This study was withdrawn because 1) we did not secure needed funding, and 2) the research question became less important while we were waiting for funding opportunities as a few similar studies got published. We have no plan to pursue this study.

Notwithstanding the rapid increase in utilization of reverse total shoulder arthroplasty (RTSA), little consensus or high-quality evidence exists regarding optimal rehab methods following the procedure. Our research question is how different rehab methods influence the clinical outcomes following RTSA. This proposal presents a prospective randomized clinical trial where 100 eligible RTSA patients will be randomly assigned to two rehab groups (Short immobilization with patient-directed therapy vs Long immobilization with supervised therapy) at a single institution. Patient-reported outcomes, objective clinical data, and complications will be compared between the groups over a period of 2 years.

The investigators will perform a prospective randomized non-blinded clinical study where the investigators will recruit a total of 100 patients undergoing RTSA at our institution according to the inclusion and exclusion criteria. Participants will be randomly assigned to one of two rehab groups (50 in each group) - short immobilization combined with patient-directed home therapy (Patient-led rehab) vs long immobilization combined with supervised physical therapy (Supervised rehab) - using a randomization protocol developed by our biostatistics department. The group assignment will be kept sealed in an envelope until the day of surgery to blind both the patients and surgeons.

Surgical procedure and postoperative rehabilitation Enrolled patients will undergo RTSA with one of the two shoulder and elbow surgeons under general anesthesia combined with interscalene block. A standard deltopectoral approach will be used, and the subscapularis tendon will be repaired using transosseous sutures if it is reparable. Tornier reverse shoulder arthroplasty system (Ascent Flex humeral stem with 132.5°inclincation angle, standard Perform Reversed glenoid components, and standard polyethylene liner) will be used in all patients. The size of the humeral stem, baseplate, glenosphere, and polyethylene liner will be chosen on the basis of the patient's anatomy. A sling with an abduction pillow will be used for postoperative immobilization in 30° abduction and 45° internal rotation in all patients. The patient-led rehab group will be educated on their postoperative rehab in detail prior to surgery. The detailed rehab methods for each group are as follows.

1. Patient-led rehab group Duration of the sling: 1 week in sling with abduction pillow for immobilization; then optional for comfort only Start of shoulder exercise: Within 1 week of surgery once pain subsides Mode of rehab: Patients performing exercises on their own following instructions Phase 1 (3-5 days to 2 weeks): Pendulum, deltoid and periscapular muscle isometric strengthening Phase 2 (2-4 weeks): passive, active assisted motion Phase 3 (4-8 weeks): Active motion Phase 4 (8-12 weeks): Strengthening
2. Supervised rehab group Duration of the sling: 4 weeks in sling with abduction pillow for immobilization; then optional for comfort only Start of shoulder exercise: At 4 weeks after surgery Mode of rehab: Patients attending outpatient physical therapy sessions led by therapists Phase 1 (0-4 weeks): Immobilization in sling Phase 2 (4-8 weeks): Passive, active assisted motion Phase 3 (8-12 weeks): Active motin Phase 4 (12-16 weeks): Strengthening

Postoperative outcome evaluation Strict adherence to the rehab protocol will be encouraged and monitored during the rehab period. The research coordinator will contact patients twice a week for the first 2 weeks and then once a week for the following 2 weeks. Patients will be given a pain diary where they will record their pain levels daily along with their daily narcotic use for the first 2 weeks. Patients will return for follow-up, undergo x-rays, and will be evaluated at 2 weeks, 6 weeks, 12 weeks, 6 months, 1 year, and 2 years. The following data will be obtained during the study period.

1. Pain measured using a visual analog scale daily for the first 2 weeks; then at each follow-up visit
2. Patient-reported outcomes: ASES, QuickDASH, SANE, and PROMIS at each visit except at 2 weeks.
3. Range of motion: forward elevation, external rotation, internal rotation at each visit except at 2 weeks
4. Radiographic findings of failure such as dislocation, implant dislodgement, migration, or loosening, scapular notching, periprosthetic humuerus fracture, or scapular fracture at any time: Will be included as complications
5. Complications such as instability, hematoma, wound healing issue, abnormal level of pain due to rehab method at any time, or reoperation
6. Overall patient satisfaction with their rehab method using a 5-point Likert-type scale at 6 months (Question: Were you satisfied with your rehab method? Very unsatisfied, slightly unsatisfied, neutral, somewhat satisfied, very satisfied)

Statistical analysis Sample size calculation: Sample size was calculated to examine the superiority of patient-led rehab over supervised rehab. The null hypothesis is that there is no difference in the primary outcome measure (ASES score at 6 months) between the two rehab methods. Considering similar protocols in the literature12-14 and a mean difference in ASES scores of 11 points with corresponding standard deviation of 18 points, a minimum of 100 patients (50 for each group) are needed to achieve at least 80% power using a two-sided type I error of 0.05. These numbers account for a 10% patient attrition.

Statistical methods: Intention-to-treat analysis will be used for this randomized clinical trial study. Patients who are lost to follow up (dropout), do not comply with their assigned rehab method based on criteria (noncompliance), or change to the other rehab method (cross-over) will be analyzed according to their original group assignment. If there is greater than 10% of data is missing, the group data will be treated according to a missing data protocol - multiple imputation method under a missing at random (MAR) assumption. Patient-reported and objective clinical outcomes will be compared between the two rehab groups using two sample t-test or Mann-Whitney U-test if assumptions are not met. Differences between rehab groups across the follow-up period will be identified using a repeated measures ANOVA, and the appropriate covariance structure will be determined using BIC, a measure of overall model fit. Categorical data including the proportions of complications will be compared between the two rehab groups using Chi-square or Fisher's Exact tests; odds ratios and 95% confidence limits will be calculated where appropriate.

Inclusion Criteria:

1. Aged 18 years or older
2. Diagnosed with rotator cuff arthropathy, glenohumeral osteoarthritis with a > 1.5-cm rotator cuff tear, or massive irreparable rotator cuff tear
3. Have a normal functioning deltoid
4. Have preserved teres minor function (a negative Hornblower's sign)
5. Have failed conservative management for >3 months

Exclusion Criteria:

1. younger than 18 years
2. chronic opiate use history
3. fibromyalgia or other conditions that increase susceptibility to pain
4. preoperative stiffness (definition: < 30 degree passive external rotation, < 90 degree passive forward elevation), neurologic disorder affecting the ipsilateral upper extremity,
5. RTSA for acute proximal humerus fracture or fracture sequelae (nonunion, malunion)
6. previous arthroplasty, fracture fixation, or open rotator cuff procedures in the same shoulder
7. deltoid dysfunction
8. inflammatory glenohumeral arthritis
9. non-English speaking or unable to understand English for consent.

No plan to share IPD