CHAPTER 66: PERICARDIAL DISEASES

Disorders of the pericardium constitute a diverse group of pathologies, ranging from benign congenital structural abnormalities, such as pericardial cyst, to life-threatening entities, such as cardiac tamponade and constrictive pericarditis. Although pericardial diseases are not uncommon in daily practice, physicians including cardiologists tend to be less familiar with managing these diseases. This can lead to a delay in the diagnosis (patients with constriction might have been treated for “refractory” heart failure for years) or incomplete or inadequate treatment. We provide herein a concise and practical approach to pericardial diseases for cardiovascular practitioners. We have also incorporated the recommendations from the 2015 European Society of Cardiology (ESC) Guidelines on the Diagnosis and Management of Pericardial Diseases¹ (a summary of the most relevant recommendations is also provided in the chapter).

The human pericardium has two distinct layers; the serosa is composed of a single column of mesothelial cells that surrounds all four cardiac chambers and the proximal great vessels and reflects on itself to form the inner surface of the fibrosa, a fibrocollagenous structure (Figs. 66–1 and 66–2). This monolayer of serosal cells covering the surface of the heart and epicardial fat is also called the visceral pericardium, whereas the fibrosa and the reflection of the serosa make the parietal pericardium. Pericardial fluid, an ultrafiltrate of the plasma, accumulates between the serosal layers; normal pericardial fluid volume ranges from 15 to 50 mL. As opposed to the visceral pericardium, the fibrosa does not cover the cardiac surface in its entirety; pericardial reflections at the level of great vessels give origin to the pericardial sinuses. The two most important are the transverse sinus, which lies between the great arteries (aorta and pulmonary artery) and the great veins (pulmonary veins and vena cavae), and the oblique sinus, which has the shape of an inverted U and lies behind the left atrium as the fibrosa reflects around the pulmonary veins. Pericardial fluid can accumulate in these sinuses, increasing pericardial volume if necessary, and therefore, constituting the pericardial reserve (Fig. 66–3).

Most of the innervation of the pericardium occurs via the phrenic nerves (C4-C6), which course anteriorly; this is particularly relevant during pericardiectomy. Blood supply to the pericardium is provided by small branches from the aorta or the internal thoracic artery.

The pericardium is not essential for life; no adverse consequences follow congenital absence or surgical removal of the pericardium.^1,2,3,4 However, the pericardium serves many important (although subtle) functions (Table 66–1).^1,2,3,4 It limits distension of the cardiac chambers and facilitates interaction and coupling of the ventricles and atria. Thus, changes in pressure and volume on one side of the heart can influence pressure and volume on the other side. Limitation of cardiac filling volumes by the pericardium may also limit cardiac output and oxygen delivery during exercise.² The pericardium also influences quantitative and qualitative aspects of ventricular filling; the thin-walled right ventricle and atrium are more subject to the influence of the pericardium than is the more resistant, thick-walled left ventricle.

Although the magnitude and importance of pericardial restraint of ventricular filling at physiologic cardiac volumes remain controversial, there is general agreement that pericardial reserve volume (ie, the difference between unstressed pericardial volume and cardiac volume) is relatively small and that pericardial influences become significant when the reserve volume is exceeded. This may occur with rapid increases in blood volume and in disease states characterized by rapid increases in heart size (eg, acute mitral and tricuspid regurgitation, pulmonary embolism, right ventricular infarction). In contrast, chronic stretching of the pericardium results in “stress relaxation”; this explains why large, but slowly developing, effusions do not produce tamponade. In addition, the pericardium adapts to cardiac growth by “creep” (ie, an increase in volume with constant stretch) and cellular hypertrophy.

The pericardium serves a variety of other important functions. It prevents excessive torsion and displacement of the heart, minimizes friction with surrounding structures, and is an anatomic and immunologic barrier to the spread of infection from contiguous structures. The thin layer of pericardial fluid reduces friction on the epicardium and equalizes gravitational, hydrostatic, and inertial forces over the surface of the heart; therefore, transmural cardiac pressures do not change during acceleration or differ regionally within cardiac chambers. The pericardium also has immunologic, vasomotor, paracrine, and fibrinolytic activities.⁵ The mesothelium of the pericardium is metabolically active and produces prostaglandin E₂, eicosanoids, and prostacyclin; these substances modulate sympathetic neurotransmission and myocardial contractility, and may influence epicardial coronary arterial tone. Epicardial mesothelial cells may modulate myocyte structure and function and gene expression. The level of brain natriuretic peptide in the pericardial fluid is a more sensitive and accurate indicator of ventricular volume and pressure than either plasma brain natriuretic peptide or atrial natriuretic factor; it may play an autocrine-paracrine role in heart failure. Finally, the pericardial space has been used as a vehicle for drug delivery and gene therapy¹; studies using radiolabeled growth factors indicate that substances more consistently and reproducibly gain access to the coronary arteries via pericardial fluid than via endoluminal delivery.⁵

Pericardial diseases may be caused by either infectious or noninfectious agents (Table 66–2) and typically may be either an isolated process or a part of a systemic disease (eg, systemic inflammatory disease).^1,2,3,4 The pericardial response to an infectious or noninfectious noxa is rather nonspecific and is usually manifested as pericardial inflammation (pericarditis) with the possible increased production of pericardial fluid giving rise to pericardial effusion. If the pericardial inflammation persists, the evolution may be toward fibrosis and calcification with the development of constrictive pericarditis. On this basis, in clinical practice, pericardial diseases are usually manifested as the following “pericardial syndromes”: pericarditis (acute and recurrent), pericardial effusion, cardiac tamponade, constrictive pericarditis, congenital diseases of the pericardium, and pericardial masses¹; these conditions will be reviewed in this chapter.

Treatment of pericardial disease is also challenging in that there is a paucity of randomized, placebo-controlled trials (Level of Evidence: A) from which appropriate therapy may be selected and important clinical decisions assisted. Table 66–3 summarizes the main recommendations of the 2015 ESC guidelines.¹

Acute pericarditis is an inflammatory pericardial syndrome with or without pericardial effusion that can be caused by infectious or noninfectious causes (see Table 66–2). The following description refers to viral and idiopathic pericarditis without significant effusion. Specific forms of pericardial heart disease are reviewed later in this chapter.

History

Acute pericarditis is usually characterized by sharp retrosternal pain (> 85% of cases) that is aggravated by lying down and relieved by sitting up; its onset is frequently heralded by a prodrome of fever, malaise, and myalgia. The pain of pericarditis is often worse with inspiration and is difficult to distinguish from pleurisy; in some cases, it may radiate to the shoulders, arms, and jaw and may simulate “ischemic chest pain” (Fig. 66–4).

Physical Examination

The most specific physical sign can be the presence of a pericardial rub, which is identifiable in no more than a third of cases. It is a superficial scratchy or squeaking sound best heard with the diaphragm of the stethoscope applied firmly and with respiration suspended over the left sternal border and the cardiac apex and is heard better with the patient sitting up and leaning forward.^2,4 It is likened to the sound of walking on dry snow or the squeak of a leather saddle (see Fig. 66–4). Most pericardial friction rubs are independent of the respiratory cycle, but on occasion, they are louder during inspiration. The pericardial rub may be confined to ventricular systole, but most often includes a component during atrial systole and occasionally during ventricular diastolic filling, resulting in biphasic and triphasic rubs, respectively. Frequent examinations are necessary to detect a rub because of its evanescent nature; pericardial fluid does not prevent a friction rub.

In uncomplicated pericarditis, the jugular venous pressure usually remains normal. Ventricular third and fourth heart sounds indicate coexisting myocardial disease. The history and physical examination are also helpful in recognizing complications and in identifying underlying diseases associated with pericarditis. Depending on the etiology, there may be fever and other signs of inflammation or systemic illness.

Evaluation and Testing

Electrocardiography

On an electrocardiogram (ECG), the classical reported feature of acute pericarditis is widespread ST-segment elevation (Fig. 66–5), although either this sign or the earlier PR depression (as an atrial current of injury that is the earliest ECG sign) are an expression of concomitant subepicardial involvement and not “pure pericarditis” because the pericardium is electrically silent.^6,7,8 As opposed to other leads, aVR will show PR elevation (“knuckle sign”), another ECG sign of acute pericarditis.⁹

Presentation times and therapies may affect the ECG presentation; early presentation may show ST-segment elevation, whereas a normal ECG or an ECG with negative T waves (reflecting an ECG in evolution) can be seen in late presentations or chronic forms. In patients with a fast response to medical therapy and in mild forms, the ECG may be absolutely normal; thus, a normal ECG does not exclude pericarditis.^6,7

The ECG may either confirm the clinical suspicion of pericardial disease or first alert the clinician to the presence of pericarditis. Serial tracings may be needed to distinguish the ST-segment elevations caused by acute pericarditis from those caused by acute myocardial infarction or normal early repolarization. The ST-T wave changes in acute pericarditis are diffuse and have characteristic evolutionary changes.

The ECG evolves according to the classical four stages reported by Imazio et al⁸ in no more than 60% of patients. In the first stage, ST-segment elevations (which differ from ischemic ST-segment elevations by their upward concavity and seldom exceed 5 mm in height) typically occur within a few hours of the onset of chest pain and persist for hours or days. Depression of the PR segment (except in lead aVR) occurs in this stage and differentiates acute pericarditis from early repolarization variants. In the second stage, the ST segments return to baseline; at this point, the T waves may appear normal or exhibit a loss of amplitude. In the third stage, tracings show inversion of T waves. T-wave inversions may persist indefinitely, particularly with chronic pericarditis. The ECG normalizes in the variably present fourth stage.

Atrial arrhythmias complicate 5% to 10% of cases of acute pericarditis. The risk of cardiac arrhythmias is very low in simple acute pericarditis, that is, in the absence of myocarditis or structural heart diseases.^6,7,10,11

The ST-segment elevation seen in acute pericarditis can usually be distinguished from that of acute myocardial infarction by several differential features, which are reported in Table 66–4.

