Direct Dynamic Kinetic Analysis and Computer Simulation of Growth of Clostridium perfringens in Cooked Turkey during Cooling

Abstract

This research applied a new 1-step methodology to directly construct a tertiary model that describes the growth of Clostridium perfringens in cooked turkey meat under dynamically cooling conditions. The kinetic parameters of the growth models were determined by numerical analysis and optimization using multiple dynamic growth curves. The models and kinetic parameters were validated using independent growth curves obtained under various cooling conditions. The results showed that the residual errors (ε) of the predictions followed a Laplace distribution that is symmetric with respect to ε = 0. For residual errors, 90.6% are within ±0.5 Log CFU/g and 73.4% are ±0.25 Log CFU/g for all growth curves used for validation. For relative growth <1.0 Log CFU/g, 88.9% of the residual errors are within ±0.5 Log CFU/g, and 63.0% are within ±0.25 Log CFU/g. For relative growth of <2.0 Log CFU/g, 92.7% of the residual errors are within ±0.5 Log CFU/g, and 70.3% are within ±0.25 Log CFU/g. The scale and distribution of residual errors clearly suggests that the models and estimated kinetic parameters are reasonably accurate in predicting the growth of C. perfringens. Monte Carlo simulation was used to estimate the probabilities of >1.0 and 2.0 Log CFU/g relative growth of C. perfringens in the final products at the end of cooling. This probabilistic process analysis approach provides a new alternative for estimating and managing the risk of a product and can help the food industry and regulatory agencies assess the safety of cooked meat in the event of cooling deviation.

A Theoretical Analysis for Assessing the Variability of Secondary Model Thermal Inactivation Kinetic Parameters

“…For example, Poschet et al [ 42 ], in a Monte Carlo analysis, represented the parameter distributions of a microbial growth model by normal distributions. Results by Huang and Vinyard [ 9 ] showed that the distributions of a primary model (a system of ordinary differential equations) parameters were approximately normal. Pénicaud et al [ 50 ] superimposed a pseudo random noise on the experimental data, performed 2000 runs using Monte Carlo technique and showed that the primary model’s parameters (apparent reaction rate constant of ascorbic acid loss) were normally distributed.…”

“…The most widely applied procedure for food kinetic modeling (either for microbiological measurements or for chemical changes) includes a step-to-step procedure [ 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 ]. In such an approach, initially, experiments are conducted under different constant temperature conditions, the most representative indices are selected, and their change is measured as a function of processing or storage time.…”

“…A reliable alternative is deemed necessary to account for the real uncertainty of model parameters, in order to be able to predict, in a more accurate way, chemical or microbiological changes, at any selected temperature and time during processing and product shelf life (at storage or distribution). The Monte Carlo simulation technique has been recently proposed in literature for the probabilistic assessment of stochastic variability and uncertainty associated with microbiological [ 9 , 29 , 30 , 31 , 32 ] or chemical/sensory [ 33 , 34 , 35 , 36 , 37 , 38 ] attributes of several food matrices.…”

Holistic Approach to the Uncertainty in Shelf Life Prediction of Frozen Foods at Dynamic Cold Chain Conditions

“…Alternatively, the model parameters can be determined in a single step considering the same isothermal dataset as a whole, by incorporating the secondary model equations into the primary model and performing a non-linear regression [15]. Current publications have proposed and employed one-step kinetic analysis [1,[16][17][18][19][20][21]. Such approach circumvents the need for statistical estimation of intermediate parameters by employing all the experimental data in a single non-linear algorithm [22], with higher number of degrees of freedom, leading to more precise parameter calculation.…”

Growth and survival of Salmonella Paratyphi A in roasted marinated chicken during refrigerated storage: Effect of temperature abuse and computer simulation for cold chain management

“…More recently, a new one-step dynamic analysis method was developed to directly construct a tertiary growth model from dynamically changing temperature profiles to predict the growth of Clostridium perfringens in cooked beef and turkey meat during cooling (Huang, 2015;Huang and Vinyard, 2016). The models developed using this approach have shown to accurately predict the growth of C. perfringens in cooked beef under both fluctuating and isothermal conditions (Huang, 2015;Huang and Vinyard, 2016). It is also a more efficient method for constructing predictive models as it avoids time-consuming and labor-intensive isothermal experiments.…”