For the first objective, an extensive set of human aortic and pulmonary valves was quantitatively analyzed in terms of geometry, structure, and material properties. Based on this information, subject-specific computational models were developed to estimate the mechanical state for both valves as a function of age. Although mechanical stress is often hypothesized to be preserved in cardiovascular tissues, this parameter was predicted to increase with age and have different magnitudes for the aortic and the pulmonary valve. Strikingly, the deformation (stretch) of the valve was fairly stable with age and remarkably similar for both the aortic and pulmonary valve. Accordingly, these data suggest that stretch serves as a homeostatic mechanical parameter in heart valves.
Next, computational algorithms were developed to understand the relative importance of growth (i.e. changes in volume) and remodeling (i.e. changes in material properties) for preserving stretch homeostasis. Systematic analyses revealed that growth is most important for preserving stretch early in life, whereas tissue remodeling, surprisingly, appeared to be unfavorable with regard to preserving stretch. In contrast, tissue remodeling was identified as being the crucial factor for maintaining stretch later in life, whereas changes in the valve geometry had a negative effect in terms of preserving stretch homeostasis.
For the second objective, computational models were developed to obtain a more fundamental understanding of the biological processes responsible for growth and homeostasis. Specifically, the effects of mechanical factors on cell-cell communication were modeled via incorporating experimental data into a theoretical framework. Predictions with these models revealed that establishment of homeostasis can arise suddenly upon certain changes in tissue geometry. This study therefore provides a biological motivation for the emergence of homeostasis depending on the mechanical state of the tissue, where the integration of mechanics and biology into computational models has been essential to understand the implications of mechanobiological mechanisms.
To predict the growth potential of tissue-engineered valves, a computational framework was developed where growth was predicted locally based on the local mechanical state. The prediction of native valve growth demonstrated that the assumption of growth in response to stretch enables the establishment of a homogeneous stretch state. Interestingly, although the amount of growth now varied locally within the valve, the total amount of growth was still in good agreement with the experimentally observed valve growth. Furthermore, it was noticed that the predicted heterogeneity in leaflet thickness matched very well with the observed geometry of native valves. Application of this framework to predicting growth of tissue-engineered heart valves demonstrated that growth does not necessarily lead to preservation of a physiological or functional valve geometry. It appears that there are many factors that contribute to valve growth, where material parameters, valve geometry, and boundary conditions have a major effect.
Dissemination of the research results was realized through 4 scientific publications (of which 1 in the prestigious PNAS journal) and 10 conference presentations. Additional publications are expected in the future. Furthermore, theory and results have been included in courses at TU/e, and formed the basis for student projects.
