Abstract
Objectives:
Shoulder instability is a common cause of shoulder pain and dysfunction, particularly in young and active individuals. While anterior instability remains the most common type, recent literature shows that posterior instability occurs more frequently than previously thought. As opposed to its anterior counterpart, posterior instability most often presents with pain, particularly in active individuals involved in dynamic posterior loading of the shoulder resulting in repetitive microtrauma of the posterior capsulolabral structures. While arthroscopic stabilization for posterior glenohumeral instability has shown excellent success preventing recurrent instability and allowing return to sport, eventual progression to glenohumeral arthritis remains a concern in these patients. However, the rate of and risk factors for arthritis post-surgery in young and high demand populations have not been described in the literature and may be important in guiding decision making when treating the young patient with posterior shoulder instability. The purpose of this study was to evaluate the rate of progression to glenohumeral arthritis and identify potential risk factors after arthroscopic posterior stabilization in a young and high demand population.
Methods:
This study included 110 active duty servicemembers identified in the Military Heath System (MHS) with ICD-9/ICD-10 posterior shoulder instability who underwent primary arthroscopic surgical stabilization and had postoperative imaging or medical records available over a twelve-year period between January 2004 and September 2016. All procedures were performed at a single institution. Patients were excluded if the they had previously undergone a stabilization procedure on the shoulder of interest, if no pre-operative imaging was available, if they did not have a minimum follow-up of four years, or if operative information regarding the number of anchors used was unavailable. Presence of arthritis (yes/no) was defined over follow-up using radiographic parameters as described by Samielson and Prieto identifying patients with at least ‘mild arthritis’. Kaplan-Meier survival curves were estimated for development of arthritis and compared by patient characteristics using log-rank tests. Cox proportional hazard models were used to calculate Hazard Ratios (HR) with 95% confidence intervals (95% CI) associated with patient characteristics as predictors of the development of glenohumeral arthritis.
Results:
Among the 110 patients with posterior shoulder instability requiring surgical fixation, 12% (13/110) developed glenohumeral arthritis. The mean age of all patients was 23.9 years (SD 6.71). The median time to diagnosis of arthritis was 7 years and the median follow-up time was 8 years (IQR 5;11). Kaplan-Meier curves showed differences in time to arthritis among patient groups stratified by age and index surgery anchor number (log rank p for each <0.05). Hazard ratios (95% CIs) estimated from separate Cox regression models of the development of glenohumeral arthritis were .344 (1.3262 to 8.431) for age in years (per SD) and 2.8280 (1.10577 to 7.233) for surgery anchor number.
Conclusions:
No previous study has reported the incidence and risk factors for glenohumeral arthritis after arthroscopic surgical stabilization for posterior shoulder instability. In the current study, generally young and high demand patient population, the rate of progression to glenohumeral arthritis after posterior stabilization surgery was 12%. Within this cohort, older age and higher number of anchors used were associated with progression to arthritis. The incidence of arthritis over follow up post surgery was lower than incidences previously described in older populations, a finding which may advocate for early surgical intervention for a young patient presenting with posterior shoulder instability.