Contribution of animal models in the search for effective therapy for endocarditis due to enterococci with high-level resistance to gentamicin.

Earlier studies suggest that ampicillin and amoxicillin are more effective than other beta-lactam agents in killing enterococci, although beta-lactam agents are slowly and incompletely bactericidal against most strains of Enterococcus faecalis. We previously showed that continuous infusion of ampicillin is more effective than intermittent administration in decreasing the number of enterococci in valvular vegetations of rats with catheter-induced endocarditis that are treated for 5 days. In this model, we found ampicillin plus sulbactam more effective than ampicillin alone against a beta-lactamase-producing enterococcal strain with high-level resistance to gentamicin. Daptomycin therapy produced results approximately equal to those of ampicillin plus sulbactam. Vancomycin and teicoplanin given for 5 days at doses producing equivalent serum levels had approximately equal efficacy. However, 10-day therapy with low-dose teicoplanin was considerably more effective than similar treatment with vancomycin. High-dose teicoplanin for 5 days produced sterile valves in 82% of the animals studied.     

Single-dose rifampin prophylaxis for experimental endocarditis induced by high bacterial inocula of viridans streptococci

In rats challenged with viridans streptococci poorly susceptible to antibiotic killing, single doses of antibiotics only prevent endocarditis induced by bacterial inoculum sizes that produce disease in 90% of control animals (ID90): additional doses are required to protect against inocula exceeding the ID90. We investigated whether single-dose rifampin would extend the efficacy of single-dose prophylaxis to inocula exceeding the ID90. We used two strains of viridans streptococci highly susceptible to killing by rifampin and two resistant strains. All rats were injected with 10-1,000 times the ID90 of the four strains. Single-dose rifampin successfully prevented endocarditis due to all four strains. A few prophylaxis failures were observed after challenge with the two poorly susceptible strains, but in vivo emergence of resistant variants did not account for these failures. Thus, rifampin was the first antibiotic given as a single dose that successfully prevented experimental streptococcus endocarditis after challenge with high bacterial inocula.     

A novel mechanism of resistance to penicillin-gentamicin synergism in Streptococcus faecalis

A patient with enterococcal endocarditis, who relapsed after repeated courses of apparently adequate treatment with ampicillin plus gentamicin, was subsequently cured with ampicillin-tobramycin therapy. The organisms isolated from this patient were strains of Streptococcus faecalis that were resistant to penicillin (or ampicillin)-gentamicin synergism but not to penicillin (or ampicillin)-tobramycin synergism. The mechanism of resistance in these strains appears to be related to a specific defect in the intracellular uptake of gentamicin (but not tobramycin) in the presence of penicillin.     

Successful prophylaxis of experimental streptococcal endocarditis with single-dose amoxicillin administered after bacterial challenge

Rats with catheter-induced aortic vegetations were challenged intravenously with various inoculum sizes of tolerant Streptococcus sanguis or Streptococcus faecalis. Single-dose amoxicillin (40 mg/kg) was given intravenously either 30 min before or 30-240 min after bacterial challenge. Prophylaxis of endocarditis against both strains was successful when the inocula used for challenge were in the range of the minimum inoculum producing bacterial endocarditis in 90% of control animals (ID90) but was less effective or failed with larger inocula or when amoxicillin administration was delayed up to 240 min after bacterial challenge with S. sanguis. In a group of rats profoundly depleted of neutrophils by a rabbit anti-rat neutrophil serum given 30 min after challenge with S. faecalis at ID90, single-dose amoxicillin administered simultaneously with the antiserum was protective, indicating that neutrophils were not required for successful endocarditis prophylaxis.     

beta-Lactamase production in experimental endocarditis due to aminoglycoside-resistant Streptococcus faecalis

We used a beta-lactamase-producing (beta L+) strain of Streptococcus faecalis that also had high levels of resistance to all aminoglycosides to induce experimental endocarditis in rats. The rats were treated for five or 10 days with procaine penicillin, vancomycin, gentamicin, rifampin, or ciprofloxacin (alone or in various combinations), or with penicillin plus clavulanic acid. The levels of penicillin in serum and vegetations declined rapidly in the beta L+-infected rats treated with procaine penicillin alone, unlike the sustained levels of penicillin in either beta L- -infected rats treated with procaine penicillin or beta L+-infected rats treated with penicillin plus clavulanic acid. For the beta L+-infected rats, the enterococcal counts in vegetations were significantly reduced (greater than 3 log10 cfu/g) only by vancomycin and by penicillin plus clavulanic acid. The efficacy of the latter regimen probably resulted from the inhibition of penicillin inactivation by clavulanic acid in vegetations infected with the beta L+ strain. Our in vivo findings document the biologic significance of beta-lactamase production.     

