Pharmacological Criteria That Can Affect The Detection of Doping Agents in Hair

See
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/12625157
Pharmacological criteria that can affect the

detection of doping agents in hair
Article in Forensic Science International February 2000
Impact Factor: 2.14 DOI: 10.1016/S0379-0738(99)00176-0 Source: PubMed
CITATIONS
READS
47
120
3 authors:
Pascal Kintz
Vincent Cirimele
X-Pertise Consulting
Chemtox Laboratory
459 PUBLICATIONS 8,397 CITATIONS
SEE PROFILE
SEE PROFILE
Bertrand Ludes
Universit Ren Descartes - Paris 5
SEE PROFILE
All in-text references underlined in blue are linked to publications on ResearchGate,

letting you access and read them immediately.
Available from: Vincent Cirimele

Retrieved on: 04 May 2016
Foreword
So far doping control procedures have focussed on the detection of doping substances or their
signal metabolites in urine. However, since a method was developed to produce peptide
hormones by genetic engineering which subsequently led to an extensive marketing of these
hormones, blood tests have been increasingly discussed as a supplementary method for
doping detection. However, from a legal point of view, blood tests must be judged differently
from urine tests.
Apart from urine and blood, specialists are discussing other endogenous excretions or parts
(e.g. saliva, sweat, hair, nails) as being suitable for the detection of doping substances.
Methods of analysing the contents of hair have been successfully used in forensic medicine
for years. It is for example possible to verify the misuse of drugs in this matrix.
For this reason, the Federal Institute of Sport Science has supported research on detecting
doping substances in human hair. First positive results were achieved by Sauerwein and
Meyer who developed a method for the detection of clenbuterol in human scalp hair.
Although clenbuterol is used for therapeutic purposes in medicine, it is a banned substance
according to the list of the IOC. Currently, additional methods are being developed in
cooperation with the Forensic Institute of the University of Munich and the Institute for
Doping Analytics and Sports Biochemistry (IDAS) in Kreischa. The connection between
scientific know-how and instrumental analytics on the one hands and the chemistry of hair
and its contents on the other hand can give new impetus to doping analytics.
I am glad that the workshop in Kreischa has also led to a cooperation between the Federal
Institute of Sports Science and the International Society for Hair Analytics. It was only last
year that this society at its congress in Martigny stated clearly in a resolution that it
considered hair analytics as an important additional contribution to doping analytics.
I would like to thank the organizers of this workshop for paying attention to hair analytics and
hope that there will be fruitful discussions among the participants.
Dr. Martin-Peter Bch

Head of the Federal Institute of Sports Science
Contents
Page
Foreword............................................................................................................................
Martin-Peter Bch
Preface...............................................................................................................................
Hans Sachs
Introduction to Hair Analysis.............................................................................................

Hans Sachs
Hair Sample Collecting Kit for the Police Forces..............................................................

Werner Bernhard, Beat Aebi, Christian Staub, Hans Sachs
11
Workshop Stations
A General Solid-Phase Extraction Procedure for Drug Testing in Hair.............................
Christian Staub, Christle Girod
15
Determination of Benzodiazepines in Hair by GC/MS-NCI..............................................

Susanna Fehn, Hans Sachs
23
Specificity and Sensitivity Requirements in Hair Analysis................................................ 35

Detlef Thieme
Identification of Buprenorphine and Methadone in Hair by HPLC/MS............................. 49
Vincent Cirimele, Pascal Kintz, Bertrand Ludes
Interpretation of Quantitative Findings and Data Evaluation in Hair Analysis.................. 61
Michael Uhl
Criteria that Can Affect the Detection of Doping Agents in Hair......................................
Pascal Kintz
79
Reviews on Detection of Anabolic Agents in Hair

Analysis of Hair Samples for 2-Agonists.......................................................................... 91
Kurt Einhellig, Michael Uhl
Testing Endogenous Steroids in Hair.................................................................................
Hans Sachs
97
Detection of Exogenous Steroids in Hair........................................................................... 103

Detlef Thieme, Joachim Grosse and R. Klaus Mueller
Testing for 19-Norsteroids in Hair..................................................................................... 111
Pascal Kintz, Vincent Cirimele, Bertrand Ludes
Progress in Hair Analysis for Illegal Drugs
Preface
The Society of Hair Testing feels delighted being able to continue the tradition of conferences and
workshops concerning analytic methods for drugs in human hair.
There are two special reasons for holding the workshop 2000 at the Institute of Doping Analysis
and Sports Biochemistry: The conference in Martigny in 1999 has already shown that hair
analysis is expanding to the field of doping analysis. Secondly, there is an excellent equipment
for trace analysis in Kreischa, and the scientists of the institute, especially Detlef Thieme and
Joachim Grosse, supported by the director of the institute, Klaus Mller, have already published
well recognized contributions concerning the analysis of doping agents in hair. The preparation
of the workshop is showing that the whole institute staff is supporting this workshop, and
obviously enjoys hosting the participants from all over the world.
This workshop will not only introduce to fields which are new to some of the hair specialists but
also give a survey on routine methods and the evaluation of their results as well as on the
acceptance of hair analysis by the courts, traffic authorities and medical doctors, especially
psychiatrists.
This booklet does not only intend guiding the participants through the different stations of the
workshop, but also giving a survey on the modern field of hair analysis, including just recently
developed methods, to the members of the Society and to other interested scientists. As a major
part of the book is dealing with doping agents, it might also be interesting to authorities and
sports associations responsible for doping control.
The workshop would not have been possible without the support of the director of the
Bundesinstitut fr Sportwissenschaft (Federal Institute of Sports Science) Dr. Peter Bch and the
special promotion of Dr. Carl Mller-Platz. Their interest in new fields of modern toxicological
and biochemical analysis have encouraged us to provide the workshop in Kreischa.
Dr. Hans Sachs

President of the Society of Hair Testing
Preface
Introduction to Hair Analysis

Hans Sachs
Institute of Legal Medicine, Munich
Regular control of drug consumers by examining their hair starts with the publication of
Baumgartner 1979. He extracted opiates with methanol by heating for 2 hrs, then evaporating and
reconstituting in buffer and examining with Abuscreen RIA. In Germany this method was
introduced by Arnold in 1980. With the introduction of the new desktop GC/MS, the mass
selective detector, which lead the capillary directly to the ion source of the MSD, a new era began
because of the improved sensitivity. After morphine and codeine could be determined separately
in hair of a heroin abuser, the other drugs were also soon detected and a regular drug
consumption could be proved. Although possible contamination, cosmetic hair treatment, and the
irregular growth speed of the hair restrict the usefulness of hair examination, hair analysis is used
today for workplace testing in the United States in thousands of cases per day. In Europe, where
workplace tests on drug consumption are not performed consequently, hair analysis is used for
forensic and clinical applications.
The general fields of hair analyses are:
Criminal court proceedings

o Reliability of the defendant
o Question of chronic drug abuse by the victim (including hair of exhumed bodies)
Clinical purposes
o Detection of psychiatric patients with chronic drug abuse
o Explanation of withdrawal symptoms of neonates.
Driving ability control
Doping control?
Prior to the evaluation of the results of hair analyses it has to be considered that hair is not a
homogenous matrix like blood or urine, and it can only be homogenised to a certain extent before
examination.
Hair has a cylindrical fibre structure consisting of mainly three chemically different layers:
Cuticle
Resistant shield of keratinized cells of fish scale-like appearance with 4 m of thickness.
Cortex
The main layer of the hair fibre containing melanin granules which are responsible for the
colour of the hair. Strong racial influence leads to differences in hair thickness of
Caucasians (40 75 m), Afro-Americans (55 75 m), Spanish (65 80 m), and
Asians (65 95 m).
Medulla
If present, in the centre of the hair, rich in protein and lipids, with insoluble fractions.
Fig.1: CZ= cuticle cells; R = Cortex; M= medulla
The possible differences in races lead to juridical problems in multi-ethnic societies. Hair of
Afro-Americans could contain higher drug concentrations than hair of blond Caucasians after the
same drug consumption.
Today, drugs in hair are analyzed in forensic cases by GC/MS. Immunochemical methods can be
taken as a pre-test. In special cases when higher sensitivity and specificity are demanded, high
resolution MS or LC/MS are used.
Table 1
Screening Procedures for the Detection of Illegal Drugs in Hair
Reference s. chapter:
Legal aspects and data
evaluation......
Kauert and Rhrich
Moeller et al.
Kintz and Mangin
Analytes
heroin, 6-MAM,
dihydrocodeine, codeine,
methadone, THC, cocaine,
amphetamine, MDMA, MDEA,
MDA
Ultrasonic bath 5 min each5 ml
H2O5 ml acetone5 ml
heroin, 6-MAM,
MDA
20 ml H2O (2x)20 ml acetone
heroin, 6-MAM,
MDA
5 ml Cl2CH2
Decontamination step
Homogenization
Extraction
Clean-up
petrolether
100 mg hair cut into small
sections in a 30 ml vial
4 ml methanol ultrasonic bath
5 h, 50C
none
(2 x 5 min)
ball mill
ball mill
20-30 mg powdered hair, 2 ml

50 mg powdered hair, 1 ml 0,1
acetate buffer + M HCl, 16 h / 56 C
glucuronidase/aryl-sulfatase, 90
min/ 40C
NaHCO3; SPE (C18) , elution
with 2 ml acetone/CH2Cl2
(3:1)
Derivatization
propionic acid anhydride
1000 L PFPA/75 L PF-npropanol; 30 min/60C;

N2/60C; 50 L ethyl acetate
GC conditions
Column: 20m x 0.25 mm x 0.25

m methyl silicone; inj. temp.:
280 C; temp. prog. 140 C, 20
C/min to 300 C, 8 min
MS conditions
EI 70 eV; SIM at m/z

297, 313, 370 (THC)
82,182, 303 (cocaine)
268, 310, 327, 341, 369, 383,
397 (opiates)
235, 250 (methaqualone)
Column: 12m x 0.2 mm x 0.33

m phenyl methyl silicone; inj.
temp.: 260 C; temp. prog. 70
C, 30 C/min to 155 C, 10
C/min to 240, 1 min
EI 70 eV; SIM at m/z
118, 190 (amphetamines)
421, 300, 182, 303 (BZE &
cocaine)
282, 284, 390, 414, 444, 447,
473, 577 (opiates)
(NH4)2HPO4;extraction 10 ml
CHCl3/2-propanol/n-heptane
(50:17:33); organic phase
purified with 0.2 M HCl; HCl
phase to pH 8.4; re-extraction
with CHCl3
40 L BSTFA/1%TMS;
20 min/70C
The development of the methods used for opiates in the past 5 years is combined with the
development of screening methods for opiates, cocaine, cannabinoids, and amphetamine
including its derivatives. Three methods are dominating in the literature, as it is briefly described
in Tab. 1. The liquid-liquid extraction after HCl-hydrolysis introduced by Kintz et al. and the
solid-phase extraction after enzymatic hydrolysis with -glucuronidase/sulfatase lead to similar
results, both with the disadvantage, that heroin and 6-O-acetylmorphine (MAM) might be
hydrolyzed to morphine. The methanol extraction and direct detection with GC/MS is known
10
since the early 90s, but published in detail by Kauert and Rhrich not before 1996. It is
undoubtedly the simplest method with high sensitivy for heroin, cocaine, and THC, but a poor
sensitivity for their metabolites morphine, benzoylecgonine, and THC-COOH.
Hair Sample Collecting Kit for the Police Forces

Werner Bernhard1, Beat Aebi1, Christian Staub2, Hans Sachs3
1
Institute fr Rechtsmedizin, Universitt Bern, Bhlstrasse 20, CH-3012 Bern
Institut de Mdecine Lgale, Centre Mdical Universitaire, Avenue de Champel 9, CH-1211 Genve
Institut fr Rechtsmedizin, Universitt Mnchen, Frauenlobstrasse 7a, D-80046 Mnchen,
In 1997 the Society of Hair Testing published a general statement concerning the examination of
drugs in human hair. This statement made the following demands on the collection of hair
samples (1):
Sample collection should be performed by a responsible authority respecting the legal, ethical
and human rights of the person to be tested for drugs of abuse. Hair samples should be obtained
in a non drug contaminated environment by an appropriately trained individual, not necessarily
a physician. A sufficient amount of sample should be collected so that a repeat analysis or a
confirmation analysis by another laboratory can be performed should it be needed.
To fulfill these demands, hair collection is performed obeying the following rules (2,3):
1. Hair sampling must not be performed in the neighbourhood of street drugs and not by persons
who were in contact to those drugs.
2. To obtain comparable results, hair should be collected from the posterior vertex region of the
scalp. If this is not possible, the place from which the sample was taken, has to be
documented.
3. In order to maintain a statistical significance of the sample and/or to perform a screening test
as well as a confirmation test, the weight of the specimen should aim to be approximately 200
mg (a suitable amount of sample could be estimated by comparison with the diameter of a
pencil).
4. The hair strand has to be cut close to the skin. The length of remaining hair has to be
documented.
5. For shipment and storage, the hair sample should be wrapped in aluminum foil to maintain
integrity and to avoid contamination. (Long hair strands can be tied together with a piece of
string or of dental floss before being cut.)
12
6. Hair tip and the cut end of the samples have to be labelled on the aluminum foil.
7. Specimens can be stored under dry conditions at room temperature
In 1997, a hair sample collecting kit was introduced and training in hair sample taking was given
to police officers in Switzerland. The training course for the police forces was effective. The
content of the kit and the chain of custody form permitted a correct sampling avoiding
contamination and ensuring the identity of the specimens. The kit provided to the officers is
shown on the photgraph on the next page.
References:
1.
Statement of the Society of Hair Testing concerning the examination of drugs in human hair, Forensic Sci. Int.84
(1997) 3-6
2.
R. Denk, I. Raff, H. Sachs, Haaranalysen bei Betubungsmittelkonsum, Kriminalistik 4 (1992) 253-255
3.
United Nations, Guidelines for Testing Drugs under International Control in Hair, Sweat, and Saliva, United
Nations Publication, 1999, Sales No. E.99.XI.14
13
14
A General Solid-Phase Extraction Procedure for Drug Testing in Hair

Christian Staub and Christle Girod
Institute of Forensic Medicine, Geneva, Switzerland
1. Introduction
Different methodological approaches have been proposed for drug testing in hair. These include
pretreatment, ranging from simple elution with solvents to complete dissolution of the protein
structure by deep alkaline hydrolysis. After this pretreatments, different extraction techniques are
described ranging from liquid-liquid extraction to the so called solid-phase extraction.
The analytical procedures generally involve the following steps:

1. Decontamination of the specimen
2. Preparation (such pulverization or segmentation in 2-3 mm)
3. Hydrolysis (acid, alkaline or enzymatic...)
4. Purification (extraction, concentration and derivatization)
5. Analysis by chromatography
In our laboratory in Geneva (Switzerland), analysis of human hair for the detection of drugs of
abuse was first performed in 1995. Initially, request for hair analysis were few, and it is only
since 1997 that these analyses have become routine. As demand grew, we developed a solidphase extraction (SPE) method for the following drugs:
! codeine
! 6-monoacetylmorphine (6-MAM)
! morphine
! cocaine
! methadone
! ecstasy (MDMA)
! eve (MDEA)
Subsequently, the use of a robot ASPEC (Gilson Medical Electronics, Villiers-le-Bel, France)
allows to drop certain fastidious manipulations, and to treat a large number of samples at the
same time.
16
2. Basic principles of solid-phase extraction (SPE).

In general SPE can be used for three important purposes in up-to-date analyses:
! concentration of the analyte
! removal of interfering substances
! changing the matrix of the analyte as needed for subsequent analyses
In most cases these three effects occur together. Since analytes can be either adsorbed on the SPE
packing material or directly flow through while the interfering substances are retained.
The first case is shown in the figure below :
17
The sample is drawn through the solid phase, and the analyte molecules are enriched on the
adsorbent. Some interfering components and solvent molecules are not retained. Then remaining
interfering components are washed from the adsorbent with a suitable washing solution. Finally,
the analyte is removed from the adsorbent by elution with a suitable solvent.
The consideration made above indicate that an optimum SPE presents a poor columnchromatographic separation. However, this method can be utilised efficiently for a preseparation
of groups of compounds or a single analyte from the matrix. It is extensively used for clean-up by
solid phase extraction.
3. Analytical Procedure
3.1. Decontamination and pulverization
At the beginning, the hair is washed sucessively by percolation with 5 ml of methylene chloride,
5 ml of water and finally 5 ml of methanol. This step is very important to eliminate possible
external contamination. Then the hair is dried at 60 oC for 30 min and pulverised for 10 min at 70
s-1.
3.2. Hydrolysis
Drugs are fixed inside the hair matrix, therefore a digestion procedure is necessary before the
extraction. About 50 mg of powdered hair are placed in a tube and 1 ml of hydrochloric acid
0.01 M is added. After incubation at 60 oC for 12 hours, the solution is then neutralised with 1
ml of NaOH 0.01 M and buffered with 1 ml of phosphate buffer pH 7.0. After centrifugation at
4000 r.p.m. for 5 min, the supernatant was removed into a special tube for the extraction.
3.3. Extraction
Isolute HCX cartridges, provided by IST (Hengoed, UK), or Bond Elut Certify cartridges,
provided by Varian (CA, USA), could be used for the extractions.
The samples are extracted in the following ten steps:

1. Column conditioning with methanol (2 ml)
2. Column conditioning with water (2 ml)
3. Dispensing sample on the column
4. Rinsing with water (2 ml)
18
5. Rinsing column with acetate buffer pH 4 (1 ml)

6. Rinsing column with methanol (2 ml)
7. Drying column at 10 in. Hg for 5 min
8. Elution with methylene chloride/isopropanol (80/20) + 2% ammonia
hydroxide (2 ml)
9. Addition of the standard nalorphine
10. The eluent is dried under nitrogen
This extract could be analysed by GC/MS or other chromatographic methods.
3.4. Derivatization
GC/MS analysis requires a procedure of derivatization, for example by propionylation. 100

pyridine and 100
l of
l of propionic anhydride (PA) is added to the extract and incubated for 30 min
at 60 oC. After evaporation under a stream of nitrogen, the extract is reconstituted in 50
l of
ethyl acetate.
3.5. Gas chromatography - mass spectrometry method
Helium is used as carrier gas with the following capillary column (DB-5MS, 15 m x 0.25 mm x
0.25
m, J&W Scientifics (Folsom, USA).
Temperatures: 85oC maintained for 1 min to 190oC at 15oC / min and maintained for 0.5 min, to
210oC at 2 oC/min maintained for 1 min, to 270oC at 20oC/min for the final 8 min. Injector
temperature is 270oC and injection is made in splitless mode.
Three
l of the sample are injected into the GC/MS system operating in selected ion monitoring
mode (SIM). The source and interface temperatures are 200 and 280oC respectively.
The detection is performed with the following ions: codeine m/z = 355 and 282, 6-MAM m/z =
383 and 327, morphine m/z = 397 and 341, cocaine m/z = 303 and 182, methadone m/z = 294
and 72, MDMA m/z = 114 and 162, MDE m/z = 162 and 72, nalorphine (chromatographic
standard) m/z = 423 and 367.
19
4. Validation data
Soaked standard hair is used for the validation process. This hair is prepared by adding an aqeous
standard opiate solution to the pulverized hair. The mixture is submitted to magnetic stirring for
at least five hours, then filtered and the hair is finally washed (see procedure described in section
3.1) in order to retain only the most strongly bound drug fraction.
In these conditions following validation data could be obtained.
4.1. Linearity
The following calibration curves and correlation coefficients are obtained for a range of 1 to 20
ng/mg.
Compound
Target ion
Equation
codeine
355
y = 0.00286x - 0.03355
0.9987
6-MAM
383
y = 0.00146x - 0.02924
0.9979
morphine
397
y = 0.00127x - 0.00468
0.9999
cocaine
303
y = 0.00055x - 0.00071
0.9984
methadone
294
y = 0.00067x - 0.00487
0.9963
MDMA
114
y = 0.00510x - 0.16091
0.9969
MDEA
162
y = 0.03426x +1.52329
0.9958
4.2. Extraction recovery and limits of detection
Extraction recovery is determined by comparing the peak areas of an extracted hydrochloric acid
solution with the peak areas of methanolic solutions both at the same concentration. A posthydrolysis recovery is measured in these conditions.
The limit of detection (LOD) is evaluated as the lowest concentration giving a chromatographic
peak with the signal to noise ratio S/N = 3.
20
Compound
(n = 6)
Recovery (%) LOD (ng/mg)
codeine
95
0.05
6-MAM
80
0.05
morphine
103
0.05
cocaine
79
0.15
methadone
94
0.10
MDMA
81
0.10
MDE
71
0.05
The recovery of 103% obtained for morphine is probably due to a slight hydrolysis of 6-MAM.
4.3. Repeatability
Six replicates of soaked hair of low concentration and high concentration are analysed through
the complete procedure. The relative standard deviation (RSD) obtained is generally inferior to
10% for the two concentrations studied.
Compound
(n = 6)
High concentration
(ng/mg) Cv(%)
Low concentration
(ng/mg)
Cv(%)
codeine
31.9
7.0
4.2
7.5
6-MAM
26.6
2.4
3.5
7.9
morphine
32.9
3.8
3.5
6.4
cocaine
41.1
6.2
5.5
11.6
methadone
31.9
6.8
3.5
9.7
MDMA
27.7
5.0
7.9
6.1
MDEA
37.1
5.2
10.9
8.0
21
5. Concentrations of drugs in hair.
This analytical technique was used in routine in our laboratory for two years.
The table below summarises the various concentrations of drugs in hair.
Compound
mean (ng/mg)
median(ng/mg)
range(ng/mg)
codeine
32
0.9
0.5
0.1 - 5.6
6-MAM
39
4.0
1.7
0.1 - 45
morphine
36
1.1
0.7
0.1 - 6.0
cocaine
101
35.5
6.3
0.2 - 370
methadone
34
4.4
3.0
0.1 - 15
MDMA
3.8
1.1
0.6 - 14
6. References
(1) P. Zimmerli, Ch. Staub, Analyse de drogues dans les cheveux par extraction
automatise en phase solide, Rev. Fr. Labor. 282 (1996) 51-55.
(2) Ch. Girod, F. De Dominicis, R. Giovannini, Ch. Staub, Analyse de drogues dans
les cheveux par extraction automatise en phase solide: validation et utilisation
dans la routine, Toxicorama 1 (1999) 46-50.
(3) Ch. Girod, Ch. Staub, Analysis of drugs of abuse in hair by automated solid-phase
extraction, GC/EI/MS and GC/ion trap/CI/MS, Forensic Sci. Int. 107 (2000) 261-271.
22
Determination of Benzodiazepines in Hair by GC/MS-NCI

Susanna Fehn und Hans Sachs
1.
Target
The described method enables the qualitative detection and quantitative determination of
Diazepam, Nordazepam, Oxazepam, Flunitrazepam, Lorazepam, Lormetazepam, Prazepam,
Flurazepam, Midazolam and Triazolam in hair.
2.
Working Principle
After addition of the deuterated standard flunitrazepam-d7 to the hair which is cut into small
pieces the sample are extracted with Serensen buffer followed by liquid-liquid extraction with
1-chlorobutane. The extract is examined by GC/MS in SIM mode and the evaluation is made by
HP software and EXCEL data sheets.
3.
Field of Application
This method is created for the determination and quantification of
Diazepam
Nordazepam
Oxazepam
Flunitrazepam
Lorazepam
Lormetazepam
Prazepam
Flurazepam
Midazolam
Triazolam
in hair. About 50-150 mg of hair, cut into small pieces, are examined.
24
Precaution and Safety Regulations
The valid safety handling instructions informs about outgoing dangers.

