Non-bacterial thrombotic endocarditis: an updated narrative review

Introduction

Non-bacterial thrombotic endocarditis (NBTE) is defined by the deposition of sterile fibrin-platelet vegetations on cardiac valve leaflets in the absence of microbial infection. First described by Ziegler in 1888 and formally named by Gross and Friedberg in 1936 (1,2), NBTE has historically been regarded as a rare post-mortem finding. Our group published a review of NBTE in the European Journal of Cardio-Thoracic Surgery in 2007 (3), when ante-mortem diagnosis relied primarily on two-dimensional transoesophageal echocardiography (TEE) and the clinical triad of a predisposing condition, cardiac murmur, and systemic embolism.

In the years since, the number of publications on NBTE has increased, with a reported peak of 35 PubMed-indexed clinical reports in 2023 (4), reflecting advances in multimodality imaging, broader clinical recognition, and new disease associations, most notably severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (4). In parallel, the rapid global expansion of transcatheter aortic valve replacement (TAVR) has drawn attention to sterile thrombus on bioprosthetic leaflets. This entity, which ranges from subclinical hypo-attenuated leaflet thickening (HALT) to clinical bioprosthetic valve thrombosis, occurs after both transcatheter and surgical valve replacement and is conventionally classified as bioprosthetic valve thrombosis rather than as NBTE.

This updated review synthesises the contemporary evidence on the epidemiology, pathogenesis, diagnosis, and management of NBTE; distinguishes native-valve NBTE from Libman-Sacks endocarditis, bioprosthetic valve thrombosis and prosthetic-valve infective endocarditis; and identifies evidence gaps to inform future research. The key questions addressed were: (I) what are the contemporary aetiological associations and clinical features of NBTE? (II) how should NBTE be distinguished from infective endocarditis? (III) what evidence supports the choice of anticoagulant? and (IV) what, if any, is the relationship between NBTE and bioprosthetic leaflet thrombosis after TAVR?

Several recent reviews have addressed aspects of NBTE (4-7). The present review differs from prior work by incorporating the 2025 guidance on blood-culture-negative endocarditis and valvular heart disease (8,9), recent randomised data on antithrombotic therapy after TAVR (10,11), and an explicit comparison of NBTE with bioprosthetic leaflet thrombosis in which any shared mechanism is treated as a hypothesis. We present this article in accordance with the Narrative Review reporting checklist (available at https://shc.amegroups.com/article/view/10.21037/shc-2026-0022/rc).

Methods

We searched PubMed/MEDLINE, Embase, the Cochrane Library and Web of Science from database inception to 25 September 2026 (original search completed in March 2025; updated search completed on 25 September 2026). Search terms combined controlled vocabulary (for example, the MeSH term “Endocarditis, Non-Infective”) and free-text terms for non-bacterial thrombotic, marantic and Libman-Sacks endocarditis with terms for malignancy, antiphospholipid syndrome (APS), coronavirus disease 2019 (COVID-19), anticoagulation, and transcatheter or surgical bioprosthetic leaflet thrombosis. Reference lists of included articles, current international guidelines and scientific statements, and ClinicalTrials.gov were hand-searched. English-language, peer-reviewed publications in adults were eligible, including randomised trials, cohort studies, case series, systematic reviews, guidelines and consensus statements; case reports were included only where they provided unique clinical or mechanistic information. Titles and abstracts were screened by S.A., and full texts were assessed independently by S.A. and S.K.O.; disagreements were resolved by discussion and consensus. The search strategy is summarised in Table 1, and the full PubMed strategy is provided in Table S1.

Because this is a narrative review, no formal risk-of-bias assessment or meta-analysis was performed, and findings are synthesised thematically. Several systematic reviews and meta-analyses of published cases draw on overlapping primary reports and institutional cohorts; we therefore did not aggregate patient numbers across sources, and each numerical estimate is attributed to the individual study from which it derives. The NBTE literature is characterised by a predominance of case reports, case series, and retrospective institutional cohorts, reflecting the rarity of the condition, and no randomised controlled trial specific to NBTE has been conducted. The two largest contemporary clinical series, from the Cleveland Clinic (42 patients over 20 years) and the Mayo Clinic (115 patients with cancer-associated NBTE), provide the most robust clinical characterisation available (12,13). By contrast, evidence on bioprosthetic leaflet thrombosis derives from large TAVR registries and randomised trial imaging sub-studies; whether it shares biology with native-valve NBTE has not been established.

Epidemiology

The autopsy-based prevalence of NBTE ranges from 0.3% to 9.3% (14). In the Cleveland Clinic series, 42 cases were identified among more than 600,000 echocardiograms, confirming its rarity in ante-mortem practice (12). The mean age at diagnosis was 54 years and 66.7% were female, reflecting the contribution of autoimmune aetiologies; malignancy was present in 40.5%, systemic lupus erythematosus (SLE) in 33.3%, and APS in 35.7% (12).