Antibiotic prophylaxis of experimental endocarditis after dental extractions

In rats with catheter-induced sterile aortic valve vegetations we studied the efficacy of single-dose amoxicillin and single-dose erythromycin prophylaxis for the prevention of bacterial endocarditis after extractions of periodontally diseased teeth. Endocarditis after extractions occurred in 89% of control animals and was due to group G streptococci, to Staphylococcus aureus, or to both organisms. A single-dose of amoxicillin or erythromycin successfully prevented endocarditis due to these bacterial species. The analysis of the bacteremia (by culturing blood drawn 1 min after extraction on penicillinase-containing blood agar plates) indicated that amoxicillin did not influence the incidence or the magnitude of circulating group G streptococci and S. aureus, while erythromycin apparently suppressed them. However, when care was taken to eliminate blood erythromycin by a lysis-centrifugation process, the incidence and magnitude of bacteremia after erythromycin prophylaxis was similar to that in control rats. We conclude that single doses of amoxicillin and erythromycin successfully prevent experimental endocarditis after dental extractions. Since this prophylaxis was operative by mechanisms other than the prevention of the circulation of bacteria before seeding the valvular vegetations, it suggests that recommendations for prevention of bacterial endocarditis should not be aimed only at providing adequate antibiotic blood levels to suppress the bacteremia produced by the invasive procedure.     

Role of tolerance in cloxacillin prophylaxis of experimental Staphylococcus aureus endocarditis.

The role of tolerance was investigated in the prophylaxis of Staphylococcus aureus endocarditis with cloxacillin in rats. The effect of a single dose of 500 mg/kg, two 80 mg/kg doses 3 h apart, and a single dose of 80 mg/kg, alone or in combination with a single dose of 4 mg of gentamicin/kg, were compared for a tolerant strain of S. aureus and its isogenic nontolerant variant. At all dosages, cloxacillin was significantly less effective in preventing endocarditis with the tolerant strain than with the nontolerant variant. With the high dose of cloxacillin or two successive lower doses, nearly complete protection could be obtained against the nontolerant strain. However, for the tolerant strain, only the combination of cloxacillin and gentamicin afforded almost complete protection. For the tolerant strain, no serum bactericidal activity was found at the time of bacterial challenge after an injection of any dose of cloxacillin. These results suggest that the in vitro phenomenon of tolerance may have relevance in vivo.     

Endocarditis due to streptomycin-susceptible Enterococcus faecalis with high-level gentamicin resistance

A case of community-acquired endocarditis caused by Enterococcus (Streptococcus) faecalis with high-level resistance to gentamicin sulfate but not to streptomycin sulfate is described. Killing curves performed using achievable serum levels showed synergistic killing when streptomycin but not gentamicin, tobramycin, or amikacin was combined with penicillin G sodium or vancomycin hydrochloride. Combination therapy with vancomycin and streptomycin resulted in cure. Serum bactericidal levels indicated activity of the synergistic, as well as a nonsynergistic (vancomycin plus gentamicin), combination. Routine screening of blood isolates for high-level resistance to streptomycin and gentamicin can provide guidance for selection of therapeutic combinations in serious enterococcal infections, including endocarditis.     

