Performance characteristics of a carbon isotope ratio method for detectingdoping with testosterone based on urine diols: Controls and athletes with elevated testosterone/epitestosterone ratios

Background: Carbon isotope ratio methods are used in doping control to dete rmine whether urinary steroids are endogenous or pharmaceutical. Methods: Gas chromatography-combustion-isotope ratio mass spectrometry (GC- C-IRMS) was used to determine the delta C-13 values for 5 beta -androstane- 3 alpha ,17 beta -diyl diacetate (5 betaA), 5 alpha -androstane-3 alpha ,17 beta -diyl diacetate (5 alphaA), and 5 beta -pregnane-3 alpha ,20 alpha -d iyl diacetate (5 betaP) in a control group of 73 healthy males and 6 athlet es with testosterone/epitestosterone ratios (T/E) >6. Results: The within-assay precision SDs for 5 betaA, 5 alphaA, and 5 betaP were +/- 0.27 parts per thousand, +/- 0.38 parts per thousand, and +/- 0.28 parts per thousand, respectively. The between-assay precision SDs ranged f rom +/- 0.40 parts per thousand to +/- 0.52 parts per thousand. The system suitability and batch acceptance scheme is based on SDs. For the control gr oup, the mean delta C-13 (SD) values were -25.69 parts per thousand (+/- 0. 92 parts per thousand), -26.35 parts per thousand (+/- 0.68 parts per thous and), and -24.26 parts per thousand (+/- 0.70 parts per thousand), for 5 be taA, 5 alphaA, and 5 betaP, respectively. 5 betaP was greater than 5 betaA and 5 alphaA (P <0.01), and 5<beta>A was greater than 5 alphaA (P <0.01). T he means - 3 SD were -28.46<parts per thousand>, -28.39 parts per thousand, and -26.37 parts per thousand for 5 betaA, 5 alphaA, and 5 betaP, respecti vely. The maximum difference between 5 betaP and 5 betaA was 3.2 parts per thousand, and the maximum 5 betaA/5 betaP was 1.13. Three athletes with chr onically elevated T/Es had delta C-13 values consistent with testosterone a dministration and three did not. Conclusions: This GC-C-IRMS assay of urine diols has low within- and betwee n-assay SDs; therefore, analysis of one urine sample suffices for doping co ntrol. The means, SDs, +/-3 SDs, and ranges of delta C-13 values in a contr ol group are established. In comparison, testosterone users have low 5 beta A and 5 alphaA, large differences between 5 betaA or 5 alphaA and 5 betaP, and high 5 betaA/5 betaP and 5 alphaA/5 betaP ratios. (C) 2001 American Ass ociation for Clinical Chemistry.