Echocardiography in Pericardial Effusion
ABDUL J. TAJIK, M.D. Rochester, Minnesota
Echocardiography has become established as the procedure of choice for the detection, confirmation and serial follow-up of patients with pericardiai effusion. in this article the technic and pitfalls of recording, and the criteria and their sensitivity for the diagnosis of pericardiai effusion are reviewed. in addition, echographic findings in special instances, such as accumulation of pericardiai effusion behind the left artium, the swinging heart syndrome and cardiac tamponade, are discussed. Although Edler [l] mentioned separation of the echo of the anterior heart wall from that of the chest wall in patients with pericardiai effusion, it was not until Feigenbaum and colleagues [2-51 published a series of papers in the mid 1960’s that the usefulness of echocardiography in pericardial effusion was established; other papers confirmed their findings [6- 141. The subsequent popularity of echocardiography in this country is, to a great extent, directly attributable to these findings. Today, echocardiography is considered to be the procedure of choice for evaluating patients suspected of having pericardial effusion. This high place afforded to echocardiography is due to its being an accurate, easy to perform and noninvasive method. It must be remembered, however, that the results obtained are entirely dependent on the knowledge, experience and technical skills of the examiner. The echocardiographer must be thoroughly familiar with the ultrasonic cardiac anatomy and various intracardiac landmarks. Many previously published false-positive and false-negative echograms in pericardial effusion can be directly attributed to faulty technic. ANATOMY AND TECHNIC OF RECORDING
From the Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Requests for reprints should be addressed to Dr. Abdul J. Tajik. Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55901.
The normal pericardium consists of two layers: the outer, fibrous pericardium and the inner, serous pericardium. The fibrous pericardium is a cone-shaped bag that has attachments anteriorly to the sternum, inferiorly to the diaphragm, posteriorly to the pulmonary veins and superiorly to the great vessels. The serous pericardium consists of two layers, the inner of which lines the myocardium and is called the visceral pericardium or epicardium. Visceral pericardium extends a short distance over the great vessels and then is reflected on itself to form the parietal layer, which lines the fibrous pericardium. The space between these two layers is the pericardial cavity, which normaliy contains an average of about 20 ml of fluid. Because posteriorly the pericardium is reflected from the back of the left atrium onto the pulmonary veins, no potential space exists for the accumulation of
July 1977
The American Journal of Medicine
Volume 63
29
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSION-TAJIK
Figure 1. Technic of recording echograms at different gain settings. Both echograms were recorded at the level of the choroae tendineae (C). At lower recording intensity (left part of figures), only the strong echo originating from the pericardial lung interface is recorded(P). As the intensity of the echograph is gradually increased, next the epicardial (EP) and lastly the weakest endocardial (EN) echo are recorded A, all three echoes (periwdial, epicardial and endocardial) are readily seen. No clear anterior echo-free space can be recognized in this figure. 6, epicardial and pericardial echoes are in close opposition, and therefore no clear separation can be seen between these two echoes during the cardiac cycle. With a higher gain setting, the endocardial and myocardial echoes can be well recorded. Note that an anterior echo-free space is readily seen in this figure. During ventricular systole, the anterior right ventricular wall echo separates from the chest wall echo. This is not an uncommon finding in normal subjects. RV = right ventricle; VS = ventricular septum; MV = mitral valve; L V = left ventricle.
pericardial fluid behind the left atrium at the level of the pulmonary veins. However, it must be remembered that there also exists posteriorly (directly behind the left atrium) a blind recess of pericardial cavity which is bordered by the pericardial reflection between the pulmonary veins. This recess is called the oblique sinus of the pericardium. It is in this region that, under special circumstances, pericardial fluid can and does accumulate. One cannot overemphasize the importance of proper recording technics when looking for pericardial effusion. Clear visualization of the anterior right ventricular wall and of the left ventricular posterior wall is necessary and
30
July 1977
The American Journal of Medicine
Volume 63
can be obtained by careful adjustments of reject, gain, intensity and damping controls. With the patient in the recumbent position and the trunk elevated 20 to 30 degrees, the examination is begun by first recording the familiar mitral valve echo. From this recording position, the transducer is tilted inferolaterally to record echoes of the left ventricular posterior wall at the chordae tendineae level (Figure 1). Special care must be taken that all the components (endocardium, epicardium and pericardium) of the left ventricular posterior wall are clearly recorded. It is a good practice to record the echogram at different gain settings. While the left ventricular posterior wall is recorded, the intensity of
ECHOCARDIOGRAPHY
IN PERICARDIAL
EFFUSION-TAJIK
RV
Figure 2. Echograms of two patients in congestive heart failure who had smallpericardial effusions. A, patient with a large atria/ septal defect. Echogram demonstrates the features of right ventricular volume overload (increased right ventricular dimension and abnormal ventricular septal motion). As the recording intensity of the echograph machine is reduced (middle of figure), the different components of the left ventricular posterior wall can be easily distinguished. The pericardial echo (P) demonstrates only minimal anterior motion in systole. An echo-free space representing a small amount of pericardial fluid is well seen as the epicardial (EP) echo separates from the pericardial echo during systole. 6, patient with severe mitral regurgitation. Echogram demonstrates the increased dimension of the left ventricle and hyperdynamic motion of both the septum and the left ventricular posterior wall. When the gain is reduced (middle of figure), o&y the strong pericardial and epicardial echoes are recorded, and a clear echo-free space between these two echoes becomes evident. It represents a small amount of pericardial effusion. No anterior echo-free space is recorded. In both of these echograms, note that the pericardial effusion could only be delineated when the gain and the intensity of recording were reduced. At higher gain settings (left and right half of figures), the echo-free space becomes obliterated by echoes originating from the pericardial fluid. RV = right ventricle; C = chordae tendineae; L V = left ventricle; VS = ventricular septum; PF = pericardial fluid; EN = endocardium.
July 1977
The American Journal of Medicine
Volume 63
31
ECHOCARDIOGRAPHY
IN PERICARDIAL
EFFUSION-TAJIK
the echograph should be reduced to obtain optimal definition of the strong posterior pericardial echo, which normally moves anteriorly during systole (Figure 1). The pericardial echo having been defined, the intensity should be gradually increased to identify, next in sequence, the epicardial and then the endocardial echoes. With proper gain settings, the relatively echo-free space between the endocardium and the epicardium can be filled in by echoes representing myocardium (Figure 1B). In normal subjects, the epicardial echo of the left ventricular posterior wall is in close contact with the parietal pericardium and pleura at the pericardial lung interface behind the heart. Similarly, the echoes from the anterior wall of the right ventricle are in close contact with the nonmoving chest wall echoes. Occasionally, in normal subjects, a small echo-free space may exist between the chest wall echoes and the anterior wall of the right ventricle; this is believed to be due to the fact that epicardial fat or loose areolar tissue or a portion of the lung is interposed between the chest wall and the anterior right ventricular wall (Figure 1B). ECHOGRAPHIC CRITERIA FOR DIAGNOSIS OF PERICARDIAL EFFUSION The following changes appear in the echocardiogram with the development of pericardial effusion. Demonstration of an echo-free space between the right ventricular anterior wall and the chest wall, and/or between the left ventricular posterior wall and the pericardial lung interface, indicates the presence of pericardial effusion. It has now been well established that the pericardial fluid first appears posteriorly in the dependent portion of the pericardial cavity. Therefore, the earliest manifestation of the presence of a tiny pericardial effusion is a slight separation, through systole and part of diastole, of the left ventricular posterior wall epicardial echo from a relatively stationary pericardial echo (Figure 2A). With increasing accumulation of fluid, this posterior echofree space becomes evident throughout the entire cardiac cycle (Figure 2B). In the absence of an anterior echo-free space, a small posterior echo-free space should suggest the presence of a small amount of pericardial effusion. With increasing amounts of pericardial fluid, the pericardial echo becomes stationary because the pericardial fluid, which acts as an insulator, attenuates the transmission of the left ventricular posterior wall motion to the parietal pericardium. Small amounts of pericardial effusion are not uncommon in patients who are in congestive heart failure. In both of the examples shown in Figure 2, a small amount of pericardial fluid is present as indicated by the small echo-free space posteriorly. No anterior echo-free space could be demonstrated. With further increase in the pericardial effusion, the posterior space widens; with larger effusions, the fluid
32
July 1977
The American Journal of Medicfne
Volume 63