Mechanisms of successful amoxicillin prophylaxis of experimental endocarditis due to Streptococcus intermedius

Prophylaxis with amoxicillin (40 mg/kg) was studied in rats with aortic valve vegetations. Bacteria on the valves were quantitated early (10 min to 6 hr) and late (three days) after intravenous challenge with tolerant Streptococcus intermedius. Amoxicillin reduced by 40% the number of bacteria per valve 10 min after intravenous challenge with 10(5) S. intermedius (P less than .05) and by 74% the incidence of endocarditis three days thereafter (P less than .0001). Bacterial multiplication started 2 hr after challenge in control rats, whereas bacteria disappeared in 6 hr in amoxicillin-treated rats. Intravenous penicillinase 30 min after challenge abolished successful amoxicillin prophylaxis, a result demonstrating the necessity of prolonged growth inhibition for protection. Growth inhibition for 18 hr (two subsequent amoxicillin doses) was necessary for protection after intravenous challenge with 10(5) S. intermedius. Thus, in the absence of bacterial killing, inhibition of valvular colonization by amoxicillin was not as important a mechanism of endocarditis prophylaxis as was prolonged inhibition of bacterial growth, which allowed adherent bacteria to be cleared from the valves.     

Apparent failure of endocarditis prophylaxis caused by penicillin-resistant Streptococcus mitis

Antibiotic resistance among viridans streptococci has increased with Streptococcus mitis being more resistant than other viridans species. In a case presented in this report, it is possible that antibiotic resistance contributed to an apparent failure of endocarditis prophylaxis. The patient had undergone periodontal surgery on 2 separate occasions and in both instances was administered 2 g of amoxicillin orally 1 hour before each procedure. He subsequently developed a subacute illness and had multiple blood cultures drawn that grew S. mitis with a minimum inhibitory concentration of 1.0 microg/mL for penicillin. Transesophageal echocardiogram provided further evidence of infective endocarditis with vegetations seen on the anterior leaflet of the mitral valve. Combination therapy with high-dose intravenous aqueous crystalline penicillin G and gentamicin sulfate for 4 weeks was curative. Clindamycin, rather than amoxicillin, has since been used as dental prophylaxis for subsequent procedures.     

Infective endocarditis caused by Streptococcus bovis resistant to the lethal effect of penicillin G.

Penicillin G alone is generally recommended for the treatment of infective endocarditis caused by Streptococcus bovis because clinical isolates of S bovis are represented as being uniformly and markedly susceptible to penicillin G. However, two strains of S bovis recovered from two patients with bacterial endocarditis were resistant to the lethal effect of penicillin G. Combination therapy, cefazolin sodium and gentamicin sulfate in patient 1 and penicillin G and gentamicin in patient 2, was necessary; synergy, as manifested by lethal activity against the infecting strains, was demonstrated in the laboratory. We stress the need to determine the minimal lethal concentration of penicillin G for clinical isolates of S bovis. Until such information is available, particularly in life-threatening infections, combination drug therapy, consisting of an aminocyclitol added to a beta-lactam antimicrobic, should be used.     

High-level penicillin resistance among isolates of enterococci. Implications for treatment of enterococcal infections

STUDY OBJECTIVE: To determine the extent and clinical significance of high-level penicillin-resistant Enterococcus faecium at our institution. DESIGN: Surveillance of clinical enterococcal isolates, in-vitro susceptibility and timed survival studies, and determination of antibiotic efficacy in an experimental model of enterococcal endocarditis. MEASUREMENTS AND MAIN RESULTS: For a 6-month period, 14% of enterococcal isolates (30 of 212) were identified as E. faecium. One third of the isolates were highly resistant to penicillin G (minimum inhibitory concentration [MIC], greater than or equal to 200 micrograms/mL) but did not produce beta-lactamase. The findings from in-vitro survival studies showed that this high-level resistance resulted in the loss of bactericidal activity normally observed when an aminoglycoside antibiotic agent is combined with penicillin. An experimental rat model of endocarditis provided in-vivo data that confirmed our in-vitro observations. After the rats received therapy for 72 hours, penicillin G either alone or in combination with gentamicin did not significantly decrease the numbers of enterococci in vegetations on heart valves compared with untreated controls (P = 0.62 and P = 0.58, respectively). CONCLUSIONS: Enterococcus faecium accounts for a notable proportion of clinical enterococcal isolates. Many strains from patients at our institution, as well as from patients at other institutions throughout the country, are highly resistant to penicillin. Because high-level penicillin resistance has important therapeutic implications, periodic surveillance and MIC testing of significant enterococcal isolates, especially E. faecium, are suggested.     

