Urinary testosterone (T) to epitestosterone (E) ratios by GC/MS. I. Initial comparison of uncorrected T/E in six...

Six laboratories in six countries collaborated to investigate the analytical method for estimating the testosterone to epitestosterone ratio (T/E) in urine by gas chromatography/mass spectrometry in the context of detecting the application of T as a doping agent in sport. The protocol specified many but not all details of reagents and instrument conditions. The design included the distribution and analysis of four urines with different T/E values, three replicates per value, and one standard. The ranges of mean T/E values for the four urines estimated by peak area (PA) were 0.32-0.42, 0.72-0.94, 0.91-1.14 and 3.19-5.48. The analyses of variance for these data and for the peak height (PH) data were significant for the laboratory factor (p < 0.0001). In addition there was a significant interaction between the urine factor and the laboratory factor which indicates the complexity of the analysis. T/E calculated using PA was not significantly different from that using PH. For within-laboratory precision all values for PH and PA were < 8.3%, and for between-laboratory precision all values were < 11.7% except for one (20.1%). The data represent a baseline for future experiments designed to elucidate the sources of within-and between-laboratory variance, and to harmonize estimates of T/E.