Evaluation of Calcium-Enhanced Media for Colistin Susceptibility Testing by Gradient Agar Diffusion and Broth Microdilution

Abstract

Despite the increasing reliance on polymyxin antibiotics (polymyxin B and colistin) for treatment of multidrug-resistant Gram-negative infections, many clinical laboratories are unable to perform susceptibility testing due to the lack of accurate and reliable methods. Although gradient agar diffusion is commonly performed for other antimicrobials, its use for polymyxins is discouraged due to poor performance characteristics. Performing gradient agar diffusion with calcium enhancement of susceptibility testing media has been shown to improve the identification of polymyxin-resistant isolates with plasmid-mediated resistance (mcr-1). We therefore sought to evaluate the broad clinical applicability of this approach for colistin susceptibility testing by assessing a large and diverse collection of resistant and susceptible patient isolates collected from multiple U.S. medical centers. Among 217 isolates, the overall categorical and essential agreement for calcium-enhanced gradient agar diffusion were 73.7% and 65.5%, respectively, compared to the results for reference broth microdilution. Performance varied significantly by organism group, with agreement being highest for Enterobacterales and lowest for Pseudomonas aeruginosa. Nevertheless, even for Enterobacterales, there was a high rate of very major errors (9.2%). Performance was similarly poor for calcium-enhanced broth microdilution. While calcium enhancement did allow for more accurate categorization of mcr-1-resistant isolates, there were unacceptably high rates of errors for both susceptible and non-mcr-1-resistant isolates, raising serious doubts about the suitability of these calcium-enhanced methods for routine colistin susceptibility testing in clinical laboratories.

Revisiting colistin susceptibility testing: will adding calcium to Mueller–Hinton agar improve the detection of colistin resistance?

“…We used reference BMD as our gold standard, but as observed by others. [4,6], as well as in our study, the improved separation between susceptible and resistant isolates by adding Ca 2+ to MH agar seemed to be related mainly to resistance mediated by mcr genes. Another limitation of our study is the small number of isolates included.…”

“…In our study, addition of 5 mM Ca 2+ to MH agar increased separation between colistin susceptible and resistant E. coli for both gradient tests and disks, but there was significant overlap between susceptible and resistant isolates of K. pneumoniae, P. aeruginosa and Acinetobacter species. This might be related to the species and/or the colistin resistance mechanism included in the respective studies [6][7][8][9][10]. In our study, some mcr-negative E. coli and K. pneumoniae with colistin MICs of 8-16 mg/L were poorly, or not at all, separated from the susceptible isolates (≤2 mg/L).…”

“…Gwozdzinski et al [4] showed that addition of 5 mM Ca 2+ to Mueller-Hinton (MH) medium for colistin MIC determination by broth microdilution and Etest resulted in better separation between resistant and susceptible isolates of Enterobacterales. Green et al [6] were not able to reproduce these results in a collection of 229 isolates of Enterobacterales, P. aeruginosa and Acinetobacter species. However, to our knowledge, no one has evaluated if addition of lower or higher concentrations of Ca 2+ to the MH medium could improve the test, or if it is possible to develop a screening test for colistin resistance with disk diffusion by adding Ca 2+ to MH agar.…”

The Minimum Inhibitory Concentration of Antibiotics: Methods, Interpretation, Clinical Relevance

“…EA for resistant strains was much lower (13%), although categorical agreement (CA) was higher and stood at 87%. A recent research study from the United States published in 2020 [ 26 ] indicates that calcium supplementation does not bring the expected improvement in the reliability of the gradient method in the determination of colistin MIC, as EA and CA were 65.5% and 73.7% respectively for all Gram-negative organisms tested, including those susceptible and resistant to colistin. Similarly, work is ongoing to assess whether the gradient test actually does not allow a reliable assessment of fosfomycin MIC.…”

Current Update on Intrinsic and Acquired Colistin Resistance Mechanisms in Bacteria

“…Until now, there was no agreement on the methodology for colistin susceptibility testing. Because of the weak diffusion of colistin in agar, the disc diffusion process and gradient tests were inaccurate (91)(92)(93)(94)(95). As a result, disc diffusion and gradient diffusion are ineffective in determining polymyxin susceptibility.…”