Development and validation of a predictive model for low cardiac output syndrome after surgical repair of tetralogy of Fallot.

Background: Low cardiac output syndrome (LCOS) remains a serious postoperative complication for children with tetralogy of Fallot (TOF), which often leads to increased morbidity and mortality. Early identification of LCOS and timely management are critical for better outcomes. This study aimed to develop a prediction model incorporating pre- and intraoperative characteristics for LCOS within 24 hours after surgical repair of TOF in children. Methods: The training dataset consisted of patients with TOF who underwent surgical repair in 2021, while the validation dataset consisted of patients in 2022. The univariable and multivariable logistic regression analyses were performed to recognize the risk factors of postoperative LCOS and a predictive model was established based on multivariable logistic regression analysis in the training dataset. Model predictive power was assessed using the area under the receiver operating characteristic curve (AUC). The calibration of the nomogram was evaluated and the Hosmer-Lemeshow test was used to assess the good fit. Decision curve analysis (DCA) was used to estimate the net benefits of the prediction model at different threshold probabilities. Results: In the multivariable logistic analysis, peripheral oxygen saturation, mean blood pressure, and central venous pressure were independent risk factors for postoperative LCOS. The AUC of the predictive model for postoperative LCOS was 0.84 (95% CI: 0.77-0.91) and 0.80 (95% CI: 0.70-0.90) in the training and validation datasets, respectively. The calibration curve for the probability of LCOS showed good agreement between the prediction by nomogram and actual observation both in the training and validation datasets. The Hosmer-Lemeshow test yielded nonsignificant statistics both in the training and validation datasets (P=0.69 and 0.54, respectively), indicating a good fit. The DCA revealed that more net benefits would be obtained by using the nomogram to predict LCOS than that achieved in either the treat-all-patient scheme or the treat-none scheme both in the training and validation datasets. Conclusions: This study is the first to incorporate pre- and intraoperative characteristics to develop a predictive model for LCOS after surgical repair of TOF in children. This model showed good discrimination, good fit and clinical benefits.