Pericarditis is defined as inflammation of the pericardium, the thin fibrous sac surrounding the heart. Pericarditis may be subclinical or present as sharp precordial pain of acute onset. Acute pericarditis has many causes (Table 30), but it is most often idiopathic or presumed to be viral in origin.
Acute pericarditis is diagnosed clinically by the presence of at least two of the following four criteria: chest pain typical for pericarditis, a pericardial friction rub, new electrocardiographic (ECG) changes, or a new pericardial effusion. The chest pain of acute pericarditis is sharp and severe. It is not typically related to exertion and is not relieved with rest or nitroglycerin, unlike anginal pain. The pain is characteristically worse in the supine position and improves with sitting up and leaning forward. These pain features may be related to tension of the pericardium at its sternal and diaphragmatic attachments.
A pericardial friction rub is frequently present on auscultation. This harsh, scratchy sound classically has three components corresponding to the cardiac cycle during normal sinus rhythm: atrial systole, ventricular systole, and ventricular filling. The three phases of the pericardial friction rub differentiate it from a pleural friction rub, which has two components that are linked to respiration. The rub of pericarditis may also be monophasic or biphasic but is not affected by respiration. Auscultation should be performed during held end-expiration with the patient in the supine position or sitting upright.
The typical ECG feature of acute pericarditis is ST-segment elevation in multiple leads that does not correspond with a single coronary distribution. PR-segment depression in lead II or reciprocal PR-segment elevation in lead aVR may also be present (Figure 30). In contrast, ECG findings of acute myocardial infarction are hyperacute T waves, ST-segment elevation consistent with a single coronary distribution, reciprocal ST-segment depression, pathologic Q waves during the evolution of myocardial infarction, and lack of PR-segment change.
Echocardiography should be used to detect a pericardial effusion, although an effusion is not found in all cases of acute pericarditis. When the diagnosis remains uncertain, cardiac magnetic resonance (CMR) imaging with gadolinium intravenous contrast may be used to identify evidence of pericardial inflammation, characterized by pericardial thickening and late gadolinium enhancement (Figure 31). Alternatively, gated cardiac CT may also demonstrate pericardial inflammation.
Additional findings that support acute pericarditis may include fever and serologic evidence of inflammation (leukocytosis and elevated erythrocyte sedimentation rate or C-reactive protein [CRP] level). Serum cardiac troponin levels in acute pericarditis are normal or may be slightly elevated if there is a component of myopericarditis.
Although most cases of acute pericarditis are idiopathic, a search for other causes is appropriate (see Table 30). Evaluation includes a thorough history and physical examination, with additional testing based on the suspected cause. Tuberculosis testing should be considered in hospital workers, patients who are incarcerated or residing in chronic care facilities, and patients who are from or have traveled to an endemic area. Treatment of tuberculosis may reduce the risk for reactive pericardial constriction from greater than 80% to less than 10%. Other causes include uremia, chest irradiation, and recent cardiac surgery. Cardiac surgery may be followed by postpericardiotomy syndrome, which is characterized by pericardial inflammation that is likely autoimmune in nature and usually follows a latent period of several weeks. The presentation and treatment of postpericardiotomy syndrome are similar to those of idiopathic pericarditis.
Most patients with acute pericarditis can be managed medically on an outpatient basis; however, patients with acute pericarditis accompanied by high-risk features may require hospitalization for treatment and monitoring. Predictors of poor prognosis include temperature higher than 38 °C (100.4 °F), subacute onset, a large pericardial effusion (>20-mm diastolic echo-free space) or tamponade at presentation, oral anticoagulation therapy, or lack of response (no improvement in symptoms and/or inflammatory markers) after 1 week of treatment.
First-line treatment for acute idiopathic pericarditis consists of aspirin (750-1000 mg) or NSAIDs (ibuprofen 600 mg) every 8 hours for 1 to 2 weeks. Colchicine (0.5 mg once or twice daily for 3 months) is recommended as adjunctive therapy to shorten the duration of symptoms and reduce the chances of treatment failure or recurrence. Patients who initially respond to therapy but develop recurrent pericarditis after treatment completion may benefit from a longer course of standard therapy with slow tapering. CRP may be useful as a marker of treatment response, with tapering initiated after the CRP level normalizes. Expert consensus opinion recommends that athletes not return to competitive exercise for 3 months from initial onset and that nonathletes restrict strenuous activity until symptoms resolve.
