30 - 268 Acute and Chronic Myocarditis

268 Acute and Chronic Myocarditis

Neal K. Lakdawala,

Lynne Warner Stevenson, Joseph Loscalzo

Acute and Chronic

Myocarditis Myocarditis is defined as inflammation of the heart muscle, which may present acutely, subacutely, or insidiously. Outcomes can include resolution, relapsing course, or progression to chronic cardiomyopathy. Myocarditis is generally considered acute when presenting with less than a month of symptoms, often only a few days. The profile of acute myocarditis is changing as the use of sensitive troponin assays and car­ diac magnetic resonance imaging (MRI) are increasing recognition of mild cases with good outcomes. Acute myocarditis is most often attrib­ uted to active infection but can be caused by multiple types of nonin­ fectious inflammation, such as sarcoidosis, giant cell and eosinophilic myocarditis, and systemic autoimmune disease or immunotherapies. It is uncertain as to how often chronic myocardial inflammation contrib­ utes to dilated cardiomyopathy. As with other cardiomyopathies, there is increasing implication of genetic predisposition to inflammatory responses and of independent pathogenic variants as primary cause of myocyte pathology, with almost one in five patients with inflamma­ tory myocarditis harboring genetic variants considered pathogenic for cardiomyopathy. MODELS OF INFECTIOUS MYOCARDITIS Infectious agents can injure the myocardium through direct invasion with disruption of normal cellular processes and through activation of different phases of immune responses with or without persistent infection. Although myocarditis has been reported with most types of infectious agents, it is most often associated with viruses and the pro­ tozoal disease Trypanosoma cruzi (Chagas’ disease). The pathogenesis of viral myocarditis has been traditionally portrayed in three phases, arising from extensive study in murine models of enteroviruses, with much less known about human infection. As defined by infection with coxsackie B virus, phase I includes direct invasion of the myocar­ dium. Cellular entry may be enhanced by some genetic variants of the coxsackie/adenovirus receptors. The enteroviral protease A degrades the myocyte structural protein dystrophin and interacts with other proteins to induce apoptosis and interfere with autophagy. Direct myocardial invasion with entry of the viral genome has been shown to occur with the enteroviruses, HIV, and dengue virus. However, most common viruses do not appear to inflict cardiac damage directly. Early cardiac effects may result primarily from cytokine storm and other nonspecific immune responses leading to myocardial depression. This initial immune response appears to be crucial to recovery, as early immunosuppression increased viral replication and worsened cardiac injury in animal models. The second phase is host response to infection in recognition of common antigenic patterns, triggering macrophage activation and expansion of specific T- and B-cell populations. Myocarditis due to this phase of immune response without direct viral-mediated injury is considered to be virally “triggered” myocarditis, implicated with respi­ ratory viruses including adenovirus, influenza, and COVID-19. Molec­ ular mimicry between viral and cardiac antigens may be the cause of some cases of myocardial injury caused by autoreactive T cells in the absence of ongoing viral infection. There is increasing study of the role of viral “hijacking” of cellular machinery to produce extracellular vesicles containing viral RNA that may then gain protected entry into other cells. A third phase of progression to chronic cardiomyopathy has been demonstrated in animal models of enteroviruses. However, myocarditis is rarely due to enteroviruses in adults. There are limited data providing direct links between chronic cardiomyopathy and previous viral infection (see “Chronic Myocarditis and Cardiomyopathy” below). The MRI pattern of intramyocardial late gadolinium enhancement

in chronic cardiomyopathy was previously considered as evidence of prior viral myocarditis, but this pattern is now recognized to be com­ mon in genetic cardiomyopathy as well.

