The purpose of this study was to assess the clinical effect of a mobile application supporting home-based rehabilitation for the patients after arthroscopic rotator cuff repair. The investigators hypothesized the clinical results of the patients using a mobile application (APP) to support the home-based rehabilitation were comparable to the patients receiving the supervised rehabilitation.

This prospective randomized case-control study was approved by the institutional review board of the Kaohsiung Veteran General Hospital (IRB No. KSVGH18-CT12-15) prior to enroll any patients. Patients were recruited if they had a small to medium-sized full-thickness rotator cuff tear or, a Lafosse type II or III subscapularis tear diagnosed and then repaired under shoulder arthroscope. After the surgery, patients were randomized either to the home-based rehabilitation (the home group) or the hospital supervised rehabilitation (the supervised group).

In the home group, patients self-managed rehabilitation exercise without supervision. Rehabilitation were supportive with the APP. Patient could communicate with the physician via the APP. In the supervised group, patients attended one-on-one instructions with therapists and exercised under supervision at hospital.

Patients' characters were recorded. Peri-operative factors associated with rotator cuff healing were assessed. The active ROM (forward elevation, abduction, external and internal rotation), the visual analogue scale (VAS) pain scores, the American shoulder and elbow surgeon shoulder (ASES) scores and the modified Constant scores were recorded pre-operatively and post-operatively 3, 6, 12 and 24 months. The isometric shoulder strength was assessed with the hand-held dynamometer. The compliance of post-operative rehabilitation was evaluated not only from patients' self-reported logs but also by physicians at post-operatively 6, 12 and 24 weeks. Tendon integrity was evaluated with MRI scan at least 6 months after the index surgery. In the pilot study, we found a mean difference of 4 points and a standard deviation of 5.5 points in the modified Constant scores. Power analysis revealed a total sample size of 62 patients (31 patients in each group) would achieve a statistical power of 0.8 with a two-tailed level of 0.05 to detect significant differences. Statistical level of significance was defined as p<0.05.

Introduction- Arthroscopic repair in the treatment of rotator cuff tear had been accepted wildly. It had been demonstrated that surgical management improved life quality and reduced economic burdens. However, joint stiffness and cuff re-tear, the two most common complications after arthroscopic cuff repair were concerned. Post-operative rehabilitation and its compliance are crucial for resilience. For rehabilitation, performing exercise too slowly caused joint stiffness, but too aggressively increased the risk of cuff re-tear. To prevent these complications, supervised physical therapy was advised after rotator cuff repair. However, the associated cost of rehabilitation under supervision exacerbated patients' financial burden. Home-based rehabilitation had been proposed and employed wildly, such as for the cardiopulmonary function after coronary artery disease, the ambulation training after joint replacement and the muscle strengthening after ACL reconstruction. In the conservative treatment of frozen shoulder and rotator cuff tear, patients with home-based physical therapy obtained comparable outcomes to supervised physical therapy. After arthroscopic rotator cuff repair, the clinical benefit of home-based rehabilitation was still controversial. In previous studies, supervised rehabilitation was not superior to home-based rehabilitation in long-term functional outcomes. However, some authors considered that the rehabilitation under supervision reduced the risk of cuff re-tear and joint stiffness. Furthermore, the patients who received the home-based rehabilitation supporting with a booklet or videotape had inferior compliance. For resolving the above-mentioned problems, supporting with a mobile application was possibly an alternative. It is a trend nowadays that telerehabilitation was assisted with mobile applications, such as for patients with stroke, spinal cord injury and musculoskeletal trauma. There was still lack of the literature about such rehabilitation method for the patients after arthroscopic rotator cuff repair. The purpose of this study was to assess the clinical effect of a mobile application supporting home-based rehabilitation for the patients after arthroscopic rotator cuff repair. The investigators hypothesized the clinical results of the patients using a mobile application to support the home-based rehabilitation were comparable to the patients receiving the supervised rehabilitation.

Patients Registries- This prospective randomized case-control study was approved by the institutional review board of the Kaohsiung Veteran General Hospital (IRB No. KSVGH18-CT12-15) prior to enroll any patients. Eligible patients were determined by the attending surgeon. After the surgery, a research assistant conducted the informed consent process before the enrollment in the study. Subjects were randomized either to the home-based rehabilitation (the home group) or the hospital supervised rehabilitation (the supervised group). Each patient was assigned a study number in a consecutive sequence. A randomization sequence using two variables was determined by the mobile application. The allocation of two rehabilitation modes according to the randomization sequence was designated to each subject in a sealed envelope. The therapist opened envelopes at the 1st day post-operatively.

