Myocarditis in the Intensive Care Unit

Myocarditis is defined as inflammation of heart muscle.¹ Many different etiologic agents have been implicated in this disease, but viral infections are the most common cause. Myocarditis is also associated with autoimmune and other systemic diseases.² The clinical picture of myocarditis varies widely, from asymptomatic patients who recover without specific therapy and suffer no long-term sequelae to critically ill patients with heart failure and cardiogenic shock. There are no standardized, specific, and widely agreed-upon criteria for making the diagnosis of myocarditis or for determining a cause in many patients.³ Lastly, there has been controversy regarding the most appropriate medical therapy for this condition.

On pathologic examination of myocardial biopsy specimens or on autopsy series, myocarditis is usually apparent as infiltration of myocardium with lymphocytes and fibroblasts, accompanied by myocyte necrosis (myocytolysis).³ It is this type of myocarditis, often termed lymphocytic myocarditis, that will be referred to in this chapter unless otherwise specified. Other types of inflammatory reactions can be seen less frequently in myocarditis, involving giant cells, eosinophils, or granulomas, which can be associated with specific clinical conditions.

In most patients with myocarditis, a specific cause is not found.⁴ It is presumed that in North America and Europe, the most common etiologic agent is viral.¹ Coxsackie B enterovirus was felt to be the most common cause up to the 1990s, but adenoviruses and parvovirus 19 have been implicated as causative agents more frequently over the past 20 years. Other viral causes include hepatitis C, cytomegalovirus, and human herpesvirus 6.² Myocarditis is a common finding in patients infected with human immunodeficiency virus (HIV). However, the causative agent responsible in these cases may be a secondary viral infection such as cytomegalovirus or other opportunistic infection such as mycobacteria, fungi, or parasites, rather than HIV itself.^1,5,6 Infectious illnesses such as Lyme disease, acute rheumatic fever, and diphtheria often have myocarditis as a prominent feature. In Central and South America, the most common cause of myocarditis is the protozoan, Trypanosoma cruzi, the cause of Chagas’ disease (Table 83-1). Systemic and autoimmune diseases such as systemic lupus erythematosus, polymyositis, scleroderma, sprue, Whipple’s disease, and sarcoidosis can be complicated by myocarditis, and myocarditis can be a feature of the infiltrative cardiomyopathies seen in hemochromatosis or amyloidosis. Idiopathic specific forms of myocarditis include hypersensitivity or eosinophilic myocarditis, which has also been reported after smallpox vaccination,⁷ and giant cell myocarditis.⁸ Lastly, myocarditis can be associated with doxorubicin cardiomyopathy or with peripartum cardiomyopathy, or it can be a manifestation of a hypersensitivity reaction to medications^9,10 (Table 83-2).

TABLE 83-1 Causes of Myocarditis^*

^* The most common causes are shown in boldface type.

From Feldman A, McNamara D. Myocarditis. N Engl J Med 2000;343:1388-98.

TABLE 83-2 Distinct Forms of Myocarditis

From Haas G. Etiology, evaluation, and management of acute myocarditis. Cardiol Rev 2001;9:88-95.

Unfortunately, it is difficult to make a clinical diagnosis of a specific viral cause of myocarditis. This usually requires measurement of antiviral antibody titers in acute and convalescent-phase sera. Viral cultures of tissue specimens are unreliable.⁴ Identification of viral genomes incorporated in myocyte DNA suggests but does not specifically prove that the virus is the cause.

Pathogenesis

Based on observations of human myocarditis, as well as murine models of the disease caused by coxsackie B3, the pathogenesis of viral myocarditis can be described in three stages.^2,11 The first stage is initiated by viral infection and replication within myocytes. Viral proteases and activation of cytokines may produce myocyte damage and apoptosis.¹² The presence of this viral replication phase is difficult to detect clinically because patients may be asymptomatic during this phase or only have nonspecific viremic symptoms. In addition, there is no rapid screening test to confirm viral infection.

The second stage involves host immune activation. Stimulation of cellular immunity and humoral responses attenuates viral proliferation and can result in recovery from the illness. However, unabated immune activation can result in activated T cells targeting myocardial antigens that cross-react with viral peptides. This leads to release of cytokines such as tumor necrosis factor (TNF), interleukin (IL)-1, and IL-6, resulting in further myocyte damage.^1,12 Activation of CD4 cells and antibody production plays a less important pathogenetic role. It is believed that this secondary immune response to viral infection plays a greater role in disease pathogenesis than the primary infection.¹²

Evidence supporting these mechanisms includes several key observations. Myocardial biopsy with recombinant DNA techniques can detect viral genomes in 20% to 35% of patients. Tissue-specific autoantibodies have been detected in 25% to 73% of patients with evidence of myocarditis on biopsy, with antibodies directed against contractile, structural, and mitochondrial myocyte proteins. Inappropriate expression of the major histocompatibility complex can frequently be demonstrated on biopsy specimens.¹ Elevated levels of inflammatory cytokines are detected in patients with active myocarditis.

Either persistent overactivation of cellular immune activity or incomplete clearing with persistent or recurrent viral replication can lead to the third stage, during which significant myocardial damage occurs. This leads to left ventricular (LV) dilatation and remodeling, LV systolic dysfunction, and manifestations of heart failure.¹² These processes can then abate, with reduction in LV size and improvement of LV function, or can continue to progress with development of dilated cardiomyopathy, worsening ventricular function, and chronic heart failure. Chronic dilated cardiomyopathy is the major long-term sequela of acute myocarditis (Figure 83-1).

Figure 83-1 Pathogenesis of viral myocarditis involves direct myocardial injury from viral infection as well as immune-mediated myocyte damage from cytokines, proteases, and autoantibodies. The outcome of these processes can be healing of inflammation and resolution, ongoing active myocarditis, or chronic dilated cardiomyopathy.

(Adapted from Blauwer LA, Cooper LT. Myocarditis. Prog Cardiovasc Dis 2010;52:274–88.)

