Comparative assessment of antimicrobial susceptibility testing for tigecycline and colistin against Acinetobacter baumannii clinical isolates, including multidrug-resistant isolates.

Acinetobacter baumannii has become a serious concern in clinical practice owing to its multiple resistance to antimicrobial agents. Tigecycline and colistin may be used as alternative therapies, although they lack practical susceptibility testing guidelines. This study assessed the reliability of commonly used methods (disc diffusion, Etest and VITEK(®) 2) for testing sensitivity to both agents compared with the reference broth microdilution (BMD) method against 290 A. baumannii clinical isolates, including multidrug-resistant isolates. For tigecycline, essential agreement and categorical agreement (CA) of minimum inhibitory concentration (MIC) testing were most correlated with BMD when using a breakpoint of susceptible (S)≤1/resistant (R)>2 mg/L; 94.8% and 84.5% (Etest) and 99.3% and 75.5% (VITEK 2), respectively. A disc diffusion zone diameter breakpoint of S≥17/R≤12 mm showed good agreement. All three methods did not show major errors or very major errors. For colistin, a BMD MIC breakpoint of S≤2/R>4 mg/L was proposed. The disc diffusion method was highly reproducible with a zone diameter breakpoint of S≥12/R≤9 mm. However, Etest results showed a different MIC range, and the MIC breakpoint should be modified to S≤0.5/R>2 mg/L, whilst a similar MIC breakpoint to BMD could be applied for VITEK 2. Both Etest and VITEK 2 showed a high CA for isolates with colistin-susceptible and -resistant results. We recommend that disc diffusion, Etest and VITEK 2 may be used with caution for testing tigecycline and colistin based on our proposed breakpoints. The reliability of individual methods will be discussed.