The rate of inactivation of metronidazole in vitro was determined during the course of time-kill curves against anaerobic and aerobic bacteria in the stationary phase of growth. Metronidazole at a concentration of 10 jig/ml, as measured by bioassay, was rapidly inactivated in broth culture by susceptible anaerobic bacteria (miniimum bactericidal concentration c 3 ,ug/ml), and this correlated closely with its bactericidal activity. In contrast, the drug was neither inactivated nor had any bactericidal activity against a resistant strain of Propionibacterium acnes (minimum bactericidal concentration > 1,500 ,ug/ml). Three of four aerobic bacteria also inactivated metronidazole, although at generally slower rates than the anaerobes, but this was not associated with a bactericidal effect against these organisms. The presence of aerobic bacteria in mixed cultures with Bacteroides fragilis did not, moreover, inhibit the bactericidal activity of metronidazole against the latter organism. However, the possibility still remains that, in vivo, aerobic bacteria capable of inactivating metronidazole could inhibit the action of the drug against anaerobes in mixed infections.It has been well established that metronidazole is consistently bactericidal in vitro against many anaerobic bacteria (2, 13) and appears to be effective clinically in the treatment of anaerobic infections (1,14). Against aerobic organisms, however, metronidazole is relatively inactive (2), and, moreover, earlier studies have shown that it is inactivated by certain aerobic bacteria (6,8).A study was devised to examine the rate of inactivation of metronidazole by both anaerobic and aerobic bacteria and the potential inhibitory effect of aerobic organisms on its activity against anaerobes.MATERIALS AND METHODS Bacterial organisms. Six clinical anaerobic isolates (Bacteroides fragilis, Fusobacterium nucleatum, Eubacterium lentum, Peptostreptococcus anaerobius, Clostridium perfringens, and Propionibacterium acnes) and four aerobic strains (Staphylococcus aureus, Escherichia coli, Proteus morganii, and Streptococcus faecalis) were used as the test organisms. AU were maintained on blood agar in an anaerobic glove box (Coy Manufacturing, Ann Arbor, Mich.). The gas mixture within the anaerobic glove box consisted of 85% nitrogen, 10% hydrogen, and 5% carbon dioxide. The bacteria were subcultured in brucella broth (Difco) containing 5 ug per ml of hemin(BB+H) 18 to 48 h before each experiment.The minimum inhibitory concentrations (MICs) of metronidazole for these organisms were determined by the tube dilution method in BB+H in the anaerobic glove box at an inoculum size of approximately 105 colony-forming units per ml (CFU/ml). The minimum bactericidal concentrations (MBCs) were determined after subculture of tubes showing no growth above to brucella agar (Difco) containing hemin (BA+H) and defined as the lowest concentration at which no growth occurred.Time-kill curves. All procedures were performed within the anaerobic glove box (AGB) except the pouring of plate...