer. Invest Ophthalmol Vis Sci. 2010;51(5):2403- 72. Elder MJ. Leber congenital amaurosis and its association with
2410. keratoconus and keratoglobus. J Pediatr Ophthalmol Strabismus.
50. Yildiz EH, Diehl GF, Cohen EJ, Hammersmith KM, Laibson PR, 1994;31(1):38-40.
Rapuano CJ. Demographics of patients older than 50 years with 73. Nelson ME, Talbot JF. Keratoglobus in the Rubinstein-Taybi syn-
keratoconus. Eye Contact Lens. 2009;35(6):309-311. drome. Br J Ophthalmol. 1989;73(5):385-387.
51. Lema I, Suárez AI, Díez-Feijoo E. Unilateral keratoconus: video- 74. McClellan KA, Billson FA. Acute hydrops in keratoglobus. Arch
keratography and Orbscan study—optical correction. Eye Contact Ophthalmol. 1987;105(10):1432-1433.
Lens. 2009;35(1):15-19. 75. Gupta VP, Jain RK, Angra SK. Acute hydrops in keratoglobus with
52. Lema I, Romero P, Mato JL, Feijóo ED. Corneal descriptive indi- vernal keratoconjunctivitis. Indian J Ophthalmol. 1985;33(2):121-
ces in the fellow eye of unilateral keratoconus. Eye Contact Lens. 123.
2009;35(2):65-68. 76. Kiire C, Srinivasan S. Management of bilateral acute hydrops sec-
53. Li X, Rabinowitz YS, Rasheed K, Yang H. Longitudinal study of ondary to keratoglobus with perfluoroethane (C2F6) pneumodesce-
the normal eyes in unilateral keratoconus patients. Ophthalmology. metopexy. Clin Experiment Ophthalmol. 2009;37(9):892-894.
2004;111(3):440-446. 77. Maguire LJ, Klyce SD, McDonald MB, Kaufman HE. Corneal
54. Holland DR, Maeda N, Hannush SB, et al. Unilateral keratoconus. topography of pellucid marginal degeneration. Ophthalmology.
Incidence and quantitative topographic analysis. Ophthalmology. 1987;94(5):519-524.
1997;104(9):1409-1413. 78. Lee BW, Jurkunas UV, Harissi-Dagher M, Poothullil AM, Tobaigy
55. Rabinowitz YS. Diagnosing keratoconus and patients at risk. Cat FM, Azar DT. Ectatic disorders associated with a claw-shaped
Refract Surg Today. May 2007. Accessed on 15th May 2010 from pattern on corneal topography. Am J Ophthalmol. 2007;144(1):154-
http://bmctoday.net/crstoday/2007/05/article.asp?f= CRST0507_ 156.
15.php 79. Oie Y, Maeda N, Kosaki R, et al. Characteristics of ocular higher-
56. Lim L, Wei RH, Chan WK, Tan DT. Evaluation of keratoconus in order aberrations in patients with pellucid marginal corneal degen-
Asians: role of Orbscan II and Tomey TMS-2 corneal topography. eration. J Cataract Refract Surg. 2008;34(11):1928-1934.
Am J Ophthalmol. 2007;143(3):390-400. 80. Tang M, Shekhar R, Miranda D, Huang D. Characteristics of kerato-
57. Schlegel Z, Lteif Y, Bains HS, Gatinel D. Total, corneal, and inter- conus and pellucid marginal degeneration in mean curvature maps.
nal ocular optical aberrations in patients with keratoconus. J Refract Am J Ophthalmol. 2005;140(6):993-1001.
Surg. 2009;25(10 Suppl):S951-S957. 81. Seiler T, Koufala K, Richter G. Iatrogenic keratectasia after laser in
58. Sabesan R, Yoon G. Visual performance after correcting higher situ keratomileusis. J Refract Surg. 1998;14(3):312-317.
order aberrations in keratoconic eyes. J Vis. 2009;9(5):6.1-6.10. 82. Probst LE, Machat JJ. Mathematics of laser in situ keratomileusis
59. Nakagawa T, Maeda N, Kosaki R, et al. Higher-order aberrations for high myopia. J Cataract Refract Surg. 1998;24(2):190-195.
due to the posterior corneal surface in patients with keratoconus. 83. Guirao A. Theoretical elastic response of the cornea to refractive
Invest Ophthalmol Vis Sci. 2009;50(6):2660-2665. surgery: risk factors for keratectasia. J Refract Surg. 2005;21(2):176-
60. Alió JL, Shabayek MH. Corneal higher order aberrations: a method 185.
to grade keratoconus. J Refract Surg. 2006;22(6):539-545. 84. Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assess-
61. Vinciguerra P, Albè E, Trazza S, et al. Refractive, topographic, ment for ectasia after corneal refractive surgery. Ophthalmology.
tomographic, and aberrometric analysis of keratoconic eyes under- 2008;115:37-50.
going corneal cross-linking. Ophthalmology. 2009;116(3):369-378. 85. Randleman JB, Russell B, Ward MA, Thompson KP, Stulting
62. Piñero DP, Alió JL, El Kady B, Pascual I. Corneal aberrometric RD. Risk factors and prognosis for corneal ectasia after LASIK.
and refractive performance of 2 intrastromal corneal ring segment Ophthalmology. 2003;110:267-275.
models in early and moderate ectatic disease. J Cataract Refract 86. Amoils SP, Deist MB, Gous P, Amoils PM. Iatrogenic keratectasia
Surg. 2010;36(1):102-109. after laser in situ keratomileusis for less than -4.0 to -7.0 diopters of
63. Piñero DP, Alio JL, El Kady B, et al. Refractive and aberrometric myopia. J Cataract Refract Surg. 2000;26:967-977.
outcomes of intracorneal ring segments for keratoconus: mechani- 87. Klein SR, Epstein RJ, Randleman JB, Stulting RD. Corneal ectasia
cal versus femtosecond-assisted procedures. Ophthalmology. after laser in situ keratomileusis in patients without apparent preop-
2009;116(9):1675-1687. erative risk factors. Cornea. 2006;25:388-403.
214 Ch a p t e r 8

88. Randleman JB, Banning CS, Stulting RD. Corneal ectasia after 104. Goto T, Zheng X, Klyce SD, et al. Evaluation of the tear film sta-
hyperopic LASIK. J Refract Surg. 2007;23:98-102. bility after laser in situ keratomileusis using the tear film stability
89. Caster AI, Friess DW, Potvin RJ. Absence of keratectasia after analysis system. Am J Ophthalmol. 2004;137(1):116-120.
LASIK in eyes with preoperative central corneal thickness of 450 to 105. Mobillia EF, Kenyon KR. Contact lens-induced corneal warpage. Int
500 microns. J Refract Surg. 2007;23:782-788. Ophthalmol Clin. 1986;26:43-53.
90. Binder PS. Analysis of ectasia after laser in situ keratomileusis: risk 106. Wang X, McCulley JP, Bowman RW, Cavanagh HD. Time to reso-
factors. J Cataract Refract Surg. 2007;33:1530-1538. lution of contact lens-induced corneal warpage prior to refractive
91. Condon PI, O’Keefe M, Binder PS. Long-term results of laser in situ surgery. The CLAO Journal. 2002;28(4):169-171.
keratomileusis for high myopia: risk for ectasia. J Cataract Refract 107. Di Girolamo N, Coroneo MT, Wakefield D. UVB-elicited induction
Surg. 2007;33:583-590. of MMP-1 expression in human ocular surface epithelial cells is
92. Randleman JB. Post-laser in-situ keratomileusis ectasia: current mediated through the ERK1/2 MAPK-dependent pathway. Invest
understanding and future directions. Curr Opin Ophthalmol. Ophthalmol Vis Sci. 2003;44(11):4705-4714.
2006;17(4):406-412. 108. Dushku N, John MK, Schultz GS, Reid TW. Pterygia pathogenesis:
93. Rao SN, Epstein RJ. Early onset ectasia following laser in situ corneal invasion by matrix metalloproteinase expressing altered
keratomileusis: case report and literature review. J Refract Surg. limbal epithelial basal cells. Arch Ophthalmol. 2001;119(5):695-
2002;18(2):177-184. 706.
94. Randleman JB, Trattler WB, Stulting RD. Validation of the Ectasia 109. Oh JY, Wee WR. The effect of pterygium surgery on contrast
Risk Score System for preoperative laser in situ keratomileusis sensitivity and corneal topographic changes. Clin Ophthalmol.
screening. Am J Ophthalmol. 2008;145(5):813-818. 2010;4:315-319.
95. Binder PS, Trattler WB. Evaluation of a risk factor scoring system 110. Maheshwari S. Pterygium-induced corneal refractive changes.
for corneal ectasia after LASIK in eyes with normal topography. Indian J Ophthalmol. 2007;55(5):383-386.
J Refract Surg. 2010;26(4):241-250. 111. Avisar R, Loya N, Yassur Y, Weinberger D. Pterygium-induced
96. Ambrósio R Jr, Dawson DG, Salomão M, Guerra FP, Caiado AL, corneal astigmatism. Isr Med Assoc J. 2000;2(1):14-15.
Belin MW. Corneal ectasia after LASIK despite low preoperative 112. Shimazaki J, Tsubota K. Analysis of videokeratography after
risk: Tomographic and biomechanical findings in the unoperated, penetrating keratoplasty: topographic characteristics and effects
stable, fellow eye. J Refract Surg. 2010;19:1-6. of removing running sutures. Ophthalmology. 1997;104(12):2077-
97. Tuli SS, Iyer S. Delayed ectasia following LASIK with no risk 2084.
factors: is a 300-microm stromal bed enough? J Refract Surg. 113. Sarhan AR, Dua HS, Beach M. Effect of disagreement between
2007;23(6):620-622. refractive, keratometric, and topographic determination of astig-
98. Altan-Yaycioglu R, Pelit A, Akova YA. Comparison of ultrasonic matic axis on suture removal after penetrating keratoplasty. Br J
pachymetry with Orbscan in corneal haze. Graefes Arch Clin Exp Ophthalmol. 2000;84(8):837-841.
Ophthalmol. 2007;245(12):1759-1763. 114. Vinciguerra P, Epstein D, Albè E, et al. Corneal topography-guided
99. Boscia F, La Tegola MG, Alessio G, Sborgia C. Accuracy of Orbscan penetrating keratoplasty and suture adjustment: new approach for
optical pachymetry in corneas with haze. J Cataract Refract Surg. astigmatism control. Cornea. 2007;26(6):675-682.
2002;28(2):253-258. 115. Geggel HS. Arcuate relaxing incisions guided by corneal topog-
100. Gayton JL. Etiology, prevalence, and treatment of dry eye disease. raphy for postkeratoplasty astigmatism: vector and topographic
Clin Ophthalmol. 2009;3:405-412. analysis. Cornea. 2006;25(5):545-557.
101. Cho P, Douthwaite W. The relation between invasive and noninva- 116. Kovoor TA, Mohamed E, Cavanagh HD, Bowman RW. Outcomes of
sive tear break-up time. Optom Vis Sci. 1995;72(1):17-22. LASIK and PRK in previous penetrating corneal transplant recipi-
102. Iskander DR, Collins MJ. Applications of high-speed videokeratos- ents. Eye Contact Lens. 2009;35(5):242-245.
copy. Clin Exp Optom. 2005;88(4):223-231. 117. Vajpayee RB, Sharma N, Sinha R, Bhartiya P, Titiyal JS, Tandon
103. Goto T, Zheng X, Okamoto S, Ohashi Y. Tear film stability analy- R. Laser in-situ keratomileusis after penetrating keratoplasty. Surv
sis system: introducing a new application for videokeratography. Ophthalmol. 2003;48(5):503-514.
Cornea. 2004;23(8 Suppl):S65-S70.
Ch a p te r 9

To p o grap hy-
Guid e d Co ntact Le ns
Fitting
Shawna Hill Vanderhoof, OD, FAAO ; Randy Kojima, OD, FAAO, FOAA;
Jason Jedlicka, OD, FAAO ; Matthew Lampa, OD, FAAO ;
and Tracy Schroeder Swartz, OD, MS, FAAO

T
he topographer has become an indispensable tool to staff comprehensively on the need to take good photos. This
understand the anterior surface shape of the cornea means slow, steady movements with the instrument, a com-
and give much information about corneal power, cur- fortably seated patient, wide fissure size, and an even tear
vature, elevation, and disease. It is also an incredible asset to film. Always check the Placido rings prior to finishing the
our contact lens practice for the fitting of specialty contact mapping process to be sure a good image has been taken.
lenses. Traditionally, topography gave us a starting point in Take multiple images when building GP contact lenses so
choosing the initial diagnostic lens. Now, using topography there are many to choose from if 1 or 2 captures should be
and its contact lens fitting software, we can actually design less than ideal.
a gas-permeable (GP) lens more efficiently. A software-gen- Most cameras require light to bounce off the object, and
erated GP lens can be created to compare with an actual trial this reflection is recorded as an image. A corneal topogra-
on the eye of the same parameters. These programs can help pher bounces its own light (the Placido rings) off the hydrau-
in the initial lens design selection and aid in the efficiency lic fluid layer on the corneal surface and records the image
of choosing the optimal parameters for each individual eye. of the Placido reflection on the tear film.1 Considering that
This is infinitely beneficial when dealing with diseased or we must capture a photo reflected from the corneal surface,
asymmetrical eyes, which are challenging to start. it becomes imperative that the fluid covering the surface
The software is not limited to topographers or GPs alone; is smooth, even, and devoid of tear break-up or dryness.2,3
we will also discuss specialty soft and GP contact lens Figure 9-1 exhibits these imperfections caused by tear film
design cases from aberrometers to tomographers. break-up, dryness, or inconsistencies in the fluid layer in
the Placido reflection. The instrument determines the shape
and elevation of the eye by analyzing the ring reflection off

R AW I MAGES TO ANALYZING the corneal surface. If any number of these rings should
distort, break, or collapse on one another, the accuracy of
M APS the instrument will be hindered.
When analyzing the maps, use an axial interpretation to
understand corneal power or astigmatism and tangential to
For this instrument to provide the practitioner with valu- measure shape or curve at a particular point. To appreci-
able information, a good capture is required. Train your ate the acuity that is derived from the anterior surface, a
Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical
215 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 215 -230)
© 2012 SLACK Incorporated
216 Ch a p t e r 9

Figure 9-1. Note the distorted rings inside the red circle. This would
result in analysis error (extrapolation error). A quality topography cap-
ture should exhibit rings that look even and parallel, as most appear
outside the red circle.

Figure 9-2. Sagittal depth.

refractive power map can be helpful to understand asym- asphericity, and lens diameter/corneal diameter.6 Scleral
metries within the virtual pupil. Last, an elevation map is contact lenses completely vault the cornea and rest upon the
the best way to discern where the peaks and depressions in bulbar conjunctiva. The difference in sagittal depth between
the cornea exist. Topographers use the elevation display to the contact lens and the eye determines this relationship.
calculate the height differences across the cornea by plac- Using anterior segment OCT at the Pacific University
ing a reference sphere and then displaying the elevation of College of Optometry, the corneal-scleral tangential angle
the cornea above or below the reference surface across the between 10.00 mm and 15.00 mm (the limbal angle) was
entire map display. The reference sphere is the curve that measured as well as the angle from 15.00 mm to 20.00 mm
aligns with the most number of points on the cornea. Each (the scleral angle).5 This would suggest that toric and/or
instrument has built into it the parameters of many contact quadrant-specific scleral designs could lead to better fit-
lenses and then takes the difference between the known sur- ting characteristics and possibly better lens comfort. While
face of the contact lens (based on the individual parameters keratometry values might be the same between patients, a
for that lens) and the elevation display. smaller eye will result in significantly less sagittal depth
Some topographers also have sagittal depth information. and marked difference in the limbal/scleral angles. This is
Corneal sagittal depth (z-value) is defined as a measure- in contrast to corneal GP lenses for which you determine the
ment in mm or µm of the distance between the geometric fit by the base curve-to-keratometry measurement because
center of the cornea and the intersection of a specified they are fit to align with the cornea.
chord length (y-value) 4 as seen in Figure 9-2. Scleral lens
fitting is based on the sagittal depth as the sclera cannot be
described in curvature or shape factors as the cornea can.
Van der Worp and colleagues found the shape of the limbus TH E “RULES” OF G AS-
and anterior sclera to be tangential in shape, showing that
the sclera is nonspherical and nonrotationally symmetric P ERMEABLE LENS FITTING
in nature.5 Factors that affect the sagittal depth for the lens
and cornea, respectively, include base curve radius/kera- The following set of rules to guide the fitting of GP lens-
tometry (K) values, peripheral curves/peripheral corneal es will facilitate fitting and minimize the time required7:
 To p o gra p hy-G u id e d Co n t a c t Le n s Fit t in g 217

• Diam eter selection: The optimum diameter should

be large enough to stabilize the lens along the
horizontal meridian while smaller than the vis-
ible iris diameter to allow for 1 mm of movement
vertically.
• A pical clearance (base curvature): The central base
curve should be steep enough to clear the corneal
apex. The base curve radii you choose should cre-
ate an apical clearance of bet ween 20 and 30 µm.
Apical clearances larger than 40 µm may produce
apical steepening and spectacle blur. When this
occurs, the pattern will exhibit the appearance of
excessive central clearance.
• H orizontal alignm ent : The GP lens should be
steep enough to land midperipherally along the
horizontal meridian. Figure 9-3. An ideal GP fit with apical clearance, landing at
• Vertical channel of tears: The lens should exhibit 3 and 9 o’clock and a channel of tear layer along the vertical
unobstructed movement along the vertical merid- meridian.
ian (Figure 9-3).

