E Matuschek1, J Åhman2, C Webster2, G Kahlmeter2. 1. EUCAST Development Laboratory, Växjö, Sweden. Electronic address: erika.matuschek@kronoberg.se. 2. EUCAST Development Laboratory, Växjö, Sweden.
Abstract
OBJECTIVE: Both EUCAST and CLSI recommend broth microdilution (BMD) for antimicrobial susceptibility testing of colistin, but BMD is rarely used in routine microbiology laboratories. The objective of this study was to evaluate five commercially available BMD products and two brands of gradient tests for colistin MIC determination using BMD according to ISO standard 20776-1 as reference. METHODS: Colistin MIC determination was performed according to the manufacturer's instructions on five commercially available BMD products (Sensititre, MICRONAUT-S, MICRONAUT MIC-Strip, SensiTest, and UMIC) and two gradient tests (Etest and MIC Test Strip). Colistin reference MICs were determined using frozen panels according to ISO standard 20776-1. An international collection of Gram-negative bacteria (n=75) with varying levels of colistin susceptibility was tested. RESULTS: The colistin BMD products correlated well with reference tests, in particular for Sensititre and the two MICRONAUT products (essential agreement ≥96%: 66/69 (96%, CI 88-99%), 72/75 (96%, CI 88-99%) and 74/75 (99%, CI 92-100%)). The results were somewhat poorer for the BMD products SensiTest and UMIC: EA 88% (51/58, CI 77-95%) and 82% (61/74, CI 72-89%), respectively), and considerably poorer for the gradient tests (EA 43-71% depending on gradient test and Mueller-Hinton agar manufacturer). The gradient tests generally underestimated colistin MICs, resulting in a significant number of false susceptible results (9-18 of total 75 tests, compared with 1-3 for the BMD products). CONCLUSIONS: Based on the results of this study, we advise laboratories not to trust gradient tests for colistin susceptibility testing and to use broth microdilution methods for this purpose. There are several commercial broth microdilution tests available and in principle they perform well.
OBJECTIVE: Both EUCAST and CLSI recommend broth microdilution (BMD) for antimicrobial susceptibility testing of colistin, but BMD is rarely used in routine microbiology laboratories. The objective of this study was to evaluate five commercially available BMD products and two brands of gradient tests for colistin MIC determination using BMD according to ISO standard 20776-1 as reference. METHODS: Colistin MIC determination was performed according to the manufacturer's instructions on five commercially available BMD products (Sensititre, MICRONAUT-S, MICRONAUT MIC-Strip, SensiTest, and UMIC) and two gradient tests (Etest and MIC Test Strip). Colistin reference MICs were determined using frozen panels according to ISO standard 20776-1. An international collection of Gram-negative bacteria (n=75) with varying levels of colistin susceptibility was tested. RESULTS: The colistin BMD products correlated well with reference tests, in particular for Sensititre and the two MICRONAUT products (essential agreement ≥96%: 66/69 (96%, CI 88-99%), 72/75 (96%, CI 88-99%) and 74/75 (99%, CI 92-100%)). The results were somewhat poorer for the BMD products SensiTest and UMIC: EA 88% (51/58, CI 77-95%) and 82% (61/74, CI 72-89%), respectively), and considerably poorer for the gradient tests (EA 43-71% depending on gradient test and Mueller-Hinton agar manufacturer). The gradient tests generally underestimated colistin MICs, resulting in a significant number of false susceptible results (9-18 of total 75 tests, compared with 1-3 for the BMD products). CONCLUSIONS: Based on the results of this study, we advise laboratories not to trust gradient tests for colistin susceptibility testing and to use broth microdilution methods for this purpose. There are several commercial broth microdilution tests available and in principle they perform well.