Corticosteroid Injection Before and After Rotator Cuff Repair: A Systematic Review and Meta-analysis of Injection Number, Timing, and Structural Failure

Registration type: Retrospective registration. Searches, screening, data extraction, risk-of-bias assessment and analysis were completed before registration (October 2026). The review was not registered in PROSPERO because registration was sought after data extraction, which PROSPERO does not accept. Review question: In adults undergoing primary rotator cuff repair, is corticosteroid injection before (within 12 months) or after repair associated with structural failure (imaging-confirmed retear, revision or reoperation), surgical site infection, or patient-reported outcomes, and does the association depend on the number and timing of preoperative injections? Condition: Rotator cuff tear treated by primary repair. Intervention/exposure: Corticosteroid injection into the shoulder (subacromial or glenohumeral) before or after repair; number of injections (1, 2, 3 or more) and interval before surgery (0-3, 3-6, 6-12 months). Comparator: No injection; or different number or timing of injection. Outcomes: Primary - structural failure (retear on MRI/MR arthrography/ultrasound, revision repair, reoperation). Secondary - surgical site infection; patient-reported outcome measures. Study designs: Randomized trials, cohort and case-control studies, comparative conference abstracts. Excluded: intraoperative, perineural or systemic steroid; no comparison group. Information sources: PubMed, Embase (incl. trial registry records), Web of Science Core Collection, Cochrane CENTRAL, inception to October 2026; supplementary PubMed search for all-shoulder-arthroscopy infection studies; reference lists, cited-by, Google Scholar. Selection and extraction: Two reviewers independently screened in Rayyan with blinding; reviewer 2 was a large language model agent (Claude, Anthropic) applying prespecified criteria, verified by the senior author; disagreements resolved by consensus. One reviewer extracted, a second checked. Risk of bias: RoB 2 (RCTs); ROBINS-I version 2 (non-randomized studies). Synthesis: Random-effects odds ratios (REML, Hartung-Knapp), Mantel-Haenszel and Peto for rare events; one study per overlapping claims database; subgroup and sensitivity analyses prespecified in the manuscript methods; absolute effects per 1,000; GRADE. Deviations from the initial plan: The review was initially framed to estimate a safe interval between preoperative injection and repair by timing. After the feasibility search and preliminary extraction, the absolute risk differences by timing of a single injection were small and inconsistent across studies, whereas the number of injections (2 or more) and the safety of postoperative injection for retear showed clinically meaningful effects. The question was therefore reframed to separate injection number from timing and to include postoperative injection as a co-primary question. All analyses in the final manuscript were specified before the final analysis was run; this reframing is reported in the manuscript. Review status: Completed; manuscript in preparation for submission to Journal of Shoulder and Elbow Surgery. Review team: Jong Pil Yoon (first author), Sung Jin Park (corresponding author); Department of Orthopedic Surgery, Kyungpook National University Hospital, Daegu, Republic of Korea. Funding: This research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number RS-2022-KH130593), and by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2026-25485638). Conflicts of interest: None. The authors declare no conflicts of interest related to this work.