Interindividual variation in the ability of aspirin to inhibit platelet cyclooxygenase-1 (COX-1) could account for some on-treatment cardiovascular events. Here, we sought to determine whether there are clinical phenotypes that are associated with a suboptimal pharmacological effect of aspirin. In a prospective, 2-week study, we evaluated the effect of aspirin (81 mg) on platelet COX-1 in 135 patients with stable coronary artery disease by measuring serum thromboxane B2 (sTxB2) as an indicator of inhibition of platelet COX-1. A nested randomized study compared enteric-coated with immediate-release formulations of aspirin. We found that sTxB2 was systematically higher among the 83 patients with metabolic syndrome than among the 52 patients without (median: 4.0 versus 3.02 ng/mL; P=0.013). Twelve patients (14%) with metabolic syndrome, but none without metabolic syndrome, had sTxB2 levels consistent with inadequate inhibition of COX (sTxB2 ≥13 ng/mL). In linear regression models, metabolic syndrome (but none of its individual components) significantly associated with higher levels of log-transformed sTxB2 (P=0.006). Higher levels of sTxB2 associated with greater residual platelet function measured by aggregometry-based methods. Among the randomized subset, sTxB2 levels were systematically higher among patients receiving enteric-coated aspirin. Last, urinary 11-dehydro thromboxane B2 did not correlate with sTxB2, suggesting that the former should not be used to quantitate aspirin's pharmacological effect on platelets. In conclusion, metabolic syndrome, which places patients at high risk for thrombotic cardiovascular events, strongly and uniquely associates with less effective inhibition of platelet COX-1 by aspirin.