Clinical and radiological outcome of conservative vs. surgical treatment of atraumatic degenerative rotator cuff rupture: design of a randomized controlled trial

The COPACABANA study is designed as a level-1 prospective randomized controlled trial. The study design, procedures, protocols and informed consent are approved by the local Medical Ethical Committee (registration number MZH2008-36). The trial is registered in the Netherlands Trial Registry (NTR TC 2343). The inclusion period is planned from April 2009 to December 2011.

An MRI scan will be done on all patients aged between 45 and 75 with a clinically suspected atraumatic rotator cuff rupture who were referred to the departments of orthopedic surgery and rehabilitation of both Martini Hospital and University Medical Center Groningen, The Netherlands. If the MRI of the affected shoulder shows - as assessed by two independent assessors (PE, FOLH) - a full thickness rotator cuff rupture with degenerative characteristics, the patient will be included in the study. Exclusion criteria are traumatic rotator cuff rupture, previous surgical treatment of the shoulder, frozen shoulder, radiological and symptomatic osteoarthritis of the glenohumeral or acromioclavicular joint, (rheumatoid) arthritis, diabetes mellitus and cognitive disorders, neurological disease or language barriers impairing participation. All patients must give informed consent before participating in the COPACABANA study.

Patients who meet the inclusion criteria will be informed about the study. After consenting to participate, patients are allocated to conservative or surgical treatment. One researcher (FOLH) will perform randomization. Randomization procedure is based on opaque sealed envelopes. Patients will be informed in accordance with the Dutch Medical Treatment Contracts Act about the treatment, the risk of complications and the anticipated period of recovery for both treatment groups.

Conservative treatment consists of a subacromial steroid infiltration. The first infiltration will be performed immediately following inclusion. Using a 21-gauge needle with the covered syringe, the treating orthopedic surgeon injects the patient's subacromial bursa via the anterolateral approach applying an aseptic technique[1]. If no relief is obtained with the first subacromial infiltration, a second infiltration will be performed under radiological or ultrasound guidance. The infiltration will be repeated at a maximum of three times. Further treatment consists of analgesic medication with NSAIDs, paracetamol, and/or tramadol. Patients will be referred to a physiotherapist that uses a standardized protocol developed by the Department of Physical Therapy of Martini Hospital (Table 1). During the first 4 weeks passive movements will be performed to preserve glenohumeral and scapulothoracic mobility. After 6 weeks active movements will be performed. Twelve weeks after commencement of treatment physiotherapy will be aimed at strength regeneration. Mobility will be further optimized. Physical therapy is continued until patients reach a full range of motion and an improvement in strength is achieved.

The surgical procedure is performed by two qualified and experienced surgeons (JvR, RD). Surgery will take place within 6 weeks after the inclusion date. If indicated, NSAIDs or other pain medications will be prescribed in the period until surgery.

Surgery is carried out under general anesthesia, if indicated supplemented with an interscalenus plexus block. The operation is performed in beach-chair position. The approach applied is the anterolateral "deltoid on" approach. The coracoacromial ligament is detached from its insertion and the subacromial bursa is excised. The anteroinferior part of the acromion is removed, as well as osteophytes on the lower surface of the acromioclavicular joint. The ruptured ends of the rotator cuff are mobilized and repaired. Reconstruction depends on the type of rupture [26], augmented by using bone anchors. If repair is not possible due to extensive degenerative changes, debridement and a biceps tenolysis at the glenoid and tenodesis in the sulcus bicipitalis tenodesis with bone anchors will be performed. No tendon transfers or artificial augmentation devices will be used. The deltoid muscle is reattached to the acromion by transosseal refixation. As no differences in outcome have been reported between open or arthroscopic repairs, we chose to use the open approach, omitting the possible bias of a learning curve[16]. Standard postoperative pain relief protocols used in the participating hospitals will be used throughout the study.

After surgery, the patient receives a sling for six weeks combined with exercise instructions (Table 2). Patients will be referred to a physiotherapist who uses a standardized protocol (Table 2). During the first 6 weeks the emphasis is on mobility, not on strength. After 6 weeks strength development will be included as well. Physical therapy after surgery is done until patients reach a full range of motion and an improvement in strength is achieved.

Outcome assessment will take place in both study groups at randomization (T0), 6 weeks post-randomization (T1), and at 3, 6 and 12 months (T2, T3, T4) post-randomization. At 12 months follow-up a second MRI of the affected shoulder will be taken (Figure 1). The outcome assessment will take place in an outpatient clinic setting and will focus on shoulder function and pain. Functional outcome (primary outcome) will be determined with the Constant Murley score. Secondary outcome measures are the Dutch Simple Shoulder Test, the Visual Analogue Scale for pain and restriction, the Goutallier score (Table 3), anatomical localization of the rotator cuff rupture and integrity of the rotator cuff. Further, a generic, patient-based cost-effectiveness analysis will be performed.

Different prospective studies on clinical outcomes following open rotator cuff repair in full-thickness rotator cuff ruptures state that the Constant Murley Score 1 to 3 years postoperatively is between 70 and 90 [19, 31‐34]. Sample size calculation was performed; primary outcome measure was the Constant-Murley Score, whereby 10 points was considered a clinically relevant difference between the two groups, with a standard deviation of 20, alpha set on 5% and power on 80%. This resulted in a required number of 49 patients in each group. Assuming a dropout rate of 10%, two groups of 54 patients will have to be included.

The means and standard deviations of the intervention and control group will be calculated for patient and outcome characteristics. In order to determine whether a difference exists between the surgical and conservative groups as far as primary outcome measurement (pain and function of the shoulder as determined with the Constant-Murley Score) over all five measurement moments is concerned, a random effect analysis shall be carried out which includes corrections for hospital and operating surgeon and for a number of patient characteristics. This analysis will be repeated with the secondary outcome measurements as dependent variables.

The prognostic value of localization and extent of the rotator cuff rupture and the degree of fatty degeneration influencing the functional recovery after 12 months will be analyzed using a linear regression model. Here too corrections will be made for hospital and operator and for a number of patient characteristics.