High failure rates and poor quality of life after modular non-fusion knee arthrodesis following failed total knee arthroplasty: a multicenter retrospective study of 85 patients.

backgroundModular non-fusion knee arthrodesis (MKA) represents a limb-salvage option for patients with failed total knee arthroplasty (TKA) when further reconstruction is no longer feasible, most commonly due to periprosthetic joint infection (PJI) or aseptic failures. However, evidence regarding implant survival, infection control, functional outcomes, and quality of life remains limited and heterogeneous. This multicenter study aimed to evaluate clinical outcomes after MKA and identify risk factors for treatment failure. MATERIAL AND

methodsThis retrospective multicenter cohort study included 85 adult patients from two centers who underwent MKA after failed TKA between 2003 and 2024, with a minimum follow-up of 12 months. The primary outcome was the overall implant survival, defined as no replacement of the MKA (total or partial) nor conversion to above-the-knee amputation (AKA). Secondary outcomes included all-cause revision-free and infection-free survival (following the Delphi-consensus criteria from 2013), amputation rate, functional outcome assessed through patient-reported outcome measures (PROM) by the Lower Extremity Function Score (LEFS), and quality of life measured using the EQ-5D-3L. Survival analyses were performed using Kaplan-Meier analysis and Cox regression.

resultsSixty out of eighty-five patients (70.6%) retained their modular arthrodesis until the latest follow-up, with patients requiring replacement of their MKA (17/85) or undergoing AKA (8/85) during follow-up. The median (IQR) follow-up was 35.0 (46.0) months. There was no difference in overall implant survival between septic and aseptic indications (71.2% vs. 69.2%; p = 0.855). Two-stage revision was associated with an improved implant survival proportion (90.9%) compared with multi-stage (67.6%; p = 0.042) and single-stage revision (57.7%; p = 0.010). Exploratory threshold analysis suggested that patients with five or more previous surgeries experienced substantially higher failure rates. Aseptic indications, particularly painful TKA without structural failure, were associated with significantly higher amputation rates than septic cases. Among patients available for PROM assessment, functional outcomes remained poor, with a mean LEFS of 21.6/80, and quality of life was substantially reduced (mean EQ-5D-3L index 0.612), with no significant differences across subgroups.

conclusionMKA following failed TKA is associated with only moderate implant survival and poor functional outcomes and quality of life. Especially painful TKAs carry an unexpectedly high risk of subsequent amputation, while extensive prior surgical history is a major predictor of failure. Two-stage revision was associated with improved implant survival; however, given the retrospective design and potential selection bias, causal conclusions cannot be drawn. Careful patient selection and thorough preoperative counseling remain essential. Further studies should focus on the comparison of MKA with AKA, as well as with additional revision total knee arthroplasty when feasible.