Bone loss is accelerated in elderly men. Little is known about the pathophysiology of senile bone loss or about the role played by relative sex steroid deficiency in the determination of bone turnover in elderly men. In a population-based sample of 283 healthy, ambulatory men, aged 71-86 years, we sought to determine whether lower bone mineral density (BMD; using dual-energy X ray absorptiometry at the hip and the forearm) is associated with higher bone turnover, and we assessed the impact of sex steroid status on bone turnover. Indices of bone formation, serum osteocalcin (s-Oc), and bone-specific alkaline phosphatase (s-bAP) and indices of bone resorption, serum and urinary telopeptide of type I collagen (s-CTx and u-CTx), and urinary free deoxypyridinoline (u-Dpd) were intercorrelated (r = 0.29-0.76, p < 0.001). Bone turnover indices were negatively associated with BMD (r = -0.17 to -0.34, p < 0.01). In univariate analyses, there was a trend toward weak negative associations of bone turnover markers with serum free testosterone (FT), significant only for s-Oc and s-CTx (r = -0.16 and -0.14, p < 0.01), and with serum free estradiol (FE(2)), significant only for u-CTx and s-CTx (r = -0.18 and -0.19; p < 0.01). The lower quartile for FE(2) was associated with higher values of u-CTx (p = 0.003) and s-CTx (p < 0.001). However, in multivariate models, for the individual markers of bone turnover a negative association between estradiol (E(2)) or FE(2) and s-CTx was the only remaining (marginally) significant association (p < 0.05) for the relationship between sex steroids and any of the bone turnover indices assessed. In community-dwelling men age >70 years, bone turnover rate, as determined by biochemical markers, is a significant negative determinant of prevalent BMD. However, the findings do not support the view that relative differences in sex steroid status, as observed among healthy elderly men, have a major impact on bone turnover.