Recent studies have demonstrated the potential diagnostic value of 4D flow MRI for evaluating altered aortic hemodynamics in patients with aortopathy and after aortic valve replacement (AVR).^1,2 In this study, 4D flow MRI was used to characterize aortic blood flow in patients who have undergone AVR with bioprosthetic vs. mechanical valves.

ECG and respiratory gated 4D flow MRI with full volumetric coverage of the thoracic aorta was acquired in 19 patients who received AVR at Northwestern Memorial Hospital (Chicago, IL, USA) between 2012 and 2015 (n = 10 bioprosthetic valves, n = 9 mechanical valves), as well as 6 healthy controls matched for age and blood pressure. All acquisitions were made on 1.5T MRI systems (Siemens, Erlangen, Germany) during free breathing with average spatial resolution of approx. 2.5 mm³ and temporal resolution of 38-42 ms. 4D flow data analysis included corrections for Maxwell terms, eddy currents and velocity aliasing (Matlab, MathWorks Inc., MA, USA) and 3D segmentation of the aortic volume (Mimics, Materialize NV, Leuven, Belgium).

Blood flow in the thoracic aorta was visualized in EnSight (CEI, Apex, NC, USA) by way of 3D pathlines showing the time-resolved evolution of velocities throughout one cardiac cycle. Nine analysis planes were placed along the aorta, perpendicular to the aortic axis, at the following anatomical landmarks as shown in Figure 1: 1) at the aortic root 1-2 cm above the valve, 2) at the proximal ascending aorta (AAo) 1-2 cm above the sinotubular junction, 4) at the distal AAo just before the brachiocephalic trunk, 3) at the mid-AAo halfway between planes 2 and 4, 5) at the mid-arch just before the common carotid artery, 6) at the distal arch just before the left subclavian artery, 7) at the proximal descending aorta (DAo), 8) at the DAo at the level of plane 2, and 9) at the distal DAo. In all analysis planes, quantification of peak systolic velocities was performed.

In addition, peak systolic 3D wall shear stress (WSS) was calculated along the entire segmented 3D aorta wall based on a previously reported method.^3,4 As shown in Figure 2, the aorta was partitioned into 10 distinct regions of interest (ROI), corresponding to the inner (along the lesser curvature) and outer (along the greater curvature) segments of the proximal AAo, distal AAo, arch, proximal DAo, and distal DAo, respectively. Mean peak systolic WSS was computed inside each ROI.

Inter-group differences in peak systolic velocity and peak WSS were studied using Welch’s t-test with a p-value less than 0.05 denoting statistical significance.

Bioprosthetic valves produced significantly higher peak systolic velocities than controls in the root (2.24 vs. 1.25 m/s, p < 0.01) and all three AAo planes (1.93 vs. 1.21 m/s, p < 0.01; 1.58 vs. 1.02 m/s, p < 0.01; 1.23 vs. 0.89 m/s, p = 0.01), while AVR with mechanical valves resulted in higher velocities than controls in only the root (2.09 vs. 1.25 m/s, p < 0.01) and proximal AAo (1.58 vs. 1.21 m/s, p = 0.03). When comparing the two AVR groups, bioprosthetic valves produced greater velocities than mechanical valves in the proximal and mid-AAo. Both bioprosthetic and mechanical valves had greater peak WSS than controls in the inner proximal AAo (2.70 and 2.22 vs. 1.40 N/m2, both p < 0.01) and outer proximal AAo (2.52 and 2.02 vs. 1.31 N/m2, both p < 0.01). There were no significant differences in peak WSS in the AAo between the two AVR groups. Additionally, mechanical valves had greater WSS than bioprosthetic valves in the inner proximal DAo, as well as in the inner and outer distal DAo. Tables 1 and 2 provide a full tabulation of results.The results of our study show that there are significant changes to aortic blood flow in AVR patients compared to healthy controls, which include elevated peak systolic velocities and WSS in the ascending aorta. Although increased WSS in the AAo has been implicated in the development of aortic pathology ⁵, the long-term clinical outcomes of AVR patients with elevated systolic velocities and WSS on 4D flow MRI are not well-understood and present opportunities for future research. Subsequent analyses will investigate the independent influence of aortic stiffness related to variable extents of graft replacement of the aorta on local blood flow. Accurate characterization of post-operative blood flow by 4D flow MRI might differentiate the efficacies of various surgical approaches to AVR in reproducing physiologic flow patterns and reducing further aortopathy that may result from derangements to aortic hemodynamics over time.