To test the hypothesis that cephalosporins resistant to beta-lactamase are preferred in the treatment of serious staphyloccal infections, the ability of four cephalosporins to eradicate bacteria from the cardiac vegetations of rabbits with experimental endocarditis was examined. Two strains of Staphylococcus aureus were chosen as pathogens: one that rapidly and completely inactivated 50 micrograms of cefazolin in vitro (beta-lactamase-positive) and another that did not inactivate any cephalosporin (beta-lactamase-negative). Rabbits with a polyethylene catheter in the left ventricle were reliably infected witih 10(5) bacteria. Similar numbers of S. aureus were recovered from the cardiac vegetations of rabbits inoculated with the beta-lactamase-positive strain after 24 hr of treatment with each of four cephalosporins. However, when the animals were treated at intervals of 6 hr for four days, significantly fewer rabbits survived after treatment with cefazolin than with cephalothin. No difference in survival was observed in the treatment of rabbits with endocarditis due to the beta-lactamase-negative strain. The failure of cefazolin in the treatment of staphylococcal endocarditis in rabbits may be due to inactivation of the drug by beta-lactamase in vivo.

Methicillin-tolerant staphylococci are organisms that are inhibited by low concentrations of the drug but are resistant to its bactericidal effects. The clinical significance of this in vitro phenomenon is unknown. An experimental model of staphylococcal endocarditis in rabbits was used to determine whether methicillin-tolerant staphylococci are particularly difficult to eradicate from infected heart valves. In vitro sensitivity testing was used to identify a tolerant and a nontolerant strain of Staphylococcus aureus . Rabbits prepared by the insertion of a polyethylene catheter into the left ventricle and injected with a tolerant strain survived significantly longer than those injected with a nontolerant strain. No significant differences were demonstrated in the prevention of endocarditis with a single prophylactic dose of methicillin, or in the treatment of established endocardial infection with multiple doses of methicillin. In this study, methicillin had the same activity against experimental endocarditis caused by a tolerant and a nontolerant strain of S. aureus .

The relative efficacy of single doses of antibiotics in modifying the development of Bacteroides fragilis subsp. fragilis endocarditis was studied in an experimental model. Antibiotics were administered 0.5 -h before intravenous injection of B. fragilis subsp. fragilis into rabbits prepared by insertion of a polyethylene catheter into the left side of the heart; 48 h later, intracardiac vegetations were excised and cultured anaerobically. B. fragilis was recovered from 92% of untreated animals. After a single dose of procaine penicillin G (250 mg/kg intramuscularly), 80% of the animals remained infected. Chloramphenicol (30 mg/kg), carbenicillin (50 mg/kg), and metronidazole (10 mg/kg) were also ineffective (76, 80, and 75% infected, respectively). Cefamandole (30 mg/kg), cefoxitin (30 mg/kg), and erythromycin (30 mg/kg) were significantly more active (50, 55, and 45% infected, respectively), as were higher doses of carbenicillin. Clindamycin (50 mg/kg) was the most effective regimen (11% infected). At present, the relevance of these results to the therapy of serious B. fragilis infections is not known, but this model may prove useful in the evaluation of the prevention of B. fragilis subsp. fragilis bacteremia.

No abstract