Patient-specific computational modelling of endovascular treatment for intracranial aneurysms

Abstract

Endovascular techniques, such as endoluminal or endosaccular reconstruction, have emerged as the preferred method for treating both ruptured and unruptured intracranial aneurysms, replacing open surgery in most cases. The minimally invasive approach has been shown to result in better surgical outcomes and lower mortality rates. Before the procedure, neuroradiologists rely only on their experience and visual aids from medical imaging techniques to select the appropriate endovascular option, device model and size for each patient. Despite the benefits of endovascular techniques, significant complications can arise during and after the procedures, including intraprocedural aneurysm perforation, delayed rupture, aneurysm regrowth, in-stent restenosis and thromboembolic events. Therefore, predictive virtual replicas of these interventions can serve as a valuable tool to assist neuroradiologists in the decision-making process and optimise treatment success, especially in cases involving complex geometries. Computational modelling can enable the simulation of different treatment strategies considering the most clinically relevant short- and long-term outcomes of the deployment and the postoperative complications that may arise over time.

Statement of significance: This review explores the state of the art in modelling the mechanics of the main neurovascular devices, their deployment within patient-specific geometries, their interaction with the vessel wall and their influence on the local hemodynamics. As it strongly affects their applicability in clinical practice, particular attention is paid to the computational accuracy and efficiency of the different modelling strategies. The aim is to evaluate how these scientific tools and discoveries can support practitioners in making informed decisions and highlight the challenges that require further study.