Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Cardiovascular Engineering

Cardiovascular Engineering - Poster Session E

Poster U5 - Investigating the Impact of Band Lengths and Circumferences on Arteriovenous Fistulas Using Computational Fluid Dynamics

Saturday, October 26, 2024
10:00 AM - 11:00 AM EST
Location: Exhibit Hall E, F & G

Introduction:  End Stage Renal Disease (ESRD) is a chronic condition that impairs the kidneys' ability to function at necessary levels, often necessitating a kidney transplant or dialysis therapy. Pulmonary hypertension (PH) is a complication that can arise in dialysis patients, characterized by a mean pulmonary artery pressure (mPAP) greater than 20 mmHg. [1] Echocardiography indicates that PH is prevalent in about 80% of patients with ESRD [2]. This condition has been linked to the high cardiac output caused by an arteriovenous fistula (AVF), a surgical connection between a vein and an artery that provides vascular access for hemodialysis. Studies show that the normal blood flow rate from an AVF is 500-800 mL/min, while rates of 1200-1500 mL/min are associated with the development of PH [3]. Fistula banding is a surgical method used to control cardiac output from arteriovenous fistulas (AVFs) by tightening a region of the AVF to restrict blood flow. This research utilizes computational fluid dynamics (CFD) to simulate blood flow through the AVF and model various banding scenarios. Here, band circumference (14.6mm, 14.1mm, and 13.6 mm) and length (2mm, 4mm, and 6mm) have been investigated to understand the impact on the overall Wall Shear Stress (WSS) and velocity of the fistulas.

Materials and

Methods:  Magnetic resonance imaging (MRI) data was obtained from a patient diagnosed with pulmonary hypertension (PH) who has an arteriovenous fistula (AVF). The MRI images were imported into Mimics software, where they were segmented to generate a 3D model of the AVF. This 3D model was then imported into 3-Matic software, where the inlets and outlets were defined: the proximal artery inlet (PA) at the top left, the proximal vein outlet (PV) at the top right, and the distal artery outlet (DA) at the bottom. Additionally, geometry editing was performed in 3-Matic using a push-and-pull tool to create models of banded fistulas with different geometries. Bands with lengths 2mm, 4mm, and 6mm were EACH created for band circumferences of 14.6mm, 14.1mm and 13.6mm.

The AVF 3D model was then imported into ANSYS Workbench software, where it was meshed, and transient blood flow was simulated over three cardiac cycles. Patient-specific velocity data from the MRI was used as the inlet boundary condition. For the outlet boundary conditions, constant pressure was applied, with pressure values collected from the outlets of the original un-banded fistula. A no-slip boundary condition was applied to the walls. Blood was used as the fluid for the simulation and was characterized as an incompressible Newtonian fluid with a density of 1060 kg/m3 and a viscosity of 3.2 cP [4].

Results, Conclusions, and Discussions: The results are displayed in Table 1 and Figures 1 to 3. Table 1 shows the values for maximum velocity, volume-weighted average velocity, average velocity from the PV outlets, maximum WSS, average WSS, and minimum WSS. Figures 1 to 3 present the WSS contours and velocity volume renderings of the AVFs, illustrating the distribution of WSS and velocity, respectively. The figure captions summarize the ranking from highest values to lowest. For band circumferences 14.6mm and 13.6mm, 4mm, the volume-weighted average velocity and WSS values go in an order of band length 4mm > 6mm > 2mm. For band circumference 14.1mm, the volume-weighted average velocity and WSS values go in an order of 6mm > 2mm > 4mm.

It was initially expected that the results of the volume weighted-average velocity and WSS would follow a predictable sequence in accordance with increasing or decreasing band lengths. However, the results were more scattered, reflecting the unpredictable nature of fistulas. Examining the contours in Figures 1 to 9, certain regions exhibit significantly higher levels of WSS and velocity than others. A noticeable trend is that the average WSS follows the same order as the average velocity from highest to lowest. Another thing noticed is that the average velocities at the PV outlets tend to reduce for smaller band circumferences and longer lengths. One limitation of the study is the outlet boundary condition, as using constant pressure may not accurately represent real-life scenarios.

With pulmonary hypertension being a significant problem for patients with ESRD, computational fluid dynamics (CFD) offers a valuable tool for simulating and comparing different fistula band geometries. In this experiment, three band lengths were tested across three banding circumferences. The results indicate a relationship between average WSS and average velocity, but no linear correlation was observed between band length and fluid flow parameters. Incorporating a Windkessel model for the outlet boundary condition will be explored to improve the accuracy of simulations.

Acknowledgements (Optional):  Department of Engineering, East Carolina University
Undergraduate Research and Creativity Awards (URCA). East Carolina University,
Kaitlin Southern