Epsilon Open Archive

Abstract

Knowledge on pharmacokinetic/pharmacodynamic (PK/PD) relationships of antimicrobials may assist in defining a dose that maximises effect while minimising side effects such as selection of resistance. The aim of this thesis was to develop a model for studies of PK/PD relationships of antimicrobials in presence of the host defences. To this end, the usefulness of subcutaneously implanted tissue cages was explored. Tissue cages implanted in calves were infected with Escherichia coli and the calves were treated iv with different doses of trimethoprim + sulfadoxine. Irrespective of dose, no effect was noted in cages infected before treatment. In vitro studies and analysis of thymidine indicate that the effect of trimethoprim was antagonised by thymidine in tissue cage fluid. When the antimicrobial is administered systemically as above, the pharmacokinetics of the drug in tissue cage fluid is typically sluggish. It was shown that if tissue cages with a constant diffusion area but different volumes are used, and drug is injected directly into the cages, the volume of the cage will determine the elimination rate of the drug. Thereby, by varying the dose and cage-type, a range of concentration-time profiles can be simulated. To explore the utility of this concept, the effects of penicillin and danofloxacin against infections with Mannheimia haemolytica in tissue cages in calves were studied. For penicillin, the area under the curve to minimum inhibitory concentration (MIC) and time above MIC were equally predictive for effect. For danofloxacin, the area under the curve to MIC predicted the effect, and the magnitude needed for near maximum effect was 240-244 h. These results are in line with information for other fluoroquinolone-bacterial combinations. The advantage of the described model is that different concentration-time profiles can be simulated, and concentrations of drug at the site of infection and bacterial counts can be monitored over time in presence of the host defences. Its validity needs to be challenged by direct comparisons with other models, and by clinical studies. The results suggest it can be a useful intermediary step between in vitro studies and clinical trials aiming at drug dosage optimisation.