Using hemodynamic modeling to improve dialysis patients’ chances of success
University of Strathclyde researchers use Simcenter to compare hemodynamics before and after AVF surgery
University of Strathclyde
The University of Strathclyde is a technological university known for conducting high-impact entrepreneurial research. It is the first university in the U.K. to win the Times University of the Year Award in 2012 and 2019, which honors its success to deliver technological and socially progressive solutions to global challenges, alongside its partners.https://www.strath.ac.uk/
- Glasgow, United Kingdom
- Simcenter Products, Simcenter STAR-CCM+
- Industry Sector:
- Medical devices & pharmaceuticals
Improving surgery outcomes
Hemodynamic modeling has the power to improve the understanding of the outcomes of arteriovenous fistula (AVF) surgeries for patients. The procedure involves connecting a vein, usually in an arm, to an artery. The goal is to increase the blood flow and robustness of the vein by making it larger and thicker so that it can better withstand the regular needling involved with dialysis, a treatment necessary for patients with renal disease. Research conducted by George Hyde-Linaker, a doctoral candidate in the Department of Biomedical Engineering at the University of Strathclyde (Strathclyde) and Dr. Asimina (Melina) Kazakidi, senior lecturer and leader of the Biofluid Mechanics (BioFLM) Research Group at the same department, in collaboration with doctors from Queen Elizabeth University Hospital, was recently published in the Journal of Medical Engineering & Physics (MEP, Hyde-Linaker G., Hall Barrientos P., Stoumpos S., Kingsmore D.B. and Kazakidi A. (2022). Patient-specific computational hemodynamics associated with the surgical creation of an arteriovenous fistula. Med. Eng. Phys. 105, 103814, doi: 10.1016/j. medengphy.2022.103814) This work shows how using simulations can predict vessel changes after AVF surgery.
Using Simcenter™ STAR-CCM+™ software to model blood flow before and after surgery will help clinicians understand the hemodynamic environment and determine the optimal sites for AVFs. Simcenter STAR-CCM+ is part of the Siemens Xcelerator business platform of software, hardware and services.
The University of Strathclyde is a technological university known for conducting high-impact entrepreneurial research. It is the first university in the U.K. to win the Sunday Times University of the Year award in 2012 and 2019, which delivers technological and socially progressive solutions to global challenges, alongside its partners. The university has a key research theme in healthcare technology and a health tech cluster (HTC) for transforming knowledge. The Department of Biomedical Engineering is recognized for its innovative and multi-disciplinary research programs between engineering and the life sciences. The university has invested £16M to refurbish the Wolfson Building, a state-of-the-art space for biomedical engineering research.
The recently published work in the MEP Journal was supported in part by the University of Strathclyde’s International Strategic Partner Research Studentships, the U.K. Research and Innovation (UKRI) Engineering and Physical Sciences Research Council (EPSRC) award via the Transformative Healthcare Technologies scheme (EP/W004860/1) and the EU H2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 749185. This work continues by building a data bank of patient demographics and completing computational fluid dynamics (CFD) studies to determine biological and hemodynamic parameters that predict favorable patient outcomes.
Oscillatory Shear Index (OSI) contour plots prior-to and after fistula creation (Image courtesy of Hyde-Linaker et al. Med Eng Phys 105 (2022), 103814)
Using simulation to address contraindication
Traditional contrast agents are substances used to increase the contrast of structures and fluids within the body in medical imaging. The problem with such agents is that they contraindicate for end-stage renal disease patients. For an AVF surgery to improve a patient’s outcome, a researcher or physician needs the right tool to accurately predict how the body will react to an AVF. This is where simulation is valuable to the overall process.
The surgery to create an AVF involves creating an anastomosis, a connection between the vein and the artery. This redirects blood flow into the vein at a high rate, enabling it to extract blood from the vein for external hemodialysis.
One aspect of this procedure that has historically been neglected is understanding how the blood flows near the point of the connection. One problem a patient can experience after AVF surgery is arterial steal syndrome, which occurs when the AVF prevents the hand from receiving an adequate blood supply. Researchers from Strathclyde’s BioFLM research group indicate that decreased blood flow throughout the rest of the body, other than the arm containing the anastomosis, is another problem.
The BioFLM researchers used Simcenter STAR-CCM+ to reveal that simulations can help physicians prevent negative impacts on the hand on the side of the anastomosis as well as the rest of the body. Using Simcenter STAR-CCM+ also helped researchers predict time-averaged wall shear stress (WSS), the force per unit area exerted by the blood in motion on the solid boundary of the vessel. The research published at MEP further shows other flow patterns at the anastomosis. The WSS levels modeled by CFD indicate the successful maturation of the AVF and aim to help physicians improve the success rates of future AVF surgeries and post-operative patient management.
The Strathclyde BioFLM’s research helps to ensure an AVF will last for a substantial period of time. AVFs are the gold standard of vascular access, but between 30 and 70 percent of AVFs fail. The two leading causes of failure are neointimal hyperplasia and inadequate outward remodeling. Neointimal hyperplasia involves the growth of smooth muscle cells and extra-cellular matrix in the intima of the vein or artery. The impact of arterial remodeling on the arm vessels proximal to the AVF may be detrimental to the patient through increased predisposition to other complications, such as cardiovascular disease and arterial steal syndrome. This research from the BioFLM group will ultimately help physicians ensure that AVFs remain viable so hemodialysis can continue.
