Geometry-hemodynamics relationships in the vertebral artery V1 segment: A patient-specific CFD and multivariable analysis

Vascular geometry influences local hemodynamic conditions in the vertebral artery (VA), but the independent contributions of different morphological characteristics remain insufficiently understood. This study aimed to identify the independent morphological determinants of hemodynamic characteristics in the VA V1 segment. Thirty patient-specific bilateral VA V1 models reconstructed from computed tomography angiography were analyzed using computational fluid dynamics. Hemodynamic parameters, including time-averaged wall shear stress (TAWSS), oscillatory shear index (OSI), and relative residence time (RRT), together with time-averaged nitric oxide concentration (TANO), were quantified. Generalized linear mixed models (GLMMs) were applied to evaluate independent geometry-hemodynamics associations while accounting for bilateral dependence and adjusting for age, sex, and vascular side. Side-specific generalized linear and Beta regression analyses were further performed. Pronounced left-right differences were observed in vascular geometry and hemodynamics. After multivariable adjustment, the proximal and distal subclavian artery (SA) diameters emerged as the most consistent independent predictors of WSS- and NO-related metrics. In contrast, the significant associations between VA tortuosity and hemodynamic metrics observed in univariate analyses were substantially attenuated after adjustment for confounding factors. Side-specific analyses revealed distinct geometry-hemodynamics relationships between the left and right VAs under the corresponding side-specific inlet conditions. Diameter-related geometric features, particularly the proximal and distal SA diameters, exert stronger and more consistent independent associations with local hemodynamic characteristics than tortuosity-related descriptors in the VA V1 segment. Geometry-hemodynamics relationships also differed between the left and right VAs under side-specific physiological inlet conditions, highlighting the need to consider both vascular anatomy and flow boundary conditions in patient-specific hemodynamic analysis. These findings improve understanding of geometry-dependent hemodynamic variation in the VA V1 segment and provide quantitative evidence for future patient-specific hemodynamic assessment.