Abstract
Objectives:
There are no studies that directly compare beach chair (BC) versus lateral decubitus (LD) position for anterior instability. In the only systematic review evaluating BC vs LD, bone loss is not accounted for in the recurrence rate. The purpose of this is to identify predictors of shoulder instability recurrence and revision after anterior shoulder stabilization surgery in a young, high demand population and evaluate surgical position and glenoid bone loss as independent predictors of the outcomes of interest, recurrence and revision.
Methods:
A consecutive series of 641 arthroscopic Bankart stabilizations were performed by sports-medicine certified and fellowship trained orthopaedic surgeons from 2009-2016 in either the BC or LD position. Patients were included if they underwent an isolated primary arthroscopic anterior capsulolabral repair. Patients were excluded if concomitant labral repair and/or Remplissage procedures were performed at the time of surgery. Shoulders were additionally excluded if magnetic resonance imaging (MRI) was not available at the time of preoperative evaluation or the patient was lost to follow-up. All shoulders were evaluated for glenohumeral bone loss using the perfect circle technique on the sagittal en-face MRI as well as for bipolar lesions according to the on/off-track method of Diagacomo et al. Glenoid bone loss was grouped into three categories: <5%, 5-13.5%, and >13.5%. The primary outcomes of interest were recurrent instability and revision stabilization. Recurrent instability was defined as the presence of a recurrent subluxation and or dislocation event and/or the presence of a positive apprehension. Multivariable logistic regression models were used to assess the relationships of outcomes with age, position, glenoid bone loss group, and, track.
Results:
641 shoulders with a mean age of 22.3 years (SD 4.46) underwent isolated arthroscopic Bankart repair and were followed for a mean 6 years. The overall recurrent instability and revision rates were 15.7% (101/641) and 11.5% (74/641). Recurrent instability was observed in 15.6% (24/154) and 15.9% (77/487) of LD and BC shoulders, respectively. After adjusting for confounders, multivariable logistic regression found no association between surgical position and recurrent instability (p=0.85). Age was an independent predictor for recurrence. The odds of recurrence were 1.58 times higher per 4.5 years (1 standard deviation) decrease in age (P<0.01, 95% CI 1.02 to 1.72). In a separate multivariable logistic regression model of revision surgery as the dependent variable, revision surgery was not associated with age, surgical position, glenoid bone loss group, or recurrence.
Conclusions:
Among fellowship-trained orthopaedic surgeons the overall failure of primary arthroscopic anterior shoulder stabilization was 15.7% in a high-demand population and equivalent outcomes may be anticipated with arthroscopic Bankart repair performed in the BC or LD position. In multivariable analysis, younger age, but not surgical position, was an independent risk factor for recurrence.