CHAPTER 268 ACUTE MYOCARDITIS ■ ■PRESENTATION AND DIAGNOSIS Acute myocarditis in adults typically occurs between 30 and 45 years and is more common in men than women, who tend to present at over 45 years of age. Prodromal symptoms typical of influenza, gas­ troenteritis, or upper respiratory infection may have occurred days to a few weeks earlier. Fever is present in over half of cases. Chest pain is the most common symptom of acute myocarditis, occurring in

80% of patients, who also often present with dyspnea or arrhythmias. Arrhythmias can manifest as palpitations or syncope but can also cause sudden cardiac death, as in an autopsy series in which inflammatory myocarditis was diagnosed 3–10% of previously healthy young adults. At presentation, about one-fourth of acute myocarditis cases include left ventricular ejection fraction (LVEF) <0.50, ventricular arrhyth­ mias, or clinical shock, which is reported in ~3–9% of cases and termed fulminant myocarditis. Acute and Chronic Myocarditis Viral titers are not often helpful to guide initial therapy of acute myocarditis. Respiratory viral panels can confirm recent infection with influenza and adenovirus, most implicated in acute myocarditis in adults with a viral prodrome. Finding of COVID-19 may alert to extracardiac involvement. Specific infectious serologies for HIV, Chagas’ disease, cytomegalovirus, dengue fever, and Lyme disease and serologies for systemic autoimmune disease should be sent in selected patients. Eosinophil counts should be routinely checked because hype­ reosinophilia is present in most cases of eosinophilic myocarditis. The common definition for “probable myocarditis” includes new appearance of least one symptom listed above and at least one sup­ porting finding, which can be elevated troponin levels, consistent ECG findings, or imaging findings of decreased LVEF or obvious wall motion abnormality. Other laboratory findings may include elevated creatine phosphokinase levels, indicating cardiac or skeletal muscle involvement, and elevated C-reactive protein levels. The most com­ mon electrocardiogram (ECG) findings are ST elevation suggesting infarction, but also include conduction block, tachyarrhythmias, and nonspecific ST-T changes. Because the combination of chest pain, ECG changes, and elevated troponin is typical of both acute myocarditis and myocardial infarction, the first step in the differential diagnosis is often coronary artery imaging. Subsequent diagnosis of acute myocar­ ditis has been reported in up to one-third of patients with myocardial infarction and nonobstructed coronary arteries (MINOCA). “Definite myocarditis” previously required positive biopsy findings of typical inflammatory lymphocytic infiltrate on myocardial biopsy (Fig. 268-1). Newer techniques of immunohistochemistry may increase sensitiv­ ity of biopsies and advance our understanding of the inflammatory cell populations. However, the diagnosis can now be confirmed noninva­ sively when typical symptomatic presentation is accompanied by com­ bination of elevated troponin levels and consistent findings on MRI, which include evidence of both myocardial edema and nonischemic myocardial injury in the absence of other known cause (Fig. 268-2). Although cardiac MRI may often be sufficient to diagnose acute myocarditis, endomyocardial biopsy should be performed when ven­ tricular arrhythmias, conduction block, elevated eosinophil count, or evidence of systemic autoimmune disease suggests causes of acute myocarditis with specific implications for therapy and prognosis, as discussed below in “Noninfective Inflammatory Myocarditis.” ■ ■THERAPY AND OUTCOMES The prognosis is good for “uncomplicated” acute myocarditis, present­ ing with LVEF ≥0.50 without ventricular arrhythmias or circulatory shock. In the largest series, this low-risk group accounted for about three-fourths of patients with acute myocarditis, in which spontaneous recovery was common without specific therapies other than diuretics and nonsteroidal analgesics for chest pain. Almost all of these patients