Some patients with acute pericarditis may develop incessant or chronic pericarditis. Incessant pericarditis has been defined as pericarditis lasting for longer than 4 to 6 weeks but less than 3 months without remission, whereas chronic pericarditis is defined as lasting longer than 3 months. Glucocorticoid therapy is reserved for patients with recurrent, incessant, or chronic pericarditis despite standard therapy (including patients with uremic pericarditis not responsive to intensive dialysis); patients with contraindications to NSAID therapy; and patients with autoimmune-mediated pericarditis. Prednisone should be initiated at a dosage of 0.25 mg/kg to 0.5 mg/kg and continue for 3 months. Tapering should not be initiated until after the first 2 to 4 weeks of therapy or until the CRP level normalizes.
Limited data suggest that the interleukin-1 receptor antagonist anakinra may be an effective therapy for idiopathic recurrent pericarditis that is refractory to standard treatment; however, its role in management is not established.
Pericardial effusion is characterized by an increased amount of fluid in the pericardial cavity. Many patients with pericardial effusion are asymptomatic, and the effusion is discovered incidentally with chest radiography, CT, or echocardiography. In asymptomatic patients, most effusions are idiopathic; however, malignancy, infections, autoimmune disease, hypothyroidism, and iatrogenic causes (medications, anticoagulation therapy) should be considered (see Table 30). In countries where tuberculosis is endemic, more than 60% of effusions are caused by tuberculosis.
If cancer or bacterial infection is strongly suspected, pericardiocentesis should be considered for diagnostic purposes. In patients with a pericardial effusion of unknown cause and elevated inflammatory markers, empiric treatment of pericarditis may be reasonable. Drainage should be considered if large idiopathic effusions are present for more than 3 months because one in three patients progress to cardiac tamponade.
Cardiac tamponade occurs when fluid accumulation within the pericardial space compresses the heart and impedes diastolic filling. When fluid accumulates rapidly (such as with trauma, aortic dissection, or invasive cardiac procedures), tamponade may occur at relatively low pericardial volumes. Subacute or chronic processes, such as neoplastic disease or hypothyroidism, may be associated with much larger effusions (several hundred milliliters in volume).
Clinical signs of tamponade include tachycardia, hypotension, muffled heart sounds, and elevation of the jugular venous pulse. The y descent of the jugular venous pulse may be absent because passive filling of the ventricles is impeded by the intrapericardial pressure. This finding may be difficult to appreciate, especially in a patient with tachycardia. Pulsus paradoxus represents exaggerated ventricular interdependence and is a key clinical feature of cardiac tamponade. It is characterized by a fall in systolic pressure of greater than 10 mm Hg during inspiration. Pulsus paradoxus is not specific for tamponade and must be interpreted in conjunction with other clinical and echocardiographic features.
The ECG may demonstrate electrical alternans (related to a swinging motion of the heart within the pericardial fluid) or low voltage in patients with tamponade. If the accumulation of fluid has occurred slowly, the cardiac silhouette is typically enlarged on chest radiography.
Echocardiography is an essential tool in the diagnosis of cardiac tamponade because it defines the presence, distribution, and relative volume of pericardial fluid (Figure 32). Early diastolic collapse of the right ventricle, late diastolic collapse of the right atrium, and abnormal interventricular septal motion are features associated with cardiac tamponade. Additionally, Doppler evaluation may demonstrate a decrease in mitral inflow velocity of more than 25% with inspiration, which is the echocardiographic equivalent of pulsus paradoxus (Figure 33).
Cardiac catheterization is rarely necessary for diagnosis. The hemodynamic hallmarks of tamponade include blunting or loss of the y descent within the right atrial pressure waveform and elevated and equalized diastolic pressures. The latter reflects the transmitted effect of the intrapericardial pressure. Invasive arterial pressure recordings also show pulsus paradoxus.
Cardiac tamponade is life-threatening, and once a diagnosis is established, fluid removal is required. Drainage is most commonly accomplished with pericardiocentesis, with fluoroscopy or echocardiographic guidance. Surgical therapy via a subxiphoid approach is indicated to drain fluid when pericardiocentesis cannot be performed safely, to obtain pericardial tissue for diagnostic purposes, or to prevent recurrent pericardial effusion by creating a pericardial window (often used in cases of malignant pericardial effusion). In hemodynamically unstable patients, intravenous normal saline is used to stabilize the patient as a temporizing measure or as a bridge to definitive therapy.
After drainage of pericardial fluid, hemodynamic and clinical evaluation may occasionally disclose findings of underlying pericardial constriction, termed effusive constrictive pericarditis. If clinical evaluation and imaging techniques (such as CMR imaging) suggest an active inflammatory process, a course of medical therapy similar to that for acute pericarditis may be considered, and the patient should be re-evaluated before surgery is contemplated.
