MIC vs.

kill-curve based pharmacokinetic/pharmacodynamic modelling of activities of cefpodoxime and Cefixime

Abstract number: 1134_02_110 Derendorf H., Liu P., Rand K., Obermann B. Objectives: Pharmacokinetic (PK)/Pharmacodynamic (PD) modeling of antibiotics usually consists of the comparison between plasma PK and the MIC, such as the T > MIC, Cmax/MIC, and AUC/MIC. These indices are limited due to the innate inaccuracy of the MIC and the fact that it does not reflect the in vivo scenario where concentrations are not static but fluctuate between doses. An alternative is to use time-kill curves that follow bacterial killing and growth as a function of time and concentration. This study provides a systematic comparison of MIC and kill-curve approaches to show the potential of both methods for antibiotic evaluation. In an example, we developed a mathematical pharmacokinetic/pharmacodynamic (PK/PD) model to integrate the in vitro antimicrobial activity with the PK profile of two oral cephalosporines at the tissue site. Methods: Kill curves could be described with different combinations of maximum kill rate (kmax), drug concentration at halfmaximum effect (EC50) and bacterial growth rate (k0) that resulted in the same MIC in each scenario. In the experimental part, bacterial time-kill curves of cefpodoxime and cefixime against four bacterial strains were compared in in vitro kinetic models in which previously measured human pharmacokinetic profiles of unbound antibiotic were integrated. Results: Different combinations of kmax, EC50 and k0 can yield the same MIC and consequently the same MIC-based index. However, depending on the combination of PD parameters, kill curves may predict quite opposite outcomes in both scenarios. EC50 values of cefpodoxime and cefixime were consistent with their respective MIC values. Both antibiotics had similar high potency against H. influenzae(EC50: 0.04 mg/L) and M. catarrhalis (EC50: 0.12 mg/L), while the potency of cefpodoxime against S. pneumoniae strains was about 10-fold higher than that of cefixime (EC50s/sensitive: 0.02 vs 0.27 mg/L; EC50s/intermediate: 0.09 vs 0.69 mg/L). Simulations showed that cefpodoxime will have higher bacteriological success against S. pneumoniae than cefixime. Conclusions: Simple comparison of exposure and MIC may not be sufficient to evaluate anti-infective efficacy. Kill curves provide a more detailed approach in predicting antimicrobial effects. The developed Emax model effectively described the pharmacodynamics of cefpodoxime and cefixime. Cefpodoxime (200 mg bid) has higher tissue penetration and antimicrobial efficacy than cefixime (400 mg qd) againstS. pneumoniae.