Dimitrios P. Sokolis

Ascending thoracic aortic aneurysms are associated with compositional remodeling and vessel stiffening but not weakening in age-matched subjects

OBJECTIVE We sought to examine in age-matched subjects the biomechanical and compositional remodeling associated with ascending thoracic aortic aneurysms according to region and direction. METHODS Whole, fresh, degenerative ascending thoracic aortic aneurysms were taken from 26 patients (age, 69 +/- 2 years; maximum aortic diameter, 5.9 +/- 0.3 cm) during elective surgical intervention, and 15 nonaneurysmal ascending thoracic aortas were obtained during autopsies (age, 66 +/- 3 years; maximum aortic diameter, 3.3 +/- 0.2 cm). These were cut into anterior, right lateral, posterior, and left lateral regions, and circumferentially and longitudinally oriented specimens were prepared. The aortic specimens were submitted to histomorphometric and biomechanical studies, including measurement of failure strain (ie, extensibility), failure stress (ie, strength), and peak elastic modulus (ie, stiffness). RESULTS Wall elastin, but not collagen content, decreased in aneurysmal specimens, displaying lower wall thickness and failure strain, higher peak elastic modulus, and equal failure stress than control specimens in the majority of regions and directions. Similar differences were noted in pooled data from all regions. Regional variations in mechanical parameters were mostly found in longitudinally oriented tissue. Circumferential specimens showed higher failure stress and peak elastic modulus but equal failure strain than longitudinal specimens. CONCLUSIONS Our findings contradict previous studies on ascending thoracic and abdominal aortic aneurysms, suggesting that the former might not cause weakening but rather only stiffening and reduction in tissue extensibility and elastin content. Marked heterogeneity was evident in healthy and aneurysmal aortas. The present data offer insight into the pathogenesis of aneurysm dissection. Information on directional and regional variations is pertinent because dissections develop circumferentially and bulging preferentially occurs in the anterior region.

Regional and directional variations in the mechanical properties of ascending thoracic aortic aneurysms

This study aimed to assess regional and directional differences in the mechanical properties of ascending thoracic aortic aneurysms (ATAA). Whole fresh ATAA were taken from twelve patients, undergoing elective surgical repair, and cut into tissue specimens. These were divided into groups according to direction and region, and subjected to uniaxial testing beyond rupture. In the majority of tests, the inner layers of the aortic wall ruptured first; failure stress (measure of tissue strength) and peak elastic modulus (measure of tissue stiffness) were significantly higher circumferentially in all regions. Marked heterogeneity was evident in the mechanical properties of ATAA, with the anterior region longitudinally being the weakest and least stiff of all regions. No correlation was found between failure stress and ATAA diameter or patient age. Failure stress showed inverse correlations with wall thickness and direct correlations with peak elastic modulus. The current information, relating to regional and directional differences, may provide a better understanding of the mechanism responsible for the development of circumferential tears of the inner aortic wall layers in ATAA dissections.

Biomechanical response of ascending thoracic aortic aneurysms: association with structural remodelling

Ascending thoracic aortic aneurysms (ATAA) were resected from patients during graft replacement and non-aneurysmal vessels during autopsy. Tissues were histomechanically tested according to region and orientation, and the experimental recordings reduced with a Fung-type strain–energy function, affording faithful biomechanical characterisation of the vessel response. The material and rupture properties disclosed that ATAA and non-aneurysmal aorta were stiffer and stronger circumferentially, accounted by preferential collagen reinforcement. The deviation of microstructure in the right lateral region, with a longitudinal extracellular matrix and smooth muscle element sub-intimally, reflects the regional differences in material properties identified. ATAA had no effect on strength, but caused stiffening and extensibility reduction, corroborating our histological observation of deficient elastin but not collagen content. Our findings may serve as input data for the implementation of finite element models, to be used as improved surgical intervention criteria, and may further our understanding of the pathophysiology of ATAA and aortic dissection.

