Colistin and Polymyxin B Minimal Inhibitory Concentrations Determined by Etest Found Unreliable for Gram-Negative Bacilli

Ochsner J. 2017 Fall;17(3):239-242.


Background: A reliable method of polymyxin B and E (colistin) susceptibility testing remains elusive. These drugs diffuse poorly into agar, creating potentially inaccurate Etest and disk diffusion results, and testing by these methods is not recommended. Broth microdilution is the reference testing method, although it can be sometimes difficult to interpret. Currently, when a colistin susceptibility test is ordered for a patient in the Ochsner Health System, our diagnostic microbiology laboratory performs the Etest. As an in-house quality assessment project, we compared colistin and polymyxin B minimal inhibitory concentrations (MICs) determined by Etest with MICs determined by broth microdilution to evaluate whether colistin MICs are accurately being reported by Etest.

Methods: A total of 143 nonduplicate clinical isolates from Ochsner patients during 2015-2016 were tested: Acinetobacter baumannii (n=60), Pseudomonas aeruginosa (n=44), and Enterobacteriaceae (n=39) (13 Escherichia coli, 15 Klebsiella spp, and 11 Enterobacter spp). Colistin and polymyxin B MICs were determined by Etest and broth microdilution.

Results: Using broth microdilution, 16/143 (11%) isolates were nonsusceptible to colistin, and 12/143 (8%) were nonsusceptible to polymyxin B. With Etest, 4/143 (3%) isolates were nonsusceptible to colistin, and 7/143 (5%) were nonsusceptible to polymyxin B. Essential agreement of colistin and polymyxin B MICs between broth microdilution and Etest was 84/143 (59%) and 87/143 (61%), respectively. Categorical agreement for colistin and polymyxin B was 127/143 (89%) and 126/143 (88%), respectively.

Conclusion: We found a high rate of discrepancy between colistin and polymyxin B Etest and broth microdilution MICs. Very major errors (colistin/polymyxin B-susceptible by Etest, colistin/polymyxin B-resistant by broth microdilution) were detected in 10% of isolates tested with colistin and 8% of polymyxin B-tested isolates. The data from this study confirm that broth microdilution should be performed for susceptibility testing of polymyxins.

Keywords: Colistin; gram-negative bacteria; polymyxin B.