Prediction model of postoperative mortality for congenital heart disease: evidence from two large-scale cohorts.

This study aimed to identify the risk factors for postoperative mortality in patients under 3 years of age with congenital heart disease (CHD) undergoing cardiac surgery and to develop a predictive nomogram for clinical use. We retrospectively analyzed data from 3409 patients under 3 years who underwent cardiopulmonary bypass surgery. Predictors including baseline characteristics, surgical details, and laboratory parameters were select via least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation. Variables with non-zero coefficients were incorporated into a multivariable logistic regression model to construct the final prediction nomogram. Six independent predictors of postoperative mortality were identified: preoperative serum creatinine (OR = 1.045, 95%CI 1.032-1.059), total protein (TP, OR =0.902[0.865-0.939]), glucose (OR = 1.201[1.095-1.317]), triglycerides (OR = 1.569[1.239-2.000]), as well as postoperative lactate (OR = 1.244[1.173-1.324]) and cystatin C (OR = 4.012[1.951-7.744]). The nomogram demonstrated excellent discrimination in the internal validation cohort (area under the receiver operating characteristic curve [AUC] = 0.952[0.932–0.972]) with good calibration. External validation yielded a AUC of 0.761 (0.621–0.901) and acceptable calibration. The decision curve analysis confirmed the clinical utility of the nomogram across a wide range of threshold probabilities. We developed and validated a nomogram incorporating sex routinely available clinical variables to stratify postoperative mortality risk in young CHD patients. Close monitoring and management of these predictors may help reduce postoperative mortality in this vulnerable population.