Neonatal heart failure is a severe condition with high mortality, posing a significant burden on families and healthcare systems. Early prediction of mortality risk is crucial for improving outcomes. To develop and validate a predictive model and scoring system for in-hospital mortality within 28 days in neonates with heart failure. This multicenter retrospective study included 579 neonates from the First Affiliated Hospital of Xinjiang Medical University (training and internal validation sets) and 118 from Beijing Anzhen Hospital (external validation set). Data included sociodemographic characteristics, clinical symptoms, medical history, medication use, laboratory results, and outcomes. Lasso was used for variable selection from a large set of candidates, followed by logistic regression on the selected variables to build the final model and scoring system. Lasso regression identified 20 key variables. This study found that fibrinogen < 2 g/L, poor postnatal response, and oliguria were associated with an increased risk of death in neonates with heart failure, while the use of digoxin, cedilanid, dopamine, and epinephrine reduced the risk of death. The model showed AUC values of 0.87 (95% CI: 0.82-0.91), 0.83 (95% CI: 0.77-0.90), and 0.85 (95% CI: 0.77-0.93) in the training, internal validation, and external validation sets, respectively. The scoring system effectively categorized patients into low, medium, and high-risk groups. This study established a predictive model and scoring system for in-hospital mortality risk in neonates with heart failure, enabling early identification of high-risk infants and guiding individualized treatment strategies to improve outcomes.
Wei et al. (2026) studied this question.