Eosinophilic Myocarditis and Nonbacterial Thrombotic Endocarditis in a Patient With Immunoglobulin G4-Related Disease: A Case Report and Review

INTRODUCTION

Immunoglobulin G4-related disease (IgG4-RD) is an emerging systemic fibro-inflammatory disorder characterized by dense lymphoplasmacytic infiltrates rich in IgG4-positive plasma cells, storiform fibrosis, and, often, elevated serum IgG4 concentrations.¹⁾ Recent literature has highlighted the potential for severe cardiac manifestations, including myocarditis, in patients with IgG4-RD, which can lead to significant morbidity and mortality if not promptly diagnosed and treated.²⁾

Eosinophilic myocarditis is a specific form of myocarditis marked by myocardial infiltration of eosinophils, often associated with peripheral eosinophilia and hypereosinophilic syndrome.³⁾ This condition can rapidly progress to severe cardiac dysfunction. Nonbacterial thrombotic endocarditis (NBTE), characterized by the presence of sterile thrombi on heart valves, often occurs in the context of hypercoagulable states and systemic inflammatory conditions.⁴⁾ This case report and review aim to shed light on these rare but significant complications, emphasizing the need for heightened clinical awareness and comprehensive diagnostic evaluation to improve patient outcomes. We hypothesize a pathophysiological sequence in which IgG4-RD drives aberrant immune activation, leading to eosinophilic myocarditis, the resulting eosinophilia and systemic inflammation then promotes a hypercoagulable state that culminates in NBTE. Recognizing this stepwise progression is essential for timely diagnosis and appropriate immunosuppressive treatment.

CASE

A 35-year-old male patient presented to our hospital with chest pain and dyspnea, which had started 1 month prior. The patient had no history of regular medication use, including non-steroidal anti-inflammatory drugs, antibiotics, or any other agents known to cause drug-induced hypersensitivity eosinophilia, effectively excluding drug hypersensitivity as a causative factor. Initial blood tests at a local clinic showed an elevated troponin I level, and an electrocardiogram revealed ST segment depression, raising suspicion of myocardial infarction. The patient was transferred to another hospital where further evaluation, including blood tests and transthoracic echocardiography (TTE), was performed. At the time of evaluation at the second hospital, the patient’s blood tests revealed the following: white blood cell (WBC) count of 10,850/mm³ (with eosinophils at 48.2%), creatinine kinase (CK) at 242 U/L (normal range, 58–348 U/L), CK-MB at 7.6 ng/mL (normal range, 0–4.94 ng/mL), high-sensitivity troponin I at 7,457.4 pg/mL (normal range, 0–34.2 pg/mL), and B-type natriuretic peptide (BNP) at 566 pg/mL (normal range, 0–100 pg/mL). TTE showed a left ventricular ejection fraction (LVEF) of 31%, with global hypokinesia and a small amount of pericardial effusion. Coronary angiography was performed to rule out acute myocardial infarction, revealing no significant coronary artery stenosis (Supplementary Videos 1 and 2). Myocarditis was suspected based on elevated cardiac enzymes and reduced left ventricular function but endomyocardial biopsy (EMB) and cardiac magnetic resonance imaging were not performed at that stage. The patient was managed conservatively with initiation of guideline-directed medical therapy for heart failure with reduced ejection fraction. The dyspnea improved, and the patient was discharged after a week.

Despite initial improvement, the patient experienced persistent dyspnea and fever (body temperature 38.9°C), leading to readmission to the previous hospital. Follow-up echocardiography showed increased pericardial effusion, prompting pericardiocentesis, which yielded negative culture results on repeated testing (3 sets of blood cultures collected on separate days, all negative). Pericardial fluid analysis revealed a low WBC count (WBC 150/mm³, eosinophils 2%) with a lactate dehydrogenase of 120 IU/L and glucose of 85 mg/dL, consistent with a transudative pericardial effusion. The adenosine deaminase level was 35 IU/L (normal range, 8–19 IU/L). Although 35 IU/L is below the commonly cited cutoff of 40 IU/L, published data have reported relatively high sensitivity and specificity for tuberculosis pericarditis even at ≥35 IU/L. To definitively exclude tuberculosis pericarditis, additional testing was performed: acid-fast bacillus smear was negative, tuberculosis polymerase chain reaction of the pericardial fluid was negative, and QuantiFERON-TB Gold test was negative, collectively ruling out tuberculosis pericarditis. Subsequently, the patient developed abdominal pain and memory impairment. Physical examination revealed Janeway lesions on both plantar surfaces of the toes. Imaging studies identified splenic and small cerebral infarctions (Supplementary Figure 1). TTE revealed vegetation on the aortic valve, raising suspicion of infective endocarditis, leading to transfer to our hospital for surgical intervention.