Clinical Presentation and Diagnosis

The incidence of myocarditis is difficult to determine; many cases are mild with subclinical disease. Myocarditis is diagnosed on clinical grounds, as there are no specific clinical diagnostic criteria. The presentation of myocarditis varies widely. Patients can be asymptomatic insofar as myocarditis has been found in 1% to 10% of autopsy specimens of young adults who had no history of cardiac illness. Myocarditis can be found at autopsy in up to 20% of cases of young, apparently healthy adults who die suddenly and unexpectedly.^1,4,10

Patients ill with myocarditis present with nonspecific symptoms of dyspnea (72%), chest pain (32%), and symptoms of arrhythmia (18%).¹³ The presentation may be indistinguishable from acute coronary syndromes due to coronary artery disease. There may have been a preceding viral prodrome with fever, malaise, and arthralgias. Physical examination can show fever, tachycardia, S₃ and S₄ gallop sounds, and a pericardial rub if myopericarditis is present. Signs of heart failure can be present, including pulmonary rales and wheezes, elevated jugular venous pulse, and peripheral edema. Murmurs of mitral regurgitation and tricuspid regurgitation may be heard. Infrequently, the presentation is fulminant and severe, with acute heart failure, pulmonary edema, and cardiogenic shock.⁴

The differential diagnosis includes acute myocardial infarction (AMI), pericarditis, or chest pain from pulmonary causes including pulmonary embolism or pneumonia. Generalized sepsis is also a consideration.

Laboratory findings can include leukocytosis, eosinophilia, and an elevated erythrocyte sedimentation rate. Cardiac biomarkers such as creatine kinase, troponin T, and troponin I may be elevated, with sensitivity of troponin I reported at 34% and specificity of 89%.¹⁴ Rheumatologic serologic markers and HIV status should be evaluated.

The 12-lead electrocardiogram (ECG) is an insensitive test for the diagnosis of myocarditis. It shows sinus tachycardia and nonspecific ST-segment depression and T-wave inversion most often. Patients may present with chest pain and ST-segment elevation, with a picture mimicking AMI. More severe cases can be associated with supraventricular or ventricular arrhythmias, conduction disturbances, and heart block.¹

Echocardiography is essential to diagnose and quantitate regional or global LV wall-motion abnormalities, left ventricular and right ventricular size and function, the presence of pericardial effusion, and valvular regurgitation. Fulminant myocarditis is characterized by a nondilated left ventricle, with severe systolic dysfunction and increased wall thickness reflecting myocardial edema.¹⁵ Findings on myocardial nuclear scintigraphy are frequently abnormal, but this test is not useful in the diagnosis of myocarditis. Cardiac catheterization and coronary angiography are often necessary to exclude acute ischemia as the cause of chest pain or acute heart failure.

There is increasing use of cardiac magnetic resonance imaging (CMR) in the diagnosis of myocarditis.^16,17,18 This technique has the potential to offer a noninvasive means to make this diagnosis. CMR should be considered in symptomatic patients with a high clinical suspicion of disease when the results are likely to affect management decisions. Diagnostic criteria include: (1) focal or diffuse myocardial edema in T2-weighted images, (2) early gadolinium enhancement indicating inflammation, and (3) late gadolinium enhancement in subepicardial or mid-myocardial areas indicating necrosis and fibrosis. Abnormalities may be diffuse or patchy, often confined to the lateral free wall of the left ventricle or the base of the interventricular septum (Figure 83-2). Diagnostic accuracy of CMR is reported at 78% when 2 or 3 criteria are present and 68% when only late gadolinium enhancement is present. CMR is more likely to be abnormal when performed more than 7 days after onset of symptoms. CMR may also detect pericardial effusion (seen in 32%-57% of patients) and gives information regarding LV function. CMR can also be used to direct myocardial biopsy in patients with patchy uptake. The value of CMR for assessing prognosis is unknown, and this presently represents a major limitation of this diagnostic technique.^19,20,21

Figure 83-2 Cardiovascular magnetic resonance (CMR) with late gadolinium enhancement—normal and abnormal findings in myocarditis. A, Normal myocardium with no evidence of irreversible myocyte injury. B, Regional subepicardial enhancement of the lateral wall (arrow). C, Subepicardial enhancement of lateral and midwall enhancement of the septal wall (arrows). D, Diffuse subepicardial enhancement.

(From Friedrich MG, Sechtem U, Schulz-Menger J, Holmvang G, Alakija P, Cooper LT et al. Cardiovascular magnetic resonance in myocarditis: a JACC White Paper. J Am Coll Cardiol 2009;53(17):1475–87.)

Endomyocardial Biopsy

Percutaneous endomyocardial biopsy (EMB) is currently used to aid in the diagnosis of myocarditis and is considered the definitive diagnostic technique. The Dallas criteria have been accepted as the standard for histopathologic diagnosis. These criteria define active myocarditis as the presence of an inflammatory myocardial infiltrate (more than five lymphocytes per high-power field) accompanied by myocyte necrosis. Borderline myocarditis is defined as inflammation without myocyte necrosis. However, there is no difference in prognosis in patients with either of these biopsy results.⁹ Thus, lymphocyte infiltration (with or without myocyte necrosis) is the most important diagnostic criterion.

Although EMB is useful for diagnostic purposes, there are a number of significant limitations. A high frequency of interobserver variation has been noted among pathologists in applying the Dallas criteria. Biopsies are not sensitive in diagnosing myocarditis; various series have reported positive right ventricular biopsy results in only 10% to 67% of patients with myocarditis suspected on clinical grounds or with recent-onset idiopathic dilated cardiomyopathy. This variability may relate to the timing of biopsies in respect to the stage or chronicity of the patient’s illness. In addition, the myocardial inflammation may not be diffuse and may be patchy, or may predominantly involve the left ventricle, so random right ventricular biopsies may miss affected myocardium.²² Thus, performing a biopsy earlier in a patient’s clinical course, taking multiple biopsy specimens, and performing LV biopsies are ways of improving diagnostic yield. In addition, immunohistochemical staining for human leukocyte antigens can improve diagnostic sensitivity.^11,23 EMB should be performed in centers with a high-volume experience, with proven safety and availability of appropriate pathologic techniques.²⁴ However, it is important to emphasize that a negative biopsy finding does not preclude the diagnosis of myocarditis.