Enterococcal endocarditis: a comparison of prosthetic and native valve disease

Between 1973 and 1987, 36 patients with 41 episodes of enterococcal endocarditis were seen at our institution. There were 22 episodes of native valve endocarditis (NVE) and 19 episodes of prosthetic valve endocarditis (PVE). The overall mortality before completion of therapy was 15% (18% due to NVE and 11% due to PVE). Among patients with NVE, involvement of the aortic valve was significantly associated with death or complicated illness (defined as the need for valve replacement before completion of antibiotic therapy or relapse of endocarditis after completion of therapy). Among patients who survived episodes of PVE, 69% were cured without surgical intervention. Gentamicin was administered in combination with a penicillin or vancomycin in the majority of episodes (mean duration of therapy with aminoglycosides: 5 weeks). Renal dysfunction occurred in 44% of patients who received gentamicin and occurred more frequently in patients with elevated serum creatinine levels before treatment. Our results suggest that enterococcal PVE can often be successfully treated with antibiotics alone, and they confirm the efficacy of gentamicin when it is administered in combination with cell wall-active agents for the treatment of endocarditis due to enterococci that lack high-level resistance to this agent.     

Effect of antimicrobial therapy for experimental infections due to group B Streptococcus on mortality and clearance of bacteria

In an effort to develop more effective antimicrobial therapy, we evaluated three alternative regimens currently available to clinicians for their efficacy against experimental bacteremia and meningitis due to group B Streptococcus (GBS) in newborn rats: various doses of penicillin G (100, 200, 400, or 800 mg/kg per day), combined penicillin G-gentamicin vs. penicillin G, and ceftriaxone vs. penicillin G. Higher doses of penicillin G and ceftriaxone exhibited significantly greater bactericidal activity in blood and cerebrospinal fluid (CSF), whereas the bactericidal activity of penicillin G plus gentamicin was not significantly different from that of penicillin G. Clearance of GBS from blood was significantly more rapid in animals receiving ceftriaxone. However, differences in death rates were not apparent with any single regimen. These findings suggest that clearance of GBS from blood and CSF can be improved by more potent antimicrobial agents, but further reduction in the death rate may be difficult to achieve by antimicrobial therapy alone.     

Endocarditis due to group D streptococci. Comparison of disease caused by streptococcus bovis with that produced by the enterococci

The group D streptococci include the nonenterococcal Streptococcus bovis in addition to the classic enterococci. Endocarditis due to Strep. bovis has received little previous attention in the medical literature. A review of all cases of group D streptococcal endocarditis seen at the Massachusetts General Hospital between 1964 and 1973 revealed 14 cases caused by Strep. bovis and 15 by enterococci. There were only minor differences in the clinical presentations of endocarditis caused by these two groups of organisms. Although it contains the group D antigen Strep. bovis behaved like Strep. viridans in producing endocarditis. Moreover, the strains of Strep. bovis in this study were much more susceptible to penicillin than the enterococci. Therapy of severe enterococcal infections requires penicillin plus an aminoglycoside antibiotic whereas the present study strongly suggests that penicillin alone is adequate therapy for endocarditis due to Strep. bovis.     

Infective endocarditis caused by Streptococcus mutans. A complication of idiopathic hypertrophic subaortic stenosis

Three patients with endocarditis caused by Streptococcus mutans were seen during a six-month period. All had clinical features of subacute bacterial endocarditis, including fever, heart murmurs, and positive blood cultures. One had underlying aortic insufficiency and two had idiopathic hypertrophic subaortic stenosis. All patients were treated with parenteral antibiotics and were cured. Streptococcus mutans is a pleomorphic, microaerophilic organism that is associated with dental caries and plaque. Differentiation of S mutans from enterococcal endocarditis is important because the former condition can be treated for a shorter period of time with penicillin alone, without the addition of aminoglycoside antibiotics.     

First Reported Human Case of Native Mitral Infective Endocarditis Caused by Streptococcus canis

A 65 year-old woman was admitted for acute heart failure and severe sepsis revealing definite mitral infective endocarditis with severe regurgitation, complicated by multiple embolisms. Three blood cultures yielded a group G Streptococcus canis strain. Urgent surgery was performed with bioprosthetic valve replacement. Polymerase chain reaction analysis of the valve found S canis DNA. Amoxicillin and gentamicin were given for 2 weeks followed by 4 weeks of amoxicillin alone. She reported contact with a dog without bite. S canis has been reported to cause zoonotic septicemia but to our knowledge, this is the first human case of native valve infective endocarditis.     