Upon transfer, blood tests showed high-sensitivity troponin T at 1,980 ng/L (normal range, <14 ng/L), N-terminal-pro-B-type natriuretic peptide (NT-proBNP) at 9,259 pg/mL (normal range, <115 pg/mL), and a WBC of 26,130/mm³ (with eosinophils at 51%), suggesting eosinophilic myocarditis. Additional laboratory workup for eosinophilia revealed an elevated serum immunoglobulin E level of 3,740 kU/L (normal range, 0–113 ku/L); however, the eosinophil cationic protein level was not measured. Previous blood tests from the other hospital showed a WBC count of 23,370/mm³ (with eosinophils at 26.7%) 10 days prior and 21,500/mm³ (with eosinophils at 48.0%) 3 days prior. TTE revealed severe left ventricular systolic dysfunction with an LVEF of 25% and global hypokinesia. A 1.2×0.8 cm mobile echogenic mass was also observed on the aortic valve, confirmed by transesophageal echocardiography, with no other valvular abnormalities (Figure 1A, Supplementary Videos 3 and 4).

Figure 1. Serial echocardiographic findings of aortic valve vegetation. (A) Baseline echocardiography. Vegetation is observed on the aortic valve (indicated by an arrow). (B) Echocardiography conducted 2 months later. Following anticoagulation therapy, the aortic valve vegetation has completely resolved (indicated by an arrow).

An EMB with hematoxylin and eosin staining confirmed eosinophilic myocarditis, showing extensive eosinophilic infiltration within myocardial fibers (Figure 2A). Additional immunohistochemical staining demonstrated CD3+ T lymphocytes and CD68+ macrophages interspersed among the eosinophilic infiltrate, consistent with the acute stage of eosinophilic myocarditis. Blood culture was negative, and considering the eosinophilic myocarditis, the vegetation on the aortic valve was diagnosed as NBTE rather than infective endocarditis. To identify the cause of eosinophilia and eosinophilic myocarditis, tests for parasitic infections and hematologic disorder were conducted. Bone marrow examination was not performed as the clinical presentation was highly suggestive of IgG4-RD, and the invasive procedure was deemed potentially risky given the patient’s unstable clinical status. Instead, other causes of hypereosinophilia were systemically evaluated. Evaluation for eosinophilic granulomatous polyangiitis (EGPA) was performed: anti-neutrophil cytoplasmic antibody (ANCA) serology (anti-MPO and anti-PR3) was negative, and clinical features (no asthma, no sino-nasal disease, no vasculopathy) were inconsistent with EGPA. Chronic eosinophilic leukemia and hypereosinophilic syndrome were indirectly excluded by confirming the absence of the FIP1L1-PDGFRA fusion gene via peripheral blood analysis, in lieu of a bone marrow study. The test for IgG4-RD showed elevated total IgG at 1,990 mg/dL (normal range, 680–1,620 mg/dL) and IgG4 at 492 mg/dL (normal range, 3–201 mg/dL). Additionally, EMB showed an IgG4/IgG ratio >0.4 with >10 IgG4-secreting plasma cells per high-power field (HPF) (Figure 2B). Cardiac magnetic resonance imaging revealed multifocal subendocardial and transmural late gadolinium enhancement consistent with myocarditis, while T2-weighted imaging demonstrated myocardial edema in the corresponding lesions (Figure 3). Although mapping parameters such as native T1 levels and extracellular volume fraction were regrettably not measured, the collective evidence from multimodality imaging and clinical evaluation remained robust, confirming definite IgG4-RD and the diagnosis of myocarditis.

Figure 2. Histopathology of the heart of interventricular septum. (A) Hematoxylin and eosin stain shows extensive eosinophilic infiltration within the muscle fiber. (B) Immunoglobulin G4 immunohistochemistry demonstrates staining of more than 10 plasma cells.

Figure 3. Heart magnetic resonance imaging of myocarditis. (A) Subendocardial late gadolinium enhancement is observed in the left ventricular inferoseptal wall. (B) Transmural endocardial late gadolinium enhancement is observed in the apicolateral wall.

Following the EMB, intravenous methylprednisolone 60 mg was administered every 12 hours for 2 days, resulting in reduction of eosinophilia to a WBC count of 28,670/mm³ (with eosinophils at 2.0%). Oral prednisolone 20 mg twice daily was initiated, intravenous methylprednisolone was discontinued, and warfarin was used for anticoagulation to manage NBTE. Ten days after starting corticosteroid therapy, follow-up blood tests showed a WBC count of 17,310 /mm³ (with eosinophils at 0.2%), total IgG 1,740 mg/dL, IgG4 392 mg/dL, high-sensitivity troponin T 105 ng/L, and NT-proBNP 2,679 pg/mL. TTE showed a slight improvement in LVEF to 32%; however, NBTE on the aortic valve remained unchanged. The patient was discharged with plans for outpatient follow-up.

