Table of Contents

Introduction

On September 8, 2026, Circulation published the new American Heart Association scientific statement on infective endocarditis (IE) — the first comprehensive AHA document on the subject since 2015 [1].

Ten years is a long time in endocarditis.

The patients have changed over that interval. Cardiovascular prostheses and transcatheter devices are now everywhere, molecular microbiology has moved out of the research laboratory and into routine diagnostics, imaging has expanded well beyond conventional echocardiography, and the old choreography — several weeks of intravenous antibiotics, surgical opinion afterwards — has been steadily undermined.

The new statement therefore lands in a different clinical world from its predecessor. It also lands three years after the 2023 European Society of Cardiology Guidelines for the management of endocarditis, which had already rewritten a good deal of diagnosis, antimicrobial treatment and surgical timing [2].

For a European reader the useful question is not:

What does the new AHA statement say?

That much can be read in the original. The more productive question is:

What does it add to the ESC 2023 framework, where do the two documents genuinely diverge, and what should change on Monday morning?

That is the angle taken here.

What changed since 2015?

Echo-centred diagnosis → multimodality imaging

Blood cultures alone → cultures plus molecular microbiology

Prolonged IV therapy for nearly everyone → selected partial oral treatment

Delayed surgical referral → early multidisciplinary Endocarditis Team

Surgery versus medical therapy → a broader spectrum of procedural options

The common denominator is a move away from a relatively linear treatment pathway towards a multidisciplinary, dynamic and individualized strategy.

1. Why do we need another major statement on infective endocarditis?

Endocarditis remains an uncommon disease. It is not becoming rarer, and it is certainly not becoming simpler.

The AHA documents a continuing rise in incidence since the 2015 statement, driven by several converging epidemiological trends: an ageing population carrying more comorbidity, wider use of prosthetic valves and cardiovascular implantable electronic devices, and the burden of injection drug use-associated endocarditis. Despite two decades of diagnostic and therapeutic refinement, one-year mortality still approaches 40% in contemporary series [1].

That figure deserves to be said out loud more often. Roughly four patients in ten are dead within a year of diagnosis.

Meanwhile the clinical phenotype has fragmented.

A young patient with isolated native tricuspid valve endocarditis; an octogenarian with suspected infection of a transcatheter aortic valve; a patient with prosthetic valve endocarditis and a periannular abscess; a patient with an infected pacemaker system. All four satisfy the same broad definition. In practice they have almost nothing in common — neither diagnostic pathway, nor therapeutic priority, nor operative strategy.

This fragmentation explains a theme shared by both the ESC and AHA documents:

infective endocarditis should no longer be managed as an isolated infectious disease of a heart valve.

It is better read as a systemic disease requiring simultaneous integration of microbiology, imaging, haemodynamics, neurological risk, surgical anatomy and source control.

The consequence is organisational as much as therapeutic. The Endocarditis Team becomes a structural component of contemporary care, and both ESC 2023 and AHA 2026 endorse multidisciplinary management, typically bringing together cardiology, cardiac imaging, infectious diseases, microbiology and cardiac surgery.

For the cardiac surgeon this changes the nature of the consultation.

The question is increasingly not:

“Has antibiotic therapy failed enough to justify surgery?”

but rather:

“Does this patient already have an anatomical, haemodynamic or infectious trajectory for which early intervention offers the best chance of definitive source control?”

Whether every centre can actually staff such a team is, of course, a different discussion.

2. What is genuinely new or particularly relevant in the AHA 2026 statement?

Several concepts in the new statement were already visible in the ESC 2023 Guidelines. A few represent a clearer evolution of practice.

Diagnosis moves beyond the classical Duke framework

The AHA incorporates the 2023 Duke–International Society for Cardiovascular Infectious Diseases (Duke-ISCVID) criteria into contemporary diagnostic practice [3].

The diagnostic toolbox widens considerably as a result.

