|Mckean, J - IOWA STATE UNIVERSITY|
|Leite, R - UNIV OF MINAS GERAIS|
Submitted to: Pig Veterinary Society International Congress Proceedings
Publication Type: Proceedings
Publication Acceptance Date: June 5, 2002
Publication Date: June 5, 2002
Citation: ROSTAGNO, M.H., HURD, H.S., GAILEY, J.K., MCKEAN, J.D., LEITE, R.C.COMPARATIVE EVALUATION OF SALMONELLA DETECTION ASSAYS ON SWINE FECAL SAMPLES. CONGRESS OF THE INTERNATIONAL PIG VETERINARY SOCIETY. 2002. P. 311. Technical Abstract: Epidemiological investigations require accurate means of discriminating between infected and non-infected animals, which is dependant on detecting evidence of the agent in clinical samples. This detection can be a major problem with Salmonella, as epidemiologic studies generally require testing large numbers of samples or pigs, and conventional bacteriological methods for its isolation and identification are labor intensive and time consuming. Several Salmonella detection assays, which use different principles for detection are commercially available. They were developed for use in the food industry. However, they may have application in preharvest food safety research. In this study, the performance of four commercially available Salmonella detection assays were comparatively evaluated for detection of Salmonella in swine fecal samples. Salmonella detection assays were comparatively evaluated for detection of Salmonella in swine feces, using 10 g samples (n = 100). For the "gold-standard," the combination of results from two conventional bacteriological isolation methods were used. The detection assays evaluated in this study included two antigen capture enzyme-linked immunossays (ELISA 1: TECRA Salmonella Visual Immunoassay; and ELISA 2: Assurance Gold Salmonella EIA), a DNA hybridization assay (Gene-Trak), and an immunochromatographic assay (Path-Stik). From the fecal samples used in this study (n = 100), 78 were positive for the isolation of Salmonella by method A, and 75 were positive by method B. The combination of the results from both methods (A + B) constituted 87 positive samples. The sensitivity and agreement (Kappa statistics) with the "gold-standard" for each evaluated detection assay were: 87.36% and 0.64 for the ELISA 1; 98.85% and 0.96 for the ELISA 2; 97.70% and 0.92 for the DNA hybridization assay; and 80.46% and 0.48 for the immunochromatographic assay. All detection assays, except by the immunochromatographic assay, detected no false-positive samples. The immunochromatographic assay detected 1% of false-positive samples. The false-negative results for each evaluated assay were: 11% for the ELISA 1; 1% for the ELISA 2; 2% for the DNA hybridization assay, and 18% for the immunochromatographic assay. Results from this study demonstrate that there are Salmonella detection assays currently available from the food industry with potential application on swine clinical samples. These assays could be useful in epidemiological investigations using clinical samples, like swine feces. These assays may allow the processing of large numbers of samples as usually required in many epidemiologic studies of Salmonella in swine populations. The use of commercially available detection assays is a reliable and useful tool, besides of being less laborious and less expensive than conventional bacteriological isolation methods.