In pericardial constriction (constrictive pericarditis), a variably thickened layer of visceral or parietal pericardium, or both, surrounds some or all of the cardiac chambers and progressively restricts filling of the ventricles. Because of anatomic pericardial encasement of the heart with resultant marked elevations in atrial pressure, 75% of ventricular filling occurs very rapidly and at high velocity during the first third of diastole, thus causing the characteristic “dip and plateau” or “square-root sign” on the left and right ventricular pressure tracings. In most cases, diastolic filling is restricted in both ventricles, which become increasingly interdependent; that is, filling of both ventricles depends on the relative motion of the interventricular septum during diastole, the phenomenon known as ventricular coupling.

Classic “chronic constriction” is conventionally thought of as a process that proceeds over many months to years. However, the clinical spectrum of constrictive pericarditis has changed since the early 1970s in the Western hemisphere, primarily because tuberculosis has become relatively infrequent, whereas cardiac surgery has become relatively commonplace. In contrast to the “chronic” variant of constriction, in which symptoms evolve over the course of many months and years and which represents progressive fibrosis of the pericardium, most cases are now “subacute,” with constriction evident in the 3- to 12-month period after the pericardial insult (e.g., viral infection of the pericardium or cardiac surgery). The patient with constriction after radiation to the mediastinum for treatment of cancer is the primary exception to this rule, when constriction develops months to years after treatment.

Two other variants also exist. In the patient with active pericarditis, the inflamed pericardium can thicken rapidly over a few weeks and produce “acute” symptoms of constriction. Finally, usually after cardiac surgery and at times with acute pericarditis, “transient” constriction develops with elevated jugular venous pressure and clinical signs of constriction. This constrictive pattern resolves after institution of antiinflammatory therapy with nonsteroidal antiinflammatory agents or steroids. Whether chronic constriction develops at a later date is not known.

Knowledge of three other clinical syndromes of constriction that may be acute, subacute, or chronic but are not “classic” is important. In “regional” or local constriction, pericardial thickening is present only over certain chambers of the heart and occurs most frequently shortly after cardiac surgery, when pericardial inflammation and thickening occur over the right side of the heart, or with neoplasm. Affected patients have evidence of pulmonary or systemic venous congestion but usually not both. Effusive-constrictive disease is the variable combination of findings of cardiac tamponade and constriction in some patients. Abnormal pulsus paradoxus is more common in these patients than in patients with constriction. A jugular venous waveform in which the x descent is steeper than the y descent is also suggestive of effusive-constrictive disease, and the presence of Kussmaul sign in
a patient with pericardial effusion is suggestive of effusive-constrictive disease rather than isolated cardiac tamponade. More often than not, the diagnosis is suspected when elevation in jugular venous pressure persists after pericardiocentesis.

Finally, latent or “low-volume” constriction should be suspected in the patient with persistent dyspnea, fatigue, and mild lower-extremity edema after aggressive diuresis normalizes the central venous pressure. Volume replacement unmasks constrictive hemodynamics.

Before the modern medical and surgical era, pericardial constriction was most commonly caused by tuberculosis or was thought to be idiopathic. Of 231 cases of constriction verified by surgery or autopsy at the Mayo Clinic from 1936 to 1982, the causes were idiopathic factors in 73%, pericarditis in 10%, pyogenic infection in 6%, radiation in 5%, and, least frequently, arthritis, cardiac surgery, or “other causes” in 2% each (1). In the period from 1985 to 1995, causes in order of incidence were idiopathic in 45 (33%) of 135 cases; after cardiac surgery events (18%); after pericarditis (16%); after radiation (13%); “other causes,” including neoplasm, trauma, and drugs (10%); inflammatory arthritides (7%); and, in rare cases, pyogenic infection (3%). These findings have been confirmed by other investigators (2,3). Other pathologic conditions reported less frequently to cause pericardial constriction include mesothelioma; uremia (chronic, in which affected patients are receiving dialysis); hemopericardium after trauma or thrombolytic therapy or related to coagulopathies; Dressler syndrome after myocardial infarction; vasculitis; drugs, including those used for lupus (hydralazine and procainamide) and migraine prophylaxis (methysergide); hypereosinophilia syndromes; amyloidosis; Whipple disease; and sarcoidosis (Table 27.1).

