This randomised controlled study assessed whether continuous positive airway pressure (CPAP) of 10 cmH2O during cardiopulmonary bypass improves post-bypass oxygenation in children compared with no CPAP during bypass. We studied children with a ventricular septal defect. CPAP of 10 cmH2O was applied during bypass in the CPAP group (n=24), whereas the lungs were left deflated in the control group (n=20). In both groups, an alveolar recruitment maneuver was performed by applying positive pressure of 30 to 40 cmH2O for five seconds before weaning from bypass. Postoperative ventilation had the peak inflation pressure set to produce an expired tidal volume of 8 ml/kg with positive end expiratory pressure of 5 cmH2O. Arterial blood gas and haemodynamic measurements were performed at skin incision, five minutes after weaning from bypass, five minutes after chest closure and four hours after arrival in the intensive care unit. In four children CPAP was discontinued because it adversely affected the operating field. There was no difference in demographic characteristics, haemodynamic data, bypass time and operation time. No difference was observed between the groups with respect to pH, PaO2, P(A-a) DO2, PaCO2, and ETCO2 at each time. Variability in the data was greater than expected, leading to a decrease in the expected power of the study. CPAP at 10 cmH2O during bypass was not found to improve the post-bypass oxygenation as compared with leaving the lung deflated during bypass in children undergoing ventricular septal defect repair who had an alveolar recruitment maneuver at the end of bypass.