Identification of clinical isolates of gram-negative nonfermentative bacteria by an automated cellular fatty acid identification system

Publisher Website
Google Scholar
Abstract
An automated cellular fatty acid (CFA) bacterial identification system, Microbial Identification System (MIS; Microbial ID, Newark, Del.), was compared with a conventional system for the identification of 573 strains of gram-negative nonfermentative bacteria. MIS identifications were based exclusively on the CFA composition following 22 to 26 h of growth at 28 degrees C on Trypticase soy agar. MIS identifications were listed with a confidence measurement (similarity index [SI]) on a scale of 0 to 1.0. A value of greater than or equal to 0.5 was considered a good match. The MIS correctly listed as the first choice 478 of 532 (90%) strains contained in the data base. However, only 314 (59%) had SI values of greater than or equal to 0.5. Of the 54 strains in which there was not agreement, 37 belonged to the genera Acinetobacter, Moraxella, or Alcaligenes or were Pseudomonas pickettii. Reproducibility studies suggest that SI variation is most likely a function of a difference in culture age at the time of analysis, which is due to the relatively low temperature and time of incubation. Other discrepancies were attributable to insufficiently characterized library entries or an inability to differentiate chemotaxonomically closely related species. The MIS, as the first automated CFA identification system, is an accurate, efficient, and relatively rapid method for the identification of gram-negative nonfermentative bacteria. The development of a CFA library with the media and incubation conditions routinely used for the isolation of clinical pathogens could further decrease the identification time and provide an increase in accuracy.