| Abstract
|
:
|
In response to continued concerns regarding Listeria cross-contamination of ready-to-eat meat and poultry products in both retail and home kitchens, a series of studies was conducted to: (1) optimize the quantitative recovery of L. monocytogenes from stainless steel surfaces, (2) determine direct and sequential transfer rates for L. monocytogenes from artificially contaminated ready-to-eat luncheon meats to a delicatessen slicer and vice versa, (3) determine the effects of cutting force, stainless steel grade, sharpness, and product composition on transfer of L. monocytogenes from artificially contaminated ready-to-eat luncheon meats to knives and vice versa, and (4) develop a mathematical model based on the transfer coefficients obtained from the previous three objectives that will predict the numbers of L. monocytogenes cells transferred during slicing of delicatessen meats. Initially, four sampling devices: (1) sterile environmental sponge (ES), (2) sterile cotton-tipped swab (CS), (3) sterile calcium alginate fiber-tipped swab (CAS), and (4) 1-ply composite tissue (CT), were evaluated for quantitative recovery of L. monocytogenes from food-grade stainless steel. Recovery was 2.70, 1.34, and 0.62 log greater using CT compared to ES, CS, and CAS, respectively. The CT device, which is inexpensive and easy to use, represents a major improvement over other methods in quantifying L. monocytogenes. Thereafter, a commercial delicatessen slicer blade and simulated kitchen knife blades were used as vectors for sequential transfer of L. monocytogenes from (a) an inoculated blade (∼108, 10 5, 103 CFU/blade) to 30 slices of uninoculated delicatessen turkey, bologna, and salami, (b) inoculated product (∼108 cm2) to the blade and (c) inoculated product (108, 105, 103 CFU/cm 2) to 30 slices of uninoculated product via the blade with cutting force and product composition also assessed for their impact on Listeria transfer. Using slicer blades inoculated at 108 CFU/blade, Listeria populations decreased logarithmically to 102 CFU/slice after 30 slices. Findings for inoculated slicer blades and products (105 CFU/blade or cm2) were similar with Listeria counts of 102 CFU/slice after 5 slices and enriched samples generally negative after 27 slices. Using 103 CFU/slicer blade, the first 5 slices typically contained ∼101 CFU/slice by direct plating with enrichments negative after 15 slices. Knife blades containing 105 and 103 CFU/blade typically yielded direct counts out to only 20 and 5 slices, respectively, with "tailing" observed thereafter. Variables that enhanced Listeria transfer during slicing and cutting included higher fat and lower moisture content, application force, blade surface roughness, and stainless steel grade with greater transfer using 304 as opposed to 316. These finding were then used to develop four fitted predictive models in the form [CFU (X) = kaX] along with a program written in GWBasic. These models can be used if any two of the following three values are known: (a) initial inoculum level, (b) total bacterial transfer, (c) fraction of bacteria remaining on blade after consecutive slicing, solving for each model parameter CFU(X), k, or a. Based on our models, the greatest number of Listeria (>90%) will be found in the first 15 slices.
|