Sunday, August 31, 2008

Antimicrobial therapy of prosthetic valve endocarditis

Antimicrobial therapy of prosthetic valve endocarditis

Author
Adolf W Karchmer, MD
Section Editor
Stephen B Calderwood, MD
Deputy Editor
Elinor L Baron, MD, DTMH



Last literature review version 16.2: May 2008 | This topic last updated: May 12, 2008 (More)


INTRODUCTION — Infection of a prosthetic heart valve can be difficult to diagnose and manage. Optimal treatment of prosthetic valve endocarditis (PVE) requires:

• Identification of the causative microorganism.
• Selection of a bactericidal antimicrobial regimen of proven efficacy.
• A clear understanding of the intracardiac pathology and attendant complications of PVE.
• Surgical intervention, especially when infection has extended beyond the valve to contiguous cardiac tissue.

The antimicrobial therapy of prosthetic valve endocarditis will be reviewed here. The pathogenesis, microbiology, pathology, clinical features, diagnosis, prevention, and surgical management of PVE are discussed separately. (See "Presentation and diagnosis of prosthetic valve endocarditis" and see "Surgery for prosthetic valve endocarditis").

GENERAL PRINCIPLES — Treatment of PVE with antimicrobial agents alone frequently fails, and invasive infection with subsequent valve dysfunction often arises before or during therapy. Thus, all treatment for PVE should be initiated in the hospital, preferably in an institution where cardiac surgery is available. Patients should remain hospitalized until fever resolves and it is clear that surgery can be safely avoided.

It is essential to isolate the causative organism in patients with suspected PVE. For patients who are hemodynamically stable with an indolent clinical course, antibiotic therapy should be delayed pending the blood culture results. This delay allows additional blood cultures to be obtained without the confounding effect of antibiotics, which is particularly important for patients who have received recent antimicrobial agents and whose initial blood cultures may be negative.

However, patients presenting with hemodynamic instability or acute disease should receive empiric antibiotics promptly after three sets of blood cultures have been obtained. Empiric antibiotic therapy with three agents should be initiated: vancomycin, gentamicin, and either cefepime or a carbapenem. Subsequent therapy should be adjusted based on culture results; if cultures remain negative, therapy as outlined for culture negative PVE should be used (See "Culture-negative" below).

Antimicrobial treatment regimens for PVE are based upon clinical experience. The antimicrobials used to treat a specific pathogen causing PVE are generally the treatment used for that organism when it causes native valve endocarditis (NVE). Staphylococci, which commonly cause PVE, are an exception to this dictum. (See "Antimicrobial therapy of native valve endocarditis").

No randomized controlled studies have evaluated the optimal duration of therapy for PVE. Treatment guidelines from the American Heart Association (AHA) and the European Society for Cardiology (ESC) recommend that PVE should be treated with an agent(s) that is bactericidal for the isolated microorganism for at least six weeks [1-3] . We generally agree with these guidelines and recommend a minimum of six weeks of treatment.

There are small differences, described under each specific microorganism below, in recommended guidelines between the AHA and the ESC [1-3] . We are in general agreement with these guidelines.

STAPHYLOCOCCI — Treatment choices for staphylococcal PVE are not contingent on whether the pathogen is coagulase-negative or S. aureus, unlike most other types of staphylococcal infections [4,5] (show table 1). The primary consideration in choosing therapy hinges upon whether or not the organism is sensitive to methicillin and other beta-lactam antibiotics. In addition, PVE caused by S. aureus frequently requires prompt surgical intervention. (See "Surgery for prosthetic valve endocarditis" section on "Microorganisms usually requiring surgery").

Antimicrobial treatment of staphylococcal PVE requires combination therapy. We agree with the AHA and ESC who recommend a triple drug regimen, as described below.

Evidence to support a triple-drug regimen (with one drug being rifampin) comes from animal models of prosthetic device infection and retrospective clinical series [6-10] . A retrospective study of valve cultures from 61 patients with staphylococcal PVE treated surgically found that valves from patients receiving combination therapy were 5.9 times more likely to be culture-negative than those receiving monotherapy when results were adjusted for duration of therapy before surgery [6] . Although the numbers are too small to analyze, all six patients who received a triple-drug regimen that included rifampin, had negative valve cultures at surgery.

Methicillin susceptibility — Vancomycin is the critical drug for isolates resistant to methicillin, while a semisynthetic penicillinase-resistant penicillin (nafcillin, oxacillin) is the mainstay of therapy for isolates susceptible to methicillin. In patients with penicillin allergy that does not involve anaphylaxis, swelling, or hives, the AHA recommends that a first generation cephalosporin can substitute for nafcillin or oxacillin. We agree with this recommendation.

