- Email: [email protected]
Phonocardiography in interventricular septal defects
Phonocardiography
in Interventricular
Ernest
Chapel Hill,
Craige, M.D.,
Septal
Defects
N. C.
Physiologic studies in patients with interventricular septal defects have, in recent years, demonstrated a whole spectrum of abnormalities.‘z2 At one end, there are the small left-to-right shunts, with pressures and resistances in the pulmonary and systemic circuits remaining normal. The magnitude of the shunt is regulated in these cases by the size of the opening and the pressure gradient across it. In the larger defects, however, the anatomic communication presents less of an obstruction. Here the magnitude of the left-to-right shunt is controlled by the relative resistance of the pulmonary and systemic arterioles.3 There may be very large shunts, several times the systemic flow, in cases in which pulmonary resistance is low. The pulmonary resistance, however, may reach such a magnitude that the left-to-right shunt will be much reduced in volume, and a right-to-left shunt may appear. The classic physical findings are altered with these modifications in flows and pressures. Recently, phonocardiographic studies have been utilized in the assessment of these patients who are now often under consideration for surgical repair of the defect.4 The present study demonstrates the relationship between the phonocardiographic findings and the physiologic abnormalities as disclosed by cardiac catheterizat.ion. MATI~RIXIS
AND
METHODS
Thirty-five patients with interventricular septal defects demonstrated at cardiac catheterization were studied. They ranged in age from 4 to 34 years; the average age was 13 years. Patients with additional abnormalities, such as pulmonary stenosis, were not included. Cardiac catheterization was performed in the standard manner, using Statham P23D strain gauges for pressure transducers, and a four-channel Sanborn Poly-Viso apparatus for pressure recording. In the younger children, the collection of air samples for the determination of oxygen consumption was impractical. The volume of the shunt, therefore, is expressed, not in cubic centimeters, but as a proportion of the systemic blood flow. For example, simple left-to-right shunts are expressed as a fraction of systemic flow: Y
=
Y SBF
PBF
-
SBFS
-
’
PBF =SE
From the Department of Medicine. University of North N. C. This investigation was supported by Grant No. HT-5033 Health Service. Received for publication Nov. 30, 1959.
Carolina
School
of Medicine,
Chapel
Hill.
51
(C4
and
C5),
United
States
Public
52
Am.
CKAIGE
Heart J. July, 1960
where Y = left-to-right shunt, PBF = pulmonary blood flow, and SBF = systemic blood flow. In order to express the left-to-right shunt in cases in which two-way shunting occurs, the following formula is used: Y = PBF - EPBF Y -=-SBF and
for the right-to-left
EPBF
SBF
SBF
shunt: Z Z -=I-___ SBF
where EPBF = effective a quantitative expression oxygen consumption.
PBF
pulmonary of the
=
SBF
-
EPBF EPBF SBF
blood flow, and 2 = right-to-left shunts is made possible without
shunt. By use of these the use of determinations
ratios of
The patients were divided according to measurements of pressures in the pulmonary artery, and blood flow, into three groups. Group I: Systolic pressures in the pulmonary artery of 4.5 mm. Hg or below, with a shunt of small to moderate size (up to 2.0 times systemic flow). Group II: Systolic pressures moderately elevated in the pulmonary artery, but with large left-to-right shunts. The hypertension here was believed to be of the hyperkinetic variety, because there was no evidence of increased pulmonary arteriolar resistance. Group III: Pulmonary hypertension of severe degree. These patients had shunts of smaller size, with often a right-to-left, as well as a left-to-right, component of the shunt. Here the pulmonary hypertension was attributed to increased pulmonary resistance. Phonocardiograms were made on all patients in the recumbent position by means of a Sanbcrn Twin-Beam oscillograph, at paper speeds of 75 mm. per second. Logarithmic recordings were taken in all instances. The bell endpiece was usually employed, although the diaphragm was used to pick up higher frequency sounds or murmurs. The heart sounds were recorded at the apex, and in the fourth, third, and second left intercostal spaces, as well as in the second right intercostal space. Lead II of a simultaneous electrocardiogram was recorded for timing purposes. An estimation of the amplitude of heart sounds was made on a grading scale of increasing intensity from l+ to 4+, with 1+ meaning a sound of diminished intensity; 2+, normal; 3+ and 4+, increased amplitude. These phonocardiographic phenomena are difficult to quantitate in a more precise manner, because of differences in thickness of the chest wall, as well as varying frequencies of the sounds being recorded. Since the details of the second sound were often obscured in the location of maximum intensity of the murmur, the details of timing of the components of the second sound were best studied in other localities, such as the pulmonic area. The figures given for splitting of the second sound were obtained during the expiratory phase of respiration. For controls, a group cf 23 normal students were examined; they ranged in age from 10 to 19 years. Splitting of the second sound was found to be present in them, varying with respiration but averaging 0.027 second during expiration, and having a range of 0.02 to 0.04 second. The ejection sound in early systole was best seen in tracings made in the second and third left intercostal spaces. It was recorded as a brief sound, lasting only 0.01 to 0.02 second. Its time was noted from onset of the QRS complex of the electrocardiogram (Q-ejection sound). The time of appearance of the first rapid vibrations of the first sound, indicating closure of the A-V valves, was also noted and is referred to in Table I as the Q-l time.
RESULTS
Group I.-Cases l-22. The patients with low systolic pressures in the pulmonary artery had relatively small left-to-right shunts (Table I). There were 22 in this group; the average age was 10 years. The second sound was not increased in amplitude in 20 out of 22 of these individuals. It was normally split (0.02
Voiume 60 Number
1
PHONOCARDIOGRAPHY
IN
INTERVENTRICULAR
SEPTAL
DEFECTS
5.4
4th L.I.C.S.
