Delayed fracture healing is the improper healing of fractures within a reasonable amount of time and is estimated to impact a sixth of all fractures that occur annually in the United States1. While blood- and imaging-based bone turnover biomarkers have been thoroughly investigated throughout the healing process of bone, there is still a lack of understanding on how well these biomarkers can predict union in individual patients. Although conventional radiography is the most common clinical practice for assessing bone healing progression, this imaging technique—as well as the other imaging methods used—fails to discern the in vivo mechanical environment of bone, and therefore the likeliness of union or nonunion. There is a need to identify mechanical biomarkers that could better differentiate between patients who undergo typical healing progression versus delayed fracture healing. In order to identify these mechanical biomarkers, a 3D in vitro cell culture platform that recapitulates the micromechanical environment must be developed and tested. Success of this in vitro platform relies on the generation of rigorous testing protocols for assessing stiffness and fluid flow within this organoid system. This study aims to develop an experimental and computational pipeline for mechanically characterizing 3D printed (3DP) scaffolds—Voronoi, IsoTruss, and Truncated Octahedron (TO) geometries—that will be the foundation for future studies to explore patient-specific mechanical biomarkers in these bone tissue engineered scaffolds.

A dynamic mechanical analysis (DMA) strain sweep was performed on the scaffolds (n=6 for 4- and 7-day 3T3 fibroblast seeded Voronoi and TO scaffolds, n=4 for 4- and 7-day seeded IsoTruss scaffolds, n=3 for 4- and 7-day soaked controls for each geometry) to measure storage modulus, loss modulus, and the damping coefficient. The Voronoi geometry increased significantly in storage modulus when seeded for seven days compared to four days (p=0.0293). There was also an overall significant decrease in stiffness when the scaffolds were seeded versus non-seeded (p<<0.001). Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) was performed to produce fluid flow experimental validation data, and this provided insights on the micromechanical environment of the IsoTruss scaffold that were consistent with the computational fluid dynamics (CFD) simulation model. The CFD model was used to calculate wall shear stresses (WSS) for various inlet velocities (0.05, 0.10, 0.15, 0.20, and 0.25 mm/s), with 0.15 mm/s producing WSS best within the range of extracellular matrix formation.

DMA, DCE-MRI, and CFD all confirmed mechanical characteristics of the IsoTruss geometry that were unique to its specific micromechanical architecture. Out of all scaffolds tested, the IsoTruss geometry achieved the maximum (3.47 MPa) and minimum (0.0631 MPa) storage modulus. The computational analysis pipeline revealed that the patterns observed in the DMA experiments could be caused by buckling due to the fourteen-strut intersections and printing infidelity issue related to the IsoTruss geometry. The protocol developed herein for the experimental and computational analyses done on the scaffolds in this thesis will be used in the future on bone organoids to study individualized fracture healing.