There are no published prospective studies that compared 2-year functional outcomes of RTSA and anatomic TSA for the treatment of primary glenohumeral osteoarthritis with intact rotator cuffs and no excessive glenoid retroversion. The primary goal of our prospective randomized study is to determine whether RTSA have at least as good results as anatomic TSA (non-inferiority), in patients with glenohumeral osteoarthritis, without rotator cuff tears nor significant glenoid retroversion.

Total anatomic shoulder arthroplasty (TSA) is an effective treatment of severe glenohumeral osteoarthritis, with significant improvement in shoulder pain and function. Concerns about glenoid loosening, associated with difficult revision procedures and disappointing outcomes, have however been raised.

Reverse total shoulder arthroplasty (RTSA) was designed with a medialized center of rotation to treat cuff tear arthropathy. Favorable early reports led to the expansion of primary indications of RTSA to proximal humeral fractures as well as osteoarthritis with poor glenoid bone stock. Recent reports revealed excellent clinical results of RTSA for primary glenohumeral arthropathy with intact rotator cuff and a low rate of complications.

Retrospective studies comparing functional results - of anatomic TSA for treating glenohumeral osteoarthritis with RTSA for rotator cuff arthropathy - found equivalent or greater improvements in American Shoulder and Elbow Surgeons score (ASES) at >2-year follow-up. In a study comparing anatomic TSA to RTSA for the treatment of glenohumeral osteoarthritis with intact rotator cuffs, Steen et al. reported equivalent functional results at >2-year follow-up. The retrospective matched cohort study could, however, not eliminate biases, such that RTSA patients had higher preoperative glenoid retroversion than anatomic TSA patients. The authors therefore hypothesize that, in patients treated for glenohumeral osteoarthritis without excessive glenoid retroversion, RTSA will render better functional outcomes than anatomic TSA, at 2 postoperative years. Many other studies confirmed in 2019 Steen's hypothesis. Moreover, several studies revealed good long-term survivorship after RTSA.

There are no published prospective studies that compared 2-year functional outcomes of RTSA and anatomic TSA for the treatment of primary glenohumeral osteoarthritis with intact rotator cuffs and no excessive glenoid retroversion. The primary goal of our prospective randomized study is to determine whether RTSA have at least as good results as anatomic TSA (non-inferiority), in patients with glenohumeral osteoarthritis, without rotator cuff tears nor significant glenoid retroversion.The secondary goals are 1) to evaluate whether RTSA eventually grants superior postoperative clinical and radiographic outcomes than anatomic TSA (superiority), 2) to determine whether RTSA is associated with fewer postoperative complications than anatomic TSA.

Inclusion Criteria:

• Primary glenohumeral arthritis,
• Intact rotator cuff,
• No important glenoid bone loss (cf exclusion criteria),
• Patients between 65 and 85 years old
• Informed Consent as documented by signature (Appendix Informed Consent Form).

Exclusion Criteria:

• B2 glenoid with > 80% posterior humeral head subluxation or greater 25 degrees neoglenoid retroversion,
• B3 and C type glenoids,
• Full thickness rotator cuff tear,
• Acute or malunited proximal humeral fracture,
• Chronic locked dislocation
• Rheumatoid arthritis,
• Revision surgery or surgical antecedents,
• Tumors,
• Axillary nerve damage,
• Non-functioning deltoid muscle,
• Glenoid vault deficiency precluding baseplate fixation,
• Infection and neuropathic joints,
• Known or suspected non-compliance, drug or alcohol abuse,
• Patients incapable of judgement or under tutelage,
• Inability to follow the procedures of the study, e.g. due to language problems, psychological disorders, dementia, contraindication for CT scan etc. of the participant,
• Enrolment of the investigator, his/her family members, employees and other dependent persons.

While the patients and care providers will be aware of the type of prosthesis, the physicians who assessed the outcomes and analyzed the data will be blinded to this information.