also appears anteriorly as an anterior echo-free space (Figure 3). In the presence of anterior effusion, the amplitude of motion of the anterior wall of the right ventricle becomes exaggerated. A large pericardial effusion is diagnosed when there is a further increase in the depth of the anterior and posterior echo-free spaces. Frequently, but not invariably, when a large anterior and posterior effusion is present, the heart may be observed to be moving freely in the pericardial sac (swinging heart). In the current-day practice of echocardiography, it is imperative that complete scans of the left ventricle be recorded in each patient [ 15,161. A typical M-mode scan of the left ventricle in a patient with moderate pericardial effusion is shown in Figure 4. As can be seen on the left of the echogram at the level of the chordae tendineae, pericardial fluid can be recorded both anteriorly and posteriorly behind the left ventricular posterior wall. As the beam is tilted superiorly and medially toward the base of the heart (aortic valve recording position with the left atrium posteriorly), the pericardial fluid is seen to disappear at the atrioventricular junction, and no fluid is recorded behind the left atrium. This feature has been a consistent and reproducible finding in pericardial effusion. Figure 5 is from another patient with a moderatesized pericardial effusion, again demonstrating that, as the sound beam is tilted toward the base of the heart, the posterior echo-free space is markedly diminished in size but does not completely disappear behind the left atrium. The possibility of echographicdemonstration of pericardial fluid posterior to the left atrium has only recently been appreciated. Figure 6 demonstrates a base to apex scan from a patient with a large pericardial effusion. This echogram was obtained with the patient’s trunk elevated approximately 30 degrees. On the left of the echogram, with the transducer tilted toward the base of the heart, an anterior echo-free space is recorded; in addition, a small echo-free space is evident behind the left atrium. In the middle of the figure, at the level of the left ventricle, moderate-size anterior and posterior echo-free spaces are recorded, and on the right of the figure with the transducer tilted toward the apex of the heart, both the anterior and the posterior echo-free spaces are seen to enlarge; these findings indicate that the bulk of this pericardial fluid was surrounding the apex of the heart. Occasionally, in patients with large pericardial effusion, and specifically in those with the swinging heart syndrome, the anterior echo-free space may become more striking than the posterior echo-free space. In such situations, difficulty may be encountered in recording posterior-lying structures because the anterior right ventricular wall becomes a very strong reflector of ultrasound and diminishes the energy
ECHOCARDIOGRAPHY
available for deeper penetration. We have also seen patients with large posterior pericardial effusions and without any significant anterior effusion. Figure 7 shows the echogram of a patient with idiopathic pericardial effusion. Note that a large posterior pericardial effusion is present, and there is little if any anterior effusion. This echogram was also obtained with the patient’s trunk elevated to 30 to 40 degrees; this ordinarily should have accentuated the anterior effusion. On the other hand, the echogram in Figure 8 demonstrates a large anterior echo-free space without any significant posterior echo-free space. Such an echogram rarely results from pericardial effusion and, unless proved otherwise, should be regarded as negative for pericardial effusion. In most instances this anterior echo-free space is a result of large amounts of epicardial fat or retrosternal extracardiac mass lesions, such as thymoma or lymphoma [ 17,181. Although such an echogram may conceivably be found in a patient with loculated anterior pericardial effusion, we have to date not seen such an example and, therefore, are reluctant to diagnose pericardial effusion on the basis of a large anterior echo-free space alone. SENSITIVITY OF METHOD AND QUANTIFICATION OF EFFUSION
The early experimental studies in dogs had indicated that a minimum infusion of 50 ml of fluid in the pericardial sac was necessary before echocardiography could reliably demonstrate an effusion [2]. Although exact data were lacking for human subjects, it was generally believed, until recently, that echocardiography could reliably detect 50 to 100 ml of pericardial fluid. In this regard, the recent publication of Horowitz et al. [ 191 has been of great importance. In their attempts at quantifying and characterizing various patterns of pericardial effusions, they established the lower limit of sensitivity of a technically optimal, properly interpreted echocardiogram as 15 to 20 ml. Using a spheroid model of the heart, they also proposed a method of quantitating pericardial effusion, but they were quick to acknowledge that such a method was not extremely accurate. In our practice, we do not attempt accurate quantitation of pericardial effusion but are satisfied in using echocardiography as a technic for the detection and gross semiquantitative (minimal, small, moderate, large) estimation of the fluid. PITFALLS OF RECORDING TECHNIC As previously emphasized, a proper echographic technic is essential for arriving at the correct diagnosis of pericardial effusion. Improper use of echographic controls (reject, depth compensation, intensity and gain settings) accounts for most of the inadequate and confusing results. In particular, excessively low gain
IN PERICARDIAL
EFFUSION-TAJIK
settings result in incomplete identification of various layers of the left ventricular posterior wall; too high gain settings can result in excessive echoes from the pericardial fluid which will obscure the recognition of an echo-free space (Figure 2). In this regard, if the technic of recording echograms at various gain settings is religiously followed, the incidence of false-positive results and, more specifically, false-negative results can virtually be eliminated. Another pitfall in the recording technic pertains to the direction of the transducer beam. In the literature it has been emphasized that when the transducer is tilted too far medially, an echo-free space behind the posterior wall can be recorded on account of structures such as the coronary sinus, pulmonary veins, descending aorta and other posterior mediastinal structures. Occasionally, with the transducer directed medially, the mitral annulus can be mistaken for the left ventricular posterior wall, and rarely the medial aspect of the left ventricular posterior wall may be recorded. In both of these instances, a false-positive diagnosis of pericardial effusion may be made. This was particularly true in the early days of echocardiography when Polaroids were in use, and the scans of the left ventricle were infrequently recorded. In current practice, with the use of continuous M-mode scans of the left ventricle, confusion of pericardial effusion with the aforementioned structures has become most unlikely. SWINGING HEART SYNDROME
With large anterior and posterior pericardial effusions, the heart may move freely within the pericardial sac. This type of cardiac motion abnormality (swinging heart) was initially described by Feigenbaum and associates [3] in 1966. The free motion of the heart is attributed to the absence of the restraining function of the pericardium. The heart, suspended by the great vessels, swings freely within the pericardial sac, causing both the anterior and the posterior walls of the heart to move synchronously, that is, anteriorly during systole and posteriorly during diastole or vice versa. Generally, but not invariably, greater swing occurs with larger effusions and especially those of malignant origin. As a consequence of the free swinging of the heart, abnormal motion of various structures, such as the septum, mitral and tricuspid valves, and aortic and pulmonary valves, has been noted [ 20-231. As the heart swings anteriorly during systcle, both the anterior and the posterior walls of the heart, as well as the ventricular septum, also move anteriorly, and therefore an abnormal ventricular septal motion is not infrequently recorded in this situation. Various types of mitral valve prolapse patterns have also been described in patients with swinging heart syndrome. The most common pattern is the typical late-systolic mitral valve prolapse.
July 1977
The American Journal of Medicine
Volume 63
33
Figure 3. Patient with moderate pericardial effusion (idiopathic). Echogram demonstrates anterior (smaller) and posterior (larger) echo-free spaces. The motibn of the anterior wall of the right ventricle, ventricular septum and left ventricular posterior wail is normal; that is, during ventricular systofe the right ventricular anterior wail and ventricular septum move posteriorly as the let? ventricular posterior wall moves anteriorly. PF = pericardial fluid; C = chordae tendineae; VS = ventricular septum; EN = endocardium; EP = epicardium; P = pericardium.
Figure 4. Conti&ous M-mode scan from the body of the left ventricle to the base of the heart from a patient with moderate pericardial effusion. At the ventridular level, moderate anterior and posterior echo-free spaces can be seen, As the beam is tilted toward the aortic root with the left atrium behind, the echo-free space disappears behind the left atrium. The motion of the right ventricular anterior wall, ventricular septum and left ventricular posterior wall is normal. PF = pericardial fluid; RV = right ventricle; LV = ieft ventricle; PW = posterior wall of the left ventricle; MV = mitral valve; A V = aortic valve; LA = left atrium. From Lemire F, Tajik AJ, Giuliani ER, et al. [ZO].
Figure 5. M-mode scan from the base of the heart (aortic root and left atrium) toward the body of the left ventiicte at the level of the chordae tendineae. This echogram is from the same patient as in Figure 3. On the left of the echogram is a suggestion of a tiny echo-free space behind the left atrial wall which appears to be continuous with the larger echo-free space behind the left ventricular posterior wall. PF = pericardial fluid; VS = ventricular septum; MV = mitral valve; EN = endocardium; EP = epicardium; P = pericardium; A,V = aortic valve; LA = left atrium.
