Background: For an optimum treatment of infections, appropriate antimicrobials should be selected according to the results of antibiotic susceptibility test (AST). However, the present AST does not take into account of antimicrobial concentrations in tissues, although different tissues have different distribution of antimicrobials. Thereby we intended to evaluate the usefulness of interpreting antimicrobial susceptibility depending on tissue concentrations of antimicrobials. 

Methods: Gram-negative bacilli isolated from clinical specimens at Yeungnam University Hospital during the period from January to July, 2006 were evaluated retrospectively. The data on blood concentration, half life and tissue distribution of antimicrobials with variable administration route and dosage were collected and arranged in the forms of previous reports. The diameters of the zone of inhibition from the disc diffusion method were converted to minimum inhibitory concentration (MIC) and the organism was regarded as resistant if the converted concentration was higher than the expected concentration in the tissue. 

Results: Among the data reported as susceptible, antimicrobial concentrations in peritoneal fluid and bile showed a relatively good relationship with AST. But, aminoglycosides and carbenicllin concentrations in wounds and respiratory tissues were shown to be inadequate, thus resulting in a low bacteriologic cure. In cerebrospinal fluid, ciprofloxacin was less effective regardless of dosage. 

Conclusion: Antimicrobial concentration is variable in different tissues and more information on antimicrobial tissue distribution is needed for the appropriate treatment of infections. Reporting of MIC rather than AST with breakpoints should be considered for selection of antimicrobials. Therefore, an interpretation of AST in consideration of the tissue concentration would be more helpful for prevention of major errors and control of infections. (Korean J Clin Microbiol 2006;9(2):125-130)