Fig. I.-Case 14. The pansystolic murmur is maximal in the fourth it has a crescendo-decrescendo con5guration. This was the most common and 1s designated Type A in Table I.
left intercostal space, where murmur in Groups I and 1 I
2nd L.I.CS
4 th L.I.C.S.
Fig. P.-Case
10. The pansystolic murmur is again maximal in the fourth constant intensity. Type B in Table I.
left intercostal
space. hut is of
54
CRAIGE
Am. Heart J. July, 1960
to 0.04 second; average 0.03 second) in 15 of the 22 in this group. Variation in the degree of separation of the two components was frequently noted with respiration. However, in 7 patients, splitting was noted up to 0.05, 0.06, and 0.07 second, persisting in expiration, although the electrocardiogram was normal, with no evidence of conduction delay, and flow through the shunt was relativei) small. A third sound was noted in the third and fourth left intercostal spaces in 5 of these patients. Its timing was that of a normal third heart sound, and it was considered to be a normal finding in these youthful subjects. One patient in this group had an apical low-frequency diastolic murmur. Its timing was early to middiastolic, and corresponded to the period of rapid filling of the ventricle. Ejection sounds, or systolic clicks, were not found in this group, with but one exception. An intense pansystolic murmur was noted in all the patients in Group I (Fig. 1). The murmur was of high frequency and began immediately after the first sound. It extended to, and often slightly beyond, the aortic component of the second sound. The location of maximal intensity of the murmur was in the third or fourth left intercostal space parasternally. In 5 patients, however, the murmur was equally intense in the second left intercostal space. The distinct peaking in the intensity of the murmur in mid-systole produced a diamond shape on the phonocardiogram in 14 cases. In 7 others, however, the murmur had a much more constant intensity and resulted in what might better be described as a cigar shape on the tracing (Fig. 2). In 1 case an atypical crescendo murmur, which reached a peak just before the second sound, was noted. In the area of maximal intensity of the murmur, details r,i the second sound were often obscured. Group II.-Cases 23-27. The 5 patients in this group were characterized by moderate to severe pulmonary hypertension. Their average age was 17 years. Their left-to-right shunts were very large (4 to 6 times systemic flow), and no right-to-left shunts were noted. The second sound was normally to widely split in these patients, with the two components 0.02 to 0.05 second apart. Respiratory variation was not present in those with the widest splitting. In 4 of the 5 patients in this group, third sounds were noted in the apical tracing. In 1 of them this was followed by an apical low-frequency diastolic murmur of brief duration. In 3 of the patients in Group II an early ejection sound or click was noted. The systolic murmur was similar to that described in the patients of Group I ; it was pansystolic and of high frequency and intensity. Its configuration was also diamond shaped, being crescendo-decresendo (Fig. 3). There was one exception, Case 27, in which the murmur was brief and of small intensity, being, therefore, of the type seen more typically in the patients of Group III. Group III.-Cases 28-35. The 8 patients in this group had pulmonary hypertension of severe degree. Their average age was 18 years. Small to moderate left-to-right shunts up to 2 times the systemic blood flow were present. In 7, there was an associated right-to-left shunt of from 0.13 to 0.5 times the systemic blood flow.
PHONOCARDIOGRAPHY
IN
INTERVENTRI(‘~~LAR
SEYTAL
.i.i
DEFECTS
In 2 of these patients the second sound was narrowly split (0.03 second:, but : in the remaining 6, no splitting could be detected. In 7 of the 8 in this group in intensity (3+ and 4+ 1. the second sound was moderately to greatlyaccentuated
Hig. 3.--Case
24. Intense
pansystolic
Fig. 4.--Case
32. The systolic sound
murmur A third
of Type A, 8een in presence sound is present.
of a large left-to-right,Ishunt.
murmur is barely visible (Type C in Table I). There (X) and au accentuated single second sound.
is a systolir
ejwrion
23/10
45/12
20/2
18/O
4415
25/2
45/4
14
8
6
5
15
12
4
6 4
9
13
3.5
7
17
5
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
16.
17.
18.
35/10
24/12
25/3
43/8
18/10 30110
40/10
18/10
20/s
20/7
36/3
43/o
23/3 30/o
23/2
30/o
22/10
22/3
23/O-5
23/8
38/O-3
30/10
23/2
30/o
5
2.
30/12
33/o
23/10
23/O
9
--
;
PA. ‘RESSURE IMM. Erg)
10
_-
(MM.
AGE
(:YR.)
wd
R.V. PRESSURE
1.
CASE NUMBER
-
.-
1.4x
1.3x
1.5x
2.0x
1.5x
0.2x 0.66X
0.25x
0.25x
1.0x
0.5x
1.5x
2.0x
2.0x
0.75x
0.7.5x
2.0x
0.2x
1LEFT-TORIGHT SHUNT*
0
0
0
0
0
0 0
0
0
0
0
0
0
0
0
0
0
0
1ZIGHT-T( LEFT SHUNT*
2nd and
B
A
A
A
A
A
LICS
LICS
LICS
4th
LICS
LICS
3rd LICS
2nd and 3rd
4th
4th
3rd and
4th
3rd LICS 4th LICS
4th LICS
B
3rd LICS
3rd LICS
B B
4th LICS
A
3rd LICS
A
3rd LICS
2nd, 3rd, and 4th LICS
2nd and
LICS
B
A
2nd and 3rd
LICS
LICS
A
3rd
MAXIMAL LOCATION
MURMUR
3rd
A
I
A
1-4)
(SEE
Crescendo B
TYPB FIGS.
