Background: HIV-1 and malaria are common infections in Africa, and cause substantial morbidity and mortality. HIV infection has been associated with an increased incidence of malaria, and more severe disease. Our aim was to assess the effect of antiretroviral treatment (ART) on the frequency of clinical malaria in people with HIV, and to measure the additive effects of co-trimoxazole (trimethoprim and sulfamethoxazole) prophylaxis, ART, and insecticide-treated bednets.
Methods: In 2001, we enrolled 466 HIV-infected individuals aged 18 years or older in Uganda in a prospective cohort study that provided co-trimoxazole prophylaxis to 399 participants after 5 months of no intervention. In 2003, we enrolled 138 survivors from the initial study, and 897 new participants from the same community, to take antiretroviral therapy (ART) in addition to co-trimoxazole prophylaxis. The ART was in most cases a combination of stavudine, lamivudine, and nevirapine or efavirenz. In 2004, we also gave participants insecticide-treated bednets. Households were visited weekly by study staff to record fever, illness, or death in the preceding 7 days. In cases of reported fever in the previous 2 days, we took blood to test for malaria parasites. We compared the frequency of clinical malaria, adjusting for CD4-cell count, age, sex, and season.
Findings: 1035 individuals were given co-trimoxazole and ART (median age 38 years, 74% female, and median CD4-cell count 124 cells/microL); 985 of these, plus four new participants, received co-trimoxazole, ART, and bednets. There were 166 cases of clinical malaria in the study. Compared with a baseline malaria incidence of 50.8 episodes per 100 person-years, co-trimoxazole prophylaxis was associated with 9.0 episodes per 100 person-years (adjusted incidence rate ratio [IRR] 0.24, 95% CI 0.15-0.38); ART and co-trimoxazole with 3.5 episodes per 100 person-years (0.08, 0.04-0.17); and co-trimoxazole, ART, and bednets with 2.1 episodes per 100 person-years (0.05, 0.03-0.08). Malaria incidence was significantly lower during ART and co-trimoxazole than during co-trimoxazole alone (IRR 0.36 [95% CI 0.18-0.74], p=0.0056).
Interpretation: A combination of co-trimoxazole, antiretroviral therapy, and insecticide-treated bednets substantially reduced the frequency of malaria in adults with HIV.