Alongside conventional blood cultures and echocardiography, the updated criteria recognize:

molecular microbiological techniques;
polymerase chain reaction and sequencing approaches;
cardiac computed tomography;
^18F-FDG PET/CT;
and, in specific circumstances, direct intraoperative evidence of endocarditis.

The Duke-ISCVID criteria also broaden the spectrum of microorganisms considered typical in particular clinical settings, and — finally — formally recognise transcatheter valves and cardiac implantable electronic devices as relevant predisposing substrates.

This goes further than a refinement of diagnostic criteria. Or more precisely: it reflects a change in how the disease is conceptualized.

A patient can have clinically important infective endocarditis even when the classical sequence of:

positive blood cultures + obvious vegetation on echocardiography

never fully materialises.

Blood culture-negative endocarditis

Nowhere does this matter more than in blood culture-negative endocarditis, where prior antimicrobial therapy and fastidious organisms conspire to blind conventional microbiology.

The dedicated 2025 AHA statement on blood culture-negative endocarditis had already made the case for molecular diagnostics and advanced imaging in this setting [4]. The 2026 document folds those concepts into the broader management framework.

TAVR-associated endocarditis

Infection involving a transcatheter aortic valve is the other situation in which conventional diagnosis struggles.

Prosthetic material, the stent frame, residual calcification and acoustic shadowing all conspire against the echocardiographer, and peri-prosthetic abnormalities may simply not be visible.

The AHA is explicit: transthoracic and transoesophageal echocardiography both show reduced diagnostic sensitivity in TAVR-associated IE, and multimodality imaging — particularly ^18F-FDG PET/CT — may be necessary when clinical suspicion remains high.

The practical message is blunt.

A negative TEE does not exclude TAVR endocarditis.

Antibiotic therapy becomes more individualized

For decades, infective endocarditis was effectively synonymous with several weeks of intravenous antibiotics.

That equation no longer holds.

The pivotal POET trial demonstrated that, in carefully selected clinically stable patients with left-sided IE, switching from intravenous to appropriately chosen oral combination therapy was non-inferior to continued intravenous treatment [5]. Long-term follow-up supported the durability of the strategy [6].

The ESC absorbed this in 2023, describing treatment in two phases.

The first is an in-hospital phase: intravenous antimicrobial therapy, evaluation of surgical indications, removal of infected foreign material where necessary, drainage of abscesses.

In selected stable patients, therapy may then be completed through outpatient parenteral treatment or partial oral therapy — with the ESC framework generally requiring clinical stability and reassessment by transoesophageal echocardiography before the transition is made.

The 2026 AHA statement reinforces the same conceptual shift and goes slightly further, acknowledging a possible role for long-acting lipoglycopeptides such as dalbavancin or oritavancin in carefully selected patients.

None of which means that endocarditis has become an outpatient disease.

What it means is narrower:

route of administration should no longer be confused with intensity or quality of treatment.

Selected patients can safely complete therapy outside the traditional prolonged inpatient-IV model — provided source control has been achieved, microbiology is appropriate, the patient is clinically stable and follow-up is genuinely reliable. That last condition is the one most often assumed rather than verified.

A notable change: gentamicin in staphylococcal prosthetic valve endocarditis

Of everything in the statement, the most immediately actionable point concerns staphylococcal prosthetic valve endocarditis (PVE).

The AHA advises avoiding gentamicin in this setting: potential toxicity outweighs expected benefit [1].

This deserves attention, because aminoglycoside-containing combination regimens have long formed part of the therapeutic approach to staphylococcal PVE and remain embedded in some guideline-based treatment schemes — and, in many hospitals, in the antibiotic order sets nobody has revisited for years.

In the cardiac surgical population the issue is sharper still.

These patients tend to arrive with several risk factors for acute kidney injury already in place:

sepsis;
haemodynamic instability;
contrast exposure;
pre-existing renal dysfunction;
cardiopulmonary bypass;
and, frequently, prolonged antimicrobial therapy.

Adding nephrotoxicity on top of that, without convincing evidence of incremental clinical benefit, becomes difficult to defend.

