Orthopaedic surgeons are faced every day with the question: “Will this fracture heal?” As clinicians, what criteria do we use to determine fracture-healing? Often this is a clinical gestalt involving symptoms such as pain at the fracture site, physical examination findings such as absence of tenderness to palpation at the fracture site, and a careful analysis of radiographs. Clinicians often use the presence of bridging callus as a radiographic sign of fracture-healing. However, there is ongoing controversy about how much callus and how many cortices with bridging callus, as well as the visibility of the fracture line, there should be to be interpreted as signs of fracture-healing. Currently, there is no one “gold standard” or commonly agreed to criterion to determine or even measure fracture-healing. A variety of scoring systems have been developed over the years, but none have become common parts of the daily life of clinicians treating fractures. Devices to measure fracture-healing by determining fracture stiffness have been developed but have not been adopted widely for clinical use. Most clinicians still rely on clinical examination and radiographs to determine fracture-healing.

Lack et al. performed a retrospective review of 176 tibial shaft fractures, open and closed, treated with intramedullary nailing. They proposed that the amount of time it takes to achieve cortical bridging predicts final healing, and they sought to determine the reliability and the accuracy of the time to bridging of one, two, or three or more cortices as predictors of fracture-healing. In this series, a final diagnosis of union or nonunion required up to fifteen months of observation. The nonunion rate was 7% overall, 3% for the closed fractures, and 11% for the open fractures. These results are consistent with previous literature. Receiver operating characteristic (ROC) curve analysis demonstrated that all of the cortical bridging criteria successfully discriminated between eventual union and nonunion. However, the predictive accuracy of the criteria differed in the early postoperative period, such that the more stringent criteria (bicortical and tricortical bridging) required a longer postoperative interval to achieve accuracy.