Mitral Cleft in Ostium Primum Atrial Septal Defect

Assessed by Cross-sectional Echocardiography

SHINTARO BEPPU, M.D., YASUHARU NIMURA, M.D., HIROSHI SAKAKIBARA, M.D.,
SEIKI NAGATA, M.D., YUNG-DAE PARK, M.D., KIYOSHI BABA, M.D.,
YASUAKI NAITO, M.D., MITSUSHIGE OHTA, M.D., TETSuo KAMIYA, M.D.,
HITOSHI KOYANAGI, M.D., AND TsUYOSHI FUJITA, M.D.

SUMMARY We attempted to detect mitral deformities in ostium primum atrial septal defect using real-
time cross-sectional echocardiography. Transverse sections of the anterior mitral leaflet echo were examined
in 11 patients with this malformation who subsequently received surgical treatment. The section for observing
the transverse view of the anterior leaflet was along the sagittal plane of the body, because of the deformity of
the mitral annulus. Each echocardiographic finding was compared with the surgical and angiographic findings.
On the echocardiogram, the superior and inferior parts of the anterior mitral leaflet separated into two parts
during diastole in all patients with mitral cleft. Thin linear echoes connected the ridges of the cleft and the ven-
tricular septum in seven patients in whom the accessory chordae at that area were revealed at surgery. The
systolic configuration of the anterior leaflet echo varied among the patients. The severity of the mitral regurgi-
tation seemed to relate not only to the size of the cleft but also to the systolic configuration of the anterior
mitral leaflet. After surgery, diastolic separation of the anterior leaflet echo was no longer observed. However,
the abnormal systolic configuration of the anterior leaflet was unchanged.

IN OSTIUM PRIMUM atrial septal defect (ASD),
absence of the membranous septum and a cleft in the
anterior mitral leaflet are the characteristic features
resulting from maldevelopment of the endocardial
cushions. The success of the surgery depends on the
repair of the mitral regurgitation due to mitral cleft,
because mitral regurgitation sometimes remains after
surgery.'-' Mitral repair in this malformation is
Downloaded from http://ahajournals.org by on September 27, 2019
multiple echoes of the mitral valve were reported as
specific findings using M-mode and multicrystal echo-
cardiography."1 12, 14 Mitral cleft has been observed
with real-time cross-sectional echocardiography.16' 17
In the present study our goal was to detect the
mitral cleft, to show the abnormal configuration of the
anterior mitral leaflet using real-time cross-sectional
echocardiography and to compare the echocardio-

