Treatment and Outcome of Periprosthetic Joint Infection in Unicompartmental Knee Arthroplasty

Abstract

Background: Periprosthetic joint infection (PJI) after unicompartmental knee arthroplasty (UKA) is a devastating but poorly understood complication, with a paucity of published data regarding treatment and outcomes. This study analyzes the largest cohort of UKA PJIs to date comparing treatment outcome, septic and aseptic reoperation rates, and risk factors for treatment failure.

Methods: Twenty-one UKAs in 21 patients treated for PJI, as defined by Musculoskeletal Infection Society criteria, were retrospectively reviewed. Minimum and mean follow-up was 1 and 3.5 years, respectively. Fourteen (67%) patients had acute postoperative PJIs. Surgical treatment included 16 debridement, antibiotics, and implant retentions (DAIRs) (76%), 4 two-stage revisions (19%), and 1 one-stage revision (5%). Twenty (95%) PJIs were culture positive with Staphylococcus species identified in 15 cases (71%).

Results: Survivorship free from reoperation for infection at 1 year was 76% (95% confidence interval, 58%-93%). Overall survival from all-cause reoperation was 57% (95% confidence interval, 27%-87%) at 5 years. Two additional patients (10%) underwent aseptic revision total knee arthroplasty for lateral compartment degeneration 1 year after DAIR and tibial aseptic loosening 2.5 years after 2-stage revision. All patients who initially failed PJI UKA treatment presented with acute postoperative PJIs (5 of 14; 36%).

Conclusion: Survivorship free from persistent PJI at 1 year is low at 76% but is consistent with similar reports of DAIRs for total knee arthroplasties. Furthermore, there is low survivorship free from all-cause reoperation of 71% and 57% at 2 and 5 years, respectively. Surgeons should be aware of these poorer outcomes and consider treating UKA PJI early and aggressively.

Keywords: 1-stage exchange; 2-stage exchange; debridement; periprosthetic joint infection; total knee arthroplasty; unicompartmental knee arthroplasty.