SYSTOLIC
TABLE
sound
0
0 sound
0
0
sound
0
sound
0
rumble
sound
Third
Third
sound
sound
Aortic diastolic murmur
Third
Third
Third
Third
Diastolic
0
0
0
DIASTOLIC PHENOMENA
‘rhird
_-
I --
zz
0.04 0.03 0.05 0.03 0.03 0.06
2+ 2+ 2+ 2+ 2+ 2+ 2+
0 0
0
0
0
0
0
ii
0
0
0
0
0
0
0
2+
2+
2+
2+
2+
2+
3+
2+
2+
0.04
2+
0
7
SPLITTING (SEC.)
0.05
--
INTENj I T Y (+
SOUND
2+
0
EJECTION SOUNDt
SECOND
r: $ 5 E
65/S
90/O-S
100/o-3
90/o-10
6
10
12
15
30
9
17
30
22
31
21
12
13 4
4
21.
22.
23.
24.
25.
26.
21.
28.
29.
30.
31.
.32.
33. 34.
35.
80/40
-3
1 S/75
125/65
110/25
125/60
120/45
90/40
100/60
90/25
65/23
SO/20
23/8
25/6
25/10
25/9
0.5x
0.25x 2.0x
2.0x
0.16X
0.4x
0.25x
2.0x
4.0X
6.0X
4.0x
4.0x
4.0X
0.2X
1.4x
0.5x
0.3x LICS
LICS
LICS
LICS
I 2nd LICS
0.13x
LICS
LICS
LICS
LICS
2nd LICS 2nd LICS
3rd
4th
4th
3rd
3rd LICS
3rd and 4th
4th
3rd LICS
4th
3rd LICS
3rd LICS
3rd LICS
3rd
4th
0.15x 0.2x
0.2X
0.5x
0.4x
0.25x
0
0
0
0
0
0
0
0
0
0
LICS
1
1
0
0
0
0
sound
sound
sound
0
0
0
0
and
/ +0.05/0.12
1 +0.06/0.12 ; +0.07/0.13
+0.06/0.13
+0.07/0.12
+O.OS/O.lZ
+0.07/0.12
+0.06/0.10
+0.06/0.16
+0.07/0.10
+0.07/0.11
0
0
+0.06/O.
0
0
0
over the Q-ejection was vonsist,ent with
0
Fi
0
Pulmonary diastolic murmur
rhird
Third
rhird
‘hidsomd
*Magnitude of shunt is expressed as a proportion of systemic flow (see text for explanation). tThe presenre of a wstolic ejecLion sound is indicated by +. The figures refer Lo t,he Q-l t,ime iin seconds) $‘l’he pulmonary art,ery was not entered. but the identity of right ventricular and radial arterial pressures necn the ventricles. lThe pulmonary artery was noti entered, but the diagnosis was confirmed at autopsy. “The intensity of wronrl so~mtl is ~rprwsed as f C diminished: 2-C normal: 3 f and +I-. arcentuatrd.
80/S
15010-5 112/O-S
135/o-15
12.5/s-10
100/o-5
125/O-S
120/O-S
58/O-6
23/O
25/o
25/3
16
20.
25/o
11
19.
sound a free
10
i
0.04 0.02
2+ 2+
0
-I+
t,ime (in seconds). communication br-
0.03
0
0.03
2+
3+
4f
0.02
0.04
2+
3+
0.05
0.03
2+ +I-
0.05
2+
58
CRAIGE
Am. Heart J. July, 1960
A systolic click or ejection sound was recorded early in systole in all of these patients with severe pulmonary hypertension. It was best seen in tracings taken in the second and third left intercostal spaces. The time of the systolic click averaged 0.12 second from the onset of the QRS (range, 0.10 to 0.13 second), The time from onset of the QRS to the beginning of the main valvular components of the first sound averaged 0.07 second (range, 0.05 to 0.08 second). A third sound was not seen in any of these patients with severe pulmonary hypertension. No apical diastolic murmur was recorded in any of this group. One, however, had a decrescendo diastolic murmur in the pulmonary area, attributed to pulmonary regurgitation. The systolic murmur was atypical in all of the patients of Group III (see Table I and Fig. 4). It was abbreviated in duration, being completed during approximately the first third of systole, in contrast to the pansystolic duration of the murmurs in the patients of Groups I and II. The intensity of the murmur was very small, again in striking contrast to those in patients of Groups I and II. DISCUSSION
The systolic murmur in ventricular septal defect results from turbulence due to the jet of blood passing through the defect. It is not surprising, therefore, that modifications in the murmur occur with heightened resistance in the pulmonary circuit, which diminishes the left-to-right flow. This effect is seen in the contrast in murmurs between the patients of Group III and those of the groups with lower pressures. The soft and brief systolic murmur along the upper left sternal border in these patients with similar pressures in the two ventricles has been attributed by Leathamc and Wood7 to ejection of blood into the dilated pulmonary artery, rather than to flow through the septal defect itself. In patients with large flows, there is increased filling of the left ventricle in early diastole. This, in turn, is usually associated with a third sound or middiastolic murmur at the apex in the patients of Group II, and some of the patients of Group I. Whether the apical rumbling murmur was associated with a gradient of pressure across the mitral valve could not be determined by the methods employed here. It was believed to be a flow murmur resulting from an increased return from the lungs to the left side of the heart.6 Because of the degree of pulmonary hypertension which existed in some of the patients of Group III, a blowing diastolic murmur of pulmonary regurgitation might have been expected. This was noted, however, in only one instance. The tendency to wide splitting of the second sound in some of the patients of Groups I and II is attributed to two factors: (1) prolongation of right ventricular systolic ejection, associated with an increased stroke volume of the right ventricle, which might delay pulmonary closure; and (2) a shortened left ventricular systolic ejection, with early closure of the aortic valve because of the existence of an alternate route of ejection from the left ventricle in addition to the aorta. Extreme elevations in pulmonary pressure were reflected in narrowing and accentuation of the second sound. This was due to an abbreviation of right ventricular systolic ejection, as has been shown to be present in pulmonary hypertension from various
ptr&~ ”
“10
PHONOCARDIOGRAPHY
IN
INTERVENTRICULAR
SEPTAL
59
DEFECTS
causes. In addition, an early systolic click or ejection sound appeared in the patients of Group III. The finding of such a sound has been well described lq I,eatham and Vogelpoels in cases of pulmonary hypertension, and has also been demonstrated in a variety of conditions resulting in dilatation of the aorta or pulmonary artery.$ Its timing is such as to coincide with the beginning of the period of maximum ejection of blood into the dilated and probably already tense
pulmonary
artery.
