gms | German Medical Science

Objectives: Bone fracture healing is known to be sensitive to the degree of fixation stability. However, it is unclear which biological processes during healing are mechano-sensitive and if mechanical stimulation is required during all stages of repair. The science of how mechanical and physical stimuli regulate biological processes is termed mechanobiology. Advancements in our understanding of mechanobiology of secondary bone healing have been hampered by our inability to control and measure the load history in an experimental fracture. In this study, a novel bone fracture model, which isolates an experimental fracture from functional loading, is applied to investigate if stimulation limited to the early proliferative phase is sufficient for timely bone healing.

Methods: The fracture model consists of a 3mm experimental fracture (proximal) and a 30mm critical size defect in an ovine tibia with dual external fixation. One external fixator stabilises the proximal and far distal fragment. A second custom active fixator bridges from the middle mobile segment to the proximal fragment and controls the magnitude of axial displacement in the experimental fracture. Furthermore, the active fixator was applied to monitor fracture stiffness over the course of healing. An in vivo study was conducted using twelve skeletally mature sheep (3-4 years) separated into two groups (n=6); control and stimulatory. The loading protocol for the stimulated group used axial compressive movements (1 mm, 500 cycles/day @1Hz, from day 5 to day 21) whilst the control group was subjected to only weekly axial movements to measure stiffness (0.1 mm, 100 cycles/day @1Hz). This latter protocol (stiffness measurements) was continued in both groups until the end of the study (week 4-9). Experimental defects were evaluated post-mortem with microCT and mechanical testing in torsion.

Results and Conclusion: The unstimulated control group was characterised in micro-CT analysis by minimal external callus formation and a lack of bone bridging at the experimental fracture. In contrast, the stimulated group exhibited advanced healing with solid bone formation across the defect. Biomechanically, the healed defects in the stimulatory group (19.8 Nm) showed significantly (p<0.05) higher torsional forces to failure than the control (7.8 Nm).

This novel experimental model permits the application of a well-defined load history to an experimental bone fracture. The poor healing observed in the control group is consistent with under-stimulation and the isolation of the impact of functional loading. This model permits to experiment with different weight bearing protocols and fixation rigidities without the interference of functional loading.