Search

A multilaboratory collaborative study was undertaken to determine whether the anaerobic disk diffusion test of Horn et al. could be performed reproducibly and accurately. Tests with nine different antimicrobial disks were evaluated. Reproducibility of the agar diffusion disk method was similar to that of the reference agar dilution test procedure. The anaerobic disk diffusion procedure was found to be a potentially useful method for testing some antimicrobial agents against rapidly growing anaerobes belonging to the Bacteroides fragilis group. These promising results warrant further investigations and validations.

To assess the in vitro activity of flomoxef (6315-S), moxalactam, cefoxitin, cefotaxime, and clindamycin against anaerobes 197 clinical isolates (27 Bacteroides fragilis, 42 B. thetaiotaomicron, 10 B. vulgatus, 7 B. ovatus, 6 B. uniformis, 6 B. distasonis, 7 Bacteroides melaninogenicus group, 11 Bacteroides oralis group, 21 Clostridium difficile, 7 C. perfringens, 3 C. sporogenes, 3 Clostridium spp., 33 Propionibacterium acnes, 14 Peptococcaceae) were studied by means of agar dilution tests. The MIC90 of B. fragilis was less than 2 micrograms/ml for flomoxef, less than 4 micrograms/ml for moxalactam, less than 16 micrograms/ml for cefoxitin, less than 128 micrograms/ml for cefotaxime and less than 2 micrograms/ml for clindamycin. The respective MIC90's of B. thetaiotaomicron were less than 64, less than 128, less than 32, less than 256 and 8 micrograms/ml. Strains of the other Bacteroides species and groups were more susceptible to flomoxef and the other antibiotics than B. thetaiotaomicron. Against Clostridium difficile flomoxef (MIC90 less than 4 micrograms/ml) proved to be superior to the other agents tested. Most of the Clostridium strains other than C. difficile were also susceptible to flomoxef; anaerobic grampositive cocci and Propionibacterium acnes were very sensitive (MIC90's less than 1 and less than or equal to 0.125 micrograms/ml, respectively). Its anti-anaerobic activity, together with its efficacy against aerobes, should make flomoxef a useful adjunct to the arsenal of modern antibiotic therapy.

In infections of polymicrobial etiology, it seems mandatory to combine an antibiotic with marginal activity against anaerobes with an anti-anaerobic drug, e.g. metronidazole or clindamycin. We investigated the effect of associations of anaerobic and facultatively anaerobic pathogens on MBCs of enoxacin, clindamycin, metronidazole, and combinations of enoxacin plus clindamycin or metronidazole. Single testing, and associations of Bacteroides fragilis with Escherichia coli or Enterococcus faecalis revealed MBCs of 64 mg/l for enoxacin. However, investigating metronidazole, MBCs for B. fragilis increased from 0.5-1 mg/l to 2-4 mg/l in association with E. coli. The combination of enoxacin with metronidazole restored MBCs of 0.5 and 1 mg/l for the B. fragilis strains. B. fragilis associated with E. faecalis showed MBCs of 2 to 64 mg/l for metronidazole, an increase of up to 64-fold. The enoxacin MBCs for E. coli in association with B. fragilis were up to eight-fold higher than for E. coli alone (2 mg/l compared to 0.25 mg/l). Association of E. coli with E. faecalis and B. fragilis showed a low to moderate increase for MBCs of E. coli when enoxacin was tested alone or in combination with metronidazole or clindamycin. In contrast, MBCs for E. faecalis did not change significantly with any of the associations or combinations tested. It is evident that in vitro antimicrobial susceptibility of associations of pathogens can be modified by interactions between strains, and/or antimicrobials.

Clavulanic acid is known to potentiate the activity of amoxycillin and ticarcillin against beta-lactamase-producing Bacteroides species. In order to assess the usefulness of the clavulanate antibiotics in mixed infections with aerobes and anaerobes, artificially associated cultures of clinically significant Bacteroides species and facultative anaerobes (Escherichia coli, Klebsiella pneumoniae, Haemophilus influenzae, Enterococcus faecalis, Staphylococcus spp. and Streptococcus spp.) were tested. The minimal bactericidal concentrations (MBC) of amoxycillin and ticarcillin, singly and in combination with clavulanate, were determined for the individual species in the associated culture and for the associated culture as a whole. In the absence of clavulanate, association experiments showed that the MBCs of susceptible strains increased in the presence of beta-lactamase-producing species, whereas the addition of clavulanate to the test systems led to MBCs which were mostly comparable to values as determined in single strain testing.

