Abstract
The etiology of inflammatory heart disease encompasses a wide array of pathologic processes, including infectious, rheumatic, autoimmune, toxic, allergic diseases, and idiopathic, among others. The manifestations of such disease usually present clinically with pericarditis or myocarditis, with or without pericardial effusion. Myocardial or pericardial involvement may be the primary, isolated component of the disease process or just one manifestation of a systemic, multiorgan disease. The cardiac involvement may be the first manifestation of a systemic disease or may occur years after the primary disease process has been identified. The cardiologist, intensivist, and surgeon must often collaborate in the management of these patients because the clinical manifestations of such disease may be associated with life-threatening cardiac dysfunction, may require invasive procedures to treat or monitor such conditions, or may progress to chronic heart failure, for which transplantation may be the only option.
Key Words:
pericarditis, myocarditis, pericardial effusion, tamponade, heart failure
Inflammatory Heart Disease—Pericardial Effusion/Tamponade
Normal Pericardium: Anatomy and Physiology
Anatomy.
The pericardium is a dual-layered structure enveloping the heart and proximal great vessels. It consists of an inner visceral pericardium (also called the epicardium when in contact with the myocardium), and an outer parietal pericardium, composed of layers of collagen fibrils and elastin fibers. The pericardium normally contains approximately 10 to 50 mL of fluid, an ultrafiltrate of plasma (the pericardial reserve volume). Pericardial fluid may contain cytokines, prostaglandins, atrial natriuretic peptide (particularly in congestive heart failure [HF]), endothelin, and growth factors that may locally influence myocytes and vascular tissue, as well as albumin and other plasma proteins. Drainage of pericardial fluid occurs via lymphatics on the surface of the heart and parietal pericardium.
The pericardium is tethered by its reflection around the great vessels and fibrous connection with the vertebral column, sternum, and diaphragm. The outer surface of the pericardium is in direct contact with the pleura. The lungs constitute a space that envelops the heart and pericardium termed the cardiac fossa . Sensory innervation to the pericardium is supplied by the phrenic nerve.
Physiology.
Various functions have been attributed to the pericardium, including (1) maintenance of the heart in a fixed position in the chest, (2) protection of the heart from infection, inflammation, or contiguous spread of malignancy from surrounding structures, (3) minimizing friction associated with cardiac motion, and (4) influencing cardiac dynamics by exerting external constraining forces over the heart. The pericardium appears to influence cardiac dynamics by maintaining an optimal shape of the heart, enhancing interactions between chambers to balance right and left ventricular output, and attenuating acute changes in preload and afterload. However, the pericardium is not essential for cardiovascular function as evidenced by the fact that both congenital and acquired deficiency states are compatible with long-term survival.
Pericardial Effusion and Tamponade
Overview.
Pericardial effusion occurs in a variety of clinical settings when excess pericardial fluid accumulates beyond the usual pericardial reserve volume. Pericardial effusion may consist of transudate due to obstruction of fluid drainage and/or low plasma oncotic pressure (e.g., HF), exudate due to inflammatory/infectious processes (e.g., purulent pericarditis), blood (e.g., chest trauma), air (e.g., pneumopericardium), chyle (idiopathic or postoperative chylopericardium), or even intravenous (IV) crystalloid solution (e.g., perforation of the right atrium or intrapericardial cava by a central venous catheter). Causes of pericardial effusion, with indication of presence of inflammation of the pericardium (pericarditis) versus noninflammatory causes, are listed in Box 28.1 .
Infectious Pericarditis
Viral
See Box 28.2
Bacterial/Tuberculous
Staphylococcus aureus, Haemophilus influenzae, Neisseria meningitidis, Streptococcus pneumoniae β-hemolytic streptococci, Mycoplasma pneumoniae, Mycobacterium tuberculosis, Enterobacter cloacae, Corynebacterium diphtheriae
Fungal
Candida, Aspergillus, Histoplasma, coccidiomycosis
Noninfectious Pericarditis
Cardiac Injury
Postpericardiotomy syndrome, post myocardial infarction
Systemic Inflammatory Disease
Rheumatic fever, juvenile rheumatoid arthritis, systemic lupus erythematosus, inflammatory bowel disease, Kawasaki disease, rheumatic fever, tumor necrosis factor receptor–associated periodic syndrome (TRAPS), familial Mediterranean fever
Neoplasms
Leukemia, lymphoma, metastatic tumor, radiation pericarditis
Renal
End-stage renal disease (uremia), dialysis
Miscellaneous
Hypersensitivity to drugs/toxins (e.g., procainamide, hydralazine, penicillins, cromolyn, dantrolene, anthracyclines), amyloidosis, radiation
Additional Causes of Pericardial Effusion ± Pericarditis
Traumatic and Postoperative
Blunt chest trauma, central line malposition, catheterization/biopsy, rupture of coronary aneurysm (such as with Kawasaki disease), early postoperative hemorrhage
Elevated Central Venous Pressure/Low Plasma Oncotic Pressure
Superior vena cava syndrome, chylous effusion, primary pulmonary hypertension and right ventricular failure, decompensated congestive heart failure, liver failure, anorexia nervosa
Bone marrow transplant/graft versus host disease
Hyperthyroidism/Hypothyroidism
Etiology.