difficult due to the complex deformities of the mitral
valve and the deformity of the mitral annulus. At pres-
ent, however, there is no adequate method to dem-
onstrate the details of the mitral deformities before
surgery or the results of the repair of the cleft after
surgery. Angiocardiography reveals the presence of
the mitral cleft and regurgitation but reveals few de-
tails about the leaflet and chordae tendineae.6-8 Al-
though the mitral anomalies can be seen at surgery,
the width of the cleft and the configuration of the
mitral apparatus are modified in the arrested heart,
and the exact orientation of the mitral apparatus is ob-
scure. Therefore, a method of evaluating the condition
of the mitral apparatus in the beating heart is needed.
As echocardiography offers unique information
about the arrangement and movement of the intracar-
diac structures, it has come to be an indispensable ex-
amination in diagnosing congenital and acquired heart
diseases. The echocardiographic findings in endocar-
dial cushion defect, including ostium primum ASD,
have been reported.9 17 In mitral cleft, the systolic
From the National Cardiovascular Center, Research Institute
and Departments of Cardiology, Pediatric Cardiology, Cardiac
Surgery and Radiology, Osaka, Japan.
Address for correspondence: Shintaro Beppu, M.D., National
Cardiovascular Center, Research Institute, 5-chome, Fujishirodai,
Suita, Osaka 565, Japan.
Received February 21, 1979; revision accepted May 6, 1980.
Circulation 62, No. 5, 1980.
1099
graphic findings with the surgical and angiocardio-
graphic findings in patients with ostium primum ASD.
Subjects and Methods
We studied only subjects who later underwent sur-
gery to permit the comparison of echocardiographic
and anatomic findings. Ten patients with ostium
primum ASD and one patient with a transitional form
of endocardial cushion defect"' were included. The
latter patient was included because her ventricular
septal defect was small and was considered clinically
to be the same as ostium primum ASD. There were
five males and six females, ages 4-45 years. At sur-
gery, 10 patients had mitral cleft and the other one did
not. For the sake of convenience, the degree of the
mitral cleft was defined as complete or incomplete, ac-
cording to its extension to the mitral annulus. Seven of
the 10 had complete cleft and the other three had in-
complete cleft. In nine patients with mitral cleft, the
cleft was sutured directly; one patient underwent
mitral valve replacement. The accessory chordae were
revealed at surgery in five patients and were resected
partially in two. The mitral regurgitation before and
after surgery was diagnosed by cineangiocardiog-
raphy using the criteria of Sellers et al.19 The angio-
graphic and surgical findings for individual patients
are listed in table 1.
All patients were examined using real-time cross-
sectional echocardiography before surgery and all
 1100
TABLE 1. Angiocardiographic and Sitrgical Findings
Age Before surgery
Pt (years) MR (cine) Cleft
1 24 2+ Complete
2 15 2+ Complete
3 37 4+ Complete
4 21 3+ Incomplete
5 13 1+ Incomplete
6 4 None Absent
7 26 4+ Incomplete
8 45 4+ Complete
9* 15 2+ Complete
10 7 3+ Complete
11 13 1+ Complete
*Transitional form of endocardial cushion defect.
tNo mitral regurgitant murmur.
tThe accessory chordae were partially resected.
Abbreviations: MR = mitral regurgitation; MVR
except the patient who underwent mitral valve replace-
ment were also examined after surgery. The echocar-
diographic findings of the mitral configuration and
motion were compared with those in healthy subjects
and patients without abnormalities in the mitral valve.
The echocardiograph used was a Toshiba SSH- 1 IA
or a Varian V-3000, which are electronic phased-array
