- Email: [email protected]
Correlation of echocardiographic and clinical findings in patients with pericardial effusion
Correlation
of Echocardiographic
Patients With Pericardial
PAUL
A.
GERALD GREGORY GORDON
VIGNOLA, M.
MD
POHOST, D.
S.
MD,
CURFMAN, MYERS,
MD,
FACC MD FACC
Boston, Massachuseffs
and Clinical Findings in
Effusion
Clinical data and echocardiographic findings were correlated in 20 patients with pericardial effusion. Moderate to large effusions were associated with increased motion of the entire heart within the pericardial sac. A correlation was found between the estimated volume of fluid and the diastolic excursion and velocity of the right ventricular and left ventricular walls (P
One of the earliest clinical applications of reflected ultrasound was in the detection of pericardial fluid.l Over the past decade, many clinical studies have confirmed the reliability of the echocardiogram in the diagnosis of pericardial effusion. 2-g However, no systematic correlation has been made between the clinical findings in patients with pericardial effusion and other more carefully defined simultaneously measured echocardiographic variables. This investigation was undertaken in an attempt to make these correlations. Special attention was given to the echocardiographic findings that were helpful in predicting the hemodynamic significance of pericardial effusion. Methods
From the Noninvasive Diagnostic Laboratory of the Cardiac Unit, Massachusetts General Hospital, Boston, Mass. This study was supported in part by a Grant from the Ambrose Monell Foundation, New York, N. Y. and in part by Grant HL 14209 for a Specialized Center of Research in Arteriosclerosis from the National Heart and Lung Institute, Bethesda, Md. Manuscript accepted September 17, 1975. Address for reprints: Paul A. Vignola, MD, Cardiac Unit, Massachusetts General Hospital, Boston, Mass. 02114.
Echocardiograms: All echocardiograms demonstrating pericardial effusion performed at the Massachusetts General Hospital between November 1, 1973 and March 1, 1975 were reviewed. Only those tracings that adequately demonstrated both the anterior and posterior mitral valve leaflets and the left ventricular septal and posterior free wall echoes were retained for study. Small isolated posterior effusions were not included. Twenty-eight echocardiograms of 20 patients demonstrating anterior and posterior pericardial fluid met these criteria and formed the basis of this report. Forty-seven tracings either showed minimal pericardial effusions or were deemed technically inadequate and were not included in the analysis. If a pericardial drainage procedure was performed repeat echocardiograms were examined when available. The echocardiograms reviewed were performed with an Ekoline 20A echocardiograph interfaced with a Honeywell 1856 Fiberoptic strip chart recorder. An Aerotech 2.25 megahertz transducer focused at 5, 7.5 or 10 cm was used. The patients were examined in the supine or semierect position. Display of the anterior and posterior mitral valve leaflets verified appropriate transducer position. The damping or reject controls were used to incrementally decrease gain. The volume of anterior and posterior fluid was assessed
April 1976
The American Journal of CARDIOLOGY
Volume 37
791
ECHOCARDIOGRAPHY
IN
PERICARDIAL
EFFUSIONS-VIGNOLA
ET
AL.
trocardiographic findings were correlated with the echocardiographic measurements. Cardiac tamponade was indicated clinically by a combination of two or more of the following criteria: elevated jugular venous pressure, pulsus paradoxus (greater than 10 mm Hg) and X-ray evidence of cardiac silhouette enlargement in patients in a low cardiac output state who showed a prompt favorable hemodynamic response to pericardiocentesis. Seven of the nine patients had technically adequate follow-up echocardiograms after pericardiocentesis. Two patients came to postmortem examination, which included a detailed gross and microscopic examination of the heart and pericardium.
in a semiquantitative fashion by the method recently outlined by Horowitz et aLlo The following variables of mitral value diastolic motion were measured: the E-F slope, C-E excursion of the anterior leaflet, E-E distance between anterior and posterior leaflets and the total duration of diastole. The measured E-F slopes were corrected for the exaggerated total cardiac swing.” The systolic excursion of both the anterior and posterior mitral valve leaflets was also examined. Motion and mean velocity of the epicardial surfaces of the anterior right ventricular and posterior left ventricular wall were measured during diastole. The left ventricular chamber dimensions were determined at end-systole and end-diastole in the manner outlined by Feigenbaum.12 The data obtained were subjected to statistical analysis using Fisher’s exact test.13 Clinical findings: The clinical records of the patients studied were reviewed independently without reference to the echocardiographic data. The clinical, X-ray and elec-
Results Clinical
Data (Table
I)
There were 10 men and 10 women. The median age was 45.6 years (range 13 to 74 years). Six of the 20 pa-
TABLE I Clinical Data
case no.
Age (yr) & Sex
Clinical Diagnosis
Pericardial Tamponade
Blood Pressure fmm Hg)
Heart Rate ibeatsimin)
Pulsus Paradoxus (mm Hg)
5
35F
Uremia ES00 cc
_ _ _ _ + + + -
6
66M
Ca pancreas
7
110/90
82
S-10
7
50F
Acute pericarditis
-
120170
100
8
8
67F
Azotemia; Ca breast, esophagus
_
120170
80
10
9
62M
Ca esophagus
t
120180
100
20
10
40M
Uremia
_
160190
60
_
1
13M
2
14F
Trauma 6 600 cc Lymphoma p1000
cc
3
44M
Ca lung p 1000cc
4
38M
Ca
lung
PI600
cc
90150 120160 118/60 120170 120170
136 130 150 106 98
10 5
120175 120180
130 103
25 5
1 OOi60 105160
160 150
20 15
160/110 160/110
67 75
15 5
; 0
_
+
t
+ _
t t t t
+ _ t + t
.
t t +
+ -
100
_ -
t
_ +
100180
100
-
+
-
140190
115
t
+
t
160190
100
-
_
130/80 120180
Et
-
13
28F
Uremia
-
160/100
120
-
14
47F
Scleroderma; uremia
_
190/l
120
_
15
74F
Lupus
_
120170
120
4
16
52F
Lupus
7
100160
100
18
17
29M
Hypothyroid; viral pericarditrs is 2000 cc
+
120170
85
-
140170
88
18
25M
TB pericarditis
-
120180
19
70M
Cardiomyopathy; congestive failure
-
20
66F
Uremia
_
20
*Patient apneic, on respirator. - = absent; f = present;. . = not determined; ? = uncertarn. Ca = carcinoma; ECG = electrocardiogram; 5 = after removal of pericardial
of CARDIOLOGY
t +
+ _ _ _ _ _ + t t
_ t
-
Viral pericarditis
Journal
t + t t _
+ _ + + _ + + + + _ _ _
t t
Uremia
41 F
The American
Pericardial Rub
t t _ _
51 M
12
April 1976
Cardiomegaly
_ t t t t t t
11
702
Low Voltage ECG
Volume
37
l
fluid; TB = tuberculous
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSIONS-VIGNOLA
ET AL.
