The testosterone : androstenedione ratio in male undermasculinization

OBJECTIVE Recent reports suggest that low testosterone:androstenedione (T:A ) ratio following hCG stimulation may be a useful method of diagnosing 17 b eta -hydroxysteroid dehydrogenase-3 (17 beta HSD3) deficiency. The aim of t his study was to establish the range of T:A ratios in cases of undermasculi nization with proven aetiologies other than 17 beta HSD3. DESIGN Register-based study of cases of male undermasculinization reported to a central database by clinicians. SUBJECTS Amongst the 421 cases of undermasculinization. 114 cases had testo sterone and androstenedione levels before and after hCG stimulation. Of the 114, there were 18 cases of abnormal testes, 17 cases of complete androgen insensitivity syndrome (CAIS), 68 cases of partial AIS (PAIS). Of the 17 c ases of CAIS, 13 had evidence of androgen receptor (AR) dysfunction; in the PAIS cohort, 26 cases had evidence of AR dysfunction. METHODS Analysis of T:A ratios in the above cohorts and comparison of these ratios to those in a group of previously described cases of 17 beta HSD3 d eficiency with a mean ratio of 0.4 (SD: 0.2) RESULTS The median age (range) for the CAIS, PAIS and abnormal testes cohor t was 1.25 years (0.06-16.5), 0.7 years (0.02-40.3) and 0.5 years (0.04-6.5 ), respectively. In CAIS, the median T:A rose from 0.4 (0.1 to 8.0) to 4.5 (0.5-16.7); in PAIS, median T:A rose from 0.7 (0.1 to 15) to 3.9 (0.3-20.5) ; in cases with abnormal testes, median T:A rose from 0.4 (0.1 to 5.6) to 0 .6 (0.1-3.6). The median post-hCG T:A ratio was significantly lower in the abnormal testes cohort (P < 0.01). None of the cases of AIS with AR mutatio n had a low T:A ratio, Only four out of 84 cases diagnosed as AIS had a T:A ratio less than 0.8 (mean +2SD in 17<beta>HSD3 deficiency). In one of the four cases, the T:A ratio rose to 3.5 following a prolonged hCG stimulation test. CONCLUSION Deficiency of 17 beta HSD3 should be considered in 46XY undermas culinization if the post-hCG stimulation T:A ratio is less than 0.8. Howeve r, low T:A ratios may be encountered in conditions such as abnormal testes. Before embarking on mutational analysis, we would also recommend careful e valuation for testicular dysgenesis including a prolonged hCG stimulation t est in cases with a low T:A ratio.