Effects of androgen administration on the growth hormone-insulin-like growth factor I axis in men with acquired immunodeficiency syndrome wasting

It is unknown whether hypogonadism contributes to decreased insulin-like gr owth factor I (IGF-I) production and/or how testosterone administration may effect the GH-IGF-I axis in human immunodeficiency virus (HIV)-infected me n with the acquired immunodeficiency syndrome (AIDS) wasting syndrome (AWS) . In this study, we investigate the GH-IGF-I axis in men with the AWS and d etermine the effects of testosterone on GH secretory dynamics, pulse charac teristics determined from overnight frequent sampling, arginine stimulation , and total and free IGF-I levels. Baseline GH-IGF-I parameters in hypogona dal men with AWS (n = 51) were compared before testosterone administration (300 mg, im, every 3 weeks vs. placebo for 6 months) with cross-sectional d ata obtained in two age-matched control groups: eugonadal men with AIDS was ting (n = 10) and healthy age-matched normal men (n = 15). The changes in G H-IGF-I parameters were then compared prospectively in testosterone- and pl acebo-treated patients. Mean overnight GH levels [1.8 +/- 0.3 and 2.4 +/- 0 .3 vs. 0.90 +/- 0.1 mu g/L (P = 0.04 and P = 0.003 vs, healthy controls)] a nd pulse frequency [0.35 +/- 0.06 and 0.37 +/- 0.02 vs. 0.22 +/- 0.03 pulse s/h (P = 0.06 and P = 0.002 vs, healthy controls)] were comparably elevated in the eugonadal and hypogonadal HIV-positive groups, respectively, compar ed to those in the healthy control group. No significant differences in pul se amplitude, interpulse interval, or maximal GH stimulation to arginine ad ministration (0.5 g/kg, iv) were seen between either the eugonadal and hypo gonadal HIV-positive or healthy control patients. In contrast, IGF-I levels were comparably decreased in both HIV-positive groups compared to the heal thy control group [143 +/- 16 and 165 +/- 14 vs. 216 +/- 14 mu g/L (P = 0.0 04 and P = 0.02 vs. healthy controls)]. At baseline, before treatment with testosterone, overnight GH levels were inversely correlated with IGF-I (r = -0.42; P = 0.003), percent ideal body weight (r = -0.36; P = 0.012), album in (r = -0.37; P = 0.012), and fat mass (r = -0.52; P = 0.0002), whereas IG F-I levels correlated with free testosterone (r = 0.35; P = 0.011) and calo ric intake (r = 0.32; P = 0.023) in the hypogonadal HIV-positive men. In a stepwise regression model, albumin (P = 0.003) and testosterone (P = 0.011) were the only significant predictors of GH [mean GH (mu g/L) = -1.82 x alb umin (g/dL) + 0.003 x total testosterone (mu g/L) + 6.5], accounting for 49 % of the Variation in GH. Mean overnight GH levels decreased significantly in the testosterone-treated patients compared to those in the placebo-treat ed hypogonadal patients (-0.9 +/- 0.3 vs. 0.2 +/- 0.4 mu g/L; P = 0.020). I n contrast, no differences in IGF-I or free IGF-I were observed in response to testosterone administration. The decrement in mean overnight GH in resp onse to testosterone treatment was inversely associated with increased fat- free mass (r = -0.49; P = 0.024), which was the only significant variable i n a stepwise regression model for change in GH [change in mean GH (mu g/L) = -0.197 x kg fat-free mass - 0.53] and accounted for 27% of the variation in the change in GH. In this study, we demonstrate increased basal GH secretion and pulse freque ncy in association with reduced IGF-I concentrations, consistent with GH re sistance, among both hypogonadal and eugonadal men with AIDS wasting. Testo sterone administration decreases GH in hypogonadal men with AIDS wasting. T he change in GH is best predicted by and is inversely related to the magnit ude of the change in lean body mass in response to testosterone administrat ion. These data demonstrate that among hypogonadal men with the AWS, testos terone administration has a significant effect on the GH axis.