Arthroscopic Glenoid Bone Grafting With Nonrigid Fixation for Anterior Shoulder Instability

Abstract

Background: The healing rate and clinical outcomes of glenoid bone grafting with nonrigid fixation for patients with recurrent anterior shoulder instability are unknown. Hypothesis: Glenoid bone grafting with nonrigid fixation can yield satisfactory results for patients with recurrent anterior shoulder instability with regard to graft healing and the restoration of shoulder stability. Study Design: Case series; Level of evidence, 4. Methods: A total of 52 patients with recurrent anterior shoulder instability underwent Bankart or bony Bankart repair as well as arthroscopic glenoid bone grafting. Allogenic bicortical iliac grafts were used. Instead of firm fixation, the grafts were tethered to the glenoid by sutures from anchors placed in the glenoid surface. Follow-up occurred at 3, 6, 12, and 24 months. Computed tomography and magnetic resonance imaging examinations were performed immediately after surgery and at each follow-up visit to evaluate the healing of the graft and the changes in the repaired capsule-labrum structure. Functional evaluations were taken at 24 months with the Oxford Shoulder Instability Score and the Rowe Score for Shoulder Instability. For the patients who underwent surgery 2.5 years earlier, an additional review was conducted to evaluate the latest stability status of the shoulder. Failure was defined as recurrence of dislocation or instability. Results: The mean follow-up time was 39 months (range, 24-64 months). In all cases, the grafts healed to the glenoid at 3 or 6 months, and glenoid remodeling was complete within 12 months; in most cases, a robust bone-capsule structure formed on the anterior side of the glenoid. The glenoid defect area changed from 32.7% ± 8.7% (range, 10.7% to 53.9%) to −16.3% ± 3.3% (range, −26.7% to 5.9%), and the glenoid defect width changed from 28.3% ± 8.7% (range, 10.4% to 54.5%) to −16.9% ± 7.3% (range, −33.4% to 2.8%). Compared with the presumed normal glenoid, the final glenoid surface area increased in 94.2% of patients and final glenoid width increased in 96.2% of patients. One patient experienced redislocation and 2 experienced a sense of instability without dislocation, which resulted in a failure rate of 5.8%. Six patients exhibited slight pain. The Oxford score improved from 29.7 ± 5.6 preoperatively to 42.4 ± 3.3 at 2 years postoperatively, and the Rowe score improved from 34.7 ± 6.1 preoperatively to 91.8 ± 2.8 at 2 years postoperatively. Conclusion: In this study, arthroscopic glenoid bone grafting with nonrigid fixation in combination with Bankart repair resulted in 100% graft healing and the satisfactory restoration of shoulder stability.