This clicking sound is easily confused with the valvul;tr
components of the first sound. The latter, however, are best heard---ancl recorded---over the apex or tricuspid areas, whereas the systolic click in the patients of our Group III was recorded over the second and third left intercostal spaces, 0.05 second after the initial valvular contributions to the first sound. Although there are some exceptions, as noted in Table I, the differences in phonocardiographic findings in the three groups can be summarized simply in Table I I. 'GABLE
II.
SUMMARY
0I; PHONOCARDIOGKAPHIC SEPTAL
FINDINGS DEFECT
IN 35
CASES
.--. SYSTOLIC
Group I
MUKMUR
I
Loud; pansystolic; crescendo-decrescendo or of constant intensitv. Maximal in third and fourth LICS
DIASTOLIC PHENOMENA
SECOND
01; INTEKVEKTRICvL.211
-. _~~ SOUND
~. ~~ EJKCTIOS SOlYlT
Zero; or normal third sound in children
Normally to widely split. UsualI> variable with respiration. ?jot accentuated
Absent
Third sound or mid-diastolic murmur
Normally to widely split, but not variable with respiration when widely split
Sometimes present
Tylp;,“, y B as in Figs. Group II
Loud ; pansystolic; crescendo-decrescendo. Maximal in third and fourth LICA Type A as in Fig. 3
Group III
Short, soft, early systolic; maximal in second, third, or fourth LICS Type C as in Fig. 4
Occasionally, pulmonary diastolic murmur
i Narrowly split or I single. .\ccentu,
~ Present
nted
i
It is of interest that, as Brotmacher and Campbelllo have pointed out, the uncomplicated ventricular septal defects with normal pressures tend to be found in patients in the first two decades of life, whereas the cyanotic types are found in patients more evenly spread through the first four decades. In our series, the average age of the patients of Group I was 10 years, and the average age of those in Groups II and III was 17 and 18 years, respectively, a finding in agreement with the data of these authors. We do not feel, however, that there is necessarily a tendency for patients to progress from one group to another, since in those patients who have been recatheterized, physiologic findings were unchanged.
60
CRAIGE
SUMMARY
AND
.41x1. Heart J. July, 1960
CONCLUSIONS
Thirty-five patients with interventricular septal defects were studied by means of cardiac catheterization and phonocardiography. They were divided into three groups on the basis of pressures and flows in the pulmonary artery. Group I were those with low pressures and flows who had harsh, pansystolic murmurs, either crescendo-decrescendo or of constant intensity, with normally to widely split second sounds. Group II were those with moderate to severe pulmonary hypertension and large left-to-right shunts who also had loud pansystolic murmurs, crescendo-decrescendo in configuration, and second sounds that were normally to widely split. Apical third sounds or diastolic murmurs were seen in this group and were attributed to rapid filling of the left ventricle. This transitional group shared some of the features of the other two types, since the systolic murmurs were similar to those in the patients with smaller shunts, whereas ejection sounds were often present, as in the patients with high pressures and mixed shunts. Group III were the patients with severe pulmonary hypertension and smaller or mixed shunts who had atypical systolic murmurs-brief and subdued. The second sound was accentuated and often single, and an ejection sound in early systole was uniformly present over the pulmonary area. It is believed that the phonocardiogram adds significantly to one’s ability to estimate the state of the physiologic disturbance in patients with interventricular septal defect. The cardiac catheterizations were performed S. Harned, Jr., Department of Pediatrics, and Dr. Department of Medicine, North Carolina Memorial
by Dr. Nelson K. Ordway and Dr. Herbert Daniel T. Young and Dr. Thomas C. Gibson, Hospital, Chapel Hill, N. C.
REFERENCES
Blount,
2. 3. 4. 5. 6. 7.
8. 9. IO.
S. G., Jr., Mueller, H., and McCord, M. C.: Ventricular Septal Defect: Clinical and Hemodynamic Patterns, Am. J. Med. 18:871, 1955. Shepherd, J. T.: The Pulmonary Circulation in the Presence of Interatrial, Interventricular and Interarterial Communications. In The Pulmonary Circulation, edited by W. R. Adams and I. Veith, New York, 1959, Grune &Stratton, Inc., p. 204. Edwards, J. E.: Functronal Pathology of the Pulmonary Vascular Tree in Congenital Cardiac Disease, Circulation 15:164, 1957. Auscultatory and Phonocardiographic Findings in Ventricular Septal DeLessof, M. H.: fect (Abstract), Circulation 18:748, 1958. Luisada, A. A., and Liu, C. K.: Cardiac Pressures and Pulses, New York, 1956, Grune & Stratton, Inc., p. 101. Leatham, A. : Auscultation of the Heart, Pediat. Clin. North America 5:839, 1958. Wood, P.: McDonald, L., and Emanuel, R.: The Clinical Picture Correlated With PhysioIogrcal Observations in the Diagnosis of Congenital Heart Disease, Pediat. Clin. North America 5:981, 1958. Leatham, A., and Vogelpoel, L.: The Early Systolic Sound in Dilatation of Pulmonary Artery, Brit. Heart J. 16:21, 1954. Minhas, K., and Gasul, B. M.: Systolic Clicks: A Clinical, Phonocardiographic, and Hemodynamic Evaluation, AM. HEART J. 57:49, 1959. History of Interventricular Septal Defect, Brotmacher, L., and Campbell, M. : The Natural Brit. Heart J. 20:97, 1958.
