Dixon Yang, MD
Aspirin monotherapy, aspirin with dipyridamole, and clopidogrel alone are the first-line antithrombotic therapies for long-term secondary prevention of non-cardioembolic stroke and transient ischemic attack (TIA). In clinical practice, we often consider an individual patient’s bleeding risks against potential benefits of reduced ischemic events from these single antiplatelet agents. Assessing risk-benefit by stratification of bleeding risk may help guide clinical decision-making. Therefore, Hilkens et al. sought to investigate the net benefit of antiplatelet treatment according to an individual’s bleeding risk through pooled analysis of six randomized control trials.
The authors pooled individual patient data from CAPRIE, ESPS-2, MATCH, CHARISMA, ESPRIT, and PRoFESS, which investigated antiplatelet therapy in the subacute or chronic phase after non-cardioembolic stroke or TIA. The authors stratified patients into quintiles by their individually calculated S2TOP-BLEED score, derived from sex, smoking, modified Rankin Scale, prior stroke, hypertension, body mass index, age, and diabetes. For each quintile, the authors determined the annual rate of major bleeding and recurrent ischemic events of: 1) aspirin monotherapy; 2) aspirin-clopidogrel versus monotherapy; 3) aspirin-dipyridamole versus clopidogrel; and 4) aspirin versus clopidogrel. In the second, third, and fourth comparisons, the authors calculated net benefit.
Over 37,000 patients were included in the analysis. In the aspirin monotherapy group, the ischemic risk outweighed the bleeding risk in all quintiles. For the dual antiplatelet group, the risk reduction of ischemic events did not outweigh the risks of major bleeds in any quintile (Figure 2). A net clinical benefit of dual antiplatelets was seen in the lowest 3 quintiles when compared to single antiplatelet therapy, but only if intracranial hemorrhage is considered. In the aspirin-dipyridamole versus clopidogrel analysis, there was no clear preference for either treatment. Lastly, in the aspirin versus clopidogrel comparison, the net benefit was positive for clopidogrel in the lowest 4 quintiles.
The study’s key message is that the risk of major bleeding and the risk of future ischemic events increased in parallel for all antiplatelet treatments across all quintiles. The authors concluded that bleeding risk assessment alone is insufficient to guide long-term antiplatelet treatment for individual patients after non-cardioembolic stroke or TIA. The study lends support to a lack of net benefit in long-term dual antiplatelet therapy. Study limitations include the possibility of excluding subgroups with higher bleeding risk, heterogeneity of bleeding risk and classification across randomized controlled trials, and lack of data on stroke subtype. Further study is needed to identify other clinical predictors of bleeding and ischemia risk in long-term antiplatelet treatment.
