Hair analysis of anabolic steroids in connection with doping control results from horse samples

JOURNAL OF MASS SPECTROMETRY J. Mass Spectrom. 2008; 43: 1001 1008 Published online 19 June 2008 in Wiley InterScience (www.interscience.wiley.com).1446 Hair analysis of anabolic steroids in connection with doping control results from horse samples P. Anielski Institute of Doping Analysis and Sports Biochemistry, Dresdner Strasse 12, D-01731 Kreischa, Germany Received 11 April 2008; Accepted 9 May 2008 Doping control of anabolic substances is normally carried out with urine samples taken from athletes and horses. Investigation of alternative specimens, e.g. hair samples, is restricted to special cases, but can also be worthwhile, in addition to urine analysis. Moreover, hair material is preferred in cases of limited availability or complicated collection of urine samples, e.g. from horses. In this work, possible ways of interpretation of analytical results in hair samples are discussed and illustrated by practical experiences. The results demonstrate the applicability of hair analysis to detect anabolic steroids and also to obtain further information about previous abuse. Moreover, the process of incorporation of steroids into hairs is described and the consequences on interpretation are discussed, e.g. on the retrospective estimation of the application date. The chosen examples deal with the detection of the anabolic agent testosterone propionate. Hair samples of an application study, as well as a control sample taken from a racing horse, were referred to. Hair material was investigated by a screening procedure including testosterone, nandrolone and several esters (testosterone propionate, phenylpropionate, decanoate,undecanoate, cypionate; nandrolone decanoate, dodecanoate and phenylpropionate; limits of detection (LODs) between 0.1 and 5.0 pg/mg). Confirmation of testosterone propionate (LOD 0.1 pg/mg) was carried out by an optimised sample preparation. Trimethylsilyl (TMS) and tert-butyl dimethylsilyl derivatives were detected by gas chromatography high-resolution mass spectrometry (GC-HRMS) and gas chromatography tandem mass spectrometry (GC-MS/MS). Copyright 2008 John Wiley & Sons, Ltd. KEYWORDS: hair analysis; anabolic steroids; testosterone propionate; horse; long-term detection; doping control INTRODUCTION Anabolic steroids are administered with the intention to enhance muscle growth and to improve performance in sports. According to The 2008 Prohibited List of the World Anti-Doping Code (WADA), 1 anabolic androgenic steroids belong to the class S1.1 and are prohibited in and out of competition. In general, doping control ofathletesand racing horses is carried out by taking urine and blood samples. Furthermore, potential abuse has been suspected in horses selected for breeding purposes. Beside the performance of the horse, the physical appearance is assessed by a selection committee, e.g. during stallion licensing. Several horse-breeder associations have been interested in a feasible procedure to control anabolic steroids which are suspected to be the mainly abused compounds. Hair material was deemed to be a suitable specimen due to quick sampling, easy storage conditions and supposed analytical preferences. Apparent advantages often associated with hair analysis are the possibilities of long-term detection and the retrospective estimation of the application date. This idealised situation Ł Correspondence to: P. Anielski, Institute of Doping Analysis and Sports Biochemistry, Dresdner Strasse 12, D-01731 Kreischa, Germany. E-mail: p.anielski@idas-kreischa.de is limited to selected analytes and to certain issues. Several authors have already discussed reasons why hair samples are not suitable to replace urine analyses for doping control. 2 6 Nevertheless, hair specimen may provide complementary information in addition to urine or blood samples. 7 9 For example, in suspicious cases the detection of the administered parent compound or the extension of detection times is worthwhile. Therefore, a method was established to detect several anabolics in hair samples. Substances included in the screening are clenbuterol, nandrolone, testosterone and eight different esters available as veterinary drugs. 10 In contrast to the facile sampling a demanding analytical procedure is required. Anabolic steroids are incorporated into hair material in a comparatively low amount, and matrix interferences may hinder detection. Following a single application, anabolic steroids have been reported to be undetectable in hair samples several times. 11 14 To achieve appropriate detection limits, extensive sample purification is necessary, e.g. carried out by solid-phase extraction and highperformance liquid chromatography (HPLC). Furthermore, sensitive detection techniques have to be applied, including gas chromatography high-resolution mass spectrometry (GC-HRMS), gas chromatography tandem mass spectrometry (GC-MS/MS) and liquid chromatography tandem mass spectrometry (LC-MS/MS). 10,12,13,15 17 Copyright 2008 John Wiley & Sons, Ltd.

