OVERVIEW: What every practitioner needs to know about native valve endocarditis
Are you sure your patient has native valve endocarditis?
Native valve endocarditis (NV-IE) is one of the most important life-threatening infectious diseases, and its timely diagnosis, antibiotic treatment, and management of complications is critical to optimal outcomes.
Typical symptoms are variable and depend on the valve involved, the age and co-morbid diseases status of the patient, the etiologic organism, the extent and location of metastatic complications, and the duration of the infection. However, virtually all patients have fever (except those who are partially treated with empiric outpatient antibiotics, antipyretics, or who have renal insufficiency).
The modified Duke criteria were introduced to provide a diagnostic schema and context for clinicians to stratify patients suspected of IE into categories of “clinically definite IE” versus “possible IE” versus “rejected for IE.” These criteria are modeled after the Jones criteria for rheumatic fever and contain a “scoring” system based on readily obtainable clinical, echocardiographic, and microbiologic yardsticks.
The “major” criteria include:
Positive blood cultures for a typical IE pathogen, such as an enterococcus or staphylococcus (usually “high-grade” = many + cultures out of many cultures obtained [e.g., 4/4 + blood cultures])
Positive echocardiogram for “signature lesions of IE” – valvular or mural vegetations or perivalvular abscesses
Clinical (auscultatory) evidence of “new” valvular regurgitation
NOTE: There are two zoonotic causes of NV-IE for which positive blood cultures are rarely found, but in which specific sero-diagnosis is required: Q Fever and Bartonella; in addition, TropherymaIE (Whipples IE) is usually diagnosed by polymerase chain reaction (PCR) of valvular tissue.
The “minor” criteria include:
Underlying valvular heart disease
Embolic complications (e.g., stroke)
Immunologic phenomena (e.g., immune complex glomerulonephritis)
Positive blood cultures – not high-grade (e.g., 2/4 +)
Echocardiogram with abnormalities (e.g., valve thickening), but no signature lesions
NOTE: With the advent of the transesophageal echo (TEE), this last “minor” criterion has been removed if this echo method is used.
The “scorecard” for diagnosing “clinically definite cases of NV-IE” is one of the following combinations:
2 major criteria, OR
1 major PLUS 3 minor criteria, OR
5 minor criteria
The “scorecard” for diagnosing “possible IE” is important, as such patients are usually treated for IE:
1 major PLUS 1 minor criteria, OR
3 minor criteria
NOTE: These latter criteria provide a specific “floor” for diagnosing definite or possible IE. Patients not fulfilling either of these metrics are categorized as “rejected for IE.”
Key Symptoms of NV-IE: Other than fever (occurring in 80-90% of proven cases), most symptoms are highly variable and non-specific. Moreover their occurrence depends on the “pace” and duration of disease, as well as complications. Sub-acute NV-IE often features fatigue, weight loss, and arthralgias. With advancing valve destruction, symptoms of CHF supervene.
Key Signs of NV-IE: Fever is the overwhelmingly most common sign. Most other signs are either due to embolic complications (e.g., stroke syndromes, Osler’s Nodes), valvular incompetence (signs of CHF), or immunologic phenomena (e.g., polyarthritis).
How did the patient develop native valve endocarditis? What was the primary source from which the infection spread?
The basic pathways by which patients develop NV-IE depend heavily on the pathogen-of-interest.
For staphylococcal IE, a number of cases can be linked to recent or concurrent skin-soft tissue syndromes or vascular catheters.
NOTE: In most cases of S. aureusNV-IE, you cannot find the incident primary source.
NOTE: A syndrome that looks exactly like tricuspid valve NV-IE can be seen in patients with MRSA infection complicated by either spontaneous septic upper extremity thrombophlebitis or septic pelvic thrombophlebitis (especially in pediatric ages).
For enterococcal IE, many cases can be linked to concomitant UTI (especially prostatic infection in older men).
For viridans group streptococcal IE, poor oral hygiene and underlying dental disease are risk factors. It is very controversial whether a single dental practitioner encounter puts patients with underlying valvular heart disease at risk for IE.
For S. bovis (now S. gallolyticus) IE, underlying gastrointestinal (GI) tract disease, especially malignancy, needs to be actively screened for.
