Constrictive pericarditis: its history and current status.

The diagnosis of constrictive pericarditis remains a challenge because it is often mimicked by restrictive cardiomyopathy. The last few years have seen numerous advances in our ability to differentiate between these two conditions which often have similar physical findings and hemodynamics. This review begins with a brief history of constrictive pericarditis; this is followed by an extensive discussion of newer etiologies, and then the classical clinical history and physical examination findings are described. Radiologic, electrocardiographic, and angiographic findings are discussed. The hemodynamics of constrictive pericarditis are reviewed. Recent results of echocardiographic and echo-Doppler investigations are presented. Emphasis is placed upon the limitations of M-mode echocardiography in the diagnosis of constrictive pericarditis. The value of echocardiographic Doppler studies of mitral and tricuspid flow velocity patterns, as well as of those in the pulmonary veins and hepatic veins, is described. Nuclear ventriculograms and angiocardiograms tend to show more rapid ventricular filling in constrictive pericarditis than in restrictive cardiomyopathy. Although only a small number of patients has been studied, these evaluations seem to have merit in separating restrictive cardiomyopathy from constrictive pericarditis. The role of computed tomography scanning and magnetic resonance imaging studies of pericardial thickness in confirming the presence of constrictive pericarditis is discussed. Abnormal pericardial thickening (> 3 mm) confirms the diagnosis of constrictive pericarditis, but only if the characteristic hemodynamic pattern is present. The usefulness of endomyocardial biopsy in recognizing specific varieties of restrictive cardiomyopathy is presented.(ABSTRACT TRUNCATED AT 250 WORDS)

Cardiac tamponade. A clinical or an echocardiographic diagnosis?

In most patients, cardiac tamponade should be diagnosed by a clinical examination that shows elevated systemic venous pressure, tachycardia, dyspnea, and paradoxical arterial pulse. Systemic blood pressure may be normal, decreased, or even elevated. The diagnosis is confirmed by echocardiographic demonstration of moderately large or large circumferential pericardial effusion and in most instances, of right atrial compression, abnormal respiratory variation in right and left ventricular dimensions, and in tricuspid and mitral valve flow velocities. Pulsus paradoxus may be absent with left ventricular dysfunction, atrial septal defect, regional tamponade, and positive-pressure breathing. Systemic venous pressure may be normal with localized tamponade of the left atrium or ventricle. Patients with moderately large or large pericardial effusions may have echocardiographic evidence of right atrial compression without clinical signs of elevated venous pressure or pulsus paradoxus. The majority of these patients have mild or moderate tamponade and if not subjected to pericardial drainage, should be observed closely. In some of these patients, when the etiology is known and the disease can be treated effectively with medication, e.g., nonsteroidal anti-inflammatory agents or adrenal corticosteroids in Dressler's syndrome or relapsing pericarditis, pericardial drainage may not be necessary.

Influence of acute right ventricular dysfunction on cardiac tamponade.

Echocardiographic and hemodynamic data were measured in nine closed chest dogs during graded cardiac tamponade (pericardial pressure 5, 10, 15 mm Hg) before and after production of diffuse acute ischemic right ventricular dysfunction. Right ventricular dysfunction was produced by intracoronary injection of nonradioactive microspheres (mean diameter +/- SD 54 +/- 4 microns) and caused a significant increase in right atrial pressure (7.6 +/- 1.4 vs. 1.6 +/- 1 mm Hg, p less than 0.001) and cross-sectional areas of both the right atrium (8.3 +/- 0.3 vs. 5.6 +/- 0.2 cm2, p less than 0.001) and right ventricle (8.8 +/- 0.4 vs. 5.7 +/- 0.4 cm2, p less than 0.001). Right atrial and ventricular collapse required a significantly larger pericardial effusion and pericardial pressure after right ventricular infarction than before. Mean aortic pressure had fallen 1.9 +/- 2% and 6.5 +/- 6.9% at the time of right atrial collapse (p = NS before vs. after right ventricular dysfunction) and 3 +/- 4.1% and 20.1 +/- 20.8% at the time of right ventricular collapse (p less than 0.03) before and after right ventricular dysfunction, respectively. In the presence of ischemic right ventricular dysfunction, echocardiographic signs of cardiac tamponade are less sensitive and occur later in the hemodynamic progression of cardiac tamponade. Pulsus paradoxus with cardiac tamponade was not prevented by coexisting ischemic right ventricular dysfunction.

Influence of splenectomy on hemodynamics during cardiac tamponade.

