Background: Platelet activation has been implicated in restenosis after percutaneous transluminal coronary angioplasty (PTCA), but previous studies may have been confounded by factors such as elastic recoil and arterial remodelling. Restenosis after coronary stenting is unlikely to be affected by these factors. Methods: Forty-nine patients who had stenting for acute or impending closure after PTCA were included in the study. Patients with restenosis (≤ 50% stenosis by angiography) and without restenosis were selected using a case-control design. Restenosis was determined by the caliper method. Patients were tested for platelet activation 1-4 years after their procedure while taking their usual medications (including aspirin). Reliability testing was conducted with 11 healthy subjects. Platelet activation was measured in blood leaving a bleeding-time wound (wound- induced platelet activation), using flow cytometry. Blood was collected from the wound site 1 and 2 min after the incision. Monoclonal antibodies were used to test for activation of glycoprotein (GP) IIb/IIIa (PAC-1), GPIIb/IIIa ligand binding (anti-ligand-induced binding site 1: anti-LIBS-1), and P- selectin expression (AC1.2). Results: Short-term intersample reliability was very good to excellent for anti-LIBS-1 and AC1.2 (intraclass correlation coefficients 0.79-0.96), but only fair for PAC-1. Patients with restenosis (n = 25) had greater activation in all measures than patients without restenosis (n = 24); the difference was significant for GPIIb/IIIa ligand binding at 1 min (P = 0.03). The correlation between GPIIb/IIIa ligand binding at 1 min and percent stenosis at follow-up was also significant (P = 0.03). Patients taking nitrates had lower activation; after eliminating these patients, GPIIb/IIIa ligand binding was greater among patients with restenosis at both 1 and 2 min (P = 0.04 for both). Conclusions: The results suggest that increased GPIIb/IIIa ligand binding may be associated with restenosis after coronary stenting. The results also suggest that the wound-induced platelet activation method is a reliable and valid measure of platelet activity.