Differentiating Outcomes and Complications Between Extraplexal Tendon Transfers and Arthrodesis for Shoulder...

Background: Glenohumeral arthrodesis (GHA) and extraplexal tendon transfers (TT) have been described as options for secondary shoulder stabilization and reanimation following adult traumatic brachial plexus injury (BPI) with delayed presentation or failure of primary nerve reinnervation. This study aimed to evaluate the outcomes and complication profiles of these two approaches to shoulder reanimation to better understand the indications, anticipated outcomes, and complication risks of each for traumatic brachial plexus injury. Methods: A systematic search of six databases (PubMed, EMBASE, SCOPUS, CINAHL, SPORTDiscus, Cochrane Library) was conducted in March 2025 following PRISMA guidelines. Studies reporting clinical outcomes in adults undergoing GHA or TT for traumatic BPI were included. Pooled mean range of motion and proportional complication and reoperation rates were calculated using random- and fixed-effects models, as appropriate. Results: A total of 22 studies involving 269 TT procedures and 194 GHA procedures were analyzed. Mean shoulder abduction was 81° (95% CI 54-108°) in the TT group and 51° (95% CI 37-65°) in the GHA group. Mean forward flexion was 88° (95% CI 51-124°) in the TT group and 56° (95% CI 44-68°) in the GHA group. The pooled complication rate was 4.8% (95% CI 2.6-8.6%) after TT and 26.4% (95% CI 18.5-36.1%) after GHA. The pooled reoperation rate was 3.2% (95% CI 1.5-6.6%) after TT and 17% (95% CI 10.8-25.7%) after GHA. Notably, TT cohorts generally had shorter follow-up durations, which may underrepresent late complications or reoperations. Conclusions: TT results in significantly lower complication and reoperation rates and demonstrates similar range-of-motion outcomes compared to GHA, suggesting that TT can be considered a first-line salvage option for motion preservation, while GHA remains an option for persistent instability, pain, or inability to achieve functional positioning of the hand in patients with traumatic BPIs. Additional comparative studies with higher levels of evidence are warranted to validate these findings.