Patients undergoing cardiac surgery often receive red-cell transfusions, along with the associated risks and costs. Early intraoperative normovolemic hemodilution (i.e., acute normovolemic hemodilution [ANH]) is a blood-conservation technique that entails autologous blood collection before initiation of cardiopulmonary bypass and reinfusion of the collected blood after bypass weaning. More data are needed on whether ANH reduces the number of patients receiving allogeneic red-cell transfusion.

In a multinational, single-blind trial, we randomly assigned adults from 32 centers and 11 countries who were undergoing cardiac surgery with cardiopulmonary bypass to receive ANH (withdrawal of ≥650 ml of whole blood with crystalloids replacement if needed) or usual care. The primary outcome was the transfusion of at least one unit of allogeneic red cells during the hospital stay. Secondary outcomes were death from any cause within 30 days after surgery or during the hospitalization for surgery, bleeding complications, ischemic complications, and acute kidney injury.

A total of 2010 patients underwent randomization; 1010 were assigned to ANH and 1000 to usual care. Among patients with available data, 274 of 1005 (27.3%) in the ANH group and 291 of 997 (29.2%) in the usual-care group received at least one allogeneic red-cell transfusion (relative risk, 0.93; 95% confidence interval, 0.81 to 1.07; P=0.34). Surgery for postoperative bleeding was performed in 38 of 1004 patients (3.8%) in the ANH group and 26 of 995 patients (2.6%) in the usual-care group. Death within 30 days or during hospitalization occurred in 14 of 1008 patients (1.4%) in the ANH group and 16 of 997 patients (1.6%) in the usual-care group. Safety outcomes were similar in the two groups.

Among adults undergoing cardiac surgery, ANH did not reduce the number of patients receiving allogeneic red-cell transfusion.