Although EMB is an insensitive test with a number of problems, a positive biopsy finding has a high positive predictive value.⁹ Some authors question the benefits of performing biopsy with standard staining techniques as a routine in suspected myocarditis cases, but this remains the best diagnostic test currently available. Other analyses such as examining specimens for viral genomes utilizing polymerase chain reaction (PCR) or using immunohistochemistry technology to identify up-regulated HLA proteins may offer improved diagnostic yield.²²

Endomyocardial biopsy should be strongly considered in cases of suspected myocarditis when pathology results will affect management decisions. A recent American Heart Association/American College of Cardiology/European Society of Cardiology (AHA/ACC/ESC) scientific statement offered recommendations concerning the appropriate use of EMB based on patients’ clinical presentations.²⁵ EMB was deemed useful, beneficial. and effective (class I indication) in patients with acute heart failure with hemodynamic compromise, after causes such as coronary artery disease are excluded. EMB is this setting is necessary to differentiate giant cell myocarditis and eosinophilic myocarditis from lymphocytic myocarditis, since immunosuppressive therapy is mandated in the first two conditions (see later). A class I indication for EMB was also recommended for patients with new-onset subacute heart failure, with duration of illness of 2 weeks to 3 months, who fail to improve with medical therapy for heart failure or who demonstrate severe ventricular arrhythmia or advanced heart block. EMB should be considered if causes such as sarcoidosis or collagen vascular disease are suspected and should be performed to diagnose giant cell myocarditis or eosinophilic myocarditis.²⁶ Endomyocardial biopsy should always be performed prior to initiating immunosuppressive therapy (Table 83-3).

TABLE 83-3 Indications for Endomyocardial Biopsy

Adapted with permission from Wu L, Lapeyre A, Cooper L. Current role of endomyocardial biopsy in the management of dilated cardiomyopathy and myocarditis. Mayo Clin Proc 2001;76:1030-8.

An algorithm has been proposed outlining the steps in evaluating patients suspected of having acute myocarditis (Figure 83-3).

Figure 83-3 Diagnostic algorithm for suspected acute myocarditis.

(Adapted from Nelson KH, Li T, Afonso L. Diagnostic approach and role of MRI in the assessment of acute myocarditis. Cardiol Rev 2009;17:24–30.)

Clinical Course and Prognosis

The clinical course and prognosis of acute myocarditis is variable. The majority of patients diagnosed with myocarditis will improve. Patients with mild symptoms most often recover without complications. Eight to 12% of young, apparently healthy adults who die suddenly from a cardiac cause are found to have myocarditis at autopsy, suggesting that patients even with apparently mild illness can suffer fatal arrhythmias.¹¹ Some patients with myocarditis will progress to chronic dilated cardiomyopathy with manifestations of systolic heart failure,³ although a precise incidence is not known. Fifteen to 25% of patients who present with new-onset dilated cardiomyopathy have evidence for antecedent myocarditis.³ Patients with heart failure and LV dysfunction will experience spontaneous resolution of their illness within 12 months in up to 40% of cases, without long-term sequelae. Roughly one-quarter of patients with acute myocarditis and ejection fraction less than 35% will improve, half will develop chronic cardiomyopathy and heart failure, and one-quarter will deteriorate and may be candidates for cardiac transplantation.²⁷

It is important to examine the patient population under study and the criteria used for diagnosing myocarditis in any series assessing prognosis and mortality. No clinical markers reliably predict which patients with myocarditis will recover or worsen.⁹ In the Myocarditis Treatment Trial, 1-year mortality rate was 20% and 5-year mortality was 56% in patients with biopsy-confirmed lymphocytic myocarditis.²⁸ A series of 21 patients with active myocarditis on biopsy was analyzed for predictors of disease course. Variables assessed included baseline hemodynamics, use of ventilatory and circulatory support, and serum cardiac biomarkers. Overall, there was a 37% mortality rate (8 of 21), with death occurring at 27.6 ± 6.9 days. Factors predicting a worse prognosis included hypotension (mean 84/49 mm Hg), higher pulmonary capillary wedge pressure (mean of 24 mm Hg), and use of mechanical ventilation. Factors that were not predictive of mortality included sex, age, heart rate, cardiac index, peak creatine kinase, or the use of intraaortic balloon counterpulsation for circulatory support.²⁹ Another trial reported 181 patients with myocarditis confirmed by EMB utilizing the Dallas criteria, immunohistochemical staining and PCR, which assesses for viral genome. LV biopsy was performed in 90% of patients. Patients were followed for an average of 59 months, and 22% died or received cardiac transplantation. Multivariate analysis concluded that functional class III and IV heart failure and a positive immunohistochemical result were the only predictors of poor outcome, and treatment with beta-blockers was associated with better outcomes²³ (Figure 83-4). Other series have reported that LV ejection fraction (LVEF) less than 40% and right ventricular dysfunction also predict a poorer prognosis.¹¹

Figure 83-4 Prognosis for patients with acute myocarditis was predicted by three factors: New York Heart Association functional class, positive immunohistology for myocarditis at endomyocardial biopsy, and therapy with beta-blockers.

(Adapted from Kindermann I, Kindermann M, Kandolf R et al. Predictors of outcome in patients with suspected myocarditis. Circulation 2008;118:639–48.)

Fulminant Myocarditis

A small percentage of patients with acute myocarditis present critically ill with acute severe heart failure and cardiogenic shock. This presentation is termed fulminant myocarditis. Most often these patients give a history of recent fever and symptoms of a viral illness, with a distinct time of onset of heart failure symptoms. This presentation can be contrasted with that of patients with myocarditis who have acute heart failure but not cardiogenic shock, who demonstrate a less distinct time of onset of heart failure symptoms and less severe hypotension.

