Comparison of culture, polymerase chain reaction (PCR), TaqMan Salmonella, and Transia Card Salmonella assays for detection of Salmonella spp. in naturally-contaminated ground chicken, ground turkey, and ground beef.

Four types of assays were evaluated for the detection of Salmonella spp. in retail ground chicken (86 packages), ground turkey (104 packages), and ground beef (54 packages). Two 25 g samples from each package were separately subjected to pre-enrichment in buffered peptone water for 20 h at 37 degrees C followed by enrichment in Rappaport Vassiliadis (RV) broth for 20 h at 42 degrees C. The RV enrichments were plated onto Rambach agar, Rainbow Agar Salmonella, and XLT4 agar, and were also tested by a PCR assay targeting the Salmonella invA gene, as well as by the TaqMan Salmonella PCR assay. Additionally, the RV enrichments were tested using the Transia Card Salmonella immunoassay. Results showed that 16.8, 24.0, 28.8, and 26.4% of turkey samples were positive for Salmonella spp. by culture, PCR, TaqMan PCR, and Transia Card Salmonella assays, respectively. Eighteen, 28.5, 35.5, and 34.9% of chicken samples were positive by culture, PCR, TaqMan PCR, and Transia Card Salmonella assays, respectively, and 6.5, 6.5, 6.5, and 18.5% of ground beef samples were positive by the four assays, respectively. Analysis of the data using the kappa statistic showed that there was substantial to excellent agreement between the PCR and TaqMan PCR assays and between the PCR and culture assays (kappa coefficients ranging from 0.67 to 0.87), while there was poor to fair agreement between the results of the Transia Card Salmonella assay and the other methods (kappa coefficients ranging from 0.28 to 0.32). Overall, results showed that the PCR-based assays were more sensitive than the culture method, and the culture and PCR-based assays were more specific than the immunoassay for detection of Salmonella in ground chicken, turkey, and beef due to the occurrence of false positive results using the immunoassay.