Determination of the accuracy and optimal cut-off point for ELISA test in diagnosis of human brucellosis in Iran.

In endemic area the most challenging problem for brucellosis is to find a reliable diagnostic method. In this case-control study, we investigated the accuracy of ELISA test for diagnosis of human brucellosis and determined the optimal cut-off value for ELISA results in Iran. The laboratory diagnosis of brucellosis was performed by blood isolation of Brucella organism with a BACTEC 9240 system and/or detection of Brucella antibodies by standard agglutination test (titer ≥ 1:160). Serum level of ELISA IgG and ELISA IgM from 56 confirmed cases of brucellosis and 126 controls were compared with each other by Box plot graph and Receiver Operating Characteristic (ROC) curve. Box plot graphs showed the high degree of dispersion for IgG and IgM data in patients compared with all controls. We observed partially overlapping for IgM data (not for IgG) between cases and controls in graphs. The area under ROC curve for distinguishing between cases and controls was larger for IgG compared to IgM. Based on results of this study, ELISA IgG test was more reliable than ELISA IgM test in diagnosis of human brucellosis in Iran. Using a cut-off of 10 IU/ml and 50 IU/ml had most sensitivity (92.9%) and most specificity (100%) for ELISA IgG test, respectively