Effect of moxalactam on human fecal microflora.

Five healthy male adults received 2 g of moxalactam every 12 h for 7 days. The alterations of fecal microflora were investigated before, during, and after treatment with moxalactam. On day 7 of treatment, the number of total bacteria was decreased in all subjects. There was marked suppression of the obligate anaerobic bacteria and enterobacteria to undetectable levels, but the counts of Streptococcus spp. and Lactobacillus spp. increased. On day 7, two subjects had Clostridium innocuum and Clostridium ramosum in their feces but not Clostridium difficile. On day 7 after treatment, the counts of Streptococcus spp., enterobacteria, Lactobacillus spp., and Clostridium spp. in all subjects were still not normal.

Several antibiotics are known to cause considerable changes in normal human fecal flora and sometimes also to permit overgrowth of resistant bacteria (7,17). Some fecal bacteria may produce toxins and can cause pathogenic conditions. Moxalactam, a 1-oxa-1-lactam derivative, was developed by Shionogi & Co., Osaka, Japan. It was demonstrated that this drug is effective against both aerobic and anaerobic bacteria (18).
Although moxalactam might be expected to produce appreciable changes in the human fecal flora, there are few published data concerning the effect of this drug on the fecal flora (2). This study was undertaken to determine the changes in normal human fecal flora when moxalactam was administered intravenously.
Five healthy male subjects between the ages of 32 and 44 years and weighing 62 to 76 kg (mean, 67.8 + 5.4 kg) participated in this study. All subjects received moxalactam in a 2-g dose every 12 h for 7 days. The drug was administered in 50 ml of saline solution over a period of 30 min.
After incubation, each plate was examined for bacterial counts. Colonies from the anaerobic agar were subcultured on Eggerth-Gagnon agar or glucose-blood-liver agar in duplicate to allow aerobic and anaerobic incubation. Microorganisms growing only on the anaerobic agar were identified as strictly anaerobic bacteria. All of the s'trains isolated were maintained on Eggerth-Gagnon liver agar slants (14) filled with 100% CO2 gas. Colonies from TS agar were also subcultured on TS agar aerobically. The aerobic isolates were kept on TS agar slants (14). Both slants were stored at 4°C. Transfers were usually made at 1-month intervals. Strictly anaerobic bacteria isolated from the feces were identified by methods described previously (4,9). Enterobacteria, staphylococci, and Candida spp. were classified by the API systems (Analytab Products, Inc., Plainview, N.Y.). The other aerobic bacteria were identified by conventional methods (6,11). For the isolates identified, the bacterial count per gram of wet feces was calculated and converted into a logarithmic equivalent. A few strains could not be fit into any of the established species; these were designated as spp. The total viable count was calculated from the sum of the counts of each bacterial species. With these methods, the lowest detectable number of microorganisms was 2 log10/g of wet feces. The bacterial counts were analyzed statistically by Student's t test. The concentration of moxalactam in the feces was determined by the band culture method (10).
'Some isolates could not be identified to the species level with currently accepted identification protocols and presently recognized species. d Statistically significant at P < 0.01 when compared with the pretreatment values (Student's t test was used for the bacterial counts). ' p < 0.001. J One isolate (one species).
After the drug treatment, the total counts in all subjects showed a recovery to pretreatment normal levels. On day 7 of treatment, the incidence of Bacteroides vulgatius, Bacteroides spp., Bifidobacteriiim adolescentis, Eubacterium aerofaciens, and Eubacterium spp. was lower than that pretreatment (Table 1). In two of the subjects, however, the counts of Clostnidiium sp. were markedly increased. The Clostridiium sp. were identified as Clostridium innocuum and Clostridiii,n rainosuim. On the other hand, the aerobic bacteria were significantly (P < 0.001) increased on day 7. The counts of Streptococcus spp. (P < 0.001) and Lactobacillus spp. were higher in this experimental period than in the pretreatment period. Significantly increased numbers of Streptococcus faecalis and Streptococcus faecium were observed on day 7. The counts of Candida spp. changed in all subjects. Especially, there were marked increases in Candida sp. in two subjects. On day 7 of treatment, the concentration of moxalactam in the feces was 110.4 + 73.9 p.g/g.
The intravenous administration of moxalactam induces marked changes in the fecal flora, with suppression of strict anaerobic bacteria and enterobacteria. Similar changes were recently reported with moxalactam except for lactobacilli and Pseudomonas spp. (2). It has also been reported that the fecal microflora of patients given cefoperazone (1) or ceftriaxone (3) was dominated by members of the family Streptococcaceae and clostridia but not by obligate anaerobes and enterobacteria. Our results with intravenous administration of moxalactam were in close agreement with these reports.
In the present study, the significant reduction of strict anaerobes, particularly Bacteroides fr)agilis group, Bifidobacteriium spp., Eubac-teriumn spp., and Peptostreptococ'c'us spp., in stool specimens during treatment was in close agreement with the report of Allen et al. (2). The increased numbers of Streptococ clus spp. and Lactobacilluls spp. in the fecal microflora during treatment may be due to the resistance of these microorganisms to moxalactam. It is of interest that two of our subjects had intestinal colonization with moxalactam-resistant Clostr-idiuiim spp. during treatment. None of the Clostnidiuirn spp. isolated during this study were identified as Clostridilim difficile.
A period of 7 days after moxalactam administration was not sufficient for the fecal flora to recover to the pretreatment levels. Normalization of aerobic bacteria and Clostridium spp. may require another week. However, the numbers and incidences of B. fragilis group, Bifidobac terium adolescentis, Eubac(teriun aerojaciens, and Peptostrepto-('O(C'ClS produictus, which were the predominant fecal bacteria before treatment, were normalized on day 7 after treatment. On the other hand, significantly increased numbers of C. innocuum, V. panri'ula, S. fJacclalis, E. c oli, and L. gasseri were observed after the treatment. On day 12 after cefoperazone administration, a reduction in the number of bifidobacteria and an increase in the number of Eutbac teritum spp., Clostnidilurn spp., Lac-tobacillus spp., and Streptococcils spp. were reported (5). Our results were closely related with regard to the increased number of Clostnidiion spp. and Streptoc-occulis spp. These alterations after the treatment may be considered a common effect of the new r-lactam antibiotics on human fecal microflora.