The diagnosis of Lyme borreliosis, except in cases characterised by pathognomonic clinical manifestation, usually requires confirmation by means of microbiological diagnostic assay, mainly by antibody detection methods. In our study antibodies to B. burgdorferi were tested in neurological patients with suspected Lyme borreliosis, depending on syndrome and clinical diagnosis. Antibodies were tested with IFT, ELISA and immunoblot. Blood samples of patients tested with IFT and ELISA tests were positive in 88 patients. Positive indirect immunofluorescence tests were found in 83 patients; in 5 patients the antibody level was borderline. Of these, 40 were positive also in ELISA but a correlation between IF titers and ELISA-positivity was not established. The immunoblot method confirmed specific antibody positivity in 36 of 88 patients (45.45%) who were positive (or borderline positive) in the indirect IF test, and in 28 of 40 (70%) ELISA-positive patients. Antibody specificity was found in 8 indirect IF-positive patients who were ELISA negative. This may be explained by the higher immunoblot sensitivity in comparison with ELISA. The Lyme borreliosis diagnosis was clinically established in 19 patients; antibodies to B. burgdorferi were only found in 13 patients in all three tests, and in 4 patients only in the indirect IF test. The results of serological tests for antibodies to B. burgdorferi should be interpreted with caution, as the tests are not standardized and may show false positive or false negative results. A two-step serological examination with the immunoblot test is recommended, whereby some nonspecific reactions may be eliminated. The results of serological tests have only supportive value and cannot be deemed conclusive when establishing an etiological diagnosis.