Danger of infection: human material has to be regarded as potentially infectious.
4. Reagents and Supporting Material
For best results, the below mentioned chemicals should be used. These chemicals seem to be
optimal and cause no interference with extraction, chromatography and detection. Other reagents
and supporting material can be used after testing.
4.1. Chemicals and Reagents
4.1.1. Water (Braun Melsungen, Aqua ad injectabilia)

4.1.2. Acetone (Merck p.a. 14)
4.1.3. Petroleum ether (Merck 1.00915)
4.1.4. Disodium hydrogen phosphate (Merck p.a. 4873)
4.1.5. Potassium dihydrogen phosphate (Merck p.a. 6580)
4.1.6. Chlorobutane (Fisher Scientific HPLC C/4765/17)
4.1.7. Ethyl acetate (Merck Licrosolv 1.00868)
4.2. Solutions
4.2.1. Solution A:
KH2PO4
9,07g/L
pH 4,5
4.2.2 Solution B:
Na2HPO4 x 2H2O
11,83g/L pH 9,3
4.2.3 Serensen buffer: 19,6 ml Solution A + 80,4 ml Solution B (pH 7,4)

4.2.1
Stock solution 1g/l [see 4.3.1]

10mg of each substance are dissolved in a graduated flask in 10 ml of methanol
[4.1.5.] and an stored in a glas tube with a teflon screw cap.
storage: refridgerator
4.2.2
purchasable ampoule:
flunitrazepam-d7 (100 mg/l)
4.3 Stock Solutions and Calibrators

4.3.1 Stock solutions
concentration 1g/L of each substance :

Diazepam
Nordazepam
Oxazepam
Flunitrazepam
Lorazepam
Lormetazepam
Prazepam
Flurazepam
Midazolam
Triazolam
Dilutions:
concentration 1mg/l:
Oxazepam
Lorazepam
Midazolam
concentration 0,5mg/l:
Diazepam
Nordazepam
Flunitrazepam
Lormetazepam
Prazepam
Flurazepam
Triazolam
4.3.2. Calibrators
The solutions for calibration were made straight before use.

A mixture of Oxazepam, Lorazepam, Midazolam (1 mg/l each) and Diazepam, Nordazepam,
Flunitrazepam, Lormetazepam, Prazepam, Flurazepam, Triazolam (0.5 mg/l each) is made.
25
26
The calibrators are prepared by adding 5 l, 10 l and 20 l of the calibrator mixture to drug-free
hair samples.
Calibrator I: 50 pg/mg Diazepam, Nordazepam, Flunitrazepam, Lormetazepam, Prazepam,

Flurazepam, Triazolam, and 100 pg/mg Oxazepam, Lorazepam, Midazolam
Calibrator II: 100 pg/mg Diazepam, Nordazepam, Flunitrazepam, Lormetazepam, Prazepam,

Calibrator III: 200 pg/mg Diazepam, Nordazepam, Flunitrazepam, Lormetazepam, Prazepam,

4.3.2.2. Internal Standard (IS)

1 l flunitrazepam-d7 (10 mg/l) is added to the calibrators, to the samples as well as to one drug
free hair sample.
5.
Instrumentation and Supporting Material
(Other instrumentation and supporting material could be used after testing)
5.1 Instrumentation:
5.1.1. Vortex (Scientific Industries Vortex-Genie 2TM)

5.1.2. Shaker KS 10 (Laborgertebau Bhler,Tbingen, Germany)
5.1.3. Heating block and evaporation station (ReactiTherm III Heating Modul and
ReactiVap III, Pierce Rockford, USA)
5.1.4. Ultrasonic bath Transsonic Digital (Elma,Singen Germany)
5.1.5. GC/MS:
Gaschromatograph HP 6890 (Hewlett Packard, Bblingen,Germany)
Mass Selektive Detector HP 5973 (Hewlett Packard, Bblingen,Germany)
GC column J&W DB 5 MS, 30m x 0,25mm i.d. x 0,25m thickness
(Chromatographiehandel Mller,Fridolfing,Germany)
5.1.6. Centrifuge
27
5.2 Supporting Material
5.2.1 Graduated flasks 10 ml

5.5.2 Polypropylene reagent vials 30 ml (Sarstedt, Nrnbrecht, Germany Nr.60.543)
5.5.3 Pipetten:
Gilson Microman M10
Gilson Distriman
Eppendorf 100 l
5.5.4 Glas tube 10 ml: Duran-Glas,Schliff NS 14, volume 10ml
5.2.5 Glas stopper: Duran-Glas mit Schliff NS 14
5.2.6 Glas tube 7 ml
5.2.7 GC vials:
Glas injections vials including 100 l inserts, (Chromatographiehandel Mller,
Fridolfing, Germany no.610025 and no.610261, respectively)
Fitting caps N 11 TB/aA (Chromatographiehandel Mller, Fridolfing, Germany)
6.
Preparation prior to Analysis
Every vial to be used has to be signed clearly with the toxicological registration number, internal
calibration or control mark with a waterproof marker.
7.
Experimental
7.1.
General
Under normal conditions 50 150 mg of hair material are used for one hair analysis. Regarding
this amount, the findings have to be re-calculated.
Normally a strand of hair of 6 cm beginning at the head skin is to be examined. Samples refering
to police are divided up into three segments 3 cm each if possible.
7.2.
Sample Preparation and Extraction
1. The hair samples are cut as close as possible to the skin from the posterior vertex with a
scissor and are stored in a piece of aluminium foil at room temperature. The hair length is
28
documented and the needed segments are prepared. 50 150 mg of each segment is weighed
into a polypropylene vial [5.2.2] and cut into small pieces with a scissor.
2. The sample is washed with 5 ml of deionized water [4.1.1], 5 ml of acetone [4.1.2] and 5 ml
of petroleum ether [4.1.3], respectively. The hair sample is rocked for 5 min at each washing
step and the supertanant is removed and disposed. After addition of 4 ml Serensen buffer
and 5 l of flunitrazepam-d7 (conc. 1mg/l) as internal standard, the sample is incubated for 5
h at 50C in an ultrasonic bath.
3. The buffer phase is put into a 10 ml glas tube and is extracted by liquid/liquid extraction with
6 ml chlorobutane (rocking the sample for 1 min and centrifugation at 3000 rpm for 3 min).
The organic layer (upper phase) is pipetted in another glas tube (7ml) and is evaporated to
dryness under a stream of nitrogen.
4. The residue is reconstituted in 50 l ethyl acetate [4.1.7] and filled in a GC-vial [5.2.7].
The drug free hair sample and the calibrators are handled the same way.
7.3.
GC/MS - Analysis
7.3.1. General
The extract is analysed in the SIM-Mode especially for

A) Diazepam, Nordazepam, Flunitrazepam, Lormetazepam, Prazepam, Flurazepam, Triazolam
(Temperature Program A, see 7.3.2)
B) Oxazepam, Midazolam, Lorazepam (Temperature Program B, see 7.3.2)
For both methods 1 l is injected. The syringe of the autosampler is rinsed with methanol and
chloroform/ethyl acetate (1:1).
7.3.2. GC-conditions (see [5.1.5]):
Column J&W DB 5-ms (30 m x 0.25 mm i.d. x 0.25 m thickness)

Carrier gas: helium (flow 1,2 ml/min)
Injection volume: 1l (splitless)
Injector temperature: 250C
29
Temperature program A:
150C/ 1min 280C/9.5 min (rate 20C/min)
duration of analysis: 17,00 min
Temperature program B:
150C/ 1min 280C/5.5 min (rate 20C/min)
duration of analysis: 13,00 min
7.3.3. MS-conditions (see [5.1.5]):
The mass spektrometer is operated in the negative chemical ionization detection mode (+ 400 eV
above NCI autotune).
Methane is used as reagent gas and its flow is set 40% of max. flow with a pressure of 2.0 x10-4
torr in the ion source. The ion source temperature and quadrupole temperature are 150C and
106C, respectively.
Temperature program A:
Solvent-delay: 7.00 min
Following masses are recorded:
Lorazepam
m/z 302,304,306
Diazepam
m/z 284,286,285
Midazolam
m/z 325,327,326
Prazepam
m/z 324,325,326
Lormetazepam
m/z 304,306,308
Triazolam
m/z 306,308,307
Flunitrazepam-d7 (IS)
m/z 320,321
Temperature program B:
Solvent-delay: 7.00 min
Following masses are recorded:
Oxazepam
m/z 268,270,269
Nordazepam m/z 270,272,271

Flunitrazepam
m/z 313,314,315
30
Flurazepam
m/z 387,389,388
Flunitrazepam-d7 (IS)
m/z 320,321
7.3.4. Recording
Data recording and evaluation is made with HP Chemstation software for GC

Recording methods : BENZO3.M und BENZO4.M.
The elution of the substances takes place in the following order:
A) BENZO3.M
Lorazepam, Diazepam, Midazolam, Prazepam, Lormetazepam, Triazolam
B) BENZO4.M.
Oxazepam, Nordazepam, Flunitrazepam, Flurazepam
7.3.5. Evaluation
The integration of the peaks is made manually. Through the standard of calibration the
concentration of the substances is calculated with an EXCEL datasheet (1 point calibration).
For quantitation the so called Target-Ions are used [siehe 7.3.3]. The other ions are qualifier (+/20% range is allowed). To get peaks as qualitative positive the signal to noise ratio has to be >3.
The signal to noise ratio has to be >10 to quantitize the peaks.
8.
Documentation and Archivation
Internal Documentation
The printed chromatograms of each sample and the values which are calculated with the
EXCEL datasheet are attached to laboratory concerning identification sheets.
The prints of the calibrators are kept in special archive boxes.
The chronological order of the injections to the GS/MS is noted in a special logbook
referring to the gaschromatograph.
The used software automatically documents the recording date and the operator.
31
The results are transfered into an EXCEL sheet with the description of the used method and
operator. For result documentation an ACCESS database is used (Hair(year)Off97).
8.2.
Storage of Data, Samples and Hair Material
The files of the sequence runs and the single runs of the GC/MS are saved first on harddisk
later on CD to clear harddisk space.
The hair samples are archived for another two years.
The autosampler vials with the extraction residues and the examined hair samples are put to
the laboratory waste.
A. Negrusz et al, J. Anal. Toxicol.
V. Cirimele et al, J. Anal. Toxicol. 20
V. Cirimele et al, Forensic Sci. Int.
V. Cirimele et al, J. Leg. Med. 108/5
23/6 (1999) 429-435
(1996) 596-598
84/1-3 (1997) 211-218
(1996) 265-267
Analytes
flunitrazepam,
7-aminoflunitrazepam
flunitrazepam,
flunitrazepam,
lorazepam
Homogenisation
50 mg of hair are pulverized
approximately 50 mg of hair are
Extraction
3 ml Methanol + 5 ng flunitrazepam-
1 ml Serensen buffer (pH 7.6) +
1 ml Serensen buffer (pH 7.6) +
Serensen buffer (pH 7.6) + 5ng
d7 and 5 ng 7-aminoflunitrazepam-d7,
sonication for 1h, supertanant stored
20ng diazepam-d5, incubation

20h/40C, LLE (diethylether/CHCl3
20ng diazepam-d5, incubation

20h/40C, LLE (diethylether/CHCl3
lorazepam-d4, incubation 2h/40C,

LLE (diethylether/CHCl3 80/20)
in refrigerator, 3 ml HCl (0,1N)
80/20)
80/20)
50 l HFBA 60C/30 min,
150 l HFBA/ethyl acetate (2:1)
150 l HFBA 60C/30 min, 25 l
evaporation, 25 l ethyl acetate
60C/30 min, evaporation, 25 l ethyl ethyl acetate

acetate
min
Column 30 m x 0.25 mm x 0.25 m
Column 30m x 0.25 mm
Column 30m x 0.25 mm x 0.25 m
Inj. tem. 240C (splitless), 1l inj.
Inj. tem. 240C (splitless), 1.5l inj.
Tem. progr.: 60C/1min
310C/3min at a rate of 30C/min
NCI (+ 400 eV above autotune),
methane (1.3 torr), ion source 200C,

quadrupole 100C

quadrupole 100C

quadrupole 100C
Reference
pulverized
incubation overnight at 55C,

combined extracts SPE
(CH2Cl2/isopropanol/ammonium
hydroxide 78:20:2)
Derivatisation
GC-conditions
MS-conditions
-4
methane (3.7 x 10 torr), ion source

250C, quadrupole 106C
Table 1: Comparison of described methods for the determination of benzodiazepines in hair
35 l BSTFA + 1%TMCS 60C/20
M. Yegles et al, Forensic Sci. Int.
Institute of Legal Medicine Munich
K. M. Hld et al, Forensic Sci. Int. 84
P.Kintz et al, J. Chromatogr. B:
84/1-3 (1997) 211-218
(2000)
(1997) 201-209
Biomed. Appl. 677 (1996) 241-244
diazepam, nordazepam, oxazepam,
diazepam, nordazepam, oxazepam
alprazolam
nordiazepam, oxazepam
lorazepam, lormetazepam,
midazolam, lorazepam, prazepam,
flunitrazepam,
lormetazepam, flunitrazepam,
flurazepam, triazolam
Homogenisation
about 100 mg of hair are cut up into

small pieces
10-20 mg of rat hair are cut up into

small pieces
approximately 50 mg of hair are

pulverised
Extraction
2 ml acetate buffer (pH 4)+ deuterated
4 ml Serensen buffer (pH 7.4) + 5ng
Digestion with 2 ml 1N NaOH + 5ng
Serensen buffer (pH 7.6) + 200ng
standards, hydrolyzation with 70 l of

-glucuronidase/arylsulfatase for
triazolam-d4, 40C/overnight, LLE

flunitrazepam-d7, 5h/50C ultrasonic
bath, LLE (Chlorobutane), evaporation (toluene/CH2Cl2 7/3)
nordiazepam-d5 + 200ng oxazepam-d5,

incubation 20h/40C, LLE
2h/40C, supernatant was removed, 2
of chlorobutane, 50 l ethyl acetate
(diethylether/CHCl3 80/20)
Reference
Analytes
ml distilled H2O were added and after

shaking also removed, SPE of the
buffer fractions (acetone/CH2Cl2 3:1),
50 l ethyl acetate
Derivatisation
GC-conditions
35 l BSTFA + 1%TMCS 90C/20
1%TMCS 80C/30 min
min
Tem. progr. 70C/2min
Tem. progr. A: 150C/1min
220C/9.5min at a rate of 25C/min 280C/9.5min at a rate of 20C/min

Tem. progr. B: 150C/1min
255C at 5C/min 300C/7min
MS-conditions
25 l ethyl acetate + 25 l BSTFA +
EI (70 eV) ,
interface 280C
280C/5.5min at a rate of 20C/min
interface 310C, transferline 300C,

ionizer 130C
NCI, methane (0.6 torr),
-4
methane (2.0 x 10 torr), ion source

150C, quadrupole 106C
Table 1: Comparison of described methods for the determination of benzodiazepines in hair (continued)

methane (0.2 kPa), ion source 200C
34
Literature
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
M. Yegles, Y. Marson, R. Wennig, Influence of bleaching on stability of benzodiazepines

in hair. Forensic Sci. Int. 107/1-3 (2000) 87-92
A. Negrusz, C. Moore, D. Deitermann, D. Lewis, K. Kaleciak, R. Kronstrand, B. Feely, R.
S. Niedbala, Highly sensitive micro-plate enzyme immunoassay screening and NCI-GSMS confirmation of flunitrazepam and ist major metabolite 7-aminoflunitrazepam in hair.
J. Anal. Toxicol. 23/6 (1999) 429-435
K. B. Van-Loon, Haaranalyseresultaten benzodiazepinen roepen vragen op (Results of
hair analysis for benzodiazepines raise questions). Pharm. Weekbl. 133/8 (1998) 348-349
V. Cirimele, P. Kintz, B. Ludes, Screening for forensically relevant benzodiazepines in
human hair by gas chromatography-negative ion chemical ionization-mass spectrometry.
J. Chromatogr. B: Biomed. Appl. 700/1-2 (1997) 119-129
J. Segura, A. Redon, G. Gonzalea, C. J. Sanchez, L. San, Deteccion del consumo de
benzodiacepinas mediante analysis de pelo por metodos inmunologicos (Immunological
analysis of hair to detect benzodiazepines consumption). Rev. Toxicol. 14/1 (1997) 30-35
M. Yegles, F. Mersch, R. Wennig, Detection of benzodiazepines and other psychotropic
drugs in human hair by GC/MS. Forensic Sci. Int. 84/1-3 (1997) 211-218
K. M. Hld, D. J. Crouch, D. G. Wilkins, D. E. Rollins, R. A. Maes, Detection of
alprazolam in hair by negative ion chemical ionization mass spectrometry. Forensic Sci.
Int. 84 (1997) 201-209
V. Cirimele, P. Kintz, C. Staub, P. Mangin, Testing human hair for flunitrazepam and 7aminoflunitrazepam by GC/MS-NCI. Forensic Sci. Int. 84/1-3 (1997) 189-200
V. Cirimele, P. Kintz, P. Mangin, Determination of chronic flunitrazepam abuse by hair
analysis using GC-MS-NCI. J. Anal. Toxicol. 20 (1996) 596-598
V. Cirimele, P. Kintz, P. Mangin, Detection and quantification of Lorazepam in human
hair by GC-MS/NCI. Int. J. Leg. Med. 108/5 (1996) 265-267
P. Kintz, V. Cirimele, F. Vayssette, P. Mangin, Hair analysis for nordiazepam and
oxazepam by gas chromatography-negative ion chemical ionization-mass spectrometry. J.
Chromatogr. B: Biomed. Appl. 677 (1996) 241-244
P. Kintz, P. Mangin, Determination of gestational opiate, nicotine, benzodiazepine,
cocaine and amphetamine exposure by hair analysis. J. Forensic Sci. Soc. 33/3 (1993)
139-142
J. J. Sramek, W. A. Baumgartner, T. N. Ahrens, V. A. Hill, N. R. Cutler, Detection of
benzodiazepines in human hair by radioimmunoassay. Ann. Pharmacother. 26/4 (1992)
469-472
Specificity and Sensitivity Requirements in Hair Analysis
Application of High Resolution and Tandem Mass Spectrometry to the