observations for evidence of tamponade because of his terminal state. Patient 16 was a 52 year old woman with lupus pericarditis who was admitted because of dyspnea. Although pulsus paradoxus (18 mm Hg) was recorded on the day of admission, the patient’s jugular venous pressure was not increased. The dose of corticosteroids was increased, and 18 hours later her dyspnea decreased and paradoxical pulse disappeared.
tients were thought to have neoplastic effusions. In four of the six patients this diagnosis was confirmed by pericardial biopsy or fluid cytologic study. Pericardial tamponade: Four patients had tamponade as defined. Three of these patients had malignant effusions. The presence of pulsus paradoxus correlated with tamponade (P
Echocardiographic
Data (Table II)
Increased cardiac motion: Moderate to large pericardial effusions were noted in 16 patients who each manifested a striking pendular swinging motion of all cardiac structures (Fig. 1). Larger effusions were associated with greater values for excursion and velocity of the anterior right ventricular and posterior left ventricular walls (P
TABLE II Echocardiographic Data
Effusion (ant/post)
Case no.
Apparent MV Prolapse
MV E-F Slope (mm/&
Posterior Wall Excursion (mm)
Diastolic Posterior Wall Velocity (mm/set)
RV Excursion (mm)
Diastolic RV Wall Velocity (mm/set)
Systolic Notch RV Wail
42.9 25.0
4 0
14.3 0
No No
128 200 164
25.0 41.4 0
25 19 0
125.0 65.5 0
No No No
12 116
60 0
12 0
60 0
Yes No
15.0 25.0
17 7
85.0 35.0
Yes
16.3 0
No No
1
Small/large None/small
Early No
172 140
2
Large/mod Modimod None/none
Late Pansystolic None
3
Mod/mod None/none
Late No
4
Large/small Mod/small
Early No
5
Mod/small None/none
Pansystolic No
6
Large/mod
Pansystolic
22
10
7
Small/small
No
156
4
8
Small/small
No
60
9
Mod/mod
No
10
Small/small
11 12
12 7
42 30 145 120
1
20.4 0
Yes
52.6
Yes
12.9
4
12.9
No
IO
25
3
7.5
No
30
29
90.6
29
90.6
Yes
No
83
8
12.3
6
9.2
No
Small/small
No
64
7
20
0
0
No
Small/mod None/small
No
10.2 0
No No
13 10
No
26.3
20
26.5 18.2
13
None/mod
Early
110
32.1
0
0
No
14
None/mod
No
58
20.8
0
0
No
15
Mod/mod
No
60
20.8
16.7
No
16
Mod/mod
No
60
24
10
33
No
17
Mod/large None/none
Pan No
45 74
20 0
19 0
42.3 0
Yes No
18
Small/mod
62
4
10.3
0
0
No
19
Mod/small
No
160
2
6.3
2
6.3
No
20
Large/large
No
55
5
93.7 t9.62
7.25 +1.10
Mean SEM Ant = anterior; the mean.
.. MV = anterior
leaflet
of the mitral
valve; post = posterior;
April 1976
4
25 23.0 k3.59 RV = anterior
No
5
25
7.32 ?I.49
25.2 +6.28
wall of the right ventricle;
SEM = standard error of
The American Journal of CARDIOLOGY
Volume 37
763
ECHOCARDIOGRAPHY
IN
PERICARDIAL
EFFUSIONS-VIGNOLA
ET
AL.
ECG
CW
ECG CW
EFF
RV
RV
/vs AMV
AMV
PL VW
P/WV
EFF PER/
PL VW II
C
B
FIGURE 1. Case 2. Serial echocardiograms
of a 14 year old girl with a neoplastic pericardial effusion but without tamponade. A, heart rate is 150 beats/min. The excursion of the right and left ventricular walls and the interventricular septum is increased. All three structures appear to move posteriorly during systole. Arrows indicate apparent late systolic prolapse of the mitral valve. B, heart rate is 106 beats/min. Prolapse appears to occur throughout systole. C, after pericardiocentesis the increased cardiac motion and apparent mitral valve prolapse are no longer present. AMV = anterior leaflet of the mitral valve: CW = chest wall; ECG = electrocardiogram: EFF = pericardial effusion; IVS = interventricular septum: PERI = pericardium; PLVW = posterior left ventricular wall: PMV = posterior leaflet of the mitral valve: RV = anterior wall of the right ventricle.
effusions when comparable volumes of fluid were present as estimated by the echocardiogram (P <0.02). Mitral valve prolapse: Mitral valve prolapse was apparent in eight of the patients studied. Three patterns were observed: early (Fig. 2), late and pansystolic posterior movement of the mitral leaflets (Fig. 1, A and B). These patterns were significantly related to the heart rate (P
effusions the two dimensions were almost equal, thereby invalidating attempted calculations of left ventricular dimensions and ejection fraction. At rapid heart rates the interventricular septum occasionally appeared to move paradoxically (Fig. 3). Again this was an artifact of the exaggerated cardiac motion and did not correlate significantly with any of the clinical or echocardiographic variables examined, including the presence of tamponade. Signs of cardiac tamponade: There was a significant correlation between a diminished E-F slope (less than 50 mm/set) and the presence of tamponade. Furthermore, a notch on the anterior right ventricular epicardial surface occurring 0.04 f 0.01 second after the QRS was noted in all four patients with tamponade (Fig. 2 and 3). The notch was intermittent but could easily be seen after at least 50 percent of the cardiac cycles. It was not observed in 15 of the 16 patients without tamponade. Three of the four patients had repeat echocardiograms after pericardiocentesis. In each instance with relief of tamponade the E-F slope increased to within the normal range and the notch disappeared (Fig. 3). In one patient both findings appeared again when tamponade recurred. Tamponade was not documented in one patient (Case 6) who demonstrated a diminished E-F slope and a right ventricular epicardial notch. As noted, this patient was in a terminal state and was not subjected to frequent observations for evidence of tamponade.
malignant
704
April 1976
The American
Journal
of CARDIOLOGY
Volume
Discussion Mechanism of increased cardiac motion: Feigenbaum et al.ll were the first to note the echocardiographic finding of increased cardiac motion in pa37
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSIONS-VIGNOLA
ET AL.