If the organism is susceptible to gentamicin by routine testing, this should be the second agent, with rifampin as the third agent. (See "Rifampin" below). The aminoglycoside should be administered for the initial two weeks of treatment, after which it can be discontinued and the other two agents continued for at least four additional weeks. If the organism is resistant to gentamicin, an alternative aminoglycoside should be sought based upon antibiotic susceptibilities.

If the isolate is resistant to all available aminoglycosides, a fluoroquinolone to which the strain is highly susceptible should be used [11-13] . If a fluoroquinolone is used in lieu of an aminoglycoside, we prefer to continue the three-drug regimen for the entire course of treatment. When the isolate is resistant to all aminoglycosides and fluoroquinolones, daptomycin [14] , linezolid [15] , or trimethoprim-sulfamethoxazole could be considered as a third drug for the initial two weeks of therapy. If breakthrough bacteremia or microbiologic failure occurs in patients receiving daptomycin, the isolate recovered from the breakthrough bacteremia should be tested for the development of daptomycin resistance [16] .

Optimal therapy of PVE caused by methicillin-resistant S. aureus with reduced vancomycin susceptibility, has not been established. Although linezolid and daptomycin are often active against these organisms, clinical experience in the treatment of PVE is limited [14,17] .

Rifampin — Rifampin appears to have the unique ability to kill staphylococci that are adherent to foreign material, based upon in vitro data, evidence from animal model experiments, and clinical observations [4,8-13,18] . This drug is an essential component of regimens used to treat staphylococcal PVE. However, bacterial cells have a relatively high intrinsic mutation rate for the gene controlling the rifampin site of action. These mutations allow the selection of a rifampin-resistant subpopulation when large numbers of staphylococci are exposed to ineffective rifampin-containing regimens [4,11] .

To protect against the emergence of resistance, the recommended regimens for staphylococcal PVE (see "Staphylococci" above) ideally contain two additional antimicrobials, which should be identified prior to the initiation of rifampin, if at all possible. Thus, a regimen with two other drugs to which the staphylococci are susceptible should be in place at the time rifampin is begun. If the isolate is not sensitive to two additional antimicrobials, therapy with a single antistaphylococcal agent should be administered for three to five days before beginning rifampin. This strategy may reduce the total number of staphylococci at the site of infection and thus diminish the probability that a rifampin-resistant subpopulation will emerge. Nevertheless, susceptibility to rifampin should be reassessed when regimens containing rifampin fail [8] .

STREPTOCOCCI — Combination therapy with a beta-lactam antibiotic and an aminoglycoside (if the isolate is susceptible) is the preferred regimen for streptococcal endocarditis due to synergistic killing of the organism when two antibiotics are used in combination [19] .

Based on in vitro studies, clinical series, and experience of experts, penicillin plus gentamicin is recommended for the therapy of PVE caused by penicillin-susceptible streptococci (minimum inhibitory concentration [MIC] <0.12 mcg/mL) (show table 2) [5,20] . Gentamicin, if the isolate does not exhibit high level resistance (see "Enterococci" below), should be given only during the initial two weeks of treatment. Streptomycin, if the isolate does not possess high level resistance, may be given in lieu of gentamicin to achieve the same effect, but gentamicin is more commonly used in clinical practice due to the wider availability of gentamicin serum levels, and because dosing regimens for gentamicin are more familiar to most clinicians than for streptomycin [21,22] . For these reasons, we recommend gentamicin if the isolate is susceptible. Penicillin, a cephalosporin, or vancomycin can be used alone if aminoglycoside therapy is relatively contraindicated [4] .

If the streptococcus is relatively resistant to penicillin (MIC ≥0.12 mcg/mL), the AHA recommends that the aminoglycoside should be continued for the entire four to six weeks of therapy, if not precluded by nephrotoxicity (show table 3). We agree with the AHA that the aminoglycoside be continued for the duration of treatment. In contrast, the ESC recommends that the aminoglycoside only be given during the initial two weeks of treatment even when the isolate is relatively resistant to penicillin. Although the AHA recommends that gentamicin be dosed once daily, we, as well as the ESC, advocate three equally divided doses.

Among patients who are allergic to penicillin, vancomycin is advised for those with immediate type reactions (urticaria or anaphylaxis). Cefotaxime or ceftriaxone may be used in non-immediate allergies. (See "Penicillin and related antibiotic allergy; skin testing; and desensitization" section on "Cephalosporins").

ENTEROCOCCI — To achieve bactericidal activity against enterococci requires the synergistic interaction of a cell wall active agent (penicillin, ampicillin, or vancomycin) and an aminoglycoside (gentamicin or streptomycin) [19] . (See "Antimicrobial therapy of native valve endocarditis"). To achieve this interaction the organism must not be resistant to the cell wall active agent at achievable serum concentrations and must not be resistant to gentamicin at 500 mcg/mL or streptomycin at 1000 mcg/mL in broth or at 2000 mcg/mL when using cultures on agar. Growth in the presence of the aminoglycoside at these concentrations indicates high-level resistance and precludes synergy when the aminoglycoside is used. Resistance to gentamicin at this concentration also indicates that synergy cannot be achieved with netilmicin, tobramycin, amikacin, or kanamycin. High-level resistance to gentamicin and streptomycin are mediated by two independently acquired genes; hence, organisms should be tested for high-level resistance to each of these drugs.

