Barnard Endocarditis

S. aureus endocarditis occurs in four clinically distinct populations [158] : intravenous drug users (IVDUs); patients with prosthetic valves; patients with health-care-acquired (nosocomial or nosohusial) endocarditis; and non-IVDU patients with community-acquired endocarditis. This chapter will focus on the latter group, as the former groups are discussed in other chapters.

Recent studies have demonstrated that S. aureus has become the leading cause of endocarditis, accounting for approximately 30% of cases [158,159]. Of these, approximately 87% are NVE [158]. Although a large proportion of cases of S. aureus IE are community-acquired [160,161], there is an increasing prevalence of health-care-associated disease, owing in part to the growing use of interventional procedures and implantable devices [159]. Community-acquired S. aureus NVE may involve right-sided and/or left-sided cardiac structures. Right-sided disease typically has high cure rates with relatively short-course medical therapy alone [52]. In non-IVDUs, S. aureus predominantly involves the left-side and is associated with mortality rates ranging from 25-50% [2,52]. S. aureus NVE is also associated with higher rates of embolization (cerebrovascular and systemic) and persistent bacteremia when compared to NVE due to other pathogens [159,162].

The management of S. aureus infections in general, and NVE in particular, has become increasingly difficult owing to evolving mechanisms of antibiotic resistance. Penicillin was introduced into clinical practice in 1941 and it was demonstrated to be an effective anti-staphy-

lococcal agent. Within one to two years of its introduction, however, highly penicillin-resistant isolates of S. aureus were found [163]. The mechanism of resistance is due to acquisition of a plasmid-mediated penicillinase. Penicillin resistance propagated rapidly, and currently, > 95% of S. aureus strains are resistant to penicillin [164]. However, in the rare instance where an isolate responsible for IE is susceptible to penicillin, it should be used in high doses (e.g., penicillin G 24 million units/day IV).

The emergence of penicillin-resistant S. aureus during the 1940s prompted the development of a new class of penicillins that were specifically targeted against these penicillin-resistant strains. The first representative of this class, methicillin, was introduced in 1951. By the mid-1950s, however, methicillin-resistant strains of S. aureus (MRSA) were prevalent. This resistance is mediated by the production of an alternate penicillin-binding protein, termed PBP-2a, which is encoded by the mecA gene [165]. PBP-2a has low affinity for P-lactams, thus allowing synthesis of the bacterial cell wall despite the presence of normally lethal P-lactam concentrations [166]. In addition to mediating resistance to methicillin (and other semi-synthetic penicillinase-resistant penicillin), it also provides resistance to cephalosporins, cephamycins, and carbapenems [166]. The mecA gene is encoded on a mobile genetic element, the staphylococcal chromosomal cassette mec (SCCmec), which also contains insertion sites for plasmids and transposons that facilitate acquisition of resistance to other antibiotics. Consequently, cross-resistance to other classes of antibiotics, such as erythromycin, clindamycin, gentamicin, trimethoprim-sulfa-methoxazole (TMP/SMX), and ciprofloxacin may occur [166]. Although MRSA was typically considered a nosocomial pathogen, typing of SCCmec has identified community-associated MRSA strains (CA-MRSA) that are distinct from the hospital strains in pathogenicity and antimicrobial susceptibility [167]. Although the majority of MRSA strains causing IE are healthcare-associated [159], IE due to CA-MRSA has also been reported [168]. There is some evidence to suggest that infections with MRSA are associated with increased morbidity and mortality, when compared to infections with methicillin-susceptible S. aureus (MSSA) [169,170]; this association has also been demonstrated in endocarditis [158,160,171]. There is some con cern, however, that the increased mortality associated with MRSA infections may be biased by confounding variables, such as length of hospitalization [172] or severity of illness [173]; in other words, the colonization/infection with MRSA represents a surrogate marker of increased length of hospitalization, which, in turn, is a reflection of multiple or severe co-morbidities. This latter factor may, in fact, be the principle reason for the higher mortality rates.

