Effect of Antibiotic Treatment on Growth of and Toxin Production by Clostridium difficile in the Cecal Contents of Mice

Antibiotics play a crucial role in the pathogenesis of Clostridium difficile-associated diarrhea (CDAD), presumably by disrupting the indigenous microflora of the colon, thereby allowing C. difficile to grow to high concentrations with the production of toxin (2, 7). Antibiotics that have inhibitory activity against C. difficile may suppress the organism during treatment; however, C. difficile overgrowth and disease may develop after completion of therapy during the period of recovery of the indigenous microflora (2, 4, 7). Antibiotics that disrupt the anaerobic component of the microflora may be particularly likely to cause CDAD (1, 2, 21). However, antibiotics which cause relatively little disruption of the anaerobic microflora, such as ciprofloxacin and trimethoprim-sulfamethoxazole, have also been associated with CDAD (7, 23). In this study, we used an in vitro mouse model to examine the effect of antibiotic treatment on growth of C. difficile in cecal contents. We hypothesized that antibiotics that disrupt the anaerobic microflora would facilitate growth and toxin production by C. difficile, whereas antibiotics that do not disrupt the anaerobic microflora would not.

Three strains of C. difficile were studied. Strain 1 was ATCC 9689, strain 2 was a clinical isolate from Cleveland, and strain 3 was ATCC 43593. Strains 1 and 2 produced toxin, whereas strain 3 did not. Broth dilution MICs of the test antibiotics for the three strains were determined using standard methods for susceptibility testing of anaerobic bacteria (11).

The in vitro model that was used was adapted from the in vitro model of colonization resistance to C. difficile infection developed by Borriello et al. (1). These investigators demonstrated that antibiotics that promoted in vitro growth and toxin production by C. difficile in cecal emulsions of hamsters also caused C. difficile disease in hamsters, whereas antibiotics that did not promote in vitro growth and toxin production did not cause disease (1). The experimental protocol was approved by the Animal Care Committee of the Cleveland Veterans Affairs Medical Center.

Female CF-1 mice weighing 25 to 30 g (Harlan Sprague-Dawley, Indianapolis, Indiana) were housed in individual cages with plastic filter tops to prevent cross-contamination among animals. We studied three antibiotics that we have previously shown to cause marked disruption of the anaerobic stool microflora of mice (clindamycin, piperacillin-tazobactam, and ceftriaxone), and three antibiotics that inhibit facultative gram-negative bacilli but cause minimal disruption of the anaerobic microflora (levofloxacin, cefepime, and aztreonam) (12, 13, 16). Mice received daily subcutaneous injections (0.2-ml total volume) of saline, levofloxacin (0.375 mg/day), cefepime (2.0 mg/day), aztreonam (3.0 mg/day), ceftriaxone (2.0 mg/day), clindamycin (1.4 mg/day), or piperacillin-tazobactam (8 mg/day) for 4 days. Three days after the final antibiotic dose, mice were euthanized by CO₂ asphyxiation. To evaluate the potential for piperacillin-tazobactam to inhibit growth of C. difficile during the course of treatment, one group of mice received daily treatment with this agent as described above but were euthanized 2 h after administration of the final antibiotic dose. The doses of antibiotics were equal to the usual human doses administered over a 24-hour period (milligrams of antibiotic per gram of body weight).

