Associations Between Chronic Kidney Disease and Outcomes With Use of Prasugrel Versus Clopidogrel in Patients With Acute Coronary Syndrome Undergoing Percutaneous Coronary Intervention: A Report From the PROMETHEUS Study

Objectives This study sought to compare clinical outcomes in a contemporary acute coronary syndrome (ACS) percutaneous coronary intervention (PCI) cohort stratified by chronic kidney disease (CKD) status. Background Patients with CKD exhibit high risks for both thrombotic and bleeding events, thus complicating decision making regarding antiplatelet therapy in the setting of ACS. Methods The PROMETHEUS study was a multicenter observational study comparing outcomes with prasugrel versus clopidogrel in ACS PCI patients. Major adverse cardiac events (MACE) at 90 days and at 1 year were defined as a composite of death, myocardial infarction, stroke, or unplanned revascularization. Clinically significant bleeding was defined as bleeding requiring transfusion or hospitalization. Cox regression multivariable analysis was performed for adjusted associations between CKD status and clinical outcomes. Hazard ratios for prasugrel versus clopidogrel treatment were generated using propensity score stratification. Results The total cohort included 19,832 patients, 28.3% with and 71.7% without CKD. CKD patients were older with greater comorbidities including diabetes and multivessel disease. Prasugrel was less often prescribed to CKD versus non-CKD patients (11.0% vs. 24.0%, respectively; p < 0.001). At 1 year, CKD was associated with higher adjusted risk of MACE (1.27; 95% confidence interval: 1.18 to 1.37) and bleeding (1.46; 95% confidence interval: 1.24 to 1.73). Although unadjusted rates of 1-year MACE were lower with prasugrel versus clopidogrel in both CKD (18.3% vs. 26.5%; p < 0.001) and non-CKD (10.9% vs. 17.9%; p < 0.001) patients, associations were attenuated after propensity stratification. Similarly, unadjusted differences in 1-year bleeding with prasugrel versus clopidogrel (6.0% vs. 7.4%; p = 0.18 in CKD patients; 2.6% vs. 3.5%; p = 0.008 in non-CKD patients) were not significant after propensity score adjustment. Conclusions Although risks for 1-year MACE were significantly higher in ACS PCI patients with versus without CKD, prasugrel use was 50% lower in patients with renal impairment. Irrespective of CKD status, outcomes associated with prasugrel use were not significant after propensity adjustment. These data highlight the need for randomized studies evaluating the optimal antiplatelet therapy in CKD patients with ACS.