The minimal inhibitory concentration (MIC) is conventionally used to define in vitro levels of susceptibility or resistance of a specific bacterial strain to an antibiotic and to predict its clinical efficacy. Along with MIC, other measures of bacteria resistance exist: the MIC determined at high bacterial inocula (MIC_HI) that allow the estimation of the occurrence of inoculum effect (IE) and the mutant prevention concentration, MPC. Together, MIC, MIC_HI and MPC represent the bacterial “resistance profile”. In this paper, we provide a comprehensive analysis of such profiles of K. pneumoniae strains that differ by meropenem susceptibility, ability to produce carbapenemases and specific carbapenemase types. In addition, we have analyzed inter-relations between the MIC, MIC_HI and MPC for each tested K. pneumoniae strain. Low IE probability was detected with carbapenemase-non-producing K. pneumoniae, and high IE probability was detected with those that were carbapenemase-producing. MICs did not correlate with the MPCs; significant correlation was observed between the MIC_HIs and the MPCs, indicating that these bacteria/antibiotic characteristics display similar resistance properties of a given bacterial strain. To determine the possible resistance-related risk due to a given K. pneumoniae strain, we propose determining the MIC_HI. This can more or less predict the MPC value of the particular strain.

OXA-48 carbapenemases are frequently expressed by Klebsiella pneumoniae clinical isolates; they decrease the effectiveness of carbapenem therapy, particularly with meropenem. Among these isolates, meropenem-susceptible carbapenemase-producers may show decreased meropenem effectiveness. However, the probability of the emergence of resistance in susceptible carbapenemase-producing isolates and its dependence on specific K. pneumoniae meropenem MICs is not completely known. It is also not completely clear what resistance patterns will be exhibited by these bacteria exposed to meropenem, if they would follow the patterns of non-beta-lactamase-producing bacteria and other than beta-lactams antibiotics. These issues might be clarified if patterns of meropenem resistance related to the mutant selection window (MSW) hypothesis. To test the applicability of the MSW hypothesis to meropenem, OXA-48-carbapenemase-producing K. pneumoniae clinical isolates with MICs in a 64-fold range (from susceptible to resistant) were exposed to meropenem in a hollow-fiber infection model; epithelial lining fluid meropenem pharmacokinetics were simulated following administration of 2 grams every 8 hours in a 3-hour infusion. Strong bell-shaped relationships between the meropenem daily dose infused to the model as related to the specific isolate MIC and both the antimicrobial effect and the emergence of resistance were observed. The applicability of the MSW hypothesis to meropenem and carbapenemase producing K. pneumoniae was confirmed. Low meropenem efficacy indicates very careful prescribing of meropenem to treat K. pneumoniae infections when the causative isolate is confirmed as an OXA-48-carbapenemase producer.

The inoculum effect (IE) is a well-known phenomenon with beta-lactams. At the same time, the IE has not been extensively studied with carbapenem/carbapenemase inhibitor combinations. The antibiotic-to-inhibitor concentration ratio used in susceptibility testing can influence the in vitro activity of the combination. To explore the role of these factors, imipenem/relebactam and doripenem/relebactam MICs were estimated against six Klebsiella pneumoniae carbapenemase (KPC)-producing Klebsiella pneumoniae strains at standard inocula (SI) and high inocula (HI) by two methods: with a fixed relebactam concentration and with a fixed, pharmacokinetic-based carbapenem-to-relebactam concentration ratio. The combination MICs at HI, compared to SI, increased with most of the tested strains. However, the IE occurred with only two K. pneumoniae strains regardless of the MIC testing method. The relationship between the MICs at SI and the respective inoculum-induced MIC changes was observed when the MICs were estimated at pharmacokinetic-based carbapenem-to-relebactam concentration ratios. Thus, (1) IE was observed with both carbapenem/relebactam combinations regardless of the MIC testing method; however, IE was not observed frequently among tested K. pneumoniae strains. (2) At HI, carbapenem/relebactam combination MICs increased to levels associated with carbapenem resistance. (3) Combination MICs determined at pharmacokinetic-based carbapenem-to-inhibitor concentration ratios predict susceptibility elevations at HI in KPC-producing K. pneumoniae.