ECG CW
./1/s .AMV
FIGURE 2. Case 4. Echocardiogram of a 38 year old man with a neoplastic pericardial effusion and pericardial tamponade. The E-F slope of the anterior leaflet of the mitral valve (AMV) is diminished and a right ventricular notch is present. Arrows indicate apparent early systolic prolapse of the mitral valve. Abbreviations as in Figure 1.
f
tients with large pericardial effusionsll and their observation was subsequently confirmed by others.7J4 Although the exact nature of this exaggerated anteroposterior displacement of the whole heart is not known, excessive rotation due to loss of the normal restraining influence of the surrounding mediastinal structures has been proposed.7J’ The correlation between the estimated volume of effusion noted in our study and the diastolic excursion and velocity of the anterior right ventricular and posterior left ventricular walls supports this mechanism and suggests that a greater swing occurs with larger effusions. The frequent association of neoplastic pericardial effusion and increased cardiac motion, previously
stressed,7J1 was confirmed in our study. Although uremia, collagen vascular disease and trauma did produce a comparable degree of fluid accumulation as estimated by the echocardiogram, patients with these conditions had less abnormality of wall excursion and velocity than was observed with neoplastic effusions. There was a significant difference in right and left ventricular diastolic wall velocity and excursion between malignant and other effusions at comparable fluid volumes. Three factors should be considered: (1) pericardial compliance, (2) fluid viscosity, and (3) ventricular inertia. Ingrowth of neoplastic tissue into the pericardium resulting in a rigid noncompliant sac that allows uniform distribution of
B
A
ECG
cw
FIGURE 3. Case 3. Serial echocardiograms of a 44 vear old man with a neoolastic effusion and tamoonade before and after pericardiocentesis. A, heart rate is 130 beats/min. The E-F slope of the anterior mitral valve (AMV) is diminished and a right ventricular notch is present. Arrows indicate apparent late systolic prolapse of the mitral valve. B, after relief of tamponade. heart rate is 103 beats/min. The E-F slope is now normal and “pseudo” prolapse is no longer apparent.
AMV AMV+%$$_
C..W.‘..
PL VW
4 EFF PER/ /’ A
April 1976
B
The American
Journal
of CARDIOLOGY
Volume
37
706
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSIONS-VIGNOLA
ET AL.
lated to the compliance of the pericardial sac. It follows that high pressures may occur with small effusions that develop rapidly whereas slowly accumulating pericardial effusions may result in no tamponade despite large volumes. 20,21 Although echocardiography can reliably detect even small volumes of intrapericardial fluid, lo the presence of critical cardiac compression must be assessed on the basis of multiple clinical and hemodynamic variables. Mitral valve E-F slope: The steepness of the diastolic closure slope of the mitral valve is generally thought to be related to the rate of the left ventricular filling1g,22-24 and therefore, indirectly, to left ventricular diastolic compliance. The significant correlation between a decreased E-F slope of the anterior leaflet of the mitral valve and clinical pericardial tamponade in our study population suggests diminished left ventricular filling consequent to critical cardiac compression. Several observations support this contention. No patient with a normal E-F slope was thought to have tamponade. Only one of the five patients who demonstrated a diminished E-F slope was thought not to have critical cardiac compression. Three of the four patients had repeat echocardiograms after pericardiocentesis. In all three instances the slope reverted to normal. A decreased E-F slope of the anterior leaflet of the mitral valve is seen in several conditions resulting in a diminution in the rate of left ventricular filling.23 This relative lack of specificity should be considered when evaluating the echocardiogram of patients with pericardial effusion. Our data suggest that a normal E-F slope in such a patient may be evidence against the presence of pericardial tamponade. Notch on the right ventricular epicardial echo: This finding was present in the four patients with tamponade. When pronounced, it could also be seen superimposed simultaneously on the interventricular septum and was associated with coarse oscillations of the left ventricular posterior wall. The notch occurred 0.04 f 0.01 second after the QRS complex. Since isovolumetric contraction begins 0.01 to 0.05 second after inscription of the QRS complex, the timing suggests that the two events may be related.25 Isovolumetric contraction is heralded by anterior displacement of the apex, which can be recorded clinically at the upstroke of the apex cardiogram. With the restraining influences of the mediastinal structures removed and the heart free to swing, augmentation of isovolumetric movement might occur and be recorded as a change in cardiac direction. However, this explanation does not account for the absence of the notch when total cardiac motion is exaggerated without tamponade. One would have to postulate a qualitative or quantitative change in isovolumetric contraction related to the presence of tamponade. No such evidence is available. Regardless of origin, the presence of the notch may serve as a useful echocardiographic marker of clinical tamponade. No patient without the notch was thought to have tamponade even though the estimated volume of fluid, degrees of
fluid and therefore unimpeded cardiac swing has been offered as a possible mechanism for this .association.7 In inflammatory nonneoplastic effusions adhesions occur that may attenuate cardiac motion. Differences in the viscosity of the pericardial fluid and differences in ventricular mass and inertia may also affect cardiac swing although there are few data on this point.15J6 Mechanism of “pseudo” mitral valve prolapse: Echocardiography has provided a reliable and noninvasive method of diagnosing mitral valve prolapse.17 Although the variable spectrum of the echocardiographic manifestations has recently been emphasized,r8 little has been written about the problem of false positive diagnosis with this technique. Several examples of apparent mitral valve prolapse were demonstrated echocardiographically in our study of patients with pericardial effusion. Mitral valve motion as observed by echocardiography is the resultant of intrinsic mitral leaflet movement and to a lesser extent mitral ring and total heart motion.lg In the presence of a large effusion cardiac motion is exaggerated. When exaggerated posterior excursion of the whole heart occurs during systole the resultant mitral echo displays posterior movement giving the appearance of systolic prolapse. In the eight patients showing this abnormality the timing of this “pseudo” prolapse appeared to be a function of the degree of cardiac swing and the heart rate. When the heart rate was 120 beatslmin or greater the posterior swing occurred during the beginning or end of systole. When the rate was less than 120 beats/min the swing occurred in mid-systole, giving the appearance of pansystolic mitral prolapse. The following findings suggest that this observation is artifactual and does not represent true mitral prolapse. Systolic clicks or murmurs, or both, were absent in all eight patients demonstrating “pseudo” prolapse. Seven of these patients had repeat echocardiograms after a pericardial drainage procedure. In every case when the exaggerated cardiac swinging motion decreased or disappeared the mitral valve echocardiogram revealed no suggestion of prolapse. Finally, two patients who demonstrated the early systolic “pseudo” prolapse pattern died. At postmortem examination of the mitral valve leaflets, the chordae tendineae and papillary muscles were completely normal both grossly and histologically. Thus, the mitral valve echocardiogram in patients with pericardial effusion may demonstrate a pattern of mitral valve prolapse that is due to exaggerated total cardiac motion rather than to intrinsic mitral valve disease. Observation of the motion of the entire heart will help to distinguish this “pseudo” prolapse from true prolapse. Pericardial
Tamponade
Pericardial tamponade is characterized by impairment in the diastolic filling of the heart with a resultant decrease in cardiac stroke output.20 It is known that elevation of pericardial pressure is primarily re-
766
April 1976
The American
Journal
of CARDIOLOGY
Volume
37
ECHOCARDIOGRAPHY
swing, and heart rates were comparable. Only one patient who was thought not to have pericardial tamponade had the notch demonstrated. Additionally, three of the four patients had repeat echocardiograms after relief of tamponade. In all three instances the notch and the coarse oscillations of the posterior left ventricular wall disappeared. The reason for the intermittency of the notch in a single tracing is unclear. It may be related to variation in the angle between the echo beam and the right ventricular wall’due to the exaggerated anteroposterior excursion of the whole heart. The right ventricular epicardial notch has not previously been observed,l-gill possibly because many of the original echocardiographic studies of pericardial effusion were performed before the strip chart recorder became generally available. Although Polar-
IN PERICARDIAL
EFFUSIONS-VIGNOLA
ET AL.