Figure 6. Condensed (slow paper speed) M-mode base (left) to apex (right) scan of the heart from a patient with a large pericardial effusion. This patient had systemic amyloidosis with pericardial as well as myocardial involvement. Echogram was performed with the patient’s trunk elevated about 30 degrees. At the level of the left ventricle (middle of figure), moderate-size anterior and posterior echo-free spaces can be seen. As the echo beam is angulated toward the apex of the heart, both the anterior and the posterior echo-free spaces widen. This demonstrates the presence of a large pericardial effusion surrounding the apex of the heart. Also note the increased thickness of the septum and left ventricular posterior wall secondary to amyloid infiltration. Abbreviations are the same as for the other figures.
Figure 7. effusion
Echogram from a patient with a large pericardial effusion anteriorly. The motion of the cardiac wails remains normal.
that is almost entirely located posteriorly
with only little
Figure 6. This echogram, from a patient with large retrosternal lymphoma involving the pericardium, demonstrates an M-mode scan from the base of the heart to the body of the left ventricle. At the level of the left ventricle, a small posterior echo-free space can be seen which represents pericardial effusion. However, note that there is a large anterior echo-free space between the chest wall (CW) echoes and the echo of the anterior wall of the right ventricle. At surgery, 60 ml of pericardial fluid was removed. The anterior echo-free space was produced as a result of a large mass (5 cm in anteroposterior diameter) Of POOrlY differentiated lymphoma.
ECHOCARDlOGFiAPHY
IN PERICARDIAL
EFFUSION-TAJIK
Vignola et al. [24] noted a distinct effect of heart rate on the type of mitral valve prolapse. They observed prolapse in early or late systole if the heart rate exceeded 120/min and a pansystolic-type prolapse when the heart rate was less than 120/min. That this mitral valve prolapse is a pseudoprolapse phenomenon has been proved by the fact that, after pericardiocentesis or pericardiectomy, the mitral valve motion returns to normal. We have similarly observed a pseudoprolapse pattern of the tricuspid valve in three patients who had malignant effusions. Occasionally, anterior motion of the mitral valve in systole may be recorded. Similarly, abnormal motion of the aortic valve and of the pulmonic valve has been noted in these patients [20,21].
--
.--
,P_
__--
-w-.
It needs to be emphasized that abnormalities of the septal and valvular motion can frequently be recorded in patients with large pericardial effusions and the swinging heart syndrome, and it is important to keep in mind that these abnormalities are “pseudo” in nature; awareness of this phenomenon should prevent incorrect diagnosis. Factors that determine the swinging of the heart are not entirely known. The size of the effusion, which clearly plays an important role, is not, however, the sole factor, because we have observed instances of equally large pericardial effusions without the swinging heart syndrome. When swinging becomes excessive so that the heart does not return to its original position before the next
_
__-
.__-_--_
This echogram, from a patient with a large malignant pericardial effusion, demonstrates the swinging of the heart. Figure 9. A, scan from the body of the left ventricle to the base of the heart is demonstrated Large anterior and posterior echo-free spaces can be well identified, the anterior space being larger than the posterior space.. Arrows point to the echographic appearance of the mitral valve prolapse on alternate beats. Toward the base of the heart, the posterior echo-free space is seen to be continuous also with the large space behind the left atria/ waif. 6, close-up view taken at the level of the left ventricle. Arrows again point to the mitral valve prolapse on alternate beats. No electrical aiternans can be appreciated on the electrocardiogram recorded here. C, this view demonstrates that every alternate aortic-valve opening excursion is smaller and the opening duration (ejection time) is shorter. See text for derails.
36
July 1977
The American Journal of Medicine
Volume 63
ECHOCARDlOGF?APHY IN PERICARDIAL EFFUSION-TAJIK
electrical depolarization, the phenomenon of electrical alternans takes place [3,25,26]. Figure 9 is from a patient with malignant effusion causing tamponade and with electrical and pulsus alternans. In Figure 9A a large anterior and posterior pericardial effusion can be seen. Note also the features of free swinging of the heart. At the time of the first electrical depolarization, the anterior wall of the right ventricle is in its most anterior location closest to the chest wall. Following this, the anterior wall moves posteriorly and does not return to its original anterior position before the next electrical depolarization. Thus, every second electrical depolarization is smaller because the heart is situated more posteriorly in the fluid-filled pericardial sac. This accounts for the phenomenon of total electrical alternans as seen in patients with large pericardial effusion. Also note that every second mitral valve echo demonstrates a pseudoprolapse pattern in systole. Figure 96 shows the aortic root and aortic valve motion, and posteriorly the left atrium is recorded. Note that every second aortic valve opening is of smaller amplitude and of shorter duration, consistent with the presence of pulsus alternans. Also note the large echo-free space posterior to the left atrium and the alternating large and small excursion of the left atrial wall. PERICARDIAL
FLUID BEHIND THE LEFT ATRIUM
Until recently it was thought that fluid behind the left atrium theoretically could not be pericardial because at that level the pericardium is tightly bound to the left atrial wall by its reflection onto the pulmonary veins. For this reason, fluid noted behind the left atrium on echocardiography had been regarded as representing pleural effusion [ 151. In January 1974 we first noted an exception to this rule. In a patient with a large pericardial effusion with no pleural effusion, M-mode scan of the left ventricle revealed the presence of an echo-free space (fluid) behind the left atrial posterior wall. We have subsequently noted similar findings in 11 additional patients, and our observations in the initial five patients formed the basis of a previous report [ 201. In all these patients, pericardial effusion was large and in most the heart was swinging. In all, the echo-free space behind the left ventricle could be demonstrated to be continuous with the echo-free space behind the left atrium (Figures 10 and 11). In these instances an abnormal motion of the left atrial wall, consisting of anterior motion in systole occurring in unison with the anterior swing of the entire heart, was also noted. In addition, the left atrial wall also had a large amplitude of excursion. These observations documented that fluid behind the left atrium is not necessarily pleural effusion but could as well be pericardial in patients with large pericardial effusion and the swinging heart syndrome. We
believe that the differentiation in such instances can readily be made, based on the motion characteristics of the left atrial posterior wall. The left atrial wall was hyperdynamic and had a large anterior excursion persisting into the early ventricular systole in patients who had pericardial effusion behind the left atrium, whereas the left atrial wall motion remained normal in patients with pleural effusion behind the left atrium (Figures 10 and 11). In some patients (three of 12) the pericardial fluid behind the left atrium could be made to disappear with slightly different angulation of the transducer, a finding that indicated selective localization. Occasionally, a retrocardiac pleural effusion may simulate pericardial effusion (false-positive study). In such cases, in addition to the criteria noted, recording the echogram in the high left lateral decubitus position (in order to drain the fluid from behind the heart) or after thoracentesis may permit differentiation from pericardial effusion [ 151. DIAGNOSIS OF CARDIAC TAMPONADE ECHOCARDIOGRAPHY
BY
Cardiac tamponade is a clinical syndrome characterized by hypotension, venous distention, tachycardia, pulsus paradoxus and a prompt, favorable hemodynamic response to pericardiocentesis. Although echocardiography in such patients allowed easy detection of a large anterior and posterior pericardial effusion, until recently no echographic criteria for the diagnosis of cardiac tamponade were available. Feigenbaum et al. [2,4], in their earlier publications, noted that the motion of the posterior wall was diminished in patients with cardiac tamponade. This finding subsequently was noted, however, in only two of six patients with such a diagnosis [3]. The next recognized echocardiographic criterion in cardiac tamponade was a swinging heart with electrical and pulsus alternans [25,26] (Figure 9). Recently, D’Cruz et al. [27] and Vignola et al. [24] reported preliminary observations suggesting that changes in mitral valve motion and in chamber dimensions may be indicative of tamponade. Decreased excursion of the mitral valve with diminished E-F slope was noted by both groups. D’Cruz et al. also noted pronounced phasic variation in the dimensions of the right and left ventricles. During inspiration the dimension of the right ventricle increased whereas that of the left ventricle decreased. This variation in chamber dimension with respiration accounts for the pulsus paradoxus that is commonly seen in these patients. This is shown in Figure 12. Vignola et al., in addition, described a notch on the epicardial surface of the right ventricle, which occurred during the isometric contraction phase of the cardiac cycle, and also noted coarse oscillations of the left ventricular posterior wall. Hence, in a patient with
July 1977
The American Journal of Medicine
Volume 63
37
Figure 10. M-mode scan from the base of the heart toward the apex of the left ventricle from a patient with a large pericardial effusion. At the base of the heart behind the left atrial wall, a moderate-size echo-free space is seen which is continuous with the echo-free space seen behind the left ventricular posterior wall. The motion of the right ventricular anterior wail, left ventricular posterior wall and mitral valve is normal. The left atrial wall, however, is hyperdynamic. From Lemire F, Tajik AJ, Giuliani ER, et al. [20].
