Hypereosinophilic endomyocardial disease

Article written and reviewed by Dr Luca Moretti

Hypereosinophilic endomyocardial disease includes cardiac lesions attributable to eosinophilic activation and infiltration in the context of hypereosinophilic syndromes and the conditions that cause them. Injury may involve myocardium, ventricular endocardium, and the atrioventricular apparatus, presenting as myocarditis, intracavitary thrombosis, or restrictive fibrotic remodeling. The designation links the heart to a systemic disease: the simultaneous presence of heart disease and a high eosinophil count is not sufficient to prove that the former is caused by the latter.

The necrotic, thrombotic, and fibrotic phenotype is classically described as Loeffler endocarditis, but cardiac involvement may be recognized before advanced endocardial lesions appear. Conversely, a patient with established fibrosis may have a normal eosinophil count at the time of assessment because of treatment or inactivity of the process that produced it. The diagnostic pathway should therefore reconstruct both the current disease and its history, distinguishing biologic activity, tissue injury, and hemodynamic consequences.

Hypereosinophilic syndromes are rare and heterogeneous. The frequency of cardiac involvement varies according to subtype, diagnostic criteria, disease duration, and screening intensity; percentages from historical series cannot be transferred indiscriminately to populations treated early or to individual molecularly defined neoplasms. The heart nevertheless retains central prognostic importance because an initially minimally symptomatic lesion may progress to embolization or irreversible fibrosis. The availability of targeted treatments makes it particularly important to identify the cause before assigning every presentation to a generic idiopathic syndrome.

Etiology, pathogenesis, and pathophysiology

Hypereosinophilia is a quantitative or tissue definition, whereas hypereosinophilic syndrome requires organ damage attributable to eosinophils. The distinction avoids two opposite errors: automatically treating an asymptomatic hematologic abnormality as organ disease and delaying treatment of severe injury because it has not persisted for months. The historical six-month interval is not required to initiate contemporary diagnosis and treatment. Moreover, eosinophilic tissue injury may be clinically relevant even when the blood count does not consistently reach the conventional threshold.

Etiologic classification separates clonal myeloid proliferations, reactive eosinophilias, and idiopathic forms after exclusion of recognizable causes. In myeloid or myeloid/lymphoid neoplasms with tyrosine-kinase fusions, eosinophils are part of the pathologic clone; genes such as PDGFRA, PDGFRB, FGFR1, and JAK2 identify categories with different behavior and therapeutic sensitivity. Chronic eosinophilic leukemia is another hematologic diagnosis to be made according to appropriate morphologic and molecular criteria. These conditions should not be merged under a label that erases their prognostic implications.

In the lymphocytic variant, a lymphocyte population with an aberrant phenotype may produce cytokines, particularly interleukin 5, that stimulate eosinophils that are not necessarily clonal. Cutaneous manifestations, lymphadenopathy, and increased immunoglobulins may raise suspicion, but diagnosis requires integration of immunophenotyping, clonality studies, and the clinical picture. Detection of a T-cell receptor rearrangement alone does not demonstrate that the clone is responsible for eosinophilia. Distinction from myeloid proliferation has therapeutic consequences and requires surveillance of the lymphocytic component in addition to control of eosinophil counts.

Reactive causes include parasitic infections relevant to exposure, drug reactions, certain neoplasms, and immune-mediated diseases. Asthma and atopy may accompany eosinophilia but do not automatically explain a marked persistent increase or cardiac injury. Eosinophilic granulomatosis with polyangiitis requires particular attention when asthma, rhinosinusitis, neuropathy, or vasculitic findings coexist; negative ANCA does not exclude it. The idiopathic form is defined after adequate investigation, whereas hypereosinophilia without organ damage and without an identified cause remains a condition to be followed, not a synonym for overt hypereosinophilic syndrome.

Cardiac toxicity depends on tissue recruitment and degranulation. Eosinophil cationic proteins and oxidant products damage cardiomyocytes and endothelium, amplifying inflammation and coagulation activation. The blood count measures neither the tissue burden nor the amount of mediators already released. This explains the possible discrepancy between modest peripheral eosinophilia and major injury, as well as persistence of dysfunction after normalization of the blood count. Organ susceptibility and biologic subtype contribute to why patients with similar counts have different clinical courses.

In the heart, inflammatory injury may produce edema, necrosis, arrhythmias, and depressed contractility. Injury to the endocardial surface promotes intracavitary thrombosis, often apical, to which stasis and cavity deformation contribute. Thrombogenicity does not, however, depend only on ejection fraction: eosinophilic activation may render a relatively well-contracting surface pathologic. During repair, organized thrombi and fibrous tissue may reduce ventricular volume and incorporate the subvalvular apparatus. The sequence is biologically plausible and well described, but its different components may be simultaneous.

