PK-PD modeling of β-lactam antibiotics: in vitro or in vivo models?

A modified E(max)-pharmacokinetic-pharmacodynamic (PK-PD) model was previously proposed in literature for describing the antimicrobial activity of β-lactam antibiotics based on in vitro experiments. However, bacteria behave differently in vitro and in vivo. Thus, the aims of this study were to model the killing effect of piperacillin (PIP) against Escherichia coli on immunocompromised infected rats using this model and to compare the parameters obtained in vitro and in vivo for the same bacteria/drug combination. The PK-PD parameters determined in vitro and in vivo were as follows: generation rate constant of 1.30 ± 0.10 and 0.76 ± 0.20 h(-1), maximum killing effect of 3.11 ± 0.27 and 1.38 ± 0.20 h(-1) and concentration to produce 50% of the maximum effect of 5.44 ± 0.03 and 1.31 ± 0.27 μg ml(-1), respectively. The comparison between the in vitro and in vivo parameters was not straightforward and had to take into consideration the intrinsic differences of the models involved. So far, the main application of the PK-PD model evaluated is for the comparison of different antimicrobial agent's potency and efficacy, under equivalent conditions.