In a Mayo Clinic series of 115 patients with cancer-associated NBTE, from which patients with antiphospholipid antibodies were excluded, lung cancer was the most frequent primary tumour (39.1%), followed by pancreatic (16.5%), gynaecological (14.8%), and gastrointestinal (8.7%) malignancies (13). A 2023 individual-patient meta-analysis of 144 published cancer-associated cases found adenocarcinoma to be the predominant histological subtype (65%), with lung cancer in 29% and pancreatic cancer in 19% (15). In that analysis, 69% had metastatic disease at diagnosis, and NBTE was the presenting manifestation of previously undiagnosed cancer in 66.4% (15); because these figures derive from published case reports, they may be influenced by publication bias.

COVID-19 association

NBTE has been reported in SARS-CoV-2 infection, with endothelial injury, cytokine-mediated hypercoagulability, and tissue factor activation proposed as mechanisms (4,7). The evidence consists of case reports, and the incidence of COVID-19-associated NBTE is unknown.

Bioprosthetic leaflet thrombosis after TAVR and surgical valve replacement

Four-dimensional computed tomography (4D-CT) after bioprosthetic aortic valve replacement may show HALT, with or without reduced leaflet motion (RLM) (16,17). HALT occurs after both TAVR and surgical aortic valve replacement (SAVR), and its reported frequency varies with the population, timing of imaging and antithrombotic regimen, from approximately 10% to almost 30% in trial imaging sub-studies (10,17-19). In the PARTNER 3 CT sub-study, HALT at 30 days was more frequent after TAVR than after SAVR (13% vs. 5%); HALT was dynamic, resolving without anticoagulation in more than half of affected patients by 1 year while appearing de novo in others (18). The Valve Academic Research Consortium 3 (VARC-3) classifies these findings as bioprosthetic valve thrombosis and distinguishes subclinical leaflet thrombosis from clinically significant valve thrombosis (20).

Proposed contributors to HALT include blood stasis in the native sinus and neosinus, incomplete frame expansion, prosthesis size and design, and the non-endothelialised surface of bioprosthetic leaflets early after implantation (16,21). Flow stasis is not unique to TAVR: leaflet-level stasis also occurs with surgical bioprostheses, which is consistent with the detection of HALT after SAVR (17,18). Both balloon-expandable and self-expanding valve designs are affected (18,19). Clinically significant (obstructive or embolic) bioprosthetic valve thrombosis is considerably less common than subclinical HALT (21), although HALT has been associated with cerebrovascular events in a meta-analysis of observational studies (22).

Pathogenesis

Three principal, often overlapping, mechanisms are thought to contribute to native-valve NBTE (Figure 1).

Figure 1 Proposed pathogenesis of native-valve NBTE. A systemic hypercoagulable and/or inflammatory state (for example, malignancy or autoimmune disease) and endothelial injury lead to platelet and fibrin deposition and sterile vegetations on the valve. Vegetations on left-sided (mitral or aortic) valves embolise to the systemic circulation (for example, brain, kidneys, and spleen); pulmonary embolism occurs only in the uncommon setting of right-sided (tricuspid or pulmonary) valve involvement, as labelled in the figure. Downstream effects include tissue ischaemia and infarction, with a secondary local inflammatory response to infarction. The figure was created by the authors with the assistance of an AI image-generation tool and was edited by the authors, who checked all content for accuracy and take full responsibility for it; it has not been published previously. AI, artificial intelligence; NBTE, non-bacterial thrombotic endocarditis.

Immune-mediated injury

In SLE and APS, antiphospholipid antibodies interact with phospholipid-binding proteins on valvular endothelium, promoting complement activation and fibrin-platelet thrombus formation (23). Libman-Sacks lesions are typically small, sessile, verrucous vegetations, most often involving the mitral valve (23,24). Triple positivity (lupus anticoagulant, anticardiolipin and anti-β2-glycoprotein I antibodies) is associated with the highest thrombotic risk (25).

Longitudinal echocardiographic studies in SLE describe the evolution of Libman-Sacks lesions, including resolution, persistence, and the development of valvular thickening and regurgitation, which occasionally requires surgical intervention (24). Hydroxychloroquine is recommended for most patients with SLE (26), but a specific effect on the progression of valvular lesions has not been demonstrated.

Hypoxia and tissue factor activation

Experimental hypoxia upregulates tissue factor and induces NBTE-like lesions in animal models (27). In COVID-19, hypoxia, endothelial injury and systemic inflammation are proposed to create a prothrombotic milieu (7), although direct evidence linking these mechanisms to valvular lesions is limited.

Hypercoagulability and malignancy

Adenocarcinomas, particularly mucin-producing tumours, release tissue factor, cancer procoagulant, and cytokines that promote endothelial tissue factor expression (28,29). Disseminated intravascular coagulation (DIC) frequently coexists with NBTE in autopsy series (30). Chemotherapy-related vascular toxicity and immobility may further increase thrombotic risk (29).