Identification of Anabolic Steroids (stanozolol) or Benzodiazepines (oxazepam)
in Complex Hair Matrices
Detlef Thieme
Institute of Doping Analysis and Sports Biochemistry,
Dresdner Str. 12, 01731 Kreischa (near Dresden) GERMANY
Abstract
Hair analysis includes all analytical problems associated with the presence of low substance
concentrations in a dirty matrix. Extensive clean-up procedures and specific identification
techniques are analytical requirements in hair testing. The implication of this workshop station is
to demonstrate the capability and limitations of different sophisticated mass spectrometric
detection techniques, whilst all other influences (matrix, extraction, derivatisation,
chromatography and ionisation) remain unchanged.
Introduction
The requirement of sensitive and selective identification techniques is a logical consequence in

hair analysis, because the analytes potentially present in a complex matrix are typically
characterised by low concentrations and high lipophilicity.
Clean-up procedures to increase the concentration of the substances in question and to suppress
the background are mandatory but often not sufficient for an unequivocal identification. Simple
routine procedures like solid phase extraction or liquid-liquid extraction have often failed to
separate the lipophilic target compounds from a fatty matrix, and more sophisticated clean-up
techniques had to be applied.
Nevertheless, requirements to specificity and sensitivity of the detection techniques are still high.
Both aspects have to be adjusted according to the analytes properties, to optimise separation
(chromatography) and detection (mass spectrometry). This does not necessarily mean the
application of expensive equipment - specific derivatisation and/or ionisation techniques may be
36
effective as well but access to robust screening procedures applicable to larger groups of
substances or even to different substance classes is difficult. Oxazepam and stanozolol where
chosen as model substances to outline the influence of enhanced specificity in hair analysis and to
demonstrate, that false negative as well as false positive results can be prevented by sophisticated
techniques.
Instrumentation and Analytical Principles
Resolution in terms of mass spectrometry defines the capability of an instrument to differentiate

between adjacent masses. High resolution mass spectrometry is mainly associated with
application of sectorfield instruments, although the resolution of other types of mass
spectrometers has recently improved considerably (time of flight).
In contrast to the high technical standard (and costs) of HRMS instruments, the basic scientific
background of sectorfield instruments is straight forward and easy to understand. The mass
specific separation of ions is based on a balance between Lorenz and centrifugal forces in a
magnetic field and the resolution of the instrument is controlled by the width of the ion beam
(fig 1).
Substances which appear to be identical at unit mass resolution may be distinguished in principle
by HRMS, if their elemental composition is different. The most efficient applications are directed
to substances containing elements with negative mass defects (dioxins, benzodiazepines,
clenbuterol etc.) whereas substances with common composition (hydrocarbons) are less
specific.
Tandem mass spectrometry is based on combinations of at least two mass spectrometers. Almost
any type of spectrometer is suitable for such a coupling in space, whereas the coupling in time
means a subsequent application of the same device (ion trap) for several separation steps. The
most common instruments are triple stage quadrupoles but there are new upcoming techniques
like quadrupole / time-of-flight coupling [1].
The general intention consists in the separation of one specific fragment in the first stage. This
precursor ions are than fragmented in a collision cell and the resulting product ions are
analysed in a second mass spectrometer.
The type of experiments practically employed in this workshop-demonstrations is Multiple
Reaction Monitoring (MRM) where the resulting information is two-dimensional and
37
interference is only possible, if both masses (precursor and product ions) would match (as well as
the retention time). This means: only ions of the same mass, which undergo the same
fragmentation process could produce identical signals. This is by far less likely than coincidence
of masses in conventional MS.
CC
D
S
Fig.1.
SL
Schematics of a hybrid sectorfield / quadrupole mass spectrometer: Ions are formed by
electron impact in the source [S] and separated in a magnetic field [M]. The mass
spectrometric resolution of the first stage is controlled by a slit [SL] limiting the width of
the beam. Selected ions enter the collision cell [CC], where product ions are formed and
analysed in the second stage (quad [Q]).
The instrument used at the workshop is the hybrid version of the mass spectrometer AutoSpec
(Micromass), which is a combination of a high resolution sectorfield instrument with a
quadrupole.
This instruments permit a realistic comparison of different techniques, because GC-separation
and ionisation is identical and the differences between low resolution MS, high resolution MS
and tandem MS can be compared immediately by a variation of method parameters.
38
Experimental
3.1
Sample Preparation
3.1.1
Extraction
There is no optimum standard hair sample preparation procedure for anabolic steroids, which
covers the requirements of all substances with adequate recovery.
We have always used an optimised procedure rather than a general unknown screening. An
overview of different approaches is summarised in figure 2. The main difference between
methanol or sodium hydroxide extraction depends on the target substances, many of the steroids
especially their esters- are not stable under conditions of a hydroxide disintegration, whereas the
methanol extraction has insufficient recovery for some analytes (stanozolol). Clean-up and
derivatisation procedures are substance-specific too. The most effective and versatile procedure
was LC cleanup, which was applicable to most of the analytes and permits a simultaneous
separation / enrichment of different substances.
50-100 mg Hair Powder
Clean-Up
Derivatisation
Detection
HPLC-APCI_MS/MS
Extraction
Methanol Extraction
2h @ 55 deg
Ultra-Sonication
HPLC
SPE
enol-TMS
GC-HRMS
Disintegration
0.5n NaOH
15 min @ 95 deg
LLE
TMS/TFA
GC-MS/MS
GC-NCI_MS
Fig. 2. Scheme for hair sample preparation
3.1.2
39
Derivatisation
The mobile phase / organic solvent is removed at 80C in a nitrogen stream. The residue is than
desiccated and derivatized 30 min at 60C using the following reagent mixture to yield enol-TMS
derivatives:
MSTFA + NH4I + dithioerythritol
3.2
Experimental Parameters
3.2.1
GC separation
(15 ml + 100 mg + 10 mg).
12.5 m Optima-5 MS (Macherey-Nagel)

(0.2 mm ID, 0.35 m film thickness, crosslinked 5% phenyl-methyl-silicone)
injection parameters
- injection mode:
splitless
- injection volume:
1.5 l
oven temperature program

- initial temperature:
150 deg C
- initial time:
0.5 min
- ramp:
12.5 deg C / min to 340 deg C
- final time:
3.2.2
2 min
MS detection
- ionisation mode:
EI, 38 eV
- trap current
500 A
- source temperature: 250 deg C

HRMS parameter
- acquisition mode
selected ion recording (SIR),
- resolution
1.000 compared to 10.000
- ions detected
stanozolol
3OH-stanozolol
oxazepam
472.3305,
545.3415,
401.1271,
560.3650
415.1065,
429.1221,
430.1300
40
MS-MS parameters
- acquisition mode
multiple reaction monitoring (MRM-Q),
- collision energy
100 eV
- resolution
1000 (magnet), 2 amu (quad)
- fragmentation reactions
3.3
stanozolol
472"143,
472"457
oxazepam
430"267,
430"341
430"401
Experimental demonstration
3.3.1 Model Substances
The specificity of a high resolution or tandem-mass spectrometric detection depends on the target
substances properties and the experimental conditions.
Tandem mass spectrometry is not very effective, if the fragmentation pattern of a molecule is
very poor (for instance molecular ion only) or too complex (high number of low abundant peaks
in unspecific regions).
Fragmentation reactions should be substance- or substance-class specific, whereas reactions like
loss of water, of trimethylsilanol or methylgroups in a TMS derivative are not very specific.
The suitability of high resolution techniques depends on the presence of hetero-atoms in the
molecule. Halogens in particular reduce the accurate mass of the fragments compared to the
average mass of the background.
This means on the other hand, that increase of mass spectrometric resolution will not improve the
identification of a substance, coeluted by a interference, which is identical in chemical
composition or produces isomeric fragments. The model substances for the experiments are
chosen according to their relevance in hair analysis [2, 3] as well as to their chemical properties.
3.3.2
41
Oxazepam
Oxazepam is a benzodiazepine containing chlorine with several diagnostic fragments in the mass
spectrum (fig. 3). It carries polar amino and hydroxy groups and is therefore critical regarding
impurities in the gas chromatographic systems. The formation of a bis TMS derivative proved to
be helpful if it should be identified under routine conditions in an instrument which is used for
injections of MSTFA (fig. 4).
Both techniques (HRMS and MS/MS) are suitable for a sensitive identification, although more
classical options (negative CI [4], and station 2 of this workshop) are applicable too.
CH3
2.8E6
429
Si
415
O
N
OSi(CH3)3
N
Cl
M = 430
[M-90]
340
[M-15]
147
313
340
329
401
415
179
0.0E0
200
Fig.3.
300
400
Structure and mass spectrum of oxazepam-bis-TMS
500
m/z
42
1.4E4
8:41
268.2
N
OH
Cl
0.0E0
6:00
429.3
10:00
4.0E6
8:48
Si (CH3)3
O
N
OSi(CH3)3
Cl
0.0E0
6:00
10:00
Time
Fig.4. Structure and selected ion recording chromatograms of oxazepam. Comparison of

intensity and peak shape of the oxazepam standard with the bis-TMS-derivative.
401.1
415.1
429.1
43
430.1
2.1E6
0.0E0
401.1272
7:00
415.1065
429.1221
11:00
430.1300
5.1E5
0.0E0
7:00
430.2->267.1 @100.0
430.2->341.2 @100.0
11:00
430.2->401.2 @100.0
4.7E4
0.0E0
7:00
11:00
Time
Fig. 5. Comparison of low resolution (top), high resolution (middle) and tandem mass
spectrometry (bottom), applied to the identification of oxazepam in a spiked hair sample.
The comparison between the techniques (fig. 5) demonstrates, that concentrations equivalent to
10pg/mg can hardly be detected with conventional MS after derivatisation, but the detection limit
will be about 100 times higher, if a resolution of 10.000 is applied. Tandem mass spectrometry
was less successful in this application.
3.3.3 Stanozolol
Stanozolol was a very popular anabolic steroid, frequently abused by athletes and bodybuilders in
the past. The dependency of the detection limit on the stability of the chromatographic system,
caused by the amino group in the A-pyrazol-ring was a crucial problem for a long period of
time. The formation of mixed O-TMS / N-perfluoroalcylamide derivatives was attempted to
improve the chromatographic stability. Moreover, this permits a very specific NCI detection of
stanozolol in complex matrices [2], because no analogue derivatives of endogenous steroids exist.
44
The situation has improved considerably due to the refined passivation of GC columns and
detection of stanozolol and its metabolites is usually carried out based on their per-TMSderivatives. Fig. 6 shows the successful identification of stanozolol in a hair sample of a
bodybuilder by MS/MS.
545.3
560.3
1.0E9
13:18
0.0E0
12:36
13:00
545.3415
13:24
13:48
560.3650
6.0E6
13:17
0.0E0
12:36
13:00
560.3-->254.1
13:24
13:48
560.3-->520.3
1.2E6
13:17
0.0E0
12:36
13:00
13:24
13:48
Fig. 6. Comparison of low resolution (top), high resolution (middle) and tandem mass
spectrometry (bottom), applied to the identification of 3OH-stanozolol in a hair sample
of a bodybuilder.
The 3OH-stanozolol metabolite was chosen as sensitive criterion for the comparison of
techniques because of its lower concentrations, although the major product incorporated into hair
was the parent compound stanozolol (fig. 7).
The required sensitivity for stanozolol identification is relatively high, and conventional mass
spectra at unit resolution clearly exhibit lower detection limits. A comparison of identifications
using different detection techniques demonstrates the problem of high amounts of background
interfering the signals at low resolution. Sophisticated techniques have to be applied to identify
45
stanozolol (per-TMS) at appropriate peak to noise ratios. Tandem mass spectrometry is more
effective than HRMS in this example.
Starting to deal with identification of steroids in hair matrix, stanozolol was a helpful tool for
understanding of principles of incorporation into hair, because the parent compound was
detectable in hair at higher amounts than their typical urinary metabolites 3-hydroxy-, 4hydroxy and epi-stanozolol. In a fatal case, the concentration of the parent compound detected in
hair was 30 times higher than the concentration of the corresponding metabolite (fig. 7).
472.3305
6.8E6
Stanozolol (9ng / 50mg Hair)
15:20
OH
3.4E6
HN
N
H
0.0E0
12:00
560.3650
13:00
14:00
15:00
16:00
17:00
18:00
545.3415
1.9E5
3'-OH-Stanozolol (300pg / 50mg Hair)
15:49
OH
HO
9.7E4
HN
N
H
12:00
0.0E0
13:00
14:00
15:00
16:00
17:00
18:00
Fig. 7. Identification of stanozolol and 3hydroxy-stanozolol.
3.3.4
Miscellaneous Results
The spectrum of anabolic substances we have detected so far in hair is rather wide [5].
Based on forensic cases of steroid trafficking, diminished responsibility or fatalities amongst
bodybuilders, the random collection of substances comprises
Clenbuterol
46
Stanozolol
(3OH metabolite)
Metandienone
(epi-metandienone, 6-OH-metabolite)
Mesterolone
Metenolone enantate
Nandrolone decanoate and laureate
Testosterone phenylpropionate, propionate, decanoate, enantate and isocaproate
The most valuable result was the identification of testosterone esters in hair (fig. 8), because the
differentiation of the endogenous steroid testosterone in urine is a crucial problem. The
identification in hair of the injected esters (which are neither endogenous, nor excreted in urine)
is a unique marker of testosterone abuse.
472.3000->457.3
472.3000->209.1 472.3000->194.1
13:49
OCOC 6H13
50 ng Standard
12:00
0.0E0
16:00
14:00
15:41
70 mg Blank Hair
12:00
14:00
13:48
5.0E4
3.2E3
0.0E0
16:00
1.5E4
86 mg Suspicious Hair Sample
12:00
14:00
0.0E0
16:00 Time
Fig. 8.
Identification of testosterone enantate in hair of a bodybuilder by tandem mass
spectrometry.
47
Conclusions and Forensic Considerations
Specificitiy differences between GC/MS, GC/HRMS and GC/MS-MS are studied using the
identical chromatographic and ionisation techniques. The improvement of specificity by
application of HRMS and MS-MS is substance dependent. The model substances applied
were easily detectable by both techniques. Detection limits were improved considerably
compared to low resolution MS.
The identification criteria defined for low resolution MS (three diagnostic ions must be
detected, signal/noise and relative abundances below cut-off) are applied to high resolution
and tandem MS as minimum requirements.
Parent compounds are often incorporated into hair although they are (practically) not excreted
in urine. This offers the opportunity to identify diagnostic key substances for longer periods
of time. Especially precursors of endogenous steroids are attractive targets of hair analyses.
Dealing with steroids there are individual variations of endogenous steroid concentrations
depending on body site, scalp region, health, age and sex, extra to the usual restrictions in hair
analysis (hair pigmentation, treatment etc.) to consider.
Practical cases of identification of steroids in hair were devoted to forensic cases, (trafficking
of steroids, diminished responsibility, steroid related fatalities) where the number of possible
analytes was reduced and analyses could be optimised in advance. A general unknown
screening of all anabolic agents seems to be too complex to be effective.
Acknowledgement
This work was funded by the Bundesinstitut fr Sportwissenschaften, Kln,
Germany (VF 0414/02)
48
References
[1]
H. Budzikiewicz, Massenspektrometrie, VCH, Weinheim, New York, Basel, Cambridge

1992.
[2]
K. M. Hold, D. G. Wilkins, D. J. Crouch, D. E. Rollins, R. A. Maes, Detection of

stanozolol in hair by negative ion chemical ionization mass spectrometry, J Anal Toxicol
20 (1996) 345.
[3]
M. Yegles, F. Mersch, R. Wennig, Detection of benzodiazepines and other psychotropic

drugs in human hair by GC/MS, Forensic Sci Int 84 (1997) 211.
[4]
P. Kintz, V. Cirimele, F. Vayssette, P. Mangin, Hair analysis for nordiazepam and

oxazepam by gas chromatography--negative-ion chemical ionization mass spectrometry, J
Chromatogr B Biomed Appl 677 (1996) 241.
[5]
D. Thieme, J. Grosse, R. K. Mueller, H. Sachs, Detection of several anabolic steroids of

abuse in human hair,Manfred Donike Workshop (Cologne), Detection of several anabolic
steroids of abuse in human hair (1998).
Identification of Buprenorphine and Methadone in Hair by HPLC/MS

Vincent Cirimele, Pascal Kintz, Bertrand Ludes
Institut de Mdecine Lgale, 11 rue Humann
67085 Strasbourg Cedex, France
1. HPLC/MS determination of buprenorphine and norbuprenorphine in hair samples
Introduction
Buprenorphine (BUP) is an hemi-synthetic opioid derivative, closely related to morphine
and congener alkaloids, which is obtained from thebaine after a 7-step chemical procedure. BUP
is a powerful analgesic (25 to 40 times more potent than morphine) that exhibits both partial
agonist activity at the -opiate receptor and antagonist activity at the
-opiate receptor [1,2].
This drug has been initially developed for the treatment of acute and chronic pain, especially of
surgical or neoplastic origin. Its main advantages over morphine are a poor respiratory depressant
activity and a lack of significant withdrawal symptoms. A high-dosage, sublingual formulation of
buprenorphine is available in France since February 1996 (Subutex, Schering-Plough Labs.) for
the management of heroin addicts. As it may be ordered by any physician and is delivered under
limited control, this alternative to the methadone substitution previously organized in specific
detoxication centers only has led to some deviations (appearance of a black market, intravenous
injection of crushed tablets). Some addiction potential, cases of abuse and fatalities attributable to
the misuse of Subutex associated to benzodiazepines have been reported in France and other
countries (3).
As a consequence, it has become necessary for every forensic laboratory to be able to
assay BUP in biological samples and in hair specimens. For the past twenty years, hair analysis
has been proposed for identifying chronic drug abusers in forensic science. Drug concentration
along the hair shaft can be correlated with the time of drug use. Hair is known to allow a drug
administration to be tracked back for months, and thus offers the possibility of determinating
long-term drug exposure (4).
This paper describes the procedure based upon HPLC hyphenated to ionspray-mass spectrometry
(HPLC/ISP-MS) for the sensitive and specific determination of BUP and norBUP in human hair.
50
Material and Methods

Liquid Chromatography
The HPLC separations were performed at ambient temperature on a 4-m NovaPak
(Waters) C18 column (150 x 2.0 mm, i.d.) protected by a 5-m Opti-Guard (Interchim) C18
guard cartridge (15 x 1.0 mm, i.d.). The mobile phase was acetonitrile/2 mM NH4COOH buffer,
pH 3.0. A 20-ml, dual syringe HPLC pump (Applied Biosystems 140 B) was employed to deliver
a continuous flow of 200 L/min. A post-column split of 1 : 3 was used to reduce at 50 L/min
the flow rate infused into the HPLC/MS interface.
Mass Spectrometry
The MS detection was carried out on a Perkin-Elmer Sciex API-100 apparatus equipped
with an Ionspray (= pneumatically assisted electrospray) interface. Nitrogen (99.95 %, 40 psi)
was employed as both the nebulizing gas (flow rate 1.16 L/min) and the 'curtain gas' (flow rate
1.08 L/min) that prevents solvent vapours and solid contaminents from entering the vacuum
chamber. The ion sampling orifice was held at a potential of + 50 V, and the electron multiplier at
+ 2400 V. MS data were collected as either 1) total ion chromatograms (TIC) by monitoring the
signal over the mass range m/z 260-475 for drug identification, or 2) multiple ion monitoring
(MIM) at m/z 414 (norBUP), 417 (norBUP-d3), 468 (BUP), and 472 (BUP-d4).
Specimens and BUP/norBUP Extraction Procedure

Hair samples (at least 50 mg cut close to the scalp at the posterior vertex) were
decontaminated by two CH2Cl2 washes (5 mL, 2 min) then pulverized (Retsch MM2 ball mill,
5-10 min); 50 mg of the resulting powder were then incubated overnight at 56 C in 1 mL of 0.1
N HCl, after addition of 15 ng of deuterated BUP and norBUP (Radian).
After neutralization (using 0.1 N NaOH, 1 mL) of the incubation medium, 2 mL of a saturated,
(NH4)2HPO4 buffer, pH 8.4, and 5 mL of CHCl3/2-propanol/n-heptane (25 : 10 : 65, v/v) were
added. After agitation and centrifugation (3500 g, 10 min), the organic phase was evaporated
(Speed Vac Concentrator, 45 C, 30 min); the dry extract was resuspended in 40 L of the mobile
phase, and after a final centrifugation (10,000 g, 5 min) 25 L of the supernatant were removed,
from which 5 L were injected onto the column.
51
Results and Discussion

The ISP mass spectra of BUP, norBUP are quite simple since they exhibit one unique
peak corresponding to the protonated molecular ion [M + H]+ at m/z 468 and 414, respectively
(fig 1 and 2). This low abundance or absence of fragmentation is a typical character common to
mass spectra generated by the different atmospheric pressure ionization (API) HPLC/MS
interfaces.
Retention times and m/z values for norBUP, norBUP-d3, BUP, and BUP-d4 are listed in
the following table.
Analyte
Retention time (min)
Ions (m/z)
Norbuprenorphine
6.33
414 (d3 : 417)
Buprenorphine
9.17
468 (d4: 472)
Results of the analytical validation are summerized in this second table.

