[Comparison of rheumatoid test procedures--value and critical interpretation of sensitivity and specificity and their effect on pre-test and post-test probability].

In this prospective study, sera of 440 patients with rheumatic and degenerative joint diseases were tested for the presence of rheumatoid factor (RF). The Latex agglutination test (LFT), Waaler-Rose hemagglutination, laser nephelometry and IgM-Enzyme immunoassay (IgM-EIA) were used for detecting IgM-rheumatoid factors. In addition, rheumatoid factor of IgA isotype was measured by an IgA-Enzyme immunoassay. Sensitivity, specificity, pre-test- and post-test-probability were evaluated based on the data obtained to compare the test systems used. Under prospective patient selection, none of the test systems used reached a sensitivity of 100% concerning its cut off level. Despite this limitation, latex agglutination and IgM-EIA reached the highest sensitivity. Waaler-Rose test (90,8%) showed the best result for specificity. The IgA-EIA held the third position in sensitivity, specificity and efficiency. By comparing sensitivity with specificity, no test system can be recognized as the absolutely best one, since the receiver operating characteristic curves (ROC) overlapped. Practically rheumatoid factor measurement should initially use a highly sensitive assay, such as LFT and IgM-EIA to screen for RF. In the case of a positive result a more specific assay should be used, for example laser nephelometry, to confirm the result.