CTA and DSA Based Computational Fluid Dynamics Models for Morphological and Hemodynamic Assessment of Intracranial Atherosclerotic Stenosis

Yang Rui¹Xulong Yin²Gaohui Li³Jianping Xiang³Qi Fang¹Hui Wang^1*Bo Li^2*

• ¹First Affiliated Hospital of Soochow University, Suzhou, China
• ²The First Affiliated Hospital of Soochow University, Suzhou, China
• ³ArteryFlow Technology Co Ltd, Hangzhou, China

Background: Intracranial atherosclerotic stenosis (ICAS) is a primary cause of ischemic stroke. Accurate assessment of anatomical and hemodynamic characteristics is crucial for treatment planning, yet current clinical evaluation primarily relies on luminal stenosis. Objective: This study aims to compare computational fluid dynamics (CFD) models based on digital subtraction angiography (DSA), computed tomography angiography (CTA) and CTA model incorporating DSA hemodynamic information (CMD) integrating DSA flow data with CTA morphological structure, evaluating their differences and consistency in ICAS assessment. Methods: 40 ICAS patients who underwent CTA and DSA were retrospectively included. Patient-specific CFD simulations were performed using standardized boundary conditions to assess morphological data and hemodynamic parameters, including pressure ratio, wall shear stress ratio, and high shear stress areas. Statistical analyses included paired comparisons, intraclass correlation coefficients (ICC), and Bland-Altman analysis. Results: CTA-based models demonstrated excellent consistency with DSA in anatomical measurements (ICC > 0.90). The CMD approach enhanced consistency in functional metrics, with CMD-derived PR and WSSR highly concordant with DSA results. When using CTA alone, WSSR was slightly underestimated, particularly in middle artery lesions. Subgroup analysis indicated that lesion location significantly influences flow and shear stress patterns. Conclusion: CTA-based CFD modeling serves as a reliable non-invasive alternative to DSA for morphological ICAS assessment. The CMD method further improves the accuracy of functional evaluation by integrating flow data. These findings support the integration of anatomical imaging with hemodynamic modeling to enhance the clinical potential for stroke risk stratification.