Characterization of Blood Flow Changes in Normal and Pathological Aortic Dilation from 4D Flow Magnetic Resonance Imaging

ARIEL PASCANER; THOMAS DIETENBECK; YASMINA CHENOUNE; SOPHIA HOURIEZ--GOMBAUD-SAINTONGE; UMIT GENCER; EMILIE BOLLACHE; GILLES SOULAT; NADJIA KACHENOURA; DAMIAN CRAIEM; ELIE MOUSSEAUX

Conferencia; International Conference in Computing in Cardiology 2019; 2019

Aim: Maximal diameter (Dmax), which is commonly used to diagnose thoracic aortic aneurysm (TAA) was previously shown to be normal in 20-30% of patients who ultimately develop dissection. Besides, inner aortic flow is associated with its wall dynamics. Thus, our aim was to quantify aortic flow changes using 4D flow MRI in the setting of ascending aorta (AA) dilation. Methods: We studied 20 patients with TAA and tricuspid aortic valve (TAV) and 56 healthy controls (30 subjects, 36±9y ≤50 years named YC, 26 subjects, 65±9y >50 years named OC). All underwent 4D flow MRI. After aortic segmentation, regional volume of backward flow (VBF) was extracted in addition to in-cross-section velocity standard deviation (SD) as well as maximal velocity jet angle (Angle) and eccentricity (Ecc). Receiver operating characteristic (ROC) analysis was performed to assess ability of flow indices to characterize dilation. Results: While AA Dmax changed by 1.4 folds between TAV and OC, VBF changed by 6.5 folds, and Ecc, Angle and SD changed by 1.3 to 1.9 folds between the two groups. Moreover VBF varied consistently with age and was able to detect AA dilation with an accuracy of 0.98. Conclusion: 4D flow MRI indices of local aortic flow disorganization, specifically backward flow, were able to accurately characterize dilation.