The etiology of pericardial effusion varies widely by population studied; however, purulent pericarditis appears to have become less common in the era of antibiotics and routine vaccination against previously common bacterial pathogens. In a 20-year review of children with pericardial effusion from a tertiary care center in the United States, bacterial infections were found in only 3% of patients, with Haemophilus influenzae , Staphylococcus aureus , and Neisseria meningitidis each reported. In this study 39% of patients with pericardial effusion had associated neoplastic disease, 9% collagen vascular disease, 8% renal disease, 5% other diagnoses (human immunodeficiency virus [HIV], viral sepsis, hypothyroidism, anorexia nervosa), and 37% were classified as idiopathic. By contrast, a study of children with pericardial effusion from a tertiary center in India demonstrated tuberculosis in 52%, other bacteria in 23% (including S. aureus and Pseudomonas aeruginosa ), viral causes in 12%, idiopathic effusion in 8%, and only 4% with neoplastic disease. Similar findings have been reported in adult studies of pericardial effusion, with 7% to 50% idiopathic, 10% to 37% neoplastic, 2% to 20% infectious, 4% to 20% uremic, 5% to 15% collagen vascular, 0% to 20% iatrogenic, 0% to 8% post myocardial infarction, and as high as 50% to 70% tuberculous in some areas of the world.
Pericardial effusion may occur after surgery for congenital heart disease and is usually transudative but may be chylous. Increased venous pressure is thought to be a principal factor in the pathogenesis. Various conditions, including septal defect repairs, cavopulmonary anastamoses, and repair of lesions with right ventricular (RV) outflow tract obstruction, have been associated with postoperative pericardial effusions, with variable clinical significance and typical detection a mean of 11 days after surgery. Prolonged drainage may indicate hemodynamic problems within the repair and should prompt consideration. Postoperative effusions may also occur in patients after heart transplantation, possibly due to the smaller donor heart within the chronically stretched recipient pericardium. Persistence of pericardial effusions after transplant has been correlated with both higher incidence and more severe histology of rejection. Postpericardotomy syndrome following heart surgery is associated with pericardial effusion, with symptoms largely related to pericarditis as discussed later.
Signs and Symptoms.
Small pericardial effusions may be asymptomatic or associated with only slightly muffled heart sounds on examination and may be discovered incidentally by widened cardiac silhouette on chest radiograph or on an echocardiogram performed in the context of a workup for an associated systemic illness. Dyspnea or chest discomfort may be present, especially if there is accompanying pericarditis. A classification system has been proposed based on size (large >20 mm in size on imaging, moderate 10 to 20 mm, small <10 mm), distribution (circumferential versus loculated), and duration (acute if <1 week, subacute 1 week to <3 months, or chronic if >3 months) ; however, clinical manifestations in children are multifactorial and do not necessarily correlate with these features.
Cardiac Tamponade.
Cardiac tamponade is a life-threatening clinical condition that occurs when accumulation of material (such as fluid, blood, or air) in the pericardium causes hemodynamic compromise. The classic presentation described by Beck in 1935 includes (1) hypotension, (2) rising central venous pressure, and (3) a small, quiet heart. As pericardial volume increases, intrapericardial pressure undergoes a slow ascent until reaching the limits of extensibility, at which point a rapid rise in pressure occurs associated with cardiac chamber compression ( Fig. 28.1 ). The rate of fluid accumulation, effectiveness of compensatory mechanisms, and stiffness of the pericardium affect the inflection point of the intrapericardial pressure curve and thus the severity and timing of clinical manifestations. Therefore a slowly developing pericardial effusion (such as in a systemic inflammatory state) may result in large volumes of fluid before significant cardiovascular effect, whereas a rapidly accumulating process such as a traumatic intrapericardial hemorrhage may quickly cause hemodynamic collapse.
Progression of hemodynamic changes in tamponade proceeds in a well-described fashion. Increased pericardial pressure leads to increased cardiac diastolic pressure, reduced myocardial transmural pressure, smaller cardiac chambers with decreased compliance, and decreased preload and cardiac output. A compensatory sympathetic response results in tachycardia, increased venous tone, and baroreflex-induced systemic arteriolar vasoconstriction to compensate for the hypotension. Jugular venous distention, hepatomegaly, and elevated central venous pressures may be evident. Pulsus paradoxus, defined as an exaggerated decrement in the arterial systolic pressure (>10 mm Hg) during inspiration, occurs in tamponade primarily due to bulging of the interventricular septum into the left ventricle (LV) as the right heart fills and the RV free wall is inhibited from expanding anteriorly due to pericardial constraint. Exceptions to this occur when abnormal communications exist between cardiac chambers (atrial septal defect; patent foramen ovale), when there is alteration of normal ventricular pressure-volume relations (hypertrophic cardiomyopathy, severe aortic stenosis, aortic insufficiency), or when there is decreased intravascular volume (low-pressure tamponade). Characteristic echocardiographic findings in pericardial effusion and tamponade are depicted in Fig. 28.2 .