TOPOGRAPH Y-G UIDED

FITTING
Wave Con t act Len s System
The Keratron Scout (Optikon 2000, Rome, Italy) and
Medmont E300 (Medmont International Pty Ltd, Vermont,
Melbourne, Australia) corneal topographers use the Wave
contact lens design software program. The Wave design
program uses the topography output from the Keratron/
Medmont topographer to design virtually any type of lens
desired. The lens designs are broken down into 3 categories:
rotationally symmetric (RSym), geometrically symmetric
(GSym), or free form (FForm). Using these categories, it
is possible to design soft or GP single vision, multifocal
(center near, center distance, or segmented), aspheric toric,
and aspheric bitoric lenses, as well as reverse geometry GP
designs.
The lens design software is powerful enough to com- Figure 9-4. WAVE design box.
bine elements of design to create virtually any lens design
imaginable. GP lens diameters can be made up to 15.0 mm,
while soft lens designs range from 13.0 to 15.0 mm. Preset can create an asymmetric correction to enhance vision in
controls allow for auto designs of standard lenses, myopic eyes with higher-order aberrations (HOAs) or asymmetric
ortho-k lenses, or hyperopic ortho-k lenses. astigmatism. At this point, the initial lens design will be
Once topography has been obtained, the user chooses complete.
the type of lens to be designed (soft or GP) and the WAVE The lens design box has various aspects that can be
program begins to design the lens. As the lens is being accessed. The upper region of the box includes the lens
designed, the lens design screen will appear (Figure 9-4). material, the lens diameter, and the lens clearance profile.
Once the initial design is complete, the program will These are all customizable. This profile is set up so as to
ask for the spectacle prescription to be entered. This will allow the user to increase or decrease lens clearance in any
allow the software to create the lens with the correct power, area of the lens, as well as to change the widths of any of
whether it be spherical or toric. In addition, a power map the respective curves.
218 Ch a p t e r 9

The midregion of the lens design box is where the

patient’s prescription is listed, as well as the option for a
power map for asymmetric power to be cut into the lens,
and the box to click to enter any over-refraction found over
a current design. The bottom third of the design box has an
area that describes the lens thickness in the center and edge
(either of which can be altered by the designer), the lens
power, the actual base curve at the region selected by the
user, and the various lens views available, which include a
simulated fluorescein pattern, the back or front of the lens
surface viewed from an axial or curvature perspective, as
well as other options. A quad view option allows for mul-
tiple views to be seen all at one time.
The initial lens design may be an RSym, GSym, or
Figure 9-5. Topography for DS, preorthokeratology O D.
FForm design, chosen by the program based upon the type
of corneal shape. The user can alter the type of lens after
the initial design. For fairly regular, spherical, or low astig-
matic corneas, an RSym design is usually acceptable. For
highly toric corneas, it may be desirable to have the lens in
a GSym design, which allows for “normal” toricity correc-
tion. In cases of irregular corneas, or where other highly
specific characteristics are desired (such as a spherical base
curve with toric peripheral curves), an FForm design may
be preferred. The FForm design allows for asymmetric lens
designs to be achieved using the flexibility of 24 zones of
curve modification.
From there, further adjustments to the lens can be made
using one of the many modification tools. There are drop-
down menus that allow for steepening or flattening the
base curve, midperiphery, or peripheral curves; changing
Figure 9-6. Topography for DS, preorthokeratology O S.
the diameter; or adding an over-refraction. If a lens design
needs modifying after dispensing, the lens design box can
be reopened, modifications can be made to the lens design,
and a new lens can be ordered with the modifications made. His corneal topography in each eye (Figures 9-5 and 9-6)
All lens designs are transmitted via e-mail to the main lab, showed enough curvature for orthokeratology, but with his
where they are fed directly into the lathe for production. correction being roughly -5.00, the patient and his mother
The Wave design system, while effective for many types were cautioned that full correction may not be attainable.
of lens fits, is particularly helpful in difficult cases where Because of the higher correction and the potential for the
standard lens fitting sets are less effective, such as reverse correction to increase due to his young age, flexibility of
geometry lens designs, orthokeratology, or for post-surgical planning to optimize parameters and change diameter and
eyes. optic zone sizes as needed in the future was desired. Rather
than fit him in a standard orthokeratology lens design, a
Ort h okeratology Applicat ion Wave designed lens was chosen.
Because he had little astigmatism, an RSym design was
One of the most useful applications of the Wave system an acceptable lens design. A larger lens diameter (11.0 mm)
is the designing of lenses for orthokeratology or corneal was chosen to accommodate his larger pupil size and to
reshaping. Because you have the ability to control all lens try to increase the treatment zone as much as possible. The
parameters, you can sometimes fit patients who would oth- lens designs for each eye are shown in Figures 9-7 and 9-8.
erwise get less-than-optimal results with one of the many Using this design, he was able to obtain uncorrected vision
lens design systems for orthokeratology available commer- of 20/20 in each eye. His topographies at 1-year post-initi-
cially. This was the case for DS, a 12-year-old patient who ation of treatment are shown in Figures 9-9 and 9-10. The
was already exhibiting significant myopia at this early age. topographies demonstrate excellent centration of correction
DS presented for examination in May 2004 with an inter- and minimal astigmatic correction. DS continues in ortho-
est in corneal reshaping. After a complete examination, we keratology now 6 years later. His correction has increased
determined a prescription of: by less than 1 D in each eye during the past 6 years with
OD: -5.50-0.25 x 20 20/20 maintenance of 20/20 BCVA.
OS: -5.00-0.25 x 180 20/20
 To p o gra p hy-G u id e d Co n t a c t Le n s Fit t in g 219

Figure 9-9. O D topography for DS, 1-year post-initiation of

orthokeratology.

Figure 9-7. O D lens design box for DS. Note lens param-
eters.

Figure 9-10. O S topography for DS, 1-year post-initiation of

orthokeratology.

Reverse Geom et ry Len ses

The postsurgical eye that is steep centrally but even
steeper in the midperiphery is a challenging fit. This is
most common in those eyes that are post-penetrating kera-
toplasty for keratoconus. ZB is a 54-year-old man who had
a history of keratoconus and corneal transplants in 1991 in
the right eye and 1993 in the left eye. He subsequently wore
soft toric lenses for several years, but found the vision to be
deteriorating.
His vision with soft toric lens correction was 20/50 in
the right eye and 20/40 in the left. Pinhole did not improve
the vision. His refraction was
R: -9.00 +5.50 x 125 20/50
L: -8.75 +4.50 x 55 20/40
His grafts were clear in each eye and were free of signs
of rejection. Corneal topography demonstrated a highly
astigmatic central area with significant steepening in the
midperiphery (Figures 9-11 and 9-12).
Figure 9-8. O S lens design box for DS. Note lens param- After explaining the status of his prescription, vision,
eters. and corneal shape, it was decided that trying a GP lens
220 Ch a p t e r 9

Figure 9-11. ZB’s corneal topography O D.

Figure 9-13. ZB’s WAVE lens design O D.

Figure 9-12. ZB’s corneal topography O S.

might offer better vision than the soft lenses he had been
using. A GP lens fitting was undertaken, which produced
visual acuities approaching 20/20, but with an unstable fit.
What was needed was a reverse geometry lens design, but
one with a smaller optic zone and steeper base curves than
most commercially available reverse geometry lens design
options. To achieve the desired fit, an FForm Wave lens was
designed for each eye.
The Wave lens design for each eye is shown in Figures
9-13 and 9-14. For the right eye, a lens diameter of 10.7 was
chosen, while the left eye, due to the steeper midperiphery,
was fitted in a 10.3 diameter. The lenses were dispensed,
and ZB was able to adapt to the lenses and achieve 20/20- Figure 9-14. ZB’s WAVE lens design O S.
2 vision in each eye. Minor adjustments were made to the
peripheral curves in each eye to facilitate tear exchange,
but the end result was an acceptable fit and vision in each but recently felt his vision deteriorating. He had seen several
eye. The actual lenses on the eye are shown in Figures 9-15 eye care providers seeking improved vision, and though the
and 9-16. lenses he had tried had provided better visual acuity, he
Another lens fitting challenge is the post-radial kera- could not tolerate those lenses. His entering acuity uncor-
totomy (RK) patient. These patients often have somewhat rected was 20/60-2 in each eye. All entrance tests were
sensitive corneas as well as a small, very flat area centrally normal. His refraction was:
with a steeper midperiphery. LJ is a 42-year-old man with a OD: +2.00 +3.25 x 90 20/25-2
history of RK in 1993. He had done well for several years, OS: -1.75 +1.75 x 75 20/30
 To p o gra p hy-G u id e d Co n t a c t Le n s Fit t in g 2 21

Figure 9-17. LJ’s corneal topography O D.

Figure 9-15. ZB’s WAVE lens O D.

Figure 9-18. LJ’s corneal topography O S.

A diameter of 11.5 was chosen for each eye for stability of

Figure 9-16. ZB’s WAVE lens O S. lens fit. The lens design boxes are shown in Figures 9-19
and 9-20.
The lenses were dispensed, and the visual acuity with
Slit-lamp examination revealed 8 radial scars from his correction was recorded at 20/25+1 in each eye and
previous RK, as well as 2 arcuate scars in each eye from 20/20-1 in both eyes. The lenses fit well and the patient’s
an astigmatic keratectomy. The temporal aspect of the initial comfort was acceptable. The lens fit is shown in
right cornea was significant for mild evidence of wound Figures 9-21 and 9-22. The patient adapted to the lenses,
gape between the arcuate cut and one of the radial cuts. was happy with the quality of vision, and was able to toler-
All other ocular findings were normal. Corneal topography ate lens wear 10 hours daily, which was acceptable to LJ.
demonstrated a fairly typical post-RK map in each eye, with
a small flat central area surrounded by a steeper midperiph-
ery (Figures 9-17 and 9-18). Med m on t E3 0 0
His contact lens history included previous fittings in
The Medmont E300 corneal topographer uses a simu-
hybrid lenses, semi-scleral GPs, as well as various corneal
lated fluorescein display for evaluating the fit of GP contact
GP lenses. His main problem with all previous lenses was
lenses. The software allows the selection of many generic
the development of photosensitivity with lens wear. He had
and specific manufacturers’ lens designs, and the param-
not worn any lenses prior to his initial visit in more than
eters are calculated based on the captured topography. The
2 years. After discussing his options, which included recon-
manifest refraction will also be specified here in order for
sideration of surgical options as well as various contact
the software to calculate the required contact lens power. It
lens designs, we decided to proceed with an RSym OD
can predict residual astigmatism by comparing the corneal
and GSym OS Wave lens design. The lenses were designed
and manifest refractive astigmatism.
based upon his corneal maps and his spectacle prescription.
222 Ch a p t e r 9

Figure 9-21. LJ’s WAVE lens design O D.

Figure 9-19. LJ’s Wave lens design O D. Note that

the lens is an RSym design, and the lens view in the
lower right area of the lens design box demonstrates
a spherical base curve lens, with a resulting toric
fluorescein pattern.

Figure 9-22. LJ’s WAVE lens design O S.

In the fluorescein display, the simulated contact lens can

be moved over the cornea to determine how the fit varies
as the lens repositions. The N and T on the sides denote the
nasal (N) or temporal (T) orientation. Dragging the lens
with the mouse to reposition the lens is left at that location
and a new simulated fluorescein pattern is generated.
RM, a 24-year-old male graduate student, presented
Figure 9-20. LJ’s Wave lens design O S. Note that the to the office complaining of blurry vision in both eyes at
lens is a GSym design, and the lens view in the lower distance. He was interested in contact lenses as he had
right area of the lens design box demonstrates a toric
only worn spectacles in the past. His last eye exam was
base curve lens, with a resulting spherical fluorescein
pattern. 1 year ago. His medical and ocular history is unremarkable
except for seasonal allergies. After a complete examination,
we determined the refraction and keratometry as listed in
Lens diameter, base curvature, and peripheral curvatures Table 9-1. His corneal topography in each eye (Figures 9-23
are customizable. The changing of some values will auto- and 9-24) showed high regular with-the-rule astigmatism,
matically cause other values to be recalculated. Changing making him a good candidate for a bitoric GP contact lens.
them manually and unchecking the auto-update option can We educated the patient on the options, and the patient
override this function. decided to proceed.
 To p o gra p hy-G u id e d Co n t a c t Le n s Fit t in g 223

TABLE 9-1.

Bitoric Gas Permeable Fitting Clinical Data

OD OS
Refraction Plano-3.75x014, 20/20 +0.25-4.75x169, 20/20
Keratometry 41.50 @ 012 / 45.00 @ 102 41.25 @ 179 / 45.50 @ 089
Initial lens parameters 40.75/44.50D 40.75/44.50D
8.30/7.60 8.30/7.60
+0.50/-3.00 +0.75/-3.00
9.5mm 9.5mm
Adjusted lens parameters 41.50/44.25 41.25/45.00
8.13/7.63 8.18/7.50
-0.25/-3.00 +0.25/-3.75
9.5mm 9.5mm

Figure 9-23. RM’s corneal topography O D showing high with- Figure 9-25. RM’s simulated fluorescein pattern O D.
the-rule astigmatism.

Computer software suggested the parameters listed in

Table 9-1 based on topography. The simulated fluorescein
patterns of these lenses can be seen in Figures 9-25 and
9-26. Note that the flat meridians in both eyes have less than
20 to 30 µm of clearance. Therefore, we steepened the flat
meridian of each eye, causing a change in the lens power.
More minus was added to each lens, -0.25 for every 0.25
of curvature change. The steep meridian was steepened
by 0.50 in the left eye to improve apical clearance. There
was no residual astigmatism calculated in this case. The
adjusted lens parameters are listed in Table 9-1.
The contact lenses were dispensed to the patient (Figures
9-27 and 9-28). The patient was able to achieve 20/20 vision
with the bitoric GPs and was able to adapt to the lenses after
building up his wear time over 2 weeks.
Figure 9-24. RM’s corneal topography O S showing high with- BW, a 31-year-old male contractor, presented to the
the-rule astigmatism. office complaining of blurred distance vision, in the left
eye more than the right eye, especially while driving. He is
224 Ch a p t e r 9

Figure 9-26. RM’s simulated fluorescein pattern O S. Figure 9-29. BW’s corneal topography O D showing keratoco-
nus. Apical radius is 53.00 D (6.35).

Figure 9-27. RM’s Medmont lens design O D.

Figure 9-30. BW’s corneal topography O S showing more

advanced keratoconus. Apical radius is 65.00 (5.20).

showed keratoconus, more in the left eye than the right eye.
After educating BW on his options, we decided that GPs
were his best option.
Computer software suggested the parameters listed in
Table 9-2 based on topography. The simulated fluores-
cein patterns of these lenses can be seen in Figures 9-31
and 9-32. The keratoconus bi-aspheric (KBA) lens from
Essilor Contact Lens is a 10.2-mm diameter GP design
with a 9.2-mm back surface optic zone. The large optic
zone combined with a custom aspheric back surface allows
Figure 9-28. RM’s Medmont lens design O S. for centration over the pupil and, thus, reduced flare and
glare.8 Also, the KBA enables you to match its eccentricity
(e-value) to that of the cornea. Aspheric lenses have a vary-
interested in contact lenses and was diagnosed with kerato- ing base curve defined by the back optic zone eccentricity.
conus at his last eye exam 2 months ago. BW’s uncorrected Corneas that flatten at a significant rate from their
vision was 20/25 in the right eye and 20/50 in the left eye. geometric center outward have high e-values, and those
Slit-lamp examination showed clear corneas in both eyes. that flatten mildly from the geometric center have low
His corneal topography in each eye (Figures 9-29 and 9-30) e-values.6
 To p o gra p hy-G u id e d Co n t a c t Le n s Fit t in g 225

TABLE 9-2.

Keratoconus Bi-Aspheric (KBA) Fitting Clinical Data

OD OS
Refraction +1.00-3.00x043, 20/25 Plano-3.00x142, 20/50
Keratometry 42.50 @ 019 / 47.75 @ 109 48.00 @ 166 / 61.50 @ 076
Initial lens parameters KBA /45.63 (7.40) / 10.2 / 1.0 ecc KBA / 50.38 (6.70) / 10.2 / 1.2 ecc
Adjusted lens parameters KBA / 45.63 (7.40) / -4.25 / 10.2 / 1.0 ecc KBA / 52.00 (6.50) / -10.50 / 10.2 / 1.3 ecc

Figure 9-31. BW’s simulated fluorescein pattern O D.

Note the GP is approximately resting at 3 and 9 o’clock.

Figure 9-33. BW’s Medmont lens design O D, exactly as simu-

lated in Figure 9-31.

Figure 9-32. BW’s simulated fluorescein pattern O S.

Note the GP is approximately resting at 3 and 9 o’clock.

The right lens (see Figure 9-31) shows 30 µm of apical

clearance and a good peripheral fit as well. The left lens
(see Figure 9-32) shows approximately 20 µm of apical
clearance, which rapidly decreases to touch, only 1 mm
temporal. Therefore, we steepened the base curve, thereby
increasing eccentricity as shown in Table 9-2.
Figure 9-34. BW’s Medmont lens design on left eye. Note
The contact lenses of the actual lenses ordered were dis- the improved fluorescein pattern centrally and increased tear
pensed to the patient (Figures 9-33 and 9-34). The patient profile inferiorly.
226 Ch a p t e r 9

TABLE 9-3.