The observation that NBTE may precede the diagnosis of cancer (13,15) supports evaluation for occult malignancy in patients with unexplained sterile valve vegetations, but the optimal extent of this evaluation has not been established. We suggest a documented, risk- and symptom-directed approach: detailed history and examination, routine laboratory tests, completion of age- and sex-appropriate cancer screening, and targeted cross-sectional imaging (for example, CT of the thorax, abdomen, and pelvis) where clinically indicated (29,31). Serum tumour markers [such as carbohydrate antigen 19-9 (CA19-9), carcinoembryonic antigen (CEA), cancer antigen 125 (CA-125) and prostate-specific antigen (PSA)] are not validated as screening tests in this setting, may yield false-positive results, and should be reserved for specific clinical indications. In unprovoked venous thromboembolism, the addition of comprehensive CT to limited screening did not significantly reduce the number of missed occult cancers (32), illustrating the limitations of indiscriminate screening.

Bioprosthetic leaflet thrombosis: a hypothesised overlap

On bioprosthetic valves, blood stasis in the native sinus and neosinus, together with the non-endothelialised leaflet surface early after implantation, promotes platelet aggregation and fibrin deposition (16,21). This process shares features with NBTE, namely sterile fibrin-platelet thrombus on a valve surface, but it occurs on prosthetic tissue, is driven substantially by prosthesis- and flow-related factors, and in most patients is not linked to a systemic prothrombotic disease. We therefore regard any mechanistic relationship between NBTE and HALT as a hypothesis. It could be tested, for example, by histopathological comparison of explanted leaflets and by examining whether systemic hypercoagulable states (such as active cancer or APS) increase the risk of leaflet thrombosis.

For clarity, we distinguish five entities (20,33): (I) native-valve NBTE, comprising sterile fibrin-platelet vegetations on native valves, typically associated with malignancy or other hypercoagulable states; (II) Libman-Sacks endocarditis, the form of NBTE associated with SLE and/or APS, in which immune-mediated valvular injury may lead to chronic thickening and dysfunction; (III) clinical bioprosthetic valve thrombosis, defined as thrombus on a surgical or transcatheter bioprosthesis associated with symptoms, thromboembolism, or haemodynamic valve dysfunction; (IV) subclinical leaflet thrombosis, defined as HALT with or without RLM detected by CT in the absence of clinical sequelae or significant haemodynamic change; and (V) prosthetic-valve infective endocarditis, which is microbial infection of a surgical or transcatheter prosthesis.

Clinical features

Systemic embolism dominates the clinical presentation of NBTE, and stroke is its most frequent manifestation in clinical series (12,13); in a systematic review of 32 published patients with lung cancer-associated NBTE, 47% presented with stroke and 61% had multi-organ infarction involving territories such as the brain, kidneys, spleen, and mesentery (34). Multiple embolic events in different vascular territories, particularly without fever or leucocytosis, in a patient with a predisposing condition should prompt evaluation for NBTE (Figure 1), although infective endocarditis must still be excluded.

After TAVR, subclinical HALT is usually an incidental CT finding and is frequently not associated with an elevated transprosthetic gradient or symptoms (18). In contrast, clinical bioprosthetic valve thrombosis presents with dyspnoea, thromboembolism (including stroke), or haemodynamic valve deterioration such as a rising gradient on echocardiography (20,21).

A key diagnostic scenario is the patient treated empirically for blood-culture-negative infective endocarditis who fails to improve despite appropriate antibiotics. This pattern should prompt reassessment of the diagnosis, including immunological testing and evaluation for malignancy, while recognising that treatment failure may also have infective explanations (8).

Cardiac murmur is frequently absent in NBTE because vegetations often do not cause significant valvular dysfunction. Where valvular regurgitation is identified, it may reflect recurrent endothelial injury with progressive fibrosis rather than acute cusp destruction. A new murmur in the context of malignancy, DIC, or APS should nonetheless prompt echocardiographic evaluation. Recurrent embolism is common: in a patient-level systematic review of 236 patients with NBTE-related stroke, recurrent ischaemic events occurred in 43% over a median follow-up of 2 months (35). Multiterritorial embolism is suggestive of NBTE but is not specific, because it also occurs in infective endocarditis and other cardioembolic conditions (36).

Diagnosis

TEE is more sensitive than transthoracic echocardiography (TTE) for detecting small NBTE vegetations. In the Cleveland Clinic series of 42 patients, vegetations were identified by TTE in 45.2% and by TEE in 97.1% of those who underwent TEE (12); these proportions derive from a small, selected retrospective cohort in which the diagnosis was defined echocardiographically, and they should not be interpreted as population sensitivities. Three-dimensional TEE and multiplanar reconstruction may further improve characterisation. In that series, the mitral valve was most commonly affected (61.9%), followed by the aortic valve (23.8%) (12); echocardiographic findings in cancer-associated NBTE have also been described in a Mayo Clinic series of 111 patients (37).

Diffusion-weighted magnetic resonance imaging (MRI) frequently shows multiple small, disseminated infarcts in several vascular territories in NBTE, whereas infective endocarditis produces more variable patterns; however, the patterns overlap, and multiterritorial infarcts also occur in infective endocarditis and other cardioembolic conditions (36). ¹⁸F-fluorodeoxyglucose positron emission tomography/CT (¹⁸F-FDG PET/CT) is an adjunct rather than a binary discriminator: its sensitivity for native-valve infective endocarditis is limited, so absent valvular uptake does not exclude infection, and valvular uptake may occur in non-infective processes, including active thrombus and, occasionally, NBTE (8,33). Whole-body PET/CT may, however, contribute to the search for occult malignancy (31).

