Computer-aided cartilage tissue-engineering : a numerical evaluation of the influence of inhomogeneities, collagen architecture and temporal culture effects

Hyaline articular cartilage has a crucial role in the distribution of joint mechanical loads and smooth movement of bones. Because of its poor healing capacity, cartilage damage is progressive and may lead to osteoarthritis (OA). Replacing damaged cartilage with tissue engineered (TE) cartilage is promising for the treatment of OA. Despite improvements, however, the inferior mechanical property of today’s TE cartilage is one of the main reasons, which causes unsatisfactory mid- to long-term outcome in clinical applications. Insufficient mechanical stiffness, likely the results of insufficient collagen content in the extracellular matrix (ECM), lack of physiological collagen architecture (i.e. vertical fibers in the deep zone and horizontal fibers in the superficial zone) and inhomogeneous distribution of ECM are among the main different characteristics of current TE cartilage, compared to those of native cartilage. In addition, the production process of TE cartilage requires cells to synthesis ECM within a few weeks. This is quite different compared to the physiological process of native cartilage formation, which is on the order of months/years. It is unknown how such characteristics of TE cartilage may influence its mechanical performance. For a better fundamental understanding of the role of these aspects, the objective of this thesis was to evaluate (1) the influence of cellular-and tissue-scale inhomogeneities in ECM distribution on the mechanical environment of chondrocytes in agarose TE constructs during and at the end of culture, (2) the relative importance of mechanical stiffness, collagen content and collagen network architecture for the post-implantation mechanical performance of TE cartilage, (3) current and newly proposed mechanical stimulations to stimulate formation of a physiological collagen architecture in TE cartilage and (4) the effects of temporal ECM deposition on the mechanical behavior of TE cartilage. A finite element approach with a composition-based non-linear fiber-reinforced poroviscoelastic swelling material model was used. It was shown that inhomogeneities in ECM distributions reduce overall construct stiffness and may significantly alter the tissue-level mechanical environment in the construct as well as the micromechanical environment of chondrocytes at the sites of inhomogeneties, in both free-swelling and under mechanical loading culture conditions. To assess post-implantation mechanical conditions of TE constructs under physiological loading, an axisymmetirc model of a medial tibia plateau was used in which a cylindrical part of the mesh at the central region of the model represented a TE implant. Results indicated that adverse implant composition and ultrastructure would lead to post-implantation excessive mechanical loads, with collagen network architecture being the most critical variable. To explore which mechanical stimulations would be likely to trigger a physiological collagen architecture, strain fields as a result of unconfined compression, and a novel loading regime of sliding indentation with an without lateral tissue compression were evaluated. Results suggested that sliding indentation is likely to stimulate formation of an appropriate superficial zone with parallel fibers. Adding lateral compression may stimulate formation of a deep zone with perpendicularly aligned fibers. Finally, a numerical framework was developed to evaluate the effects of temporal ECM deposition on the mechanical behavior of TE cartilage. The effects of differences in the rate of proteoglycans and collagen synthesis during cartilage TE were quantitatively evaluated. The range of predicted construct stiffness was compared to those reported in the literature. It was shown that alterations in the synthesis rates of proteoglycans and collagen would significantly change the mechanical behavior of TE cartilage. This indicated that even with similar final ECM contents, constructs would have different mechanical properties depending on the history of development. The insights provided in this dissertation should be considered in future cartilage tissue-engineering studies, as they may indicate directions that would result in the development of mechanically superior TE cartilage with enhanced long-term survival. Furthermore, in the present study we showed how numerical simulations can be used to assist in evaluating and designing loading protocols for tissue engineering and provide useful insights for experimental studies to discriminate promising protocols from those with poor potential, which is a step forward towards successful tissue-engineering of cartilage