Regional age-related changes in aortic pulse wave velocity measured using magnetic resonance imaging

Objective: Aortic pulse wave velocity (PWV) is an independent predictor of cardiovascular risk. However, PWV is usually determined between the carotid and femoral sites, and ignoring the proximal ascending aorta which may stiffen most with age. We tested this hypothesis by assessing regional PWV along the length of the entire aorta using MRI. Design and Method: 147 healthy subjects aged 18 to 79 years, all free of cardiovascular disease and medication were recruited from the ACCT study. Seated and supine blood pressure was measured and carotid-femoral pulse wave velocity (PWVSphyg) was assessed by applanation tonometry (SphygmoCor, Australia). Phase contrast MRI (PCMRI) (1.5T scanner, GE) was then performed in all patients to determine PWV (PWVMRI) in 4 aortic regions, the arch (R1), descending-thoracic (R2), mid-thoracic (R3) and abdominal aorta (R4). Repeatability of PWVMRI was assessed across 2 visits, in a subset of 10 subjects. PWVMRI over the length of the whole aorta (aPWVMRI) was compared with PWVSphyg as validation. Results: The average PWVMRI for R1 to R4 were 4.6 ± 1.5m/s, 5.5 ± 2.0, 5.7 ± 2.3, and 6.1 ± 2.9m/s respectively. There was no difference between genders. The greatest age-related increase in PVWMRI occurred in R4 (0.9m/s per decade, P < 0.001) followed by R2 (0.7m/s, P < 0.001), R3 (0.6m/s, P < 0.001) and R1 (0.4m/s, P < 0.001). PWVMRI showed good repeatability (mean difference = -0.4 ± 2.1m/s). aPWVMRI and PWVSphyg were strongly correlated (r = 0.71, P < 0.001), although overall aPWVMRI was significantly lower than PWVSphyg (5.7 ± 1.8vs.7.3 ± 1.8m/s, P < 0.001). Conclusions: Age-related increases in arterial stiffness are greatest in the descending and abdominal aorta. Measurements of PWV using PCMRI are reproducible and correlate with surface measurements of PWV.