PART 6 Disorders of the Cardiovascular System FIGURE 268-1  Acute myocarditis. Microscopic image of an endomyocardial biopsy showing massive infiltration with mononuclear cells and occasional eosinophils associated with clear myocyte damage. The myocyte nuclei are enlarged and reactive. Such extensive involvement of the myocardium would lead to extensive replacement fibrosis even if the inflammatory response could be suppressed. Hematoxylin and eosin–stained section, 200× original magnification. (Image courtesy of Robert Padera, MD, PhD, Department of Pathology, Brigham and Women’s Hospital, Boston.) still had LVEF ≥0.50 on follow-up imaging and survived without trans­ plantation at 5 years of follow-up. In the one-fourth of patients with complicated presentations, ~18% died or underwent heart transplanta­ tion in the next 5 years. Patients with reduced LVEF are generally dis­ charged on the recommended therapies for heart failure with reduced ejection fraction (HFrEF). There are no data supporting routine immu­ nosuppression in patients with acute myocarditis with presumed viral etiology. Fulminant myocarditis requiring high-dose inotropic therapy or mechanical circulatory support occurs in 3–9% of acute myocarditis, with early outcomes of refractory cardiogenic shock but recovery to near-normal LVEF in >50% of affected patients. CHRONIC MYOCARDITIS AND CARDIOMYOPATHY Patients with reduced LVEF at the time of presentation with acute myocarditis are at higher risk of developing chronic cardiomyopathy. When the LVEF does not resolve in the initial months after diagnosis, FIGURE 268-2  Magnetic resonance image of myocarditis showing the typical midwall location (arrow) for late gadolinium enhancement from cardiac inflammation and scarring. (Image courtesy of Ron Blankstein, MD, and Marcelo Di Carli, MD, Division of Nuclear Medicine, Brigham and Women’s Hospital, Boston.)

particularly with persistent elevation of troponin, cardiac magnetic resonance imaging (CMRi) or endomyocardial biopsy may reveal ongoing inflammation and developing fibrosis. Evaluation should again be considered for specific etiologies of myocarditis including noninfectious causes of inflammatory disease and for other causes of dilated cardiomyopathy (Chap. 266, Table 266-2). Evidence of chronic inflammatory disease may also be found in some patients presenting with gradual onset of dilated cardio­ myopathy without prior acute myocarditis. Immunosuppression for nonspecific lymphocytic myocarditis in the chronic setting has been tried in multiple series but has not demonstrated convinc­ ing benefit. Additional findings of anticardiac antibodies and viral genome fragments are of uncertain significance, as they could be either causes of ongoing cardiac injury or neutral bystanders. Assay for anticardiac antibodies and viral genome analysis in chronic cardiomyopathy continue to be investigated but are not in routine clinical use. In the past, dilated cardiomyopathy without other cause was gener­ ally assumed to result from previously unrecognized viral myocarditis, in part because CMRi patterns of fibrosis suggested prior inflamma­ tion. However very similar patterns are typical of genetic cardiomy­ opathies. The expanding access to genetic testing reveals that 20–40% of patients with dilated cardiomyopathy have pathogenic variants suf­ ficient to cause their cardiomyopathy, and the prevalence and nature of these variants are similar in acute myocarditis. Mapping of the interconnections between genetic predisposition, inflammation, and cardiomyopathy is evolving rapidly. ■ ■SPECIFIC CAUSES OF INFECTIOUS MYOCARDITIS The first viruses implicated in infectious myocarditis were the picor­ navirus family of RNA viruses, principally the enteroviruses, coxsackie virus, echovirus, and poliovirus. Influenza, another RNA virus, is implicated with varying frequency every winter and spring as epitopes change. Of the DNA viruses, adenovirus, vaccinia, and the herpes­ viruses (varicella-zoster virus, cytomegalovirus, Epstein-Barr virus, and human herpesvirus 6 [HHV6]) can cause myocarditis but also occur commonly in the healthy population. Polymerase chain reaction (PCR) detects viral genomes in some patients with dilated cardio­ myopathy and persistent lymphocytic infiltrates defined as “chronic myocarditis,” such as the DNA viruses parvovirus B19 and HHV6, but these are also found in normal “control” hearts, so their contribution to chronic cardiomyopathy is uncertain. COVID-19 is an RNA virus for which infection was associated with substantial rates of cardiac involvement in patients hospitalized with severe disease, including acute coronary syndrome, thromboemboli, pericarditis, and myocarditis, for which the diagnosis was limited to clinical findings and elevated enzymes rather than demonstration of lymphocytic infiltrates, as biopsies were rarely performed. From broad population data, the rate of myocarditis during the pandemic was estimated at 150/100,000 people, a 15-fold increase over pre-COVID incidence. As for most causes of viral myocarditis, the most common adult patients are young men under the age of 40. As for other viruses, there are multiple potential mechanisms of cardiac injury. Viral par­ ticles were identified by PCR in myocardial samples but in low levels at which pathogenicity was uncertain. Other mechanisms include cytokine storm, activated T cells and macrophages, and potential molecular mimicry by which autoimmune reactions may be triggered, as for some animal models of myocarditis with other viruses. As with multiple other viruses, extracellular vesicles containing viral RNA have been identified in myocardium during COVID infection, potentially promoting protected entry into other cells for further injury. In addi­ tion, the endothelium is also vulnerable, leading to prothrombotic endotheliopathy that may contribute to myocardial ischemia and stroke, along with the systemic coagulopathy. HIV was associated with a dilated cardiomyopathy incidence of 1–2%, but this is declining with the availability of highly active anti­ retroviral therapy. Cardiomyopathy in HIV may also result from other associated viruses, such as cytomegalovirus and hepatitis C. Antiviral drugs to treat chronic HIV can cause cardiomyopathy, both directly