The patient with pericardial constriction most often is initially seen with symptoms of venous congestion resembling right-sided heart failure with normal left- and right-ventricular systolic function. Most common complaints by patients include lower-extremity swelling, dyspnea, and effort intolerance related to pulmonary venous congestion, and abdominal discomfort from hepatic distention.

In the patient with the acute variant of pericarditis with constrictive physiology, chest pain may be the predominant complaint. In
advanced cases of constriction, ascites occurs and may be more remarkable than the lower-extremity edema, although some patients have anasarca. Most patients cannot recall any history of antecedent pericarditis. Many of these patients have undergone extensive prior evaluation for hepatic disease with cirrhosis or congestive heart failure before the diagnosis of constriction is suspected. Unsuspected findings on two-dimensional and Doppler echocardiography performed for the diagnosis of congestive heart failure prompted an evaluation for constrictive pericarditis in 40% of patients who underwent pericardiectomy at the Mayo Clinic in the 10 years before 1997 (4).

Most patients with constriction have at least mild tachycardia, especially with even minor exertion. In these patients, stroke volume cannot increase with exercise, and the major compensatory mechanism is an increase in heart rate. Patients with chronic constriction may exhibit atrial fibrillation. Blood pressure is usually normal, but it may be low or even hypertensive.

The patient who has undergone aggressive diuresis may have orthostatic hypotension. Abnormal pulsus paradoxus is rare, occurring only if effusive-constrictive disease or chronic obstructive pulmonary disease is present. The patient with severe limitations of cardiac output may manifest peripheral cyanosis with cool extremities. The patient with hepatic failure or cirrhosis from increased hepatic venous pressure may be jaundiced. Funduscopic examination may reveal that retinal veins are engorged. The hallmark of pericardial constriction is elevated jugular venous pressure with rapid and sharp x and y descents, which produce the characteristic W wave on the jugular venous tracing. Often, the jugular venous pressure is so elevated that the veins must be examined with the patient in the sitting position, or even standing, to see the top of the venous column. In some patients who have undergone overdiuresis, the jugular venous pressure may not be elevated, but the classic physical findings are unmasked with volume infusion. Kussmaul sign (the paradoxic increase in jugular venous pressure during inspiration) is present in most patients. A cardiac impulse may not be palpable, leaving the patient with a “quiet precordium,” although an early diastolic impulse corresponding to the pericardial “knock” is occasionally detected. The first and second heart sounds are normal. The loud early diastolic S₃ or “knock” may be confused with splitting of S₁. Hepatomegaly, ascites, and even splenomegaly may be detected on abdominal examination, and lower-extremity edema is common. If constriction is regional and restricted to the right side of the heart, systemic venous congestive symptoms may be present, whereas pulmonary congestion may not.

The electrocardiogram almost universally displays nonspecific T-wave abnormalities. Voltage is usually normal but may be decreased or increased. The chest radiograph shows normal cardiac size but may reveal an enlarged superior vena cava or azygous vein, or both, and often shows bilateral pleural effusions. In the modern era, pericardial calcification is only very rarely identified on chest radiographs.

Transthoracic echocardiography with spectral Doppler imaging may be the first clue to the diagnosis of pericardial constriction. Transthoracic echocardiography is unreliable for the detection of pericardial thickening. Pericardial thickness as measured by transesophageal echocardiography correlates well with measurements by computed tomography (r > 0.95; p < 0.0001), but this technique has not been widely accepted (5). Two-dimensional echocardiography reveals the restricted motion of the myocardium and the ventricular interdependence with a flat left ventricular posterior wall, absence of diastolic ventricular expansion, and bowing of the interventricular septum toward the left ventricle during inspiration, with an abnormal septal “bounce” in early systole. The inferior vena cava is usually dilated, but this can represent elevated central venous pressure from any cause.