Risk of cardiovascular events among patients with HIV treated with atazanavir-containing regimens: a retrospective cohort study

BACKGROUND: A previous cohort study indicated that atazanavir (ATV), a protease inhibitor used for HIV treatment, is not associated with an increased risk of cardiovascular (CV) events. The objective of this study was to compare the risk of CV events among antiretroviral-naïve patients initiating ATV-containing versus ATV-free ARV regimens. METHODS: Patients with HIV who newly initiated antiretroviral therapy were selected from MarketScan Commercial and Multi-State Medicaid databases. The first claim for an antiretroviral medication between 1/1/2007 and 12/31/2013 was known as the index date. Patients were categorized as initiating an ATV-containing or an ATV-free regimen. Patients who did not have 6 months of continuous enrollment prior to the index date or who had evidence of a CV event during this time period were excluded. Myocardial infarction, stroke, percutaneous coronary intervention, and coronary artery bypass graft were identified through diagnosis and procedure codes. Patients were followed from index date until a CV event, continuous gap of >30 days without initiated ARV, a claim for ATV in the ATV-free cohort, disenrollment, or study end, whichever occurred first. Unadjusted incidence rates (IR) were calculated and propensity-score-weighted Cox proportional hazards models were fit to compare hazards of CV events between the two cohorts. RESULTS: A total of 22,211 patients (2437 ATV-containing and 19,774 ATV-free) were identified in the Commercial Database and 7136 patients were identified (1505 ATV-containing and 5631 ATV-free) in the Medicaid Database. CV events were uncommon (Commercial IR per 1000 person-years for a CV event: ATV-containing = 3.01, ATV-free = 3.26; Medicaid IR: ATV-containing = 10.9, ATV-free = 9.9). In propensity-score-weighted models combining the two populations, there was no significant difference in the hazards of a CV event for patients initiating an ATV-containing regimen compared with those initiating an ATV-free regimen (hazard ratio = 1.16, 95 % confidence interval 0.67–1.99). CONCLUSIONS: In this real-world analysis, there was no significant increase in the risk of CV events associated with exposure to ATV-containing regimens. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12879-016-1827-1) contains supplementary material, which is available to authorized users.