A Comparison of the Responsiveness of 4 Commonly Used Patient-Reported Outcome Instruments at 5 Years After Matrix-Induced Autologous Chondrocyte Implantation

Abstract

Background:

Patient-reported outcome (PRO) instruments are employed to assess outcomes after matrix-induced autologous chondrocyte implantation (MACI), although the PRO most responsive to change after surgery remains unknown.

Purpose:

To compare the responsiveness of 4 commonly used PRO instruments at 5 years after MACI.

Study Design:

Cohort study (diagnosis); Level of evidence, 2.

Methods:

The Knee Injury and Osteoarthritis Outcome Score (KOOS), the Lysholm score, the Tegner activity scale, and the 36-item Short Form Health Survey (SF-36) were administered to 104 patients before and at 5 years after MACI knee surgery. The Self-administered Patient Satisfaction Scale was employed at 5 years to investigate each patient’s overall level of satisfaction as well as satisfaction with relieving pain and improving the ability to perform daily activities, partake in recreational activities, and participate in sport. The effect size (ES) and standardized response mean (SRM) were used to compare PRO responsiveness. A receiver operating characteristic (ROC) curve analysis was performed to evaluate the extent to which PRO changes were associated with satisfaction. The minimal clinically important difference (MCID) according to the criterion of patient satisfaction was identified as the PRO instrument change score maximizing classification accuracy.

Results:

The most responsive PRO measures were the KOOS sport/recreation (ES, 1.63; SRM, 1.43) and quality of life (QOL) (ES, 1.37; SRM, 1.18) subscales. The least responsive were the SF-36 mental component summary (MCS) (ES, 0.38; SRM, 0.40) and the Tegner activity scale (ES, 0.91; SRM, 0.59). Of the 104 patients, 54 (51.9%) reported being “very satisfied,” 38 (36.5%) “somewhat satisfied,” 8 (7.7%) “somewhat dissatisfied,” and 4 (3.9%) “very dissatisfied.” A ROC curve analysis was performed using “very satisfied” as the responder criterion. The strongest association was between the change in KOOS sport/recreation with satisfaction in improving the ability to perform recreational activities (area under the curve, 0.756; 95% confidence interval, 0.663-0.849), and the change score maximizing prediction accuracy (MCID) was 40 (sensitivity, 69%; specificity, 76%).

Conclusion:

The KOOS sport/recreation and QOL subscales were the most responsive PRO measures and were most predictive of satisfaction. This information will provide a guide as to the improvements required in pertinent PRO measures to produce a satisfied patient, while allowing researchers to better structure trials in these patients using the most relevant PRO instruments.