Study design
An investigator-initiated blinded (patient and data analyser) non-inferiority randomized controlled clinical trial.
Material
78 patients with shoulder osteoarthritis randomized to either stemmed or stemless shoulder prosthesis.
Inclusion criteria
1. Patients aged ≥ 18 years.
2. Patients have been referred to the Orthopedic Department or Zealand University Hospital, Koege or Hvidovre University Hospital for shoulder pain during the study period.
3. Patients have the ability to read trial information in Danish and give informed consent.
4. The diagnosis will be based on plain radiographs (at least two perpendicular views), reduction of joint space, and/or osteophyte formation.
5. Clinical presentation with pain at night and/or daily pain, pain in overhead activity, and consumption of pain medication.
Exclusion criteria
1. Previous shoulder surgery that involves the humeral head and/or the glenoid cavity.
1. Patients with alcohol or drug abuse problems that can compromise rehabilitation and follow-up appointments as assessed by the recruiting surgeon at the first visit.
1. Patients unable to understand instructions in Danish, follow the rehabilitation protocol, or answer the questionnaires because of physical or cognitive inabilities as evaluated by the recruiting surgeon at the first visit.
1. Brachial plexus palsy.
1. Patients with previous fractures around the shoulder (clavicle, scapula, and proximal humerus fractures).
1. Patients with MRI scan-verified full thickness total tear of one or more of the rotator cuff tendons.
1. Patients with CT scan-verified glenoid retroversion ≥ 20° that does not allow glenoid component fixation without bone graft or need an augmented glenoid component [12].
Eligibility
Patients referred with the diagnosis of "glenohumeral osteoarthritis" will be examined by clinical examination and plain radiographs to confirm the diagnosis. In the case of glenohumeral osteoarthritis, which is to be treated with shoulder arthroplasty, the shoulder surgeon will offer the patient participation in the trial at the first visit, and the patient will be referred to routine shoulder CT and MRI scans (which are already a part of standard shoulder osteoarthritis protocol) and receive a folder with patient participant information and "Researchers rights in a health science research project" folder.
Patients will have time to consider participation before the next visit (approximately one month) after the first visit.
At the second visit, if the patients give their informed consent for participation in the study, they will receive three questionnaires to fill out while undisturbed in a room next door before handing them to a secretary who will check that the questionnaires are completed before the patient leaving the hospital.
The informed consent gives the primary investigator access to information about the patient's health condition from the medical record, including data on age, sex, American Society of Anesthesiologists (ASA) score, previous shoulder operations, and results from shoulder CT and MRI scans.
All surgical interventions will be undertaken at the Zealand University Hospital Koege and Hvidovre Hospital. A two-year recruitment period from 01.03.2019 to 01.01.2022 is planned.
Figure 1
Randomization and concealment of allocation
A computerized irreversible randomization application, Research Electronic Data Capture (REDCap)© [13], will allocate patients into two equal groups:
· Operated with stemmed shoulder arthroplasty, Biomet Comprehensive® Total Shoulder System.
· Operated with stemless shoulder arthroplasty, Biomet Comprehensive® Nano Shoulder System.
The randomization sequence will be computer generated by block randomization (4 in each block) and balanced according to age in years (18-49, 50-59, 60-69, 70-79, 80-89, 90-110), sex, and ASA-score. The randomization (patient allocation) will be performed on the trial laptop computer in the operation room while the patient is being anesthetized in the preparation room next door. The operating surgeon will obtain the allocation code by logging onto the REDCap website after positioning the anesthetized patient in a beach chair position. After randomization, the patient's allocation will be revealed to the operation team to unpack the necessary operation tools and prosthesis.
Blinding
Patients will be blinded to which implant type they are receive. This will be done by not sharing the operation chart files with the patient’s online file records; the inserted shoulder prosthesis will be noted by a unique code in operation file, and the radiologist will be asked not to describe the post-operation x-ray controls.
The primary investigator will conduct the blinded statistical analysis supervised by the biostatistician.
