Antimicrobial susceptibility of flavobacteria as determined by agar dilution and disk diffusion methods

A total of 106 clinical isolates of flavobacteria, including 41 isolates of Flavobacterium meningosepticum, 59 of Flavobacterium indologenes, and 6 of Flavobacterium odoratum were collected from January 1992 to December 1995 from patients in Taiwan. The in vitro activities of antimicrobial agents were determined concomitantly by the standard agar dilution and disk diffusion methods. More than 90% of the flavobacterial isolates were resistant to cephalothin, cefotaxime, ceftriaxone, moxalactam, aztreonam, imipenem, aminoglycosides, erythromycin, and glycopeptides. The majority of F. meningosepticum isolates were susceptible to piperacillin and to minocycline but resistant to ceftazidime, with MICs at which 90% of the isolates are inhibited being 8, 4, and >128 mu g/ml, respectively. Approximately half of the F. indologenes isolates were susceptible to piperacillin, cefoperazone, ceftazidime, and minocycline, with MICs at which 50% of the isolates are inhibited being 4, 16, 8, and 4 mu g/ml, respectively. The majority of F. odoratum isolates were resistant to all the antimicrobial agents tested except minocycline, to which five of six isolates were susceptible. With least-squares regression analysis and error rate-bounded analysis methods, the following resistant and susceptible zone diameter breakpoints were established: less than or equal to 12 and greater than or equal to 17 mm, respectively, for piperacillin against F. meningosepticum and F. indologenes; less than or equal to 13 and greater than or equal to 18 mm, respectively, for ceftazidime against F. meningosepticum and F. indologenes, less than or equal to 17 and greater than or equal to 21 mm, respectively, for ofloxacin against F. indologenes; less than or equal to 16 and greater than or equal to 20 mm, respectively, for ciprofloxacin against F. meningosepticum. Valid breakpoints for the disk diffusion method could not be established for cefoperazone and ofloxacin against F. meningosepticum and for minocycline against F. meningosepticum and F. indologenes due to a poor correlation coefficient for the regression line or for cefoperazone and ciprofloxacin against F. indologenes due to the presence of remarkable error rates.