Based on in vitro studies, animal models, and clinical series [23] , we, along with the AHA and ESC recommend combination therapy with a cell wall active agent (penicillin, ampicillin, or vancomycin) plus an aminoglycoside (usually gentamicin or streptomycin) for treatment of PVE caused by enterococci (as long as the strain is confirmed to be susceptible).

Cephalosporins are not usually active against enterococci and also do not interact with aminoglycosides to result in bactericidal synergy. The cephalosporins should not be used alone as the cell wall active agent in the treatment of enterococcal PVE. In past years, the regimens outlined for treatment of enterococcal PVE in the following tables had been predictably bactericidal (show table 4 and show table 5) [23] . However, antibiotic resistance among enterococci has become significantly more common, necessitating that each strain causing endocarditis be carefully tested in order to select a synergistic regimen [24,25] .

Penicillin/ampicillin resistance will most commonly be due to alterations in penicillin-binding proteins. In that situation, vancomycin is the cell-active agent of choice. Occasionally, E. faecalis may be resistant to penicillin and ampicillin by virtue of beta-lactamase production. In this instance vancomycin or ampicillin-sulbactam could be used. (See "Mechanisms of antibiotic resistance in enterococci").

If the enterococcus has high-level resistance to both streptomycin and gentamicin, synergy is not feasible and an aminoglycoside should not be administered. When resistance precludes bactericidal therapy, a prolonged course of 8 to 12 weeks of one of the cell wall active agents should be administered instead, but therapy in patients with native valve endocarditis has only a 40 percent chance of being successful [26] .

Although the data are limited, in the setting of progressive nephrotoxicity, the duration of aminoglycoside administration may be reduced to less than six weeks with no decrease in cure rates. This was illustrated in a prospective study from Sweden of 93 episodes of enterococcal endocarditis that included 27 cases of prosthetic valve endocarditis [27] . Clinical cure was achieved in 75 of 93 episodes (81 percent) overall, and in 21 of 27 (78 percent) with PVE. In patients who achieved clinical cure, a cell wall-active antimicrobial therapy was given for a median of 42 days, and a synergistic aminoglycoside was added for a median of 15 days.

Optimal therapy of PVE caused by vancomycin-resistant E. faecium (VRE) organisms, which often are also resistant to penicillin and ampicillin, and highly resistant to gentamicin and streptomycin, has not been established. VRE are occasionally susceptible to penicillin and ampicillin and may not have high-level resistance to streptomycin and gentamicin. A full evaluation of the isolates resistance profile is required to select optimal therapy.

Although quinupristin-dalfopristin (E. faecium only) and linezolid (E. faecium and E. faecalis) are often active against these organisms, their effectiveness in the treatment of PVE caused by VRE is not known [28] . The following table outlines possible regimens for PVE caused by VRE (show table 6). Surgical intervention during suppressive bacteriostatic therapy should be strongly considered when PVE is caused by highly resistant enterococci. (See "Treatment options for infections caused by vancomycin-resistant enterococci — Human studies").

HACEK — Because some of these organisms are ampicillin-resistant due to the production of beta-lactamase, and all are highly susceptible to third generation cephalosporins, we, along with the AHA recommend therapy for HACEK PVE with one of the following antibiotics: ceftriaxone, cefotaxime, or a comparable third generation cephalosporin; ampicillin-sulbactam; or ciprofloxacin (recommended only for patients unable to tolerate cephalosporin or ampicillin therapy), administered for six weeks (show table 7). The ESC recommends ampicillin (if the organism is susceptible) or a third generation cephalosporin. Patients with HACEK PVE, who do not have valvular dysfunction, generally can be cured with antibiotics alone [29] .

CORYNEBACTERIA (DIPHTHEROIDS) — If the strain is susceptible to gentamicin (MIC <4.0 mcg/mL), penicillin plus gentamicin will result in synergistic bactericidal activity and is recommended as therapy. Gentamicin resistance precludes bactericidal synergy [30] . Vancomycin is bactericidal against diphtheroids and is recommended for therapy when strains are resistant to gentamicin or when patients are allergic to penicillin.

GRAM-NEGATIVE BACILLI — We recommend treatment of PVE caused by gram-negative bacilli be based upon the susceptibility of the causative organism. Where possible, a synergistic bactericidal regimen should be used. Treatment for Pseudomonas aeruginosa is based upon experience in patients with NVE. If the organism is susceptible, high dose tobramycin (8 mg/kg per day in three equally divided doses IV or IM to achieve peak concentrations approaching 15 mcg/mL) plus ticarcillin, piperacillin, cefepime, or ceftazidime is recommended. (See "Antimicrobial therapy of native valve endocarditis").

