Risk of Revision Shoulder Arthroplasty After Anatomic and Reverse Total Shoulder Arthroplasty.

INTRODUCTION

The objective of this study was to determine the survivorship of anatomic total shoulder arthroplasty (aTSA) and reverse TSA (rTSA) over a medium-term follow-up in a large population-based sample and to identify potential risk factors for revision surgery.

METHODS

The State Inpatient Database from the Healthcare Cost and Utilization Project was used to identify patients who underwent aTSA or rTSA from 2011 through 2015 using ICD9 codes. We modeled the primary outcome of time to revision or arthroplasty using the Cox proportional hazards model. The predictors of revision surgery in the model include aTSA versus rTSA, indication for surgery, age, sex, race, urban versus rural residence, hospital length of stay zip code-based income quartile classification, and Elixhauser comorbidity readmission score.

RESULTS

Among 43,990 patients in this study, 1,141 (4.0%) underwent revision or implant removal over the 4-year study period. The median age was 71 years, and 57% of patients were female. Indications for the index surgery included primary osteoarthritis (75.2%), cuff tear (8.5%), acute fracture (7.0%), malunion/nonunion (1.4%), and other (7.8%). Among these indications for surgery, the risk of revision or removal was greatest in patients who underwent the primary procedure for malunion/nonunion (hazard ratio [HR] 2.39, 95% confidence interval [CI] 1.69 to 3.39) compared with the reference of primary osteoarthritis. Male patients who underwent aTSA were less likely to need revision surgery than male patients who underwent rTSA (HR: 0.59, 95% CI 0.49 to 0.71), and the opposite relationship was observed in female patients (HR: 1.41, 95% CI 1.18 to 1.69). Age, length of stay, and Elixhauser comorbidity score were predictive of revision surgery (P < 0.0001, P = 0.0005, P < 0.0001, respectively), whereas race, urban versus rural, and zip code-based income quartile were not.

DISCUSSION: aTSA and rTSA showed excellent 4-year survivorship of 96.0% in a large population-based sample. aTSA and rTSA survivorships were similar at the 4-year follow-up.