Choice of endpoints in antiplatelet trials

Publisher Website
Google Scholar
Abstract
Article abstract The relative efficacies of different antiplatelet agents for stroke prevention are unclear because of differences in clinical trial design, a lack of direct comparisons between individual agents, and differences in the choice of primary endpoints. Individual endpoints in a clinical trial are often combined into a single primary endpoint cluster. Theoretically, a combined endpoint may reduce the sample size required to demonstrate significant benefits of an effective therapy. However, unless all components of a composite endpoint are affected in the same direction and to a similar degree, their inclusion may not provide the anticipated increase in statistical power. In fact, the use of a combined endpoint may lead to an underestimate of therapeutic benefits when patients at high risk for one particular endpoint are studied. For patients with stroke or TIA, the single outcome of stroke is particularly important because these patients have a higher risk of recurrent stroke than any other vascular outcome during the initial years after a stroke. Because of the low incidence of myocardial infarction (MI) in stroke trials, the inclusion of MI in the primary endpoint will usually have minimal influence on trial outcome, and antiplatelet therapy has not been shown to be beneficial in preventing nonvascular death. Chance variations in the incidence of MI or death may detract from the benefit of the agent for stroke prevention when it is included in a combined endpoint. The benefit of antiplatelet therapies for patients with recent cerebrovascular events is determined most accurately if stroke alone is chosen as the primary endpoint.