| Ralph Shabetai, MD |
May 1, 2000 |
Pericardial constriction results from scarring with the consequent loss of elasticity of the pericardial sac. This leads to impairment of ventricular filling, usually affecting both ventricles, but less commonly only one or the other, as they enlarge in mid to late diastole. As a result, the majority of ventricular filling occurs rapidly in early diastole. The basic aspects of this disorder are discussed elsewhere. (See "Pericardial compressive syndromes"). This card will review in detail the clinical features of constrictive pericarditis and the hemodynamically similar condition restrictive cardiomyopathy.
VENOUS PRESSURE ！ Using only a systemic venous pressure tracing, it is not possible to distinguish between constrictive pericarditis, restrictive cardiomyopathy, tricuspid regurgitation with an enlarged compliant right atrium, right heart failure, or circulatory overload with systemic congestion.
Kussmaul's sign ！ Kussmaul's sign represents the lack of the expected inspiratory decline in jugular venous pressure. It is often sought diligently in pursuit of the diagnosis of constrictive pericarditis. Even during normal breathing, however, Kussmaul's sign takes a forme fruste; respiratory variation of the mean central venous pressure is absent rather than reversed. Furthermore, Kussmaul's sign is also observed in right heart failure or systemic venous congestion of any cause, and in severe tricuspid regurgitation.
Careful inspection of right atrial pressure tracings in patients with constrictive pericarditis discloses that although the mean venous pressure fails to decline during inspiration, the y descent becomes deeper and steeper (show figure 1).
In addition, the y descent, classically the dominant wave of the venous pulse, may be equaled or exceeded in depth by the x descent.
Thus, the venous pulse of constrictive pericarditis should be considered an exaggerated version of the normal venous pulse. In contrast, the venous pulse in cardiac tamponade has a truly abnormal waveform characterized by attenuation of the y descent, or replacement of it by an upwardly sloping segment [1].
Equalization of pressures ！ Equalization (equilibration) of pressures in the four cardiac chambers is commonly said to be a major hemodynamic criterion for the diagnosis of constriction. This statement may be an oversimplification, because the pulmonary wedge pressure falls during inspiration while systemic venous pressure remains constant. Thus, it is not reasonable to expect the two pressures to be precisely equal throughout the respiratory cycle, and frequently equilibration is present only during inspiration (show figure 2).
Furthermore, equilibration of diastolic pressures occurs in some patients who have heart failure or an acute volume overload, but who do not have constrictive pericarditis. The dilated heart engaging the pericardium, and thereby simulating the hemodynamics of constrictive pericardial disease, is the most probable explanation for this phenomenon.
Left ventricular diastolic pressure that exceeds that in the right ventricle by 5 mmHg or more favors a diagnosis of cardiomyopathy rather than constrictive pericarditis. However, a significant proportion of patients restrictive cardiomyopathy show equalization of diastolic pressures as in constrictive pericarditis. Hemodynamic maneuvers, such as exercise, fluid loading and observing postextrasystolic beats, have been proposed as a means to cause diastolic pressures (equal in the two ventricles) to deviate. There is limited published experience with this approach.
Dip and plateau ！ The first observations of the dip and plateau configuration of ventricular pressure (also called the square root sign) during diastole led investigators to theorize that ventricular filling during early diastole must be unusually rapid and perhaps aided by augmented suction, and that ventricular filling must be halted by pericardial restraint from the end of the first third of diastole onward [2]. This conjecture was verified after it became possible to record the pattern of ventricular filling by ventriculography and subsequently by other imaging techniques. Although some disagree [3], most authors believe (as we do) that the same dip and plateau waveform characterizes restrictive cardiomyopathy [4].
It has been suggested that the dip and plateau pattern of ventricular diastolic pressure may not be a true pathophysiologic phenomenon, but an artifact of underdamping of a conventionally recorded pressure signal. As an example, ventricular pressure tracings from a diagnostic cardiac catheterization laboratory will frequently show significant oscillations when pressure changes are slow, such as during ejection and in diastole. The latter situation may simulate the early diastolic dip, and be followed by what passes for a plateau (show figure 3).