Atlas Topography With the MasterFit Software

OD OS
Refraction -5.00 -1.50 x 009; 20/20 -3.00 -2.50 x 174; 20/20
Keratometry 41.75 @ 010 / 43.50 @ 100 43.00 @ 175 / 46.00 @ 085
Initial lens parameters 42.25 / -4.50 / 9.5 mm 43.25 / -3.00 / 9.8 mm
Adjusted lens parameters 42.50 / -4.50 / 9.5 mm 43.50 / -3.00 / 9.5 mm

was able to achieve 20/20 vision with the aspheric GPs and
was able to adapt to the lenses after building up his wear
time over the first 2 weeks. He also understands that refit-
ting will be necessary if the keratoconus progresses.

Zeiss At la s 9 0 0 0 a n d
Ma sterFit II Soft w a re
The Atlas 9000 topographer (Carl Zeiss Meditec, Jena,
Germany) and optional MasterFit II Contact Lens Software
enable the user to design GP contact lenses and to simulate
the resulting fluorescein pattern on the cornea.9 MasterFit
II features both topographic and keratometric-based lens-
fitting methods. Based on customizable fitting preferences,
Figure 9-35. FC’s corneal topography O D showing with-the-
MasterFit II will recommend the most appropriate lens
rule astigmatism.
type (including spherical, aspheric, front toric, back toric,
and bitoric designs) and will design a custom lens that best
achieves the desired fitting preferences. MasterFit II also
Her last eye exam was 2 years ago. Her ocular and medical
features region-specific databases of stock (finished) lenses
history is unremarkable. After a complete examination, we
that may be used as an alternative to creating custom lens
determined her refraction as described in Table 9-3.
designs.
Her corneal topography in each eye (Figures 9-35 and
The user may create multiple lens designs per topogra-
9-36) showed with-the-rule astigmatism, greater in the left
phy exam to simulate the effect of adjusting the available
eye. Computer software suggested the parameters based on
lens parameters on the fluorescein pattern. After finalizing
topography, which are listed in Table 9-3. The simulated
the lens design, the user may print the lens design using the
fluorescein patterns of these lenses can be seen in Figures
order form templates provided, or the order form may be
9-37 and 9-38. A Computer Aided Design (CAD) GP lens
saved as a PDF and sent by e-mail to the preferred labora-
from Valley Contax (Springfield, OR) was selected due to
tory.
her low to moderate with-the-rule astigmatism. This lens
The MasterFit II software also displays tear film thick-
has a spherical optical zone and an aspheric periphery.
ness profile graphs. To the right of the fluorescein map is
The right eye shows a lens 0.50 D steeper than flat K,
the vertical (90 degrees) tear film thickness profile graph,
and the apical clearance is right under the 20-µm mark so
which plots the tear layer thickness (µm; solid green area)
the final lens is steepened by 0.25 (see Figure 9-37). The
along the corneal surface. Below the fluorescein map are
left eye shows a lens virtually on flat K with the horizontal
the horizontal (180 degrees) tear film thickness profile
and vertical tear film thickness less than adequate (see
graph, which plots the tear layer thickness (µm) along the
Figure 9-38). The left lens was also steepened. Actual lens
corneal surface, and white profile meridian (white graph)
parameters are listed in Table 9-3. The contact lenses of the
on the fluorescein map.
actual lenses ordered were dispensed to the patient (Figures
This case demonstrates the software in a patient having
9-39 and 9-40). The patient was able to see 20/20 with her
with-the-rule astigmatism. A 36-year-old woman presented
GPs. If there was residual with-the-rule astigmatism in
to the office complaining of blur at distance in both eyes.
either eye, a bitoric could be ordered.
 To p o gra p hy-G u id e d Co n t a c t Le n s Fit t in g 2 27

Figure 9-36. FC’s corneal topography O S showing with-the-

rule astigmatism. Figure 9-39. FC’s MasterFit CAD lens design O D.

Figure 9-37. FC’s simulated fluorescein pattern O D. Note the

Flat K is 41.87 D.

Figure 9-40. FC’s MasterFit CAD lens design O S with appro-

priate amount of tears in the vertical channel.

Finally, some systems incorporate aberrometry with

topography to guide contact lens fitting. An irregular
topography results in irregular aberrations. Correction of
the aberrations may yield improvement in vision.

Ab er rat ion Cor rect ion

Con t act Len ses
Wavetouch Hyd rogel Con tact Len ses
Figure 9-38. FC’s simulated fluorescein pattern O S. Note the Wavetouch hydrogel lenses use wavefront aberrometry
Flat K is 43.12 D.
to correct for both refraction and optical aberration as mea-
sured by a certified aberrometer, such as the iTrace Visual
Function Analyzer (Tracey Technologies, Houston, TX).
228 Ch a p t e r 9

TABLE 9-4.

Wavetouch Hydrogel Lens Parameters

Base curve supplied MDN (median)
in the WaveTouch STP (steep)
Acquisition set*
Diameter 14.7mm
Material Methafilcon A, 55% water
content
Daily wear schedule Supplied for the patient as
quarterly replacement lens
Figure 9-41. Aberrometry measurement is performed over the after the initial fit is finalized
contact lens. The system uses the dots to determine where the *Practitioners can get a ”made to order” BC with additional informa-
visual center should be placed on the lens. tion, through the Consultation Department

Wavefront custom contact lenses may provide superior

quality of vision compared to standard contact lenses, par- A
ticularly for those patients with RMS values greater than
0.3 µm. Because aberrometry measurement is sensitive to
ocular surface disease and corneal instability, contact lenses
(soft and GP) should be discontinued, and ocular surface
disease should be addressed prior to fitting.
These lenses are manufactured from patient informa-
tion gathered from aberrometry readings to correct both
lower- and higher-order aberrations. This is accomplished
by performing the measurements prior to fitting and then
again with the lens on the eye (Figure 9-41). This allows the
prescription to include the aberrations induced by the lens
and incorporates the change the lens creates. Data are for-
warded to WaveTouch Technologies, where individualized
B
lenses are manufactured and returned to the practitioner.
Lens parameters are described in Table 9-4. The lenses are
dispensed and evaluated similar to traditional lenses. Minor
adjustments are made using customary spherical changes to
improve the vision.
The following case illustrates the fitting process. A
40-year-old man with a history of keratoconus greater in
the right eye and difficulty wearing GP lenses presented for
evaluation. He wore contact lenses to the initial visit despite
severe discomfort in his right eye, and results are listed
in Table 9-5. We discussed management options, and he
desired Intacs implantation. Unfortunately, we learned his
insurance would not cover this procedure. He discontinued Figure 9-42. (A) Corneal topography O D. (B) Corneal topog-
the lenses due to comfort issues and returned the following raphy O S.
month for fitting with Wavetouch lenses.
Topography and aberrometry were obtained for both
eyes (Figure 9-42 and 9-43). The aberrometry refrac- were found to be satisfactory. After this, aberrometry was
tion was checked by traditional refraction, and his vision performed twice for each eye over the trial lenses.
improved to 20/50 OD, 20/25 OS from 20/100 OD, Lenses were ordered and dispensed. The patient reported
20/30 OS. Steep trial contact lenses with a power of -6.50 D exceptional comfort with the new lenses and good vision.
OD and -5.50 D OS were chosen based on his keratometry Vision upon dispensing was 20/50 in the right eye and
readings and refractive status. These lenses were worn for 20/25 in the left eye, and the patient was released for
20 minutes. The movement and centration of the trial lenses 1 week. After 1 week, the vision was 20/50 with an
 To p o gra p hy-G u id e d Co n t a c t Le n s Fit t in g 229

TABLE 9-5.

Patient’s Clinical Information

OD OS
Vision with GP lens 20/50 20/40-
Refraction March 2010 -8.00, 20/100 -8.00 +4.25 x 89, 20/30+
iTrace refraction April 2010 -8.50 +0.50 x 124 -8.12+4.00 x 85
Prefitting
HOA Value 0.786 0.253
Coma 0.678 0.178
Astigmatism 0.352 1.345
Spherical +0.087 -0.042
With Contact Lens
Vision at 1 week: 20/70 20/25
HOA Value 0.460 0.173
Coma 0.139 0.053
Astigmatism 0.548 0.318
Spherical -0.098 -0.020

Figure 9-43. Visual function analysis for each eye prior to lens Figure 9-44. Visual function analysis for each eye with the lens
fitting. in each eye.

over-refraction of +0.50 in the right eye and 20/25 in the closely agree in a high percentage of cases. If this is true,
left eye. The patient reported significant improvement in diagnostic lens fitting may not be necessary on normal
vision, illustrated by Figures 9-44 and 9-45. The lenses corneas. Instead, the practitioner can design lenses with the
were adjusted by +0.50 in the right eye, and the patient is best possible tear layer profile from the GP (or soft) fitting
doing well with 10 to 12 hours of daily wear long-term. software and custom order. However, if the theoretical and
actual do not consistently agree, then the instrument still
serves to offer a diagnostic lens starting point and will
C ONCLUSION reduce the number of modifications or remakes and, there-
fore, the number of office visits and lost chair time.
Software can be used to design the most simple as well
If your aberrometer, topographer, and/or tomographer as the most complex lens designs. Quality of the topo-
are accurate, the theoretical fitting and actual fitting will graphical image is crucial to success with the lens design
230 Ch a p t e r 9

R EFERENCES
1. Maeda N. Topcon KR-9000PW. In: Wang M, ed. Corneal
Topography in the Wavefront Era: A Guide for Clinical Application.
Thorofare, NJ: SLACK Incorporated; 2006:259-268.
2. Kojima R. Validating corneal topography maps. Contact Lens
Spectrum. July 2007. http://www.clspectrum.com/article.
aspx?article=100638. Accessed July 20, 2011.
3. Erdélyi B, Csákány B, Németh J. Spontaneous alterations of the cor-
neal topographic pattern. J Cataract Refract Surg. 2005;31(5):973-
978.
4. Anderson D. Map out your lens fitting. Optometric
Management. October 2008. http://www.optometric.com/article.
aspx?article=102235. Accessed July 20, 2011.
5. Van der Worp E, Graf T, Caroline P. Exploring beyond the corneal
Figure 9-45. Wavefront aberration comparison O D. borders. Contact Lens Spectrum. June 2010. http://www.clspec-
trum.com/article.aspx?article=104343. Accessed July 20, 2011.
6. Sindt C. Basic scleral lens fitting and design. Contact Lens
software. For the user who wishes to make full use of this Spectrum. October 2008. http://www.clspectrum.com/article.
aspx?article=102163. Accessed July 20, 2011.
sophisticated lens design system, time is required to study 7. Kojima R, Caroline P. Designing GPs from corneal topography.
lens design videos and programs, as well as some trial and Contact Lens Spectrum. October 2009. http://www.clspectrum.
error in the early stages. com/article.aspx?article=103490. Accessed July 20, 2011.
8. Anderson D, Kojima R. “Eccentricity” is in against thin. Optometric
Management. November 2008. http://www.optometric.com/article.
aspx?article=102288. Accessed July 20, 2011.
9. Atlas Corneal Topography System User Manual System Software
Version 3.0. Dublin, CA: Carl Zeiss Meditec, Inc.
Ch a p te r 10

Pre -re fractive Surge ry

To p o grap hic Evaluatio n

J. Bradley Randleman, MD

E
valuation of corneal curvature and biomechanical shape and strength; in these patients, a combination of
integrity is critical in preoperative patient evaluation technologies may be needed to generate a thorough overall
for a variety of surgeries, including keratorefractive evaluation.
and intraocular lens-based refractive procedures. Operating
on irregular corneal curvatures can produce unpredict-
able and unstable refractive outcomes, while operating on
biomechanically weak corneas can have untoward results, C ORNEAL TOPOGRAPH Y
including unstable refractions, induced irregular astigma-
tism, and postoperative ectasia.1 Corneal topographic images are generated from Placido
As we have seen in previous chapters, there are a vari- ring reflections from the corneal surface; the distance
ety of different techniques and technologies for obtaining between these rings indicates focal curvature changes.
unique information about the cornea that prove useful This ring pattern is then converted into color-coded maps
in producing a thorough picture of preoperative corneal (Figure 10-1). Appropriate topographic evaluation requires
characteristics. These include corneal topography, corneal a systematic approach to the scans produced (Tables 10-1
tomography, and ocular hysteresis measurements. Corneal and 10-2).
topography is essential to evaluate anterior corneal cur- When evaluating a topographic image, it is first essen-
vature. Corneal tomography (with scanning slit beam or tial to determine the color scale used. Most surgeons use
Scheimpflug systems and high-resolution anterior segment a 0.5-D to 1.0-D color scale for keratorefractive surgical
optical corneal tomography) provides regional corneal epi- evaluations, although a 1.5-D scale has been advocated for
thelial and stromal thickness profiles and elevation data. consistency.2 Color step identification is critical because
Corneal hysteresis measurement may provide direct mea- identical topographies can look significantly different based
surements of corneal “strength” via hysteresis measures on the color step used (Figure 10-2).
obtained with the ocular response analyzer. Once the color step is identified, the topographic pattern
Corneal topography is essential for keratorefractive can be evaluated. There are a variety of established pat-
evaluation and remains the primary technology used in terns3 (Figure 10-3) that can be grouped into 4 major pat-
screening. However, in some cases, additional testing may terns in an otherwise healthy cornea (Figures 10-4 through
provide important supplementary details on overall cornea 10-11) 4:
Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical
231 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 231-242)
© 2012 SLACK Incorporated
232 Ch a p t e r 1 0

TABLE 10-1.

Corneal Topographic Evaluation:

Step-By-Step Approach
1. Identify color scale
2. Assess image quality
3. Identify topographic pattern
4. Additional considerations for topographic pattern
evaluation
a. Age
b. Between-eye symmetry
c. Dry Eye
d. Corneal Scars
e. Contact lens corneal warpage
f. Unusual pattern location

TABLE 10-2.

Corneal Tomographic Evaluation:

Step-By-Step Approach
1. Identify color scales/units of measurement
2. Assess image quality
3. Evaluate regional corneal thickness profiles
4. Evaluate focal anterior and posterior elevation changes
Figure 10-1. Placido image showing Placido rings
reflected off the corneal surface (top) with the resul-
tant color-coded generated image (bottom).
Most practitioners are comfortable identifying clearly
normal and abnormal patterns. However, the challenge lies
in discriminating between the subtle patterns that exist on
1. Normal (including round, oval, or regular bowtie pat-
a continuum between “normal” (acceptable for surgery)
terns)
and “abnormal” (not acceptable for surgery). Some level of
2. Asymmetric bowtie topographic asymmetry can be found in at least one-third
3. Focal steep or skewed radial axis patterns of individuals presenting for refractive surgical evaluation.8
4. Abnormal patterns, including keratoconus, forme In these cases, the following information can provide clues
fruste keratoconus (FFKC), or pellucid marginal cor- in the analysis: patient age, between-eye pattern symmetry,
neal degeneration. against-the-rule topographic patterns, unusual pattern loca-
Normal patterns include round or oval patterns and those tion, contact lens history, and corneal health.
with symmetric bowtie patterns. Asymmetric bowtie pat-
terns include asymmetric steepening between 0.5 D and
1.0 D in any meridian. Focal steep patterns are those with Pat ien t Age
asymmetric steepening of 1.0 D or more in any merid-
Young patient age is a risk factor for the development
ian (that do not meet published criteria for FFKC). Forme
of ectasia.4,9,10 This is likely due to the fact that eyes pre-
fruste patterns are those that meet published FFKC crite-
disposed to naturally occurring corneal ectasia will have
ria.5 Traditional pellucid corneal marginal degeneration
evolving topographic patterns that advance over time.11,12
(PCMD) patterns are those that demonstrate a crab-claw
Thus, when young patients are evaluated, their “final”
pattern or those with steepening in the far inferior periphery
naturally occurring topographic pattern may not yet have
between the 4 and 8 o’clock meridian.6 The traditional crab-
become manifest. Extra caution is therefore necessary when
claw pattern can be found in either keratoconus or PCMD
evaluating topographic patterns in these individuals, and
eyes.7
more weight is given to subtle patterns.
 Pr e -r e fra c t ive Su r ge r y To p o gra p h ic Eva lu a t io n 233

A B

C D

Figure 10-2. Placido image. The same topographic image is shown in the following color steps: (A) 0.25 D, (B) 0.5 D, (C) 1.0 D,
and (D) 1.5 D. Note the apparent difference in pattern based on the color step used.

Figure 10-3. Patterns of

steepening, with an Srax
over 21 degrees and
an Abs(I-S) value over
0.8 D, may correspond
to FFKC. (Reprinted
with permission from
Wang M, ed. Corneal
Top ography in the
Wavefront Era: A Guide
for Clinical Application.
Thorofare, NJ: SLACK
Incorporated; 2006).

Figure 10-4. Placido image: normal pattern.

234 Ch a p t e r 1 0

Figure 10-8. Placido image: skewed radial axis pattern.

Figure 10-5. Placido image: symmetric bowtie pattern.

Figure 10-9. Placido image: forme fruste keratoconus.

Figure 10-6. Placido image: asymmetric bowtie pattern.

Figure 10-10. Placido image: pellucid marginal corneal

degeneration suspect pattern.
Figure 10-7. Placido image: focal steep pattern.
 Pr e -r e fra c t ive Su r ge r y To p o gra p h ic Eva lu a t io n 2 35

Figure 10-11. Placido image: pellucid marginal corneal

degeneration pattern. Figure 10-12. Placido image: against-the-rule astigmatism
pattern. Note that the bow-tie pattern is displaced inferiorly
toward the periphery (also known as “sagging”). This is consis-
tent with an early crab-claw type pattern.
Bet w een -Eye
Pat ter n Asy m m et r y
Topographic patterns are usually symmetric in normal
corneas, and significant between-eye asymmetry should
raise suspicion in the clinician. Further, keratoconus should
be considered a bilateral disease for the purposes of kera-
torefractive surgical evaluation, and if either eye has a high-
ly suspicious topographic pattern, then both eyes should be
considered abnormal and should be excluded from surgery.

Aga in st-t h e-Ru le

Top ograp h ic Pat ter n s
Against-the-rule patterns should also raise suspicion
in the clinician (Figure 10-12). While these patterns can
be a normal variant, they may also be the initial signs of
an evolving PCMD “suspect” topography.7,13,14 Special Figure 10-13. Placido image: superior steepening pattern.
attention should be paid to the far periphery of the image
to make sure that the pattern does not begin to “droop”
downward, indicating the early stages of a crab-claw type
pattern.
Ot h er Con sid erat ion s a n d
Top ograp h ic Ar t ifact s
Un u su a l Pat ter n Lo cat ion s Many other situations may generate irregular corneal
topographies. These include corneal scars and opacities
Traditional patterns suspicious for ectatic disorders, from previous corneal ulcers or epithelial basement mem-
including keratoconus and PCMD suspects, have inferior brane dystrophy (EBMD; Figure 10-15). In these situations,
steepening to varying degrees; however, superior kerato- it is generally reasonable to proceed with keratorefractive
conus and PCMD can occur (Figure 10-13).15-21 Focal surgery if all other parameters are within range; however,
steepening in any meridian should be considered abnor- for most scars, opacities, and EBMD, surface ablation is
mal unless other causes for this steepening can be eluci- preferable to LASIK.
dated, such as significant corneal pannus or corneal scars
(Figure 10-14).13,22-24
236 Ch a p t e r 1 0

Figure 10-14. Placido image: Lateral steepening pattern

(also called a “vertical D” pattern).
B
A

B
Figure 10-16. (A) Placido image: irregular astigmatism pattern
secondary to dry eye (B) that appears more normalized follow-
ing aggressive lubrication and repeat topographies.

and if an irregular pattern is found in a contact lens wearer,

lenses should be discontinued for a longer period.
Dry eyes and blepharitis can also significantly impact
topographic pattern.26-28 These conditions should be evalu-
ated, and in many instances the patient will benefit from
treatment prior to repeat topographies (Figure 10-16).
Effective treatment strategies include topical artificial tear
use, topical cyclosporine to increase tear production, and
topical azithromycin or oral tetracycline antibiotic use for
Figure 10-15. (A) Clinical photo and (B) Placido image: blepharitis management.
irregular astigmatism pattern associated with EBMD. Finally, while it may seem intuitively obvious, the qual-
ity of the scan will directly affect the pattern, and this
should not be overlooked. Technicians should be properly
Extensive contact lens wear, including years of wear, trained not only to obtain quality scans but also to rec-
number of hours/day, and overnight wear, can affect topo- ognize poor-quality scans and automatically repeat these
graphic patterns long after the contact lenses have been to give the practitioner high-quality data to evaluate. The
removed.25 Soft toric and rigid gas-permeable contact most common causes of poor quality scans include patient-
lenses may have even more effect on topographic pattern, topographer misalignment, small images generated from
 Pr e -r e fra c t ive Su r ge r y To p o gra p h ic Eva lu a t io n 2 37

Figure 10-17. Scanning slit

beam image (O rbscan II):
Note the abnormal Placido
pattern (lower left). There are
focal areas of anterior and
posterior elevations infero-
temporally (upper left and
upper right, respectively).
There is also a corresponding
area of focal corneal thick-
ness inferotemporally (lower
right) that is thinner than the
central corneal thickness.

poor corneal exposure, data drop-out from debris in the tear thickness. While no absolute ratio between peripheral cor-
film or eyelashes, and tear lake irregularities. neal thickness has been determined, in one study, popula-
tion average inferior corneal thickness measurements were
more than 60 µm thicker than central corneal measurements
C ORNEAL TOMOGRAPH Y in a normal population.33 A reduction in the peripheral/
central corneal thickness ratio may be an early indicator of
an abnormal cornea.34 Ambrosio and colleagues reported
In addition to anterior curvature data generated by significant differences in corneal thickness spatial profiles
topography, technologies that generate corneal tomographic and corneal volume distribution between keratoconic and
data may be useful for preoperative screening. There are normal corneas,35,36 and these metrics may be useful in
numerous machines available clinically today that measure screening, especially in eyes with borderline topographies.
the cornea using slightly different techniques, including Corneal thickness profiles and elevation data can also
scanning slit beam (Orbscan II, Bausch & Lomb, Rochester, be generated by Scheimpflug imaging technology (Figure
NY), Scheimpflug imaging (Pentacam, OCULUS, Inc, 10-18).37-39 The Pentacam generates corneal thickness
Wetzlar, Germany and Galilei, Ziemer Ophthalmics, Port, spatial profiles that may indicate early corneal pathology
Switzerland), and high-resolution ocular coherence tomog- (Figure 10-19).
raphy (OCT). With scanning slit beam and Scheimpflug Corneal thickness profiles can also be generated by
technologies, both corneal thickness profiles and corneal high-resolution ultrasound40-42 and ocular coherence
elevations as compared to best-fit spheres are generated in tomography (Figure 10-20).43,44 Subtle changes in epi-
an attempt to better characterize overall corneal shape. thelial and stromal thickness may have implications for
The Orbscan II device incorporates both Placido imag- screening, as compensatory epithelial thinning overlying
ing and scanning slit beam-based imaging to generate an ectatic region may be an early sign of a corneal ectatic
both standard topographic and tomographic data. Overall, disorder.45-47
corneal thickness profiles are measured, and from these The relative weight given to tomographic data ver-
anterior and posterior elevation data are generated. These sus topographic data is still hotly debated; however, the
are measured against a “best-fit sphere” to give an overall 2 together are likely additive. Anterior curvature changes
impression of focal elevations (Figure 10-17). There have with concomitant elevation abnormalities and abnormal
been reports discussing the value of these individual param- corneal thickness profiles increases the odds that a cornea
eters, especially posterior float metrics29-32 ; however, these is biomechanically weaker and should be excluded for kera-
remain to be validated clinically. Another useful metric torefractive surgery.
is the relationship between central and peripheral corneal
238 Ch a p t e r 1 0

Figure 10-18. Scheimpflug image

(Pentacam). Note the abnormal
sagittal curvature pattern, with
focal steepness laterally (lower
left). There are focal areas of
front and back elevations later-
ally (upper left and upper right,
respectively).

Figure 10-19. Corneal thickness spatial profile from the

Scheimpflug image (Pentacam). Note the drop-off in values in
the corneal peripheral diameter.

Figure 10-20. High-resolution ante-

rior segment O CT image. Note the
numerical values (left) and the cor-
responding color-coded mapping
(right).
 Pr e -r e fra c t ive Su r ge r y To p o gra p h ic Eva lu a t io n 239

A B

Figure 10-21. Placido image of (A) right and (B) left eyes. In the right eye, note the symmetric bowtie pattern oriented along the
110-degree meridian in the right eye combined with the moderate asymmetry oriented along the 30-degree meridian. In the left
eye, note the focal steepening inferotemporally.

Eva lu at in g Top ograp h ic PATIENT EVALUATION:

a n d Tom ograp h ic Dat a C ASE EXAMPLE
In d ep en d en t ly From Ot h er
Pat ien t Factor s A 22-year-old man presented for refractive surgery
evaluation. His manifest refraction was -3.25 sphere and
While a complete surgical evaluation requires the cli- -4.00 +1.00 x 055 in the right and left eyes, respectively,
nician to evaluate and consider a variety of information with best spectacle-corrected visual acuities of 20/20 in
in addition to topographic pattern, including refraction, both eyes. Central corneal thickness was 560 µm in the
corneal thickness, overall ocular health, and patient age, right eye and 570 µm in the left eye. The anterior segment
it is critical that topographic pattern evaluation initially be and dilated fundus exam was unremarkable in both eyes.
done independently. This statement is essential and bears Topographies are shown (Figure 10-21). Additional images
repeating: First evaluate topographic patterns independent of the left eye are shown (Figures 10-22 through 10-24).
of other patient information. Abnormal topographic pat- Topographic evaluation of the right eye Placido image
terns can occur in any age group, with any refraction or (in 0.5 D color scale) demonstrates what initially appears to
corrected visual acuity, and with any corneal thickness. be a symmetric bowtie pattern (with some inferior artifact,
Some clinicians may be tempted to devalue subtle abnor- likely from tear film) oriented in the 110-degree meridian.
malities if the patient has a low myopic or hyperopic refrac- However, on further inspection, there is subtle asymmetry
tion, their correctable acuity is “excellent,” or the cornea is oriented along the 30-degree meridian, with the inferior
“thick enough”; however, abnormal topography can occur temporal cornea approximately 0.8 D steeper in some areas
in any eye and is an independent risk factor for ectasia. than the corresponding superior nasal area. Topographic
Ectasia has occurred in hyperopic eyes, thick corneas, evaluation of the left eye Placido image (in 0.5 D color
and eyes that otherwise appeared “normal” but that could scale) demonstrates more pronounced asymmetry oriented
have been excluded by close topographic pattern analysis along the 150-degree meridian, with the inferior temporal
alone.13,14,34,48,49 cornea approximately 1.5 D steeper in some areas than
the corresponding superior nasal area. In comparing the
2 eyes, there is some between-eye asymmetry between the
patterns.
240 Ch a p t e r 1 0

Figure 10-22. Scanning

slit beam image (O rbscan
II). Note the Placido pat-
tern that is quite simi-
lar to the Placido pattern
from Figure 10-21. Also
note the focal increased
anterior and posterior
elevations inferiorly in
the far periphery (upper
left and right images,
respectively).

Figure 10-23. Scheimpflug

image (Pentacam). Note
the sagittal curvature
map (lower left) that
corresponds in pattern
to the Placido images
from Figures 10-21 and
10-22. There is more pro-
nounced focal steepening
in both the front and back
elevation maps (upper
left and right images,
respectively).

Further evaluation of the left eye with scanning slit beam and appropriate peripheral thickening in relation to the cen-
technology (Orbscan II) confirms the Placido pattern (see tral thickness. Neither anterior nor posterior elevations are
Figure 10-22, lower left, also in 0.5 D color scale). The cor- overly remarkable (see Figure 10-22, upper left and right,
neal thickness profile (see Figure 10-22, lower right) appears respectively); however, there is some increased elevation in
normal, with the thinnest portion of the cornea in the center the far inferior periphery that warrants further evaluation.
 Pr e -r e fra c t ive Su r ge r y To p o gra p h ic Eva lu a t io n 2 41

Figure 10-24. Scheimpflug image (Pentacam): The corneal thickness spatial profile (upper middle image) exhibits a gradual rela-
tive reduction toward the periphery.

Scheimpflug imaging technology (Pentacam) demon-

strates a sagittal curvature pattern (see Figure 10-23, lower C ONCLUSION
left) that correlates well with the Placido patterns. The
corneal thickness profile (see Figure 10-23, lower right) There are a variety of technologies available for patient
appears normal, with the thinnest portion of the cornea in tomographic and topographic screening. Careful evaluation
the center and apparently normal peripheral thickening in of these modalities, sometimes in combination for suspect
relation to the central thickness. There is a more focal area topographic patterns, can improve the refractive screening
of elevation visible on both the front and back elevation process and thereby improve the safety and efficacy of
maps (see Figure 10-23, upper left and upper right, respec- keratorefractive surgery.
tively). Further, mean corneal thickness values (see Figure
10-24, upper middle box) demonstrate a gradual relative
reduction toward the periphery.
So, the overall topographic and tomographic evaluation R EFERENCES
for this patient reveals an asymmetric pattern with between-
eye asymmetry, borderline corneal thickness spatial pro- 1. Randleman JB, Russell B, Ward MA, et al. Risk factors and
files, and subtle increased anterior and posterior elevations prognosis for corneal ectasia after LASIK. Ophthalmology.
more notable in the left eye. The patient’s young age makes 2003;110(2):267-275.
2. Smolek MK, Klyce SD, Hovis JK. The Universal Standard Scale:
this pattern highly suspicious; therefore, LASIK is not rec-
proposed improvements to the American National Standards
ommended, and a reasonable approach would be to delay Institute (ANSI) scale for corneal topography. Ophthalmology.
surgery for 1 to 2 years to re-evaluate the topographic and 2002;109(2):361-369.
tomographic data before considering surface ablation. 3. Rabinowitz YS, Yang H, Brickman Y, et al. Videokeratography data-
base of normal human corneas. Br J Ophthalmol. 1996;80(7):610-
616.
242 Ch a p t e r 1 0

4. Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assess- 29. Quisling S, Sjoberg S, Zimmerman B, et al. Comparison of
ment for ectasia after corneal refractive surgery. Ophthalmology. Pentacam and Orbscan IIz on posterior curvature topography mea-
2008;115(1):37-50. surements in keratoconus eyes. Ophthalmology. 2006;113(9):1629-
5. Rabinowitz YS, McDonnell PJ. Computer-assisted corneal topogra- 1632.
phy in keratoconus. Refract Corneal Surg. 1989;5(6):400-408. 30. Rao SN, Raviv T, Majmudar PA, Epstein RJ. Role of Orbscan II in
6. Maguire LJ, Klyce SD, McDonald MB, Kaufman HE. Corneal screening keratoconus suspects before refractive corneal surgery.
topography of pellucid marginal degeneration. Ophthalmology. Ophthalmology. 2002;109(9):1642-1646.
1987;94(5):519-524. 31. Sonmez B, Doan MP, Hamilton DR. Identification of scanning slit-
7. Lee BW, Jurkunas UV, Harissi-Dagher M, et al. Ectatic disorders beam topographic parameters important in distinguishing normal
associated with a claw-shaped pattern on corneal topography. Am J from keratoconic corneal morphologic features. Am J Ophthalmol.
Ophthalmol. 2007;144(1):154-156. 2007;143(3):401-408.
8. Varssano D, Kaiserman I, Hazarbassanov R. Topographic patterns 32. Tabbara KF, Kotb AA. Risk factors for corneal ectasia after LASIK.
in refractive surgery candidates. Cornea. 2004;23(6):602-607. Ophthalmology. 2006;113(9):1618-1622.
9. Klein SR, Epstein RJ, Randleman JB, Stulting RD. Corneal ectasia 33. Liu Z, Huang AJ, Pflugfelder SC. Evaluation of corneal thickness
after laser in situ keratomileusis in patients without apparent preop- and topography in normal eyes using the Orbscan corneal topogra-
erative risk factors. Cornea. 2006;25(4):388-403. phy system. Br J Ophthalmol. 1999;83(7):774-778.
10. Randleman JB, Trattler WB, Stulting RD. Validation of the Ectasia 34. Randleman JB, Banning CS, Stulting RD. Corneal ectasia after
Risk Score System for preoperative laser in situ keratomileusis hyperopic LASIK. J Refract Surg. 2007;23(1):98-102.
screening. Am J Ophthalmol. 2008;145(5):813-818. 35. Ambrosio R Jr, Alonso RS, Luz A, Coca Velarde LG. Corneal-
11. Fogla R, Padmanabhan P. Bilateral keratectasia after unilateral laser thickness spatial profile and corneal-volume distribution: tomo-
in situ keratomileusis. J Cataract Refract Surg. 2004;30(10):2033- graphic indices to detect keratoconus. J Cataract Refract Surg.
2034. 2006;32(11):1851-1859.
12. Wang JC, Hufnagel TJ, Buxton DF. Bilateral keratectasia after uni- 36. Ambrosio R Jr, Klyce SD, Wilson SE. Corneal topographic and
lateral laser in situ keratomileusis: a retrospective diagnosis of ectatic pachymetric screening of keratorefractive patients. J Refract Surg.
corneal disorder. J Cataract Refract Surg. 2003;29(10):2015-2018. 2003;19(1):24-29.
13. Ambrosio R Jr, Wilson SE. Early pellucid marginal corneal degen- 37. Khachikian SS, Belin MW. Posterior elevation in keratoconus.
eration: case reports of two refractive surgery candidates. Cornea. Ophthalmology. 2009;116(4):816, e1; author reply -7.
2002;21(1):114-117. 38. Belin MW, Khachikian SS. An introduction to understanding eleva-
14. Fogla R, Rao SK, Padmanabhan P. Keratectasia in 2 cases with tion-based topography: how elevation data are displayed—a review.
pellucid marginal corneal degeneration after laser in situ keratomi- Clin Experiment Ophthalmol. 2009;37(1):14-29.
leusis. J Cataract Refract Surg. 2003;29(4):788-791. 39. Ciolino JB, Khachikian SS, Belin MW. Comparison of corneal
15. Sridhar MS, Mahesh S, Bansal AK, Rao GN. Superior pellucid thickness measurements by ultrasound and Scheimpflug photogra-
marginal corneal degeneration. Eye (Lond). 2004;18(4):393-399. phy in eyes that have undergone laser in situ keratomileusis. Am J
16. Bower KS, Dhaliwal DK, Barnhorst DA Jr, Warnicke J. Pellucid Ophthalmol. 2008;145(1):75-80.
marginal degeneration with superior corneal thinning. Cornea. 40. Reinstein DZ, Archer TJ, Gobbe M, et al. Epithelial thickness in the
1997;16(4):483-485. normal cornea: three-dimensional display with Artemis very high-
17. Cameron JA, Mahmood MA. Superior corneal thinning with frequency digital ultrasound. J Refract Surg. 2008;24(6):571-581.
pellucid marginal corneal degeneration. Am J Ophthalmol. 41. Reinstein DZ, Couch DG, Archer T. Direct residual stromal thick-
1990;109(4):486-487. ness measurement for assessing suitability for LASIK enhancement
18. Chiang CC, Lin JM, Tsai YY. Superior keratoconus with inferior by Artemis 3D very high-frequency digital ultrasound arc scanning.
paracentral corneal thinning and inferior peripheral pellucid mar- J Cataract Refract Surg. 2006;32(11):1884-1888.
ginal degeneration. Eye (Lond). 2007;21(2):266-268. 42. Reinstein DZ, Archer TJ, Silverman RH, Coleman DJ. Accuracy,
19. Weed KH, McGhee CN, MacEwen CJ. Atypical unilateral repeatability, and reproducibility of Artemis very high-frequen-
superior keratoconus in young males. Cont Lens Anterior Eye. cy digital ultrasound arc-scan lateral dimension measurements.
2005;28(4):177-179. J Cataract Refract Surg. 2006;32(11):1799-1802.
20. Prisant O, Legeais JM, Renard G. Superior keratoconus. Cornea. 43. Li Y, Meisler DM, Tang M, et al. Keratoconus diagnosis with opti-
1997;16(6):693-694. cal coherence tomography pachymetry mapping. Ophthalmology.
21. Eiferman RA, Lane L, Law M, Fields Y. Superior keratoconus. 2008;115(12):2159-2166.
Refract Corneal Surg. 1993;9(5):394-395. 44. Li Y, Shekhar R, Huang D. Corneal pachymetry mapping with
22. Bansal AS, Randleman JB. Corneal ectasia after LASIK in a patient high-speed optical coherence tomography. Ophthalmology.
with superior pellucid marginal degeneration. J Refract Surg. 2006;113(5):792-799, e2.
2009;25(2):172-174. 45. Reinstein DZ, Gobbe M, Archer TJ, et al. Epithelial, stromal, and
23. Rao SK, Fogla R, Padmanabhan P, Sitalakshmi G. Corneal total corneal thickness in keratoconus: three-dimensional display
topography in atypical pellucid marginal degeneration. Cornea. with Artemis very-high frequency digital ultrasound. J Refract
1999;18(3):265-272. Surg. 2010;26(4):259-271.
24. Abad JC, Rubinfeld RS, Del Valle M, et al. Vertical D: a novel 46. Reinstein DZ, Archer TJ, Gobbe M. Corneal epithelial thick-
topographic pattern in some keratoconus suspects. Ophthalmology. ness profile in the diagnosis of keratoconus. J Refract Surg.
2007;114(5):1020-1026. 2009;25(7):604-610.
25. Wilson SE, Lin DT, Klyce SD, et al. Topographic changes in contact 47. Reinstein DZ, Archer TJ, Gobbe M. Stability of LASIK in topo-
lens-induced corneal warpage. Ophthalmology. 1990;97(6):734-744. graphically suspect keratoconus confirmed non-keratoconic by
26. De Paiva CS, Harris LD, Pflugfelder SC. Keratoconus-like Artemis VHF digital ultrasound epithelial thickness mapping: 1-
topographic changes in keratoconjunctivitis sicca. Cornea. year follow-up. J Refract Surg. 2009;25(7):569-577.
2003;22(1):22-24. 48. Malecaze F, Coullet J, Calvas P, et al. Corneal ectasia after
27. Dursun D, Piniella AM, Pflugfelder SC. Pseudokeratoconus caused photorefractive keratectomy for low myopia. Ophthalmology.
by rosacea. Cornea. 2001;20(6):668-669. 2006;113(5):742-746.
28. de Paiva CS, Lindsey JL, Pflugfelder SC. Assessing the severity 49. Randleman JB, Caster AI, Banning CS, Stulting RD. Corneal
of keratitis sicca with videokeratoscopic indices. Ophthalmology. ectasia after photorefractive keratectomy. J Cataract Refract Surg.
2003;110(6):1102-1109. 2006;32(8):1395-1398.
Ch a p te r 11