Downloaded from http://ahajournals.org by on September 27, 2019
sector systems. In the Toshiba SSH-1 1A, the trans-
ducer has 32 elements with a frequency of 2.4 MHz.
The scanning angle is 780 and the line density is 224
lines/frame. In the Varian V-3000, the transducer has
32 elements with a frequency of 2.25 MHz. The scan-
ning angle is 840 and the line density is 85, 128 or 255
lines/frame at the depth level of 21, 15 or 7 cm,
respectively. Real-time cross-sectional echocardio-
grams were recorded on 8-mm cine film or a video-
tape recorder. The cross-sectional echocardiograms at
a given cardiac phase were recorded on Polaroid film
or obtained later from 8-mm cine film.
Echocardiographic examinations were performed
with the patient in the supine position. The cross-
sectional plane was set to observe the transverse view
of the anterior mitral leaflet. In healthy subjects and
patients with cardiac disease other than ostium
primum ASD, the section was set along the short axis,
as the mitral annulus is parallel to the short axis.
However, in ostium primum ASD, the right side of the
mitral annulus is displaced toward the apex and the
anterior part of the annulus is parallel not to the short
axis but to the sagittal plane of the body, as shown in
the left ventriculogram (fig. IA). Therefore, the sec-
tion for observing the transverse view of the anterior
mitral leaflet was set to be the sagittal plane of the
body so that the section was parallel to the anterior
part of the mitral annulus. First, the transducer was
placed in the third, fourth or fifth interspace, and the
echocardiographic section was set along the long axis
of the left ventricle (fig. 1B-1). Then, the transducer
CI RCULATION VOL 62, No 5, NOVEMBER 1980
During surgery
Accessory After surgery
chordae Method MR (cine)
+ Suture 2-
+ Suture 1+
- MVR
Not checked Suture 3+
Not checked Suture 1+
- - None
- Suture None
- Suture -t
+ Suture -t
+ Suturet 2+
+ Suturet
= mitral valve replacement; 4 = present;- = absent.
was rotated clockwise to the sagittal plane of the body
while the echo of the anterior mitral leaflet was set to
be the center of the cross-sectional image (fig. IB-2).
When the echo of the posterior mitral leaflet or the left
ventricular posterior wall was observed, the
transducer was tilted medially to ensure that the chor-
dae tendineae were not being examined (fig. 1 B-3).
Though this section was oblique to the left ventricle, it
was perpendicular to the anterior mitral leaflet in os-
tium primum ASD. To scan the entire area of the
arnterior leaflet, the transducer was tilted carefully
from the direction of the mitral annulus to that of the
tip of the anterior leaflet.
In three patients, contrast studies with cross-sec-
tional echocardiography at cardiac catheterization
were available. Isotonic saline was injected into the
left ventricular cavity and the location of the
regurgitation was examined.
Results
Normal Anterior Mitral Leaflet
In healthy subjects and patients with cardiac
diseases other than ostium primum ASD, the echo of
the anterior mitral leaflet was a curved line parallel to
the ventricular septum echo during diastole and
straight or slightly concave toward the left atrium and
lying between the left ventricular outflow tract and the
left atrium during systole (fig. 2). There was neither in-
terruption of the leaflet echo nor asynchronous motion
in any part of the leaflet throughout the cardiac phase.
Ostium Primum ASD
The superior and inferior portions of the anterior
mitral leaflet separated into two parts during diastole,
and the superior portion narrowed the left ventricular
outflow tract (figs. 3 and 4). Each part of the leaflet
 ECHO OF MITRAL CLEFT IN OSTIUM PRIMUM ASD/Beppu et al. 1 101

