Meropenem (MEPM, SM-7338), a novel parenteral carbapenem antibiotic, was examined for its effect on intestinal flora in children. Seven children with infectious diseases (3 male and 4 female children of age's ranging from 4 months to 8 years and 9 months weighing from 7.3 to 23.0 kg) were treated with MEPM at doses ranging 10.3 to 40.5 mg/kg 3 or 4 times a day for 6 to 12 days. Before, during and after the treatment, identities and numbers of various bacteria contained in 1 g of feces were determined and fecal beta-lactamase activity and Clostridium difficile D-1 antigen were also assayed. Changes in fecal flora during MEPM treatment was somewhat different depending on cases. Regarding Enterobacteriaceae among aerobes, all of 7 cases exhibited moderate or pronounced reductions in Escherichia coli. Some of the cases exhibited the tendency to increase in Klebsiella oxytoca. Enterobacter cloacae and Citrobacter freundii. E. coli which was reduced during the treatment increased rapidly after the treatment in 5 out of 7 cases, and the initial bacterial counts were restored. Diverse strains were observed within the genus Enterococcus, while the overall bacterial counts of this genus exhibited the tendency to increase during the treatment. As a result, no significant change in total aerobe count was observed in any case except 1 case where Enterococcus count was somewhat reduced. Among anaerobes, major bacteria such as Bacteroides, Bifidobacterium, Eubacterium and Peptococcaceae exhibited tendencies to decrease in some cases during the antibiotic treatment. Two infants and 1 child exhibited significant decreases in total anaerobe counts. In most of the cases, such changes in major anaerobes were transient and bacterial counts recovered to their initial values rapidly after completion of the treatment. In no cases, glucose non-fermentative Gram-negative bacilli or fungus became predominant. Although C. difficile D-1 antigen was observed in 4 cases, its changes had no relationship with characteristics of feces. C. difficile was not detected in any of the cases. MEPM was detected in feces in 4 cases being treatment, in concentrations ranging from 0.35 to 66.0 micrograms/g. Fecal MEPM levels were very low except in 1 case in which beta-lactamase was negative. From these results, effects of MEPM on intestinal flora in children were relatively minor compared to other new beta-lactam drugs. However, a care should be taken to minimize diarrhea and bacterial turnover when a prolonged use of the antibiotic, was practiced because of potential significant effects on intestinal flora.