Development of a novel murine delayed secondary fracture healing in vivo model using periosteal cauterization.

INTRODUCTION: Delayed union and nonunion development remain a major clinical problematic complication during fracture healing, with partially unclear pathophysiology. Incidences range from 5 to 40% in high-risk patients, such as patients with periosteal damage. The periosteum is essential in adequate fracture healing, especially during soft callus formation. In this study, we hypothesize that inducing periosteal damage in a murine bone healing model will result in a novel delayed union model. MATERIALS AND METHODS: A mid-shaft femoral non-critically sized osteotomy was created in skeletally mature C57BL/6 mice and stabilized with a bridging plate. In half of the mice, a thin band of periosteum adjacent to the osteotomy was cauterized. Over 42 days of healing, radiographic, biomechanical, micro-computed tomography and histological analysis was performed to assess the degree of fracture healing. RESULTS: Analysis showed complete secondary fracture healing in the control group without periosteal injury. Whereas the periosteal injury group demonstrated less than half as much maximum callus volume (p < 0.05) and bridging, recovery of stiffness and temporal expression of callus growth and remodelling was delayed by 7-15 days. CONCLUSION: This paper introduces a novel mouse model of delayed union without a critically sized defect and with standardized biomechanical conditions, which enables further investigation into the molecular biological, biomechanical, and biochemical processes involved in (delayed) fracture healing and nonunion development. This model provides a continuum between normal fracture healing and the development of nonunions.

Fixation compliance in a mouse osteotomy model induces two different processes of bone healing but does not lead to delayed union.

Delayed unions are a problematic complication of fracture healing whose pathophysiology is not well understood. Advanced molecular biology methods available with mice would be advantageous for investigation. In humans, decreased fixation rigidity and poor reduction are generally associated with delayed unions. In this study, these two factors were combined to observe their effect on bone healing in mice. Two plates with locking screws, one with 14 the bending stiffness of the other, were used to stabilize a 0.45mm gap osteotomy. muCT, radiographs, 4pt-bending tests and histological analysis demonstrated that the different plate types led to two different healing pathways. The less flexible bridging plate induced only intramembranous ossification whereas the more flexible bridging plate induced a mixture of endochondral and intramembranous ossification. However, the different plates led to a delay in healing of only 3-5 days in the period between 14 and 21 post-operative days. In mice, considerable fixation flexibility is necessary to induce secondary bone healing similar to that which occurs in humans, but this was not sufficient to induce a substantial delay in bone healing as would be expected in humans.

Remodeling of fracture callus in mice is consistent with mechanical loading and bone remodeling theory.

During the remodeling phase of fracture healing in mice, the callus gradually transforms into a double cortex, which thereafter merges into one cortex. In large animals, a double cortex normally does not form. We investigated whether these patterns of remodeling of the fracture callus in mice can be explained by mechanical loading. Morphologies of fractures after 21, 28, and 42 days of healing were determined from an in vivo mid-diaphyseal femoral osteotomy healing experiment in mice. Bone density distributions from microCT at 21 days were converted into adaptive finite element models. To assess the effect of loading mode on bone remodeling, a well-established remodeling algorithm was used to examine the effect of axial force or bending moment on bone structure. All simulations predicted that under axial loading, the callus remodeled to form a single cortex. When a bending moment was applied, dual concentric cortices developed in all simulations, corresponding well to the progression of remodeling observed experimentally and resulting in quantitatively comparable callus areas of woven and lamellar bone. Effects of biological differences between species or other reasons cannot be excluded, but this study demonstrates how a difference in loading mode could explain the differences between the remodeling phase in small rodents and larger mammals.