in Interventricular
Ernest
Chapel Hill,
Craige, M.D.,
Septal
Defects
N. C.
Physiologic studies in patients with interventricular septal defects have, in recent years, demonstrated a whole spectrum of abnormalities.‘z2 At one end, there are the small left-to-right shunts, with pressures and resistances in the pulmonary and systemic circuits remaining normal. The magnitude of the shunt is regulated in these cases by the size of the opening and the pressure gradient across it. In the larger defects, however, the anatomic communication presents less of an obstruction. Here the magnitude of the left-to-right shunt is controlled by the relative resistance of the pulmonary and systemic arterioles.3 There may be very large shunts, several times the systemic flow, in cases in which pulmonary resistance is low. The pulmonary resistance, however, may reach such a magnitude that the left-to-right shunt will be much reduced in volume, and a right-to-left shunt may appear. The classic physical findings are altered with these modifications in flows and pressures. Recently, phonocardiographic studies have been utilized in the assessment of these patients who are now often under consideration for surgical repair of the defect.4 The present study demonstrates the relationship between the phonocardiographic findings and the physiologic abnormalities as disclosed by cardiac catheterizat.ion. MATI~RIXIS
AND
METHODS
Thirty-five patients with interventricular septal defects demonstrated at cardiac catheterization were studied. They ranged in age from 4 to 34 years; the average age was 13 years. Patients with additional abnormalities, such as pulmonary stenosis, were not included. Cardiac catheterization was performed in the standard manner, using Statham P23D strain gauges for pressure transducers, and a four-channel Sanborn Poly-Viso apparatus for pressure recording. In the younger children, the collection of air samples for the determination of oxygen consumption was impractical. The volume of the shunt, therefore, is expressed, not in cubic centimeters, but as a proportion of the systemic blood flow. For example, simple left-to-right shunts are expressed as a fraction of systemic flow: Y
=
Y SBF
PBF
-
SBFS
-
’
PBF =SE
From the Department of Medicine. University of North N. C. This investigation was supported by Grant No. HT-5033 Health Service. Received for publication Nov. 30, 1959.
Carolina
School
of Medicine,
Chapel
Hill.
51
(C4
and
C5),
United
States
Public
52
Am.
CKAIGE
Heart J. July, 1960
where Y = left-to-right shunt, PBF = pulmonary blood flow, and SBF = systemic blood flow. In order to express the left-to-right shunt in cases in which two-way shunting occurs, the following formula is used: Y = PBF - EPBF Y -=-SBF and
for the right-to-left
EPBF
SBF
SBF
shunt: Z Z -=I-___ SBF
where EPBF = effective a quantitative expression oxygen consumption.
PBF
pulmonary of the
=
SBF
-
EPBF EPBF SBF
blood flow, and 2 = right-to-left shunts is made possible without
shunt. By use of these the use of determinations
ratios of
The patients were divided according to measurements of pressures in the pulmonary artery, and blood flow, into three groups. Group I: Systolic pressures in the pulmonary artery of 4.5 mm. Hg or below, with a shunt of small to moderate size (up to 2.0 times systemic flow). Group II: Systolic pressures moderately elevated in the pulmonary artery, but with large left-to-right shunts. The hypertension here was believed to be of the hyperkinetic variety, because there was no evidence of increased pulmonary arteriolar resistance. Group III: Pulmonary hypertension of severe degree. These patients had shunts of smaller size, with often a right-to-left, as well as a left-to-right, component of the shunt. Here the pulmonary hypertension was attributed to increased pulmonary resistance. Phonocardiograms were made on all patients in the recumbent position by means of a Sanbcrn Twin-Beam oscillograph, at paper speeds of 75 mm. per second. Logarithmic recordings were taken in all instances. The bell endpiece was usually employed, although the diaphragm was used to pick up higher frequency sounds or murmurs. The heart sounds were recorded at the apex, and in the fourth, third, and second left intercostal spaces, as well as in the second right intercostal space. Lead II of a simultaneous electrocardiogram was recorded for timing purposes. An estimation of the amplitude of heart sounds was made on a grading scale of increasing intensity from l+ to 4+, with 1+ meaning a sound of diminished intensity; 2+, normal; 3+ and 4+, increased amplitude. These phonocardiographic phenomena are difficult to quantitate in a more precise manner, because of differences in thickness of the chest wall, as well as varying frequencies of the sounds being recorded. Since the details of the second sound were often obscured in the location of maximum intensity of the murmur, the details of timing of the components of the second sound were best studied in other localities, such as the pulmonic area. The figures given for splitting of the second sound were obtained during the expiratory phase of respiration. For controls, a group cf 23 normal students were examined; they ranged in age from 10 to 19 years. Splitting of the second sound was found to be present in them, varying with respiration but averaging 0.027 second during expiration, and having a range of 0.02 to 0.04 second. The ejection sound in early systole was best seen in tracings made in the second and third left intercostal spaces. It was recorded as a brief sound, lasting only 0.01 to 0.02 second. Its time was noted from onset of the QRS complex of the electrocardiogram (Q-ejection sound). The time of appearance of the first rapid vibrations of the first sound, indicating closure of the A-V valves, was also noted and is referred to in Table I as the Q-l time.
