Guide détaillé à venir
Nous préparons un guide éducatif complet pour le Duke Criteria for Infective Endocarditis. Revenez bientôt pour des explications étape par étape, des formules, des exemples concrets et des conseils d'experts.
The Duke Criteria are a standardised diagnostic framework for infective endocarditis (IE), a potentially life-threatening infection of the endocardium — the inner lining of the heart — most commonly affecting the cardiac valves. First described by Durack et al. at Duke University in 1994 and subsequently refined by Li et al. in 2000 (the Modified Duke Criteria), the system classifies patients as having Definite IE, Possible IE, or Rejected IE based on a combination of major and minor clinical, microbiological, and echocardiographic findings. The criteria were developed because IE is notoriously difficult to diagnose: its clinical presentation is protean, blood cultures are negative in up to 31% of cases (culture-negative IE), and echocardiographic findings require expert interpretation. Before the Duke Criteria, diagnostic practice was highly variable and relied heavily on clinical gestalt. The Modified Duke Criteria achieved a sensitivity of approximately 80% and specificity of 99% for pathologically confirmed IE in validation studies. Major criteria emphasise microbiological evidence (specific organisms from qualifying blood cultures) and echocardiographic evidence of endocardial involvement. Minor criteria encompass predisposing conditions, fever, vascular and immunological phenomena, and microbiological findings that fall short of major status. A score is computed and the case classified as Definite (2 major, 1 major + 3 minor, or 5 minor), Possible (1 major + 1 minor, or 3 minor), or Rejected (firm alternate diagnosis, resolution with antibiotics ≤4 days, no pathological evidence at surgery/autopsy). The 2023 ESC guidelines incorporated FDG-PET/CT and cardiac CT as additional major imaging criteria.
Definite IE: 2 Major criteria OR 1 Major + 3 Minor OR 5 Minor | Possible IE: 1 Major + 1 Minor OR 3 Minor | Rejected: Alternate diagnosis, resolution ≤4d antibiotics, or no pathological evidence
- 1Obtain at least 3 sets of blood cultures (aerobic and anaerobic bottles) from separate venepuncture sites before starting antibiotics; timing intervals and organism identity determine whether microbiological criteria are major or minor.
- 2Classify microbiological findings: typical IE organisms (Streptococcus viridans group, S. gallolyticus, HACEK group, S. aureus, enterococci in absence of primary focus) from 2 separate cultures = 1 major criterion; persistently positive cultures (>12 h apart, or 3 of 4 cultures) = 1 major criterion; single positive for Coxiella burnetii or anti-phase 1 IgG titre >1:800 = 1 major criterion.
- 3Perform echocardiography (TTE first, TOE if TTE non-diagnostic or high suspicion): look for vegetation (oscillating intracardiac mass on valve or supporting structure), abscess or pseudoaneurysm, new partial dehiscence of prosthetic valve, new valvular regurgitation — any of these constitutes 1 major criterion.
- 4Consider new imaging major criteria (Modified 2023 ESC): FDG-PET/CT with abnormal activity around prosthetic valve (implanted >3 months) or cardiac CT showing paravalvular lesions.
- 5Assess minor criteria: predisposing heart condition (MVP, congenital, previous IE, prosthetic valve) or intravenous drug use; fever >38°C; vascular phenomena (emboli, septic pulmonary infarcts, mycotic aneurysm, intracranial haemorrhage, Janeway lesions); immunological phenomena (glomerulonephritis, Osler nodes, Roth spots, rheumatoid factor positive); microbiological evidence not meeting major criteria (single positive culture for non-typical organism).
- 6Sum the criteria and apply the classification rule: Definite = 2 major OR 1 major + 3 minor OR 5 minor; Possible = 1 major + 1 minor OR 3 minor; Rejected = none of the above, or firm alternative, or resolution ≤4 days on antibiotics.
- 7Always apply clinical judgement alongside the score — a case with a convincing clinical picture and one positive blood culture for S. aureus bacteraemia with TOE vegetation should be treated as IE even before formal classification threshold is met.
