Michael Roland Stefan Diebels Kerstin Wickert Tim Pohlemann Bergita Ganse ^*

• Saarland University, Saarbrücken, Saarland, Germany

Bone healing can be improved by axial micromovement, as has been shown in animals and human patients with external fixators. In the development of smart fracture plates, the ideal amount of stroke for different fracture types in the differentand healing stages is currently unknown. It was hypothesized that the resulting strain in the fracture gap of a simple tibial shaft fracture does not vary with the amount of axial stroke in the plate, the fracture gap size, and the fracture angle. With finite element simulations based on body donation CT data, the second invariant of the deviatoric strain tensor (J2), strain energy density, hydrostatic strain, octahedral shear strain, and percentage of the fracture gap in the 'perfect healing window' were computed for different gap sizes (1 to 3 mm), angles (5° to 60°), and plate stroke levels (0.05 to 0.60 mm) in three healing stages. Multiple linear regression analyses were performed. Results showed that an active fracture plate should deliver an axial stroke in the range of 0.10 to 0.45 mm. Different optimal stroke values were found for each healing phase, namely 0.10 to 0.25 mm for the first, 0.10 mm for the second, and 0.35 to 0.45 mm for the third healing phase, depending on the fracture gap size and less on the fracture angle. J2, hydrostatic strain, octahedral shear strain and the strain energy density correlated with the fracture gap size and angle (all P < 0.001). The influence of the fracture gap size and angle on the variability (adjusted R2) in several outcome measures in the fracture gap was shown to vary throughout healing. The contribution to the variability of the percentage of the fracture gap in the perfect healing window was greatest during the second healing phase. For J2, strain energy density, hydrostatic strain, and octahedral shear strain, the fracture gap size showed the greatest contribution in the third fracture healing phase, while the influence of fracture angle was independent of the healing phase. The present findings are relevant for implant development and to design clinical studies that aim to accelerate fracture healing using axial micromovement.