StressMAP: A Surrogate Model for Predicting Pacemaker Dependency in TAVI In-Silico Trials

Abstract

Transcatheter aortic valve implantation (TAVI) is a minimally invasive procedure for treating severe aortic stenosis. A key complication of this procedure is conduction abnormalities (CAs), caused by injury to the atrioventricular conduction system and may necessitate permanent pacemaker implantation (PPI). While computational models have examined TAVI-induced stress on the aortic root and its impact on CAs, they neglect balloon aortic valvuloplasty and fail to assess the risk of conduction system injury during the entire procedure. Additionally, the lack of mapping different types of CAs to PPI outcomes limits their application to in-silico trials of TAVI devices. In this work, we present StressMAP, a surrogate model for predicting pacemaker dependency in TAVI in-silico trials. We developed finite element analysis models of the balloon aortic valvuloplasty and the deployment of mechanically expanded TAVI devices. We implemented the models using the explicit numerical solver ANSYS LS-DYNA. Calculation verification was performed to confirm the spatio-temporal discretisation and solver parameters of the structural mechanics models. An automatic workflow was built to deploy the TAVI devices in patient-specific aortic root anatomies, reconstructed from the computed tomography angiography images from a clinical trial of the same device. Using the workflow, we will have performed TAVI deployment simulations in the patient-specific anatomies and developed StressMAP for predicting post-TAVI pacemaker dependency based on the stress distribution on the inner walls of aortic root anatomies. The model validation has been conducted by comparing simulated PPI predictions with the clinical trial outcomes at the individual and population levels.