Measures of bioavailable serum testosterone and estradiol and their relationships with muscle strength, bone density, and body composition in elderlymen

In the present cross-sectional study of 403 independently living elderly me n, we tested the hypothesis that the decreases in bone mass, body compositi on, and muscle strength with age are related to the fall in circulating end ogenous testosterone (T) and estrogen concentrations. We compared various m easures of the level of bioactive androgen and estrogen to which tissues ar e exposed. After exclusion of subjects with severe mobility problems and signs of deme ntia, 403 healthy men (age, 73-94 yr) were randomly selected from a populat ion-based sample. Total T (TT), free T (FT), estrone (E-1), estradiol (E-2) , and sex hormone-binding globulin (SHBG) were determined by RIA. Levels of non-SHBG-bound T (non-SHBG-T), FT (calc-FT), the TT/SHBG ratio, non-SHBG-b ound E-2, and free E-2 were calculated. Physical characteristics of aging i ncluded muscle strength measured using dynamometry, total body bone mineral density (BMD), hip BMD, and body composition, including lean mass and fat mass, measured by dual-energy x-ray absorptiometry. In this population of healthy elderly men, calc-FT, non-SHBG-T, E-1, and E- 2 (total, free, and non-SHBG bound) decreased significantly with age. T (to tal and non-SHBG-T) was positively related with,muscle strength and total b ody BMD (for non-SHBG-T, respectively, beta = 1.93 +/- 0.52, P < 0.001 and beta = 0.011 +/- 0.002 P < 0.001). An inverse association existed between T and fat mass (beta = -0.53 +/- 0.15, P < 0.001). Non-SHBG-T and calc-FT we re more strongly related to muscle strength, BMD, and fat mass than TT and were also significantly related to hip BMD. E-1 and E-2 were both positivel y, independently associated with BMD (for E-2, beta = 0.21 +/- 0.08, P < 0. 01). Non-SHBG bound E-2 was slightly strongly related to BMD than total E-2 . The positive relation between T and BMD was independent off,. E-1 and E-2 were not related with muscle strength or body composition. In summary, bioavailable T, E-1, total E-2, and bioavailable E-2 all decrea se with age in healthy old men. In this cross-sectional study in healthy el derly men, non-SHBG-bound T seems to be the best parameter for serum levels of bioactive T, which seems to play a direct role in the various physiolog ical changes that occur during aging. A positive relation with muscle stren gth and BMD and a negative relation with fat mass was found. In addition, b oth serum E-1 and E-2 seem to play a role in the age-related bone loss in e lderly men, although the cross-sectional nature of the study precludes a de finitive conclusion. Non-SHBG-bound E-2 seems to be the best parameter of s erum bioactive E-2 in describing its positive relation with BMD.