Robert H. Hollis, MD; Brandon A. Singletary, MPH; James T. McMurtrie, BS; Laura A. Graham, MPH; Joshua S. Richman, MD, PhD; Carla N. Holcomb, MD; Kamal M. Itani, MD; Thomas M. Maddox, MD MSc; Mary T. Hawn, MD, MPH
JAMA Surg. Published online October 07, 2015. doi:10.1001/jamasurg.2015.3420
There are currently 8 large randomized controlled trials supporting restrictive transfusion triggers (7 or 8 g/dL) in a variety of different patients groups including: 1) critically ill ICU patients, 2) critically ill pediatric ICU patients, 3) postoperative cardiac surgery patient, 4) gastrointestinal bleeding patients, 5) traumatic brain injury patients, 6) elderly orthopedic patients with cardiovascular disease, and 7) patients with septic shock. There is however, a paucity of evidence supporting the ideal hemoglobin threshold for patients with ischemic coronary syndromes.
Hollis et. al set out to study transfusion thresholds in patients with coronary artery disease undergoing non-cardiac surgery who had a post-operative myocardial infarction (MI) in a retrospective cohort study. Of the 7,361 patients undergoing non-cardiac surgery, 371 (3.7%) had an MI. In MI patients who had a nadir hematocrit of 20%-24%, blood transfusion was protective against mortality (OR: 0.28; 95% CI: 0.13-0.64). On the other hand, in MI patients with a nadir hematocrit of 24%-30%, transfusion was not significantly associated with a change in mortality. Such findings support a restrictive transfusion trigger (7-8 g/dL) even in patients with acute MI.
Being a retrospective cohort study, there are of course some limitations. For example, the nadir hemoglobin was used as a surrogate for the true transfusion trigger, patients who received greater than 4 units intra-operatively or 4 units post-operatively were excluded, and the cohort only included patients from Veterans’ Affairs hospitals. Furthermore, the investigators could not determine the timing of, or a causal relationship between anemia, transfusion and myocardial infarction. Until an adequately powered randomized trial is conducted, we can only rely on such observational studies to determine best practices. This current study suggests that MI patients may not benefit from RBC transfusions when the nadir hemoglobin is > 8 g/dL.
Daniel J. Johnson BS, Steven M Frank MD
Department of Anesthesiology and Critical Care Medicine
Johns Hopkins Medical Institutions