FIGURE 1. (A) Frontal view of the angiocardiogram in os-

titt11z prinium atrial .eptal defect showing that the right side
o/ft/he contour of the left ventricle (L V) consists of the mitral
anndlus. The correct section at which to observe the
transverse 'yiew of the anterior mitral leaflet by cross-
sectional echocardiographv is indicated by the triangle as a
spatial plane. The white arrowsi. indicate mitral regurgita-
tionl. The black arrowt, points to the nonopaque notch that in-
dicates mzitral cleft. (B) Schenmatic drawings indicating the
procedure to set the section. A 0 aorta; Ao.v = aortic
ial vel

"I.. '

1r
Downloaded from http://ahajournals.org by on September 27, 2019

/( t I
B 1 2 3

FIGURE 2. Transverse views of the anterior mitral leaflet (AML) in a healthy subject showing ani uninter-
rupted curve during diastole (left) and a straight line during systole (right). R V = right ventricle; I VS = ven-
tricular septunm; L V left ventricle; LA = left atrium.
 1102 ClIRCULATION VOL 62, No 5, NOVEMBER 1980

DIASTOLE SYSTOLE

(HEAD)

RV
v

L
L.A
Downloaded from http://ahajournals.org by on September 27, 2019

FIGURE 3. Trans verve views of the anterior mitral leaflet (AML) in ostium primunm atrial septal defect
(palici?t 2) showing diastolic separation of the anterior leaflet indicated by the arrowt (left) and thin linear
echoe.s indicat ed byi the arrows between the ventricular septum (I VS) and the anterior leaflet, which was bent
toward the ventricular septum during systole (right). RV = right ventricle; LA = left atrium.

