LAUSR

PURPOSE Differences in HIV prevalence, access to antiretrovirals and ICU resources may result in wide variation in sepsis mortality in HIV patients. The aim of this study was to perform a meta-analysis to quantify the excess risk of sepsis mortality in HIV patients. MATERIALS AND METHODS A systematic review was performed using three databases. The systemic inflammatory response syndrome criteria was used for the presumptive diagnosis of sepsis. We only included studies that stratified sepsis mortality by HIV serostatus. A meta-analysis was performed using random effects models, with subgroup analyses performed using country income, sepsis severity, and time periods. RESULTS 17 studies were included, containing 82,905 patients. Sepsis mortality was found to be 28% higher in the HIV positive patients (95% CI 1.13-1.46, p < .01). Relative risk of mortality was higher in patients treated in low-income countries (RR 1.43 in low-income vs. 1.29 in high-income countries). Mortality was more pronounced in HIV patients with severe sepsis (RR 1.32 in severe sepsis vs. RR 1.15 in sepsis). CONCLUSIONS HIV increases the risk of sepsis mortality compared to seronegative individuals across all time periods and geographic areas. We note that this effect is more pronounced in patients with organ dysfunction.