Personalising the decision for prolonged dual antiplatelet therapy: Development, validation and potential impact of prognostic models for cardiovascular events and bleeding in myocardial infarction survivors

Aims: To develop models to aid the decision to prolong dual antiplatelet therapy (DAPT) requires balancing an individual patient’s potential benefits and harms. Methods and results: Using population-based electronic health records (CALIBER, England, 2000-2010), of patients evaluated 1 year after acute MI we developed (n=12,694 patients) and validated (n=5,613) prognostic models for cardiovascular (cardiovascular death, MI or stroke) and three different bleeding endpoints. We applied trial effect estimates to determine potential benefits and harms of DAPT and individuals net clinical benefit. Prognostic models for cardiovascular events (c-index: 0.75 (95% CI: 0.74, 0.77)) and bleeding (c index 0.72 (95% CI: 0.67, 0.77)) were well calibrated: 3-year risk of cardiovascular events was 16.5% overall (5.2% in lowest and 46.7% in highest-risk individuals) while for major bleeding it was 1.7% (0.3% in lowest and 5.4% in highest-risk patients). For every 10,000 patients treated per year, we estimated 249 (95% CI: 228, 269) cardiovascular events prevented and 134 (95% CI: 87,181) major bleeding events caused in highest-risk patients, and 28 (95% CI: 19, 37) cardiovascular events prevented and 9 (95% CI: 0, 20) major bleeding events caused in lowest-risk patients. There was a net clinical benefit of prolonged DAPT in 63% to 99% patients depending on how benefits and harms were weighted. Conclusions: Prognostic models for cardiovascular events and bleeding using population-based electronic health records may help personalise decisions for prolonged DAPT 1 year following acute MI.