Red Cell Transfusion in Critically Ill and Noncritically Ill Patients With Acute Myocardial Infarction and Anemia-A Subanalysis of the Myocardial Ischemia and Transfusion (MINT) Trial.

Abstract

Anemia is common in critically ill patients. In prior trials, a restrictive transfusion strategy was at least as effective as, and possibly superior to, a liberal strategy in critically ill adult patients. However, the Transfusion Requirements in Critical Care (TRICC) trial suggested that a restrictive strategy might be harmful in critically ill patients with ischemic heart disease. Our objective was to compare outcomes for restrictive and liberal transfusion strategies in critically ill and noncritically ill patients with acute myocardial infarction (AMI) and anemia. Subgroup analysis of a parallel, two-arm randomized, controlled trial. One hundred forty-four sites in the United States, Canada, France, Brazil, New Zealand, and Australia. Adult patients with AMI and anemia enrolled in the Myocardial Ischemia and Transfusion (MINT) trial. Patients being treated in an ICU at randomization were considered critically ill. Patients were randomly assigned to a restrictive (transfuse when hemoglobin < 7-8 g/dL) or liberal (transfuse when hemoglobin < 10 g/dL) transfusion strategy. We determined the effect of transfusion strategy on death and the composite of death/myocardial infarction (MI) at 30 days among critically ill and noncritically ill patients. Of 3504 patients enrolled in MINT, 1679 (47.9%) were critically ill. At 30 days, death (11.9% vs. 6.5%) and death/MI (18.9% vs. 12.7%) occurred more frequently in critically ill patients compared with noncritically ill patients. A restrictive, compared with a liberal, transfusion strategy resulted in a nonsignificant increase in the risk of death by 9% (risk ratio [RR], 1.09; 95% CI, 0.77-1.54) and death/MI by 9% (RR, 1.09; 95% CI, 0.85-1.38) in noncritically ill patients and a nonsignificant increase in the risk of death by 24% (RR, 1.24; 95% CI, 0.95-1.61; interaction p = 0.55) and death/MI by 21% (RR, 1.21; 95% CI, 0.99-1.47; interaction p = 0.52) in critically ill patients. There is no evidence that the overall MINT trial results, which generally favor a liberal transfusion strategy in AMI patients, apply differently to noncritically ill and critically ill patients.