The treatment of choice for MRSA, both nosocomial and community-acquired, is the glycopeptide class of antimicrobials. In North America, vancomycin is the glycopeptide commercially available. Teicoplanin has been used in other parts of the world. At appropriate doses, the efficacy of these glycopeptides in the management of S. aureus IE is comparable [174]. However, the efficacy of the glycopeptides is inferior to that of the P-lactams for the management of IE with S. aureus isolates that demonstrate in vitro susceptibility to both classes of antimicrobials [173,175,176]. This inferiority is reflected in a delayed clearance of bacteremia (i.e., >6 days), higher rates of treatment failure, and higher rates of relapse [177-179]. These effects are due to vancomycin's suboptimal pharmacokinetics (i.e., poor vegetation penetration) and pharmacody-namics (i.e., slower in vitro bactericidal effect [180] ) when compared to P-lactams. Thus, in IE with MSSA, P-lactams are the drug of choice.

More recently, strains of S. aureus with decreased susceptibility to vancomycin have been recognized. These isolates are inhibited by vancomycin concentrations of 8-16 ^g/mL, which is interpreted as "intermediate susceptibility" by CLSI (formerly NCCLS) criteria [181]. Despite this in vitro classification, infections caused by these vancomycin-intermediate S. aureus (VISA) strains have not responded well clinically when treated with vancomycin, including cases of endocarditis [182-185]. These strains appear to develop from preexisting MRSA strains under the selective pressure of prolonged and/or suboptimal administration of vancomycin [186,187]. In addition to VISA, there has also been increased recognition of het-erogeneously vancomycin-intermediate S. aureus (h-VISA) strains; these are strains of S. aureus containing subpopulations of vancomycin-resistant daughter cells, typically at a rate of one organism per 105-106 organisms, for which the apparent vancomycin MICs of the parent strain are only 1-4 mg/L (i.e., susceptible) [188]. These subpopulations typically have MICs that are two- to eightfold higher than that for the original clinical isolate. The clinical significance of h-VISA isolates remains to be fully elucidated. It has been reported in association with IE [182,189]. As well, evidence suggests that infections with such strains are associated with clinical evidence of vancomycin treatment failure (defined as persistent fever and bacteremia for >7 days after commencement of vancomycin therapy) with high bacterial load infection [190], although another study found that heteroresistance is not a common cause of persistent or recurrent bac-teremia [191]. Therefore, further studies are required to determine the frequency of h-VISA in endocarditis, as well as the significance of heterogeneity in its management.

In addition to VISA and h-VISA, there have been reports of infections with strains of S. aureus that demonstrate complete resistance to vancomycin, defined as an MIC of vancomycin > 32 |g/mL [181]. These vancomycin-resistant S. aureus (VRSA) strains remain, thankfully, relatively uncommon in the clinical setting. VRSA strains appear to differ from VISA strains with respect to their mechanisms of resistance. VISA strains undergo changes in pep-tidoglycan synthesis after prolonged van-comycin exposure, resulting in an irregularly shaped, thickened extracellular matrix on electron microscopy [192]. There is also decreased cross-linking of the peptidoglycan strands, which allows increased exposure of D-Ala-D-Ala residues [185]. These residues bind and sequester vancomycin outside the cell wall, blocking its effect within the cytoplasmic membrane. VRSA strains, on ther other hand, develop vancomycin resistance via the acquisition of the vanA operon, presumably from surrounding vancomycin-resistant E. faecalis [185,193]. These isolates produce cell wall precursors with D-Ala-D-Lac, instead of D-Ala-D-Ala, that have low affinity for vancomycin, conferring resistance.

Isolated right-sided NVE accounts for only 5-10% of cases of infective endocarditis [194]. The majority of cases occur in patients with IVDU, but 5-10% of cases occur in nonusers [195-197]. The major pathogen is S. aureus [194,197,198]. A previous major cause was rheumatic tricuspid valve disease. With medical progress, it is predominantly occurring as a complication of other cardiac anomalies, as well as from central venous/intracardiac catheteriza-tion [194,199]. Of course, it can also occur as a component of multi-valvular IE [200]. The majority of the clinical literature on the management and prognosis of isolated right-sided S. aureus NVE has been extrapolated from the experience in patients with IVDU, which is discussed in chapter 3.