Although the ST-segment elevation in the early repolarization variant (common in young individuals, especially blacks, athletes, and psychiatric patients) may simulate the ECG of acute pericarditis, the former is distinguished by the absence of PR-segment depression and evolutionary ST-T wave changes. Moreover, the height of the J-point can be used to differentiate early repolarization from acute pericarditis: if the ratio between the height of the J-point and the height of the T wave is < 0.25, early repolarization is diagnosed.⁴

Laboratory Studies and Imaging

Nonspecific blood markers of inflammation, such as the erythrocyte sedimentation rate, C-reactive protein, and white blood cell count, usually increase in cases of acute pericarditis. Patients with extensive epicarditis and/or concomitant myocarditis occasionally have increases in serum cardiac isoenzymes (eg, troponins).

The chest-x-ray may be completely normal in the absence of concomitant pleuropulmonary disease or large pericardial effusion (> 300 mL); in such cases, the heart may assume a water bottle–shaped configuration (Fig. 66–6).¹ The chest radiograph may provide evidence of concomitant pleuropulmonary diseases (eg, tuberculosis, fungal disease, pneumonia, neoplasm).

The use of echocardiography for the evaluation of all patients with suspected pericardial disease was given a Class I recommendation by a 2003 task force of the American College of Cardiology, the American Heart Association, and the American Society of Echocardiography¹² and 2015 ESC guidelines on the diagnosis and management of pericardial diseases.¹ Echocardiographic identification of pericardial effusion confirms the clinical diagnosis of acute pericarditis (Fig. 66–7), but is not necessary for the diagnosis because pericarditis may occur without pericardial effusion (“dry pericarditis”).

A semiquantitative assessment of pericardial effusion is recommended by the 2015 ESC guidelines.¹ The assessment is performed by the end-diastolic distance of the echo-free space between the epicardium and parietal pericardium: small (< 10 mm), moderate (10-20 mm), and large (> 20 mm). A large (> 20 mm) pericardial effusion or cardiac tamponade identifies high-risk patients with an increased risk of a nonviral etiology or complications during follow-up.¹³

The clinical diagnosis of acute pericarditis is established when two of four clinical criteria are satisfied (Table 66–5). In atypical cases, additional supporting criteria are provided by elevation of markers of inflammation and evidence of pericardial inflammation by an imaging technique (eg, computed tomography [CT] or cardiac magnetic resonance [CMR]). The inflamed pericardium may be contrast enhanced on CT and show increased signal on T2-weighted imaging on CMR (as evidence of pericardial edema) and late enhancement after administration of gadolinium (as evidence of increased extracellular space that may be caused by edema and/or organizing pericarditis).^14,15

Treatment

Hospitalization is warranted for many patients who present with an initial episode of acute pericarditis and high-risk features such as elevated temperature (> 38°C), subacute onset, large pericardial effusion, cardiac tamponade, immunosuppression, recent trauma, oral anticoagulant therapy, and aspirin or nonsteroidal anti-inflammatory drug (NSAID) failure¹³ to determine the etiology and for monitoring (Table 66–6 and Fig. 66–8); close follow-up is important in the remainder of patients, and outpatient management is warranted.¹⁰

Establishing the exact cause of acute pericarditis is especially important in patients with high-risk features, a suspicion of a bacterial or neoplastic etiology, or failure of empiric anti-inflammatory therapy. An extensive evaluation is generally unnecessary in a young, previously healthy adult who presents with a viral syndrome, typical pericardial chest pain, and a pericardial friction rub and has a prompt response to empiric anti-inflammatory therapy.

The mainstay of therapy of simple uncomplicated pericarditis is aspirin or a NSAID plus colchicine to fasten the response to anti-inflammatory therapy and prevent recurrences.^{16,17,18,19,20,21,22,23} Colchicine is now a well-established therapy even for the first episode of pericarditis.^17,18,19,24 It is recommended without a loading dose using weight-adjusted dosing (eg, colchicine 0.5-0.6 mg once daily for patients < 70 kg and 0.5-0.6 mg twice a day for patients > 70 kg with a duration of therapy of 3 months) (Class I recommendation; Level of Evidence [LOE]: A).¹ These recommendations are based on randomized data from the Colchicine for Acute Pericarditis (COPE)¹⁸ and Investigation on Colchicine for Acute Pericarditis²⁴ trials, which showed that the use of colchicine in addition to anti-inflammatory therapy reduces recurrence rates as well as persistence of symptoms at 72 hours after initiation of therapy.

Proposed therapeutic schemes for acute pericarditis are reported in Table 66–7. The attack dose should be maintained until symptom resolution and normalization of markers of inflammation (eg, C-reactive protein).^21,22 Tapering is recommended after initial clinical remission according to 2015 ESC guidelines.¹

Corticosteroids should be prescribed only as second-line therapy in case of contraindication or failure of aspirin and NSAID and for specific indications (eg, pregnancy, systemic inflammatory disease already on steroids, renal failure, need to prevent interactions with other drugs, especially oral anticoagulants).

Prophylaxis against gastrointestinal bleeding with proton pump inhibitors is warranted in all patients to allow a safe therapy with high-dose aspirin or NSAID.

As a nonpharmacologic measure, avoidance of physical exertion and exercise is recommended until remission in nonathletes and for an arbitrary defined time of 3 months in athletes,^25,26 also according to the 2015 ESC guidelines.¹

Prognosis

The overall prognosis is related to the etiology.²⁷ Recurrences are reported to occur in 20% to 50% of patients, but the recurrence rate may be halved if empiric anti-inflammatory therapy of pericarditis is provided with the adjunct of colchicine.^18,24

The most feared complication is the evolution toward constrictive pericarditis that rarely occurs in idiopathic and viral pericarditis (< 1%), and it is usually more common for specific etiologies, such as autoimmune and neoplastic pericarditis (2%-5%) and especially tuberculous and purulent pericarditis (20%-30%).²⁷

Recurrent pericarditis is one of the most troublesome complications of pericarditis, occurring in one-third of cases. Use of colchicine and restriction of corticosteroid use in acute pericarditis may help to reduce the recurrence rate.

Most recurrences remain “idiopathic” without a definite etiology, although most recurrences are considered to be immune mediated. In clinical practice, a significant number of cases are related to the inappropriate therapy of the first episode of pericarditis, including low anti-inflammatory doses, short courses, nonprescription of colchicine, or use of corticosteroids, especially high doses (eg, prednisone 1.0 mg/kg/d).

A true recurrence occurs when there is a symptom-free interval from the previous episode of pericarditis. This time is usually 4 to 6 weeks in order to allow the completion of the anti-inflammatory therapy of pericarditis. In the absence of this symptom-free interval, the term incessant pericarditis is proposed rather than recurrent pericarditis because incessant pericarditis is characterized by continuous symptoms without remission. Such forms may directly progress to constrictive pericarditis. The term chronic is generally referred, especially for pericardial effusions, to disease processes lasting for > 3 months, and chronic pericarditis is an arbitrary term used by experts for disease lasting > 3 months, according to the 2015 ESC guidelines (Table 66–8).¹

The mainstay of the therapy for recurrences is similar to that of acute pericarditis: aspirin or an NSAID plus colchicine at the same doses recommended for acute pericarditis, but in this case, colchicine is recommended for 6 months (Class I recommendation; LOE: A).¹

Colchicine as adjunct to aspirin or an NSAID has been demonstrated to be safe and efficacious to improve the response to conventional anti-inflammatory therapy (aspirin/NSAID) and halve the recurrence rate either in first recurrences (data from Colchicine for Recurrent Pericarditis [CORE] and Colchicine for Recurrent Pericarditis [CORP] trials)^20,28 or in patients with multiple recurrences.²⁹

Corticosteroids should be considered as a second-line therapy for patients with contraindications for or failure of aspirin/NSAID or patients with specific indications (eg, pregnant patients, patients with a systemic inflammatory disease already on corticosteroids, patients with renal failure, patients on oral anticoagulant therapies to avoid interference with aspirin/NSAIDs). If used, low-to-moderate doses (eg, prednisone 0.2-0.5 mg/kg/d or equivalent) are indicated because high doses (eg, prednisone 1.0-1.5 mg/kg/d) are associated with a high rate of severe side effects (up to 25%) with more drug withdrawals, drug-related hospitalization, and recurrences. A slow tapering is also critical in order to allow corticosteroid withdrawal after remission (Table 66–9).³⁰

In more difficult cases, aspirin or NSAID, colchicine, and corticosteroids may be used together as a triple therapy to achieve a better control of symptoms.³¹

A small subset of patients with recurrent pericarditis (also called refractory recurrent pericarditis) develop corticosteroid dependence (ie, unable to withdraw or taper corticosteroids without a new relapse) and do not respond to colchicine. In such cases, anecdotal evidence supports the use of alternative drugs (eg, azathioprine, but especially human intravenous immunoglobulins and anakinra; Table 66–10).^32,33,34,35

After failure of medical therapy, the last option is pericardiectomy, which should be performed in centers with specific expertise in such surgery to achieve the best outcomes.³⁶

Idiopathic Pericarditis

In North America and western Europe (countries with a low prevalence of tuberculosis), acute pericarditis is most often idiopathic (80%-55% of unselected cases) and is typically a self-limited disease usually lasting 2 to 6 weeks. Small pericardial effusions occur commonly (up to 60% of cases), but cardiac tamponade is unusual (< 5% of cases). Heart failure caused by associated myocarditis and constrictive pericarditis is uncommon. These complications usually can be detected by clinical and echocardiographic evaluation. The clinical course and prognosis of individuals with pericarditis are otherwise largely determined by the presence and nature of any underlying disease.