1002 P. Anielski However, the suitability and validity of the method is mainly affected by substance-specific properties. The process of incorporation is strongly associated with the polarity of the analyte and has to be taken into consideration for an exact interpretation of the results. Furthermore, the concentrations detectable in hair depend on several parameters like pharmacokinetic, bioavailability, dosage, cosmetic treatment and individual variations. 18 20 In this work, possible ways of interpretation are discussed and illustrated by practical experiences and analytical results in hair. The chosen examples deal with the detection of the anabolic compound testosterone propionate. Horsehair samples originated from an application study as well as from a racing horse, which was controlled for anabolic substances. Regularly, two segments of each hair sample were analysed (hair length 0 6 and 6 12 cm, measured from roots) and to begin with, processed according to the screening procedure. In case of a suspicious finding the confirmation procedure was subsequently applied and, additionally, the corresponding wash solutions were analysed. MATERIALS AND METHODS Standards, chemicals and reagents The following reference compounds were used: testosterone, testosterone propionate, nandrolone (Jenapharm, Germany); testosterone phenylpropionate, decanoate, cypionate (Sigma-Aldrich, Germany); testosterone undecanoate (NARL, Australia); nandrolone decanoate, dodecanoate, phenylpropionate (Steraloids, USA). Testosterone-d 3 (NARL, Australia), metenolone enanthate (Schering, Germany) and nandrolone phenylpropionate (Steraloids, USA) were added as internal standards (ISs), either for screening or confirmation analysis. Solvents used for extraction and liquid chromatography were methanol (KMF, Germany), n-pentane (freshly distilled, Fisher Scientific, UK), acetonitrile and water (gradient grade, Fisher Scientific, UK). The derivatisation reagent 1 consists of 1 ml of N-methyl- N-trimethylsilyl-trifluoroacetamide (MSTFA; Macherey- Nagel, Germany), 5 mg of ammonium iodide and 2 µl of propanethiol (Merck, Germany). Reagent 2 was prepared with 1 ml N-methyl-N-tert-butyldimethylsilyltrifluoroacetamide (MBDSTFA; Macherey-Nagel, Germany), 2mgiodineand20µl propanethiol (Merck, Germany). An aqueous buffer solution (ph 9) was applied, containing sodium bicarbonate and potassium carbonate (KMF, Germany) at a concentration of 1.2 M each. Hair specimen Blank hair samples were collected from eight untreated geldings, pulverised and pooled. Control samples containing testosterone propionate were prepared using blank material spiked to a concentration of 1 and 2 pg/mg. Due to the lack of reference hair material, the relevant amount of testosterone propionate was added separately to each weighted aliquot of the blank specimen. Positive hair material originated from three stallions treated with testosterone propionate (i.m. 4 ð 60 mg, at intervals of 3 weeks). Hair samples were collected from the tail 4 weeks after the fourth injection. The test specimen originated from a gelding and was taken from the mane of the horse. Sample preparation Hair screening procedure For analysis, a strand of the horsehair is usually cut into segments of 6 cm length, measured from the roots. Two sections were investigated of each sample (0 6 and 6 12 cm). The segments were decontaminated