Accuracy of Commercial and Reference Susceptibility Testing Methods for Detecting Vancomycin-Intermediate Staphylococcus aureus

We compared the results obtained with six commercial MIC test systems (Etest, MicroScan, Phoenix, Sensititre, Vitek Legacy, and Vitek 2 systems) and three reference methods (agar dilution, disk diffusion, and vancomycin [VA] agar screen [VScr]) with the results obtained by the Clinical and Laboratory Standards Institute broth microdilution (BMD) reference method for the detection of VA-intermediate Staphylococcus aureus (VISA). A total of 129 S. aureus isolates (VA MICs by previous BMD tests, ≤1 μg/ml [n = 60 strains], 2 μg/ml [n = 24], 4 μg/ml [n = 36], or 8 μg/ml [n = 9]) were selected from the Centers for Disease Control and Prevention strain collection. The results of BMD with Difco Mueller-Hinton broth were used as the standard for data analysis. Essential agreement (percent ±1 dilution) ranged from 98 to 100% for all methods except the method with the Vitek Legacy system, for which it was 90.6%. Of the six commercial MIC systems tested, the Sensititre, Vitek Legacy, and Vitek 2 systems tended to categorize VISA strains as susceptible (i.e., they undercalled resistance); the MicroScan and Phoenix systems and Etest tended to categorize susceptible strains as VISA; and the Vitek Legacy system tended to categorize VISA strains as resistant (i.e., it overcalled resistance). Disk diffusion categorized all VISA strains as susceptible. No susceptible strains (MICs ≤ 2 μg/ml) grew on the VScr, but all strains for which the VA MICs were 8 μg/ml grew on the VScr. Only 12 (33.3%) strains for which the VA MICs were 4 μg/ml grew on VScr. The differentiation of isolates for which the VA MICs were 2 or 4 μg/ml was difficult for most systems and methods, including the reference methods.In January 2006, the Clinical and Laboratory Standards Institute (CLSI) published new interpretive criteria for vancomycin and Staphylococcus aureus. The breakpoints were lowered from ≤4 μg/ml to ≤2 μg/ml for susceptible, 8 to 16 μg/ml to 4 to 8 μg/ml for intermediate, and ≥32 μg/ml to ≥16 μg/ml for resistant (2). The vancomycin breakpoints for coagulase-negative staphylococci were not changed. The rationale for lowering the S. aureus intermediate breakpoint to 4 μg/ml was (i) that intermediate S. aureus isolates, although they are rare, likely represented a population of organisms that demonstrate heteroresistance, and (ii) limited outcome data suggested that infections with these isolates are likely to fail vancomycin therapy (9). The results of broth microdilution performed by use of the CLSI reference method were the primary S. aureus susceptibility data evaluated before the CLSI breakpoint change was made. We undertook the study described here to determine the accuracy of commercial systems and reference methods for the detection of decreased vancomycin susceptibility among isolates of S. aureus.(This work was presented in part at the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, IL, 17 to 20 September 2007.)