Infectious Pericarditis

Viral Pericarditis

Viral pericarditis is the most common infectious type, although a definitive diagnosis from acute and convalescent (3 weeks) viral neutralizing antibodies is generally not helpful in a sporadic case of pericarditis. Viral isolation from pericardial fluid and in situ hybridization techniques have been used to identify a specific etiology. However, viral infection is often presumed rather than proved, and many cases are classified as idiopathic. Common viral infections causing acute pericarditis are those resulting from echovirus and coxsackievirus; however, several other viruses may cause pericarditis (see Table 66–2).

However, in immunocompetent patients, there are no proven antiviral therapies to offer, and the disease is usually self-limiting and resolving, while pericardial inflammation and pain are treated by aspirin or a NSAID and colchicine. Thus, it is absolutely not necessary to reach a specific viral etiology in clinical practice.

Bacterial Pericarditis

The most common cause of bacterial (purulent) pericarditis is tuberculous pericarditis. Tuberculosis is the leading cause of pericardial diseases in developing countries and all over the world. Purulent pericarditis is now relatively rare, especially in countries with a low prevalence of tuberculosis.³⁷

In developed countries, bacterial pericarditis is most often caused by streptococci, staphylococci, gram-negative rods, and, increasingly, anaerobic bacteria.³⁸ Haemophilus influenzae is an important cause in children. The increasing frequency of cardiac surgery and percutaneous interventions, selection-induced changes in the flora responsible for hospital-acquired infections, and prolonged survival of immunocompromised hosts (eg, human immunodeficiency virus [HIV], steroids) have changed the incidence and bacterial spectrum of purulent pericarditis. Predisposing factors include pericardial effusion, immunosuppression, chronic diseases (eg, alcohol abuse, rheumatoid arthritis, chronic kidney disease, malignancy), cardiac surgery, and chest trauma. Pericardial involvement often is unrecognized when it complicates systemic infection; unusually high fever and high white blood cell counts are clues to the presence of pericarditis. Children and immunosuppressed patients of all ages are most vulnerable, and the characteristic features of acute pericarditis are frequently absent. The course of bacterial pericarditis is fulminant, often presenting with cardiac tamponade; adhesive and constrictive pericarditis are common sequelae in survivors and may develop suddenly and early. Less common complications of purulent pericarditis include mycotic aneurysms (most often Staphylococcus and Salmonella species in the thoracic aorta) and left ventricular and submitral pseudoaneurysms.³⁸ Pericarditis complicating systemic infection and sepsis may go unrecognized and misdiagnosed. Many patients lack the typical findings of pericarditis, and the diagnosis of purulent pericarditis often is made either at autopsy or after cardiac tamponade develops; empyema is a common antecedent. The threshold for echocardiography in the septic patient should be low, and whenever purulent pericarditis is suspected, the pericardial space should be explored and fluid and blood sent for extensive microbiologic testing. Bacterial pericarditis is treated with surgical exploration and drainage and appropriate systemic antibiotics. Fibrinolytics may be used to lyse fibrous adhesions, liquefy purulent exudate, and prevent constrictive pericarditis. Pericardiectomy may be required for dense adhesions, loculated and thick effusions, persistent infection, recurrence of tamponade, and constriction. Although fatal if untreated, mortality remains high (~40%) in those receiving proper therapy.^37,38

Legionella infections account for approximately 10% of community-acquired pneumonias and may be associated with pericarditis more often than previously was appreciated. Studies suggest that patients with pericardial involvement tend to be younger and healthier than those without it. Recurrent pericarditis, effusion, and chronic constriction occur in approximately 20% to 30% of cases.²⁷ Pericarditis may be an early complication of Lyme disease.

Tuberculous Pericarditis

Tuberculosis is a major cause of pericarditis in nonindustrialized countries, but an uncommon cause in developed countries with a low prevalence of tuberculosis.¹ Nevertheless, its incidence is increasing because of HIV infection; consequently, tuberculosis should be considered in the differential diagnosis of pericardial heart disease.^1,39

Tuberculous pericarditis results from hematogenous spread of primary tuberculosis or from the breakdown of infected mediastinal lymph nodes, with the result that affected individuals generally lack the typical symptoms and signs of pulmonary tuberculosis. A delayed hypersensitivity response to protein antigens of the bacilli that penetrate the pericardium is responsible for the morbidity associated with tuberculous pericarditis.

Fever, weight loss, and night sweats occur early; pericardial pain and friction rubs are often absent. Enlargement of mediastinal lymph nodes is not routinely seen on chest radiography (CT or magnetic resonance imaging [MRI] is required), although an enlarged cardiac silhouette is common. Patients may present with tamponade or constriction, which may be subacute. A fibrinous pericarditis with caseating necrosis and mononuclear infiltrate gives rise to an effusive phase, which is often voluminous and hemodynamically significant. An adhesive phase follows resolution of the effusion and eventuates in dense, calcific adhesions with clinical constriction in > 30% of patients.^1,39

A definite diagnosis of tuberculous pericarditis is based on the demonstration of the presence of tubercle bacilli in the pericardial fluid or tissue. However, a probable diagnosis of tuberculous pericarditis can be achieved with evidence of the disease elsewhere (eg, pulmonary tuberculosis) and concomitant pericarditis, a lymphocytic pericardial exudate with elevated unstimulated interferon-gamma, adenosine deaminase, or lysozyme levels. An ex juvantibus diagnosis is admitted only in countries with a high prevalence of tuberculosis with the demonstration of the response to empiric antituberculous therapy.^1,39,40 A presumptive diagnosis generally requires a history of contact and/or purified protein derivative conversion (although the latter lacks sensitivity and specificity).

A regimen consisting of rifampicin, isoniazid, pyrazinamide, and ethambutol for at least 2 months, followed by isoniazid and rifampicin (total of 6 months of therapy), is effective in treating extrapulmonary tuberculosis. Treatment for 9 months or longer gives no better results and has the disadvantages of increased cost and increased risk of poor compliance. Tuberculous pericarditis has a high risk of evolving in constrictive pericarditis, usually within 6 months in effusive forms. Prompt antibiotic therapy is essential to prevent this progression, which occurs in 20% (especially in developed countries) to 40% of cases. Additional treatments that may be useful to prevent constriction include intrapericardial urokinase and adjunctive prednisolone for 6 weeks, which may halve this complication (to be avoided in HIV-infected patients because it increase the risk of HIV-associated malignancies).^1,39,40,41

Pericardiectomy is recommended if the patient’s condition is not improving or is deteriorating after 4 to 8 weeks of antituberculosis therapy (Class I recommendation; LOE: C).¹

Human Immunodeficiency Virus Pericarditis

HIV is an important cause of pericardial heart disease. The presence of pericardial effusion was considered a negative prognostic predictor in past years. However, today, patients with HIV who are treated with highly active antiretroviral therapies (HAART) may have an etiologic spectrum of pericardial diseases that is similar to that of a non–HIV-infected population.⁴²

Pericardial involvement may be a result of associated malignancies (eg, lymphoma and Kaposi sarcoma), viruses (including HIV), and opportunistic infections (eg, mycobacteria, cytomegalovirus, Nocardia, and cryptococci) and, irrespective of its cause, predicts a poor prognosis in patients with HIV infection if they are not treated by HAART. Large symptomatic pericardial effusion in patients with HIV infection should be aggressively investigated because two-thirds of these cases have an identifiable cause. Tamponade in patients with HIV is mycobacterial (Mycobacterium tuberculosis or Mycobacterium avium-intracellulare) in origin in approximately one-third of patients.⁴² Tuberculous pericarditis in those with HIV infection is associated with a lower risk of constriction.¹

Neoplastic Pericarditis

Neoplastic pericardial involvement is manifested by pericarditis or simple pericardial effusion (usually moderate to large, cardiac tamponade) related to metastatic lymphatic involvement (especially for lung cancer) or hematogenous spread (especially for breast cancer). In addition, lymphomas, leukemias, melanoma, and cancer of contiguous organs (eg, esophageal cancer) may also affect the pericardium. Only rarely is neoplastic disease primary (usually pericardial mesothelioma).^1,43,44,45

Neoplastic pericardial disease may be manifested as pericarditis, pericardial effusion, effusive-constrictive pericarditis, or constrictive pericarditis. Masses may be diagnosed by imaging techniques.

The definite diagnosis is based on the confirmation of the malignant infiltration within the pericardium by cytology or pericardial biopsy. A probable diagnosis may be achieved by the detection of tumor markers in pericardial fluid (eg, carcinoembryonic antigen [CEA], cytokeratin 19 fragment [CYFRA 21-1]), although none of the tumor markers has been proven to be accurate enough for distinguishing malignant from benign effusions. Evidence of malignant disease elsewhere and concomitant pericarditis or pericardial effusion are also suggestive, although in almost two-thirds of patients with documented malignancy, pericardial effusion is caused by nonmalignant diseases (eg, radiation pericarditis, other therapies, opportunistic infections). The management of these patients require a multidisciplinary approach with oncologists, radiotherapists, and other subspecialty experts.⁴³

Pericardial Diseases in Systemic Inflammatory Diseases and Post–Cardiac Injury Syndromes

Systemic inflammatory diseases (especially systemic lupus erythematosus, Sjögren syndrome, rheumatoid arthritis, scleroderma, systemic vasculitides, Behçet syndrome, sarcoidosis, and inflammatory bowel diseases) are common causes of pericarditis or “apparently” isolated pericardial effusion. Up to 10% of patients with pericarditis (often recurrent) have a known systemic inflammatory disease, but rarely, pericarditis/pericardial effusion may be the first manifestation of the systemic disease. Usually the degree of pericardial involvement is related to the activity of the systemic disease. Moreover, concomitant myocarditis may be present as well, because it is also a cause of myocardial inflammatory involvement.^43,46

A specific subgroup of these patients, especially children, may be affected by periodic fevers or autoinflammatory diseases. Periodic fevers are genetic disorders characterized by mutations of genes involved in the regulation of the inflammatory response, without involvement of specific T cells or autoantibodies. The most common autoinflammatory syndromes include familial Mediterranean fever, in which serositis episodes last only 1 to 3 days, and tumor necrosis factor receptor–associated periodic syndrome, in which the episodes last weeks. Mutations associated with these disorders are rare in recurrent pericarditis. A positive family history for pericarditis or periodic fevers, a poor response to colchicine, and the need for immunosuppressive agents are clues to the possible presence of an autoinflammatory disease. Genetic testing is required for the definitive diagnosis.⁴³

The term post–cardiac injury syndromes includes a group of inflammatory pericardial syndromes (post–myocardial infarction pericarditis, post-pericardiotomy syndrome, and post–traumatic pericarditis). Such syndromes are presumed to have an autoimmune pathogenesis triggered by an initial damage to pericardial and/or pleural tissues, caused by either myocardial necrosis (late post–myocardial infarction pericarditis or Dressler syndrome), surgical trauma (post-pericardiotomy syndrome), accidental thoracic trauma (traumatic pericarditis), or iatrogenic trauma with or without bleeding (pericarditis after invasive cardiac interventions).^2,47 Such forms are increasing, especially in developed countries, as a result of the aging of the population and the widespread use of percutaneous coronary intervention.

