In vitro pharmacodynamics of various antibiotics in combination against extensively drug-resistant Klebsiella pneumoniae

Abstract

Extensively drug-resistant (XDR) Klebsiella pneumoniae is an emerging pathogen in Singapore. With limited therapeutic options available, combination antibiotics may be the only viable option. In this study, we aimed to elucidate effective antibiotic combinations against XDR K. pneumoniae isolates. Six NDM-1-producing and two OXA-181-producing K. pneumoniae strains were exposed to 12 antibiotics alone and in combination via time-kill studies. A hollow-fiber infection model (HFIM) with pharmacokinetic validation was used to simulate clinically relevant tigecycline-plus-meropenem dosing regimens against 2 XDR K. pneumoniae isolates over 240 h. The emergence of resistance against tigecycline was quantified using drug-free and selective (tigecycline at 3× the MIC) media. The in vitro growth rates were determined and serial passages on drug-free and selective media were carried out on resistant isolates obtained at 240 h. Both the polymyxin B and tigecycline MICs ranged from 1 to 4 mg/liter. In single time-kill studies, all antibiotics alone demonstrated regrowth at 24 h, except for polymyxin B against 2 isolates. Tigecycline plus meropenem was found to be bactericidal in 50% of the isolates. For the isolates that produced OXA-181-like carbapenemases, none of the 55 tested antibiotic combinations was bactericidal. Against 2 isolates in the HFIM, tigecycline plus meropenem achieved a >90% reduction in bacterial burden for 96 h before regrowth was observed until 10(9) CFU/ml at 240 h. Phenotypically stable and resistant isolates, which were recovered from tigecycline-supplemented plates post-HFIM studies, had lower growth rates than those of their respective parent isolates, possibly implying a substantial biofitness deficit in this population. We found that tigecycline plus meropenem may be a potential antibiotic combination for XDR K. pneumoniae infections, but its efficacy was strain specific.

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

Figures

FIG 1

Representative time-kill studies against K. pneumoniae 3229, depicting single antibiotics with bactericidal activity.

FIG 2

Representative time-kill studies against K. pneumoniae strains 1883 (A), 3229 (B), 7433 (C), 7522 (D), 43320 (E), and 44591 (F) depicting bactericidal combinations.

FIG 3

Microbiologic response observed in the following infection models: meropenem, tigecycline (T), and meropenem plus tigecycline against K. pneumoniae strains 7522 (A) and 44591 (B). q8h, every 8 h.

FIG 4

Typical observed pharmacokinetic profiles in the following infection models: tigecycline at 100 mg every 12 h (A) and meropenem at 2 g every 8 h as a 3-h infusion (B).