When bioprosthetic valve thrombosis is suspected, TEE or 4D-CT is recommended (9); CT is particularly useful after TAVR, when the stent frame may limit echocardiographic assessment of the leaflets. Routine CT screening for HALT in asymptomatic TAVR recipients is not recommended outside research settings (9,38).

Exclusion of infective endocarditis should follow a contemporary blood-culture-negative endocarditis algorithm and the 2023 Duke-ISCVID criteria (8,33,39). At least three sets of blood cultures should be obtained, ideally before antibiotics are started, because prior antibiotic exposure is the most common cause of culture negativity (8). HACEK (Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella and Kingella) organisms are generally recovered by modern continuous-monitoring culture systems or identified by molecular methods, and routine extended incubation is usually unnecessary (8). Further testing should be exposure-directed, including serology for Coxiella burnetii and Bartonella species, Brucella testing where there is relevant exposure, and consideration of Tropheryma whipplei and fungal infection in appropriate clinical contexts (8,33). When valve tissue is excised, it should undergo histopathology, culture, and molecular testing [for example, 16S ribosomal RNA (rRNA) gene polymerase chain reaction], which can establish both infective and non-infective diagnoses (8). Antiphospholipid antibody testing (lupus anticoagulant, anticardiolipin and anti-β2-glycoprotein I antibodies) is appropriate when NBTE is suspected, but results require cautious interpretation: acute illness and anticoagulant therapy can interfere with lupus anticoagulant assays (40), antiphospholipid antibodies may be transiently positive during infection, including infective endocarditis, and APS requires persistent positivity on repeat testing at least 12 weeks apart (41). A single acute panel should not therefore be equated with APS. SARS-CoV-2 testing should be considered when the clinical timeline is consistent with recent infection. No single clinical, laboratory, or imaging feature independently excludes infective endocarditis (Table 2).

Bioprosthetic valve thrombosis after TAVR: relationship to NBTE

With the continuing growth of TAVR, clinicians increasingly encounter valve-level lesions after TAVR whose differential diagnosis includes subclinical HALT, clinical bioprosthetic valve thrombosis, and prosthetic-valve infective endocarditis. Infective endocarditis after TAVR has been reported at approximately 1% per person-year, is most commonly caused by enterococci and Staphylococcus aureus, and carries in-hospital mortality of approximately 36% to 47% in registry series (42-44). Prosthetic-valve infective endocarditis is usually, but not invariably, associated with positive blood cultures and systemic inflammation; ¹⁸F-FDG PET/CT is more informative in prosthetic than in native-valve endocarditis but may show non-specific uptake early after implantation (33).

HALT and RLM frequently regress with oral anticoagulation, supporting a thrombotic mechanism, although spontaneous regression also occurs (16-18). Current guidelines recommend single antiplatelet therapy after TAVR in patients without an independent indication for oral anticoagulation and advise against routine oral anticoagulation or dual antiplatelet therapy to prevent valve thrombosis (9). In GALILEO, a rivaroxaban-based strategy was associated with higher rates of death or thromboembolic events and of bleeding than an antiplatelet-based strategy (45), and in ATLANTIS, apixaban was not superior to standard care (46). More recently, ACASA-TAVI reported that 12 months of direct oral anticoagulant (DOAC) monotherapy reduced HALT compared with aspirin and was non-inferior for a composite of bleeding, thromboembolic events and death (10), whereas in NOTION-4 a 3-month course of DOAC therapy reduced HALT at 3 months but not at 12 months and was associated with more adverse clinical events (11). Whether reducing HALT improves clinical outcomes or valve durability remains unknown.

The complexity of antithrombotic management in patients with cancer is illustrated by a reported case of clinical valve thrombosis and arterial embolism after TAVR in a patient with metastatic pancreatic cancer and permanent atrial fibrillation who was receiving single antiplatelet therapy (47). The authors attributed the use of aspirin monotherapy, despite atrial fibrillation, to concern about cancer-related bleeding risk. The patient developed acute limb ischaemia requiring surgical thrombectomy, and the valve thrombosis improved with warfarin (47). A single case cannot establish how CHA₂DS₂-VASc or HAS-BLED perform in patients with cancer; rather, it illustrates that, in patients with an independent indication for anticoagulation, a decision to withhold anticoagulation after TAVR requires explicit multidisciplinary weighing of thrombotic and bleeding risks and close follow-up. Such scenarios should be captured in prospective registries.

The distinction between bioprosthetic valve thrombosis and prosthetic-valve infective endocarditis carries important therapeutic implications: anticoagulation is the principal treatment for clinical valve thrombosis, whereas infective endocarditis requires prolonged antimicrobial therapy with Heart Team assessment for surgery in selected patients. Misclassification exposes patients to either unnecessary prolonged antibiotic therapy or inadequate treatment of life-threatening infection. An integrated assessment combining blood cultures, inflammatory markers, TEE, cardiac CT, and, where appropriate, ¹⁸F-FDG PET/CT is recommended for suspected valve-level pathology after TAVR (9,33).