One of the few areas, then, where AHA 2026 may reach local antimicrobial protocols almost immediately.

Valve cultures and molecular tests after surgery: do not simply restart the clock

A second clarification, less conspicuous but practically important, concerns patients operated on during active IE.

The AHA states that a positive culture or molecular test from resected valve tissue does not, by itself, determine postoperative antibiotic duration.

With molecular methods the point becomes critical.

PCR and sequencing detect microbial nucleic acids long after viable organisms have disappeared. The Duke-ISCVID authors themselves caution that tissue-based molecular tests may remain positive after successful treatment and require clinical interpretation [3].

In other words:

detection of microbial DNA does not necessarily indicate persistent viable infection.

Postoperative duration should instead integrate:

the causative microorganism;
duration and adequacy of effective preoperative therapy;
blood-culture clearance;
operative findings;
presence or absence of periannular extension;
microbiological and pathological findings;
adequacy of surgical source control;
and the subsequent clinical course.

One sentence in a long statement — and potentially a real change at the bedside, where a single positive PCR has been known to buy a patient another six weeks of intravenous therapy.

3. AHA 2026 versus ESC 2023: where do they really differ?

Before comparing them, one distinction has to be made. These are not documents of the same kind.

The ESC 2023 document is a formal Clinical Practice Guideline, built around recommendations, classes and levels of evidence.

The AHA 2026 publication is a Scientific Statement, intended to synthesize contemporary evidence and provide expert scientific guidance.

They are not competing rulebooks, and reading them as such generates disagreements that do not exist.

For European practice the ESC Guidelines remain the operational framework. The AHA statement is most useful as an updated layer of evidence and interpretation on top of it.

Clinical domain	ESC 2023	AHA 2026	Practical interpretation
Endocarditis Team	Central component of care, particularly for complicated IE	Strongly recommended and associated with improved outcomes	Near-complete convergence
Diagnostic criteria	ESC diagnostic algorithms integrate multimodality imaging	Explicit use of Duke-ISCVID 2023 criteria	AHA adopts the newer international framework
Molecular microbiology	Recognized, particularly in difficult diagnostic settings	Strong emphasis when cultures are negative or inconclusive	Growing clinical relevance
Cardiac CT / PET-CT	Integrated into diagnostic algorithms, particularly PVE	Strongly emphasized, especially for TAVR-IE	Echo alone no longer suffices in selected cases
Partial oral antibiotics	Accepted in clinically stable selected patients after initial IV treatment	Supported after the acute phase in carefully selected patients	Strong convergence
Long-acting antibiotics	Not a major component of the core strategy	Dalbavancin/oritavancin may be considered in selected patients	AHA expands therapeutic flexibility
Staphylococcal PVE	Traditional combination regimens remain represented	Explicit advice to avoid gentamicin, risks outweighing benefits	The one clinically relevant divergence
Surgical timing	Emergency, urgent or non-urgent surgery according to indication	Strong emphasis on early procedural intervention and multidisciplinary assessment	Same direction of travel
Positive valve microbiology after surgery	A new full course should be started if valve cultures are positive	Positive valve culture or molecular testing does not by itself alter postoperative antibiotic duration	A genuine difference between ESC 2023 and AHA 2026
Right-sided IE	Surgery/intervention in selected complicated cases; percutaneous approaches already emerging	PMA explicitly incorporated for selected high-risk patients failing medical therapy	Rapidly evolving, evidence still thin

Taken together: convergence rather than conflict.

Both documents describe the same pathway — early multidisciplinary assessment, multimodality diagnosis, individualized antimicrobial therapy, timely source control.

The interesting differences are in the details, as they usually are.

4. What matters most for the cardiac surgeon?

Six aspects deserve particular attention.

4.1 Surgical evaluation should occur early, not after antibiotic failure

The traditional sequence:

diagnosis → weeks of antibiotics → persistent problem → surgical consultation

is increasingly hard to defend in complicated left-sided endocarditis.