RESULTS
Group I.-Cases l-22. The patients with low systolic pressures in the pulmonary artery had relatively small left-to-right shunts (Table I). There were 22 in this group; the average age was 10 years. The second sound was not increased in amplitude in 20 out of 22 of these individuals. It was normally split (0.02
Voiume 60 Number
1
PHONOCARDIOGRAPHY
IN
INTERVENTRICULAR
SEPTAL
DEFECTS
5.4
4th L.I.C.S.
Fig. I.-Case 14. The pansystolic murmur is maximal in the fourth it has a crescendo-decrescendo con5guration. This was the most common and 1s designated Type A in Table I.
left intercostal space, where murmur in Groups I and 1 I
2nd L.I.CS
4 th L.I.C.S.
Fig. P.-Case
10. The pansystolic murmur is again maximal in the fourth constant intensity. Type B in Table I.
left intercostal
space. hut is of
54
CRAIGE
Am. Heart J. July, 1960
to 0.04 second; average 0.03 second) in 15 of the 22 in this group. Variation in the degree of separation of the two components was frequently noted with respiration. However, in 7 patients, splitting was noted up to 0.05, 0.06, and 0.07 second, persisting in expiration, although the electrocardiogram was normal, with no evidence of conduction delay, and flow through the shunt was relativei) small. A third sound was noted in the third and fourth left intercostal spaces in 5 of these patients. Its timing was that of a normal third heart sound, and it was considered to be a normal finding in these youthful subjects. One patient in this group had an apical low-frequency diastolic murmur. Its timing was early to middiastolic, and corresponded to the period of rapid filling of the ventricle. Ejection sounds, or systolic clicks, were not found in this group, with but one exception. An intense pansystolic murmur was noted in all the patients in Group I (Fig. 1). The murmur was of high frequency and began immediately after the first sound. It extended to, and often slightly beyond, the aortic component of the second sound. The location of maximal intensity of the murmur was in the third or fourth left intercostal space parasternally. In 5 patients, however, the murmur was equally intense in the second left intercostal space. The distinct peaking in the intensity of the murmur in mid-systole produced a diamond shape on the phonocardiogram in 14 cases. In 7 others, however, the murmur had a much more constant intensity and resulted in what might better be described as a cigar shape on the tracing (Fig. 2). In 1 case an atypical crescendo murmur, which reached a peak just before the second sound, was noted. In the area of maximal intensity of the murmur, details r,i the second sound were often obscured. Group II.-Cases 23-27. The 5 patients in this group were characterized by moderate to severe pulmonary hypertension. Their average age was 17 years. Their left-to-right shunts were very large (4 to 6 times systemic flow), and no right-to-left shunts were noted. The second sound was normally to widely split in these patients, with the two components 0.02 to 0.05 second apart. Respiratory variation was not present in those with the widest splitting. In 4 of the 5 patients in this group, third sounds were noted in the apical tracing. In 1 of them this was followed by an apical low-frequency diastolic murmur of brief duration. In 3 of the patients in Group II an early ejection sound or click was noted. The systolic murmur was similar to that described in the patients of Group I ; it was pansystolic and of high frequency and intensity. Its configuration was also diamond shaped, being crescendo-decresendo (Fig. 3). There was one exception, Case 27, in which the murmur was brief and of small intensity, being, therefore, of the type seen more typically in the patients of Group III. Group III.-Cases 28-35. The 8 patients in this group had pulmonary hypertension of severe degree. Their average age was 18 years. Small to moderate left-to-right shunts up to 2 times the systemic blood flow were present. In 7, there was an associated right-to-left shunt of from 0.13 to 0.5 times the systemic blood flow.
PHONOCARDIOGRAPHY
IN
INTERVENTRI(‘~~LAR
SEYTAL
.i.i
DEFECTS
In 2 of these patients the second sound was narrowly split (0.03 second:, but : in the remaining 6, no splitting could be detected. In 7 of the 8 in this group in intensity (3+ and 4+ 1. the second sound was moderately to greatlyaccentuated
Hig. 3.--Case
24. Intense
pansystolic
Fig. 4.--Case
32. The systolic sound
murmur A third
of Type A, 8een in presence sound is present.
of a large left-to-right,Ishunt.
murmur is barely visible (Type C in Table I). There (X) and au accentuated single second sound.
is a systolir
ejwrion
23/10
45/12
20/2
18/O
4415
25/2
45/4
14
8
6
5
15
12
4
6 4
9
13
3.5
7
17
5
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
16.
17.
18.
35/10
24/12
25/3
43/8
18/10 30110
40/10
18/10
20/s
20/7
36/3
43/o
23/3 30/o
23/2
30/o
22/10
22/3
23/O-5
23/8
38/O-3
30/10
23/2
30/o
5
2.
30/12
33/o
23/10
23/O
9
--
;
PA. ‘RESSURE IMM. Erg)
10
_-
(MM.
AGE
(:YR.)
wd
R.V. PRESSURE
1.
CASE NUMBER
-
.-
1.4x
1.3x
1.5x
2.0x
1.5x
0.2x 0.66X
0.25x
0.25x
1.0x
0.5x
1.5x
2.0x
2.0x
0.75x
0.7.5x
2.0x
0.2x
1LEFT-TORIGHT SHUNT*
0
0
0
0
0
0 0
0
0
0
0
0
0
0
0
0
0
0
1ZIGHT-T( LEFT SHUNT*
2nd and
B
A
A
A
A
A
LICS
LICS
LICS
4th
LICS
LICS
3rd LICS
2nd and 3rd
4th
4th
3rd and
4th
3rd LICS 4th LICS
4th LICS
B
3rd LICS
3rd LICS
B B
4th LICS
A
3rd LICS
A
3rd LICS
2nd, 3rd, and 4th LICS
2nd and
LICS
B
A
2nd and 3rd
LICS
LICS
A
3rd
MAXIMAL LOCATION
MURMUR
3rd
A
I
A
1-4)
(SEE
Crescendo B
TYPB FIGS.