had a different fluttering motion in diastole. These
findings during diastole were found in all patients with
mitral cleft but not in the patient without cleft (patient
6). The extent of the cross section in which the
diastolic separation of the anterior mitral leaflet echo
was observed varied among the patients. In patients
with a complete cleft, the diastolic separation was
easily detected in the wide area. In patients with an in-
complete cleft, the diastolic separation was not
detected near the annulus.
During systole, the echo of the two parts of the
leaflet met at the center and formed a knot echo,
somewhat resembling two hands pressed together. The
size of the knot echo varied among the subjects from
well-defined (fig. 4) to scarcely defined (fig. 3). In the
patients with a well-defined knot echo, the angiocar-
diograms showed a wide nonopaque notch indicating
mitral cleft. In patient 8 there was an opening between
the ridges of the two parts of the leaflet during systole.
In five patients, the echo line of the anterior mitral
leaflet was not straight, but bent toward the ven-
tricular septum with the angle at the contact site (fig.
3). The echo of the anterior leaflet was almost straight
in four patients (fig. 5), one of whom did not have the
cleft. In one of the other two patients, the edge of one
part of the leaflet slipped down the edge of the other
part so that they did not meet symmetrically, and the
echo line of the leaflet was not straight (patient 3) (fig.
4). In the other, patient 4, the prolapse of the anterior
mitral leaflet was shown by echocardiography and the
configuration of the anterior leaflet during systole was
not assessed.
In seven patients, thin linear echoes were detected
between the ventricular septum and the ridges of the
two parts of the divided leaflet echoes (fig. 3). At sur-
gery, accessory chordae were found in the area con-
cerned. In some patients, these echoes were not very
strong and could not be recognized clearly on the real-
time image, but were seen on the 8-mm frame-by-
frame display. All patients with systolic bending of the
anterior leaflet echo had accessory chordae, which
looked as if they were preventing the anterior leaflet
 ECHO OF MITRAL CLEFT IN OSTILUM PRIMUM ASD/Beppu et al. 1103

DIASTOLE SYSTOLE
Downloaded from http://ahajournals.org by on September 27, 2019

2a LA AML LA

FIGURE 4. Transverse views of the anterior nmitral leaflet (A ML) in ostium prinium atrial septal deJtct
(patient 3). Tlhe diastolicfinidings are the same as in figure 3 (left). There is a gap in the coaptation between
t he tituo parts of th e an7terior leaflet (right). I VS ventricular septum ; R V right ventricle; L V left en-
t(rile; left atrilnn7i.

from being straight during systole (fig. 3). Conversely,
the patients with the accessory chordae did not always
show the systolic bending of the anterior leaflet echo.
The cross-sectional echocardiographic findings before
surgery are listed in table 2.
Eight patients had complete cleft, but the severity of
mitral regurgitation varied. In patient 11, whose
echocardiogram showed the thin linear echoes and
systolic bending of the anterior mitral leaflet echo,
angiocardiography showed minimal mitral regurgita-
tion. In patient 7, who did not have the accessory
chordae and whose anterior leaflet echo was straight,
angiocardiography showed severe mitral regurgita-
tion. In this patient, a contrast study showed
regurgitation from the site between the two parts of
the leaflet (fig. 5). The patient who had an opening
between the ridges of the divided leaflet echo had
severe mitral regurgitation and the same finding in the
contrast study as in patient 7.
In all patients, after surgical repair of the cleft, the
anterior mitral leaflet echo did not show the diastolic
separation (fig. 6). It seemed that the motion of the
 1104 CIRCULATION VOL 62, No 5, NOVEMBER 1980

TABLE 2. Echocardiographic Findings

Before surgery After surgery
Dliastolic Thirn linear Systolic l)iastolic Thini linear Systolic
Pt separation* echoes bendinlg separation echoes bending
1 ++ + + + +
2 ++ + ++ + +
3 ++ Not assessed MVR
4 + + Undetermiinedt + Undeterminedt
5 ++ + + ? +
6
7 +
8 ++
9 ++ ++ - ++
Pt ++ + + ±
+
11 ++ + + + i
*++ and + inean that this finding was observed from near the anniulus to the tip of the leaflet and only
near the tip of the leaflet, respectively.
tThe systolic configuration was not assessed because of the prolapse of the anterior mitral leaflet.
Abbreviation: MVR mitral valve replacement.
=

anterior mitral leaflet was not restrained by the suture
site. The knot echo was observed throughout the car-
diac phase, indicating the suture site. In three patients
who showed systolic bending of the anterior leaflet
echo before surgery and whose accessory chordae
were not resected, the thin linear echoes were detected
Downloaded from http://ahajournals.org by on September 27, 2019
changed in two and diminished in one. In one of two
patients who showed systolic bending of the leaflet
before surgery and whose accessory chordae were par-
tially resected at surgery, the echo of the anterior
leaflet was almost straight and mitral regurgitation
had decreased (patient 10). In two patients who had

at the same area as in the preoperative state, and the neither the accessory chordae nor systolic bending of
degree of systolic bending of the anterior leaflet was the anterior leaflet mitral regurgitation was not
unchanged (fig. 6). In these patients, the severity of the revealed. The severity of mitral regurgitation was not
mitral regurgitation by angiocardiography was un- reduced in the patient with mitral valve prolapse

