Abstract
Purpose: The purpose of this study was to determine psychometric properties of the Lysholm score and Tegner activity scale as patient-administered outcomes scores for anterior cruciate ligament injuries of the knee. We hypothesized that these two outcomes measures remain valid by today’s standards twenty-five years after they were originally introduced as physician-administered outcomes instruments.
Methods: One thousand seven hundred eighty-three (1783) patients were included in this study. There were 749 females and 1034 males. Average age was 37 years (range, 18 to 77). Isolated ACL tears were documented in 593 patients, and 1190 patients had concurrent injuries including meniscus pathology and/or cartilage damage. Patients with multiple ligamentous injuries were excluded. All patients in this study were diagnosed with an ACL tear at arthroscopy. For responsiveness, scores were measured preoperatively and at a minimum of two years postoperatively. For test-retest, scores were measured at a minimum of two years postoperatively and again within four weeks of the original postoperative questionnaire. For criterion validity, patients completed the short form (SF-12®) of the health related quality-of-life scale and the IKDC score in addition to Lysholm and Tegner instruments. For all other analyses, preoperative Lysholm score or Tegner activity levels were used.
Results: There was acceptable test-retest reliability for overall Lysholm score (ICC=0.94 [95% confidence interval=0.88 to 0.96]) and Tegner (ICC=0.82 [95% confidence interval=0.66 to 0.89]). The minimum detectable change for Lysholm was 8.9 and for Tegner was 1.4. The Lysholm score demonstrated acceptable internal consistency (Cronbach’s alpha=0.72). The Lysholm score correlated with IKDC (r=0.78) and the physical function domain of SF-12® (r=0.43). The Tegner scale correlated with physical function domain of SF-12® (r=0.2) and IKDC (r=0.22). Both scores had acceptable floor and ceiling effects, and all hypotheses were significant. The Lysholm score and Tegner scale both had a large overall effect size. There were no differences between isolated and combined ACL injuries.
Conclusion: After 25 years of changes in treatment of ACL injuries and postoperative rehabilitation protocols, the Lysholm knee score and Tegner activity scale continue to demonstrate acceptable psychometric parameters. The Lysholm score and Tegner scale both had acceptable test-retest reliability, construct validity, criterion validity, content validity, and responsiveness when patient-administered similar to the physician-administered results when they were originally validated. Our hypothesis was affirmed.
Correspondence should be addressed to: EFORT Central Office, Technoparkstrasse 1, CH – 8005 Zürich, Switzerland. Email: office@efort.org