Treatment of these forms is similar to that of idiopathic pericarditis, with use of aspirin/NSAID plus colchicine as a first-line therapy and corticosteroids as a valid option especially when patients are on oral anticoagulants and interaction with anti-inflammatory drugs is not wanted.^43,47

Radiation-Induced Pericardial Disease

Prior chest radiation is an important cause of pericardial disease. Radiation therapy may affect not only the pericardium, but also the myocardium, heart valves, coronary arteries, and all mediastinal structures inducing fibrosis. Most cases are secondary to radiation therapy for Hodgkin lymphoma or breast or lung cancer, and serious radiation-induced pericardial disease is most often a result of radiation therapy of Hodgkin lymphoma. Today, lower doses and modern radiation therapy (shielding and dose calculation) have reduced the complications; radiation pericarditis has now dropped from 20% to 2.5% of cases.

Radiation can induce an early disease, with pericarditis with or without effusion, and a late disease, with constrictive pericarditis after 2 to 20 years and not necessarily preceded by pericarditis. This late disease may affect a variable number of patients (4%-20% of patients) and appears to be dose dependent and related to the presence of late pericardial effusion in the delayed acute phase. The effusion may be serous or hemorrhagic and has a high probability to develop fibrous adhesions. Therapies are similar to those employed in pericarditis and pericardial effusion.

The concomitant possible myocardial damage affects the prognosis because radiation-induced constrictive pericarditis has a worse outcome after pericardiectomy.^1,43,48

Traumatic Pericardial Disease

Blunt and penetrating traumas are important causes of pericarditis, particularly among young men. Acute tamponade beginning very soon after trauma (pericardial injury syndrome), recurrent pericardial effusion, recurrent acute pericarditis (postpericardial syndrome), and chronic constrictive pericarditis are well-recognized complications.

Severe blunt trauma, as with a fall from a considerable height, can rupture the pericardium, resulting in the potentially fatal complication of herniation of the heart. This event should be suspected if the chest radiography shows pneumothorax or air in the mediastinum distributed between the heart and diaphragm.

The application of echocardiography in the trauma unit rapidly and accurately diagnoses hemopericardium in patients with potentially penetrating cardiac wounds. Failure to repair the injury responsible for tamponade is associated with a poor clinical outcome. Constrictive pericarditis may be delayed, presenting weeks or years after the injury.

Chylopericardium

Chylopericardium is a milky-white pericardial effusion comprised of chyle, the normal content of the lacteals (lymphatics of the small intestine) and thoracic duct; the composition is variable, but generally has a high content of chylomicrons, protein, and lymphocytes. The disorder is rare, but morbidity (a result of nutritional, metabolic, and immunologic abnormalities) and mortality are considerable. Although the majority of cases are asymptomatic, cardiac tamponade can occur. Acute pericarditis and chronic constriction may result from the irritant effects of chyle.

Chylous pericardial effusions generally follow traumatic (blunt or penetrating) or surgical injury (thoracic or cardiac) to the thoracic duct, but may result from neoplastic obstruction of the duct (secondary chylopericardium); less commonly, they may be idiopathic (primary). Drainage and dietary manipulation are effective in approximately 55% of cases. Failure to respond to a diet rich in medium-chain triglycerides and pericardiocentesis warrants ligation of the thoracic duct and pericardiectomy. In cases deemed inappropriate for aggressive therapy, implantation of a valved pericardioperitoneal conduit has been helpful.

Pericardial Diseases in Renal Failure

Pericardial diseases in renal failure have been become less common than in the past, but should be considered in the differential diagnoses for pericarditis and pericardial effusion. There are three main presentations of pericarditis in renal failure: (1) uremic pericarditis, occurring before renal replacement therapy or within 8 weeks from its initiation and related to retention of toxic metabolites; (2) dialysis pericarditis, occurring on dialysis (usually ≥ 8 weeks after its initiation); and (3) constrictive pericarditis, occurring only rarely.^1,4,49,50

Myxedema Pericardial Disease

Pericarditis with effusion (sometimes containing cholesterol) occurs in approximately one-third of patients with myxedema. Effusions develop slowly and may reach a prodigious size; slow resolution usually follows the institution of thyroid replacement therapy. Pericardial drainage is generally not indicated because myxedema effusions seldom cause tamponade.¹

Pericardial effusions are commonly encountered in routine cardiovascular practice, with clinical presentation ranging from an incidental finding to cardiac tamponade. The incidence of pericardial effusion is unknown; it has been reported to be present in 9% of echocardiograms performed at a European tertiary care center specializing in diseases of the pericardium.⁵¹ The differential diagnosis of a pericardial effusion is extensive, and the management of pericardial effusion depends on disease progression and its hemodynamic consequences.

Clinical Presentation and Etiologies

Pericardial effusions can be classified according to their temporal development: acute, subacute, and chronic (typically longer than 3 months).³ Pericardial fluid accumulates when there is an increase in its production as a result of inflammation of the serosal layers; alternatively, pericardial effusions can occur in the setting of impaired lymphatic drainage of the pericardial space as a result of increased central venous pressure (as observed in heart failure or severe pulmonary hypertension). Effusions will have characteristics of an exudate in the former and of a transudate in the latter. The anatomy of pericardial lymphatics and their communication with other thoracic structures, such as the pleura and the breasts, explain pericardial involvement in pulmonary/pleural disorders and malignancies.

The volume of a pericardial effusion can range from trivial to large. The assessment of the size of pericardial effusions is most frequently done by transthoracic echocardiography. Quantification of pericardial effusion size is subjective and varies as the pericardial fluid shifts according to the patient’s body position (for example, increase in the size of a posterior pericardial effusion in the supine positon). The size of the pericardial effusion has prognostic implications; a large effusion in the setting of acute pericarditis is associated with a high risk of developing cardiac tamponade.¹³ Echocardiography also differentiates circumferential from loculated effusions.

Data from multiple studies suggest that idiopathic pericardial effusions are the most common (approximately 50% in some series) in developed countries, followed by infectious causes and cancer (15%-20% each) and connective tissue disorders.^52,53,54 In developing countries, tuberculosis continues to be the predominant etiology, accounting for 50% to 60% of the cases.⁵⁴ Essentially all causes of pericarditis (as depicted in Table 66–2) can manifest as pericardial effusion. Additional causes include iatrogenic causes (after intervention or device placement); postinjury causes (following myocardial infarction, post-pericardiotomy syndrome), and less commonly, drugs or systemic disorders (such as hypothyroidism). Transudative effusions might also occur as a result of hypoalbuminemia in the setting of liver or kidney diseases or as a result of high hydrostatic pressure, as seen in heart failure. The presence of pericardial effusions in patients with pulmonary hypertension should be noted because it has been described as a poor prognostic feature in that population.⁵⁵

The clinical presentation of patients with pericardial effusion will vary according to the underlying process, rate of accumulation, and hemodynamic effect. Patients presenting with acutely developing pericardial effusion, as in thoracic trauma, cardiac perforation, or aortic dissection, might present with rapid hemodynamic deterioration secondary to cardiac tamponade despite relatively small pericardial effusions. Conversely, in chronic pericardial effusions, symptoms might develop over weeks or even months, and patients might be entirely asymptomatic, with the effusion being an incidental finding during thoracic imaging or transthoracic echocardiography. Symptoms may include atypical chest pain or fullness, exertional dyspnea, orthopnea (rarely), or other less specific findings such as cough, malaise, or fatigue. Compression of local structures can lead to dysphagia, hoarseness, or hiccups.

Physical Examination

The diagnosis of pericardial effusion at the bedside is difficult and essentially of historical interest. The roles of physical exam in patients with pericardial effusion are, first, to assess the hemodynamic consequences of the pericardial effusion and, second, to identify signs suggestive of underlying causes (such as signs of myxedema). Heart rate and blood pressure should be manually recorded, and the presence of pulsus paradoxus (Fig. 66–9) should be sought. A thorough examination of the jugular veins is mandatory to determine whether cardiac tamponade is present (the details of the physical examination in the diagnosis of cardiac tamponade are discussed below). In uncomplicated pericardial effusions, distant heart sounds might be present, but cardiac examination might be otherwise unremarkable. Lastly, pericardial rubs may be auscultated in patients with concomitant pericarditis.

