A low response to clopidogrel has been associated with an increased risk of stent thrombosis. However, the definition of a nonresponse to clopidogrel remains controversial, and different tests have been used to assess the clopidogrel response. The present study was designed to assess the predictive value of adenosine diphosphate (ADP)-induced platelet aggregation (ADP-Ag) and the Platelet Reactivity Index of vasodilator-stimulated phosphoprotein for the occurrence of stent thrombosis in patients admitted for non-ST-elevation acute coronary syndrome undergoing percutaneous coronary intervention. A total of 598 consecutive patients with non-ST-elevation acute coronary syndrome undergoing coronary stenting were prospectively included. They received 600 mg of clopidogrel >or=12 hours before percutaneous coronary intervention. Acute or subacute definite or probable stent thrombosis occurred in 11 patients (1.8%). These patients had significantly greater ADP-Ag compared to patients free of stent thrombosis (68 +/- 14% vs 56 +/- 19%, p = 0.002) but only a trend toward a greater Platelet Reactivity Index of vasodilator-stimulated phosphoprotein (62 +/- 14% vs 53 +/- 23%, p = 0.19). The construction of receiver operating characteristic curves to examine the most predictive value of ADP-Ag for stent thrombosis gave a threshold of ADP-Ag of >67% to identify low responders. These patients were at a greater risk of stent thrombosis than the clopidogrel responders (4.3% vs 0.8%, odds ratio 5.8, 95% confidence interval 1.9 to 24.6, p = 0.003). In conclusion, in patients with non-ST-elevation acute coronary syndrome undergoing percutaneous coronary intervention, ADP-Ag is a good parameter to identify clopidogrel nonresponders who are at increased risk of stent thrombosis, with a cutoff value of ADP-Ag of >67%.