By Michael Crawford, MD, Editor

SYNOPSIS: A randomized trial of a restrictive blood transfusion strategy (hemoglobin < 8 g/dL; goal 8-10 g/dL) vs. a more liberal strategy (hemoglobin < 10 g/dL; goal > 11 g/dL) in patients with acute myocardial infarction and anemia showed the restrictive strategy is noninferior to the liberal strategy for preventing the primary outcome of death, reinfarction, stroke, or emergency revascularization.

SOURCE: REALITY trial presented by Philippe G. Steg at the European Society of Cardiology Virtual Congress, Sept. 1, 2020.

Which patients with cardiac disease will benefit from blood transfusion for anemia is controversial. The authors of the REALITY trial randomized 666 patients with acute myocardial infarction (AMI) and anemia to a restrictive blood transfusion strategy (hemoglobin [Hgb] < 8 g/dL; goal Hgb of 8-10 g/dL (n = 342) or a more liberal strategy (Hgb < 10 g/dL; goal Hgb > 11 g/DL (n = 324). The strategies were maintained until hospital discharge or 30 days, whichever came first. Follow-up was for 30 days after AMI. The patient’s mean age was 77 years, and 43% were women.

Both those with and without ST elevation were included if their last ischemic symptoms were less than 48 hours before admission. An elevated troponin also was required. Anemia was defined as Hgb less than 10 g/dL but greater than 7 g/dL. Generally, less than 7 g/dL is regarded as a criterion for transfusion. Patients with cardiogenic shock, elective PCI- or CABG-associated MI, hematologic disease, transfusion in the 30 days preceding the AMI or massive bleeding compromising the patient’s prognosis were excluded. The primary composite endpoint was all-cause death, reinfarction, stroke, and emergency revascularization prompted by ischemia. Of note, 70% of patients experienced a non-ST elevation AMI; of those, 80% underwent coronary angiography. Of the coronary angiography group, 59% received a PCI and 4% underwent coronary bypass surgery.

The mean units of packed red blood cells per patient was similar for the restrictive vs. liberal strategies (2.9 vs. 2.8). The primary outcome occurred in 11% of the restrictive group and 14% of the liberal group (HR, 0.77; 95% CI, 0.50-1.18; P < 0.05 for noninferiority and P = 0.22 for superiority). All-cause mortality was 5.6% vs. 7.7%, recurrent MI was 2.1% vs. 3.1%, and emergency revascularization was 1.5% vs. 1.9% (all P = NS). Among secondary outcomes, a restrictive strategy led to lower infection rates (0% vs. 1.5%; P = 0.03) and lower acute lung injury rates (0.3% vs. 2.2%; P = 0.03). The authors concluded that a restrictive blood transfusion strategy in patients with AMI and anemia is noninferior regarding 30-day outcomes vs. a more liberal strategy.

COMMENTARY

Several previous studies in hospitalized medicine patients revealed a Hgb cutoff of 7 g/dL for blood transfusions provides the ideal equipoise regarding outcomes. Thus, many hospitals made this a policy to save money on blood transfusion costs. However, most of those studies included few (if any) patients with acute ischemic heart disease syndromes. Accordingly, there was a concern that ischemic myocardium might perform better with higher Hgb levels. After all, the myocardium cannot extract more oxygen from the blood when it is needed as many other organs can. The myocardium is solely dependent on the oxygen-carrying capacity and the volume of blood it receives. Hence, many argued transfusions should be performed at higher cutoff values (e.g., 10 g/dL). The results of small observational studies of heart failure and postoperative patients supported this hypothesis. Therefore, this randomized study from Europe is important.

The prior studies and the ischemic myocardium hypothesis must have influenced the REALITY trial investigators’ study design, since they chose a cutoff value of 8 g/dL for the restrictive group rather than 7 g/dL. The liberal group cutoff was 10 g/dL, which would be supported by some of the previous studies and is the most liberal policy recommended. The lack of superiority of the liberal strategy, the reduced blood utilization, and lower costs of the restrictive strategy make a strong argument for the restrictive policy at a Hgb cutoff of 8 g/dL in AMI patients. Whether this should be the policy for other forms of acute heart disease is unclear. The results of other observational studies of acute heart failure patients with anemia have suggested that 9 g/dL or 10 g/dL may be the ideal cutoff in such patients.

We have all seen heart failure patients markedly improve after anemia is treated, but the target Hgb level to transfuse and achieve with transfusion is poorly defined. In the REALITY trial, there was a trend toward more acute kidney injury in the restrictive group (9.7% vs. 7.1%; P = 0.24). One can imagine this could be a bigger problem in heart failure patients, especially those with ischemic cardiomyopathy. I hope the REALITY trial researchers are planning a similar study of an acute heart failure cohort.