To determine the prevalence of platelet nonresponsiveness to aspirin treatment for secondary stroke prophylaxis, the authors studied consecutive patients during a 29-month period. Information regarding their ischemic events, risk factors, and medications was collected. Platelet aggregation in response to collagen and arachidonic acid was used to determine platelet responsiveness to aspirin. A total of 653 patients were evaluated. Of these, 129 patients (20%) were determined to be nonresponsive to aspirin based on continued platelet aggregation in response to collagen, arachidonic acid, or both. A total of 87 (13%) of the 653 patients were clinical aspirin failures (ie, presented with new focal cerebral ischemic symptoms while taking aspirin). Of the patients with new cerebral ischemic symptoms, 57 (66%) were determined to be platelet nonresponsive to aspirin. The odds ratio for platelet nonresponsiveness to aspirin in patients who suffered a recurrent ischemic event while taking aspirin was 14.25 (95% confidence interval: 8.5-23.7; P < .5). Continued platelet aggregation despite aspirin treatment occurred in 20% of ambulatory patients treated for secondary stroke prophylaxis. The prevalence of nonresponsiveness to aspirin was statistically higher in those patients who suffered recurrent cerebral ischemia while taking aspirin (P < .5) compared with patients who remained without new ischemic symptoms.