The patient was a middle-aged woman with a history of anxiety, chronic bronchitis, asthma, hypertension, and gastroesophageal reflux disease who had a recent history of cough, fever, sinusitis and chills, for which she tested negative for COVID-19 (coronavirus disease 2019) and was treated with antibiotics. She was admitted to an outside hospital after a partial recovery with shortness of breath and leg weakness, which progressed to cardiogenic shock, non–ST-elevation myocardial infarction, acute kidney injury, and respiratory distress. She was transferred to a medical center and was found to have lactic acidosis and hypotension. Left heart catheterization showed normal coronary arteries and coronary flow with a left ventricular end-diastolic pressure of 26 mm Hg. A heart biopsy was performed, which showed eosinophilic myocarditis. She was treated with venoarterial extracorporeal membrane oxygenation and pressors but died 3 days after biopsy from cardiogenic shock. Peripheral eosinophil count ranged from 3 %to 10%. Serum antineutrophil cytoplasm antibody (ANCA) was negative.
The heart biopsy demonstrated interstitial eosinophils, granulomatous reaction to eosinophil breakdown products, thrombosis, and myocyte necrosis (Fig. 1). A diagnosis of necrotizing eosinophilic myocarditis was made. The patient did not survive long enough in order for a definitive clinical diagnosis to be made. However, given the history of asthma and sinusitis, eosinophilic granulomatosis with polyangiitis (EGPA) was considered the most likely cause. The presence of a granulomatous reaction around eosinophil breakdown products was thought to be nonspecific and not necessarily a manifestation of extravascular EGPA.
DIFFERENTIAL DIAGNOSIS OF EOSINOPHILIC MYOCARDITIS
The differential diagnosis of eosinophilic myocarditis includes hypersensitivity myocarditis, EGPA, and Loeffler endomyocarditis (hypereosinophilic syndrome [HES]). The distinction among these entities is generally made based on clinical grounds. A meta-analysis of 179 cases of pathologically documented eosinophilic myocarditis was recently published with emphasis on clinical findings.1 In this study, a definite cause was not identified in 36% of patients (as the case in this case report); 34% of patients had hypersensitivity myocarditis traced to a drug, and EGPA accounted for 13%, and HES 8%. In the remaining 9%, there were a variety of causes including parasites and malignancies.
Drug-induced eosinophilic myocarditis had a lower association with asthma and a higher incidence of fever than other types, and mural thrombi were rare. Peripheral eosinophilia was absent in 25% of cases and had the highest rates in HES and EGPA.
The most common form of eosinophilic myocarditis is caused by an allergic or hypersensitivity reaction, usually to drugs, most commonly antibiotics. It takes 2 forms: nonnecrotizing interstitial myocarditis characterized by sparse interstitial infiltrates of macrophages and eosinophils (Fig. 2), and necrotizing myocarditis with a dense inflammatory infiltrate (Fig. 3). Nonnecrotizing eosinophilic myocarditis (classic hypersensitivity myocarditis) is usually an incidental finding and is not seen on biopsy, because patients are asymptomatic and the infiltrates are usually sparse and focal. In contrast, necrotizing eosinophilic myocarditis triggered by a drug is a life-threatening condition that is treated with steroids and cardiac support, as well as cessation of the drug. Histologically documented necrotizing hypersensitivity myocarditis is rare; a meta-analysis of all reported cases in 2017 yielded 61 total cases.1 In contrast, the incidence of incidental hypersensitivity myocarditis diagnosed initially at autopsy is high, approximately 2% to 5%, depending on the population and history of medication use. In patients awaiting transplant, who are often on multiple medications, the incidence is 2% to 7%, as diagnosed histologically in the explanted heart or left ventricular apex removed at time of assist device insertion.2,3
Hypersensitivity is due to a delayed hypersensitivity reaction, and not a toxic effect of the offending drug. It has been postulated that modified collagen may be a trigger for eosinophilic reaction and degranulation.
