Journal of Analytical Toxicology,Vol. 23, October 1999
Detection of Nandrolone, Testosterone,and their Esters
in Rat and Human Hair Samples
Karin M. HiJld1, Chad R. Borges1, Diana G. Wilkins 1, Douglas E. Rollins1,~ and Robert E. Joseph, Jr.2
1Centerfor Human Toxicology, Department of Pharmacology and Toxicology, University of Utah, Salt Lake City, Utah 84112
and 2Addiction ResearchCenter, National Institute on Drug Abuse, Baltimore, Maryland 21224
I Abstract I
Nandrolone and testosterone are anabolic androgenic steroids
occasionally abused by athletes. A sensitive, specific, and
reproducible gas chromatography-mass spectrometry method for
the quantitative determination of nandrolone, testosterone,and
their esters in hair has been developed. The limits of quantitation
of this method, based on 20 mg of hair, were 50 pg/mg for
nandrolone and testosterone, 100 pg/mg for testosterone acetate,
and 200 pg/mg for nandrolone-decanoate. Nandrolone-d3 and
testosterone-d3 were used as internal standards. This method has
been applied to the analysis of these compounds incorporated into
rat and human hair. Male Long-Evans rats were given nandrolone
decanoate 60 mg/kg intraperitoneally (ip) once daily for 10 days
over a time period of 14 days. Two of the three rats contained
nandrolone in the pigmented hair collected at day 21 at a
concentration of 63 and 76 pg/mg, respectively. No drug was
found in the corresponding nonpigmented hair. The rat hair
samples that tested positive for nandrolone contained also
nandrolone decanoate in concentrations of 0.9 and 1.2 ng/mg,
respectively. In a separate experiment rats were given testosterone
acetate 10 mg/kg ip once daily for five days. No testosterone or
testosterone acetate was detected in the rat hair samples. Hair
specimens were also obtained from four self-reported steroid
users. The hair of two subjects were determined to be positive for
testosterone in concentrations of 54 and 81 pg/mg. These data
demonstrate that it is possible to detect the steroids nandrolone,
testosterone, and nandrolone decanoate in hair after systemic
administration.
Introduction
Doping with endogenous steroids is one of the most serious
issues facing sports today. When cleverly administered, these
compounds are very difficult to detect. Quadrupole mass spectrometers (MS) cannot distinguish between pharmaceutical
testosterone and natural testosterone because their spectra
are identical (1).Although much progress has been made in de* Address correspondenceto Douglas E. Rollins, Center for Human Toxicology, University
of Utah, 20 S 2030 E RM 490, Salt LakeCity, UT 84112-9457.
Email [email protected].
416
tecting exogenous steroid abuse, detecting the abuse of these
endogenous substances like testosterone has been more difficult. The traditional method is measurement of the ratio of
testosterone to epitestosterone. Bowers and Sanaullah (2) developeda method to directly detect the sulfate and glucuronide
conjugates of steroid metabolites by using high-performance
liquid chromatography-tandem mass spectrometry
(HPLC-MS-MS), instead of indirect documentation by hydrolyzing urinary metabolites. A second new technique in use
is combustion isotope ratio MS, which renders it possible to
distinguish endogenous from synthetic testosterone because of
their difference in 13Ccontent (3). Another potential method
would be to characterize the intact testosterone esters in
plasma because the short-chain esters commonly used in drug
preparation are not synthesized in the body, in contrast to
long-chain fatty acid steroid esters (4,5). de la Torre et al. (6)
reported the gas chromatographic-mass spectrometric
(GC-MS) analysis of testosterone esters in plasma and found
that levelsas low as I ng/mL could be detected. Shackleton and
colleagues (7) developedan electrosprayMS method for the detection of testosterone esters in plasma. Using this technique,
testosterone enanthate and undecanoate were detected after intramuscular injection or oral administration of the drugs.