SYSTOLIC
TABLE
sound
0
0 sound
0
0
sound
0
sound
0
rumble
sound
Third
Third
sound
sound
Aortic diastolic murmur
Third
Third
Third
Third
Diastolic
0
0
0
DIASTOLIC PHENOMENA
‘rhird
_-
I --
zz
0.04 0.03 0.05 0.03 0.03 0.06
2+ 2+ 2+ 2+ 2+ 2+ 2+
0 0
0
0
0
0
0
ii
0
0
0
0
0
0
0
2+
2+
2+
2+
2+
2+
3+
2+
2+
0.04
2+
0
7
SPLITTING (SEC.)
0.05
--
INTENj I T Y (+
SOUND
2+
0
EJECTION SOUNDt
SECOND
r: $ 5 E
65/S
90/O-S
100/o-3
90/o-10
6
10
12
15
30
9
17
30
22
31
21
12
13 4
4
21.
22.
23.
24.
25.
26.
21.
28.
29.
30.
31.
.32.
33. 34.
35.
80/40
-3
1 S/75
125/65
110/25
125/60
120/45
90/40
100/60
90/25
65/23
SO/20
23/8
25/6
25/10
25/9
0.5x
0.25x 2.0x
2.0x
0.16X
0.4x
0.25x
2.0x
4.0X
6.0X
4.0x
4.0x
4.0X
0.2X
1.4x
0.5x
0.3x LICS
LICS
LICS
LICS
I 2nd LICS
0.13x
LICS
LICS
LICS
LICS
2nd LICS 2nd LICS
3rd
4th
4th
3rd
3rd LICS
3rd and 4th
4th
3rd LICS
4th
3rd LICS
3rd LICS
3rd LICS
3rd
4th
0.15x 0.2x
0.2X
0.5x
0.4x
0.25x
0
0
0
0
0
0
0
0
0
0
LICS
1
1
0
0
0
0
sound
sound
sound
0
0
0
0
and
/ +0.05/0.12
1 +0.06/0.12 ; +0.07/0.13
+0.06/0.13
+0.07/0.12
+O.OS/O.lZ
+0.07/0.12
+0.06/0.10
+0.06/0.16
+0.07/0.10
+0.07/0.11
0
0
+0.06/O.
0
0
0
over the Q-ejection was vonsist,ent with
0
Fi
0
Pulmonary diastolic murmur
rhird
Third
rhird
‘hidsomd
*Magnitude of shunt is expressed as a proportion of systemic flow (see text for explanation). tThe presenre of a wstolic ejecLion sound is indicated by +. The figures refer Lo t,he Q-l t,ime iin seconds) $‘l’he pulmonary art,ery was not entered. but the identity of right ventricular and radial arterial pressures necn the ventricles. lThe pulmonary artery was noti entered, but the diagnosis was confirmed at autopsy. “The intensity of wronrl so~mtl is ~rprwsed as f C diminished: 2-C normal: 3 f and +I-. arcentuatrd.
80/S
15010-5 112/O-S
135/o-15
12.5/s-10
100/o-5
125/O-S
120/O-S
58/O-6
23/O
25/o
25/3
16
20.
25/o
11
19.
sound a free
10
i
0.04 0.02
2+ 2+
0
-I+
t,ime (in seconds). communication br-
0.03
0
0.03
2+
3+
4f
0.02
0.04
2+
3+
0.05
0.03
2+ +I-
0.05
2+
58
CRAIGE
Am. Heart J. July, 1960
A systolic click or ejection sound was recorded early in systole in all of these patients with severe pulmonary hypertension. It was best seen in tracings taken in the second and third left intercostal spaces. The time of the systolic click averaged 0.12 second from the onset of the QRS (range, 0.10 to 0.13 second), The time from onset of the QRS to the beginning of the main valvular components of the first sound averaged 0.07 second (range, 0.05 to 0.08 second). A third sound was not seen in any of these patients with severe pulmonary hypertension. No apical diastolic murmur was recorded in any of this group. One, however, had a decrescendo diastolic murmur in the pulmonary area, attributed to pulmonary regurgitation. The systolic murmur was atypical in all of the patients of Group III (see Table I and Fig. 4). It was abbreviated in duration, being completed during approximately the first third of systole, in contrast to the pansystolic duration of the murmurs in the patients of Groups I and II. The intensity of the murmur was very small, again in striking contrast to those in patients of Groups I and II. DISCUSSION
The systolic murmur in ventricular septal defect results from turbulence due to the jet of blood passing through the defect. It is not surprising, therefore, that modifications in the murmur occur with heightened resistance in the pulmonary circuit, which diminishes the left-to-right flow. This effect is seen in the contrast in murmurs between the patients of Group III and those of the groups with lower pressures. The soft and brief systolic murmur along the upper left sternal border in these patients with similar pressures in the two ventricles has been attributed by Leathamc and Wood7 to ejection of blood into the dilated pulmonary artery, rather than to flow through the septal defect itself. In patients with large flows, there is increased filling of the left ventricle in early diastole. This, in turn, is usually associated with a third sound or middiastolic murmur at the apex in the patients of Group II, and some of the patients of Group I. Whether the apical rumbling murmur was associated with a gradient of pressure across the mitral valve could not be determined by the methods employed here. It was believed to be a flow murmur resulting from an increased return from the lungs to the left side of the heart.6 Because of the degree of pulmonary hypertension which existed in some of the patients of Group III, a blowing diastolic murmur of pulmonary regurgitation might have been expected. This was noted, however, in only one instance. The tendency to wide splitting of the second sound in some of the patients of Groups I and II is attributed to two factors: (1) prolongation of right ventricular systolic ejection, associated with an increased stroke volume of the right ventricle, which might delay pulmonary closure; and (2) a shortened left ventricular systolic ejection, with early closure of the aortic valve because of the existence of an alternate route of ejection from the left ventricle in addition to the aorta. Extreme elevations in pulmonary pressure were reflected in narrowing and accentuation of the second sound. This was due to an abbreviation of right ventricular systolic ejection, as has been shown to be present in pulmonary hypertension from various
ptr&~ ”
“10
PHONOCARDIOGRAPHY
IN
INTERVENTRICULAR
SEPTAL
59
DEFECTS
causes. In addition, an early systolic click or ejection sound appeared in the patients of Group III. The finding of such a sound has been well described lq I,eatham and Vogelpoels in cases of pulmonary hypertension, and has also been demonstrated in a variety of conditions resulting in dilatation of the aorta or pulmonary artery.$ Its timing is such as to coincide with the beginning of the period of maximum ejection of blood into the dilated and probably already tense
pulmonary
artery.
