Prednisone Concentrations in Human Hair

27 downloads 0 Views 214KB Size Report
Prednisone Concentrations in Human Hair. Vincent Cirimele 1, Pascal Kintz TM, Jean Pierre Goull~ 2, and Bertrand tudes 1. 11nstitut de M~decine L~gale, ...
Journal of Analytical Toxicology, Vol. 26, March 2002

Technical Note

Prednisone Concentrations in Human Hair Vincent Cirimele 1, Pascal Kintz TM, Jean Pierre Goull~ 2, and Bertrand tudes 1 11nstitutde M~decine L~gale, Strasbourg, Franceand 2Laboratoire de Pharmacologieet de Toxicologie, Le Havre, France

Introduction In 1999, France passed a law against doping, the so-called Loi Buffet. This law was designated to protect athletes against doping consequences, particularly after long-term abuse. In 2001, this law was implemented with its practical aspects. Among them, hair sampling to document chronic exposure was considered as a valid procedure. Drug administration can be tracked by hair analysis for months or even years and thus, offers the possibility of determinating long-term drug exposure (1). Corticosteroids are known to have numerous side effects. During the treatment of specific deseases, these compounds are allowed in urine for a short period of time, but in the case of doping practices, it may be of interest to evaluate chronic administration of corticosteroids. Scientific literature contains only a few reports on the identification of corticosteroids in hair (2-4), but these substances can be added to the list of xenobiotics incorporated into hair. However, the interpretation of the results with unknown dosages of corticosteroids appears very difficult. Among the problems associated with the use of hair in doping control, the unknown dose-concentration relationship can lead to controversies about the potential applications of the specimen (5). To evaluate the existance of a such a correlation, this article reports the identification and quantitation of prednisone in human hair of patients treated with Cortancyl|

Experimental Hair specimens were obtained from 10 patients treated with prednisone. A hair strand was cut with scissors as close as possible to the skin in the vertex posterior region and stored at room temperature. Before extraction, hair strands were washed twice (2 rain each) in 5 mL methylene chloride at room temperature. The hair section was choosen in accordance with the treatment period duration and then pulverized in a ball mill. Prednisone was tested by our previous procedure (4) with some modifications. Briefly, 100 mg of the powdered hair was incubated in I mL Soerensen buffer, pH 7.6 for 16 h at 40~ in * Author to whom correspondenceshould be addressed.E-mail: [email protected].

110

presence of 50 ng cortisol-d3 as internal standard. For further purification, SPE C18 Isolute extraction columns were used. Column activation was performed with 3 mL methanol (MeOH) followed by 3 mL deionized water. The incubation medium was centrifuged, and the supernatant was removed and added to the activated column, then rinsed with 1 mL acetone/deionized water (2:8, v/v), followed by I mL of deionized water, and finally with 1 mL of hexane. Columns were dried for 30 rain, and the corticosteroids were eluted with three successive volumes of MeOH (0.5 mL each). The eluates were evaporated to dryness and resuspended in 1 mL 0.2N NaOH and re-extracted with 3 mL of diethylether. After agitation and centrifugation, the organic phase was removed and evaporated to dryness, and the dry extract was resuspended in 30 pL of MeOH. A 2-pL volume of the extract was injected onto the column (4-pm Novapak C18 Waters, 150 x 2.0-mm i.d.) protected by a 5-pro Opti-Guard C18 fingertight guard cartridge (15 x 1.0ram i.d.). Each 10-min chromatographic run was carried out with a binary mobile phase of acetonitrile and 2raM NH4COOH pH 3.0 buffer, using a linear gradient (acetonitrile 20 to 60% in 4 rain) generated by a 20-mL dual-syringe high-performance liquid chromatography (HPLC) pump (Applied Biosystems model 140B). The flow rate was 200 pL/min with a postcolumn split of 1:3 (flow rate infused into the ionspray: 50 pL/min). Detection was carried out by a PerkinElmer Sciex API-100 mass spectrometer (MS). Nitrogen (purity grade 99.95%) was employed as nebulizing gas. The instrument was operated in the positive ionization mode (ionspray +4500V). Ions generated in the ion source were sampled into the mass analyzer through a 25-pro orifice held at +20V. MS data were recorded in the single ion monitoring mode. Prednisone was identified on the basis of its retention time and ion ratio (m/z 359 and 341). Quantitative results were obtained after determination of the response factor of prednisone (m/z 359) against cortisol-d3 (mlz 366). Because of the lack of reference material (control positive hair samples for prednisone), validation parameters were determined on spiked hair specimens. Standard calibration curves were obtained by adding 2, 5, 25, 100, and 200 ng of prednisone to 100 mg of pulverized blank control hair (prednisone free) and 50 ng cortisol-d3 to obtain final concentrations of 20, 50, 250, 1000, and 2000 pg/mg of hair. Extraction

Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.

Journal of Analytical Toxicology, Vol. 26, March 2002

recoverywas determined by adding 25 ng of prednisone to 100 mg of powderedblank control hair (prednisonefree) and 50 ng of cortisol-d3 (n = 3) to obtain a final concentration of 250 pg/mg. Precision was determined by adding 25 ng of prednisone to 100 mg of powdered blank control hair (prednisone free) and 50 ng of cortisol-d3 (n -- 8), corresponding to a final concentration of 250 pg/mg. The limit of detection was determined by decreasing the prednisone concentration in order to obtain a signal-to-noise ratio of 2. All these validation parameters were determined after extraction of the spiked blank

control hair by the established procedure, that is, HPLC separation of the analytes and MS detection.

