Submitted to: Food Microbiology
Publication Type: Peer reviewed journal
Publication Acceptance Date: 8/28/2007
Publication Date: 11/1/2007
Citation: Juneja, V.K., Marks, H., Thippareddi, H.H. 2007. Predictive model for growth of clostridium perfringens during cooling of cooked uncured beef. Food Microbiology. 25:42-55. Interpretive Summary: One of the most common types of food poisoning in the United States is caused by the bacterium, Clostridium perfringens. Illnesses have been traditionally associated with inadequate cooling practices in retail food service operations. Thus, there was a need to determine the time and temperature for cooked uncured meat products to remain pathogen-free and provide vital data for performing risk assessment on cooked meat. We developed a model that can be used to predict the growth of C. perfringens at temperatures relevant to the cooling of cooked products. The predictive model will be of immediate use to the retail food service operations and regulatory agencies to aid with the disposition of products subject to cooling deviations and therefore, ensure the safety of the cooked foods.
Technical Abstract: This paper considers two models that have been used for modeling growth of C. perfringens during cooling. Using a common approach or methodology for constructing models, there was no appreciable difference between the model predictions when the population of cells was within the lag or exponential phases of growth. For a temperature decline from 54.4 degrees C to 27 degrees C in 1.5 h, the models predicted a log10 relative growth of about.1.1, with a standard error of about 0.08 log10, while observed results for two replicates were 0.43 and 0.90 log10. For the same temperature decline in 3 h, the predicted log10 relative growth was about 3.6 log10 (with a standard error of about 0.07), and the observed log10 relative growths were 2.4 and 2.5 log10. When the cooling scenarios were extended to lower temperature, the predictions were somewhat better, taking into account the larger relative growth: for a cooling scenario of 54.4 degrees C to 27 degrees C in 1.5 h and 27 degrees C to 4 degrees C in 12.5 h, the average predicted and observed log10 relative growths were 3.2 log10 and 2.7 log10, respectively. When cooling was extended from 27 degrees C to 4 degrees C in 15 h, the average predicted and observed log10 relative growths were 3.7 log10 and 3.6 log10, respectively. For the latter cooling scenario, the levels were greater than 6 log10, still less than stationary levels of about 7 or 8 log10. The range for which C. perfringens can grow is estimated at 10 degrees C to 53 degrees C. These findings, while providing estimates of growth that could be used for designing safe cooling processes, together with other reported results using the same methodology as used in this paper, point to a possible inadequacy of the derived models. In particular, predicting growth in dynamic cooling scenarios using these models is based on the appropriateness of applying results obtained for isothermal environments. Incorporating a different assumption that growth kinetics depends in an explicit way on the prior history of cells might provide a more accurate estimate of growth. More research is needed to concentrate on growth of pathogens in dynamic environments.