To p o grap hy-
Guid ed Custo m Ab latio n

Tracy Schroeder Swartz, OD, MS, FAAO

T
here are a variety of etiologies for irregular astigma- Correcting irregular astigmatism caused by refractive
tism, directly measured as irregularity on topograph- procedures is complex, as most devices used to perform
ical maps and manifesting as increased higher-order refractive procedures are unable to correct the problems they
aberrations (HOAs) using wavefront analysis. Mild ocular may cause. These problems include a small optical zone,
surface changes such as dry eye and contact lens warpage decentered ablation, severely prolate or oblate corneas, and
are usually correctable by treating the dry eye and discon- irregularly irregular astigmatism induced by laser-assisted
tinuing the lenses. More severe ocular surface changes such in situ keratomileusis (LASIK), photorefractive keratec-
as scarring, corneal dystrophies, and degenerations such tomy (PRK), radial keratotomy, conductive keratoplasty,
as anterior basement membrane dystrophy and pterygium cataract surgery, and deep lamellar keratoplasty. Several
may require surgery to remove the scarring, address the systems have been used to correct these problems.1-5 The
dystrophy, or remove the pterygium. Refractive procedures following sections illustrate how several systems surgically
including radial keratotomy, excimer ablation, and conduc- address correction of irregular astigmatism.
tive keratoplasty can also create irregular astigmatism.

LASERSIGH T ASTRAP RO
David D. Liu, PhD; Zhengjun Fan, MD; Junhua Xiao, MD; and Bing Liu, MD

E
ven with the advancement of technologies and clini- such as decentered ablations and other irregular corneas.
cal applications, difficult corneal cases do occasion- AstraPro 2.2Z uses precise 3-dimensional elevation data
ally prove to be a challenge in laser refractive sur- obtained from AstraMax, a 3-dimensional topographer,
gery. To address such a challenge, LaserSight Technologies, and subjective refraction data to calculate an ablation pat-
Inc, has developed a 3-dimensional topography-based tern that allows the restoration of both the desired shape of
custom ablation product, AstraPro 2.2Z, to repair problems, the postenhancement cornea and the desired visual axis of

Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical

243 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 243-252)
© 2012 SLACK Incorporated
244 Ch a p t e r 1 1

a problem cornea, resulting in tissue-saving ablation that

creates an effective refractive optical zone on the most
applicable optical axis.
In problem corneas, such as those with a decentered
ablation, the corneal elevation is asymmetrical. When
using a topographer to obtain the elevation measurement,
the acquired data are misaligned with the visual axis of
the eye (for normal cornea, the topography center is a good
approximation to the visual axis center). If we design an
ablation profile based on the misaligned axis, any attempts
to fix the problem cornea will not only be ineffective, but
also wasteful in cornea tissues. The new ablation algorithm
used in AstraPro 2.2Z is based on this observation and
understanding. The algorithm involves an optimized ellip-
soidal model that targets a prolate postoperative surface (for
better visual acuity) and at the same time also allows the
axis of the postoperative ellipsoid of the corneal surface to
be optimized with respect to the preoperative visual axis of
the cornea. In the case of an asymmetrical cornea, its corne- Figure 11-1. Decentered ablation as measured by the
al topography-indicated visual axis is biased and therefore AstraMax 3-D topographer. It is important to properly
should not be used as the target axis for the postoperative adjust the patient’s head and eye position prior to mea-
corneal surface. The optimization criterion is to find a new surement of topography. The patient must properly fixate
axis that minimizes the tissue ablated while achieving the on the fixation target.
refractive correction needed using a prolate ablation model.
The optimized ablation profile is a point-by-point abla-
tion pattern based on asymmetrical elevation data and can
only be achieved with a point-by-point scanning refractive
laser. LaserSight AstraScan XL is equipped with a high-
speed eye tracker and high-speed micro spot scanning
system, ideally suited for executing this type of precise,
point-by-point, 3-dimensional asymmetrical ablation pro-
file. The following case reports demonstrate the system’s
applications.

Ca se 1
A 24-year-old female patient underwent LASIK in May
2009 in a different hospital on the right eye. The patient
complained of poor vision with ghost images. Uncorrected
visual acuity was 40/100. The patient was referred to our Figure 11-2. The arrowed line from the origin indicated the
hospital in May 2010. Upon examination with the AstraMax shift direction of the predicted post-enhancement visual axis.
and manifest refraction, it was observed that a signifi- The shift amount is very small, typically well below 0.1 mm by
cantly decentered ablation existed, as shown in Figure 11-1. 0.1 mm, and never exceeds 0.2 mm by 0.2 mm. The shift is
always in the direction of the flat area of the pre-enhancement
We evaluated the patient’s cornea and decided to use
cornea, indicating that the measurement bias for the problem
LaserSight’s AstraPro 2.2Z to plan an enhancement on the cornea is toward the steep axis.
cornea. LaserSight AstraPro 2.2Z allows significant tissue
saving over a standard algorithm. More importantly, it is
effective in optimally choosing the post-enhancement tar-
and the patient achieved 20/20 uncorrected visual acuity.
get corneal visual axis, typically along the direction of the
The corrected topography map and the difference map of
flat portion of the cornea, as indicated in Figure 11-2. The
the pre-enhancement and post-enhancement are displayed
relatively small change of predicted postoperative visual
in Figure 11-3. As can be seen, the enhancement ablation
axis allows for significant tissue saving and dramatically
was custom designed to just compensate for the ablation
improved enhancement effectiveness. After the enhance-
deficiency caused by the decentration.
ment, the major complaint of double vision was resolved,
 To p o gra p hy-G u id e d Cu st o m Ab la t io n 24 5

Figure 11-5. In this 3-dimensional ablation profile, the ablation

volume distribution is calculated precisely from the measured
Figure 11-3. The upper right is the pre-enhancement and the 3-dimensional corneal elevation data. The planned ablation
lower right is the post-enhancement topography map. The profile is optimized to correct the missed ablation tissues to
left is the difference of the two. The post-enhancement map achieve a symmetrical and smooth postoperative corneal
illustrates the restored center optical zone, and the difference surface.
map indicates approximately the amount of refractive power
correction achieved for this problem cornea.

Figure 11-6. The post-enhancement map of this cornea is

smooth and symmetrical, with resolution of the ghost image
Figure 11-4. A significantly decentered ablation. and poor visual acuity complaint while saving unnecessary
and excessive ablation. The topography-guided ablation tech-
nique, due to the high precision of the elevation data (on the
order of 1 µm) and the advanced ellipsoidal model in free
Ca se 2 space, with optimization in both radius, shape factor, and axis,
was able to repair severe irregularity of the cornea, achieving
This patient presented with a pre-enhancement UCVA high confidence for the patients as well as for the surgeons.
20/100. The decentered ablation is shown in Figure 11-4.
AstraPro was used to create the ablation profile from the
3-dimensional corneal elevation data (Figure 11-5). Due to ghost image and poor visual acuity complaint while sav-
the high precision of the anterior corneal elevation data, the ing unnecessary and excessive ablation. After applying
predicted postoperative target surface can be achieved with AstraPro 2.2Z, the decentration was corrected, and the
high precision with a smooth and precision micro-scanning post-enhancement UCVA was 20/20.
ablation laser. As can be seen, the planned ablation profile A topography data-guided ablation technique, due to the
is optimized to touch up on the missed ablation tissues to high precision of the elevation data and advanced ellipsoi-
achieve a symmetrical and smooth postoperative corneal dal model in free space, with optimization in radius, shape
surface. Figure 11-6 shows that the post-enhancement map factor, and axis, was successfully able to repair the severe
of this cornea is smooth and symmetrical, resolving the irregularity of the cornea.
246 Ch a p t e r 1 1

TOPOGRAPH Y-GUIDED CUSTOM ABLATION W ITH IVIS SUITE

Aleksandar Stojanovic, MD and Xiangjun Chen, MD, MS

i Ca se 1
VIS Technologies (Taranto, Italy) has pioneered topogra-
phy-guided custom ablation since the mid-1990s by intro-
ducing Corneal Interactive Programmed Topographic A 57-year-old woman had her right eye treated with
Ablation (CIPTA) software. The software was used for PRK in 1997. She was retreated because of regression/
custom ablation planning based on Orbscan topography undercorrection with LASIK in 1999. The baseline refrac-
(Bausch & Lomb, Rochester, NY) and for treatment on tive error was -4.25 -0.25 x 90 degrees, and best-corrected
LaserSight 100-Hz, small-flying-spot laser (LaserSight distance visual acuity was 20/20. Blurred vision, starburst
Technologies Inc, Orlando, FL). Since then, the company halos, and multiple images were noted after the PRK and
developed its own Scheimpflug-based topographer, the increased after her LASIK. Irregular astigmatism due to
Precisio, as well as a dynamic pupillometer (pMetrics) and decentered ablation and oblate asphericity were diagnosed.
a 1000-Hz, 0.6-mm flying-spot excimer laser (iRES), com- Topography-guided cTEN ablation with iVIS suite was
pleting the “iVIS Suite” topography-guided custom ablation planned and performed in October 2008.
system. Prior to corrective treatment, the patient’s uncorrected
In the iVIS Suite, corneal anterior and posterior eleva- distance visual acuity (UDVA) was 20/30, and subjective
tion, pachymetry, pupil as well as iris and scleral vessel refraction of +1.50 -0.25 x 90 degrees failed to improve
registration-data from the Precisio, pupillometry-data from vision. Central ultrasound pachymetry (Corneo-Gage Plus,
the pMetrics, and the patient’s refraction are fed to the Sonogage Inc, Cleveland, OH) was 517 µm. Slit-lamp
CIPTA software. CIPTA then compiles a custom ablation microscopy was unremarkable except for central grade 1
plan with the aim of transforming the actual corneal shape subepithelial haze within the LASIK flap. Precisio floating-
into a regular aspheric shape of desired curvature, within elevation and axial curvature maps showed irregular astig-
the treatment zone suggested by pupillometry. The volume matism (Figure 11-7). Asymmetry within the central 3 mm
of the ablation is defined by the intersection between the was 1.5 D, while the Q-value was +0.56. Wavefront aber-
corneal anterior surface as detected by Precisio and the rometry measured at 6 mm with the Allegretto Analyzer
targeted aspheric surface, determined by CIPTA. The cur- (Wave-Light, Erlangen, Germany) showed total root-mean-
vature of the targeted surface is determined by subtracting square HOA of 0.95 µm (coma 0.67 µm and spherical aber-
the amount of desired dioptric change from the preoperative ration 0.44 µm). pMetrics dynamic pupillometry showed
curvature. Hence, the corneal topography and the subjective 3.95, 6.59, and 8.75 mm as photopic, “ideal,” and scotopic
refraction are used as bases for treatment of higher- and pupil sizes.
lower-order aberrations, respectively (the former is limited The treatment goal was an aberration-free corneal
to the HOAs originating from the corneal surface). surface and emmetropia using a single ablation involving
The CIPTA-generated ablation plan is then transferred the epithelium and the stroma (including the haze). After
to the iRES excimer laser, which employs a synchronized importing the patient’s Precisio topography/tomography
x,y and cyclotorsional tracking. To reduce the high-frequen- and manifest spherocylindrical refraction, the CIPTA soft-
cy laser’s thermal effect and the ablation-rate difference ware computed the desired regular aspheric postoperative
between the corneal epithelium and the stroma, delivery of surface with a curvature change, which compensated for
constant local frequency (number of pulses per second) per the manifest spherocylindrical error within the optical zone,
area is implemented into iRES, allowing the use of inte- matching the “ideal” pupil size. The ablation plan showed
grated transepithelial ablation as the system’s recommended maximum ablation depth of 34 µm and the outer transition
operative technique. The technique merges laser epithelium zone of 8.75 mm. The ablation plan is shown on Figure 11-8.
removal and stromal ablation into a single, uninterrupted A detailed surgical protocol is described elsewhere.7
custom ablation, termed “cTEN” (custom transepithelial Twelve months after the treatment, slit-lamp micros-
“no-touch”). Such integrated, topography-guided ablation copy showed a clear cornea, with no trace of haze. The
seems to be especially suited for therapeutic purposes for patient’s uncorrected and corrected visual acuity were
cases with irregular astigmatism, where epithelial remodel- 20/25 and 20/20, respectively, and manifest refraction was
ing plays a significant role and has to be addressed.6 +0.50 -0.25 x 50 degrees in the right eye. Wavefront aber-
The following case illustrates the use of iVIS Suite’s rometry showed total root-mean-square HOA of 0.33 µm
topography-guided cTEN ablation in treatment of irregular (coma 0.22 µm and spherical aberration 0.18 µm) for
astigmatism after decentered PRK/LASIK. 6-mm diameter. Floating elevation and axial curvature
 To p o gra p hy-G u id e d Cu st o m Ab la t io n 24 7

Figure 11-7. Floating elevation (left) and axial curvature map Figure 11-9. Floating elevation (left) and axial curvature map
(right) before the topography-guided ablation using the iVIS (right) 12 months after the topography-guided ablation with
Suite. iVIS Suite.

unaffected by an ablation on the stroma beneath the flap.