NOTE: For S. aureus,it is no longer prudent to ascribe nosocomial or health care-associated bacteremia as “low-risk” for NV-IE. Such patients should be actively evaluated and worked up for IE; in contrast, nosocomial enterococcal IE is rarely related to IE. Moreover, coagulase-negative staphylcoccal bacteremia, which is community-acquired, should not be routinely ascribed as a “contaminant.” A number of recent cases of NV-IE caused by such organisms have been observed. In cases of particularly aggressive NV-IE caused by coagulase-negative staphylococci, microbiologic screening either by PCR or by biochemical assays should be done to diagnose S. lugdenensisIE.
Which individuals are of greater risk of developing native valve endocarditis?
In terms “valvular” risk factors for acquiring NV-IE, the number one “offender,” statistically appears to be prior cases of NV-IE, followed by mitral valve prolapse with significant mitral regurgitation, usually accompanied by thick, redundant leaflets. Now that children with complex congenital heart disease are living well into their adult years, this is representing an increasingly important risk factor for NV-IE (although some of these cases are actually “mural IE” at sites of “jet lesion flow.”) Although rare in the developed world, rheumatic heart disease remains a pivotal underlying correlate of NV-IE in underdeveloped countries.
As noted, in terms of “co-morbid disease” risk factors, it has become increasingly clear that “health care-contact” is a critical parameter in patients in the industrialized world, especially for S. aureusNV-IE. This is usually in the context of dialytic therapy (particularly hemodialysis), use of immunosuppressive therapies, and placement of central indwelling vascular catheters.
Beware: there are other diseases that can mimic native valve endocarditis:
Because of the non-specificity of many of the signs and symptoms of NV-IE, a number of other diseases can mimic this syndrome, including:
Rheumatologic disorders (especially systemic lupus erythematosis [SLE])may be accompanied by ‘fever of unknown origin’ (FUO), arthritis, skin lesions, and “marantic” (sterile) vegetative IE.
Malignancy: May see FUO and marantic IE.
Drug reactions to agents that can cause the “SLE syndrome” or those causing serum sickness
Hemoglobinopathies: Sickle cell anemia patients may present with fever, joint pains, heart murmurs, and high WBC counts, thus, mimic NV-IE.
What laboratory studies should you order and what should you expect to find?
Results consistent with the diagnosis
NOTE: Most routine lab tests are not very helpful in diagnosing NV-IE. Abnormalities of these tests usually track with the duration of disease (e.g., anemia, thrombocytopenia). However, there are a few routine labs that may provide “classic clues” to an underlying NV-IE, including:
Positive rheumatoid factor
“Active urine sediment,” especially hematuria, with or without pyuria
Hypocomplementemia (especially in NV-IE patients with renal or joint involvement)
Other routine tests, such as chest X-ray and EKG, may be helpful in select circumstances as a good hint of underling NV-IE (e.g., chest X-ray consistent with multiple pulmonary emboli or showing specific chamber enlargements; EKG showing heart block).
Results that confirm the diagnosis
As noted, positive blood cultures are the sine quo nonof IE.
Occasionally, culture of embolic tissue yields the organism, when blood cultures are negative. This is usually seen in fungal IE.
The most important causes of blood culture-negative NV-IE include:
Recent prior outpatient antibiotics (Note: such antibiotics can be oral and not particularly active against the pathogen-of-interest)
Zoonotic IE, most commonly Q Fever; Bartonella IE (especially B. quintana); Brucella IE
NOTE: some key associations here: Livestock means Q Fever and Brucella; homeless-alcoholism means Bartonella; Southern Europe and Mediterranean areas means Q Fever.
Confirming zoonotic IE: Q Fever and Bartonella mean serologic diagnosis (although Bartonellawill grow slowly on enriched media).
Serologies of B. quinatana and B. hensalae may cross-react.
Serologic diagnosis of BrucellaIE is not as important as growing the organism. Alert your Microbiology lab that you are thinking of Brucellaepidemiologically for use of enriched media.
NOTE: Although high phase 1 IgG or IgM antibody (≥ 1:800) is classically used to diagnose Q Fever IE. Recent studies from Raoult’s lab question the accuracy of using these antibodies, especially to follow treatment efficacy.
