Computational hemodynamic study of endovascular stenting in patient-specific cerebral aneurysms

Abstract

Stroke is the leading cause of death after heart disease and cancer and the number one cause of long-term disability in America. About 80% of hemorrhagic stroke are produced by the rupture of cerebral aneurysms. Surgical clipping and coil embolization are the most common methods of treating these aneurysms. However, both these treatments have some limitations for wide neck aneurysms. Recently, there has been an increased interest in the use of stents as flow diverters. With an effective design, the blood flowing into the aneurysm can be disrupted thereby promoting intra-aneurysmal thrombus formation and thus preventing rupture. Modeling blood flow around these endovascular devices in intracranial aneurysms is important for designing better devices and to personalize and optimize endovascular stenting procedures in the treatment of these aneurysms. Numerous studies have been conducted in the past using animal models, patient-specific in-vitro models and idealized computational models. Nevertheless, all of them have significant limitations. The main disadvantage of using animal models is that they fail to replicate the variable anatomy of diseased human arteries. And the problem with in-vitro models is that they are not suitable for large population studies. Patient-specific image-based numerical models have demonstrated to be a fast, reliable and inexpensive way of simulating blood flow inside these aneurysms. Moreover, studies using these models have the potential to replicate the exact anatomy of specific patients in order to connect specific hemodynamic factors to clinical events. Furthermore, large patient population study is also possible. A methodology was previously developed in the CFD Lab, GMU, to conduct patient-specific studies but without any endovascular devices. It includes image processing and segmentation algorithms, unstructured 3D grid generation, finite element solver for Navier-Stokes equations, rheological models and visualization techniques. However, the main difficulty in using these models for endovascular stent simulations lies in the generation of acceptable computational grids inside the blood vessels and around these devices. An adaptive embedded gridding technique originally developed for fluid-structure interaction problems at the CFD lab, GMU, tremendously simplifies this impediment. In this doctoral thesis the computational modeling pipeline has been extended to model patient-specific hemodynamics of stented cerebral aneurysms. The methodology was evaluated and demonstrated with a number of imagebased models and different stent aneurysms and applied to the study of the effects of stents on the flow in aneurysms and side arterial branches.