The resulting restrictive physiology combines high filling pressures with limited volume reserve. Stroke volume may be low even with preserved ejection fraction, while mitral or tricuspid regurgitation amplifies congestion. Systemic involvement introduces additional mechanisms: anemia, lung disease, thromboembolism, vasculitis, and treatment toxicity may contribute to dyspnea or reduced functional capacity. Attributing every symptom to the heart, or every cardiac abnormality to eosinophilia, prevents recognition of treatable components requiring different interventions.

Clinical manifestations

Cardiac involvement may initially be silent. Elevated troponin, an electrocardiographic abnormality, or an echocardiographic finding may precede heart failure, justifying baseline assessment even when the patient presents for cutaneous or hematologic manifestations. Acute forms cause chest pain, dyspnea, palpitations, and syncope; hypotension, oliguria, and impaired perfusion suggest possible severe myocarditis. The speed of progression matters more than the documented duration of eosinophilia: threatening organ injury requires immediate decisions.

In chronic disease, dyspnea and congestion reflect restriction, systolic dysfunction, or atrioventricular regurgitation. Ascites, hepatomegaly, and edema may dominate in right-sided forms, whereas orthopnea and pulmonary congestion point to elevated left-sided pressures. Exercise capacity may worsen before resting signs appear because the stiff ventricle cannot adequately increase output. A new episode of atrial fibrillation or persistent tachycardia may reveal a previously compensated limitation.

Embolic ischemic events may involve the brain, limbs, and viscera; venous or microvascular thrombosis may coexist as an expression of systemic disease. Event topography and imaging help distinguish an intracardiac source from local thrombosis without assuming that every ischemic event is caused by an apical thrombus. In patients with neurologic symptoms, peripheral neuropathy, vasculitis, and embolism have different implications: sudden focal deficits, asymmetric sensory disturbances, and a subacute course should be reconstructed precisely.

Extracardiac assessment looks for pruritus, urticaria, angioedema, infiltrative lesions, respiratory and gastrointestinal symptoms, neurologic disorders, thrombosis, splenomegaly, and lymphadenopathy. Meaning derives from the combination and chronology: isolated dermatitis does not identify the lymphocytic variant, splenomegaly does not prove a myeloid neoplasm, and asthma alone is insufficient to diagnose vasculitis. Drugs, supplements, occupational exposures, travel, and previous blood counts help narrow the hypotheses. Response to corticosteroids is also a descriptive finding, not an independent etiologic proof.

Physical examination includes rhythm, signs of regurgitation, perfusion and congestion, together with skin, lymph nodes, spleen, and neurologic assessment. Searching for systemic manifestations should not delay recognition of a cardiac emergency. Conversely, after stabilization, an examination limited to the heart risks missing decisive clues for causal treatment. The multidisciplinary approach is most valuable when it allows these findings to be assembled into a single verifiable diagnosis rather than treating each organ as an independent problem.

Investigations and diagnosis

The first step is to confirm the hematologic abnormality and assess its significance. Absolute eosinophil count, repeat complete blood count, and peripheral smear are used, comparing the results with previous tests and current therapies. An isolated percentage can be misleading, especially when the total leukocyte count changes. Contemporary definitions distinguish blood elevation from tissue demonstration and, for the syndrome, require a reasonable link with organ dysfunction. Any need for temporal confirmation must not become mandatory waiting in the face of threatening myocarditis or thrombosis.

Elements of the proposed definition of hypereosinophilia and hypereosinophilic syndrome:

• Blood hypereosinophilia: absolute count of at least 1.5 × 10⁹/L, equivalent to 1,500/µL, documented on at least two determinations separated by at least two weeks.
• Tissue hypereosinophilia: marked increase in tissue eosinophils on pathologic assessment, or eosinophils exceeding 20% of nucleated cells in bone marrow, or extensive tissue deposition of eosinophil granule proteins.
• Hypereosinophilic syndrome: blood and/or tissue hypereosinophilia with organ damage or dysfunction attributable to eosinophils, after exclusion of a predominant alternative explanation.

Initial cardiac screening includes ECG, troponin, natriuretic peptides, and echocardiography. Echocardiography evaluates ventricular function, apices, thrombi, endocardial thickening, atrial dilatation, and regurgitation; a normal result does not exclude early injury if troponin, symptoms, or course maintain a high level of suspicion. Cardiac magnetic resonance is then considered to define inflammation, scar, and thrombosis. The frequency of subsequent surveillance depends on activity, subtype, and initial findings: negative screening is not a permanent guarantee in the presence of persistent hypereosinophilia.