Slow growing organisms: HACEK or Abiotrophia (and other nutritionally variant oral streptococci)
NOTE: The value of “holding routine blood cultures” in the Micro lab for 2-4 weeks to pick up unusual slow-growing organisms has been hotly debated and probably not cost-effective or diagnostically-relevant. Using the correct media is more important!
T. whippeli: As noted, this is usually a histopathological and/or PCR diagnosis on biopsied or resected tissues.
NOTE: Some articles have pushed the use of broad-range PCR based on consensus 16S ribosomal RNA sequences for common IE pathogens as a diagnostic adjunct (especially in blood culture-negative cases). Such approaches are not widely available, but you can seek out these labs in the literature to run your serum samples in this regard.
What imaging studies will be helpful in making or excluding the diagnosis of native valve endocarditis?
The trademark imaging study for NV-IE is the cardiac ECHO. There are many “urban legends” about the use of ECHO in IE. However, recent data would make the following concepts valid:
The sensitivity of the transthoracic ECHO (TTE; about 60%) is significantly inferior to the transesophageal ECHO (TEE; >90%) for defining “vegetations” in adult cases of NV-IE.
A technically “good” TTE is not the equivalent of a TEE.
The right-sided valves are not seen as well by the TTE as the TEE (especially the pulmonic valve).
Periannular complications of NV-IE are best interrogated by the TEE.
“Mobile vegetations” do not necessarily portend higher embolic rates versus sessile lesions.
“Size” matters: Vegetations larger than 1 cm in dimension are associated with increased embolic risks; frank valve disruption and increased risk of developing CHF within the 30d hospitalization (or outpatient treatment) period.
Stable vegetation size on therapy does not mean a failing antibiotic regimen.
More than two TEEs in suspected NV-IE patients does not further increase the yield.
There are compelling prognostic implications of having a case of IE, especially in terms of long-term follow-up and risk of eventual congestive heart failure (CHF) and need for valve replacements; also, there are obvious prophylactic considerations when such patients undergo selected invasive procedures in the future.
NOTE: key size correlates: 1) Vegetations larger than 1 cm PLUS IE of anterior MV leaflet PLUS S. aureus equal very increased risk of embolic event (especially central nervous system [CNS]). In this scenario, some surgeons have recommended early “preemptive” mitral valve (MV) repair if feasible to forestall the CNs embolic risk. 2) Vegetations larger than 1 cm PLUS aortic valve PLUS S. aureus equal increased risk of periannular extension of IE.
NOTE: The biggest controversy viz-a-vizthe ECHO in NV-IE involves S. aureusbacteremia (SAB):
IDSA guidelines recommend that all patients with catheter-associated SAB get a TEE (controversial).
Frequency of “occult S. aureus NV-IE is all-comers” with SAB from such foci equals about 15% (defined by finding vegetations on ECHO)
Recent studies suggest that such risks are very small in “nosocomial” SAB from catheters (especially).
Current paradigm for routine TEE in SAB in the absence of other clues to NV-IE:
Community-acquired or health care contact/non-nosocomial
Persistent bacteremia after catheter removal
Indwelling catheters being used for hemodialysis
Intracardiac devices (pacemaker; ICD)
NOTE: As you “overlay” patients with underlying cardiac valve disease onto SAB in the absence of clues to NV-IE, you ratchet up the frequency of finding occult vegetative IE to about40%.
NOTE: Diagnostically, we find it very useful to perform “routine” abdominal CT scans on suspected or bona fide cases of NV IE even in the absence of abdominal symptoms/signs:
High frequency of typical (small) embolic type lesions, especially in the spleen or kidneys
Occasionally will find an occult abscess in one of these target organs
Finding splenomegaly in “iffy” cases of NV-IE can help the diagnostic dilemma
What consult service or services would be helpful for making the diagnosis and assisting with treatment?
If you decide the patient has disease native valve endocarditis, what therapies should you initiate immediately?
NOTE: We consider IE a classic “multidisciplinary” syndrome. Cardiology and Infectious Diseases should always be involved in care of NV-IE. In severe cases (e.g., complicated by CHF), the cardiac surgeons should be involved early on. The Microbiology lab (as noted) can be invaluable for diagnosis and susceptibility testing. Other selected services are frequently called in to help manage complications of NV-IE when they arise, especially CNS and renal complications.