Management of leaflet thrombosis after TAVR should be phenotype-directed. Clinical bioprosthetic valve thrombosis, with symptoms, thromboembolism, or haemodynamic valve dysfunction, generally warrants oral anticoagulation, usually with a vitamin K antagonist (VKA), and repeat imaging to confirm resolution (9,38). For isolated subclinical HALT without haemodynamic consequence, the benefit of treatment is uncertain, spontaneous regression is common, and routine anticoagulation is not recommended; management should take account of bleeding risk and whether an independent indication for anticoagulation, such as atrial fibrillation or venous thromboembolism, is already present (9,18). Patients with active cancer or APS were largely excluded from the randomised trials, so decisions in these groups should be individualised. The optimal agent and duration of anticoagulation for clinical bioprosthetic valve thrombosis, and the role of DOACs, remain uncertain.

Management

Management must be individualised. No randomised controlled trial specific to NBTE has been conducted, and recommendations therefore rely on observational data, indirect evidence, and expert opinion (4,48).

Treatment of the underlying condition

Treatment of the underlying condition remains central. Effective oncological treatment has been associated with regression of vegetations and fewer embolic events in case reports and series (29); in the systematic review of lung cancer-associated NBTE, mortality was 38% among patients who received chemotherapy and 91% among those who did not, a comparison that is confounded by performance status and disease stage (34). In SLE and APS, the underlying disease should be managed according to current recommendations (25,26).

Anticoagulation

In the absence of contraindications, therapeutic anticoagulation is generally recommended for NBTE with systemic or pulmonary embolism; the 2012 American College of Chest Physicians guideline suggested full-dose unfractionated heparin (UFH) or subcutaneous low-molecular-weight heparin (LMWH) (Grade 2C) (48). UFH is useful in the acute phase when rapid reversal may be needed, for example around invasive procedures, in patients at high bleeding risk, after haemorrhagic transformation of a cerebral infarct, or in severe renal impairment. LMWH is commonly preferred for longer-term treatment of malignancy-associated NBTE, by extrapolation from trials in cancer-associated venous thromboembolism such as CLOT (49), which provide indirect evidence only, and from observational NBTE data (28,29); dose adjustment or monitoring may be required in renal impairment, thrombocytopenia, and at extremes of body weight. VKAs are an alternative when long-term oral therapy is required but may be difficult to manage in active cancer because of drug interactions, nutritional variability, and unstable international normalised ratio (INR) control.

Comparative data on anticoagulants in NBTE are limited to observational studies and systematic reviews of published cases. In the systematic review of 32 patients with lung cancer-associated NBTE, of whom 16 received anticoagulation, mortality was 20% with LMWH, 75% with DOACs and 85.7% with UFH (34). These figures derive from very small, uncontrolled, treatment-stratified subgroups of case reports and are subject to confounding by indication, selection bias and publication bias; they cannot be interpreted as causal treatment effects. In a patient-level systematic review of NBTE-related stroke, recurrent infarction was reported in 56% (10/18) of patients treated with DOACs, 29% (35/120) of those treated with UFH or LMWH, and 18% (11/60) of those treated with VKAs (35), with the same limitations. Recurrent embolism during DOAC therapy has been reported (50), whereas other reports describe regression of vegetations during DOAC therapy (51). These data justify caution with DOACs as first-line therapy for NBTE, particularly in malignancy, but do not establish the superiority of any agent. Mechanistically, heparins act indirectly through antithrombin, LMWH predominantly by inhibiting factor Xa and UFH by inhibiting both factor Xa and thrombin (52); experimental effects of heparins on selectin-mediated tumour-cell adhesion and metastasis are of mechanistic interest but have not been shown to improve outcomes in NBTE.

APS

In thrombotic APS, VKAs are the anticoagulants of choice (25). After a first venous thrombosis, a target INR of 2–3 is recommended; after arterial thrombosis, options include a VKA with a target INR of 2–3, a VKA with a target INR of 3–4, or a VKA (INR 2–3) plus low-dose aspirin, chosen according to the individual risks of bleeding and recurrence, and recurrent events despite adequate anticoagulation require specialist review (25). The TRAPS trial, which enrolled triple-positive patients, was terminated early because of an excess of arterial thromboembolic events with rivaroxaban compared with warfarin (53). DOACs should therefore be avoided in triple-positive patients and in those with arterial events, whereas their use in selected lower-risk patients with venous events remains debated (25). The diagnosis of APS requires persistent antiphospholipid antibody positivity confirmed at least 12 weeks apart (41).

Surgical intervention

Surgical intervention is seldom required. Potential indications include haemodynamically significant valvular dysfunction, recurrent embolism despite adequate anticoagulation in patients with a controllable underlying disease, and diagnostic uncertainty requiring valve excision and histological analysis (4,5). In the individual-patient meta-analysis of cancer-associated cases, 21 patients underwent surgery, with a trend toward improved survival that is likely to be influenced by patient selection (15).

