Revision Shoulder Arthroplasty: Predictors of Subsequent Revision Surgery and Economic Burden amongst Medicare Beneficiaries

Abstract

Revision shoulder arthroplasty continues to add an increasing burden on patients and the healthcare system. This study aimed to delineate long-term shoulder arthroplasty revision incidence, quantify associated Medicare spending, and identify relevant predictors of both revision and spending. The complete 2016-2022(Q3) Medicare fee-for-service inpatient and outpatient claims data was analyzed. Patients receiving a primary total shoulder arthroplasty for osteoarthritis, rotator cuff pathology, or inflammatory arthropathy were included and subsequent ipsilateral revision surgeries were identified. The time to revision was modeled using the Prentice, Williams, and Peterson Gap Time Model. Medicare spending within 90 days post-discharge was modeled using a generalized linear model. The analysis was subdivided by index procedure type: anatomic total shoulder arthroplasty (TSA) and reverse shoulder arthroplasty (RSA). A total of 82,949 primary TSAs and 172,524 RSAs were identified. Compared to index TSA cases, index RSA cases had a lower first revision rate in an observation window of nearly 7 years (1.9% vs. 3.5%, p<0.001), but a higher rate of second (11.4% vs. 4.9%, p<0.001) as well as third revision (13.8% vs. 13.8%, p=0.449). TSA spending was significantly lower than RSA spending for the index procedure ($21,531 vs. $23,267, p<0.001), first ($23,096 vs. $26,414, p<0.001), and second ($25,060 vs. $29,983, p<0.001) revision. There was no statistically significant difference in third revision between TSA and RSA groups ($31,313 vs. $30,829, p=0.860). Age, sex, race, and rheumatoid arthritis were among the top predictors of revisions. Top predictors of Medicare spending included having a non-osteoarthritis surgical indication, a hospital stay of three or more days, a discharge to a setting other than home, malnutrition, dementia, stroke, major kidney diseases, and being operated on in a teaching hospital. Compared with TSA, RSA was associated with a lower first revision rate, but a higher subsequent revision rate. An index RSA procedure was also associated with higher initial Medicare spending as well as subsequent revision surgery spending compared with an index TSA procedure. Demographics and comorbid medical conditions were among the top predictors of revisions, while procedure-related factors predicted Medicare spending.