Abstract
Objectives:
Nonoperative management of posterior shoulder instability is common, however there is limited data available to assess the pathomorphology of nonoperative management. The purpose of this study is to evaluate glenohumeral pathomorphology in shoulders with posterior glenohumeral instability treated nonoperatively.
Methods:
We conducted a retrospective review of a consecutive series of patient with isolated posterior shoulder instability defined as an isolated posterior labral tear with a corresponding positive Jerk or Kim test. Patients were excluded if they had prior shoulder surgery or absence of a Jerk or Kim test. Non-operative management was defined as a trial of formal physical therapy pursuit of nonsurgical modalities for a minimum 6 months. Patients who underwent non-operative management and subsequently had a repeat MRI of the initially injured shoulder were identified, and the two studies were compared to evaluate for changes in glenoid bone loss, glenoid morphology, cartilage injuries, and the presence of concurrent pathology. Our primary outcome was glenoid changes associated with failure of non-operative management, which was defined as reoperation and/or medical separation from the military due to the injured shoulder. Secondary outcomes included evaluation of potential risk factors for failure of non-operative management including glenoid bone loss, glenoid version, and posterior humeral head subluxation. Continuous variables were compared with student’s t-test or Fisher exact test when appropriate. Categorical variables were analyzed using Chi-squared. Multivariable regression analysis was used to evaluate risk factors for failure.
Results:
42/90 (46.7%) patients failed a 6-month trial of nonoperative management after being diagnosed with posterior glenohumeral instability and went onto receive an arthroscopic stabilization procedure. The failure group demonstrated a significantly greater humeral head subluxation ratio than the cohort of patients who survived nonoperative management (0.65 ± 0.2 vs 0.62 ± 0.2; p = 0.0375). Of those who failed nonoperative management only 17 had repeat MRI’s for comparison with initial MRI’s, which revealed a significantly greater increase in glenoid bone loss (6.54 ± 1.59 vs 2.68 ±1.71; p = 0.00274). The mean time from index MRI and repeat MRI was 488 days (95% CI 317 to 658).
Conclusions:
In patients that underwent 6-months of nonoperative management for isolated posterior glenohumeral instability, failure occurred approximately 47% of the time and was associated with a greater posterior humeral head subluxation ratio on index MRI than those who did not fail. Additionally, those who had repeat MRI on average 1.3 years later demonstrated greater glenoid bone loss when compared to the index MRI.