Buprenorphine
Norbuprenorphine
Linearity
0.02 to 5 ng/mg (r 0.993)
0.01 to 3 ng/mg (r 0.998)
Recovery
62 %
54 %
Repetability (0.2 ng/mg)
8.8 %
7,5 %
LOD
4 pg/mg
2 pg/mg
Hair concentrations for BUP and norBUP measured by HPLC/ISP-MS in the hair of 71 chronic
users ranged from 0.01 to 8.16 ng/mg and from not detected to 1.47 ng/mg, respectively.
For Vincent and colleagues (5), BUP ranged from 60 to 361 pg/mg and norBUP from 29 to 785
pg/mg in the hair of 5 subjects.
For Wilkins and colleagues (6), BUP ranged from 4.5 to 156.8 pg/mg and norBUP from 4.8 to
1438.5 pg/mg in the hair of 4 subjects.
52
53
54
55
As an example, fig. 3 shows the SIM chromatogram obtained by extracting a hair sample from a
25-year-old addict. BUP and norBUP concentrations were 3.15 and 0.64 ng/mg, respectively.
BUP and norBUP have been also reported to be detectable in hair from treated subjects by means
of RIA or HPLC with coulometric detection [7]. For this application HPLC/ISP-MS appeared at
least as sensitive (LODs in drug-free, spiked hair powder about 4 pg/mg for BUP, and 2 pg/mg
for norBUP), and much more specific due to the selected-ion detection.
In conclusion, the coupling of HPLC to mass spectrometry via API interfaces, that has
been already presented as a complement of choice to GC/MS for the determination of nonvolatile
and/or thermolabile substances with high sensitivity and specificity, thereby appears as an
interesting alternative in the case of BUP and norBUP dosage in human hair.
Dosage of buprenorphine and norbuprenorphine in human hair
Extraction of hair specimen
1. Decontamination
Two successive washes of the hair strand (100 mg approximately) in 5 ml of methylene chloride
for 2 min.
Pulverization in a ball mill
2. Hydrolysis
50 mg of powdered hair
15 ng of deuterated internal standards (buprenorphine-d4 and norbuprenorphine-d3)
1 ml of HCl 0.1M
Incubation overnight at 56 C
56
3. Extraction
1 ml NaOH 0.1 M
2 ml of NH4Cl buffer pH 9.5
5 ml chloroforme/2-propanol/n-heptane (25:10:65, v/v/v)
Agitation (15 min), centrifugation (15 min at 3000 rpm)
Organic phase removed and evaporated to dryness
Dry extract reconstituted in 35 ml of methanol
Analyses by HPLC-ISP-MS
Injection volume : 2 ml
Column : 5-mm Novapak C18 Waters (150 x 2.0 mm i.d.)
Binary mobile phase of ACN-TMA 50 mg/L/2mM NH4COOH pH 3.0 buffer
ACN : 35 %, 70 % at 9 min
Flow rate : 200 ml/min (50 ml/min infused)
Detection carried out using positive ionization mode

IS + 4500 V; OR +75 V; RNG + 350 V; Qo -10 V
SIM acquisition
Analytical parameters (tab. 1)
Analyte
Ions (m/z)
Norbuprenorphine
6.33
414 (d3 : 417)
Buprenorphine
9.17
468 (d4: 472)
57
Analytical validation
Buprenorphine
Norbuprenorphine
Linearity
0.02 to 5 ng/mg (r 0.993)
0.01 to 3 ng/mg (r 0.998)
Recovery
62 %
54 %
Repetability (0.2 ng/mg)
8.8 %
7,5 %
LOD
4 pg/mg
2 pg/mg
Positive hair specimen (tab. 2)
Analyte
Concentrations
Norbuprenorphine
0.64 ng/mg
Buprenorphine
3.15 ng/mg
2. Chiral separation of methadone and EDDP extracted from human hair
Extraction of hair specimen
1. Decontamination
Two successive washes of the hair strand (100 mg approximately) in 5 ml of methylene chloride
for 2 min.
Pulverization in a ball mill
2. Hydrolysis
50 mg of powdered hair
200 ng of deuterated internal standards (methadone-d3 and EDDP-d3)
0.5 ml b-glucuronidase from bovine liver (20 000 units prepared in water)
1 ml of NH4Cl buffer pH 9.5
Incubation overnight at 40 C
58
3. Extraction
5 ml chloroforme/2-propanol/n-heptane (25:10:65, v/v/v)
Agitation (15 min), centrifugation (15 min at 3000 rpm)
Organic phase removed and evaporated to dryness
Dry extract reconstituted in 35 ml of methanol
Analyses by HPLC-IS-MS
Injection volume : 2 ml
Column : 5-mm CHIRAL-AGP Chromtech (100 x 4.0 mm i.d.)
Binary mobile phase of 2-propanol/2mM NH4COOH pH 5.8 buffer
2-propanol : 8 % for 8 min, 20 % at 9 min, 20 % at 16 min
Flow rate : 500 ml/min (50 ml/min infused)
Detection carried out using positive ionization mode

IS + 4500 V; OR +30 V; RNG + 375 V; Qo -10 V
SIM acquisition
Analytical parameters (tab. 1)
Analyte
Ions (m/z)
(R)-Methadone
14.5
310-265 (d3 : 313-268)
(S)-Methadone
17.0
310-265 (d3 : 313-268)
(R)-EDDP
12.5
278 (d3 : 281)
(S)-EDDP
13.5
278 (d3 : 281)
59
Analytical validation
Methadone
EDDP
Linearity
0.5 to 20 ng/mg (r 0.997)
0.2 to 10 ng/mg (r 0.995)
Recovery
86 %
82 %
Repetability (1 ng/mg)
9.5 %
8.3 %
LOD
0.2 ng/mg
0.1 ng/mg
Positive hair specimen (tab. 2)
Analyte
Concentrations
(R)-Methadone
10.22 ng/mg
(S)-Methadone
5.93 ng/mg
(R)-EDDP
1.38 ng/mg
(S)-EDDP
1.38 ng/mg
References
1. Buprenorphine. In: Baselt RC, Cravey RH, editors. Disposition of toxic drugs and
chemicals in man. Foster City: Chemical Toxicology Institute, 1995; 93-4.
2. Takemori AE, Ikeda M, Portoghese PS. The mu, kappa and delta properties of various
opioid agonists. Eur J Pharmacol 1986; 123: 357-61.
3. Tracqui A, Kintz P., Ludes B. Buprenorphine-related deaths among drug addicts in
France: A report on 20 fatalities. J. Anal. Toxicol. 1998; 22: 430-34.
4. Drug testing in hair. In P. Kintz (Ed.), CRC Press, Boca Raton, 1996.
5. Vincent F, Bessard J, Vacheron J, Mallaret M, Bessard G. Determination of

buprenorphine and norbuprenorphine in urine and hair by gas chromatography-mass
spectrometry. J. Anal. Toxicol. 1999; 23: 270-279.
60
6. Wilkins D, Slawson M, Valdez A, Sison C, Huber A, Ling W, Rollins D. Buprenorphine,

norbuprenorphine and melanin in human hair. Proceedings of the SOFT-TIAFT joint
Meeting, Albuquerque, USA, 1998.
7. Kintz P, Cirimele V, Edel Y, Jamey C, Mangin P. Hair analysis for buprenorphine and its
dealkylated metabolite by RIA and confirmation by LC/ECD. J Forensic Sci 1994; 39:
1497-503.
Interpretation of Quantitative Findings

and Data Evaluation in Hair Analyses
Michael Uhl
Bayerisches Landeskriminalamt, Maillingerstr. 15, 80636 Munich, Germany
Abstract
Hair analysis for illegal and therapeutic drugs provides a long-term information on an
individual`s drug use. Hair retains drugs in a relatively inert matrix, its surveillance window is
limited by the type and length of hair. The first steps of toxicological hair analysis are sample
collection followed by drug isolation from the hair structure. Drug identification and
quantification will supply numerical data, but the question is: how could a simple information
about a concentration of a drug or a metabolite be evaluated? Some biochemical, mechanical and
empirical aspects are assumed to assist the task of interpretation of quantitative findings.
1. Introduction
A pioneer in hair analysis is Werner Baumgartner who decided 1979 to apply immunoassays to
the detection of morphine in hair [1]. His efforts led to great interest and activity by many
scientists entering the unknown and fascinating territory hair analysis. Later, gas chromatography
coupled with mass spectrometry (GC/MS) or coupled with tandem mass spectrometry
(GC/MSMS) were introduced as important analytic tools for hair testing.
An early period of uncritical euphoria was discontinued when simple mechanistic models for
drug incorporation into the hair matrix were replaced by more complex considerations. Modern
concepts about environmental exposure [2], passive contamination, multi-compartment model for
drug incorporation [3], irregular hair growth, cosmetical treatment of hair [4], and ethnic head
hair types [5] were published.
An understanding of these new data lead in the history of hair analysis to the present period of
critical evaluation.
In addition to determination of illegal and therapeutic drugs, hair testing was utilized for different
goals also:
62
Interpretation of Quantitative Findings and Data Evaluation in Hair Analysis
Determination of heavy metals [6], hair as a mirror of the environment
In diagnosis of disease and assessment of nutritional status [6]
For doping control in special cases, in addition to urine testing [7]
Different legal situations in different countries lead to different questions. Orders for hair
analyses for drugs of abuse come from a various kind of mandators: coming from the court, from
the prosecutor`s office, from the police, from the army, from employers, or from administration
authories. Applications of hair analysis for illegal and therapeutic drugs fall into the following
categories:
In criminal court proceedings:
-
Confirmation of chronic abuse to order a medical and physiological examination.
Examination of the reliability of an accused or of a witness.
Testing a supposed victim`s hair sample, intoxication over a long period?
Detoxification control in addicts
General and special workplace testing:

-
Preemployment and workplace testing (e.g. police candidates or officers)
Drug testing in military court-martials
Additional scopes of application are:

-
Drug related fatalities
Examination of the reliability of a father or a mother in child custody and adoption cases,
prenatal and neonatal hair analysis
Testing a hair sample of a former chronic drug user for restoration of his driving licence
These heterogenous orders require individual evaluations, exceeding the simple statement yes
or no to a question like: was this person exposed to drugs or did this person use drugs?
2. Basic aspects
Hair is an unique matrix because no active metabolism and excretion is present to remove drugs
once deposited.
The first steps of toxicological hair analysis are sample collection followed by drug isolation
from the hair structure. Drug identification and quantification will mostly be performed by
63
chromatographic techniques. After determination of a correct concentration there is still a

question:
How could a simple information about a concentration of a drug or a metabolite be evaluated? Is
it responsible that the same numerical data about a certain illegal drug always conducts to the
same conclusion? Is 1,0 ng benzoylecgonine per mg hair determinated in different samples
always due to a similar drug history? It is likely a difference if this concentration of
benzoylecgonine was detected in a proximal 6 cm segment, in a proximal 1 cm segment, in a
proximal 1 cm segment of a dyed strand or in a distal (14-15 cm) 1 cm segment.
For careful interpretation of quantitative findings and data evaluation several important aspects
should be considered.
2.1 Concentration decrease in normally treated hair

Washing the hair several times per week could be seen as normal treatment. Repeated
shampooings have no significant influence on drug content in hair [8]. It was found in a
controlled study [9] that a single dose of deuterated cocaine could be detected for 2-6 months and
it could be detected longer the higher the dose was applied.
With increasing distance from the root the cuticula will be more and more damaged, and the
natural barrier against the loss of entrapped drugs is broken. In experiments with the model
compound methoxyamphetamine a decrease of about 50% in drug concentration was observed
after five months [10].
Polar substances with a tight affinity to the hair matrix are extracted more slowly than unpolar
substances [11]. In segmental analysis the concentration ratio of the polar metabolite to the parent
drug increases with growing distance from the root.
According to personal experience with the detection of the acidic and polar substance 9-carboxyTHC this specific, low concentrated metabolite is washed out and decomposed quickly. In the
great majority of cases the highest concentration of 9-carboxy-THC will rather be in the proximal
than in the distal segment [12].
64
2.2 Effects of cosmetically treated hair

Every strong chemical, physical and mechanical influence could have harmful effects on the
cuticula: perming, straightening, dyeing, bleaching, excessive washing, intensive illumination
with ultraviolet radiation, unusual exposure to the sunlight.
Bleaching, blonding or lightening involve the irreversible destruction of melanin by oxidation, a
partial or even complete degradation of melanin is possible. When strong bleach is used the
physical properties of hair (e.g. a higher porosity) will be altered.
It is important to know that there are different techniques to modify hair color or to mask white
hair. Hair dyeing includes the use of permanent, semipermanent, and temporary dyes. A
permanent dye lasts through any number of washings (as well as permanent perming). A
semipermanent dye is removed after 2-10 washings, and a temporary dye is largely eliminated
after 1 washing [13].
The best method of achieving a permanent hair color is the use of oxidation hair dyes. This
drastic method seems to have a harming influence on the drug molecules entrapped within the
hair. Direct dyes impart a semipermanent or temporary color that lasts a variable time. A
historically and well known representative of this class is henna.
Commercially available bleaching and perming formulas were applied in vitro to the hair strands,
and a decrease in opiate concentration was observed [4]. In authentic hair samples, the drug
levels of the formerly positive hair fibers had been reduced but distinct tendencies couldn`t be
scrutinized.
The concentration of drugs in hair of an addict with bleached and natural colored hair was
investigated [14]. The concentrations found in bleached compared with natural hair were about
33-34 % for cocaine, benzoylecgonine, and methylecgonine, for codeine 25 %, for morphine 11
%, and for 6-MAM 14 %.
2.3 Different specificities of head hair and nonhead hair

The growth rate of head hair could range widely from 0,2 mm/d-1,12 mm/d [15], about 80-95%
of the follicles remain in anagen stage. The smallest part of resting hair is found in the vertex
region. Growth rate could be influenced by therapeutic drugs and is also depending on age, sex,
race and even on seasonal fluctuation [16,17]. In many cases a medium head hair growth rate of 1
65
cm/month was taken. For daily routine we recommend to assume a growth rate of 0,8-1,5
cm/month.
Nonhead hair has a quite similar growth rate (0,9-1,1 cm/month [17]) but a different growth cycle
in comparison to scalp hair. In beard, axillary and pubic hair 40-60% remains in the resting
phase. Therefore, nonhead hair are not suitable for segmental analysis. In case of a positive result
no realistic decision on the period of drug exposure (use, consumption) is possible: maybe just
the last month before sample collection, maybe for a whole year, maybe for a short period, but
nearly 1 year in the past.
2.4 Drug binding to hair

Hair may be thought as an ion exchange resin, it contains three major functionalities whose
characteristics can vary as a function of pH. Acidic groups will be produced by oxidation of
cysteine [2]. Sulfate groups will be expected to be more prevalent in damaged or cosmetically,
chemically treated hair. The isoelectric point of hair decreases when the hair is damaged or when
hair ages. Negatively charged drugs or metabolites like a acidic compound should bind poorly to
the hair matrix.
2.5 Metabolites, pseudometabolites, quality of drugs in street samples

There are three different kind of drugs suitable for the determination in hair samples:
-
-Drugs without a specific metabolite:
-amphetamine
-
-Drugs with nonspecific metabolites:
-Cocaine
For cocaine, the metabolites benzoylecgonine (BE), methylecgonine (ME) and ecgonine are also
products of hydrolysis. It is published in literature [2] that in USA the ratio of BE/cocaine
determined in street samples of cocain-hydrochoride could be 0,86 and even 2,39. On the other
hand, concentrations of these products of hydrolysis (pseudometabolites) found in Germany are
different. According to data of several thousand street samples of cocain-hydrochloride seized in
Bavaria more than 99,9% have an absolute concentration of BE considerably less than 5% [18].
Data of drug concentrations must be transformed to the local situation of each country.
66
-Heroin
In addition to 6-monoacetylmorphine and morphine acetylcodeine could be utilized as additional
marker for the use of illicit heroin [19].
-Methylenedioxymetamphetamine (MDMA), methylenedioxyethylamphetamine (MDEA)
The detection of methylenedioxyamphetamine (MDA) could be due to hydrolysis and metabolisation of MDMA or MDEA or use of MDA
-
Drugs with specific metabolites:
-Hashish and marijuana
11-nor-delta-9-tetrahydrocannabinol-9-carboxylic acid is a metabolite of THC [12,20] and will

not be found in smoke of hashish or marijuana.
2.6 What is a drug positive hair? What indicates a negative result? Relation between drug
dose and hair concentration
Positive test result demonstrates that the individual has used/consumed drugs during the time
period represented by the hair length analyzed. The additional detection of specific metabolites
provides confidence in the evaluation of the findings. The combination of washing techniques,
cut off levels and specific metabolites is recommended.
So far, there are two important statements dealing with the basis issue: what are criteria for
obtaining a positive result:
A consensus initiated by the Society of Hair Testing (SHT) obtained during the 1st European
Meeting on Hair Analysis in Genoa, Italy in June 1996 and a final consensus of the Hair Testing
Working Group (HTWG) documented after its third meeting in Texas, May, 1999.
These two statements will be cited partly.
-
Statement of the Society of Hair Testing concerning the examination of drugs in human
hair [21].
Criteria for obtaining a positive hair test result
1) In order to evaluate the possibility of passive contamination, four criteria are recommended:
.
The identification of metabolites
The use of metabolite-to-parent drug ratios
The assay values of the decontamination washes
Threshold values
67
2) The determination of the following metabolites can be recommended:

Cocaine: Benzoylecgonine and Cocaethylene;
.
.
Heroin: 6-monoacetylmorphine and morphine;
THC: Carboxy-THC;
Amphetamines: none
3) The following metabolite-to-parent drug ratios may be used:

-
Cocaine: benzoylecgonine/cocaine 0.05 (since BE is not always present, hydrolysis

controls should be used)
Heroin: 6-monoacetylmorphine / morphine 1.3 (corrected for hydrolysis)
The positive result of a hair analysis may be used to confirm if a person has used or was exposed
to a drug.
What indicates a negative result?

