Evaluation of the automicrobic system for susceptibility testing of Pseudomonas aeruginosa to gentamicin, tobramycin, and amikacin.

The AutoMicrobic system (AMS; Vitek Systems, Inc., Hazelwood, Mo.) was studied for its ability to produce accurate and precise MIC interpretations for Pseudomonas aeruginosa susceptibility to gentamicin, tobramycin, and amikacin. MICs were determined in parallel on 200 selected P. aeruginosa isolates by using the AMS discrete-integer MIC program AMS p12.ROB for interpretation of the AMS Gram-Negative General Susceptibility Urinary Card, and a reference small-integer broth microdilution test. Parallel AMS and broth microdilution MICs were also replicated for three selected strains of P. aeruginosa for which MICs were representative of the dilution test ranges. For the 200 P. aeruginosa isolates, mean AMS MICs were significantly larger than the reference test mean MICs, coefficients of variation were approximately double those of the reference test, and correlation coefficients were unacceptably low for each antimicrobial agent. MIC replication studies for the three selected P. aeruginosa strains showed comparable AMS and reference mean MICs in the lower portions of the dilution ranges, significantly higher AMS mean MICs in the upper portions, and mean coefficients of variation of 63 and 9.6%, respectively, for replicated AMS and reference MICs. These results indicate that the AMS, in its present stage of development, does not produce acceptable MIC measurements for P. aeruginosa susceptibility to gentamicin, tobramycin, and amikacin.