Hair has been proposed as an alternative matrix to urine or
plasma for detecting drug use and may be particularly useful
for the detection of anabolic steroids. An important feature of
hair analysis is the possibility of detecting drugs in hair for a
long period of time after ingestion. In a previous study we
were able to detect stanozolol in rat hair after intraperitoneal
administration (8). In this study, a sensitive, selective, and reproducible GC-MS method for the detection of nandrolone
and testosterone and their esters, nandrolone decanoate and
testosterone acetate, was developed.
Material and Methods
Reagents
Nandrolone, testosterone, and the internal standard testosterone-d3 were obtained from Radian Corp. (Austin, TX). Nandmlone decanoate and testosterone acetate were obtained from
Reproduction(photocopying)of editorialcontentof thisjournalis prohibitedwithoutpublisher'spermission.
Journal of Analytical Toxicology, Vol. 23, October 1999
Sigma (St. Louis, MO). Nandrolone-d3was obtained from Cambridge Isotope Laboratories (Andover, MA). Capillary
GC/GC-MS solvent grade methanol, ethyl acetate, chloroform,
iso-octane and hexane were obtained from Baxter (McGaw
Park, IL). Analytical reagent-grade potassium phosphate, potassium hydroxide, sodium hydroxide, and hydrochloric acid were
obtained from Mallinckrodt (St. Louis, MO). Heptafluorobutyrylimidazole (HFBI), Florox TM (2.5 mg/mL O-(pentafluorobenzyl)hydroxylamine hydrochloride in pyridine), and
heptafluorobutyric acid anhydride (HFAA)were obtained from
Pierce (Rockford, IL). Sephadex LH-20 was obtained from
Pharmacia (Piscataway, NJ).
Stock solutionsand preparation of standard curves
A drug reference solution (1 mg/mL) was diluted in
methanol to obtain mixed working solutions containing 0.1
ng/mL, 1.0 ng/mL, and 10.0 ng/mL of nandrolone and testosterone or nandrolone decanoate and testosterone acetate. Stock
and working solutions were stored at -20~ until use. Daily
standard curves were prepared by fortifying drug-free human
hair with known concentrations of standards. To fortify hair,
drug-free hair was carefully cut into small segments, 20 mg
was placed into silanized glass vials, and the working solution
was added. The concentrations of the standards were 50, 100,
200, and 500 pg/mg and 1, 5, and 10 ng/mg.
Preparation of quality-control samples
Positive quality-control samples (0.5 and 5.0 ng/mg in fortified hair) were prepared daily. Stock solutions used to prepare
quality-control samples were prepared independently from
those used to prepare standards. Drug-free hair was also digested, extracted, and analyzed as a negative control in each
assay.
Nandrolone decanoate and testosteroneacetate
administration and hair collection
Male, hooded, Long-Evans rats (120-150 g) obtained from
Harlan Sprague-Dawley (Indianapolis, IN) were kept in a constant room-temperature environment with an alternating
12-h light-dark cycle with food and water available ad libitum.
Animals were housed individually in hanging wire cages to
prevent contamination from their bedding or the urine or
saliva of other rats. Testosterone acetate or nandrolone decanoate dissolvedin corn oil was administered intraperitoneally
(n = 3) at a dose of 10 mg/kg once per day for five days. On day
0 (prior to dosing) a 1-in. x 1-in. area on the nonpigmented
sides of the animal and the pigmented stripe on the animal's
back were shaved to the skin using an electric animal shaver.
The same areas were again shaved on day 14 and day 28 after
the start of the dosing. Because no steroids or steroid esters
were found in the hair of these rats a higher dose of nandrolone decanoate dissolved in corn oil was administered intraperitoneally (n = 3) at a dose of 60 mg/kg once per day for
10 days over a time period of 14 days. The time interval between dosing and hair shaving was also increased. Hair was collected on days 0, 21, and 35 after the start of the dosing. Hair
was stored at -20~ until analysis. Hair collected from the animals was not washed prior to analysis.
Subject hair samples
Hair samples were donated from self-reported steroid users.
