Numerical Modeling of Flow in the Cerebral Vasculature: Understanding Changes in Collateral Flow Directions in the Circle of Willis for a Cohort of Vasospasm Patients Through Image-Based Computational Fluid Dynamics

Angela Straccia; Michael C Barbour; Fanette Chassagne; David Bass; Guilherme Barros; Daniel Leotta; Florence Sheehan; Deepak Sharma; Michael R Levitt; Alberto Aliseda

doi:10.1007/s10439-024-03533-w

Numerical Modeling of Flow in the Cerebral Vasculature: Understanding Changes in Collateral Flow Directions in the Circle of Willis for a Cohort of Vasospasm Patients Through Image-Based Computational Fluid Dynamics

Ann Biomed Eng. 2024 May 17. doi: 10.1007/s10439-024-03533-w. Online ahead of print.

Authors

Affiliations

¹ Department of Mechanical Engineering, University of Washington, Seattle, WA, USA. straccia@uw.edu.
² Department of Mechanical Engineering, University of Washington, Seattle, WA, USA.
³ INSERM U1059 Sainboise, Mines Saint-Étienne, Saint-Étienne, France.
⁴ Department of Neurological Surgery, University of Washington, Seattle, WA, USA.
⁵ Applied Physics Laboratory, University of Washington, Seattle, WA, USA.
⁶ Department of Medicine, University of Washington, Seattle, WA, USA.
⁷ Department of Anesthesiology and Pain Medicine, University of Washington, Seattle, WA, USA.

PMID: 38758460
DOI: 10.1007/s10439-024-03533-w

Abstract

The Circle of Willis (CoW) is a ring-like network of blood vessels that perfuses the brain. Flow in the collateral pathways that connect major arterial inputs in the CoW change dynamically in response to vessel narrowing or occlusion. Vasospasm is an involuntary constriction of blood vessels following subarachnoid hemorrhage (SAH), which can lead to stroke. This study investigated interactions between localization of vasospasm in the CoW, vasospasm severity, anatomical variations, and changes in collateral flow directions. Patient-specific computational fluid dynamics (CFD) simulations were created for 25 vasospasm patients. Computed tomographic angiography scans were segmented capturing the anatomical variation and stenosis due to vasospasm. Transcranial Doppler ultrasound measurements of velocity were used to define boundary conditions. Digital subtraction angiography was analyzed to determine the directions and magnitudes of collateral flows as well as vasospasm severity in each vessel. Percent changes in resistance and viscous dissipation were analyzed to quantify vasospasm severity and localization of vasospasm in a specific region of the CoW. Angiographic severity correlated well with percent changes in resistance and viscous dissipation across all cerebral vessels. Changes in flow direction were observed in collateral pathways of some patients with localized vasospasm, while no significant changes in flow direction were observed in others. CFD simulations can be leveraged to quantify the localization and severity of vasospasm in SAH patients. These factors as well as anatomical variation may lead to changes in collateral flow directions. Future work could relate localization and vasospasm severity to clinical outcomes like the development of infarct.

Keywords: Cerebral flow; Circle of Willis; Collateral pathways; Computational fluid dynamics; Numerical modeling; Patient-specific modeling; Vasospasm.

Abstract

Grants and funding