Impact of growth matrix on pharmacodynamics of antimicrobial drugs for pig pneumonia pathogens

BACKGROUND The most widely used measure of potency of antimicrobial drugs is Minimum Inhibitory Concentration (MIC). MIC is usually determined under standardised conditions in broths formulated to optimise bacterial growth on a species-by-species basis. This ensures comparability of data between laboratories. However, differences in values of MIC may arise between broths of differing chemical composition and for some drug classes major differences occur between broths and biological fluids such as serum and inflammatory exudate. Such differences must be taken into account, when breakpoint PK/PD indices are derived and used to predict dosages for clinical use. There is therefore interest in comparing MIC values in several broths and, in particular, in comparing broth values with those generated in serum. For the pig pneumonia pathogens, Actinobacillus pleuropneumoniae and Pasteurella multocida, MICs were determined for three drugs, florfenicol, oxytetracycline and marbofloxacin, in five broths [Mueller Hinton Broth (MHB), cation-adjusted Mueller Hinton Broth (CAMHB), Columbia Broth supplemented with NAD (CB), Brain Heart Infusion Broth (BHI) and Tryptic Soy Broth (TSB)] and in pig serum. RESULTS For each drug, similar MIC values were obtained in all broths, with one exception, marbofloxacin having similar MICs for three broths and 4-5-fold higher MICs for two broths. In contrast, for both organisms, quantitative differences between broth and pig serum MICs were obtained after correction of MICs for drug binding to serum protein (fu serum MIC). Potency was greater (fu serum MIC lower) in serum than in broths for marbofloxacin and florfenicol for both organisms. For oxytetracycline fu serum:broth MIC ratios were 6.30:1 (P. multocida) and 0.35:1 (A. pleuropneumoniae), so that potency of this drug was reduced for the former species and increased for the latter species. The chemical composition of pig serum and broths was compared; major matrix differences in 14 constituents did not account for MIC differences. Bacterial growth rates were compared in broths and pig serum in the absence of drugs; it was concluded that broth/serum MIC differences might be due to differing growth rates in some but not all instances. CONCLUSIONS For all organisms and all drugs investigated in this study, it is suggested that broth MICs should be adjusted by an appropriate scaling factor when used to determine pharmacokinetic/pharmacodynamic breakpoints for dosage prediction.