The locks of hair were cut as close as possible to the scalp at the
time of collection. Sample 1 contained brown hair collected
from a 22-year-old white male. This individual received
1.0-1.75 g of testosterone compounds/week and 0.5-0.8 g of
nandrolone decanoate/week for 12 weeks. Samples 2, 3, and 4
were donated from drug users who reported the ingestion of
nandrolone, trenbolone, and/or testosterone up to 2-3 months
prior to sampling. Steroids purchased on the black market
may or may not contain the drug listed on the label. Sample 2
was brown curly hair, sample 3 was black hair, and sample 4
was brown hair.
Digestion and extraction for the detection of nandrolone
and testosterone
Hair (10-50 rag) was carefully cut into small pieces and
mixed thoroughly. Prior to digestion, deuterated internal standards were added to the hair standards and controls (prepared
as described previously) and rat or subject samples. The hair
was completely solubilized (digested) with I mL of 1N NaOH at
70~ for 15 min. After digestion, the tubes were cooled in a
freezer for 10 min, the pH of the samples was adjusted to 6.0
with 1N HCI, and I mL 100raM phosphate buffer (pH 6.0) was
added. Five milliliters of ethyl acetate was then added, and analytes were extracted by mechanical shaking for 30 rain. The
tubes were centrifuged at 2500 rpm for 10 min. The organic
phase was transferred to a silanized tube and evaporated to dryness at 40~ in a water bath under air. The residue was derivatized according to a previously developed method by de Boer
(9) with minor adjustments. First, 50 ~L of Florox reagent
was added. The mixture was then heated overnight at 80~
After cooling to room temperature, 25 lJL of HFBI were added
to the mixture and it was heated for 30 rain at 80~ Excess
derivatization reagents were removed by filtration through a
column of Sephadex LH-20 slurry (chloroform/hexane, 1:1)
packed in a pasteur pipette with a glass bead on the bottom
using chloroform/hexane (1:1) as an eluant. The steroid derivatives were eluted in the first 2 mL of eluant, and the solvent was
removed under a gentle stream of air at 40~ The residue was
dissolved in 50 IJL of iso-octane and ethyl acetate (4:1, v/v) for
GC-MS analysis.
Digestion and extraction for the detection of nandrolone
decanoate and testosterone acetate
Hair (10-50 rag) was carefully cut into small pieces and
mixed thoroughly. Prior to digestion, nandrolone-d3 and testosterone-d3 were added to all the samples as the internal standards for the respective steroid esters. The hair was incubated
overnight with I mL of methanol at 50~ After evaporation of
the methanol and reconstitution in phosphate buffer (pH 6.0),
5 mL of ethyl acetate was added and analytes extracted by mechanical shaking for 30 rain. The tubes were then centrifuged
at 2500 rpm for 10 min. The organic phase was transferred to
a silanized tube and evaporated to dryness at 40~ in a water
bath under air. The residue was derivatized with 200 lJL of
chloroform and 100 lJL of HFAA.The mixture was heated for 30
min at 70~ Excess of derivatization reagents was removed by
417
Journal of Analytical Toxicology, Vol. 23, October 1999
filtration through a Sephadex LH-20 column as described earlier. The residue was dissolved in 50 pL of iso-octane and ethyl
acetate (4:1, v/v) for GC-MS analysis.
GC--MS analysis
Analyseswere performed on a Finnigan-MATrM4500 GC-MS
equipped with INCOS| software (Finnigan MAT,tan Jose, CA).
The chromatographic column was an HP Ultra 1 (crosslinked
methyl siloxane) capillary column (25 m x 0.20-ram i.d., 0.33
pro). The initial column temperature of 180~ was held for
0.1 rain, then programmed to 320~ at the rate of 20~
and held for 5.4 rain at the final temperature. The carrier gas
was helium with a head pressure of approximately 10 psi. Temperatures of the injection port, interface, transfer line, and
ionizer were 275~ 290~ 290~ and 130~ respectively.