ESC and AHA both favour early multidisciplinary assessment.

The classical major indications for surgery, meanwhile, are essentially unchanged:

heart failure or severe haemodynamic consequences of valve dysfunction;
uncontrolled infection;
prevention of embolic events in appropriately selected patients.

ESC 2023 is unusually explicit about timing, classifying surgery as:

emergency: generally within 24 hours;
urgent: within approximately 3–5 days;
non-urgent: during the same hospitalization,

according to the clinical indication.

So the conceptual shift is not a new list of indications.

It is a change in when the surgical question gets asked.

A patient with severe aortic regurgitation, progressive periannular extension and a large vegetation is not an “antibiotic patient” who may later become a “surgical patient.”

He or she is a patient with infective endocarditis in whom antimicrobial therapy and surgical source control must be evaluated in parallel.

4.2 Neurological complications should not automatically postpone surgery

Neurological complications remain the hardest part of IE management, and the part where practice varies most between centres.

Historically, stroke meant delay — a relatively fixed interval before cardiac surgery, driven by concern about haemorrhagic transformation during cardiopulmonary bypass and systemic anticoagulation.

Contemporary ESC practice is more individualized.

Where a strong surgical indication persists, an uncomplicated ischaemic stroke does not necessarily justify postponement. The balance becomes:

risk of neurological deterioration from early surgery

versus

risk of continued embolization, heart failure or uncontrolled infection while waiting.

This is precisely the decision the Endocarditis Team exists to make.

Neurology, neuroradiology, cardiology and cardiac surgery need to read the cerebral imaging alongside the urgency of cardiac source control, rather than applying a fixed waiting period inherited from an earlier era. In practice this conversation tends to happen at the bedside, on a Friday afternoon, with incomplete information — which is exactly why the team should already have been convened.

4.3 Prosthetic valve endocarditis requires early anatomical thinking

PVE is not native valve endocarditis occurring on a different surface.

The possibility of:

periannular abscess;
prosthetic dehiscence;
pseudoaneurysm;
fistula;
conduction abnormalities;
and extensive destruction of the annulus or aortic root

makes anatomical progression the dominant variable.

Cardiac CT and PET/CT complement echocardiography here, particularly where prosthetic artefacts defeat conventional imaging.

For the surgeon, the presence of a vegetation is rarely the decisive information.

The question that shapes the operation is:

How far has the infection extended outside the valve itself?

That determines not only whether to operate, but what operation is actually being planned — and whether the root will need replacing.

4.4 A negative TEE is less reassuring after TAVR

TAVR-associated endocarditis is becoming more common simply because transcatheter implantation keeps expanding into larger and younger populations.

The AHA warns explicitly that both TTE and TEE lose diagnostic sensitivity in this setting, and supports multimodality imaging — PET/CT in particular — when suspicion persists.

The clinical consequence is direct:

high clinical suspicion + negative echocardiography should not terminate the diagnostic work-up.

4.5 Surgery does not automatically restart six weeks of antibiotics

Postoperative antimicrobial strategy has to be individualized.

A technically successful operation may achieve radical source control: infected prosthetic material removed, vegetations excised, infected periannular tissue debrided.

And positive molecular findings from that excised tissue may reflect nothing more than residual microbial genetic material.

The postoperative plan belongs to the surgical and infectious disease teams jointly. Not to a single laboratory result.

A good illustration of a wider point: contemporary endocarditis management increasingly demands integration rather than algorithms built on isolated variables.

4.6 Percutaneous mechanical aspiration enters the therapeutic landscape

The most intriguing procedural development concerns right-sided infective endocarditis.

Percutaneous mechanical aspiration (PMA) uses large-bore aspiration systems to reduce intracardiac vegetation burden.

The AHA published a dedicated Science Advisory on PMA in May 2026, and the new comprehensive statement incorporates the technique as a possible option in selected patients with right-sided disease, inadequate response to antimicrobial therapy and high surgical risk [7].

