The Arthroscopic Treatment of Recurrent Anterior Shoulder Instability

Study Objective

The objective of this study is to investigate the outcomes of arthroscopic treatments of recurrent anterior shoulder instability. Arthroscopic bone grafting of the glenoid will be compared to the gold standard, arthroscopic Bankart repair. The investigators hypothesize that arthroscopic anatomic glenoid reconstruction with bone graft will provide better functional outcomes and decreased risk of recurrent dislocation and subluxation. The null hypothesis is that there is no difference in change in WOSI scores between arthroscopic anatomic glenoid reconstruction and the current standard treatment, arthroscopic Bankart repair, at two years of follow up in patients with recurrent shoulder dislocation.

Subject Selection

The objective of this study is to determine if glenoid bone grafting provides superior functional outcomes as determined by WOSI score to Bankart repair in patients with anterior shoulder instability. This will be a single center, double blinded, randomized controlled trial of a minimum of 200 patients performed at the Halifax Infirmary, Queen Elizabeth II Health Sciences Centre in Halifax, Nova Scotia, Canada. The primary outcome measured will be WOSI score. This is a validated measure assessing quality of life in patients with this condition. Secondary outcomes will be subluxation, re-dislocation, strength and ROM, and other validated functional outcome scores, including: the VR-12 questionnaire, the DASH questionnaire, and the Marx Shoulder Activity Scale (appendix). The ISIS (instability severity index score) (appendix) score will be calculated.

The Western Ontario Shoulder Instability Index (WOSI) is a tool designed for self-assessment of shoulder function for patients with instability problems. It is a disease specific quality of life score. Its high sensitivity makes it able to monitor an individual patient's progress. The WOSI score has good validity (0.768), a high degree of reliability (0.911), and a high degree of responsiveness.

The Veterans RAND 12-Item Health Survey (VR-12) is a brief, generic, multi-use, self-administered health survey comprised of 12 items. It has been shown to have high internal consistency reliability (alpha > .80) and high convergent validity for EQ-5D items (r > .56). . The instrument is primarily used to measure health-related quality of life, to estimate disease burden, and to evaluate disease-specific benchmarks.

The DASH Outcome Measure is a 30-item, self-report questionnaire designed to measure physical function and symptoms in people with any of several musculoskeletal disorders of the upper limb. DASH stands for "Disabilities of the Arm, Shoulder and Hand." In a Norweigan study examing those with shoulder conditions the reliability was shown to be 0.93 and Construct validity was 0.75. The tool gives clinicians and researchers the advantage of having a single, reliable instrument that can be used to assess any or all joints in the upper extremity.

The Marx Shoulder Activity Scale is a short, easy to administer measure of shoulder activity used to predict outcome of shoulder disorders. The activity rating is a numerical sum of scores for five activities rated on a five-point frequency scale from never performed (0 points) to daily (4 points). The activity scale has excellent reliability (intraclass correlation coefficient 0.92) and construct validity (r = 0.66). It can be completed quickly and used in conjunction with patient-based measures of shoulder outcome to define patient populations for cohort studies, and to assess activity level as a prognostic factor in patients with shoulder disorders.

The ISIS is a simple scoring system based on factors derived from a pre-operative questionnaire, physical examination and AP radiographs. It may help to distinguish between patients who will benefit from an arthroscopic anterior stabilization using suture anchors and those who will not. It has good reliability (r>0.8). The score has high validity, as patients with a score of six points or less have an acceptable recurrence risk of 10%, and are therefore potentially good candidates for this procedure. By contrast, those patients with more than six points have a recurrence risk of 70%.

Safety of both procedures will be monitored and any adverse events will be recorded. Preoperative CT and MRI findings of bone loss will be compared with each other and also with arthroscopic intra-operative measurements. The investigators hypothesize that arthroscopic bone grafting of the glenoid provide superior functional outcomes as determined by WOSI score compared to Bankart repair in patients with anterior shoulder instability.