Key principles of therapy
See Table I, Table II, Table III, Table IV.
1. Anti-infective agents
Antimicrobial therapy for the most common organisms causing NV-IE is outlined in , , , ,
It is strongly advised that where ever possible treatment, whether administered in-hospital or in the ambulatory setting conform to the recommended therapy. There is extensive experience with most of the recommended regimens and they are thought to provide optimal therapy for what could be a life threatening infection.
There have been several recent significant changes in recommended regimens:
S. aureus :For NV-IE caused oxacillin susceptible or resistant S aureus the optional use of concurrent gentamicin for the initial 3-5 days of therapy is no longer recommended. Data suggest that not only is the clinical benefit of this therapy unclear, but also even this brief exposure to gentamicin is associated with nephrotoxicity. For NV-IS caused by Oxacillin susceptible S. aureus that is confined to the right heart valves (no systemic emboli, peripheral signs of IE, nor left heart valve vegetations on TEE) a two week course of nafcillin/oxacillin plus gentamicin is generally effective in those patients who respond with prompt resolution of fever and negative blood cultures. Right sided NV- IE caused by oxacillin resistant S. aureus is not reliably treated with a 2 week course of vancomycin even when combined with gentamicin. Right sided S. aureus NV- IE is largely seen in parenteral drug abusers.
Enterococcus faecalis: Several studies have demonstrated that for NV-IE caused by enterococci (primarily E. faecalis) a full 6 weeks of the gentamicin component of combination therapy may not be required for effective treatment. Thus some experts have abbreviated the gentamicin component of therapy to 2-3 weeks while continuing the cell wall active agent (penicillin, ampicillin or vancomycin) for the total 6 week course. Limiting gentamicin exposure may avoid nephrotoxicity or lessen nephrotoxicity if such begins to emerge early in treatment.
Enterococcus faecalis: A new regimen has been recommended for treatment of NV-IE caused by E. faecalis. This so-called “double beta lactam regimen” combines high dose ceftriaxone and ampicillin to achieve synergistic killing of enterococci. Studies indicate that the regimen has efficacy comparable to the gentamicin containing combination regimens. Because of reduced risk of nephrotoxicity this regimen is preferred treatment of patients with a baseline creatinine clearance of <50 mL/min. It is effective against infection caused by enterococci with high level resistance to gentamicin (organisms wherein bactericidal synergism does not result by combining gentamicin with cell wall active agent).
The HACEK organism [Hemophilus spp, Aggregatibacter aphrophilus, Aggregatibacter actinomycetemcomitans, Cardiobacterium hominis, Eikenella spp, Kingella spp.] can be resistant to ampicillin because of betalactamase production or intrinsic resistance. Hence ceftriaxone is recommended for therapy for HACEL NV-IE unless susceptibility to ampicillin can be conclusively demonstrated.
For treatment of typical organisms with unusual susceptibility profiles or unusual organisms, please refer to the Guidelines Baddour et al. and Habib et al. see below).
Empiric antimicrobial therapy for native valve endocarditis (Non-IVDA)
For patients with indolent disease the focus should be on oral streptococci and enterococci (non-penicillin allergic)
Ampicillin + ceftriaxone (with or without gentamicin at synergy doses)
Penicillin-allergic patient: vancomycin + gentamicin (at synergy doses)
For patients with more toxic or acute presentations the focus should be be on S aureus as well as streptococci and enterococci.
Vancomycin plus ceftriaxone and oxacillin/nafcillin
Whether coverage for MRSA or MSSA should be added to empiric therapy for all patients with suspected NV-IE has not been clearly defined. However, given that S. aureus is now the number one pathogen in NV-IE in the industrialized world, many experts recommend adding MRSA/MSSA empiric coverage. If followed, such a regimen should contain both vancomycin and a B-lactam, such as nafcillin or oxacillin. Vancomycin is clearly inferior to nafcillin or oxacillin for invasive MSSA infections, especially bacteremia and endocarditis.
Patients who have had recent health care contact (e.g. recent hospitalizations; hemodialysis; PICC lines, etc), patients who have recently been in jail, are military recruits or provide a history of a recent skin “infection” (pimple, boil, cellulitis, etc) would all be considered higher risk for MRSA-MSAA empiric coverage.
