The rose bengal test is often used for rapid diagnosis of human brucellosis in endemic areas. However, autoantibodies have never been investigated as a reason for false-positive or false-negative results. Therefore, the aim of this study was to show the effect of autoantibody detection on the rapid diagnosis of human brucellosis in an endemic area. The study included 2 groups: antinuclear antibody (ANA)-positive and ANA-negative groups. Diagnosis of brucellosis was established by isolation of Brucella spp. from blood culture. The overall sensitivity and specificity of the rose bengal test were 100 % and 90.8 %, respectively. The specificity (100 % versus 89 %) and positive predictive value of the test (100 % versus 8 %) fell markedly from the ANA-negative to the ANA-positive group. As a conclusion, this study verified our suspicion about the effect of autoantibodies on rose bengal test results to the diagnosis of human brucellosis. However, to have definite decisions, extensive studies with larger populations are needed. (c) 2007 Elsevier Inc. All rights reserved.