Background:  Hypogonadism is the most common endocrine disorder among HIV-infected men, but little is known about its cause(s). We performed a prospective study to determine the prevalence and risk factors for hypogonadism at a large HIV clinic.

Methods:  All adult HIV-infected men in our clinic were offered enrollment. Data collected included demographics, hormone supplements, CD4 counts, viral load, duration of HIV infection, CDC stage, HIV medications, and use of alcohol, tobacco, or drugs. Questionnaires (ADAM) regarding hypogonadism, testicular examination, and early morning testosterone levels were performed. Those with hypogonadism (<300 ng/dL) underwent an evaluation for its cause and were offered treatment. Statistical analysis included Fisher’s exact and rank sum tests for qualitative and quantitative data, respectively.  Multivariate logistic models were constructed (STATA software 8.0).

Results:  Of 296 study patients, 50 (17%) had hypogonadism. The ADAM was positive among 93% of the entire cohort; the sensitivity of the test was 84% with a specificity of only 11%. Predictors of hypogonadism in the univariate analyses included increasing age (OR 1.30 per 5 years, p <0.001), CDC stage C (OR 3.15, p = 0.004), longer duration of HIV infection (OR 1.08, p = 0.002), lower CD4 nadir per 100 cells (OR 0.83, p = 0.03), longer duration of receiving HIV medications (OR 1.11, p = 0.002), and lipodystrophy (OR 2.24, p = 0.011). Tobacco use appeared protective (OR 0.03, p = 0.005). In the multivariate model, increasing age (OR 1.24, p = 0.002) was associated with hypogonadism while smoking was protective (OR 0.44, p = 0.02); there was no association with the use of HIV medications or any other variable. All patients had secondary hypogonadism; 1 patient had hypothyroidism; 72% were given testosterone therapy (the rest declined or had medical conditions precluding its use). Overall, 63% reported satisfaction with the replacement therapy and 77% plan to continue the testosterone supplementation.

Conclusions:  Hypogonadism is common among HIV-infected men in the HAART era and is usually attributable to hypothalamic/pituitary dysfunction. Given its association with increasing age, the prevalence of hypogonadism may rise as HIV-infected men survive to older ages. Smoking is protective of hypogonadism mirroring data in HIV-negative men; this may be due to nicotine’s effect on aromatase activity.