Estimation of the Prevalence of Clostridium perfringens in Raw, Heat-Treated, and Fully Cooked Processed Meat Products in the USA

The goal of this study was to determine the probability of finding Clostridium perfringens organisms in raw and finished ready-to-eat (RTE) meat products, which could have a practical implication to food safety. In 2001, the Food Safety Inspection Service (FSIS) reported that the levels of C. perfringens were greater than 4-log CFU/g in a limited number of raw meat samples. This was the basis for creating the current 9CFR 318.17(a) (2) rule limiting permissible potential growth of C. perfringens during thermally processed product chilling (stabilization) to <1 log CFU/g. Data from raw materials (chicken), not-ready-to-eat (NRTE) bacon, and cooked products (submitted by three US commercial meat processors) were gathered. The cooked ready-to-eat (RTE) product data involved 3,141 cooked products not related to process deviations, while data collected from 2,886 products originated from process deviations determined not to meet the "safe harbors" outlined in USDA, FSIS Appendix B. Although five bacon (not RTE) samples exceeded 4-log CFU/g C. perfringens, none of the cooked samples (regardless of process deviations association) exceeded 2-log CFU/g, suggesting that the high initial levels in the raw product were vegetative cells which were eliminated during the cooking process and therefore would not grow during cooling. The probability of having counts ranging from 2 to 3 log CFU/g in nondeviation samples and deviation RTE samples was estimated to be 0.0012 and 0.0013, respectively. The data further suggest that the probability of C. perfringens growing to populations causing illness (8-log CFU per serving, assuming a 100-g serving) during product stabilization of commercial meat products is highly unlikely, even with cooling-related deviations.