Introduction: In healthy men, body weight and total fat content increase with advancing age, while serum testosterone levels decrease. In order to elucidate whether a causal relationship between these phenomena exists, we investigated the influence of testosterone or human chorionic gonadotrophin substitution on body mass index (BMI), total fat mass and serum leptin in testosterone-treated and untreated hypogonadal patients in comparison with ageing eugonadal men. Methods: In a cross-sectional study, the inter-relationships of body weight, total fat mass, serum sex hormones and leptin were analysed in untreated hypogonadal men (n ¼ 24; age 19-65 years), treated hypogonadal men (n ¼ 61; age 20-67 years) and healthy eugonadal men (n ¼ 60; age 24-78 years). Total fat mass was assessed by bioimpedance measurement. Univariate and multiple linear regression analysis was used to detect possible differences. Results: In eugonadal men, serum testosterone levels decreased with advancing age (correlation coefficents: r ¼ 20:71; P , 0:0001), while BMI ðr ¼ 0:39; P ¼ 0:002Þ; total fat content ðr ¼ 0:51; P , 0:0001Þ and leptin ðr ¼ 0:48; P , 0:0001Þ increased significantly. In untreated hypogonadal patients, an increase in BMI ðr ¼ 0:50; P ¼ 0:013Þ and total fat mass ðr ¼ 0:41; P ¼ 0:044Þ was also observed with advancing age. However, in substituted hypogonadal patients, no age-dependent change in BMI ðr ¼ 0:067; P ¼ 0:606Þ; body fat content ðr ¼ 20:083; P ¼ 0:522Þ; serum testosterone ðr ¼ 20; 071; P ¼ 0:59Þ or serum leptin ðr ¼ 20:23; P ¼ 0:176Þ was found. Conclusion: Since testosterone-substituted older hypogonadal men show BMI and fat mass similar to those of younger eugonadal men and since non-treated hypogonadal men are similar to normal ageing men, testosterone appears to be an important factor contributing to these changes. Thus ageing men should benefit from testosterone substitution as far as body composition is concerned.