Iatrogenic Nerve Palsy Occurs With Anterior and Posterior Approaches for Humeral Shaft Fixation

Objective: To determine if surgical approach impacts the rate of nerve palsy after plate fixation of humerus shaft fractures and whether or not iatrogenic nerve palsy recovers in similar ways to preoperative palsy. Design: Retrospective. Setting: Two trauma centers. Patients: Patients 18+ years of age with nonpathologic, extra-articular humerus shaft fractures (OTA/AO 12A/B/C and 13A2-3) treated with plate fixation. Intervention: Plate fixation of humerus shaft fractures, from 2008 to 2016. Main Outcome Measurement: Rate of iatrogenic nerve palsy by a surgical approach and injury characteristics. Results: Two hundred sixty-one humeral shaft fractures were included. The rate of preoperative palsy was 19%. Radial nerve palsy (RNP) was present in 18%. Iatrogenic RNP occurred in 12.2% and iatrogenic ulnar palsy in 1.2%. Iatrogenic palsy occurred in 15.6% of middle and 15% of distal fractures, with fracture location significantly different in those developing RNP (P = 0.009). Iatrogenic RNP occurred in 7.1% of anterolateral, 11.7% of posterior triceps-splitting, and 17.9% of posterior triceps-sparing approaches (P = 0.11). Follow-up data were available for 139 patients at an average of 12 months. Preoperative RNP resolved less often than iatrogenic RNP, in 74% versus 95% (P = 0.06). Time to resolution was longer for preoperative RNP, at 5.5 versus 4.1 months (P = 0.91). Twenty-two percent with preoperative RNP underwent tendon transfer or wrist fusion, versus 0% after iatrogenic RNP (P = 0.006). Conclusion: Iatrogenic RNP is not uncommon with humeral fracture fixation and occurs at similar rates in anterior and posterior approaches and with midshaft and distal fractures. Iatrogenic RNP had a high rate of recovery. Preoperative RNP more often requires surgery for unresolved palsy.