Many drugs have been histologically associated with hypersensitivity myocarditis.4 In the meta-analysis of Brambatti et al,1 the drugs implicated included antibiotics (36.5%, mainly minocycline and β-lactam antibiotics), central nervous system agents (21.1%; primarily clozapine followed by carbamazepine), vaccines (7.7%), antitubercular agents (1.9%), and a variety of other drugs or undetermined drug (32.8%). Hypersensitivity myocarditis causing severe heart failure has been reported complicating Stevens-Johnson syndrome.5
In incidentally found hypersensitivity myocarditis at autopsy and explants, a specific drug is rarely identified because multiple medications have usually been administered. Hypersensitivity myocarditis is seen in 3% to 10% of cardiac explants2,3 and may develop while the patient is on a ventricular assist device.6 There has been an association with increased risk for rejection, but not with decreased graft survival.3
Patients with symptomatic hypersensitivity myocarditis often present with signs and symptoms of myocarditis, often with heart failure. In a study by Brambatti et al,1 54% of patients had fever, and only 64% peripheral eosinophilia. The symptoms generally occur within weeks after exposure to the offending agent and generally resolve after withdrawal of the drug. The cardiac symptoms may be accompanied by rash.7–10
Most patients respond to steroids and drug cessation.11–15 Irreversible heart failure may lead to cardiogenic shock and death or heart transplantation.5,16,17
Histologically, asymptomatic disease is manifested by sparse interstitial infiltrates of macrophages and lymphocytes, with occasional plasma cells; eosinophils may be difficult to find. At autopsy, infiltrates in other organs such as lung and liver are present in more than half of patients.4 Symptomatic myocarditis is characterized by cellular infiltrates of eosinophils, macrophages, and lymphocytes, with myocyte destruction and abundant myocyte necrosis. The degree of necrosis in some cases is such that there is a macrophage giant cell reaction that can resemble giant cell myocarditis.
EOSINOPHILIC GRANULOMATOSIS WITH POLYANGIITIS (CHURG-STRAUSS SYNDROME)
Overall, more than 50% of patients with EGPA have cardiac involvement, which is the major cause of morbidity and mortality.18 Eosinophilic myocarditis is the pathologic manifestation of heart involvement in endomyocardial biopsy. Cardiac involvement generally indicates that ANCA is negative in most cases. Cross-sectional studies have shown that two-thirds of patients show subclinical evidence of heart involvement, especially in younger patients who are ANCA-negative.19
Autopsy samples have a larger range of findings than biopsies of living patients (Fig. 4). There may be necrotizing granulomatous arteritis, usually of intramural muscular arteries, extravascular granulomas, and eosinophilic myocarditis,20 which may be the predominant finding even at autopsy.21 An explant study of antineutrophil antibody–negative patients with end-stage cardiomyopathy secondary to EGPA showed arteritis with fibrinoid necrosis in one-third, eosinophilic myocarditis in one-half (one with giant cells as in the case reported here), and the remainder with nonspecific findings including interstitial fibrosis.22 Endomyocardial biopsies performed prior to transplant showed eosinophilic infiltrates in 5 of 7, 2 of which had arteritis on biopsy.
Hypereosinophilic syndrome with primary myocardial involvement has been termed “Loeffler endocarditis” by pathologists, although clinicians often restrict the use of the term to chronic heart disease resulting in a restrictive cardiomyopathy (Table 1).
TABLE 1 -
Causes of Eosinophilic Myocarditis as Seen on Heart Biopsy
||May have some giant cells, but not typically around degranulated eosinophils; myocyte necrosis typical
||Most common with known etiology; may have delayed presentation after drug exposure
||May have some giant cells, but not typically around degranulated eosinophils; myocyte necrosis typical
||Often presents with heart involvement, second most common with known etiology
||Arteritis not usually seen on biopsy but is essentially diagnostic if present; myocyte necrosis variable and may be less than idiopathic and hypersensitivity
||Rare, often presents with chronic features of restrictive cardiomyopathy (Loeffler endomyocarditis)
||Mural thrombi in the ventricles more frequent, also vascular thrombi in biopsy specimen, myocyte necrosis often less than idiopathic, and hypersensitivity
|Malignancy, parasites (especially Toxocara)
||<10% of all cases of eosinophilic myocarditis
||Very rare, pathologic features variable
Degranulation of eosinophils within the eosinophil infiltrated myocardium causes myocardial necrosis from the release of toxic cationic proteins and mural thrombi, which can occur anywhere in the ventricles.
Hypereosinophilic syndrome can be described according to 3 categories: primary, secondary, and idiopathic forms. Primary HES is a myeloproliferative disorder associated with the chromosomal rearrangement resulting in PDGFRA fusion kinase. Secondary HES is a polyclonal expansion of eosinophils in reaction to lymphoproliferative disease. The idiopathic form is most common. It includes the “overlap” form, a phrase coined for cases that are limited to a specific organ, such as the heart or digestive tract.23
In the acute phase, mural thrombi (in either ventricle) occur in up to one-third of cases.1 In more chronic disease, the thrombus may encroach on mitral valve leaflets resulting in insufficiency. The tricuspid and pulmonic valves may also be affected by the mural thrombus.24 Endocardial changes typical of HES, as noted previously, may be seen in patients with malignancy (Fig. 5).
