Bone Fragment Union and Remodeling After Arthroscopic Bony Bankart Repair for Traumatic Anterior Shoulder Instability With a Glenoid Defect

Abstract

Background: Although good clinical outcomes have been reported after arthroscopic bony Bankart repair, the extent of bone union is still unclear. Purpose: To investigate bone union after arthroscopic bony Bankart repair and its influence on postoperative recurrence of instability. Study Design: Cohort study; Level of evidence, 3. Methods: Among 113 consecutive shoulders that underwent arthroscopic bony Bankart repair, postoperative evaluation of bone union by computed tomography (CT) was performed at various times in 81 shoulders. Bone union was investigated during 3 periods: 3 to 6 months postoperatively (first period), 7 to 12 months postoperatively (second period), and 13 months or more postoperatively (third period). The influence of the size of the preoperative glenoid defect and the size of the bone fragment on bone union was investigated, as well as the influence of bone union on postoperative recurrence of instability. In shoulders with bone union, bone fragment remodeling and changes in the glenoid defect size were also investigated. Results: The bone union rate was 30.5% in the first period, 55.3% in the second period, and 84.6% in the third period. Among 53 shoulders with CT evaluation in the second period or later and follow-up for a minimum of 1 year, there was complete union in 33 shoulders (62.3%), partial union in 3 (5.7%), nonunion in 8 (15.1%), and no fragment on CT in 9 (17.0%). The complete union rate was 50% for 22 shoulders with small bone fragments (<5% of the glenoid diameter), 56.3% for 16 shoulders with medium fragments (5%-10%), and 86.7% for 15 shoulders with large fragments (>10%). The recurrence rate for postoperative instability was only 6.1% for shoulders with complete union, while it was 50% for shoulders with partial union, nonunion, no fragment, and no fragment on CT. The recurrence rate was significantly higher (36.4%) in shoulders with small fragments, but it was significantly lower in shoulders with bone union. In shoulders with bone union, the bone fragment frequently became larger over time, while the size of the glenoid defect decreased significantly from 18.6% preoperatively to 4.7% postoperatively. Conclusion: Bone union was not always achieved after arthroscopic bony Bankart repair, and union was often delayed. Recurrence of instability was significantly more frequent when bone union failed. The size of the glenoid defect decreased significantly in shoulders with bone union.