A predictive model that describes the effect of prolonged heating at 70 to 90 degrees and subsequent incubation at refrigeration temperature on growth from spores and toxigenesis by nonproteolytic Clostridium botulinum in the presence of lysozyme

Refrigerated processed foods of extended durability such as cook-chill and sous-vide foods rely on a minimal heat treatment at 70 to 95 degrees C and then storage at a refrigeration temperature for safety and preservation. These foods are not sterile and are intended to have an extended shelf life, often up to 42 days. The principal microbiological hazard in foods of this type is growth of and toxin production by nonproteolytic Clostridium botulinum. Lysozyme has been shown to increase the measured heat resistance of nonproteolytic C. botulinum spores. However, the heat treatment guidelines for prevention of risk of botulism in these products have not taken into consideration the effect of lysozyme, which can be present in many foods. In order to assess the botulism hazard, the effect of heat treatments at 70, 75, 80, 85, and 90 degrees C combined with refrigerated storage for up to 90 days on growth from 10(6) spores of nonproteolytic C. botulinum (types B, E, and F) in an anaerobic meat medium containing 2,400 U of lysozyme per ml (50 micrograms per ml) was studied. Provided that the storage temperature was no higher than 8 degrees C, the following heat treatments each prevented growth and toxin production during 90 days; 70 degrees C for greater than or equal to 2,545 min, 75 degrees C for greater than or equal to 463 min, 80 degrees C for greater than or equal to 230 min, 85 degrees C for greater than or equal to 84 min, and 90 degrees C for greater than or equal to 33.5 min. A factorial experimental design allowed development of a predictive model that described the incubation time required before the first sample showed growth, as a function of heating temperature (70 to 90 degrees C), period of heat treatment (up to 2,545 min), and incubation temperature (5 to 25 degrees C). Predictions from the model provided a valid description of the data used to generate the model and agreed with observations made previously.