Bakground: In recent years, knowledge of bacterial resistance to antimicobials has expanded in important ways. Availability of an increasing number of antibiotics allows more precise individualization of resistance phenotypes and recording susceptibility results as patterns or phenotypes is valuable for both surveillance and patient care. If the patterns of resistance to panels of related antimicrobials are considered the underlying mechanisms can often be inferred. And the inferred mechanisms make the clinician to be advised to use alternative treatment. Interpretation of resistance phenotypes is based on the comparison of clinical isolates with prototype susceptible bacteria belonging to the same species. But interpretative reading of antimicrobial susceptibility tests requires an immense knowledge of antibiotics. Such interpretative reading is best achieved by computerized expert systems.

Methods: The authors attempt to determine phenotypes for the clinically isolated strains for each class of drugs tested by the Vitek 2 system^TM(bioMerieux, Marcy I’Etoile, France) using the Advanced Expert System^TM(AES, bioMerieux, Marcy I’Etoile, France). A total of 91, 107, 89, 65, 251, 113, 47, 33, 23, 122 and 110 isolates of Staphylococcus aureus, coagulase negative staphylococci, Enterococcus faecalis, Enterococcus facium, Escherichia coli, Klebsiella pneumoniae, Serratia marcescens, Enterobacter cloacae, Enterobacter aerogenes, Pseudomonas aeruginosae and Acinetobacter baumannii, were examined respectively.

Results: Biological correction based on the phenotype was recommended from 2.2% of E. faecalis to 46.8% of S. marcescens and therapeutic correction, from 7.3% of A. baumannii to 60.9% of E. aerogenes. A total of 25, 26, 18, 19, 22, 22, 15, 15, 17, 19, 19 phenotypes of S. aureus, coagulase negative staphylococci, E. faecalis, E. facium, E. coli, K. pneumoniae, S. marcescens, E. cloacae, E. aerogenes, P. aeruginosa and A. baumannii, were detected respectively. Association of resistance mechanism from S. aureus, coagulase negative staphylococci, E. coli, K. pneumoniae, S. marcescens, show 10, 11, 6, 4 and 3 pairs from resistant phenotypes, respectively.

Conclusion: Vitek AES potentially provides a tool to assist the development of antimicrobial susceptibility interpretation in the clinical microbiology laboratory. The inferred mechanisms make the clinician to be advised to use alternative treatment. (Korean J Clin Microbiol 2002;5(2):84-96)