Background: An association between transfusion during index hospitalization and increased subsequent mortality has been reported in acute myocardial infarction (AMI). Whether this reflects the prognostic role of transfusion per se, or the impact of the index event leading to transfusion, remains unclear. We sought to evaluate the impact of transfusion on mortality in patients with AMI.
Methods: Using the nationwide FAST-MI 2005 AMI registry, we recorded anemia on admission, Thrombolysis in Myocardial Infarction major or minor bleeding, and transfusions during hospital stay. Multivariable analyses were performed to identify independent predictors of in-hospital and 5-year mortality. Cohorts of patients matched for propensity to receive transfusion were compared.
Results: Among 3541 patients with AMI, 827 (23.4%) had anemia on admission, 114 (3.2%) had minor or major bleeding, and 151 (4.3%) underwent transfusion. After multivariable analysis, both anemia and bleeding were independently associated with 5-year mortality (hazard ratio [HR] 1.4, 95% CI 1.2-1.6 and HR 1.4, 95% CI 1.1-1.8, respectively), whereas transfusion did not appear to be an independent predictor (HR 1.1, 95% CI 0.8-1.5). Mortality at 5 years did not differ between cohorts matched for propensity to receive transfusion.
Conclusions: In this cohort, anemia on admission and bleeding during hospitalization were both associated with increased 5-year mortality in patients with myocardial infarction. Conversely, transfusion per se was not associated with lower survival. Further work is needed to clarify the optimal transfusion strategy in patients with bleeding or anemia and myocardial infarction.
Copyright © 2015 Elsevier Inc. All rights reserved.