Other characteristic findings in pericardial effusion and tamponade include low-voltage QRS complexes on electrocardiogram and electrical alternans, which is beat-to-beat alteration of the QRS complex amplitude and axis due to the swinging motion of the heart within a large pericardial effusion ( Fig. 28.3 ). Electrical alternans is considered specific but not sensitive for tamponade. The central venous pressure waveform in tamponade classically shows an exaggerated x descent in ventricular systole with a diminished or absent y descent in ventricular diastole ( Fig. 28.4 ). This is because in tamponade the intraluminal atrial or ventricular diastolic pressures (i.e., the downstream pressures for venous return) are effectively determined by the pericardial pressure. Ejection of blood in ventricular systole causes a decrease in total intrapericardial volume and subsequent antegrade venous flow, whereas pericardial and atrial pressures remain elevated and unchanged in ventricular diastole as intrapericardial volume transfer occurs through the atrioventricular (AV) valves.
Management.
Management of pericardial effusion consists of treating the underlying disease, with drainage performed when needed for diagnostic purposes or if progression to tamponade appears imminent. Management of tamponade consists of immediate drainage of pericardial fluid via percutaneous catheter pericardiocentesis or surgical pericardotomy/pericardectomy. Volume loading with crystalloid or colloid solution may be considered as a temporizing measure to augment preload, but definitive therapy should not be delayed. Diuretics may worsen the patient’s condition by further reducing preload. Inotropic agents may be considered but may have limited effect in the setting of an already brisk endogenous sympathetic response. Endotracheal intubation and mechanical ventilation should be approached with extreme caution because further cardiovascular compromise can occur with increased intrathoracic pressure due to increased lung volumes, further reducing systemic venous return and increasing total external constraint of the heart.
Pericardiocentesis should be performed in an intensive care unit (ICU), operating room, or catheterization laboratory with experienced personnel and resuscitation supplies present and echocardiographic guidance if possible. An illustration of the technique of pericardiocentesis using the Seldinger technique and a pigtail catheter is depicted in Fig. 28.5 . The patient should be positioned in the semiupright position. Local anesthesia with 1% or 2% lidocaine may be used to facilitate puncture; systemic sedation/anesthesia should be used judiciously due to risk of cardiovascular collapse even with administration of agents such as fentanyl or ketamine. An 18-gauge needle long enough to reach the effusion (often 6 cm) is used, with the best site of puncture and angulation determined echocardiographically as the point at which the largest fluid accumulation is closest to the body surface. In an emergency the subxiphoid approach may be used with approximately 15-degree angulation off the skin directed toward the left shoulder. In a retrospective series of 94 procedures in 73 pediatric patients, the optimal site of puncture by echocardiographic guidance was determined to be para-apical (68.1%), subxiphoid (17%), left axillary (5.3%), left parasternal (3.2%), right parasternal (2.1%), posterolateral (1.1%), or unspecified (3.2%).
Continuous suction is held until fluid is aspirated, at which point a 0.035- or 0.038-inch guide wire is advanced, followed by dilation and insertion of a 7 Fr or 8 Fr pigtail catheter, typically left in place for ongoing drainage until the underlying cause is resolved (see Fig. 28.5 ). Fluid may be sent to the laboratory for cell count with differential, determination of glucose, protein, and lactate dehydrogenase levels, and bacterial/fungal culture if further characterization is needed. A cardiac surgeon should be immediately available during the procedure because complications of pericardiocentesis can include laceration/perforation of the myocardium or coronary vessels with hemopericardium; however, intrapericardial fluid may often be bloody even without cardiac puncture (56% of cases in one study). Bloody pericardial fluid is nonclotting; additionally, bloody pericardial fluid usually sinks to the bottom of a gauze sponge, whereas blood forms clots on the surface of the gauze. Approximately 2 to 5 mL of agitated saline may be instilled under echocardiographic guidance to confirm placement. Other complications of pericardiocentesis include air embolism, pneumothorax, pulmonary edema, arrhythmias (most commonly vasovagal bradycardia), or puncture of the peritoneal cavity.
After placement the pericardial catheter may then be flushed with saline to prevent catheter plugging, and intermittent drainage (frequency dependent on reaccumulation rate and hemodynamics; usual minimum, every 2 hours) is followed by sterile saline catheter flush to maintain catheter patency. This is preferred over continuous drainage for maintenance of catheter patency. Balloon pericardiotomy has been used in adults, and there is some experience in children. Pericardial effusions also may be drained surgically via a subxiphoid approach or laparoscopy. A pericardial window is a communication created surgically, typically between the pericardial and pleural space. This allows ongoing drainage of effusion and may be used for recurrent effusions (such as in malignancy) or to prevent tamponade after cardiac surgery.
Inflammatory Heart Disease—Pericarditis
Overview
Pericarditis is inflammation of the pericardium, which may be acute, incessant (lasting for ≥4 weeks to <3 months without remission), recurrent (symptoms after a documented first episode and a symptom-free period of ≥4 weeks), or chronic (>3 months). Acute pericarditis is an infrequent cause of chest pain among pediatric emergency department visits (<0.2%) and among pediatric cardiology consultations for chest pain at a tertiary center (≈5%) and may come to the attention of an intensivist only when associated with a severe systemic disease, a postoperative state, and/or a significant pericardial effusion. Proposed diagnostic criteria for pericarditis include two or more of the following: (1) chest pain, classically pleuritic and relieved by sitting up and leaning forward, (2) pericardial friction rub, (3) new widespread ST elevation or PR depression on electrocardiogram (ECG) (60% or more), and (4) new or worsening pericardial effusion (approximately 80%). Additional supporting features include elevated markers of inflammation (white blood cell count, C-reactive protein [CRP], erythrocyte sedimentation rate [ESR]) and evidence of pericardial inflammation on imaging tests such as computed tomography (CT) or cardiac magnetic resonance imaging (MRI), if performed. Assessment of markers of myocardial injury (creatine kinase [CK]-MB, troponin) as well as transthoracic echocardiography is important when pericarditis is strongly suspected in critically ill patients due to the risk of concurrent coronary syndrome and/or myocarditis. Chest x-ray examination results are generally normal in isolated pericarditis but may show an enlarged cardiac silhouette (water bottle sign) if a significant (generally >300 mL in an adolescent/adult) pericardial effusion is present.
