Does the relation between red blood cell transfusion and mortality vary according to transfusion indication? A case–control study among patients with acute coronary syndromes

Abstract

Red blood cell (RBC) transfusions are common among patients with acute coronary syndromes (ACS). We sought to determine their impact on mortality according to indication. We conducted a case-control analysis of ACS patients admitted to a single coronary care unit. Transfused patients were matched to nontransfused patients for age, estimated glomerular filtration rate, and nadir hematocrit. Conditional logistic regression analysis was used to determine the independent relationship between RBC transfusion and inhospital mortality. Results were stratified according to the indication for transfusion. Of the 3190 patients admitted with ACS, 206 (6.4%) received at least one RBC transfusion. Of these, 103 patients were matched to 185 nontransfused patients. The most common indications for transfusion were nonspecific anemia (48%) and overt blood loss (42%). Transfused patients had a significantly lower ejection fraction and were more likely to present in Killip class IV, undergo pulmonary artery catheterization, and require an intra-aortic balloon pump. On univariate analysis, RBC transfusion was associated with a doubling of the risk of inhospital mortality [odds ratio: 2.0; 95% confidence interval: 1.0-3.9; P=0.046]. However, after adjustment for confounders, transfusion was no longer significantly associated with higher inhospital mortality (odds ratio: 1.8; 95% confidence interval: 0.6-5.1; P=0.3). On stratified analysis, RBC transfusion was not significantly associated with inhospital mortality among patients transfused for either nonspecific anemia or overt blood loss. Among coronary care unit patients with ACS, RBC transfusion does not seem to have a significant impact on inhospital mortality irrespective of the indication for transfusion.