Shoulder periprosthetic joint infection is associated with increased mortality

Abstract

Background: Periprosthetic joint infections (PJIs) are a catastrophic complication after shoulder arthroplasty and may be associated with an increased mortality risk, as seen in hip and knee arthroplasty. Shoulder PJI organisms differ from hip and knee infections, as lower-virulence organisms, such as Cutibacterium acnes, are more commonly encountered. This study evaluated the association between shoulder PJI and mortality.

Methods: We retrospectively identified 411 patients who underwent revision shoulder arthroplasty from 2007 to 2020 at a single institution. 2018 International Consensus Meeting on Musculoskeletal Infection criteria were used to categorize each case as definite, probable, possible, or unlikely PJI. Mortality rate was assessed by performing chart reviews and an obituary search. Revision cases were grouped into a septic cohort (definite and probable PJI) and an aseptic cohort (possible and unlikely PJI). Kaplan-Meier analyses were performed to compare survival between septic and aseptic groups. The log-rank test was used to compare cumulative survival distributions and survival rates at 90 days, 6 months, 1 year, 2 years, and 5 years. Demographic information, Charlson Comorbidity Index (CCI), and culture data were collected. PJI organisms were categorized as virulent or nonvirulent per an infectious disease specialist, and predictors of mortality were determined by performing stepwise logistic regression analyses.

Results: The overall mortality rate was significantly greater (P < .001) in the septic group (20.5%) than in the aseptic group (6.6%). When evaluating time from revision surgery to death, patients with PJI had significantly greater mortality compared with those undergoing aseptic revision at 2 years (7.7% vs. 2.1%, P = .01) and 5 years (17% vs. 5.1%, P < .001). Body mass index, CCI, race, sex, and age were not significantly different between groups. Groups differed in utilization of 2-staged procedures (65% septic, 9% aseptic, P < .001). Multivariate regression analysis found that the variables most associated with mortality were septic revision, 2-staged procedures, and CCI. In the septic group, patients with C acnes PJI had a significantly lower mortality rate when compared to PJI from any other offending organism (3.1% vs. 48.4%, P = .001).

Conclusion: Revision shoulder arthroplasty in the setting of PJI not only carries severe functional consequences for patients but is also associated with an increased mortality risk. Furthermore, C acnes infections are associated with a more favorable survival profile than PJI resulting from other organisms. Continued efforts to decrease shoulder arthroplasty infection rates are warranted and may influence long-term survival.

Keywords: Periprosthetic joint infection; aseptic shoulder infection; revision shoulder arthroplasty; septic shoulder infection; shoulder arthroplasty mortality.