A pharmacodynamic model of the response of Staphylococcus aureus in an in vitro system simulating the pharmacokinetics of penicillin G in the interstitial fluid of sheep

The purpose of this study was to investigate the pharmacokinetic/pharmacodynamic (PK/PD) interaction of penicillin G and Staphylococcus aureus in an in vitro kinetic system based on the pharmacokinetics of penicillin G in the interstitial fluid (ISF) dialysate in sheep.A new technique using implanted dialysis fiber bundles provided a simple and convenient method to obtain ISF dialysate from sheep for the determination of penicillin G concentrations in ISF. Penicillin G was administered intravenously to sheep at 26.4 mg/kg and 52.9 mg/kg. The disposition of penicillin G was best described by a two-compartment open model defined by the sum of two exponentials equation Cp = 170.64e$\sp{\rm -7.16t}$ + 31.04e$\sp{\rm -1.56t}$ for the low dose and Cp = 418.19e$\sp{\rm -6.15t}$ + 59.85e$\sp{\rm 1.56t}$ for the high dose. The rate constants that described the distribution of penicillin G in plasma and ISF dialysate were subsequently used to simulate a similar penicillin G profile in an in vitro kinetic system. The pharmacodynamic response of Staphylococcus aureus to equal doses of penicillin G administered either as a bolus dose or continuous infusion was studied in the in vitro system during a 6-hour observation period. There was a greater percent decrease in the mean number of viable bacteria during the continuous infusion (92.77%) as compared to the bolus dose (83.73%). The PK/PD interaction between penicillin G and Staphylococcus aureus in the in vitro kinetic system was best described by a two-compartment open model with an effect compartment. The effect compartment was added to provide for the temporal aspects of the pharmacodynamic interaction. These included a prolonged post-antibiotic effect and a lag phase for onset of antibacterial action after penicillin G was added. Due to physical limitations of the in vitro kinetic system, a complete bacterial profile including bacterial regrowth was not observed. Therefore, although the model adequately described the data obtained, it could not be used to predict optimal dosage regimens.PK/PD models based on the pharmacokinetics of the drug at the site of infection and changing drug concentrations in vitro could be used for the study of pharmacokinetic and pharmacodynamic interactions. Additional testing needs to be completed to obtain the complete bacterial profile after administration of an antibacterial agent, including bacterial regrowth. Such a PK/PD model could then be used to predict antibacterial effects of various dosage regimens.U of I OnlyETDs are only available to UIUC Users without author permissio