Predicting Long-Term Childhood Survival of Newborns with Congenital Heart Defects: A Population-Based, Prospective...

Area of Science:

• Pediatrics
• Cardiology
• Public Health

Background:

• Congenital heart defects (CHDs) are the most common major congenital anomalies, affecting nearly 1% of births.
• CHDs present significant heterogeneity in severity, management, and etiology, necessitating precise classification for accurate prognostication and research.
• The Anatomic and Clinical Classification of CHD (ACC-CHD) was developed to standardize CHD coding and data analysis.

Purpose of the Study:

• To describe the long-term childhood survival rates of newborns diagnosed with CHDs.
• To develop and validate predictive models for infant mortality utilizing the ACC-CHD classification system.

Main Methods:

• A prospective, population-based cohort study (EPICARD) included 1881 newborns with CHDs, excluding those with chromosomal or other anomalies.
• Statistical analyses employed non-parametric and flexible parametric survival models.
• Model performance was evaluated using Harrell's C index and Royston-Sauerbrei RD2, with internal validation via bootstrap.

Main Results:

• The overall 8-year survival rate for isolated CHDs was 0.96 (95% CI [0.93-0.95]).
• Significant survival differences were observed across ACC-CHD categories, with rates ranging from 0.995 for interatrial/ventricular septal defects to 0.34 for functionally univentricular hearts.
• The ACC-CHD classification demonstrated strong predictive ability, with Harrell's C at 87% (alone) and 89% (full model), and RD2 at 45% (alone) and 50% (full model).

Conclusions:

• The ACC-CHD classification serves as a robust foundation for a highly discriminant survival model for newborns with CHDs.
• The classification system exhibits good predictive accuracy for 8-year survival.
• Individual outcome prediction for CHDs remains a complex clinical and statistical challenge.