Figure 11. This echogram, from a patient with postradiation pericarditis, demonstrates large anterior and posterior echo-free spaces. The posterior echo-free space is continuous also behind the left atrial posterior wall. In addition, note the late systolic mitral valve prolapse pattern (large arrows). An abnormal closure of the aortic valve in systole can also be seen. These abnormalities of the mitral valve and the aortic valve motion returned to normal after pericardiectomy. An electrocardiogram is buried in the crystal artifact, and the tops of the ORS complexes are marked by the sma// arrows on the top of the figure.
Figure 12. This echogram, from a patient with a large malignant pericardial effusion, demonstrates pronounced variation in cardiac dimensions with respiration. During inspiration (IN) the dimension of the right ventricle appears to increase whereas that of the left ventricle decreases. The opposite changes occur during expiration (EXP). A large posterior and a moderate anterior pericardial effusion are present.
ECHOCARDIOGRAPHY
large pericardial effusion, these changes in mitral valve motion, notching of the right ventricular anterior wall or coarse oscillations of the left ventricular posterior wall, may be indicative of cardiac tamponade. However, it should be remembered that these findings are preliminary and will need substantiation. CONCLUSIONS Although careful bedside examination of patients’ symptoms, pulse, jugular veins and precordium, combined with a roentgenogram of the chest and an electrocardiogram, will permit the correct diagnosis of pericardial effusion to be made in most cases, the clinical distinction from an enlarged heart may not always be possible at the bedside. In this regard, the advantages of echocardiography, a safe, noninvasive,
IN PERICARDIAL
EFFUSION-TAJIK
sensitive, accurate and easy to perform bedside technic, become clearly apparent. None of the other technics (radioisotopes, angiography, carbon dioxide injection) used for confirmation of pericardial effusion can be performed at the bedside. Echocardiography can easily provide the answer as to the cause of an enlarging cardiac silhouette. This is of great practical importance in clinical practice and, especially, in the management of patients with valvular heart disease. Moreover, echocardiography not only provides the diagnosis of pericardial effusion but also offers good insight into underlying or associated cardiac disorders. This technic also provides an opportunity to determine the true incidence of pericardial involvement in various systemic diseases, for example, rheumatoid arthritis (Figure 13) collagen vascular diseases, myxedema
This echogram is from a patient with juvenile rheumatoid arthritis. A, demonstrates the presence of a small posterior pericardial effusion. Note that pericardial echo (P) is f/at. 6, was repeated after 12 days of therapy, and now there is no longer any evidence of pericardial effusion, and the normal systolic anterior motion of the pericardial echo has been restored.
Figure 13.
July 1977
The American Journal of Medicine
Volume 63
39
ECHOCARDIOGRAPHYIN PERICARDIAL EFFUSION-TAJIK
[28] and others. With a technically good, properly interpreted echocardiogram, the incidence of true false-positive and true false-negative studies is extremely low, and so echocardiography has become the method of choice for the detection, confirmation and serial follow-up of patients with pericardial effusion.
ACKNOWLEDGMENT
I am grateful to Drs. James B. Seward and Emilio R. Giuliani for their review and helpful criticism, and to Joan L. Hargrove and Nancy K. Buzza for their technical assistance.
REFERENCES Edler I: Diagnostic use of ultrasound in heart disease. Acta Med Stand 152 (suppl308): 32, 1955. 2. Feigenbaum H, Waldhausen JA, Hyde LP: Ultrasound diagnosis of pericardial effusion. JAMA 191: 711, 1965. 3. Feigenbaum H, Zaky A, Grabhorn LL: Cardiac motion in patients with pericardial effusion: a study using reflected ultrasound,Circulation 34: 611, 1966. 4. Feigenbaum H, Zaky A, Waldhausen JA: Use of ultrasound in the diagnosis of pericardial effusion. Ann Intern Med 65: 443, 1966. 5. Feigenbaum H, Zaky A, Waldhausen JA: Use of reflected ultrasound in detecting pericardial effusion. Am J Cardiol 19: 84, 1967. 6. Pate JW, Gardner HG, Norman RS: Diagnosis of pericardial effusion by echocardiography. Ann Surg 165: 826, 1967. 7. Moss AJ, Bruhn F: The echocardiogram: an ultrasound technic for the detection of pericardial effusion. N Engl J Med 274: 380, 1966. a. Soulen RL, Lapayowker MS, Gimenez JL: Echocardiography in the diagnosis of pericardial effusion. Radiology 86: 1047, 1966. 9. Rothman J, Chase NE, Kricheff II, et al.: Ultrasonic diagnosis of pericardial effusion. Circulation 35: 358, 1967. 10. Goldberg BB, Ostrum BJ, lsard HJ: Ultrasonic determination of pericardial effusion. JAMA 202: 103, 1967. 11. Klein JJ, Segal BL: Pericardial effusion diagnosed by reflected ultrasound. Am J Cardiol 22: 57, 1968. 12. Feigenbaum H: Ultrasonic cardiology: diagnostic ultrasound as an aid to the management of patients with pericardial effusion. Dis Chest 55: 59, 1969. 13. Goldschlager AW, Freeman LM, Davis PJ: Pericardial effusions and echocardiography: false results with ultrasound reflection method. NY State J Med 7: 1854, 1967. 14. Klein JJ, Raber G, Shimada H, et al.: Evaluation of induced pericardial effusion by reflected ultrasound. Am J Cardiol 22: 49, 1968. 15. Feigenbaum H: Pericardial effusion. Echocardiography, Philadelphia, Lea & Febiger, 1972, p 163.
40
July 1977
The American
Journal of Medicine
Volume 63
16.
17.
ia.
19.
20.
21.
22.
23.
24.
25. 26. 27.
28.
Abbasi AS, Ellis N, Flynn JJ: Echocardiographic M-scan technique in the diagnosis of pericardial effusion. J Clin Ultrasound 1: 300, 1973. Tingelstad JB, McWilliams NB, Thomas CE: Confirmation of a retrosternal mass by echocardiogram. J Clin Ultrasound 4: 129. 1976. Child JS, Abbasi AS, Pearce ML: Echocardiographic differentiation of mediastinal tumors from primary cardiac disease. Chest 67: 108, 1975. Horowitz MS, Schultz CS, Stinson EB, et al.: Sensitivity and specificity of echocardiographic diagnosis of pericardial effusion. Circulation 50: 239, 1974. Lemire F, Tajik AJ, Giuliani ER, et al.: Further echocardiographic observations in pericardial effusion. Mayo Clin Proc 51: 13, 1976. Nanda NC, Gramiak R, Gross CM: Altered systolic motion of cardiac valves in pericardial effusion: echocardiographic studies (abstract). Circulation 52 (suppl 2): 134, 1975. Levisman JA, Abbasi AS: Abnormal motion of the mitral valve with pericardial effusion: pseudo-prolapse of the mitral valve. Am Heart J 91: 18, 1976. Owens JS, Kotler MN, Segal BL, et al.: Pseudoprolapse of the mitral valve in a patient with pericardial effusion. Chest 69: 214, 1976. Vignola PA, Pohost GM, Curfman GE, et al.: Correlation of echocardiographic and clinical findings in patients with pericardial effusion. Am J Cardiol 37: 701, 1976. Gabor GE, Winsberg F, Bloom HS: Electrical and mechanical alteration in pericardial effusion. Chest 59: 34 1, 197 1. Usher BW, Popp RL: Electrical alternans: mechanism in pericardial effusion. Am Heart J 83: 459, 1972. D’Cruz IA, Cohen HC, Prabhu R, et al.: Diagnosis of cardiac tamponade by echocardiography. Changes in mitral valve motion and ventricular dimensions, with special reference to paradoxical pulse. Circulation 52: 460, 1975. Kerber RE, Sherman B: Echocardiographic evaluation of pericardial effusion in myxedema. Incidence and biochemical and clinical correlations. Circulation 52: 823, 1975.
