Abstract
PURPOSE: Medial unicompartmental knee arthroplasty (UKA) is a treatment option for medial knee osteoarthritis, with an increase in surgeries over the last few years. However, the results of revision total knee arthroplasty (TKA) after a UKA vary greatly. The purpose of the study was to examine the survival after revision TKA of a failed UKA. METHODS: This is a retrospective single-centre analysis that includes 35 revision TKA procedures after the failed UKA performed from 2004 to 2019. The median follow-up after revision TKA was 39 months (interquartile range [IQR]: 32-52). The indication for revision of the UKA was aseptic loosening in 49% of patients (17/35). We evaluated demographic factors, reason for revision and revision implant used with descriptive statistics. Implant survival analysis with a focus on re-revision-free survival and potential re-revisions was performed using Kaplan-Meier survival curves. Differences in survival were analyzed using the log-rank test. p Value was set at 0.05. RESULTS: Forty per cent (14/35) of revision implants were posterior stabilized revision TKA, followed by 34% (12/35) of condylar constrained designs and 23% of rotating hinged TKA (8/35). Only one patient was revised to a cruciate retaining primary implant (3%). The re-revision-free survival after revision TKA amounted to 94% (95% confidence interval [CI]: 91%-100%) after 1 year, 80% (95% CI: 67%-93%) after 2 years and 74% (95% CI: 56%-90%) at 5 years. Twenty-three per cent of patients (8/35) underwent re-revision after the initial UKA revision after a median time period of 21 months (IQR: 12-24). The reasons for repeat revision were tibial aseptic loosening in 9% of patients (3/35), periprosthetic joint infection (PJI) in 9% of patients (3/35) and instability in 5% (2/35). Rotating hinge knee implants showed reduced survivorship. CONCLUSIONS: Revision of medial UKA is associated with an increased use of more elaborate and complex revision implants. There is a substantial risk of repeat revision, with aseptic tibial loosening and PJI being the main reasons for the failure of this series. LEVEL OF EVIDENCE: Level III.