I am starting this post with a vignette from an article by Henry F. Chambers, M.D., and Arnold S. Bayer, M.D. August 6, 2020
N Engl J Med 2020; 383:567-576.
A 72-year-old man with type 2 diabetes mellitus, stage 2 chronic kidney disease, and a history of mild aortic stenosis is admitted to the hospital with fever, dysuria, and urinary frequency. His temperature is 38.9°C, the pulse is regular at 110 beats per minute, and the blood pressure is 145/95 mm Hg. His lungs are clear; a grade 3/6 systolic ejection murmur is heard at the right upper sternal border. Laboratory tests are notable for a hemoglobin level of 12 g per deciliter, a white-cell count of 13,500 per cubic millimeter (with 80% polymorphonuclear cells), a serum glucose level of 340 mg per deciliter (18.7 mmol per liter), a serum creatinine level of 1.7 mg per deciliter (150 μmol per liter), and a urinalysis with 3+ protein, 20 to 50 white cells per high-power field, and 4+ glucose. Two blood cultures and a urine culture are positive for ampicillin-susceptible Enterococcus faecalis. How would you further evaluate and treat this patient?
Clinically Dukes criteria are used to determine whether the patient is likely to have bacterial endocarditis.
Major clinical criteria
- Positive blood culture
- Typical microorganisms (Staphylococcus aureus, viridans streptococci, Streptococcus gallolyticus, HACEK [haemophilus species, aggregatibacter (formerly actinobacillus) species, cardiobacterium species, Eikenella corrodens, and kingella species], and community-acquired enterococci in the absence of a primary focus) consistent with infective endocarditis from two separate blood cultures
- Microorganisms consistent with infective endocarditis from persistently positive blood cultures, defined as ≥2 positive cultures from blood samples drawn >12 hr apart or all of 3 or a majority of ≥4 separate cultures of blood (with first and last sample drawn at least 1 hr apart)
Single positive blood culture for Coxiella burnetii or phase I IgG antibody titer >1:800
- Positive echocardiography
- Vegetation (defined as an oscillating intracardiac mass on a valve or supporting structure), abscess, or new partial dehiscence of a prosthetic valve
- New valvular regurgitation (an increase or change in preexisting murmur is not sufficient
- Minor clinical criteria
- Presence of predisposing cardiac condition or intravenous drug use
- Temperature ≥38.0°C (100.4°F)
- Vascular phenomena such as systemic arterial emboli, septic pulmonary emboli, mycotic aneurysm, intracranial hemorrhage, conjunctival hemorrhages, or Janeway lesions
Immunologic phenomena such as glomerulonephritis, Osler nodes, Roth spots, or rheumatoid factor
- Positive blood cultures that do not meet major criteria, or serologic evidence of active infection with organism consistent with infective endocarditis
* Adapted from Li et al.
A definite diagnosis is based on two major criteria, five minor criteria, or one major criterion plus three minor criteria. Possible endocarditis is based on three minor criteria or one major criterion plus one minor criterion. If criteria for either definite or possible endocarditis are not met, the diagnosis of infective endocarditis is rejected.
What is a vegetation?
A presumed injury to the valvular endothelium or endocardium leads to the exposure of collagen tissue to which platelets adhere and a network of fibrin forms. This forms a micro-thrombus or sterile vegetation. In other words a vegetation is a micro-thrombus at the site of exposed collagen tissue where the endothelium has been denuded.
Why do the vegetations form an excellent culture medium for bacteria to grow on?
There is usually a poor immune response to the presence of bacteria in the host: neutrophils do not penetrate the vegetation well nor do antibodies and other host defence mechanisms which activate the alternative pathway of the complement cascade: bind the acute-phase reactant C-reactive protein: activate procoagulant activity on the surface of endothelial cells: upon binding to epithelial and endothelial cells and macrophages, induce production of cytokines, nitric oxide, and chemokines: initiate the influx of neutrophils. Hence bacteria can grow unhindered inside the vegetation.
What is the usual source of the bacteria?
The mouth is where most of the bacteria reside in the body. They lie under the dental plaque and in infected gums and gum abscesses. Dental procedures lead to the invasion of blood with oral bacteria. Infected skin lesions can also be the source. Invasive procedures like injections (specially when drug addicts do not adhere to sterile techniques), biopsies can be the source, indwelling transcutaneous devices such as IV lines, CV lines can all serve as conduits for the entry of bacteria into the blood.
What are the characteristics of vegetations that lead to most of the clinical features of bacterial endocarditis?
High bacterial densities, rapid growth of the vegetation, and friability and fragmentation of the growing vegetation, drive the four mechanisms that are responsible for most of the clinical features of infective endocarditis and its complications.
What are the clinical features driven by the vegetations?
