Résumé : BACKGROUND: Blood transfusion is frequently required in children undergoing cardiac surgery and is associated with altered postoperative outcome. This may be due to alterations in red blood cell properties related to the storage process. OBJECTIVE: To evaluate the effect of blood storage duration on postoperative morbidity and mortality in children undergoing cardiac surgery. DESIGN: A retrospective review of a paediatric cardiac surgery database. SETTING: Department of Anaesthesiology, Queen Fabiola Children's University Hospital, Brussels, Belgium. PARTICIPANTS: Children transfused with one or two units of blood in the perioperative period. INTERVENTIONS: None. MAIN OUTCOME MEASURES: Storage duration was used to allocate children to the Group 'Young' or the Group 'Old' (cut-off=7 days). The primary endpoint was a composite based on the incidence of hospital mortality and/or the incidence of at least one organ failure. RESULTS: From 1014 children in the database, 570 were included in the final analysis. One hundred and eighteen patients were included in the Group 'Young' [median (interquartile range, IQR) storage duration 6 (5 to 7) days] and 452 in the Group 'Old' [storage duration 14 (11 to 19) days]. No difference was found in mortality, length of ICU stay, mechanical ventilation duration, postoperative infection and major organ dysfunction. Duration of storage used as a continuous variable did not influence the incidence of the composite endpoint when evaluated by univariate or multivariate logistic regression analyses. CONCLUSION: Red blood cell storage duration did not influence postoperative morbidity and mortality in paediatric cardiac surgery patients transfused with one or two units of blood. © 2014 Copyright European Society of Anaesthesiology.