OBJECTIVES: The reduction of postoperative acute kidney injury in patients undergoing cardiopulmonary bypass surgery using an oxygen delivery-guided perfusion strategy (oxygen delivery strategy) for cardiopulmonary bypass management compared with a fixed flow perfusion (conventional strategy) remains controversial. The purpose of this study was to determine whether a oxygen delivery strategy would reduce the incidence of postoperative acute kidney injury in patients undergoing cardiopulmonary bypass surgery. METHODS: We randomly enrolled 300 patients undergoing cardiopulmonary bypass surgery. Patients were randomly assigned to a oxygen delivery strategy (maintaining a oxygen delivery index value >300 mL/min/m2 through pump flow adjustments during cardiopulmonary bypass) or a conventional strategy (a target pump flow was determined on the basis of the body surface area). The primary end point was the development of acute kidney injury. Secondary end points were the red blood cell transfusion rate and number of red blood cell units, intubation time, postoperative length of stay in the intensive care unit and the hospital, predischarge estimated glomerular filtration rate, and hospital mortality. RESULTS: Acute kidney injury occurred in 20 patients (14.6%) receiving the oxygen delivery strategy and in 42 patients (30.4%) receiving the conventional strategy (relative risk, 0.48; 95% confidence interval, 0.30-0.77; P = .002). The secondary end points were not significantly different between strategies. In a prespecified subgroup analysis of patients who had nadir hematocrit less than 23% or body surface area less than 1.40 m2, the oxygen delivery strategy seemed to be superior to the conventional strategy and the existence of quantitative interactions was suggested. CONCLUSIONS: An oxygen delivery strategy for cardiopulmonary bypass management was superior to a conventional strategy with respect to preventing the development of acute kidney injury.