Surgical procedures- Arthroscopic rotator cuff repair for all patients were performed by a single senior surgeon (F.H) at Kaohsiung Veteran General Hospital. Under general anesthesia, patients sit on the beach-chair. The posterior, posterolateral, lateral, and anterolateral portals were created. Intra-articular lesions including subscapularis and biceps long-head tendon were evaluated via the posterior viewing portal. For a Lafosse type II or III tear, subscapularis was repaired with one or two anchors under posterior visualization. Tenotomy, tenodesis or labrum repair for biceps lesions was performed according to labrum tear pattern, age, and functional demand. Rotator cuff lesion was addressed while viewing from the lateral portal. The dimension of involved cuff from anterior to posterior was measured with a probe. Single- or double-row repair was performed based on tear size and pattern. Acromioplasty or partial distal clavicle resection was performed if indicated.

Rehabilitation protocol- All patients received the one standard rehabilitation protocol and used the booklet that described the rehabilitation exercise in detail. The 6-month rehabilitation protocol consisted of four stages. The first stage was the immobilization phase that patients wear the arm sling for 2-4 weeks. The second stage was the ROM phase that patients started from passive to assisted-active ROM for 4-6 weeks. The third stage was the low-tensity training phase that patients allowed active ROM and isokinetic elastic-band exercise for 4-8 weeks. The fourth stage was the joint stabilization phase that patients focused on training the muscle for scapular stabilization and continued isokinetic exercise at least 2 months post-operatively. All patients were follow-up in the clinic every 4-6 weeks after the surgery. After the clinical visit, they were instructed by therapists to perform exercise at home while proceeding the next rehabilitation stage. The exercise program was adjusted according to the recovery status of each patient.

Intervention- In the home group, patients managed rehabilitation exercise by themselves after the surgery without supervision. Rehabilitation protocol and exercise instruction video were available in the APP. Patient could communicate with the physician and therapist via the APP. In the supervised group, subjects attended one-on-one instructions with a therapist (once a week at post-operative 2nd~12th week and once two weeks at post-operative 13th~24th week; totally 16 sessions in the 6-month rehabilitation program) and exercised under supervision at hospital. Subjects would receive additional sessions based on the rehabilitation progress. Subjects could cancel scheduled sessions by personal reasons. We provided telephone consultation without APP communication for subjects in the supervised group.

Outcome evaluation- Patients characters including age, sex, body mass index, arm dominance, trauma history, education background and functional demand were recorded. Peri-operative factors associated with rotator cuff healing were assessed. The active ROM (forward elevation, abduction, external and internal rotation), the visual analogue scale (VAS) pain scores, the American shoulder and elbow surgeon shoulder (ASES) scores and the modified Constant scores were recorded by a research assistant preoperatively and post-operatively 3, 6, 12 and 24 months. Biceps, subscapularis, infraspinatus, and supraspinatus index were measured post-operatively 3, 6, 12 and 24 months. The compliance of post-operative rehabilitation was assessed at post-operatively 6, 12 and 24 weeks. Tendon integrity was evaluated with MRI scan (a 1.5-T unit) at least 6 months after the surgery.

Sample size analysis and statistics methods- In the pilot study, the investigators found a mean difference of 4 points and a standard deviation of 5.5 points in the modified Constant scores. Power analysis revealed a total sample size of 62 patients (31 patients in each group) would achieve a statistical power of 0.8 with a two-tailed level of 0.05 to detect significant differences. All statistical analyses were performed using the SPSS (version 23; IBM, Armonk, NY). Categorical data between groups were compared with the Pearson's chi-squared test or Fisher's exact test. Comparisons of normally distributed continuous data with the independent t-test and non-normally distributed continuous data with the rank sum test were performed. The analysis of covariance was used to compare the subjective scores (the VAS pain scores, ASES scores and modified Constant scores) and objective outcomes (active ROM and muscle index) between two groups. Statistical level of significance was defined as p<0.05.

Inclusion Criteria:

• patients had a small to medium-sized (less than 3 cm) full-thickness rotator cuff tear or, a Lafosse type II or III subscapularis tear diagnosed and repaired by shoulder arthroscope
• patients were willing to consent to post-operative rehabilitation randomization and commit to the two-year clinical follow-up period.

Exclusion Criteria:

• pre-operative shoulder stiffness (defined as passive forward elevation less than 100 degrees and external rotation loss over 50% comparing to the contralateral side)
• an intra-operative irreparable cuff tear
• a prior ipsilateral shoulder surgery
• a history of systemic auto-immune disease or septic arthritis of the ipsilateral shoulder

IPD will not be shared.