Abstract
BACKGROUND: Plain radiographs play a key role in the diagnosis of knee osteoarthritis (OA), with the Kellgren-Lawrence (KL) grading system being widely used to assess radiographic severity of the disease. Previous studies have suggested that radiographically less severe knee OA may be associated with inferior outcomes following total knee replacement (TKA). This study aimed to evaluate whether preoperative radiological findings are predictive of postoperative Oxford Knee Score (OKS) and residual knee pain after primary TKA. METHODS: We retrospectively included all patients who underwent TKA at our high-volume joint replacement center in 2018. The primary outcome was the OKS collected at one year postoperatively. The secondary outcome was patient-reported knee pain at one year. Preoperative plain radiographs were graded using the KL classification, and joint space height was measured. Additional covariates included age, gender, BMI, ASA score, and specific comorbidities. Multivariable analyses were performed, and model performance was assessed using R(2) and variable importance metrics. RESULTS: A total of 1,401 patients were included. Patients with less severe osteoarthritis had lower postoperative OKS compared to those with more advanced disease. Similarly, a greater joint space height was associated with smaller improvements in OKS following TKA. In multivariable models, the R(2) was 0.125 for postoperative OKS and 0.075 for residual pain. KL grade, joint space height, and mechanical axis were combined as an interaction term. CONCLUSION: In this large single-center cohort, comprehensive clinical and radiographic data were analyzed using advanced statistical methods. Our findings support previous evidence suggesting that comparatively less severe radiographic OA is associated with less favorable outcomes after TKA. However, the limited predictive capacity of the models highlights the multifactorial nature of postoperative outcomes and the challenges faced in forecasting patient-reported results following TKA.