and through drug hypersensitivity. The clinical picture may be com­ plicated by pericardial effusions and pulmonary hypertension. There is an increased frequency of lymphocytic myocarditis found at autopsy; HIV viral particles have been demonstrated in the myocardium in some cases. For any serious infection, the systemic inflammatory response can cause nonspecific depression of cardiac function, which is generally reversible if the patient survives. Other viruses implicated specifically in myocarditis include mumps, respiratory syncytial virus, the arbovi­ ruses (dengue fever and yellow fever), and arenaviruses (Lassa fever). ■ ■OTHER INFECTIOUS CAUSES Parasitic Myocarditis  Chagas’ disease is the third most com­ mon parasitic infection in the world and the most common infective cause of cardiomyopathy. The protozoan T. cruzi is transmitted by the bite of the reduviid bug, endemic in the rural areas of South and Central America. Transmission can also occur through blood transfu­ sion, organ donation, from mother to fetus, and occasionally orally. Programs to eradicate the insect vector have decreased the global prevalence from about 18 million in 1990 to 6 million, with the highest number of cases in Brazil and Argentina and increased prevalence also in Spain. It is estimated that >300,000 residents in the United States are infected with Chagas’ disease, a minority of whom have acquired it locally. Initial infection with T. cruzi is usually silent but, in 5–10% of cases, can present with systemic illness and acute myocarditis with systemic dense parasitic infiltration in the myocardium forming pseudocysts that rupture and lead to diffuse myocyte necrosis. After initial infection, patients enter an “indeterminate phase,” which can last for decades. The silent progression during this phase was at one time attributed to secondary immune activation, but a degree of per­ sistent parasitemia can now be detected and is recognized as the major determinant of ongoing inflammation and progression to chronic heart failure. Depopulation of parasympathetic neurons in the heart and gastrointestinal tract likely contributes to clinical disease. Under­ standing of the disease patterns has improved in part from the ability to test blood donors for infection. Better detection of the organism has revealed that both the prevalence and survival of patients in the indeterminate phase are higher than previously recognized, with over half of patients remaining asymptomatic. T. cruzi infection should be considered during evaluation of acute myocarditis or cardiomyopathy in all patients from endemic regions and for other patients with suggestive features. These include sinus and atrioventricular node conduction system disease and right bundle branch block, with frequent atrial and ventricular arrhythmias, some of which may cause sudden death before symptomatic disease. The dilated ventricles are thrombogenic and sometimes have aneurysms. For acute myocarditis, PCR is the most sensitive test and trypomasti­ gotes may also be detected in blood when parasite levels are high. In the chronic phase with low parasite burden, multiple serologic tests may be necessary for diagnosis, but their sensitivity and specificity are improving. Antiparasitic treatment with nifurtimox or benznidazole is recom­ mended for acute myocarditis with Chagas’ cardiomyopathy. During the indeterminate phase prior to clinical symptoms, therapy is widely recommended for children. For the indeterminate phase, there is controversy over the routine use of antiparasitic treatment, which is associated with significant toxicity. Therapy is more often advocated for patients <50 years old, who are more likely to tolerate chronic suppressive therapy. Treatment is recommended for premenopausal women and for patients on immunosuppression, such as after cardiac transplantation. A large multinational trial showed no benefit for benznidazole given to patients with chronic heart failure from Chagas’ cardiomyopathy. In these patients, general therapy is as indicated for other HFrEF, with increased indications for anticoagulation and for pacemaker defibrillators. Improvement of LVEF is uncommon and survival is lower than for other cardiomyopathies after the onset of overt clinical heart failure.

