This is a randomized controlled trial comparing the Comprehensive Nano stemless total shoulder arthroplasty (intervention group) with the Comprehensive un-cemented stemmed total shoulder arthroplasty (control group). All Danish citizens with glenohumeral osteoarthritis indicating a total shoulder arthroplasty referred to the orthopedic department at Copenhagen University Hospital in Gentofte will be offered participation in the trial. The following exclude from participation in the study: below 18 years of age; cognitive or linguistic impairment; insufficient function of the rotator cuff; poor bone quality and ASA group 4-5. A total of 122 patients will be included of which 56 will be part of radiostereometric analysis (RSA) study of humeral component migration. This will allow a maximum of 20% drop out. The primary outcome is magnitude and pattern of migration of the humeral component assessed by RSA. The secondary outcomes are patient-reported outcomes, functional outcome, readmission, complications, revisions, changes in bone mineral density (BMD) of the proximal humerus assessed by duel energy x-ray absorptiometry (DXA) and economy (cost utility analysis). To avoid bias in some of these outcomes, the patient will not know what type of humeral component was used. The patients are examined before the operation and 3, 6, 12 and 24 months after the operation.

Background Osteoarthritis (wear and tear) of the shoulder can, in the most severe cases, require insertion of shoulder prosthesis (an artificial shoulder joint, a total shoulder arthroplasty). This is an effective treatment with pain-relief and significant improvement in shoulder function. A design with a metal stem in the humeral bone canal is currently regarded as the best treatment option, but complications related to the stem including humeral fractures can have devastating consequences. By using a stemless humeral component stem-related complications can be reduced. Furthermore, the canal preserving design may also facilitate further surgery should the need of a revision prosthesis arise.

Currently, little is known about the results of the stemless design. The initial results have been promising, but as with the stemmed design migration and eventually loosening of the prosthesis can lead to poor results and, in some cases, revision.

Using radiostereometric analysis (RSA) a special x-ray examination technique based upon insertion of small metal beads into the bone surrounding an implant and attachment of beads to the prosthesis (or using 3-dimensional models of the implant (CAD models) = model-based RSA (MB-RSA)), implant migration can be measured extremely accurate. The RSA technique has previously been used to study migration and loosening of hip and knee prostheses. It has been suggested that RSA should be used to monitor small series of new prostheses the first two years after surgery as a part of a safe phased introduction.

RSA has never been used to access migration of a stemmed or a stemless humeral component nor has the stemmed and the stemless humeral component been compared with regard to pain relief and shoulder function.

Purpose The purpose is to compare the Comprehensive Nano stemless total shoulder prosthesis (intervention group) with the Comprehensive stemmed total shoulder prosthesis (control group).

Method All Danish citizens with shoulder osteoarthritis indicating a total shoulder arthroplasty referred to the orthopedic department at Copenhagen University Hospital in Gentofte will be offered participation in the trial. The following exclude from participation in the study: below 18 years of age, cognitive or linguistic impairment, insufficient function of the rotator cuff (muscles which stabilize the shoulder joint) and severe co-morbidity (ASA-score 4-5).

A total of 122 patients will be included of which 56 will be part of the RSA analysis. This will allow a maximum of 20% drop out. The study will be conducted as a randomized clinical trial (RCT) where each participant by lot will be allocated to one of the two designs. Patients will subsequently be carefully monitored to determine whether there is a difference. The primary outcome is magnitude and pattern of migration of the humeral component assessed by MB-RSA. The secondary outcomes are pain relief, questionnaires (patient-reported outcome), range of movement and strength (functional outcome), readmission, complications, revisions, changes in bone density round the shoulder assessed by duel energy x-ray absorptiometry (DXA) and economy (cost utility analysis). To avoid bias in some of these outcomes, the patient will not know what type of prosthesis was used. The patients are examined before the operation and 3, 6, 12 and 24 months after the operation. The investigators expect short-term publications of early results at about 2 years after inclusion of the last patient.

Ethical considerations Apart from the differences in design, the two groups are treated alike. Today, the two designs are considered equal in the treatment of osteoarthritis. With this study, all patients are at risk of being treated with a prosthesis that subsequent analyses will deem inferior. This is implicit in the study design and there is nothing a priori to suggest which prosthesis is the better.

The standard treatment at the orthopedic department at Copenhagen University Hospital in Gentofte includes preoperative planning with plain radiographs, CT and MRI scans of the glenohumeral joint and follow-up examinations with plain radiographs postoperatively and at 3 months. In addition to this standard treatment the participating patients will have additional radiographic examinations including:

RSA examination at 1 week, 3 months, 6 months, and at 1 and 2 years: 0.04 mSv Plain radiographs at 1 and 2 years: 0.02 mSv DXA scans preoperatively, at 1 week, 3 months, 6 months, and at 1 and 2 years: 0.04 mSv

The background radiation in Denmark is 3 mSv each year. The extra radiation dosage of 0.10 mSv from this study corresponds to the background radiation during approximately 12 days.

The standard treatment at the orthopedic department at Copenhagen University Hospital in Gentofte includes a follow-up examination at 3 months. The participating patients will have additional follow-up examinations at 1 week, 6 months, and at 1 and 2 years. Besides being time consuming this may maintain the patients in a role of being ill. However, the patients will also have the advantage of feeling secure and any uncertainty or problems can be addressed more easily.

The results of the study should be of benefit to future patients with osteoarthritis of the shoulder and thereby also for the use of resources in healthcare. Neither the coordinating investigator nor participating investigators or surgeons have personal economic interests in the study results.

Inclusion Criteria:

• All Danish citizens with glenohumeral osteoarthritis indicating a total shoulder arthroplasty referred to the orthopedic department at Copenhagen University Hospital in Gentofte

Exclusion Criteria:

• below 18 years of age
• cognitive or linguistic impairment
• insufficient function of the rotator cuff
• poor bone quality
• ASA group 4-5

It is not the plan to make IPD available to other researchers.