Validation of the Knee Injury and Osteoarthritis Outcome Score (KOOS) for the treatment of focal cartilage lesions

Abstract

Objective: To validate the Knee Injury and Osteoarthritis Outcome Score (KOOS) for the treatment of focal cartilage lesions.

Methods: A total of 40 patients (mean age 35+/-12 years), treated for a focal cartilage lesion in the knee were included in this study. Test-retest data were collected with an intermediate period of 2 days. Patients were asked to complete the Dutch KOOS and complementary questionnaires [short form-36 (SF-36), Lysholm, EuroQol-5D (EQ-5D)] to evaluate the clinimetric properties of the KOOS in terms of internal consistency (Cronbach's alpha), reliability [intra-class-correlation (ICC) and Bland and Altman plots], construct validity (Spearman's rank correlation), floor and ceiling effects and responsiveness.

Results: The Cronbach's alpha of the KOOS subdomains and total score ranged from 0.74 to 0.96. The overall ICC of the KOOS was 0.97 while the subscales ranged from 0.87 to 0.95. The Bland and Altman plots showed a small individual variance between the two assessments in time. Spearman's rank correlations between the subscales of the KOOS and representative subscales of the SF-36, Lysholm and EQ-5D were high to moderate ranging from 0.43 to 0.70. We observed no floor effect while the largest observed ceiling effect was 10.3%. The responsiveness was moderate to large with the effect size ranging from 0.70 to 1.32 and the standardized response mean 0.61 to 0.87.

Conclusion: This study illustrates the validity and reliability of the KOOS in measuring the clinical condition of patients after treatment of focal cartilage lesions. This study provides a basis for the use of the KOOS for future clinical research in cartilage repair.