oid@ photographs of oscilloscopic tracings were adequate in identifying the presence of pericardial fluid, they did not lend themselves as well to careful examination of subtle changes in cardiac wall and mitral valve motion. The small number of patients with pericardial tamponade in this study make any correlation with a diminished E-F slope and the right ventricular notch tenuous. Our findings should be considered preliminary; their sensitivity and specificity need to be evaluated further in a larger series of patients with tamponade. Acknowledgment We thank Drs. Robert A. Johnson and Edward W. Bough for reviewing the manuscript, Mr. John Newall for the statistical analysis and Mry. Harriet Walker for her help in preparing the manuscript.
References 1. Feigenbaum H, Waldhausen JA, Hyde HP: Ultrasound diagnosis of pericardial effusion. JAMA 191:711-714, 1965 2. Soulen RH, Lapayowker MS, Gimener JL: Echocardiography in the diagnosis of pericardial effusion. Radiology 86: 1047-105 1, 1966 3. Rothman J, Chase NE, Kricheff II, et al: Ultrasonic diagnosis of pericardial effusion. Circulation 35:358-364, 1967 4. Feigenbaum H: Echocardiographic diagnosis of pericardial effusion. Am J Cardiol 26:475-479, 1970 5. Feigenbaum H, Zaky A, Waldhausen JA: Use of ultrasound in the diagnosis of pericardial effusion. Ann Intern Med 65:443452, 1966 6. Goldberg BB: Ultrasonic determination of pericardial effusion. JAMA 202:103-106, 1967 7. Klein JJ, Segal BL: Pericardial effusion diagnosed by reflected ultrasound. Am J Cardiol 2257-64, 1968 8. Moss A, Bruhn F: The echocardiogram: an ultrasound technic for the detection of pericardial effusion. N Engl J Med 274: 380-384, 1966 9. Pate JW, Gardner HC, Norman RS: Diagnosis of pericardial effusion by echocardiography. Ann Surg 165826-829, 1967 10. Horowitz MS, Schultz CS, Stinson ES, et al: Sensitivity and specificity of echocardiographic diagnosis of pericardial effusion. Circulation 50:239-247, 1974 11. Feigenbaum H, Zaky A, Grabhorn LL: Cardiac motion in patients with pericardial effusion. Circulation 34:611-619, 1966 12. Feigenbaum H: Echocardiography. Philadelphia, Lea & Febiger, 1972, p 103 13. Keeping ES: Introduction to Statistical Inference. Princeton, D Van Nostrand, 1962, p 319 14. Gabor GE, Winberg F, Bloom HS: Electrical and mechanical alternation in pericardial effusion. Chest 59:341-344, 1971
15.
Davidsohn I, Henry JB: Clinical Diagnosis by Laboratory Methods, 14th edition. Philadelphia, WB Saunders, 1969, p 11751176 16. Levison SA, MacTate RP: Clinical Laboratory Diagnosis, 7th edition. Philadelphia, Lea 8 Febiger, 1969, p 1084-1090 17. Dillon JC, Haine CL, Chaing S, et al: Use of echocardiography in patients with prolapsed mitral valve. Circulation 43503-507, 1971 18. DeMaria AN, Kind JF, Bogren H, et al: The variable spectrum of echocardiographic manifestations of the mitral valve prolapse syndrome. Circulation 50:33-41, 1974 19. Zaky A, Nasser WK, Feigenbaum H: Study of mitral valve action recorded by reflected ultrasound and its application in the diagnosis of mitral stenosis. Circulation 37:789-799. 1968 20. Shabetai R, Fowler NO, Guntheroth WG: The hemodynamics of cardiac tamponade and constrictive pericarditis. Am J Cardiol 26:480-489, 1970 21. lsaacs JP, Bergulung E, Sarnoff SJ: Ventricular function. III. The pathologic physiology of acute cardiac tamponade studied by means of ventricular function curves. Am Heart J 48:66-76, 1954 22. Layton C, Gent G, Pridie R, et al: Diastolic closure rate of normal mitral valve. Br Heart J 35:1066-1074, 1973 23. Laniado S, Yellin E, Kotler M, et al: A study of the dynamic relations between the mitral valve echogram and phasic mitral flow. Circulation 51:104-113, 1975 24. Zaky A, Grabhorn L, Feigenbaum H: Movement of the mitral ring: A study of ultrasoundcardiography. Cardiovasc Res 1: 121-127, 1967 25. Tavel ME: Clinical Phonocardiography and External Pulse Recording, 2nd edition. Chicago, Year Book Medical Publishers, 1972, p 48-49
April 1976
The American
Journal
of CARDIOLOGY
Volume
37
767
of Echocardiographic
Patients With Pericardial
PAUL
A.
