Background: HIV infection has been associated with increased risk of myocardial infarction (MI), yet the use of aspirin (ASA) for primary prevention has not been studied in this group. We therefore investigated ASA use and risk of MI in ASA users compared to non-users, stratified by HIV infection and coronary heart disease (CHD) risk factors.

Methods: We conducted an observational cohort study of 3698 HIV⁺ and 33,348 matched HIV^– patients without known baseline CHD in a large healthcare system in Boston from 2000-2009. We developed an algorithm to ascertain non-episodic ASA use employing medication data and electronic health record free text search. A definition of at least 2 ASA prescriptions, 2 text strings, or 1 of each, all occurring more than 30 days prior to MI, was validated by medical record review and yielded an area under the relative operating characteristics (ROC) curve of 0.78. We used Cox proportional hazard modeling to evaluate the relationship between ASA use and incident MI within the HIV⁺ and control groups, adjusting for demographics, cardiovascular risk factors, and HIV-related variables when applicable. We further assessed the effect of ASA in models stratified by cardiovascular risk.

Results: ASA use was lower among HIV⁺ compared to HIV^– patients for the overall group (12.4% vs 15.3%, p <0.001), among patients with 0-1 CHD risk factors (5.5% vs 6.7%, p = 0.037) and among patients with 2 or more CHD risk factors (22.1% vs 42.4%, p <0.001) (all comparisons HIV⁺ vs HIV^–). ASA use was not associated with decreased MI risk (hazard ratio [HR] 0.97, 95% confidence interval [CI], 0.64-1.49, p = 0.90) in multivariate modeling controlling for traditional risk factors among HIV⁺ patients but was associated with a significantly decreased risk of MI (HR 0.29, 95% confidence interval [CI] 0.24-0.34, p <0.001) among HIV^– patients.

Conclusions: Aspirin use was lower among HIV⁺ patients compared to HIV^– patients, with a greater relative difference among those with significant CHD risk. Moreover, ASA use was associated with reduced risk of incident MI among HIV^– patients but not among HIV⁺ patients. Further studies are needed to investigate optimal indications and strategies for ASA use among HIV⁺ patients.