Defining thresholds for the Kellgren-Lawrence grading between 2 and 3 for the Japanese knee osteoarthritis measure questionnaire for urban inhabitants more than 65-year old

Abstract

Purpose: According to the population-based cohort study in Japan, over 80 percent of 80-year old and above Japanese women have had a prevalence rate of knee osteoarthritis (OA). Knee OA is one of the most prevalent conditions resulting in disability particularly in the elder population, and it also leads to high economic costs and death further. To assess knee OA, Kellgren-Lawrence (KL) grading has been commonly used in many epidemiological studies, and the KL grade 2 is usually used as the disease threshold. Prevent knee joint degeneration progressing from KL grade 2 to KL grade 3 in healthy individuals may reduce high economic costs by receiving knee replacement surgery or interventional therapy, ensure elder inhabitant’s quality of life further. In addition, patient-reported outcome measures (PROMs) of the Japanese Knee Osteoarthritis Measure (JKOM) with its high reliability and validity, has been widely used to evaluate OA-related health domain. Physicians can assess knee OA with the KL grading and JKOM objectively. However, what JKOM scores indicating the process of knee OA were unclear. The purpose of this present study was to determine the KL grading thresholds between KL grade 2 and KL grade 3 for the JKOM for urban inhabitants more than 65-year old. Methods: Participants who completed the JKOM questionnaire, received X-ray and diagnosed as KL grade 2 or KL grade 3 of knee OA by the expert physician were included in this study. To determine the score that best distinguished those who were in KL grade 2 and who were in KL grade 3, the sensitivities and specificities for each of the potential scores were plotted as a receiver operating characteristic (ROC) curve. The maximal product of specificity and sensitivity was taken to be the KL grade 2-3 threshold. Results: A total of 754 out of 795 participants (mean age, SD= 73.0, 5.5; male vs female = 321 vs 433) were included in this study. Of these, 645 participants (85.5%, male vs female = 299 vs 346) were in KL grade 2 with the mean JKOM score (SD) of 5.6 (6.8) points, which was significantly lower than KL grade 3 group (mean score, SD= 11.9, 11.5, P value < 0.0001). In addition, mean score of the subscale of JKOM between KL grade 2 and KL grade 3 were also statistically significant (P value < 0.05). The KL 2-3 threshold (sensitivity, specificity) was 8 (0.54, 0.78) for the JKOM total score, 8 (0.57, 0.75) for the degree of knee pain (JKOM I), 1 (0.73, 0.62) for the pain and stiffness in knees (JKOM II), 4 (0.19, 0.89) for the condition in daily life (JKOM III), 1 (0.70, 0.62) for the general activities (JKOM IV), and 2 (0.50, 0.70) for the health conditions (JKOM V). Conclusions: Our results should provide additional information to interpretation what score indicates the progressing of knee joint degeneration in the healthy population.