Emission current, electron energy, conversion dynode, and
multiplier were set at 0.16 mA, -70 eV, -3 kV, and -1700 V,
respectively.One to two microliters was injected splitless onto
the column.
GC-MS analysisfor nandrolone and testosterone
The MS was operatedin the negative-ionchemical ionization
(NICI) detection mode for the analysis of nandrolone and
testosterone. The reagent gas was methane adjusted to a
source pressure of 0.60 Torr. The MS was operated in the selected ion monitoring (tiM) mode and programmed for detection of m/z 645 (nandrolone-d0),m/z 648 (nandrolone-d3),
m/z 659 (testosterone-d0), and m/z 662 (testosterone-d3).
Peak-height ratios, based upon the ratios of peak to height of
the two isomers to the corresponding internal standards, were
calculated. The concentration was determined from leastsquares regression equations generated from peak-height
ratios of the calibrators. Peak-height ratios were used instead
of area-height ratios for calculations because the correlation
coefficients of the linear curves were slightly better using the
peak-heights.
GC-MS analysisfor nandrolone and testosterone esters
The MS was operated in the positive-ion chemical ionization
(PICI) detection mode for the analysis of the esters. The
reagent gas was methane/ammonia (1:2) adjusted to a source
pressure of 0.60 Torr. The MS was operated in the selected ion
monitoring (tiM) mode and programmed for detection of m/z
642 (nandrolone decanoate), rn/z 491 (nandrolone-d3), rn/z
544 (testosterone acetate), and m/z 505 (testosterone-d3).Peakheight ratios, based upon the ratios of peak-height of each
analyte to the corresponding internal standard were calculated. The concentration was determined from least-squares regression equations generated from peak-height ratios of the
calibrators.
Recovery
Extraction recovery was experimentally determined by
preparing two sets of samples (A and B). Set A consisted of
samples at two different concentrations: 0.5 ng/mg and 5
ng/mg (n = 5). This set was digested and extracted using the
method as described previously. Samples in set B were digested and extracted as described previously, except the com418
pounds of interest were added after evaporation of the organic
solvent, followed by evaporation. The ratio of analyte peak
height to the corresponding internal standard peak height
was calculated, and the mean ratios were determined for samples in sets A and B. The mean ratio for set A was dividedby the
mean ratio for the corresponding concentration in set B to obtain the recovery ratio. The recovery ratio was multiplied by
100 to calculate the percent recovery.
Results and Discussion
Analytical method
The structures of nandrolone and testosterone together with
their PFBO-HFB ester derivatives are shown in Figure 1. ~vo
resolved GC peaks, corresponding to the E- and Z-isomers,
which are characteristic for oxime derivatives, are seen on the
chromatograms (9). Formation of the PFBO derivative was
slow. However,because NICI results in a very sensitiveand specific analytical procedure, this type of derivativewas chosen in
A
.Carl
OH
nandrolone
~~--
9,
C--C~F7
F - ~ CH20-N
F F
PFBO-HFB ester derivative of nandrolone
testosterone
~ ~
R F
F - - ~ CH20--N
F F
9,
-c-C'
3F7
C ~
PFBO-HFB ester derivative of testosterone
Figure 1. Structures of nandrolone and testosterone together with their
PFBO-HFB ester derivatives.
Journal of Analytical Toxicology, Vol. 23, October 1999
order to introduce several electron capture groups. Figure 2A
shows the mass spectrum of the PFBO-HFBderivative of nandrolone. The ion at m/z 645 corresponds to the loss of an HF
molecule. This HF molecule originated from the PFBO group
(9). A similar loss was seen for testosterone with m/z 659
(Figure 2B).
IN, I
I
A
i
~=
,,,
=,
~ - --".'.'.'.'.'.'.'~--- - - - ~ . . . . .
-,~ ......
m/z
.2
Ira.O-
7
~s
tt3 ~w.l~ . . . . . -;. ....... -&/ ....... ~ ....... ~/...
"
B
-~; .......