Interventions
The operations will be performed by four shoulder surgeons familiar with the procedures of both shoulder arthroplasty systems. All patient will be operated on with the deltopectoral approach with subscapularis and biceps tenodesis and instrumentation as described by the manufacture [14, 15].
All patients will receive standard pre-and peri-operative pain management, including general anesthesia and interscalene peripheral nerve block performed by the anaesthesiologist.
Post-operative pain management will be adjusted regarding each patient’s needs and recorded. We do not expect that individualized pain management will affect our primary outcome measure one year after surgery.
Physiotherapy and self-training.
All patients will receive a standardized rehabilitation program at their local municipality service. The physiotherapist in the hospital will demonstrate training protocol and instructions the day after surgery.
From 0–6 weeks, postoperative passive-to-active unloaded movements in all directions are allowed, with the following exceptions:
1. No external rotation allowed the first three weeks. After three weeks, external rotation is allowed to feeling of capsular tightening.
2. No internal rotation of the operated arm behind the back the first six weeks.
Outcome measures
The primary outcome will be the WOOS score at 12 months[16].
The secondary outcomes will measure the following:
1. WOOS score at 3 months.
2. OSS at 3 and 12 months postoperatively [17].
3. General health status questionnaire EQ-5D at 3 and 12 months [18].
At baseline, three-, and twelve-month visits, patients will fill out questionnaires, WOOS, OSS, and EQ-5D, alone before leaving the hospital (at the second visit after accepting participation) and in the waiting room (at three and twelve months visit) without interference from the medical staff.
Follow-up
To ensure that the prosthesis is in situ, all patients will undergo a conventional x-ray in two perpendicular views before discharge. The discharge will be on day one or day two postoperatively, depending on the patient's pain and the need for recruiting social help service at home.
All patients will be followed actively for one year postoperatively. Further follow-up including revision rates will be recorded at five and ten years after surgery using data from the Danish Shoulder Register (DSR).
Complications and adverse events that may have developed after the 12-month evaluation but that did not lead to revision surgery will be extracted from the patient records (e.g. discharge letters, discharge diagnosis, needed journal notes).
Outcome assessment tools
Western Ontario Osteoarthritis of the Shoulder Score (WOOS)
The WOOS is a patient-administered, disease-specific questionnaire for measurement of the quality of life of patients with osteoarthritis. It provides scores for four domains: (1) physical symptoms; (2) sport, recreation, and work; (3) lifestyle; and (4) emotions. Patients answer each question using the visual analogue scale. The WOOS score is calculated by measuring the distance from the left side of the line and calculating the possible score ranging from 0 to 100 (recorded to the nearest 0.5 mm.) It consists of 19 questions, and the total score ranges from 0 to 1900. A maximum score of 1900 signifies that the patient has an extreme decrease in shoulder-related quality of life, whereas a score of 0 signifies that the patient has no decrease in shoulder-related quality of life. The raw scores are converted to a percentage of the maximum score for simplicity of presentation. The questionnaire has been translated into Danish and validated and tested on patients with shoulder osteoarthritis [16, 19, 20].
Oxford Shoulder Score (OSS)
Oxford Shoulder Score (OSS) is a measurement tool for the assessment of outcomes of shoulder surgery [17, 21]. It has been tested and validated in patients with primary or secondary osteoarthritis. The OSS is a 12-item questionnaire, with each item scored from 0–4; thus, the overall score is the sum of the scores received for individual questions. This results in a continuous score ranging from 0 (most severe symptoms) to 48 (least severe symptoms)[17]. For simplicity of presentation, the raw scores will be converted to a percentage of the maximum score. We will use a validated Danish version [21].
EQ-5D
The EQ-5D, the health status component of the EuroQol assessment (EuroQol Group, Rotterdam, The Netherlands), is a generic instrument for describing and evaluating health-related quality of life. The EuroQol instrument has been designed for self-completion by the respondent.