GERALD GREGORY GORDON
VIGNOLA, M.
MD
POHOST, D.
S.
MD,
CURFMAN, MYERS,
MD,
FACC MD FACC
Boston, Massachuseffs
and Clinical Findings in
Effusion
Clinical data and echocardiographic findings were correlated in 20 patients with pericardial effusion. Moderate to large effusions were associated with increased motion of the entire heart within the pericardial sac. A correlation was found between the estimated volume of fluid and the diastolic excursion and velocity of the right ventricular and left ventricular walls (P
One of the earliest clinical applications of reflected ultrasound was in the detection of pericardial fluid.l Over the past decade, many clinical studies have confirmed the reliability of the echocardiogram in the diagnosis of pericardial effusion. 2-g However, no systematic correlation has been made between the clinical findings in patients with pericardial effusion and other more carefully defined simultaneously measured echocardiographic variables. This investigation was undertaken in an attempt to make these correlations. Special attention was given to the echocardiographic findings that were helpful in predicting the hemodynamic significance of pericardial effusion. Methods
From the Noninvasive Diagnostic Laboratory of the Cardiac Unit, Massachusetts General Hospital, Boston, Mass. This study was supported in part by a Grant from the Ambrose Monell Foundation, New York, N. Y. and in part by Grant HL 14209 for a Specialized Center of Research in Arteriosclerosis from the National Heart and Lung Institute, Bethesda, Md. Manuscript accepted September 17, 1975. Address for reprints: Paul A. Vignola, MD, Cardiac Unit, Massachusetts General Hospital, Boston, Mass. 02114.
Echocardiograms: All echocardiograms demonstrating pericardial effusion performed at the Massachusetts General Hospital between November 1, 1973 and March 1, 1975 were reviewed. Only those tracings that adequately demonstrated both the anterior and posterior mitral valve leaflets and the left ventricular septal and posterior free wall echoes were retained for study. Small isolated posterior effusions were not included. Twenty-eight echocardiograms of 20 patients demonstrating anterior and posterior pericardial fluid met these criteria and formed the basis of this report. Forty-seven tracings either showed minimal pericardial effusions or were deemed technically inadequate and were not included in the analysis. If a pericardial drainage procedure was performed repeat echocardiograms were examined when available. The echocardiograms reviewed were performed with an Ekoline 20A echocardiograph interfaced with a Honeywell 1856 Fiberoptic strip chart recorder. An Aerotech 2.25 megahertz transducer focused at 5, 7.5 or 10 cm was used. The patients were examined in the supine or semierect position. Display of the anterior and posterior mitral valve leaflets verified appropriate transducer position. The damping or reject controls were used to incrementally decrease gain. The volume of anterior and posterior fluid was assessed
April 1976
The American Journal of CARDIOLOGY
Volume 37
791
ECHOCARDIOGRAPHY
IN
PERICARDIAL
EFFUSIONS-VIGNOLA
ET
AL.
trocardiographic findings were correlated with the echocardiographic measurements. Cardiac tamponade was indicated clinically by a combination of two or more of the following criteria: elevated jugular venous pressure, pulsus paradoxus (greater than 10 mm Hg) and X-ray evidence of cardiac silhouette enlargement in patients in a low cardiac output state who showed a prompt favorable hemodynamic response to pericardiocentesis. Seven of the nine patients had technically adequate follow-up echocardiograms after pericardiocentesis. Two patients came to postmortem examination, which included a detailed gross and microscopic examination of the heart and pericardium.
in a semiquantitative fashion by the method recently outlined by Horowitz et aLlo The following variables of mitral value diastolic motion were measured: the E-F slope, C-E excursion of the anterior leaflet, E-E distance between anterior and posterior leaflets and the total duration of diastole. The measured E-F slopes were corrected for the exaggerated total cardiac swing.” The systolic excursion of both the anterior and posterior mitral valve leaflets was also examined. Motion and mean velocity of the epicardial surfaces of the anterior right ventricular and posterior left ventricular wall were measured during diastole. The left ventricular chamber dimensions were determined at end-systole and end-diastole in the manner outlined by Feigenbaum.12 The data obtained were subjected to statistical analysis using Fisher’s exact test.13 Clinical findings: The clinical records of the patients studied were reviewed independently without reference to the echocardiographic data. The clinical, X-ray and elec-
Results Clinical
Data (Table
I)
There were 10 men and 10 women. The median age was 45.6 years (range 13 to 74 years). Six of the 20 pa-
TABLE I Clinical Data
case no.
Age (yr) & Sex
Clinical Diagnosis
Pericardial Tamponade
Blood Pressure fmm Hg)
Heart Rate ibeatsimin)
Pulsus Paradoxus (mm Hg)
5
35F
Uremia ES00 cc
_ _ _ _ + + + -
6
66M
Ca pancreas
7
110/90
82
S-10
7
50F
Acute pericarditis
-
120170
100
8
8
67F
Azotemia; Ca breast, esophagus
_
120170
80
10
9
62M
Ca esophagus
t
120180
100
20
10
40M
Uremia
_
160190
60
_
1
13M
2
14F
Trauma 6 600 cc Lymphoma p1000
cc
3
44M
Ca lung p 1000cc
4
38M
Ca
lung
PI600
cc
90150 120160 118/60 120170 120170
136 130 150 106 98
10 5
120175 120180
130 103
25 5
1 OOi60 105160
160 150
20 15
160/110 160/110
67 75
15 5
; 0
_
+
t
+ _
t t t t
+ _ t + t
.
t t +
+ -
100
_ -
t
_ +
100180
100
-
+
-
140190
115
t
+
t
160190
100
-
_
130/80 120180
Et
-
13
28F
Uremia
-
160/100
120
-
14
47F
Scleroderma; uremia
_
190/l
120
_
15
74F
Lupus
_
120170
120
4
16
52F
Lupus
7
100160
100
18
17
29M
Hypothyroid; viral pericarditrs is 2000 cc
+
120170
85
-
140170
88
18
25M
TB pericarditis
-
120180
19
70M
Cardiomyopathy; congestive failure
-
20
66F
Uremia
_
20
*Patient apneic, on respirator. - = absent; f = present;. . = not determined; ? = uncertarn. Ca = carcinoma; ECG = electrocardiogram; 5 = after removal of pericardial
of CARDIOLOGY
t +
+ _ _ _ _ _ + t t
_ t
-
Viral pericarditis
Journal
t + t t _
+ _ + + _ + + + + _ _ _
t t
Uremia
41 F
The American
Pericardial Rub
t t _ _
51 M
12
April 1976
Cardiomegaly
_ t t t t t t
11
702
Low Voltage ECG
Volume
37
l
fluid; TB = tuberculous
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSIONS-VIGNOLA
ET AL.
