Bayesian Validation of the Indirect Immunofluorescence Assay and Its Superiority to the Enzyme-Linked Immunosorbent Assay and the Complement Fixation Test for Detecting Antibodies against Coxiella burnetii in Goat Serum.

Abstract

Although many studies have reported the indirect immunofluorescence assay (IFA) to be more sensitive in detection of antibodies to Coxiella burnetii than the complement fixation test (CFT), the diagnostic sensitivity (DSe) and diagnostic specificity (DSp) of the assay have not been previously established for use in ruminants. This study aimed to validate the IFA by describing the optimization, selection of cutoff titers, repeatability, and reliability as well as the DSe and DSp of the assay. Bayesian latent class analysis was used to estimate diagnostic specifications in comparison with the CFT and the enzyme-linked immunosorbent assay (ELISA). The optimal cutoff dilution for screening for IgG and IgM antibodies in goat serum using the IFA was estimated to be 1:160. The IFA had good repeatability (>96.9% for IgG, >78.0% for IgM), and there was almost perfect agreement (Cohen's kappa > 0.80 for IgG) between the readings reported by two technicians for samples tested for IgG antibodies. The IFA had a higher DSe (94.8%; 95% confidence interval [CI], 80.3, 99.6) for the detection of IgG antibodies against C. burnetii than the ELISA (70.1%; 95% CI, 52.7, 91.0) and the CFT (29.8%; 95% CI, 17.0, 44.8). All three tests were highly specific for goat IgG antibodies. The IFA also had a higher DSe (88.8%; 95% CI, 58.2, 99.5) for detection of IgM antibodies than the ELISA (71.7%; 95% CI, 46.3, 92.8). These results underscore the better suitability of the IFA than of the CFT and ELISA for detection of IgG and IgM antibodies in goat serum and possibly in serum from other ruminants.