In TAVR recipients with clinical bioprosthetic valve thrombosis causing significant valve dysfunction refractory to anticoagulation, surgical or transcatheter reintervention may be considered by the Heart Team (9). In native-valve NBTE, the decision to intervene surgically must weigh the embolic risk of the vegetation against the operative risk in patients who are frequently debilitated by advanced malignancy or autoimmune disease. In patients with potentially curable or well-controlled malignancy, removal of the embolic source may enable continuation of oncological therapy that would otherwise be compromised by recurrent cerebral embolism.

Summary of developments, limitations and future research

The clinical landscape of NBTE has changed substantially since our 2007 review. Important developments include recognition of SARS-CoV-2 infection as a potential precipitant; larger contemporary series characterising malignancy associations; recognition that sterile thrombus also occurs on bioprosthetic valves after TAVR and SAVR, an entity that is distinct from NBTE but may share some mechanisms; and expansion of the multimodality diagnostic toolkit.

The quality of the evidence reviewed is a fundamental limitation. The primary literature consists predominantly of retrospective single-centre series and case reports, with inconsistent definitions, selective reporting, and publication bias favouring unusual or successfully treated cases; systematic reviews and meta-analyses of published cases inherit these biases, and several include overlapping patients. Comparisons between anticoagulants are confounded by indication and disease severity. The evidence on bioprosthetic leaflet thrombosis is more robust, deriving from randomised trial imaging sub-studies, but patients with active cancer or APS were largely excluded. As a narrative review, this article is also subject to selection bias in the identification and interpretation of studies. International prospective registries structured around standardised diagnostic criteria are the most feasible path toward improving the evidence base.

Several specific research priorities can be identified. First, the optimal anticoagulant for NBTE, including the role of DOACs, requires prospective evaluation in registries with standardised outcome definitions or in pragmatic randomised trials. Second, the optimal evaluation for occult malignancy in patients presenting with NBTE should be defined. Third, whether systemic hypercoagulable states such as active cancer or APS increase the risk, or modify the natural history, of leaflet thrombosis after TAVR or SAVR is unknown, as is the optimal antithrombotic strategy in such patients. Fourth, histopathological studies comparing NBTE vegetations with bioprosthetic leaflet thrombus could test the hypothesised mechanistic overlap. Fifth, whether treating isolated subclinical HALT improves clinical outcomes or valve durability remains unresolved (10,11).

Conclusions

NBTE remains diagnostically and therapeutically challenging. Diagnosis requires TEE and a rigorous, contemporary exclusion of infective endocarditis, recognising that no single feature is discriminatory, together with risk-directed evaluation for malignancy and autoimmune disease. Management is individualised and centres on treatment of the underlying condition and anticoagulation: heparins are generally favoured on observational and indirect evidence, VKAs are preferred in thrombotic APS, and DOACs should be used with caution. Bioprosthetic leaflet thrombosis after TAVR or SAVR is a distinct entity; any biological relationship with NBTE remains hypothetical, and routine CT screening or anticoagulation after TAVR is not recommended. Prospective international registries with standardised definitions are needed to inform clinical practice and future trials.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://shc.amegroups.com/article/view/10.21037/shc-2026-0022/rc

