The Effect of Arthroscopic Bankart Repair on Anterior-posterior Glenohumeral Translation and Shoulder Proprioception in Patients With Traumatic Anterior Shoulder Instability: a Prospective Cohort Study

• Aim: To investigate the effect of arthroscopic Bankart repair on shoulder biomechanics and proprioception, and increase understanding of traumatic anterior shoulder instability.
• Main research questions: A (biomechanics): Does arthroscopic Bankart repair have a stabilising effect on the anterior-posterior glenohumeral translation in patients with traumatic anterior shoulder instability? B (proprioception): Does arthroscopic Bankart repair improve neuromuscular control in patients with traumatic anterior shoulder instability?
• Objectives and hypotheses research question A: In patients with traumatic anterior shoulder instability undergoing arthroscopic Bankart repair, to investigate the anterior-posterior glenohumeral translation and the scapular rotations before and six and twelve months after surgery and whether the ranges are restored to the same as the non-injured contralateral shoulder.

Hypotheses:

• Arthroscopic Bankart repair results in a ≥2.5 mm decrease in anterior-posterior glenohumeral translation, remaining both six and twelve months after surgery.
• Arthroscopic Bankart repair reduces anterior-posterior glenohumeral translation to the same range as measured in the non-injured shoulder (±2.5 mm).
• Arthroscopic Bankart repair reduces superior-inferior glenohumeral translation significantly, as measured six and twelve months after surgery.
• Scapular rotations and tilt remain unchanged after arthroscopic Bankart repair.
• Objectives and hypotheses research question B:

In patients with traumatic anterior shoulder instability undergoing arthroscopic Bankart repair, to investigate the neuromuscular control before and six and twelve months after surgery and whether the neuromuscular control is restored to the same level as the non-injured contralateral shoulder.

Hypotheses:

• Arthroscopic Bankart repair improves neuromuscular control, remaining both six and twelve months after surgery.
• Arthroscopic Bankart repair improves neuromuscular control to the same range as the non-injured shoulder.
• Other objectives:

In patients with traumatic anterior shoulder instability undergoing arthroscopic Bankart repair:

1. To investigate patient-reported outcome measures (PROM) before and six and twelve months after surgery.
2. To determine the recurrence rates (radiographically confirmed or manually reduced dislocation) in the first twelve months after surgery.
3. To investigate the shoulder range of motion (ROM) before and six and twelve months after surgery.
4. To assess the joint instability by manual testing before and six and twelve months after surgery.
5. To quantify potential bone loss before surgery.
6. To investigate if there are correlations between a) the shoulder biomechanics and b) the neuromuscular control, and PROM, ROM, and bone loss, respectively.

• Type of study:

Multi-clinical prospective cohort study including 30 patients. Patients' contralateral shoulders will be used as controls.

• Time schedule: Recruitment and inclusion began 1 April 2022 and is expected to last for 24 months. With a one-year follow-up the complete study period is expected to last three years.
• Set-up: Five sports orthopaedics clinics in the Capital Region of Denmark will recruit patients for the study. While the responsibility for treatment remains at the recruiting center, all study-related activities will take place at the Copenhagen University Hospital Hvidovre. Patients who have signed an informed consent form will undergo examinations prior to, as well as six and twelve months after surgery. All examinations will include biomechanical and proprioceptive testing, accompanied by a clinical shoulder examination and two validated electronic questionnaires regarding shoulder-related function and quality of life. A pre-interventional 3D computed tomography scan will be performed to evaluate bone loss. Study participation will not influence the treatment course.
• Sample size calculation:

The sample size calculation is made to allow for assessment of the primary outcomes within each category of outcomes; For research question A concerning biomechanics, the study is powered to detect a mean change in anterior-posterior glenohumeral translation of ≥2.5 mm with a standard deviation (SD) of 2.3 mm. For research question B concerning neuromuscular control, the study is powered to detect an effect size of 0.8 for change in reaction time. For the clinical and patient-reported outcomes, the study is powered to detect an effect size of 0.8 for the WOSI score.

For power of 90% and type I error rate of 0.017 (0.05/3), correcting for three tests, sample size for a one-sample t-test is 15, 24 and 24 for anterior-posterior glenohumeral translation, reaction time and WOSI score, respectively. As such, the largest sample size of 24 will be used. To account for expected dropout rate of 25% a total of 30 patients will be included.

Originally, the study was powered to detect a between group difference in anterior-posterior glenohumeral translation of ≥2.5 mm with a standard deviation (SD) of 2.3 mm, and a power of 80%. With six variables in the analysis (sex, height, BMI, dominant/non-dominant side affected, bone loss, clinical score) and an estimated 15% dropout rate, the calculation resulted in 55 patients. During the writing of a protocol article, it was realized that since the same shoulders are compared pre and post intervention no between groups analysis is performed and hence the setup controls for the six mentioned variables in itself. The sample size was thus re-calculated as stated above.