The EQ-5D is a descriptive system comprising five dimensions in each of which the respondents describe their health state: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension has five levels: no problems, slight problems, moderate problems, severe problems, and extreme problems. The patients are asked to indicate their health state by ticking the box next to the most appropriate statement in each of the five dimensions, which corresponds to a one-digit number that expresses the level selected for that dimension. The patient choose one of five levels for each dimension, and thus a five-digit number (EuroQol Group 1990) can define the resulting health state. The reliability and validity of the EQ-5D have been evaluated in different patient populations, including the Danish population, with the conclusion that the assessment had good validity, reliability, and responsiveness [18, 22]. The EQ-5D demonstrated good internal and external responsiveness in patients with shoulder injuries and can therefore be used as an outcome measure for evaluating the HRQOL in both clinical studies and health-care assessments [23–25]. We received permission to use the Danish version (permission ID number 27296) [26].
Patient dropout and protocol violations
Patients who drop out of the trial will be recorded, and the reason for dropout will be noted. The principal investigator will record and report any protocol violation.
Side effects and adverse events reporting
The department routinely uses both types of shoulder prostheses. The treatments’ risks and disadvantages are therefore not expected to differ from normal patient application outside the trial. They include superficial infection, deep infection, loosening of the prosthesis, nerve or vascular injury during surgery, lack of healing, and prosthesis migration. The amount of ionizing radiation to which the subjects are exposed during the CT scan and X-rays does not differ from that for patients outside the study.
To assess possible complications after prosthesis insertion, we will record any case of deep infection, nerve injury, implant loosening (based on radiographic images), heterotopic ossification, instability, dislocation, implant removal, and revision surgery or any adverse event that leads to hospitalization one year after surgery. All complications will be reviewed at the end of the trial and reported.
Statistical analysis plan
Sample size and power calculation
No existing studies report the minimal clinically important difference (MCID) of WOOS, but the MCID of the OSS was investigated and found to be 6 points, or 12.5% of a maximum score [27]. By extrapolating MICD from OSS to WOOS, we calculated the MICD of WOOS to 237.5 points (12.5%).
With a MICD of 12.5% and a standard deviation (SD) of ±20.0, which we found from our retrospective evaluation of patients operated with stemless TSA in our department, we need to include 64 patients, 32 in each group, to be 80% sure that the lower limit of a one-sided 95% confidence interval will be above the non-inferiority threshold of -12.5%. With loss of follow up assumption of 20%, a total of 78 patients is required [28].
Hypotheses
We hypothesize the following:
· A one-year postoperative WOOS score for patients operated with stemless TSA is inferior to that of patients operated with stemmed TSA by at least MCID = 237.5 (12.5%).
(H0: μstemmed > μstemless+ MCID)
· A one-year postoperative WOOS score for patients operated with stemless TSA is not inferior to that of patients operated with stemmed TSA by MCID = 237,5 (12.5%).
(HA: μstemmed ≤ μstemless+ MCID)
Data analysis
No interim data analysis will be carried out. The primary investigator will enter data in the REDCap database, and the principle investigator before the data analysis will perform proofreading of a random sample of 20 subjects.
The primary investigator will conduct the data analysis, supervised by a biostatistician. Patients who drop out after the three-month evaluation will be included in the analysis based on three months of data ('last observation carried forward'). Patient dropouts before the three months’ evaluation will not be included in the analysis unless they participate in the one-year questionnaire evaluation. As suggested in the extension of the CONSORT statement for non-inferiority trial design, a per-protocol analysis will be conducted and results obtained [29, 30]. A questionnaire containing one or more missing answers will be marked as incomplete, and the patients will be contacted by letter to answer the missing questions.
Descriptive statistics will be used to report demographic data. Differences in demographic data and outcome measures between the two groups will be compared using chi- square test for categorical variables, and parametric (student´s test) or non-parametric (Mann Withney U test) test for continues variables depending on the nature of data. Data will be checked for possible extreme values prior to the analysis. The statistical analyses will be performed using SPSS version 25.0 (IBM Corp, Armonk, NY, USA) [31]. The level of statistical significance is set at P < .05, and all P values are two tailed. All complications, including glenoid and humeral component loosening, glenoid wear instability, rotator cuff tear, intraoperative and postoperative periprosthetic fracture, nerve injury, infection, deltoid injury, and deep venous thrombosis will be reported. Our sample size does not give us power enough to allow comparison.