Nerve injury in revision total elbow arthroplasty: a systematic review and meta-analysis.

Abstract

BACKGROUND: Revision total elbow arthroplasty (TEA) is technically demanding and carries a substantial risk of postoperative neurological complications because of scarring, altered anatomy, implant removal, and repeated humeral and ulnar exposure. The incidence, nerve distribution, and recovery profile of nerve injury after revision TEA remain incompletely defined. This study aimed to systematically review the literature to define the incidence, recovery profile, and risk factors for nerve injury after revision TEA. METHODS: A systematic review of the literature was performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidance. Thirteen retrospective case series were included, comprising 282 revision TEAs in 271 patients. Random-effects meta-analysis of proportions was undertaken where the data permitted. The primary outcome was postoperative nerve injury following revision TEA. Secondary outcomes included nerve type, recovery, secondary nerve-related procedures, infection, triceps insufficiency, metallosis, periprosthetic fracture, and re-revision. RESULTS: The pooled incidence of postoperative nerve injury was 22.3% (95% confidence interval [CI] 16.3 to 29.6; I2 = 34.6%). The crude incidence was 60 of 282 revisions (21.3%, 95% CI 16.6 to 26.5). The ulnar nerve was involved in 66.7% of all nerve injuries, the radial nerve in 31.7%, and the median nerve in 1.7%. No significant difference in pooled nerve-injury incidence was identified between studies published before 2010 and those published from 2010 onwards (22.9% vs. 21.4%, P = .837). Recovery reporting was heterogeneous; among injuries with numerically extractable outcomes, 85.4% improved partially or completely (95% CI 72.2 to 93.9). Secondary nerve-related procedures were reported in 7 studies. Pooled complication estimates were 10.2% for infection, 13.6% for triceps insufficiency, 25.5% for metallosis, 15.9% for periprosthetic fracture, and 14.0% for re-revision. CONCLUSIONS: Postoperative nerve injury is a common and clinically important complication of revision TEA, affecting approximately 1 in 5 cases. The ulnar nerve is most frequently involved, although radial nerve injury accounts for a substantial proportion of cases. Many neuropathies improve during follow-up, but persistent deficits and the need for secondary nerve-related procedures are not uncommon. Future studies should adopt standardized neurological definitions and reporting to improve comparability and guide preventive surgical strategies.