New radiographic-based surrogate outcome measures for osteoarthritis of the knee

Abstract

Objective: Minimum joint space width (mJSW) between the femoral condyle and tibial plateau is currently the principal radiographic outcome measure for knee osteoarthritis (OA). While mJSW measurement has been proven effective, there is a need for quantitative measures with greater sensitivity to OA so studies of disease modifying trials can be performed more quickly and with fewer subjects. Also, mJSW is a single outcome measure that does not reflect all disease changes. Here we present a study of new radiograph-based outcome measures to quantify OA progression.

Design: Our goal is made possible by a software algorithm that delineates both the edge of the tibial plateau and the femoral condyle on digitized knee radiographs. We also developed a new coordinate system based on anatomical landmarks to facilitate measurement of the outcome measures. We studied measurements of joint space width (JSW) and average JSW at fixed locations along the joint interface. The outcome measures were tested using duplicate acquisitions made from 16 OA and 18 normal knees. Reproducibility was quantified by the root-mean square standard deviation and the coefficient of variation.

Results: We found that the reproducibility for all outcome measures was better than or similar to that for mJSW. Despite a slightly higher reproducibility error for some new outcome measures compared to mJSW, they are potentially valuable since they reflect independent disease changes.

Conclusion: We have demonstrated new reproducible radiographic outcome measures to quantify OA progression with the potential to be an improvement over standard minimum joint space width.