Abstract

Within the existing methods for carbon isotope ratio (CIR) determinations one of the main challenges was the separation of steroids with 5α- and 5β-conformation like androsterone (AND) and etiocholanolone (ETIO) or 5α-androstanediol (5a-Diol) and 5β-androstanediol (5b-Diol). Switching to a different HPLC column allowed for complete separation of ETIO and AND and 5a-Diol and 5b-Diol. This change in sample preparation necessitated the determination of a new reference population to establish reliable thresholds for the improved method. Therefore n = 67 males and females were investigated enclosing the glucuronidated steroids ETIO, AND, 5a-Diol, 5b-Diol, pregnanediol, 16-androstenol, 11-ketoetiocholanolone, testosterone, epitestosterone and dehydroepiandrosterone (DHEA). Additionally, to corroborate earlier findings on steroids excreted sulfated, the following steroids from the sulfate fraction were included: ETIO, AND, DHEA, epiandrosterone and 17α- and 17β-androstenediol. Regarding the glucuronidated steroids, over all the new limits are lower than the existing ones. For ETIO and AND the differences are small suggesting that the co-elution of both compounds had no significant influence on the measured CIR. A significant difference was found for 5a-Diol and 5b-Diol, suggesting that the old method introduced a systematic offset on these values. For sulfated steroids, the values were comparable but showed smaller standard deviations, now equal to the ones for glucuronidated steroids. This investigation demonstrates the necessity to re-evaluate reference based thresholds after changing the method for CIR determinations even though most changes are mainly due to improved measurements resulting in smaller standard deviations. Verf.-Referat