Comparison of Self-Reported Knee Injury and Osteoarthritis Outcome Score to Performance Measures in Patients After Total Knee Arthroplasty

Objective To characterize patient outcomes after total knee arthroplasty (TKA) by (1) examining changes in self-report measures (Knee Injury and Osteoarthritis Outcome Score [KOOS]) and performance measures over the first 6 months after TKA, (2) evaluating correlations between changes in KOOS self-report function (activities of daily living [ADL] subscale) and functional performance (6-minute walk [6MW]), and (3) exploring how changes in pain correlate with KOOS ADL and 6MW outcomes. Design Retrospective cohort evaluation. Setting Clinical research laboratory. Patients (or Participants) Thirty-nine patients scheduled for a unilateral, primary TKA for end-stage unilateral knee osteoarthritis. Methods Patients were evaluated 2 weeks before surgery and 1, 3, and 6 months after surgery. Main Outcome Measurements KOOS, 6MW, timed-up-and-go (TUG), and stair climbing tests (SCT), quadriceps strength. Results Three of 5 KOOS subscales significantly improved by 1 month after TKA. All 5 KOOS subscales significantly improved by 3 and 6 months after TKA. In contrast, performance measures (6MW, TUG, SCT, and quadriceps strength) all significantly declined from preoperative values by 1 month after TKA and significantly improved from preoperative values by 3 and 6 months after TKA; yet, improvements from preoperative values were not clinically meaningful. Pearson correlations between changes in the KOOS ADL subscale and 6MW from before surgery were not statistically significant at 1, 3, or 6 months after TKA. In addition, KOOS Pain was strongly correlated with KOOS ADL scores at all times, but KOOS Pain was not correlated with 6MW distance at any time. Conclusions Patient self-report by using the KOOS did not reflect the magnitude of performance deficits present after surgery, especially 1 month after TKA. Self-report KOOS outcomes closely paralleled pain relief after surgery, whereas performance measures were not correlated with pain. These results emphasize the importance of including performance measures when tracking recovery after TKA as opposed to solely relying on self-reported measures.