Objectives
Pharmacodynamic parameters evaluated under conditions that simulate an infection site volume and microbial load might reveal hidden risks of resistance selection and subsequent treatment failure. The study aimed to investigate the predictive potential of MICs determined at various conditions on the antimicrobial effect and emergence of resistance.
Methods
We assessed meropenem MICs (microdilution: 0.2 mL, 5 × 105 cfu/mL; macrodilution: 2 mL, 5 × 105 cfu/mL), MICHVs (220 mL, 5 × 105 cfu/mL), MICHIs (0.2 mL, 5 × 107 cfu/mL) and MICHVIs (220 mL, 5 × 107 cfu/mL) for five Klebsiella pneumoniae strains and analysed these values alongside the results of experiments in a dynamic in vitro model. A clinically relevant meropenem dosing regimen was simulated and the starting bacterial inocula were 106 and 108 cfu/mL.
Results
The effectiveness of meropenem agreed with MICHVs for the 106 cfu/mL inoculum and with MICHIs or MICHVIs for the 108 cfu/mL inoculum. Strains characterized as resistant according to these values grew during meropenem exposure, and resistant mutants were selected.
Conclusions
Our results suggest that MICHV-based parameters may be suitable for predicting antibacterial effects and the risk of resistance development when the inoculum is 106 cfu/mL, while MICHI- or MICHVI-based parameters are suitable for these purposes when the inoculum is 108 cfu/mL. Also, the correlation between resistance selection and the MICHI-based parameter was as high as one that corresponds with a mutant prevention concentration (MPC)-based parameter; this suggests that the MPC can be replaced by the more easily determined alternative parameter MICHI.