Anne A Smartt, Ryan R Wilbur, Bryant M Song, Aaron J Krych, Kelechi Okoroha, Jonathan D Barlow, Christopher L Camp
Purpose: To compare the clinical outcomes of operative and nonoperative management, identify risk factors for recurrent instability, and identify risk factors for progression to surgery after failed nonoperative management for patients with first-time anterior shoulder dislocation after the age of 50 years.
Methods: An established geographic medical record system was used to identify patients who experienced a first-time anterior shoulder dislocation after the age of 50 years. Patient medical records were reviewed to identify treatment decisions and outcomes of interest, including rates of frozen shoulder and nerve palsy, progression to osteoarthritis, recurrent instability, and progression to surgery. Outcomes were evaluated using Chi-square tests and survivorship curves were generated using Kaplan-Meier methods. A Cox model was developed to evaluate for potential risk factors of recurrent instability and progression to surgery after an initial trail of at least 3 months of nonoperative treatment.
Results: 179 patients were included with a mean follow-up of 11 years. 14% ( = 26) underwent early surgery within 3 months and 86% ( = 153) were initially treated nonoperatively. Mean age (59 years), was similar for both groups, but those that underwent early surgery had an increased rate of full-thickness rotator cuff tears (82% vs 55%; = .01), labral tears (24% vs 8.0%; = .01), and humeral head fracture (23% vs 8.5%; = .03). When comparing the early surgery group to the nonoperative group, there were similar rates of persistent moderate-severe pain (19% vs 17%; = .78) and frozen shoulder (8 vs 9%, respectively; = .87) at final follow-up. Although nerve palsy (19% vs 8%; = .08) and progression to osteoarthritis (20% vs 14%; = .40) were more common in surgical patients, they experienced lower rates of recurrent instability after surgical intervention (0% vs 15%; = .03) compared to nonoperatively treated patients. Increasing number of instability events prior to presentation was the greatest risk factor for recurrent instability (HR 232; < .01). Fourteen percent ( = 21) failed initial nonoperative treatment and proceeded to surgical intervention at an average of 4.6 years after the initial instability event, and the greatest risk factors for progression to surgery were recurrent instability (HR 3.41; < .01).
Conclusions: Although the majority of patients >50 years that experience ASI are treated nonoperatively, those that require surgery tend to have more significant injury pathology, a lower risk of recurrent instability after surgery, but a higher progression to osteoarthritis compared to patients that do not require surgical intervention. There was no difference in pain severity at final follow-up, rates of frozen shoulder or nerve palsy between patients who underwent initial nonoperative treatment after instability and those who underwent surgery. A history of multiple instability episodes prior to presentation was the greatest predictor of recurrent instability and failure of nonoperative treatment and progression to surgery.
Level of Evidence: Level III, retrospective cohort study.