In a study of 147 patients presenting with heart failure due to biopsy-positive active myocarditis with ejection fraction less than 40%, 10% of patients were diagnosed with fulminant myocarditis and 90% with acute lymphocytic myocarditis.⁸ The patients with fulminant myocarditis needed hemodynamic support with high-dose vasopressors or left ventricular assist devices (LVADs). The acute myocarditis patients had more stable hemodynamics and did not require vasopressors or received them at low doses. Patients with fulminant myocarditis tended to be younger and had higher heart rates and lower systemic blood pressure. There was no difference between the groups in mean pulmonary capillary wedge pressure or cardiac index.

With aggressive treatment, patients with fulminant myocarditis actually had better survival rates: 93% at 1 year and 93% at 11 years. Patients with acute myocarditis had an 85% 1-year survival rate and a 45% survival rate at 11 years. Patients with lower pulmonary capillary wedge pressure or higher cardiac index at presentation also had better survival.

In summary, fulminant myocarditis has a distinct clinical course, with critical illness at presentation but with excellent long-term survival once patients recover from the acute phase of their illness. Healing of myocardial injury and significant improvement of LV systolic function can be expected. Therefore, an aggressive approach to therapy, including the use of ventricular assist devices or other mechanical assist devices, without resorting to early cardiac transplantation, is warranted (Figure 83-5).⁹

Figure 83-5 Unadjusted transplantation-free survival according to clinicopathologic classification. Patients with fulminant myocarditis were significantly less likely to die or require heart transplantation during follow-up than were patients with acute myocarditis (P = 0.05 by the log-rank test).

(From McCarthy RE III, Boehmer JP, Hruban RH, Hutchins GM, Kasper EK, Hare JM et al. Long-term outcome of fulminant myocarditis as compared with acute (nonfulminant) myocarditis. N Engl J Med 2000;342:690–5.)

Giant Cell Myocarditis

Giant cell myocarditis is a distinct form of myocarditis, generally with a rapidly progressive course without significant likelihood of spontaneous resolution. On endomyocardial biopsy, infiltration with inflammatory giant cells is seen. Although the pathogenesis is not clear, it is believed to be an autoimmune disorder, and CD4 T lymphocytes are thought to play an important role. A total of 63 patients with biopsy-confirmed giant cell myocarditis were studied retrospectively.³⁰ Heart failure was the presentation in 75% of cases; 14% presented with ventricular arrhythmias, and 11% presented with chest pain, an abnormal ECG, or heart block. There was an association with inflammatory bowel disease in 8% of cases. Survival was poor, with a median time of 5.5 months to death or cardiac transplantation (Figure 83-6). In this uncontrolled series, immunosuppressive therapy was associated with prolonged survival from 3 months in 30 patients not given immunosuppressive drugs and 3.8 months in patients treated with prednisone, to 11.5 months in patients given prednisone plus azathioprine and 12.6 months in patients who were given cyclosporine as part of their regimen. Prognosis after cardiac transplantation was also worse when compared with other forms of heart disease, with a 30-day mortality rate of 15% and a 26% mortality rate during the 3.7-year posttransplant follow-up period. Twenty-six percent of patients had giant cell infiltrates seen in their transplanted heart at an average time of 3 years after transplant.

Figure 83-6 Line graphs showing the Kaplan-Meier survival curves for patients with giant cell myocarditis and lymphocytic myocarditis from the onset of symptoms (A) and from time of presentation to the referring center (B). In each case, survival was significantly shorter among those with giant-cell myocarditis.

(From Cooper LT Jr, Berry GJ, Shabetai R. Idiopathic giant cell myocarditis—natural history and treatment. Multicenter Giant Cell Myocarditis Study Group Investigators. N Engl J Med 1997;336:1862.)

Therapy

General Management of Heart Failure

Treatment of myocarditis is based on the clinical presentation. Patients with mild disease can be treated expectantly, with dietary sodium restriction and avoidance of strenuous exercise for several weeks or months.³ Animal models indicate that strenuous exercise can worsen myocarditis. Elimination of unnecessary medications is important in patients with eosinophilia.

Nonsteroidal antiinflammatory drugs should be avoided because they may worsen myocarditis.⁴ The routine use of anticoagulants for prophylaxis of systemic emboli is not recommended. Patients who present with symptoms of arrhythmia or heart failure should be hospitalized, with continuous cardiac rhythm monitoring performed for evaluation of potential life-threatening arrhythmias or conduction abnormalities. If these are diagnosed, they are treated in a similar matter as in patients with other causes of heart disease, utilizing antiarrhythmic drugs or pacemakers. However, a period of observation is recommended to assess for improvement of cardiac function prior to implantation of an implantable cardioverter-defibrillator (ICD).

There are data in murine models of myocarditis supporting the use of ACE inhibitors, angiotensin blockers, and beta-blockers. These drugs reduce inflammation and lessen necrosis and fibrosis.^2,3,11,19 There are convincing data in humans supporting the use of these medications, as well as aldosterone antagonists in patients with dilated cardiomyopathy. Therefore, in patients with myocarditis and heart failure, the use of standard multidrug medical therapy for heart failure and LV systolic dysfunction is indicated.^3,10 These medications have been shown to improve symptoms, prolong life, and regress the adverse LV remodeling in patients with dilated cardiomyopathy of various causes.^36,37,38

Treatment with ACE inhibitors should be initiated at low doses, with upward titration to maximally tolerated doses. Patients should be closely monitored for potential side effects including renal insufficiency, hyperkalemia, and angioedema. Relative contraindications to the use of ACE inhibitors include renal failure, hyperkalemia, bilateral renal artery stenosis, and hepatic failure. Patients with hypotension should be treated with parenteral vasopressors or circulatory assist devices prior to initiation of low-dose ACE inhibitor therapy.

