[False resistance to imipenem in gram negative bacilli with and automatized system]

The aim of this study was to evaluate the reliability of MIC values of imipenem against gramnegative rods obtained with the automated system WalkAway-98 (MicroScan, Dade, USA).

One-hundred and seventy three consecutive clinical isolates of Gram-negative rods for which the MIC of imipenem were > or = 4 mg/l (Urine-Combo 6I panels, U6I) or > or = 8 mg/l (Neg-Combo 6I panels, N6I) were evaluated, including 104 non-fermenting gram-negative rods (NFGNR) and 69 enterobacteria. Panels were inoculated and read according to manufacturer's instructions. Microdilution, according to NCCLS guidelines, was used as the method of reference. MIC of imipenem determined by WalllAway-96 and microdilution differing > or = 2 dilution steps from those obtained with mirodilution were considered as discrepant results. Discrepancies in clinical categories were also evaluated by calculating three types of errors: very major (false susceptibility), major (false resistance) and minor (either susceptible or resistant by one method and intermediate by the other one).

The percentages of discrepancies in the MIC of imipenem determined with U6I panels were 74% and 84% for NFGNR and enterobacteria, respectively. No very major errors were detected. Major errors were observed for 6% and 12% of the strains with U6I panels in NFGNR and enterobacteria, respectively, and in 12% (NFGNR) and 50% (enterobacteria) with N61 panels. With U61 panels minor errors were observed in 11% and 25% of NFGNR and enterobacteria, respectively, while with N61 panels minor error were observed in 39% and 45% of both groups, respectively.

MIC of imipenem > or = 4 mg/l obtained with the WalkAway-96 system against gramnegative rods, particularly in the case of enterobacteria, should be confirmed with a reference susceptibility method.