Related Experiment Videos
The Pediatric Pulmonary Hypertension International Risk Score: A Prediction Model for Outcomes Using Machine Learning
Megan Griffiths1, Bhargava K Chinni2, Chantal Lokhorst3
1Heart Center, Children's Health, Division of Cardiology, Department of Pediatrics, UT Southwestern Medical Center, Dallas, TX (M.G.).
Insights
A new data-driven model accurately predicts 1-year risk in pediatric pulmonary hypertension (PH). This tool aids clinical decisions for children with PH, improving care by stratifying risk effectively.
Area of Science:
- Pediatric Cardiology
- Pulmonary Hypertension Research
- Biostatistics and Predictive Modeling
Background:
- Pediatric pulmonary hypertension (PH) care lacks specific risk prediction models, often relying on adult data or expert opinion.
- Current models are limited, hindering effective risk stratification and treatment guidance for pediatric PH patients.
- A need exists for validated, data-driven tools tailored to the unique characteristics of pediatric PH.
Purpose of the Study:
- To develop and externally validate a data-driven 1-year risk prediction model for pediatric pulmonary hypertension (PH).
- To identify key predictors for a 1-year composite outcome including death, transplant, or specific surgical interventions.
- To provide a robust tool for clinical risk stratification in pediatric PH.
Main Methods:
- A cohort of 345 pediatric PH patients (World Symposium on Pulmonary Hypertension groups 1 and 3) was used for model development (80% training, 20% testing).
- BorutaSHAP and random forest identified 16 predictors from 176 variables; extreme gradient boosting modeled the 1-year outcome.
- External validation was performed using the Dutch National Registry for Pulmonary Hypertension in Childhood (n=155) and the Spanish Registry of Pediatric Pulmonary Hypertension (n=327).
Main Results:
- The model achieved an area under the receiver operating characteristic curve (AUC) of 0.90 in the test cohort, with a 99% negative predictive value.
- External validation demonstrated good performance with AUCs of 0.76 and 0.77 in the Dutch and Spanish registries, respectively.
- Kaplan-Meier analysis confirmed significant differentiation of outcomes among risk groups defined by the model.
Conclusions:
- A multicenter, validated model offers reliable 1-year risk prediction for pediatric PH (World Symposium on Pulmonary Hypertension groups 1 and 3).
- This data-driven tool addresses a critical gap in pediatric PH care by enabling accurate clinical risk stratification.
- The model serves as a robust instrument to guide therapy decisions and improve outcomes for children with PH.
Background:
Risk prediction is fundamental to pulmonary hypertension (PH) guideline-based care, yet pediatric-specific risk prediction models remain limited, relying primarily on single predictors, expert opinion, or application of adult models to children. The authors developed and externally validated a data-driven 1-year risk prediction model for pediatric PH.
Methods:
Pediatric patients with PH (n=345; World Symposium on Pulmonary Hypertension groups 1 and 3) enrolled in the Pediatric Pulmonary Hypertension Network Registry (2014-2020; 50.4% male; median age, 4.9 years [interquartile range, 1.9-10.3]) were split into training (80%) and test cohorts (20%). The Dutch National Registry for Pulmonary Hypertension in Childhood (n=155 [1993-2020]) and the Spanish Registry of Pediatric Pulmonary Hypertension (n=327 [2009-2023]) were used for external validation. From 176 variables, BorutaSHAP feature selection with random forest identified 16 predictors for a 1-year outcome of time to death, transplant, Potts shunt, or atrial septostomy, modeled using extreme gradient boosting. Performance was assessed with the area under the receiver operating characteristic curve, confusion matrices, calibration, and Kaplan-Meier event-free survival.
Results:
The final model achieved an area under the receiver operating characteristic curve of 0.90 (0.79-0.97) and 99% (96%-99%) negative predictive value in testing, dividing participants into 3 groups with strong outcome discrimination. External validation showed an area under the receiver operating characteristic curve of 0.76 (Dutch National Registry for Pulmonary Hypertension in Childhood, 0.70-0.81) and 0.77 (Spanish Registry of Pediatric Pulmonary Hypertension, 0.73-0.82) with negative predictive values of 93% (93%-97%) and 96% (93%-97%), respectively. Kaplan-Meier analysis significantly differentiated outcomes by risk group.
Conclusions:
This multicenter, validated model provides good 1-year risk prediction in pediatric PH across World Symposium on Pulmonary Hypertension groups 1 and 3, providing a robust tool for clinical risk stratification to guide therapy and addressing a gap in pediatric PH care.