Evaluation and Testing

The chest radiograph reveals enlargement of the cardiac silhouette in large effusions, but chest radiographs might be entirely normal in patients with pericardial effusions. ECG findings include low QRS voltage and electrical alternans (Fig. 66–10; beat-to-beat variations in the QRS amplitude as a result of changes in cardiac position in the setting of the effusion). In patients with pericardial effusion related to acute pericarditis, typical ECG changes of pericarditis might be present. Some patients with pericardial effusions might have entirely normal ECGs.

Pericardial effusions are most often identified by transthoracic echocardiography, but might also be incidentally found during radiologic thoracic imaging (for example, during staging or follow-up of patients with known cancer). Echocardiography is the main diagnostic tool in the evaluation of pericardial effusion and should be performed in all patients (Class I recommendation; LOE: C)¹ because it allows the assessment of its size, location, and hemodynamic consequences (presence of tamponade physiology). The presence of trivial effusions (only noticeable during ventricular systolic) is not uncommon and of no clinical consequence (“physiologic” effusions). Although objective quantitation of the size of the pericardial effusion is recommended¹ (free pericardial space in diastolic < 10 mm in small, 10-20 mm in moderate, and > 20 mm in large effusions), estimation of the size of pericardial effusion is mainly qualitative in clinical practice. Echocardiography is also indicated during pericardiocentesis.

The determination of the cause of the pericardial effusion might be challenging, and additional testing should be performed, taking into account the clinical scenario and the patient’s social and travel history. Given the prevalence of tuberculosis in developing countries (and among immigrants in developed countries), its diagnosis should always be considered and ruled out in those patients; HIV testing should also be performed. If bacterial pericarditis is suspected, prompt diagnostic pericardiocentesis is mandatory. Serum inflammatory markers (C-reactive protein and erythrocyte sedimentation rate) should be measured (Class I recommendation; LOE: C).¹ The need for pericardiocentesis in the evaluation of the etiology should be individualized. A diagnostic algorithm is proposed in Fig. 66–11.³ Patients presenting with acute pericarditis, elevated inflammatory markers, and no signs of cardiac tamponade might be medically managed. Patients with small and moderate effusions can be safely observed. Diagnostic pericardiocentesis should be reserved for patients with large pericardial effusions or where the diagnosis of pericardial involvement would have management implications (such as in some malignancies).

Table 66–11 summarizes routine pericardial fluid testing and accompanying etiologies. Analysis of pericardial fluid can discriminate exudates from transudates, but this distinction has limited diagnostic implications. Low glucose levels are more frequently encountered in infectious or malignant effusions. Total white blood cell count and differential white blood cell count rarely have meaningful diagnostic or management implications. A pericardial fluid–to–serum hemoglobin ratio ≥ 0.5 of the plasma hemoglobin is diagnostic of hemopericardium.

Bacterial and mycobacterial cultures should be performed if bacterial infection or tuberculosis is suspected, respectively. Acid-fast bacilli staining, adenosine deaminase, pericardial lysozyme, and interferon-γ levels, as well polymerase chain reaction testing, should be added in the evaluation of tuberculous pericarditis.⁵⁶ Patients with tuberculous pericarditis have a high risk of developing constrictive pericarditis (30%-50% in some series^57,58); thus, a high index of suspicion is needed in those at risk.

Pericardial effusion cytology should be performed (the largest volume possible sent to increase its yield) when malignancy is either known or suspected. Specific tumor marker levels (eg, CEA, cancer antigen [CA]-125) can also be measured in the pericardial fluid if malignant effusion is a possibility. Lastly, although some groups include pericardioscopy and pericardial biopsy in the workup of pericardial effusions, this is rarely performed by our group.

Additional imaging can also provide further information in the workup of patients with pericardial effusions (Class IIa recommendation; LOE: C).¹ CT allows additional assessment of the size and location of the fluid collection, as well as differentiation between pericardial fluid versus pericardial thickening (which can be challenging echocardiographically). It also permits detection of associated thoracic abnormalities (such as tumors) or identification of pericardial masses. CMR has the advantage of assessing the presence and degree of pericardial inflammation and concomitant myocardial involvement in the case of myopericarditis. The cost and availability of MRI (especially in the acute setting) are potential limiting factors; the use of gadolinium is required to assess inflammation; however, this is contraindicated in some patients (as in those with advanced renal insufficiency).

Treatment

Therapy should be directed at the underlying cause (Class I recommendation; LOE: C)¹; if pericarditis is present, treatment of pericarditis is the goal. For patients with idiopathic effusions (especially if inflammatory markers are elevated), a combination of an NSAID and colchicine is recommended (Class I recommendation; LOE: C)¹; viral pericardial effusions are also often responsive to this regimen. In cases of relapsing pericardial effusions, a longer treatment course should be prescribed. Corticosteroids should not be used as first-line therapy and should be limited to special situations (treatment of concomitant systemic disease) or in patients for whom NSAID are contraindicated.¹ The use of other immunosuppressant agents (eg, azathioprine) should be reserved to a multidisciplinary team (eg, cardiologists, rheumatologists) with expertise in the management of refractory pericarditis. Limited data are available on the use of topical (intrapericardial) steroid therapy, and it is currently not routinely recommended.

Aspirin is recommended as first-line therapy in post–acute myocardial infarction pericarditis (Class I recommendation; LOE: C)¹ given the risk of impaired myocardial healing and expansion of the infarct zone seen with other agents.⁵⁹ Corticosteroids should be reserved for refractory cases because it might affect myocardial healing.

The details regarding pericardiocentesis are described in the section on cardiac tamponade; postprocedurally, the intrapericardial catheter should be kept in place until output is lower than 30 mL over 24 hours; this promotes adhesion of pericardial layers, limiting fluid reaccumulation and reducing the need for repeat pericardiocentesis. Pericardial window or balloon pericardiostomies can be performed in patients with refractory recurrent pericardial effusions and cardiac tamponade (most commonly secondary to malignant effusions).³ Pericardiectomy might also be an option in selected cases, but should only be performed in centers with surgical expertise.

The prognosis of pericardial effusions is related to its underlying etiology. It has been suggested that large pericardial effusions carry a 30% chance of cardiac tamponade⁶⁰; close follow-up is recommended in those circumstances. The risk of developing constrictive pericarditis is very small in idiopathic pericarditis,²⁷ even if relapse is present. Given the higher risk in bacterial or tuberculous pericarditis, those patients also require more frequent re-evaluation. Pericardial guidelines do not provide specific recommendations regarding follow-up. Patients with small effusions who respond to therapy do not need long-term follow-up. Patients with moderate-to-large effusions should be evaluated at the end of their therapy course with repeat imaging; at that point, the need for continuous follow-up (every 3 or 6 months) is reassessed.

Cardiac tamponade corresponds to a corollary of hemodynamic derangements that are secondary to increased intrapericardial pressure and impaired cardiac filling. Essentially any form of disease that affects the pericardium can potentially lead to cardiac tamponade. Although normally associated with pericardial fluid (effusion), tamponade may also occur when blood, pus, or clots occupy the pericardial space.

Pathophysiology and Clinical Presentation

Although usually considered in the differential diagnosis of patients presenting with shock and acute hypotension, the clinical presentation of cardiac tamponade depends on the tempo with which pericardial fluid accumulates and intrapericardial pressures rise. Intrapericardial pressures will directly reflect pericardial compliance, as illustrated in pericardial pressure-volume curves (see Fig. 66–3). When the pericardial volume changes acutely (eg, cardiac perforation secondary to a procedure or as a result of aortic dissection), a small amount (as little as 150 mL) of fluid in the pericardial space results in a significant, abrupt elevation in intrapericardial pressure and can lead to sudden hemodynamic decompensation. Conversely, in the setting of chronic pericardial effusions, the pericardium “stretches out” and is sometimes able to accommodate liters of fluid without major hemodynamic consequences. As illustrated on curve B in Fig. 66–3, the pressure-volume curve is much less steep if the increase in pericardial pressure develops subacutely or chronically.

A good understanding of the pericardial pressure-volume curves is not merely academic; it is critical for the proper management of patients with pericardial effusion. First, it shows that cardiac tamponade is part of a clinical spectrum. Patients might present acutely, but cardiac tamponade can also be an insidious process, where subtle, subclinical hemodynamic abnormalities develop as patients approach the steep part of the curve (decompensated cardiac tamponade). Second, it illustrates that once at the steepest portion of the pressure-volume curve, very small changes in pericardial volume can lead to profound changes in intrapericardial pressure and overall hemodynamics. This is of extreme clinical importance; patients presenting with decompensated cardiac tamponade who “stabilize” with vasopressors or intravenous fluids are not truly stable, and pericardiocentesis should not be delayed. A positive clinical response to removal of a small amount of pericardial fluid is not a reassuring finding; the pericardial space should be fully drained. If the former approach is pursued, prompt reaccumulation of a small volume of pericardial fluid might have catastrophic consequences.

As intrapericardial pressure rises in patients with cardiac tamponade, elevated pericardial pressure is transmitted to all four chambers, affecting diastolic filling. As a result, right atrial transmural pressure decreases markedly and systemic venous pressure increases. As a result of pericardial restraint from a now distended, inelastic pericardium, right ventricular filling occurs at the expense of left ventricular filling—the so-called enhanced ventricular interaction or ventricular interdependence. Although some degree of ventricular interdependence is always present, the changes that occur in cardiac tamponade correspond to marked, pathologic exaggeration of normal pericardial physiology. As a result, with inspiration, right ventricular stroke volume increases (secondary to increased venous return) while left ventricular stroke volume decreases (this is the origin of pulsus paradoxus); the opposite occurs during expiration. As right ventricular diastolic filling is further compromised, both ventricles become underfilled, and cardiac output decreases. Tachycardia and increased peripheral vascular resistance initially compensate for the reduction in stroke volume, but as pericardial pressures increase further, the reduction in cardiac output leads to hypotension, shock, and ultimately death unless the pericardial effusion is drained.