FIGURE 5. Conitrast study in us tiu'n prio?lum atrial septal defect (patient 7). The echo of the anterior leaflet is almost straight.
An ar-ros indicates co ntac,t site of the separated anterior leaflet (left). Immediately after saline injection into the left ventricle
(L V), c nt/rast echoes filled almios t the entire ventricular cavity and sonie of themi are observed in the left atrium (LA ) from the
cot7lact site oJ the two parts, as indicated by the arrow (middle). In the following phase, the contrast echoes in the left atrium
h ae inecreas ed (right). RV right ventricle; IVS =ventricular septum.
 ECHO OF MITRAL CLEFT IN OSTIUM PRIMUM ASD/Beppu et al.
FIGURE 6. Trans verse views of the anterior mizitral leaflet (A ML) after surgical suture of the cleft showing the uninterrupted
aniiterior leafltct anid the kniot echo indicating the suture site during diastole (left) and the bending of the leaflet and the thin linear
ec ho Act ween the ventric ular septuuni (I VS) and the leafle t (arrow) during sys tole (right) (patient 2). R V = right ventricle; LA
icit atrium.
(patient 4). The individual angiographic and echocar-
diographic findings after surgery are listed in tables 1
and 2, respectively.
Discussion
The Transverse View of the Anterior
Downloaded from http://ahajournals.org by on September 27, 2019
Mitral Leaflet in Ostium Primum ASD
In examining the transverse view of the anterior
mitral leaflet by cross-sectional echocardiography, it
was important to choose the section plane carefully.
First, in ostium primum ASD, the section should be
sagittal with respect to the body to cross the anterior
mitral leaflet in the transverse plane. This is because
the right side of the annulus of the mitral valve is dis-
placed toward the apex, angulating parallel to the
sagittal plane of the body, as the membranous septum
is not developed in this malformation.6,8, Second,
the left atrium should be visualized behind the mitral
valve to ensure that the ultrasound beam crosses the
anterior leaflet. When the ultrasound beam is directed
toward the left ventricle, the echo of the mitral ap-
paratus may not be that of the anterior leaflet, but that
of the chordae tendineae. As the chordae tendineae
are divided rightward and leftward and connect with
the papillary muscles, the cross-sectional echocar-
diogram apparently shows the presence of the mitral
cleft.
Echocardiographic Features of the Mitral Cleft
Diastolic separation of the anterior mitral leaflet
echo is thought to indicate the mitral cleft because this
finding was not detected in any patient without mitral
cleft.'6' 1' The degree of the cleft was suggested from
the extent of the cross section in which the diastolic
separation was observed. Angiocardiographically, the
1105
mitral cleft is diagnosed by the nonopaque notch in the
mitral area.68 This notch would correspond to the
knot echo formed by the ridges of the divided leaflet in
the cross-sectional echocardiogram.
The Echo of the Accessory Chordae
The thin linear echoes between the ridges of both
parts of the anterior leaflet and the ventricular septum
are believed to be the echo of the accessory chordae
because there was neither a false-positive nor a false-
negative case. To detect these echoes in some patients
required careful observation. However, the presence
of accessory chordae could not be detected by any
means except echocardiography.
In some patients, the accessory chordae seemed to
restrain the anterior mitral leaflet from being straight
during systole, causing it to bend toward the ventric-
ular septum. This condition was shown patho-
logically.20 The accessory chordae may prevent the
two parts of the anterior leaflet from prolapsing into
the left atrium at the site of the cleft.
Apex four-chamber echocardiography has been
recommended as being possibly superior for showing
the accessory chordae, but it does not show the mitral
cleft at the same time. Because it is necessary to show
the presence of the accessory chordae and their
relationship to the mitral cleft, we consider the sagittal
plane the superior view.
Mitral Regurgitation in Ostium Primum ASD
Mitral regurgitation in ostium primum ASD is
thought to be produced at the site of the cleft. In this
study, the degree of the mitral regurgitation was not
directly proportionate to the size of the cleft. In
patient 11, the mitral regurgitation was minimal even
though the cleft was complete; in patient 7, however,
 1106 CIRCULATION
TABLE 3. Possible Factors Causing Mitral Regurgitation in
Ostium Primum Atrial Septal Defect
1. Imperfect coaptation of the separated AML parts
1) Width of the cleft in systole
2) Absence of the accessory chordae
3) Other mitral apparatus
2. Imperfect coaptation of AML and PML
1) Presence of the accessory chordae
2) Mitral valve prolapse
3) Other mitral apparatus
Abbreviations: AML = anterior mitral leaflet; PML =
posterior mitral leaflet.
the mitral regurgitation was severe even though the
cleft was incomplete. Therefore, several possible fac-
tors causing regurgitation must be considered (table
3). In ostium primum ASD, there are two coaptation
parts in the mitral valve. First, regurgitation depends
upon the coaptation of the two parts of the anterior
mitral leaflet divided by the cleft. The width of the gap
between the two parts during systole is the prime fac-
tor. In patient 8, who had a complete cleft and a