The development and implementation of diagnostic methods that allow rapid assessment of antibiotic activity against pathogenic microorganisms is an important step towards antibiotic therapy optimization and increase in the likelihood of successful treatment outcome. To determine whether fluorescence microscopy with acridine orange can be used for rapid assessment (≤8 h) of the meropenem activity against Klebsiella pneumoniae, six isolates including three OXA-48-carbapenemase-producers were exposed to meropenem at different levels of its concentration (0.5 × MIC, 1 × MIC, 8 or 16 µg/mL) and the changes in the viable counts within 24 h were evaluated using fluorescence microscopy and a control culture method. The approach was to capture the regrowth of bacteria as early as possible. Within the first 8 h fluorescence microscopy allowed to categorize 5 out of 6 K. pneumoniae strains by their meropenem susceptibility (based on the MIC breakpoint of 8 mg/L), but meropenem activity against three isolates, two of which were OXA-48-producers, could not be accurately determined at 8 h. The method proposed in our study requires improvement in terms of accelerating the bacterial growth and regrowth for early meropenem MIC determination. Volume-dependent elevation in meropenem MICs against OXA-48-producers was found and this phenomenon should be studied further.

Resistance to carbapenems has become a problem due to Klebsiella pneumoniae (K. pneumoniae), harboring carbapenemases. Among them, there are isolates that are recognized as carbapenem-susceptible; however, these carbapenemase-producing strains with low meropenem minimal inhibitory concentrations (MICs) may pose a threat to public health. We aimed to investigate the impact of the ability to produce carbapenemases by a bacterial isolate on the effectiveness of meropenem in the hollow-fiber infection model. K. pneumoniae and Escherichia coli (E. coli) strains with equal meropenem MICs but differing in their ability to produce carbapenemases were used in pharmacodynamic simulations with meropenem. In addition to standard MIC determination, we assessed the MICs against tested strains at high inoculum density to test if the inoculum effect occurs. According to pharmacodynamic data, the carbapenemase-producing strains were characterized with a relatively decreased meropenem effectiveness compared to non-producers. Meanwhile, the effect of meropenem perfectly correlated with the meropenem exposure expressed as the DOSE/MIC ratio when high-inoculum (HI) MICs but not standard-inoculum (SI) MICs were used for regression analysis. It could be concluded that meropenem-susceptible carbapenemase-producing strains may not respond to meropenem therapy; the antibiotic inoculum effect (IE) may have a prognostic value to reveal the meropenem-susceptible Enterobacterales that harbor carbapenemase genes.

Background/Objectives: Assessing antibiotic MICs at high bacterial counts is likely to disclose hidden bacterial resistance and the inoculum effect if present and therefore also reveal potential decreased antibiotic effectiveness. In the current study, we evaluated the predictive potential of MICs determined at high bacterial inocula to evaluate meropenem effectiveness and emergence of resistance in Klebsiella pneumoniae. Methods: Nine carbapenemase-free or carbapenemase-producing K. pneumoniae strains were exposed to meropenem in an in vitro hollow-fiber infection model (HFIM). The treatment effects were correlated with simulated antibiotic ratios of the area under the concentration–time curve (AUC) to the MIC (AUC/MIC) and to MICs determined at high inocula (AUC/MIC_HI). Results: Based on MICs determined at standard inocula, meropenem effects at different AUC/MIC ratios for both carbapenemase-free and carbapenemase-producing K. pneumoniae strains were stratified and could not be described by a single relationship. In contrast, when AUC/MIC_HI ratios were used, a single relationship with the antibiotic effect was obtained for all tested strains. Similarly, the emergence of meropenem resistance in HFIM was concordant with AUC/MIC_HI, but not with AUC/MIC ratios. Conclusions: MICs determined at high bacterial inocula enable the prediction of meropenem effects both for carbapenemase-free and for carbapenemase-producing K. pneumoniae strains. Also, MICs at standard and high inocula can identify carbapenemase-producing strains by revealing the inoculum effect.