Constrictive pericarditis is characterized by pericardial thickening, fibrosis, and sometimes calcification that impair diastolic filling and limit total cardiac volume. Within developed countries, most cases are viral or idiopathic in origin. Other causes include cardiac surgery, chest irradiation, autoimmune disease, and tuberculosis or other bacterial infection. Tuberculosis remains a major cause of constrictive pericarditis within developing countries.
Patients with constrictive pericarditis most commonly present with indolent progression of right-sided heart failure symptoms, including peripheral edema, abdominal swelling, and fatigue. Dyspnea and fatigue limit exertion and are caused by increased diastolic pressures and limited ability to augment cardiac output due to the fixed stroke volume.
On physical examination, the jugular veins are distended, with prominent x and y descents. The height of the waveform does not fall or may increase during inspiration (Kussmaul sign), reflecting the fixed diastolic volume of the right heart. Early diastolic filling is unimpaired or even accentuated and is followed by sudden cessation when total acceptable volume is met, resulting in a high-frequency early diastolic sound (the pericardial knock). Characteristics that may be used to differentiate a pericardial knock from other diastolic sounds are listed in Table 31. Pulsus paradoxus is less frequent in constrictive pericarditis than in cardiac tamponade. Peripheral edema, ascites, hepatomegaly, and pleural effusions are common. Muscle wasting may be evident in advanced cases.
Constrictive pericarditis is diagnosed with imaging studies and hemodynamic evaluation. Chest radiography or CT may demonstrate partial or circumferential pericardial calcification, and CT or CMR imaging may demonstrate pericardial thickening (>3 mm). Importantly, constriction may exist in the absence of these findings. In one case series, 18% of cases of hemodynamically proven constrictive pericarditis occurred with normal pericardial thickness. CMR imaging may also demonstrate an inspiratory septal shift, a sign of ventricular interdependence.
Transthoracic echocardiography in constrictive pericarditis reveals normal right and left ventricular size and systolic function despite prominent symptoms and examination findings suggestive of heart failure. The echocardiographic finding of dilatation of the inferior vena cava reflects elevated right-sided filling (right atrial) pressure. Myocardial relaxation is impaired in myocardial disease, such as restrictive cardiomyopathy, but is unimpaired or even enhanced in constriction, in which early diastolic filling is rapid and unimpeded. Doppler echocardiography and tissue Doppler velocity are required to differentiate constrictive pericarditis from restrictive cardiomyopathy.
When constrictive pericarditis is suspected but not confirmed by echocardiography, cardiac catheterization can be performed. Invasive hemodynamic findings of constrictive pericarditis include a prominent y descent in the right atrial waveform, which corresponds with the dip of the right ventricular dip-and-plateau waveform (“square root sign”). The significant y descent and the right ventricular dip both represent unimpeded or rapid early diastolic ventricular filling. As inflow volume reaches the fixed pericardial constraint, pressure rises rapidly until maximum volume is achieved; pressure then remains constant, causing the plateau phase of the square root sign. A more specific finding is ventricular interdependence during simultaneous right and left ventricular systolic pressure measurement. During inspiration, right ventricular inflow is enhanced, and right ventricular systolic pressure rises; however, these changes occur with a concomitant decrease in left ventricular filling and reduction in left ventricular stroke volume and systolic pressure (Figure 34). The converse is seen during expiration.
Increased pericardial thickening and impaired distensibility may occur without fibrosis or calcification in the setting of acute or subacute inflammation. In these patients, constriction may be transient and resolve spontaneously. Patients with transient constrictive pericarditis present most commonly with symptoms of right-sided heart failure, although fever and chest pain may indicate the active inflammatory condition. Most cases are idiopathic; other causes include recent cardiac surgery, acute pericarditis, autoimmune disease, or chemotherapy. Systemic markers of inflammation (erythrocyte sedimentation rate and CRP) may be elevated in patients with transient constriction but are generally normal in patients with fixed constriction. Echocardiographic features are similar to those of fixed constriction; however, pericardial effusion is more likely to be present in patients with transient constrictive pericarditis.
Treatment of transient constrictive pericarditis is the same as for acute pericarditis. Treatment with anti-inflammatory agents for 2 to 3 months is reasonable in hemodynamically stable patients before recommending surgical pericardiectomy. Response to therapy is monitored clinically, echocardiographically, and, if initial inflammatory markers are elevated, serologically.
Patients with chronic pericardial constriction should be referred for surgical pericardial stripping (pericardiectomy performed via median sternotomy). In advanced cases, adequate resection of the pericardium may be difficult, leading to incomplete resolution of symptoms. Diuretic therapy to relieve symptoms of congestion may be useful in patients who are not deemed surgical candidates or in whom stripping was incomplete.