The effect of prior splenectomy on the hemodynamics of cardiac tamponade was investigated in 15 closed-chest pentobarbital sodium-anesthetized dogs. Hemodynamics were compared at baseline and during staged cardiac tamponade (pericardial pressures of 5, 10, and 15 mmHg) at control (n = 15) and after splenectomy (n = 8) and sham operation (n = 7). The fall in mean arterial pressure with cardiac tamponade was significantly greater in splenectomized dogs than in either sham-operated or control dogs (P less than 0.001). Cardiac output was more depressed at the third level of cardiac tamponade in splenectomized than in sham-operated or control dogs (12.8 +/- 14.5 vs. 29.3 +/- 8.7 and 25.4 +/- 9.4 ml.min-1.kg-1, respectively; both P less than 0.05 vs. splenectomy). Hemodynamic failure, defined as an inability to maintain mean arterial pressure greater than 50 mmHg for 5 min, occurred at a lower pericardial pressure in splenectomized than in sham-operated dogs (13.1 +/- 3.8 vs. 18.1 +/- 3.5 mmHg, P less than 0.05). Hematocrit increased significantly with cardiac tamponade in controls and sham-operated but not splenectomized dogs. The percent increase in hematocrit from baseline to the third stage of cardiac tamponade was 19.6 +/- 9.8 and 22.3 +/- 5.6% in control and sham dogs, respectively. Thus the canine spleen plays an important role in cardiovascular compensation to cardiac tamponade. Parallel changes in hematocrit suggest that a part of this response is due to splenic autotransfusion.

Effects of pericardiectomy on canine pulmonary blood volume during hypervolaemia.

STUDY OBJECTIVE: The aim was to determine the effect of the pericardium on the pulmonary blood volume response to intravascular volume loading. DESIGN: Changes in pulmonary lung volume were measured from radioactive counts over the lung during radionuclide ventriculography. Baseline measurements, and repeat measurements after infusion of 21 ml.kg-1 of the dog's own blood, were made both before and after a pericardiectomy. SUBJECTS: Ten closed chest, anaesthetised dogs were studied. MEASUREMENTS AND MAIN RESULTS: Prior to and following pericardiectomy, volume loading produced equivalent and significant increases in left atrial, mean pulmonary artery, and right atrial pressures (all p less than 0.05). Before pericardiectomy, radionuclide lung counts increased from 1606(SEM 348) to 1870(402) with volume loading, corresponding to a 16% rise in lung counts from baseline (p less than 0.05). Following pericardiectomy, a similar volume load did not result in a significant rise in lung counts [1588(245) to 1697(255), 9%, p = 0.16], but was accompanied by an increase in left ventricular diastolic volume, from 39.7(6.6) to 58.7(6.4) ml, p less than 0.05, and a decrease in systemic vascular resistance index, from 122,600(14,600) to 86,600(10,000) dynes.s.cm-5 x kg, p less than 0.05. CONCLUSIONS: These data support the concept that removal of the pericardium is accompanied by reduced pulmonary blood volume overload in response to intravascular volume loading. The mechanism appears to be related to improved left ventricular diastolic filling, perhaps the result of diminished ventricular interaction, and to redistribution of excess intravascular volume from the pulmonary to the systemic circuit.

Tuberculous pericarditis.

Tuberculosis is responsible for approximately 4% of cases of acute pericarditis, 7% of cases of cardiac tamponade, and, in older studies, 6% of instances of constrictive pericarditis. However, in some nonindustrialized countries, tuberculosis is a leading cause of pericarditis. The diagnosis is based on demonstration of tubercle bacilli in pericardial fluid or on histologic section of the pericardium, or proof of tuberculosis elsewhere in a patient with otherwise unexplained pericarditis. Treatment consists of triple drug therapy for at least 9 months (isoniazid, rifampin, and streptomycin or ethambutol). Pyrazinamide can be used for the first 2 months, and the total therapeutic period can then be shortened to 6 months after culture conversion. Three months of corticosteroid therapy may be useful in patients in whom pericardial effusion persists or recurs despite the use of antituberculous drugs. Surgical resection of the pericardium is indicated for recurrent or life-threatening tamponade, or when there is persistent elevation of systemic venous pressure unrelieved by pericardiocentesis. As many as one third to one half of patients will eventually require pericardiectomy despite adequate drug therapy.

Hemodynamic efficacy of rapid saline infusion and dobutamine versus saline infusion alone in a model of cardiac rupture.