CMR can distinguish edema, expansion of extracellular space, and avascular masses by integrating cine imaging, mapping, and late contrast enhancement. A nonterritorial subendocardial distribution supports suspicion of endomyocardial disease but does not by itself demonstrate eosinophilic origin. Thrombus may be identified even when flattened against the wall and poorly visible on echocardiography. Serial comparison should separate resolution of edema, evolution of fibrosis, and thrombus resolution because these phenomena have different time courses and implications.

Endomyocardial biopsy is indicated mainly when myocarditis typing changes an urgent decision or when the picture remains uncertain despite noninvasive testing. A fulminant presentation, ventricular arrhythmias, or advanced conduction blocks make histologic recognition more relevant. The result depends on the site, number and quality of samples, and previous treatment; a focal infiltrate may be missed. Biopsy is not automatically needed to confirm every stable scar and should be planned with thrombi and procedural risk in mind.

The search for reactive causes begins with history and organ clues. The medication list includes nonprescription products and treatments already discontinued; investigation for parasitic disease is guided by exposures and probability, using methods appropriate to the organism. Investigation for Strongyloides is particularly important if corticosteroids are planned because persistent infection may be complicated by hyperinfection. When vasculitis is suspected, ANCA, urinalysis, renal function, imaging, and neurologic or histologic assessment are integrated according to site. An isolated positive test does not replace the clinical picture, and a negative result does not close every hypothesis.

Myeloid assessment includes evaluation of the other blood cell lines, smear, vitamin B12, tryptase, and, when indicated, bone marrow examination, cytogenetics, and molecular testing. The FIP1L1::PDGFRA fusion may result from a cryptic abnormality, so a normal karyotype does not exclude it. The choice among targeted tests, FISH, and broader methods depends on suspicion and specialist availability. A single negative PDGFRA test does not exhaust the work-up for neoplasms with eosinophilia, particularly when splenomegaly, morphologic abnormalities, or changes in other cell lines persist.

Flow cytometry and T-cell receptor studies investigate a possible lymphocytic variant, while suspicious lymphadenopathy may require dedicated tissue sampling. Sequencing panels may identify useful variants, but a mutation associated with clonal hematopoiesis, particularly in an older patient, does not by itself demonstrate that eosinophilia is neoplastic or causes cardiac injury. The result should be integrated with morphology, variant allele fraction, other abnormalities, and criteria for the specific entity. On the lymphocytic side, demonstrating an aberrant T-cell population means identifying a possible stimulus for eosinophil production; on the myeloid side, demonstrating a relevant fusion means identifying a lesion of the neoplastic clone. This distinction explains why eliminating eosinophils may control injury without eliminating the population that sustains them, whereas a drug directed against a sensitive tyrosine kinase may modify the proliferative process at its source. It is therefore the biologic coherence between the finding and the phenotype, not positivity of a clonality test alone, that makes the result useful.

Attribution of cardiac injury finally requires consideration of ischemia, valve disease, genetic cardiomyopathies, infection, and drug toxicity. Elevated troponin during a hypereosinophilic crisis may support a causal link, but demonstration of coronary stenosis or a different myocarditis changes the reasoning. In cases without histologic proof, convergence of course, imaging, and treatment response may make the diagnosis clinically convincing without converting it into pathologic certainty. The final report should make explicit the systemic subtype, degree of cardiac attribution, active components, and residual damage.

Treatment and prognosis

Acute cardiac injury is an indication for prompt treatment regardless of formal completion of a temporal criterion. In appropriate forms, glucocorticoids rapidly reduce eosinophilic activity; in organ-threatening presentations, specialist recommendations include prednisone around 1 mg/kg/day, optionally preceded by intravenous methylprednisolone in the most severe cases. Dose, route, and duration depend on severity, subtype, and response. Useful diagnostic samples should be obtained before treatment when possible, but without compromising stabilization of a patient in shock or with life-threatening arrhythmias.

Management of infectious risk is part of causal treatment. In patients with compatible exposure, investigation and possible treatment for Strongyloides should be coordinated with urgent immunosuppression; passively waiting for every result may be inappropriate when probability is significant and the heart is threatened. In drug reactions the responsible agent is withdrawn, whereas infections, vasculitides, and neoplasms require specific treatment. A good initial corticosteroid response does not eliminate the need to complete this pathway because it may temporarily control biologically different conditions.

In sensitive neoplasms, imatinib acts on the proliferative mechanism. Diseases with FIP1L1::PDGFRA may respond to lower doses than those used for other fusions; specialist regimens commonly use 100 mg/day for PDGFRA and 400 mg/day for PDGFRB, with adjustments according to entity, response, and tolerance. These regimens cannot be transferred to all clonal eosinophilias. In the presence of cardiac injury, initiation requires close monitoring and may include corticosteroid coverage. Efficacy should also be documented at the molecular level when a marker is available, avoiding equating normalization of the blood count with eradication of the clone.

