We hypothesized that single-nucleotide polymorphisms (SNPs) associated with heightened in vitro platelet function during aspirin exposure (which we define as “laboratory aspirin resistance”) would be associated with greater risk for death, myocardial infarction (MI) or stroke among patients with coronary artery disease regularly using aspirin.

Duke Databank for Cardiovascular Disease patients with (n = 3,449, CATHeterization GENetics cohort) or without (n = 11,754, nongenetic cohort) banked DNA with ≥1 coronary stenosis >75% were followed up at 6 months, then annually for death, MI, or stroke occurring during periods of reported aspirin use. We evaluated associations of candidate SNPs from GNB3, PEAR1, ITGB3, VAV3, ITGA2, GPVI, PTGS1, F2R, THBS1, A2AR, and GP1BA with events during follow-up using Cox proportional hazards modeling adjusted for clinical characteristics associated with outcomes in the nongenetic cohort.

Over a median of 3.5 years, 2,762 (24%) nongenetic cohort patients and 648 (19%) CATHeterization GENetics cohort patients had the composite outcome during reported aspirin use. No candidate SNPs were significantly associated with death, MI, or stroke in either univariable or multivariable analyses. A prospective analysis demonstrated 80% to 88% power to detect a hazard ratio of ≥1.3 for minor allele carriers.

Patients with angiographically significant coronary artery disease regularly using aspirin and carrying SNPs associated with laboratory aspirin resistance were not at higher risk for death, MI, or stroke. Using these SNPs to guide more aggressive antiplatelet therapy is not justified by these results. Direct extrapolation from in vitro findings to the clinical setting should be avoided.