Abstract: SA-PO0451
Intravascular Ultrasonography Demonstrates Accuracy over Angiography in Arteriovenous Hemodialysis Dysfunction
Session Information
- Dialysis: Vascular Access
November 08, 2025 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 803 Dialysis: Vascular Access
Authors
- Wu Wong, David Jasen, Boston University Chobanian & Avedisian School of Medicine, Boston, Massachusetts, United States
- Decker, James Whelan, Boston University Chobanian & Avedisian School of Medicine, Boston, Massachusetts, United States
- Banihani, Dima Hussein, University of South Carolina, Columbia, South Carolina, United States
- Shazly, Tarek, University of South Carolina, Columbia, South Carolina, United States
- Chitalia, Vipul C., Boston University Chobanian & Avedisian School of Medicine, Boston, Massachusetts, United States
Background
Intravascular ultrasound (IVUS) has been shown to be more sensitive than conventional angiography in detecting the number and extent of arteriovenous (AV) access stenoses. However, in previous studies, IVUS was evaluated against angiography, which has several limitations. Therefore, it was necessary to compare the accuracy of both techniques against a ground truth.
Methods
Via Computer Aided Design (CAD), 33 vascular conduits modeled after patient fistulograms were 3D printed using stereolithography. Randomly generated stenoses and aneurysms were introduced. Conduits were then subjected to IVUS and contrast venography.
Results
Compared to CAD dimensions (ground truth), the % error was higher in angiography versus IVUS (p = 0.026). The CAD-based minimum diameter of stenosis (MDS) correlated strongly with IVUS (R = 0.96), but not with angiography (R = 0.29). For concave stenotic geometries, the average % error of MDS based on Gaussian curvature values was higher in angiography compared to IVUS (p = 0.006). For convex geometries of aneurysms, the average % error of maximum diameter of aneurysm (MDA) was higher in angiography compared to IVUS (p = 0.018). No significant difference was observed for flat stenotic geometries (p = 0.48). The two techniques were significantly different for stenoses > 50% (which is generally considered hemodynamically significant, P = 0.01) compared to <50% stenosis (P = 0.21). IVUS also demonstrated smaller % error when underestimating stenoses (p = 0.03).
Conclusion
For the first time, this study demonstrates the accuracy of IVUS over angiography using a ground truth and paves a way to validate imaging techniques utilized to evaluate dialysis access functionality.
Funding
- NIDDK Support