For maintenance control of appropriate syndromes, mepolizumab provides an option directed against interleukin 5. In the randomized phase III trial in patients with FIP1L1::PDGFRA-negative HES, 300 mg subcutaneously every four weeks, added to existing therapy, reduced over 32 weeks the proportion of patients with at least one flare or withdrawal from the study: 28% versus 56% with placebo. The outcome concerns syndrome control in the studied population; it does not demonstrate regression of fibrosis or efficacy as a substitute for urgent treatment of fulminant myocarditis. Patient selection and the objective of reducing corticosteroid burden remain essential.

Hydroxyurea, interferon alfa, and other strategies may be considered in specific refractory forms or when targeted options are inappropriate, with monitoring of toxicity and organ response. In the lymphocytic variant, reducing eosinophils does not necessarily mean controlling the aberrant T-cell population; in neoplasms with other fusions, treatment may require different drugs and hematologic pathways. Therapies directed against eosinophils and their signals should therefore be placed within disease classification without presenting all biologics or cytoreductive drugs as equivalent alternatives.

Cardiac treatment proceeds in parallel. Diuretics and heart-failure measures are adapted to congestion, systolic function, pressure, and perfusion; in restrictive physiology, excessive depletion may abruptly reduce output. Arrhythmias require specific management, while shock may require temporary support in an experienced center. Control of eosinophilia does not replace these measures: an injury already produced continues to generate hemodynamic consequences even when the biologic stimulus is rapidly suppressed.

Anticoagulation is considered for intracardiac thrombus, embolization attributable to a cardiac source, and other defined indications. Hypereosinophilia alone, without thrombosis or additional clinical elements, is not a universal indication. Choice accounts for renal function, bleeding, interactions, and procedures; duration depends on thrombus evolution and persistence of risk, not merely on the blood count. Recurrence requires verification of drug exposure, disease activity, and the nature of the event before concluding that the entire antithrombotic strategy has failed.

When obliteration and regurgitation due to fibrosis predominate, surgical assessment considers endocardiectomy and valve treatment. Surgery addresses mechanical damage and should be coordinated with control of the cause because persistent systemic activity may compromise the result. In advanced heart failure, mechanical support and transplantation require assessment of anatomy, extracardiac complications, and hematologic prognosis. A normalized count is not enough to define operability, just as extensive scar does not automatically exclude every possibility of benefit.

Remission should be described on separate levels: clinical, hematologic, molecular when relevant, and cardiac. Follow-up monitors symptoms and previously involved organs, eosinophil count, markers of the specific disease, and toxicity; ECG, biomarkers, and imaging are repeated according to risk and evolution. Discontinuation or reduction of targeted therapy requires a specialist surveillance plan because molecular relapse may precede hematologic relapse. From the cardiac perspective, persistent fibrosis or regurgitation requires continuity of care even in complete systemic remission.

Complications

Thrombosis may involve the heart, arteries, and veins and produce recurrent events while procoagulant mechanisms remain active. Neurologic or visceral injury may become the main determinant of independence and prognosis even after eosinophilia is controlled. In patients with cerebral embolism, infarct extent and hemorrhagic transformation affect antithrombotic treatment; in cases with multiple thromboses, investigation of the source and systemic factors should remain open. Coexistence of bleeding and thrombotic risk requires reassessment rather than automatic application of a single rule.

Cardiac injury may consolidate into restrictive cardiomyopathy, with chronic congestion, valvular insufficiency, and reduced ability to increase output. Arrhythmias and conduction disturbances may complicate both inflammation and scar. When the heart remains dysfunctional after remission, the burden of residual damage should be recognized without automatically intensifying eosinophil-directed treatment. Conversely, new injury biomarkers or rapid worsening require investigation for relapse, ischemia, infection, and other acute causes.

Complications of systemic disease include pulmonary, cutaneous, neurologic, and gastrointestinal injury in addition to the risks intrinsic to clonal forms. In the lymphocytic variant, changes in lymphadenopathy, constitutional symptoms, or the T-cell population require reassessment; in myeloid neoplasms, surveillance follows the risk of the specific entity. Cardiac stability does not make hematologic follow-up unnecessary, and stability of the blood count does not guarantee absence of injury in other organs.

Corticosteroids expose patients to infection and metabolic, bone, and muscle toxicity; cytoreductive agents, immunomodulators, and tyrosine-kinase inhibitors have specific profiles that may interfere with heart-failure treatment. Iatrogenic myopathy or anemia may reduce functional capacity without indicating cardiac progression, while fluid retention and drug interactions may favor decompensation or bleeding. Assessment of complications should therefore distinguish activity, damage, and toxicity, maintaining a shared plan for treatment reduction, procedures, and relapse control.