The antiplatelet effect of aspirin varies individually. This study evaluated whether the antiplatelet effect of aspirin associates with polymorphisms in the genes coding for cyclo-oxygenase-1 (COX-1) and several platelet glycoprotein (GP) receptors in patients with stable coronary artery disease (CAD). Blood samples were collected from 101 aspirin-treated (mean 100 mg/d) patients. Compliance to treatment was assessed by plasma salicylate measurement. Platelet functions were assessed by two methods: 1) Response to arachidonic acid (AA, 1.5 mmol/L in aggregometry, and 2) PFA-100, evaluating platelet activation under high shear stress in the presence of collagen and epinephrine (CEPI). Aspirin non-response was defined as: 1) slope steeper than 12%/min in AA-aggregations, and 2) by closure time shorter than 170 s in PFA-100. The methods used detected different individuals as being aspirin non-responders. Five and 21 patients, respectively, were non-responders according to AA-induced aggregation and PFA-100. Increased plasma thromboxane B2 levels correlated with poor aspirin-response measured with both AA-induced aggregations and PFA-100 (P = 0.02 and P = 0.003, respectively). Of the non-responders detected by AA, 3 of 5 (60%) carried the rare G allele for the -A842G polymorphism of COX-1 in contrast to 16 of 96 (17%) responders (P = 0.016). Diabetes was associated with poor response. Aspirin non-response detected by PFA-100 associated with C13254T polymorphism of GP VI and female gender (P = 0.012 and P = 0.019, respectively). Although two patients were possibly non-compliant, this did not effect present conclusions. Evaluation of aspirin efficacy by AA-induced aggregation and PFA-100 detected different individuals, with different genotypic profiles, as being aspirin non-responders.