''''311
m/z
IN,l,
SO,O-
IO.I,
. .2
m/z
. === ~,,= ~,,==~
T
/1=,
.=T.-- m
lid
m/Z
RB
Figure 2. A: Mass spectrum of the PFBO-HFB ester derivative of nandrolone. B: Mass spectrum of the PFBO-HFB ester derivative of testosterone.
CH3
9, /X/X/X/~CH3
Figure 4. A: Mass spectrum of the HFB ester derivative of nandrolone
decanoate. B: Mass spectrum of the HFB ester derivative of testosterone
acetate.
Table I. Intra-assay Accuracy and Precision for the
Analysis of Hair Fortified with Nandrolone, Nandrolone
Deconoate, Testosterone, and Testosterone Acetate
(n = 5)
O- C
Target
concentration
(ng/mg)
C 3F 7-0. O
HFB ester derivative of nandrolone decanoate
,o,
--C --CH 3
C3F7-C" O"
HFB ester derivative of testosterone acetate
Figure 3. Structures of the HFB ester derivatives of nandrolone decanoate and testosterone acetate.
Measured
concentration
(ng/mg)
Nandrolone
0.5
5.0
Testosterone
0.5
5.0
Nandrolone decanoate
0.5
5.0
Testosterone acetate
0.5
5.0
Accuracy
(%)
Precision
(% CV)*
0.5
5.1
97.1
101.7
1.9
3.1
0.5
5.1
103.0
101.7
3.3
4.4
0.6
4.4
116.0
88.2
11.0
3.5
0.4
4.5
83.4
89.6
5.9
3.4
* %CV, Coefficient of variation.
419
Journal of Analytical Toxicology, Vol. 23, October 1999
Figure 3 shows the structures of the formed HFB derivatives
of nandrolone decanoate and testosterone acetate. The mass
spectra of these HFB esters are presented in Figure 4A and 4B.
PICI was used for this analysis because the perfluoro acid anhydrides gave no molecular anion for esters in NICI. The ions
at m/z 642 and 544 correspond to the formation of an ammonia
adduct. The possibility of analyzing TMS enol derivatives was
also studied. However, several main products were formed
with masses close together. Most likely these compounds originated from one product by losing hydrogens. These derivatives
were therefore not suitable for selected ion monitoring.
Peak-height ratios were calculated for each standard and
plotted against the known concentration of the standard. The
limit of quantitation (LOQ) based on 20 mg of hair was established at 50 pg/mg for nandrolone and testosterone, 100 pg/mg
for testosterone acetate, and 200 pg/mg for nandrolone decanoate. Simple linear regression of the standard curves was calculated using Cricket Graph software. Pearson product moment
correlation coefficients were typically 0.99, and the assay was
linear from the LOQ of the individualanalytes to 10 ng/mg hair.
Intra- and interassay precisions of the analytical method
were determined by analyzing two different concentrations, 0.5
Table II. Intra-assay Accuracy and Precision, and
Recovery for the Analysis of Hair Fortified with
Nandrolone, Nandrolone Decanoate, Testosterone, and
Testosterone Acetate (n = 5)
Target
Measured
concentration concentration Accuracy Precision Recovery
(ng/mg)
(n~mg)
Nandrolone
0.5
5.0
Testosterone
0.5
5.0
Nandrolone decanoate
0.5
5.0
Testosterone acetate
0.5
5.0
(%)
(% CV)*
(%)
0.5
5.1
105.3
101.7
9.0
3.4
47.9
46.1
0.5
5.1
100.0
102.4
2.0
1.6
86.9
84.0
0.6
4.9
127.3
98.7
8.0
13.3
87.7
69.0
0.5
5.1
91.3
102.2
10.3
12.2
72.6
72.0
* %CV,Coefficientof variation.
A
N2
B
645
~.-
IS 2
IS 2
100'{)1
,
I050
"r..~;
l
ItO0
0:13
.
9
i
1150
I:S6
.
.
.
.
