Characterization of 3-D flow structures in the stenosed carotid bifurcation with plaque ulceration

Carotid plaque ulcerations, or irregularities in plaque surface morphology, have been identified as an independent risk factor for ischemic stroke. Our previous studies using Doppler ultrasound (DUS) have indicated significant flow disturbances distal to ulceration in the atherosclerotic carotid bifurcation, as characterized by parameters such as turbulence intensity (TI). Additional tools are needed to understand the implications of such flow abnormalities on the risk of thrombogenesis and cerebral ischemic events. Numerical simulations using computational fluid dynamics (CFD) can supplement experimental DUS studies, providing higher resolution, time-resolved models of 3-D flow fields. CFD is also able to quantify hemodynamic factors that indicate thromboembolic or plaque rupture potential. We report a CFD analysis of an ellipsoidal ulcer model and a matched non-ulcerated model in a moderately stenosed carotid bifurcation, with the same vessel geometries and flow conditions used in our previous DUS studies. The CFD models used a spatial finite element discretization of over 160,000 quadratic tetrahedral elements to adequately resolve the flow field. Pulsatile flow simulations with boundary conditions and flowrate waveforms matching DUS experimental conditions were iterated for ten cardiac cycles. Turbulence intensity was calculated for the CFD models and compared with DUS experimental results. The CFD models were able to capture differences in flow patterns between cardiac cycles. As observed in the empirical DUS results, the CFD ulcer model displayed higher levels of TI in the post-stenotic region than the CFD non-ulcerated model. The extent and magnitude of TI was comparable to the DUS results, after modeling for the effects of sample volume geometry, intrinsic spectral broadening, and a high pass filter. Furthermore, the CFD results indicate that sampling volume size and location have an effect on DUS quantification of TI in the post-stenotic carotid artery. Small elevations in maximum shear stress, transit time, shear exposure and level of threshold activation were observed in the ulcerated model, as compared to the non-ulcerated model. CFD facilitates the comparison of hemodynamic parameters between ulcer models and may help to demonstrate the risks of embolism or plaque rupture posed by ulcerated atherosclerotic plaques in the carotid bifurcation.