Custom PRK on top of the flap would remove the subepi-
thelial haze, but in that case, the epithelial remodeling could
become an issue of concern. The epithelial remodeling,
which would have most likely occurred in the current case
due to the irregular astigmatism, would cause a mismatch
between the epithelial surface morphology (the data source
for topography-guided custom ablation planning) and the
stromal morphology (the surface to be treated). In such a
case, topography-guided ablation delivered on the stroma
after the epithelial removal, as a part of PRK or other
surface ablation technique that involves epithelial removal
prior to the ablation, would lead to a potentially significant
ablation error caused by the mentioned mismatch.
Figure 11-8. Ablation planning based on imported elevation With the current cTEN technique, the epithelial removal
topography (left) and the compiled ablation map (right). is programmed into a single ablation together with the
stromal ablation and is executed as an integral part of an
uninterrupted laser treatment, thanks to the similar ablation
maps showed significantly reduced irregular astigmatism rates between the epithelium and the stroma, achieved by
(Figure 11-9) with asymmetry reduced to 0.7 D within the the iRES laser. This allows the topography-detected surface
central 3 mm and prolate asphericity, with a Q value of to be used as a true basis for custom ablation, circumventing
-0.14. Most importantly, the patient’s complaints of visual the problem of the epithelial remodeling and the stromal
disturbances were fully resolved. mismatch. Moreover, the area of the epithelial ablation in
It may be undisputable that the corneal topography- cTEN fits exactly the outer edges of the stromal ablation
guided custom ablation is the most logical approach in area, and therefore only the absolutely necessary amount of
addressing iatrogenic corneal surface irregularity, but the epithelium is removed, minimizing the trauma to the cornea
choice of the operative technique to deliver the ablation and and increasing the speed of re-epithelialization. One impor-
treat a post-LASIK case where visually disturbing pathol- tant desired improvement to the cTEN approach would be
ogy (eg, haze) resides within the flap may not be so clear. the use and integration of technology that would accurately
Flap relift, the most commonly used technique in LASIK measure the epithelial thickness, such as Artemis VHF
retreatment, could not address the current case because the scanner (ArcScan Inc, Golden, CO), mentioned elsewhere
subepithelial haze residing within the flap would remain in this book.
248 Ch a p t e r 1 1

ARTEMIS VH F D IGITAL ULTRASOUND EVALUATION OF

TOPOGRAPH Y-G UIDED R EPAIR
Dan Z. Reinstein, MD, MA(Cantab), FRCSC, FRCO phth; Timothy J. Archer,
MA(O xon), DipCompSci(Cantab); and Marine Gobbe, MST(O ptom), PhD

T
opography-guided ablation profiles are derived from B-scans, comprising 8 hemimeridians. This is our standard
front surface corneal topography, and the algorithms scanning protocol as it provides a sufficiently high density
are designed to calculate the tissue removal required of information in the central cornea with a lower density of
to return the cornea to a regular aspheric shape. Therefore, information in the periphery, where it is less needed. The
the ablation profiles are usually irregular and nonsymmet- Artemis has been described in detail previously.8-14
ric. The results of topography-guided treatments are usu- Contrast sensitivity was well below the normal range;
ally evaluated by inspection of the change in front surface the patient was not able to see more than 2 patches at 3, 6,
corneal topography; however, the shape of ablated stromal 12, or 18 cpd. Analysis of front surface topography showed
tissue has not been confirmed by direct measurement of that the optical zone was decentered inferotemporally rela-
stromal change. The Artemis very high-frequency digital tive to the visual axis; the Atlas axial map is presented in
ultrasound arc-scanner (ArcScan Inc, Morrison, CO) is Figure 11-10. Wavefront analysis showed significant HOAs
capable of obtaining a map of the stromal thickness across with a higher-order RMS of 1.63 µm and spherical aber-
the central 10-mm diameter in vivo.8-10 The change in stro- ration of 1.30 µm (OSA notation15). The Artemis epithe-
mal thickness can be calculated as the difference in stromal lial thickness profile showed that the epithelium was up to
thickness before and after surgery to evaluate the achieved 84-µm thick centrally, significantly thicker than normal,14
central ablation depth in conventional myopic ablations,11-13 surrounded by an annulus of thinner epithelium (down to
as demonstrated by the following case. 44 µm) at a radius of 5 mm. This annulus of thinner epithe-
A 45-year-old man was referred to our clinic complain- lium, over the midperipheral “knee” characteristic of radial
ing of ghosting and increasing blur in both eyes, but more keratotomy, was also decentered inferotemporally, match-
pronounced in the right eye. Prior to undergoing radial ing the irregularity seen on topography.
keratotomy, the preoperative refraction was approximately A topography-guided PRK procedure was performed
-9.50 D in both eyes. The left eye was treated first using using the MEL80 excimer laser and CRS-Master custom
the “American” method, and the right eye was treated ablation programming software (version 2.1), which has
2 weeks later using the “Russian” method. Sixteen incisions been described previously.1 The CRS-Master algorithm
were created in both eyes. Clinical records were not avail- uses corneal elevation data, captured by the Atlas topogra-
able postoperatively, but the patient explained that he did pher, to calculate the ablation required to obtain an aspheric
not need to wear glasses for the first 10 years and was able surface. The algorithm also analyzes the corneal wavefront,
to obtain a private pilot’s license. After 4 years, the patient derived from the corneal topography data, together with the
noticed his vision deteriorating, and the patient sought sur- clinical manifest refraction to calculate the refractive error
gical correction of the right eye. correction. The algorithm takes into account that the mani-
Uncorrected distance visual acuity of 20/40+1 improved fest refraction possesses a component that compensates for
with a manifest refraction of +0.75 -2.00 x (20/32+2). corneal HOAs and is able to discriminate this component to
Further testing included topography, whole eye wavefront determine the HOA-free manifest refractive error.
using the WASCA aberrometer (Carl Zeiss Meditec, Jena, Numerous Atlas examinations were obtained, and the
Germany) both undilated and after cycloplegia using tropi- exam to use for treatment was selected by the surgeon
camide 1%, pupillometry, corneal biomechanical analysis, (DZR). The criteria for selecting the examination to use
handheld ultrasonic pachymetry, and tonometry. Vertical for treatment were for the exam to a) be in focus, b) have
sinusoidal grid contrast sensitivity testing was obtained smooth, regular mires rings, and c) have continuous data
at 3, 6, 12, and 18 cycles per degree (cpd) using the CSV- within sufficient diameter. The Atlas topographic data were
1000 (VectorVision Inc, Greenville, OH). The Artemis 1 imported into the CRS-Master. The CRS-Master automati-
VHF digital ultrasound arc-scanner was used to determine cally selects the center of the pupil as found by the Atlas to
the thickness profile of each corneal layer. The reproduc- use as the center for the ablation generation algorithms. The
ibility of pachymetric topography of the Artemis has CRS-Master also allows the surgeon to shift this location,
been shown to be 0.58 µm for the epithelium, 1.78 µm for and the corneal vertex was chosen for this case according
the stroma, and 1.68 µm for the cornea.10 Pachymetric to our standard protocol of centering all ablations on the
profiles were calculated based on data from 4 meridional corneal vertex.1,16,17 The corneal vertex was determined
 To p o gra p hy-G u id e d Cu st o m Ab la t io n 24 9

hydration of the corneal surface in advance of the asym-

metric ablation being performed. After the ablation was
performed, a series of 3 “wet” 20-µm PTK ablations were
performed to smooth out the incisional scarring; this
involved flooding the exposed stroma with BSS solution to
isolate ablation selectively only on the steep ridges of the
edges of the radial keratotomy incisions and keratin plugs
jutting above the general stromal surface. Typical postop-
erative treatment followed, including a bandage lens and
topical medications.
Nine months after the procedure, the uncorrected dis-
tance visual acuity in the patient’s right eye was 20/32
with a manifest refraction of +0.75 -0.75 x 20, and the
CDVA had improved to 20/20-1. The slit-lamp examina-
tion showed no haze. Figure 11-10 demonstrates the Atlas
topography, WASCA HOAs, and Artemis epithelial and
stromal thickness, as well as difference maps for each. Front
corneal surface topography showed that the corneal surface
was significantly more regular and the optical zone was
centered on the corneal vertex. The Atlas difference map
demonstrated an area of flattening superiorly, which was
the location of maximum ablation. WASCA showed that
a significant reduction in HOAs had been achieved, with
the higher-order RMS reduced by 60% from 1.63 µm to
0.65 µm, and spherical aberration reduced by 90% from
1.30 µm to 0.13 µm. The Artemis epithelial thickness pro-
Figure 11-10. Artemis display illustrating preoperative and file showed that the epithelium had remodeled according to
postoperative axial maps (top row), epithelial thickness profiles the new stromal surface; the effective optical zone on the
(second row), stromal thickness (third row), and wavefront epithelial thickness map was centered about the corneal
aberrations (bottom row). vertex, and the diameter of the annulus of thin epithelium
had increased to just over 6 mm. The epithelial change map
showed that the epithelium had become thinner centrally
as the center of the topography mires, which is calculated and thicker in a paracentral annulus to achieve the new
automatically by the Atlas and displayed in the CRS-Master profile. The Artemis stromal thickness change map repre-
software. During surgery, the corneal vertex was determined sented the profile of stromal tissue removed by the irregular
as the first Purkinje reflex, seen as the patient fixated coaxi- topography-guided ablation; there was an area of maximum
ally with the aiming beam and the view of the surgeon’s tissue removal superiorly with a tongue of tissue removed
contralateral eye through the Carl Zeiss operating micro- inferiorly, which matched the intended ablation profile. The
scope in the MEL 80. A 6.50-mm optical zone was used maximum change in stromal thickness was 160 µm, which
with a transition zone out to 8.50 mm. The ablation profile was 40 µm more than the predicted ablation. It is unlikely
was irregular with an area of deepest ablation superiorly and that the difference between the predicted ablation depth and
a tongue of ablation inferiorly. The maximum ablation depth the measured stromal thickness change was due to the wet
was 120 µm located 1.4-mm superiorly and 0.8-mm nasally, PTK ablation.
where the Artemis pachymetry was 518 µm, which resulted Despite the dramatic changes in topography and HOAs,
in a predicted postoperative corneal thickness of 398 µm. there was very little change in contrast sensitivity, which
The ablation depth at the corneal vertex was 86 µm, where was still below the normal range.
the Artemis pachymetry was 532 µm, which resulted in a In summary, evaluation of the change in stromal thick-
predicted postoperative corneal thickness of 446 µm. ness using Artemis VHF digital ultrasound demonstrated
The ablation profile included a pre-ablation of 6-µm excellent agreement with the pattern of the theoretical
phototherapeutic keratectomy (PTK), which was pro- planned ablation.
grammed into the same treatment profile, to homogenize the
250 Ch a p t e r 1 1

SCH EIMPFLUG -BASED TOPOGRAPH Y-G UIDED C USTOM

ABLATION TREATMENT
Renato Ambrósio Jr, MD, PhD; Frederico P. Guerra, MD;
and Cristiana de Moraes Ramalho, MD

T
he use of OCULUS tomography with Allegretto fourth-generation fluoroquinolone and prednisolone acetate
excimer treatment has reportedly been successful 1% six times a day until complete epithelialization. One
in cases of irregular astigmatism.4 A 42-year-old week after surgery, uncorrected distance visual acuity was
female physician with a corneal scar following contact lens- 20/30, the contact lens was removed, and the patient started
related Acanthamoeba keratitis presented with uncorrected topical cyclosporine twice daily and was instructed for ste-
distance visual acuity of 20/400 and manifest refraction of roid tapering.
+0.50 -4.50 x 20, improving the vision to 20/60 in the left One month after surgery, uncorrected distance visual
eye. acuity was 20/25+. Four months after surgery, uncorrected
Advanced Therapeutic Surface Ablation was performed distance visual acuity was 20/25, and manifest refraction
with the Alcon-WaveLight Allegretto Eye Q. Epithelial was plano -1.25 x 165, giving 20/20.
removal was accomplished using a 50-µm PTK ablation Figure 11-11 illustrates the subtraction sagittal curvature
with a 7.0-mm optical zone and transition to 8.9 mm. map from preoperatively and 4 months postoperatively. Note
Topography-guided ablation was planned based on Oculyzer the reduction of keratometric astigmatism from 5.4@178 to
data. Mitomycin C at 0.005% was applied for 1 minute, fol- 0.7@55. Note the subtraction map is very similar as the
lowed by 50 cc of chilled BSS irrigation. A soft contact ablation TCAT plan (Figure 11-12).
lens was placed, and the patient was instructed to use a

Figure 11-11. The difference in

the sagittal curvature map from
preoperative (B) and 4 months
postoperative (A) corrective treat-
ment. Note the reduction of
astigmatism on the cornea from
-5.40 D to 0.70 D.
 To p o gra p hy-G u id e d Cu st o m Ab la t io n 251

Figure 11-12. Ablation profile

demonstrating the irregular abla-
tion to address the variable astig-
matism.

R EFERENCES
1. Reinstein DZ, Archer TJ, Gobbe M. Combined corneal topography 9. Reinstein DZ, Archer TJ, Gobbe M, Silverman R, Coleman DJ.
and corneal wavefront data in the treatment of corneal irregularity Stromal thickness in the normal cornea: three-dimensional display
and refractive error in LASIK or PRK using the Carl Zeiss Meditec with Artemis very high-frequency digital ultrasound. J Refract
MEL 80 and CRS-Master. J Refract Surg. 2009;25(6):503-515. Surg. 2009;25:776-786.
2. Wu L, Zhou X, Ouyang Z, Weng C, Chu R. Topography-guided 10. Reinstein DZ, Archer TJ, Gobbe M, Silverman RH, Coleman DJ.
treatment of decentered laser ablation using LaserSight’s excimer Repeatability of layered corneal pachymetry with the Artemis
laser. Eur J Ophthalmol. 2008;18(5):708-715. very high-frequency digital ultrasound arc-scanner. J Refract Surg.
3. Lin DT, Holland SR, Rocha KM, Krueger RR. Method for 2009;Nov:1-14.
optimizing topography-guided ablation of highly aberrated eyes 11. Reinstein DZ, Srivannaboon S, Archer TJ, Silverman RH, Sutton H,
with the ALLEGRETTO WAVE Excimer Laser. J Refract Surg. Coleman DJ. Probability model of the inaccuracy of residual stro-
2008;24(4):S439-S445. mal thickness prediction to reduce the risk of ectasia after LASIK
4. Jankov MR 2nd, Panagopoulou SI, Tsiklis NS, Hajitanasis GC, part I: quantifying individual risk. J Refract Surg. 2006;22:851-860.
Aslanides M, Pallikaris G. Topography-guided treatment of irregu- 12. Reinstein DZ, Srivannaboon S, Archer TJ, Silverman RH, Sutton H,
lar astigmatism with the WaveLight excimer laser. J Refract Surg. Coleman DJ. Probability model of the inaccuracy of residual stro-
2006;22(4):335-344. mal thickness prediction to reduce the risk of ectasia after LASIK
5. Toda I, Yamamoto T, Ito M, Hori-Komai Y, Tsubota K. Topography- part II: quantifying population risk. J Refract Surg. 2006;22:861-
guided ablation for treatment of patients with irregular astigmatism. 870.
J Refract Surg. 2007;23(2):118-125. 13. Reinstein DZ, Archer TJ, Gobbe M. Corneal ablation depth readout
6. Reinstein DZ, Silverman RH, Sutton HF, Coleman DJ. Very high- of the MEL80 excimer laser compared to Artemis three-dimension-
frequency ultrasound corneal analysis identifies anatomic correlates al very high-frequency digital ultrasound stromal measurements.
of optical complications of lamellar refractive surgery: anatomic J Refract Surg. 2010;26(12):949-959.
diagnosis in lamellar surgery. Ophthalmology. 1999;106:474-482. 14. Reinstein DZ, Archer TJ, Gobbe M, Silverman RH, Coleman DJ.
7. Stojanovic A, Jankov MR. Treatment of irregular astigmatism: Epithelial thickness in the normal cornea: three-dimensional dis-
developing an ideal corneal surface with the iVIS suite. In: Wang play with Artemis very high-frequency digital ultrasound. J Refract
M, ed. Irregular Astigmatism: Diagnosis and Treatment. Thorofare, Surg. 2008;24:571-581.
NJ: SLACK Incorporated; 2008:211-218. 15. Applegate RA, Thibos LN, Bradley A, et al. Reference axis selec-
8. Reinstein DZ, Silverman RH, Raevsky T, et al. Arc-scanning very tion: subcommittee report of the OSA Working Group to establish
high-frequency digital ultrasound for 3D pachymetric mapping of standards for measurement and reporting of optical aberrations of
the corneal epithelium and stroma in laser in situ keratomileusis. the eye. J Refract Surg. 2000;16:S656-S658.
J Refract Surg. 2000;16:414-430.
252 Ch a p t e r 1 1

16. Reinstein DZ, Couch DG, Archer TJ. LASIK for hyperopic astig- 17. Reinstein DZ, Archer TJ, Gobbe M. LASIK for myopic astigma-
matism and presbyopia using micro-monovision with the Carl Zeiss tism and presbyopia using non-linear aspheric micro-monovision
Meditec MEL80. J Refract Surg. 2009;25:37-58. with the Carl Zeiss Meditec MEL 80 Platform. J Refract Surg.
2011;27(1):23-37.
Ch a p te r 12

Planning
Ke rato re fractive
Treatme nts Using Wave fro nt
and Co rneal To p o grap hy
Data
Noel Alpins, FRANZCO, FRCO phth, FACS
and George Stamatelatos, BScO ptom

T
he differences that frequently exist between corneal because treatment by refractive parameters alone will leave
parameters measured by keratometry or topography the neutralization of this internal optical aberration (ORA)
and the refractive astigmatism values, as measured on the cornea. In approximately 7% of cases, this amount
by manifest or wavefront techniques, have become increas- can be worse than the preoperative corneal astigmatism,1
ingly evident to the practicing refractive surgeon. The potentially leading to increased aberrations and associated
importance of preoperatively quantifying this parameter is visual symptoms, particularly under mesopic conditions.
a key ingredient in successful refractive astigmatism treat- Equally, treatment by topographic parameters alone will
ment. The ocular residual astigmatism (ORA), defined as leave all the ORA postoperatively remaining in the manifest
the vectorial difference between the corneal and refractive refraction, again reducing the potential satisfactory visual
astigmatism, singly takes into account the angular as well outcome.
as the magnitude difference.1-5 This should form part of Ideally, a treatment paradigm incorporating both refrac-
any routine eye assessment prior to performing refractive tive and topographic parameters in a systematic manner
surgery that includes astigmatism treatment. would leave the minimum possible total astigmatism in the
In the excimer laser field, the perplexing decision wheth- eye, distributed in an optimal proportion between the cor-
er to treat based on refractive, topographic, or a combination nea and refraction.
of both parameters continues to be of relevance.6-8 This is

Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical

253 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 253 -258)
© 2012 SLACK Incorporated
254 Ch a p t e r 1 2

A B

Figure 12-1. (A) Polar diagram of refractive cylinder (corneal C

plane) at positive axis and simulated keratometry. (B) Double
angle vector diagram (DAVD) showing a “doubling” of the
positive cylinder axes without a change in the astigmatic
magnitudes. (C) Polar diagram displaying the O RA as it would
appear on the eye.

