Abstract
BACKGROUND: Nerve blocks are commonly used for analgesia following shoulder arthroscopy, but the optimal technique remains uncertain. This network meta-analysis (NMA) evaluates the effect of peripheral nerve blocks in adults undergoing shoulder arthroscopy. METHODS: We searched databases including PubMed, Embase, Web of Science, and the Cochrane Library from their inception to July 2025. Randomized controlled trials (RCTs) to evaluate the effect of peripheral nerve blocks after arthroscopic shoulder surgery were included. The primary outcome was the pain scores at 6 hours postoperatively. RESULTS: We included 25 RCTs involving 2039 patients and assessed ten techniques. The Visual Analog Scale (VAS) scores at 6 hours were lower with suprascapular and axillary nerve blocks (SANB), followed by coracoid approach brachial plexus (CPB) + suprascapular nerve block (SSB) and supraclavicular brachial plexus block (SCPB). The VAS scores at 12 hours were lower with erector spinae plane block (ESPB), followed by SANB and infraclavicular-suprascapular blocks (ICSB). The VAS scores at 24 hours were lower with SANB, followed by upper trunk block (UTB) and CPB + SSB. The morphine consumption within 24 hours was significantly reduced by UTB, followed by SSB and SCPB. For hemidiaphragmatic paralysis (HDP), ICSB had the lowest incidence (0%), followed by CPB+SSB (21.7%) and SCPB (29.5%). For hoarseness, ICSB and SSB had the lowest incidence (0%), followed by UTB (4.8%). For Horner's syndrome, ICSB and SSB had the lowest incidence (0%), followed by UTB (5.7%). For motor block, SSB had the lowest incidence (1.2%). For postoperative nausea and vomiting (PONV), ESPB had the lowest incidence (9.4%), followed by SSB (15.4%) and SANB (14.3%). LIMITATION: The included studies have differences in local anesthetic agents and postoperative analgesic regimen, which may affect the generalizability of the findings. Our NMA did not include continuous nerve block which have also been shown to provide significant analgesic benefits. There was a lack of data on long-term analgesia and functional outcomes. CONCLUSION: The SANB was more likely to improve pain scores within 24 hours and reduce the incidence of PONV, HDP, hoarseness, and Horner's syndrome. The SSB was proven to have the weakest analgesic effect.