observations for evidence of tamponade because of his terminal state. Patient 16 was a 52 year old woman with lupus pericarditis who was admitted because of dyspnea. Although pulsus paradoxus (18 mm Hg) was recorded on the day of admission, the patient’s jugular venous pressure was not increased. The dose of corticosteroids was increased, and 18 hours later her dyspnea decreased and paradoxical pulse disappeared.
tients were thought to have neoplastic effusions. In four of the six patients this diagnosis was confirmed by pericardial biopsy or fluid cytologic study. Pericardial tamponade: Four patients had tamponade as defined. Three of these patients had malignant effusions. The presence of pulsus paradoxus correlated with tamponade (P
Echocardiographic
Data (Table II)
Increased cardiac motion: Moderate to large pericardial effusions were noted in 16 patients who each manifested a striking pendular swinging motion of all cardiac structures (Fig. 1). Larger effusions were associated with greater values for excursion and velocity of the anterior right ventricular and posterior left ventricular walls (P
TABLE II Echocardiographic Data
Effusion (ant/post)
Case no.
Apparent MV Prolapse
MV E-F Slope (mm/&
Posterior Wall Excursion (mm)
Diastolic Posterior Wall Velocity (mm/set)
RV Excursion (mm)
Diastolic RV Wall Velocity (mm/set)
Systolic Notch RV Wail
42.9 25.0
4 0
14.3 0
No No
128 200 164
25.0 41.4 0
25 19 0
125.0 65.5 0
No No No
12 116
60 0
12 0
60 0
Yes No
15.0 25.0
17 7
85.0 35.0
Yes
16.3 0
No No
1
Small/large None/small
Early No
172 140
2
Large/mod Modimod None/none
Late Pansystolic None
3
Mod/mod None/none
Late No
4
Large/small Mod/small
Early No
5
Mod/small None/none
Pansystolic No
6
Large/mod
Pansystolic
22
10
7
Small/small
No
156
4
8
Small/small
No
60
9
Mod/mod
No
10
Small/small
11 12
12 7
42 30 145 120
1
20.4 0
Yes
52.6
Yes
12.9
4
12.9
No
IO
25
3
7.5
No
30
29
90.6
29
90.6
Yes
No
83
8
12.3
6
9.2
No
Small/small
No
64
7
20
0
0
No
Small/mod None/small
No
10.2 0
No No
13 10
No
26.3
20
26.5 18.2
13
None/mod
Early
110
32.1
0
0
No
14
None/mod
No
58
20.8
0
0
No
15
Mod/mod
No
60
20.8
16.7
No
16
Mod/mod
No
60
24
10
33
No
17
Mod/large None/none
Pan No
45 74
20 0
19 0
42.3 0
Yes No
18
Small/mod
62
4
10.3
0
0
No
19
Mod/small
No
160
2
6.3
2
6.3
No
20
Large/large
No
55
5
93.7 t9.62
7.25 +1.10
Mean SEM Ant = anterior; the mean.
.. MV = anterior
leaflet
of the mitral
valve; post = posterior;
April 1976
4
25 23.0 k3.59 RV = anterior
No
5
25
7.32 ?I.49
25.2 +6.28
wall of the right ventricle;
SEM = standard error of
The American Journal of CARDIOLOGY
Volume 37
763
ECHOCARDIOGRAPHY
IN
PERICARDIAL
EFFUSIONS-VIGNOLA
ET
AL.
ECG
CW
ECG CW
EFF
RV
RV
/vs AMV
AMV
PL VW
P/WV
EFF PER/
PL VW II
C
B
FIGURE 1. Case 2. Serial echocardiograms
of a 14 year old girl with a neoplastic pericardial effusion but without tamponade. A, heart rate is 150 beats/min. The excursion of the right and left ventricular walls and the interventricular septum is increased. All three structures appear to move posteriorly during systole. Arrows indicate apparent late systolic prolapse of the mitral valve. B, heart rate is 106 beats/min. Prolapse appears to occur throughout systole. C, after pericardiocentesis the increased cardiac motion and apparent mitral valve prolapse are no longer present. AMV = anterior leaflet of the mitral valve: CW = chest wall; ECG = electrocardiogram: EFF = pericardial effusion; IVS = interventricular septum: PERI = pericardium; PLVW = posterior left ventricular wall: PMV = posterior leaflet of the mitral valve: RV = anterior wall of the right ventricle.
effusions when comparable volumes of fluid were present as estimated by the echocardiogram (P <0.02). Mitral valve prolapse: Mitral valve prolapse was apparent in eight of the patients studied. Three patterns were observed: early (Fig. 2), late and pansystolic posterior movement of the mitral leaflets (Fig. 1, A and B). These patterns were significantly related to the heart rate (P
effusions the two dimensions were almost equal, thereby invalidating attempted calculations of left ventricular dimensions and ejection fraction. At rapid heart rates the interventricular septum occasionally appeared to move paradoxically (Fig. 3). Again this was an artifact of the exaggerated cardiac motion and did not correlate significantly with any of the clinical or echocardiographic variables examined, including the presence of tamponade. Signs of cardiac tamponade: There was a significant correlation between a diminished E-F slope (less than 50 mm/set) and the presence of tamponade. Furthermore, a notch on the anterior right ventricular epicardial surface occurring 0.04 f 0.01 second after the QRS was noted in all four patients with tamponade (Fig. 2 and 3). The notch was intermittent but could easily be seen after at least 50 percent of the cardiac cycles. It was not observed in 15 of the 16 patients without tamponade. Three of the four patients had repeat echocardiograms after pericardiocentesis. In each instance with relief of tamponade the E-F slope increased to within the normal range and the notch disappeared (Fig. 3). In one patient both findings appeared again when tamponade recurred. Tamponade was not documented in one patient (Case 6) who demonstrated a diminished E-F slope and a right ventricular epicardial notch. As noted, this patient was in a terminal state and was not subjected to frequent observations for evidence of tamponade.
malignant
704
April 1976
The American
Journal
of CARDIOLOGY
Volume
Discussion Mechanism of increased cardiac motion: Feigenbaum et al.ll were the first to note the echocardiographic finding of increased cardiac motion in pa37
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSIONS-VIGNOLA
ET AL.
