Purpose: Implantable left ventricular assist devices (LVAD) can restore blood circulation for patients with end-stage heart failure. However, significant complications such as stroke may occur due to the unfavorable hemodynamic environments produced by suboptimal implantation configuration. High-speed angiography (HSA) was performed at 1000 fps to resolve previously unseen vascular flow dynamics in-vitro for one LVAD outflow graft configuration.
Methods: Flow detail was evaluated within a 3D-printed aortic vascular model. The imaging ROI focused on the anastomosis of a 10-mm diameter LVAD graft with the wall of the ascending aorta, at an insertion angle of 90 degrees with respect to the centerline of the aorta. The model was connected to a constant flow loop, with a system flow rate set to 5 L/min, as measured at the graft. The aortic valve was assumed to remain closed, so all flow into the aorta originated from the LVAD graft mimicking complete LVAD support. Iodine-based contrast media was slowly injected with an automated syringe injector and imaged at 1000 fps with the new 5cm x7.5cm FOV CdTe-based Aries photon-counting detector.
Results: Examination of the image sequences illustrates detailed flow dynamic patterns made available through use of the high-frame-rate-enabled, larger-FOV detector. The iodine contrast-labeled fluid can be seen exiting the LVAD graft as a high-velocity jet, which directly impinges on the aortic wall opposite the graft. The 90-degree configuration causes a strong recirculation pattern to form inferior to the anastomosis zone.
Conclusion: This investigation presents a novel HSA application, where the severe hemodynamic conditions produced by the high velocity jet from a 90-degree LVAD graft configuration are directly visualized in-vitro. This initial study sets the groundwork for further investigation into other LVAD graft configurations, where better alignment of the outflow graft anastomoses with the aortic axis may produce more favorable flow patterns in the aorta.
Funding Support, Disclosures, and Conflict of Interest: This research was supported in part by Canon Medical Systems Corporation and NIH Grant 1R01EB030092.