Results and Discussion

Under the determined analytical conditions, no interference with prednisone and the internal standard was observed by any extractable endogenous materials present in hair. The correlation coefficientof the calibration curve was 0.998 for prednisone, indicating linearity between 20 and 2000 pg/mg. PreTable I. Analytical Results for Prednisone in the Hair of cision at 250 pg/mg (CVvalues) was 8% for prednisone with a Treated Patients relative extraction recovery of 70%. The limit of detection was 15 pg/mg for prednisone, but the limit of quantitation was Tested Prednisone fixed at 30 pg/mg. Patient Dosage segment concentration Hair specimens were obtained from 10 patients treated with prednisone with doses ranging from 5 to 60 mg/day.Analytical I* 5 mg/day 0 to I 0 cm none detected for 24 months results are reported in Table I. Prednisone was identified in 2 5 mglday 0 to 13 cm 30 pglmg the hair of nine patients. The concentrations ranged from 30 to for 72 months 130 pg/mg. The active metabolite prednisolone was never de3 5 mg/day 0 to 5 cm 39 pg/mg tected, which is consistent with the fact that the major comfor 12 months pound found in hair is the parent drug. Although the limited 4 10 mg/day 0 to 10 cm 30 pg/mg number of subjects precludes generalization, results are sugfor 8 months gestive of a low incorporation rate of prednisone in human 5 10 mg/day 0 to 8 cm 57 pg/mg hair. For dosages ranging from 5 to 60 mg/day, prednisone > 30 months concentrations were always largely under 1 ng/mg. The only 6 20 to 6 mg/day 0 to 8 cm 45 pg/mg negative hair sample was obtained from a woman with lightfor 9 months colored hair dyed black and treated with prednisone at the 7 55 to 9 mg/day 0 to 4 cm 63 pg/mg for 9 months lowest dosage (5 rag/day). This was certainly due to the col8 40 to 35 mg/day 0 to 1 cm 103 pg/mg oration of her hair, as it is well established that cosmetic treatfor 1 months ments such as coloration, bleaching, dyeing,and waving lead to 9 50 to 25 mg/day 0 to 5 cm 130 pg/mg the loss of xenobiotic content of the hair (6). for 3 months The developedprocedure was able to detect prednisone in the 10 60 to 35 mg/day 0 to 6 cm 90 pg/mg hair of patients treated with the lowest dose of prednisone, for 5 months which is 5 mg/day. When the total amount of ingested prednisone was compared to the measured concentrations in hair, a * Light-coloredhairdyed black. low but not insignificant correlation (R2 = 0.578, p < 0.03) was noticed (Figure 1). In this observation, the y ,, . 0 2 2 x 2.947, R-squared: .578 result obtained for patient no. 8 (103 mg of , , , i , , 1 401 prednisone in the proximal 1-cm long hair seco tion after a treatment period of one month) was 1 2o.I omitted for two reasons: (1) the small amount Jr of sample (30 mg instead of the 100 mg reoo quired), which can lead to a more approximative quantitative determination and (2) the rel80 atively short treatment period (one month) in contrast with the 3-13-cm long hair section 60 analyzed for the other patients where pred0 o g 40 nisone concentration reflects the mean expoO sure over several months (dilution due to the 2 length of the analyzed strand). 10'00 15'00 20'00 25'00 " 30'00 35'00" 40'00 4500 50o Although the limited number of subjects (n = Total amount of ingested prednisone during the corresponding hair segment analyzed with 1 cm/month growth (mg) 8) preclude generalization, results are suggestive of a dose-concentration correlation. Figure 1. Regression curve between total amount of ingested prednisone and prednisone In the future, it would be necessary to access concentration in hair. more treated patients and to analyze shorter n

111

Journal of Analytical Toxicology, Vol. 26, March 2002

hair sections corresponding to an unique dosage to confirm this observation.

References 1. R Kintz. Hair testing and doping control in sport. Toxicol. Lett. 102-103:109-113 (1998). 2. F. Bevalot, Y. Gaillard, M.A. Lhermitte, and G. P~pin. Analysis of corticosteroids in hair by liquid chromatography-electrospray ionisation mass spectrometry. J. Chromatogr. B 740:227-236 (2O00). 3. V. Cirimele, P. Kintz, V. Dumestre, J.R Goull~, and B. Ludes. Iden-

112

tification of ten corticosteroids in human hair by liquid chromatography-ionspray mass spectrometry. Forensic Sci. Int. 107: 381-388 (2000). 4. V. Cirimele, P. Kintz, A. Tracqui, and 8. Ludes. First identification of prednisone in human hair by liquid chromatography-ionspray mass spectrometry. J. Anal. Toxicol. 23:225-226 (1999). 5. P. Kintz, V. Cirimele, and B. Ludes. Pharmacological criteria that can affect the detection of doping agents in hair. Forensic ScL Int. 107:325-334 (2000). 6. V. Cirimele, P. Kintz, and P. Mangin. Drug concentrations in human hair after bleaching. J. Anal. Toxicol. 19:331-332 (1995). Manuscript received March 8, 2001; revision received July 26, 2001.