Two-minute k-space and time–accelerated aortic fourdimensional flow MRI: Dual-center study of feasibility and impact on velocity and wall shear stress quantification

Purpose To investigate the two-center feasibility of highly k-space and time (k-t)–accelerated 2-minute aortic four-dimensional (4D) flow MRI and to evaluate its performance for the quantification of velocities and wall shear stress (WSS). Materials and Methods This cross-sectional study prospectively included 68 participants (center 1, 11 healthy volunteers [mean age ± standard deviation, 61 years ± 15] and 16 patients with aortic disease [mean age, 60 years ± 10]; center 2, 14 healthy volunteers [mean age, 38 years ± 13] and 27 patients with aortic or cardiac disease [mean age, 78 years ± 18]). Each participant underwent highly accelerated 4D flow MRI (k-t acceleration, acceleration factor of 5) of the thoracic aorta. For comparison, conventional 4D flow MRI (acceleration factor of 2) was acquired in the participants at center 1 (n = 27). Regional aortic peak systolic velocities and three-dimensional WSS were quantified. Results k-t–accelerated scan times (center 1, 2:03 minutes ± 0:29; center 2, 2:06 minutes ± 0:20) were significantly reduced compared with conventional 4D flow MRI (center 1, 12:38 minutes ± 2:25; P <.0001). Overall good agreement was found between the two techniques (absolute differences ≤15%), but proximal aortic WSS was significantly underestimated in patients by using k-t–accelerated 4D flow when compared with conventional 4D flow (P ≤.03). k-t–accelerated 4D flow MRI was reproducible (intra- and interobserver intraclass correlation coefficient ≥0.98) and identified significantly increased peak velocities and WSS in patients with stenotic (P ≤.003) or bicuspid (P ≤.04) aortic valves compared with healthy volunteers. In addition, k-t–accelerated 4D flow MRI–derived velocities and WSS were inversely related to age (r ≥−0.53; P ≤.03) over all healthy volunteers. Conclusion k-t–accelerated aortic 4D flow MRI providing 2-minute scan times was feasible and reproducible at two centers. Although consistent healthy aging- and disease-related changes in aortic hemodynamics were observed, care should be taken when considering WSS, which can be underestimated in patients