As described earlier, β-adrenergic blockade was associated with improved survival in a multivariate analysis of patients with acute myocarditis.²³ Large randomized controlled clinical trials, which included patients with idiopathic dilated cardiomyopathy, have unequivocally shown benefit from beta-blockers in patients with LV systolic dysfunction,^39–43 and these agents should also be used in patients with heart failure due to myocarditis. Beta-blockers should be initiated after patients are on a stable dose of ACE inhibitors and when signs of fluid overload have resolved. Contraindications to beta-blocker therapy include bronchospastic disease or severe chronic obstructive lung disease, heart block, or significant underlying bradycardia. Hypotension should be corrected prior to initiating beta-blocker therapy.

Digoxin has been shown in animal models to decrease levels of cytokines, but digoxin was associated with adverse outcomes in one murine model of myocarditis. Digoxin can be useful in helping to control ventricular rates in patients with atrial fibrillation. After ACE inhibitors and beta-blockers have been initiated, the use of digoxin should be considered in patients with significant LV systolic dysfunction. However, no survival benefit for digoxin has ever been shown in patients with heart failure due to dilated cardiomyopathy.⁴⁴ Contraindications to the use of digoxin include renal failure or heart block.

Use of the aldosterone antagonist, spironolactone, has been shown to have symptomatic and survival benefit in patients with class III-IV chronic systolic heart failure.⁴⁵ In experimental models, these agents can reverse the progressive myocardial fibrosis that occurs in the remodeling process of dilated cardiomyopathy. These agents have not been studied in patients with myocarditis, but their use should be strongly considered in patients with severe LV dysfunction (ejection fraction less than 35%) and symptomatic heart failure.² Contraindications to the use of aldosterone antagonists include renal insufficiency, serum creatinine levels above 2.0 mg%, or hyperkalemia. Serum potassium levels must be carefully monitored during initiation and dose titration.

In critically ill patients with severe heart failure and low cardiac index, parenteral vasodilators should be used. Intravenous (IV) nitroprusside is a powerful venous and arterial dilator which significantly reduces systemic vascular resistance, mean systemic arterial pressure, and pulmonary capillary wedge pressure, raising cardiac index. It must be administered in the intensive care unit (ICU), with invasive hemodynamic monitoring with a pulmonary artery catheter, to best gauge the appropriate dose of medication and accurately assess response to therapy. Prolonged use of nitroprusside is associated with accumulation of the toxic metabolites thiocyanate and cyanide, and serum levels of these compounds must be monitored. Intravenous nitroglycerin is also an effective venodilator and coronary vasodilator, with less arterial dilating property than nitroprusside. The use of nitroglycerin in cases of myocarditis has not been studied. Patients often develop tolerance to this drug.^46–48

Patients with severe myocarditis may develop cardiogenic shock, with hypotension, respiratory failure, and signs of end-organ hypoperfusion. In these instances, initial treatment with inotropic agents or vasopressors is indicated. Dobutamine is a potent β₁-agonist with less β₂– and α-agonist properties. Dobutamine has favorable short-term hemodynamic effects with increasing myocardial contractility, reducing systemic vascular resistance and reducing pulmonary capillary wedge pressure. However, dobutamine can be proarrhythmic, and patients can develop tolerance to the drug. Routine use of dobutamine in patients with exacerbations of chronic systolic heart failure was associated with increased mortality rates when compared with placebo.⁴⁹

Milrinone is another parenteral inotropic agent that works by inhibiting phosphodiesterase. This drug leads to increased inotropy and decreased systemic vascular resistance and pulmonary capillary wedge pressure, with resultant increased stroke volume and cardiac index. Milrinone may cause hypotension. It is less proarrhythmic than dobutamine, and it does not induce tolerance.^50,51

Arterial vasoconstrictors such as norepinephrine and dopamine can be used in patients with refractory hypotension for short-term urgent blood pressure support. However, these agents cause increased myocardial oxygen consumption and can have deleterious effects on myocardial function.

In patients with fulminant myocarditis or cardiogenic shock not responding to pharmacologic therapy, intraaortic balloon counterpulsation should be utilized. Mechanical ventricular assist devices (VADs) are used for patients requiring greater hemodynamic support. These devices are mechanical pumps which provide physiologic cardiac output and LV afterload reduction and may provide time for spontaneous improvement or recovery of normal LV function. VADs are usually univentricular but can be biventricular, supporting both right and LV function. With improved technology, these devices are smaller and can be implanted through smaller incisions or percutaneously. VADs are connected to an external power pack via a driveline through the skin. The power pack is now small and portable, so patients have freedom of movement and can participate in rehabilitation efforts during VAD use. Routine anticoagulation therapy is not required. Complications of VADs include local site infection, sepsis, thromboemboli, and device failure.^52,53

In patients with myocarditis, VADs can be used to provide circulatory needs and improve coronary flow during the time necessary for spontaneous resolution of myocarditis to occur. Beneficial reverse remodeling may occur while patients are on VAD support, resulting in improved myocyte structure and function. VADs can provide support for months or even years. Some authors believe that patients with fulminant myocarditis should be given every opportunity to recover ventricular function with VAD use, and that cardiac transplantation should be used only as a last resort when severe heart damage is irreversible.⁵⁴

There are several unresolved issues regarding VAD usage in patients with myocarditis. These include appropriate patient selection, timing of VAD placement, best medical therapy during VAD support, and optimal duration of VAD support. A 50-day course of VAD support in the study described earlier allowed identification of 50% of those patients who ultimately recovered, and a 90-day course identified 80% of patients who recovered. The optimal means of serial assessment of native heart function while on VAD support needs to be delineated, and the best weaning protocol also needs definition.

Cardiac transplantation is the final option for treating critically ill patients with myocarditis. However, these patients have a higher rate of transplant rejection and a lower survival rate when compared with patients transplanted for ischemic or other causes of cardiomyopathy. Myocarditis has been reported to recur in the transplanted heart (Figure 83-7).¹⁰

Figure 83-7 Graph showing actuarial survival duration of heart transplant recipients with active lymphocytic myocarditis (green) compared with that of age-matched (red) and age- plus sex-matched (purple) control patients.

(From Haas G. Etiology, evaluation, and management of acute myocarditis. Cardiol Rev 2001;9:88–95.)