The symptoms of isolated right-sided S. aureus NVE are predominantly nonspecific constitutional symptoms, i.e., fever, chills, night sweats, and malaise, which may contribute to a delay in diagnosis. The major reason for seeking medical attention is the deveopment of respiratory symptoms (e.g., dypnea, pleuritic chest pain, productive cough, hemoptysis), usually the result of septic pulmonary emboli [197]. One study suggests that the triad of recurrent pulmonary events, anemia, and microscopic hematuria (termed "the tricuspid syndrome") should raise clinical suspicion of tricuspid valve endocarditis [194]. Typically, there is a paucity of cardiac signs and symptoms, although right-sided congestive heart failure may occur.

Isolated right-sided S. aureus NVE has a low mortality. Relatively abbreviated courses of medical therapy alone produces cure rates >90% [201]. In the absence of any intracardiac or extra-pulmonary metastatic disease, right-sided NVE with MSSA may be successfully treated with as little as two weeks of a variety of intravenous anti-staphyloccocal therapies, typically a penicillinase-resistant penicillin with or without an aminoglycoside (e.g., nafcillin plus tobramycin) [202-204]. An alternative successful regimen has been ciprofloxacin (IV then oral) plus oral rifampin for four weeks [205,206]. It should be remembered, however, that this literature is based on the experience in patients with IVDU, where such regimens produced a relapse rate of ~6% [180,207], necessitating prolongation of treatment (e.g., to four weeks) for cure. Furthermore, such short-course regimens may not be appropriate in patients with cardiac or extra-cardiac complications, fever lasting > 7 days, or advanced HIV infection (i.e., CD4 count <200 cells/mm3) [208].

In right-sided NVE due to MRSA, van-comycin is currently the standard treatment, typically at doses of 30 mg/kg/24 hours in divided doses, with monitoring of serum levels [180,208]. The efficacy of vancomycin treatment for MRSA IE, however, is less than that for P-

lactams for MSSA IE, even in the management of right-sided disease [180]. As such, when van-comycin needs to be used, a more prolonged course of intravenous therapy is required. In a retrospective review of 300 cases of S. aureus right-sided NVE, chiefly composed of IVDUs, a 28-day course of vancomycin was adequate for most patients, producing a cure rate of ~70-80% [180]. However, when compared to treatment with P-lactams, the use of vancomycin was associated with delayed clearance of bacteremia and higher rates of complications.

Most of the experience with S. aureus right-sided NVE is based on patients with IVDU and suggests that valve replacement is rarely indicated. Surgery should, however, be considered in patients with vegetations >1.0 cm, as these patients are at increased risk for developing new-onset and recurrent emboli [199] and right-sided heart failure [209]. Vegetations >2.0 cm are associated with increased risk of death [210]. Persistent fever, clinically evident right-sided heart failure [198], or increased right ventricular end-diastolic dimension by echocardiography [209] have also defined subgroups of patients who subsequently required valvular surgery. The occurrence of septic pulmonary emboli, despite antimicrobial therapy, is not considered an indication for surgery if the patient is clinically improving [208,211,212]. It should be noted, however, that the experience with surgical intervention in non-IVDU patients with this infection is limited.

In general, tricuspid valve replacement has been avoided in patients with right-sided IE because of the high likelihood of contamination of the prosthetic valve with ongoing IVDU. In patients without drug use, this fear should not preclude such intervention. Alternatively, vege-tectomy (i.e., excision of the vegetation only) or tricuspid valvuloplasty can be performed. However, the preferred type of surgery remains to be determined.

Left-sided S. aureus NVE is by far more common than right-sided infection. Furthermore, it is a more virulent disease. The overall mortality rate for this infection ranges from 20-65% [201]. Even when diagnosed correctly and managed with appropriate antimicrobial therapy, the complication rate ranges from 20% to 50% [201]. Congestive heart failure is the most common complication, and it portends a poor prognosis. Neurologic manifestations occur in 20-35% of patients [158,213]. These typically occur early in the disease, either before or shortly after the administration of antibiotics [214]. Recurrent emboli are infrequent if the infection is adequately controlled with antimicrobial therapy [213,214]. Neurological complications are accompanied by high mortality rates. Therefore, rapid diagnosis and initiation of antimicrobial therapy may still be the most effective means to prevent neurologic complications.