A negative result does not refute use of or exposure to the drug.
Relation between drug dose and hair concentration

To ascertain the time of exposure and the extent of drug use is difficult and needs further
research. For some drugs, data now exist which indicate that, in very controlled clinical studies, a
positive dose-concentration relationship exists.
The most extensively discussed point was the introduction of threshold values or cut-offs. Different legal
situations in the different countries lead to different questions in court proceedings which require different
cut-offs. The society will therefore set up a committee in which every concerned country will be
represented by one member. This committee will have the task to develop a list of cut-offs with respect to:
-
country
purpose of hair analysis
substance of abuse
GC/MS method
68
A final consensus of the Hair Testing Working Group documented after its third meeting in
Texas, May, 1999.
For confirmation, the the following target analytes and cutoff levels are recommended:
Drug Class
Target analyte
Cutoff
THC
9-Carboxy-THC
0,05 pg/mg hair
Cocaine
cocaine
1,0 ng/mg hair
and
benzoylecgonine (BE)
0,1 ng/mg hair
and
BE/cocaine ratio
0,1
cocaine shall not be reported unless BE
0,1 ng/mg hair
or
cocaine
0,5 ng/mg hair
and
cocaethylene (CE)
0,1 ng/mg hair
or
norcocaine (NC)
0,1 ng/mg hair
cocaine shall not be reported unless CE or NC
Amphetamines methamphetamine
0,3 ng/mg hair
amphetamine
Amphetamine must be present
Opiates
0,1 ng/mg hair
0,3 ng/mg hair

0,05 ng/mg hair to report methamphetamine
morphine
0,2 ng/mg hair
codeine
0,2 ng/mg hair
6-MAM
0,2 ng/mg hair
6-MAM cannot be reported unless another analyte in the opiate class is present at
a concentration above its appropiate cutoff
PCP
PCP
0,3 ng/mg hair
69
3. Case examples
Most of the institutes experienced in hair testing will use GC/MS as analytical instrument for the
detection of drugs in hair. For determination of 9-carboxy-THC it is necessary to enhance the
capabilities of a conventional instrumentation by using GC/MS/MS .
Some of the following authentic case examples should demonstrate that the capability of a good
analytical instrument like a tandem mass spectrometer can be a helpful tool for a rendering of an
expertise.
3.1 Case 1: Testing of pubic hair as adjunctive specimen
A man with short and bleached head hair was seized with a lot of ecstasy (MDMA) tablets and 10
g of cocaine. Only a trace of cocaine (0,35 ng/mg hair), but no benzoylecgonine was detectable in
the head hair. According to the definition of the SHT and HTWG this sample could not be
valuated as positive.
The pubic hair of the same person (fig.1) was tested positive for MDMA (1,3 ng/mg hair), its
potential metabolite MDA (0,35 ng/mg hair), cocaine (4,1 ng/mg hair), benzoylecgonine (0,3
ng/mg hair) and amphetamine (1,1 ng/mg hair).
No statement should be made about the overall dosage and and time of drug use/exposure.
3.2 Case 2: unusual ratio of cocaine/benzoylecgonine
In cases where the time of chronical abuse lies more than several months in the past, the
evaluation of the results will be more and more complicated. According to the court it was to be
proven that a defendant with 43 cm long hair had only abused cocaine during a time period
earlier than 15 month in the past. The sample, a 43 cm long strand of hair was cut into four
segments of equal length.
Two proximal segments were tested negative for cocaine and metabolites. As presented in Fig.2
two distal segments had a concentration of 0,2 ng/mg cocaine and 1,5 and 4,3 ng/mg
benzoylecgonine. A cocaine/benzoylecgonine ratio of that kind was found so far in the hair of
ancient peruvian coca leaf chewers [22].
70
3.3 Case 3: detection of acetylcodeine as additional marker for heroin abuse
A person taken into custody for possession of 100 grams of heroin claimed severe addiction to
this opiate in order to attenuate penalty. Chronic heroin abuse could be confirmed by the result of
a hair analysis (Fig.3). Concentrations of the parent drug, metabolites and of an impurity due to
manufacturing procedure [7] were high and might be due to heroin addiction: heroin: 8,2 ng/mg;
6-mono-acetylmorphine: 36 ng/mg; morphine: 1,4 ng/mg and acetylcodeine 3,6 ng/mg hair.
3.4 Case 4: detection of 9-carboxy-THC
Two examples should demonstrate capabilities and limits of GC/MS/MS. The scalp hair of a
supposed hashish dealer was to be tested for cannabinoids. According to the dealer`s self
admitted use he characterizes himself as a moderate cannabis smoker. The sample was cut into a
3 cm root and a 3 cm tip segment: the concentration of 11-nor-delta-9-THC-9-carboxylic-acid in
the root segment was very close to the LOQ of 0.16 pg (fig.4)
However, the result for this metabolite was not positive in the tip segment. (fig.5)
According to our available data we don`t realize a significant correlation between the
concentration of THC and 9-carboxy-THC. It is a preliminary observation that in segmential hair
analysis the highest concentration of 9-carboxy-THC is detected in the root segment, a lower
concentration in the middle segment (if available), the minimum concentration in the tip segment.
3.5 Case 5: hair testing of a sample coming from cosmetically untreated hair
A man already imprisoned for 11 months being a trafficker of cannabis claimed that he consumed
a lot of marijuana in a period more than one year ago. The hair should be tested for 9-carboxyTHC. He expressed his sympathy for the jamaican religious community rastafari by presenting a
special hair-dress. Because he never combed his hair and used shampoo the hair matted
completely. A distal segment (12-16 cm) of the dreadlock should correspond with that past period
1 year before sample collection. However, due to the helical appearence the actual lenght of
every single hair within this lock is longer. The right choice of segmental length was not possible.
The result for 9-carboxy-THC in the distal segment was negative. So, the rastafari`s affirmation
couldn`t be refuted.
71
3.6 Case 6: sample of a robber with repeatedly dyed hair.

A female haircutter and pretended addict to opiates tried to improve her financial situation by
committing a bank robbery. She affirmed during her interrogation that she consumed in this past
period of this crime about 0,3 g dihydrocodeine and 2-3 g heroin in normal street quality (515 %) daily. Furthermore, she claimed that she bleached and dyed her hair every week. Her hair
sample was negative for 6-monoacetylmorphine, morphine, heroin, and acetylcodeine. Only
traces of dihydrocodeine (0.9 ng/mg hair) were detected. It is hard to imagine that there was only
a selective oxidation of heroin, its metabolites, and of acetylcodeine.Therefore, the extent of her
heroin addiction seemed to be exaggerated.
72
Fig 1: Testing of pubic hair as adjunctive specimen: MDMA (1,3 ng/mg hair), MDA (0,35 ng/mg
hair), cocaine (4,1 ng/mg hair), benzoylecgonine (0,3 ng/mg hair) and amphetamine (1,1 ng/mg
hair).
73
Fig 2: Unusual cocaine/benzoylecgonine ratio determinated in a distal segment: cocaine

0,2 ng/mg hair, benzoylecgonine 4,3 ng/mg hair.
74
Fig 3: Confirmation of a claimed and confirmed addiction to heroin: 6-mono-acetylmorphine:

36 ng/mg; morphine: 1,4 ng/mg, heroin: 8,2 ng/mg and acetylcodeine 3,6 ng/mg hair.
Fig. 4: Detection of 9-carboxy-THC in low concentration 0,16 pg/mg hair
75
76
Fig. 5: Futile detection of 9-carboxy-THC: an attempt exceeding the capabilities
4. Conclusion
Each analytical toxicologist will appreciate the solid bases of reliable data. He will
probably prefer two, three or even more data in order to evaluate the drug history of a
certain person.
Generally, a particular information from the mandator about the individual subject can be
important for a rendering of an expertise. Aspects that could be considered are for example:
When is the person taken into custody? What are the special allegations? Is there a self reported
use?
77
The intended application of the data must be carefully considered. Mostly, a high concentration
of metabolites will mostly harmonize with use of a great amount of drugs. As seen in chapter 2 a
low concentration could have different reasons.
Possible criteria for obtaining a positive hair result are recommended by the SHT and
HTWG.
The whole pattern of metabolites should be evaluated in order to diagnose drug consumption. It is
essential that all these analytical results are in good harmony.
For rendering an expertise based on hair analysis an analytic expert with diligence, experience
and the capabilities of a excellent analytical instrument is a reliable combination. Many
unanswered questions remain. The qualitative results are valid, but the interpretation of these
results and the quantitative data is still debatable.
References
[1] A.M. Baumgartner, P.F.Jones, W.A. Baumgartner, and C.T. Black, Radioimmunassays of hair
for determination of opiate-abuse histories. J. Nucl. Med. 20 (1979) 748-752.
[2] D.A. Kidwell and D.L. Blank, Environmental exposure - the stumbling block of hair testing.
In P. Kintz (ed.), Drug Testing in Hair, CRC Press, Boca Raton New York London Paris, 1996,
pp. 17-68.
[3] G.L. Henderson, Mechanism of drug incorporation into hair, Forensic Sci. Int. 63 (1993) 1929.
[4] L. Ptsch and G. Skopp, Stability of opiates in hair fibers after exposure to cosmetic
treatment, Forensic Sci. Int. 81 (1996) 95-102.
[5] E.J. Cone and R.E. Joseph Jr.,The potential for bias in hair testing for drugs of abuse. In P.
Kintz (ed.), Drug Testing in Hair, CRC Press, Boca Raton New York London Paris, 1996, pp. 6993.
[6] A. Chatt and S.A.Katz., Hair Analysis: Applications in the biomedical and environmental
sciences, VCH Publishers, New York, 1988 pp 14-16 and pp. 77-81
[7] Consensus of the Society of Hair Testing on hair testing for doping agents, Forensic Sci. Int.
107 (2000) 3.
[8] W.A. Baumgartner, V.A. Hill, and W.H. Blahd, Hair analysis for drugs of abuse, J. Forensic
Sci 34 (1989) 1433-1453.
78
[9] G.L. Henderson, M.R. Harkey, C. Zhou, R.T. Jones and, P. Jacob III, Incorporation of
isotopically labeled cocaine and metabolites into hair: 1. Dose response relationships, J. Anal.
Toxicol. 20 (1996), 1.
[10] Y. Nakahara, M. Shimamine, and K. Takahashi: Hair Analysis for drugs of abuse III.
Movement and stability of methoxyamphetamine (as a model compound of methamphetamine)
along the hair shaft with hair growth, J. Anal. Toxicol. 16 (1992) 253-257.
[11] F. Pragst, M. Rothe, J. Hunger, and S. Thor, Structural and concentration effects on the
deposition of tricyclic antidepressants in human hair, Forensic Sci Int. 84 (1997) 225-235.
[12] M.Uhl, Tandem mass spectrometry: a helpful tool in hair analysis for the forensic expert,
Forensic Sci. Int. 107 (2000) 169-179.
[13] T. Clausen, Hair coloring preparations. Ullmann`s Encyclopedia of Industrial Chemistry; 5
th edition Vol. A12, VCH Verlagsgesellschaft, Weinheim (1989) 583-588.
[14] V. Cirimele, P. Kintz, and P. Mangin, Drug concentration in human hair after bleaching, J.
Anal. Toxicol 19 (1995) 331-332.
[15] P. Mangin, Drug analysis in nonhead hair. In P. Kintz (ed.), Drug Testing in Hair, CRC
Press, Boca Raton New York London Paris, 1996, pp. 279-287.
[16] F. Pragst, M.Rothe, K. Spiegel, and F. Sporkert, Illegal and Therapeutic Drug
Concentrations in Hair Segments A Timetable of Drug Exposure ? Forensic Sci Rev 10 (1998)
81-111.
[17] M.R. Harkey, Anatomy and physiology of hair, Forensic Sci. Int. 63 (1993) 9-18.
[18] K. Stein, personal communication, May 2000.
[19] P. Kintz, C. Jamey, V. Cirimele, R. Brenneisen, and B. Ludes, Evaluation of acetylcodeine
as a specific marker of illicit heroin in human hair. J. Anal. Toxicol. 22 (1998) 425-429.
[20] T. Cairns, D. Kippenberger, H. Scholtz and W. Baumgartner, Determination of carboxyTHC in hair by mass spectrometry/mass spectrometry. In R.A. de Zeeuw, I. Al Hosani, S. Al
Munthiri and A. Maqbool (eds.), Proceedings of the 1995 International Conference and
Workshop for Hair Analysis in Forensic Toxicology, Abu Dhabi, Nov. 19-23, 1995, pp 185-193.
[21] Society of Hair Testing, Forensic Sci. Int. 84 (1997) 3-6.
[22] A.C. Springfield, L. W. Cartnell, A.C. Aufderheide, J. Buikstra, and J. Ho, Cocaine and
metabolites in the hair of ancient Peruvian coca leaf chewers, Forensic Sci. Int. 63 (1993) 269275.
Criteria that Can Affect the Detection of Doping Agents in Hair

Pascal Kintz
Institut de Mdecine Lgale, 11 rue Humann
67000 Strasbourg, France
1. Introduction
During the last two years in France, several forensic cases involving doping agents were reported
in the popular press, including the nandrolone use by of Djamel Bouras (a judo olympic winner),
use by the cyclists participating in the Tour de France, unapproved use in animals slated for fodd
consumption, and several trafficking cases in humans. To document the pattern of drug use, the
judges in charge of these cases requested toxicological analyses based on blood, urine, and hair
tests (1-3). Although hair is not yet a valid specimen for the International Olympic Committee
(IOC), it is accepted in most courts of justice. During the same period, some conflicting results
were observed, all involving athletes that tested positive in urine in accredited IOC laboratories
and negative in hair in forensic certified laboratories.
A lot of experience has been aquired in the detection of opiates and cocaine in hair. In contrast,
there is a serious lack of suitable references to interpret the analytical findings for doping agents.
In hair, doping agents concentrations, such as anabolic steroids, corticosteroids, or 2-agonists,
are in the range of picograms per milligram, whereas cocaine, amphetamines or opiates are
generally found in the range of several nanograms per milligram (4-7). Therefore, it was the
feeling of the Society of Hair Testing to obtain a consensus on hair testing for doping agents (8).
It is clear that there is a great deal of research to be performed before the scientific questions and
curiosity surrounding hair drug testing is satisfied. Some of this is due to a lack of consensus
among the active investigators on how to interpret the results on an analysis of hair. Among the
unanswered questions, five are of critical importance: 1. What is the minimal amount of drug
detectable in hair after administration? 2. What is the relationship between the amount of the drug
used and the concentration of the drug or its metabolites in hair? 3. What is the influence of hair
color? 4. Is there any racial bias in hair testing? 5. What is the influence of cosmetic treatments?
80
Criteria That Can Effect the Detection of Doping Agents in Hair
The present report documents scientific findings on these questions, with particular attention to
the applications of hair in doping control.
2. What is the minimal amount of drug detectable in hair ?
When using hair in a suspected doping case, particularly when urine of the athlete was positive
and hair negative (several cases were reported during the past two years), the question of
importance is to know whether the analytical procedure was sensitive enough to identify traces of
drugs. It has been always accepted in the forensic community that a negative hair result cannot
exclude the administration of the detected drug or one of its precursors (such as norandrostendiol
or norandrostendione for the metabolites of nandrolone) and should not overrule a positive urine
result. Nevertheless, the negative hair findings lends enough ambigity to the positive urine result,
coupled with the sporting consequences for the athlete, that substancial justice refereeing occurs.
For drugs of abuse like cocaine and opiates, several controlled studies were conducted during the
past years to evaluate the threshold dose. According to Henderson, et al. (9), the threshold dose
for detecting cocaine in hair appears to be approximately 25-35 mg cocaine, administered
intravenously. Once incorporated into hair, a single dose of cocaine can be detected for 2 to 6
months. Codeine was detected in hair for 8 weeks after a single oral dose of 60 mg (10). In
contast, with anabolic steroids we were not able to identify nandrolone nor nandrolone
undecanoate in the hair of a 37-year old man, receiving a single intramuscular injection of 50 mg
nandrolone undecanoate, although his urine remained positive for norandrosterone and
noretiocholanolone, the nandrolone metabolites, for at least 8 months. Hair was tested 2 and 6
months after administration, and the limit of detection for nandrolone in hair was 1 pg/mg. The
same observations were recently made by Segura, et al. (11) who did not detect nandrolone,
nandrolone decanoate nor testosterone esters (enanthate, propionate, undecanoate) after a single
dose administration. However, in the same study, hair analysis revealed an increase of more than
2 fold in testosterone concentration in hair from subjects treated with the enanthate and
propionate esters as compared with the pre-dose samples.
81
The physiochemical properties of drugs on the incorporation rates into hair are of paramount
importance. The influence on drug incorporation of melanin affinity, lipophilicity, and membrane
permeability was evaluated by Nakahara, et al. (12). Despite their high lipophilicity, steroids have
quite low incorporation rate into hair, suggesting that the higher incorporation rates of basic drugs
(cocaine, amphetamines ...) than neutral (steroids, benzodiazepines, cannabinoids ) or acidic
ones are strongly related to the membrane permeability of the drug based on the pH gradient
between blood and the acidic hair matrix. This can perhaps explain the low steroids
concentrations that are reported in hair, and therefore, it is not surprising that a single exposure
cannot be demonstrated. When compared, clenbuterol (with a primary amine function) and
salbutamol, two 2-agonists, showed 2 different pattern of accumulation in guinea-pigs hair,
largely in favour of clenbuterol, even with a 10 times lower dose (13).
Therefore, until laboratories will have sensitive enough methodologies to detect a single use of
steroids, care should be taken to compare urine and hair findings. For anabolic steroids to have an
appreciable performance enhancing effect, they must be chronically administered, in contrast to
the immediate stimulant properties of cocaine or amphetamine.
Repeated use would favor
identification by hair analysis.
The knowledge of the minimal quantity that can be identified in hair is useful when intrepretating
the analytical results. Although the following example was reported in newspapers, it is
indicative of the differences between sport rules and forensic cases and the place that hair has in
such situations. During the summer 1999, the Cuban high jumper Javier Sotomayor tested
positive in urine for cocaine. The athlete claimed that he only drunk one cup of tea coming from
Peru (Reuters agency, Wednesday August 4, 1999). For the International Athletic Federation and
IOC, the presence of cocaine or metabolites in urine is enough to be considered as doping,
irrespective of the manner of use. Our laboratory was involved in a similar case. The judge asked
if it is possible to find cocaine metabolites in urine after tea consumption. It was necessary to
establish the urinary excretion of cocaine and metabolites after oral administration of tea (Mate
de Coca, Zurit, ASA Alimentos) containing 3.9 mg of cocaine. The hair of the same subject was
tested 1 month later (first 1-cm section) and was negative (LOD: 10 pg/mg), suggesting that the
minimal detectable dose in hair is higher than 4 mg of cocaine. A hair test of Sotomayor would
aid in confirming his claim (no cocaine abuse and negative hair test) or, at the opposite (positive
82
hair test) would demonstrate cocaine abuse, as a single "line" of street cocaine (e.g., 50-100 mg)
would be detectable.
3. What is the relationship between the amount of the drug used and the concentration of
the drug or its metabolites in hair ?
A common interpretative question posed when positive drug test results are reported is if it is
possible to put a quantitative finding into context. In other words, "Is this a lot of drug?" or " Did
this person use a lot of drug or just a little ? " A standard answer to either of these inquiries is that
a positive hair test result can be interpreted as meaning that the donor has chronically or
repetitively used the drug identified in the hair; however, chronic or repetitive are not defined in
the same way by all individuals.
To document hair findings, in term of amount and frequency of consumption, essentially two
approaches have been developed. One (14) is to compare the measured concentration with a level
(low, medium or high) of consumption obtained from the habitual use declared by a large number
of addicts. For example, according to Ppin and Gaillard (14), a cocaine concentration in hair
lower than 4 ng/mg indicates a low cocaine consumption. The second method, used in our
laboratory, is to compare the positive result with the distribution of concentrations for the drug in
positive hair that are regularly tested. However, this can only be done for drugs that are tested in
a routine basis (more than 100 samples per year).
In case of doping agents, historical data are not available, and therefore its not possible to
estimate the pattern of drug use. At this time, only data for testosterone (5, 15-17) and
dehydroepiandrosterone, DHEA (18), two endogenous substances for which physiological
concentration ranges in human hair are available in the literature. These concentrations are
generally in the range 1-10 pg/mg, including hair specimens obtained from high level (Olympic)
athletes. A higher testosterone concentration can be indicative of use, and further investigations
are necessary to identify in hair the esters of testosterone (2, 4, 11, 17).
83
There is a need to put a quantitative interpretation into context, particularly for forensic
applications. In a case of suspected anabolic trafficking by bodybuilders (3), the judge asked us if
the seized drugs (tablets and ampoules) supported personal use or if it was for resale. Hair
analysis was used to demonstrate abuse by the bodybuilders, without the possibility of indicating
the frequency of exposure and the dosage. Another example of the medico-legal aspects of
doping (19, 20) is cardiovascular complications, that can be fatal, or psychiatric problems
associated with enhanced aggressivity. In such cases, toxicological investigations are necessary to
document the side-effects of steroids, and the pattern of exposure is often requested. The
quantitative findings are of less importance in comparison with qualitative identification in
doping control. However, it can be useful to estimate the habit of consumption.
Some papers reported for drugs of abuse or pharmaceuticals, that there is a significant doseconcentration relationship between dosage and concentration in hair, while some other failed
(21). The lack of correlation can be explained by the following factors : variability of hair-growth
cycle, influence of cosmetic treatments and hygienic practices, uncertainty of dosages ingested by
abusers, typical underestimation of self-reported doses, unknown purity of compounds, and
considerable variation in uptake of drug from blood to hair, rate of sweating and amount of
apocrine and sebaceous gland secretions between individuals. Until controlled studies are done
for doping agents (is this possible?), it seems difficult to use hair analysis to calculate the dose,
time or duration of drug use.
84
4. What is the influence of hair color?
Melanins are responsible for the color of hair, as determined by the quantity of pheomelanin and
eumelanin present in hair. Black and brown hair contain more eumelanin than red and blond hair.
Current evidence suggests that melanin is the principal component of binding of drugs. Several
studies performed with animals and humans reported that organic compounds are excreted
preferentially in dark hair and suggested that excretion of drug is closely linked to the presence of
melanin (22).
In 1996, Hld et al (23) demonstrated that stanozolol is incorporated preferentially into rat
pigmented hair. This was also noticed for nandrolone in guinea pigs (13) or clenbuterol in calves
(24). After administration of the same daily dose of 10 g clenbuterol during 25 days to 9
volunteers, Gleixner et al (25) observed that the clenbuterol residues were lower in light hair
(blond, grey) than in dark hair (black, brown).
More recently, Kronstrand et al (26), after single oral administration of 100 mg of codeine to 9
subjects, concluded that the incorporation of codeine into hair is affected by its melanin content
and that the relationship is exponential.
All these studies suggest that the hair color or melanin content may be the major determinant of
drug binding and, consequently, may result in color bias in hair testing.
5. Is there any racial bias in hair testing ?
Several researchers have demonstrated that different hair types incorporate differing amounts of
drugs when exposed under identical conditions (27-29). These studies have suggested that coarse,
dark hair may incorporate more drug than fine brown or blond hair. Henderson et al (27) showed
that non-Caucasian subjects (n=9) incorporate 2.9 (in average) times more deuterated cocaine
into their hair as did the Caucasian subjects (n=6), in equivalent experimental conditions. These
findings were consistent with those of Cone et al (30), who found that cocaine concentrations in
both head and arm hair of African-American subjects were significantly higher (p < 0.05) than
those found in Caucasian subjects.
85
Bias in drug testing can be defined as an increased likelihood of detecting one group of
individuals over another when both have comparable use. There is a debate (31, 32) among the
forensic community about racial bias as a problem may arise in the use of hair analysis
procedures that identify more African-American drug users than Caucasian drug users. Thus, for
a low-use population, such as athletes, there is a potential risk to identify more frequently a
population than another. The introduction of positive cut-offs concentrations will not solve this
problem. Kidwell (31) has recently proposed that the uptake and retention of drugs into hair is
culturally (cosmetic hair treatments, hair care habits, removal by personal hygiene) rather than
genetically (porosity of hair) influenced. In consequence, the author has suggested methods that
use either the incorporation of dyes or of drug homologs to reduce bias in hair testing. This
means that a correction for potential bias can be achieved in a special requested situation.
In conclusion, for the interpretation of results, the higher accumulation of several substances in
black hair as compared with blond hair will need to be addressed in order not to discriminate
specific populations due to their hair color, resulting in inequity during doping control (33).
6. What is the influence of cosmetic treatments?
Interpreting the results from hair analysis for drugs can be difficult, because this is a relatively
new methodology and there are many issues that still remain to be answered. One important issue
is the stability of drugs in human hair. Many factors are likely to affect the concentration of drugs
in hair and thus complicate the interpretation of results from the analyses. These factors include
cosmetic treatments of hair (34, 35). In all the studies, the authors observed that drug
concentrations declined dramatically after cosmetic treatment (bleaching or permanent waving)
and UV exposure, concluding that both the chemical composition of different cosmetic agents
and the methods applied in hair treatment affect drug stability. When comparing treated hair by
bleaching or dyeing before sample collection with the corresponding non-treated hair strand, the
mean differences in drug concentrations were in the range 30-80 % for opiates, cocaine, cannabis
and nicotine. These differences depended not only on the type of cosmetic treatment, as
bleaching produced higher decreases than dyeing, but also on the degree of hair damage, i.e. the
more damaged the hair, the larger the differences in the concentration levels of drugs.
86
In case of cosmetic treatment, the hair analysis results appear to be biased rather towards false
negatives than towards false positives. A recent study performed by Cirimele, et al. (36)
demonstrated that the daily use of a shampoo containing cannabinoids, including THC, would
not lead to a positive hair test.
During the past years, numerous athletes, particularly cyclists, have colored their hair, but it was
rather considered as an aesthetic approach rather than an evasive procedure. However, these
treatments have consequences on the hair test results. Gaillard and coworkers (2) only found
positives for anabolic agents (nandrolone, testosterone undecanoate) in two non-bleached hair
samples. Our laboratory, missed identifing testosterone (LOD 0.5 pg/mg) in several specimens
bleached with hydrogen peroxide, and therefore we recommend the collection of pubic hair in
case of bleached or colored head hair. However, care should be taken when sampling hair from
other anatomic regions, as the concentrations can be highly variable according to the specimens.
We were very surprised when comparing the physiological concentrations of DHEA and
testosterone in hair collected from the head, pubis, and axillae. An unusual high amount of
DHEA was sequestered in axillary hair. Generally, the overall pattern of drug is similar in head
and non-head hair, and with some exceptions, the highest drug and metabolite concentrations are
observed in pubic hair, while the lowest are found in axillary hair (37).
7. Perspectives and conclusion
In case of doping control, drugs are screened in urine specimens according to validated standard
operating procedures in accredited laboratories. As forensic laboratories can be involved in
testomony dealing with doping agents, the idea of using hair for doping control has emerged as
hair analysis has been accepted in court in other cases. Courts can request additional informations
on the pattern of use of doping substances, such as during the 1998 cycling Tour de France where
blood, urine, and hair were simultaneously collected. Hair can both confirm repetitive abuse and
identify the exact nature of the parent compound (e.g., nandrolone, norandrostendiol or
norandrostendione, in case of positive urine for norandrosterone). Moreover, long-term use (over
several months) of restricted compounds (only authorized under specific conditions and for a
short period), such as salbutamol or corticoids, can be documented through hair analysis. The
87
determination of testosterone esters in hair should allow a definitive unambiguous confirmation

