Novel Computational Model for Recapitulating Patent Ductus Arteriosus Stenting Interventions.

Abstract

Patent Ductus Arteriosus (PDA) stenting is a minimally invasive procedure used to maintain pulmonary blood flow in patients with ductal-dependent cyanotic congenital heart disease. However, because of its high complexity, anatomic variability, and frequent need for reinterventions, improved pre-procedural planning is necessary. In this retrospective study, we developed a physics-based computational framework to simulate PDA stenting using patient-specific data. We applied this method to two patients from a single center: Patient 1 had a type I PDA with a single stent implantation, while Patient 2 had a type II PDA and underwent placement of two stents. We segmented pre-procedural CT scans, modeled guidewire tracking and the bent pre-deployment configurations of the angioplasty balloon and stent, and simulated stent deployment within patient-specific PDA anatomies. Quantitative validation against post-procedural segmentations showed an average distance error of less than 1 mm, demonstrating high accuracy in replicating real-world outcomes. The pipeline effectively captured key mechanical interactions among the stent, balloon, guidewire, and PDA, highlighting phenomena, such as PDA straightening, changes in diameter and orientation, and the displacement of surrounding vasculature during deployment. Future work will integrate semiautomatic tools to predict the best-suited procedural parameters, including stent length, diameter, positioning, and vascular access and predicted risk of complications. Ultimately, our goal is to develop a predictive platform that enhances clinical decision-making, optimizes procedural efficiency, and reduces complications and reinterventions, thereby improving outcomes for pediatric patients undergoing PDA stenting.