Cardiac tamponade should always be considered in patients presenting with shock and elevated venous pressure, unexplained hypotension, clinical decompensation in the setting of known pericardial effusion, recent intracardiac manipulation (cardiac catheterization, transseptal puncture, pacemaker implantation, electrophysiology ablation) or intrapericardial (pericardial access) procedures, or after trauma or cardiac or thoracic surgery. Its diagnosis can be challenging, particularly when several other possible causes for hypotension/shock are present (eg, following trauma or cardiac operation). Thus, a high index of suspicion is needed, and the diagnosis of cardiac tamponade should be actively sought by clinical and imaging techniques. Conversely, patients with insidious development of cardiac tamponade might complain of reduced functional capacity, nonspecific chest pain, or simply fatigue. We discourage the use of the terms medical cardiac tamponade and surgical cardiac tamponade because they generate confusion and have no etiologic or management implications.

Physical Examination

Tachycardia (heart rate > 100 bpm in adults) is usually present, but some patients might have heart rates at the upper end of normal (90-100 bpm). Hypotension is a classic sign of cardiac tamponade, but relative hypotension might be present in patients who are hypertensive at baseline (hypertensive tamponade). Blood pressure should be manually measured in patients with possible tamponade with special attention to the presence of pulsus paradoxus (see Fig. 66–9). In normal patients, systolic blood pressure should not drop more than 10 mm Hg upon inspiration. The presence of “pulsus” might be subtle, and proper technique needs to be applied. The blood pressure cuff should be inflated to 15 to 20 mm Hg above the systolic blood pressure (determined by palpation and then auscultation). The cuff should then be slowly deflated and the Korotkoff sounds meticulously auscultated while the patient breathes normally; the sounds will show respirophasic changes, decreasing (or even disappearing) with inspiration. At a certain pressure, these respirophasic changes will be no longer noticeable, and the expiratory and inspiratory Korotkoff sounds will be similar. The “pulsus” will correspond to the pressure difference between the two points and should be recorded (eg, 12 mm Hg).

Inspection of the jugular veins will show elevation in central venous pressure; venous distention may or may not be present. Analysis of the venous wave contour will reveal absence or blunting of the y descent, representing impaired ventricular diastolic filling. If pericardiocentesis is performed, pre- and post-estimated venous pressure and venous contour should be recorded. Persistence of elevation in central venous pressure following pericardial effusion drainage suggests that the cardiac tamponade has not been totally relieved or that effusive-constrictive pericarditis is present (in this case, blunting will be replaced by deepening of the y descents).⁶¹

The apical impulse might not be palpable or might be hyperkinetic. Cardiac auscultation might reveal distant heart sounds. Because early diastolic filling is profoundly impaired, a third heart sound should never be present; its occurrence suggests an alternative diagnosis. For unclear reasons, tamponade does not lead to pulmonary edema; the lungs should be clear on auscultation in patients with isolated cardiac tamponade.

Although classically described as a sign of cardiac tamponade, Beck’s triad—hypotension, distended neck veins, and distant heart sounds—is rarely present and, therefore, its absence should not influence the evaluation of patients with suspected cardiac tamponade.⁶²

Evaluation and Testing

ECG frequently shows sinus tachycardia; electrical alternans (see Fig. 66–10) might be present, where QRS amplitude will vary with each cardiac beat as a result of the swinging motion of the heart in the large effusion; changes in P and QRS axes might also be present. Low QRS amplitude might be seen and can be caused by the effusion or underlying myocardial disease. Lastly, ST-T changes associated with underlying pericarditis might be also be noted.

Chest radiography might reveal a very enlarged cardiac silhouette in the setting of a large pericardial effusion, but might be entirely normal. In isolated cardiac tamponade, pulmonary parenchymal abnormalities should not be seen; pleural effusions might be found in the case of a pleuropericarditic process.

Two-dimensional and Doppler transthoracic echocardiography is the key diagnostic tool in the evaluation and management of patients with cardiac tamponade (Class I recommendation; LOE: C).¹ First, echocardiography confirms the presence of pericardial fluid (or clots; Fig. 66–12). Although the size and location of an effusion are important for planning and feasibility of pericardiocentesis, the size of the effusion (small vs large) does not favor or exclude the presence of cardiac tamponade. An important part of the echo-Doppler examination in patients with suspected tamponade is the evaluation of the inferior vena cava; it is typically dilated (plethoric) and shows diminished (< 50%) or absent collapse with “sniffing maneuver.” An entirely normal inferior vena cava makes tamponade unlikely. Signs of impaired diastolic filling can be evidenced by two-dimensional echocardiography, illustrated by collapse of the right ventricular or right atrial walls during diastole (collapse of the left atrium is less common). The hallmark of pericardial restraint, ventricular interdependence is manifested as respirophasic changes in the interventricular septum motion (see Fig. 66–12). Doppler signs of tamponade physiology included respirophasic changes in mitral inflow (see Fig. 66–12) and left ventricular stroke volume (echocardiographic equivalent of pulsus paradoxus). The most specific finding, however, is the presence of expiratory flow reversals in the hepatic veins (see Fig. 66–12).^63,64,65 The use of a respirometer facilitates timing of inspiratory and expiration, and should always be performed in the evaluation of patients suspected of having tamponade or constriction.

Cardiac catheterization will show elevation in right- and left-sided filling pressures with equalization of end-diastolic pressures. In acute tamponade, however, left-sided filling pressures might be lower as a result of significant underfilling of the left ventricle. Although discordance of right and left ventricular systolic pressures might be present, as opposed to constrictive pericarditis, rapid filling waves will not be seen on ventricular tracings. The main diagnostic feature of cardiac tamponade is blunting or absence of y descents on right atrial tracings (Fig. 66–13) as a result of impaired right ventricular early diastolic filling⁶⁶; the y descents should normalize after pericardiocentesis.

Classical teaching states that cardiac tamponade is a clinical diagnosis and should not be based on testing alone. Once the clinical diagnosis is made, patients should undergo pericardiocentesis. However, cardiac tamponade is part of continuum and not an all-or-none phenomenon; patients in earlier stages may have “tamponade physiology” (typically by echocardiography) on complementary testing before overt (decompensated) clinical signs develop. This can also be the case in chronic cardiac tamponade. The management of those patients should be individualized, and intervention should be based on clinical status, risks of the pericardiocentesis (location and size of the effusion), and expertise of the proceduralists.

Treatment

Cardiac tamponade is a cardiac emergency and should be treated with prompt needle pericardiocentesis unless it is caused by aortic dissection. The procedure should be done with transthoracic echocardiographic guidance by an experienced operator. Blind pericardiocentesis should only be performed in extreme, life-threatening circumstances. If proper technique is applied, complications rates are low.⁵¹ The goal should be aspiration of all pericardial fluid. The use of simultaneous right atrial pressure measurement is not mandatory and varies between groups; at our institution, invasive pressure recording is not routinely performed during pericardiocentesis.

During echo-guided pericardiocentesis, the largest collection of fluid in closest proximity to the chest wall should be identified, defining the optimal site for needle entry. Echocardiography also confirms the absence of interposed lung or liver tissue. The angle of the transducer should orient the trajectory of the needle. In our published experience, major complications occurred in 3% of cases; the most common site of entry was para-apical (70% of cases).⁴⁴ Complications include cardiac or coronary perforation, hemothorax or liver injury, pneumothorax, and pneumopericardium.⁶⁷

The daily output of pericardial fluid is recorded. The pigtail catheter should not be removed until the output is less than 30 mL over a 24-hour period. It is our practice to repeat a limited echocardiogram to reassess the effusion size and areas of loculation before the intrapericardial catheter is removed. The need for surgical pericardiocentesis is rare and involves less common circumstances (eg, removal of clots or pus) or simultaneous repair of damaged cardiac structures.

Constrictive pericarditis results from inflammation and/or scarring of the pericardium, leading to impairment of cardiac filling.⁶⁸ Therefore, constrictive pericarditis is a form of diastolic heart failure. The importance of the diagnosis lies in the fact that pericardial constriction is a potentially curable disease and carries a very poor prognosis if left untreated.⁶⁹

Pathophysiology and Clinical Presentation

Constrictive pericarditis develops after an injury to the pericardium results in ongoing pericardial inflammation and ultimately fibrosis. This process might be acute (over the course of days), subacute, or chronic; in acute cases, inflammation prevails, whereas fibrosis tends to be predominant in chronic forms of the disease. Pericardial inflammation may resolve spontaneously or with anti-inflammation therapy, but the pericardium typically becomes fibrotic or scarred if not resolved within 3 months. The reason why some patients will evolve to have constrictive pericarditis after an acute event is unclear; however, certain etiologies (eg, tuberculous pericarditis) are associated with higher risk of developing constrictive pericarditis.^27,58 The disease also appears to favor men over women (2:1 ratio).⁶⁸ Although, constrictive pericarditis is primarily a pericardial disease, subepicardial myocardial atrophy might be present⁷⁰ (particularly in long-standing cases), leading to concomitant myocardial systolic and diastolic dysfunction. A mixed pattern is also seen in patients who develop constriction after cardiac surgery or radiation therapy.

Given the abnormal pericardial compliance, diastolic filling is significantly impaired in patients with constrictive pericarditis (steeper increase in pressure per change in volume in the pressure-volume curve). As a consequence, cardiac filling pressures increase and cardiac output falls as stroke volume decreases. The thickened pericardium (Fig. 66–14) has limited expansion, and cardiac chambers “compete” for pericardial space during diastolic filling. With respiration, an increase in preload in one ventricle occurs at the expense of filling in the other (ventricular interdependence). The abnormal pericardium insulates the heart from respirophasic changes in intrathoracic pressures, and negative pressures generated by inspiration are not transmitted to cardiac chambers. This contributes to decreased filling of the left atrium and ventricle upon inspiration as a result of decreased pressure gradient between pulmonary veins and left atrium (pulmonary veins are located outside the pericardium). These physiologic changes can be appreciated by both invasive hemodynamic and noninvasive hemodynamic assessment, and are used for the diagnosis of constrictive pericarditis.

