Impact of Bleeding on Mortality After Percutaneous Coronary Intervention: Results From a Patient-Level Pooled Analysis of the REPLACE-2 (Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events), ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy), and HORIZONS-AMI (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction) Trials

Abstract

This study sought to develop a risk score predictive of bleeding in patients undergoing percutaneous coronary intervention (PCI) and to investigate the impact of bleeding on subsequent mortality. Bleeding complications after PCI have been independently associated with early and late mortality. This study represents a patient-level pooled analysis including 17,034 patients undergoing PCI from 3 large, randomized trials of bivalirudin versus heparin plus glycoprotein IIb/IIIa inhibitors, including the REPLACE-2 (Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events), ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy), and HORIZONS-AMI (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction) trials. We developed a risk score to predict noncoronary artery bypass graft (CABG)-related TIMI (Thrombolysis In Myocardial Infarction) major bleeding and evaluated the impact of various types of bleeding on 1-year mortality. A non-CABG-related TIMI major bleed occurred within 30 days in 267 patients (1.6%), and death occurred in 497 patients (2.9%) within 1 year. A risk score was developed to predict the bleeding risk of patients undergoing PCI, consisting of 7 variables (serum creatinine, age, sex, presentation, white blood cell count, cigarette smoking, and randomized treatment). The TIMI major bleeding rates increased by bleeding risk score groups: from 0.4% for those in the lowest to 5.8% for those in the highest risk group. Non-CABG-related TIMI major bleeding and the occurrence of myocardial infarction within 30 days were independent predictors of subsequent mortality, with respective hazard ratios of 4.2 and 2.9, each p < 0.001. Ranked in order of severity, TIMI major bleeding, blood transfusion without TIMI bleed, TIMI minor bleeding requiring blood transfusion, and TIMI minor bleeding not requiring blood transfusion were independent predictors of subsequent mortality with hazard ratios of 4.89, 2.91, 2.73, and 1.66, respectively. Isolated hematomas were not predictive of subsequent mortality. Non-CABG-related bleeding within 30 days is strongly associated with an increased risk of subsequent mortality at 1 year in patients undergoing PCI for all indications. A risk score was established to calculate the bleeding risk for patients undergoing PCI, allowing therapeutic decision making to minimize the incidence of bleeding.