Purpose: This study developed kinetic mathematical models to predict S. aureus growth in ready-to-eat salads under constant and dynamic temperatures.
Methods: A five-strain mixture of S. aureus was inoculated in 5 g potato and sweet potato salads at 5 log CFU/g. The samples were stored at 10 (216 h), 15 (216 h), 20 (120 h), 25 (48 h), and 30°C (48 h). Cell counts of total bacteria (tryptic soy agar) and S. aureus (mannitol salt agar) were enumerated. The growth data were fitted to the modified Gompertz model and the Baranyi model to calculate lag phase duration (LPD; h), maximum specific growth rate (μmax; log CFU/g/h), lower asymptote (log CFU/g), and upper asymptote (log CFU/g). The parameters were fitted to the Davey model, the square root model, and a polynomial equation. Accordingly, S. aureus growth was simulated under constant (20 and 27°C) and dynamic temperatures, and the data was compared with observed data. To evaluate the model performance, root mean square error (RMSE) was calculated.
Results: Growth of total bacteria and S. aureus was observed in potato salad and sweet potato salad at 15-30°C. For both primary models, μmax values were increased, but LPD values were decreased as temperature increased. R2 values (0.789-0.976) of the developed secondary models were acceptable. RMSE values were 0.67-0.83 and 0.67-0.7 for 20 and 27°C, respectively, indicating that the model performance was acceptable. Moreover, the predicted S. aureus cell counts calculated by the model simulation under dynamic storage temperature were close to the observed S. aureus cell counts.
Significance: The results indicate that the developed models for potato and sweet potato salad should be useful in predicting S. aureus growth.