The relicensing of aprotinin in Europe and Canada has stimulated discussions on its usefulness in paediatric cardiac surgery. To systematically evaluate the available evidence on the efficacy and safety of aprotinin in paediatric cardiac surgery. Systematic review of all randomised and observational studies comparing aprotinin with tranexamic acid, epsilon aminocaproic acid, placebo or no drug in paediatric cardiac surgery. Meta-analyses were performed on efficacy and safety outcomes. PubMed, Cochrane Central Register of Controlled Trials, Web of Science and Embase were searched from January 2000 to March 2021. Studies that enrolled children under 18 years undergoing cardiac surgery with cardiopulmonary bypass. Thirty-two studies enrolling a total of 63 894 paediatric cardiac procedures were included. Aprotinin significantly reduced total blood loss [mean difference -4.70 ml kg-1, 95% confidence interval (CI), -7.88 to -1.53; P = 0.004], postoperative transfusion requirements and the incidence of surgical re-exploration for bleeding [odds ratio (OR) 0.74, 95% CI, 0.56 to 0.97; P = 0.03]. Aprotinin had no effects on 30-day mortality (OR 1.02, 95% CI, 0.93 to 1.11; P = 0.73) and on other safety outcomes, except for the incidence of renal replacement therapy (RRT), which was significantly increased in patients given aprotinin (OR 1.29, 95% CI, 1.08 to 1.54; P = 0.006). Findings from observational and randomised controlled trials did not largely differ. A sub-group analysis in neonates showed that aprotinin significantly reduced packed red blood cell transfusions and the incidence of postoperative surgical re-exploration for bleeding and/or tamponade. When compared with lysine analogues, aprotinin was more effective at reducing bleeding and transfusion without increasing the risk of side effects. This meta-analysis suggests that aprotinin is effective and well tolerated in paediatric cardiac surgery. Given the large heterogeneity of the results and the risk of selection bias in observational studies, large randomised controlled trials are warranted.