See the reference list for updated AHA recommendation on prevention of endocarditis not covered in this chapter.
2. Other key therapeutic modalities.
Valve replacement therapy – major indications in NV-IE
Persistent bacteremia or fungemia
Periannular extension of infection
Multiple systemic emboli, especially to CNS
NOTE: Large vegetations (>1cm) without other indications are controversial. Increasing vegetation sizes on serial TTE or TEE is controversial.
NOTE: Most surgical studies suggest that “early” valve replacement surgery in NV-IE is “protective” against in-hospital and subsequent mortality. However, how early is early?
NOTE: Mitral valve vegectomy and repair has become a popular alternative to valve replacement if the valve competence is not compromised pre-operatively.
What complications could arise as a consequence of native valve endocarditis?
The major complications of NV-IE are either valvular, embolic, periannular or immunologic.
Notes on valvular complications include:
The most common complication of NV-IE is that it presents as various degrees of CHF.
It can be acute or sub-acute in presentation.
ECHO is a reasonable metric for defining the degrees of valvular regurgitation and CHF in NV-IE patients.
Onset of CHF should trigger detailed Cardiology and Cardiac Surgery consultations.
Notes on embolic complications include:
The overall symptomatic embolic risk is about 20-25% in most studies.
Emboli to the CNS are the most frequent and most prognostically important symptomatic lesions.
The range of syndromes of CNS emboli is broad in NV-IE, from multiple small lesions causing headache without focal findings to single lesions causing a stroke syndrome to hemorrhagic bleed syndrome to mycotic aneurysm.
Embolic risk falls dramatically with initiation of therapy (over the first week), although never reaching zero during therapy.
Contrast CT or contrast MRI is the best method to define the presence, size, and nature of the CNS embolic lesions. Some groups have recommended “routine” MRI/MRA to detect occult microbleeds, especially pre-cardiac surgery, even in patients without CNS symptoms.
Timing of cardiac valvular surgery after a CNS “embolic event” is controversial and depends on the size and nature of the CNS lesion. These are the overall notions about this gleaned from the literature and discussions with other experts:
Small (<2 cm) non-hemorrhagic need not delay cardiac surgery.
Hemorrhagic complications within CNS emboli irrespective of size should prompt a delay in cardiac surgery from 2 to 4 weeks if possible.
Large embolic lesions (>2 cm) seem to be at increased risk of hemorrhagic transformations during cardio-pulmonary bypass. Cardiac surgery should be delayed for 2-4 weeks if possible in this circumstance.
Management of small, peripheral “microbleeds” on MRI/MRA remains to be determined.
Notes on periannular complications:
Should be suspected in NV-IE patients with prolonged fever, persistent bacteremia, or unexplained heart block.
Early TEE is the method of choice for such complications.
Notes on immunologic complications:
Renal involvement is the most common immunologic syndromes (wide spectrum of syndromes equals hematuria/pyuria/proteinuria to glomerulonephritis to renal failure).
“Tip-off” is concomitant hypocomplementemia or positive serum rheumatoid factor.
Early Renal Service consultation is recommended to consider renal biopsy to guide therapy.
Syndromes often resolve with appropriate antibiotics.
Occasionally require steroids.
WHAT'S THE EVIDENCE for specific management and treatment recommendations?
Baddour, LM, Wilson, WR, Bayer, AS, Fowler, VG. “Infective Endocarditis in Adults: Diagnosis, antimicrobial therapy, and management of complications: A statement for healthcare professionals from the the American Heart Association: Endorsed by the Infectious Diseases Society of America. Circulation”. vol. 132. 2015. (This guideline statement, issued jointly from the American Heart Association and Infectious DiseasesSociety of America, provides a detailed approach to the treatment of endocarditis, with discussions of antibiotic therapy and surgical intervention.)
Habib, G, Lancellotti, P, Antunes, MJ. “2015 ESC Guidelines for the Management of Infective Endocarditis.”. European Heart Journal. 2015. (This is a detailed guideline statement on the treatment of endocarditis from the European Society for Cardiology. Some recommendations differ from those of the American Heart Association (Baddour et al))
Chu, VH, Bayer, AS. “Use of echocardiography in the diagnosis and management of infective endocarditis.”. Curr Infect Dis Reports. vol. 9. 2007. pp. 283-90. (Meta-analysis of the use of ECHO in IE from a number of recent publications.)