Absence of a postantibiotic effect in experimental Pseudomonas endocarditis treated with imipenem, with or without gentamicin

We found an in vitro postantibiotic effect (PAE) of 3-4 h for imipenem and of approximately 5 h for imipenem plus gentamicin against Pseudomonas aeruginosa. We therefore evaluated these antibiotics in a rat model of pseudomonas endocarditis. A rapid bactericidal effect was initially observed in vegetations from rats treated with imipenem alone or in combination with gentamicin. Bacterial counts rose rapidly, however, as soon as levels of imipenem in vegetations fell below the minimal inhibitory concentration (i.e., no PAE was demonstrated). Levels of gentamicin in vegetations were similar to those that had enhanced the bactericidal effect of imipenem and had resulted in a 5-h PAE in vitro. The presence in vitro of a PAE for imipenem, with or without gentamicin, does not necessarily predict its presence in vivo in pseudomonas endocarditis in the rat.     

Efficacy of teicoplanin for enterococcal infections: 63 cases and review.

Sixty-three cases of monomicrobial enterococcal infections treated with teicoplanin in two open clinical studies in Europe from 1982 to 1989 are presented. Infections were documented as endocarditis (n = 18); septicemia (n = 8); and urinary tract (n = 29), skin/soft-tissue (n = 6), or bone/joint (n = 2) infections. A total of 63 enterococcal strains were isolated; all of 29 strains tested were susceptible to teicoplanin (geometric mean MIC, 0.16 micrograms/mL; range, 0.06-0.5 micrograms/mL). Forty-eight patients were treated with teicoplanin alone and 15 were treated with teicoplanin in combination with an aminoglycoside. The rate of clinical cure was 84.1%; 4.8% of patients clinically improved, 7.9% had clinical recurrence, and 3.2% did not respond to therapy. Bacteriologic eradication was observed in 87.2% of patients; persistence, in 3.2%; recurrence, in 3.2%; and reinfection, in 4.8%. One case was not evaluable bacteriologically. Of 18 patients with endocarditis, 15 were cured with a mean daily dose of 5.4 mg/kg--six with monotherapy and nine with combination therapy. All patients with urinary tract infections were treated with monotherapy, and 89.7% were cured (mean daily dose, 4.6 mg/kg). Lower rates of clinical cure and bacteriologic eradication were observed in septicemic patients without endocarditis (62.5%). This study demonstrated a good efficacy of teicoplanin for the treatment of enterococcal infections due to susceptible strains, but further clinical studies would be useful for establishing optimal dosage and the indications for combination therapy, especially for severe infections.     

Penicillin therapy for treatment of experimental endocarditis caused by viridans streptococci in animals

We studied the efficacy of penicillin and penicillin combined with streptomycin in the treatment of experimental endocarditis caused by viridans streptococci that are susceptible, tolerant, or relatively resistant to penicillin. Rabbits with experimental endocarditis were treated with procaine penicillin (1.5 X 10(5) U/kg) administered twice daily or with procaine penicillin (1.5 X 10(5) U/kg) plus streptomycin (20 mg/kg) administered twice daily for five days. Compared with control animals, animals treated with penicillin alone experienced a significant reduction (P less than .001) of colony forming units per gram of cardiac valve vegetations when infected with streptococci that are susceptible, tolerant, or resistant to penicillin. This antibiotic alone was less effective against streptococci that were tolerant or resistant to penicillin than against streptococci susceptible to the drug (P less than .01). The combination of penicillin and streptomycin was more effective therapy than was penicillin alone in animals with penicillin-tolerant or penicillin-resistant streptococci causing endocarditis (P less than .01). Penicillin-streptomycin therapy was less active against penicillin-resistant strains than against either penicillin-tolerant (P less than .04) or penicillin-susceptible (P less than .01) strains. The results of our study suggest that tolerance or relative resistance to penicillin in strains of viridans streptococci influences the response to therapy with penicillin alone or penicillin combined with streptomycin in the treatment of experimental endocarditis caused by viridans streptococci.