To conduct a meta-analysis to synthesize the evidence regarding the effect of co-infection with GB virus C (GBV-C) on survival of HIV-infected individuals, and to estimate the effect. A Bayesian meta-analysis was conducted to synthesize evidence from eligible studies. Prospective survival studies of HIV-1-infected individuals, with outcome defined as time from baseline to all-cause death, were included and classified by whether GBV-C status was determined in early or late HIV disease. The primary measure was the hazard ratio (HR) of death for HIV-infected individuals with GBV-C infection versus those without GBV-C infection. Eleven studies from eight publications met the inclusion criteria. For studies with GBV-C status measured 2 years or less after HIV seroconversion (912 subjects), the combined HR was 0.88 [95% credible interval (CI) 0.30, 1.50]. For studies with GBV-C status measured more than 2 years after HIV seroconversion (1294 subjects), the combined HR was 0.41 (95% CI 0.23, 0.69). No conclusive evidence was found of an association between survival and GBV-C infection early in HIV disease. However, when GBV-C infection was present later in HIV disease, a significant reduction in the hazard for mortality was observed for those with co-infection. Potential explanations for this difference include a non-proportional benefit of GBV-C over time, possibly related to clearance of GBV-C infection early in HIV disease. The timing of GBV-C infection appears to account for the contradictory results of studies on the effect of GBV-C coinfection on survival of HIV-infected people.
