Sexually transmitted infections (STIs) have increased among men who have sex with men (MSM) and are associated with unsafe sex practices, intrinsic morbidity, and enhanced genital shedding and transmission of HIV. Screening for asymptomatic STIs is recommended as part of the HIV prevention efforts, however, optimal screening strategies among HIV-infected MSM have not been well defined. In this study, conducted from April 2004 to September 2006, 212 HIV-infected MSM from two urban HIV clinics were screened for asymptomatic STIs. Testing for Neisseria gonorrhea and Chlamydia trachomatis from pharynx, rectum, and urine, as well as serologic testing for syphilis were performed initially, and then after 6 and 12 months. A self-administered questionnaire was used to assess possible predictors of incident asymptomatic STIs. A cost analysis was performed to assess different screening strategies for detecting incident STIs. The baseline prevalence of STIs was 14% (n = 29; 95% confidence interval [CI] 9%-19%) and the incidence of new infections was 20.8 cases per 100 person years (95% CI 14.8-28.4 cases per 100 person years). Younger age, higher CD4 cell count, and marijuana use were associated with increased risk of acquiring an asymptomatic STI. The laboratory cost to detect one positive STI did not significantly differ between once- and twice-yearly screening. However, almost half of all incident STIs were detected at the 6-month screening visit, potentially resulting in an increased duration of infectivity if these cases remained undiagnosed. In conclusion, prevalent and incident asymptomatic STIs are common among HIV-infected MSM. Our data support current Center for Disease Control and Prevention STI guidelines that recommend routine screening at increased frequency for HIV-infected MSM.

The study provides data to support the current recommendation for pneumococcal immunization of all HIV-infected persons. Although this conclusion could lead to renewed enthusiasm for increasing pneumococcal immunization rates in HIV-infected persons, it must be recognized that the study is observational and ascertainment bias cannot be excluded.