C ALCULATION OF ORA
Figure 12-1A shows on a polar display the second-order
spherocylinder of the wavefront or the positive cylinder of
the manifest refraction measurements, together with the
topography or keratometry measurements for corneal astig-
matism. Doubling the axes of the astigmatism while leaving
the magnitudes unchanged allows for the conversion of polar
coordinates to Cartesian coordinates (Figure 12-1B). The
ORA, determined on the double-angle plot, is then trans-
ferred to the origin (x=0, y=0) and halved to simulate how it
would exist within the eye (Figure 12-1C—polar diagram). Figure 12-2. The iAssort software program for corneal astigma-
This vectorial difference, measured in diopters and degrees tism analysis. Simulated keratometry parameters are exported
and calculated using simple trigonometric principles, has from the topography system (in this example, 2.95 D @ 130)
a proportional relationship to astigmatism—meaning that, and are compared to the refractive astigmatism (-1.29 D x 37
while the astigmatic differences between refractive and cor- at the corneal plane) to calculate the O RA, which in this case
neal astigmatism increases in either magnitude and/or angle is 1.68 D Ax 42. This is high in magnitude as displayed by the
red cross, which indicates that it lies outside the normal range
difference, so too increases the magnitude of the ORA. As a
of 0.00 D to 1.00 D.
result, treatment using refractive parameters alone neutral-
izes the internal ocular astigmatism quantified by the ORA
on the front corneal surface, leading to increased aberra-
simulated keratometry parameters exported directly from
tions and a reduction in the quality of vision achieved.9,10
topography and refractive cylinder at the corneal plane.
There are currently software programs available that will
Conveniently, these can be used directly from the analytical
calculate the ORA (Figure 12-2, www.assort.com) from
outputs generated within the topography device.
 Pla n n in g Ke ra t o r e fra c t ive Tr e a t m e n t s U sin g Wa ve fr o n t D a t a 2 55

By quantifying the vectorial difference between the with normal (emmetropic) vision can suffer from aberra-
corneal and the refractive astigmatism using the ORA, tions that affect functional vision.14
the maximum possible correction of astigmatism can be Alternatively, topography-guided treatments do not con-
determined. The distribution of the remaining ORA, which sider that the amount of corneal astigmatism often differs
is the best outcome possible for an individual eye, needs from the refractive (second-order) astigmatism. Omitting
to be considered carefully. Caution needs to be given to this phenomenon during planning can result in refractive
whether to treat the eye as is customary practice, leaving the astigmatism surprises.
astigmatism totally on the cornea by treating with manifest The advantages of preoperatively addressing both cor-
or wavefront refraction, or ascertaining if it is better to dis- neal and refractive astigmatism reduces the level of astig-
tribute the unavoidable remaining astigmatism between the matism that is left on the cornea compared to using refrac-
two in an optimized manner to achieve the most favorable tive parameters alone, and, as a result, fewer second- and
outcome. third-order aberrations are likely to remain.2,3,9

VECTOR P LANNING Vector Pla n n in g St u d ies

Ap p lied to Kerato con u s a n d
The Alpins method of vector planning directs the treat- Wavefron t-Gu id ed Treat m en t s
ment closer to the principal corneal meridian, thereby
correcting a greater amount of corneal astigmatism. This Two studies have shown the benefits of combining corne-
results in less corneal induced aberrations and a greater al and refractive parameters.9,12 The first is a retrospective
potential for improvement in best-corrected visual acu- study of 45 eyes with forme fruste and mild keratoconus.
ity.5,9,11,12 The reduction in corneal astigmatism substan- These patients underwent surface ablation laser surgery.12
tially exceeds the increase in measurable refractive cylinder All photoastigmatic refractive keratectomy (PARK) treat-
that might theoretically occur. This has the overall effect ments were optimized to leave minimum remaining corneal
of minimizing the total amount of astigmatism (refractive astigmatism, favoring a bias to a with-the-rule orientation.
plus topographic) remaining after laser surgery, effectively Postoperative results at 12 months showed that, on average,
gaining “something for nothing.” for every eye in the group, the corneal cylinder was reduced
This method of vector planning can be used employ- by an additional 0.59 D compared to results that would
ing any combination, either of keratometry or topography have been attained by treating refractive values alone. A
values and manifest wavefront or cycloplegic refraction mean 1.61 D for each patient in the study group would have
parameters, to determine the ORA amount present. The remained on the cornea postoperatively compared to the
importance of vector planning when preoperatively assess- 1.02 D achieved. This was achievable without compromis-
ing an astigmatic patient for refractive laser surgery can- ing the refractive outcome. In fact, it was enhanced with a
not be overemphasized. In cases where the ORA is high better-than-predicted refractive astigmatism reduction. The
(>1.00 D), the patient needs to be advised that all of the expected mean refractive target was 0.59 D compared to
astigmatism in the optical system cannot be corrected no the mean achieved at 12 months postoperatively of 0.43 D,
matter how perfect the surgical treatment applied so that demonstrating a clear net gain when assessing both corneal
expectations of a perfect outcome are adjusted to a realistic and refractive modes.
level. No problems or adverse signs such as increase in corneal
There are several limitations of wavefront-guided treat- irregularity and curvature or progression of ectasia result-
ments. Wavefront-guided treatments do not address the ing in a reduction of uncorrected or best-corrected visual
excessive and unnecessary amount of corneal astigmatism acuity were detected.
that would be left remaining (equal to the ORA) on the In the second study, 21 eyes of 14 patients were dis-
cornea postoperatively. This would potentially compromise tributed into 2 groups in a prospective double-masked
visual acuity and contrast sensitivity outcomes, particularly study.9 One group was treated by wavefront parameters
under mesopic conditions. No recognition of topographic alone, the other by wavefront combined with topography
astigmatism values are taken into account during treatment values (WF&VP) using vector planning. For the WF&VP
with this approach. combined group, the treatment profile was calculated
They also do not consider the patient’s subjective appre- using simulated keratometry readings from the topography
ciation of astigmatism. The inclusion in the treatment of a (Humphrey Atlas) and second-order Zernike coefficients
patient’s conscious perception of his or her refractive astig- Defocus 4, Astigmatism 3, and 5 from the wavefront display
matism is likely to benefit satisfaction.3,13 (WaveScan Wavefront) of the entire eye.
Wavefront-guided treatments could result in corneal sur- Results showed a trend to greater correction of corneal
face irregularities3 if all aberrations measured are corrected astigmatism in the WF&VP combined group, with bet-
on the corneal surface. It is important to note that even eyes ter visual outcomes in low-contrast mesopic conditions, a
256 Ch a p t e r 1 2

Figure 12-3. The ASSO RT treatment screen displays both the Figure 12-4. The emphasis shown here relates to topography-
manifest refraction and the corneal astigmatism by topography. guided treatment where all the O RA (1.68 D) remains in the
The emphasis shown here is that of conventional treatment, manifest refraction (100% topography/0% wavefront refrac-
0% topography/100% wavefront refraction leaving all the tion), shown as the “Target = 0.84/-1.68 x 132.”
O RA on the cornea as displayed by the “Target = 1.68 @ 132.”

from the ORA axis to neutralize the internal (noncorneal)

greater reduction in horizontal coma, and greater potential error and results in the calculated zero astigmatism in the
for improvement in best-corrected visual acuity. postoperative refraction (shown as the light blue “Target”).
The target-induced astigmatism vector (TIA) is the amount
of astigmatism that we are attempting to correct, and for this
Vector Pla n n in g: A Ca se St u d y treatment to sphericize the refraction is 1.29 D Ax 37.
At the other extreme in the planning process, if we treat
The Alpins Statistical System for Ophthalmic Refractive this eye emphasizing the complete reduction of topographic
Surgery Techniques (ASSORT) program developed at astigmatism values alone (Figure 12-4), 1.68 D of the
NewVision Clinics uses vector planning and analysis in a unavoidable residual ORA will theoretically remain in the
paradigm that maximally reduces overall astigmatism and postoperative refraction in conjunction with a spherical cor-
has the option to favor a bias to with-the-rule astigmatism. nea. However, taking an optimized view of the situation and
This software has been used in this chapter for all calcula- incorporating a proportion of each into the overall treatment
tions. by shifting the emphasis for astigmatism reduction “to the
Using the example referred to in Figure 12-2, the left” increases the proportion of a preference for complete
ASSORT laser treatment module in Figure 12-3 displays corneal astigmatism correction, resulting in the maximum
the second-order spherocylindrical wavefront refraction, as ablation of treatment being more closely aligned to the
measured using aberrometry, to be -3.63 DS/-1.43 DC x 37 principal corneal meridian. Consequently, there is less “off
at vertex distance of 12.5 mm and -3.47 DS/-1.29 DC x 37 axis” effect to the corneal astigmatism, with more astigma-
at the corneal plane. tism reduction and less torque and meridian shift.15
The topographic data of the same astigmatic eye display Figure 12-5 shows the optimal treatment as determined
a simulated keratometry value of 2.95 D of astigmatism at from the surgical emphasis graph (Figure 12-6), which
the steepest meridian of 130 degrees. The total of corneal displays a linear relationship between the emphasis and
and refractive (corneal plane) together is 4.24 D, the corneal the orientation of the target astigmatism. Figure 12-6 is
astigmatism making up 70% of it. based on the notion that with-the-rule corneal astigmatism
The amount of uncorrectable astigmatism in this patient’s is favorable and against-the-rule astigmatism is relatively
eye is 1.68 D Ax 42 (ORA). The distribution of this is unfavorable, so that more corneal astigmatism is reduced
reflected in the “emphasis” bar, where 100% indicates a when its orientation is not favorable, which might also
goal of completely eliminating refractive astigmatism and include oblique astigmatism. The placement of the emphasis
0% shows the emphasis on completely reducing topographic on total corneal or refractive astigmatism correction is a
astigmatism by the treatment. decision that is guided by the surgeon.
If we treat conventionally, that is, with 100% reduction In this example, the meridian of the target topography
of manifest refractive astigmatism as shown in Figure 12-3, is 132 degrees. As it lies 42 degrees from a favorable
all of this (ocular) residual astigmatism will remain neutral- with-the-rule orientation of 90 degrees, 42/90 or 47% can
ized on the cornea. This is shown as the “Target” 1.68 D preferentially be apportioned to a topography-based goal
at a meridian of 132 degrees, which is 90 degrees away of zero astigmatism. By the same process, the refractive
 Pla n n in g Ke ra t o r e fra c t ive Tr e a t m e n t s U sin g Wa ve fr o n t D a t a 2 57

Figure 12-5. ASSO RT Treatment Planning shows how the Figure 12-6. Linear relationship of surgical emphasis for
O RA of 1.68 D Ax 42 is apportioned 47% to eliminating the sphericity versus orientation of target topography based on
topography astigmatism and 53% to the refractive cylinder. the notion that with-the-rule astigmatism is favorable and
The O RA is neutralized by an equivalent 0.89 D at the cornea against-the-rule is unfavorable. This graph displays that, as
and -0.79 DC at the spectacle refraction but at an orientation the targeted corneal astigmatism approaches 90 degrees, the
90 degrees away of 132 degrees. target refractive cylinder is reduced, so if the targeted corneal
astigmatism is at 90 degrees, all the O RA will be corrected on
the cornea.

astigmatism power axis of 42 degrees is 48 degrees away

from 90 degrees and hence places 53% emphasis to a cor- initial ORA, the surgery then fails to achieve the maximum
rection of refractive astigmatism. The TIA here is greater astigmatism treatment, and the astigmatism is either under-
than before at 2.07 D Ax 39. or overtreated, leaving an excess amount of astigmatism
The overriding surgical principle is to approach corneal remaining.1
sphericity when the orientation of the target corneal astig- As technology advances with future possibilities, each
matism to remain becomes increasingly unfavorable. It is and every point that is measurable on the cornea by topog-
a surgical decision guided by which orientation is more raphy, together with wavefront aberrometry, could be “vec-
favorable than others. Thus, the surgical emphasis needs tor planned,” resulting in the ultimate ablation profile with
to be apportioned accordingly, using a linear relationship smooth, even transitions and an optimized amount of cor-
or other formula. For most astigmatism treatment, the neal astigmatism remaining with the maximal treatment.
emphasis of 60% to refractive astigmatism (rather than
100% conventional) is a safe choice, gaining a reduction
in remaining corneal astigmatism of 40%. In some cases,
another option is maintaining the proportion of corneal C ONCLUSION
to refractive astigmatism (70%/30%) after the surgery as
existed before surgery was undertaken with an emphasis of Refractive treatments based completely on either extreme
70% to refractive astigmatism. of topographic or refractive data alone can not attain the
The patient’s ORA is still 1.68 D but is now apportioned same optimal results in most astigmatic patients as vector
between the refraction and the corneal parameters. The planning, which combines both parameters in a balanced
corneal astigmatism targeted in this example is 53% of optimized manner. Wavefront-guided laser refractive sur-
1.68 D (0.89 D) and the remaining 47% (0.79 DC) of the gery has certainly been of benefit in correcting, on the cor-
emphasis placed refractively in a spherical equivalent of nea, the internal aberrations of the eye. However, the cor-
zero (+0.39 DS/-0.79 DC Ax 132). This remaining refrac- rection of higher-order aberrations in cases where there is a
tive astigmatism with a spherical equivalent of zero may not significant ORA needs to be more fully explored to increase
be fully perceptually evident to the patient, as shown in a overall patient satisfaction. The ORA is the key parameter
previous study,12 providing a net benefit of less remaining that quantifies the amount of excess corneal astigmatism
astigmatism. to neutralize the internal ocular second-order aberrations.
It is important to highlight that no matter what the Topographic-guided lasers provide comprehensive map-
percentage chosen on the “emphasis” bar, the maximal ping in situations of corneal irregularity where manifest
treatment enables the minimum amount of astigmatism to and wavefront refractions may be inadequate to provide a
be targeted, equal to the ORA. If the combined magnitude smoother corneal surface.
of the targeted remaining astigmatism is greater than the
258 Ch a p t e r 1 2

Using the Alpins method of vector planning to combine 6. Holladay JT, Bains HS. Optimized prolate ablations with the
information from the refraction (manifest or wavefront) NIDEK CXII excimer laser. J Refract Surg. 2005;21(5 Suppl):S595-
S597.
as well as the topography can help minimize astigmatism 7. Mrochen M, Jankov M, Bueeler M, Seiler T. Correlation between
remaining on the cornea and in the optical system of the corneal and total wavefront aberrations in myopic eyes. J Refract
eye. This optimizes visual outcomes, particularly in low Surg. 2003;19(2):104-112.
light and reduced-contrast environments. The ORA and 8. Kohnen T. Combining wavefront and topography data for excimer
the resultant astigmatism remaining on the cornea are the laser surgery: the future of customized ablation? J Cataract Refract
Surg. 2004;30:285-286.
essential parameters to guide the astigmatism optimization 9. Alpins NA, Stamatelatos G. Clinical outcomes of laser in situ
in the vector planning process. keratomileusis using combined topography and refractive wave-
front treatments for myopic astigmatism. J Cataract Refract Surg.
2008;34:1250-1259.
10. Kugler L, Cohen I, Haddad W, Wang MX. Efficacy of laser in
R EFERENCES situ keratomileusis in correcting anterior and non-anterior cor-
neal astigmatism: comparative study. J Cataract Refract Surg.
2010;36(10):1745-1752.
1. Alpins NA. New method of targeting vectors to treat astigmatism. 11. Alpins NA, Stamatelatos G. Combined wavefront and topography
J Cataract Refract Surg. 1997;23:65-75. approach to refractive surgery treatments. In: Wang M, ed. Corneal
2. Alpins NA. Astigmatism analysis by the Alpins method. J Cataract Topography in the Wavefront Era: A Guide for Clinical Application.
Refract Surg. 2001;27:31-49. Thorofare, NJ: SLACK Incorporated; 2006:139-143.
3. Alpins NA. Wavefront Technology: a new advance that fails to 12. Alpins NA, Stamatelatos G. Customized PARK treatment of myo-
answer old questions on corneal vs. refractive astigmatism correc- pia and astigmatism in forme fruste and mild keratoconus using
tion. J Cataract Refract Surg. 2002;18:737-739. combined topographic and refractive data. J Cataract Refract Surg.
4. Alpins NA, Walsh G. Aberrometry and topography in the vec- 2007;33:591-602.
tor analysis of refractive laser surgery. In: Boyd BF, Agarwal A, 13. Lipshitz I. Thirty-four challenges to meet before excimer laser tech-
eds. Wavefront Analysis, Aberrometers and Corneal Topography. nology can achieve super vision. J Refract Surg. 2002;18:740-743.
Panama: Highlights of Ophthalmology; 2003:313-322. 14. Williams D, Yoon GY, Porter J, Guirao A, Hofer H, Cox I. Visual
5. Alpins NA, Schmid L. Combining vector planning with wave- benefit of correcting higher-order aberrations of the eye. J Refract
front analysis to optimize laser in-situ keratomileusis outcomes. Surg. 2000;16:S554-S559.
In: Krueger RR, Applegate RA, MacRae SM, eds. Wavefront 15. Alpins NA. Vector analysis of astigmatism changes by flattening,
Customized Visual Correction; The Quest for Super Vision II. steepening, and torque. J Cataract Refract Surg. 1997;23:1503-
Thorofare, NJ: SLACK Incorporated; 2004:317-328. 1514.
Ch a p te r 13

Co rneal To p o grap hy
What Will the Up co ming Decad e Bring?