ABDUL J. TAJIK, M.D. Rochester, Minnesota
Echocardiography has become established as the procedure of choice for the detection, confirmation and serial follow-up of patients with pericardiai effusion. in this article the technic and pitfalls of recording, and the criteria and their sensitivity for the diagnosis of pericardiai effusion are reviewed. in addition, echographic findings in special instances, such as accumulation of pericardiai effusion behind the left artium, the swinging heart syndrome and cardiac tamponade, are discussed. Although Edler [l] mentioned separation of the echo of the anterior heart wall from that of the chest wall in patients with pericardiai effusion, it was not until Feigenbaum and colleagues [2-51 published a series of papers in the mid 1960’s that the usefulness of echocardiography in pericardial effusion was established; other papers confirmed their findings [6- 141. The subsequent popularity of echocardiography in this country is, to a great extent, directly attributable to these findings. Today, echocardiography is considered to be the procedure of choice for evaluating patients suspected of having pericardial effusion. This high place afforded to echocardiography is due to its being an accurate, easy to perform and noninvasive method. It must be remembered, however, that the results obtained are entirely dependent on the knowledge, experience and technical skills of the examiner. The echocardiographer must be thoroughly familiar with the ultrasonic cardiac anatomy and various intracardiac landmarks. Many previously published false-positive and false-negative echograms in pericardial effusion can be directly attributed to faulty technic. ANATOMY AND TECHNIC OF RECORDING
From the Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Requests for reprints should be addressed to Dr. Abdul J. Tajik. Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55901.
The normal pericardium consists of two layers: the outer, fibrous pericardium and the inner, serous pericardium. The fibrous pericardium is a cone-shaped bag that has attachments anteriorly to the sternum, inferiorly to the diaphragm, posteriorly to the pulmonary veins and superiorly to the great vessels. The serous pericardium consists of two layers, the inner of which lines the myocardium and is called the visceral pericardium or epicardium. Visceral pericardium extends a short distance over the great vessels and then is reflected on itself to form the parietal layer, which lines the fibrous pericardium. The space between these two layers is the pericardial cavity, which normaliy contains an average of about 20 ml of fluid. Because posteriorly the pericardium is reflected from the back of the left atrium onto the pulmonary veins, no potential space exists for the accumulation of
July 1977
The American Journal of Medicine
Volume 63
29
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSION-TAJIK
Figure 1. Technic of recording echograms at different gain settings. Both echograms were recorded at the level of the choroae tendineae (C). At lower recording intensity (left part of figures), only the strong echo originating from the pericardial lung interface is recorded(P). As the intensity of the echograph is gradually increased, next the epicardial (EP) and lastly the weakest endocardial (EN) echo are recorded A, all three echoes (periwdial, epicardial and endocardial) are readily seen. No clear anterior echo-free space can be recognized in this figure. 6, epicardial and pericardial echoes are in close opposition, and therefore no clear separation can be seen between these two echoes during the cardiac cycle. With a higher gain setting, the endocardial and myocardial echoes can be well recorded. Note that an anterior echo-free space is readily seen in this figure. During ventricular systole, the anterior right ventricular wall echo separates from the chest wall echo. This is not an uncommon finding in normal subjects. RV = right ventricle; VS = ventricular septum; MV = mitral valve; L V = left ventricle.
pericardial fluid behind the left atrium at the level of the pulmonary veins. However, it must be remembered that there also exists posteriorly (directly behind the left atrium) a blind recess of pericardial cavity which is bordered by the pericardial reflection between the pulmonary veins. This recess is called the oblique sinus of the pericardium. It is in this region that, under special circumstances, pericardial fluid can and does accumulate. One cannot overemphasize the importance of proper recording technics when looking for pericardial effusion. Clear visualization of the anterior right ventricular wall and of the left ventricular posterior wall is necessary and
30
July 1977
The American Journal of Medicine
Volume 63
can be obtained by careful adjustments of reject, gain, intensity and damping controls. With the patient in the recumbent position and the trunk elevated 20 to 30 degrees, the examination is begun by first recording the familiar mitral valve echo. From this recording position, the transducer is tilted inferolaterally to record echoes of the left ventricular posterior wall at the chordae tendineae level (Figure 1). Special care must be taken that all the components (endocardium, epicardium and pericardium) of the left ventricular posterior wall are clearly recorded. It is a good practice to record the echogram at different gain settings. While the left ventricular posterior wall is recorded, the intensity of
ECHOCARDIOGRAPHY
IN PERICARDIAL
EFFUSION-TAJIK
RV
Figure 2. Echograms of two patients in congestive heart failure who had smallpericardial effusions. A, patient with a large atria/ septal defect. Echogram demonstrates the features of right ventricular volume overload (increased right ventricular dimension and abnormal ventricular septal motion). As the recording intensity of the echograph machine is reduced (middle of figure), the different components of the left ventricular posterior wall can be easily distinguished. The pericardial echo (P) demonstrates only minimal anterior motion in systole. An echo-free space representing a small amount of pericardial fluid is well seen as the epicardial (EP) echo separates from the pericardial echo during systole. 6, patient with severe mitral regurgitation. Echogram demonstrates the increased dimension of the left ventricle and hyperdynamic motion of both the septum and the left ventricular posterior wall. When the gain is reduced (middle of figure), o&y the strong pericardial and epicardial echoes are recorded, and a clear echo-free space between these two echoes becomes evident. It represents a small amount of pericardial effusion. No anterior echo-free space is recorded. In both of these echograms, note that the pericardial effusion could only be delineated when the gain and the intensity of recording were reduced. At higher gain settings (left and right half of figures), the echo-free space becomes obliterated by echoes originating from the pericardial fluid. RV = right ventricle; C = chordae tendineae; L V = left ventricle; VS = ventricular septum; PF = pericardial fluid; EN = endocardium.
July 1977
The American Journal of Medicine
Volume 63
31
ECHOCARDIOGRAPHY
IN PERICARDIAL
EFFUSION-TAJIK
the echograph should be reduced to obtain optimal definition of the strong posterior pericardial echo, which normally moves anteriorly during systole (Figure 1). The pericardial echo having been defined, the intensity should be gradually increased to identify, next in sequence, the epicardial and then the endocardial echoes. With proper gain settings, the relatively echo-free space between the endocardium and the epicardium can be filled in by echoes representing myocardium (Figure 1B). In normal subjects, the epicardial echo of the left ventricular posterior wall is in close contact with the parietal pericardium and pleura at the pericardial lung interface behind the heart. Similarly, the echoes from the anterior wall of the right ventricle are in close contact with the nonmoving chest wall echoes. Occasionally, in normal subjects, a small echo-free space may exist between the chest wall echoes and the anterior wall of the right ventricle; this is believed to be due to the fact that epicardial fat or loose areolar tissue or a portion of the lung is interposed between the chest wall and the anterior right ventricular wall (Figure 1B). ECHOGRAPHIC CRITERIA FOR DIAGNOSIS OF PERICARDIAL EFFUSION The following changes appear in the echocardiogram with the development of pericardial effusion. Demonstration of an echo-free space between the right ventricular anterior wall and the chest wall, and/or between the left ventricular posterior wall and the pericardial lung interface, indicates the presence of pericardial effusion. It has now been well established that the pericardial fluid first appears posteriorly in the dependent portion of the pericardial cavity. Therefore, the earliest manifestation of the presence of a tiny pericardial effusion is a slight separation, through systole and part of diastole, of the left ventricular posterior wall epicardial echo from a relatively stationary pericardial echo (Figure 2A). With increasing accumulation of fluid, this posterior echofree space becomes evident throughout the entire cardiac cycle (Figure 2B). In the absence of an anterior echo-free space, a small posterior echo-free space should suggest the presence of a small amount of pericardial effusion. With increasing amounts of pericardial fluid, the pericardial echo becomes stationary because the pericardial fluid, which acts as an insulator, attenuates the transmission of the left ventricular posterior wall motion to the parietal pericardium. Small amounts of pericardial effusion are not uncommon in patients who are in congestive heart failure. In both of the examples shown in Figure 2, a small amount of pericardial fluid is present as indicated by the small echo-free space posteriorly. No anterior echo-free space could be demonstrated. With further increase in the pericardial effusion, the posterior space widens; with larger effusions, the fluid
32
July 1977
The American Journal of Medicfne
Volume 63
