Different treatment options are available for massive or irreparable rotator cuff tears. An arthroscopic or an open repair approach is possible in the majority of cases and functional outcomes are improved, particularly when a complete arthroscopic repair can be achieved. However, the healing rate of massive rotator cuff tears after repair may remain low and failure of healing is associated with progression of arthritis. An alternative to arthroscopic or open repair is reverse shoulder arthroplasty which decreases pain and improves function, active shoulder elevation and quality of life. The primary goal of this prospective multicentric randomized study is to determine if there is a difference of functional outcomes between rotator cuff repair (RCR) repair and reverse shoulder arthroplasty (RSA).

The majority of degenerative rotator cuff tears occur in individuals over 60 years of age. Therefore, as our population increases in size and advances in age, the incidence of rotator cuff tears is also increasing. A growing number of people are remaining active at this age, and continue to place substantial physical demands on their shoulders notably into their seventh and eighth decades of life. At the same time, the rotator cuff undergoes intrinsic degeneration and the prevalence of osteoporosis increases. Consequently, a significant and growing number of arthroscopic rotator cuff repairs are performed in individuals with poor soft tissue or bone quality. Moreover, whereas most rotator cuff tears occur at the tendon-bone insertion, fixation quality can be challenged by a tear that occurs more medially, leaving only a small amount of tendon for fixation by suture.

Different treatment options are available for massive or irreparable rotator cuff tears, including debridement and subacromial decompression, repair (partial or complete), transfer of the subscapularis tendon, transfer of the teres major muscle, deltoid flap reconstruction, transfer of the latissimus dorsi or the pectoralis major, superior capsule reconstruction, augmented cuff repair, subacromial balloon and reverse total shoulder replacements. None of these treatments has proved superiority on other ones, particularly when the rotator cuff is massively torn.

Massive degenerative rotator ruff tears are a challenge. An arthroscopic or an open repair approach is possible in the majority of cases and functional outcomes are improved, particularly when a complete arthroscopic repair can be achieved. However, the healing rate of massive rotator cuff tears after repair may remain low and failure of healing is associated with progression of arthritis. An alternative to arthroscopic or open repair is reverse shoulder arthroplasty which decreases pain and improves function, active shoulder elevation and quality of life. Recently, Liu et al. demonstrated that both rotator cuff repair (RCR) and reverse shoulder arthroplasty (RSA) are effective and reliable options for massive rotator cuff tears (RCT), but revealed a better shoulder function for patients in the rotator cuff repair (RCR) group. While these results are interesting, this study remains retrospective and call for new studies with a higher level of evidence.

The primary goal of this prospective multicentric randomized study is to determine if there is a difference of functional outcomes between rotator cuff repair (RCR) repair and reverse shoulder arthroplasty (RSA).

Inclusion Criteria:

• 1 Patient voluntarily consents to participate
• 2 Patient with a massive and reparable non traumatic rotator cuff tears after failure of conservative treatment
• 3 Patients with the following pre-operative images: Three standardized X-rays series and Magnetic Resonance Imaging (MRI) or arthro Magnetic Resonance Imaging without sign of traumatic lesions

Exclusion Criteria:

• 1 Patient has known intentions, obligations, or co-morbidity that would inhibit them from participating in the study
• 2 Revision rotator cuff repair
• 3 Patient consent withdrawal
• 4 Glenohumeral arthritis defined as stage > 3 Hamada classification
• 5 Infection and neuropathic joints
• 6 Known or suspected non-compliance, drug or alcohol abuse
• 7 Patients incapable of judgement or under tutelage
• 8 Inability to follow the procedures of the study
• 9 Enrolment of the investigator, his/her family members, employees and other dependent persons

Under reasonable request from scientists (with accepted research protocol)

The statistician will be blinded to patient allocation groups.