Trichinellosis (trichinosis) is caused by the Trichinella genus of nematodes (roundworms). The larvae are ingested with undercooked meat, with an initial intestinal phase after which larvae migrate into skeletal muscles, causing myalgias, weakness, and fever. Periorbital and facial edema and conjunctival and retinal hemorrhage may also be seen. Although the larva may occasionally invade the myocardium, clinical heart failure is rare and, when observed, attributed to the eosinophilic inflammatory response. The diagnosis is made from the serologies and is further supported by the presence of eosinophilia. Treatment includes anthelminthic drugs and glucocorticoids if inflam­ mation is severe.

CHAPTER 268 Acute and Chronic Myocarditis Bacterial Infections  Most bacterial infections can involve the heart occasionally through direct invasion and abscess formation but do so rarely. More commonly, systemic inflammatory responses to severe infection and sepsis depress myocardial contractility. Diphtheria is caused by bacillus Corynebacterium diphtheriae, usu­ ally presenting with an upper respiratory illness particularly affecting the pharynx, where a pseudomembrane is formed in response to the diphtheria toxin. This toxin interferes with protein synthesis in the heart and may particularly affect the conduction system. Cardiac involvement is the most common cause of death from this infection but rarely occurs with the occasional dominant cutaneous presenta­ tion. The prevalence of vaccines has shifted the incidence of diphtheria to countries without routine immunization and to older populations who have lost their immunity. The diagnosis is made from pharyngeal bacterial culture, which requires special culture media and a positive assay for the toxin but can be suspected from gram-positive rods on a Gram stain. Clinical suspicion is sufficient indication for the specific antitoxin, which should be administered as soon as possible, with higher priority than antibiotic therapy. Streptococcal infection with β-hemolytic streptococci is most com­ monly associated with acute rheumatic fever and is characterized by inflammation and fibrosis of cardiac valves and systemic connective tissue, but it can also lead to a myocarditis with focal or diffuse infil­ trates of mononuclear cells. Other systemic bacterial infections that can involve the heart include brucellosis, legionella, meningococcus, mycoplasma, psittacosis, and salmonellosis, for which specific treat­ ment is directed at the systemic infection. Tuberculosis can involve the myocardium directly as well as through tuberculous pericarditis but rarely does so when the disease is treated with antibiotics. Whipple’s disease is caused by Tropheryma whipplei. The usual manifestations are in the gastrointestinal tract, but pericarditis, coronary arteritis, valvular lesions, and occasionally clini­ cal heart failure may also occur. Multidrug antituberculous regimens are effective, but the disease tends to relapse even with appropriate treatment. Tick-Borne Infections  Spirochetal myocarditis has been diag­ nosed from myocardial biopsies containing Borrelia burgdorferi, which causes Lyme disease. Lyme carditis most often presents with arthritis and conduction system disease that resolves within 1–2 weeks of anti­ biotic treatment and is only rarely implicated in chronic heart failure. Other borrelia species carried by either ticks or lice can cause relapsing fever. Additional tick-borne illnesses associated with febrile illnesses and myocarditis include Rocky Mountain spotted fever, Q fever, and ehrlichiosis, all of which are treated with doxycycline alone or in com­ bination with other agents. ■ ■NONINFECTIVE INFLAMMATORY MYOCARDITIS Myocardial inflammation can occur in the absence of infectious causes. The paradigm of noninfective inflammatory myocarditis is cardiac transplant rejection, from which we have learned that myocardial depression can develop and reverse quickly, that noncellular mediators such as antibodies and cytokines play a major role in addition to lym­ phocytes, and that myocardial antigens are exposed by prior physical injury and viral infection. In the largest registry of acute myocarditis, chronic systemic inflammatory diseases were implicated in 7% of cases and in 15% of those with high-risk features.

