Development and validation of animal models for fracture-related infection

Fracture-related infections (FRIs) are a challenging complication in orthopaedics for all people and systems involved. The incidence of these infections is considerably high, particularly in open fractures. FRI management typically involves prolonged antibiotics therapies, irrigation and debridement of the fracture site, and/or implant exchange. Unfortunately, this regimen is often ineffective at infection eradication resulting in poor patient outcomes and inefficient use of healthcare resources such that improved preventative and therapeutic interventions are needed. To effectively address these gaps, valid preclinical animal models for FRIs are needed; the purpose of this thesis was to review, develop, and validate animal models for this condition. First, we systematically reviewed the literature to analyze the translational rigor of the previously utilized preclinical FRI animal models. A variety of methods were utilized to establish an FRI among the 77 articles reviewed. Ultimately, few models incorporated clinically relevant interventions at clinically relevant time points while also validating key components of FRI pathogenesis such as bacterial biofilm on fracture implants. Considering this, we developed and validated preclinical canine models for early-onset FRI of the fibula bone and delayed-onset FRI of the ulna bone. In these models, a proximal fibular or distal ulnar 1 cm ostectomy was established, then internally fixated with plates and screws that were pre-incubated in a 1x105 MRSA suspension. After a period of infection development for 7 days in the early-onset fibular FRI model and 3 weeks in the delayed-onset ulnar FRI model, all animals underwent irrigation and debridement of the fracture site. Validating these models, clinical, bacteriologic, radiographic, and histologic outcomes were assessed. Clinically, all fracture sites displayed consistent evidence of local infection while the animals presented with minimal evidence of systemic infection. Quantitative microbial cultures of fracture-site bone and soft tissue indicated a clinically realistic positive culture rate for MRSA given the respective time points. Implant-associated osteolysis, implant loosening, and delayed union were observed radiographically. Lastly, bacterial biofilms were detected on all animals' fracture implants histologically. Comparatively, the early-onset fibular FRI model involves a bone with a greater soft tissue coverage, whereas the delayed-onset ulnar FRI model involves a bone with greater biomechanical load. Additionally, the early-onset model results a lesser degree of bone healing and osteomyelitis compared to the delayed-onset model. All considered, these animal models provide a translationally accurate approach to test and develop novel, onset-specific strategies aimed at improving outcomes for patients with FRIs.