Abstract P233: Selecting Antiplatelet Therapy After Percutaneous Intervention for an Acute Coronary Syndrome: Weighing the Risks and Benefits of Prasugrel Versus Clopidogrel

Introduction: Clinical trials report the mean benefits and risks of a treatment for a population, which can mask the heterogeneity of risks and benefits for individual patients. Recognizing and communicating individual patients’ risks may support translation of trial results into practice. Methods: Using data from 12,907 ACS patients undergoing percutaneous coronary intervention (PCI) in TRITON-TIMI 38, we fit separate multivariable logistic regression models to predict major ischemia (cardiac death, MI & stroke) and bleeding (TIMI major/minor) risk with both prasugrel and clopidogrel. We built full models using variables selected a priori , then created a more parsimonious model by sequentially removing variables explaining the least variance until the reduced model R 2 in prediction of the full model fell below 0.95. Potential interactions were tested between treatment and all covariates. Results: Models are presented in the Figure; the ischemia risk model included interactions of treatment with age, heparin, diabetes and Killip class. Predicted risk of ischemia varied substantially across the population and was lower with prasugrel (mean 8.9%±4.2%, range 1%-50.6%) vs. clopidogrel (mean 11.3%±5.2%, range 1.5%-61.7%), while bleeding risk was greater with prasugrel (mean 5.0%±3.1%, range 0.5%-26.8%) vs. clopidogrel (mean 3.7%±2.4%, range 0.4%-20.8%). Conclusions: We found substantial variability in patients’ predicted benefit and risk. Risk models that leverage data clinical trial data to estimate each patient's risk and benefit with prasugrel vs. clopidogrel at the time of decision making should be tested to see if they support safer, more cost-effective care.