Immunosuppressive Therapy

Because autoimmune mechanisms are responsible for myocardial injury and the clinical manifestations of myocarditis, therapy with immunosuppressive drugs has been studied. However, given the high rate of spontaneous recovery of LV function, placebo-controlled trials are essential to properly evaluate the effects of therapy. In addition, heterogeneous patient populations consisting of patients with acute myocarditis and chronic dilated cardiomyopathy have been included in immunosuppressive trials, confounding the interpretation of results.

High-dose daily prednisone therapy was used for a 3-month course in 102 patients with dilated cardiomyopathy, 59% of whom were classified as having “reactive” myocarditis on endomyocardial biopsy.⁵⁵ The authors found a significant improvement in LVEF at 3 months in treated patients with reactive myocarditis (Figure 83-8), but this improvement was not sustained at 9 months. Improvement did not occur in patients with nonreactive biopsies treated with prednisone. No significant mortality benefit from immunosuppressive treatment was noted, although this was not a prespecified primary endpoint.

Figure 83-8 A, Ejection fraction in reactive dilated cardiomyopathy patients at 3 months. B, Prednisone does not change ejection fraction in nonreactive patients in 3 months.

(From Parrillo JE, Cunnion RE, Epstein SE, Parker MM, Suffredini AF, Brenner M et al. A prospective, randomized, controlled trial of prednisone for dilated cardiomyopathy. N Engl J Med 1989;321:1061–8.)

The Myocarditis Treatment Trial enrolled 111 patients with a positive endomyocardial biopsy finding and LVEF less than 45%, with a duration of illness of less than 2 years.²⁸ Three treatment groups were compared: daily prednisone plus azathioprine, prednisone plus cyclosporine, and placebo. Mortality was 20% at 1 year and 56% at 3 years. These investigators found no difference in ejection fraction at week 28 or week 52, no change in LV size at week 28, and no difference in 1-year mortality between treated and untreated groups. Their conclusion was that these immunosuppressive strategies were not beneficial. Significant limitations of this study include a 30% dropout rate and significant interobserver variability among pathologists’ diagnoses of biopsy specimens, despite utilizing the Dallas criteria (Figure 83-9).

Figure 83-9 Actuarial mortality (defined as deaths and cardiac transplantations) in immunosuppression and control groups. Numbers of patients at risk are shown at the bottom. There was no significant difference in mortality between the two groups.

(From Mason JW, O’Connell JB, Herskowitz A, Rose NR, McManus BM, Billingham ME et al. A clinical trial of immunosuppressive therapy for myocarditis. The Myocarditis Treatment Trial Investigators. N Engl J Med 1995;333:269–75.)

In view of the limitations of histopathologic diagnosis using the Dallas criteria, another group of investigators used immunohistologic markers of inflammation, up-regulation of HLA, to diagnose active myocarditis as an indication for immunosuppressive therapy.⁵⁶ This criterion has the advantage of indicating that autoimmunity is playing a role in pathogenesis. Also, since HLA is distributed throughout the entire myocardium, biopsy sampling error is eliminated as a confounding variable in assessing response to therapy. In this study, 84 of 202 patients with chronic (>6 months) idiopathic dilated cardiomyopathy (ejection fraction < 40%) were found to have strong expression of HLA in biopsy specimens and were randomized to receive placebo or prednisone plus azathioprine for 3 months. At 3 months’ follow-up, a significant improvement in the prespecified secondary endpoints of LVEF, LV volumes, and functional capacity was seen in the treated group, and this improvement was maintained at 2 years (71.8% improvement in the treated group versus 30.8% in the untreated group). However, there was no improvement in the prespecified composite primary endpoint of death, cardiac transplant, or hospital readmission. This study was limited by a 31% dropout rate.

In another study, patients with positive endomyocardial biopsy specimens and progressive heart failure who responded to 6 months of therapy with prednisone and azathioprine were more likely to have circulating cardiac autoantibodies and no viral genome in their myocardium as compared with nonresponders.⁵⁷

Studies have suggested that in patients with heart failure and low ejection fraction, IV immunoglobulin has a pronounced antiinflammatory effect, as measured by circulating levels of inflammatory markers.⁵⁸ Uncontrolled studies suggested benefit in patients with myocarditis from treatment with IV immunoglobulin.^59,60 However, a placebo-controlled double-blind trial of IV immunoglobulin in patients with myocarditis or idiopathic dilated cardiomyopathy of less than 6 months’ duration showed no significant improvement with therapy, as assessed by ejection fraction or functional capacity at 6 and 12 months.⁵⁰ In this study, average LVEF improved from 25% ± 8% at baseline to 41% ± 17% at 6 months in both treated and untreated groups. One-year event-free survival rate was 91.9% in both groups, indicating a favorable prognosis.

In summary, there is no evidence that patients with lymphocytic myocarditis or idiopathic dilated cardiomyopathy benefit from routine use of immunosuppressive therapy. However, this treatment approach should be considered in patients with myocarditis and positive endomyocardial biopsy findings, those who develop early signs of severe heart failure, and those who are shown to experience progressive worsening of LV function. Lastly, immunosuppressive therapy should be used in patients with myocarditis associated with connective tissue diseases such as systemic lupus erythematosus (SLE), eosinophilic or granulomatous forms of the disease, and in giant cell myocarditis (Figure 83-10).

Figure 83-10 Algorithm describing a reasonable approach to myocarditis management based on currently available data. ACE, angiotensin-converting enzyme; CAD, coronary artery disease; CHF, congestive heart failure; CMP, cardiomyopathy; EF, ejection fraction; LV, left ventricular.

(From Parrillo J. Myocarditis: how should we treat in 1998? J Heart Lung Transplant 1998;17:941–4.)

Current investigations are evaluating antiviral therapies in the acute stage of myocarditis as well as the use of antiviral vaccine in the prevention of disease. Appropriately powered, controlled, prospective studies of homogeneous patient groups utilizing immunosuppressive therapy are still needed. Evaluating the mechanisms of myocardial recovery during VAD support may also help direct research toward novel approaches to the treatment of myocarditis.