,
.
.
.
.
1200
,
.
.
t250
1:541
.
i/k
.
Illl)
6 ~l
1300 SIP.AN
9:43 TIME
9:21
1150
1200
0:30
o' t
.
.
D
,
1:1iA
~
.
-
.
",
-,.
,
182
IS2
.
N2
1.9
~
.
7:$1
I~10
[1:41 TIM
Time (mln)
Time (rain)
10001
1250
9:21
tl:fdl
.
i
ICI~
8:N
.
.
.
.
i
~
Time (rain)
.
.
.
.
9
1~21
.
.
.
.
,
0:1:3
.
1060
1:$1
.
.
.
I100
0:13
.
.
.
.
It r
8::14
.
.
i
t~
G:N
1250
I~l
.
.
.
.
J
~
IK~J4
IN8
Time (mln)
Figure 5. Representativeselected ion chromatograms of derivatized extracts of blank rat hair (A), rat hair fortified with 50 pg/mg nandrolone (B), pigmented
rat hair from rat 2 before dosing (day 0) (C), and pigmented rat hair from rat 2 21 days after dosing containing 76 pg/mg nandrolone (D). Peak identification:
IS 1, nandrolone-d3-1; IS 2, nandrolone-d3-2; N 1, nandrolone-do-1; N 2, nandrolone-do-2.
420
Journalof Analytical Toxicology,Vol. 23, October 1999
and 5.0 ng/mg, of fortified hair controls in several batches
(TablesI and II). For the determination of intra-assayprecision,
each concentration was analyzed in replicates of five. For the
determination of interassay precision, samples were analyzed
in triplicate on three separate days. The mean measured concentration for each batch was then used to calculate interassay
Table III. Summary of Nandrolone, Testosterone, and
Nandrolone Decanoate Concentrations Found in Rats
and Human Hair
Nandrolone
Testosterone
Nandrolonedecanoate
Subject
(pg/mg)
(pg/mg)
(ng/mg)
1
2
3
ND
76
63
ND
1.2
0.9
Human
1
54
81
ND
ND
ND
ND
ND
ND
2
3
4
ND
ND
ND
ND
precision. The mean, accuracy, and percent coefficient of variation (%CV) were calculated at each concentration of nandrolone, testosterone, nandrolone decanoate, and testosterone
acetate. The %CV of both intra- and interassay precision experiments was less than 15%. The accuracy was within 15%
for all analytes and concentrations, except for nandrolone
decanoate and testosterone acetate at the 0.5 ng/mg concentrations. At this lowest concentration the accuracy was within
30% for these two analytes.
Using 1N NaOH at 95~ for 10 rain to digest the hair samples, recovery data in fortified hair at low (0.5 ng/mg) and
high (5.0 ng/mg) concentrations were only around 17.0% and
15.4%, respectively,for nandrolone. However,the recovery of
nandrolone could be increased to 47.9% and 46.1%, respectively, if the digestion temperature was lowered to 70~ (Table
II). Testosterone was less susceptible to the digestion temperature, probably because the difference between nandrolone
and testosterone is one methylgroup, which makes testosterone more stable by eliminating the acidic proton to the
carbonyl group.
The digestion solvent was changed for the analysis of the esters to methanol at 50~ instead of 1N NaOH because the esters are unstable in NaOH. After derivatization of the esters
A
T2
659!
v~,,
.
.
.
.
IS 2
.
.
.
IS 2
IS I
662
m
tldO
7"r
"
"
"
'"
l'50t "
11::16
"
.
I
.
1200
.
.
.
B
.
.
.
.
i
12~
-
9CAN
13~
7:51
a:l~
I:~
Time (mln)
T2
C
-
-
,
. . . .
1200
11:54
i
19:~0
~
. . . .
