ARTHROSCOPIC BANKART REPAIR FOR ADOLESCENT UNIDIRECTIONAL SHOULDER INSTABILITY: CLINICAL AND RADIOGRAPHIC PREDICTORS OF REVISION SURGERY AND INSTABILITY

INTRODUCTION: The treatment of adolescent shoulder instability can be a challenging problem, with multiple patient and radiographic risk factors for recurrent instability. Although glenoid bone loss has been well described, humeral bone loss has gained more recent attention. The purpose of this study is to evaluate the incidence and clinical outcomes of “off track” shoulder lesions and their association with clinical outcomes. METHODS: A retrospective IRB approved study was performed to identify patients less than 19 years of age treated with isolated arthroscopic anterior labral repair for unidirectional shoulder instability. Radiographic measurements of glenoid diameter, % glenoid bone loss, glenoid track, hill-sachs interval (HIS), HS/glenoid track ratio, and intact anterior articular angle (IAAA) were performed for all patients with magnetic resonance imaging (MRI). All patients were contacted at final follow-up to collect outcome scores (PASS and SANE scores, activity level). A multivariable logistic regression analysis was performed to identify predictors of revision surgery or subjective instability (RI). RESULTS: 86 patients were identified to meet inclusion criteria and 69 of these patients, 53 males and 16 females with a median age of 16 years [15, 17], had minimum 1-year clinical follow-up and were included in assessment of clinical outcomes and multivariable analysis. 12 patients (17%) had revision surgery and 10 patients had subjective instability without revision (14%) In univariate analysis (Table 1), patient height was the only patient factor which was significantly different between the revision/instability (RI) cohort and the non-revision/instability (NRI) cohort. All 86 patients were included in radiographic analysis. Intra- and inter-rater reliability (ICCs) for radiographic measurements are presented in table 2. Agreement was good and excellent for all measurements with the exception of inter-rater reliability of IAAA. Multivariable regression analysis demonstrated that HS/glenoid track ratio was not a predictor for RI. PASS and SANE scores at final follow-up were significantly lower in the RI cohort (79.5 and 70) respectively) as compared to the NRI cohort (94 and 90 respectively), p<0.001. 42 patients (69%) returned to the same or higher level of sports following surgery and this was no different between RI and NRI cohorts. CONCLUSIONS: 31% of adolescent patients in our cohort experienced RI following arthroscopic bankart repair and this was associated with inferior PASS and SANE scores. Off-track lesions were not predictive of failure with primary repair.