Nonsteroidal Anti-Inflammatory Drug-Induced Fracture Nonunion: An Inhibition of Angiogenesis?

BACKGROUND: Approximately 5% to 10% of fractures may result in delayed union or nonunion. The results of research done over the past three decades have shown that the use of nonsteroidal anti-inflammatory drugs (NSAIDs) has an inhibitory effect on fracture repair, but the exact mechanism of action remains to be elucidated. Cancer research has identified that NSAIDs impede cell proliferation by inhibiting angiogenesis. It is proposed that a similar mechanism occurs in the induction of NSAID-induced nonunions. This hypothesis was investigated in a randomized placebo-controlled trial of the NSAID rofecoxib with use of a murine femoral fracture model. METHODS: Two hundred and forty mice were randomized to receive either the nonsteroidal anti-inflammatory drug rofecoxib (5 mg/kg orally) in a 0.5% methylcellulose solution (the NSAID group) or the 0.5% methylcellulose solution only (the control group). Two hundred and thirty-five of the 240 mice underwent surgery to induce an open transverse middiaphyseal femoral fracture, which was then treated with use of a custom-made external fixator. Five additional animals underwent sham surgery with no fracture induced. Outcomes measures included radiographic assessment, histologic analysis, biomechanical testing, and use of laser Doppler flowmetry to assess blood flow across the fracture gap. RESULTS: Radiography revealed similar healing patterns in both groups; however, at the later stages (day 32), the NSAID group had poorer healing. Histological analysis demonstrated that the control animals healed quicker (at days 24 and 32) and had more callus and less fibrous tissue (at days 8 and 32) than the NSAID animals did. Biomechanical testing found that the control animals were stronger at day 32. Both groups exhibited a similar pattern of blood flow; however, the NSAID group exhibited a lower median flow from day 4 onward (significant at days 4, 16, and 24). Positive correlations were demonstrated between both histological and radiographic assessments of healing and increasing blood flow. NSAID-treated animals exhibited lower blood flow and poorer healing by all parameters. Regression analysis, however, demonstrated that the negative effect of NSAIDs on fracture repair is independent of its inhibitory action on blood flow. CONCLUSIONS: Following the development of a novel method of analyzing functional vascularity across a fracture gap, we have demonstrated that the cyclooxygenase-2 (COX-2) inhibitor rofecoxib has a significant negative effect on blood flow across the fracture gap as well as an inhibiting effect on fracture repair. CLINICAL RELEVANCE: COX-2 inhibitors are marketed as having low side-effect profiles. We propose that these drugs should be used with caution in all patients following osseous trauma and, in particular, after injuries that may already predispose a fracture to a delayed union due to osseous, vascular, or patient-related factors.