View Report Details

Development and validation of a versatile model for predicting growth of Clostridium perfringens during cooling of meat

Project Code: B14008

30/11/2003

University of Reading, School of Food Biosciences, Food Microbial Sciences Unit.
Loder, C ;
Kingston University
Kelly, A; Mackey, B;
University of Reading, School of Food Biosciences, Food Microbial Sciences Unit.
Robbins, P; Fryer, P

Clostridium perfringens is the main single cause of food poisoning associated with red meat, and the Food Standards Agency has identified it as a target organisms in its strategy for reducing the incidence of foodborne illness by 20% by April 2006. Between 500 and 1500 cases are reported each year in England and Wales but the real total may be much higher than this (CDR 1997). Spores of C. perfringens are able to survive normal cooking and can germinate and outgrow if food is not properly cooled. Outbreaks of foodborne
illness typically occur in hotels, restaurants, hospitals or residential homes and are often caused by poor temperature control during the handling of cooked meats in these establishments. There is also concern that large commercial cooked meat products that take several hours to cool may also be at risk from growth of C. perfringens. There is little prospect of completely preventing contamination of meat by C. perfringens because the organism occurs ubiquitously in the soil and is found in the gut of healthy animals. The
Food Standards Agency has therefore identified cooking and cooling as critical points for controlling this pathogen.

Guidance on the handling of cooked meats under domestic or commercial conditions is contained in various regulations, guidelines and codes of practice and has been summarised by Gaze et al. (Campden and Chorleywood Food Research Association. Report no. MB/REP/16286/1, 1996). Despite the usefulness of these guidelines they do not provide specific information on the actual extent of growth that is likely to occur under different cooling regimes. It is therefore difficult to evaluate and compare the risks associated with different commercial or domestic cooking and cooling practices. A model is therefore required that can predict growth of C. perfringens under a wide range of cooling profiles including more complex ones.

The aim of this work was to develop and test a model for predicting growth of C. perfringens at different regions within a cooked food during cooling. It was seen as being complementary to related projects at the Institute of Food Research Norwich, CCFRA and the Health Protection Agency. The work at Reading was conducted in collaboration with the Chemical Engineering Department at Birmingham University,

The project consisted of the following main elements: (i) development and validation of a mathematical model to predict temperature profiles at different regions of a food during heating and cooling (ii) validation of the Norwich C. perfringens growth model under appropriate changing temperature conditions (iii) combining the temperature and growth models to be able to predict growth at different regions in a food during cooling and (iv) validation of the combined model.

Model food systems were developed consisting of (i) clostridial growth medium solidified with a gelling agent (gelrite) contained in cylindrical plastic `sausage’ casings or (ii) minced beef packed into casings of the same size. The gelling agent remained solid above 100ºC and allowed experiments to be conducted in this temperature range if necessary. Growth of C. perfringens could be followed after being subjected to different heating and cooling regimes. Two sizes of sausage were used: large (13-14 cm diameter) or small
(3 cm diameter).

Computer code was written in MATLAB to solve engineering equations for predicting the dynamically changing temperature profiles within the cylinders during heating. The resultant predictive model was tested experimentally with thermocouples placed in the gelrite or minced beef sausages. Temperatures measured within the sausages at the centre or at
different radii were very close to those predicted by the model under a range of different heating and cooling conditions.

Growth of C. perfringens in liquid medium (Reinforced clostridial medium containing minced beef) was followed at a range of constant or changing temperatures and the viable counts compared with predicted values. The agreement between predicted growth and experimental results was very good provided the adjustment index parameter was set at 0.01 or in some cases 0.001. There were no major differences in the extent of the lag phase between spore and vegetative cell inocula under these conditions.

The C. perfringens growth model was incorporated in the MATLAB temperature predictor model and simulation studies were carried out comparing growth at different regions of a large cylindrical food product under different cooling regimes. Preliminary validation studies with the small sausage gave good agreement between model and experiment.

The ability to calculate internal temperatures and consequent microbial growth as a function of external temperatures and product size permits simulation exercises that will help to identify critical factors in risk assessment. The model can be extended to allow temperature profiles and microbial growth to be simulated in other in other shaped bodies using appropriate numerical methods. This results of this project together with those of colleagues at IFR Norwich, CCFRA and HPA will provide the Food Standards Agency with a powerful tool for evaluating the risk of C. perfringens growth under different heating and cooling conditions, and will be an educational resource for the Food and Catering Industries and Enforcement Agencies.

Main File

Some of the files on this site may be in a format that your computer can't read. However, you can download Readers and Viewers for the following document types below: