Evaluation and comparison of fluorescence polarization assay with three of the currently used serological tests in diagnosis of human brucellosis.

Fluorescence polarization assay (FPA) is a method that has been used for the diagnosis of brucellosis in animals for many years. To test its possible usefulness for the diagnosis of human brucellosis, 230 sera from patients with clinical signs of brucellosis and positive serological tests (Rose Bengal, Standard Agglutination Test, iELISA), and 305 sera from a healthy population with no clinical/epidemiological/serological evidence were examined with FPA. By using ROC analysis, the cut-off value was estimated at 99 mP, with 93.5% sensitivity (95% CI 89.5-96.3) and 96.1% specificity (95% CI 93.2-97.9). The pairwise comparison of ROC curves between FPA and iELISA and between FPA and RBT revealed no significant statistic difference (P < 0.05). On the contrary it revealed a significant statistic difference between FPA and SAT (P > 0.05). SAT also had the lowest sensitivity (81.7%) among the three tests used in case definition while iELISA had a sensitivity of 90.8% and RBT a sensitivity of 88.7%. The Kappa analysis showed that FPA has a very good agreement (0.92) with the "status of the disease" and with iELISA (0.837). According to our results, FPA seems to be a valuable method for the diagnosis of brucellosis in humans. Taking into consideration the advantages of the method such as the speed of results obtaining, the objectivity of results interpretation, as well as the cost, FPA could be considered as a replacement for other established methods. However, further studies are needed to assess the reproducibility of FPA.