Transient Apical Ballooning Syndrome

A distinctive cardiomyopathy with acute onset, frequently precipitated by emotional or physical stress, is termed transient apical ballooning syndrome (TABS) owing to a distinctive LV wall-motion abnormality. This cardiomyopathy was first described in patients in Japan in 1991,⁶¹ and the syndrome has subsequently been described in the United States and Europe.^62,63 It is characterized by the sudden onset of chest pain and/or dyspnea, ECG changes mimicking AMI, and mild elevation of serum myocardial biomarkers. The syndrome is precipitated by extreme emotional or physical stress in over 70% of cases.⁶⁴ The characteristic LV wall-motion abnormality is akinesis or dyskinesis of a large area of the LV apex (Figure 83-11 and 83-12). Coronary artery stenosis is not present. TABS is also known as stress cardiomyopathy or takotsubo cardiomyopathy, so named because the takotsubo pot used by Japanese fishermen to trap octopus has a shape similar to the left ventricle in this condition (“short neck, round flask”).^63–66

Figure 83-11 End-diastolic and end-systolic apical four-and-two-chamber echocardiographic views demonstrating typical apical and mid-ventricular LV wall-motion abnormalities of a patient with transient apical ballooning syndrome.

(From Gianni M, Dentali F, Grandi AM, Sumner G, Hiralal R, Lonn E. Apical ballooning syndrome or takotsubo cardiomyopathy: a systematic review. Eur Heart J 2006;27:1523–9.)

Figure 83-12 Left ventriculogram showing typical left ventricular wall-motion abnormalities in transient apical ballooning syndrome. Arrows in systole indicate hyperkineses of basal inferior and anterior segments, with severe hypokinesis of remaining wall segments.

(From Sharkey SW, Lesser JR, Zenovich AG, Maron MS, Lindberg J, Longe TF et al. Acute and reversible cardiomyopathy provoked by stress in women from the United States. Circulation 2005;111:472–9.)

There is a marked preponderance for elderly females to be affected by this condition—86% to 100% in reported series, with a mean age of 63 to 67 years. Between 66% and 90% of patients will present with chest pain, and 15% to 20% will present with dyspnea, pulmonary edema, or shock. The most common ECG changes seen are ST-segment elevation or marked T-wave inversions in the precordial leads. These findings are indistinguishable from AMI. Elevation of creatine kinase MB (CK-MB) and troponin is seen in the majority of patients, but the enzyme rise is typically milder than would be expected, given the marked ECG and LV wall-motion abnormalities.

Precipitators of TABS have included arguments with family members, the death of loved ones, or sudden financial setbacks. Physical stresses have included medical procedures such as thoracentesis or biopsy, institution of cancer chemotherapy or hemodialysis, and hip fracture and noncardiac surgeries.

Echocardiography and left ventriculography show moderate to severe LV dysfunction in these patients, with characteristic hyperkinesis of inferior-basal and basal-septal segments, with severe hypokinesis or dyskinesis involving mid-anteroseptal, apical, and inferior-apical wall segments. Acutely, LVEF is reduced to 20% to 40%.^64,65 Up to 20% of patients may demonstrate a LV outflow tract gradient due to basal septal hyperkinesis and transient systolic anterior motion of the anterior leaflet of the mitral valve.^61,62,66

Patients with TABS often present critically ill, with pulmonary edema, hypotension, and shock. Cardiogenic shock develops secondary to marked LV systolic dysfunction and decreased stroke volume. Shock can also be exacerbated by the development of an LV outflow tract gradient.⁶⁸ Cardiogenic shock has been reported in 5% of patients at presentation and has occurred during the course of the illness in 6% to 46% of patients in different series.^63–6669

Suspicion of TABS and urgent diagnosis are important, since therapy and prognosis differ substantially from AMI. TABS should not be treated with thrombolytic therapy, as coronary occlusion is not involved in the pathogenesis. If cardiogenic shock develops, treatment with intraaortic balloon pump (IABP) counterpulsation is indicated. Inotropic therapy should be used judiciously or not at all. Dobutamine and other β-agonists may worsen cardiogenic shock by increasing hyperkinesis of the basal portion of the heart and causing or aggravating an LV outflow tract gradient. There have been several case reports of patients with TABS with hypotension who develop frank cardiogenic shock after initiation of inotropic therapy. Since a hyperadrenergic state has been proposed to be a major pathogenic mechanism, empirical use of beta-blockers while patients are being supported with IABP counterpulsation has been used successfully. Echocardiography can be useful to guide therapy. For those with extensive wall-motion abnormalities but no outflow obstruction, IABP support without beta-blockers is recommended. Administration of the α-agonist, phenylephrine, can also be considered in cases with a high LV outflow tract gradient, because this drug increases afterload, causing LV dilatation and a decrease in mitral valve systolic anterior motion and lowering of intraventricular gradients.⁶⁷

TABS is associated with a good prognosis; therefore aggressive therapy of hemodynamic compromise and cardiogenic shock is indicated. In almost all patients, the marked apical wall-motion abnormalities begin to improve within days, and LV function can be expected to recover to normal during the ensuing weeks or months. Follow-up in various series has shown recovery of LVEF to normal in most instances. In-hospital mortality in larger series has been reported at 0% to 4%.^{62,64–66,69,70} The large majority of survivors will recover completely, with normal functional status. The long-term prognosis is good. In one series, only 2 out of 72 patients had recurrence of TABS within 13 months.⁶³ In another series, the recurrence of TABS was calculated at 2.9% per year. Over a 4-year follow-up, long-term survival of patients who recovered from TABS was equivalent to sex- and age-matched control groups without a history of TABS.⁷⁰