After the mice were euthanized, the cecum was removed and opened longitudinally. Cecal contents were collected and transferred to an anaerobic chamber (Coy Laboratories, Grass Lake, Michigan) within 5 min. The cecal contents were diluted threefold (volume/volume) in sterile prereduced phosphate-buffered saline (PBS). A final concentration of 10⁴ CFU/ml of each strain was added to separate aliquots of the cecal contents of individual mice. The C. difficile strains were prepared for inoculation by serially diluting 72-hour broth cultures in sterile prereduced PBS. After incubation for 24 h, the samples were diluted in sterile PBS and plated on prereduced cefoxitin-cycloserine-fructose agar containing 1% taurocholate to quantify C. difficile. To determine toxin production, a commercially available kit for detection of C. difficile toxin (Diagnostic Hybrids, Inc., Athens, Ohio) was utilized as recommended by the manufacturer. The cecal content supernatants were serially diluted 10-fold in specimen diluent. Following dilution, samples were added to microtiter plates containing human fibroblast cells and observed by bright-field microscopy, at 24 and 48 h, for evidence of C. difficile toxin cytopathic effect. The concentrations of antibiotics in cecal contents were determined by an agar diffusion assay with strains of Clostridium perfringens (for clindamycin, piperacillin-tazobactam, and ceftriaxone) or Escherichia coli (for cefepime, levofloxacin, and aztreonam) as the indicator strains (14). Experiments were performed twice with 6 total mice per group.

Because levofloxacin may be administered orally or parenterally, we performed an additional set of experiments to compare the effects of oral versus subcutaneous levofloxacin. Mice (4 per group) received daily levofloxacin (0.375 mg/day) by subcutaneous injection (0.2-ml total volume) or by orogastric inoculation (0.5-ml total volume) for 4 days. Orogastric inoculation was performed by use of a stainless steel feeding tube (Perfektum; Popper & Sons). The remainder of the experiment was performed as described above.

One-way analysis of variance was performed to compare C. difficile densities and toxin production among the treatment groups. P values were adjusted for multiple comparisons using the Scheffe correction. Computations were performed with the use of Stata software (version 5.0; Stata, College Station, Texas). P < 0.05 was considered significant.

The broth dilution MICs of the antibiotics for the three strains are shown in Table 1. Clinical and Laboratory Standards Institute (formerly National Committee for Clinical Laboratory Standards) criteria for interpreting susceptibilities are not available for C. difficile. None of the cecal contents had detectable concentrations of C. difficile prior to inoculation of the test strains (level of detection, ∼2 log₁₀ CFU/ml). Figure 1A demonstrates the effect of antibiotic treatment on growth of the C. difficile strains in cecal contents. The cecal contents of mice treated with saline, levofloxacin, cefepime, or aztreonam were inhibitory to growth of C. difficile and consistently yielded a ≥1-log decrease in density in comparison to density at the time of inoculation (4 log₁₀ CFU/ml). Cecal contents of mice treated with orogastric levofloxacin inhibited C. difficile to the same degree as cecal contents of mice receiving subcutaneous levofloxacin (i.e., a ≥1-log decrease in density of C. difficile was observed in cecal contents of all mice receiving oral levofloxacin) (data not shown). The cecal contents of mice that received prior treatment with ceftriaxone, clindamycin, and piperacillin-tazobactam supported growth of C. difficile (P < 0.001 for each group in comparison to saline controls). Ceftriaxone, clindamycin, and piperacillin-tazobactam were detectable in cecal contents 3 days after the final dose (mean ± standard deviation, 3.3 ± 0.6 μg/ml, 22.9 ± 3.2 μg/ml, and 2 ± 0.8 μg/ml, respectively), whereas levofloxacin, cefepime, and aztreonam were not detectable at that time (limit of detection, 1 μg/ml). Cecal contents prepared during the course of piperacillin-tazobactam treatment had higher concentrations of piperacillin-tazobactam (mean ± standard deviation, 31.2 ± 9.7 μg/ml) and were inhibitory to growth of C. difficile.

Figure 1B shows the toxin levels of the two toxin-producing C. difficile strains (strains 1 and 2). High levels of toxin were produced in cecal contents of mice that received prior treatment with ceftriaxone, clindamycin, and piperacillin-tazobactam, whereas minimal or no toxin was detectable in the saline, levofloxacin, and aztreonam groups (P < 0.01). In addition, minimal or no toxin was detectable in cecal contents prepared during the course of piperacillin-tazobactam treatment.