Antimicrobial therapy, for reasons discussed previously, should include a P-lactam when possible. For the uncommon situation caused by penicillin-susceptible S. aureus, benzyl penicillin at maximal doses is the preferred agent. The treatment of choice for MSSA NVE is a penicillinase-resistant semi-synthetic penicillin (e.g., cloxacillin 2 gm intravenously every four hours). Although for other types of S. aureus infections, such as cellulitis, first-generation cephalosporins have proven useful as alternatives, the use of such agents (e.g., cefazolin) in the treatment of MSSA NVE is with caution. There have been three previously reported cases of cefazolin failure in patients with such infection. The infecting strain isolated produced P-lactamase type A, which has very high rates of cefazolin hydrolysis. Furthermore, these strains produced high amounts of the enzyme. As such, these isolates demonstrated high MICs to cefa-zolin. In the context of a cardiac vegetation, where the number of residing organisms can be as high as 1010 CFU/gram of tissue, Nannini and colleagues propose that an inoculum effect mediated clinical failure. That is, the high quantity of bacteria results in the production of large amounts of enzyme with inherently augmented cefazolin hydrolysis rates, leading to inactiva-tion of the drug and persistence of the infection. As such, the authors caution that cefazolin usage for treatment of MSSA NVE may be associated with clinical failure. It is unclear what the frequency of such isolates is in clinical practice. Therefore, semi-synthetic penicillins (or penicillin itself) should be used whenever possible. In the absence of any complications, four weeks of therapy is usually sufficient [12,52].

The addition of aminoglycosides to P-lactams produces an enhanced bactericidal effect in vitro, as well as in a rabbit experimental model of endocarditis. However, several clinical studies have failed to demonstrate a clinical benefit, as evidence by equivalent efficacy of cure rates when compared to P-lactam monotherapy, when the total length of therapy was four to six weeks. There was demonstration, though, that combination therapy did result in significantly faster clearance of bacteremia, but this did not correlate with a more rapid clinical response, as both groups of patients were febrile for approximately the same length of time. There was, however, an increased incidence of nephrotoxicity in the group receiving the aminoglycoside. As such, the use of aminoglycosides (e.g., gen-tamicin) in the management of MSSA NVE should be limited. The BSAC does not recommend it use in this setting [12], whereas the AHA recommends that if it is used, it be done only for the first three to five days of therapy for left-sided disease [52]. Furthermore, the latter group recommends regular administration of gentamicin, such as two or three times daily, rather than once-daily therapy, with a total daily dose not to exceed 3 mg/kg in patients with normal renal function.

For MRSA NVE, vancomycin is the drug of choice. However, it may be associated with suboptimal outcomes [178,179]. Optimization of dosage to achieve a one-hour serum peak concentration of 30-45 |g/mL and trough concentration of 10-15 |g/mL may be beneficial [12,52]. The BSAC recommends the use of a second antibiotic, in addition to vancomycin, either rifampicin (300-600 mg 12 hourly by mouth), gentamicin (1 mg/kg body weight eight hourly, modified according to renal function), or sodium fusidate (500 mg eight-hourly by mouth), based on susceptibility testing [12]. This suggestion, though, is based on expert opinion. Although rifampin demonstrates potent activity against S. aureus in vitro, the in vitro effect when combined with semi-synthetic penicillins, vancomycin, or amingly-cosides is highly variable [173]. As well, one study of patients with MRSA IE comparing van-comycin monotherapy to vancomycin plus rifampin showed no statistically significant difference in clinical outcome [179]. Similarly, there is insufficient published evidence robustly to demonstrate a clinical benefit for fusidic acid-based combination therapy [215].

The other major indication to use van-comycin has traditionally been in patients who are unable to tolerate P-lactams. Because of the superior efficacy of this class of antimicrobials, for patients with a questionable history of type 1, immediate-type hypersensitivity reaction to penicillin (e.g., urticaria, angioedema), skin testing should be performed to penicillin [52]. If negative, P-lactams should be instituted. Alternatively, a cephalosporin may be considered [52]; first-generation cephalosporins should be used with caution.

