Arthroscopic Versus Mini-open Rotator Cuff Repair: A Randomized Trial and Meta-analysis

Abstract

Background: Patients with complete rotator cuff tears who fail a course of nonoperative therapy can benefit from surgical repair. Purpose: This randomized trial compared mini-open (MO) versus all-arthroscopic (AA) rotator cuff repair. Study Design: Randomized controlled trial; Level of evidence, 1. Methods: Patients with rotator cuff tears were randomized to undergo MO or AA repair at 9 centers by 23 surgeons. The primary outcome (Western Ontario Rotator Cuff Index [WORC]) and secondary outcomes (American Shoulder and Elbow Surgeons [ASES] score, Shoulder Pain and Disability Index [SPADI] pain subscale, 12-Item Short Form Health Survey [SF-12], reported medication use, adverse events), as well as measurements of range of motion and strength, were collected at 1 month before surgery; at 2 and 6 weeks postoperatively; and at 3, 6, 12, 18, and 24 months postoperatively. A blinded radiologist evaluated rotator cuff integrity on magnetic resonance imaging (MRI) at baseline and 1 year. Intention-to-treat analysis of covariance with the preoperative WORC score, age, and tear size as covariates assessed continuous outcomes. Sex differences were assessed. A meta-analysis synthesized the primary outcome between MO and AA repair with previous trials. Results: From 954 patients screened, 411 were ineligible (276 because of recovery with physical therapy), 449 were screened at surgery (175 ineligible), and 274 completed follow-up (138 MO and 136 AA). The AA and MO groups were similar before surgery. WORC scores improved from 40 preoperatively to 89 (AA) and 93 (MO) at 2 years, for an adjusted mean difference of 3.4 (95% CI, –0.4 to 7.2). There were no statistically significant differences between the AA and MO groups at any time point. All secondary patient-reported outcomes were not significantly different between the MO and AA groups, except the 2-year SPADI pain score (8 vs 12, respectively; P = .02). A similar recovery in range of motion and strength occurred in both groups over time. MRI indicated minimal improvement in muscle relative to fat (AA: n = 3; MO: n = 2), with most worsening (AA: n = 25; MO: n = 24) or remaining unchanged (AA: n = 70; MO: n = 70). Opioid use was significantly reduced after surgery (from 21% to 5%). The meta-analysis indicated no significant standardized mean difference between groups in the primary outcome across all pooled studies (standardized mean difference, –0.06 [95% CI, –0.34 to 0.22]). Conclusion: Both AA and MO rotator cuff repair provide large clinical benefits, with few adverse events. There is strong evidence of equivalent clinical improvements. Trial Registration: NCT00128076.