By Michael H. Crawford, MD, Editor

SOURCE: Murphy GJ, et al. Liberal or restrictive transfusion after cardiac surgery. N Engl J Med 2015;372:997-1008.

Due to the cost of blood transfusions and the lack of data supporting liberal transfusion policies, newer guidelines recommend more restrictive transfusion thresholds (e.g., hemoglobin < 7 g/dL). However, many believe cardiac surgery is an exception due to the myocardium’s high oxygen requirement. Thus, these investigators performed the Transfusion Indication Threshold Reduction (TITRe2) trial to determine if a restrictive transfusion threshold would reduce postoperative morbidity and health care cost. In 17 United Kingdom (UK) cardiac surgery centers, 2003 patients were randomized to a restrictive threshold of < 7.5g/dL or a liberal threshold of < 9g/dL, with transfusions administered to help the hemoglobin above these thresholds. The patients were followed for 3 months for the primary outcome of serious infections or an ischemic event in the brain, heart, gut, or kidney. Median age of the patients was 70 years old, and 69% were men. Most had coronary bypass (41%) or valve surgery (31%). One-quarter of the patients received a transfusion before being enrolled. A median of one unit of red cells was transfused in the restrictive group and two in the liberal group. Transfusions were given to 64% of the restrictive group and 95% of the liberal group.

The primary outcome was seen in 35% of the restrictive group and 33% of the liberal group (P = NS). The rates of pulmonary complications and the length of ICU stay did not differ between groups. There were more deaths in the restrictive group (4.2% vs 2.6%; HR, 1.64; 95% CI, 1.0-2.7, P = 0.045). Overall costs were no different between the two groups. The authors concluded that a restrictive transfusion policy after cardiac surgery was not superior to a more liberal policy, and overall costs were similar.

COMMENTARY

Currently, transfusion rates in the UK and the United States are highly variable between cardiac surgery centers (8-93%). Part of this wide variation in practice is the controversy in the literature. Observational studies that showed higher risks of mortality and morbidity with liberal transfusion policy were confounded by patient characteristics that influenced transfusion decisions. The few comparative trials done lacked statistical power. This study largely avoided these pitfalls and failed to prove the hypothesis that bad outcomes and costs would be higher with a liberal transfusion policy. Concerning was the results of the secondary outcome variable of death, which were almost twice as high in the restrictive group. Also, this difference persisted despite sensitivity analyses. In addition, patients who already got a transfusion before being enrolled in the trial were excluded. When rising creatinine was added to the primary endpoint, the results favored a liberal policy.

The more liberal approach is clinically plausible when you consider that ischemic myocardium needs oxygen delivery, which makes the cardiac surgery setting different from other settings with blood loss. Thus, I believe allowing clinicians to use their own judgment with cardiac surgery patients, rather than being constrained by hospital policy, makes the most sense for now.