Patient-Reported Outcome Measures in the Repair of Chondral Defects: How Well Do the Scores Relate to Each Other?

Abstract

Introduction: In the repair of focal chondral defects, there are several patient-reported outcome measures (PROMs) that are used to assess the patient's well-being. However, the question remains as to how well one scoring system relates to another, which may restrict the comparison of results from different studies. Therefore, we examined the strength of correlations between the Lysholm and KOOS scores.

Materials and methods: The data for this analysis was obtained from the Autologous Matrix-Induced Chondrogenesis (AMIC®; Geistlich Pharma AG, Wolhusen, Switzerland) knee registry, which is an ongoing, multicentre database designed to record changes over time in knee function and symptoms. This is done using the Lysholm score, the Visual Analogue Scale (VAS) for pain, and the five domains of the Knee injury and Osteoarthritis Outcome Score (KOOS). All patients had preoperative and postoperative scores at one-year follow up. The results were evaluated using the Spearman's rank correlation test.

Results: We identified 79 patients in the registry, all of whom were treated by the co-authors and had preoperative scores and postoperative scores at one year for the Lysholm, VAS, and the KOOS domains. The Lysholm score demonstrated a significant correlation (p<0.0001) to all KOOS domains. The correlation coefficients were 0.81, 0.82, 0.83, 0.84, and 0.76 for the KOOS domains of symptoms, pain, activities of daily living (ADL), quality of life (QoL), and Sport, respectively. The correlation between VAS pain and the KOOS domain for pain was significant (p<0.0001) but notably lower, with a correlation coefficient of 0.71.

Conclusion: Our data provides evidence that the outcome of the Lysholm knee score is strongly correlated with the KOOS scores, with the KOOS domains of ADL and pain exhibiting the highest correlation. Thus, it may be possible, through formulae calculations, to predict a KOOS score from the Lysholm score. With regard to assessment of outcomes over larger numbers of studies, the pooling of substantially more data could facilitate the conduct of systematic reviews and meta-analyses pertaining to the surgical treatment of chondral injuries of the knee.