Development and validation of a nomogram prediction model for mild pulmonary arterial hypertension risk in children with left-to-right shunt congenital heart disease.

Mild pulmonary arterial hypertension (PAH) represents an early and potentially reversible complication stage in children with congenital structural heart disease (CHD). Early identification of high-risk children is crucial for improving prognosis. This study aimed to develop and validate a nomogram model for predicting the risk of mild PAH in children with left-to-right shunt CHD. This single-center retrospective study enrolled 340 children diagnosed with CHD who underwent cardiac catheterization at the Department of Pediatrics, The First Affiliated Hospital of Guangxi Medical University, between January 2005 and April 2022. Patients were divided into a training set (n = 237) and an internal validation set (n = 103) at a 7:3 ratio using stratified random sampling. In the training set, predictors were selected from 35 candidate variables using least absolute shrinkage and selection operator regression combined with 10-fold cross-validation, and the final prediction model was constructed using multivariate logistic regression. Model performance was comprehensively evaluated through Bootstrap internal validation (1000 resampling iterations) and on the independent validation set, including discrimination (area under the curve [AUC]), calibration (calibration curve), and clinical utility (decision curve analysis). The model ultimately identified 5 independent predictive factors: loud P2, decreased exercise tolerance, defect size, left ventricular systolic diameter Z SCORE, and age. The nomogram constructed based on these factors achieved a Bootstrap-corrected AUC of 0.871 (95% confidence interval = 0.824-0.919) in the training set and an AUC of 0.861 (95% confidence interval = 0.769-0.938) in the independent validation set. The calibration curve indicated high consistency between predicted probabilities and actual observed values. Decision curve analysis demonstrated that the model provided clinical net benefit across a wide range of threshold probabilities. This study successfully developed and validated a well-performing nomogram model for predicting the risk of mild PAH in children with left-to-right shunt CHD. This model integrates easily accessible clinical, imaging, and demographic indicators, facilitating the early identification of high-risk children and providing a practical tool for personalized intervention and clinical decision-making.