Growth comparison and model validation for growth of Shiga toxin-producing Escherichia coli (STEC) in ground beef

Abstract

The aim of this study was to compare the growth of non-O157 Shiga toxin-producing Escherichia coli (STEC) with E. coli O157:H7 in ground beef and modified Tryptic Soy Broth (TSB). Additionally, the performance of three available models (ComBase, Huang et al. model, and Cepeda et al. model), for predicting STEC growth in ground beef was evaluated. To achieve this, eleven groups of STEC, including E. coli O157:H7, were inoculated into irradiated ground beef (7% or 27% fat) or modified TSB, and then subjected to programmed water baths to follow time-varying sinusoidal temperature profiles for 300 h (low temperature, 5–15 °C) and 25 h (high temperature, 10–40 °C). The growth data of STEC in ground beef were fitted to the three models, and model performance indices such as mean relative percentage error (MRPE), mean absolute relative error (MARE), root mean squared error (RMSE), bias factor (BF) and accuracy factor (AF) were calculated to assess their accuracy in predicting STEC growth. The study found that the growth of non-O157 STEC was similar to that of E. coli O157:H7, indicating that growth models for E. coli O157:H7 can be used to predict the growth of non-O157 STEC. The fat content did not significantly affect STEC growth in ground beef. Among the three models tested, the Huang et al. model underestimated the growth of all STEC in ground beef, with an RMSE of up to 2.14 log CFU/g. The ComBase model also under-predicted STEC growth in ground beef at high temperatures (10–40 °C) and could not predict growth at temperatures below 10 °C, where STEC might grow slowly. In contrast, the Cepeda et al. model accurately predicted STEC growth in ground beef and is considered an conservative growth model for STEC in commercial ground beef products. In summary, this study provides important insights into the growth of non-O157 STEC and E. coli O157:H7 in ground beef and their modeling, which can aid in predicting the safety of commercial ground beef products.