Food Protection Trends - May 2010 - 270

FIGURE 1. D values for existing TDT data. Temperatures for each data set are shown on the graph in oC. Data for three replicates at the indicated times for 50oC (122oF, octagons), 52oC (126oF, triangles), 54oC (129oF, squares), 56oC (133oF, circles), 58oF (136oF, diamonds), and 60oC (140oF, inverted triangles) are shown. Regression lines for each data set are shown next to the corresponding temperature number.

(1), and the D and Z values were determined by two approaches. The first approach was to develop a linear model based on the published five log reduction times that were originally determined by a non-linear (Weibull) model (4). A log-linear transformation of the published five log reduction times (generated by use of the Weibull model) was plotted against temperature. Linear regression was then carried out to determine a Z value from the slope of the regression line. In the second approach, the existing thermal death time (TDT) raw data (4) was re-examined using a linear model instead of a Weibull model. Linear regression of log10 surviving cells vs. time, and log10 D values vs. temperature, to determine D and Z values (respectively), was carried out. All regression calculations were performed with SigmaPlot software (Version 10.0, Systat Software, Inc, Chicago IL). For both methods, 160oF was chosen as a reference temperature.

RESULTS
The D values for a temperature range of 50oC (122oF) to 60oC (140oF) were generated by use of a linear model with the existing TDT data (Fig. 1). Based on these D values (Table 1), a Z value of 11.98oF was determined for E. coli O157:H7. For a reference temperature of 160oF, a processing time of 0.016 min. was determined for the linear model (Fig. 2). The five log reduction line from these data has the same slope and Z value, but the processing time at the reference temperature of 160oF was 0.08 min (Table 2). The R2 value for the log-linear regression to determine the Z value was 0.96 (Fig.2). Previously, Breidt et al. (4) used a Weibull model and a exponential decay function to predict five log reduction times. The predicted values from an exponential decay model were used to determine safe processing times for temperatures between 160oF and 180oF (4). Recommendations for safe processing times for industry included the addition of five times the standard error to the predicted processing times. From these data, a linear model was used to fit the predicted values from the exponential decay model by taking the log10 of the predicted

FIGURE 2. Processing times based on three different models. The triangles represent the log10 of the D values determined for the linear model as described in the text. The squares represent the published Weibull five log reduction data with five times the standard error added. The regression lines are as follows: dotted line, from the one log reduction values (triangles); solid black line, five log reduction times; dashed black line, exponential decay model with five times the standard error added.

270 FOOD PROTECTION TRENDS | MAY 2010
Food Protection Trends - May 2010

Table of Contents for the Digital Edition of Food Protection Trends - May 2010
