Evaluation of specificity of indirect enzyme-linked immunosorbent assay for diagnosis of human Q fever.

Ninety-five acute- and convalescent-phase serum specimens from 48 patients suspected of having rickettsial or Legionella infections were assayed for antibodies to Coxiella burnetii, the causative agent of Q fever. To evaluate the specificity of the indirect enzyme-linked immunosorbent assay (ELISA) for human Q fever, we compared the ELISA results with those of the indirect immunofluorescence antibody (IFA) test. The ELISA data were analyzed by two different criteria for a positive test. The first criterion for positive results by ELISA was based upon diagnostic titers established in a study of 150 subjects who had no demonstrable cellular or humoral immune responses to C. burnetii phase I or phase II whole cells or phase I lipopolysaccharide. The second criterion was based upon diagnostic antibody titers in a study of 51 subjects who had been diagnosed as having clinical Q fever and had fourfold or greater rises in humoral immune responses to C. burnetii phase I and phase II whole-cell antigens. A comparison of the ELISA and IFA test results of the 95 serum specimens indicated excellent agreement between the tests (Kappa = 92.9%; P < 0.05). None of the 38 patients whose etiologies were confirmed serologically as Legionnaires' disease or rickettsial diseases other than Q fever were classified as positive for C. burnetii by the ELISA. Only one patient identified by the IFA test as having Q fever was not scored positive by the ELISA. These results suggest that the ELISA is useful for epidemiologic screening and as a diagnostic test for human Q fever.