Automated time-lapse microscopy a novel method for screening of antibiotic combination effects against multidrug-resistant Gram-negative bacteria

OBJECTIVES

Antibiotic combinations are often used for carbapenemase-producing Enterobacteriaceae (CPE) but more data are needed on the optimal selection of drugs. This study aimed to evaluate the feasibility of a novel automated method based on time-lapse microscopy (the oCelloScope, Philips BioCell A/S, Allerød, Denmark) to determine in vitro combination effects against CPE and to discuss advantages and limitations of the oCelloScope in relation to standard methods.

METHODS

Four Klebsiella pneumoniae and two Escherichia coli were exposed to colistin, meropenem, rifampin and tigecycline, alone and in combination. In the oCelloScope experiments, a background corrected absorption (BCA) value of ≤8 at 24 h was used as a primary cut-off indicating inhibition of bacterial growth. A new approach was used to determine synergy, indifference and antagonism based on the number of objects (bacteria) in the images. Static time-kill experiments were performed for comparison.

RESULTS

The time-kill experiments showed synergy with 12 of 36 regimens, most frequently with colistin plus rifampin. BCA values ≤8 consistently correlated with 24-h bacterial concentrations ≤6 log₁₀ CFU/mL. The classification of combination effects agreed with the time-kill results for 33 of 36 regimens. In three cases, the interactions could not be classified with the microscopy method because of low object counts.

CONCLUSIONS

Automated time-lapse microscopy can accurately determine the effects of antibiotic combinations. The novel method is highly efficient compared with time-kill experiments, more informative than checkerboards and can be useful to accelerate the screening for combinations active against multidrug-resistant Gram-negative bacteria.

Evaluation of antibacterial activities of colistin, rifampicin and meropenem combinations against NDM-1-producing Klebsiella pneumoniae in 24 h in vitro time-kill experiments

OBJECTIVES

To investigate the activity of colistin alone or in double and triple combination with rifampicin and meropenem against NDM-1-producing Klebsiella pneumoniae.

METHODS

Eight isolates of NDM-1-producing K. pneumoniae were exposed to clinically relevant antibiotic concentrations in 24 h time-kill experiments. Three colistin concentrations were used for two of the strains. Resistance development was assessed with population analysis and sequencing of the mgrB and pmrB genes.

RESULTS

Initial killing was achieved with colistin alone, but with considerable regrowth at 24 h. Combinations including colistin and rifampicin were bacteriostatic or bactericidal against all strains. Colistin plus meropenem was bactericidal against one strain, but, overall, meropenem showed little additive effects. Higher concentrations of colistin did not enhance antibacterial activity. Resistant populations and deletion or mutations in the mgrB and pmrB genes were frequently detected in endpoint samples after exposure to colistin alone.

CONCLUSIONS

Based on the results of this and previous studies, the combination of colistin and rifampicin seems promising and should be further explored in vivo and considered for clinical evaluation. Meropenem seems less useful in the treatment of infections caused by high-level carbapenem-resistant NDM-1-producing K. pneumoniae. Higher colistin concentrations did not result in significantly better activity, suggesting that combination therapy might be superior to monotherapy also when colistin is prescribed using high-dose regimens in accordance with current recommendations.