Fluid Flow Within Porous Scaffolds for Bone Tissue Engineering: a Computational and Experimental

Fluid FLOW WITHIN BONE TISSUE ENGINEERING SCAFFOLDS: A COMPUTATIONAL AND EXPERIMENTAL STUDY

Silvia Truscello(1), Sebastian de Boodt(2), Toon Leroy(2), Jan Schrooten(1), Daniel Berckmans(2), Hans Van Oosterwyck(3)

(1)Katholieke Universiteit Leuven, Department of Metallurgy and Materials Engineering, Belgium; (2)Katholieke Universiteit Leuven,Division M3-BIORES: Measure, Model & Manage Bioresponses, Belgium; (3)Katholieke Universiteit Leuven, Division of Biomechanics and Engineering Design, Belgium

Introduction

As part of a bone tissue engineering therapy cell-seeded scaffolds can be cultured in perfusion bioreactors prior to in vivo implantation in the bone defect.The flow mediated mechanical stimuli inside the scaffold appear to be a factor that affects the in vitro proliferation and osteogenic differentiation [Cartmell 2003]. Quantification of the fluid dynamic microenvironment inside scaffolds is needed in order to assess this effect. The goal of this study was to develop a combined computational and experimental approach to characterise the fluid flow within porous scaffolds.

Methods

Experimental setup: laboratory experiments were carried out in a perfusion bioreactor designed for live imaging through the transparent cover. Distilled water was perfused at a flow rate of 0.018ml/min through a titanium scaffold (6 mm width, 20 mm length, 0.5 mm height) with a regular structure consisting of 0.5 mm by 0.5 mm struts and 0.3 mm wide channels. The water was seeded with red polystyrene 10 µm diameter particles. Images were captured at 10 fps with a CCCD (Cooled Charge Coupled Device) camera mounted on a stereomicroscope. A sequence of 20 images (ROI 600x600 pixels) was pre-processed to maximize the contrast between the moving particles and the background. For every two subsequent images a vector field of 24 by 24 grid points was calculated with an optical flow algorithm using the Lucas Kanade constraint [Bruhn 2005]. The resulting 19 vector fields were averaged excluding the 10% extreme values.

CFD model: a computational fluid dynamic (CFD) model of the medium flow was developed to predict the distribution of fluid velocity and wall shear stress. The ACIS-based solid modeller Gambit was used to build the 3D model and the mesh. Due to symmetry only half of the fluid domain was modelled. The mesh presented 1.845.916 tetrahedral elements and the element edge length was 25 µm .The finitevolumes code Fluent was used to set up and solve the problem.

Results

Both experimental and CFD calculations show similar flow patterns, with a peak velocity of 0.4 mm/s (fig.1). The average and the median of the shear stress at the scaffold surfacewere calculated by the CFD model to be 3.4 and 2.3 mPa respectively.

Figure 1: Perfusion bioreactor with part of the scaffold design in top view (top) – Fluid velocity vector plots of CFD calculations (bottom left) and measurements (bottom, right).

Conclusion

The coupling of experimental and numerical analysis provides a complete flow characterisation and represents the basis to quantify the relation between the hydrodynamic environment and cell growth and differentiationwithin bone scaffolds.

Reference

Bruhn, A.et al., International Journal of ComputerVision 61:211-231, 2005

Cartmell S.H.et al., Tissue Engineering, 9(6), 1197-203, 2003