Risk Stratification fi cation and Prognosis of Pulmonary Arterial Hypertension Associated with Congenital Heart Disease

Background: Current guidelines for managing pulmonary arterial hypertension (PAH) recommend a risk stratification approach. However, the applicability and accuracy of these strategies for PAH associated with congenital heart disease (PAH-CHD) require further validation. This study aims to validate the reliability and predictive accuracy of a simplified stratification strategy for PAH-CHD patients over a three-year follow-up. Additionally, new prognostic variables are identified and novel risk stratification methods are developed for assessing and managing PAH-CHD patients. Methods: This retrospective study included 126 PAH-CHD patients. Clinical and biochemical variables across risk groups were assessed using Kruskal-Wallis and Fisher's exact tests. Independent risk factors were identified using ordered logistic regression, while Kaplan-Meier and Cox proportional hazards regression analyses evaluated their impact on all-cause mortality. A new stratification model for the PAH-CHD population was constructed based on these analyses. Results: Significant survival differences across stratified risk groups were observed (p < 0.001), validating the effectiveness of the simplified risk stratification method in PAH-CHD patients. Prothrombin activity was a strong independent predictor of adverse outcomes of PAH-CHD patients (Hazard ratio 0.95, p < 0.001, C-index 0.70). A model combining N-terminal pro-brain natriuretic peptide, prothrombin activity, albumin, and right atrial area achieved an area under the curve of 0.89 and a C-index of 0.85. Conclusions: The simplified risk stratification method is applicable to PAH-CHD patients. Prothrombin activity is a strong independent predictor of adverse outcomes. A comprehensive risk stratification approach, incorporating both established and novel biomarkers, enhances accessibility and offers predictive efficacy during follow-up for PAH-CHD patients, comparable to established models.