using 5 ml of a solution of methanol/water (1 : 1, v/v). The samples were shaken for 1 min utilising an overhead shaker. The hair segments were removed from the wash solution which was stored at 5 C until further preparation. After drying under a stream of nitrogen at 80 C, the hair samples were pulverised using a swing mill (Retsch MM301). One hundred milligrams of the hair powder were weighed and the ISs added (1 ng testosterone-d 3, 10 ng metenolone enanthate). Extraction was carried out by incubation with 2.5 ml of methanol in an ultrasonic bath (4 h, 50 C). The methanol layer was separated and evaporated to dryness under nitrogen at 80 C. The residue was dissolved in 0.5 ml of the buffer solution (ph 9) and extracted twice with 2.5 ml n- pentane/methanol (25 : 1, v/v). The combined extracts were dried under nitrogen at 55 C and reconstituted with 16 µl of acetonitrile and 16 µl of water. Further purification was carried out by HPLC, utilising a reversed phase column (Zorbax XDB-C18, 3.0mmð 150 mm, 5 µm). Water (A) and acetonitrile (B) were applied as mobile phases at a flow rate of 0.6 ml/min. A linear gradient was employed starting at 20% B (held from 0 to 1 min) up to 100% B at 11 min (100% B held from 11 to 20 min). Injection volume was 25 µlandthree different fractions were collected: 6.0 to 8.0 min (containing nandrolone, testosterone), 10.8 to 17.3 min (testosterone propionate, phenylpropionate, decanoate, cypionate; nandrolone decanoate, phenylpropionate) and 17.3 to 19.5 min (testosterone undecanoate, nandrolone dodecanoate). After evaporation to dryness (nitrogen, 80 C), samples were derivatised with 25 µl of reagent 1 at 55 C for 25min. Analyses of the formed TMS derivatives were carried out by GC-HRMS. Confirmation of testosterone propionate Samples were prepared in a manner similar to the screening procedure, except for the following modifications. All samples were prepared in duplicate, and nandrolone phenylpropionate (5 ng) was added as IS to each hair sample. HPLC fractions containing testosterone propionate and nandrolone phenylpropionate (IS) were collected from 10.6 to 12.5 min. One batch was derivatised with reagent 1 to form TMS derivatives and the other samples were converted to tert-butyl dimethylsilyl derivatives by addition of 25 µl of reagent 2. Wash solutions An aliquot of 1 ml was evaporated to dryness (nitrogen, 80 C), reconstituted using 0.5 ml of the buffer solution (ph 9) and extracted twice with 2.5 ml n-pentane/methanol (25 : 1, v/v). After evaporation of the organic phase under nitrogen

Hair analysis of anabolic steroids 1003 at 55 C, the dry residue was derivatised with reagent 1 (25 µl) andmeasuredbygc-hrms. Instrumental parameters and detection Analyses were carried out utilising a GC-HRMS system (gas chromatograph HP 5890; mass spectrometer AutoSpec Ultima Q, Fisons). A 12.0-m column (VF-1MS, 0.2 mm ID ð 0.33 µm, Varian) was used for GC; temperature programme: 150 C (held 0.5 min), 12.5 C/min to 325 C (held 3 min); carrier gas helium (constant pressure ¾19 psi; flow 0.8 1.5 ml/min). Injection volume was 1 µl on column (splitless mode). MS parameters: electron energy 35 ev, trap current 500 µa, maximum acceleration voltage 8000 V. Figure 1. EI mass spectrum, structure and characteristic fragmentations of testosterone propionate; (a) trimethylsilyl derivative, (b) tert-butyl dimethylsilyl derivative.

