(219e) Validation of Stroke Prediction in Patients with Carotid Artery Disease Using CFD

Stroke remains one of the leading causes of death in the United States accounting for one out of every twenty deaths.[1] Carotid artery stenosis (CAS) is a major cause of both stoke and transient ischemic attacks (TIAs), and it has been documented that the risk of development of these illnesses are directly related to the degree of stenosis.[2] Research has proven that carotid endarterectomies (CE) can be beneficial in treating symptomatic patients with high-grade internal carotid artery (ICA) stenosis. However, treatment of lower risk and asymptomatic patients remains controversial due to high risks associated with the procedure. Some of these risks include myocardial infarction, cranial nerve injury, hemodynamic instability, infection, and bleeding.[3] Therefore, minimally, or non-invasive procedures are currently being studied to assess patients and reduce risks of thromboembolic and ischemic events. The carotid artery carries oxygenated blood from the heart to the head. There are two carotid arteries on each side of the neck. Each common carotid artery branches into two divisions: the internal carotid, supplying blood to the brain, and the external carotid that provides blood to the face. As the carotid ascends towards the head, it branches out and creates a bifurcation. The internal carotid is especially interesting as it contains a bulge, the carotid sinus, with a diameter greater than that of the common carotid.[4,5] This geometry and the resulting increase in adverse pressure gradient create a flow reversal region during the cardiac cycle. Velocity profiles in the carotid sinus may be linked to stroke, as this site is prone to plaque deposition.[6,7] The goal of this project is to investigate velocity profiles and pressure gradients in the carotid artery to understand the biological significance of the carotid sinus and its relation to stroke risk.

Using Computational Fluid Dynamics (CFD) simulations of the carotid artery, medicine and engineering intersect. CFD uses finite-element analysis to determine flow characteristics while paying specific attention to the geometry, velocity profiles, streamlines, and pressure gradients. This method has an advantage over solely relying on ultrasound data because it provides insight into the unique flow patterns in a patient’s individual carotid artery geometry. Once a mesh, a collection of vertices, edges, and faces that define the shape of an object in computational simulations, has been created, physiological boundary conditions are defined.[6,7] These boundary conditions include important parameters for blood density, blood viscosity (Carreau model), outlet pressures, body temperature, and peak systolic (PSV) and end diastolic (EDV) velocities. The Navier-Stokes Equations are solved simultaneously for flow simulations at nodes created in the meshing portion of the application. A solution is generated when residual limits are met.

This project consists of two Phases. In Phase 1, forty-five patients whose outcomes were known to the analysis team were provided by the Department of Neurosurgery at the University of Rochester Medical Center, along with their respective ultrasound data. These patient data were analyzed using CFD to determine the parameters that can be used to predict stroke. In this study, it was found that high pressure and velocity gradients near the wall of the internal carotid artery were clear indicators of the potential for stroke, and have been previously presented.[8] Now, we present the results of Phase 2 of the project where an additional forty-five patients were analyzed in a double-blind study where the patient outcomes regarding stroke were not known until after the analysis was completed. The results suggest that CFD can be used to accurately predict the outcome of stroke or no-stroke for patients with carotid artery disease. The results were consistent with our previous findings that stroke patients have irregular velocity gradients and more extreme velocities in the carotid sinus, where there is less fluid due to plaque deposit from atherosclerosis. It was also found that comparisons of patients with different stroke outcomes require more than a comparison of velocities and geometries in a steady model. A pulsatile flow user-defined function (UDF) for a transient set-up has been developed to consider velocity differences during diastole and systole. In theory, a pulsatile flow model simulates physiological conditions more realistically, as the heart pumps blood in a pulsatile fashion. Validation of the model confirmed that mass conservation is obeyed. A scaled ratio, in which the maximum velocity obtained in the CFD measurements at the peak of the velocity profile at the carotid bulge was compared to the lowest velocity at the reversal region, was also created. Preliminary comparison has shown that stroke patients have higher ratios of flow reversal. A UDF that instantly extracts velocity and pressure data at the bulge is being developed to streamline the process of correlating patient stroke outcomes. Finally, improvements to the model include accounting for the fluid-structure interaction between the artery and the blood, in which the vessel wall is modeled as a homogenous, isotropic, viscoelastic material. Ultimately, this project creates a prediction model for stroke clinical outcomes, with preventative medicine benefits.

References
[1] "Stroke | cdc.gov", Centers for Disease Control and Prevention, 2020. [Online]. Available: https://www.cdc.gov/stroke/index.htm.
[2] "Types of Stroke | cdc.gov", Cdc.gov, 2020. [Online]. Available: https://www.cdc.gov/stroke/types_of_stroke.htm#ischemic.
[3] A. Greenstein et al., "Association between minor and major surgical complications after carotid endarterectomy: Results of the New York Carotid Artery Surgery study", Journal of Vascular Surgery, vol. 46, no. 6, pp. 1138-1146, 2007.
[4] Bluestein D. Utilizing Computational Fluid Dynamics in Cardiovascular Engineering and Medicine—What You Need to Know. Its Translation to the Clinic/Bedside. Artificial organs. 2017;41:117-121.
[5] J. Krejza et al., "Carotid Artery Diameter in Men and Women and the Relation to Body and Neck Size", Stroke, vol. 37, no. 4, pp. 1103-1105, 2006.
[6] Foster, D.; Salerno, D. “A Computational Fluid Dynamics Model of the Carotid Artery”. Presented at the 2017 AIChE Annual Meeting, Minneapolis, MN, October 29-November 3, 2017.
[7] Foster, D.; Weldy, A. “Computational Fluid Dynamics Simulations of the Human Carotid Artery to Predict Strokes in Patients with Carotid Artery Disease”. Presented at the 2019 AIChE Annual Meeting, Chemical Engineering Principles Advancing Medicine Session, Orlando, FL, November 12, 2019.
[8] Maruicio Araiza Canizales, Priscila Passerotti Vaciski Barbosa, Lauren E. Redus, Leonor N. Teles, Hon Sum A. Lee, Jonathan Stone, and David G. Foster, “Patient-Specific CFD Simulations of the Carotid Artery to Predict Stroke,” 2021 AICHE Annual Meeting, Boston, MA, November 2021.