A minimum of 200 patients will be enrolled in the study as determined by a priori sample size calculations. The sample-size calculation was based on the mean WOSI score (85 of 100; standard deviation [SD] = 20) from a sample of 133 patients with traumatic anterior shoulder instability who were followed postoperatively for a minimum of one year. An estimate of change (10% change from the mean WOSI score) was used to ensure sufficient sample size and power. With the use of a two-sided test, an α value of 0.05, and power of 0.90, the sample size was determined to be eighty-six patients per group. This was inflated to one hundred patients per group to account for an estimated 16.8% loss to follow-up (this value demonstrated in a study with Canadian patients with similar characteristics as the target population). Each of the five surgeons participating in the study see approximately 40 patients with this pathology per year, giving an estimated recruitment time of 200 patients in one year.

Inclusion Criteria 2 or more documented incidences of anterior glenohumeral dislocation

Exclusion Criteria Uncontrolled diabetes (Hgb A1C >7% Age 16-60 Prior surgery of affected shoulder Pregnancy Multidirectional instability Posterior instability Paralysis of the shoulder Cancer Severe systemic illness

Patients with recurrent anterior shoulder instability who are referred to any of the orthopedic surgeons at Halifax Infirmary or the Dartmouth General Hospital will be initially seen by the research coordinator, who will confirm the diagnosis and screen patients for study eligibility. Following a detailed discussion of the risks/benefits/alternatives of the study, the patient will be provided with a copy of the informed consent to review if they demonstrate interest in the study. If the patient then wishes to be enrolled in the study, the surgeon or research assistant will obtain patient consent for the study according to the Division of Orthopaedic Surgery and Capital District Health Authority (CDHA) institutional protocols.

Research Plan

In the pre-operative period, there will be no additional time requirements for the patient, with the exception of the informed consent process for the study. Routine radiographs including an AP, axillary, and Y view of the shoulder as well as a pre-operative CT with 3D reconstruction and MRI are obtained for all patients, as is the standard of care for patients with recurrent shoulder instability. The patient will not be exposed to any additional radiation above the usual standard of care.

The patients will return to clinic to complete a pre-operative assessment and collection of baseline data. Structured clinical examination will be conducted by a physiotherapist. and consist of active and passive range of motion testing by goniometer as well as strength testing by hand dynamometer. The patient will also complete baseline questionnaires ( WOSI questionnaire, the VR-12 questionnaire, the DASH questionnaire, the Marx Shoulder Activity Scale) and the ISIS score will be calculated. Any complications or events of dislocation or subluxation will be recorded. Primary outcome will be WOSI score. Secondary outcome will be subluxation/dislocation, and other quality of life measures and functional clinical assessment (ROM and strength) will be assessed.

Randomization For eligible patients who meet the inclusion criteria, the research coordinator will obtain informed consent and randomly allocate patients to one of two surgical procedure groups: Bankart repair or Bankart repair and anatomic glenoid reconstruction with allograft bone. Randomization will be based on surgeon skill, with one surgeon (Dr. Wong) performing the anatomic glenoid reconstruction, and four Surgeons (Drs. Trenholm, Coady, Urquhart, \& Legay) performing Bankart repairs. Allocation will be determined through the use of computer-generated, variable-block-size randomization and consecutively numbered opaque envelopes. Patients will then meet the assigned surgeon, who will discuss the repair technique in a typical clinical context, reconfirm eligibility, and address any patient concerns regarding consent and randomization. The expertise-based randomization design allowed the surgeons to perform their preferred surgical procedure. Two groups (Bankart repair and anatomic glenoid reconstruction) will be selected with 100 subjects each. After patients have been randomized, the treating surgeon will have final decision to exclude patient.

Surgical Technique A diagnostic arthroscopy is performed according to the method described by Snyder (2003). The amount of glenoid bone loss and any Hill-Sachs lesions are visualized and assessed according to a standardized protocol using a calibrated probe. This technique will be as described by Burkhart. In his cadaveric study, he showed that the bare spot is almost exactly in the center of the inferior glenoid, with very small standard deviations. A calibrated probe is introduced through the posterior portal, and the distance from the bare spot to the posterior glenoid rim is measured. The surgeon infers that the pre-dislocation intact measurement from the bare spot to the anterior glenoid rim would have been the same as the posterior measurement. The surgeon then uses the calibrated probe to measure the distance from the bare spot to the anterior glenoid rim. Hill-Sachs lesions will also be measured and documented with photography. Intraoperative photos and video will be recorded of this technique, and the photographs will be reviewed by an independent observer. The measurement will be quantified by the independent observer to remove any observer bias. Any bone loss is identified, quantified, and compared with the pre-operative CT and MRI.