Lance J. Kugler, MD and Ming Wang, MD, PhD

I
n the 5 years since the first edition of this book was pub- predictable result. Despite these limitations, there is still
lished, there have been tremendous advances in the tech- room for improvement in the precision of excimer laser
nology used and understanding of corneal topography, ablation profiles. Such improvement will depend upon our
wavefront analysis, ocular biometry, and biomechanics. Such ability to combine preoperative topographic, wavefront, and
advances have led to our ability to determine the anatomi- biomechanical data to generate enhanced ablation profiles
cal shape and optical behavior of the eye in unprecedented and then link such profiles to excimer lasers that can deliver
detail. The volume of data generated by modern diagnostic the profile in a consistent, reproducible manner.
instruments is staggering and, at times, overwhelming. By
combining technologies, such as wavefront with topography
or biomechanics with Scheimpflug imaging, we increase
the value and clinical relevance of the data. Despite the R EFINEMENT OF
volume of diagnostic data at our disposal, the precision
with which we can surgically alter the optical performance M EASUREMENT
of the cornea has not kept up with our ability to define its
shape. It seems archaic in this era of advanced topography, Devices that combine corneal topography with wave-
tomography, aberrometry, biometry, and biomechanics that front aberrometry are uniquely able to define the source
the major determination of an ablation profile remains of an optical aberration (Figure 13-1).1,2 AMO’s iDesign
refraction. Modern excimer lasers are remarkably precise, platform is another example of this technology that has not
yet when used perioperatively, modern corneal topographers yet been released commercially. These systems subtract the
and aberrometers illustrate the limitations of current excimer corneal aberrations from the total or whole eye aberrations
laser surgery. Postoperative aberrometry reminds us of our measured to precisely identify the source of the optical
limited ability to remove higher-order aberrations (HOAs) so aberration and guide the clinician as to the best approach
beautifully described preoperatively by our devices. for surgical correction. In the context of refractive surgery
The technology gap is partly due to the speed by which screening, current devices are dependent upon relatively
diagnostic devices gain clinical and regulatory acceptance crude pattern recognition on the part of the clinician and
relative to surgical devices. However, it has been sug- are not able to reliably quantify the magnitude of the aber-
gested that excimer lasers have reached their peak in terms rations in question to the degree required for precise surgi-
of precision of laser-tissue interaction, and wound heal- cal correction. The future will undoubtedly lead to further
ing remains the largest and least-controllable barrier to a refinement of these combined devices.

Wang M. Co rn e al To p o g rap h y: A Gu id e f o r Clin ical

259 A p p licat io n in t h e W ave f ro n t Era, Se co n d Ed it io n (p p . 259 -262)
© 2012 SLACK Incorporated
260 Ch a p t e r 1 3

Figure 13-1. Combination of topography and wavefront

aberrometry allows location of aberrations.

Sp ecificit y of Screen in g others are currently being developed to analyze corneal

stiffness and physical shape changes induced by applana-
Posterior corneal surface analysis using tomography tion. Prior to the advent of these devices, we relied upon
is increasingly used to screen patients for subclinical corneal architectural characteristics to approximate corneal
ectatic disease. Modern tomographers such as the Pentacam biomechanical strength. For example, corneal thickness is
(OCULUS, Wetzlar, Germany) are able to detect subclinical an approximate representation of corneal strength in the
keratoectatic disease3-5 and measure corneal power6 with assessment of keratoconus. By incorporating measured
increasing sensitivity. As with any new diagnostic technol- biomechanical properties with physical measurements, an
ogy, however, increased sensitivity brings with it an impor- enhanced understanding of an eye’s response to keratore-
tant limitation, namely decreased specificity. Despite the fractive surgery may more accurately be defined, and can-
attention it receives in the literature and lay media, postker- didacy better determined.
atorefractive ectasia remains a rare condition and one that Dynamic measurement of deformation may be per-
drives further investigation using biomechanics. The danger formed using rasterstereographic corneal topography
of increased sensitivity in screening prior to refractive sur- (RCT), Scheimpflug imaging, and anterior segment OCT.
gery is that we may inadvertently deny refractive surgery to RCT employs a grid projected onto the fluorescein-covered
patients who do not carry significantly increased risk. corneal surface. The changes to the grid pattern induced by
Future advancements in the field of corneal topography an air puff similar to NCT are monitored by an imaging
will include enhancing efficiency by which we identify camera. The images are then processed by a computer to
clinically relevant trends and outliers within the massive generate data regarding the stiffness of the cornea.
volume of data at our disposal. Doing so will increase the Dynamic ultra high-speed Scheimpflug corneal imag-
specificity of our technology, thus improving our ability to ing of the CorVis (Oculus, Wetzlar, Germany) combines
separate patients at increased risk from those who are not. air-puff tonometry with dynamic Scheimpflug images to
Important work is already being done in this area, as exem- measure various descriptors, including applanation time,
plified by the BAD (Belin-Ambrosia) Enhanced Ectasia applanation lengths, highest deformation, central concave
scale on the OCULUS Pentacam. Future work in the area of curvature at the highest concavity point, and central thick-
refractive surgery screening will continue to further refine ness (Figure 13-2). Similar methods are being investigated
the specificity of screening parameters. using AS-OCT. Our approach to treatment of keratoconus
may change as well as we gain understanding regarding the
use of various surgical procedures to improve biomechani-
cal properties.
I NTEGRATION OF C ORNEAL
TOPOGRAPH Y AND Cu stom izin g IOL Ca lcu lat ion s
BIOMECH ANICAL P ROPERTIES As cataract surgery continues to evolve and merge with
refractive surgery, so must the instruments at our disposal.
Instruments that evaluate biomechanical characteristics A simple K reading and axial length is no longer sufficient
of the cornea are currently evolving. The Ocular Response to provide the precise results required for success with pres-
Analyzer first identified characteristics of elasticity, and byopic lenses. Eyes that have undergone keratorefractive
 Co r n e a l To p o gr a p hy: Wh a t Will t h e U p c o m in g D e ca d e Br in g? 2 61

Figure 13-2. An image compos-

ite of CorVis ST measurements.
Frames are captured around
the monitored the air jet pulse.
(Reprinted with permission of
Renato Ambrósio Jr, MD, PhD,
and Diogo Caldas, MD).

surgery now needing cataract extraction present further Direct measurement of the structure in question is the
challenges in the realm of IOL calculations. It is becoming goal of topography-guided excimer ablation. As the ocular
increasingly clear that our traditional methods of IOL cal- structure most responsible for refraction of light, the ante-
culation are not robust enough to handle postkeratorefrac- rior cornea is the surface from which the majority of HOAs
tive eyes.7-10 Traditional formulas require a few measured arise. By directly measuring this surface with corneal
biometry values and derive the rest of the anatomy based topography, and applying treatment based on the results, an
upon assumptions. These assumptions are no longer valid optically superior surgical result is possible. Topography-
following keratorefractive surgery and result in refractive guided ablations are promising for the treatment of irregu-
surprises following surgery. lar astigmatism, which remains one of the most elusive
As our understanding of ocular biometry and its relation- post-LASIK complications to correct. A valiant effort was
ship to IOL power and IOL position continue to improve, made toward this treatment modality by the Topolink sys-
so must our ability to accurately measure these values. tem in the late 1990s; however, results were disappointing.
For example, the refractive power of the posterior cornea A more recent generation of topography-guided platforms,
becomes clinically relevant for IOL calculations in eyes including the iVIS (LIGI Tecnologie Medicali, Taranto,
that are postkeratorefractive surgery. Initial attempts at Italy), Allegretto WaveLight (Alcon, Fort Worth, TX), and
incorporating posterior corneal power in the determination Zeiss MEL 80 (Carl Zeiss Meditec, Jena, Germany) show
of corneal power, such as the Holladay EKR report on the promise toward refinement of topo-guided technology.
Oculus Pentacam, will continue to be refined as the instru- As promising as topography-guided treatments may be,
ments allow for more reliable and reproducible posterior excimer ablation guided by a combination of corneal topog-
power measurements. Biometry may be incorporated into raphy and wavefront aberrometry will be the treatment
topographic systems, and tomographers will likely develop approach of the future.
improved algorithms for IOL power calculation. Overall, the theme of the next decade will be integra-
tion. As cataract surgery merges with refractive surgery, so
will the devices and technology we use to achieve the best
Top ograp h y-Dr iven possible outcomes. Diagnostic equipment will continue to
merge with excimer lasers to deliver more refined ablation
Keratorefract ive Su rger y patterns. In order to manage the vast amount of data that
will result from these instruments, improved methods of
Wavefront-guided excimer laser surgery continues to
data analysis and risk assessment will continue to be an
be a popular and well-established treatment modality.
important area of research and development.
However, this technology suffers from an important limita-
tion in that all aberrations of the entire visual system are
treated on the anterior corneal surface, regardless of their
site of origin. For example, treatment of a spherical cornea
with lenticular astigmatism results in an astigmatic cornea.
R EFERENCES
Similarly, HOAs arising from the crystalline lens or pos-
1. Won JB, Kim SW, Kim EK, Ha BJ, Kim TI. Comparison of internal
terior cornea are treated on the anterior corneal surface. and total optical aberrations for 2 aberrometers: iTrace and OPD
This approach may not provide ideal optical performance scan. Korean J Ophthalmol. 2008;22(4):210-213.
postoperatively.11 Furthermore, the excimer laser treatment 2. Fontes BM, Ambrósio R Jr, Jardim D, Velarde GC, Nosé W. Ability
required to correct HOAs is calculated based on wavefront of corneal biomechanical metrics and anterior segment data in
analysis rather than direct measurement of the curvature or the differentiation of keratoconus and healthy corneas. Arq Bras
Oftalmol. 2010;73(4):333-337.
irregularity of the corneal surface. In general, data that are 3. Fu J, Wang X, Li S, Wu G, Wang N. Comparative study of anterior
directly measured are more reliable and robust than math- segment measurement with Pentacam and anterior segment optical
ematically derived data. coherence tomography. Can J Ophthalmol. 2010;45(6):627-631.
262 Ch a p t e r 1 3

4. Oliveira CM, Ribeiro C, Franco S. Corneal imaging with slit- 9. Savini G, Hoffer KJ, Carbonelli M, Barboni P. Intraocular lens power
scanning and Scheimpflug imaging techniques. Clin Exp Optom. calculation after myopic excimer laser surgery: clinical comparison
2011;94(1):33-42. of published methods. J Cataract Refract Surg. 2010;36(9):1455-
5. Tang M, Chen A, Li Y, Huang D. Corneal power measurement with 1465.
Fourier-domain optical coherence tomography. J Cataract Refract 10. Kugler L, Cohen I, Haddad W, Wang MX. Efficacy of laser in
Surg. 2010;36(12):2115-2122. situ keratomileusis in correcting anterior and non-anterior cor-
6. Rosa N, De Bernardo M, Borrelli M, Lanza M. New factor to neal astigmatism: comparative study. J Cataract Refract Surg.
improve reliability of the clinical history method for intraocular lens 2010;36(10):1745-1752.
power calculation after refractive surgery. J Cataract Refract Surg. 11. Alpins N, Stamatelatos G. Clinical outcomes of laser in situ ker-
2010;36(12):2123-2128. atomileusis using combined topography and refractive wavefront
7. Ghanem AA, El-Sayed HM. Accuracy of intraocular lens power cal- treatments for myopic astigmatism. J Cataract Refract Surg.
culation in high myopia. Oman J Ophthalmol. 2010;3(3):126-130. 2008;34(8):1250-1259.
8. Naseri A, McLeod SD. Cataract surgery after refractive surgery.
Curr Opin Ophthalmol. 2010;21(1):35-38.
 Financial Disclosure s

Dr. Ashkan M. Abbey has no financial or proprietary inter- Dr. Diogo L. Caldas has no financial or proprietary interest
est in the materials presented herein. in the materials presented herein.

Dr. Amar Agarwal is a paid consultant to Abott Medical Dr. Massimo Camellin has no financial or proprietary inter-
Optics, STAAR Surgical, and Bausch & Lomb. est in the materials presented herein.

Dr. Noel Alpins is part of the ASSORT outcome and analy- Dr. Ana Laura C. Canedo has no financial or proprietary
ses program. interest in the materials presented herein.

Dr. Renato Ambrósio Jr is a consultant for OCULUS and Dr. Shihao Chen has no financial or proprietary interest in
Alcon. the materials presented herein.

Dr. Dianne Anderson has no financial or proprietary inter- Dr. Xiangjun Chen has no financial or proprietary interest
est in the materials presented herein. in the materials presented herein.

Dr. Timothy J. Archer has no financial or proprietary inter- Dr. Y. Ralph Chu has no financial or proprietary interest in
est in the materials presented herein. the materials presented herein.

Dr. Amin Ashrafzadeh is a consultant to Carl Zeiss Meditec. Dr. Ilan Cohen has no financial or proprietary interest in the
materials presented herein.
Dr. Harkaran S. Bains is a consultant for NIDEK.
Ms. Sonya M. Dakin is a contracted employee of Tracey
Dr. Michael W. Belin is a consultant to OCULUS GmbH. Technologies.

Dr. Megan Buliano has no financial or proprietary interest Dr. Michael J. Endl has no financial or proprietary interest
in the materials presented herein. in the materials presented herein.

Dr. Phillip M. Buscemi is a paid consultant for Nidek Co, Dr. Zhengjun Fan has no financial or proprietary interest in
Ltd. the materials presented herein.

263
264 Fin a n c ia l D isc lo su r e s

Dr. Claus M. Fichte has no financial or proprietary interest Dr. Katherine E. Paton has no financial or proprietary
in the materials presented herein. interest in the materials presented herein.

Dr. Marine Gobbe has no financial or proprietary interest in Dr. Leonardo N. Pimentel has no financial or proprietary
the materials presented herein. interest in the materials presented herein.

Dr. Frederico P. Guerra has no financial or proprietary Dr. Gaurav Prakash has no financial or proprietary interest
interest in the materials presented herein. in the materials presented herein.

Dr. Doug Horner has no financial or proprietary interest in Dr. Jia Qu has no financial or proprietary interest in the
the materials presented herein. materials presented herein.

Dr. Mirko R. Jankov II is a consultant for Alcon. Dr. J. Bradley Randleman has no financial or proprietary
interest in the materials presented herein.
Dr. Jason Jedlicka has no financial or proprietary interest in
the materials presented herein. Dr. Dan Z. Reinstein is a consultant for Carl Zeiss Meditec
and has a propietary interest in the Artemis technology
Dr. Paul M. Karpecki is a consultant for Topcon and on the (ArcScan, Inc) through patents administered by the Cornell
speaker’s bureau for Oculus. Center for Technology Enterprise and Commercialization
(CCTEC).
Dr. Stephen S. Khachikian has no financial or proprietary
interest in the materials presented herein. Dr. Karolinne Maia Rocha has no financial or proprietary
interest in the materials presented herein.
Dr. Stephen D. Klyce has no financial or proprietary interest
in the materials presented herein. Dr. George Stamatelatos is a consultant for ASSORT.

Dr. Randy Kojima is Director of Technical Affairs for Dr. Aleksandar Stojanovic has no financial or proprietary
Precision Technology Services. This is a GP manufacturing interest in the materials presented herein.
company and distributor of the Medmont topographer in
North America. Dr. Tracy Schroeder Swartz is a consultant for Tracey
Technologies.
Dr. Lance J. Kugler is a consultant for Alcon, Allergan, and
Refocus Group. Dr. Bradford L. Tannen has no financial or proprietary
interest in the materials presented herein.
Dr. Matthew Lampa has no financial or proprietary interest
in the materials presented herein. Dr. Nancy K. Tripoli is a parttime consultant for Optikon.

Mr. Sissimos Lemonis is an employee of Wavelight GmbH. Dr. Bruno F. Valbon has no financial or proprietary interest
in the materials presented herein.
Dr. Ray-Ann Lin has not disclosed any relevant financial
relationships. Dr. Shawna Hill Vanderhoof has no financial or proprietary
interest in the materials presented herein.
Dr. Bing Liu has no financial or proprietary interest in the
materials presented herein. Dr. Joe S. Wakil has financial interest in Tracey Technologies
and Eyesys Vision.
Dr. David D. Liu is Chief Technology Officer for LaserSight
Technologies, Inc. Dr. Ming Wang has no financial or proprietary interest in
the materials presented herein.
Dr. Naoyuki Maeda has a research grant from Topcon.
Dr. Junhua Xiao has no financial or proprietary interest in
Dr. Renzo Mattioli is a full time employee of Optikon 2000. the materials presented herein.

Dr. Murray McFadden has no financial or proprietary inter- Dr. Sonia H. Yoo has no financial or proprietary interest in
est in the materials presented herein. the materials presented herein.

Dr. Cristiana de Moraes Ramalho has no financial or pro-

prietary interest in the materials presented herein.