[An approach for the quality control in clinical microbiology laboratory by antibiotic susceptibility]

It is important to develop a system for the prevention of various errors and the maintenance of the high quality level in the department of laboratory medicine. We developed a computer program to infer the species of microorganisms by assessing the minimal inhibitory concentration of respective microorganisms to various antibiotics, antibiotic susceptibility patterns(ASP). Retrospective analysis of 3,405 strains from 39 species using this program resulted in a sensitivity of 91.4% and specificity of 98.9%. Furthermore, in 1,660 strains isolated from September 1999 to August 2000, the efficacy of this program was evaluated using both this program and conventional identifying method such as VITEK, api and ATB. The discrepancy between this inferring method and conventional method was noted in 179 strains, revealing errors in 12 strains. We conclude that this ASP inferring program is useful for preventing errors and the maintenance of high quality examination by combining with conventional methods in the laboratory.