Etiology
Pericarditis may be infectious (including bacterial, viral, mycobacterial, or fungal) or noninfectious (see Box 28.1 ). Viral causes (especially coxsackievirus or echovirus) remain commonly implicated in acute pericarditis in otherwise healthy children, although the true incidence is difficult to determine due to incomplete detection. Purulent pericarditis may be associated with other infections such as septic arthritis, osteomyelitis, pneumonia, and pyelonephritis, with organisms discussed previously in discussion of purulent pericardial effusion. Tuberculosis is a common cause of pericarditis worldwide, especially in patients who also have HIV. A large proportion of patients with pericarditis are labeled idiopathic in multiple series (40% to 90% of all cases of pericarditis) ; however, in one multicenter database study of hospitalized children with pericarditis and pericardial effusion, only 5% were classified as idiopathic after an inpatient workup. In this study, cardiac causes (such as post cardiac surgery) were most commonly identified (54%), followed by neoplastic and renal causes (each 13%), contiguous infection or trauma (6%), and rheumatologic disease (5%). Antiheart, antiintercalated disk, antimyolemmal, antifibrillary, and antinuclear antibodies have been detected in patients with idiopathic recurrent pericarditis and/or chronic pericardial effusion, suggesting that autoimmune responses develop in predisposed individuals following exogenous triggers. Recurrent pericarditis, especially in the context of a family history of inflammatory disorders, is being increasingly recognized as a prominent feature of tumor necrosis factor receptor–associated periodic syndrome (TRAPS) and familial Mediterranean fever.
The postpericardiotomy syndrome is an important cause of noninfectious pericarditis and pericardial effusion in the ICU that may occur after any open-heart procedure that involves opening of the pericardium. Features include signs and symptoms of pericarditis (chest pain, pericardial friction rub, patient irritability), pericardial effusion, and fever occurring more than 72 hours postoperatively. In a single-center retrospective study of children following atrial septal defect repair, 28% were diagnosed with postpericardiotomy syndrome within the first year, with a median time of diagnosis 8 days postoperatively. Adult studies describe an incidence of postpericardiotomy syndrome of 9% to 24% after cardiac surgery, with more than 50% of cases occurring within 2 weeks postoperatively and more than 80% to 90% within 3 months. Pleural incision is described as a risk factor for postpericardiotomy syndrome, and pleural effusion is also commonly present. Pathophysiology of the postpericardiotomy syndrome is hypothesized to involve an autoimmune inflammatory response to pleuropericardial antigens, with perioperative viral infections also having a potential role. Treatment is similar to that for other patients with pericarditis and pericardial effusion, with additional attention to any postoperative concerns regarding the surgical repair or cardiac function. Corticosteroids and aspirin have been shown to be ineffective as prophylaxis for postpericardiotomy syndrome in children, whereas colchicine has shown a preventative effect in adults.
Management
Among critically ill patients, immediate management of pericarditis is directed at draining symptomatic pericardial effusions and treating associated systemic disease. For idiopathic, immune-mediated, or viral pericarditis, the classic first-line therapy has been nonsteroidal antiinflammatory medication (such as aspirin or ibuprofen) for 1 to 2 weeks or until symptom resolution. Colchicine and anti–interleukin-1 therapy with anakinra may be considered in recurrent or refractory cases. Corticosteroids are not recommended unless indicated for a concurrent disease process such as an autoimmune disorder, due to concerns for adverse effects, prolonged disease course, and increased risk of recurrence. Azathioprine or IV immunoglobulin (IVIG) have also been used; however, data are limited regarding these therapies.
Outcome
Pericarditis often resolves with treatment of the initial insult or underlying condition; however, recurrence may occur in approximately 15% to 30% of cases in adults. In one multicenter series of 110 children with recurrent pericarditis, median age was 13 years, and the most common symptoms at recurrence were chest pain (94%) and fever (76%). The initial cause of pericarditis in this series was most often idiopathic or viral (89%) or postpericardiotomy (9%).
In some patients, pericardial inflammation may progress to fibrosis, resulting in constrictive pericarditis. In a large prospective cohort of adults after acute pericarditis, 1.8% developed constrictive pericarditis over a median 6-year follow-up, with less than 0.5% of patients with viral or idiopathic pericarditis progressing to constrictive pericarditis and 8.3% of patients with other causes, with the highest risk for those with bacterial or tuberculous pericardial disease. In limited data in children, bacterial pericarditis also appears to be the most common cause of constrictive pericarditis. Constrictive pericarditis may be similar to pericardial effusion clinically with signs and symptoms of systemic venous hypertension (such as jugular venous distention, hepatomegaly, ascites, protein-losing enteropathy); however, cardiomegaly is not seen on radiograph, and pulmonary edema or pulsus paradoxus typically does not occur. The central venous pressure waveform in constrictive pericarditis exhibits a prominent y descent (see Fig. 28.4 ) with filling in early ventricular diastole, in contrast to the filling primarily in ventricular systole (prominent x descent) in tamponade. This difference is thought to be related to the complex physiology caused by the fibrotic pericardium forming a tight sac around the heart, which causes more “uncoupled” constraint of local pericardial surface pressures rather than the more uniform or “coupled” pericardial surface pressure across chambers that occur in tamponade.
