Rheumatology 2000; 39: 1218-1221
© 2000 British Society for Rheumatology
Detecting radiographic knee osteoarthritis: what combination of views is optimal?
Boston University Arthritis Center,
1 Department of Radiology at the Boston VA Medical Center, Boston, MA,
2 Center for Health Outcomes and Economic Research of the HSR&D Field Program, VAMC Bedford, MA, and Boston University School of Public Health, Boston, MA, USA
Objective. The failure to image the patellofemoral joint or the posterior knee compartment when evaluating persons for knee osteoarthritis may result in missed cases. While the skyline view has been recommended due to more reproducible assessment of the patellofemoral joint space, the lateral view may be easier to acquire and provides different information. We evaluated the sensitivity of different combinations of X-ray views (anteroposterior and lateral; anteroposterior and skyline; all three views) in 377 persons with knee symptoms who had all three views available and had a definite osteophyte on at least one view.
Results. Of the different views, skylines had to be excluded most often because the image of the patellofemoral joint was technically unsatisfactory. In the remaining knees, adding either a lateral or a skyline view to an anteroposterior view yielded roughly equal and high sensitivity (94–97%) when compared with the gold standard of a positive X-ray on any of the three views.
Conclusion. As long as at least an anteroposterior view and one image of the patellofemoral joint is obtained (either skyline or lateral), few cases with radiographic disease will be missed. For clinical or epidemiological studies the lateral view may be easier to acquire with high quality than the skyline view.
KEY WORDS: Knee, Osteoarthritis, Imaging, Epidemiological methods.
Correspondence to: D. T. Felson, Boston University, 715 Albany Street, Arthritis Center A-203, Boston, MA 02118, USA.