Assessment of abnormal transvalvular flow and wall shear stress direction for pediatric/young adults with bicuspid aortic valve: A cross-sectional four-dimensional flow study

Fig. 1. Retrospective collection of 4D flow cardiovascular magnetic resonance (CMR) data. Bicuspid aortic valve (BAV) patients were identified from our 4D flow database (N = 669). Out of 108 pediatric/young adult BAV patients, those who underwent Fontan/Norwood/Ross surgeries (N = 6), data with quality issues, such as phase data with uncorrectable aliasing, were excluded (N = 35). After further exclusion of complex congenital heart defects and genetic defects (Turner syndrome, N = 12; hypoplastic left heart syndrome with patent ductus arteriosus, N = 1; double outlet right ventricle, N = 1), 35 BAV patients and 18 BAV patients with coarctation were finally included in this study. 4D four dimensional

Fig. 2. Analysis strategy for diameter/flow measurements using 4D flow cardiac magnetic resonance. (A) A typical phase-contrast magnetic resonance angiography (PCMRA) image used for aortic segmentation. (B) After planes were placed along the aortic centerline, sinotubular junction (STJ), mid-ascending aorta (AAo), and distal-AAo were determined. (C) Based on the landmarks specified, planes were divided into three domains and averaged hemodynamics were used to investigate the correlation between hemodynamics and diameters. 4D four dimensional

Fig. 3. Computation of abnormal flow directionality using control atlases. Aortic segmentations of controls were registered to determine averaged shape. Velocity and wall shear stress (WSS) vectors of the controls were mapped to the average geometry by affine registration and mean vectors were calculated at all the voxels of the aorta to create control atlas. The control atlas was then registered by affine registration to each bicuspid aortic valve (BAV) patient to calculate angle differences of the velocity/WSS vectors on patient’s aortic geometry, with angle difference >60 deg and >120 deg defined as moderately and severely abnormal, respectively (yellow and red volumes in the directionality map). Examples showing the peak systolic streamlines illustrate the distinct difference in flow directions in the regions where abnormal directionality was detected

Fig. 4. Correlation plots between mid-ascending aortic (AAo) diameter Z-scores and wall shear stress (WSS) (A: WSS magnitude, B: moderately abnormal WSS direction) for the BAV subgroup. Peak systolic WSS magnitude and streamlines in the aorta are shown for three cases with different levels of Z-scores (a-c, red dots). Representative flow/WSS directions at the AAo were illustrated by black dotted arrows. As the aortic Z-score increases, direction of AAo flow becomes more abnormal due to vortical and/or helical flow while WSS magnitude varies independent of the aortic Z-score