International Journal of Food Microbiology, [Epub ahead of print], May 25, 2012, DOI: 10.1016/j.ijfoodmicro.2012.05.019

1. HOELZER,^aR. POUILLOT^a, D. GALLAGHER^b, M.B. SILVERMAN^c, J.KAUSE^d, and S. DENNIS^a
   

^a Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, Maryland 20740, USA
^b Civil and Environmental Engineering Department, 409 Durham Hall, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061
^c Joint Institute for Food Safety and Applied Nutrition, University of Maryland, College Park, Maryland 20740
^d Risk Assessment Division, Office of Public Health Science, Food Safety and Inspection Service, U.S. Department of Agriculture, 1400 Independence Avenue S.W., Washington, DC 20228

ABSTRACT

Listeria monocytogenes is readily found in the environment of retail deli establishments and can occasionally contaminate food handled in these establishments. Here we synthesize the available scientific evidence to derive probability distributions and mathematical models of bacterial transfers between environmental surfaces and foods, including those during slicing of food, and of bacterial removal during cleaning and sanitizing (models available at www.foodrisk.org).

Transfer coefficients varied considerably by surface type, and after log10 transformation were best described by normal distributions with means ranging from -0.29 to -4.96 and standard deviations that ranged from 0.07 to 1.39. 'Transfer coefficients' during slicing were best described by a truncated logistic distribution with location 0.07 and scale 0.03. In the absence of protein residues, mean log inactivation indicated a greater than 5 log10 reduction for sanitization with hypochlorite (mean: 6.5 log10; 95% confidence interval (CI): 5.0 - 8.1 log10) and quaternary ammonium compounds (mean: 5.5 log10; 95% CI: 3.6 - 7.3 log10), but in the presence of protein residues efficacy reduced dramatically for hypochlorite (mean: 3.8 log10; 95% CI: 2.1 - 5.4 log10) as well as quaternary ammonium compounds (mean: 4.4log10; 95% CI: 2.5 - 6.4 log10). Overall, transfer coefficients are therefore low, even though cross-contamination can be extremely efficient under certain conditions. Dozens of food items may consequently be contaminated from a single contaminated slicer blade, albeit at low concentrations. Correctly performed sanitizing efficiently reduces L. monocytogenes contamination in the environment and therefore limits cross-contamination, even though sanitization is only performed a few times per day. However, under unfavorable conditions reductions in bacterial concentration may be far below 5 log10.

The probability distributions and mathematical models derived here can be used to evaluate L. monocytogenes cross-contamination dynamics in environments where foods are handled, and to assess the potential impact of different intervention strategies.

Supplemental Materials

Models

Mathematical models of bacterial transfer and bacterial removal during cleaning and sanitizing(provided in @Risk).

NOTE: The software program @Risk (Palisade Corporation) is required to display and use these models.