Prospective Evaluation of Posterior Glenoid Bone Loss After First-time and Recurrent Posterior Glenohumeral Instability Events

Background: Although posterior glenohumeral instability is becoming an increasingly recognized cause of shoulder pain, the role of posterior glenoid bone loss on outcomes remains incompletely understood. Purposes: To prospectively determine the amount of bone loss associated with posterior instability events and to determine predisposing factors based on preinstability imaging. Study Design: Cross-sectional study; Level of evidence, 3. Methods: A total of 1428 shoulders were evaluated prospectively for ≥4 years. At baseline, a subjective history of shoulder instability was ascertained for each patient, and bilateral noncontrast magnetic resonance imaging (MRI) scans of the shoulders were obtained regardless of any reported history of shoulder instability. The cohort was prospectively followed during the study period, and those who were diagnosed with posterior glenohumeral instability were identified. Postinjury MRI scans were obtained and compared with the screening MRI scans. Glenoid version, perfect-circle-based bone loss was measured for each patient's pre- and postinjury MRI scans using previously described methods. Results: Of the 1428 shoulders that were prospectively followed, 10 shoulders sustained a first-time posterior instability event and 3 shoulders sustained a recurrent posterior instability event. At baseline, 11 of 13 shoulders had some amount of glenoid dysplasia and/or bone loss. The change in glenoid bone loss was 5.4% along the axis of greatest loss (95% CI, 3.8%-7.0%; P = .009), 4.4% at the glenoid equator (95% CI, 2.7%-6.2%; P = .016), and 4.2% of total glenoid area (95% CI, 2.9%-5.3%; P = .002). Recurrent glenoid instability was associated with a greater amount of absolute bone loss along the axis of greatest loss compared with first-time instability (recurrent: 16.8% ± 1.1%; 95% CI, 14.6%-18.9%; first-time: 10.0% ± 1.5%; 95% CI, 7.0%-13.0%; P = .005). Baseline glenoid retroversion ≥10° was associated with a significantly greater percentage of bone loss along the axis of greatest loss (≥10° of retroversion: 13.5% ± 2.0%; 95% CI, 9.6%-17.4%; Conclusions: Posterior glenohumeral instability events were associated with glenoid bone loss of 5%. The amount of glenoid bone loss after a recurrent posterior glenohumeral instability event was greater than that after first-time instability. Glenoid retroversion ≥10° was associated with a greater amount of posterior glenoid bone loss after a posterior instability event.