Radiographic grading for knee osteoarthritis. A revised scheme that relates to alignment and deformity.

Abstract

OBJECTIVE: To develop a radiographic grading scheme for osteoarthritis of the knee that would relate to arthritic progression and deformity. METHODS: The scheme of Scott, et al was revised to include new fields of Tibial Erosion and Subluxation; the fields of Tibial Osteophytes and Sclerosis were removed. The worst affected compartment only was scored on frontal Standardized Knee Images, which were used to define knee alignment variables. RESULTS: The interreader reliability (kappa = 0.92) and correlation for total scores (r = 0.94) were both excellent (p < 0.001). Individual field scores and total knee scores both correlated well with most limb alignment variables, especially the hip-knee-ankle angle (r = 0.77, p < 0.001). CONCLUSION: These findings encourage further evaluations for outcome measurement, diagnostic sensitivity, and sensitivity to define arthritic change over time.