Serum adiponectin levels in hypogonadal males: influence of testosterone replacement therapy

OBJECTIVE Adiponectin is an adipocyte-specific secretory protein which exhibits antiatherogenic, anti-inflammatory and antidiabetic properties. We hypothesized that testosterone plays an important role in the regulation of its secretion in humans, as adiponectin concentrations are higher in women than in men and as testosterone administration is accompanied by a reduction in serum adiponectin in animals and by reduced protein secretion in cultured adipocytes. This study aimed to evaluate adiponectin levels in hypogonadal men prior to and during testosterone replacement therapy. SUBJECTS AND METHODS In a retrospective study, adiponectin, total and free testosterone, oestradiol, SHBG, total cholesterol and triglyceride levels were evaluated in 31 hypogonadal men [HM; age, mean +/- SEM: 36.5 +/- 2.4 years; body mass index (BMI) 24.6 +/- 0.8 kg/m(2)] and 29 weight-matched eugonadal men (EM; age 30.8 +/- 1.5 years; BMI 23.4 +/- 0.6 kg/m(2)). In 13 HM (age 33.9 +/- 3.2 years; BMI 24.2 +/- 0.9 kg/m(2)) the same parameters were also evaluated after 6 months of testosterone replacement therapy. Correlation analysis between adiponectin and hormonal, biochemical and anthropometric parameters was performed in all subjects. RESULTS Testosterone, free testosterone and oestradiol concentrations were significantly lower in HM than in EM (4.4 +/- 0.4 nmol/l, 78.4 +/- 10.9 pmol/l and 36.1 +/- 3.0 pmol/l, respectively, in HM vs. 21.9 +/- 0.7 nmol/l, 507.9 +/- 13.8 pmol/l and 65.2 +/- 1.8 pmol/l, respectively, in EM, P < 0.0001), while SHBG levels in HM were higher than in EM (54.4 +/- 7.5 vs. 30.9 +/- 2.2 nmol/l, P < 0.005). Serum adiponectin levels in HM were significantly higher than in EM (9.53 +/- 0.73 vs. 6.80 +/- 0.55 mug/ml, P < 0.01). Calculation of the Pearson coefficient showed that adiponectin levels in HM were not correlated with any of the anthropometric and hormonal parameters examined, but showed a significant negative correlation with serum triglycerides (r = -0.38, P < 0.05). Serum adiponectin levels were negatively correlated with body weight (r = -0.41, P < 0.05) in EM but not with other anthropometric, hormonal or biochemical parameters. Six months after initiation of testosterone replacement therapy, which increased testosterone and free testosterone levels to the normal range, adiponectin levels were significantly reduced in HM (6.37 +/- 0.93 vs. 9.26 +/- 1.01 mug/ml, P < 0.01) and similar to those recorded in EM. CONCLUSIONS Compared to eugonadal subjects, hypogonadal men show higher adiponectin levels which are reduced by testosterone replacement therapy. This study indicates that testosterone exerts a regulatory role on adiponectin secretion in humans.