However, high fidelity pressure tracings obtained from a catheter with a micromanometer at its tip (show figure 4)
have shown conclusively that the dip and plateau pattern is real. Even though the high fidelity record never dips to subatmospheric levels, or even to zero, the wave remains sharply defined.
However, the dip and plateau is often grossly exaggerated by underdamping when recorded using a conventional fluid filled catheter and external transducer. Equilibration should be sought during the plateau phase, and may not be tight throughout the respiratory cycle (show figure 5).
DIFFERENTIATION BETWEEN CONSTRICTIVE PERICARDITIS AND RESTRICTIVE CARDIOMYOPATHY ！ Restrictive cardiomyopathy may be defined as a subset of diastolic myocardial dysfunction that simulates constrictive pericarditis both physically and hemodynamically. Because of their differing causes and treatments, cardiologists and other physicians must learn to differentiate the two conditions. In most cases, use of a simple algorithm yields the correct answer. An occasional case, however, cannot be diagnosed before biopsy or even surgical exploration.
History ！ A prior history of pericarditis, a systemic disease that affects the pericardium (eg, tuberculosis, connective tissue disease, malignancy), trauma, or radiation therapy makes the diagnosis of constrictive pericarditis more likely. On the other hand, a history of an infiltrative disease that may involve the heart muscle (eg, amyloidosis) favors the diagnosis of restrictive cardiomyopathy.
Physical examination ！ As mentioned above, observation of the jugular venous pulse or any other physical finding is of no help in distinguishing these two conditions.
Electrocardiogram ！ The electrocardiogram may be more helpful. Depolarization abnormalities (eg, bundle branch block), ventricular hypertrophy, pathologic Q waves, or impaired atrioventricular conduction strongly favor restrictive cardiomyopathy. Isolated repolarization abnormalities can occur in both conditions, but are more common in constrictive pericarditis.
Chest x-ray ！ Calcification of the pericardium (excepting scattered plaques) strongly suggests constrictive pericarditis (show radiograph 1).
However, the absence of calcification is equally compatible with either diagnosis. In one retrospective review of 135 patients with constrictive pericarditis that was confirmed surgically or at autopsy, 27 percent had pericardial calcification [5]. The cause was more likely to be indeterminant when calcification was seen (67 versus 21 percent in the absence of calcification).
Pericardial imaging ！ Many forms of constrictive pericarditis are associated with increased thickness of the pericardium detectable by MRI or CT imaging (show radiograph 2).
A recent study of 11 patients with constrictive pericarditis found an excellent correlation (r
0.95)
between measurement of pericardial thickness with transesophageal
echocardiography and that obtained by CT imaging [6].Pericardial thickness exceeding 3 to 5 mmHg is very suggestive of constrictive pericarditis; however, constriction is rarely caused by a thin tight peel of visceral pericardium. Thus, a normal appearing parietal pericardium does not rule out constrictive pericarditis.
To understand the difference in the hemodynamics of constrictive pericarditis and restrictive cardiomyopathy, it is important to note that in constrictive pericarditis, total cardiac volume is fixed by the noncompliant pericardium, and ventricular interaction is thereby greatly enhanced. In addition, changes in intrathoracic pressure are not transmited to the cardiac chambers because the pericardial space is obliterated. In restrictive cardiomyopathy, pericardial compliance is normal, and left ventricular function is normal or even reduced; repiratory variation of intrathoracic pressure is transmitted normally to the cardiac chambers.
Doppler echocardiography ！ As previously mentioned, restrictive cardiomyopathy and constrictive pericarditis share important hemodynamic characteristics and therefore have a number of Doppler characteristics in common. Most notable is a restrictive mitral inflow or ventricular filling pattern with striking E dominance and a short deceleration time (show echocardiogram 1A-1B).
Direct comparison of the two conditions, however, results in some
differences in the Doppler signals that are due to the enhanced ventricular
interaction and sparing of the ventricular septum in constrictive pericarditis.
The respiratory variation in ventricular filling velocity is usually minimal
(less than 10 percent). Patients with constrictive pericarditis may have
variation as high as 30 percent, and almost always at least 15 percent in
patients with severe constriction (show
echocardiogram 2).
