The association of structural versus load-dependent large artery stiffness mechanisms with cerebrovascular damage and cortical atrophy in humans

Abstract

Large central arterial stiffness is a risk factor for cerebrovascular damage and subsequent progression of neurodegenerative diseases, including Alzheimer's disease and dementia. However, arterial stiffness is determined by both the intrinsic components of the arterial wall (structural stiffness) and the load (i.e., arterial blood pressure) exerted upon it by the blood (load-dependent stiffness). This study aimed to determine the degree to which structural and/or load-dependent mechanisms of central arterial stiffness are associated with cerebrovascular damage. Among 128 healthy individuals (aged 63±6, age range: 50-80 years, 42% men), aortic and carotid artery stiffness was measured via carotid-femoral pulse wave velocity and B-mode ultrasonography, respectively. Using participant-specific exponential models, both aortic and carotid artery stiffness were standardized to a reference blood pressure to separate their structural and load-dependent stiffness mechanisms. Magnetic resonance imaging was used to derive total, periventricular, and deep cerebral white matter lesion volume (WMLV) and global cortical thickness. After adjusting for common cardiovascular disease risk factors, a 1 m/s increase in structural aortic stiffness was associated with 15% greater total WMLV (95% confidence interval [CI] = 0.01, 0.27, P = 0.036), 14% greater periventricular WMLV (95%CI = 0.004, 0.25, P = 0.044) and 0.011mm lower cortical thickness (95%CI = -0.022, -1.18, P = 0.028). No association was observed between structural carotid stiffness and WMLVs (total, periventricular, and deep), and neither aortic nor carotid load-dependent stiffness was associated with WMLVs or cortical thickness. Structural, not load-dependent, mechanisms of aortic stiffness are related to cerebrovascular-related white matter damage.

Keywords: Blood pressure; Brain; Cognitive decline; Hypertension; Vascular age.

Fig. 1

Example of neuroimaging biomarkers of cerebrovascular damage and cortical atrophy derived from magnetic resonance imaging. Cortical thickness (panel A) is calculated using a T1-weighted scan as the average distance from the pial (red outline) to the white matter boundary (yellow). White matter lesions (panel B), including deep white matter lesions (blue) and periventricular white matter lesions (pink) segmented on a T2-weighted FLAIR scan

Fig. 2

Bivariate correlations of aortic and carotid structural mechanisms of arterial stiffness with total (TWMLV), periventricular (PWMLV) and deep (DWMLV) white matter lesion volume and cortical thickness

Fig. 3

Bivariate correlations of aortic and carotid load-dependent (LD) mechanisms of arterial stiffness with total (TWMLV), periventricular (PWMLV) and deep (DWMLV) white matter lesion volume and cortical thickness