S. aureus bacteraemia with a new valvular vegetation represents one of the most classic and virulent presentations of IE.
Both major criteria (microbiological + echocardiographic evidence of endocardial involvement) are met independently. Classification is Definite regardless of minor criteria.
Even with a borderline microbiological finding, the combination of echo evidence and 3 minor criteria secures a Definite classification.
The single culture of S. viridans is classified as minor (microbiological evidence not meeting major criteria). The TOE vegetation is the sole major criterion. Three minor criteria complete the Definite threshold.
Prosthetic valve IE has notoriously low TTE sensitivity — TOE or FDG-PET/CT should be urgently arranged.
Coagulase-negative staphylococci (CoNS) are common contaminants but also genuine prosthetic valve pathogens. A single positive culture is minor. The clinical picture demands TOE and further workup.
The Duke Criteria explicitly reject IE if symptoms resolve within 4 days of antibiotic therapy, suggesting a non-endocardial infection.
Fever and a single culture not from a typical IE organism are minor criteria only. With an obvious alternative diagnosis (pneumonia) and rapid resolution, IE is rejected.
Emergency medicine and infectious disease: standardised classification of suspected IE to guide antibiotic initiation and investigation intensity, enabling practitioners to make well-informed quantitative decisions based on validated computational methods and industry-standard approaches
Cardiology ward: identifying Definite vs. Possible IE to determine urgency of cardiac surgery referral and cardiac surgical timing, helping analysts produce accurate results that support strategic planning, resource allocation, and performance benchmarking across organizations
Microbiology and blood culture stewardship: interpreting significance of specific organisms in blood cultures in the context of the Duke score, allowing professionals to quantify outcomes systematically and compare scenarios using reliable mathematical frameworks and established formulas
Research and audit: consistent endpoint definition in IE epidemiology studies and quality improvement projects, supporting data-driven evaluation processes where numerical precision is essential for compliance, reporting, and optimization objectives, necessitating robust computational methods that deliver consistent and verifiable results suitable for reporting, auditing, and long-term trend analysis in professional environments
Antibiotic prophylaxis decisions: determining which patients are at highest risk and warrant pre-procedural prophylaxis, which requires precise quantitative analysis to support evidence-based decisions, strategic resource allocation, and performance optimization across diverse organizational contexts and professional disciplines
Prosthetic Valve Endocarditis (PVE)
PVE accounts for 10–30% of IE cases and carries higher mortality (20–40%) than native valve IE. TTE sensitivity for vegetations on prosthetic valves is very poor due to acoustic shadowing; TOE is mandatory. FDG-PET/CT is now a major criterion for prosthetic valves implanted >3 months. Early PVE (within 1 year of surgery) is typically caused by staphylococci acquired perioperatively; late PVE has a similar microbiology to native valve IE.
Cardiac Device IE (CDRIE)
IE involving pacemaker leads, ICD leads, or cardiac resynchronisation devices (CIED) is increasingly common as device implantation rates rise. Vegetations on device leads are often best visualised by TOE; FDG-PET/CT is valuable. Treatment typically requires complete device system extraction in addition to antibiotics. The Duke Criteria were not originally designed for CDRIE but have been adapted in current guidelines.
Right-Sided IE (IVDU-associated)
Intravenous drug users predominantly develop right-sided IE, particularly tricuspid valve IE with S. aureus. TTE sensitivity is better for right-sided vegetations (which are typically large). Clinical presentation includes septic pulmonary emboli causing multiple round consolidations on chest X-ray/CT. Right-sided IE has a better prognosis than left-sided disease; surgical indications are less common except for persistent bacteraemia or large vegetations.
Q Fever Endocarditis (Coxiella burnetii)
Coxiella burnetii is an obligate intracellular bacterium that cannot be grown on standard cultures. It is the most common cause of culture-negative IE in endemic regions. A single positive blood culture or high anti-phase-1-IgG titre (>1:800) satisfies a major criterion in the Modified Duke system. Treatment requires prolonged combination therapy with doxycycline + hydroxychloroquine for a minimum of 18 months due to the organism's intracellular niche.
Non-Bacterial Thrombotic Endocarditis (NBTE/Marantic)
NBTE consists of sterile fibrin-platelet vegetations on heart valves, most commonly associated with systemic lupus erythematosus (Libman-Sacks), antiphospholipid syndrome, malignancy, and uraemia. Blood cultures are negative and the vegetations may look similar to infective vegetations on echo. The Duke Criteria will classify NBTE as Possible or Rejected once blood cultures remain negative, steering clinicians away from unnecessary antibiotics.
| Category | Criterion | Classification |
|---|---|---|
| Major — Microbiological | Typical organism (Streptococcus viridans/gallolyticus, HACEK, S. aureus, Enterococcus) from 2 separate blood cultures | 1 Major |
| Major — Microbiological | Persistently positive blood cultures (>12 h apart OR 3 of 4 cultures) | 1 Major |
| Major — Microbiological | Single positive Coxiella burnetii or IgG phase I titre >1:800 | 1 Major |
| Major — Echo | Vegetation, abscess, pseudoaneurysm, intracardiac fistula, valvular perforation, new prosthetic valve dehiscence | 1 Major |
| Major — Echo | New valvular regurgitation (worsening or change of pre-existing murmur not sufficient) | 1 Major |
| Major — Imaging (2023) | FDG-PET/CT: abnormal activity around prosthetic valve (>3 months post-implant) | 1 Major |
| Major — Imaging (2023) | Cardiac CT: definite paravalvular lesion | 1 Major |
| Minor | Predisposing heart condition or intravenous drug use | 1 Minor |
| Minor | Fever >38°C | 1 Minor |
| Minor | Vascular phenomena: major emboli, septic pulmonary infarcts, mycotic aneurysm, intracranial haemorrhage, conjunctival haemorrhage, Janeway lesions | 1 Minor |
| Minor | Immunological phenomena: glomerulonephritis, Osler nodes, Roth spots, positive rheumatoid factor | 1 Minor |
| Minor | Microbiological evidence not meeting major criteria (single positive culture for non-typical organism) | 1 Minor |
What is infective endocarditis and why is it hard to diagnose?
Infective endocarditis is a microbial infection of the endocardium, most commonly the cardiac valves. It is difficult to diagnose because its presentation is highly variable — ranging from subacute malaise and low-grade fever over weeks (subacute IE, often streptococcal) to fulminant sepsis with acute valve destruction (acute IE, typically staphylococcal). Blood cultures may be negative (up to 31%) due to prior antibiotics or fastidious organisms. Echocardiographic findings require expert interpretation and may be absent in early disease.
What are the typical IE organisms that count as a major criterion?
The following organism-isolation combinations qualify as a major microbiological criterion: Streptococcus viridans group, Streptococcus gallolyticus (bovis), HACEK organisms (Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, Kingella), Staphylococcus aureus, or Enterococcus species (in the absence of a primary focus) — all from two separate blood culture sets. Single positive cultures for Coxiella burnetii or anti-phase-1-IgG titre >1:800 also qualify.
Why is transesophageal echocardiography (TOE) preferred over TTE?
Transoesophageal echocardiography (TOE/TEE) has approximately 90–95% sensitivity for vegetations compared to ~60–70% for transthoracic echocardiography (TTE). TOE is particularly superior for prosthetic valves (where TTE is severely limited by acoustic shadowing), for detecting paravalvular abscesses and fistulae, and for small vegetations (<5 mm). ESC and AHA guidelines recommend TOE for all suspected IE in patients with prosthetic valves or intracardiac devices, and when TTE is non-diagnostic.
What are Janeway lesions and Osler nodes?
