Comparison of E test with standard broth microdilution for determining antibiotic susceptibilities of penicillin-resistant strains of Streptococcus pneumoniae.

We compared the E test (AB Biodisk North America, Inc., Culver City, Calif.) with the National Committee for Clinical Laboratory Standards broth microdilution method for the determination of MICs of penicillin and cefotaxime for 108 isolates of Streptococcus pneumoniae. The E test was performed following manufacturer's recommendations with Mueller-Hinton blood agar, and the broth microdilution procedure was performed with lysed horse blood-supplemented Mueller-Hinton broth. The microdilution method classified 26 isolates as highly penicillin resistant (MIC, > or = 2 micrograms/ml), 33 as intermediately resistant to penicillin (MIC, > or = 0.1 < 2.0 micrograms/ml), and 49 as susceptible to penicillin (MIC, < 0.1 micrograms/ml). Discordant results obtained with the E test for penicillin susceptibility testing compared with broth microdilution occurred for 19 of the 108 isolates tested. Cefotaxime MICs for 90% of isolates found highly resistant, intermediately resistant, and susceptible to penicillin by broth microdilution were 2.0, 0.5, and 0.06 micrograms/ml, respectively. There were 16 susceptibility category changes when the E test was used to determine cefotaxime MICs. All of the discrepancies in the penicillin and cefotaxime MICs determined by the E test occurred at the susceptibility category breakpoints, and all represented differences of only one twofold dilution factor. Properly performed and controlled, the E test should be a reliable quantitative procedure for more accurately predicting the susceptibility of S. pneumoniae to several antibiotics.