The clinical features of infective endocarditis and its complications driven by the vegetations are:
- valvular destruction,
- paravalvular extension of infection, and
- heart failure;
- microvascular and large-vessel embolization; metastatic infection of target organs (e.g., the brain, kidneys, spleen, and lungs);
- immunologic phenomena such as hypocomplementemic glomerulonephritis and
- false positive serologic findings of rheumatoid factor,
- antineutrophil antibodies, or syphilis.
What are the cardiac conditions that predispose to infective endocarditis?
These include congenital disease (e.g., ventricular septal defect and bicuspid aortic valve) and acquired valvular disease (e.g., degenerative valvular disease, aortic stenosis, and rheumatic heart disease).
Rheumatic heart disease, the most common predisposing condition for infective endocarditis in developing countries, is uncommon in developed countries, where the most frequent predisposing cardiac conditions are degenerative valvular diseases, congenital valvular abnormalities, and intracardiac devices.
What are the non-cardiac risk factors predisposing to endocarditis?
These include poor dentition, intravenous drug use, hemodialysis, chronic liver disease, diabetes, compromised immunity, neoplastic disease, and indwelling intravascular devices.
What is the clinical presentation of endocarditis?
If you see long term fever without an obvious cause and heart murmur, these are the two signature features of infective endocarditis, and are present in approximately 90% and 75% of patients, respectively.
Infective endocarditis may present acutely with a rapidly progressive course complicated by congestive heart failure, stroke, systemic or pulmonary embolization, severe sepsis or septic shock,
subacutely with nonspecific symptoms such as low-grade fever, malaise, chills, sweats, dyspnea, back pain, arthralgias, and weight loss over a period of weeks or sometimes months.
Microembolic or immunologic phenomena such as splinter hemorrhage, conjunctival hemorrhage, Osler nodes (distal vasculitic lesions of the fingers and toes), Janeway lesions (vasculitic lesions of the palms and soles), and Roth spots (hemorrhagic retinal lesions) are present in 5 to 10% of patients only.
What bacteria are likely to be isolated from the blood?
Worldwide, gram-positive bacteria account for approximately 80% of cases of native-valve infective endocarditis.
These bacteria include Staphylococcus aureus in 35 to 40% of cases of native-valve infective endocarditis, streptococci in 30 to 40% (viridans streptococci in approximately 20% and Streptococcus gallolyticus [formerly S. bovis] and other streptococci in approximately 15%), and enterococci in 10%.
Coagulase-negative staphylococci, a common cause of prosthetic-valve infective endocarditis, are uncommon in native-valve infective endocarditis, except for S. lugdunensis, which resembles S. aureus clinically.
HACEK species (haemophilus species, aggregatibacter [formerly actinobacillus] species, cardiobacterium species, Eikenella corrodens, and kingella species), fungi, polymicrobial infection, and, rarely, aerobic gram-negative bacilli are isolated in 5% of cases.
When can a definite pathological diagnosis be made of endocarditis?
A definite pathological diagnosis can be made if organisms are identified on histologic analysis or culture of the vegetation, intracardiac abscess, or peripheral embolus, or if evidence of a vegetation or intracardiac abscess is confirmed by histologic analysis showing active endocarditis.
A definite or possible clinical diagnosis of infective endocarditis is based on a combination of major and minor criteria that are rooted in microbiologic, echocardiographic, and clinical metrics.
How sensitive are the Dukes criteria?
The sensitivity of the modified Duke criteria for infective endocarditis is approximately 80% for definite cases and higher if possible cases are included.
These criteria have lower sensitivity in infections related to a prosthetic valve or cardiac device, endocarditis on the right side of the heart, and culture-negative infective endocarditis.
The negative predictive value is approximately 90% when criteria are not met for either definite or possible infective endocarditis.
How useful are blood cultures in bacterial endocarditis?
- Blood cultures are the most important microbiological tests for the diagnosis and treatment of infective endocarditis, and they fulfill a major Duke criterion.
- Antimicrobial therapy largely depends on the blood-culture isolate and its antimicrobial susceptibility.
- Approximately 90 to 95% of cases of native-valve infective endocarditis are blood culture–positive. To maximize recovery of a pathogen, three separate sets of blood cultures drawn 30 minutes apart are recommended before the initiation of antibiotics.
- Blood culture–negative cases are most commonly caused by recent administration of antimicrobial agents or by organisms that grow poorly or not at all in standard blood culture media (e.g., bartonella species, Coxiella burnetii, Tropheryma whipplei, and legionella). Serologic and molecular testing for likely pathogens should be performed if blood cultures are negative; this testing is guided by epidemiologic clues (e.g., C. burnetii infection may be associated with exposure to farm animals, and Bartonella quintana infection may be associated with homelessness).
On what is molecular diagnosis based on?
- Molecular diagnosis is based on nucleic acid amplification by polymerase chain reaction (PCR),
- For PCR diagnostic tests, the reported sensitivities are 33 to 90% and the reported specificities are 77 to 100%. The preferred specimen for molecular assays is an excised valve or vegetation.