Biomechanical and histological characteristics of passive esophagus: Experimental investigation and comparative constitutive modeling

Information on the passive biomechanical properties of two-layered esophagus is still limited, although this would enhance our understanding of its physiology/pathophysiology and help to address problems in surgery, medical-device applications, and for the optimal design of prostheses. In this study, rabbit esophagi were excised and dissected into mucosa-submucosa and muscle layers that were submitted to histological quantification of elastin and collagen content and orientation, as well as to inflation-extension testing and geometrical analysis, i.e. delineation of the zero-stress state serving as a reference configuration for biomechanical analysis. The pressure-radius data of both layers displayed a monotonically rising slope with inflating pressure, unlike the sigma shape characterizing elastin-rich tissues, for which biphasic constitutive models were initially postulated. Three phenomenological expressions of strain-energy function (SEF), commonly appearing in soft-tissue biomechanics literature, were used in an attempt to model the pseudoelastic response of esophageal tissue, namely the exponential Fung-type SEF, and the combined neo-Hookean (isotropic) or quadratic (anisotropic) and exponential Fung-type SEF. Accurate fits were attained for the pressure-radius-force data, spanning a wide range of longitudinal stretch ratios, when using the exponential form; the biphasic SEFs failed to generate improved fits, being also over-parameterized. According to the calculated material parameters, mucosa-submucosa was stiffer than muscle in both directions, justified by our histological observation of increased collagen content in that layer, and tissue was stiffer longitudinally, substantiated by the increased elastin and collagen contents and their preferential alignment towards that direction. Our results demonstrate that the passive response of esophagus is best modeled with an exponential Fung-type SEF.

Experimental investigation and constitutive modeling of the 3D histomechanical properties of vein tissue

Numerous studies have provided material models of arterial walls, but limited information is available on the pseudo-elastic response of vein walls and their underlying microstructure, and only few constitutive formulations have been proposed heretofore. Accordingly, we identified the histomechanics of healthy porcine jugular veins by applying an integrated approach of inflation/extension tests and histomorphometric evaluation. Several alternate phenomenological and microstructure-based strain-energy functions (SEF) were attempted to mimic the material response. Evaluation of their descriptive/predictive capacities showed that the exponential Fung-type SEF alone or in tandem with the neo-Hookean term did not capture the deformational response at high pressures. This problem was solved to a degree with the neo-Hookean and two-fiber (diagonally arranged) family SEF, but altogether the least reliable fit was generated. Fitting precision was much improved with the four-fiber (diagonally, circumferentially, longitudinally arranged) family model, as the inability of neo-Hookean function with force data was alleviated by use of the longitudinal-fiber family. Implementation of a quadratic term as a descriptor of low-pressure anisotropy facilitated the simulation of low-pressure and force data, and the four-fiber families simulated more faithfully than the two-fiber families the physiologic and high-pressure response. Importantly, this SEF was consistent with vein angioarchitecture, namely the occurrence of extensive elastin fibers along the longitudinal axis and few orthogonal fibers attached to them and of three collagen sets with circumferential, longitudinal, and diagonal arrangement, respectively. Our findings help to establish the relationship between vein microstructure and its biomechanical response, yet additional observations are obligatory prior to endeavoring generalizations to other veins.

Spectral decomposition of the compliance tensor for anisotropic plates

The spectral decomposition of compliance S is extended to the principal stress planes offering a possibility of characterization of the elastic properties of anisotropic media under plane-stress conditions. It is shown that the three eigenvalues of S, together with a “new” dimensionless parameter ωp, called the plane eigenangle, constitute the essential parameters for an invariant description of the elastic behaviour of anisotropic plates. Both the variational limits of the eigenangle ωp and the restrictive bounds to the values of the Poissons ratios imposed by thermodynamics are considered. Finally, it is shown that the plane eigenangle ωp may be employed as a monoparametric indication of the anisotropy of the material.

Local Hemodynamics and Intimal Hyperplasia at the Venous Side of a Porcine Arteriovenous Shunt

Venous anastomotic intimal hyperplasia (IH) observed in the arteriovenous shunt (AVS) has been associated with disturbed hemodynamics. This study aims to correlate hemodynamics with wall histology and wall mechanics by examining the flow field in AVS with computational fluid dynamics using experimental data taken from in vivo experiments. Input data to the computational model were obtained in vivo one month after AVS creation; adjacent vessels were submitted to histological and mechanical examination. The 3-D shunt geometry was determined using biplane angiography. Ultrasound measurements of flow rates were performed with perivascular flow probes and pressures were recorded through intravascular catheters. These data were considered as boundary conditions for calculation of the unsteady flow field. Numerical findings are suggestive of strong Dean vortices toward both vein flow exits, verified by color Doppler. The high wall shear stresses (WSSs) and their gradients appear to be related to areas of IH and vessel wall stiffening, as evidenced in preliminary histological and mechanical studies of the venous wall. Additionally, suture line hyperplasia seems to be aggravated by the high WSS gradients noted at the transition line from graft to vein.

