Rates and predictors of testosterone supplementation in HIV-infected men in Centers for AIDS Research Network of Integrated Clinical Systems (CNICS)

Background: Testosterone supplementation in US men is increasing. Little is known about testosterone supplementation in HIV-infected(HIV+) men.

Methods: We studied adult HIV+ men enrolled at 7 sites within CNICS from 1996-2011, excluding men on testosterone at entry. We calculated testosterone supplementation incidence as number of events per follow-up time (person-years, py) from entry to supplementation, loss to follow-up, or death. Incidence was age-adjusted based on population distribution from the 2010 US Census. We assessed factors associated with supplementation using log-rank and multivariable Cox regression.

Results: We studied 14,454 men with 75,173py of follow-up. Median(IQR) age was 39(32-45)y, 50% were White, 69% were men who have sex with men(MSM), and 4% had AIDS wasting. At entry, median(IQR) viral load and nadir CD4+ T-lymphocyte count(CD4) were 3.6(3.3-3.8)log₁₀copies/ml and 252(89-420)cells/µl, respectively, and 70% received antiretroviral therapy(ART), with 37% on protease inhibitors(PI). Supplementation occurred in 1482(10%) men at median(IQR) age of 44(38-51)y, with age-adjusted rate of 19.5/1,000py(95%CI14.9-29.4). In bivariable comparisons, supplementation was associated with age ≥50y, White race, MSM, AIDS wasting, low nadir CD4, and PI (all p<0.01). In multivariable analyses, supplementation was independently associated with age (per decade; HR 1.28, 95%CL1.21-1.37; p<0.01), White race (HR 1.67, 95%CL 1.47-1.91; p<0.01), AIDS wasting (HR 2.04, 95%CL 1.63-2.57; p<0.01), nadir CD4 (per 100 cells/µl; HR 0.97, 95%CL 0.94-0.99; p=0.02), and PI (HR 1.41, 95%CL 1.21-1.65; p<0.01).

Conclusions: Age-adjusted testosterone supplementation incidence in HIV+ men (19.5/1,000py) was higher compared to the general US male population (7.6/1,000py). Older age, White race, and advanced HIV were associated with supplementation.