Shoulder arthroplasty for locked anterior shoulder dislocation: a role for the reversed design

Introduction Locked anterior shoulder dislocation (LASD) is an uncommon condition associated with bone, articular cartilage, and soft tissue damage. In selected cases, shoulder arthroplasty (SA) may be the best treatment. The purpose of this study was to assess outcomes of SA for LASD. Materials and methods Between 1976 and 2013, 19 SAs [three hemiarthroplasties (HA), seven total shoulder arthroplasties (TSA), and nine reverse shoulder arthroplasties (RSA), mean age 62 years] were performed for LASD. Shoulders were followed for at least two years (range, 2-30 years, mean 7.1). Clinical and radiographic outcomes were studied. Results/discussion Three SAs required re-operation, two TSAs for early redislocation and one HA for late, painful glenoid arthrosis. Four additional shoulders (two TSA, two HA) were unstable at most recent follow-up. Pain improved from 4.7 to 2.2 (p < 0.0001) out of 5, elevation from 51 to 94 degrees (p = 0.004), and external rotation from 1 to 34 degrees (p = 0.01). There were two excellent, seven satisfactory, and ten unsatisfactory modified Neer ratings. Compared to TSA/HA, RSA experienced fewer re-operations (0 vs. 3, hazard ratio, 2.03*10^-9, p = 0.0844) and instability (0 vs. 6, p = 0.0108). Similar post-operative pain (p = 0.2192), range of motion (p = 0.2432-0.5299), strength (p = 0.2099), satisfaction (p = 0.6563), outcomes scores (p = 0.0683-0.0933), and complication rate (p = 0.3698) were seen with RSA vs. TSA/HA. Conclusions RSA for the treatment of chronic LASD provides greater pain relief, and improvement in range of motion (ROM) compared to TSA/HA. Anatomic SA is associated with a high rate of instability not seen with RSA. Therefore, anatomic SA (TSA/HA) is likely not indicated in these difficult circumstances. Level of Evidence: IV