of the administration of exogenous testosterone.
However, some issues have to be discussed before considering hair as a valid specimen by the
I.O.C. and the International Sport Federations. The relationship between urine and hair results is
not yet established and negative hair result does not mean "no doping". The potential ethnic
discrimination must be evaluated to avoid inequality during doping control.
In contrast with the problems associated with cosmetic treatments or the absence of specimen
(bald or fully shaved subject), external contamination does not constitute a major trouble when
testing for doping agents.
References
1. P. Kintz, Hair testing and doping control in sport, Toxicol. Letters 102-103 (1998) 109-113.
2. Y. Gaillard, F. Vayssette, G. Ppin, Compared interest between hair analysis and urinalysis in
doping control: results for amphetamines, corticosteroids and anabolic steroids in racing
cyclists, 2nd European Meeting on Hair Analysis, Martigny, Switzerland, June 14-16, 1999.
3. P. Kintz, V. Cirimele, H. Sachs, T. Jeanneau, B. Ludes, Testing for anabolic steroids in hair
from two bodybuilders, Forensic Sci. Int. 101 (1999) 209-216.
4. D. Thieme, J. Grosse, H. Sachs, R.K. Mueller, Detection of several anabolic steroids of abuse
in human hair, Proceedings of the 16th Cologne Workshop on Dope Analysis, Sport und
Buch, Kln, Germany, 1999, pp 9-29.
5. M.J. Wheeler, Y.-B. Zhong, A.T. Kicman, S.B. Coutts, The measurement of testosterone in
hair, J. Endocrinol. 159 (1998) R5-R8.
6. M. Machnick, H. Geyer, S. Horning, A. Breidbach, P. Delahaut, W. Schnzer, Long-term
detection of clenbuterol in human scalp hair by gas chromatography-high resolution mass
spectrometry, J. Chromatogr. B 723 (1999) 147-155.
7. X.-S. Deng, A. Kurosu, D.J. Pounder, Detection of anabolic steroids in head hair, J. Forensic
Sci. 44 (1999) 343-346.
8. Society of Hair Testing, Consensus on hair testing for doping agents, Forensic Sci. Int. (1999)
in press.
9. G.L. Henderson, M.R. Harkey, C. Zhou, R.T. Jones, P. Jacob III, Incorporation of
isotopically labeled cocaine and metabolites into human hair: 1. Dose-response relationships,
J. Anal. Toxicol. 20 (1996) 1-12.
10. D.E. Rollins, D.G. Wilkins, G.G. Krueger, Codeine disposition in human hair after single and
multiple doses, Eur. J. Clin. Pharmacol. 50 (1996) 391-397.
11. J. Segura, S. Pichini, S.H. Peng, X. de la Torre, Hair analysis and detection of single dose
administration of androgenic steroids esters, 2nd European Meeting on Hair Analysis,
Martigny, Switzerland, June 14-16, 1999.
88
12. Y. Nakahara, K. Takahashi, R. Kikura, Hair analysis for drugs of abuse. X. Effects of
physiochemical properties of drugs on the incorporation rates into hair, Biol. Pharm. Bull 18
(1995) 1223-1227.
13. J. Segura, Possibilities of ELISA methodologies for hair analysis. In : R.A. de Zeeuw, I. Al
Hosani, S. Al Munthiri, A. Maqbool (eds), Hair analysis in forensic toxicology, Proceedings
of the 1995 International Conference and Workshop. Abu Dhabi, 1995, pp 351-369.
14. G. Ppin, Y. Gaillard, Concordance between self-reported drug use and findings in hair about
cocaine and heroin, Forensic Sci. Int. 84 (1997) 37-41.
15. C. Scherer, U. Wachter, S.A. Wudy, Determination of testosterone in human hair by gas
chromatography selected ion monitoring mass spectrometry, Analyst 123 (1998) 2661-2663.
16. A. Gleixner, H.H.D. Meyer, Methods to detect anabolics in hair : use for food hygiene and
doping control, Int. Lab. July issue (1998) 20-23.
17. P. Kintz, V. Cirimele, T. Jeanneau, B. Ludes, Identification of testosterone and testosterone
esters in human hair, J. Anal. Toxicol. (1999) in press.
18. P. Kintz, V. Cirimele, B. Ludes, Physiological concentrations of DHEA in human hair, J.
Anal. Toxicol. (1999) in press.
19. R. Hausmann, S. Hammer, P. Betz, Performance enhancing drugs (doping agents) and sudden
death a case report and review of the literature, Int. J. Leg. Med. 111 (1998) 261-264.
20. B. Madea, W. Greener, F. Mussoff, R. Dettmeyer, Medico-legal aspects of doping, J. Clin.
Forensic Med. 5 (1998) 1-7.
21. P. Kintz, Clinical applications of hair analysis, in : Drug Testing in Hair, CRC Press, Boca
Raton, Fl, (1996) pp 267-277.
22. E.J. Cone, Mechanisms of drug incorporation into hair, Ther. Drug Monit. 18 (1996) 438443.
23. K.M. Hld, D.G. Wilkins, D.J. Crouch, D.E. Rollins, R.A. Maes, Detection of stanozolol in
hair by negative ion chemical ionization mass spectrometry, J. Anal. Toxicol. 20 (1996) 345349.
24. Y. Gaillard, A. Balland, F. Doucet, G. Ppin, Detection of illegal clenbuterol use in calves
using hair analysis. Application in meat quality control, J. Chromatogr. B 703 (1997) 85-95.
25. A. Gleixner, H. Sauerwein, H.H.D. Meyer, Detection of the anabolic 2-adrenoceptor agonist
clenbuterol in human scalp hair by HPLC/EIA, Clin. Chem. 42 (1996) 1869-1871.
26. R. Kronstrand, S. Frsberg, B. Kagedal, J. Ahlner, G. Larson, Relationship between melanin
and codeine concentrations in hair after oral administration, Proceedings of the annual
meeting of the American Academy of Forensic Sciences, Orlando, Fl, February 15-20, 1999,
abstract K12.
27. G.L. Henderson, M.R. Harkey, C. Zhou, R.T. Jones, P. Jacob III, Incorporation of
isotopically labeled cocaine and metabolites into human hair: race as a factor, J. Anal.
Toxicol. 22 (1998) 156-165.
28. E.J. Cone, R.E. Joseph, The potential for bias in hair testing for drugs of abuse, in : Drug
Testing in Hair, CRC Press, Boca Raton, Fl, (1996) pp 69-93.
29. T. Uematsu, R. Sato, Human scalp hair as evidence of individual dosage history of
haloperidol : longer-term follow-up study, Ther. Drug Monit. 12 (1990) 582-583
30. E.J. Cone, W.D. Darwin, W.L. Wang, The occurrence of cocaine, heroin and metabolites in
hair of drug abusers, Forensic Sci. Int. 63 (1993) 55-68.
31. D.A. Kidwell, Is hair testing culturally biased and why is this a concern in the US ?, 2nd
European Meeting on Hair Analysis, Martigny, Switzerland, June 14-16, 1999.
89
32. T. Mieczkowski, Further examination of race and hair color as biasing factors in hair analysis
for cocaine, 2nd European Meeting on Hair Analysis, Martigny, Switzerland, June 14-16,
1999.
33. J. Segura, http://nodoping.org
34. C. Jurado, P. Kintz, M. Menendez, M. Repetto, Influence of cosmetic treatment of hair on
drug testing, Int. J. Leg. Med. 110 (1997) 159-163.
35. L. Ptsch, G. Skopp, Stability of opiates in hair fibers after exposure to cosmetic treatment,
Forensic Sci. Int. 81 (1996) 95-102.
36. V. Cirimele, P. Kintz, C. Jamey, B. Ludes, Are cannabinoids detected in hair after washing
with Cannabio shampoo?, J. Anal. Toxicol. (1999) in press.
37. P. Mangin, Drug analyses in nonhead hair, in : Drug Testing in Hair, CRC Press, Boca Raton,
Fl, (1996) pp 279-287.
90
Analysis of Hair Samples for 2-Agonists

Kurt Einhellig, Michael Uhl *
(Bayerisches Landeskriminalamt, Maillingerstr. 15, 80636 Munich, Germany)
Introduction
The International Olympic Committee`s Medical Code outlines classes of banned substances
that are subject to restrictions and provides a list for each category. 2-agonists are listed in the
category anabolic agents.
Associated with the name of the former world champion in athletics Kathrin Krabbe clenbuterol
became wellknown in the public. Despite her urine sample was positive for clenbuterol she was
not banned for a doping offence but she was convicted for a violation against the German drug
law.
The potent 2-agonist clenbuterol (4-amino- -[(tert-butylamino)methyl]-3,5-dichlorobenzyl
alcohol; or 1-(4-amino-3,5-dichlorophenyl)-2-(tert-butylamino)-1-ethanol,) is used as a
sympathomimetic bronchodilator in the management of asthma and obstructive lung disease.
Proprietary names are Broncodil, Ventolase, Spiropent.
Moreover, when used at doses being several times higher than the therapeutic application,
clenbuterol has anabolic effects stimulating protein deposition in striated muscle (20% increase).
The 2-agonist increases simultaneously the energy expenditure hence reducing muscle glycogen
and body fat deposition (20% decrease) [1,2]. Clenbuterol induces a true hypertrophy of muscle
(type II fibers) and an increase in the 1-glycolytic capacity of a muscle in a whole, giving rise to
an increase in force and a reduction in relaxation time [3].
Administration of this 2-agonist in veterinary therapy was allowed in Germany until june 1997.
The brominated analog of clenbuterol is brombuterol, being an illegal substance in Germany.
92
Detection of clenbuterol in hair
2.1 Detection of clenbuterol in human hair samples [4].

Hair samples from 67 volunteers were analyzed for clenbuterol by enzyme immunoassay (EIA)
and by HPLC. Clenbuterol residues could be detected in hair from 9 volunteers who took a
known therapeutic dosage. The concentrations varied from 25-160 pg/mg hair. The accumulation
in dark hair (150 pg/mg hair) was markedly higher than in light hair (30 pg/mg hair),
demonstrating the importance of hair pigmentation of the binding of clenbuterol to the hair
matrix.
2.2 Detection of clenbuterol in hair samples of calves [5,6]

9 calves received oral doses of clenbuterol. 6.7 mg of the 2-agonist was administered daily for a
period of 15 days, the overall dose being 100 mg [5]. The hair samples were pulverized, after
clean-up by solid phase extraction the detection was performed by GC/MS. The concentrations
were in the range of 20-4372 pg/mg hair.
In another study [6] 3 calves were fed with clenbuterol in doses of 5g/kg body weight twice
daily during 3 weeks. The hair samples were analyzed using GC/MS, the concentrations ranged
between 135-1133 pg/mg hair.
2.3 Detection of clenbuterol in hair samples of guinea pigs [7]

Guinea pigs received a dose of 120 g/kg clenbuterol intraperitoneally for 15 days. Analysis of
the collected hair samples were performed by GC/MS.
Detection of brombuterol in hair
An authentic case example will demonstrate the detection of brombuterol in hair [8].
The owner of a big calf breeding farm was suspected by the prosecutor of using brombuterol
(Fig.1) illicitly to increase growth of calves. Several seized samples were to be analyzed in the
laboratories of the Bavarian State Bureau of Investigation (Bayerisches Landeskriminalamt) to
ensure that these animals were treated with brombuterol. Among the seized items were some
vials containing a solution, some used syringes and hair samples taken from 61 calves.
93
The results will be discussed.
Fig.1: Molecular structure of brombuterol
3.1 Experimental
3.1.1 Preparation of hair samples
The hair samples were placed into a Eppendorf microtube and washed by vortexing in an
ultrasonic bath twice with 1ml n-hexane (60 s), twice with 1 ml acetone (60 s). The prewashed
samples were transferred into a folded filter, remaining residues of solvents evaporated at room
temperature. Each specimen was then cut into 1-2 mm sections, and a 15 mg quantity was
weighed into an Eppendorf microtube. After addition of 0,5 ml methanol, the mixtures were
vortexed in an ultrasonic bath (40C) for 60 min and following centrifugation for 15 min. The
mixtures were allowed to stand overnight. The methanolic phase was decanted, transferred into a
vial and evaporated at 40C under a gentle stream of nitrogen.
The residues obtained were then derivatized with 100l of a mixture (1:1, v/v) of
acetonitrile/hexamethyldisilazane (HMDS) for 45 min at 75C. The mixture was dried again
under nitrogen at 25C and reconstituted in 100l methylenechloride/HMDS (95:5, v/v). A 1l
quantity of the derivatized samples were then injected into the GC/MS/MS system.
3.1.2 Instrumentation
The analyses were performed with a Finnigan triple stage quadrupole mass spectrometer (TSQ
700) coupled with a DEC station 2100 (GC/MS/MS). The tandem mass spectrometer is linked to
a Varian 3400 gas chromatograph equipped with an A 200 S autosampler.
A fused-silica capillary column (J&W, DB-5, 30 m x 0,32 mm i.d., 0,25 m film thickness) was
used for GC separation.
The injector was controlled with a cold injection system and with solvent purging (Gerstel,
Mhlheim, Germany). Injection temperature was 40C, 40-60C at 2C/min, 60s at 60, 60-
94
280C at 12C/s and temperature was kept at 280C for 60s. The interface was maintained at
280C. Helium was used as carrier gas with a flow rate of 1,0 ml/min.
The mass spectrometer was operated in the negative ionisation mode (NCI) using methane as the
reagent gas set at 220 000 millitorr. The electron multiplier was operated at 2000 V.
3.2 Discussion of the results

Among the seized items from the calf breeding farm were vials containing a solution and syringes
contaminated with unknown material. Analyses of these items were conducted with thin-layer
chromatography, gaschromatography (ECD) and gaschromatography/mass spectrometry. The
component identified was brombuterol.
Testing of the hair samples collected from 61 calves was utilized as a corroborative evidence. A
positive result for residues of brombuterol would be a sufficient evidence.
After derivatization by a silylating compound (HMDS) brombuterol could be detected in 13 hair
samples in the SIM mode using negative chemical ionisation (Fig.2). The semiquantitative
determination of the analyte is conducted with external reference standard brombuterol. The
concentrations ranged between 10 pg and 100 pg/mg hair. These findings are of the same
magnitude comparable with data determinated in hair specimens of calves being subjected
illegally clenbuterol administration [5,6]. It is about 5-10 fold above the dosage used for
therapeutical application.
Human consumption of the 2-agonist brombuterol from treated animals is presumed to be a
significant health issue [9].
It should be mentioned that there is a difference in reactivity between clenbuterol and
brombuterol. Unlike its chlorinated analog clenbuterol producing a bistrimethylsilyl substituted
derivate, only a single trimethylsilyl group is attached to the brombuterol moiety.
95
Fig.2: SIM chromatograms and reconstructured ion chromatogram (RIC) of a hair sample
collected from a calf positive for brombuterol.
96
Despite the opportunity of utilizing a GC/MS/MS (tandem mass spectrometry) and selected
reaction (SRM) monitoring analyses were conducted by GC/MS in the SIM mode.
These results show that also hair analysis can be utilized for the detection of 2-agonists in sports
doping.
Acknowledgement
The authors would like to thank Mrs. G. Sichelstiel for sample preparation and Mr. X. Neumeier
for performing the GC/MS analyses.
References
[1] E. Edelhuser, E. Scherbaum, Deutsche Lebensmittel-Rundschau (1991) 37[2] Muscling in on clenbuterol, Lancet 340 (1992) 403.
[3] M.I. Delday, P.E. Williams, and C.A. Martin, Effect of clenbuterol on immobilized muscle. J.
Neurol. Sci. 89 (1990) 376.
[4] A. Gleixner, H. Sauerwein, and H.H.D. Meyer, Nachweis von Clenbuterol in Kopfhaar: eine
Methode zur Trainingsdopingkontrolle, Deutsche Zeitschrift fr Sportmedizin 48 Nr. 2 (1997)
50-55.
[5] Y. Gaillard, A. Balland, F. Doucet, and G. Pepin, Detection of illegal clenbuterol in calves
using hair analysis. Application in meat quality control, J. Chromatogr. B 703 (1997) 85-95.
[6] L.-E. Appelgreen, U. Bondesson, E. Frederiksson, C. Ingvast Larson, and D.S. Jansson,
Untersuchung von Haarproben von Klbern auf Clenbuterol, Fleischwirt-schaft 76, 3 (1996) 314316.
[7] A. Polettini, M. Montagna, J. Segura, and X. De La Torre, Mass Spectrom. 31 (1996) 47.
[8] K. Einhellig, Vergiftete Lebensmittel aus der Praxis eines forensischen Labors, In F. Pragst
und R. Aderjan (eds.), Beitrge zum XI. Symposium der Gesellschaft fr Toxikologie und
Forensische Chemie, Mosbach 22.-24. April 1999, pp 74-82.
[9] J.F. Martinez-Navarro, Food poisoning related to consumption of illicit -agonist in liver.
Lancet 336 (1990) 1311.
* corresponding author
Testing Endogeneous Steroids in Hair

