Prediction of Ischemic and Bleeding Events Using the Dual Antiplatelet Therapy Score in an Unrestricted Percutaneous Coronary Intervention Population Analysis From the ADAPT-DES Registry

BACKGROUND: The dual antiplatelet therapy (DAPT) risk score was developed from the DAPT trial to inform the optimal duration of DAPT after percutaneous coronary intervention. We assessed the performance of the DAPT score in the ADAPT-DES (Assessment of Dual AntiPlatelet Therapy with drug-eluting stents) all-comers registry and tested the utility of additional predictors of adverse events. METHODS AND RESULTS: Outcomes between 1 and 2 years were examined according to DAPT score >= 2 versus <2, adjusted for DAPT continuation as a time-dependent variable. To assess the incremental utility of variables not included in the DAPT score, baseline high platelet reactivity was added to the ischemia model, and high platelet reactivity, baseline hemoglobin, and warfarin use 1 year after percutaneous coronary intervention were added to the bleeding model. Among 8582 patients enrolled in ADAPT-DES, 5397 were event-free after 1 year. Between 1 and 2 years, ischemic (myocardial infarction or stent thrombosis) and bleeding events occurred in 75 (1.5%) and 124 (2.3%) patients, respectively. Patients with higher DAPT scores (>= 2 versus <2) had higher rates of myocardial infarction or stent thrombosis (1.9% versus 1.1%; P=0.01) and similar rates of bleeding (2.2% versus 2.4%, respectively; P=0.79). For the prediction of myocardial infarction or stent thrombosis, bleeding and death, DAPT score 2 had sensitivities of 57%, 41%, and 56%, respectively; specificities of 58%, 57%, and 58%, respectively; positive predictive values of 1.9%, 2.2%, and 2.1%, respectively; and negative predictive values of 99%, 98%, and 99%, respectively. Addition of baseline high platelet reactivity to the DAPT score did not improve discrimination for ischemic events. Addition of high platelet reactivity and 2 other bleeding covariates to the DAPT score marginally improved discrimination. CONCLUSIONS: In ADAPT-DES, the DAPT score was predictive of ischemic events between 1 and 2 years after drug-eluting stents. Prediction of bleeding improved marginally after addition of variables not incorporated in the DAPT score.