The effects of testosterone administration on the GH axis in androgen-deficient HIV-infected women are unknown. In this study, we determined the effects of transdermal testosterone administration on GH secretory dynamics and pulse characteristics in this population. GH-IGF-I parameters were determined in response to testosterone (4.1 mg/patch, twice a week; estimated delivery rate, 150 μg/d) vs. placebo over 6 months in 31 HIV-infected women. IGF-I increased significantly in the testosterone-treated compared with the placebo-treated patients [37 (−4, 73) vs. −30 (−98, 39) ng/ml, P = 0.01; 4.8 (−0.5, 9.6) vs. −3.9 (−12.8, 5.1) nmol/liter]. GH pulse frequency increased significantly in the testosterone-treated compared with the placebo-treated subjects [1.0 (1.0, 2.0) vs. 0.0 (−0.5, 1.5) peaks per 12 h, respectively; P = 0.02]. Before testosterone administration, overnight GH pulse amplitude was significantly related to IGF-I in univariate (r = 0.41, P = 0.03) and multivariate regression analysis; however, free testosterone, estradiol, and body mass index were not significantly correlated with baseline IGF-I. In contrast, after 6 months of treatment with testosterone, the change in IGF-I was significantly correlated to the change in free testosterone in univariate (r = 0.40, P = 0.04) and multivariate regression analysis. For each 1.0 pg/ml (3.5 pmol/liter) increase in free testosterone, IGF-I increased 19 ng/ml (2.5 nmol/liter), controlling for estradiol, body mass index, and GH pulse parameters (r2 = 0.64). We demonstrate that IGF-I increases in response to physiologic, transdermal testosterone in HIV-infected women. The mechanism of this effect is unknown, but may involve a direct effect of testosterone on IGF-I, independent of changes in GH pulse dynamics.