Comparison of different methods for the measurement of serum testosterone in the aging male.

Abstract

Circulating testosterone (T) levels, mainly non-SHBG bound fractions, decline with advancing age. Free (FT) and bioavailable (BT) testosterone levels have been suggested to represent more reliably the bioactive hormone at the tissue level as compared to total testosterone (TT) levels. However, there is an ongoing controversy as to which T assay is most appropriate in defining T deficiency in the aging male. In a prospective observational study, TT, FT and BT levels were measured in 51 elderly men (55 to 70 years). TT levels were determined on two different days using 2 different assays. FT was calculated using TT, albumin and SHBG levels as well as measured directly by radioimmunoassay. The free androgen index (FAI) was calculated using the formula 100 x TT/SHBG. BT was obtained by precipitation of SHBG-bound T with ammonium sulphate. We found good correlations within different assays for TT (r = 0.87 to 0.91, p <0.001) and between TT and calculated FT (r = 0.66 to 0.84, p <0.001). In contrast, the correlation between TT and BT was poor (r = 0.17 to 0.19, p = ns). TT was equivalent to calcFT and was better in mirroring clinical sings of androgen deficiency in elderly men as compared to BT. The intra-individual day-to-day variance of TT levels was 14.8% (range 0.1-79%) with a coefficient of variance of 12.6%. Serum TT and the calculated FT fraction correlated well with each other and were superior in defining a group of elderly men with suspected androgen deficiency. In contrast, FT measured by direct RIA and BT reflected gonadal function poorly. Our data favour the repetitive use of TT when screening for androgen deficiency in elderly men.