Correlation of mechanical stresses with arterial disease frequency in animal models

Abstract

Atherosclerosis is a complex inflammatory disease which may be triggered by an elevated transport of lipid-carrying macromolecules from the blood into the arterial wall. Its non-uniform distribution has been attributed to local variations in wall shear stress. Intramural stresses and strains are regularly overlooked, despite their heterogeneous distribution and direct effect on vascular cell morphology and function.

Stenotic flow models of atherosclerosis inadvertently alter both fluid and solid stresses. Using a tapered flow modifying cuff, wall uptake of plasma macromolecules was investigated in the murine carotid artery. The greatest uptake occurred just upstream, the site which has previously been observed to later develop vulnerable atherosclerotic plaques. Uptake was also elevated, but to a significantly lesser extent, just downstream. Flow simulations and the effects of reversing cuff orientation on wall permeability, indicated a role for solid, as well as fluid stresses. Structural simulations revealed that steep spatial and cyclic stress/strain gradients occur at the cuff margins. When fluid-structure interaction effects are included, transmural pressure and circumferential stress and strain are reduced downstream.

Arterial branch sites exhibit large variations in biomechanical factors and a predilection for atherosclerotic lesions. Motivated by the need to accurately map strains around branch openings, a novel method was developed to determine in vivo strain in animal models using vascular corrosion casts. A study focused on the descending thoracic rabbit aorta demonstrated its efficacy.

Assessing the anatomical correlation of atherosclerosis with biomechanical localising factors is hindered by spatial autocorrelation, which tends to exaggerate significance, and by the use of aggregate data, which artificially inflates correlation coefficients. A comparison of four statistical tests for assessing spatial correlations highlighted substantial differences in obtained significance when applied to maps of wall shear stress and three putatively related arterial properties. A fifth test specific to aggregate data was proposed and applied. Appropriate application of these techniques will help to establish the relative importance of fluid and solid mechanics in atherogenesis.