BackgroundPostoperative cardiac arrest (CA) with or without need for extracorporeal cardiopulmonary resuscitation (ECPR) is one of the most significant complications in the early postoperative period after pediatric cardiac operation. The objective of this study was to develop and to validate a predictive model of postoperative CA with or without ECPR. MethodsIn this retrospective cohort study, we reviewed data from patients who underwent cardiac surgery with cardiopulmonary bypass (CPB) between July 20, 2020, and December 31, 2021. Variables included demographic data, presence of preoperative risk factors, The Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery mortality categories, perioperative data, residual lesion score (RLS), and vasoactive-inotropic score (VIS). We used multivariable logistic regression analysis to develop a predictive model. ResultsThe incidence of CA with or without ECPR was 4.4% (n = 24/544). Patients who experienced postoperative CA with or without ECPR were younger (age, 130 [54-816.5] days vs 626 [127.5-2497.5] days; P < .050) and required longer CPB (253 [154-332.5] minutes vs 130 [87-186] minutes; P < .010) and cross-clamp (116.5 [75.5-143.5] minutes vs 64 [30-111] minutes; P < .020) times; 37.5% of patients with an outcome had at least 1 preoperative risk factor (vs 16.9%; P < .010). Our multivariable logistic regression determined that the presence of at least 1 preoperative risk factor (P = .005), CPB duration (P = .003), intraoperative residual lesion score (P = .009), and postsurgery vasoactive-inotropic score (P = .010) were predictors of the incidence of CA with or without ECPR. ConclusionsWe developed a predictive model of postoperative CA with or without ECPR after congenital cardiac operation. Our model performed better than the individual scores and risk factors.