Effects of clindamycin and metronidazole on the intestinal colonization and translocation of enterococci in mice.

The intestinal colonization and translocation of enterococci was studied in mice treated intramuscularly with metronidazole or clindamycin, with or without oral streptomycin. Treatment with metronidazole resulted in selective elimination of strictly anaerobic cecal bacteria, with a 100-fold increase in the numbers of aerobic and facultative gram-negative bacilli and a 10,000-fold increase in the numbers of aerobic and facultative gram-positive species. Clindamycin had a similar effect on the cecal flora except that the numbers of aerobic and facultative gram-positive bacteria decreased at least 10-fold. The predominating gram-positive species in the cecal flora or metronidazole-treated mice was an enterococcus, but this organism could not be recovered from the ceca of clindamycin-treated mice. Translocating bacteria (primarily gram-negative enteric bacteria) were recovered from the mesenteric lymph nodes of the majority of mice given metronidazole or clindamycin. Gram-positive bacteria were not recovered from the mesenteric lymph nodes of 20 clindamycin-treated mice, whereas 26% of 19 metronidazole-treated mice had translocating enterococci. With addition of streptomycin to the metronidazole and clindamycin regimens, mice treated with metronidazole-streptomycin became colonized predominantly with an enterococcus, and this was the only translocating species recovered from 13% of 23 mice; however, enterococci could not be detected in the ceca of clindamycin-streptomycin-treated mice, and Bacillus spp. were recovered from the mesenteric lymph nodes of 8% of 24 mice, reflecting the composition of the cecal flora. The apparent elimination of enterococci from the ceca of clindamycin and clindamycin-streptomycin-treated mice was inconsistent with the observation that the average (n=6) peak levels of clindamycin in blood and ceca were 25 and 21 microgram/ml, respectively, whereas the in vitro MIC was 128 microgram/ml. However, this apparent in vivo activity of clindamycin against enterococci was not evident in mice given 10(9) oral enterococci; the concentrations of cecal enterococci in both clindamycin-streptomycin- and metronidazole-streptomycin-treated mice were 10(10) to 10(11) enterococci per g, with translocating enterococci recovered from approximately half of these antibiotic-treated mice. Thus antibiotic therapy with metronidazole, clindamycin, metronidazole-streptomycin, and clindamycin-streptomycin resulted in a wide variation in the cecal population levels and translocation frequencies of enterococci. This variation appeared to be related to the discrepancy between the in vivo and in vitro activities of clindamycin against enterococci.