Evaluation of serological diagnostic tests for human Brucellosis in an endemic area

Objectives: The clinical utility of complementary tests for brucellosis are not clear in many situation. This study aimed to evaluate value of these tests for brucellosis in an endemic area in Turkey. Materials and methods: This study was performed at Canakkale General Hospital in 2009. In a retrospective approach, records of the patients who evaluated for brucellosis were collected. During the study period, 236 people (131 symptomatic and 105 non-symptomatic) were evaluated for diagnosis of brucellosis. All of the samples from these patients were tested for Brucella antibody seropositivity by RB slide agglutination, standard serum agglutination, Brucella Coombs, BrucellaCapt, and ELISA IgG and IgM tests. Results: In total, 49 symptomatic patients were hospitalized and blood cultures were obtained. Brucella spp. were isolated from nine of them (18.4%).The BrucellaCapt test was found to be the most sensitive for Brucella (74.0%) and close behind it was the Coombs test (72.5%). The sensitivity for the RB test was 48.1%. The ELISA IgG test was found more sensitive for brucellosis than the ELISA IgM test was (65.6% and 49.6%, respectively). All examined tests were found about 100% specific for brucellosis but the RB test was found less specific than the others were (96.1%) Positive predictive value for all tests was about 1 but negative predictive values were only valuable for the Coombs and Brucella Capt test (0.744 and 0.755, respectively). The other serological tests were around and below 0.50, which was weak for negative results. Conclusions: The ELISA IgG and IgM tests were no superior to the other tests. By assessment of receiver operating characteristics (ROC) analysis, the Brucella Coombs and BrucellaCapt tests were found to be the most valuable tests for serological diagnosis of brucellosis in endemic areas. The seronegative tests in the symptomatic patients should be evaluated and repeated in short time. J Microbiol Infect Dis 2012; 2(2): 50-56