ECG CW
./1/s .AMV
FIGURE 2. Case 4. Echocardiogram of a 38 year old man with a neoplastic pericardial effusion and pericardial tamponade. The E-F slope of the anterior leaflet of the mitral valve (AMV) is diminished and a right ventricular notch is present. Arrows indicate apparent early systolic prolapse of the mitral valve. Abbreviations as in Figure 1.
f
tients with large pericardial effusionsll and their observation was subsequently confirmed by others.7J4 Although the exact nature of this exaggerated anteroposterior displacement of the whole heart is not known, excessive rotation due to loss of the normal restraining influence of the surrounding mediastinal structures has been proposed.7J’ The correlation between the estimated volume of effusion noted in our study and the diastolic excursion and velocity of the anterior right ventricular and posterior left ventricular walls supports this mechanism and suggests that a greater swing occurs with larger effusions. The frequent association of neoplastic pericardial effusion and increased cardiac motion, previously
stressed,7J1 was confirmed in our study. Although uremia, collagen vascular disease and trauma did produce a comparable degree of fluid accumulation as estimated by the echocardiogram, patients with these conditions had less abnormality of wall excursion and velocity than was observed with neoplastic effusions. There was a significant difference in right and left ventricular diastolic wall velocity and excursion between malignant and other effusions at comparable fluid volumes. Three factors should be considered: (1) pericardial compliance, (2) fluid viscosity, and (3) ventricular inertia. Ingrowth of neoplastic tissue into the pericardium resulting in a rigid noncompliant sac that allows uniform distribution of
B
A
ECG
cw
FIGURE 3. Case 3. Serial echocardiograms of a 44 vear old man with a neoolastic effusion and tamoonade before and after pericardiocentesis. A, heart rate is 130 beats/min. The E-F slope of the anterior mitral valve (AMV) is diminished and a right ventricular notch is present. Arrows indicate apparent late systolic prolapse of the mitral valve. B, after relief of tamponade. heart rate is 103 beats/min. The E-F slope is now normal and “pseudo” prolapse is no longer apparent.
AMV AMV+%$$_
C..W.‘..
PL VW
4 EFF PER/ /’ A
April 1976
B
The American
Journal
of CARDIOLOGY
Volume
37
706
ECHOCARDIOGRAPHY
IN PERICARDIAL EFFUSIONS-VIGNOLA
ET AL.
lated to the compliance of the pericardial sac. It follows that high pressures may occur with small effusions that develop rapidly whereas slowly accumulating pericardial effusions may result in no tamponade despite large volumes. 20,21 Although echocardiography can reliably detect even small volumes of intrapericardial fluid, lo the presence of critical cardiac compression must be assessed on the basis of multiple clinical and hemodynamic variables. Mitral valve E-F slope: The steepness of the diastolic closure slope of the mitral valve is generally thought to be related to the rate of the left ventricular filling1g,22-24 and therefore, indirectly, to left ventricular diastolic compliance. The significant correlation between a decreased E-F slope of the anterior leaflet of the mitral valve and clinical pericardial tamponade in our study population suggests diminished left ventricular filling consequent to critical cardiac compression. Several observations support this contention. No patient with a normal E-F slope was thought to have tamponade. Only one of the five patients who demonstrated a diminished E-F slope was thought not to have critical cardiac compression. Three of the four patients had repeat echocardiograms after pericardiocentesis. In all three instances the slope reverted to normal. A decreased E-F slope of the anterior leaflet of the mitral valve is seen in several conditions resulting in a diminution in the rate of left ventricular filling.23 This relative lack of specificity should be considered when evaluating the echocardiogram of patients with pericardial effusion. Our data suggest that a normal E-F slope in such a patient may be evidence against the presence of pericardial tamponade. Notch on the right ventricular epicardial echo: This finding was present in the four patients with tamponade. When pronounced, it could also be seen superimposed simultaneously on the interventricular septum and was associated with coarse oscillations of the left ventricular posterior wall. The notch occurred 0.04 f 0.01 second after the QRS complex. Since isovolumetric contraction begins 0.01 to 0.05 second after inscription of the QRS complex, the timing suggests that the two events may be related.25 Isovolumetric contraction is heralded by anterior displacement of the apex, which can be recorded clinically at the upstroke of the apex cardiogram. With the restraining influences of the mediastinal structures removed and the heart free to swing, augmentation of isovolumetric movement might occur and be recorded as a change in cardiac direction. However, this explanation does not account for the absence of the notch when total cardiac motion is exaggerated without tamponade. One would have to postulate a qualitative or quantitative change in isovolumetric contraction related to the presence of tamponade. No such evidence is available. Regardless of origin, the presence of the notch may serve as a useful echocardiographic marker of clinical tamponade. No patient without the notch was thought to have tamponade even though the estimated volume of fluid, degrees of
fluid and therefore unimpeded cardiac swing has been offered as a possible mechanism for this .association.7 In inflammatory nonneoplastic effusions adhesions occur that may attenuate cardiac motion. Differences in the viscosity of the pericardial fluid and differences in ventricular mass and inertia may also affect cardiac swing although there are few data on this point.15J6 Mechanism of “pseudo” mitral valve prolapse: Echocardiography has provided a reliable and noninvasive method of diagnosing mitral valve prolapse.17 Although the variable spectrum of the echocardiographic manifestations has recently been emphasized,r8 little has been written about the problem of false positive diagnosis with this technique. Several examples of apparent mitral valve prolapse were demonstrated echocardiographically in our study of patients with pericardial effusion. Mitral valve motion as observed by echocardiography is the resultant of intrinsic mitral leaflet movement and to a lesser extent mitral ring and total heart motion.lg In the presence of a large effusion cardiac motion is exaggerated. When exaggerated posterior excursion of the whole heart occurs during systole the resultant mitral echo displays posterior movement giving the appearance of systolic prolapse. In the eight patients showing this abnormality the timing of this “pseudo” prolapse appeared to be a function of the degree of cardiac swing and the heart rate. When the heart rate was 120 beatslmin or greater the posterior swing occurred during the beginning or end of systole. When the rate was less than 120 beats/min the swing occurred in mid-systole, giving the appearance of pansystolic mitral prolapse. The following findings suggest that this observation is artifactual and does not represent true mitral prolapse. Systolic clicks or murmurs, or both, were absent in all eight patients demonstrating “pseudo” prolapse. Seven of these patients had repeat echocardiograms after a pericardial drainage procedure. In every case when the exaggerated cardiac swinging motion decreased or disappeared the mitral valve echocardiogram revealed no suggestion of prolapse. Finally, two patients who demonstrated the early systolic “pseudo” prolapse pattern died. At postmortem examination of the mitral valve leaflets, the chordae tendineae and papillary muscles were completely normal both grossly and histologically. Thus, the mitral valve echocardiogram in patients with pericardial effusion may demonstrate a pattern of mitral valve prolapse that is due to exaggerated total cardiac motion rather than to intrinsic mitral valve disease. Observation of the motion of the entire heart will help to distinguish this “pseudo” prolapse from true prolapse. Pericardial
Tamponade
Pericardial tamponade is characterized by impairment in the diastolic filling of the heart with a resultant decrease in cardiac stroke output.20 It is known that elevation of pericardial pressure is primarily re-
766
April 1976
The American
Journal
of CARDIOLOGY
Volume
37
ECHOCARDIOGRAPHY
swing, and heart rates were comparable. Only one patient who was thought not to have pericardial tamponade had the notch demonstrated. Additionally, three of the four patients had repeat echocardiograms after relief of tamponade. In all three instances the notch and the coarse oscillations of the posterior left ventricular wall disappeared. The reason for the intermittency of the notch in a single tracing is unclear. It may be related to variation in the angle between the echo beam and the right ventricular wall’due to the exaggerated anteroposterior excursion of the whole heart. The right ventricular epicardial notch has not previously been observed,l-gill possibly because many of the original echocardiographic studies of pericardial effusion were performed before the strip chart recorder became generally available. Although Polar-
IN PERICARDIAL
EFFUSIONS-VIGNOLA
ET AL.