Bankart Repair Arthroscopic Bankart repair procedures will be performed according to each individual surgeon's usual technique. Procedures will be performed with the patient in the lateral or beach-chair position. Repairs for associated or conjoined superior labral anterior-to-posterior (SLAP) tears will be documented and performed at the surgeon's discretion. Labral detachments will be repaired with the use of suture-anchor fixation and arthroscopic tying techniques. Either two or three suture anchors will be used. Capsular redundancy will be addressed with arthroscopic suture plication at the surgeon's discretion. Surgeons will mobilize the capsulolabral tissue as deemed necessary. Surgical time and video of the operation will be recorded, and photographs will be taken documenting any bone loss.

Anatomic Glenoid Reconstruction The surgical technique was the lateral decubitus all-arthroscopic anatomic glenoid reconstruction procedure for treatment of anterior shoulder instability as described by Wong et al. (In press) The procedure is done in a semi-lateral decubitus position that assists with optimal graft placement on the native glenoid. The investigators utilize the cannulated Bristow-Latarjet Instability Shoulder System (Depuy-Mitek, MA, USA). The surgical technique is identical to that of arthroscopic Bankart repair with one additional step. Prior to insertion of anchors, one additional medal portal is created for insertion of the bone graft. The distal tibia allograft is prepared; the cannulated guide is attached and advanced through the rotator interval and secured with two cannulated screws. Finally the Bankart repair is performed above the graft. Surgical time and video of the operation will be recorded, and photographs will be taken documenting any bone loss.

Postoperative protocol

Patients in both groups will follow a standardized rehabilitation protocol as described by The American Society of Shoulder and Elbow Therapists' Consensus Rehabilitation Guideline for Arthroscopic Anterior Capsulolabral Repair of the Shoulder. Briefly, the protocol includes a 4-week period of absolute immobilization, a staged recovery of full range of motion over a 3-month period, a strengthening progression beginning at postoperative week 6, and a functional progression for return to athletic or demanding work activities between postoperative months 4 and 6. The patient will follow-up with the attending surgeon at 2 weeks post-operatively for a wound check. Then further follow-up appointments will be at 6 weeks, 3 months, 6 months, 12 and 24 months, as per current clinical practice.

At the 3, 6, 12, and 24 month visits the patient will again undergo the structured clinical examination conducted by a physiotherapist who is blinded to the patients' treatment group. The patient will be wearing clothing to cover all incisions, and will be given clear instructions to not reveal to the assessor which group they were allocated to. This will be monitored by the research assistant. The patient will also complete the WOSI, VR-12, Marx Shoulder Activity Scale, and DASH questionnaires at each of these follow-up appointments. Complications, and subluxation/dislocation events will be documented at each follow-up evaluation. On the basis of a clinical examination and patient history, the surgeon will diagnose recurrent instability and categorize it as a traumatic or atraumatic subluxation or dislocation. Safety monitoring will be in place to document and review all adverse events.

Analysis of Data:

Two independent assessors blinded to the surgical treatment group will view the intra-operative photos of measurement of glenoid size taken by each surgeon. The videos will not be associated with any PHI and only identified by unique coding number. The distance from the bare area to the anterior margin of the glenoid will be subtracted from the distance from the bare area to the posterior margin of the glenoid. The resultant value will be identified as the absolute anterior glenoid bone loss, and will be compared to the total diameter of the glenoid to estimate percent of glenoid bone loss. Summary statistics for measured bone loss will be recorded as mean +/- standard deviation, 95% confidence intervals will be calculated. The measured bone loss will be compared to pre-operative diagnostic imaging estimates via independent samples t-tests.

The outcome measures for each group will be reported as mean+/- standard deviation and 95% confidence interval will be calculated. Two-Tailed T-tests will be computed for the WOSI, VR-12, Marx Shoulder Activity Scale, and DASH scores, range of motion, and strength measures comparing the 2 experimental groups at each time point. Repeated-measures ANOVA will be used to perform a within-subjects analysis to assess differences in WOSI, VR-12, DASH, Marx Shoulder Activity Scale scores, range of motion and strength within each group between time points.