1004 P. Anielski The following ions of the screening method were used in the study selected ion recording (SIR): m/z 432.288 (testosterone), m/z 435.3115 (testosterone-d 3, IS), m/z 416.2747 (testosterone propionate), m/z 486.3529 (metenolone enanthate, IS). Limits of detection (LODs) were 0.3 pg/mg (testosterone) and 0.1 pg/mg (testosterone propionate). For confirmation, TMS derivatives were analysed using GC-HRMS and the following ions were monitored: m/z 416.2747, 401.2512 (testosterone propionate; fragmentation and spectrum, see Fig. 1(a)) and m/z 478.2903 (IS). Detected ions of the tert-butyl dimethylsilyl derivatives were m/z 458.3216, 443.2982 (testosterone propionate; see Fig. 1(b)) and m/z 445.2958 (IS). Additionally, TMS derivatives were measured by GC-MS/MS (gas chromatograph CP-3800, triple quadrupole mass spectrometer 320-MS, Varian). LOD of testosterone propionate (mono-tms) was 10 pg/mg, detecting the following reactions: m/z 416.3! 401.3, 416.3! 209.1, 416.3! 247.2 (collision energy 15 ev). Validation of the method was carried out using negative (hair blank) and positive control samples (spiked at three concentration levels: 1.0, 2.0, 5.0 pg/mg) that were processed in triplicate on 3 different days. Recovery (testosterone propionate 59 74%) was calculated by comparison of the control samples (5 pg/mg) and the response of the corresponding amount (20 pg of testosterone propionate), injected to GC-HRMS without prior sample preparation. Calibration was carried out within a range of 0 10.0 pg/mg and turned out to be linear (correlation coefficient 0.98 0.99). The LOD was estimated at a signal-to-noise ratio greater than 3. RESULTS AND DISCUSSION Long-term detection in hair samples compared to urine Doping control of horses in sports is regularly carried out in urine and blood samples. To analyse testosterone in urine samples, thresholds of 20 ng/ml (geldings) and 55 ng/ml (fillies and mares) have been established (FEI, Equine Prohibited List 21 ). The maximum detection window of an abuse of testosterone depends on different parameters, e.g. on the applied substance and the application route. Several preparations for intra-muscular injection are available to achieve a long-term anabolic effect. For this purpose, steroids are esterified with different fatty acids, e.g. propionic acid. Release and excretion of the steroid are delayed with increasing chain length of the fatty acid. 22 24 Therefore detection in urine might be possible up to several weeks or even months. Compared to other esters, testosterone propionate is a relatively short-acting compound and hence can be eliminated more rapidly. In contrast to the analysis of urine and blood, investigation of hair samples enables a longer detection time as demonstratedby the followingcase. The analyses of anabolic steroids in several horsehair samples were requested, due to the suspicion of the abuse of testosterone propionate months before the sample collection. This presumption resulted from a positive testosterone finding in urine from another horse that belonged to the same stable and had been coached by the same person. Actually, in one of the collected hair samples, the presence of the synthetic anabolic substance testosterone propionate was confirmed. Both of the investigated 6-cm hair segments tested positive (Figs 2 and 3). The finding indicates that testosterone propionate had been administered within a period of 6 months before sampling, if a growth rate of about 2 cm per month is provided for (discussed below). The results demonstrate the applicability of the method to achieve a longer detection period. In addition to doping control in urine samples, further information about the applied parent compound is available. However, the detection of steroids is time-limited and depends on the retention and stability in hair material. For example, testosterone and testosterone esters are incorporated into hair due to their lipophilic properties, but are not linked to the pigment melanin. Therefore, the compounds can be washed out, resulting in a continuous decrease of the detectable amount in hair. The loss in concentration is intensified by structural damages of the hair fibre, cosmetic treatment (bleaching, dyeing, permanent waving) or extensive hair washings. 19,25,26 Detection of the applied parent compound instead of metabolites In contrast to urine samples, analysis of hair specimen is focussed on the administered parent compounds. Metabolites and conjugates detectable in urine after application of anabolic steroids are usually not incorporated into the hair material in sufficient amounts, due to their high polarity. Instead, the incorporation of non-polar or basic molecules preferably occur. 2,15,27 30 Following an administration of anabolic steroids, the detection of lipophilic parent compounds (e.g. intact esters) or the released anabolic agent (e.g. testosterone) should be possible in hair samples. Therefore, analysis of hair specimen is worthwhile, if the identification of the applied substance is requested. Concerning the case described, the application of the synthetic testosterone propionate was ascertained by detection of the intact ester in the hair sample of the horse. Alternatively, investigation of the released anabolic active agent could be suitable, e.g. testosterone or nandrolone. In cases of an exogenous origin of the steroid, a positive finding in hair would indicate an illegal application, e.g. nandrolone in geldings. Otherwise, a quantitative analysis and interpretation of the concentrations in hair are required. Due to a lack of thresholds for endogenous steroids in hair specimens, e.g. testosterone, the results may only provide additional information and are not suitable to confirm a forbidden administration. Detection of the intact steroid ester should therefore be preferred and would prove the synthetic origin of the applied parent compound without any doubt. Boyer et al. analysed testosterone in hair samples taken from geldings after an oral application of testosterone propionate, but the intact ester was undetectable by GC- MS/MS. 14 Due to the investigation of different hair lengths, the results are not comparable to our study in detail: Boyer et al. investigated only segments which had grown during the period of treatment; in our study more distal sections were analysed (results are discussed below).

