Effect of the biological matrix on the urinary testosterone/epitestosterone ratio measured by gas chromatography/mass spectrometry in doping analysis

Testosterone doping in sport is detected by measurement of an increased testosterone/epitestosterone (T/E) ratio in urine. The critical limit is 6. The present study concerns calibration curves for the T/E ratio measured by gas chromatography/mass spectrometry (electron impact) according to the guidelines of the International Olympic Committee. Testosterone (T) and epitestosterone (E) are measured as trimethylsilyl (TMS)‐enol‐TMS ethers in selected ion monitoring mode using m / z 432 with methyltestosterone (MT) ( m / z 446) as internal standard. Calibration curves corresponding to T/E = 1, 6 and 12 prepared directly, i.e. without extraction of T and E, were non‐linear. The non‐linearity was caused by an increase of the relative molar response of T with respect to the internal standard MT with increasing concentration level. A mean increase of 82% was observed from T/E = 1 to T/E = 12 (E fixed). Adding T/E corresponding to 1/1, 6/1 and 12/1 to urine without endogeneous hormone content resulted in an almost linear calibration curve along the diagonal, with only a slight increase of the relative molar response of testosterone (16% from T/E = 1 to 12). Apparently, the biological matrix stabilizes the relative molar response over a wide concentration range. At a molar ratio of about 1/1 for T/MT, the relative molar response for direct measurement of T is identical to that observed in the presence of urine matrix, which is explained on the basis of a simple mathematical model. The practical conclusion of this study is that, contrary to the present‐day practice, calibration curves for the T/E ratio should be based on T/E added to blank urine taken through the extraction procedure. Otherwise, the T/E ratio of urine sample is systematically easily underestimated by 30% or more.