Computational Modeling of Arteriovenous Fistula Hemodynamics in Pulmonary Hypertension Patients
Author
Southern, Kaitlin M
Access
This item will be available on: 2025-07-01
Abstract
In the United States, approximately 50 million people suffer from chronic kidney disease, which over time progresses to end-stage renal disease (ESRD). Care for these patients is typically managed by dialysis with the surgical creation of an arteriovenous fistula. Even though fistula formation is a common and an effective treatment it has been suggested as a risk factor for developing pulmonary hypertension (PH). The objective of this research was to understand the relationship between fistula and pulmonary artery hemodynamics in patients with pulmonary hypertension. The completion of this objective was achieved through three aims: 1) develop protocol for creating subject-specific computational fluid dynamics (CFD) models of fistulas in patients with PH; 2) correlate fistula hemodynamics with clinical measures of pulmonary hypertension; and 3) investigate impact of fistula banding on fistula hemodynamics. Magnetic resonance images (MRI) of the fistula were obtained and processed to build a model of the fistula using Mimics software. Velocity measurements from MRI were used as boundary conditions for the CFD model. The model was then meshed, and the 3D velocity field was solved using ANSYS Workbench. Fistula data, geometric and hemodynamic, was compared with clinical data from the pulmonary artery, amongst the same patient. Multiple mini studies were performed during this project including geometry length comparisons, boundary condition comparisons, and fistula banding comparisons. For the preliminary geometry length comparison study, a shortened model was chosen and utilized throughout the duration of the simulations for reduced computational cost with little to no effect on the results. For the boundary condition comparison study, the split flow outlet condition was deemed the most physiologically accurate, but minor differences were observed between the split flow and pressure with targeted mass flow rate simulations. Zero pressure outlet conditions were found to be physiologically inaccurate and do not depict realistic flow. Computational arteriovenous fistula results indicated areas of low wall shear stress and recirculation along the anastomosis may cause endothelial dysfunction and fistula failure. Additionally, high output fistulas may lead to cardiac overload and elevated pulmonary artery pressures, from increased venous return and cardiac output. Lastly, the banding comparison study was able to mimic the role of fistula banding by restricting excessive flow and diverting more flow distally. It was concluded that the flow management technique's effectiveness depends on the band's size and location. This project presented a unique opportunity to study both the pulmonary artery and fistula within the same patient, simultaneously. Eventually, this type of modeling will provide insight to the link between fistulas and pulmonary hypertension; thus, identifying key monitoring parameters. Improved monitoring will allow physicians to intervene; thereby preventing the development of pulmonary hypertension.
Date
2023-07-21
Citation:
APA:
Southern, Kaitlin M.
(July 2023).
Computational Modeling of Arteriovenous Fistula Hemodynamics in Pulmonary Hypertension Patients
(Master's Thesis, East Carolina University). Retrieved from the Scholarship.
(http://hdl.handle.net/10342/13179.)
MLA:
Southern, Kaitlin M.
Computational Modeling of Arteriovenous Fistula Hemodynamics in Pulmonary Hypertension Patients.
Master's Thesis. East Carolina University,
July 2023. The Scholarship.
http://hdl.handle.net/10342/13179.
April 29, 2024.
Chicago:
Southern, Kaitlin M,
“Computational Modeling of Arteriovenous Fistula Hemodynamics in Pulmonary Hypertension Patients”
(Master's Thesis., East Carolina University,
July 2023).
AMA:
Southern, Kaitlin M.
Computational Modeling of Arteriovenous Fistula Hemodynamics in Pulmonary Hypertension Patients
[Master's Thesis]. Greenville, NC: East Carolina University;
July 2023.
Collections
Publisher
East Carolina University