The most commonly diagnosed noninfective inflammatory process affecting the myocardium is sarcoidosis. Sarcoidosis, as discussed in Chap. 379, is a multisystem granulomatous disease most commonly affecting the lungs, but involving many organs including skin, eyes, liver, nervous system, and bones, as well as the heart. The epidemiology appears to be changing, now recognized in men and women of all races and ethnic groups, typically between 30 and 50 years old but often after the age of 50 in women. Sarcoidosis is now understood as a com­ bination of foreign antigen presentation and a dysregulated immune response that leads to ongoing inflammation, including activated macrophages. Occupational exposures include agriculture, firefight­ ing, metal-working, and construction work, with silica dust, pesticides, mold, and other inhaled particles as examples of implicated antigens. The occurrence of sarcoidosis is increased in family members, reflect­ ing potential sharing of environments and of genetic variants, which have been identified in loci affecting immune responses.

PART 6 Disorders of the Cardiovascular System Cardiac sarcoidosis often accompanies pulmonary sarcoidosis but frequently occurs without detectable lung disease. The time course, burden and activity of cardiac granulomata, and the degree of extra­ cardiac involvement are remarkably variable. Patients may present with rapid-onset heart failure and ventricular tachyarrhythmias, conduction block, chest pain syndromes, or minor cardiac findings in the setting of pulmonary sarcoidosis, ocular involvement, an infiltrative skin rash, or a nonspecific febrile illness. Chronic sarcoidosis may go unrecognized for months or years. When ventricular tachycardia or conduction block dominates the initial presentation of heart failure without coronary artery disease, suspicion should be high for sarcoidosis as a cause of cardiomyopathy. Right bundle branch block should raise suspicion for sarcoidosis but is uncommon with other cardiomyopathies, in which left bundle branch block is more common. The pathology of cardiac sarcoidosis often shows a spectrum from metabolically active granulomas to bland fibrosis at the sites of previ­ ous inflammation. Regional wall-motion abnormalities are common, but global ventricular function may be preserved early during mild disease. When left ventricular function is only mildly reduced, ventric­ ular size may be near normal, sometimes described as “nondilated” or “minimally dilated cardiomyopathy.” Although occasionally listed with restrictive cardiomyopathies, sarcoidosis with a reduced ejection frac­ tion generally has the phenotype of dilated or minimally dilated car­ diomyopathy. There may be a right ventricular predominance of both dilation and ventricular arrhythmias, in which case potential diagnoses include genetic arrhythmogenic right ventricular cardiomyopathy, with which sarcoidosis shares multiple features. Diagnosis of cardiac sarcoidosis is easiest in the presence of biopsyproven sarcoidosis of other organs. Imaging of the heart can show regional wall motion abnormalities or small ventricular aneurysms. MRI of the heart can identify late gadolinium enhancement in a pat­ tern of fibrosis not compatible with myocardial infarction. Computed tomography of the chest often reveals pulmonary lymphadenopathy even in the absence of clinical lung disease. Positron emission tomog­ raphy (PET) of the whole chest can highlight active sarcoid lesions that are avid for glucose in heart or lung. When metabolically active adenopathy is detected, biopsy is often necessary to rule out malig­ nancy or chronic granulomatous infections such as tuberculosis or his­ toplasmosis before treating with immunosuppression for sarcoidosis. The scattered granulomata of sarcoidosis are commonly missed on cardiac biopsy (Fig. 268-3). Immunosuppression for sarcoidosis is generally initiated with glucocorticoids. Initial dosing was traditionally high but now is more often started at 30–40 mg daily or less, with early addition or substitu­ tion of steroid-sparing agents like methotrexate or mycophenolate. Third-line treatment may include tumor necrosis factor-alpha inhibi­ tors or other immunomodulators under investigation. The impact of treatment is usually more apparent for suppression of arrhythmias than for improvement of markedly impaired left ventricular dysfunction. Devices are often indicated for tachyarrhythmias or for conduction disease and usually include both pacing and defibrillation capability. Patients with sarcoidosis are immunosuppressed, and their manage­ ment should be supervised by an experienced multidisciplinary team.