The combination of echocardiography and CT scan or magnetic resonance imaging (MRI) may help establish the diagnosis of constrictive pericarditis. Pericardial biopsy is sometimes needed. Pericardiectomy is the only effective therapy and should be undertaken early before the fibrotic process also has affected the myocardium. The patient’s rate of recovery will depend on the extent of comorbid disease and especially the degree of coexistent myocardial fibrosis.
Inflammatory Heart Disease—Myocarditis
Overview
Myocarditis is defined clinically as inflammation of the heart muscle that most often results from common viral infections and postviral immune-mediated responses. There is no clear universally accepted clinical definition of acute myocarditis, and often the diagnosis is presumed upon clinical presentation and noninvasive diagnostic tests such as cardiac echocardiography or cardiac MRI. Endomyocardial biopsy (EMB) remains the gold standard for in vivo diagnosis of myocarditis, but its application in pediatrics is inconsistent because there is concern due to the perceived potential associated complications in such vulnerable patients. The Dallas criteria established in 1986 serve to define active myocarditis based upon histopathologic criteria, the presence of infiltrating lymphocytes, and myocytolysis; however, these criteria have been criticized as underestimating the incidence of myocarditis due to large interobserver variability. Nonetheless, the diagnostic and prognostic benefits of EMB results have been demonstrated. The development of new molecular techniques such as micro-RNA (miRNA) profiling, nested polymerase chain reaction (PCR), and in situ hybridization has improved the accuracy of diagnosis and prognostic value of EMB sampling, allowing for improved definitions of myocarditis, including the less prevalent subtypes such as eosinophilic and giant cell myocarditis. Despite the value of EMB in adult studies we are seeing a trend away from its use in pediatrics with an increased reliance upon cardiac MRI.
The clinical manifestations of myocarditis are heterogeneous, ranging from virtually asymptomatic cases with vague signs and symptoms to severe myocardial destruction by virus and immune cells with New York Heart Association (NYHA) class IV symptoms to cardiogenic shock and arrhythmia with reports of mechanical support in as much as 23% and mortality as high as 7.2%. Manifestations include flu-like symptoms with LV systolic dysfunction associated with signs of inflammation such as leukocytosis, elevated ESR, and increased levels of cardiac troponin and creatine kinase.
A study using International Classification of Diseases (ninth revision) codes estimated the global prevalence of myocarditis to be approximately 22 of 100,000 patients annually. The potential of myocarditis to be an insidious disease with lack of symptoms makes it difficult to ascertain its true epidemiology. Myocarditis is a major cause of sudden unexpected death ; the American Heart Association (AHA) and the American College of Cardiology (ACC) rank myocarditis as the third leading cause of sudden cardiac death in competitive athletes. Approximately 1% to 9% of routine postmortem examinations demonstrate evidence of myocarditis. A significant discrepancy exists between myocarditis diagnosed in autopsy studies and myocarditis noted clinically because many cases are subclinical and never manifest symptoms severe enough to warrant medical care.
Neonates and children exhibit a more fulminant myocarditis and are typically more susceptible to virus-induced pathogenesis than adults. Many studies report a greater prevalence and severity in men, speculated to be caused by a protective effect of natural hormonal influences on immune responses in women when compared to men. In a retrospective study that reviewed 5 years of data from the Pediatric Health Information System database, there was a bimodal age distribution for acute myocarditis, with a peak in infancy and a similar peak in mid–teenage years.
The outcome of myocarditis can be predicted based upon initial clinical presentation. Patients with acute myocarditis and preserved LV ejection fraction have a good prognosis and a high rate of spontaneous improvement without sequelae. Patients with fulminant viral myocarditis as defined by severe hemodynamic compromise requiring inotropic support or mechanical support also have an excellent long-term prognosis and are more likely to experience complete recovery if aggressive management is initiated in a timely manner. Nonfulminant myocarditis may be acute or chronic and so may progress in a more insidious manner with a less predictable long-term outcome with potential for complete resolution or progression to dilated cardiomyopathy (DCM), for which a subset may require transplantation. Virus has been isolated from the hearts of both adults and infants with acute cardiac disease. With the identification of the enterovirus genome in the myocardium of patients with myocarditis and DCM, a causal link between viral myocarditis and DCM has been widely accepted. It is estimated that roughly 10% to 15% of cases of DCM are due to biopsy-proven myocarditis. It is presumed that it is the adaptive and maladaptive response to such myocardial inflammation that determines the fate of those with myocarditis as to whether resolution occurs or a chronic ongoing immune response commits the patient to chronic HF. This is also the area where more recent attention has been focused toward the application of antiviral therapies to eradicate persistent viral states with immunosuppression that may potentially lead to treatment of such patients with chronic HF. Understanding the cellular and molecular processes responsible for such responses may provide further insight into developing such therapeutic algorithms.