Two months post-discharge, the patient’s blood tests showed improvement with a WBC count of 6,280/mm³ (eosinophils 1.9%), high-sensitivity troponin T 24 ng/L, and NT-proBNP 920.2 pg/mL. TTE showed LVEF at 35%, and the NBTE on the aortic valve had resolved (Figure 1B). Anticoagulation therapy was discontinued, and the steroid regimen was tapered, maintaining prednisolone 5 mg once daily for underlying IgG4-RD. The patient continued guideline-directed medical therapy for heart failure with reduced ejection fraction (Table 1).

Table 1. Timeline of key clinical events and treatment response.

Time point	Hospital	Clinical events and key findings	Management
Month 0 (symptom onset)	Local clinic	Elevated troponin I; ECG: ST depression
~Month 0 (1st admission)	Second hospital	WBC 10,850 (Eos 48.2%); hs-TnI 7,457 pg/mL; BNP 566 pg/mL; TTE: LVEF 31%, global hypokinesia; CAG: no stenosis	Conservative; HFrEF-directed GDMT; discharged after 1 week
~Week 2 (2nd admission)	Second hospital	Dyspnea, fever 38.9℃; increased pericardial effusion; pericardiocentesis: ADA 35 IU/L, negative cultures/AFB/PCR/QFT; Janeway lesions; splenic & cerebral infarction; TTE: aortic vegetation	Transferred to our hospital
~Week 3 (3rd admission)	Our hospital	hs-TnT 1,980 ng/L; NT-proBNP 9,259 pg/mL; WBC 26,130 (Eos 51%); TTE: LVEF 25%, aortic vegetation 1.2×0.8 cm (confirmed by TEE); EMB: eosinophilic myocarditis, IgG4/IgG >0.4, >10 IgG4+ cells/HPF; IgG4 492 mg/dL; CMR: LGE + T2 edema	IV methylprednisolone 60 mg q12h ×2 days; then oral prednisolone 20 mg BID; warfarin anticoagulation; continued HFrEF GDMT
~Week 5 (10 days after steroids)	Our hospital	WBC 17,310 (Eos 0.2%); IgG4 392 mg/dL; hs-TnT 105 ng/L; NT-proBNP 2,679 pg/mL; TTE: LVEF 32%; NBTE unchanged	Discharged; outpatient follow-up planned
Month 2 (follow-up)	Our hospital outpatient clinic	WBC 6,280 (Eos 1.9%); hs-TnT 24 ng/L; NT-proBNP 920 pg/mL; TTE: LVEF 35%, NBTE resolved; splenic infarction resolved; memory improved	Anticoagulation discontinued; prednisolone tapered to 5 mg daily; continued GDMT

ECG = electrocardiogram; WBC = white blood cell count; hs-TnI/T = high-sensitivity troponin I/T; BNP = B-type natriuretic peptide; TTE = transthoracic echocardiography; LVEF = left ventricular ejection fraction; CAG = coronary angiography; HFrEF = heart failure with reduced ejection fraction; GDMT = guideline-directed medical therapy; ADA = adenosine deaminase; AFB = acid-fast bacillus; PCR = polymerase chain reaction; QFT = QuantiFERON; NT-proBNP = N-terminal pro-B-type natriuretic peptide; TEE = transesophageal echocardiography; EMB = endomyocardial biopsy; IgG = immunoglobulin; HPF = high-power field; CMR = cardiac magnetic resonance; LGE = late gadolinium enhancement; IV = intravenous; q12h = every 12 hours; BID = twice a day; Eos = eosinophils; NBTE = nonbacterial thrombotic endocarditis.

DISCUSSION

Eosinophilic myocarditis is a disease characterized by the infiltration of eosinophils into the myocardium. The causes of eosinophilic myocarditis include drug-induced hypersensitivity; immune-related diseases, such as eosinophilic granulomatosis with polyangiitis; myeloproliferative diseases, such as chronic eosinophilic leukemia; idiopathic conditions, such as complex hypereosinophilic syndrome; and IgG4-RD, which can also cause eosinophilia.¹⁾ In cases such as the present one, eosinophilia in IgG4-RD arises due to an immunologic trigger that leads to the overexpression of interleukin 4, 5, 10, and 13, and transforming growth factor β. This overexpression triggers immune reactions involving regulatory T cells, particularly type 2 helper T cells, resulting in the characteristic eosinophilia and elevated serum IgG4 associated with IgG4-RD.

