The question of whether and to what extent the sex steroid deficiency in elderly men contributes to the pathogenesis of bone loss has not been fully explored. The aim of the present study was to assess the association of serum bioavailable (Bio) estradiol (E(2)) with the evolution of bone mineral density (BMD) in 214 community-dwelling men aged 71-86 yr as well as the possible modulation of estrogen effects by a tetranucleotide (TTTA)(n)-repeat polymorphism of the CYP19 gene, which encodes the aromatase enzyme that converts androgens into estrogens. BMD was measured at yearly intervals over a period of 4 yr using dual x-ray absorptiometry. Fasting blood was analyzed at baseline for testosterone (T), E(2), and SHBG; the respective bioavailable fractions, BioT and BioE(2), were calculated. Serum BioE(2) was associated with baseline BMD at different assessed skeletal sites, with correlation coefficients ranging between 0.23 and 0.37 (P < 0.001). Estimated annual percentage change of BMD (%BMD) was -0.39% [95% confidence index (CI), -0.56, -0.22] at the total hip, -0.04% (95% CI, -0.29, 0.21) at the femoral neck, and -0.37% (95% CI, -0.45, -0.29) at the total distal forearm. Higher circulating BioE(2) levels were associated with less bone loss at the forearm and the hip (P < 0.05). The CYP19 gene (TTTA)(n)-repeat length (determined by fragment analysis) was not associated with baseline BMD in the total group of elderly men. However, a significant association was observed between the CYP19 genotype and BMD change at the distal forearm; the highest bone loss was observed in subjects homozygotic for the shortest observed allele length of (TTTA)(7)-repeats (P < 0.02). The CYP19 (TTTA)(n)-repeat length was not associated with either baseline BioE(2) or the BioT/BioE(2) ratio. In multiple linear regression models, the CYP19 genotype and serum BioE(2) were determinants of %BMD change at the forearm (P < 0.05). No significant contribution of BioT to %BMD change was evident. As to fracture risk, the allele containing the shortest (TTTA)(n)-repeat length was more represented not only in elderly men with a positive personal fracture history (Pearson's chi(2) test = 4.03; df = 1; P = 0.05) but also in study subjects with a positive fracture history in their first-degree relatives (Pearson's chi(2) test = 6.48; df = 1; P = 0.01). In conclusion, the results of this prospective observational study support the view that BioE(2) is a determinant of bone density changes in elderly men and, furthermore, provide an indication that the aromatase enzyme may exert a direct modulatory action on bone metabolism at the tissue level in elderly men.

No abstract

Aging is accompanied by involutional changes in testicular function; limited data suggest a decrease in bilateral testicular volume (BTV). We studied BTV by ultrasonography in relation to serum gonadal hormones in 115 healthy elderly men (median age, 78 yr) and 42 young men (median age, 26.5 yr). Elderly men had a clearly smaller BTV (mean, 20.6 vs. 29.7 ml; P < 0.001), whereas serum inhibin B was slightly but significantly decreased (mean, 176.8 vs. 212.8 ng/liter; P = 0.04); lower values in the elderly were observed for bioavailable (Bio) testosterone (T), Bio 17 beta-estradiol, inhibin B/FSH (mean, 18 vs. 58 ng/mU; P < 0.001), and T/LH ratios. In the elderly and the young, respectively, BTV was associated with inhibin B (r = 0.53, P < 0.001; r = 0.41, P < 0.01), FSH (r = -0.53, P < 0.001; r = -0.48, P < 0.01), and inhibin B/FSH ratio. Only in the old men was BTV significantly associated with LH (r = -0.32; P < 0.001), Bio T (r = 0.26; P < 0.01), and T/LH (r = 0.48; P < 0.001). In a multivariate analysis, FSH, inhibin B, and Bio T were independently associated with BTV in the elderly (R(2) = 0.34). Receiver operating characteristics curve analysis indicated that BTV at a criterion value of 14.3 ml had a sensitivity of 46% and a specificity of 79% to predict low serum Bio T levels in the elderly. In conclusion, the moderately decreased BTV observed in elderly men, strongly associated with a decrease of the inhibin B/FSH ratio, is consistent with a reduced Sertoli cell mass, compensated by increased FSH stimulation resulting in only limited decrease of Sertoli cell function. Finding of a low testicular volume in elderly men can contribute to the diagnosis of hypogonadism, but this criterion has low sensitivity to detect decreased T production.

The findings of the present study do not support the view that in community-dwelling, healthy elderly men the androgen receptor gene CAG-repeat polymorphism has a substantial impact on interindividual variability of serum testosterone levels or on the determination of bone turnover and bone mineral density.

No abstract