A clinical role for fluid dynamics modelling of arteriovenous fistulae

Disease of the vascular system often manifests through inward remodelling, forming a narrowing in the vessel, called a stenosis. The appearance of stenoses in certain critical vessels, such as those in the heart and brain can have morbid consequences as the vessels approach total occlusion. This disease process is due to inflammatory responses instigated by endothelial cells at the inner wall surface, which transduce wall shear stress from the blood flow in their regulatory capacity, driving vascular remodelling in response to changes in flow. In arteriovenous fistulae (AVFs), formed as a vascular access for hemodialysis, rates of stenoses are very high. This is understood to be due to the unusual geometric configuration, in combination with supraphysiological flowrates. Significant disturbance to the flow occurs at the anastomosis, which propagates into the juxta-anastomotic vein segment, a region where stenosis is often observed in AVFs. This thesis sought to identify pathogenic WSS patterns within AVFs and improve clinical monitoring through patient-specific CFD (PS-CFD) modelling. This began with the development of an ultrasound-based 3D geometry scanning methodology, as well as a data pipeline for generation of PS-CFD models. A diverse range of patients with AVFs were scanned during a one year period, from which eight different patient cases were formed, capturing events across multiple scans over varying timelines. Longitudinal analysis of these scans was carried out, with four exhibiting signs of disease development between the scans, and the remaining four involving alterations following stenting procedures. Within the four cases of disease, WSS behaviours with transient multidirectionality over the pulse cycle, identified through the transWSS metric, were found to localise in regions where inward remodelling occurred. In two cases where stenosis was seen to progress in severity, transWSS grew in magnitude, whilst the other cases with much less severe stenoses demonstrated a reduction, which may have halted further development. Four cases following patients who had undergone stenting procedures were also analysed, with one featuring re-stenosis, again coinciding with high transWSS. The calculation of hydrodynamic resistance was found to be a useful identifier for the presence of disease, as well as the subsequent removal following these procedures, as significant reductions in resistance were seen following successful procedures, whilst other cases which developed disease elsewhere remained high in resistance. Extending the analysis to a larger dataset of scans, resistance was found to be a useful metric for assessment of AVFs. High resistance in the vein was indicative that a stenosis was present, whilst high artery resistance was due to poor diameters. Well performing AVFs, including those which had undergone successful stenting procedures were found to have much lower resistance, making resistance an ideal indicator of current AVF state. The scalability and applicability of this method, alongside the transWSS and resistance indicators shows utility for clinical usage, encouraging regular AVF scanning and PS-CFD modelling within clinical protocols, thereby allowing for the detection and prognosis as well as tracking the progression of disease within AVFs.

Grantor dc:publisher

UNSW, Sydney

Year dc:date

2018

Author dc:creator

• Carroll, John

• Computational Fluid Dynamics
• Vascular Disease
• Arteriovenous Fistula
• Wall Shear Stress
• 3D Ultrasound
• Patient-Specific
• Stent

Statement dc:rights

• open access
• CC BY-NC-ND 3.0
• free_to_read

Licence

• https://purl.org/coar/access_right/c_abf2
• https://creativecommons.org/licenses/by-nc-nd/3.0/au/

Language dc:language

EN

Identifier

https://doi.org/10.26190/unsworks/21488

OAI identifier oai:identifier

oai:unsworks.library.unsw.edu.au:1959.4/64227