Aspirin is the most commonly used antiplatelet drug for prevention of ischaemic stroke. In order to determine the prevalence and nature of aspirin failure, we studied 51 adults admitted with suspected ischaemic stroke and already prescribed daily aspirin. Within 48 hours (h) of onset, blood and urine samples were collected to assess platelet aggregation, activation and aspirin response by a range of methods. All tests were then repeated on a second sample taken 24 h after witnessed administration of 75 mg or 150 mg aspirin. At entry to the study, incomplete response to aspirin, measured by arachidonic acid (AA)-stimulated platelet aggregation, was found in 43% of patients. Following in-hospital aspirin administration, there was a significant decrease in AA-aggregation (p=0.001) suggesting poor adherence to therapy prior to admission. However, residual aggregation (10–15%) persisted in 11 subjects – suggesting alternative causes. In incomplete responders on admission, platelet aggregation with adenosine diphosphate (ADP) was significantly higher compared with responders (p<0.05) but there were no significant differences in collagen aggregation, platelet fibrinogen binding or P-selectin expression, plasma von Willebrand factor, fibrinogen, high-sensitivity C-reactive protein, or the urinary metabolite, 11-dehydro-TxB2. Incomplete platelet inhibition is common around the time of acute cerebrovascular ischaemic events in patients prescribed aspirin. Up to 50% of these observations appear due to incomplete adherence to aspirin therapy. Intervention studies are required to determine the clinical relevance of measured platelet response to aspirin in terms of outcome, and the effectiveness of improved pharmacotherapy for stroke prevention.