Given the suboptimal efficacy of glycopep-tides in the management of MRSA NVE, as well as the emergence of VISA/h-VISA/VRSA, alternative antimicrobial therapy is desired. The newer agents with the potential to address this need are the following: quinupristin/dalfopristin (Q/D), linezolid (LZL), daptomycin, and minocycline. Trimethoprim-sulfamethoxazole (TMP/SMX) may have activity as well, and thus antimicrobial susceptibility testing should be performed. The clinical experience with these agents in the management of MRSA or VISA/VRSA NVE, however, is limited. Q/D, a streptogramin antibiotic, demonstrates variable in vitro activity against MRSA isolates. Most MRSA strains possess the MLSb phenotype, rendering them cross-resistant to macrolides, lin-cosamides, and streptogramin B, mediated by methylation of the ribosomal target [216]. Expression of this phenotype may be constitutive or inducible; when it is constitutive, strains are resistant to quinupristin. The combination, Q/D, retains activity, although the bactericidal activity is reduced [216]. Furthermore, although quinupristin demonstrates homogeneous penetration into cardiac vegetations in an experimental endocarditis model, dalfopristin demonstrated a significantly decreased concentration gradient between the periphery and the core of the vegetation, implying poor penetration of the agent that maintains activity of the Q/D combination [217]. There have been few reported clinical cases in the English literature of Q/D in the treatment of MRSA NVE. It has been used successfully in 1 patient when used alone [218], and in another patient when used in combination with vancomycin and cardiac surgery [219]. However, when used in a worldwide emergency-use protocol for patients with MRSA infections intolerant of or failing prior therapy, the response rates among the few patients with endocardits was suboptimal. Only about half of the patients had a clinical response, but among patients that could be bacteriologically evaluated, both were clinical failures, suggesting that Q/D as monotherapy may not be able to consistently sterilize cardiac vegetations [220]. Further data is certainly needed.

The data supporting the use of LZL is conflicting. In a rabbit model of staphylococcal endocarditis, LZL significantly reduced bacterial vegetation densities [221]. The antimicrobial activity of LZL is not affected by inoculum size [222]. As well, there have been several cases described in which LZL was successfully used to treat MRSA or VISA endocarditis (both native and prosthetic) in cases of glycopeptide failure or intolerance [182,183,185,189,223]. However, this enthusiasm is tempered by experimental data demonstrating suboptimal activity [224], and clinical data demonstrating clinical failure and LZL-non-susceptibility [225-229]. As such, LZL may represent a therapeutic option in the management of MRSA/VISA NVE in certain populations, but emergence of resistance with clinical failure may occur.

Daptomycin is the most effective and rapidly bactericidal of the novel anti-MRSA antimicrobial agents; it produces clearance of bacteremia faster than vancomycin and the other agents

[230]. In a rat model of MRSA endocarditis, dap-tomycin produced significant decreases in the residual bacterial counts in cardiac vegetations

[231]. Similar results were obtained using simulated endocardial vegetations [232]. One case report describes the successful use of dapto-mycin the treatment of MRSA prosthetic valve endocarditis complicated by perivalvular aortic abscess with persistent MRSA bacteremia unresponsive to vancomycin therapy; surgery was not required [233]. The clinical experience is, however, limited, and an experimental model suggests that daptomycin may have limited diffusion in fibrin clots [234]. Hence, it may be predicted to be associated with clinical failure; future studies are needed.

Owing to the aggressive nature of the disease, with its associated complications, a more aggressive treatment approach has been advocated. Therefore, valve replacement surery has become an important adjunct in the management of S. aureus NVE, allowing for a higher likelihood of successfully eradicating the infection. Indications for cardiac surgical intervention have emerged and are discussed in the section "The Role of Surgery" below. Briefly, these indications include congestive heart failure, persistent bacteremia, hemodynamically significant valvular dysfunction, perivalvular extension of infection (abscess or fistula), persistent (uncontrolled) infection (e.g., increase in vegetation size after four weeks of antimicrobial therapy), and lack of effective antimicrobial therapy available (or alternatively, difficult-

to-treat pathogens). Several studies have demonstrated the beneficial role of surgery in these situations, with relatively low operative mortality rates when compared to in-hospital mortality rates with medical therapy alone, and good long-term results [158,235-238]. Although patient selection bias may contribute to the observed effect, large prospective randomized studies have not been performed, largely because they represent ethical and methodological challenges.

All Natural Yeast Infection Treatment

All Natural Yeast Infection Treatment

Ever have a yeast infection? The raw, itchy and outright unbearable burning sensation that always comes with even the mildest infection can wreak such havoc on our daily lives.

Get My Free Ebook


Post a comment