Patients undergoing cardiac surgery are among the highest recipients of red-cell transfusion. Investigators in the randomized, controlled, noninferiority TRICS III trial aimed to determine whether a restrictive red-cell transfusion strategy in patients undergoing cardiac surgery with cardiopulmonary bypass could achieve similar outcomes to a more liberal transfusion surgery. Patients assigned to the restrictive strategy received red-cell transfusion if their haemoglobin levels were <7.5 g/dl, whereas patients assigned to the liberal strategy received a transfusion if their haemoglobin levels were <9.5 g/dl in the intensive care unit (ICU) or <8.5 g/dl in a non-ICU ward. In total, 4,860 patients were included in the per-protocol analysis (n = 2,430 in each group). Among the patients in the restrictive-threshold group, 11.4% had a primary composite outcome event (composite of all-cause death, myocardial infarction, stroke, or renal failure), compared with 12.5% in the liberal-threshold group (OR 0.90, 95% CI 0.76–1.07, P < 0.001 for noninferiority). The investigators concluded that a restrictive red-cell transfusion strategy was noninferior to a more liberal strategy with regard to major death and disability in patients undergoing cardiac surgery.