Surgery to excise the infected valve is often required in gram-negative bacillus endocarditis, especially that caused by P. aeruginosa or when infection involves the left-sided heart valves. (See "Surgery for prosthetic valve endocarditis").

FUNGI — No randomized, controlled studies have evaluated the optimal therapy for fungal PVE. We, along with most infectious disease specialists, recommend a combined approach that utilizes both antifungal agents and valve replacement [31] . Amphotericin B (daily doses ranging from 0.7 to 1.0 mg/kg per day) is the antimicrobial of choice for treatment of fungal PVE as the greatest clinical experience in treating fungal PVE is with this agent.

For the treatment of endocarditis caused by mycelial fungi, such as Aspergillus or Mucor species, somewhat larger doses are used (1.0 to 1.5 mg/kg per day). We recommend for the treatment of fungal endocarditis, amphotericin B be combined with flucytosine (150 mg/kg per day divided into four doses with adjustments for renal dysfunction) in an attempt to achieve a synergistic effect. This initial phase of treatment is usually given for at least six weeks. (See "Clinical use of flucytosine").

The use of a lipid formulation of amphotericin B in lieu of amphotericin B desoxycholate for the treatment of fungal endocarditis has not been evaluated. Nevertheless, if renal dysfunction complicates amphotericin B treatment, substitution of a lipid formulation is justified.

Early surgical intervention is considered by most experts to be a standard element of treatment for fungal PVE. (See "Surgery for prosthetic valve endocarditis" section on "Microorganisms usually requiring surgery").

Since the potential for relapse is high in Candida PVE, even with surgical intervention, we along with most infectious disease specialists recommend a suppressive second phase of oral therapy with fluconazole (200 to 400 mg daily or another triazole) for prolonged periods, if not indefinitely [32-34] . (See "Candida endocarditis").

Successful treatment of Candida PVE without surgery has been reported in a few case reports using a combination regimen of fluconazole and caspofungin or fluconazole and amphotericin B [35,36] .

CULTURE-NEGATIVE — Many native and prosthetic valve endocarditis patients with negative blood cultures have been rendered culture-negative by virtue of prior antibiotic therapy. The therapy to which they have been exposed is a clue and consideration in selecting empiric treatment. (See "Culture-negative endocarditis").

In the absence of clinical clues to a specific etiology, we along with the AHA recommend that treatment for culture-negative PVE, with onset within the first year following valve surgery, should include vancomycin, gentamicin, cefepime, and rifampin [1] . For patients with the onset of disease 12 months or more after valve implantation, the AHA and we recommend treatment with ceftriaxone, gentamicin, and doxycycline [1] . Aggressive efforts must be made to identify a causative agent. (See "Diagnostic approach to infective endocarditis" section on "Culture-negative IE"). Epidemiologic considerations should be weighed carefully. As an example, in some regions of the world Coxiella burnetii is a common cause of culture-negative PVE. The possibility of fungal endocarditis should be considered, especially in patients with a complex perioperative course. If unexplained fever persists in the face of empiric therapy, surgery to obtain a vegetation for microbiologic evaluation should be considered [4,20] . (See "Q fever endocarditis").

SUMMARY AND RECOMMENDATIONS

• Treatment of prosthetic valve endocarditis is more difficult than treatment of native valve endocarditis and may require surgical replacement of the prostheses in addition to antibiotic therapy. (See "Introduction" above).
• The antimicrobial regimens used are targeted to a specific pathogen, thus identification of the causative organism is critical. (See "Introduction" above and see "Diagnostic approach to infective endocarditis").
• We recommend the same treatment regimens for a specific pathogen causing PVE as is used for that organism when it causes native valve endocarditis (Grade 1B). An exception is staphylococcal endocarditis; we recommend treatment with three agents for this microorganism. (Grade 1B). (See "General principles" above and see "Staphylococci" above).
• We recommend treatment of PVE with an agent(s) that is bactericidal for the isolated microorganism for at least six weeks (Grade 1C). (See "General principles" above).
• Treatment choices for staphylococcal PVE are the same regardless of whether the pathogen is coagulase-negative staphylococci or S. aureus. The primary consideration in choosing therapy hinges upon whether or not the organism is sensitive to methicillin and other beta-lactam antibiotics. (See "Staphylococci" above).
• We recommend a treatment regimen for enterococcal PVE that includes the synergistic interaction of a cell wall active agent (penicillin, ampicillin, or vancomycin) and an aminoglycoside (gentamicin or streptomycin). (Grade 1B). (See "Enterococci" above).

No comments: