Chapter 24 — Infective Endocarditis

📚 Guideline basis

2023 ESC Guidelines for the management of endocarditis (Delgado V, et al., Eur Heart J 2023); 2023 Duke-ISCVID criteria (Fowler VG, et al., Clin Infect Dis 2023); ASE/EACVI recommendations on endocarditis imaging; Zoghbi WA, et al. ASE prosthetic valve guideline (JASE 2024;37:2–63); 2025 ESC/EACTS valvular heart disease guidelines.

Pathophysiology & Mechanisms

Endothelial injury exposes subendothelial collagen and tissue factor, producing a sterile platelet-fibrin nidus (non-bacterial thrombotic endocarditis). Circulating organisms adhere to this nidus; those with adhesins for fibronectin and fibrinogen — Staphylococcus aureus, viridans streptococci, enterococci — colonise preferentially. Bacteria within the growing vegetation are shielded from phagocytes and antibiotic penetration by the fibrin matrix and enter a low-metabolic state, which is why bactericidal therapy is prolonged and why source control frequently requires surgery.

Three destructive processes generate the findings that determine management:

1. Leaflet destruction — perforation, aneurysm formation, chordal rupture, producing acute regurgitation into an unremodelled chamber (Chapters 19, 20).
2. Perivalvular extension — abscess, pseudoaneurysm, fistula. This occurs preferentially at the aortic annulus and in prosthetic valves, and produces conduction disturbance as the infection tracks toward the AV node.
3. Embolisation — fragments detach and embolise systemically (left-sided) or to the lungs (right-sided).

Prosthetic valve infection differs mechanically: in mechanical valves the infection almost always begins at the sewing ring, producing paravalvular leak, abscess and dehiscence rather than leaflet destruction; in bioprostheses it begins in the leaflet cusps and may extend to the ring.

Epidemiology relevant to the ICU

Right-sided endocarditis in the ICU is predominantly catheter-, device- and injection-drug-related, presents with septic pulmonary emboli rather than systemic embolism, and is frequently missed because the murmur is soft and the chest imaging is read as pneumonia.

Diagnostic Synthesis

Echocardiography within the Duke-ISCVID criteria

Echocardiography supplies the imaging major criterion. The 2023 Duke-ISCVID revision broadened imaging to include cardiac CT (for perivalvular or periprosthetic anatomical lesions) and ¹⁸F-FDG PET/CT (for prosthetic valves and intracardiac devices), reflecting the poor sensitivity of echocardiography for perivalvular complications in prosthetic material.

Echocardiographic findings constituting the major criterion: vegetation, valvular perforation or aneurysm, abscess, pseudoaneurysm, intracardiac fistula, new partial dehiscence of a prosthetic valve, or new valvular regurgitation.

A vegetation is defined as an oscillating or non-oscillating intracardiac mass on a valve or other endocardial structure, or on implanted material, in the path of a regurgitant jet or on the upstream side of the valve, with motion independent of the underlying structure and no alternative anatomical explanation.

TTE versus TEE

TEE sensitivity for native valve vegetations exceeds 90% in most series, but it is not perfect and it does not obviate repeat imaging.

The negative study is not an exclusion. In a patient with a high clinical probability and an initially negative TTE and TEE, guidelines recommend repeating imaging at 5–7 days, because vegetations grow into detectability. Documenting "TEE negative — repeat planned if clinically indicated" is materially different from "endocarditis excluded".

Findings that change surgical urgency

Anterior mitral leaflet vegetations carry a higher embolic risk than other locations.

Perivalvular complications on imaging

• Abscess — a thickened, non-homogeneous perivalvular area with an echodense or echolucent appearance and no colour flow within it. Aortic root thickening after implantation of a stentless valve is a haematoma/oedema that resolves over 3–6 months and is a classic mimic; comparison with the intra- or post-operative study resolves it.
• Pseudoaneurysm — a pulsatile, echo-free perivalvular cavity with colour flow communication, expanding in systole.
• Fistula — a communication between two adjacent cavities demonstrated by colour Doppler.
• Dehiscence — a paravalvular jet with rocking motion of the prosthesis (excursion > 15° indicates significant dehiscence).

CT angiography outperforms TEE for the anatomical extent of root abscess and pseudoaneurysm; TEE outperforms CT for small vegetations and leaflet perforations. ¹⁸F-FDG PET/CT shows intense perivalvular uptake in prosthetic valve endocarditis but must be interpreted with caution, as low-to-intermediate paravalvular uptake is a normal post-operative finding for up to a year.

Mimics — what a mass is not

Every suspected vegetation must be confirmed in at least two orthogonal imaging planes and must persist across gain settings.

ICU-Specific Limitations

Therapeutic Logic

Echocardiography contributes to four decisions:

1. Diagnosis — the imaging major criterion.
2. Surgical indication and timing — heart failure from acute regurgitation, uncontrolled infection with perivalvular extension, and embolic prevention are the three indication groups.
3. Embolic risk stratification — vegetation size, mobility and location.
4. Baseline for the post-operative prosthesis — a post-operative study establishing normal prosthetic haemodynamics is essential for all subsequent assessments (Chapter 23).

Antibiotic duration and choice follow microbiology, not imaging. Vegetation size does not shrink predictably on treatment, and serial imaging to demonstrate "resolution" is not a therapeutic endpoint; repeat imaging is indicated for new murmurs, new conduction disturbance, embolic events, persistent fever, or heart failure.

🛑 Critical pitfall: Documenting "no vegetation seen, endocarditis excluded" after a transthoracic study. TTE sensitivity is roughly 50–70% for native valves and substantially lower for prosthetic valves and abscess. This wording will be relied upon by the next clinician.

🛑 Critical pitfall: Missing acute severe regurgitation because the patient is hypotensive and the jet is small. Endocarditis with acute severe regurgitation is the commonest indication for emergency surgery and is graded by mechanism, not jet size.

🛑 Critical pitfall: Attributing new PR-interval prolongation or heart block in a patient with aortic valve endocarditis to a drug effect. It signals an aortic root abscess tracking toward the conduction system and is a surgical emergency.

• 💡 Clinical pearl: In an ICU patient with prosthetic material and positive blood cultures, go directly to TEE. Serial non-diagnostic transthoracic studies delay the diagnosis without adding information.
• 💡 Clinical pearl: New conduction disturbance is the most useful non-imaging clue to perivalvular extension. Check the ECG on every study day in aortic valve endocarditis.
• 💡 Clinical pearl: Aortic root thickening after stentless valve or homograft implantation resolves over 3–6 months and is routinely mistaken for abscess. Compare against the intraoperative study before escalating.

References

1. Delgado V, Ajmone Marsan N, de Waha S, et al. 2023 ESC guidelines for the management of endocarditis. Eur Heart J 2023;44:3948–4042.
2. Fowler VG, Durack DT, Selton-Suty C, et al. The 2023 Duke-International Society for Cardiovascular Infectious Diseases criteria for infective endocarditis. Clin Infect Dis 2023;77:518–26.
3. Zoghbi WA, Jone PN, Chamsi-Pasha MA, et al. Guidelines for the evaluation of prosthetic valve function with cardiovascular imaging. J Am Soc Echocardiogr 2024;37:2–63.
4. Habib G, Badano L, Tribouilloy C, et al. Recommendations for the practice of echocardiography in infective endocarditis. Eur J Echocardiogr 2010;11:202–19.
5. ESC/EACTS. 2025 Guidelines for the management of valvular heart disease. Eur Heart J 2025;46:4635–.