The pathogenesis of TABS is unknown. Transient multivessel coronary spasm has been proposed, but this has not been demonstrated at the time of acute coronary angiography in the vast majority of patients. In most patients, the extent of LV wall-motion abnormality is larger than the distribution of a single coronary artery.^62,63 Cardiac MRI has not shown evidence of infarction or myocarditis.⁷¹ In our judgment, a hyperadrenergic state precipitated by acute stress and causing myocardial stunning is the most attractive hypothesis. One study documented supraphysiologic levels of serum catecholamines and stress neuropeptides in patients during the acute phase of TABS, likely due to adrenal and sympathoneuronal hyperactivity. The apex of the left ventricle may be more sensitive than other LV wall segments to the deleterious effects of adrenergic hyperstimulation.⁶⁵

TABS has been reported to occur in approximately 1.7% to 2.2% of admissions for acute coronary syndrome in Japan and 2% of cases of acute heart failure due to acute coronary syndrome.^69,71 TABS may be more common than currently recognized. Correct diagnosis is more likely to be made in centers where emergency coronary angiography and primary percutaneous coronary intervention are used in the treatment of acute coronary syndrome and ST-segment elevation myocardial infarction (STEMI).

In summary, TABS should be suspected in patients who present with symptoms and ECG findings consistent with AMI, who have a large apical wall-motion abnormality seen on echocardiography or left ventriculography, and whose symptoms were precipitated by severe emotional or physical stress. Diagnosis is confirmed when urgent cardiac catheterization and coronary angiography demonstrate no significant coronary artery occlusion or stenosis.

Tachycardia-Induced Cardiomyopathy

A sustained rapid heart rate can lead to the acute development of LV dilation and dysfunction with symptoms of heart failure. This is termed tachycardia-induced cardiomyopathy (TICMP) and can occur in otherwise normal hearts or can exacerbate heart failure in patients with preexisting cardiomyopathy. Supraventricular or ventricular arrhythmias of any type can lead to this syndrome. Arrhythmias which may be responsible for TICMP include atrial fibrillation, atrial flutter, automatic atrial tachycardia, AV node reentry tachycardia, supraventricular tachycardia involving accessory pathways, accelerated junctional tachycardia, ventricular tachycardia (from RV and LV sites) and even prolonged, sustained ventricular bigeminy.⁷² It is not known how long the tachycardia needs to be present in order to cause LV dysfunction, but sustained arrhythmia for days to weeks is likely necessary. The presence of an underlying predisposing substrate has been postulated, as not all patients with sustained tachycardia will develop cardiomyopathy.⁷³

Animal models of TICMP have been established and studied to elucidate pathophysiologic mechanisms and clinical correlates. In these models, sustained, rapid atrial or ventricular pacing leads to severe biventricular systolic and diastolic dysfunction with four-chamber dilation. Within 24 hours of initiating rapid pacing, there is a fall in cardiac output and an increase in ventricular filling pressures. Neurohormonal activation occurs, typical for dilated cardiomyopathy. Cardiac output, ejection fraction, and ventricular volume continue to deteriorate over 3 to 5 weeks. When pacing is discontinued, cardiac output improves to near normal in 48 hours, and hemodynamics are normal within 4 weeks. Ejection fraction recovers to normal in 1 to 2 weeks, although end-diastolic volume remains high at 12 weeks, suggesting persistent remodeling. Structural cardiac changes seen include myocyte hypertrophy and apoptosis and altered extracellular matrix. Proposed pathophysiologic mechanisms include myocardial energy depletion, ischemia, and altered myocyte handling of calcium.^73–75

There currently are no data in humans regarding the time course, mechanisms, or cellular biochemical alterations. TICMP can occur at any age, from infants to the elderly. TICMP has been reported to occur in fetuses with sustained supraventricular tachycardia, which resolved with correction of the arrhythmia.⁷⁴ It is not known if there is a minimal heart rate necessary to induce cardiomyopathy. The longer the duration of arrhythmia, the more likely cardiomyopathy is to occur and the more severe it will tend to be. The incidence and prevalence of TICMP are not known.

The diagnosis of TICMP should be suspected in any patient with impaired ventricular function in the setting of sustained supraventricular tachycardia or ventricular tachycardia. The diagnosis is clear when LV function prior to the onset of tachycardia was demonstrated to be normal, and no intercurrent illness other than the arrhythmia has occurred. The diagnosis is confirmed when LV function rapidly improves with correction of the arrhythmia.

Treatment of TICMP is to rapidly restore normal heart rate. This can be done with parenteral rate-slowing medication, including beta-blockers such as esmolol or metoprolol or calcium channel blockers such as diltiazem. Verapamil may aggravate hypotension and LV dysfunction and should be avoided. Adenosine can rapidly convert atrioventricular nodal reentry tachycardia to sinus rhythm. Intravenous digoxin can also be considered, although its onset of action is delayed. Type I drugs such as procainamide can be prescribed for supraventricular tachycardia associated with accessory pathways. Electrical cardioversion can rapidly terminate supraventricular and ventricular tachycardia and restore sinus rhythm. In patients with atrial flutter or atrial fibrillation, reliable control of the heart rate to a range of 60 to 90 bpm is a reasonable alternative to conversion of the arrhythmia to sinus rhythm.

In patients with TICMP who have received appropriate arrhythmia therapy, heart failure symptoms improve rapidly. LV systolic function will generally recover to normal within 4 weeks if there is no other underlying heart disease. Cardiac rhythm monitoring for 24 to 48 hours is often necessary to ensure that heart rate is controlled during activity as well as at rest.⁷⁴ In a report of 11 patients with atrial flutter and abnormal systolic function who underwent atrial flutter ablation, ejection fraction improved from an average of 31% at baseline to 41% within 7 months of ablation. Lack of resolution of cardiomyopathy was predicted by a lower baseline ejection fraction.⁷⁶ A series of 24 patients with TICMP was reported whose cardiomyopathy initially resolved with arrhythmia control, but who experienced repeated rapid decline in LV function and recurrent heart failure when their arrhythmias reoccurred. These patients again had improvement or normalization of ejection fraction following repeated arrhythmia control within 6 months. However, three of the patients died suddenly and unexpectedly, emphasizing that structural and electrical abnormalities may persist on a chronic basis.⁷⁷

Key Points