We evaluated the Vitek2, Etest, and MIC Test Strip (MTS) methods of tigecycline susceptibility testing with 241 expanded-spectrum cephalosporin-resistant and/or carbapenem-resistant Enterobacteriaceae and Acinetobacter baumannii clinical isolates by using dry-form broth microdilution (BMD) as the reference method. The MIC 50/90 s were as follows: BMD, 1/4 g/ml; Vitek2, 4/>8 g/ml; Etest, 2/4 g/ml; MTS, 0.5/2 g/ml. Vitek2 produced 9.1/21.2% major errors, Etest produced 0.4/0.8% major errors, and MTS produced no major errors but 0.4/3.3% very major errors (FDA/EUCAST breakpoints). Vitek2 tigecycline results require confirmation by BMD or Etest for multidrug-resistant pathogens.
Carbapenem resistance has steadily increased in several regions, representing the most significant resistance issue among multidrug-resistant (MDR) Gram-negative pathogens (2, 17, 18). Tigecycline and colistin are among the few antimicrobials active and are commonly used for infections caused by carbapenem-resistant (CR) Enterobacteriaceae and Acinetobacter baumannii (2,4,16,21).The increasing clinical use of tigecycline necessitates rapid, simple, and accurate susceptibility testing methods. Several methods have been evaluated for routine tigecycline susceptibility testing (1,12,14,19,22) with broth microdilution (BMD) as the reference method. It has been previously noted that discrepancies may exist when different methods are used (3,7,14,19,22). It was also reported that disk diffusion and Etest have a poor correlation with BMD, especially for A. baumannii (3,10,12,22). Furthermore, the accuracy of the Vitek2 automated system (bioMérieux, Marcy l' Etoile, France) has not been adequately evaluated for tigecycline (15).In this study, we evaluated three routine tigecycline susceptibility methods, Vitek2, Etest (bioMérieux), and MIC Test Strip (MTS; Liofilchem SRL, Roseto degli Abruzzi, Italy) in comparison with BMD. These methods were applied to a large collection of CR Enterobacteriaceae and A. baumannii and expanded-spectrum cephalosporin-resistant (ESCR) Enterobacteriaceae isolates.Bacterial isolates. This study included 241 clinical isolates recovered during 2008 to 2011 from patients in five tertiary-care hospitals located in different regions of Greece, consisting of CR (K. pneumoniae carbapenemase [KPC]-producing K. pneumoniae, n ϭ 73; VIM-producing K. pneumoniae, n ϭ 39; KPCand VIM-producing K. pneumoniae, n ϭ 13; OXA-58-producing A. baumannii, n ϭ 56) and ESCR isolates (extended-spectrum -lactamase [ESBL]-producing Escherichia coli, n ϭ 20; ESBLproducing K. pneumoniae, n ϭ 20; Enterobacter spp., n ϭ 20). Identification was performed with Vitek2. Phenotypic carbapenemase detection was performed as described previously (23). Common broad-spectrum -lactamase genes (for KPC, OXA-48, OXA-58, metallo--lactamases, and ESBLs) were sought by PCR (20).
Susceptibility testing methods.The FDA-cleared commercial dry-form microtiter panels and cation-adjusted Mueller-Hinton broth with N-Tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid (Sen...