Treatment of experimental endocarditis due to ampicillin-susceptible or ampicillin-resistant Salmonella enteritidis.

Using two strains of Salmonella enteritidis, one susceptible and one resistant to ampicillin, we studied the efficacies of ampicillin, gentamicin, ampicillin plus gentamicin, ofloxacin, and cefotaxime for the treatment of experimental salmonella endocarditis. Rabbits were treated for 3 days with dosages of antibiotic selected to achieve concentrations in serum equivalent to those obtained in humans during therapy. Aortic salmonella endocarditis seemed to be very difficult to treat, and all antimicrobial regimens failed to achieve the complete sterilization of cardiac vegetations. In vitro studies did not accurately predict the in vivo response to therapy, and no correlations regarding the synergistic activity of the combination of ampicillin plus gentamicin were observed. For the ampicillin-susceptible S. enteritidis isolate, ampicillin and cefotaxime produced the greatest reduction in the number of organisms in vegetations, with no significant differences between them. For the ampicillin-resistant strain, the combination of ampicillin with gentamicin produced a synergistic effect that was not anticipated by the in vitro studies. Both cefotaxime and ofloxacin were effective in reducing the number of microorganisms in the vegetations, although the reduction produced by cefotaxime was less that that produced against the ampicillin-susceptible strain. Monotherapy with gentamicin exhibited only modest activity against the ampicillin-susceptible S. enteritidis strain.

Treatment of experimental endocarditis due to penicillin-resistant Streptococcus pneumoniae.

Using two strains of pneumococci for which MICs of penicillin were 1 and 4 micrograms/ml, those of cefotaxime were 0.01 and 0.5 micrograms/ml, and those of teicoplanin were 0.01 and 0.1 micrograms/ml, we studied the efficacy of different dosages of penicillin, cefotaxime, and teicoplanin in the treatment of experimental pneumococcal endocarditis in rabbits. Animals treated with dosages of penicillin G procaine needed to achieve levels in serum near the MIC for pneumococci showed a significant reduction in log10 CFU per gram of vegetation, as compared with the control (P < 0.001), although only 20% of the animals showed sterile vegetations. When levels of penicillin in serum were in the range of three- to fourfold the MIC, a greater reduction in log10 CFU per gram of vegetation was seen, and 88% of the animals showed sterile vegetations. Only the regimen of penicillin that provided concentrations in serum above the MIC throughout the interval between two doses provided constant sterilization of the cardiac vegetations. Dosages of cefotaxime and teicoplanin selected to achieve concentrations in serum equivalent to that obtained in humans during treatment resulted in levels of antimicrobial agents in serum hundreds or thousands of times higher than the MICs for the infecting strains. In terms of antimicrobial efficacy, cefotaxime and teicoplanin were equivalent to regimens with high dosages of penicillin.

In vivo significance of the inoculum effect of antibiotics on Escherichia coli.

The minimum dosage of antibiotics which reduced mortality in rats intraperitoneally inoculated with an Escherichia coli isolate was determined. Low mortality rates (0-10%) were obtained when antibiotics with minimal or no inoculum effect (cefoxitin, cefmetazole and gentamicin) were administered to yield serum levels 3 to 20 times the MIC, while antibiotics with a pronounced inoculum effect (cefotaxime and aztreonam) had to be administered to yield serum levels 200 to 1,000 times the MIC determined with a standard (low) inoculum. Thus, it seems that the inoculum effect observed in vitro with some antibiotics for Escherichia coli may have clinical significance.