Etiology
The spectrum of possible infectious and noninfectious causes of myocarditis is similar to that of pericarditis ( Box 28.2 ). Infectious agents may produce cardiac damage by direct invasion of myocardial cells, an abnormal immunologic response of the host, or in rare cases, toxin production by the invading organism (e.g., Corynebacterium diphtheriae ). Despite refinements in identification and viral-isolation techniques and the increased use of PCR, many cases of myocarditis remain idiopathic with no identifiable infectious agent isolated.
RNA Viruses
Coxsackievirus A, B
Echovirus
Hepatitis A
Influenza A, B
Lymphocytic choriomeningitis virus
Measles
Mumps
Poliovirus
Respiratory syncytial virus
Rhabdovirus
Rhinovirus
Rubella
DNA Viruses
Adenovirus
Cytomegalovirus
Epstein-Barr
Hepatitis B
Herpesviruses
Parvovirus
Varicella
Variola
Retroviruses
Human immunodeficiency virus
Viral Myocarditis
In cases in which the cause can be established, the majority of cases of myocarditis in North America have a viral cause. Although several viral agents have been implicated as possible causes of myocarditis ( Table 28.1 ), enteroviruses (members of the Picornaviridae family) have been the most frequently identified. Of these agents, coxsackievirus B is the culprit in more than half of the cases. Other enteroviruses associated with myocarditis include enteric cytopathic human orphan (ECHO) E6, E9, E11, and E22 serotypes. As more techniques have been developed, such as PCR and in situ hybridization, we have seen a shift in the typically identified enterovirus and adenovirus to mainly parvovirus B19 and human herpesvirus 6. Our understanding of the molecular mechanisms of myocarditis, however, largely come from animal and tissue cell culture studies of coxsackievirus B-3 (CVB3). Using CVB3 as a model system, we will discuss our current understanding of disease progression.
| Infectious Agents | Noninfectious Causes |
|---|---|
Viruses
Bacteria
Parasitic Diseases
| Toxins
Autoimmune Diseases
|
Pathogenesis.
The pathogenesis of viral myocarditis is both caused by direct injury mediated by viral infection and secondary to the immune response of the host. At the cellular and tissue levels the pathologic progression of viral myocarditis can be divided into three stages. The acute viremic stage, which typically occurs over the first 4 days of disease, is characterized by viral entry and replication with consequential activation of the innate immune response. The subacute phase is characterized by the adaptive immune response with inflammatory cell infiltration over days 4 to 14, with cardiac remodeling being the major feature of the chronic phase, which may lead to DCM in some patients.
The acute viremic stage is characterized by early cardiomyocyte damage associated with prominent viral replication. Viral invasion first begins when the circulating virion enters the myocyte via receptor-mediated endocytosis ( Fig. 28.6 ). The first step in this process is the attachment of the virus to a myocyte cell-surface adhesion molecule. The prototype adhesion molecule in myocarditis is the coxsackie adenovirus receptor (CAR). The predominance of coxsackievirus B group and adenovirus as etiologic agents for myocarditis probably results from the fact that they share a common myocardial cell-surface receptor. After the virus gains entry into the myocyte (see Fig. 28.6 ), viral genome replication occurs, producing a negative-strand viral RNA intermediate that serves as a template for transcription of progeny genomes. The release of progeny virion requires that cell destruction be orchestrated, which the virus achieves by disrupting cell integrity and architecture and taking control of cellular apoptosis ( Fig. 28.7 ).
The immune response to viral infection is crucial in the pathogenesis of myocarditis. Several observations support this theory, including the demonstration of immune complexes in the myocardium, the occurrence of myocarditis in autoimmune diseases such as systemic lupus erythematosus (SLE), and the clinical response of some patients to immunosuppressive and other immune-modulating therapies. The details of the immune response to myocardial enteroviral infection have emerged from numerous experiments using murine models.
The initial phase of acute myocarditis that occurs within the first 4 days of viral infection is characterized by viremia and the activation of the innate immune response.
Cell-mediated immunity affects the host response during the subacute phase of myocarditis (days 4 to 14) ( Fig. 28.8 ). Whereas cell-mediated immunity is necessary for viral clearance, persistent T-cell activation may aggravate myocardial damage and lead to chronic myocarditis or DCM.
Clinical Manifestations.
Manifestations of myocarditis range from an asymptomatic patient to obvious signs of congestive HF. Infants with HF may present with history of poor feeding, respiratory distress, listlessness, poor weight gain, or irritability. Common adult symptoms such as paroxysmal nocturnal dyspnea and orthopnea are uncommon in pediatric patients. Most older children are somewhere in between these two extremes, with an acute febrile illness and a minor degree of cardiovascular involvement, such as tachycardia or isolated ECG changes. Abdominal pain, anorexia, nausea, and vomiting are often observed and are likely due to liver capsule distention from hepatomegaly and/or intestinal venous congestion. The most common cardiovascular symptom is chest pain, which may be the sole complaint in some patients. Other presentations include isolated ventricular ectopy without obvious signs of cardiac failure. Syncope may result from ventricular tachycardia or less frequently from high-grade AV block. The ECG may be normal in between spells, making it more difficult to establish the diagnosis. Sudden death is the most severe presentation, and in such cases the diagnosis of myocarditis is made only at autopsy. Myocarditis and cardiomyopathy also can be causes of sudden death during general anesthesia and surgery among patients not previously suspected of having heart disease.
