Radiographic Scoring of Humeral Shaft Fractures Helps Identify Patients at Risk of Surgery for Delayed Union

Objectives: Assessing external validity and clinical relevance of modified radiographic union score (mRUS) to predict delayed union in closed humeral shaft fractures initially treated with conservative management. Design: Retrospective cohort. Setting: Single urban academic level 1 trauma center. Patients: Patients undergoing initial nonoperative treatment of a humeral shaft fracture with a minimum of 3 months follow-up and at least one set of follow-up orthogonal x-rays within 12-weeks of injury. Main Outcome Measurements: Interobserver and intraobserver reliability of the (mRUS) system for humeral shaft fractures, and establishing an mRUS threshold at 6 and 12 weeks postinjury to predict surgery for delayed union. Results: mRUS demonstrated substantial interobserver agreement on all assessments. Intraobserver agreement was nearly perfect for all reviewers on repeat assessment. mRUS of <= 7 at 6 +/- 1 weeks follow-up was associated with surgery for delayed union with an odds ratio of 4.88 (95% CI, 2.52-9.44, P < 0.01), sensitivity of 0.286, and specificity of 0.924. At 12 +/- 1 weeks follow-up, the same threshold demonstrated a stronger association with an odds ratio of 14.7 (95% CI, 4.9-44.1, P < 0.01), sensitivity of 0.225, and specificity of 0.981. Conclusions: The mRUS for humeral shaft fractures is reliable and reproducible providing an objective way to track subtle changes in radiographs over time. An mRUS of <= 7 at 6 or 12 weeks postinjury is highly specific for delayed union. This can be helpful when counseling patients about the risk of nonunion and potential early surgical intervention.