relatively wide space between the two parts of the
anterior leaflet during systole, mitral regurgitation
was severe. The width of the cleft in diastole did not
correlate with the degree of mitral regurgitation.
Another factor is the absence of accessory chordae. If
they are absent, the edges of the cleft may slip out of
Downloaded from http://ahajournals.org by on September 27, 2019
the proper position for the coaptation of the two parts
of the anterior leaflet. In patients 3, 7 and 8, who did
not have the accessory chordae, the mitral regurgita-
tion was severe. In patient 7, the echo of the anterior
mitral leaflet was almost straight, as in the normal
leaflet. However, this finding may indicate that there
is no coaptation between two parts of the anterior
leaflet. This hypothesis was supported by the contrast
study, which showed that contrast echoes flowed into
the left atrium from the site of the cleft. Conversely,
even in the patients with a complete cleft, the
accessory chordae seemed to aid the coaptation
between the two parts of the anterior leaflet and
reduce mitral regurgitation (patients 1, 2, 5, 9 and 11).
Other factors, such as the site of the papillary muscle
or the deformity of the annulus, may cause imperfect
coaptation of the two parts of the anterior mitral
leaflet. In this study, no suggestive finding was ob-
tained concerning these factors.
Second, in ostium primum ASD, as well as in other
cardiac diseases, the coaptation between the anterior
and posterior leaflets seems to be related to the
regurgitation. While perhaps preventing the edges of
the cleft from prolapsing, as mentioned above, the
accessory chordae restrain the anterior leaflet from
being straight during systole.20 This condition may in-
terfere with the proper coaptation between the ante-
rior and posterior leaflets. If the mitral regurgitation
occurred only from the site of the cleft, it would be
corrected merely by suture of the cleft. However,
mitral regurgitation remains in some patients even
VOL 62, No 5, NOVEMBER 1980
after surgery.'-' In this study, the patients with systolic
bending of the anterior leaflet after surgery were
shown to have mitral regurgitation, and all but one pa-
tient without systolic bending had no mitral regurgita-
tion, even if the accessory chordae were not resected.
In addition, as in other cardiac diseases, mitral valve
prolapse may be the cause of the mitral regurgitation,
as in patient 4.
Conclusion
This study showed that real-time cross-sectional
echocardiography demonstrates not only the presence
of the mitral cleft, but also the configuration of the
anterior mitral leaflet and the presence of the acces-
sory chordae. This information has not been shown in
the beating heart. Not only the cleft but all these con-
ditions may relate to the mitral regurgitation in this
malformation. Although we did not study a large
number of patients, one explanation of the cause of
mitral regurgitation was proposed from the echo-
cardiographic viewpoint. Real-time cross-sectional
echocardiography should become indispensable in
assessing mitral deformities before surgery.
References
1. Braunwald NS, Morrow AG: Incomplete persistent atrioven-
tricular canal: operative methods and the results of pre- and
postoperative hemodynamic assessment. J Thorac Cardiovasc
Surg 51: 71, 1966
2. Griffiths SP, Ellis K, Burris JO, Blumenthal S, Bowman FO Jr,
Maim JR: Postoperative evaluation of mitral valve function in
ostium primum defect with cleft mitral valve (partial form of
atrioventricular canal). Circulation 40: 21, 1969
3. McMullan MH, McGoon DC, Wallace RB, Danielson GK,
Weidman WH: Surgical treatment of partial atrioventricular
canal. Arch Surg 107: 705, 1973
4. Hardesty RL, Zuberbuhler JR, Bahnson HT: Surgical treat-
ment of atrioventricular canal defect. Arch Surg 110: 1391,
1975
5. Losay J, Rosenthal A, Castaneda AR, Bernhard WH, Nadas
AS: Repair of atrial septal defect primum. Results, course, and
prognosis. J Thorac Cardiovasc Surg 75: 248, 1978
6. Baron MG, Wolf BS, Steinfeld L, Van Mierop LHS: Endocar-
diol cushion defect. Specific diagnosis by angiocardiography.
Am J Cardiol 13: 162, 1964
7. Somerville J, Lefferson K: Left ventricular angiocardiography
in atrioventricular defects. Br Heart J 30: 446, 1968
8. Baron M: Abnormalities of the mitral valve in endocardial
cushion defect. Circulation 45: 672, 1972
9. Gramiak R, Nanda N: Echocardiographic diagnosis of ostium
primum septal defect. (abstr) Circulation 46 (suppl II): 11-37,
1972
10. Williams RG, Rudd M: Echocardiographic features of en-
docardial cushion defects. Circulation 49: 418, 1974
11. Sahn DJ, Terry RW, O'Rourke R, Leopold G, Friedman WF:
Multiple crystal echocardiographic evaluation of endocardial
cushion defect. Circulation 50: 25, 1974
12. Pieroni DR, Homey E, Freedom RM: Echocardiography in
atrioventricular canal defect. A clinical spectrum. Am J Cardiol
35: 54, 1975
13. Beppu S, Nimura Y, Nagata S, Tamai M, Matsuo H, Mat-
sumoto M, Kawashima Y, Sakakibara H, Abe H: Diagnosis of
endocardial cushion defect with cross-sectional and M-mode
scanning echocardiography. Differentiation from secundum
atrial septal defect. Br Heart J 38: 911, 1976
14. Komatsu Y, Nagai Y, Shibuya M, Takao A, Hirosawa K:
Echocardiographic analysis of intracardiac anatomy in en-
docardial cushion defect. Am Heart J 91: 210, 1976
 DICROTIC PULSE/Orchard and Craige 1107

