Osteoarthritis (OA) is the most common form of arthritis, affecting millions of people worldwide and it is common in an aging population. Surgical shoulder replacement (arthroplasty) is typically considered when non-surgical measures, such as physiotherapy or medication, have failed. There are two commonly performed surgical replacement procedures in patients who have advanced shoulder OA, and are 65 years of age and older: "Total Shoulder replacement or Arthroplasty (TSA)" and "Reverse Total Shoulder Arthroplasty (RTSA)".

Few studies have compared the two procedures. Surgeons face uncertainty regarding which procedure to perform in patients 65 years of age and older. This pilot Randomized Controlled Trial (RCT) will compare the "TSA" and "RTSA" procedures, in patients 65 years of age and older. Participants will be assigned at random, (like flipping a coin), to one of the two groups (TSA or RTSA). The overall goal of this pilot study is to determine which procedure produces better functional and quality of life outcomes with fewer complications within the first 12-months after surgery. Moreover, pilot data will help determining the feasibility of conducting a larger trial comparing TSA versus RTSA surgical management in 65 years of age and older participants with advanced shoulder OA.

1. Statement of objectives

   Primary objective: To compare clinical and health related quality of life (HRQL) outcomes of Total Shoulder Arthroplasty (TSA) or Reverse Total Shoulder Arthroplasty (RTSA) management in 65 years of age and older patients with advanced shoulder osteoarthristis (OA) over the first 12-months post-surgery.

   Secondary objective: To determine the feasibility of conducting a definitive randomized control trial (RCT) comparing TSA versus RTSA surgical management in 65 years of age and older patients with advanced shoulder OA.

2. Background, Rationale, and Present state of Knowledge

   Shoulder OA is common in aging population and may substantially impact quality of life.1-3 Seniors older than 65 years may surpass 2 billion by 20504 increasing the impact of chronic conditions such as OA on both health services and society.5 Patients with OA who do not respond to non-operative management usually require shoulder arthroplasty such as TSA and RTSA.3

   TSA is considered the gold standard treatment for younger patients with advanced shoulder OA, intact rotator cuff and preserved glenoid bone stock.6 However, for patients older than age 65 years with an intact rotator cuff (RC), optimal management is unclear. Even though RC muscles may be preserved at time of TSA, RC pathology increases with age.9 Further, RC dysfunction has been linked to increased numbers of TSA failures.10-12 Young et al.10 found that secondary RC dysfunction developed in a significant number of participants undergoing TSA and was associated with worse clinical and radiographic outcomes.10 Melis et al.11 reported 65% of patients undergoing revision TSA had associated RC tears at time of revision surgery. Chronic secondary RC pathology is seen in 55% of patients who retained their implants after TSA for over 15 years.12 The available evidence regarding outcomes after TSA compared to RTSA is mostly based on non-randomized cohort studies and may pose some challenges in interpreting the data, due to variation in patient characteristics and surgeon preference.

   Given the lack of consensus on optimal management, and conflicting, low quality evidence reports among patients older than 65 years with advanced OA, this pilot study will compare clinical and HRQL outcomes of TSA and RTSA surgical management in 65-79 year old participants with advanced shoulder OA over the first 6-months post-surgery. Secondarily, we will determine the feasibility of conducting a definitive RCT in this population.

3. The Proposed Trial Design: Parallel-arm, double-blinded, randomized clinical trial.

Enrollment Procedures: Surgeons will screen potential participants. The Research Associates (RAs) will discuss the study in detail and obtain informed consent.

Baseline Evaluation: The RAs will measure bilateral shoulder range of motion (ROM) using a goniometer, bilateral shoulder strength using a dynamometer, and collect Constant score measurements. Participants will complete the following HRQL questionnaires: Western Ontario Osteoarthritis of the Shoulder (WOOS), Subjective Shoulder Value (SSV), The American Shoulder and Elbow Surgeon's (ASES) questionnaire, and EuroQol EQ-5D-5L quality of life questionnaire. The participant will also provide socio-demographics (e.g., age, sex, height, weight, handedness, smoking status, recreational sports, symptom duration, type of onset [insidious / traumatic]), Visual Analogue Scale (VAS) for pain and current medications. The surgeon will review pre-operative CT scans to determine rotator cuff muscle quality.