Hair analysis of anabolic steroids 1005 Nielen et al. used LC-MS/MS to investigate coat hair samples of bovines. Several steroid esters were detectable following single and repeated administrations, with the exception of testosterone propionate. 13 The authors suggest that the propionate was probably not incorporated or the given dose was too low. In contrast to the referred publications, in our study testosterone propionate was identified as the intact ester in hair samples. Additionally, testosterone concentrations were quantified but no significant increase was obtained in the stallions hair after application of testosterone propionate. Statements regarding the application time In order to acquire information about the application date, the hair strand is divided into sections of definite length prior to analysis. With regard to the known growth rate of hairs, the position of a positive segment could be related to the time of exposure. For the correct interpretation of analytical findings, the mechanism of incorporation too has to be taken into consideration. 19 After administration, a substance can be integrated into hair material by different routes: via Figure 2. Detection of testosterone propionate in hair samples as mono-tms derivative (GC-HRMS, SIR; retention time 11 : 13 min; internal standard 15 : 29 min). Chromatograms of a blank hair sample, a control sample spiked to a concentration of 1.0 pg/mg and two segments of a suspicious hair sample of length 0 6 and 6 12 cm (normalised to ion m/z 416).

1006 P. Anielski Figure 3. Detection of testosterone propionate in hair samples as mono-(tert-butyl dimethylsilyl) derivative (GC-HRMS, SIR; retention time 12 : 50 min; internal standard 13 : 17 min). Chromatograms of a blank hair sample, a control sample spiked to a concentration of 1.0 pg/mg and two segments of a suspicious hair sample of length 0 6 and 6 12 cm. bloodstream into the root during hair growth or via sweat and sebum excreted at the surface of the skin. The exuded compounds can then diffuse into the hair fibre and, therefore, can be incorporated in the distal regions of the hair strand. Following the application of nandrolone prohormones, an extensive excretion of 19-norsteroids via sweat was obtained. 10 This process was identified to be the major mechanism of incorporation of steroids into hair material. The compounds were found not only in the growing segment, but also in the distal sections of the hair sample. Regarding the results of this study, the deposition via blood into the hair roots seems to be negligible. So far a similar effect has not yet been reported following the administration of steroid esters. During a study, three stallions were injected with testosterone propionate (i.m. 4 ð 60 mg, at intervals of 3 weeks). Tail hair samples were taken 4 weeks after treatment and analyses were carried out in two 6-cm segments. The investigated total hair length represents a growing period of the 6 months that passed before sampling,

Hair analysis of anabolic steroids 1007 Table 1. Concentrations of testosterone propionate in horsehair after repeated applications (i.m. 4 ð 60 mg, at intervals of 3 weeks); samples were taken 4 weeks after the last injection (equivalent to a period of 12 weeks after beginning of the study) Horse Segment 1: hair length 0 6 cm Concentration (pg/mg) Segment 2: hair length 6 12 cm 1 0.2 0.5 2 0.1 0.2 3 0.1 0.5 because mane and tail hair of horses show a permanent growth of about 2 cm per month. 31 33 Therefore, the applied testosterone propionate should only be detectable in the proximal segment of the hair samples (0 6 cm). Instead, the intact ester has been identified in both segments and the highest concentrations were measured in section 6 12 cm (up to 0.5 pg/mg; see Table 1). This segment clearly does not correspond to the hair length grown during the period of treatment, but however would represent a period of 3 months before the first application. The results also demonstrate that intact steroid esters can be excreted by sweat and sebum and thus are integrated in the distal sections of the hair. Similar results have been obtained in all horses included in the study, indicating that the mechanism is reproducible and not caused by an individual anomaly or an accidental occurrence. It is likely that the process is pre-destined for testosterone propionate compared to long-chain esters, due to the higher polarity of the propionate. Hair samples of the treated stallions have also been analysed for testosterone, but have resulted