This is not a percutaneous replacement for tricuspid valve surgery, and should not be read as one.

The evidence remains largely observational. Randomized trials are absent.

A 2025 multicentre registry of 256 patients did show technical feasibility, with procedural success reported in approximately 89% of cases — while also documenting relevant complications and the need for better prospective data [8].

Conceptually, PMA inserts an intermediate step:

medical therapy ↔ PMA ↔ surgery, with PMA used selectively as an alternative or staged bridge

in place of the traditional binary:

medical therapy versus surgery.

For some high-risk patients PMA may function as definitive source-control support; in others, as a bridge to later valve intervention.

Its precise place remains undefined, and the registry data do not yet allow us to say which patients belong in which category.

5. From guidelines to bedside: a practical pathway

Memorizing individual recommendations is not how either document becomes useful. Building a reproducible pathway is.

Step 1 — Suspect infective endocarditis

Clinical presentation, bloodstream infection, embolic phenomena, new valve dysfunction, prosthetic material or intracardiac devices set the pre-test probability.

↓

Step 2 — Obtain microbiological evidence

Multiple blood cultures before antimicrobial therapy, whenever that is clinically feasible. It often is not, and the consequences follow the patient for weeks.

If cultures stay negative or inconclusive against convincing clinical suspicion:

think beyond conventional cultures.

Targeted serology, PCR, sequencing or tissue-based testing, according to the setting.

↓

Step 3 — Define the anatomy

TTE → TEE remains the basic sequence.

But where clinical suspicion and echocardiography disagree — particularly in:

prosthetic valve endocarditis;
TAVR-associated IE;
CIED infection;
suspected periannular complications,

repeat TTE/TEE when indicated, while moving early to cardiac CT and/or metabolic/nuclear imaging when prosthetic material, periannular complications or persistent diagnostic uncertainty make multimodality imaging necessary.

↓

Step 4 — Activate the Endocarditis Team

Whether the patient fulfils diagnostic criteria is not the essential question.

The team should assess, simultaneously:

microorganism + valve anatomy + haemodynamics + embolic risk + neurological status + extracardiac infection + operative risk.

↓

Step 5 — Is there an indication for intervention?

Heart failure

Severe acute valve dysfunction, obstruction, prosthetic dysfunction or haemodynamic deterioration.

Uncontrolled infection

Persistent sepsis or bacteraemia, abscess, fistula, pseudoaneurysm, progressive periannular extension or difficult-to-eradicate infection.

Embolic risk

Large or highly mobile vegetations, recurrent embolization, selected high-risk anatomical situations.

↓

Step 6 — Decide timing, not simply “surgery yes/no”

The real decision:

operate now?

operate within a few days?

operate during the same admission?

continue medical treatment with close reassessment?

Neurological events belong inside this decision — not as an automatic reason to defer it.

↓

Step 7 — Reassess antimicrobial strategy after source control

After the acute phase and, where indicated, surgery:

confirm microbiological response;
reassess anatomy;
ask whether prolonged IV treatment is still necessary;
identify patients suitable for oral or outpatient therapy;
interpret postoperative tissue microbiology in context rather than in isolation.

↓

Step 8 — Plan structured follow-up

Discharge does not close the endocarditis episode. It moves it somewhere less well observed.

Follow-up should address:

recurrent infection;
residual or recurrent valve dysfunction;
heart failure;
complications of prosthetic material;
antimicrobial toxicity;
oral health and prevention;
and, where relevant, addiction treatment and the social determinants of reinfection.

6. Take-home messages

1. The Endocarditis Team is no longer optional in complex disease

The most consequential change is organisational, not technological. Endocarditis needs simultaneous evaluation by clinicians who understand infection, cardiac anatomy, imaging and surgery.

2. Echocardiography remains fundamental, but it has lost diagnostic exclusivity

Cardiac CT, PET/CT, molecular microbiology and the updated Duke-ISCVID criteria increasingly resolve the cases where traditional investigation stalls.