also appears anteriorly as an anterior echo-free space (Figure 3). In the presence of anterior effusion, the amplitude of motion of the anterior wall of the right ventricle becomes exaggerated. A large pericardial effusion is diagnosed when there is a further increase in the depth of the anterior and posterior echo-free spaces. Frequently, but not invariably, when a large anterior and posterior effusion is present, the heart may be observed to be moving freely in the pericardial sac (swinging heart). In the current-day practice of echocardiography, it is imperative that complete scans of the left ventricle be recorded in each patient [ 15,161. A typical M-mode scan of the left ventricle in a patient with moderate pericardial effusion is shown in Figure 4. As can be seen on the left of the echogram at the level of the chordae tendineae, pericardial fluid can be recorded both anteriorly and posteriorly behind the left ventricular posterior wall. As the beam is tilted superiorly and medially toward the base of the heart (aortic valve recording position with the left atrium posteriorly), the pericardial fluid is seen to disappear at the atrioventricular junction, and no fluid is recorded behind the left atrium. This feature has been a consistent and reproducible finding in pericardial effusion. Figure 5 is from another patient with a moderatesized pericardial effusion, again demonstrating that, as the sound beam is tilted toward the base of the heart, the posterior echo-free space is markedly diminished in size but does not completely disappear behind the left atrium. The possibility of echographicdemonstration of pericardial fluid posterior to the left atrium has only recently been appreciated. Figure 6 demonstrates a base to apex scan from a patient with a large pericardial effusion. This echogram was obtained with the patient’s trunk elevated approximately 30 degrees. On the left of the echogram, with the transducer tilted toward the base of the heart, an anterior echo-free space is recorded; in addition, a small echo-free space is evident behind the left atrium. In the middle of the figure, at the level of the left ventricle, moderate-size anterior and posterior echo-free spaces are recorded, and on the right of the figure with the transducer tilted toward the apex of the heart, both the anterior and the posterior echo-free spaces are seen to enlarge; these findings indicate that the bulk of this pericardial fluid was surrounding the apex of the heart. Occasionally, in patients with large pericardial effusion, and specifically in those with the swinging heart syndrome, the anterior echo-free space may become more striking than the posterior echo-free space. In such situations, difficulty may be encountered in recording posterior-lying structures because the anterior right ventricular wall becomes a very strong reflector of ultrasound and diminishes the energy
ECHOCARDIOGRAPHY
available for deeper penetration. We have also seen patients with large posterior pericardial effusions and without any significant anterior effusion. Figure 7 shows the echogram of a patient with idiopathic pericardial effusion. Note that a large posterior pericardial effusion is present, and there is little if any anterior effusion. This echogram was also obtained with the patient’s trunk elevated to 30 to 40 degrees; this ordinarily should have accentuated the anterior effusion. On the other hand, the echogram in Figure 8 demonstrates a large anterior echo-free space without any significant posterior echo-free space. Such an echogram rarely results from pericardial effusion and, unless proved otherwise, should be regarded as negative for pericardial effusion. In most instances this anterior echo-free space is a result of large amounts of epicardial fat or retrosternal extracardiac mass lesions, such as thymoma or lymphoma [ 17,181. Although such an echogram may conceivably be found in a patient with loculated anterior pericardial effusion, we have to date not seen such an example and, therefore, are reluctant to diagnose pericardial effusion on the basis of a large anterior echo-free space alone. SENSITIVITY OF METHOD AND QUANTIFICATION OF EFFUSION
The early experimental studies in dogs had indicated that a minimum infusion of 50 ml of fluid in the pericardial sac was necessary before echocardiography could reliably demonstrate an effusion [2]. Although exact data were lacking for human subjects, it was generally believed, until recently, that echocardiography could reliably detect 50 to 100 ml of pericardial fluid. In this regard, the recent publication of Horowitz et al. [ 191 has been of great importance. In their attempts at quantifying and characterizing various patterns of pericardial effusions, they established the lower limit of sensitivity of a technically optimal, properly interpreted echocardiogram as 15 to 20 ml. Using a spheroid model of the heart, they also proposed a method of quantitating pericardial effusion, but they were quick to acknowledge that such a method was not extremely accurate. In our practice, we do not attempt accurate quantitation of pericardial effusion but are satisfied in using echocardiography as a technic for the detection and gross semiquantitative (minimal, small, moderate, large) estimation of the fluid. PITFALLS OF RECORDING TECHNIC As previously emphasized, a proper echographic technic is essential for arriving at the correct diagnosis of pericardial effusion. Improper use of echographic controls (reject, depth compensation, intensity and gain settings) accounts for most of the inadequate and confusing results. In particular, excessively low gain
IN PERICARDIAL
EFFUSION-TAJIK
settings result in incomplete identification of various layers of the left ventricular posterior wall; too high gain settings can result in excessive echoes from the pericardial fluid which will obscure the recognition of an echo-free space (Figure 2). In this regard, if the technic of recording echograms at various gain settings is religiously followed, the incidence of false-positive results and, more specifically, false-negative results can virtually be eliminated. Another pitfall in the recording technic pertains to the direction of the transducer beam. In the literature it has been emphasized that when the transducer is tilted too far medially, an echo-free space behind the posterior wall can be recorded on account of structures such as the coronary sinus, pulmonary veins, descending aorta and other posterior mediastinal structures. Occasionally, with the transducer directed medially, the mitral annulus can be mistaken for the left ventricular posterior wall, and rarely the medial aspect of the left ventricular posterior wall may be recorded. In both of these instances, a false-positive diagnosis of pericardial effusion may be made. This was particularly true in the early days of echocardiography when Polaroids were in use, and the scans of the left ventricle were infrequently recorded. In current practice, with the use of continuous M-mode scans of the left ventricle, confusion of pericardial effusion with the aforementioned structures has become most unlikely. SWINGING HEART SYNDROME
With large anterior and posterior pericardial effusions, the heart may move freely within the pericardial sac. This type of cardiac motion abnormality (swinging heart) was initially described by Feigenbaum and associates [3] in 1966. The free motion of the heart is attributed to the absence of the restraining function of the pericardium. The heart, suspended by the great vessels, swings freely within the pericardial sac, causing both the anterior and the posterior walls of the heart to move synchronously, that is, anteriorly during systole and posteriorly during diastole or vice versa. Generally, but not invariably, greater swing occurs with larger effusions and especially those of malignant origin. As a consequence of the free swinging of the heart, abnormal motion of various structures, such as the septum, mitral and tricuspid valves, and aortic and pulmonary valves, has been noted [ 20-231. As the heart swings anteriorly during systcle, both the anterior and the posterior walls of the heart, as well as the ventricular septum, also move anteriorly, and therefore an abnormal ventricular septal motion is not infrequently recorded in this situation. Various types of mitral valve prolapse patterns have also been described in patients with swinging heart syndrome. The most common pattern is the typical late-systolic mitral valve prolapse.
July 1977
The American Journal of Medicine
Volume 63
33
Figure 3. Patient with moderate pericardial effusion (idiopathic). Echogram demonstrates anterior (smaller) and posterior (larger) echo-free spaces. The motibn of the anterior wall of the right ventricle, ventricular septum and left ventricular posterior wail is normal; that is, during ventricular systofe the right ventricular anterior wail and ventricular septum move posteriorly as the let? ventricular posterior wall moves anteriorly. PF = pericardial fluid; C = chordae tendineae; VS = ventricular septum; EN = endocardium; EP = epicardium; P = pericardium.
Figure 4. Conti&ous M-mode scan from the body of the left ventricle to the base of the heart from a patient with moderate pericardial effusion. At the ventridular level, moderate anterior and posterior echo-free spaces can be seen, As the beam is tilted toward the aortic root with the left atrium behind, the echo-free space disappears behind the left atrium. The motion of the right ventricular anterior wall, ventricular septum and left ventricular posterior wall is normal. PF = pericardial fluid; RV = right ventricle; LV = ieft ventricle; PW = posterior wall of the left ventricle; MV = mitral valve; A V = aortic valve; LA = left atrium. From Lemire F, Tajik AJ, Giuliani ER, et al. [ZO].
Figure 5. M-mode scan from the base of the heart (aortic root and left atrium) toward the body of the left ventiicte at the level of the chordae tendineae. This echogram is from the same patient as in Figure 3. On the left of the echogram is a suggestion of a tiny echo-free space behind the left atrial wall which appears to be continuous with the larger echo-free space behind the left ventricular posterior wall. PF = pericardial fluid; VS = ventricular septum; MV = mitral valve; EN = endocardium; EP = epicardium; P = pericardium; A,V = aortic valve; LA = left atrium.
