Background: There has been increasing concern that HIV-infected individuals may be more at risk for cardiovascular events in the highly-active antiretroviral therapy (HAART) era. This study examined the risk of thromboembolic events in HIV-infected and non-infected individuals and the effect of macrolide prophylaxis on those outcomes. Methods: A subcohort analysis was undertaken using data collected in the Multicenter AIDS Cohort Study to examine the relative risk of vascular events (myocardial infarction, unstable angina and ischaemic stroke). Cox proportional hazard model using age as the time scale with time varying cofactors obtained at each semi-annual visit were used to assess the independent effect of macrolide use. Results: Controlling for other significant effects including race and smoking, HIV-infection was not independently associated with vascular events. Increased risk was observed among those who used HAART (relative hazard 1.09, 95% confidence intervals 1.00–1.19 in multivariate model), antihypertensive treatment (1.81 [1.26–2.60]), lipid-lowering medication (1.65 [1.12–2.42]), and antibiotics (1.72 [1.25–2.36]). The protective association of macrolide use for a vascular event in the HAART era was also significant (0.10 [0.01–0.75]). Conclusions: Traditional risk factors are important in the pathogenesis of vascular events in HIV-infected individuals. Macrolide antibiotics may have a protective effect in the HIV-infected individuals in the HAART era.