Correlation of pharmacodynamic parameters of five beta-lactam antibiotics with therapeutic efficacies in an animal model.

The MIC is the main microbiologic parameter used to predict the efficacies of antibiotics. However, it is well known that MICs may vary according to the inoculum size used (inoculum effect), especially with some beta-lactam antibiotics. In order to correlate the pharmacologic and microbiologic properties of some beta-lactams, an experimental model of intraperitoneal infection caused by Escherichia coli in nonneutropenic and neutro-penic mice was developed. The animals were treated with three different doses of either ampicillin, piperacillin, aztreonam, cefazolin, or cefotaxime. The linear regression analysis obtained in our model shows a better correlation between in vitro activity and efficacy when the MICs considered were those obtained with a large inoculum (ca. 1 x 10(8) CFU/ml) instead of the standard inoculum (5 x 10(5) CFU/ml). The correlations for the MICs obtained with the large inoculum were 0.78 for log2 maximum concentration of drug in serum (Cmax)/ MIC, 0.72 the time that the concentrations exceeded the MIC, and 0.79 for log2 area under the serum concentration-time curve (AUC)/MIC at 24 h in nonneutropenic mice. The corresponding values in neutropenic mice, also for the MICs obtained with the large inoculum, were 0.54, 0.68, and 0.64, respectively, at 24 h. A good correlation was also obtained for the same parameters in nonneutropenic mice at 48 h. The values of Cmax, AUC, and the time that the concentrations exceeded the MIC were parallel among the antibiotics studied, and our study confirms that the time that the levels in serum exceed the MIC is a significant parameter determining the efficacies of beta-lactam antibiotics, but the correlation is much better when the MICs obtained with the large inoculum instead of those obtained with the standard (low) inoculum are considered.