Automatic procedures for measuring post-antibiotic effect and determining random errors

Four methods are described for determining the post-antibiotic effect (PAE) automatically using the high resolution capabilities of the conductance measurement of bacterial growth. The time resolution of each method is derived by a statistical procedure for calculating random error from the component errors involved in each determination. The procedures are illustrated using data generated from the measurements of the growth of Escherichia coli NCTC 4174 after exposure of cultures to five antibiotic concentrations. Excellent agreement was found to exist between three of the four methods. However, that based on the determination of the time of peak growth rate is shown to provide better resolution than the other methods, i.e., those based on growth to 10(7) organisms/mL, the duration of the lag phase or the mean retardation of growth averaged over the entire period of observation. The results strongly suggest that PAE may be ascribed entirely to a prolongation of lag time.