Peer Review File: Available at https://shc.amegroups.com/article/view/10.21037/shc-2026-0022/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://shc.amegroups.com/article/view/10.21037/shc-2026-0022/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Ziegler E. Ueber den Bau und die Entstehung der endocarditischen Efflorescenzen. Verh Kongr Inn Med 1888;7:339-43.
2. Gross L, Friedberg CK. Nonbacterial thrombotic endocarditis: classification and general description. Arch Intern Med 1936;58:620-40.
3. Asopa S, Patel A, Khan OA, et al. Non-bacterial thrombotic endocarditis. Eur J Cardiothorac Surg 2007;32:696-701. [Crossref] [PubMed]
4. Ahmed O, King NE, Qureshi MA, et al. Non-bacterial thrombotic endocarditis: a clinical and pathophysiological reappraisal. Eur Heart J 2025;46:236-49. [Crossref] [PubMed]
5. Hurrell H, Roberts-Thomson R, Prendergast BD. Non-infective endocarditis. Heart 2020;106:1023-9. [Crossref] [PubMed]
6. Zmaili M, Alzubi J, Lo Presti Vega S, et al. Non-bacterial thrombotic endocarditis: A state-of-the-art contemporary review. Prog Cardiovasc Dis 2022;74:99-110. [Crossref] [PubMed]
7. Al Chalaby S, Makhija RR, Sharma AN, et al. Nonbacterial Thrombotic Endocarditis: Presentation, Pathophysiology, Diagnosis and Management. Rev Cardiovasc Med 2022;23:137. [Crossref] [PubMed]
8. DeSimone DC, Garrigos ZE, Marx GE, et al. Blood Culture-Negative Endocarditis: A Scientific Statement From the American Heart Association: Endorsed by the International Society for Cardiovascular Infectious Diseases. J Am Heart Assoc 2025;14:e040218. [Crossref] [PubMed]
9. Praz F, Borger MA, Lanz J, et al. 2025 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J 2025;46:4635-736. [Crossref] [PubMed]
10. Dodgson CS, Herstad J, Kløve SF, et al. Anticoagulation Monotherapy vs Antiplatelet Monotherapy After Transcatheter Aortic Valve Implant: The ACASA-TAVI Randomized Clinical Trial. JAMA 2026. [Epub ahead of print]. doi: 10.1001/jama.2026.17036.
11. Jørgensen TH, Larsen AF, Jensen JM, et al. Short-Term Anticoagulant Therapy and Subclinical Leaflet Thickening in Transcatheter Aortic Valves: The NOTION-4 Trial. J Am Coll Cardiol 2026. [Epub ahead of print]. doi: 10.1016/j.jacc.2026.08.023.
12. Zmaili MA, Alzubi JM, Kocyigit D, et al. A Contemporary 20-Year Cleveland Clinic Experience of Nonbacterial Thrombotic Endocarditis: Etiology, Echocardiographic Imaging, Management, and Outcomes. Am J Med 2021;134:361-9. [Crossref] [PubMed]
13. Patrzalek P, Wysokinski WE, Kurmann RD, et al. Cancer-associated non-bacterial thrombotic endocarditis-Clinical series from a single institution. Am J Hematol 2024;99:596-605. [Crossref] [PubMed]
14. Lopez JA, Ross RS, Fishbein MC, et al. Nonbacterial thrombotic endocarditis: a review. Am Heart J 1987;113:773-84. [Crossref] [PubMed]
15. Rahouma M, Khairallah S, Dabsha A, et al. Lung Cancer as a Leading Cause among Paraneoplastic Non-Bacterial Thrombotic Endocarditis: A Meta-Analysis of Individual Patients' Data. Cancers (Basel) 2023;15:1848. [Crossref] [PubMed]
16. Makkar RR, Fontana G, Jilaihawi H, et al. Possible Subclinical Leaflet Thrombosis in Bioprosthetic Aortic Valves. N Engl J Med 2015;373:2015-24. [Crossref] [PubMed]
17. Chakravarty T, Søndergaard L, Friedman J, et al. Subclinical leaflet thrombosis in surgical and transcatheter bioprosthetic aortic valves: an observational study. Lancet 2017;389:2383-92. [Crossref] [PubMed]
18. Makkar RR, Blanke P, Leipsic J, et al. Subclinical Leaflet Thrombosis in Transcatheter and Surgical Bioprosthetic Valves: PARTNER 3 Cardiac Computed Tomography Substudy. J Am Coll Cardiol 2020;75:3003-15. [Crossref] [PubMed]
19. Blanke P, Leipsic JA, Popma JJ, et al. Bioprosthetic Aortic Valve Leaflet Thickening in the Evolut Low Risk Sub-Study. J Am Coll Cardiol 2020;75:2430-42. [Crossref] [PubMed]
20. Généreux P, Piazza N, et al. Valve Academic Research Consortium 3: updated endpoint definitions for aortic valve clinical research. Eur Heart J 2021;42:1825-57. [Crossref] [PubMed]
21. Yanagisawa R, Tanaka M, Yashima F, et al. Early and Late Leaflet Thrombosis After Transcatheter Aortic Valve Replacement. Circ Cardiovasc Interv 2019;12:e007349. [Crossref] [PubMed]
22. Rashid HN, Gooley RP, Nerlekar N, et al. Bioprosthetic aortic valve leaflet thrombosis detected by multidetector computed tomography is associated with adverse cerebrovascular events: a meta-analysis of observational studies. EuroIntervention 2018;13:e1748-55. [Crossref] [PubMed]
23. Hojnik M, George J, Ziporen L, et al. Heart valve involvement (Libman-Sacks endocarditis) in the antiphospholipid syndrome. Circulation 1996;93:1579-87. [Crossref] [PubMed]
24. Moyssakis I, Tektonidou MG, Vasilliou VA, et al. Libman-Sacks endocarditis in systemic lupus erythematosus: prevalence, associations, and evolution. Am J Med 2007;120:636-42. [Crossref] [PubMed]
25. Tektonidou MG, Andreoli L, Limper M, et al. EULAR recommendations for the management of antiphospholipid syndrome in adults. Ann Rheum Dis 2019;78:1296-304. [Crossref] [PubMed]
26. Fanouriakis A, Kostopoulou M, Andersen J, et al. EULAR recommendations for the management of systemic lupus erythematosus: 2023 update. Ann Rheum Dis 2024;83:15-29. [Crossref] [PubMed]