Patients with constrictive pericarditis typically present with features of right-sided heart failure, manifested by elevated venous pressure, ascites, and leg edema. Pericardial constriction must also be considered when right heart failure is disproportional to the degree of pulmonary congestion or in patients presenting with heart failure following cardiac surgery. For unknown reasons, patients with constrictive pericarditis are more prone to the development of ascites than other patients with right heart failure; the ascites is often more striking than lower extremity edema. Additional common complaints are fatigue, decreased functional capacity, and head fullness. Orthopnea and paroxysmal nocturnal dyspnea are not usually observed and suggest another etiology. In more advanced stages, liver and renal dysfunction might be present.

Etiologies

Causes of constrictive pericarditis are illustrated in Table 66–12. The epidemiology of the disease has substantially changed over the past 50 to 60 years, with the advances in, and increased use of, cardiac surgery. Data from the Mayo Clinic showed that the 72% of cases of constrictive pericarditis seen between 1936 and 1982 were idiopathic, whereas only 2% involved patients with prior cardiac surgery.⁷¹ Between 1996 and 2006, postoperative cases accounted for 34%, whereas the prevalence of idiopathic cases decreased to 18%.⁷² This shift in epidemiology has also been shown by other US and European centers.^73,74,75 Constrictive pericarditis occurs in 0.2% to 0.4% of patients undergoing cardiac surgery and, on average, occurs 2 years after the surgical procedure.^76,77

In areas where tuberculosis is still prevalent (such as Africa and India), tuberculous pericarditis continues to be most common cause of pericardial constriction.^78,79 Thus, a thorough travel and social history should be obtained in the evaluation of patients with constrictive pericarditis. A history of radiation therapy is also of great importance, given the risk of concomitant myocardial and valvular disease, which has diagnostic and therapeutic implications.

Physical Examination

General examination will typically reveal muscle wasting, increased abnormal girth, and leg edema, with patients usually appearing ill. In patients with a subacute course, those findings can be absent, and the complaints will appear out of proportion to the patient’s “healthy” appearance. Abdominal exam will show signs of liver enlargement and ascites; signs of cirrhosis should be noted because the presence of liver failure has implications regarding surgical risk. Pleural effusions are commonly present, but rales should not be expected.

Detailed assessment of jugular venous pressure is mandatory in the evaluation of constrictive pericarditis. Elevated central venous pressure is essentially always present, unless the patient has been aggressively diuresed. If the venous pulse cannot be seen with the patient sitting, he or she should be asked to stand; failure to do this will preclude proper analysis of venous pressure and contour in patients with very elevated filling pressures. The Kussmaul sign corresponds to an increase in venous pressure associated with inspiration; however, this is not pathognomonic of constriction and can be seen in other situations where severe elevation in right atrial pressure is present. Jugular venous waveforms will show prominent y descents or sometimes prominent x and y descents (if sinus rhythm is present and there is absence of significant concomitant myocardial disease)—the W or M pattern. It should be noted that if constrictive pericarditis is present, the predominant waves are the descents (down strokes), which is the opposite of normal. In severe tricuspid regurgitation, although a deep y descent will be present, the predominant wave is a positive one, the large v wave. Restrictive cardiomyopathy can be associated with deep y descents, and the distinction between restriction and constriction at the bedside can be extremely challenging.

Auscultation might reveal the presence of low-pitched third heart sounds, and these can sometimes only be heard during inspiration or expiration. The presence of a high-pitched early diastolic sound (slightly earlier than a typical third heart sound)—the pericardial knock—is specific, but not a sensitive sign that constrictive pericarditis is present. A loud pulmonary component of the second heart sound is not usually heard and should suggest the presence of a concomitant myocardial process.

Evaluation and Testing

ECGs usually shows nonspecific ST-T wave abnormalities; atrial fibrillation occurs commonly in patients with constrictive pericarditis. P mitrales (broad, notched P waves in lead II) in the absence of mitral disease has also been described as a sign of constrictive pericarditis.⁸⁰ Chest radiography may reveal pericardial calcification (Fig. 66–15), although this is not a sensitive or specific sign and is seen in only 25% to 30% of cases.⁸¹ Pleural effusions are frequently present. Plasma brain natriuretic peptide levels appear to be lower in patients with constrictive pericarditis than in patients with myocardial restrictive disease.⁸²

Echocardiography is now the main diagnostic tool in the evaluation of constrictive pericarditis; it also helps to assess the patient for classic mimickers of constrictive physiology, such as severe tricuspid regurgitation and myocardial restrictive disease. Guidelines recommend transthoracic echocardiography to be performed in all patients with suspected constrictive pericarditis (Class I recommendation; LOE: C).¹ In our practice, transthoracic echocardiography can diagnose constrictive pericarditis in 75% of cases.⁷² Echocardiographic features of constrictive pericarditis^15,68,83 include septal notching and flattening of the posterior left ventricular wall seen on M-mode, diastolic bounce (or shudder) of the interventricular septum, respirophasic variations in tricuspid and mitral inflow echo-Doppler (Fig. 66–16), restrictive mitral valve inflow pulse-wave Doppler, and “annulus reversus” (medial e′ velocity greater than lateral e′ on mitral annular tissue Doppler) (see Fig. 66–16). Given the multitude of echocardiographic findings and with the aim of simplifying the diagnosis of constrictive pericarditis, our group proposed the use of the following three echocardiographic criteria for its diagnosis (Mayo Clinic Echocardiographic Diagnostic Criteria⁸⁴) (Fig. 66–17): diastolic respirophasic shift of the interventricular septum, medial mitral annulus e′ ≥ 9 cm/s, and diastolic expiratory hepatic vein flow ratio > 0.8, as well as plethoric inferior vena cava and restrictive mitral inflow velocities. If septal shift is present in conjunction with one of the other two criteria, sensitivity and specificity approach 90%. If all three criteria are present, specificity increased to 97%, but sensitivity decreased to 67%. A proposed algorithm of the echocardiographic diagnosis of constrictive pericarditis is illustrated in Fig. 66–18.

Cardiac catheterization findings^85,86 (Fig. 66–19) include elevated filling pressures with equalization of end-diastolic pressures (within 5 mm Hg). Right atrial pressures typically show prominent x and y descents with no drop in mean right atrial pressure with inspiration (Kussmaul sign). Other classic findings include right ventricular systolic pressures < 55 mm Hg (ie, absence of pulmonary hypertension), presence of rapid ventricular filling waves (≥ 5 mm Hg; so-called square root sign), and right ventricular end-diastolic pressures greater than a third of right ventricular systolic pressures. However, in more than a third of the contemporary patients with constriction, pulmonary artery systolic pressure is greater than 50 mm Hg, and the classic hemodynamic features are probably correct only in patients with pure constrictive pericarditis, but not in patients with concomitant myocardial disease or additional diastolic dysfunction. A normal cardiac output argues against the presence of clinically significant constrictive pericarditis. Simultaneous left ventricular and pulmonary artery wedge pressures can show evidence of dissociation of intrathoracic and intracardiac pressures. Lastly, discordance between right and left ventricular tracings during inspiration is of great diagnostic importance given its sensitivity and specificity.^87,88 If atrial fibrillation is present, temporary pacing might be necessary during the catheterization procedure to appreciate respirophasic changes. Lastly, it should be noted that most of these findings are load dependent and might not be present if filling pressures are normal or only mildly elevated; fluid challenge (rapid infusion of 1 L of fluid) should be considered in patients with right atrial pressures < 15 mm Hg.

Additional imaging with cardiac CT or MRI might be performed to assess pericardial thickness (Fig. 66–20). A thickened pericardium, defined as more than 2 mm in maximal thickness, is present is approximately 80% of the patients with constrictive pericarditis.⁸⁹ Thus, normal pericardial thickness makes the diagnosis of constrictive pericarditis less likely, but does not exclude the diagnosis. CMR imaging provides valuable incremental information regarding the presence of pericardial inflammation, which has therapeutic implications, as well the presence of constrictive physiology (based on septal motion).

The evaluation of patients with constrictive pericarditis can be rather complex, in particular in patients with previous cardiac surgery, severe concomitant myocardial/valvular disease, or radiation heart disease. The workup should be individualized, and the threshold to intervene (most often with therapeutic pericardiectomy) will depend on concordant/discordant data. If the history, physical exam, and echocardiogram findings are typical for the disease, no additional testing is typically required. However, this requires familiarity with bedside and echocardiographic assessment of constrictive pericarditis. If necessary, additional imaging (CT and MRI) and/or cardiac catheterization are the next steps (Class I recommendation; LOE: C). In our practice, cardiac catheterization is only performed in one-third of patients undergoing pericardiectomy. If the diagnosis is still unclear despite extensive evaluation, exploratory surgery and surgical inspection of the pericardium (gold standard in the hands of an experienced surgeon) can be performed; today, the need for diagnostic thoracotomy is rare.