in negative findings. Additionally, investigations of the corresponding wash solutions gave no indications of an external contamination of the hair material. As shown in the study, the exact application date of anabolic steroids cannot be ascertained by the position of the positive section, solely with regard to the growth rate of hairs. That is why the time of incorporation of lipophilic compounds into hair does not necessarily correspond to the period of growing. Therefore, an estimation of the application time should be made carefully. In particular, the findings in hair might lead to misinterpretations concerning the beginning of a medication. Without a detailed knowledge of the incorporation process, the administration of testosterone propionate would have been dated 3 months too early in the study. Nevertheless, to control periods of abstinence, hair analysis is suitable and would be unaffected by the excretion of steroids by sweat. CONCLUSIONS The results presented in this paper demonstrate the applicability of hair analysis to detect anabolic steroids, but also point to the limitations of the procedure. Hair material is characterised by a quick sampling procedure and uncomplicated conditions for transport and storage. In particular, if samples are taken from horses these practical conditions play an important role. Another advantage is the investigation of the unchanged parent compound, which is incorporated into hair following an application. For example, after the injection of testosterone propionate, only typical metabolites would be detectable in urine. The identification of the intact ester in the hair sample could substantiate the urine finding and would verify the exogenous origin of the applied steroid. Thus, further information could be obtained by testing hair material in addition to doping control in urine. Interpretation of hair analyses concerning the date of a steroid application depends on certain issues. With regard to the growth rate of hairs, periods without treatment can be investigated, e.g. testing terms of abstinence. Due to the incorporation of steroids via sweat and sebum, it is not possible to verify an exact application date and the beginning of a repeated abuse, respectively. Furthermore, the analysis of hair samples could enable an extension of the detection period compared to urine. Even though further investigations indicated a loss of steroids from hair material with increasing time, detection was possible up to several months. Acknowledgements This work was partially financed by the German Federal Ministry of the Interior (IIA1-2506DL0001). The application study was carried out by the University of Veterinary Medicine Hannover, Clinic for Horses. The author thanks B. Zingrebe and Prof. E. Klug for providing the hair samples. REFERENCES 1. World Anti-Doping Agency. The 2008 prohibited list international standard, 2007; 1. 2. Thieme D, Grosse J, Sachs H, Mueller R. Analytical strategy for detecting doping agents in hair. Forensic Science International 2000; 107: 335. 3. Rivier L. Is there a place for hair analysis in doping controls? Forensic Science International 2000; 107: 309. 4. Midio A, de Moraes Moreau R, Silva O. The possibilities of hair analysis in the determination of involuntary doping in sports. Sports Medicine 2001; 31: 321. 5. Kintz P, Villain M, Cirimele V. Hair analysis for drug detection. Therapeutic Drug Monitoring 2006; 28: 442. 6. Gambelunghe C, Sommavilla M, Ferranti C, Rossi R, Aroni K, Manes N, Bacci M. Analysis of anabolic steroids in hair by GC/MS/MS. Biomedical Chromatography 2007; 21: 369. 7. Gaillard Y, Vayssette F, Pepin G. Compared interest between hair analysis and urinalysis in doping controls. Results for amphetamines, corticosteroids and anabolic steroids in racing cyclists. Forensic Science International 2000; 107: 361. 8. Dumestre-Toulet V, Cirimele V, Ludes B, Gromb S, Kintz P. Hair analysis of seven bodybuilders for anabolic steroids, ephedrine, and clenbuterol. Journal of Forensic Sciences 2002; 47: 211. 9. Kintz P. Testing for anabolic steroids in hair: a review. Legal Medicine (Tokyo, Japan) 2003; 5(suppl1): S29. 10. Anielski P, Thieme D, Schlupp A, Grosse J, Ellendorff F, Mueller RK. Detection of testosterone, nandrolone and precursors in horse hair. Analytical and Bioanalytical Chemistry 2005; 383: 903. 11. Höld KM, Borges CR, Wilkins DG, Rollins DE, Joseph RE Jr. Detection of nandrolone, testosterone, and their esters in rat and human hair samples. Journal of Analytical Toxicology 1999; 23: 416.