27. Nakanishi K, Tajima F, Nakata Y, et al. Tissue factor is associated with the nonbacterial thrombotic endocarditis induced by a hypobaric hypoxic environment in rats. Virchows Arch 1998;433:375-9. [Crossref] [PubMed]
28. el-Shami K, Griffiths E, Streiff M. Nonbacterial thrombotic endocarditis in cancer patients: pathogenesis, diagnosis, and treatment. Oncologist 2007;12:518-23. [Crossref] [PubMed]
29. Itzhaki Ben Zadok O, Spectre G, Leader A. Cancer-associated non-bacterial thrombotic endocarditis. Thromb Res 2022;213:S127-S132.
30. Kim HS, Suzuki M, Lie JT, et al. Nonbacterial thrombotic endocarditis (NBTE) and disseminated intravascular coagulation (DIC): autopsy study of 36 patients. Arch Pathol Lab Med 1977;101:65-8. [PubMed]
31. Tonutti A, Scarfò I, La Canna G, et al. Diagnostic Work-Up in Patients with Nonbacterial Thrombotic Endocarditis. J Clin Med 2023;12:5819. [Crossref] [PubMed]
32. Carrier M, Lazo-Langner A, Shivakumar S, et al. Screening for Occult Cancer in Unprovoked Venous Thromboembolism. N Engl J Med 2015;373:697-704. [Crossref] [PubMed]
33. 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. [Crossref] [PubMed]
34. Kamel M, Hussain F, Leung C, et al. Non-bacterial Thrombotic Endocarditis in Lung Cancer: A Systematic Review. Curr Cardiol Rev 2025;21:e1573403X343187. [Crossref] [PubMed]
35. Aljabali A, Alhayali D, Taylor-Wedge H, et al. Stroke in non-bacterial thrombotic endocarditis: a patient-level systematic review of clinical characteristics, management, and outcomes. Front Stroke 2026;5:1915602. [Crossref] [PubMed]
36. Singhal AB, Topcuoglu MA, Buonanno FS. Acute ischemic stroke patterns in infective and nonbacterial thrombotic endocarditis: a diffusion-weighted magnetic resonance imaging study. Stroke 2002;33:1267-73. [Crossref] [PubMed]
37. Kurmann RD, Klarich KW, Wysokinska E, et al. Echocardiographic findings in cancer-associated non-bacterial thrombotic endocarditis: clinical series of 111 patients from a single institution. Eur Heart J Cardiovasc Imaging 2024;25:1255-63. [Crossref] [PubMed]
38. Otto CM, Nishimura RA, et al. 2020 ACC/AHA Guideline for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol 2021;77:e25-e197. [Crossref] [PubMed]
39. Fowler VG, 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;77:518-26. [Crossref] [PubMed]
40. Devreese KMJ, de Groot PG, de Laat B, et al. Guidance from the Scientific and Standardization Committee for lupus anticoagulant/antiphospholipid antibodies of the International Society on Thrombosis and Haemostasis: Update of the guidelines for lupus anticoagulant detection and interpretation. J Thromb Haemost 2020;18:2828-39. [Crossref] [PubMed]
41. Barbhaiya M, Zuily S, Naden R, et al. The 2023 ACR/EULAR Antiphospholipid Syndrome Classification Criteria. Arthritis Rheumatol 2023;75:1687-702. [Crossref] [PubMed]
42. Regueiro A, Linke A, Latib A, et al. Association Between Transcatheter Aortic Valve Replacement and Subsequent Infective Endocarditis and In-Hospital Death. JAMA 2016;316:1083-92. [Crossref] [PubMed]
43. Amat-Santos IJ, Messika-Zeitoun D, Eltchaninoff H, et al. Infective endocarditis after transcatheter aortic valve implantation: results from a large multicenter registry. Circulation 2015;131:1566-74. [Crossref] [PubMed]
44. Butt JH, Ihlemann N, De Backer O, et al. Long-Term Risk of Infective Endocarditis After Transcatheter Aortic Valve Replacement. J Am Coll Cardiol 2019;73:1646-55. [Crossref] [PubMed]
45. Dangas GD, Tijssen JGP, Wöhrle J, et al. A Controlled Trial of Rivaroxaban after Transcatheter Aortic-Valve Replacement. N Engl J Med 2020;382:120-9. [Crossref] [PubMed]
46. Collet JP, Van Belle E, Thiele H, et al. Apixaban vs. standard of care after transcatheter aortic valve implantation: the ATLANTIS trial. Eur Heart J 2022;43:2783-97. [Crossref] [PubMed]
47. Sawano S, Miura M, Higashikuni Y, et al. Clinical valve thrombosis and arterial embolism in a cancer patient after transcatheter aortic valve replacement. Oxf Med Case Reports 2023;2023:omad125. [Crossref] [PubMed]
48. Whitlock RP, Sun JC, Fremes SE, et al. Antithrombotic and thrombolytic therapy for valvular disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e576S-600S. [Crossref] [PubMed]
49. Lee AY, Levine MN, Baker RI, et al. Low-molecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003;349:146-53. [Crossref] [PubMed]
50. Mantovani F, Navazio A, Barbieri A, et al. A first described case of cancer-associated non-bacterial thrombotic endocarditis in the era of direct oral anticoagulants. Thromb Res 2017;149:45-7. [Crossref] [PubMed]
51. Donato G, Castor F, Fernandes JVA, et al. Cancer-associated Marantic Endocarditis and Direct Oral Anticoagulants: Case Report and Systematic Review. Curr Cardiol Rev 2026;22:e1573403X371101. [Crossref] [PubMed]
52. Garcia DA, Baglin TP, Weitz JI, et al. Parenteral anticoagulants: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e24S-43S. [Crossref] [PubMed]
53. Pengo V, Denas G, Zoppellaro G, et al. Rivaroxaban vs warfarin in high-risk patients with antiphospholipid syndrome. Blood 2018;132:1365-71. [Crossref] [PubMed]