RESUMO
Ready-to-eat lettuce is a food commodity prone to contamination by pathogenic microorganisms if processing and distribution conditions as well as handling practices are not effective. A challenge testing protocol was applied to ready-to-eat iceberg-lettuce samples by inoculating different initial contamination levels (4.5, 3.5 and 2.5 log cfu/g) of Escherichia coli strain (serotype O158:H23) subsequently stored at 8, 12, 16, 20 and 24°C for 6h. A polynomial regression model for log difference (log(diff)) was developed at each inoculum level studied through the calculation of the effective static temperature (T(eff)). Furthermore, the developed model was integrated within a risk-based approach with real time/Temperature (t/T) data collected in three Spanish foodservice centers: school canteens, long-term care facilities (LTCF) and hospitals. Statistical distributions were fitted to t/T data and estimated log(diff) values were obtained as model outputs through a Monte Carlo simulation (10,000 iterations). The results obtained at static conditions indicated that the maintenance of the lettuce at 8°C slightly reduced the E. coli population from -0.4 to -0.5 log cfu/g. However, if chill chain is not maintained, E. coli can grow up to 1.1 log cfu/g at temperatures above 16°C, even at low contamination levels. Regarding log(diff) estimated in foodservice centers, very low risk was obtained (log(diff)<1.0 log cfu in all cases). Mean T(eff) values obtained in hospitals were the lowest ones (11.1°C) and no growth of E. coli was predicted in >92% of simulated cases. The results presented in this study could serve food operators to set time/Temperature requirements for ready-to-eat foods in foodservice centers, providing a scientific basis through the use of predictive modeling. These findings may also serve to food safety managers to better define the control measures to be adopted in foodservice centers in order to prevent food-borne infections.