Eosinophilia also possesses inflammatory, oxidative, and prothrombotic components that increase the risk of thrombosis.⁵⁾ This thrombosis can manifest as venous or arterial thrombosis; however, it can also appear through intracardiac cavity thrombus, resulting from Loeffler endocarditis due to chronic myocardium fibrosis, or as valve thrombosis.⁶⁾

Given the diverse causes of eosinophilia and the different treatment approaches required for each cause, identifying the underlying reason for eosinophilia is crucial. In the differential diagnosis, several conditions associated with hypereosinophilia were systematically evaluated and excluded. Drug-induced hypersensitivity was ruled out based on the absence of any culprit medication and characteristic clinical features such as rash, fever, or lymphadenopathy. EGPA was also considered unlikely given the absence of asthma and sino-nasal disease, along with negative ANCA serology. Primary hematologic malignancies, including chronic eosinophilic leukemia, were excluded by the absence of clonal eosinophilic proliferation and the FIP1L1-PDGFRA fusion gene. Additionally, hypereosinophilic syndrome was ruled out as the systemic manifestations were attributable to a secondary identifiable cause (IgG4-RD) rather than idiopathic eosinophilia exceeding 6 months without an underlying etiology.

Ultimately, the 2019 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) classification criteria for IgG4-RD require histopathological evidence of dense lymphoplasmacytic infiltration with storiform fibrosis, an IgG4-positive plasma cell count exceeding 10 cells per HPF, and an IgG4/IgG ratio greater than 0.4 in the biopsy specimen, supported by elevated serum IgG4 above 135 mg/dL. In the present case, all these criteria were met: EMB demonstrated >10 IgG4-positive plasma cells per HPF with IgG4/IgG ratio >0.4, and serum IgG4 was markedly elevated at 492 mg/dL.

In IgG4-RD, the first line of treatment is glucocorticoids. Response to glucocorticoids is usually seen within days or weeks, and remission can be achieved within months in most patients with IgG4-RD. Slow tapering of the glucocorticoids should begin 2–4 weeks after induction of therapy and last for 3–6 months; nonetheless, approximately 40% of patients either fail to achieve complete remission or experience a relapse within a year of diagnosis.⁷⁾ In such cases, second-line treatments include immunosuppressive agents, such as cyclophosphamide, mycophenolate mofetil, and azathioprine.⁸⁾ If these treatments fail, medications such as rituximab, a CD20 targeting agent, or abatacept, a cytotoxic T-lymphocyte-associated antigen 4 agonist, may be used.⁹⁾

Additionally, it is crucial to first determine whether NBTE is due to bacterial vegetation. The presence of infective endocarditis is assessed based on blood cultures and imaging studies to confirm definite infective endocarditis. NBTE can occur in advanced-stage malignancy or autoimmune diseases, such as antiphospholipid syndrome or systemic lupus erythematosus. Eosinophilia can also cause NBTE, though its prevalence is very rare. NBTE occurs in these conditions due to a hypercoagulation state.

Only a small number of case reports have described the co-occurrence of eosinophilic myocarditis with IgG4-RD or of eosinophilic myocarditis with NBTE, and none, to our knowledge, have reported all 3 conditions simultaneously. Woo et al.²⁾ described a case of IgG4-RD presenting with myocarditis and eosinophilic infiltration, demonstrating that IgG4-RD can directly drive eosinophilic cardiac involvement. Several reports of Loeffler endocarditis and eosinophilia-associated valve thrombosis have illustrated the prothrombotic potential of eosinophilia, providing indirect support for the NBTE observed in the present case. The simultaneous presence of IgG4-RD, eosinophilic myocarditis, and NBTE in our patient therefore represents an exceptionally rare clinical phenotype.

Therefore, it is necessary to check for the involvement of other organs in patients with eosinophilia, identify the cause of eosinophilia, and determine whether a thrombotic event is associated with eosinophilia. In conclusion, this case demonstrates that IgG4-RD can also cause eosinophilia and eosinophilic myocarditis, and such conditions can be accompanied by thrombotic events. Thus, identifying the cause of eosinophilia is paramount, as treatment strategies differ depending on the underlying cause.

SUPPLEMENTARY MATERIALS

Supplementary Figure 1

Clinical photography of Janeway lesions on the foot.

Supplementary Video 1

Left coronary angiography.

Supplementary Video 2

Right coronary angiography.

Supplementary Video 3

Transesophageal echocardiography showing aortic valve vegetation (short axis view).

Supplementary Video 4

Transesophageal echocardiography showing aortic valve vegetation (long axis view).