Figure 6. Condensed (slow paper speed) M-mode base (left) to apex (right) scan of the heart from a patient with a large pericardial effusion. This patient had systemic amyloidosis with pericardial as well as myocardial involvement. Echogram was performed with the patient’s trunk elevated about 30 degrees. At the level of the left ventricle (middle of figure), moderate-size anterior and posterior echo-free spaces can be seen. As the echo beam is angulated toward the apex of the heart, both the anterior and the posterior echo-free spaces widen. This demonstrates the presence of a large pericardial effusion surrounding the apex of the heart. Also note the increased thickness of the septum and left ventricular posterior wall secondary to amyloid infiltration. Abbreviations are the same as for the other figures.
Figure 7. effusion
Echogram from a patient with a large pericardial effusion anteriorly. The motion of the cardiac wails remains normal.
that is almost entirely located posteriorly
with only little
Figure 6. This echogram, from a patient with large retrosternal lymphoma involving the pericardium, demonstrates an M-mode scan from the base of the heart to the body of the left ventricle. At the level of the left ventricle, a small posterior echo-free space can be seen which represents pericardial effusion. However, note that there is a large anterior echo-free space between the chest wall (CW) echoes and the echo of the anterior wall of the right ventricle. At surgery, 60 ml of pericardial fluid was removed. The anterior echo-free space was produced as a result of a large mass (5 cm in anteroposterior diameter) Of POOrlY differentiated lymphoma.
ECHOCARDlOGFiAPHY
IN PERICARDIAL
EFFUSION-TAJIK
Vignola et al. [24] noted a distinct effect of heart rate on the type of mitral valve prolapse. They observed prolapse in early or late systole if the heart rate exceeded 120/min and a pansystolic-type prolapse when the heart rate was less than 120/min. That this mitral valve prolapse is a pseudoprolapse phenomenon has been proved by the fact that, after pericardiocentesis or pericardiectomy, the mitral valve motion returns to normal. We have similarly observed a pseudoprolapse pattern of the tricuspid valve in three patients who had malignant effusions. Occasionally, anterior motion of the mitral valve in systole may be recorded. Similarly, abnormal motion of the aortic valve and of the pulmonic valve has been noted in these patients [20,21].
--
.--
,P_
__--
-w-.
It needs to be emphasized that abnormalities of the septal and valvular motion can frequently be recorded in patients with large pericardial effusions and the swinging heart syndrome, and it is important to keep in mind that these abnormalities are “pseudo” in nature; awareness of this phenomenon should prevent incorrect diagnosis. Factors that determine the swinging of the heart are not entirely known. The size of the effusion, which clearly plays an important role, is not, however, the sole factor, because we have observed instances of equally large pericardial effusions without the swinging heart syndrome. When swinging becomes excessive so that the heart does not return to its original position before the next
_
__-
.__-_--_
This echogram, from a patient with a large malignant pericardial effusion, demonstrates the swinging of the heart. Figure 9. A, scan from the body of the left ventricle to the base of the heart is demonstrated Large anterior and posterior echo-free spaces can be well identified, the anterior space being larger than the posterior space.. Arrows point to the echographic appearance of the mitral valve prolapse on alternate beats. Toward the base of the heart, the posterior echo-free space is seen to be continuous also with the large space behind the left atria/ waif. 6, close-up view taken at the level of the left ventricle. Arrows again point to the mitral valve prolapse on alternate beats. No electrical aiternans can be appreciated on the electrocardiogram recorded here. C, this view demonstrates that every alternate aortic-valve opening excursion is smaller and the opening duration (ejection time) is shorter. See text for derails.
36
July 1977
The American Journal of Medicine
Volume 63
ECHOCARDlOGF?APHY IN PERICARDIAL EFFUSION-TAJIK
electrical depolarization, the phenomenon of electrical alternans takes place [3,25,26]. Figure 9 is from a patient with malignant effusion causing tamponade and with electrical and pulsus alternans. In Figure 9A a large anterior and posterior pericardial effusion can be seen. Note also the features of free swinging of the heart. At the time of the first electrical depolarization, the anterior wall of the right ventricle is in its most anterior location closest to the chest wall. Following this, the anterior wall moves posteriorly and does not return to its original anterior position before the next electrical depolarization. Thus, every second electrical depolarization is smaller because the heart is situated more posteriorly in the fluid-filled pericardial sac. This accounts for the phenomenon of total electrical alternans as seen in patients with large pericardial effusion. Also note that every second mitral valve echo demonstrates a pseudoprolapse pattern in systole. Figure 96 shows the aortic root and aortic valve motion, and posteriorly the left atrium is recorded. Note that every second aortic valve opening is of smaller amplitude and of shorter duration, consistent with the presence of pulsus alternans. Also note the large echo-free space posterior to the left atrium and the alternating large and small excursion of the left atrial wall. PERICARDIAL
FLUID BEHIND THE LEFT ATRIUM
Until recently it was thought that fluid behind the left atrium theoretically could not be pericardial because at that level the pericardium is tightly bound to the left atrial wall by its reflection onto the pulmonary veins. For this reason, fluid noted behind the left atrium on echocardiography had been regarded as representing pleural effusion [ 151. In January 1974 we first noted an exception to this rule. In a patient with a large pericardial effusion with no pleural effusion, M-mode scan of the left ventricle revealed the presence of an echo-free space (fluid) behind the left atrial posterior wall. We have subsequently noted similar findings in 11 additional patients, and our observations in the initial five patients formed the basis of a previous report [ 201. In all these patients, pericardial effusion was large and in most the heart was swinging. In all, the echo-free space behind the left ventricle could be demonstrated to be continuous with the echo-free space behind the left atrium (Figures 10 and 11). In these instances an abnormal motion of the left atrial wall, consisting of anterior motion in systole occurring in unison with the anterior swing of the entire heart, was also noted. In addition, the left atrial wall also had a large amplitude of excursion. These observations documented that fluid behind the left atrium is not necessarily pleural effusion but could as well be pericardial in patients with large pericardial effusion and the swinging heart syndrome. We
believe that the differentiation in such instances can readily be made, based on the motion characteristics of the left atrial posterior wall. The left atrial wall was hyperdynamic and had a large anterior excursion persisting into the early ventricular systole in patients who had pericardial effusion behind the left atrium, whereas the left atrial wall motion remained normal in patients with pleural effusion behind the left atrium (Figures 10 and 11). In some patients (three of 12) the pericardial fluid behind the left atrium could be made to disappear with slightly different angulation of the transducer, a finding that indicated selective localization. Occasionally, a retrocardiac pleural effusion may simulate pericardial effusion (false-positive study). In such cases, in addition to the criteria noted, recording the echogram in the high left lateral decubitus position (in order to drain the fluid from behind the heart) or after thoracentesis may permit differentiation from pericardial effusion [ 151. DIAGNOSIS OF CARDIAC TAMPONADE ECHOCARDIOGRAPHY
BY
Cardiac tamponade is a clinical syndrome characterized by hypotension, venous distention, tachycardia, pulsus paradoxus and a prompt, favorable hemodynamic response to pericardiocentesis. Although echocardiography in such patients allowed easy detection of a large anterior and posterior pericardial effusion, until recently no echographic criteria for the diagnosis of cardiac tamponade were available. Feigenbaum et al. [2,4], in their earlier publications, noted that the motion of the posterior wall was diminished in patients with cardiac tamponade. This finding subsequently was noted, however, in only two of six patients with such a diagnosis [3]. The next recognized echocardiographic criterion in cardiac tamponade was a swinging heart with electrical and pulsus alternans [25,26] (Figure 9). Recently, D’Cruz et al. [27] and Vignola et al. [24] reported preliminary observations suggesting that changes in mitral valve motion and in chamber dimensions may be indicative of tamponade. Decreased excursion of the mitral valve with diminished E-F slope was noted by both groups. D’Cruz et al. also noted pronounced phasic variation in the dimensions of the right and left ventricles. During inspiration the dimension of the right ventricle increased whereas that of the left ventricle decreased. This variation in chamber dimension with respiration accounts for the pulsus paradoxus that is commonly seen in these patients. This is shown in Figure 12. Vignola et al., in addition, described a notch on the epicardial surface of the right ventricle, which occurred during the isometric contraction phase of the cardiac cycle, and also noted coarse oscillations of the left ventricular posterior wall. Hence, in a patient with
July 1977
The American Journal of Medicine
Volume 63
37
Figure 10. M-mode scan from the base of the heart toward the apex of the left ventricle from a patient with a large pericardial effusion. At the base of the heart behind the left atrial wall, a moderate-size echo-free space is seen which is continuous with the echo-free space seen behind the left ventricular posterior wall. The motion of the right ventricular anterior wail, left ventricular posterior wall and mitral valve is normal. The left atrial wall, however, is hyperdynamic. From Lemire F, Tajik AJ, Giuliani ER, et al. [20].
