(470f) On the Numerical Validity of Commonly-Employed Scaffold Mimicks for Shear Stress and Flow Field Calculations in Tissue Engineering Models

Tissue engineering is an interdisciplinary field involving the replacement of damaged tissues or organs with artificial tissue substitutes. In bone tissue engineering, stem cells are typically seeded within 3-D scaffolds, made of synthetic biodegradable polymers, and followed by dynamic culture in perfusion bioreactors to reproduce tissue. Stresses that are generated by the flow of the culture media play an extremely important role in stimulating cell proliferation and differentiation. Yet, they have to be in the physiologically relevant range to promote stimulation; while applying excessive shear can result in cell death and detachment.

Unfortunately, the complex internal structure of porous scaffolds makes estimation of required stimulatory shear stresses via experimental and analytical techniques difficult. Hence, computational fluid dynamics (CFD) models, based on either idealized pore geometries or on actual scaffold images, are utilized. Such studies have become common: (Porter, Zauel et al. 2005, Lacroix, Chateau et al. 2006, Cioffi, Küffer et al. 2008, Maes, Van Ransbeeck et al. 2009, Voronov, VanGordon et al. 2010, Maes, Claessens et al. 2012, Guyot, Luyten et al. 2015, Hendrikson, Deegan et al. 2017). Yet, due to the computationally-expensive nature of the problem most approximate the scaffold geometry using smaller representative volumes (RV) cut from whole scaffolds. However, these simpletons differ from the actual scaffolds in shape, size and boundary conditions. Although this approach saves computation time and bypasses the inherent complexity of simulating large domains, the shear stress estimates in scaffold areas not included in the RV domain remain unaccounted for. Moreover, the flow field established within the scaffolds may be significantly different due to the inconsistent use of boundary conditions implemented by the two approaches: periodic BCs for the RV vs scaffold surrounded by a pipe for the whole scaffold representation. And since there is no established guide or reference for estimating the error associated with the RV approximation, this study investigates the numerical differences between the two approaches and the validity of each.

To achieve this, an in-house Lattice-Boltzmann Method code is used to simulate fluid flow through 3D reconstructions of µCT scaffold images for both RV and whole scaffold samples. Shear stress estimates obtained in both cases are compared for poly (L-lactic acid), PLLA porous foam and nonwoven fiber tissue scaffolds with varying pore size and porosities (10 total). Results of the study show that mean wall shear stress is overestimated by the RV approximation compared to the whole scaffold simulations. It was also observed that the approximation does not always adequately capture scaffold irregularities thereby contradicting an underlying assumption of RV simulations. Finally, it was found that the two approaches generated significantly different flow fields, meaning that although the mean stress values may be of similar magnitudes, the localized stresses experienced by the cells in the scaffolds may be completely different. This is particularly important since the commonly made assumption of uniform cell coverage of scaffold surface is rarely true. Therefore, modeling the full scaffold will be especially important for future tissue engineering models which take cells and tissue growth into account.

REFERENCES

Guyot, Y., F. P. Luyten, J. Schrooten, I. Papantoniou and L. Geris (2015). "A three-dimensional computational fluid dynamics model of shear stress distribution during neotissue growth in a perfusion bioreactor." Biotechnology and Bioengineering 112(12): 2591-2600.

Hendrikson, W. J., A. J. Deegan, Y. Yang, C. A. van Blitterswijk, N. Verdonschot, L. Moroni and J. Rouwkema (2017). "Influence of Additive Manufactured Scaffold Architecture on the Distribution of Surface Strains and Fluid Flow Shear Stresses and Expected Osteochondral Cell Differentiation." Frontiers in Bioengineering and Biotechnology 5(6).

Lacroix, D., A. Chateau, M.-P. Ginebra and J. A. Planell (2006). "Micro-finite element models of bone tissue-engineering scaffolds." Biomaterials 27(30): 5326-5334.

Maes, F., T. Claessens, M. Moesen, H. Van Oosterwyck, P. Van Ransbeeck and P. Verdonck (2012). "Computational models for wall shear stress estimation in scaffolds: A comparative study of two complete geometries." Journal of Biomechanics 45(9): 1586-1592.

Maes, F., P. Van Ransbeeck, H. Van Oosterwyck and P. Verdonck (2009). "Modeling fluid flow through irregular scaffolds for perfusion bioreactors." Biotechnology and Bioengineering 103(3): 621-630.

Porter, B., R. Zauel, H. Stockman, R. Guldberg and D. Fyhrie (2005). "3-D computational modeling of media flow through scaffolds in a perfusion bioreactor." Journal of Biomechanics 38(3): 543-549.

Voronov, R., S. VanGordon, V. I. Sikavitsas and D. V. Papavassiliou (2010). "Computational modeling of flow-induced shear stresses within 3D salt-leached porous scaffolds imaged via micro-CT." Journal of Biomechanics 43(7): 1279-1286.