Microbial Inactivation Kinetics Models, Survival Curves Shapes, and the Temporal Distributions of the Individual Germs Deactivation

Regardless of the targeted microbe type, a thermal or nonthermal food preservation or disinfection method’s efficacy is primarily assessed based on its kinetics. Yet, there is growing realization that inactivation kinetics and the individual microbes’ spectrum of vulnerabilities or resistances to a lethal agent are two sides of the same coin. This creates the possibility to convert traditional survival data plotted on linear or semilogarithmic coordinates to temporal distributions of the individual microbes’ deactivation, or vice versa. Such conversions are demonstrated with simulated microbial survival patterns generated with different kinds of survival models: the two-parameter Weibull distribution of which the single-parameter loglinear model is a special case, the normal, lognormal, and Fermi distribution functions, which imply that complete microbial inactivation is theoretically impossible, the three-parameter Gompertz survival model which allows for definite residual survival, and the three-parameter version of the beta distribution function, allowing for a definite thermal death time beyond which no survivors will ever be found. Also provided are simulated examples of the survival patterns of mixed microbial populations, and they all demonstrate that the common shapes of microbial survival curves do not contain enough information to infer whether the targeted microbial population is genetically or physiologically uniform or a mixture of subpopulations. The presented analysis lends support to the notion that any proposed microbial survival kinetic model’s validity should be tested by its ability to predict survival patterns not used in its formulation and not by statistical fit criteria.