By contrast, respiratory variation in peak ventricular filling
velocities remains normal in patients with restrictive cardiomyopathy. This
difference in the two diseases may be explained by the following mechanisms:
• In patients with constrictive pericarditis, the
pulmonary wedge pressure is influenced by the inspiratory fall in thoracic
pressure, while the left ventricular pressure is shielded from respiratory
pressure variations by the pericardial scar. Thus, inspiration lowers
pulmonary wedge pressure, and presumably left atrial pressure, more than left
ventricular diastolic pressure, thereby decreasing the pressure gradient for
ventricular filling. The less favorable filling pressure gradient during
inspiration explains the decline in filling velocity. Reciprocal changes occur
in the velocity of right ventricular filling [7,8].
These changes are mediated by the ventricular septum, not by increased
systemic venous return.
• In patients with restrictive cardiomyopathy,
inspiration lowers pulmonary wedge and left ventricular diastolic pressures
equally, thereby leaving the pressure gradient for ventricular filling and
filling velocity virtually unchanged.
A lesser left ventricular filling pressure gradient with constrictive
pericarditis also leads to a delay in mitral valve opening and therefore a
longer isovolumic relaxation time during inspiration. Prolonged isovolumic
relaxation of the left ventricle is a feature of both conditions, but this
finding varies with respiration in constrictive pericarditis but not
restrictive cardiomyopathy.
Some investigators have suggested that early rapid filling is even more rapid
than normal in constrictive pericarditis, but slower than normal in
restrictive cardiomyopathy [9].
This has not been our experience, nor that of other investigators who have
found identical patterns of ventricular filling in both conditions [10].
Atrial and ventricular diastolic compliance are determined by the pericardium
in patients with constrictive pericarditis. Furthermore, blood can transfer
easily from atrium to ventricle because no change in total cardiac volume
occurs. By contrast, the ventricles of patients with restrictive
cardiomyopathy are much stiffer than the atria. The atria typically enlarge
considerably and sometimes massively, and ultimately fail. Thus, late
ventricular filling velocity is reduced in patients with restrictive
cardiomyopathy and diastolic flow reversals occur; in comparison, flow
reversal in constrictive pericarditis occurs either in systole or in both
systole and diastole [8].
Differentiation from chronic
obstructive lung disease ！ Respiratory variation in mitral E velocity 15
percent is the main diagnostic criterion for constrictive pericarditis on
Doppler echocardiography, like pulsus paradoxus, but can also be present in
patients with chronic obstructive pulmonary disease. In an attempt to
distinguish between these disorders, the pulse-wave Doppler recordings of
mitral and superior vena cava flow velocities were compared in 20 patients
with chronic obstructive pulmonary disease who had a 25
percent respiratory variation in mitral E-wave velocity and 20 patients with
surgically proven constrictive pericarditis [11].
The patients with pulmonary disease had a marked increase in inspiratory
superior vena cava systolic flow velocity which was not seen in those with
constrictive pericarditis.
Coronary blood flow ！ Patients with either
constrictive pericarditis or restrictive cardiomyopathy have reduced coronary
flow reserve and peak hyperemic flow velocity compared to normals (show
figure 6).
However, coronary flow in constrictive pericarditis shows a rapid
acceleration and more rapid deceleration (velocity half-time <260 msec or
that corrected by sq rt RR <9.5) of diastolic blood flow compared to
restrictive cardiomyopathy (show
figure 7) [12].
This variance may be based upon differences in the pathogenesis of these
diseases, such as pericardial and epicardial versus myocardial involvement.
OCCULT CONSTRICTIVE PERICARDITIS ！ A
report in 1977 described 19 patients with a syndrome called occult
constrictive pericardial disease [13].
The symptom complex of this proposed syndrome was comprised of chest pain,
dyspnea, and fatigue, which may appear nonspecific. However, 12 of the 19 had
a history of prior acute pericarditis (which was recurrent in five). In
addition, two patients had pericardial calcification and, in 16, the ECG
showed nonspecific repolarization changes. A reasonable cause for pericardial
disease was present in 10.
