Total Knee Arthroplasty in Post-Traumatic Knee Osteoarthritis Shows Higher Failure Rates Depending on Implant Constraint: An Evaluation of the German Arthroplasty Register.

Abstract

Total knee arthroplasty (TKA) performed for post-traumatic osteoarthritis (PTOA) is associated with higher complication and revision rates than TKA for primary osteoarthritis (OA). This study aimed to compare implant survival and identify risk factors for revision in TKA performed for PTOA versus OA using registry data. The TKA procedures from 2012 to 2022 in the German Arthroplasty Register (EPRD) were analyzed (n = 289,382). The incidence of PTOA was 2.4% (n = 6,982). Implants were categorized as unconstrained or constrained (varus-valgus stabilized or hinge). Kaplan-Meier survival analysis and multivariable Cox proportional hazards models were used to assess implant survival, adjusted revision risk, and independent risk factors for revision. A total of 10,341 complications were recorded, with PTOA showing a higher complication rate than OA (5.7 versus 3.5%; P < 0.001). Aseptic revision rates were higher for unconstrained TKA in PTOA (P < 0.001), but not for constrained TKA (P = 0.1), whereas septic revision rates were higher for PTOA with both implant designs (both P < 0.001). The most common reasons for revision were infection (19%) and instability (8%). After adjustment for age, sex, body mass index, weighted Elixhauser score, fixation method, patellar resurfacing, and hospital case volume, a 45% higher risk of revision was observed after TKA for PTOA compared with OA. For unconstrained TKA in PTOA, the revision risk was 36% higher compared with OA (hazard ratio = 1.36; P < 0.001). Independent risk factors for revision in unconstrained TKA for PTOA included hospital volumes ≤ 200 annual cases and TKA without patellar resurfacing. An increased adjusted risk of revision was observed after TKA for PTOA compared with OA. While aseptic revision rates varied by implant constraint, septic revision rates were higher in PTOA independent of constraint. Lower hospital volume and the absence of patellar resurfacing in unconstrained implants further increased revision risk, whereas higher implant constraint was protective regarding the risk of aseptic revision.