Relative risk of cardiovascular disease among people living with HIV: a systematic review and meta-analysis

Objectives The aim of this study was to estimate the relative risk of cardiovascular disease (CVD) among people living with HIV (PLHIV) compared with the HIV-uninfected population. Methods We conducted a systematic review and meta-analysis of studies from the peer-reviewed literature. We searched the Medline database for relevant journal articles published before August 2010. Eligible studies were observational and randomized controlled trials, reporting CVD, defined as myocardial infarction (MI), ischaemic heart disease, cardiovascular and cerebrovascular events or coronary heart disease among HIV-positive adults. Pooled relative risks were calculated for various groupings, including different classes of antiretroviral therapy (ART). Results The relative risk of CVD was 1.61 [95% confidence interval (CI) 1.431.81] among PLHIV without ART compared with HIV-uninfected people. The relative risk of CVD was 2.00 (95% CI 1.702.37) among PLHIV on ART compared with HIV-uninfected people and 1.52 (95% CI 1.351.70) compared with treatment-naive PLHIV. We estimate the relative risk of CVD associated with protease inhibitor (PI)-, nucleoside reverse transcriptase inhibitor- and nonnucleoside reverse transcriptase inhibitor-based ART to be 1.11 (95% CI 1.051.17), 1.05 (95% CI 1.011.10) and 1.04 (95% CI 0.991.09) per year of exposure, respectively. Not all ART was associated with increased risk; specifically, lopinavir/ritonavir and abacavir were associated with the greater risk and the relative risk of MI for PI-based versus non-PI-based ART was 1.41 (95% CI 1.201.65). Conclusion PLHIV are at increased risk of cardiovascular disease. Although effective in prolonging survival, ART (in particular PI-based regimens) is related to further increased risk of CVD events among people at highest initial absolute risk of cardiovascular disease.