RESUMO
OBJECTIVES: To determine the radial nerve palsy (RNP) rate and predictors of injury after humeral nonunion repair in a large multicenter sample. DESIGN: Consecutive retrospective cohort review. SETTING: Eighteen academic orthopedic trauma centers. PATIENTS/PARTICIPANTS: Three hundred seventy-nine adult patients who underwent humeral shaft nonunion repair. Exclusion criteria were pathologic fracture and complete motor RNP before nonunion surgery. INTERVENTION: Humeral shaft nonunion repair and assessment of postoperative radial nerve function. MAIN OUTCOME: Measurements: Demographics, nonunion characteristics, preoperative and postoperative radial nerve function and recovery. RESULTS: Twenty-six (6.9%) of 379 patients (151 M, 228 F, ages 18-93 years) had worse radial nerve function after nonunion repair. This did not differ by surgical approach. Only location in the middle third of the humerus correlated with RNP (P = 0.02). A total of 15.8% of patients with iatrogenic nerve injury followed for a minimum of 12 months did not resolve. For those who recovered, resolution averaged 5.4 months. On average, partial/complete palsies resolved at 2.6 and 6.5 months, respectively. Sixty-one percent (20/33) of patients who presented with nerve injury before their nonunion surgery resolved. CONCLUSION: In a large series of patients treated operatively for humeral shaft nonunion, the RNP rate was 6.9%. Among patients with postoperative iatrogenic RNP, the rate of persistent RNP was 15.8%. This finding is more generalizable than previous reports. Midshaft fractures were associated with palsy, while surgical approach was not. LEVEL OF EVIDENCE: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.