Randomization: Participants will be randomized to either Group A (TSA) or Group B (RTSA) preoperatively; participants will be blinded to group allocation. The randomization sequence will be computer-generated in uneven blocks with a 1:1 ratio. The surgeons will use a Research Electronic Data Capture (REDCap) software to randomize the participant. REDCap is password protected and only accessibly for research personnel. The RAs will not have access to the randomization information in REDCap which will maintain blinding to participant group allocation.

Participant Management:

Post-operative study evaluations will occur at:

• 6 weeks, 3-, 6- and 12-months: RAs will measure bilateral shoulder ROM and collect Constant data. Participants will fill the following HRQL questionnaires: WOOS, SSV, ASES, and EuroQol EQ-5D-5L as well as VAS for pain and current medications. At 6- and 12-months, bilateral shoulder strength will also be collected. Surgeons will assess participants for complications and/or adverse events.

Sample Size: For this pilot trial, 40 participants (20/group) will allow us to assess participant recruitment and retention rates. We will also refine our selection criteria as necessary for the definitive RCT. This pilot evaluation will provide for calculation of the sample size for the definitive RCT based on confirmation of the primary outcome measure (WOOS). Although 40 participants will likely be inadequate to detect significant group differences, it should be adequate to determine if previously established minimal clinically important difference (MCID) occur in clinical outcomes between groups.

Statistical Analysis:

Primary: Our primary objective is to compare clinical and HRQL outcomes of TSA and RTSA surgical managements in 65 years of age and older participants with advanced shoulder OA over the 12-months post-surgery. Statistical analysis will be performed using Intention-to-Treat with all outcomes attributed to the assigned group. Descriptive statistics will be used for group comparisons with independent t-tests for continuous and Chi-square tests for categorical variables at baseline. Linear Mixed Modeling (LMM), adjusted for age and sex, will compare outcomes between groups over the 12-months evaluation period. The level of significance will be set at α=0.05. Statistical analysis will use the following software: SPSS version 25.0 or SAS version 9.4.

Secondary: Descriptive analysis will assist us in refining eligibility criteria, recruitment/retention rates and study timelines prior to undertaking a large scale RCT.

Recruitment, enrollment and retention rates: All 65 years of age and older patients with advanced shoulder OA presenting at the recruiting sites will be screened so that eligibility, and recruitment rates can be assessed; including the proportion of patients 1) willing to participate and 2) appropriate for randomization. We will also assess retention rates at the end of study follow-up and reasons for losses to follow-up.

Sample Size Calculation: The intervention analysis will also allow us to determine if we should perform a superiority or inferiority RCT. If there are no MCID detected in the pilot study (expected), the definitive RCT will be powered as a non-inferiority trial (i.e. similar clinical outcomes can be achieved with both surgical management (TSA and RTSA) in this age group. If MCID are found in clinical measures, the definitive RCT will be powered as a superiority trial (e.g. RTSA is better than TSA surgical management).

Inclusion Criteria:

1. Patients who have failed standard non-surgical management of their shoulder osteoarthritis who would benefit from a shoulder arthroplasty. Failed medical management will be defined as persistent pain and disability despite adequate standard non-operative management for 6 months. Medical management will be defined as:

   1. The use of drugs including analgesics and non-steroidal anti-inflammatory drugs
   2. Physiotherapy consisting of stretching, strengthening and local modalities (ultrasound, cryotherapy, etc.)
   3. Activity modification

2. Imaging, and intra-operative findings confirming advanced gleno-humeral cartilage loss

3. Patients may present with a glenoid deficiency and >15 degrees of retroversion

4. 65 years of age and older

Exclusion Criteria:

1. Active joint or systemic infection
2. Rotator cuff arthropathy
3. Need for an augmented glenoid component or a bone graft to correct version to within 10 degrees of neutral
4. Retroversion cannot be surgically corrected to within 10 degrees of neutral with a "high- side" ream technique
5. Significant muscle paralysis
6. Charcot's arthropathy
7. Major medical illness (life expectancy less than 1 year or unacceptably high operative risk)
8. Active Workers Compensation Board (WCB) claim
9. Unable to understand the consent form/process
10. Psychiatric illness that precludes informed consent
11. Unwilling to be followed for the duration of the study
12. History of previous shoulder surgery on affected side

No plan developed

Participants and outcome assessors will be blinded to group allocation. The surgeons will use a Research Electronic Data Capture (REDCap) software to randomize the participant. REDCap is password protected and only accessibly for research personnel. The outcome assessors will not have access to the randomization information in REDCap which will maintain blinding to participant group allocation. Surgeons will not informe participant about procedure received until the end of last follow-up.