Related Experiment Video
Updated: May 26, 2026
In vitro Assessment of Aortic Regurgitation Using Four-Dimensional Flow Magnetic Resonance Imaging
Published on: February 25, 2022
Impact of an aortic nitinol stent graft on flow measurements by time-resolved three-dimensional velocity-encoded MRI
Fabian Rengier1, Michael Delles, Roland Unterhinninghofen
1Department of Diagnostic and Interventional Radiology, University Hospital Heidelberg, Germany. fabian.rengier@web.de
Rationale And Objectives:
Three-dimensional (3D) velocity-encoded cine (VEC) magnetic resonance imaging (MRI) has the potential to quantify 3D hemodynamic aspects known from computational fluid dynamics and to be used to identify hemodynamic risk factors for complications of endovascular aortic repair. The purpose of this study was to investigate the impact of an aortic nickel-titanium (nitinol) stent graft on the accuracy of flow measurements by 3D VEC MRI.
Materials And Methods:
A pump generated pulsatile aortic flow in an elastic tube phantom mimicking the aorta. Stacked two-dimensional three-directional VEC MRI (stacked-2D-3dir-MRI), 3D three-directional VEC MRI (3D-3dir-MRI), and gold-standard 2D through-plane VEC MRI were applied before and after the insertion of an aortic nitinol stent graft. Six equidistant levels were analyzed twice by the same reader. The percentage difference of the measured flow rate from the gold standard was defined as the parameter of accuracy.
Results:
The overall accuracy of in-stent flow measurements related to the gold standard was -5.4% for stacked-2D-3dir-MRI and -4.1% for 3D-3dir-MRI, demonstrating significant overall underestimation compared to the gold standard (P = .016 and P = .013). However, flow measurements with the stent graft were significantly overestimated by 4.1% using stacked-2D-3dir-MRI (P < .001) and by 5.4% using 3D-3dir-MRI (P = .003) compared to identical measurements without the stent graft. In stacked-2D-3dir-MRI, this positive bias was significantly greater at the proximal and distal ends of the stent graft (P = .025). In 3D-3dir-MRI, measurements along the whole length of the stent graft were affected (P = .006). Intraobserver agreement was excellent, with intraclass correlation coefficients of 0.94 for stacked-2D-3dir-MRI (P < .001) and 0.90 for 3D-3dir-MRI (P < .001).
Conclusions:
Flow measurements within an aortic nitinol stent graft by 3D VEC MRI are feasible, but stent grafts may cause a significant positive bias.
