Lhee 2013.

Methods	Design: Randomised, double‐blind single‐centre 1:1 parallel group trial Setting: Hospital in South‐Korea Timing: Recruitment from March 2009 to October 2010 Interventions: arthroscopic subscapularis tendon repair versus debridement Sample size: sample size calculation not provided; 256 participants recruited Analysis: not clearly reported
Participants	Number of participants 394 assessed for eligibility 256 randomised (139 to repair and 117 to debridement) At minimum of 24 months data available for 101/139 (73%) in repair group and 90/117 (77%) debridement group Inclusion criteria Full‐thickness subscapular tear Exclusion criteria normal subscapularis tendon during arthroscopic surgery. Baseline data Rotator cuff repair group Mean (SD) Age, years 61 (5.3) Number (%) females 35 (35%) Tear size small to medium 46 (33%); medium to large 65 (47%); large to massive 28 (20%) Hidden partial tear (full thickness but not whole tendon; only visible in external rotation) 39 (28%) Debridement group Mean (SD) Age, years 62 (4.6) Number (%) females 58 (64%) Tear size small to medium 39 (33%); medium to large 55 (47%); large to massive 23 (20%) Hidden partial tear (full thickness but not the whole tendon; only visible in external rotation) 31 (27%)
Interventions	All surgeries were performed by a single surgeon at a single hospital Rotator cuff repair group Lateral‐anterosuperior portal (“Miracle Portal”) was used to repair subscapularis tendon. Bursa anterior to the subscapularis tendon was usually removed for the accurate positioning of the suture‐hook. Subscapularis tendon was released, pulled and sutured with suture‐hook. One or two suture anchors of Modified Mason‐Allen technique was used to secure the tendon. If the subscapularis tendon was not sufficiently mobile, further anterior interval release between subscapularis and scapula was performed. LHB (long head of biceps tendon) was either treated with a biceps tenodesis or by tenotomy when there was tear or subluxation of it. The footprint area of the subscapularis tendon, which is trapezoidal in shape on the proximal part of the lesser tuberosity, was thoroughly cleaned of soft tissue and meticulous bone preparation was done prior to placement of anchor sutures. Debridement Anterosuperior portal was made for debridement (capsulectomy and anterior bursectomy). A systematic release of the glenohumeral ligaments and the overlying subscapularis bursa was also performed. The superior aspect of the tendon was freed. from the surrounding structures (the coracohumeral and superior glenohumeral ligaments). The middle glenohumeral ligament was always released to identify the upper border of the subscapularis tendon. Both groups Postoperative rehabilitation regimen not described.
Outcomes	The participants were assessed at 3 month‐intervals The primary outcome ASES score (0 to 100, higher score indicates better function) Outcomes used in this review Mean function; ASES score
Notes	Trial registration: NCT01996904 Data analysis: the authors report ASES score at final follow‐up with a minimum of 24 months follow‐up. Withdrawals: reasons for loss in follow‐up not given. Cross‐overs: none reported Adverse events: no events in the groups (reported in ClinicalTrials.gov) Serious adverse events: no events in the groups(reported in ClinicalTrials.gov)

ASES: American Shoulder and Elbow Surgeons; SD: standard deviation.