Characterization of Rate Dependency and Inhomogeneity of Aortic Tissue

Abstract

Traumatic aortic rupture (TAR) is one of the leading causes of morbidity and mortality in motor-vehicle accidents with the majority of injuries occurring in the peri-isthmus region. To date, the mechanisms of aorta injury are poorly understood as this injury cannot be replicated reliably in cadaver crash tests. Due to inconclusiveness of the experimental tests, finite element (FE) modeling is often used to gain a better insight into the mechanisms of TAR. However, the FE models are also hindered by many unknowns particularly the soft tissues biomechanical responses. A crucial step to improve the FE models of blunt chest trauma is to advance our understanding of the local mechanical properties of aortic tissue subject to high loading rates associated with TAR. The objective of this dissertation was to investigate the effects of tissue rate dependency and inhomogeneity in the modeling of loading conditions that lead to TAR. The material properties of human aorta in large deformations and high loading rates were characterized based on oscillatory biaxial tests. It was shown that a quasilinear viscoelastic (QLV) model with the instantaneous elastic response of the second order and the reduced relaxation function with one exponentially decaying term could describe the experimental results between 20 Hz and 130 Hz. The obtained decay rates (in the range of 70 to 550 s-1) were 10 to 100 folds higher than previously reported values and showed significant rate dependence within 10 ms after the loading. It was shown that the rate dependent properties, similar to the elastic properties, were anisotropic with generally higher decay rate and stiffness observed in the circumferential direction compared to the longitudinal direction. The inhomogeneity of porcine descending thoracic aorta was characterized in three dimensions using a nano-indentation technique and QLV modeling approach. The tests were conducted in the axial, circumferential, and radial orientations with about 100 micrometer spatial resolution. Aortic tissue was divided into 10 regions across the thickness, 4 quadrants in the circumferential direction, and 3 sections in the longitudinal direction. While across the thickness, the results in different orientations were significantly different, four distinct layers were identified that were matched with the anatomical features. In the axial direction, the medial quadrant, and in all directions, the proximal DTA had the lowest stiffness. The results predict that under equal stresses, the inner layers of the medial quadrant in upper DTA would undergo more strains and will be therefore more prone to failure. This prediction is in agreement with clinical observations. The inhomogeneity and rate dependency of aorta were implemented in the Global Human Body Models Consortium full-body FE model. It was demonstrated that in a simulation of blunt chest impact, both features significantly affected the tissue strain levels particularly in the isthmus, arch, and ascending aorta. Accurate quantifications of these features are essential to assess the risk of aortic injury based on FE models.