In most cases the multisystem effects and symptoms of the underlying viral illness outweigh the symptoms related to the cardiovascular system. However, signs and symptoms related to cardiac involvement may be seen. A tachycardia out of proportion to the fever may be present. Signs of cardiac failure such as cardiomegaly, pulsus alternans, hepatomegaly, and third or fourth heart sounds may be present. Decreased exercise tolerance, dyspnea, and easy fatigability may be the only complaints in some children. Associated symptoms such as myalgias, pneumonitis, exanthems, or lymphadenopathy may be suggestive of a viral cause.
On physical examination the child may appear anxious, and sinus tachycardia is usually present. Sweating is common in infants. Pallor and cool extremities may be present and are often associated with poor peripheral pulses and prolonged capillary refill. Resting tachypnea and retractions are common. Crackles are exceedingly rare in infants and young children with HF, unlike adults, even when pulmonary edema is present. Wheezing is more likely to be present, and periorbital edema is more common in children than peripheral edema.
The presentation in infants stands out because it is often very acute and life threatening. Irritability and poor feeding give way after a few hours to signs of overwhelming shock and HF. Respiratory distress, cyanosis, thready pulses, and gallop rhythm all suggest an infant in extremis. Because of the similarity of this presentation to ductal-dependent left-sided obstructive lesions such as coarctation, prostaglandin E 1 is sometimes given before the diagnosis of myocarditis in very young infants.
Laboratory and Electrocardiographic Findings.
The majority of laboratory and ECG findings of myocarditis are nonspecific. Elevations of creatine phosphokinase and lactate dehydrogenase are present in 50% to 75% of cases with ST-segment abnormalities. CK has been previously considered a preferred biomarker to screen for myocarditis; however, its sensitivity in late diagnosis is poor because studies show that CK decreases steadily to basal levels by 36 hours. Nonspecific serologic markers of inflammation, including leukocyte count and CRP, may be elevated in suspected myocarditis. The complete blood count reveals a leukocytosis in 50% of patients. The presence of eosinophilia should prompt the search for a parasitic cause of the myocarditis. Another nonspecific indicator of an inflammatory process is an elevated ESR. Nonelevated levels of leukocytes, CRP, and ESR fail to rule out an inflammatory response within the myocardium. Levels of B-type natriuretic peptide and N-terminal pro-B-type natriuretic peptide can be elevated in myocarditis, and elevated levels may aid in distinguishing cardiac from noncardiac reasons for children who present with respiratory symptoms. Cardiac troponin I (cTnI) levels now are the most sensitive diagnostic test for biopsy-proven myocarditis. In addition, cTnI levels may be of prognostic value because they correlate with ejection fraction and clinical HF.
ECG changes in myocarditis also are nonspecific and only support the diagnosis. The ECG most commonly reveals nonspecific ST-T-wave abnormalities. ECGs provide a convenient tool for risk stratification and initial screening, but provide little in diagnostic value. Ukena et al. investigated the prognostic value of ECG in patients suspected of having myocarditis and reported that a prolonged QRS duration may be an independent predictor for cardiac death or heart transplantation. Other abnormalities include ventricular and atrial ectopy, conduction defects, sinus tachycardia, and T-wave inversion. The presence of ventricular ectopy in a febrile, irritable infant represents myocarditis until proven otherwise and should prompt immediate admission to an ICU. In infants with an ECG pattern consistent with acute myocardial infarction, anomalous origin of the left coronary artery from the pulmonary artery should be considered and excluded. Occasionally the ECG may point toward an alternate diagnosis. For example, Pompe disease has a characteristic ECG pattern of a short PR interval with very large precordial voltages. Similarly, the presence of ventricular preexcitation (Wolff-Parkinson-White syndrome) may raise the question of unrecognized sustained tachyarrhythmias as a cause of ventricular dysfunction.
Diagnostic Approach.
Because of the wide clinical spectrum in presentation and the diversity of etiologic agents, the diagnostic approach to the patient with myocarditis must include a strategy that attempts to include several etiologic possibilities. A thorough history is imperative and should focus on the possibility of toxic ingestion (including illicit and over-the-counter medications), trauma, and possible exposure to infectious agents (including a complete travel history). Past history should include immunization status, because several infectious diseases of childhood (diphtheria, polio virus) are included in the differential diagnosis. Additionally, the history may reveal signs or symptoms of previously undiagnosed collagen vascular disorder or inflammatory bowel disease.
Echocardiography is a useful measurement tool in the diagnostic assessment of suspected myocarditis and aids in the process of ruling out other causes of HF, such as structural heart disease. Echocardiography may even contribute to the assessment of prognostication in such patients through two-dimensional speckle tracking that evaluates myocardial strain and strain rate. Echocardiography investigates cardiac chamber size, wall thickness, and systolic and diastolic function, but it cannot provide direct evidence to confirm myocarditis. Patients with acute myocarditis often show only a poorly functioning LV with minimal dilation with or without regional wall motion abnormalities. There may be ventricular thickening secondary to myocardial edema, and left atrial enlargement may not be prominent, even when mitral valve regurgitation is present, suggesting an acute disease process.