oid@ photographs of oscilloscopic tracings were adequate in identifying the presence of pericardial fluid, they did not lend themselves as well to careful examination of subtle changes in cardiac wall and mitral valve motion. The small number of patients with pericardial tamponade in this study make any correlation with a diminished E-F slope and the right ventricular notch tenuous. Our findings should be considered preliminary; their sensitivity and specificity need to be evaluated further in a larger series of patients with tamponade. Acknowledgment We thank Drs. Robert A. Johnson and Edward W. Bough for reviewing the manuscript, Mr. John Newall for the statistical analysis and Mry. Harriet Walker for her help in preparing the manuscript.
References 1. Feigenbaum H, Waldhausen JA, Hyde HP: Ultrasound diagnosis of pericardial effusion. JAMA 191:711-714, 1965 2. Soulen RH, Lapayowker MS, Gimener JL: Echocardiography in the diagnosis of pericardial effusion. Radiology 86: 1047-105 1, 1966 3. Rothman J, Chase NE, Kricheff II, et al: Ultrasonic diagnosis of pericardial effusion. Circulation 35:358-364, 1967 4. Feigenbaum H: Echocardiographic diagnosis of pericardial effusion. Am J Cardiol 26:475-479, 1970 5. Feigenbaum H, Zaky A, Waldhausen JA: Use of ultrasound in the diagnosis of pericardial effusion. Ann Intern Med 65:443452, 1966 6. Goldberg BB: Ultrasonic determination of pericardial effusion. JAMA 202:103-106, 1967 7. Klein JJ, Segal BL: Pericardial effusion diagnosed by reflected ultrasound. Am J Cardiol 2257-64, 1968 8. Moss A, Bruhn F: The echocardiogram: an ultrasound technic for the detection of pericardial effusion. N Engl J Med 274: 380-384, 1966 9. Pate JW, Gardner HC, Norman RS: Diagnosis of pericardial effusion by echocardiography. Ann Surg 165826-829, 1967 10. Horowitz MS, Schultz CS, Stinson ES, et al: Sensitivity and specificity of echocardiographic diagnosis of pericardial effusion. Circulation 50:239-247, 1974 11. Feigenbaum H, Zaky A, Grabhorn LL: Cardiac motion in patients with pericardial effusion. Circulation 34:611-619, 1966 12. Feigenbaum H: Echocardiography. Philadelphia, Lea & Febiger, 1972, p 103 13. Keeping ES: Introduction to Statistical Inference. Princeton, D Van Nostrand, 1962, p 319 14. Gabor GE, Winberg F, Bloom HS: Electrical and mechanical alternation in pericardial effusion. Chest 59:341-344, 1971
15.
Davidsohn I, Henry JB: Clinical Diagnosis by Laboratory Methods, 14th edition. Philadelphia, WB Saunders, 1969, p 11751176 16. Levison SA, MacTate RP: Clinical Laboratory Diagnosis, 7th edition. Philadelphia, Lea 8 Febiger, 1969, p 1084-1090 17. Dillon JC, Haine CL, Chaing S, et al: Use of echocardiography in patients with prolapsed mitral valve. Circulation 43503-507, 1971 18. DeMaria AN, Kind JF, Bogren H, et al: The variable spectrum of echocardiographic manifestations of the mitral valve prolapse syndrome. Circulation 50:33-41, 1974 19. Zaky A, Nasser WK, Feigenbaum H: Study of mitral valve action recorded by reflected ultrasound and its application in the diagnosis of mitral stenosis. Circulation 37:789-799. 1968 20. Shabetai R, Fowler NO, Guntheroth WG: The hemodynamics of cardiac tamponade and constrictive pericarditis. Am J Cardiol 26:480-489, 1970 21. lsaacs JP, Bergulung E, Sarnoff SJ: Ventricular function. III. The pathologic physiology of acute cardiac tamponade studied by means of ventricular function curves. Am Heart J 48:66-76, 1954 22. Layton C, Gent G, Pridie R, et al: Diastolic closure rate of normal mitral valve. Br Heart J 35:1066-1074, 1973 23. Laniado S, Yellin E, Kotler M, et al: A study of the dynamic relations between the mitral valve echogram and phasic mitral flow. Circulation 51:104-113, 1975 24. Zaky A, Grabhorn L, Feigenbaum H: Movement of the mitral ring: A study of ultrasoundcardiography. Cardiovasc Res 1: 121-127, 1967 25. Tavel ME: Clinical Phonocardiography and External Pulse Recording, 2nd edition. Chicago, Year Book Medical Publishers, 1972, p 48-49
April 1976
The American
Journal
of CARDIOLOGY
Volume
37
767