In healthy subjects, platelet hyperreactivity is a global phenomenon--as opposed to agonist-specific--and epinephrine-induced platelet aggregation (EPA) is a reliable marker of this phenotype. Few data are available on platelet reactivity and the relationship between EPA and aggregation induced by other agonists in cardiovascular patients. It was the objective of this study to characterise platelet reactivity in stable cardiovascular patients treated with aspirin and to derive a composite index integrating several aggregation pathways, suitable for selecting patients with extreme phenotypes for further proteomics analysis. Platelet reactivity to agonists was assessed in 110 patients twice, two weeks apart. Factorial analysis was used to determine whether the results obtained with the different agonists could be summarised in a single composite index. EPA correlated with the aggregation values obtained with each of the other agonists, with correlation coefficients of 0.44 to 0.55 (p < 0.001). We constructed a composite "platelet reactivity" index that included 60% of the information provided by each agonist. The results obtained at the first patient visit were consistent with those obtained at the second visit (r = 0.78, p<0.01). No clinical or biological parameters correlated with the composite index. The extreme phenotypes of six selected subjects were confirmed 12 months after the second visit. In conclusion, platelet reactivity in aspirin-treated cardiovascular patients is a global phenomenon that can be summarised by a composite index based on the aggregation responses to various agonists and integrating several activation pathways. This index is not dependent on clinical or biological variables, suggesting that genetic factors regulate platelet reactivity in these patients.