Figure 11. This echogram, from a patient with postradiation pericarditis, demonstrates large anterior and posterior echo-free spaces. The posterior echo-free space is continuous also behind the left atrial posterior wall. In addition, note the late systolic mitral valve prolapse pattern (large arrows). An abnormal closure of the aortic valve in systole can also be seen. These abnormalities of the mitral valve and the aortic valve motion returned to normal after pericardiectomy. An electrocardiogram is buried in the crystal artifact, and the tops of the ORS complexes are marked by the sma// arrows on the top of the figure.
Figure 12. This echogram, from a patient with a large malignant pericardial effusion, demonstrates pronounced variation in cardiac dimensions with respiration. During inspiration (IN) the dimension of the right ventricle appears to increase whereas that of the left ventricle decreases. The opposite changes occur during expiration (EXP). A large posterior and a moderate anterior pericardial effusion are present.
ECHOCARDIOGRAPHY
large pericardial effusion, these changes in mitral valve motion, notching of the right ventricular anterior wall or coarse oscillations of the left ventricular posterior wall, may be indicative of cardiac tamponade. However, it should be remembered that these findings are preliminary and will need substantiation. CONCLUSIONS Although careful bedside examination of patients’ symptoms, pulse, jugular veins and precordium, combined with a roentgenogram of the chest and an electrocardiogram, will permit the correct diagnosis of pericardial effusion to be made in most cases, the clinical distinction from an enlarged heart may not always be possible at the bedside. In this regard, the advantages of echocardiography, a safe, noninvasive,
IN PERICARDIAL
EFFUSION-TAJIK
sensitive, accurate and easy to perform bedside technic, become clearly apparent. None of the other technics (radioisotopes, angiography, carbon dioxide injection) used for confirmation of pericardial effusion can be performed at the bedside. Echocardiography can easily provide the answer as to the cause of an enlarging cardiac silhouette. This is of great practical importance in clinical practice and, especially, in the management of patients with valvular heart disease. Moreover, echocardiography not only provides the diagnosis of pericardial effusion but also offers good insight into underlying or associated cardiac disorders. This technic also provides an opportunity to determine the true incidence of pericardial involvement in various systemic diseases, for example, rheumatoid arthritis (Figure 13) collagen vascular diseases, myxedema
This echogram is from a patient with juvenile rheumatoid arthritis. A, demonstrates the presence of a small posterior pericardial effusion. Note that pericardial echo (P) is f/at. 6, was repeated after 12 days of therapy, and now there is no longer any evidence of pericardial effusion, and the normal systolic anterior motion of the pericardial echo has been restored.
Figure 13.
July 1977
The American Journal of Medicine
Volume 63
39
ECHOCARDIOGRAPHYIN PERICARDIAL EFFUSION-TAJIK
[28] and others. With a technically good, properly interpreted echocardiogram, the incidence of true false-positive and true false-negative studies is extremely low, and so echocardiography has become the method of choice for the detection, confirmation and serial follow-up of patients with pericardial effusion.
ACKNOWLEDGMENT
I am grateful to Drs. James B. Seward and Emilio R. Giuliani for their review and helpful criticism, and to Joan L. Hargrove and Nancy K. Buzza for their technical assistance.
REFERENCES Edler I: Diagnostic use of ultrasound in heart disease. Acta Med Stand 152 (suppl308): 32, 1955. 2. Feigenbaum H, Waldhausen JA, Hyde LP: Ultrasound diagnosis of pericardial effusion. JAMA 191: 711, 1965. 3. Feigenbaum H, Zaky A, Grabhorn LL: Cardiac motion in patients with pericardial effusion: a study using reflected ultrasound,Circulation 34: 611, 1966. 4. Feigenbaum H, Zaky A, Waldhausen JA: Use of ultrasound in the diagnosis of pericardial effusion. Ann Intern Med 65: 443, 1966. 5. Feigenbaum H, Zaky A, Waldhausen JA: Use of reflected ultrasound in detecting pericardial effusion. Am J Cardiol 19: 84, 1967. 6. Pate JW, Gardner HG, Norman RS: Diagnosis of pericardial effusion by echocardiography. Ann Surg 165: 826, 1967. 7. Moss AJ, Bruhn F: The echocardiogram: an ultrasound technic for the detection of pericardial effusion. N Engl J Med 274: 380, 1966. a. Soulen RL, Lapayowker MS, Gimenez JL: Echocardiography in the diagnosis of pericardial effusion. Radiology 86: 1047, 1966. 9. Rothman J, Chase NE, Kricheff II, et al.: Ultrasonic diagnosis of pericardial effusion. Circulation 35: 358, 1967. 10. Goldberg BB, Ostrum BJ, lsard HJ: Ultrasonic determination of pericardial effusion. JAMA 202: 103, 1967. 11. Klein JJ, Segal BL: Pericardial effusion diagnosed by reflected ultrasound. Am J Cardiol 22: 57, 1968. 12. Feigenbaum H: Ultrasonic cardiology: diagnostic ultrasound as an aid to the management of patients with pericardial effusion. Dis Chest 55: 59, 1969. 13. Goldschlager AW, Freeman LM, Davis PJ: Pericardial effusions and echocardiography: false results with ultrasound reflection method. NY State J Med 7: 1854, 1967. 14. Klein JJ, Raber G, Shimada H, et al.: Evaluation of induced pericardial effusion by reflected ultrasound. Am J Cardiol 22: 49, 1968. 15. Feigenbaum H: Pericardial effusion. Echocardiography, Philadelphia, Lea & Febiger, 1972, p 163.
40
July 1977
The American
Journal of Medicine
Volume 63
16.
17.
ia.
19.
20.
21.
22.
23.
24.
25. 26. 27.
28.
Abbasi AS, Ellis N, Flynn JJ: Echocardiographic M-scan technique in the diagnosis of pericardial effusion. J Clin Ultrasound 1: 300, 1973. Tingelstad JB, McWilliams NB, Thomas CE: Confirmation of a retrosternal mass by echocardiogram. J Clin Ultrasound 4: 129. 1976. Child JS, Abbasi AS, Pearce ML: Echocardiographic differentiation of mediastinal tumors from primary cardiac disease. Chest 67: 108, 1975. Horowitz MS, Schultz CS, Stinson EB, et al.: Sensitivity and specificity of echocardiographic diagnosis of pericardial effusion. Circulation 50: 239, 1974. Lemire F, Tajik AJ, Giuliani ER, et al.: Further echocardiographic observations in pericardial effusion. Mayo Clin Proc 51: 13, 1976. Nanda NC, Gramiak R, Gross CM: Altered systolic motion of cardiac valves in pericardial effusion: echocardiographic studies (abstract). Circulation 52 (suppl 2): 134, 1975. Levisman JA, Abbasi AS: Abnormal motion of the mitral valve with pericardial effusion: pseudo-prolapse of the mitral valve. Am Heart J 91: 18, 1976. Owens JS, Kotler MN, Segal BL, et al.: Pseudoprolapse of the mitral valve in a patient with pericardial effusion. Chest 69: 214, 1976. Vignola PA, Pohost GM, Curfman GE, et al.: Correlation of echocardiographic and clinical findings in patients with pericardial effusion. Am J Cardiol 37: 701, 1976. Gabor GE, Winsberg F, Bloom HS: Electrical and mechanical alteration in pericardial effusion. Chest 59: 34 1, 197 1. Usher BW, Popp RL: Electrical alternans: mechanism in pericardial effusion. Am Heart J 83: 459, 1972. D’Cruz IA, Cohen HC, Prabhu R, et al.: Diagnosis of cardiac tamponade by echocardiography. Changes in mitral valve motion and ventricular dimensions, with special reference to paradoxical pulse. Circulation 52: 460, 1975. Kerber RE, Sherman B: Echocardiographic evaluation of pericardial effusion in myxedema. Incidence and biochemical and clinical correlations. Circulation 52: 823, 1975.