I
1300
9:43 T I M E
Time (mln)
T2
D
659
-~"
182
A
lOO.o]
-. '~.'-" . ~-:~--'i'C-:-~ :~',:"-.-:.~ ~.,"~ -:-~--':
iS 2
IS 1
/
lifo
1~1
1100
tl$
1150
~
1~
11:51
I
I~
11:21
#
13~e 8C, A N
1:.4~
1050
7:51
I1~
1:13
tl~
8~145
1200
1:58
12~0
g:'.21
I~
8CAN
0;4IS T I M E
Time (min)
Time (min)
Figure6. Representativeselected ion chromatograms of derivatized extracts of blank nonpigmented rat hair (A), nonpigmented rat hair fortified with 200 pg/mg
testosterone (B), hair from subject 1 containing 54 pg/mg testosterone (C), and hair from subject 2 containing 81 pg/mg testosterone (D). Peak identification:
IS 1, testosterone-@1; IS 2, testosterone-d3-2; T 1, testosterone-do-1; T 2, testosterone-do-2.
421
Journal of Analytical Toxicology, Vol. 23, October 1999
with HFAA,the excess of defivatization reagent was removed by
filtration through a Sephadex column instead of evaporating
the solvent off at 40~ with air (which is more common). Almost no HFB ester derivatives of nandrolone decanoate and
testosterone acetate were left after evaporating the solvent
without Sephadex. The reason for this is not clear.
and 1.7 pg/mg in female hair (10). Gleixner and Meyer (11) reported a mean value in male humans of 3.8 pg/mg. The LOQ for
testosterone in this assay was 50 pg/mg hair. We therefore were
unable to detect abnormally elevated levels between 3 and 50
pg/mg testosterone with this method. However,this limitation
may be overcome by using more hair. Twoof the hair samples of
the four self-reported steroid users contained testosterone in
concentrations of 54 and 81 pg/mg. Figure 6 shows the selected
ion chromatograms of these two hair samples together with
blank rat hair and fortified rat hair. Kintz et al. (12) reported
testosterone concentrations in the hair of two bodybuildersat 46
and 71 pg/mg, which is in the same range as our findings.
The rat hair samples that tested positive for nandrolone also
contained nandrolone decanoate in concentrations of 0.9 and
1.2 ng/mg, respectively.No nandrolone decanoate was found in
any other samples (Figure 7). The two human hair samples
positive for testosterone were analyzed for testosterone acetate
and were found negative for testosterone acetate. The most
popular testosterone ester compound on the market is testosterone enanthate. Therefore, it is very likely that these steroid
users have taken a compound other than testosterone acetate.
Quantitation of nandrolone, testosterone, and
their esters in hair
Because no nandrolone or nandrolone decanoate was found in
the rat hair samples dosed with 10 mg/kg/day a higher dose of
nandmlone decanoate was given (60 mg/kg/day). Selected ion
chromatograms of derivatized extracts of rat hair from before
dosing (day 0) and 21 days after dosing with nandmlone decanoate together with blank rat hair and fortified rat hair are
shown in Figure 5. Two out of the three rats contained nandrolone in the pigmented hair at a concentration of 63 and 76
pg/mg (Table III). No drug was found in the corresponding nonpigmented hair. No testosterone was found in the rat hair samples. Normal concentrations of testosterone reported to be found
in human hair with high-resolution MS are 2.7 pg/mg in male
A
..; . . . . ~_' "
+"tA
6 4 2 ~ .
9 -~'~_,.: 9 - + - _ i ~ -
-
:
- _~:
:
~-_,_="-'-:~; -
--:
,
~h.,,,..,J~
B
18
.'..
++]
-
, ~..,,.,~, . . . .
..
: .-. ~ ~ ~ , p . . . .
, . . , :"-,
NAD
6421 "" A
141
Time (rain)
+,~
Time (rain)
is
. . . . .
c
~,
.-:..
,
:
.'.
__...--.
;-;+.-.-.
,
.
.
.
.~"~;, . - :
Is
D
+,ot x
. . . .
.~,,
~:;
. .