3. Antibiotic treatment is becoming more flexible — not less rigorous

Partial oral therapy and selected long-acting agents widen the options, but only downstream of careful patient selection and adequate source control.

4. Surgical referral should precede therapeutic failure

The major indications are unchanged — heart failure, uncontrolled infection, embolic risk. The timing of the question is what has moved.

5. Source control is outgrowing the medical-versus-surgical dichotomy

Percutaneous mechanical aspiration in selected right-sided IE shows how treatment may become hybrid and staged. Whether it survives contact with prospective data is another matter.

AHA and ESC: more convergence than disagreement

The 2026 AHA statement does not overturn the 2023 ESC Guidelines.

It largely confirms their direction of travel.

The ESC had already pulled infective endocarditis away from the rigid model built on echocardiography, prolonged intravenous antibiotics and delayed intervention.

The AHA reinforces that movement, while adding what has accumulated since: molecular diagnostics, antimicrobial alternatives, catheter-based source control.

For European practice I would therefore read:

ESC 2023 as the principal operational framework

and

AHA 2026 as a contemporary update that should modify how parts of that framework are interpreted.

The most important change, though, is conceptual.

Infective endocarditis is not usefully described as a valve infection requiring several weeks of antibiotics.

It is a dynamic systemic disease in which microbiology, anatomy and physiology evolve simultaneously — and rarely at the same pace.

The strategy has to do likewise.

Diagnosis, antimicrobial therapy and source control proceed in parallel. And the decision about surgery begins when the diagnosis is made, not when medical treatment has already failed.

Clinical perspective

This article offers a clinical interpretation of the 2026 American Heart Association Scientific Statement in the context of the 2023 ESC Guidelines. It is intended for educational purposes and does not replace consultation of the complete source documents or individualized multidisciplinary clinical decision-making.

References

DeSimone DC, Marks L, Dayer MJ, et al. Infective endocarditis: diagnosis, antibiotic therapy, and management: a scientific statement from the American Heart Association. Circulation. Published online September 8, 2026. doi:10.1161/CIR.0000000000001466.
Delgado V, Ajmone Marsan N, de Waha S, et al. 2023 ESC Guidelines for the management of endocarditis. Eur Heart J. 2023;44(39):3948-4042. doi:10.1093/eurheartj/ehad193.
Fowler VG Jr, Durack DT, Selton-Suty C, et al. The 2023 Duke-International Society for Cardiovascular Infectious Diseases Criteria for Infective Endocarditis: Updating the Modified Duke Criteria. Clin Infect Dis. 2023;77(4):518-526. doi:10.1093/cid/ciad271.
DeSimone DC, Esquer Garrigos Z, Marx GE, et al. Blood Culture-Negative Endocarditis: A Scientific Statement From the American Heart Association. J Am Heart Assoc. 2025;14:e040218. doi:10.1161/JAHA.124.040218.
Iversen K, Ihlemann N, Gill SU, et al. Partial Oral versus Intravenous Antibiotic Treatment of Endocarditis. N Engl J Med. 2019;380:415-424. doi:10.1056/NEJMoa1808312.
Bundgaard H, Ihlemann N, Gill SU, et al. Long-Term Outcomes of Partial Oral Treatment of Endocarditis. N Engl J Med. 2019;380:1373-1374. doi:10.1056/NEJMc1902096.
El Sabbagh A, Yucel E, Sethi SS, et al. Percutaneous Mechanical Aspiration in Patients With Right-Sided Infective Endocarditis: A Science Advisory From the American Heart Association. J Am Heart Assoc. 2026;15:e050115. doi:10.1161/JAHA.126.050115.
El Sabbagh A, Hibbert B, Bangalore S, et al. Outcomes of Percutaneous Mechanical Aspiration in Right-Sided Infective Endocarditis: A Multicenter Registry. J Am Coll Cardiol. 2025;86(12):846-856. doi:10.1016/j.jacc.2025.06.054.