Box and whisker plots of the Remax values at the outlets of simulated vasculature induced over the analyzed (5th) cardiac cycle in the commonly simulated vasculature between the pre- and post-AVF cases (Image courtesy of Hyde-Linaker et al. Med Eng Phys 105 (2022), 103814)
Using FeMRA datasets with CFD
Hyde-Linaker and Dr Kazakidi of Strathclyde’s BioFLM research group generated 3D reconstructions of arterial vasculature from patient populations. The work published in MEP involves pioneering reconstructions of the entire proximal vasculature to the arteriovenous fistula, before and after surgery, using ferumoxytol-enhanced magnetic resonance angiography (FeMRA) data previously obtained by the collaborating doctors. FeMRA is a method that allows physicians to obtain high-resolution patient-specific data. The Strathclyde research team used Simcenter STAR-CCM+ to generate high-fidelity simulations of the hemodynamics from multiple FeMRA scans with an established finite-volume implementation.
“We used Simcenter STAR-CCM+ to simulate scale-resolving hybrid models, which proved to be very useful,” says George Hyde-Linaker, a doctoral candidate in the Department of Biomedical Engineering at the University of Strathclyde. “It selects RANS or LES depending on the flow rate.”
A scale-resolving hybrid (SRH) turbulence model is a hybrid Reynolds‐averaged Navier‐Stokes (RANS) Large-eddy simulation (LES) model that allows users to compute unsteady large-scale turbulence structures while having a computational expense compared to a RANS model. SRH models create fewer challenges for the user because they calibrate the mesh and time step for a problem. This helps the user make the most of a given mesh and timestep. SRH models thereby provide a sufficient level of accuracy at a reasonable computational cost and turnaround time.
(a-c) Velocity streamlines and flow directions (arrows) in the radiocephalic fistula anastomosis at (a) T1 (peak systole), (b) T2 (mid-deceleration), and (c) T3 (peak diastole). (d) Normalised TAWSS, (e) OSI. (Image courtesy of Hyde-Linaker et al. Med Eng Phys 105 (2022), 103814, doi: 10.1016/j.medengphy.2022.103814)
In addition, coupling CFD with FeMRA demonstrates the potential for a more detailed analysis of AVF maturation and its impact on the heart. “The results of the CFD simulations were validated based on phase-contrast FeMRA-obtained data in the left radial artery following anastomosis creation.” says Hyde-Linaker. They found that the mean percentage difference between the CFD and phase-contrast FeMRA in the radial (feeding) artery was 2.52 percent.
“We were able to obtain flow rates from the FeMRA scanning that was used for informing the outlets of the post-AVF simulation,” says Hyde-Linaker. “You can see that the flow rates we observed in the radial artery here corresponded fairly well. That can be a marker of our replication of the flow conditions in the patient.”
Polyhedral mesh generation of patient-specific aorta created in Simcenter Star-CCM+.
Integrating simulations into medicine
Kazakidi’s goal, who is also the director of MSc/MRes programs in biofluid mechanics at Strathclyde, is to assist physicians with simulations such as those using Simcenter STAR-CCM+ providing useful hemodynamic information and predicting patient outcomes. This can provide valuable insights for clinicians so they can get a better understanding of the diagnosis and intervention planning.
The BioFLM team found that the time to complete a workflow from initial image segmentation to CFD post-processing was approximately six hours for an experienced user, including the Simcenter STAR-CCM+ software. They explain that the benefit of using CFD in addition to FeMRA is profiling hemodynamic metrics that are difficult to measure in-vivo. “In-silico modeling can transform clinical research, informing surgeons about future personalized interventions,” says Kazakidi. “We’ve used Simcenter STAR-CCM+ for robust hemodynamic CFD studies of an AVF pre- and post-surgery patient case to help clinicians make future informed decisions based on detailed hemodynamic patterns.”
Hyde-Linaker and Kazakidi expect their further work to look at differing hemodynamic environments in a cohort of patients with varying AVF anastomoses configurations and outcomes. This will include brachiocephalic and radiocephalic fistulae. They will also account for the arterial wall motion by incorporating flow-structure interaction (FSI) simulations into the investigation. The Strathclyde research team believes that using Simcenter STAR-CCM+ to study AVF hemodynamics is a useful tool that can help in the understanding of AVF dysfunction and failure.
“Replacing aspects of clinical trials with either patient-specific models or virtual data sets and populations allows us to accumulate evidence of surgical interventions. This can be referenced in accreditation and certification. We’re interested in doing this to reduce patient numbers and lead times, not to mention costs,” says Kazakidi.
Framework used for the segmentation, reconstruction, geometry unification and volume extraction of the patient-specific geometry based on FeMRA multi-stack images and subsequent CFD analysis (Image courtesy of Hyde-Linaker et al. Med Eng Phys 105 (2022), 103814)