Fowler, VG, Miro, JM, Hoen, B. ” endocarditis throughout the world: a consequence of medical progress. From the International Collaboration on Endocarditis – Prospective Cohort Study.”. JAMA. vol. 293. 2005. pp. 3012-21. (First major publication from ICE-PCS showing the shift in microbiology of IE toward S. aureus in the industrialized countries.)
Fowler, VG, O, lsen MK, Corey, GR. “Clinical identifiers of complicated bacteremia.”. Arch Intern Med. vol. 163. 2003. pp. 2066-72. (One of many publications from the Fowler group at Duke stratifying patients presenting with S. aureus bacteremia into relative risks of hematogenous complications based on routine bedside clinical findings and results of serial blood cultures.)
Fowler, V, Scheld, WM, Bayer, AS, Mandell, GL, Bennett, JE, Dolin, R. “Principles and practice of infectious diseases. Philadelphia, PA: Churchill-Livingstone”. 2010. pp. 1067-122.. (Exhaustive textbook review of the entire field of IE, including NV-IE.)
Kaasch, AJ, Fowler, VG, Rieg, S. “Use of simple criteria set for guiding echocardiography in nosocomial bacteremia.”. Clin Infect Dis. vol. 53. 2011. pp. 1-9. (Important study on the limited utility of ECHO to "screen" patients with nosocomial S. aureus bacteremia for IE.)
Lalani, T, Cabell, CH, Benjamin, BK. “Analysis of the impact of early valve surgery on in-hospital mortality of native valve endocarditis: use of propensity scores and instrumental variable methods to correct for treatment-selection bias.”. Circulation. vol. 121. 2010. pp. 1005-13. (One of a number of analyses of the impact of "early" valve surgery in IE, with the preponderance of recent data leaning towards this approach to improve salutary outcomes in left-sided IE.)
Murdoch, D, Corey, GR, Hoen, B. “Regional variation in the presentation and outcome of patients with infective endocarditis. International Collaboration on Endocarditis – Prospective Cohort Study (ICE-PCS).”. Arch Intern Med. vol. 169. 2009. pp. 463-73. (A systematic analysis of the "face" of IE as viewed from a recent multinational prospective collaboration of IE.)
Rasmussen, RV, Host, U, Arpi, M. “Prevalence of infective endocarditis in patients with bacteremia: the value of screening with echocardiography.”. Eur J Echocardiogr. vol. 12. 2011. pp. 414-20. (Increased utility of screening ECHO in patients with S. aureus bacteremia as the prevalence of underlying valvular heart disease increases in that population.)
Wilson, W, Taubert, KA, Gewitz, M. “Prevention of infective endocarditis: guidelines from the American Heart Association.”. Circulation. vol. 116. 2007. pp. 1736-54. (Although this topic is not covered in detail in this chapter, this article systematically discusses the most recent iteration of these recommendations. This is a very controversial arena that is changing rapidly.)
Fernandez-Hidalgo, N, Almirante, B, Gavalda, J. “Ampicillinplus ceftriaxone Is as Effective as Ampicillin plus Gentamicin for Treating Enetrococcus faecalis Infective Endocarditis.”. Clin Infect Dis. vol. 56. 2013. pp. 1261-8. (This paper compares patients treated with the ampicillin-ceftriaxone regimen for E faecalis IE ( including PVE) to a cohort receiving standard therapy and demonstrated that outcomes are comparable with reduced renal dysfunction among the non-aminoglycoside exposed patients.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- OVERVIEW: What every practitioner needs to know about native valve endocarditis
- Are you sure your patient has native valve endocarditis?
- How did the patient develop native valve endocarditis? What was the primary source from which the infection spread?
- Which individuals are of greater risk of developing native valve endocarditis?
- Beware: there are other diseases that can mimic native valve endocarditis:
- What laboratory studies should you order and what should you expect to find?
- What imaging studies will be helpful in making or excluding the diagnosis of native valve endocarditis?
- What consult service or services would be helpful for making the diagnosis and assisting with treatment?
- What complications could arise as a consequence of native valve endocarditis?