Computational Analysis of Blood Flow Characteristics in Intracrainal Aneurysms

Abstract

Image-based computational fluid dynamics modeling of cerebral aneurysms has matured over the last decade. These techniques have allowed the study of patient-specific hemodynamics to better understand the mechanisms responsible for aneurysm formation, progression and rupture, and the effects of different devices and therapeutic options. The purpose of this study was to enhance and use a computational modeling chain previously developed to construct patient-specific models of cerebral aneurysms from threedimensional (3D) medical images and pulsatile physiologic flows and use them to investigate and compare the hemodynamic characteristics in ruptured and unruptured aneurysms. Using this methodology, a database containing over 200 patient-specific models of brain aneurysms has been constructed from 3D rotational angiography images. The models have been run under different physiologic flow conditions, including pulsatile and steady flows. The resulting blood flow fields were visualized using a variety of techniques in order to qualitatively classify the aneurysms according to the following characteristics: a) flow complexity, b) flow stability, c) inflow concentration, and d) size of impingement. Secondly, a number of quantitative hemodynamic measures that attempt to capture these qualitative characteristics were defined. The aneurysms were then divided into two groups according to their previous history of subarachnoid hemorrhage, i.e. ruptured and unruptured aneurysms, and the hemodynamic characteristics of each group were analyzed and statistically compared. The results indicate that ruptured aneurysms are more likely to have concentrated inflow jets, small impaction zones, complex and unstable intra-aneurysmal flow patterns, and large maximum wall shear stress. In contrast, unruptured aneurysms tend to have diffuse inflow streams, large zones of flow impaction, simple and stable flow patterns, and lower maximum wall shear stress. The importance of these results is that these associations could be used to discriminate between mechanisms of aneurysm wall weakening based on low or high flow theories and to improve aneurysm risk evaluation which is currently only based on aneurysm geometry.