Strain-energy function and three-dimensional stress distribution in esophageal biomechanics

Knowledge of the transmural stress and stretch fields in esophageal wall is necessary to quantify growth and remodeling, and the response to mechanically based clinical interventions or traumatic injury, but there are currently conflicting reports on this issue and the mechanical properties of intact esophagus have not been rigorously addressed. This paper offers multiaxial data on rabbit esophagus, warranted for proper identification of the 3D mechanical properties. The Fung-type strain-energy function was adopted to model our data for esophagus, taken as a thick-walled (1 or 2-layer) tubular structure subjected to inflation and longitudinal extension. Accurate predictions of the pressure-radius-force data were obtained using the 1-layer model, covering a broad range of extensions; the calculated material parameters indicated that intact wall was equally stiff as mucosa-submucosa, but stiffer than muscle in both principal axes, and tissue was stiffer longitudinally, concurring our histological findings (Stavropoulou et al., Journal of Biomechanics. 42 (2009) 2654-2663). Employing the material parameters of individual layers, with reference to their zero-stress state, a reasonable fit was obtained to the data for intact wall, modeled as a 2-layer tissue. Different from the stress distributions presented hitherto in the esophagus literature, consideration of residual stresses led to less dramatic homogenization of stresses under loading. Comparison of the 1- and 2-layer models of esophagus demonstrated that heterogeneity induced a more uniform distribution of residual stresses in each layer, a discontinuity in circumferential and longitudinal stresses at the interface among layers, and a considerable rise of stresses in mucosa, with a reduction in muscle.

A passive strain-energy function for elastic and muscular arteries: correlation of material parameters with histological data

A plethora of phenomenological and structure-motivated constitutive models have thus far been used as pseudoelastic descriptors in arterial biomechanics, but their parameters have not been explicitly correlated with histology. This study associated biaxial histological data with strain-energy function (SEF) parameters derived from uniaxial tension data of arteries from different topographical sites (carotid artery vs. thoracic aorta vs. femoral artery). A two-term SEF fitted the passive stress–strain data of healthy porcine tissue, justified by the biphasic response characterizing elastin-rich tissues. Selection of a quadratic (orthotropic) over the neo-Hookean (isotropic) term was dictated by the directional dissimilarities in low-stress mechanical response, consistent with our histological data indicating orthotropic symmetry for unstressed elastin. Use of the exponential term was dictated by mechanical dissimilarities at high stresses and variations in unstressed collagen composition and orientation. Accurate fits were attained; topographical variations and anisotropy in material parameters were accounted by respective variations in histomorphometrical data.

Differential histomechanical response of carotid artery in relation to species and region: mathematical description accounting for elastin and collagen anisotropy

The selection of a mathematical descriptor for the passive arterial mechanical behavior has been long debated in the literature and customarily constrained by lack of pertinent data on the underlying microstructure. Our objective was to analyze the response of carotid artery subjected to inflation/extension with phenomenological and microstructure-based candidate strain-energy functions (SEFs), according to species (rabbit vs. pig) and region (proximal vs. distal). Histological variations among segments were examined, aiming to explicitly relate them with the differential material response. The Fung-type model could not capture the biphasic response alone. Combining a neo-Hookean with a two-fiber family term alleviated this restraint, but force data were poorly captured, while consideration of low-stress anisotropy via a quadratic term allowed improved simulation of both pressure and force data. The best fitting was achieved with the quadratic and Fung-type or four-fiber family SEF. The latter simulated more closely than the two-fiber family the high-stress response, being structurally justified for all artery types, whereas the quadratic term was justified for transitional and muscular arteries exhibiting notable elastin anisotropy. Diagonally arranged fibers were associated with pericellular medial collagen, and circumferentially and longitudinally arranged fibers with medial and adventitial collagen bundles, evidenced by the significant correlations of SEF parameters with quantitative histology.