Quantification of Salmonella from porcine samples to identify potential critical control points in pre-harvest pork production

Currently there is a need for detailed enumeration data in the study of Salmonella. Enumeration of Salmonella is generally accomplished using the most probable number (MPN) technique. While generally accepted, the labor and media intensive nature of this technique discourages enumeration and limits the number of samples that can be enumerated. In an effort to reduce the labor associated with enumeration and to increase the number of samples that can be enumerated, a modified most probable number technique, termed the tube, was designed and is evaluated in this paper. This technique includes commonly used media (modified semi-solid Rappaport-Vassiliadis agar, XLT4 agar, and tetrathionate broth) in a tube format. The RX tube was capable of detecting approximately 1 Salmonella cell for 18/21 strains. Using artificially contaminated fecal samples of a known Salmonella concentration there was a correlation of 0.97 between tryptic soy agar plate counts (TSA) and MPN, 0.95 between RX and TSA, and 0.95 between RX and MPN. When naturally contaminated samples were used, a correlation of 0.77 between RX and MPN results was found. Additionally, RX and MPN were found to be equivalent within a 0.5 log tolerance. Finally, the RX tube used approximately one-fourth of the labor required by MPN. The RX method appears to be equivalent with the MPN method and requires significantly less labor, making enumeration of Salmonella more feasible.