Key result
Seven-predictor nomogram discriminates non-PFO congenital heart disease from PFO with ~0.73 AUC.
Why the study?
Previous studies of congenital heart disease often combined patent foramen ovale with structural cardiac malformations, potentially increasing phenotypic heterogeneity when evaluating maternal gestational exposures and child or family characteristics.
Do maternal gestational exposures and child/family characteristics predict non-PFO structural congenital heart disease compared to patent foramen ovale in children from Hainan Province?
Case-Control (n=208)
Yes
Do maternal gestational exposures and child/family characteristics predict non-PFO structural congenital heart disease compared to patent foramen ovale in children from Hainan Province?
By separating PFO controls from structural CHD cases, this study identified seven key maternal and familial predictors of non-PFO CHD and developed a preliminary risk-stratification nomogram (AUC 0.732), though external validation in general populations is required.
Answers are grounded in this paper and cite their sources.
May aid preliminary differentiation of non-PFO CHD from PFO in children; leaves open clinical utility pending prospective validation.
Background Congenital heart disease (CHD) is a major cause of infant morbidity and mortality. Previous studies often combined patent foramen ovale (PFO) with structural cardiac malformations, potentially increasing phenotypic heterogeneity. Objectives To identify maternal gestational exposures and child/family characteristics associated with non-PFO CHD and develop a preliminary prediction model. Methods This multicenter retrospective case-control study included 208 children from Hainan Province: 130 with non-PFO CHD and 78 with PFO. Fifty-five candidate variables were assessed. Least absolute shrinkage and selection operator (LASSO) regression was used for feature selection, followed by five machine learning models. SHapley Additive exPlanations (SHAP) analysis was used to interpret feature importance, and a nomogram was constructed. Results LASSO selected eight features, of which seven had nonzero global SHAP importance: child age in months, maternal protein intake frequency during pregnancy, history of CHD in other offspring, folic acid supplementation, frequency of family conflicts during pregnancy, maternal preconception smoking, and first-trimester hazardous substance exposure. Linear discriminant analysis (LDA) showed the best test-set performance, with an area under the receiver operating characteristic curve (AUC) of 0.702 and a Brier score of 0.232. The nomogram achieved an AUC of 0.732, and decision curve analysis indicated potential net benefit across the evaluated threshold range. Conclusions Candidate predictors associated with non-PFO CHD were identified, and an internally evaluated nomogram was developed for preliminary risk stratification. External validation is required before clinical use.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) conducted a case-control in Non-PFO congenital heart disease (n=208). Maternal gestational exposures and child/family characteristics vs. Patent foramen ovale (PFO) controls was evaluated on Non-PFO structural congenital heart disease classification (AUC 0.702 (LDA test set); AUC 0.732 (Nomogram)). Linear discriminant analysis and a 7-predictor nomogram discriminated non-PFO congenital heart disease from patent foramen ovale in children with AUCs of 0.702 and 0.732, respectively.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: