|
|
Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromesRao SV, Jollis JG, Harrington RA, et al. JAMA 292: 1555-1562, 2004.Reviewer: Michael H. Wall, MD
Abstract Excerpt: This retrospective study examined the association between blood transfusion and 30-day all-cause mortality in patients with non-ST elevation acute coronary syndromes (ACS) who developed bleeding and/or anemia during hospitalization. Twenty-four thousand, one hundred, twelve patients in three large international multicenter prospective trials of patients with ACS were evaluated. The three trials were: 1) Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb, which randomized patients to receive intravenous heparin or hirudin, 2) Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT), which randomized patients to receive eptifibatide or placebo and, 3) Platelet IIb/IIIa Antagonism for Reduction of Acute Coronary Syndrome Events in Global Organization Network (PARAGON) B trials, which randomized patients to lamifiban or placebo. Two thousand, four hundred, one (10%) patients who received at least one unit of packed red blood cells were compared to 21,711 patients who were not transfused. Transfusion was not prospectively controlled in these trials, but transfusion data was collected prospectively in all three clinical trials. Patients who received transfusion were older, more often black, female, and thinner. The transfused group also was more likely to have diabetes, hypertension, hyperlipidemia, prior stroke, congestive heart failure, and be Killip class ≥ II. The transfused patients also had a significantly lower baseline hematocrit (39.9% vs 41.7%) and significantly lower nadir hematocrit (29% vs 37.6%) than the non-transfused group. Unadjusted 30-day mortality and 30-day mortality plus myocardial infarction rates were significantly higher in the transfused group. Following appropriate statistical adjustment for numerous factors, blood transfusion was associated with a hazard ratio for death and death plus myocardial infarction greater than 2.8 (95% confidence interval 2.45 - 3.18) in all comparisons. There was also a trend toward transfusion and increased 30-day mortality during the first seven days of hospitalization. Finally, when the adjusted predicted probability of 30-day mortality with and without transfusion by nadir hematocrit was examined, there was no significant association between transfusion and 30-day mortality with a nadir hematocrit &le 25%; however, at nadir hematocrits > 25%, transfusion was associated with increased 30-day mortality. The authors conclude that because this is a post hoc analysis of prospective trials, this study "should not be considered as evidence to change practice; rather, it should be considered as evidence that caution is warranted when making transfusion decisions." Reviewer's Comments: This interesting paper presents somewhat different results than a recent similar study by Wu et al1 that evaluated 30-day mortality in patients ≥ 65 years old with acute myocardial infarction and showed that (in contrast to Rao), transfusion was associated with a decreased mortality if the hematocrit on admission was ≤ 30%. The "comment" section of the paper by Rao, and the accompanying Editorial by Hebert and Fergusson2 should be "required reading" for physicians who care for patients with coronary artery disease. Briefly, the three main differences between the Wu and Rao studies are: 1) Wu used admission hematocrit and Rao used nadir hematocrit during hospitalization; 2) Wu used a Medicare database and Rao used prospectively collected data from three large international clinical trials; and 3) Wu excluded patients < 65 years old, those who bled within 48 hours of admission, or who underwent coronary artery bypass grafting (these patients were included in the Rao study). Considering all of the results or various clinical investigations, Hebert and Fergusson in their Editorial2 recommend blood transfusion for patients with ACS when the hemoglobin level is less than 7 g/dL; transfusion "may be beneficial" in patients with ACS when the hemoglobin is less than 8 g/dL; and accepting a hemoglobin level around 9 g/dL in such patients. In summary, this clinical investigation provides additional information to help guide blood transfusion therapy in patients with coronary artery disease. However, much remains unclear regarding this controversial topic, supporting the need for further well-controlled prospective clinical investigation. References:
Table of Contents:
© Society of Cardiovascular Anesthesiologists Questions or comments? Please send email to webmaster@scahq.org |