RESUMEN
OBJECTIVES: To determine whether the initial radiographic displacement of humeral shaft fractures is associated with failure of nonoperative management. DESIGN: Retrospective cohort study. SETTING: Urban level 1 trauma center. PATIENTS/PARTICIPANTS: 106 patients with humeral shaft fractures (OTA/AO 12) initially managed nonoperatively. INTERVENTION: Functional bracing. MAIN OUTCOME MEASUREMENTS: Failure of nonoperative management, defined as conversion to surgery, malunion, and delayed union/nonunion. RESULTS: Nonoperative management failed in 33 (31%) of 106 included patients with 27 patients (25%) requiring surgery. On multivariate analysis, female sex [odds ratio (OR): 3.50, 95% confidence interval (CI): 1.09 to 11.21], American Society of Anesthesiologist classification >1 (OR: 7.16, CI: 1.95 to 26.29), initial fracture medial/lateral (ML) translation (OR: 1.09, CI: 1.01 to 1.17, per unit change), and initial fracture anterior-posterior (AP) angulation (OR: 1.09, CI: 1.02 to 1.15, per unit change) were independently associated with failure of nonoperative management. Initial fracture displacement values that maximized the sensitivity (SN) and specificity (SP) for failure included an AP angulation >11 degrees (SN 75%, SP 64%) and ML translation >12 mm (SN 55%, SP 75%). The failure rate in patients with none, 1, or both of these fracture parameters was 3.1% (1/32), 35.6% (20/56), and 66.6% (12/18), respectively. CONCLUSIONS: Nearly one-third of patients experienced failure of initial nonoperative management. Failure was found to be associated with greater initial fracture AP angulation and ML translation. Fracture displacement cut-off values were established that may be used by surgeons to counsel patients with these injuries. LEVEL OF EVIDENCE: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.