Citations

INTRODUCTION: Despite more than a decade of endovascular coil treatment, the effects of coils on cerebral aneurysm (CA) hemodynamics are still poorly understood. Coils present several challenges to in vivo and in vitro flow measurement techniques and previous in silico methods have suffered from large assumptions on coil geometry. Here we present the first fluid dynamic simulations of coiled CAs that consider the structure and deployment mechanics of embolic coils. We also investigate the influence of coil packing density, design, and configuration on CA fluid dynamics. Methods: Coil deployment was modeled using a novel finite element approach that realistically simulates coil dynamics during deployment. Two coil designs were investigated: helical and 3D. Coil design and material properties were matched to manufacturer specifications. Five deployment sequences of each coil design, at different microcatheter placements, were simulated in two idealized CA models with variable neck sizes. Blood flow was simulated using computational fluid dynamics. Simulated deployments and fluid dynamics were then compared to deployments of actual coils in identical physical CA models and in vitro particle image velocimetry flow measurements. Results: Simulated results closely matched in vitro data. Reductions in aneurysmal velocity magnitudes were largest for 3D coils and in a narrow-neck model. In that model, 3D coil deployments reduced average aneurysmal velocity magnitudes by a 51% - 69% range at packing densities less than 20% and by a 74% - 84% range at packing densities greater than 30%. Linear regression results showed reductions were strongly dependent on the spacing between coil loops within the aneurysm and packing density, with correlations of 0.6 and 0.7 respectively. Conclusion: Coil design and packing density may play equally important roles in determining CA hemodynamics. Results in an anatomical model will also be presented. The proposed virtual coiling approach represents a novel and effective method for realistically simulating coiled CAs, and is an important step towards clinical preoperative planning of coil treatment.

Embolic coiling is one of the most effective treatments for cerebral aneurysms (CAs), largely due to the hemodynamic modifications that the treatment effects in the aneurysmal environment. However, coiling can have very different hemodynamic outcomes in aneurysms with different geometries. Previous work in the field of biofluid mechanics has demonstrated on a general level that geometry is a driving factor behind aneurysmal hemodynamics. The goal of this study was to relate two specific geometric factors that describe CAs (i.e., dome size (DS) and parent-vessel contact-angle (PV-CA)) and one factor that describes treatment (i.e., coil packing density (PD)) to three clinically relevant hemodynamic responses (i.e., aneurysmal root-mean-square velocity (Vrms), aneurysmal wall shear stress (WSS), and cross-neck flow (CNF)). Idealized models of basilar tip aneurysms were created in both virtual and physical forms to satisfy two-level multifactorial experimental designs. Steady and pulsatile flow hemodynamics were then evaluated in the virtual models using computational fluid dynamics (CFD) (before and after virtual treatment with finite element (FE) embolic coil models), and hemodynamics were also evaluated in the physical models using particle image velocimetry (PIV) (before and after treatment with actual embolic coils). Results showed that among the factors considered, PD made the greatest contributions to effects on hemodynamic responses in and around the aneurysmal sac (i.e., Vrms and WSS), while DS made the greatest contributions to effects on hemodynamics at the neck (i.e., CNF). Results also showed that while a geometric factor (e.g., PV-CA) may play a relatively minor role in dictating hemodynamics in the untreated case, the same factor can play a much greater role after coiling. We consider the significance of these findings in the context of aneurysmal recurrence and rupture, and explore potential roles for the proposed methods in endovascular treatment planning.

Endovascular coiling is the most common treatment for cerebral aneurysms. The treatment consists of deploying coils of different shapes and sizes into the aneurysm with the intent of reducing blood flow therein. While the treatment has been shown to be more effective than conventional surgical methods, it is successful in only 50–80% of cases, depending on the type of aneurysm [1]. Treatment success has been correlated with favorable post-treatment hemodynamics [2]. Understanding of the effects of coils on post-treatment hemodynamics is limited, however, due to limited in-vivo and in-vitro flow measurement techniques. Previous computational studies have also suffered from drastic assumptions about coil geometry.