Treatment

Although patients presenting with constrictive pericarditis and evidence of active pericardial inflammation (either by CMR or elevated inflammatory markers) might respond to anti-inflammatory therapy (Class IIb recommendation; LOE: C),¹ therapeutic pericardiectomy is the recommended treatment for symptomatic patients with constrictive pericarditis (Class I recommendation; LOE: C).¹

Pericardiectomy

Pericardiectomy is the surgical removal of the pericardium, and the procedure is applicable to all variants of pericardial disease.^{75,90,91,92,93,94,95,96,97} Successful operation is predicated on two fundamental caveats: obtain adequate operative exposure and remove the appropriate amount of pericardium. Although median sternotomy and thoracotomy both allow for safe removal of the pericardium, thoracotomy exposure may require extension across the midline into the right chest to allow for resection of all right-sided pericardium (Fig. 66–21).^90,92,98 The Mayo Clinic Rochester experience includes 513 patients operated with isolated pericardiectomy from 1993 through 2013. During that time period, the majority (n = 412, 80%) of operations were performed through a median sternotomy.⁹⁹

From a surgical perspective, pericardiectomy can be classified as either radical or complete (also referred to as total). The nomenclature is controversial and can be confusing.⁹⁸ It is probably best to frame the discussion about amount of pericardial resection in the context of the underlying disease process. For example, effusive and relapsing pericarditis represent inflammatory pathologies that can involve the entire pericardium.⁹³ Therefore, when operation is undertaken, total or complete pericardiectomy is indicated.⁹⁰ Constrictive pericarditis, on the other hand, is a disease of the ventricles.^68,90 Operation in that setting mandates resection of the pericardium overlying the ventricles (ie, radical pericardiectomy).⁹⁰ In the Mayo Clinic Rochester experience, operation was more commonly performed for constrictive (n = 355, 69%) rather than for inflammatory etiology (n = 158, 31%).⁹⁹

The key procedure-related elements of pericardiectomy are to remove the pericardium and to preserve phrenic nerve function. From the median sternotomy approach, the pericardial resection typically begins up near the great vessels, where the pericardium is usually less inflamed and less scarred to the underlying cardiac structures (see Fig 66–21). The “anterior” pericardium is resected from the great vessels to the diaphragm, and from the left phrenic nerve to the atrioventricular groove (at a minimum for constriction) or right phrenic nerve (for inflammatory etiologies), as needed or allowed (see Fig 66–21). To prevent injury to the phrenic nerves, a 1- to 2-cm pedicled strip of pericardium should be left undisturbed along the course of the nerves.

Resection of the anterior pericardium is also referred to as a “phrenic-to-phrenic” pericardiectomy. Specific attention is brought to this nomenclature to make the point that such limited resection (see Fig 66–21) is inadequate surgical therapy for essentially all pericardial disease processes; furthermore, it puts the patient at risk of treatment failure and disease recurrence.^90,98,100 As such, it is imperative that the pericardium be removed off the diaphragm and posterior to the left phrenic nerve (see Fig 66–21). Exposure to facilitate removal of that pericardium can require cardiopulmonary bypass support and sometimes even aortic occlusion (ie, aortic cross-clamp).^75,97,100 In the Mayo Clinic Rochester experience, 207 patients (40%) received cardiopulmonary bypass support during pericardiectomy, but only 4 patients (0.8%) received aortic occlusion.⁹⁹

Constrictive pericarditis is a surgical disease that is best treated with pericardiectomy.^1,68 For this condition, the ESC recommends removal of as much of the pericardium as possible.^1,68 This is ideally and theoretically correct, but in practice, it can be misguided thinking because the constricting peel often is densely fibrosed to the heart and calcification may even penetrate into the myocardium. This is especially true in the area of the atria, inferior vena cava, and atrioventricular groove. Attempts to resect the densely adhered pericardium can be dangerous, and injury to the thin-walled structures can even prove fatal. The pathophysiology requires release of the ventricles (see Fig 66–21), which must include the visceral pericardium.^98,101 Tricuspid valve regurgitation can be a confounding condition and should be addressed prior to the completion of the operation.¹⁰²

Pericardiectomy can be performed with an associated low operative morbidity and mortality and results in significant improvement in quality of life out to 10 years.⁹⁹ In the Mayo Clinic Rochester experience, overall mortality was 2.3%; for the constrictive pericarditis group, mortality was 2.5%, and for the inflammatory group, it was 1.9%.⁹⁹ Variables associated with an increased operative mortality have been well described by several investigators and commonly include a delay in treatment, advanced heart failure (eg, increased New York Heart Association class or reduced ejection fraction), and need of completion pericardiectomy (ie, for inadequate first operation or recurrence of disease).^{75,96,97,100,103,104} The take-home message is that early and adequate pericardiectomy is both life changing and life saving. Referral for operation should not be denied.

Most patients with constrictive pericarditis present with the classic, chronic form of the disease. However, other “uncommon” presentations have been recognized.¹⁰⁵ Although typically described as two separate entities,^106,107 transient and effusive-constrictive pericarditis are most likely part of a clinical spectrum, and distinction between the two might be academic.

Some patients presenting with cardiac tamponade demonstrate features of constrictive pericarditis immediately after pericardiocentesis is performed and tamponade relieved⁶¹ (Fig. 66–22). This is felt to be secondary to decreased pericardial compliance in the setting of acute inflammation. The reported incidence of effusive-constrictive pericarditis has varied according to the population analyzed and the most common etiologies (ranging from 8% in developed countries to more than 50% in African patients with tuberculous pericarditis).^106,108 Echocardiography shows features of constrictive pericarditis with a small amount of fluid or after resolution of tamponade. Therapy should be focused on treating active inflammation⁶¹; a subset of patients might develop chronic constrictive pericarditis and require pericardiectomy. Therefore, we suggest close follow-up of patients with effusive-constrictive pericarditis.

The definition of transient constrictive pericarditis has been variable in the literature. Initially described by Sagrista-Sauleda et al,¹⁰⁷ this group included patients presenting initially with acute pericarditis (not necessarily tamponade) that developed constrictive features when the acute process was resolving and the pericardial effusion was small or absent. The long-term prognosis for affected patients was good, with none requiring surgical treatment. Others have described transient constriction as constrictive pericarditis that resolved with corticosteroids or anti-inflammatory therapy.¹⁰⁹ Pericardial effusions are common in patients with transient constriction, reported in two-thirds of cases. Affected patients tend to have signs of inflammation with elevated inflammatory markers and gadolinium enhancement on MRI at the time of the diagnosis of constrictive pericarditis (Fig. 66–23).¹¹⁰ Optimal type and duration of NSAID versus corticosteroids with or without colchicine for transient constrictive pericarditis are still unknown. However, our experience has taught us that clinical response to medical therapy is relatively quick (within a week), and at least 3 months of medical therapy are required for a complete or clinically satisfactory response.

Occult constrictive pericarditis was described in a group of patients in whom constrictive features were only noticeable after fluid challenge.¹¹¹ Although we do use rapid saline infusion in the catheterization laboratory for patients with suspected constrictive pericarditis and low filling pressures related to overdiuresis, we do not use fluid challenge routinely to identify occult constrictive pericarditis. Given the lack of data, we would not recommend the diagnosis and management of constrictive pericarditis solely based on post–saline infusion changes for patients who do not otherwise have diagnostic features of the constrictive pericarditis.

Congenital abnormalities of the pericardium are rare. The two most common abnormalities encountered are pericardial cysts and congenital absence of the pericardium (typically partial); other rarer abnormalities include pericardial diverticula and pericardial bands. These abnormalities are usually found incidentally when chest radiography or other cardiac imaging is performed. Although most patients remain entirely asymptomatic, a small group of patients might require percutaneous or surgical intervention for symptomatic relief.

Pericardial Cysts

Pericardial cysts are smooth, thin-walled structures filled with clear fluid (hence the term spring water cysts). If there is a communication between the cavity and the pericardial space, a pericardial diverticulum is said to be present. Although typically found in the right costophrenic angle¹¹² (Fig. 66–24), they can be located at the left costophrenic angle or anteriorly. Pericardial cysts usually measure 3 to 4 cm in diameter, but larger cysts can also be seen, mimicking large pericardial effusions on chest radiography. The diagnosis is usually made incidentally when imaging of the chest is performed. If the diagnosis is unclear, CT of the chest is a helpful diagnostic tool and is preferred over transthoracic echocardiography.

Patients with pericardial cysts are usually asymptomatic, and the cysts tend not to increase in size. Rarely, there is resolution of the cyst, which may be related to spontaneous rupture. Some patients with pericardial cysts complain of atypical chest pain or dyspnea, most likely as a result of compression of contiguous structures. Because the symptoms are often nonspecific, it is difficult to attribute them to the pericardial cyst.

For patients with symptoms related to the pericardial cyst, percutaneous drainage or surgical removal can be performed.¹¹³ However, given the risks associated with both procedures and the benign, well-tolerated course of the disease, we would recommend observation for most patients with pericardial cysts.

Congenital Absence of the Pericardium

Congenital absence of the pericardium can be divided into partial (most common) or total (ie, bilateral) congenital absence of the pericardium. Although a wide spectrum can be found, including absence of the right portion and diaphragmatic components of the pericardium, the most typical presentation is total absence of the left portion of the pericardium. The diagnosis is usually made by chest radiography, when there is leftward displacement of the cardiac silhouette and aortic knob with no distinct right heart border seen to the left of the thoracic spine.¹¹⁴ As a result of the absence of the pericardium, lung tissue may be visible between the pulmonary artery and the aorta, appearing as the ear of a “Snoopy dog” (Fig. 66–25). As a result of the abnormal location of the heart within the chest, congenital absence of the pericardium should be suspected in patients with “very unusual” echocardiographic windows.¹¹⁵ CT nicely illustrates the displacement of the heart and shows elongation of the sternopericardial ligament with the presence of lung tissue interposed between the pulmonary artery and the sternum.

Patients with congenital absence of the pericardium are typically asymptomatic; rarely herniation or incarceration of cardiac structures might occur, leading to arrhythmias, angina, or syncopal spells. These symptoms might be precipitated by a change in body position; however, the most common complaint is vague chest pain. The need for surgical repair of congenital defects of the pericardium is extremely rare.¹¹⁶

Although some autopsy studies reported pericardial metastatic involvement in 10% to 20% of patients with known malignancies,¹¹⁷ pericardial masses are rare in clinical practice. Primary pericardial tumors are even less common, with mesothelioma being the most frequently encountered. Others include sarcomas and malignant teratomas. Nonmalignant tumors include lipomas, hamartomas, and benign teratomas.