Background: In view of the paucity of clinical evidence, in vitro studies are needed to find antibiotic combinations effective against multidrug-resistant Gram-negative bacteria. Interpretation of in vitro effects is usually based on bacterial growth after 24 h in time-kill and checkerboard experiments. However, the clinical relevance of the effects observed in vitro is not established. In this study we explored alternative output parameters to assess the activities of colistin and meropenem against Pseudomonas aeruginosa and Acinetobacter baumannii.
Methods: Four strains each of P. aeruginosa and A. baumannii were exposed to colistin and meropenem, alone and in combination, in 8 h dynamic time-kill experiments. Initial (1 h), maximum and 8 h bacterial reductions and the area under the bacterial time-kill curve were evaluated. Checkerboards, interpreted based on fractional inhibitory concentration indices after 24 h, were performed for comparison.
Results: In the time-kill experiments, the combination resulted in enhanced 1 h, maximum and 8 h bacterial reductions against 2, 3 and 5 of 8 strains, respectively, as compared to the single drugs. A statistically significant reduction in the area under the time-kill curve was observed for three strains. In contrast, the checkerboards did not identify synergy for any of the strains.
Conclusions: Combination effects were frequently found with colistin and meropenem against P. aeruginosa and A. baumannii in time-kill experiments but were not detected with the checkerboard method. We propose that the early dynamics of bacterial killing and growth, which may be of great clinical importance, should be considered in future in vitro combination studies.
Keywords: Gram-negative bacteria; checkerboards; synergy; time-kill experiments.