A novel computational framework to predict patient-specific hemodynamics after TEVAR: Integration of structural and fluid-dynamics analysis by image elaboration

Although Thoracic EndoVascular Aortic Repair (TEVAR) is a consolidated procedure to treat thoracic aortic diseases, it still has relevant complications mainly related to suboptimal wall apposition of the stent-graft, impairing the post-operative hemodynamics and the clinical outcomes. Accurate stent-graft sizing and patient selection are the key aspects to minimize drawbacks. Unfortunately, current TEVAR planning is only based on geometrical measurements performed on static images, completely neglecting the biomechanical interplay between the stent-graft and the aorta. Despite an extensive literature dealing with bioengineering simulation of endovascular implants, studies on the prediction of the post-TEVAR hemodynamics based on both pre-operative patient-specific aortic anatomy and stent-graft mechanical features are still missing. The present study aims at providing a realistic and robust computational framework to support TEVAR planning in the clinical practice by predicting the post-operative hemodynamics given a selected stent-graft model and pre-operative medical images of the aorta to be treated. A novel approach based on a distance image aimed at transforming the result of the structural analysis of stent-graft deployment in a volume mesh suitable for a computational study of the post-TEVAR hemodynamics is presented. The study discusses two clinical cases as illustrative examples of the framework application and, for one of the two cases, a comparison of the predicted hemodynamics with a simulation based on real post-operative images is shown. Such a comparison proves that the proposed computational framework is able to capture the main hemodynamic aspects related to the stent-graft implant. In particular, the use of simulations has confirmed the unsuitability for the endovascular repair of one of the two patients due to the short proximal landing zone, leading to a high risk of the so-called bird-beak. The proposed computational framework is shown to be a useful tool to support planning of elective TEVAR, especially in those borderline cases when the sole geometrical analysis of static images is not exhaustive.