Loss of bone and muscle mass are major findings of male hypogonadism. In order to determine the long-term effect of testosterone replacement therapy on spinal bone and muscles, the trabecular and cortical bone mineral density, vertebral body area and paraspinal muscle area were assessed by quantitative computed tomography in 32 testosterone-substituted patients, aged 18-74 years, with idiopathic hypogonadotropic hypogonadism (n=6), pituitary insufficiency (n=5), Klinefelter syndrome (n=12) or other forms of primary hypogonadism (n=9). They were followed for a mean period of 3.2+/-1.7 years (mean+/-S.D.), ranging from 1 to 7 years. A significant correlation between initial serum testosterone levels and bone mineral density was found in patients with congenital forms (r=0.58; P<0.05) but not in those with acquired forms. A significant increase in trabecular and cortical bone mineral density (P<0.001) was documented in the course of replacement therapy in all patients regardless of the type of hypogonadism and age of patients. A slight but significant increase in paraspinal muscle area was observed if all patients were taken together (P<0.01). The area of paraspinal muscle correlated with body weight (r=0.58; P<0.001) and moderately with trabecular bone mineral density (r=0.4; P<0.01). Its increase did not correspond to the change observed for trabecular and cortical bone mineral density. Vertebral body area did not change over time. It correlated only with height and weight but not with bone mineral density. In conclusion, testosterone therapy of hypogonadal men improves both trabecular and cortical bone mineral density of the spine independently of age and type of hypogonadism while vertebral area remains unchanged. The effects seen on paraspinal muscles emphasize the clinical benefit of adequate replacement therapy for the physical fitness of hypogonadal men.