In-vivo mechanical properties of aortic wall tissue

Thoracic aortic aneurysm (TAA) is a localized dilatation of the thoracic aorta resulting from progressive structural damage of the aortic wall. Focal weakening of the wall plays a key role in the evolution of the aneurysmal disease and is a concomitant factor that leads to aortic rupture. Mechanical characterization of the aneurysmal wall is critically important to locate spots of altered mechanical properties, which may highlight areas at high risk of rupture. Tissue characterization usually involves an invasive procedure requiring tissue harvesting and tensile testing. This procedure is not only invasive, but also not ideal to provide local constitutive properties. In this work we combine medical imaging with continuum mechanics principles to noninvasively assess local wall material properties from dynamic computed tomography (CT) images. As a proof of concept, we applied the methodology to a clinical case study of TAA. A triangular mesh of a patient-speci c TAA wall was obtained from a stack of low-dose dynamic 10-phase CT images. An iterative grey-scale feature-tracking approach was used to estimate velocities at each node of the aortic surface. Displacements were derived from the nodal velocities, which in turn allowed the calculation of the principal stretches for each element throughout the whole cardiac cycle. An estimate of the Cauchy stress throughout the cardiac cycle was obtained using the Law of Laplace, where the local curvatures were evaluated by an extended nite di erence approach. Once the strains and the stresses are known, the planar biaxial problem was solved analytically using a isotropic Neo-Hookean strain energy function with the shear modulus being the only unknown quantity. The constitutive parameter was then identi ed through a statistical parameter estimation procedure. Finally, we validated our approach by comparing the estimates of local material properties from cineMRI on porcine aortas with the material properties measured from bi-axial and uni-axial tensile tests on samples of tissue excised from the same porcine aortas. We found good agreement between the porcine aortas material properties estimated from our method and those measured from ex-vivo tensile tests (mean error of 13.4% 14.6%). The mechanical properties obtained for the patient speci c model were ii distributed in a complex pattern with large local variability with an average shear modulus of 593 505 kPa. The results showed a remarkable spatial variation with the aneurysmal ascending aorta exhibiting sti er behavior (average shear modulus 714 579 kPa) than the healthy descending aorta (average shear modulus 563 433 kPa). This mirrors earlier observations that indicated that the initial remodeling for the aneurysmal pathology is mediated by the deposition of collagen bers in order to sustain an increase in pressure. The aneurysmal arch showed areas of sti er tissue together with areas with low shear modulus (average shear modulus 576 556 kPa), which may suggest severe in ammation or discordant remodelling mechanisms typical for an advanced stage aneurysm. Our results demonstrate that local material properties for complex vascular pathologies may potentially be predicted noninvasively. Areas with decreased mechanical properties (low shear modulus) as identi ed using our approach may indicate local wall weakening, which could indicate an aorta at risk. In vivo focal weakening evaluation for any given patient is clinically feasible. Future studies will statistically correlate these ndings with intraoperative strength testing and histologic examination of the sample. Estimation of mechanical areas of weakness may play a critical role in the future evaluation of the aorta at risk.