Evaluation of radiologic criteria in rheumatoid arthritis
Henrard, J.C.; Schoen, E.; Verret, J.M.
Revue du Rhumatisme et des Maladies Osteo-Articulaires 44(2): 105-114
1977
ISSN/ISBN: 0035-2659 PMID: 847360 Document Number: 111836
The authors compared the hand Xrays of 53 patients with rheumatoid arthritis (AR) with those of 53 control subjects matched for age and s. x. Each AR patient conformed to the New York clinical criteria. Assesment was carried out on Xrays of both hands, search being made for erosions, geodes, and joint narrowing, the severity being graduated from 0 to 4, according to data from the international Atlas of Radiology. The sensitivity, sepcificity and severity of each of these abnormalities was studied joint by joint (18 for the hand). Analysis of the results shows that study of all 18 joints in the hand is not useful; account may be taken only of the first three metacarpo-phalangeal joints, the carpo-metacarpal joints and the radio-carpal joint. The proximal interphalangeal joints, contrary to the most commonly held opinion, are more a source of errors than of diagnosis. Erosion is the most specific sign, especially if one is only considering the characteristic sites. With a specificity of the order of 98 per cent, this abnormality has sufficient weight to counteract the very low incidence of the disease in a population in comparison with degenerative disorders. Geodes should be studied more by their severity than by their frequency; this is high in the controls, which diminishes their specificity (45 for the wrist, 62 for the first carpo-metacarpal, and 75 for the first metacarpo-phalangeal joint). Joint narrowing is a difficult sign to read and its value is all at the radiocarpal and carpal joints.