Serum dihydrotestosterone and testosterone concentrations in human immunodeficiency virus-infected men with and without weight loss.

Weight loss is an important determinant of disease outcome in human immunodeficiency virus (HIV)-infected men. Others have suggested that a defect in dihydrotestosterone (DHT) generation contributes to weight loss in HIV-infected men. To determine whether DHT levels correlate with weight loss independently of changes in testosterone levels, we prospectively measured serum total- and free-testosterone and DHT levels in 148 consecutive HIV-infected men and 42 healthy men. Thirty-one percent of HIV-infected men had serum testosterone levels less than 275 ng/dL, the lower limit of the normal male range; of these, 81% had normal or low LH and FSH levels (hypogonadotropic), and 19% had elevated LH and FSH levels (hypergonadotropic). Overall, serum testosterone, free-testosterone, and DHT levels were lower in HIV-infected men than in healthy men, but serum DHT-to-testosterone ratios were not significantly different between the two groups. Serum total- and free-testosterone levels were lower in HIV-infected men who had lost 5 lb or more of weight in the preceding 12 months than in those who had not lost any weight. Serum DHT levels and DHT-to-testosterone ratios did not differ between those who had lost weight and those who had not. Serum testosterone and free-testosterone levels, but not DHT levels, correlated with weight change and with Karnofsky performance status. We also performed a retrospective analysis of data from a previous study in which HIV-infected men with serum testosterone levels less than 400 ng/dL had been treated with placebo or testosterone patches designed to nominally release 5 mg testosterone over 24 hours. Serum testosterone-to-DHT ratios did not change after testosterone treatment. Changes in fat-free mass were correlated with changes in both serum testosterone (r = 0.42, P = 0.018) and DHT (r = 0.35, P = 0.049) levels. Serum total- testosterone and DHT levels were highly correlated with one another, and when the change in serum testosterone was taken into account, serum DHT levels no longer showed a significant correlation with change in fat-free mass. We conclude that DHT levels are lower in HIV-infected men than in healthy men but that neither DHT levels nor DHT-to-testosterone ratios correlate with weight loss. During testosterone treatment, serum DHT levels increase proportionately, but the increments in serum testosterone correlate with the change in fat-free mass. Our data do not support the hypothesis that a defect in DHT generation contributes to weight loss in HIV-infected men independently of changes in testosterone levels; it is possible that such a defect might exist in HIV-infected men with more severe weight loss.