RESUMO
OBJECTIVE: The detection of autoantibodies directed toward nuclear antigens is one of the main criteria for the diagnosis of systemic lupus erythematosus (SLE), for which the most commonly used techniques are the enzyme immunoassay and immunofluorescence assay (IFA). However, the sensitivity and specificity of these tests vary between different techniques. Thus, in this study, we aimed to determine the superior method for detecting antinuclear antibodies (ANAs) and compare the accuracy of tests ordered by rheumatologists versus non-rheumatologists. MATERIALS AND METHODS: We compared the sensitivity and specificity of the two assays in 149 patients from a non-selected population, who were sent to the immunology laboratory of King Abdulaziz Medical City, Jeddah from 2018 to 2019. RESULTS: The sensitivity and specificity of the indirect IFA were 77.78 % and 58.65%, respectively. The positive and negative predictive values of IFA for SLE were 44.87% and 85.92%, respectively. The sensitivity and specificity of the enzyme-linked immunosorbent assay (ELISA) were 77.78% and 80.77%, respectively. The negative and positive predictive values of ELISA for SLE were 63.64% and 89.36%, respectively. The highest number of false-positive IFA tests was requested by family physicians and the lowest was requested by rheumatologists. CONCLUSION: Our data show that IFA has a higher negative predictive value, while ELISA has a higher positive predictive value. The positive predictive value of the test can be improved by pre-selecting patients by specialist rheumatologists.