Summary

The critical size of the Hill-Sachs lesion which may cause recurrence after an arthroscopic Bankart repair may be lower than previously reported in our series of patients.

Abstract

Aims

The absolute size of bony deficiencies which preclude an arthroscopic Bankart repair for recurrent shoulder dislocation vary. The aim of this study is to describe our centre’s experience with arthroscopic Bankart repair and the influence of Hill-Sachs lesions and glenoid deficiencies on functional outcome.

Methods

IRB approval was obtained. 107 patients who underwent arthroscopic Bankart repair alone from 2008-2013 with minimum 2 years follow-up were reviewed by an independent observer. Patients who underwent Remplissage or bone grafting were excluded. 82 patients consented to interview and scoring with the Oxford Instability Score (OIS). Pre-operative Magnetic Resonance Imaging (MRI) scans of the shoulders were reviewed measuring Hill-Sachs lesion size with the modified Cetik method and glenoid defect size with the Gyftopoulos circle method.

Results

The mean age at first dislocation was 19.4±3.4 (12.0 to 31.0) years. Duration of follow-up was 3.6±1.3 (2.2 to 8.3) years. 2.6±3.0 (0.1 to 15.3) years elapsed from first dislocation to surgery. 42/82 (51.2%) played overhead or contact sports and 44/82 (53.7%) played competitive sports before injury.
8/82 (9.8%) reported recurrence of dislocation. 5/82 (6.1%) underwent revision surgery. 23/82 (28.0%) report mild instability after surgery. 48/82 (58.5%) returned to playing sports after surgery. 74/82 (90.2%) are satisfied with their surgery. 79/82 (96.3%) are willing to undergo the same surgery again.
All patients had glenoid defects on MRI with a mean size of 17.0±7.0 (6.0 to 35.0) %. 68 (82.9%) had Hill-Sachs lesions on MRI with measured defect width of 11.7±3.5 (5.5 to 19.3) mm, depth 5.5±19.3 (1 to 7.1) mm occupying 9.9±3.5 (0.7 to 18.9) % of the humeral head. There was no significant correlation between number of dislocations and Hill-Sachs size (p<0.29). The MRI derived size of the glenoid deficiency was associated with recurrence of dislocation (23.0±7.0 vs 17.0±6.0%, p<0.006). The width and percentage of humeral head involvement of the Hill-Sachs lesion on MRI too was associated with recurrence (14.4±3.1 vs 11.4±3.4 mm, p<0.019 and 12.3±3.5 vs 9.6±3.4%, p<0.044 respectively).
Recurrence of dislocation is not associated with atraumatic dislocation, presence of Hill-Sachs lesion, ligamentous laxity, contact or overhead sports, bony Bankart lesion on arthroscopy, SLAP lesion, number of dislocation, age of dislocation, completion of physiotherapy, return to sports, patient satisfaction, competitive sports prior to injury but is associated with operative time (86.3±44.8 vs 63.3±28.2 min, p<0.043).
74/82 (90.2%) have a 2 year Good/Excellent OIS which is not associated with number of dislocations before surgery, return to sports, age of dislocation, ligamentous laxity, operative time, absence of recurrence, completing physiotherapy, playing competitive sports before surgery and size of glenoid defects and Hill-Sachs lesion on MRI.
Good/Excellent OIS is significantly associated with self-reported stability after surgery (p<0.036), satisfaction with surgery (p<0.028) and willingness to undergo surgery again (p<0.024).

Conclusion

Arthroscopic Bankart repair produces good functional outcome and high patient satisfaction, even though not all return to sports. The critical size of a Hill-Sachs lesion which may cause recurrence of dislocation after an arthroscopic Bankart repair alone was lower than previously reported in our series.