The diagnosis of viral myocarditis is generally based on circumstantial evidence such as a recent viral infection and the sudden onset of cardiac dysfunction while ruling out other diagnostic possibilities. Although viral cultures (pharyngeal, rectal) should be obtained, the “window of opportunity” for viral isolation is narrow, and therefore isolation of the virus is often not possible. Use of PCR may allow identification of the viral genome even when cultures are negative. Supportive evidence for a viral cause can be obtained by a fourfold or greater increase in antibody titer between acute and convalescent sera. In addition to these titers, the presence of Epstein-Barr virus infection should be investigated with serologic study. Routine bacterial and fungal cultures of blood should be obtained in addition to serologic examination to rule out treponemal infection (Venereal Disease Research Laboratory or rapid plasma regimen).
Pathologic confirmation of myocardial inflammation continues to be required for definitive diagnosis of myocarditis. Edwards and associates suggested a criterion for myocarditis that includes five or more lymphocytes per 20 high-power fields during histologic examination of the endomyocardial specimen. Several biopsy specimens (5 to 10) may be needed because the inflammatory process is focal. These criteria were subsequently refined with the introduction of the Dallas criteria, which state that active myocarditis is present when routine light-microscopy examination of the biopsy specimen reveals lymphocytic infiltration and myocytolysis. Borderline or ongoing myocarditis exists in the presence of lymphocytic infiltration alone without myocytolysis. The specimen is considered negative if both lymphocytic infiltration and myocytolysis are absent. Although the Dallas criteria remain in widespread use, most clinicians believe that they underestimate the true incidence of myocarditis, probably because of the patchy nature of myocardial inflammation and the high degree of interobserver variability.
EMB provides relatively high diagnostic and prognostic value in suspected myocarditis patients with reported complication rates of less than 1% among adults. Furthermore, it has demonstrated benefits in aiding in clinical decision making and in ruling out disorders important in the differential, such as endocardial fibroelastosis, Pompe disease, and eosinophilic and giant cell myocarditis. Additional testing applied to the histologic sample such as RNA/DNA hybridization and PCR allow for more sensitive means to detect the offending virus. Despite these benefits, there has been a trend away from the use of EMB in pediatrics. The major concerns leading to its decrease in use are about its safety and potential for adverse events in a very vulnerable patient. Pediatric patients who are considered for EMB are typically younger, have more impaired hemodynamics, and typically require ICU-level support. In a retrospective multi-institutional study looking at safety and efficacy of EMB among pediatric patients undergoing cardiac catheterization for suspected myocarditis, cardiomyopathy, and new-onset HF found a threefold increase in adverse events versus a posttransplant comparison group with a high-severity adverse event rate as high as 5%. A major concern for EMB is cardiac perforation, which has been reported to be as high as 4% in children less than 1 year of age with primary risk factors of weight less than 8 kg and age less than 6 months.
A report analyzing the Pediatric Health Information System database shows a statistical trend toward fewer EMBs and more cardiac MRI scans in children with myocarditis in recent years. In a survey of centers assessing the practice of EMB, less than 33% of patients with suspected myocarditis were referred for EMB, with a post hoc survey that reflected such a trend in practice in favor of a greater tendency to use cardiac MRI. The 2007 AHA/ACC/ European Society of Cardiology (ESC) scientific statement also does not support the routine use of EMB for the diagnosis of myocarditis. So despite literature that demonstrates a role for and benefit from the use of EMB, neither is it recommended nor is there evidence that it is commonly performed in pediatric patients.
Cardiac MRI has evolved as a valuable noninvasive clinical tool for the diagnosis of myocarditis. MRI gained acceptance as it demonstrated reliability in tissue characterization of cardiac allograft rejection, which histologically appears identical to acute myocarditis. It is the myocardial inflammation that presents itself as an attractive target for cardiac MRI-based imaging. Contrast enhancement (CE) is a more sensitive technique of cardiac MRI and can detect areas of myocardial damage in myocarditis. The International Consensus Group on CMR Diagnosis of Myocarditis published recommendations on the indication, implementation, and analysis of appropriate cardiac MRI techniques for noninvasive diagnosis of myocarditis (Lake Louise criteria). The combined use of three different cardiac MRI techniques is suggested. The T2-weighted edema imaging is a tool for evaluating the presence of acute myocardial edema ( Fig. 28.9 ). Further evaluation is performed looking at hyperemia and muscular inflammation using ECG-triggered T1-weighted images obtained within the first few minutes after gadolinium-diethylenetriaminepentaacetate (Gd-DTPA) infusion, deemed “myocardial early gadolinium enhancement.” This sequence has been prone to artifactual interference that decreases specificity. Lastly, a method known as late gadolinium enhancement (LGE) imaging looks at evidence of irreversible myocardial injury (i.e., necrosis and fibrosis) by a T1-weight segmented inversion-recovery gradient-echo sequence. LGE images reveal two common patterns of myocardial damage: either an intraluminal rim-like pattern in the septal wall or a subepicardial (patchy) distribution in the free LV lateral wall (see Fig. 28.9 ). In patients with acute myocarditis, areas of CE are often located in the lateral wall originating from the epicardial quartile, even if the pattern of myocardial injury is influenced by viral infection. It has been suggested that because of the ability of CE to identify areas of myocardial inflammation it can serve as a guide for EMB sampling when necessary, thus enhancing the diagnostic accuracy of biopsy specimens. LGE imaging does not allow differentiation between acute and chronic inflammation; it only represents damaged myocardium, making interpretation dependent upon clinical context.