Effect of a Combination of Ampicillin and Sulbactam on Clinical Isolates of Anaerobic Bacteria. The antimicrobial susceptibility of 182 recent clinical isolates of anaerobic bacteria to ampicillin alone, ampicillin plus 1 mg/l sulbactam, ampicillin plus 5 mg/l sulbactam, and cefoxitin was studied by means of agar dilution tests. The ampicillin-sulbactam combination (Unacid) was most effective against species of the Bacteroides fragilis group, the MIC90 of ampicillin plus 5 mg/l sulbactam for B. fragilis being less than or equal to 1 mg/l, compared to 256 mg/l of ampicillin, 4 mg/l of ampicillin plus 1 mg/l sulbactam, and 8 mg/l of cefoxitin. No significant difference between ampicillin alone and in combination with sulbactam was observed against gram-positive anaerobic rods or cocci.

Two disposable carbon dioxide/hydrogen gas-generating systems (GasPak Plus and Anaerocult A) were compared by assessing growth of obligate anaerobic bacteria. Eighty strains representing 28 species of anaerobic bacteria commonly occurring at various body sites were seeded onto 4 brain heart chocolate agar plates using a spiral plater; and 1 plate each was subsequently incubated in 2 Anaerocult A and 2 GasPak Plus systems. Bacterial growth was expressed as colony-forming units per ml (CFU/ml), reproducibility of the seeding procedure was checked, and the potential interference of H2S-producing bacteria with operation of the carbon dioxide/hydrogen systems was investigated. The presence of H2S only inhibited bacterial growth in the case of Peptostreptococcus asaccharolyticus - an effect observed in both systems. Reproducibility of the seeding procedure using the spiral plater was within acceptable range. Differences between the systems were not apparent when comparing total CFU/ml of a given species and the systems therefore provide equally effective environments for incubation of anaerobic bacteria. In both systems, however, growth varied from one species or strain to another. In mixed infections, detection of certain species of anaerobes may therefore be difficult using either system.

The reliability of four commercially available broth microdilution systems (ATB ANA, MHK-Anaerob-Biotest, Dynatech MIC system, Sceptor Anaerobe MIC system) for routine susceptibility testing was evaluated using agar dilution and broth macrodilution as reference methods. Using the categories "susceptible", "moderately susceptible" and "resistant", the rate of essential agreement (complete agreement plus minor errors) of all four systems compared to the two reference methods was satisfactory, ranging from 93% (Dynatech system and agar dilution) to 98% (Sceptor system and agar dilution). The rate of complete agreement compared to the agar dilution and broth dilution method respectively was 61% and 60% for the MHK-Anaerob-Biotest, 65% and 63% for the Dynatech system, 72% and 72% for ATB ANA, and 85% and 80% for the Sceptor system. The Sceptor system was thus superior to the other systems. The systems were easy to operate and inoculate, although arrangement of antimicrobial agents in the Dynatech panels was apt to be confusing.

The antimicrobial susceptibility of 195 recent clinical isolates of anaerobic bacteria was studied to ampicillin alone, ampicillin + 1 mg/l sulbactam, ampicillin + 5 mg/l sulbactam, and cefoxitin by means of agar dilution tests. The ampicillin-sulbactam combinations were the most effective drugs against species of the Bacteroides fragilis group, the MIC90 of ampicillin + 5 mg/l sulbactam for B. fragilis being less than 1 mg/l, compared to 256 mg/l of ampicillin, 4 mg/l of ampicillin + 1 mg/l sulbactam, and 8 mg/l of cefoxitin. No significant difference between ampicillin alone and in combination with sulbactam was observed against gram-positive anaerobic rods, Peptococcus spp. and Peptostreptococcus spp. with MIC's less than 2 mg/l.

Susceptibility of anaerobic clinical isolates from the United States, Canada, and Germany to sulbactam/ampicillin (1 + 2) (SBT/AMP), ampicillin alone (AMP), and cefoxitin (CFX) was determined with a standard agar dilution test. The isolates included 192 strains of Bacteroides fragilis, 132 strains of other Bacteroides spp., and 19 strains of Clostridium spp. Against all species tested, SBT/AMP was more active than AMP or CFX. Results obtained by a new model of associative susceptibility testing indicated that susceptibility testing of single pathogens from polymicrobial infections is not necessarily reflective of the susceptibility of the pathogens at the site of infection. The results suggest that sulbactam is effective in associations of pathogens producing beta-lactamases of Richmond-Sykes types II-V and beta-lactamases of anaerobic bacteria.