FIGURE 268-3  Sarcoidosis. Microscopic image of an endomyocardial biopsy showing a noncaseating granuloma and associated interstitial fibrosis typical of sarcoidosis. No microorganisms were present on special stains, and no foreign material was identified. Hematoxylin and eosin–stained section, 200× original magnification. (Image courtesy of Robert Padera, MD, PhD, Department of Pathology, Brigham and Women’s Hospital, Boston.) Cardiac sarcoidosis may respond to the initial immunosuppression course without recurrence, may fluctuate over a chronic course, or occasionally may lead to end-stage heart failure requiring heart transplantation. Giant cell myocarditis is less common than sarcoidosis, but accounts for 10–20% of biopsy-proven myocarditis, typically presenting acutely with rapidly progressive heart failure and tachyarrhythmias. Giant cell myocarditis occurs equally in men and women, usually of older age than those with acute viral myocarditis, and is more often associated with systemic autoimmune disorders. Unlike sarcoidosis, the more dif­ fuse involvement with giant cell myocarditis usually leads to diagnostic endomyocardial biopsy, revealing granulomatous lesions surrounded by extensive inflammation infiltrate, often with eosinophilic infiltra­ tion. However, the occasional finding of giant cell myocarditis in explanted hearts after a previous diagnosis of sarcoidosis suggests that they may share the same disease spectrum. Glucocorticoid therapy alone is rarely effective, but in combination with other immunosup­ pression therapies similar to those used for severe transplant rejection, the rate of death or heart transplantation has decreased for patients who are stable at the time of presentation. Most patients presenting with cardiogenic shock from giant cell myocarditis progress to need urgent mechanical support or transplantation, for which they may be rendered ineligible by severe infection resulting from intensive immunosuppression. Eosinophilic myocarditis is one of the causes of fulminant myocarditis but may be missed on milder presentation that does not include assess­ ment for hypereosinophilia, which is present in about three-fourths of diagnosed cases. Myocardial toxicity of eosinophils may lead to apical thrombi on cardiac imaging, and MRI reveals inflammation acutely and subendocardial fibrosis in chronic cases. Endomyocardial biopsies show infiltration with lymphocytes, neutrophils, and a high proportion of eosinophils. Hypersensitivity to chronic medications accounts for up to one-third of cases and is often cured by withdrawal of the agent and acute steroid therapy. Particularly in Mediterranean and African coun­ tries, hypereosinophilia can result from chronic parasitic infection, for which treatment may cure the myocarditis. About 12% of cases are due to eosinophilic granulomatosis with polyangiitis (EGPA, formerly Churg-Strauss syndrome), which responds to immunosuppression but may have a relapsing course. The hypereosinophilic syndrome (HES) accounts for 10–15% of cases, resulting from myeloproliferative vari­ ants or overproduction of eosinophil production, for which combined immunosuppression may soon include therapies under investigation to decrease eosinophil production. About 30% of HES cases are currently