Despite recent reports describing survival after cardiac rupture, the effectiveness of circulatory support while awaiting definitive surgical treatment is controversial. To assess the efficacy of volume expansion and pharmacologic support in cardiac tamponade due to cardiac rupture, a model of hemorrhagic cardiac tamponade was developed and treatment with rapid saline infusion and dobutamine was compared with rapid saline infusion alone in 15 closed chest dogs. A right ventricular wound of reproducible size was produced by deflating an aortic valvuloplasty balloon that had previously been passed by way of the internal jugular vein into the pericardial space and through a stab wound in the right ventricular free wall. Hemodynamic values were compared at baseline, during tamponade and after a rapid infusion (1 liter at 100 ml/min) of either saline solution alone or saline solution plus dobutamine (20 micrograms/kg per min). Atrial and pericardial pressures increased significantly in both groups. Mean arterial pressure, cardiac output and stroke volume increased with combined saline and dobutamine infusion to values similar to those at baseline (91 +/- 19%, 114 +/- 43% and 94 +/- 37% of baseline, respectively). In contrast, saline infusion alone caused a small increase in cardiac output but failed to significantly increase mean arterial pressure or stroke volume (76.8 +/- 14.2%, 55 +/- 18% and 51 +/- 17% of baseline, respectively). Combined rapid infusion of saline solution and dobutamine infusion has a more beneficial hemodynamic effect and may be more effective than rapid saline infusion alone in resuscitating patients with hemorrhagic cardiac tamponade due to cardiac rupture.

Recurrent pericarditis.

The most common background for recurrent pericarditis is that of acute nonspecific pericarditis. Relapsing pericarditis also may follow cardiac trauma, cardiac operations, myocardial infarction, and intrapericardial bleeding. The exact recurrence rate after initial attacks of idiopathic pericarditis is unknown but appears to be in the range of 15% to 32%. The mechanism of recurrent pericarditis is uncertain. An autoimmune response has been proposed, but this concept is unproved. Yoneda and coworkers, in a case of pericarditis due to coxsackie B virus, found no rise in antibody titer to this virus during recurrences. The prognosis, except for disabling pain and malaise, is good, and constrictive pericarditis, chronic myocardial disease, and cardiac tamponade are unusual complications. Although constrictive pericarditis may follow an initial attack of idiopathic pericarditis, it was reported in neither two other series of patients with relapsing pericarditis nor in this series. Cardiac tamponade has been reported as an occasional complication of relapses but did not occur in our patients. None of our patients died. Most patients with recurrent pericarditis respond to adrenal steroid therapy, but many times there is difficulty in weaning the patient from the drug. Because it is suspected that adrenal steroids may prolong attacks and promote tendency to further recurrences, initial therapy should be offered with aspirin or NSAIDs, and adrenal steroid therapy should be used only when there is no response to these agents. Recurrences may take place over a period lasting as long as 15 years, and patients with as many as 19 recurrences have been described.(ABSTRACT TRUNCATED AT 250 WORDS)

Cardiac tamponade in left ventricular dysfunction.

Echocardiographic and hemodynamic data were measured in closed-chest dogs during graded cardiac tamponade (pericardial pressure 5, 10, and 15 mm Hg) before and after production of diffuse ischemic left ventricular dysfunction. Left ventricular dysfunction was produced by intracoronary injection of nonradioactive microspheres (54 +/- 3.9 mm diameter). Changes in left atrial pressure with cardiac tamponade were influenced by coexisting left ventricular dysfunction. Left atrial pressure increased with tamponade and was equal to pericardial pressure before left ventricular dysfunction was produced. However, after left ventricular dysfunction was produced, left atrial pressure was significantly higher than pericardial pressure before tamponade, but it fell toward pericardial pressure when tamponade was produced. Pulsus paradoxus (greater than 10 mm Hg) was present in all animals with cardiac tamponade before left ventricular dysfunction but in only one animal afterward. During each level of tamponade, the inspiratory fall of aortic systolic pressure was greater before than with left ventricular dysfunction. The slope of the linear regression between pericardial pressure and millimeters of mercury of inspiratory fall in aortic systolic pressure was significantly greater before than with left ventricular dysfunction (0.74 +/- 0.12 versus 0.32 +/- 0.12, p less than 0.05). Left ventricular dysfunction caused a leftward and upward shift of the pericardial pressure-volume relation. As a result, right atrial and ventricular collapse occurred with significantly smaller volumes of pericardial fluid after than before left ventricular dysfunction. We conclude that pulsus paradoxus may be absent in cardiac tamponade with coexisting left ventricular dysfunction and unequal filling pressures. Echocardiographic signs of cardiac tamponade may occur with small effusions in the presence of left ventricular dysfunction.

Differentiation of constrictive pericarditis and restrictive cardiomyopathy by radionuclide ventriculography.

