Evaluation of bioprosthetic heart valve failure using a matrix-fibril shear stress transfer approach

New results in the theory of plane strain flexure of incompressible isotropic hyperelastic materials

New results on the classical problem of bending by end moments for incompressible isotropic hyperelastic materials within the framework of nonlinear elasticity are investigated and presented in this paper. The particular results of concern here include (i) the adaptation of Rivlin's standard analysis to the case where one end of the beam is fixed and the other end is subjected to a bending moment; and (ii) results on the finite bending of (infinitesimally) thin isotropic hyperelastic plates which are valid for large deformations , extending the classical results from the linear elasticity theory which are restricted to small deformations. An interesting feature observed in this context is that a flexed thin plate develops an oscillatory surface along the circular arc near the free end, due to local (small) deviations of the radius of curvature. A potential application to the bending of a biological soft tissue, namely the aortic valve leaflet, is briefly described by way of an example. Finally, some new results are obtained for finite bending of hyperelastic materials that exhibit limiting chain extensibility at the molecular level and involve constraints on the deformation. The amount of bending that such materials can sustain is limited by the constraint. On using a limiting chain extensibility model, closed-form solutions for the Cauchy stress components, the bending moment and the normal out-of-plane force required to sustain the bending deformation are derived.

Evaluation of Pericardial Tissues from Assorted Species as a Tissue-Engineered Heart Valve Material

Decellularized pericardial tissue is a strong candidate for a TEHV material as ECM is present to guide cellular infiltration and fixed porcine and bovine pericardial tissue have existing use in bioprosthetic heart valves. In this work, we compare the mechanical and microstructural properties of decellularized-sterilized (DS) porcine, bovine, and bison pericardial tissues with respect to use as a TEHV. H&E staining was used to verify removal of cellular content post-decellularization and to evaluate collagen fiber structure. Additionally, uniaxial and biaxial tension testing were used to compare mechanical performance and, for the latter, acquire constitutive model parameters for subsequent finite element (FE) modeling. H&E staining revealed complete removal of cellular content and good collagen fiber structure. Tensile testing showed comparable mechanical strength between the three DS pericardial tissues and considerably stronger mechanical properties compared to native tissues. Bovine and bison DS pericardial tissues showed the strongest mechanical performance in the FE models with bison demonstrating the overall best mechanical characteristics. The increased thickness of bovine and bison tissues coupled with the strong mechanical behavior and ECM structure indicates that these materials will be resistant to damage until sufficient cellular infiltration has occurred such that damaged tissue can be repaired.

Different Passive Viscoelastic Properties Between the Left and Right Ventricles in Healthy Adult Ovine

Ventricle dysfunction is the most common cause of heart failure, which leads to high mortality and morbidity. The mechanical behavior of the ventricle is critical to its physiological function. It is known that the ventricle is anisotropic and viscoelastic. However, the understanding of ventricular viscoelasticity is much less than that of its elasticity. Moreover, the left and right ventricles (LV&RV) are different in embryologic origin, anatomy, and function, but whether they distinguish in viscoelastic properties is unclear. We hypothesized that passive viscoelasticity is different between healthy LVs and RVs. Ex vivo cyclic biaxial tensile mechanical tests (1, 0.1, 0.01 Hz) and stress relaxation (strain of 3, 6, 9, 12, 15%) were performed for ventricles from healthy adult sheep. Outflow track direction was defined as the longitudinal direction. Hysteresis stress-strain loops and stress relaxation curves were obtained to quantify the viscoelastic properties. We found that the RV had more pronounced frequency-dependent viscoelastic changes than the LV. Under the physiological frequency (1 Hz), the LV was more anisotropic in the elasticity and stiffer than the RV in both directions, whereas the RV was more anisotropic in the viscosity and more viscous than the LV in the longitudinal direction. The LV was quasi-linear viscoelastic in the longitudinal but not circumferential direction, and the RV was nonlinear viscoelastic in both directions. This study is the first to investigate passive viscoelastic differences in healthy LVs and RVs, and the findings will deepen the understanding of biomechanical mechanisms of ventricular function.

Modeling the Deformation of the Elastin Network in the Aortic Valve

This paper is concerned with proposing a suitable structurally motivated strain energy function, denoted by Weelastin network, for modeling the deformation of the elastin network within the aortic valve (AV) tissue. The AV elastin network is the main noncollagenous load-bearing component of the valve matrix, and therefore, in the context of continuum-based modeling of the AV, the Weelastin network strain energy function would essentially serve to model the contribution of the “isotropic matrix.” To date, such a function has mainly been considered as either a generic neo-Hookean term or a general exponential function. In this paper, we take advantage of the established structural analogy between the network of elastin chains and the freely jointed molecular chain networks to customize a structurally motivated Weelastin network function on this basis. The ensuing stress–strain (force-stretch) relationships are thus derived and fitted to the experimental data points reported by (Vesely, 1998, “The Role of Elastin in Aortic Valve Mechanics,” J. Biomech., 31, pp. 115–123) for intact AV elastin network specimens under uniaxial tension. The fitting results are then compared with those of the neo-Hookean and the general exponential models, as the frequently used models in the literature, as well as the “Arruda–Boyce” model as the gold standard of the network chain models. It is shown that our proposed Weelastin network function, together with the general exponential and the Arruda–Boyce models provide excellent fits to the data, with R2 values in excess of 0.98, while the neo-Hookean function is entirely inadequate for modeling the AV elastin network. However, the general exponential function may not be amenable to rigorous interpretation, as there is no structural meaning attached to the model. It is also shown that the parameters estimated by the Arruda–Boyce model are not mathematically and structurally valid, despite providing very good fits. We thus conclude that our proposed strain energy function Weelastin network is the preferred choice for modeling the behavior of the AV elastin network and thereby the isotropic matrix. This function may therefore be superimposed onto that of the anisotropic collagen fibers family in order to develop a structurally motivated continuum-based model for the AV.

A transverse isotropic viscoelastic constitutive model for aortic valve tissue

A new anisotropic viscoelastic model is developed for application to the aortic valve (AV). The directional dependency in the mechanical properties of the valve, arising from the predominantly circumferential alignment of collagen fibres, is accounted for in the form of transverse isotropy. The rate dependency of the valve's mechanical behaviour is considered to stem from the viscous (η) dissipative effects of the AV matrix, and is incorporated as an explicit function of the deformation rate ([Formula: see text]). Model (material) parameters were determined from uniaxial tensile deformation tests of porcine AV specimens at various deformation rates, by fitting the model to each experimental dataset. It is shown that the model provides an excellent fit to the experimental data across all different rates and satisfies the condition of strict local convexity. Based on the fitting results, a nonlinear relationship between η and [Formula: see text] is established, highlighting a 'shear-thinning' behaviour for the AV with increase in the deformation rate. Using the model and these outcomes, the stress-deformation curves of the AV tissue under physiological deformation rates in both the circumferential and radial directions are predicted and presented. To verify the predictive capabilities of the model, the stress-deformation curves of AV specimens at an intermediate deformation rate were estimated and validated against the experimental data at that rate, showing an excellent agreement. While the model is primarily developed for application to the AV, it may be applied without the loss of generality to other collagenous soft tissues possessing a similar structure, with a single preferred direction of embedded collagen fibres.