Abstract
OBJECTIVE: The Lysholm score and the Tegner activity scale are widely used for assessing knee function and activity level after knee ligament injuries. This study aimed to asses validity and reliability of the German versions of the Lysholm score (Lysholm-G) and the Tegner activity scale (Tegner-G) in patients after total knee arthroplasty (TKA).
METHODS: Two patient groups (<3 months postoperative; 3 to 12 months postoperative) and a healthy control group completed both questionnaires twice within 3 to 7 days. On the first occasion, the SF-12 questionnaire was also administered. The instruments' absolute and relative reliability, internal consistency, construct validity (discriminative validity and exploratory factor analysis), criterion validity, and floor/ceiling effects were determined.
RESULTS: Both instruments showed acceptable relative reliability (Lysholm-G: ICC2,1 = 0.87; Tegner-G: ICC2,1 = 0.79), and the Lysholm-G had a Cronbach's alpha of 0.93. Kruskal-Wallis test and post-hoc Mann-Whitney U tests showed significant differences in both scores among the groups. Factor analysis of Lysholm-G yielded three factors that together explained 76% of the variance. Correlations between the Lysholm-G/Tegner-G and the physical component of the SF-12 were moderate (ρ = 0.60/0.59). Item analysis of the Lysholm-G revealed a low relative reliability for the item "instability" and low discriminative validity for the items "locking" and "instability".
CONCLUSIONS: This study demonstrated acceptable psychometric performances for the Lysholm-G and the Tegner-G scales as outcome measures for patients with TKA. A short version of the Lysholm score without the "locking" and "instability" items might be more appropriate for TKA patients.