Background HIV infection disproportionately affects men who have sex with men (MSM) in the industrialised world. Rectal infections are objective markers of HIV risk behaviour. We examined the association between rectal chlamydia/gonorrhoea (CT/GC) infections and HIV seroconversion.

Methods MSM attending New York City public STD clinics who report receptive anal sex are offered rectal testing using GC culture and CT nucleic acid amplification tests (NAAT); patients not known to have HIV are offered HIV testing. We created a retrospective cohort of MSM diagnosed with rectal CT or GC in 2008–2009 at STD clinics who tested HIV-negative by pooled NAAT at that visit. The outcome was time to HIV infection, defined as a STD clinic diagnosis or identified through a match against the citywide HIV/AIDS registry (HARS) for HIV diagnosed elsewhere during the analytic period. For MSM who seroconverted, HIV-free time-at-risk was from rectal infection to date of positive HIV test; those not reported with HIV were presumed uninfected and censored on 31 December 2010. Cox proportional hazards models were used to explore demographic and behavioural factors associated with HIV acquisition.

Results A total of 229 HIV-negative MSM diagnosed with rectal infections contributed 368.29 person-years of follow-up; 22/229 (9.6%) were diagnosed with HIV (16 in STD clinics and an additional 6 found in HARS), for an annual HIV incidence of 5.97% (95% CI 3.84 to 8.90). Median time from rectal infection to HIV diagnosis was 290 days (range 98–748). The small subgroups of black and <20-year-old MSM had markedly high incidence (14.19% and 10.79%, respectively) (see Abstract P1-S5.23 table 1). MSM reporting inconsistent condom use had an annual HIV incidence of 6.33% (95% CI 3.43 to 10.75). Black race was associated with a 6.5-fold increased risk of HIV; after adjusting for age this finding did not reach statistical significance (HR=5.05, 95% CI 1.00 to 25.68).

Conclusions More than 1 in 20 MSM with rectal infections are diagnosed with HIV within a year; risk is higher for subgroups such as young and black MSM. Local data on risk for seroconversion may be more compelling than national data in risk-reduction counselling. As the majority of rectal infections are due to CT, and CT is associated with substantial HIV risk, routine rectal CT screening is indicated for MSM. STD/HIV registry matching/integration permit more accurate incidence estimates and definition of affected populations with which to focus prevention activities.