Patterns of left ventricular diastolic filling in five patients with unoperated constrictive pericarditis, the same five patients following pericardiectomy, five patients with restrictive cardiomyopathy, and 14 healthy control subjects were studied by radionuclide ventriculography. Patients with constrictive pericarditis had more rapid peak left ventricular filling rates (mean 5.62, range 4.23 to 7.32 end-diastolic volumes [EDV] per second) compared to control subjects 3.44, range 2.62 to 4.45 EDV/sec, p less than 0.05). Heart rate-corrected first one-third and first one-half diastolic filling fractions were greater in patients with preoperative constrictive pericarditis compared to members of the restrictive cardiomyopathy and control groups p less than 0.05). Following pericardiectomy, patients with constrictive pericarditis had significant decreases in peak filling rate and corrected filling fractions, and all diastolic filling measurements were indistinguishable from those of control subjects. These noninvasively obtained data indicate that patients with preoperative constrictive pericarditis have an increased rate of left ventricular early diastolic filling compared to patients with restrictive cardiomyopathy and control subjects, and that these findings return to normal following surgical removal of the pericardium.

Cardiac tamponade: a comparison of right versus left heart compression.

It has been postulated that in cardiac tamponade, the hemodynamic effects of compression of the right heart chambers and great veins are more important than are the effects of left heart compression. In 10 anesthetized dogs with surgically compartmented pericardium, the hemodynamic effects of right atrial and right ventricular compression were compared with the hemodynamic effects of left atrial and left ventricular compression. The effects of right heart compression, left heart compression, and then effects of combined right and left heart compression, were compared at three levels of intrapericardial pressure: 10, 15 and 20 mm Hg. Aortic mean pressure decreased significantly at each level of intrapericardial pressure with right-sided tamponade but not with left-sided tamponade. Left atrial mean pressures decreased significantly with right-sided tamponade and increased with left-sided and combined tamponade. Right atrial mean pressures increased significantly with right-sided and combined tamponade, but not with left-sided tamponade. Heart rate increased significantly with each of the three varieties of tamponade. Cardiac output and stroke volume, which decreased with each variety of tamponade, were significantly lower during right-sided than during left-sided tamponade. Combined tamponade lowered stroke volume more than did right-sided tamponade, and lowered cardiac output more at 15 and 20 mm Hg intrapericardial pressure. It is concluded that, in this preparation, right-sided cardiac compression has more important hemodynamic effects than does left-sided compression. However, left-sided tamponade still makes a significant contribution to the total hemodynamic picture of cardiac tamponade.

Regional cardiac tamponade: a hemodynamic study.

In 10 dogs, atrial tamponade, ventricular tamponade and then combined atrioventricular (AV) tamponade were produced at 10, 15 and 20 mm Hg intrapericardial pressure. Cardiac output decreased significantly at each level of cardiac tamponade; the changes in cardiac output and mean aortic pressure were comparable with atrial and ventricular tamponade. Combined atrial and ventricular tamponade produced significantly greater increases of right and left atrial pressure and significantly greater decreases of cardiac output than did either atrial or ventricular tamponade. During atrial tamponade only, a significant pressure gradient developed between the venae cavae and the right atrium. Compression of both ventricles by tamponade has a much greater hemodynamic effect than does compression of either ventricle alone. Compression of the entire heart has a greater hemodynamic effect than does compression of the atria alone or the ventricles alone. Compression of the great veins has a potential effect in tamponade, demonstrable when the ventricles could fill normally.

Recurrent acute pericarditis: follow-up study of 31 patients.

Thirty-one patients with recurrent pericarditis were observed for periods of 2 to 19 years. Twenty-four had idiopathic pericarditis; four had postoperative or posttraumatic pericarditis, two had postinfarction pericarditis and one had recurrent pericarditis after anticoagulant-induced intrapericardial bleeding. In 24 patients (Group I), recurrences were documented by electrocardiographic changes, echocardiographic evidence of pericardial fluid or a pericardial rub as well as chest pain. In seven patients (Group II), recurrences were documented only by increased white blood cell count, increased erythrocyte sedimentation rate or fever in addition to pain. In 19 patients, the duration of the active or recurrent process was 5 years or more; in 7, it was 8 years or more. Three patients had cardiac tamponade in the initial attack; none had tamponade during recurrences. No patient developed congestive heart failure, constrictive pericarditis or cardiac arrhythmias with recurrences. Immunoelectrophoresis showed normal findings or minor deviations in 11 patients studied; B cell and T cell lymphocyte counts were normal in 10 patients and showed minor deviations in 3. Antinuclear antibody studies were normal in 19 of 22 patients and positive in low titer in 2. Most patients required adrenal steroid therapy for pain relief; steroid withdrawal was often difficult. Pericardiectomy was done in nine patients; in only two was this followed by clear-cut relief. In this group of 31 patients, 22 of whom were observed for 5 years or more, recurrent attacks of chest pain were the only major disabling feature of their pericarditis.