whether treatment was continued for 3 or 7 days. The addition of gentamicin was not associated with increased killing in rabbits infected with the aminoglycoside-resistant isolate. Both high-dose ampicillin-sulbactam and vancomycin regimens demonstrated significant, continued reduction in bacterial titers with the longer periods of treatment (P c 0.05); 7-day treatment with high-dose ampicillin-sulbactam produced a greater reduction in bacterial titers in vegetation than 7-day treatment with vancomycin (P c 0.05). We conclude that ampicillinsulbactam and vancomycin are equally effective in the treatment of experimental endocarditis due to 13-lactamase-producing, highly gentamicin-resistant E. faecalis. The optimum therapy for such infections in humans is not known.Increasing antibiotic resistance has complicated the treatment of serious infections due to enterococci, particularly those due to Enterococcus faecalis. Such infections require a bactericidal therapy, which is usually obtained only with a synergistic combination of a cell-wall-active agent, such as a P-lactam antibiotic or vancomycin, and an aminoglycoside for cure (9,13,(20)(21)(22)24). The emergence of high-level aminoglycoside resistance and the appearance of 1-lactamase-producing strains of E. faecalis (15) and Enterococcus faecium (1) are noteworthy for two reasons: (i) the number of treatment options is reduced because synergism is not observed when aminoglycosides to which the organism has demonstrated high-level resistance are used (12), and (ii) ,-lactamase production has been shown to influence the clinical outcome of infections due to these organisms (23). We have previously shown that ampicillin-sulbactam was bactericidal by the time-kill method for 13 of 20 ,-lactamaseproducing, highly gentamicin-resistant strains of E. faecalis, while bactericidal activity was absent with vancomycin against all 20 strains (10). Currently, the optimum treatment of serious infections caused by I-lactamase-producing, highly gentamicin-resistant E. faecalis has not been determined. In the studies reported herein, we evaluated the efficacy of ampicillin-sulbactam and vancomycin (each alone * Corresponding author. and in combination with gentamicin) in the treatment of experimental aortic valve endocarditis caused by a 3-lactamase-producing, highly gentamicin-resistant isolate of E. faecalis and compared these data with the results obtained in the treatment of experimental endocarditis caused by a non-p-lactamase-producing, gentamicin-susceptible E. faecalis isolate.(This study was presented in part at the 92nd General Meeting of the American Society for Microbiology, 1992 [8].) MATERIALS AND METHODS Bacterial strain. E. faecalis E-48, a ,-lactamase-producing, gentamicin-resistant isolate, was used to inoculate rabbits. The MICs for E. faecalis E-48 were as follows: ampicillin, 4 ,ug/ml; sulbactam, .8 ,ug/ml; ampicillin-sulbactam, 1 ,ug of ampicillin per ml; vancomycin, 4 ,ug/ml; and gentamicin, .2,000 ,ug/ml. In addition, several rabbits were inoculated wi...