15. Dillon JC, Weyman AE, Feigenbaum H, Eggleton RC,
Johnston K: Cross-sectional echocardiographic examination of
the interatrial septum. Circulation 55: 115, 1977
16. Lieppe W, Scallion R, Behar VS, Kisslo JA: Two-dimensional
echocardiographic findings in atrial septal defect. Circulation
56: 447, 1977
17. Hagler DJ, Tajik AJ, Seward JB, Mair DD, Ritter DG: Real-
time wide-angle sector echocardiography: atrioventricular
canal defects. Circulation 59: 140, 1979
18. Wakai CS, Edwards JE: Pathologic study of persistent com-
mon atrioventricular canal. Am Heart J 56: 779, 1958
19. Sellers RD, Levy MJ, Amplatz K, Lillehei GW: Left retrograde
cardioangiography in acquired cardiac disease. Technic, in-
dications and interpretations in 700 cases. Am J Cardiol 14:
437, 1964
20. Edwards JE: The problem of mitral insufficiency caused by
accessory chordae tendineae in persistent common atrioven-
tricular canal. Mayo Clin Proc 35: 299, 1960

Dicrotic Pulse After Open Heart Surgery

ROBERT CHARLES ORCHARD, M.D., AND ERNEST CRAIGE, M.D.

SUMMARY Pre- and postoperative echophonocardiograms (EPCGs) and preoperative hemodynamic data
of 108 patients who underwent valve replacement were reviewed to establish the frequency and significance of a
dicrotic pulse (DP) postoperatively. DP occurred almost exclusively in patients who underwent valve replace-
ment for regurgitant lesions (20 of 28 with aortic regurgitation, nine of 25 with mitral regurgitation, and four
of six with both aortic and mitral regurgitation). These patients were divided into dicrotic and nondicrotic
groups. Preoperatively, the dicrotic group had significantly larger end-diastolic volumes (p < 0.01) and end-
systolic volumes (p < 0.01) and significantly lower ejection fractions (p < 0.01). Echocardiographically, the
dicrotic group had larger left ventricular dimensions, both systolic (p < 0.01) and diastolic (p < 0.05), reduced
percentage fractional shortening of the left ventricular cavity (p < 0.01) and poor thickening properties of the
left ventricular posterior wall (%ATh-LVPW) (p < 0.01). Postoperatively the dicrotic group had a slightly
larger end-diastolic dimension (p = NS) and markedly depressed %ATh-LVPW (p < 0.001) compared with
the nondicrotic group.
On follow-up EPCG the persistence of a DP correlated with continued left ventricular dysfunction by
echocardiographic and hemodynamic studies and an extremely poor clinical course. DP after valve replace-
ment is therefore an important prognostic sign.
Downloaded from http://ahajournals.org by on September 27, 2019

A DICROTIC PULSE is characterized by two
palpable pulsations (fig. 1). As opposed to anacrotic
and bisferiens pulses, in which the two pulsations are
systolic in timing, the second pulsation of the dicrotic
pulse is diastolic, immediately after the dicrotic notch.
The dicrotic pulse has long been recognized, par-
ticularly as a feature of febrile states such as typhoid
fever.' More recently the dicrotic pulse has been
associated with a variety of cardiac diseases that have
in common a low cardiac output. These include car-
diomyopathies,2' pericardial tamponade,2' constric-
tive pericarditis2 and pulmonary embolism.2 There is
only one report, however, of dicrotism after open
heart surgery, and the significance of this finding is un-
known.4 At our institution, pre- and postoperative
echophonocardiograms are routinely performed on
From the Department of Medicine and the C.V. Richardson
Laboratory, University of North Carolina School of Medicine,
Chapel Hill, North Carolina.
Dr. Orchard is a Research Fellow in the Department of Medi-
cine. His present address: University Hospital, Saskatoon,
Saskatchewan, Canada.
Dr. Craige is the Henry A. Foscue Distinguished Professor of
Cardiology.
Address for correspondence: Ernest Craige, M.D., 338 Clinical
Sciences Building 229H, UNC School of Medicine, Chapel Hill,
North Carolina 27514.
Received April 23, 1979; revision accepted May 9, 1980.
Circulation 62, No. 5, 1980.
most adult patients who undergo cardiac surgery, so
we had the opportunity to observe the frequent,
although not uniform, presence of a dicrotic pulse
after open heart surgery. The present study was under-
taken to establish the frequency and significance of
this clinical sign in patients who undergo cardiac sur-
gery for valvular disease.
Methods
Subjects
The patients were identified by a computerized data
retrieval system as having undergone pre- and post-
operative echophonocardiograms for valvular surgery
between January 1973 and December 1978. The
patients studied were all age 45 years or younger, and
any patient who developed prosthetic malfunction was
excluded. After technically inadequate carotid pulse
tracings were excluded, 108 patients remained.
Postoperative tracings were performed 7-10 days
after surgery.
Echophonocardiography
Phonocardiograms were obtained from the four
standard precordial areas using either an Irex 101 or a
Cambridge MC IV multichannel recorder and
Leatham suction microphones. Indirect carotid artery
pulse recordings were made with an air-filled funnel