Changes in Segmental Pulse Wave Velocity of the Thoracic Aorta With Age and Left Ventricular Remodelling. An MRI 4D Flow Study

Journal of Hypertension Manuscript
Authors Gilles Soulat, Umit Gencer, Nadjia Kachenoura, Olivier Villemain, Emmanuel Messas, Pierre Boutouyrie, Stéphane Laurent, Elie Mousseaux

Objectives: Pulse wave velocity (PWV) of the aortic arch is usually estimated by using 2D phase contrast in MRI. Thanks to 4D flow MRI, segmental PWV of the ascending and descending aorta, as well as PWV of the entire thoracic aorta can now be estimated within the same examination. Our objective is to compare PWVs obtained by 2D and 4D PC, through their relationships with carotid-femoral PWV (cf-PWV), age and left ventricular remodelling.

Basic methods: MRI examinations were performed at 3 Tesla, including 2D PC acquisitions with through-plane velocity encoding and sagittal 4D phase contrast acquisitions covering the thoracic aorta volume. PWVs were calculated after estimating aortic lengths and flow transit times between the ascending aorta and descending aorta in 2D and between valve, isthmus and diaphragm in 4D resulting in 2D-PWV, 4D-TA-PWV; 4D-AA-PWV, 4D-DA-PWV.

Main results: Fifty-seven healthy volunteers (25 men, age 51 years ± 17) were studied. All MRI-PWVs were correlated with cf-PWV (r = 0.67; r = 0.63: r = 0.47; r = 0.61 for 2D-PWV, 4D-TA-PWV; 4D-AA-PWV, 4D-DA-PWV, respectively, P < 0.001). 2D-PWV and 4D-TA-PWV were strongly related with age (r = 0.76 and r = 0.77, respectively). The highest correlation, between left ventricular thickness or LV mass/end diastolic volume (EDV) ratio and segmental PWVs of the thoracic aorta was found with 4D-AA-PWV (r = 0.43, P < 0.01 and r = 0.48, P < 0.01).

Principal conclusions: Global and segmental PWV analysis of the thoracic aorta can be accurately assessed using 4D flow MRI. 4D-PWVs were highly correlated with ageing and cf-PWV. The strong association between the ascending aorta stiffness and the left ventricular remodelling in healthy volunteers is encouraging to better estimate left ventricular afterload.