Expansion of response surface models for the growth of Escherichia coli O157:H7 to include sodium nitrite as a variable.

The previously published (Buchanan et al., 1993a) response surface models for estimating the aerobic and anaerobic growth of Escherichia coli O157:H7 as a function of temperature, initial pH, and sodium chloride content have been expanded to include sodium nitrite as a further variable. A fractional factorial design was employed to quantitate the effect of NaNO2 in conjunction with the four other variables by culturing a three-strain mixture in brain heart infusion broth. The activity of NaNO2 was strongly pH-dependent, with inhibition being significant at pH values < or = 5.5 and enhanced by lowering the incubation temperature. The effects of the variables on Escherichia coli O157:H7 growth kinetics were modeled by response surface analysis using quadratic and cubic polynomial models of the natural logarithm transformation of both the Gompertz B and M parameters (Gompertz parameters) and the lag phase duration (LPD) and generation time (GT) values (kinetics parameters) calculated for individual growth curves. All models provided reasonable estimates for most variable combinations; however, comparisons of predicted versus observed values indicated that overall the most useful models were the cubic models based on LPD and GT values. Although additional validation of the models is required, comparisons of predicted times to a 1000-fold increase in population density against those calculated from previously published growth studies indicate that the models are an effective means for acquiring 'first estimates' of the growth characteristics of E. coli O157:H7.