.-:...+...-.+,
..'.-",
. . ."~-......,
NAD
642
642
Time (rain)
Time (mln)
Figure 7. Representative selected ion chromatograms of derivatized extracts of blank rat hair (A), rat hair fortified with 2.0 ng/mg nandro]one decanoate (B),
pigmented rat hair from rat 2 before dosing (clay 0) (C), and pigmented rat hair from rat 2 21 days after dosing containing 1.2 n~mg nandrolone decanoate
(D). Peak identification: ]5, nandrolone-d3; NAD, nandrolone decanoate.
422
Journal of Analytical Toxicology, Vol. 23, October 1999
Conclusions
Nandrolone and nandrolone decanoate are detected in pigmented rat hair after systemic administration. The concentration of nandrolone decanoate is much higher in rat hair than
nandrolone itself. However, the detection limit for nandmlone
decanoate in this specific assay is higher. Thus, it is difficult to
predict whether nandrolone or its ester would be easier to detect in users. Testosterone has been detected in human hair
samples in concentrations in the same quantitative range that
other researchers have (12). Because the ester assay was developed for testosterone acetate and not for testosterone enanthate (which is more commonly used), the corresponding ester
was not identified.
Acknowledgments
This research was supported by NIDA Grant Nos. DA07280
and DA09096.
References
1. D.H. Catlin, C.K. Hatton, and S.H. Starcevic. Issues in detecting
abuse of xenobiotic anabolic steroids and testosterone by analysis
of athletes' urine. Clin. Chem. 43:1280-1288 (1997).
2. L.D. Bowers and Sanaullah. Direct measurement of steroid sulfate
and glucuronide conjugates with high-performance liquid chromatography-mass spectrometry. J. Chromatogr. B 687:61-68
(1996).
3. C.H.L. Shackleton, A. Phillips, T. Chang, and Y. Li. Confirming
testosterone administration by isotope ratio mass spectrometric
analysis of urinary androstanediol. Steroids 62:379-387
(1997).
4. J.M. Lamer, S.L. Pahuja, C.H. Shackleton, W.J. McMurray,
G. Giordano, and R.B. Hochberg. The isolation and characterization of estradiol-fatty acid esters in human ovarian follicular
fluid. J. Biol. Chem. 268:13893-13899 (1993).
5. W. Borg, C.H.L Shackleton, S.L. Pahuja, and R.B. Hochberg.
Long-lived testosterone esters in the rat. Proc. Natl. Acad. Sci. 92:
1545-1549 (1995).
6. X. de la Torte, J. Segura, A. Polettini, and M. Montagna. Detection
of testosterone esters in human plasma. J. Mass Spectrom. 30:
1393-1404 (1995).
7. C.H.L. Shackleton, H. Chuang, J. Kim, X. de la Torre, and J. Segura.
Electrospray mass spectrometry of testosterone esters: potential
for use in doping control. Steroids 62:523-529 (1997).
8. K.M. HOld, D.G. Wilkins, D.J. Crouch, D.E. Rollins, and R.A. Maes.
Detection of stanozolol in hair by negative ion chemical ionization
mass spectrometry. J. Anal. Toxicol. 20:345-349 (1996).
9. D. de Boer. Profiling anabolic androgens and corticosteroids in
doping analysis. Doctoral thesis, University of Utrecht, Utrecht,
The Netherlands, 1992.
10. C. Scherer, U. Wachter, and S.A. Wudy. Determination of testosterone in human hair by gas chromatography-selected ion monitoring mass spectrometry. Analyst 123:2661-2663 (1998).
11. A. Gleixner and H.H.D. Meyer. Methods to detect anabolics in
hair: use for food hygiene and doping control. Int. Lab. luly:
20-23 (1998).
12. P. Kintz, V. Cirimele, H. Sachs, and B. Ludes. Anabolic steroids in
hair of 2 bodybuilders. Abstract at the SOFT-TIAFT 1998 Meeting,
October 5-9, 1998, Albuquerque, New Mexico.
Manuscript received March 19, 1999;
revision received May 24, 1999.
423