1008 P. Anielski 12. Segura J, Pichini S, Peng S, de la Torre X. Hair analysis and detectability of single dose administration of androgenic steroid esters. Forensic Science International 2000; 107: 347. 13. Nielen M, Lasaroms J, Mulder P, Van Hende J, van Rhijn J, Groot M. Multi residue screening of intact testosterone esters and boldenone undecylenate in bovine hair using liquid chromatography electrospray tandem mass spectrometry. Journal of Chromatography B 2006; 830: 126. 14. Boyer S, Garcia P, Popot M, Steiner V, Lesieur M. Detection of testosterone propionate administration in horse hair samples. Journal of Chromatography B 2007; 852: 684. 15. Thieme D, Grosse J, Sachs H, Mueller RK. Detection of several anabolic steroids of abuse in human hair. 16th Cologne workshop on dope analysis. Recent Advances in Doping Analysis 1998; 6: 9. 16. Kintz P, Cirimele V, Jeanneau T, Ludes B. Identification of testosterone and testosterone esters in human hair. Journal of Analytical Toxicology 1999; 23: 352. 17. Gaillard Y, Vayssette F, Balland A, Pepin G. Gas chromatographic-tandem mass spectrometric determination of anabolic steroids and their esters in hair. Application in doping control and meat quality control. Journal of Chromatography B 1999; 735: 189. 18. Balikova M. Hair analysis for drugs of abuse. Plausibility of interpretation. Biomedical Papers of the Medical Faculty of the University Palacky, Olomouc, Czechoslovakia 2005; 149: 199. 19. Pragst F, Balikova M. State of the art in hair analysis for detection of drug and alcohol abuse. Clinica Chimica Acta 2006; 370: 17. 20. Gratacós-Cubarsí M, Castellari M, Valero A, García-Regueiro J. Hair analysis for veterinary drug monitoring in livestock production. Journal of Chromatography B 2006; 834: 14. 21. Fédération Equestre Internationale. Equine prohibited list veterinary regulations, 2006; 31. 22. Junkmann K. Long-acting steroids in reproduction. Recent Progress in Hormone Research 1957; 13: 389. 23. Nieschlag E, Cüppers HJ, Wiegelmann W, Wickings EJ. Bioavailability and LH-suppressing effect of different testosterone preparations in normal and hypogonadal men. Hormone Research 1976; 7: 138. 24. van der Vies J. Pharmacokinetics of anabolic steroids. Wien Med Wochenschr 1993; 143: 366. 25. Cirimele V, Kintz P, Mangin P. Drug concentrations in human hair after bleaching. Journal of Analytical Toxicology 1995; 19: 331. 26. Jurado C, Kintz P, Menéndez M, Repetto M. Influence of the cosmetic treatment of hair on drug testing. International Journal of Legal Medicine 1997; 110: 159. 27. Henderson G. Mechanisms of drug incorporation into hair. Forensic Science International 1993; 63: 19. 28. Nakahara Y, Takahashi K, Kikura R. Hair analysis for drugs of abuse. X. Effect of physicochemical properties of drugs on the incorporation rates into hair. Biological and Pharmaceutical Bulletin 1995; 18: 1223. 29. Deng X, Kurosu A, Pounder D. Detection of anabolic steroids in head hair. Journal of Forensic Sciences 1999; 44: 343. 30. Kintz P, Cirimele V, Sachs H, Jeanneau T, Ludes B. Testing for anabolic steroids in hair from two bodybuilders. Forensic Science International 1999; 101: 209. 31. Popot M, Boyer S, Maciejewski P, Garcia P, Dehennin L, Bonnaire Y. Approaches to the detection of drugs in horse hair. Proceedings of the 13th International Conference of Racing Analysts and Veterinarians, Cambridge, 2000; 115. 32. SchluppA, Anielski P, Thieme D, MuellerR, MeyerH, Ellendorff F. The beta-agonist clenbuterol in mane and tail hair of horses. Equine Veterinary Journal 2004; 36: 118. 33. Dunnett M, Lees P. Hair analysis as a novel investigative tool for the detection of historical drug use/misuse in the horse: a pilot study. Equine Veterinary Journal 2004; 36: 113.