PREDICTIVE FACTORS FOR OUTCOMES AND PATIENT SATISFACTION 2-YEARS AFTER TOTAL KNEE ARTHROPLASTY

Objectives: There is a lack of strong correlation between functional outcomes and patient satisfaction following total knee arthroplasty (TKA). This study evaluated whether the change in these scores from pre- to post-operative periods predicted patient satisfaction 2 years after TKA. Methods: We evaluated data from 374 TKAs, assessing knee society score, Oxford knee score (OKS), and Short-Form 36 (SF-36) scores. A satisfaction questionnaire was completed, and logistic regression was performed. Receiver operating characteristic (ROC) curves assessed each scoring system’s ability to predict satisfaction at 2 years. Results: The ROC curve analysis between the physical component of SF-36 and the mean difference in pre-operative and post-operative scores showed that the area under the curve (AUC) was highest for the OKS (0.837, 95% confidence interval [CI]=0.708–0.967). The cutoff value was 12.5. At this threshold, the sensitivity and specificity of the OKS in predicting patient satisfaction were 80% and 77%, respectively. The AUC for Knee society knee score (KSKS) was 0.782 (95% CI=0.649–0.916). The cutoff value for the KSKS difference was 50.5. At this threshold, the sensitivity and specificity of KSKS for predicting patient satisfaction were 77.8% and 83%, respectively. The AUC for Knee society function score (KSFS) was 0.733 (95% CI=0.590–0.877). The cutoff value for the KSFS difference was 22.5. At this threshold, the sensitivity and specificity of KSKS for predicting patient satisfaction were 77.8% and 58.3%, respectively. Conclusion: The pre- to post-operative change in OKS more accurately predicts patient satisfaction at 2 years than KSKS (cut-off value: 50.5) and KSFS (cut-off value: 22.5), respectively.