PURPOSE: To determine the capability of fellowship trained Orthopaedic Trauma surgeons to predict union or non-union of femoral and tibial shaft fractures.

METHODS: A series of 50 patients with femur or tibia shaft fractures were evaluated. Patients were prospectively followed at 2,6,12, and 18 weeks after surgical intervention. At each interval surgeons evaluated factors related to fracture healing on AP and lateral radiographs and predicted the probability of union on a visual analog scale. Union was defined as radiographic evidence of healing three of four cortices, no tenderness with palpation of the fracture site, and full weight bearing without the use of assistive devices.

RESULTS: Eight patients missed initial visits or were lost to follow-up, making for a total of 42 patients that were included in the results. Average patient age was 31 years. Eighty-one percent of the patients went onto union (N=34) and 19% went onto nonunion (N=8). Early clinical prediction for nonunion at 2 weeks had a sensitivity of 50%, a specificity of 91%, a positive predictive value (PPV) of 57%, and a negative predictive value (NPV) of 89%. At 6 weeks, there was a sensitivity of 75%, a specificity of 100%, a PPV of 100%, and a NPV of 94%. One patient treated with intramedullary nailing was 15 years old and despite minimal callous formation the physician incorrectly predicted future union given the young age. The other patient had a severely comminuted femur fracture and required a quad cane to ambulate and should perhaps have been predicted to go onto nonunion. At 12 and 18 weeks, sensitivity, specificity, PPV, and NPV were both 100%.

CONCLUSIONS: Fellowship trained orthopaedic trauma surgeons at 6-week follow-up can predict union with a sensitivity of 75% and specificity of 100% and a PPV of 100%. Early clinical prediction at 6 weeks can be used to provide the patient with a secondary intervention such as a bone graft or bone stimulator and avoid months of delay.