In-silico modelling of the deployment of left atrial appendage occluder devices: A sensitivity analysis

Abstract

Percutaneous left atrial appendage (LAA) occlusion interventions are an alternative to oral anticoagulants for cardioembolic stroke prevention in atrial fibrillation patients. Preoperative insights into how the occluder device will interact with the patient’s LAA anatomy are limited. Choosing the optimal device type, size and position remains challenging, leading to procedural complications such as device embolization, device-related thrombosis and leaks. Computational tools could provide a personalised preoperative planning to minimise possible deployment problems. The present study aims at finding the principal steps for the in-silico deployment of LAA occluders in patient-specific anatomies considering the myocardial tissue of the LAA a deformable body. The most relevant factors in finite element modelling (FEM) to ensure a realistic interaction between the device and the anatomy were studied, including mesh discretization, material properties and boundary conditions. A sensitivity analysis study was designed and executed on a geometrical model built based on Computed Tomography images of a patient who underwent a LAA occlusion at Hospital de la Santa Creu i Sant Pau (Barcelona, Spain). Due to the high computational burden associated to the computational solid mechanics simulations required in this project, the study was performed using codes optimised for High-Performance Computing (HPC) infrastructures, including the Alya FEM solver developed by the Barcelona Supercomputing Center, which also gave access to HPC facilities to run the experiments. The current project has led to some relevant conclusions regarding the geometry characteristics, methodology and material selection to model LAA occluder deployment. It has set up the basis for future developments where a realistic device deployment could be offered as an option to clinicians for a better pre-procedural planning.