The authors proposed that very mild constrictive pericarditis can cause these
symptoms in the absence of abnormal physical or hemodynamic findings when the
patient is evaluated in the basal state. To test this hypothesis, they
measured hemodynamics invasively before and after infusing a liter or warm
saline over a period of six to eight minutes to determine if occult
constriction would then become overt. Six patients known not to have heart
disease and 12 patients with myocardial disease served as controls.
The results can be summarized as follows:
• Saline infusion caused an elevation and
equalization of ventricular filling pressures, and development of pressure
waveforms in diastole characteristic of constriction (show
figure 8)
in the patients with occult constriction.
• Ventricular filling pressures and diastolic
waveform were unaltered in the subjects free from heart disease.
• The patients with myocardial disease had
elevated ventricular filling pressures, but unequally on the two sides.
Eleven of the symptomatic patients underwent pericardiectomy with dramatic
improvement. All eleven cases had mild gross or histologic evidence of
pericardial disease. The fluid challenge was repeated postoperatively in five
of the patients with negative results.
Based upon these findings, the authors recommended pericardiectomy for
disabling symptoms. However, we have strong reservations about this study,
even though it was conducted using high fidelity pressure tracings in a
laboratory well known for high quality hemodynamic studies. It is unclear why
or how such mild constriction could cause disabling symptoms, and why chest
pain was a feature. Furthermore, the dramatic relief by pericardiectomy is
also unexplained in these patients with low normal venous pressures that were
modestly elevated by a rapid large fluid challenge, and in whom the cardiac
output was normal at rest and was not changed by the infusion.
My recommendation is that this frequently performed saline infusion test is
seldom required, and if performed, results should seldom provide the reason
for pericardiectomy. Instead, patients suspected of having occult pericardial
constrictive disease should undergo cardiac catheterization, including
measurement of oxygen consumption during progressive bicycle exercise. A study
of this nature would document exertional dyspnea and fatigue, and may help
clarify the responsible mechanism.
EFFUSIVE CONSTRICTIVE PERICARDITIS ！ The
pericardial cavity is typically obliterated in patients with constrictive
pericarditis. Thus, even the normal physiological amount of pericardial fluid
is absent. However, pericardial effusion may be present in some cases. In this
setting, the scarred pericardium not only constricts the cardiac volume but
can also put the pericardial effusion under increased pressure leading to
signs suggestive of cardiac tamponade. (See
"Pericardial compressive syndromes", section on Tamponade versus
constriction). This combination is called effusive constrictive pericarditis [14,15],
a condition that used to be common in the subacute phase of tuberculous
pericarditis [16].
Hancock is primarily responsible for our current understanding of effusive
constrictive pericarditis [14,15].
He based his description on his experience with 24 patients undergoing
pericardiectomy for constrictive pericarditis, nine of whom had concurrent
effusion. Six of the nine had undergone hemodynamic studies before surgery. A
number of clinical clues suggested that a patient considered to have
constrictive pericarditis may actually have effusive constrictive pericarditis:
• Pulsus paradoxus (rare in classical
constrictive pericarditis because of the absence of transmission of the
inspiratory decline in pressure to the right heart chambers)
• Absence of a pericardial knock
• The Y descent less dominant than expected
• Kussmaul's sign frequently absent
In Hancock's experience, and in ours, the diagnosis often becomes apparent
during pericardiocentesis in patients initially considered to have
uncomplicated cardiac tamponade. Despite lowering the pericardial pressure to
normal, the persistence of elevated right atrial pressure and the development
of y dominance and impaired respiratory variation suggest that effusive
constrictive pericarditis may be present. However, a persistently elevated
right atrial pressure after pericardiocentesis may also be due cardiac
tamponade complicating right heart failure or tricuspid regurgitation. Thus,
appropriate studies should be performed to exclude these disorders before
making the diagnosis of effusive constrictive pericarditis.
The case illustrated in Figure 9 presented with clinical and hemodynamic signs
compatible with cardiac tamponade (show
figure 9).
After pericardiocentesis, however, the underlying constrictive
pericarditis became apparent.
Treatment ！ Effusive constrictive
pericarditis is important to recognize since it is the visceral, not the
parietal layer, that constricts the heart. This feature was apparent to
surgeons who did the early series of pericardiectomy for tuberculous disease [17].
Thus, if surgery is required, it is visceral pericardiectomy that must be
performed.