Hans Sachs
During the examining of human hair on illegal drugs, some testostosterone metabolites like
androsterone and epi-androsterone could also be detected by gaschromatography/mass
spectrometry (1). In 1989 it did not seem promising to use the quantification of hormones in hair
to prove the additional application of testosterone or its derivatives as anabolic agent.
Many steroids which are close to testosterone in the biosynthetic pathway of testosterone (fig.1)
are commercially available and abused as anabolics (2).
1)
HO
Dehydroepiandrosterone
(DHEA)
2)
Androstendione
2)
H
OH
OH
1)
HO
5-Androstendiol
Testosterone
3)
H
OH
O
H
Reductases:
1) 3Hydroxysteroid-R.
2) 17Hydroxysteroid-R.
3) 5 R.
5-Dihydrotestosterone
98
Testing Endogenous Steroids in Hair
In 1995 Scherer (3) determined the androgens androstenediol, testosterone, androstenedione,

dehydro-epiandrosterone, and dihydro-testosterone as well 17-hydroxy-progesterone, a
precursor of the androgens. After enzymatic digestion the solution was extracted on Sephadex
LH-20. The extracts were derivatized with HFBA and determined by GC/MS. The following
results were obtained:
Androstenediol
Testosterone
Androstenedione
Dehvdro-epiandrosterone
9.1 - 18.7 ng/g

12.9 - 24.3 ng/g
4.9 - 14.6 ng/g
21.3 - 56.0 ng/g
Dihydro-testosterone
2.1 -7.9 ng/g
17-hydroxy-progesterone
1.3 - 6.9 ng/g
These preliminary studies did not lead to a procedure to distinguish between hair of male and
female subjects, or of children before and after puberty. In a subsequent study the author found a
significant sex difference with 2.7 ng/g (2.5 - 4.2) in male and 1.7 ng/g (1.0 - 3.4) in female hair
(4).
In 1998 Wheeler et al. (5) determined testosterone levels in the hair fo 22 male and 19 female
subjects. The sample was digested in sodium hydroxide. The solution was cleaned up by HPLC,
which provided a separation between testosterone, androstenedione, and dihydrotestosterone,
before the analytes were quantified by radioimmunoassay. The authors found a clear difference
between testosterone concentrations found in hair collected from men (12.9 - 77.7 pmol/g) and
those found in hair from women (<0.9 - 10.8 pmol/g). The concentrations of children lay in the
range of the female subjects.
Deng et al. (6) published results of anabolic steroids including some endogenous compounds in
head hair of 7 adult white male abusers. The hair was digested in 1 N NaOH, the pH adjusted to
5.6 with 6 N HCl and extracted over a C18 Isolute SPE cartridge. The compounds were
quantified by GC/MS after derivatisation with BSTFA/1% TMCS. Among other compounds the
following steroids were detected:
99
Dehydromethyltestosterone
up to 0.02 ng/mg
Testosterone
up to 0.15 ng/mg
Progesterone
Up to 0.04 ng/mg
Epitestosterone
0.05 in 1 case
Methyltestosterone
Up to 0.17 ng/mg
Dihydromethyltestosterone
In traces
Estradiol
Up to 0.11 ng/mg
The authors stated that the testosterone concentration of 0.15 ng/mg clear reflects abuse.
Hair samples of numerous steers ("Fleckvieh" with white and black hair), cows, bulls, female and
male calves were examined on estradiol and testosterone by Gleixner et al. (7). The compounds
were extracted from hair with liquid-liquid and solid-phase extraction procedures. The extracts
were cleaned up by HPLC and estradiol and testosterone were quantified by an enzyme
immunoassay. The estradiol concentrations in the hair of steers, cows, and bulls lay in the range
of 1 ng/g or below. The testosterone concentrations in the hair of steers was about 3 ng/g, in hair
of cows 6 ng/g, and in the hair of bulls in average 15 ng/g. Significant differences were found in
white hair (8 ng/g) and black hair (33 ng/g). There was no significant difference between the
concentrations in hair of male or female calves, or between the concentrations of hair with
different colours. The estradiol concentrations as well as the testosterone concentrations lay
below 10 ng/g.
Together with testosterone-esters Kintz et al. (8) determined the testosterone concentrations of 26
control subjects. After decontamination with dichloromethane, 100 mg of hair was incubated in
1M NaOH in the presence of 1 ng of testosterone-d-3. After neutralization, the extract was
purified using solid-phase extraction with isolute C-18 columns followed by liquid- liquid
extraction with pentane. After silylation, testosterone was analyzed by gas chromatography-mass
spectrometry. Concentrations were in the range 1.2 to 11.4 pg/mg with a mean value of 3.8
pg/mg.
Dehydroepiandrosterone (DHEA) is a steroid hormone naturally produced by the adrenal glands

and by the ovaries. DHEA can be converted into other hormones, including estrogen and
100
testosterone (fig. 1). In the United States, DHEA is classified as a nutritional supplement. This is
not the case in France, where the drug is listed as a doping agent. As athletes can abuse DHEA to
benefit from its conversion to testosterone, there is a need to establish the physiological range of
DHEA concentrations in human hair. DHEA was investigated by Kintz et al. (9) in hair obtained
from 27 control subjects, including 15 males and 12 females aged 17-42 years. After
decontamination with dichloromethane, 100 mg of hair was incubated in I M NaOH in presence
of 1 ng of testosterone-d3. After neutralization, the extract was purified using solid-phase
extraction with Isolute C18 columns and subsequent liquid-liquid extraction with pentane. After
silylation, DHEA was analyzed by gas chromatography/mass spectrometry. Results were linear in
the range 1-20 pg/mg. Relative extraction recovery was 91.6% with a limit of detection of 0.5
pg/mg. Concentrations were in the range 1.2-6.7 pg/mg (mean value of 4.3 pg/mg) and 0.5 to
10.6 pg/mg (mean value of 5.3 pg/mg) for the males and females, respectively.
In a further study, Kintz et al. (10) determined the testosterone and DHEA levels in hair of
different origin, head, axillary, and pubis. In accordance with results of hair examinations after
drug abuse, axillary and pubic hair show often higher concentrations than head hair, as shown in
the following table.
Subject
Testosterone
DHEA
Head
Axillary
Pubis
Head
Axillary
Pubis
4.1
182
3.9
3.2
1984
1363
3.3
39
3.5
6.7
719
90
2.2
17
4.7
4.9
2735
60
Table Ref (10)
101
Conclusion
Endogenous steroids, including precursors and metabolites of testosterone can be detected in

human hair. The concentrations lay in the range up to 100 pg/mg in head hair, but different
examiners found different levels of testosterone in control groups. Testosterone concentrations of
more than 100 pg/mg indicate an abuse of testosterone-derivatives, but more experience is
needed to prove exogenous testosterone. The introduction of a ratio-limit of
testosterone/epitestosterone has failed until now. The most elegant way to prove an intake of
substances which are also endogenous, is to detect the parent drug, e.g. the testosterone esters.
References
1. H. Sachs, R.M. Mller, Detection of drugs in hair by GC/MS, Fres Z Anal Chem 334, (1989)
713
2. C. Shackleton, E. Roitman, A. Phillips, T. Chang, Androstanediol and 5-androstenediol
profiling for detecting exogenously administered dihydrotestosterone, epitestosterone, and
dehydroepiandrosterone - potential use in gas chromatography isotope ratio mass
spectrometry, Steroids 62 (1997) 665
3. C. R. Scherer, G. Reinhardt, Nachweis sechs endogener Steroide in menschlichen Haaren mit
GC/MS und Isotopenverdnnungsanalyse, Kongressbericht der 74. Jahrestagung der
Deutschen Gesellschaft fr Rechtsmedizin, (Ed.H.Althoff), Aachen, 1995
4. C. Scherer, U. Wachter, S.A. Wudy, Determination of testosterone in human hair by gas
chromatography-selected ion monitoring mass spectrometry, Analyst 123 (1998) 2661-2663
5. M.J. Wheeler, Y.-B. Zhong, A.T. Kicman, S.B. Coutts, The measurement of testosterone in
hair, J. Endocrinol. 159 (1998) R5-R8.
6. X.S. Deng, A. Kurosu, D.J. Pounder, Detection of anabolic steroids in head hair, J Forensic
Sci, 44 (1999) 343-346
7. A. Gleixner, H.H.D. Meyer, Detection of estradiol and testosterone in hair of cattle by
HPLC/EIA, Fresenius J. Anal. Chem. 357 (1997) 1198
8. P. Kintz, V. Cirimele, T. Jeanneau, B. Ludes, Identification of testosterone and testosterone
esters in human hair, J. Anal. Toxicol. 23 (1999) 424-428
9. P. Kintz, V. Cirimele, B. Ludes, Physiological concentrations of DHEA in human hair, J.
Anal. Toxicol. 23 (1999) 424-428
10. P. Kintz, V. Cirimele, B. Ludes, Pharmacological criteria that can affect the detection of
doping agents in hair, to be submitted
102
Detection of Exogenous Anabolic Steroids in Hair

Detlef Thieme, Joachim Grosse and R. Klaus Mueller
Institute of Doping Analysis and Sports Biochemistry,
Dresdner Str. 12, D-01731 Kreischa (near Dresden) GERMANY
Introduction
The definition of exogenous steroids is certainly arbitrary. Typical exogenous steroids of abuse
may be considered endogenous in trace amounts in certain individuals (boldenone) or under
specific circumstances (nandrolone in pregnancy), and the endogenous character of anabolic
steroids is doubtful, if they are abused in excessive overdose amounts. However any possible
chemical modification (incl. esterification) of endogenous substances outside the biosynthetic
pathways of steroids was included into the considerations.
The purpose of this review was to summarize and compare technical aspects, analytical results,
data evaluations and resulting conclusions published in recent papers.
Extraction
There is a common agreement that extraction of hair sample is done by
Sodium hydroxide disintegration for all steroids with sufficient chemical stability. Hydroxide
concentration, temperature and digestion time applied to enable complete homogenisation of
hair sample differ considerably.
Methanol extraction for all steroids, which may undergo hydrolysis if treated with hydroxide,
especially steroid esters. The procedures were adapted by elevation of temperature, ultrasonication and addition of modifiers (trifluoroacetic acid [1]).
Sample pre-treatment was applied in some cases although a passive contamination is less
probable. Considerable wash-out effects have been observed after sample washing with aqueous
dodecyl sulfate [2] diminishing the concentration of the steroid in hair.
104
Clean-Up Procedure
The application of clean-up procedures includes
liquid-liquid extraction, (organic solvents: ethyl acetate, ether, hexane)
solid phase extraction, using

C18-silica as most versatile technique for routine purposes, [3], [4],
NH2 - for fixation of polar impurities, [5]
XAD2 unspecific removal of polar substances, [6]
sephadex to clean the extract
[7]
lipidex - enrichment and clean-up of derivatives
HPLC clean-up procedures using reverse phase liquid chromatography [6] permits to obtain
clean fractions which may be analysed independently.
Derivatisation
There is a wide spectrum of derivatisation techniques [8] which may be summarised with respect
to steroids analysis in hair to the classes:
per-TMS derivatives of 17-hydroxy,3-keto-steroids
OSi(CH3)3
H
(incl. stanozolol) and mono TMS derivatives of the

respective esters, [3] [1] [9]
(CH3)3SiO
MO-TMS derivatives 17-hydroxy,3-keto-steroids [2],
[7], [4]
OSi(CH3)3
O-TMS,N-HFB-derivative of stanozolol, [10] [4]

CH3ON
bis-HFB or HFB-PFBO esters of 3-keto-steroids and mono PFBO derivatives of the

respective esters [11]
H
OOCC 3F7
applied to enable NCI-MS.

F
CH2OON
F
105
The advantages of per-TMS derivatives are fast and robust reactions with single reactions
resulting in uniform products (only few exceptions forming isomers: mesterolone [9],
oxymetolone, mestanolone, androstnolone [4]). The mass spectrum is often poor and
characterised by intense molecular ions. Methoxime or oxime derivatives are less stable and form
two isomers (syn and anti), but their MS/MS sensitivity is often improved due to diagnostic
fragmentation patterns and combined with perfluoroacylic-derivatisations they are suitable for
NCI/MS.
Abbreviations of derivatives:
TMS
trimethylsilyl
MO
methoxime
HFB
heptafluorobutyryl
PFBO
pentafluorobenzyl oxime
Detection Techniques
Early attempts of identification of steroids in hair were based on immunoanalysis techniques like
ELISA [12] or EIA after HPLC separation [13]. The combination of HPLC with scintillation
detection of radio-labelled compounds is probably exceptional, [2] whereas the combination of
GC with conventional (unit resolution) mass spectrometry appeared to be the most frequent
detection technique. Electron impact ionisation was applied as versatile principle for detection of
many exogenous steroids including steroid esters [3], [7], [2]. Detection based on chemical
ionisation was applied in positive (steroid esters [11]) and negative mode (steroids especially
stanozolol [11]). High resolution mass spectrometry in EI ionisation mode was employed to
identify anabolic steroids with improved sensitivity and specificity [6, 9]. The application of
tandem mass spectrometry is another option to improve detection limits, were all kind of steroids
were investigated in EI ionisation mode with triple stage quadrupole [5], sectorfield-quadrupole
[6], or ion trap [1] coupling.
106
Qualitative Results
There are many anabolic steroids included into screening experiments and detectable at adequate
detection limits, like
17-alkyl-derivatives (bolasterone, metandienone, methyltestosterone, stanozolol,

mestanolone etc.)
17-esters (acetates, propionates, isocaproates, cypionates, phenylpropionates, enantates,

decanoates, undecanoates and dodecanoates of testosterone, nandrolone and metenolone)
A-ring modifications of testosterone (clostebol, mesterolone, metenolone, oxymetolone etc.)
Others (nandrolone, trenbolone).
Compared to these exogenous steroids, which are potentially detectable, there are only some of
them identified in real cases (humans, usually bodybuilders). The spectrum of these depends
mainly on the popularity of / access to the steroids and occurrence in real cases is more or less
by chance:
Methandienone [7]
methyltestosterone and metandienone [4]
stanozolol and nandrolone [14]
metandienone, stanozolol, metenolone enantate, nandrolone decanoate, testosterone enantate,

propionate, isocaproate, phenylpropionate and decanoate [6]
testosterone undecanoate [5]
Few studies to control the incorporation of steroids into hair were directed to stanozolol,
nandrolone and testosterone esters. [1, 2, 10-12]
Little is known about the identification of metabolites in hair. 3-OH-stanozolol was detected in
concentrations 30 times lower than stanozolol and epi-methandienone and 6-OH-methandienone
were detected in trace amounts of a bodybuilders hair sample which was clearly metandienonepositive. [6]
107
Quantification
Quantitation limits examined in systematic studies are in the low pg/ml range. The published
numbers vary between 0.08 pg/mg for nandrolone GC/MS-MS and 200 pg/mg, logically
correlated to sample preparation, derivatisation and especially to the mass spectrometric detection
method.
The concentration of the steroids detected range from 13.4 pmol/g up to several ng/mg. However
they are in general in the order of magnitude of the detection limits. For the random findings in
forensic cases, the concentration of exogenous steroids were in the low pg/mg concentration
range too.
There is a general tendency that concentration in steroid esters are higher than the corresponding
concentrations in the hydrolysed steroid. Moreover the fact that concentrations in pigmented hair
are higher than unpigmented samples is undisputed. There are clear experimental evidences for
an accumulation of stanozolol [10], nandrolone-laurate[2] and nandrolone [11] in pigmented
tissues.
Species and Dosages
The major part of human studies was based on forensic cases, abusers or volunteers amongst
bodybuilders. Typical dosage recommendations as well as urinary concentrations of positive
cases proved to be much higher (at least by factor 10) in this group compared to other
populations. Therefore it is doubtful that this group represents a reliable model population, for
clinical consideration or doping control.
Controlled animal studies under defined conditions had been carried out with guinea pig (single
dosage of 10 mg/kg or sequence of 20 mg/kg intramuscular application of nandrolone [1, 8]),
Long-Ewans rat (20 mg/kg intraperitoneal ingestion of stanozolol, 60 mg/kg daily intraperitoneal
ingestion on 10 subsequent days, [10, 11] and steer (2 mg/kg intramuscular injection of
nandrolone laurate [2]).
Most of the applied dosages were far beyond the therapeutic ranges; only in one study practically
relevant dosages of a radio-labelled steroids were administered to steer [2], administration of
single dosages of steroids were not detectable in hair [1] .
108
Conclusions
The identification of anabolic steroids in hair is a well established experimental procedure. There
is no obvious qualitative limitation for an incorporation of exogenous steroids into hair, if they
are administered at high concentrations. Application in forensic cases, like
evaluation of self-consumption in case of possession or trafficking of anabolics
diminished responsibility of steroid abuser
unexpected fatalities amongst bodybuilders
are realistic because the presumptive dosages are high and target substances are known.
On the other hand, the similarity of detection limits of procedures with realistic target
concentrations demonstrates the general unknown dilemma:
The detection and quantitation of anabolic steroids in hair required hyphenated techniques, in
spite of high dosages and methods optimised to few substances only. Screening procedures
covering hundreds of potential steroids with adequate sensitivity are not available.
Many phenomena are not yet examined very well and cannot be explained conclusively.
Hair concentrations of esters are expected to be higher than these of corresponding free steroids.
This is described in some cases [1, 6, 11] whereas an increase of testosterone was described in
parallel with undetectable testosterone esters [1]. Controversial results like these are clearly due
to technical limitations.
The knowledge about the influence of metabolism of steroids is still unclear. Hair growth is
primarily controlled by activity of endogenous steroids, the presence of receptors, activities of
enzymes, (especially 5a-steroid reductase) and corresponding blockers. These mechanisms
depend at least on age, race, sex, health condition and body site (even regions of scalp hair) [15,
16] and are therefore expected to exhibits a great biovariability.
Nevertheless, suggestions to apply hair analysis range from a complementary method to its
application in routine analysis, even replacement of urine testing. Expectations of an application
to problems like testosterone doping in sports are high, although the knowledge of basal values in
hair are based on some hundred evaluations (compared to a database in urine of several hundred
thousands).
An attempt to establish a harmonisation was made by the Society of Hair Testing in 1999 [17].
109
References
[1]
J. Segura, S. Pichini, S. H. Peng, X. de la Torre, Hair analysis and detectability of single

dose administration of androgenic steroid esters, Forensic Sci Int 107 (2000 Jan 10) 347.
[2]
M. J. Sauer, T. P. Samuels, L. G. Howells, M. A. Seymour, A. Nedderman, E. Houghton,

S. J. Bellworthy, S. Anderson, N. G. Coldham, Residues and metabolism of 19nortestosterone laurate in steers, Analyst 123 (1998) 2653.
[3]
P. Kintz, V. Cirimele, T. Jeanneau, B. Ludes, Identification of testosterone and

testosterone esters in human hair, J Anal Toxicol 23 (1999) 352.
[4]
X. S. Deng, A. Kurosu, D. J. Pounder, Detection of anabolic steroids in head hair, J

