Inter- and Intraobserver Reliabilities of Four Different Radiographic Grading Scales of Osteoarthritis of the Knee Joint

摘要

The purpose of this study was to investigate the interobserver and intraobserver reliabilities of four different radiographic grading scale of osteoarthritis (OA) of the knee joint, namely, Kellgren-Lawrence (KL), Ahlback, Brandt, and OA Research Society International (OARSI). One consultant orthopaedic surgeon and one rheumatologist who were familiar with these OA grading scales participated in the study. Radiological assessments were performed in random order by each observer on two separate occasions, at least 2 weeks apart. K-statistics were used to establish a relative level of agreement between the observers for the two readings and between separate readings by the same observer. A total of 140 patients (53 males and 87 females) with a mean age of 61.2 years (range, 50-72 years) were included in the study. Intraobserver reliability for KL was substantial for observer A (kappa: 0.753) and moderate for observer B (kappa: 0.573). Interobserver reliability for KL was moderate for both observers (kappa: 0.499 and 0.458, respectively). Intraobserver reliability for Ahlback was substantial for observer A (kappa: 0.768) and moderate for observer B (kappa: 0.561). Interobserver reliability for Ahlback was fair for both observers (kappa: 0.365 and 0.204, respectively). Intraobserver reliability for Brandt was substantial for observer A (kappa: 0.741) and moderate (kappa: 0.425) for observer B. Interobserver reliability for Brandt was fair for both observers (kappa: 0.308 and.: 0.246, respectively). Intraobserver reliability for OARSI was substantial for observer A (kappa: 0.792) and moderate for observer B (kappa: 0.508). Interobserver reliability for OARSI was moderate for observer A (kappa: 0.425) and slight for observer B (kappa: 0.175). None of the studied OA grading scales showed acceptable reliability (kappa 0.80). The evaluation of patients with OA should not be dependent on radiographic findings alone; clinical findings should also guide the treatment and follow-up.