- Plasma DNA amplification assays may assist in microbiologic diagnosis in cases in which the pathogen is difficult to determine.
What is the role of the echocardiogram in the management of bacterial endocarditis?
Echocardiography is an essential tool in the diagnosis and management of infective endocarditis. The sensitivity for detection of vegetations in native-valve infective endocarditis is 50 to 60% with transthoracic echocardiography (TTE) and 90% or more with transesophageal echocardiography (TEE). The specificities of both are approximately 95%. Because TTE is also less sensitive than TEE for detecting intracardiac complications (e.g., paravalvular abscess), TEE is preferred to rule out infective endocarditis in patients in whom this condition is suspected and to assess intracardiac complications.
Among newer forms of imaging, the most widely studied is 18F-fluorodeoxyglucose cardiac positron-emission tomography (PET) plus computed tomography (CT). PET-CT is most applicable to the diagnosis and evaluation of prosthetic-valve infective endocarditis; its role in native-valve infective endocarditis is poorly studied and unclear.
What antibiotic therapy will you choose and why?
Always use combination therapy. In general, vancomycin plus ceftriaxone is a reasonable choice for empirical therapy to cover likely pathogens while cultures are pending in patients with native-valve infective endocarditis.
Infective endocarditis that is caused by penicillin-non susceptible strains of viridans streptococci, S. gallolyticus, abiotrophia species, or granulicatella species can be treated with a combination of penicillin or ceftriaxone plus gentamicin; vancomycin monotherapy is an option, although there is less overall experience with this agent.
An antistaphylococcal penicillin (e.g., oxacillin) is the drug of choice for infective endocarditis that is caused by methicillin-susceptible strains of S. aureus (MSSA). Randomized, controlled trials have shown that combination therapy with an antistaphylococcal penicillin and either gentamicin or rifampin does not improve outcomes and is associated with adverse events; therefore, this combination is not recommended. Cefazolin is a reasonable alternative for patients with MSSA who cannot receive penicillin without adverse effects. One concern with cefazolin is that some strains have an “inoculum effect,” which is defined as an increase in the broth dilution minimum inhibitory concentration (MIC) to 16 μg per milliliter or greater at an inoculum of 5×107 CFU per milliliter (100 times the standard inoculum of approximately 5×105 CFU per milliliter). This inoculum effect, which is due at least in part to hydrolysis of cefazolin by staphylococcal penicillinase, may be associated with clinical failure.
Daptomycin or vancomycin monotherapy is recommended for treatment of native-valve infective endocarditis caused by methicillin-resistant S. aureus (MRSA)
Combination therapy is recommended for the treatment of enterococcal infective endocarditis. Penicillin or ampicillin in combination with low-dose, synergistic gentamicin has been the standard treatment for decades. The usefulness of this regimen is limited by gentamicin toxicity and an increasing incidence of high-level resistance to gentamicin that indicates a lack of synergy. Observational data suggest that a 6-week course of ampicillin plus ceftriaxone is an acceptable alternative for treatment of infective endocarditis caused by ampicillin-susceptible strains of E. faecalis. If the ampicillin–gentamicin combination is used, the efficacy of combination therapy for 2 weeks followed by ampicillin alone for 4 to 6 weeks may be similar to that of the standard combination regimen for 4 to 6 weeks and is less toxic.
What are the the three main indications for surgery in bacterial endocarditis?
The three main indications for surgery in patients with native-valve infective endocarditis are heart failure due to valvular dysfunction or perforation, uncontrolled endocardial infection (e.g., paravalvular extension or persistent bacteremia), and prevention of systemic embolization, especially to the brain.
One small randomized, controlled trial compared early surgery during the initial hospitalization and within 48 hours after randomization (in 37 patients) with conventional treatment (in 39 patients) in patients with endocarditis on the left side of the heart, severe valvular regurgitation (without heart failure), and large vegetations (>10 mm in diameter). Early surgery significantly reduced the risk of the combined end point of in-hospital death or embolic events within 6 weeks after randomization, but this decreased risk was driven entirely by decreases in the risk of systemic embolism.
Modified Duke criteria for the clinical diagnosis of infective endocarditis are not based on the results of molecular diagnostic testing. As these methods improve in accuracy and become routinely available, their role in diagnosis will need to be taken into account.
Whether routine brain magnetic resonance imaging (MRI) and other advanced imaging techniques such as PET-CT improve the diagnosis, treatment, and outcomes in patients with native-valve infective endocarditis is unclear. MRI is more sensitive than CT for detecting central nervous system (CNS) lesions, and the presence of asymptomatic embolic lesions in patients with suspected infective endocarditis is a minor criterion in support of the diagnosis. Routine brain MRI has been recommended to detect silent CNS emboli in patients who are candidates for valvular surgery, although whether this improves outcomes is unknown.