TaqMan Salmonella enterica Detection Kit. Performance Tested Method 020803

Peanut butter spiked with Salmonella enterica ser. Typhimurium was prepared by an independent laboratory and sent to Applied Biosystems to determine the sensitivity and specificity of the TaqMan Salmonella enterica Detection Kit for detecting Salmonella in peanut butter. The samples were spiked at three levels: five no-spike (0 CFU/25 g); 20 low-spike (0.2 CFU/25 g); and 20 high-spike (2 CFU/25 g). They were coded to create a blind set of 45 samples. The samples were processed based on an unpaired test design that included enrichment in buffered peptone water for the candidate method and lactose broth for the reference method. In the candidate method, a 1 mL aliquot of enriched sample was extracted using PrepMan Ultra Sample Preparation Reagent; the sample was amplified on the Applied Biosystems 7500 real-time PCR system, and analyzed for detection of Salmonella using RapidFinder Version 1.0 software. All samples processed by the candidate method were confirmed by culture according to the U.S. Food and Drug Administration's Bacteriological Analytical Manual procedures. Sensitivity, specificity, and Chi-square analysis were calculated by combining candidate method results with those of the reference method that were collected by the independent laboratory. The TaqMan Salmonella enterica Detection Kit showed 40% sensitivity, 100% specificity, and a Chi-square value equal to 1.52. Chi-square analysis indicated the candidate method and the reference method were comparable. Although the candidate method sensitivity was only 40% when compared with the reference method (unpaired samples), the sensitivity was > 100% when the candidate method results were compared with those of the confirmation method (same sample enrichment).