On endomyocardial biopsy, any biopsy with eosinophils should warrant a comment including HES in the differential diagnosis.25 The presence of mural or vascular thrombi is more a feature of HES than other causes of eosinophilic myocarditis.
1. Brambatti M, Matassini MV, Adler ED, et al. Eosinophilic myocarditis: characteristics, treatment, and outcomes. J Am Coll Cardiol
2. Gravanis MB, Hertzler GL, Franch RH, et al. Hypersensitivity myocarditis in heart transplant candidates. J Heart Lung Transplant
3. Kanai-Yoshizawa S, Sugiyama Kato T, Mancini D, et al. Hypersensitivity myocarditis and outcome after heart transplantation. J Heart Lung Transplant
4. Burke AP, Saenger J, Mullick F, et al. Hypersensitivity myocarditis. Arch Pathol Lab Med
5. Mitoff PR, Mesana TG, Mielniczuk LM, et al. Giant cell myocarditis in a patient with a spondyloarthropathy after a drug hypersensitivity reaction. Can J Cardiol
6. Pereira NL, Park SJ, Daly RC, et al. De novo development of eosinophilic myocarditis with left ventricular assist device support as bridge to transplant. Ann Thorac Surg
7. Zhang SN, He QX, Yang NB, et al. Isoniazid-induced drug rash with eosinophilia and systemic symptoms (DRESS) syndrome presenting as acute eosinophilic myocarditis. Intern Med
8. Park Y, Ahn SG, Ko A, et al. Hypersensitivity myocarditis confirmed by cardiac magnetic resonance imaging and endomyocardial biopsy. Korean J Intern Med
9. Pereira CM, Vaz M, Kotha S, et al. Dapsone hypersensitivity syndrome with myocarditis. J Assoc Physicians India
10. Kanno K, Sakai H, Yamada Y, et al. Drug-induced hypersensitivity syndrome due to minocycline complicated by severe myocarditis. J Dermatol
11. Li H, Dai Z, Wang B, et al. A case report of eosinophilic myocarditis and a review of the relevant literature. BMC Cardiovasc Disord
12. Allen SF, Godley RW, Evron JM, et al. Acute necrotizing eosinophilic myocarditis in a patient taking Garcinia cambogia
extract successfully treated with high-dose corticosteroids. Can J Cardiol
13. Eppenberger M, Hack D, Ammann P, et al. Acute eosinophilic myocarditis with dramatic response to steroid therapy: the central role of echocardiography in diagnosis and follow-up. Tex Heart Inst J
14. Aslan I, Fischer M, Laser KT, et al. Eosinophilic myocarditis in an adolescent: a case report and review of the literature. Cardiol Young
15. Komatsu T, Isoda K, Niida T, et al. A good outcome of a case with eosinophilic myocarditis. Intern Med
16. Jeremic I, Vujasinovic-Stupar N, Terzic T, et al. Fatal sulfasalazine-induced eosinophilic myocarditis in a patient with periodic fever syndrome. Med Princ Pract
17. Huston B, Froloff V, Mills K, et al. Death due to eosinophilic necrotizing myocarditis despite steroid treatment. Am J Forensic Med Pathol
18. Nadeau PL, Kumar A, O'Connor K, et al. Usefulness of cardiac resonance imaging in Churg-Strauss syndrome. J Cardiovasc Med (Hagerstown)
19. Vinit J, Bielefeld P, Muller G, et al. Heart involvement in Churg-Strauss syndrome: retrospective study in French Burgundy population in past 10 years. Eur J Intern Med
20. Sasaki A, Hasegawa M, Nakazato Y, et al. Allergic granulomatosis and angiitis (Churg-Strauss syndrome). Report of an autopsy case in a nonasthmatic patient. Acta Pathol Jpn
21. Setoguchi M, Okishige K, Sugiyama K, et al. Sudden cardiac death associated with Churg-Strauss syndrome. Circ J
22. Groh M, Masciocco G, Kirchner E, et al. Heart transplantation in patients with eosinophilic granulomatosis with polyangiitis (Churg-Strauss syndrome). J Heart Lung Transplant
23. Hsieh FH. Hypereosinophilic syndrome
. Ann Allergy Asthma Immunol
24. Garg A, Nanda NC, Sungur A, et al. Transthoracic echocardiographic detection of pulmonary valve involvement in Loeffler's endocarditis. Echocardiography
25. Kawagoshi H, Shimizu M, Suematsu T, et al. A case of eosinophilic heart disease diagnosed by endomyocardial biopsy findings. Jpn Heart J