Our findings provide support for the hypothesis that disruption of the anaerobic microflora facilitates overgrowth and toxin production by C. difficile. As noted previously, we have shown that clindamycin, ceftriaxone, and piperacillin-tazobactam cause significant disruption of the anaerobic microflora of mice, whereas levofloxacin, aztreonam, and cefepime do not (12, 13, 16). The fact that clindamycin, ceftriaxone, and piperacillin-tazobactam remained detectable in cecal contents 3 days after discontinuation of treatment suggests that these agents were excreted in relatively high concentrations into the intestinal tract. In addition, the persistent detection of these agents could be attributable in part to the fact that inhibition of the anaerobic microflora may cause significant reductions in intestinal motility (20).

Although clindamycin and expanded-spectrum cephalosporins have frequently been associated with CDAD, clinical studies suggest that β-lactam and β-lactamase inhibitors such as piperacillin-tazobactam are a relatively infrequent cause of CDAD (5, 15). Our data suggest the possibility that agents such as piperacillin-tazobactam may be relatively infrequently associated with CDAD because they possess inhibitory activity against many C. difficile strains (i.e., C. difficile may be inhibited during the course of treatment with piperacillin-tazobactam but not cephalosporins or clindamycin). Alternatively, it has been suggested that piperacillin-tazobactam could be infrequently associated with CDAD because it may cause relatively modest disruption of the intestinal microflora of humans in comparison to expanded-spectrum cephalosporins and may stimulate less toxin production than agents such as cefotaxime (5, 15, 18, 19).

Although aztreonam, cefepime, and levofloxacin do not disrupt anaerobic microflora in the stool of mice, we have previously shown that they inhibit total aerobic and facultative gram-negative bacilli (13). These data suggest that a portion of each of these antibiotics is excreted into the intestinal tract of mice. These agents were not detected in cecal contents in the present study; however, the levels were not measured until 3 days after completion of treatment. Our findings in mice are consistent with previous studies in humans (17). Aztreonam, cefepime, and levofloxacin inhibit facultative gram-negative bacilli in the stool of humans but cause minimal disruption of anaerobes (9, 17). The failure of aztreonam to inhibit anaerobes is attributable to the fact that it has no appreciable in vitro activity against anaerobic bacteria (9). Cefepime is excreted in minimal concentrations in bile and has relatively little activity against intestinal anaerobes (2, 17). Because aztreonam and cefepime had minimal in vitro activity against the C. difficile test strains (Table 1) and were not detectable in cecal contents of mice, the absence of growth in cecal contents was unlikely to be due to inhibitory activity of these antibiotics. Rather, the inhibitory activity of the cecal contents is presumably due to the fact that the anaerobic microflora that suppress C. difficile were preserved.

Fluoroquinolone antibiotics such as levofloxacin achieve relatively high concentrations in the stool of humans after oral or intravenous administration (17). The minimal impact of these agents on intestinal anaerobes of humans has been attributed to high degrees of reversible binding of these agents to fecal matter, reduced susceptibility of anaerobes to fluoroquinolones under strictly anaerobic conditions, and an inoculum effect for inhibition of anaerobes (3). Some recent clinical studies have shown an association between fluoroquinolones such as ciprofloxacin and levofloxacin and CDAD (10, 23). In contrast, other studies have suggested that monotherapy with these agents is rarely associated with CDAD (8, 22). Our findings are consistent with the hypothesis that fluoroquinolones such as ciprofloxacin and levofloxacin are likely to present a relatively low risk for CDAD when administered as monotherapy. Additional research is needed to clarify whether fluoroquinolones with enhanced antianaerobic activity (e.g., gatifloxacin and moxifloxacin) are associated with an increased risk of CDAD in comparison to fluoroquinolones with less antianaerobic activity. Results of a recent study suggest that gatifloxacin may present a higher risk for CDAD than levofloxacin (6).