Tracking geometric and hemodynamic alterations of an arteriovenous fistula through patient-specific modelling.

Background and Objective: The use of patient-specific CFD modelling for arteriovenous fistulae (AVF) has shown great clinical potential for improving surveillance, yet the use of imaging modes such as MRI and CT for the 3D geometry acquisition presents high costs and exposure risks, preventing regular use. We have developed an ultrasound based procedure to bypass these limitations.
Methods: A scanning procedure and processing pipeline was developed specifically for CFD modelling of AVFs, using a freehand ultrasound setup combining B-mode scanning with 3D probe motion tracking. The scanning procedure involves sweeping along the vasculature to create a high density stack of B-mode frames containing the lumen geometry. This stack is converted into a continuous volume and transient flow waveforms are recorded at the boundaries, synchronised with ECG and automatically digitised, forming realistic boundary conditions for the CFD models. This is demonstrated on a diseased patient-specific AVF.
Results: The three scans obtained using this procedure varied in geometry and flow behaviour, with regions of disease located in the first two scans. The outcome of the second procedure seen in the third scan indicated successful restoration with no sites of disease and higher flow. The models gave insight into the lumenal changes in diameter for both the artery and vein segments, as well as characterising hemodynamic behaviours in both the diseased and restored states. Vascular segment resistances obtained from the CFD models indicate a significant reduction once disease was removed, resulting in much higher flows enabling the patient to resume dialysis.
Conclusion: The methodology described in this study allowed for a multifaceted analysis and high level tracking in terms of both geometry and flow behaviours for a patient case, demonstrating significant clinical utility and practicality, as well as enabling further research into vascular disease progression in AVFs through longitudinal analysis.
(Copyright © 2019. Published by Elsevier B.V.)