Forensic Sci 44 (1999) 343.
[5]
Y. Gaillard, F. Vayssette, A. Balland, G. Pepin, Gas chromatographic-tandem mass

spectrometric determination of anabolic steroids and their esters in hair. Application in
doping control and meat quality control, J Chromatogr B Biomed Sci Appl 735 (1999)
189.
[6]
D. Thieme, J. Grosse, R. K. Mueller, H. Sachs, Detection of several anabolic steroids of

abuse in human hair,Manfred Donike Workshop (Cologne), Detection of several anabolic
steroids of abuse in human hair (1998).
[7]
C. Scherer, S. A. Wudy, Nachweis von Dianabol in Haaren,DGRM-Jahrestagung

(Zrich), Nachweis von Dianabol in Haaren (1996).
[8]
J. Segura, R. Ventura, C. Jurado, Derivatization procedures for gas chromatographic-mass

spectrometric determination of xenobiotics in biological samples, with special attention to
drugs of abuse and doping agents, J Chromatogr B Biomed Sci Appl 713 (1998) 61.
[9]
D. Thieme, J. Grosse, H. Sachs, R. K. Mueller, Analytical strategy for detecting doping

agents in hair, Forensic Sci Int 107 (2000 Jan 10) 335.
[10]
K. M. Hold, D. G. Wilkins, D. J. Crouch, D. E. Rollins, R. A. Maes, Detection of

stanozolol in hair by negative ion chemical ionization mass spectrometry, J Anal Toxicol
20 (1996) 345.
[11]
K. M. Hold, C. R. Borges, D. G. Wilkins, D. E. Rollins, R. E. Joseph, Jr., Detection of

nandrolone, testosterone, and their esters in rat and human hair samples, J Anal Toxicol 23
(1999) 416.
110
[12]
J. Segura, Possibilities of ELISA methodologies for hair analysis,Hair Analysis in

Toxicology (Abu Dhabi), Possibilities of ELISA methodologies for hair analysis, pp. 351
(1995).
[13]
A. Gleixner, H. Sauerwein, H. H. Meyer, Detection of the anabolic steroid

methyltestosterone in hair by HPLC-EIA, Chromatographia 45 (1997) 49.
[14]
P. Kintz, V. Cirimele, H. Sachs, T. Jeanneau, B. Ludes, Testing for anabolic steroids in

hair from two bodybuilders, Forensic Sci Int 101 (1999) 209.
[15]
N. Obana, C. Chang, H. Uno, Inhibition of hair growth by testosterone in the presence of

dermal papilla cells from the frontal bald scalp of the postpubertal stumptailed macaque,
Endocrinology 138 (1997) 356.
[16]
S. Itami, S. Kurata, T. Sonoda, S. Takayasu, Interaction between dermal papilla cells and
follicular epithelial cells in vitro: effect of androgen, Br J Dermatol 132 (1995) 527.
[17]
H. Sachs, P. Kintz, Consensus of the society of hair testing on hair testing for doping
agents, Forensic Sci Int 107 (2000) 3.
Testing for 19-Norsteroids in Hair

P. Kintz , V. Cirimele, B. Ludes
Institut de Mdecine Lgale, 11 rue Humann,
F-67000 Strasbourg, France
Introduction
Androgen use was widespread among athletes at the time of the 1964 Olympic Games, and,
despite a ban by the International Olympic Committee (IOC) in 1974, the use of anabolic steroids
increased during the last decade.
Nandrolone or 19-nortestosterone, or 17-hydroxy-19-nor-4-androsten-3-one, has been one of the
most abused anabolic steroids and doping practices are increasing, as revealed by numerous
positive cases during the last 3 years in judo, tennis, football or athletism. The drug accelerates
muscle growth by an anabolic effects. Athletes use nandrolone because it has been claimed that it
increases lean body mass, increases strength, increases aggressiveness and leads to a shorter
recovery time between workouts.
Nandrolone is metabolized into norandrosterone (NA) and noretiocholanolone (NE) (1). Other
19-norsterods, such as norandrostenedione or norandrostenediol, classified as dietary
supplements, are available over-the-counter or through the Internet and have the same
metabolites as nandrolone (1-3).
Long-term effects (severe cardio-vascular side-effects, liver diseases, etc.) and fatalities have
been reported in young steroid abusers (4, 5).
The standard of testing for anabolic steroids for doping control is gas chromatography coupled to
mass spectrometry conducted on an urine sample, and performed in accredited laboratories.
Nandrolone doping is claimed when norandrosterone is detected in urine with a chromatographic
112
response higher than that obtained with a standard spiked at 2 or 5 ng/ml for men or women,
respectively (6). Even if norandrostenediol and norandrostenedione are banned by the IOC, there
is an great interest, at least in forensic science and for survey of the athletes, to discriminate
nandrolone abuse from other 19-norsterods. This is obviously not possible in urine, as the
metabolites are common. The ratio between NA and NE has been recently proposed to
differentiate the nature of the doping agent, but this seems extremely cautious, as mixtures can be
ingested (3).
As forensic laboratories can be involved in testomony dealing with doping agents, the idea of
using hair for doping control has emerged as hair analysis has been accepted in court in other
cases. Courts can request additional informations on the pattern of use of doping substances, such
as during the 1998 cycling Tour de France where blood, urine, and hair were simultaneously
collected. Hair can both confirm repetitive abuse and identify the exact nature of the parent
compound (e.g., nandrolone, norandrostenediol or norandrostenedione, in case of positive urine
for norandrosterone), as it has been accepted by the scientific community that the parent
compound is the major analyte that is incorporated in hair (7, 8). Thus, hair analysis would
discriminate nandrolone abuse from over-the-counter preparations containing 19-norsterods.
Case reports
Recently, our laboratory was requested by an attorney to evaluate potential doping practices from
an athlete. The subject was 30-year old and tested positive for norandrosterone in urine at 230
ng/ml. The analysis was done in an accredited laboratory, but the athlete denied the result.
Nandrolone, but also norandrostenediol and norandrostenedione were then tested in hair,
according some modifications of our previous procedure for testosterone (9). Full-length hair
113
samples (12 cm long) were taken at the surface of the skin from the vertex of the athlete and
stored in plastic tubes at room temperature. Only the first 6-cm from the root were analyzed.
Hair strands were obtained from 3 bodybuilders who were arrested by the French customs,
accused of trafficking. In their luggage, the officers discovered a large amount of anabolics in
ampoules. Full-length hair samples (3 to 5-cm long) were taken at the surface of the skin from the
vertex and stored in plastic tubes at room temperature.
Controlled hair specimens were obtained from laboratory personal.
Analytical procedure
One hundred milligrams of decontaminated hair (with dichloromethane) were incubated in 1 ml
1N NaOH, 15 min at 95C, in presence of 10 ng of nandrolone-d3 used as internal standard. After
cooling, the homogenate was neutralized with 1 ml 1M HCl, and 2 ml of 0.2 M phosphate buffer
(pH 7.0) were added. The drugs were extracted by solid-phase extraction. The Isolute C18
columns were conditioned with 3 ml of methanol, followed by 2 ml of deionized water. After
sample addition, the columns were washed twice with 1 ml of deionized water. After column
drying, analytes elution occured with the addition of 3 aliquots of 0.5 ml of methanol. The eluant
was evaporated to dryness under nitrogen flow, and the residue reconstitued in 1 ml of 0.2 M
phosphate buffer (pH 7.0). A further purification step was achieved by addition of 100 mg of
Na2CO3/NaHCO3 (1:10, w/w) and 2 ml of pentane. After agitation and centrifugation, the
organic phase was removed and evaporated to dryness. The residue was derivatized by adding 50
l MSTFA/NH4I/2-mercaptoethanol (1000 : 2 : 5 ; v/v/v), then incubated for 20 min at 60 C.
A 4-l aliquot of the derivatized extract was injected into the column of a Hewlett Packard (Palo
Alto, CA) gas chromatograph (6890 Series) via a Hewlett Packard (7673) autosampler. The flow
114
of carrier gas (helium, purity grade N 55) through the column (HP5-MS capillary column, 5 %
phenyl-95 % methylsiloxane, 30 m x 0.25 mm i.d. x 0.25 mm film thickness) was 1.0 ml/min.
The injector temperature was 270 C and splitless injection was employed with a split valve offtime of 1.0 min, using the pulsed mode. The column oven temperature was programmed to rise
from an initial temperature of 150 C, maintained for 1 min, to 295 C at 30 C/min and
maintained at 295 C for the final 8 min.
The detector was a Hewlett Packard 5973 operated in the electron impact mode. The electron
multipler voltage was set at 600 V above the EI-tune voltage.
Results and discussion

Under the chromatographic conditions used, there was no interference with the analytes by any
extractable endogenous materials present in hair. Selected ions, retention times and analytical
parameters of the drugs are reported in Table 1. Analytes were identified and quantified on the
basis of their retention times and the abundance of three confirming ions.
Responses for the 3 norsterods were linear in the range 1 to 50 pg/mg, with a correlation
coefficient in the range of 0.992 to 0.997.
The within-run precisions were in the range 11.2 to 17.6 %, as determined by analyzing 6
replicates of 100 mg of drug-free hair obtained from the same subject and spiked with a final
concentration at 10 pg/mg. The extraction recovery (n = 3) was determined to be in the range
76.1 to 86.4 %. The limits of detection (S/N > 3) of the doping agents were in the range 0.5 to 1
pg/mg.
The analysis of a strand of hair obtained from the athlete revealed the presence of 19norandrostenedione at the concentration of 7 pg/mg. Extensive chromatographic procedures (2
purification steps by solid phase and liquid-liquid extraction, combined with injection of 4 l
115
through the column in pulsed mode) were analytical prerequisites for successful identification of
norandrostenedione in hair due to the low target concentration. The analysis of 3 strands of hair,
obtained from 3 bodybuiders revealed the presence of nandrolone at the concentrations of 1, 3.5
and 7.5 pg/mg.
The sensitive, specific and reproducible method developed seems to be suitable for the detection
and quantification of 19-norsterods in human hair. The determination of one 19-norsterod in
hair should allow a definitive unambigous confirmation of the exact nature of the abused agent.
This seems one of the major advantage of hair over urine, as it was also clearly demonstrated that
the identification of testosterone esters in hair can confirm the administration of exogenous
testosterone (9, 10).
The international literature is very poor in papers dealing with the identification of nandrolone in
hair. One paper, published in 1995, has been focused on guinea pigs (11) with unrealistic
nandrolone dosages, 10 or 20 mg/kg. In 1999, Hld et al (12) demonstrated that it is possible to
detect nandrolone in rat hair after systemic administration. However, the growing interest of
scientists to the detection of anabolic drugs in human was observed in the late 1998.
In their early work, Cirimele et al (13), published results from 2 bodybuilders preparing the next
world championship, with positive hair for nandrolone at 196 and 260 pg/mg. Deng et al (14),
among other anabolics, identified nandrolone in the hair of a steroid abuser at 20 pg/mg. Gaillard
et al (15) detected nandrolone in the hair of a cyclist at 5.1 pg/mg.
When using hair in a suspected doping case, particularly when urine of the athlete was positive
and hair negative (several cases were reported during the past 3 years), the question of
importance is to know whether the analytical procedure was sensitive enough to identify traces of
drugs. It has been always accepted in the forensic community that a negative hair result cannot
116
exclude the administration of the detected drug or one of its precursors (such as norandrostendiol
or norandrostendione for the metabolites of nandrolone) and should not overrule a positive urine
result. Nevertheless, the negative hair findings lends enough ambigity to the positive urine result,
coupled with the sporting consequences for the athlete, that substancial justice refereeing occurs.
This laboratory was not able to identify nandrolone in the hair of a 37-year old man, receiving a
single intramuscular injection of 50 mg nandrolone undecanoate, although his urine remained
positive for norandrosterone and noretiocholanolone, the nandrolone metabolites, for at least 8
months. Hair was tested 2 and 6 months after administration (16). The same observations were
recently made by Segura, et al (17) who did not detect nandrolone after a single dose
administration. Therefore, until laboratories will have sensitive enough methodologies to detect a
single use of steroids, care should be taken to compare urine and hair findings. For anabolic
steroids to have an appreciable proformance enhancing effect, they must be chronically
administered, in contrast to the immediate stimulant properties of cocaine or amphetamine.
Repeated amount of drug used per hair growth length would favor identification by hair analysis.
Conclusion
The sensitive, specific and reproducible method developed seems to be suitable for the detection
and quantification of nandrolone in human hair.
Hair analysis may be a useful adjunct to conventional drug testing in sports. Methods for evading
urine analysis do not affect the drug concentrations in hair. Specimens can be more easily
obtained with less embarrassment, and hair can provide a more accurate history of drug use. This
technology may find useful applications in doping control, if accepted by the International
Olympic Commitee (18).
Table 1
117
Selected ions, retention times and analytical parameters for nandrolone,
norandrostenediol and norandrostenedione in human hair
Parameters
norandrostenediol
norandrostendione
nandrolone
6.99
7.19
7.27
Ions (m/z)
330, 405, 420
344, 401, 416
287, 403, 418
Linearity (1-50 pg/mg)
0.992
0.995
0.997
Within run precision (%)
17.6
14.3
11.2
Recovery (%)
76.1
86.4
81.7
LOD (pg/mg)
0.5
0.5
The underlined ions were used for quantification.

nandrolone-d3 was eluted at 7.26 min, and the ion (m/z) 421 was used for quantification.
118
References
1.
P. Kintz, V. Cirimele, B. Ludes, Norandrosterone and noretiocholanolone :

metabolite markers, Acta Clin. Belg. suppl 1 (1999) 68-73.
2.
V.P. Uralets, P.A. Gillette, Over-the-counter anabolic steroids 4-androsten-3,17dione ; 4-androsten-3,17-diol ; and 19-nor-4-androsten-3,17-dione : excretion
studies in men, J. Anal. Toxicol . 23 (1999) 357-366.
3.
V.P. Uralets, P.A. Gillette, Over-the-counter anabolic steroids 5-androsten-3,17dione ; 5-androsten-3,17-diol ; dehydroepiandrosterone ; and 19-nor-5-androsten3,17-dione : excretion studies in men, J. Anal. Toxicol . 24 (2000) 188-193.
4.
R. Hausmann, S. Hammer, P. Betz, Performance enhancing drugs (doping agents)

and sudden death-a case report and review of the literature, Int. J. Leg. Med.
111 (1998) 261-264.
5.
B. Madea, W. Greener, F. Mussoff, R. Dettmeyer, Medico-legal aspects of doping.

J. Clin. Forensic Med. 5 (1998) 1-7.
6.
International Olympic Committee. IOC Olympic Movement Antidoping Code, IOC,

Lausanne, Switzerland (1999).
7.
E.J. Cone, Mechanisms of drug incorporation into hair, Ther. Drug Monit. 18
(1996) 438-443.
8.
G.L. Henderson, M.R. Harkey, C. Zhou, R.T. Jones, P. Jacob P, Incorporation of

isotopically labeled cocaine and metabolites into human hair : 1. Dose-response
relationships, J. Anal. Toxicol. 20 (1996) 1-12.
9.
P. Kintz, V. Cirimele, T. Jeanneau, B. Ludes, Identification of testosterone and

testosterone esters in human hair, J. Anal. Toxicol. 23 (1999) 352-356.
10.
D. Thieme, J. Grosse, H. Sachs, R.K. Mueller, Analytical strategy for detecting

doping agents in hair, Forensic Sci. Int. 107 (2000) 335-345.
11.
J. Segura, Possibilities of ELISA methodologies for hair analysis. In R.A de Zeeuw,

I. Al Hossani, S. Al Munthiri and A. Maqbool (Eds) Proceedings of the 1995
international conference and workshop on hair analysis in forensic toxicology. Abu
Dhabi Police, Abu Dhabi, (1995) pp 351-369.
12.
K.M. Hld, C.R. Borges, D.G. Wilkins, D.E. Rollins, R.E. Joseph, . Detection of
nandrolone, testosterone, and their esters in rat and human hair samples J. Anal.
Toxicol. 23 (1999) 416-423..
13.
V. Cirimele, P. Kintz, T. Jeanneau, B. Ludes, Dosage de strodes anabolisants dans

les cheveux de deux culturistes, Toxicorama, 11 (1999) 30-34.
14.
X.-S. Deng, A. Kurosu, D.J. Pounder, Detection of anabolic steroids in hair, J.

Forensic Sci. 44 (1999) 343.
15.
Y. Gaillard, F. Vayssette, A. Balland, G. Ppin, Gas chromatographic-tandem mass

spectrometric determination of anabolic steroids and their esters in hair. Application
in doping control and meat quality control, J. Chromatogr. B 735 (1999) 189-205.
16.
P. Kintz, V. Cirimele, B. Ludes, Pharmacological criteria that can affect the

detection of doping agents in hair, Forensic Sci. Int. 107 (2000) 325-334.
17.
J. Segura, S. Pichini, S.H. Peng, X. de la Torre, Hair analysis and detectability of

single dose administration of androgenic steroid esters, Forensic Sci. Int. 107 (2000)
347-359.
18.
H. Sachs, P. Kintz, Consensus of the Society of Hair Testing on hair testing for
doping agents. Forensic Sci Int 107 (2000) 3.
119

Pharmacological Criteria That Can Affect The Detection of Doping Agents in Hair

Uploaded by

Copyright:

Available Formats

Pharmacological Criteria That Can Affect The Detection of Doping Agents in Hair

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pharmacological Criteria That Can Affect The Detection of Doping Agents in Hair

Uploaded by

Copyright:

Available Formats

See

Pharmacological criteria that can affect the

459 PUBLICATIONS 8,397 CITATIONS

173 PUBLICATIONS 3,292 CITATIONS

All in-text references underlined in blue are linked to publications on ResearchGate,

Available from: Vincent Cirimele

Dr. Martin-Peter Bch

Introduction to Hair Analysis.............................................................................................

Hair Sample Collecting Kit for the Police Forces..............................................................

Determination of Benzodiazepines in Hair by GC/MS-NCI..............................................

Specificity and Sensitivity Requirements in Hair Analysis................................................ 35

Reviews on Detection of Anabolic Agents in Hair

Detection of Exogenous Steroids in Hair........................................................................... 103

Progress in Hair Analysis for Illegal Drugs

Dr. Hans Sachs

Progress in Hair Analysis for Illegal Drugs

Introduction to Hair Analysis

The general fields of hair analyses are:

Criminal court proceedings

Driving ability control

Introduction to Hair Analysis

Fig.1: CZ= cuticle cells; R = Cortex; M= medulla

Progress in Hair Analysis for Illegal Drugs

Introduction to Hair Analysis

Kauert and Rhrich

Kintz and Mangin

20-30 mg powdered hair, 2 ml

propionic acid anhydride

1000 L PFPA/75 L PF-npropanol; 30 min/60C;

Column: 20m x 0.25 mm x 0.25

EI 70 eV; SIM at m/z

Column: 12m x 0.2 mm x 0.33

Introduction to Hair Analysis

Progress in Hair Analysis for Illegal Drugs

Hair Sample Collecting Kit for the Police Forces

Institute fr Rechtsmedizin, Universitt Bern, Bhlstrasse 20, CH-3012 Bern

Institut fr Rechtsmedizin, Universitt Mnchen, Frauenlobstrasse 7a, D-80046 Mnchen,

Hair Sample Collecting Kit for the Police Forces

R. Denk, I. Raff, H. Sachs, Haaranalysen bei Betubungsmittelkonsum, Kriminalistik 4 (1992) 253-255

Progress in Hair Analysis for Illegal Drugs

Hair Sample Collecting Kit for the Police Forces

Progress in Hair Analysis for Illegal Drugs

Hair Sample Collecting Kit for the Police Forces

Progress in Hair Analysis for Illegal Drugs

A General Solid-Phase Extraction Procedure for Drug Testing in Hair

The analytical procedures generally involve the following steps:

A General Solid-Phase Extraction Procedure for Drug Testing in Hair

2. Basic principles of solid-phase extraction (SPE).

Progress in Hair Analysis for Illegal Drugs

A General Solid-Phase Extraction Procedure for Drug Testing in Hair

The samples are extracted in the following ten steps:

A General Solid-Phase Extraction Procedure for Drug Testing in Hair

5. Rinsing column with acetate buffer pH 4 (1 ml)

This extract could be analysed by GC/MS or other chromatographic methods.

GC/MS analysis requires a procedure of derivatization, for example by propionylation. 100

at 60 oC. After evaporation under a stream of nitrogen, the extract is reconstituted in 50

3.5. Gas chromatography - mass spectrometry method

m, J&W Scientifics (Folsom, USA).

Progress in Hair Analysis for Illegal Drugs