Braoios, A.Pizzolitto, Antonio Carlos [UNESP]2014-05-272014-05-272004-12-01Revista de Ciencias Farmaceuticas, v. 25, n. 1, p. 41-45, 2004.0101-3793http://hdl.handle.net/11449/68056A hundred and ninety-five (195) strains of Enterobacter spp., isolated from diverse clinical specimens - urine, feces, cateter, blood, wound, tracheal aspirate, vaginal fluid - were submitted to the conventional identification by biochemical tests, and were also submitted to the identification by panels NegCombo 20 of the system automated MicroScan - AutoScan- 4 (Dade Behring Inc., West Sacramento, CA, USA). The samples were from patients of the Clinical Laboratory from the School of Pharmacy and Biochemistry of UNOESTE, Presidente Prudente, SP, and from patients hospitalized at the University Hospital Domingos Leonardo Cerávolo, UNOESTE. Of the total of strains tested, 191 (97.9%) presented agreement between the two approaches utilized and 4 strains (2.1%) presented identification disagreement, that is, the genus identified was different in each approach. By this study, the conclusion is that both the approaches utilized for the identification presented advantages and disadvantages related to the cost, facility of execution, quickness, reliability and some other characteristics. Even so, our results showed that conventional methods represent a reliable tool for Enterobacter identification.41-45porAutomated identificationBiochemical identificationEnterobacterEnterobacteriaceaeMicroscanbacterial strainbacterium detectionbacterium identificationbacterium isolatebacterium isolationbloodcontrolled studyfeces analysisgenusintermethod comparisonlaboratory automationnonhumantracheobronchial toiletuniversity hospitalurineurine cathetervagina dischargewoundArtigoAcesso restrito2-s2.0-27444439542