Both are dermatological manifestations of IE but with different pathophysiology. Janeway lesions are painless, haemorrhagic macular lesions on the palms and soles caused by septic emboli — they are therefore a vascular phenomenon (major-adjacent, counted under vascular minor criteria). Osler nodes are painful, raised nodules on the fingers and toes caused by immune complex deposition — they are an immunological phenomenon (minor criterion). A simple mnemonic: Osler nodes are painfull and immunological; Janeway lesions are painless and embolic.
What is culture-negative endocarditis and how is it managed?
Culture-negative IE occurs in 2–31% of cases, most commonly due to prior antibiotic therapy, fastidious organisms (Bartonella, Brucella, Legionella, Tropheryma whipplei, fungi), or non-infectious mimics. Workup includes serology for fastidious organisms, 16S rRNA PCR on blood or excised valve tissue, and broad-range culture with prolonged incubation. Empirical antibiotic coverage must be broadened. The 2023 ESC guidelines recommend FDG-PET/CT in suspected culture-negative IE to detect metabolically active foci.
When is surgery indicated in infective endocarditis?
Urgent/emergency surgery (within 24 h to 1 week) is indicated for: heart failure due to valve dysfunction (most common indication, especially aortic IE), uncontrolled infection (perivalvular extension — abscess, fistula, or false aneurysm, or persistent bacteraemia despite appropriate antibiotics >7 days), and prevention of embolism (large vegetation >10 mm with embolic event or very high embolic risk). Approximately 40–50% of IE patients ultimately require surgery, with in-hospital mortality around 15–25% even with optimal management.
How do the 2023 ESC modifications differ from the original 2000 Li criteria?
The 2023 ESC IE guidelines added two new major imaging criteria to the Modified Duke system: (1) abnormal metabolic activity around a prosthetic valve or intracardiac device on FDG-PET/CT (implanted >3 months), and (2) definite paravalvular lesions on cardiac CT. These additions were driven by evidence that FDG-PET/CT significantly improves sensitivity for prosthetic valve IE, which is notoriously difficult to diagnose by echo alone. The 2023 ESC guidelines also expanded the definition of positive echo to include CT-detected lesions.
What antibiotic prophylaxis is recommended to prevent IE?
Current ESC (2023) and AHA (2021) guidelines restrict IE prophylaxis to the highest-risk patients undergoing dental procedures that involve manipulation of gingival tissue or the periapical region of teeth. High-risk categories include patients with prosthetic heart valves (including transcatheter valves), previous IE, unrepaired cyanotic congenital heart disease, and repaired CHD with residual defects. Prophylaxis is amoxicillin 2 g (or clindamycin 600 mg if penicillin-allergic) orally 30–60 min before the procedure. Prophylaxis is NOT recommended for patients with native valve disease, MVP, or other lower-risk conditions.
Conseil Pro
In any patient with unexplained bacteraemia (especially S. aureus), always perform echocardiography even before Duke classification — it changes management regardless of the score. A resting-state TOE negative for vegetation in the context of S. aureus bacteraemia still warrants at least 2 weeks of intravenous antibiotics, and a repeat TOE at 5–7 days if clinical suspicion remains. Endocarditis teams (cardiologist, cardiac surgeon, infectious disease specialist, microbiologist) should be involved from day 1.
Le saviez-vous?
The original 1994 Duke Criteria paper by Durack et al. was rejected by multiple journals before publication in the American Journal of Medicine. The reviewers felt that the criteria were too complicated and unlikely to be adopted. Within five years, the criteria had been externally validated in over 20 studies across multiple countries and had become the universal standard — a classic example of delayed scientific recognition.
Références
- ›Durack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis. Am J Med 1994
- ›Li JS et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis 2000
- ›Delgado V et al. 2023 ESC Guidelines for the management of endocarditis. Eur Heart J 2023
- ›Baddour LM et al. Infective Endocarditis in Adults: Diagnosis, Antimicrobial Therapy, and Management of Complications. AHA Scientific Statement. Circulation 2015
- ›Cahill TJ, Prendergast BD. Infective endocarditis. Lancet 2016