A Patient-Specific Analysis Framework for Assessing Stroke Risk in Pediatric Moyamoya Disease

Texas Heart Institute, Houston TX

Investigators

Linked publications, trials & patents

Abstract

PROJECT SUMMARY: Pediatric MoyaMoya Disease (MMD) is a cerebrovascular disease characterized by progressive narrowing of the major arteries in the Circle of Willis (CoW), leading to recurring ischemic and hemorrhagic stroke. Clinical strategies to prevent or reverse vessel occlusion through medical management are not available. Neurosurgical interventions are used to augment blood flow to the affected region and avert future stroke events; but carry the risk of perioperative stroke, infection and intracranial hemorrhage. Treatment is an absolute necessity however, since a morbidity rate of > 70% in untreated patients is currently realized. While atypical vessel straightening, narrowing and collateralizations are commonly observed on the CT- angiograms, there is a lack of understanding how these characteristic vascular alterations affect local hemodynamics and disease progression, and there has been no study of the effect of surgical interventions on future stroke events. Computational simulations, adjusted with patient-specific attributes, have been successfully used in other pathological conditions to predict local hemodynamics, potentially informing therapy and interventional strategies. The long-term objective is therefore to develop a predictive patient-specific analysis framework to assess stroke risk in pediatric MMD patients treated by surgical intervention, and delineate scenarios affecting disease severity. In a pilot study involving ACTA2(-/-) knockout mouse models, which develop many of the phenotypic features of MMD, computational fluid dynamic analysis of the authentic CoW vasculature was performed applying state-of-the-art isogeometric analysis technology. Locations of critical wall shear rate (above the coagulation limit) that were at a greater risk of clot formation were predicted. Results show that occlusion in one of the major arteries in the CoW increases stroke risk in mouse models of MMD. If a similar or equivalent behavior is realized in the human condition, it could have profound implications for patient care. The goal for the proposed research is to perform 3D patient-specific analysis on a pilot cohort (n=6) of human CoW to identify susceptible regions that could evolve into severe stenosis or complete vessel occlusion, leading to stroke in pediatric MMD patients, and compare the predictions to follow up clinical observations. The central hypothesis is that the simulations will accurately predict the occurrence and location of the stroke. Once the analysis framework is established and the predictive capability assessed in this pilot; a comprehensive study on a population of at least 50 cases will be performed to reliably assess the utility of computational predictions of stroke risk in pediatric MMD patients. The proposed research is significant, as it will provide predictive insight into complex interplay between vascular geometry and hemodynamic environment altered by MMD in a patient-specific sense. If confirmed in a larger cohort, the presented analysis framework could enable clinicians to predict patients at risk of stroke prior to the imaging assessments of severe hemodynamic impairment/collaterization that is used currently, potentially leading to earlier intervention.