This clicking sound is easily confused with the valvul;tr
components of the first sound. The latter, however, are best heard---ancl recorded---over the apex or tricuspid areas, whereas the systolic click in the patients of our Group III was recorded over the second and third left intercostal spaces, 0.05 second after the initial valvular contributions to the first sound. Although there are some exceptions, as noted in Table I, the differences in phonocardiographic findings in the three groups can be summarized simply in Table I I. 'GABLE
II.
SUMMARY
0I; PHONOCARDIOGKAPHIC SEPTAL
FINDINGS DEFECT
IN 35
CASES
.--. SYSTOLIC
Group I
MUKMUR
I
Loud; pansystolic; crescendo-decrescendo or of constant intensitv. Maximal in third and fourth LICS
DIASTOLIC PHENOMENA
SECOND
01; INTEKVEKTRICvL.211
-. _~~ SOUND
~. ~~ EJKCTIOS SOlYlT
Zero; or normal third sound in children
Normally to widely split. UsualI> variable with respiration. ?jot accentuated
Absent
Third sound or mid-diastolic murmur
Normally to widely split, but not variable with respiration when widely split
Sometimes present
Tylp;,“, y B as in Figs. Group II
Loud ; pansystolic; crescendo-decrescendo. Maximal in third and fourth LICA Type A as in Fig. 3
Group III
Short, soft, early systolic; maximal in second, third, or fourth LICS Type C as in Fig. 4
Occasionally, pulmonary diastolic murmur
i Narrowly split or I single. .\ccentu,
~ Present
nted
i
It is of interest that, as Brotmacher and Campbelllo have pointed out, the uncomplicated ventricular septal defects with normal pressures tend to be found in patients in the first two decades of life, whereas the cyanotic types are found in patients more evenly spread through the first four decades. In our series, the average age of the patients of Group I was 10 years, and the average age of those in Groups II and III was 17 and 18 years, respectively, a finding in agreement with the data of these authors. We do not feel, however, that there is necessarily a tendency for patients to progress from one group to another, since in those patients who have been recatheterized, physiologic findings were unchanged.
60
CRAIGE
SUMMARY
AND
.41x1. Heart J. July, 1960
CONCLUSIONS
Thirty-five patients with interventricular septal defects were studied by means of cardiac catheterization and phonocardiography. They were divided into three groups on the basis of pressures and flows in the pulmonary artery. Group I were those with low pressures and flows who had harsh, pansystolic murmurs, either crescendo-decrescendo or of constant intensity, with normally to widely split second sounds. Group II were those with moderate to severe pulmonary hypertension and large left-to-right shunts who also had loud pansystolic murmurs, crescendo-decrescendo in configuration, and second sounds that were normally to widely split. Apical third sounds or diastolic murmurs were seen in this group and were attributed to rapid filling of the left ventricle. This transitional group shared some of the features of the other two types, since the systolic murmurs were similar to those in the patients with smaller shunts, whereas ejection sounds were often present, as in the patients with high pressures and mixed shunts. Group III were the patients with severe pulmonary hypertension and smaller or mixed shunts who had atypical systolic murmurs-brief and subdued. The second sound was accentuated and often single, and an ejection sound in early systole was uniformly present over the pulmonary area. It is believed that the phonocardiogram adds significantly to one’s ability to estimate the state of the physiologic disturbance in patients with interventricular septal defect. The cardiac catheterizations were performed S. Harned, Jr., Department of Pediatrics, and Dr. Department of Medicine, North Carolina Memorial
by Dr. Nelson K. Ordway and Dr. Herbert Daniel T. Young and Dr. Thomas C. Gibson, Hospital, Chapel Hill, N. C.
REFERENCES
Blount,
2. 3. 4. 5. 6. 7.
8. 9. IO.
S. G., Jr., Mueller, H., and McCord, M. C.: Ventricular Septal Defect: Clinical and Hemodynamic Patterns, Am. J. Med. 18:871, 1955. Shepherd, J. T.: The Pulmonary Circulation in the Presence of Interatrial, Interventricular and Interarterial Communications. In The Pulmonary Circulation, edited by W. R. Adams and I. Veith, New York, 1959, Grune &Stratton, Inc., p. 204. Edwards, J. E.: Functronal Pathology of the Pulmonary Vascular Tree in Congenital Cardiac Disease, Circulation 15:164, 1957. Auscultatory and Phonocardiographic Findings in Ventricular Septal DeLessof, M. H.: fect (Abstract), Circulation 18:748, 1958. Luisada, A. A., and Liu, C. K.: Cardiac Pressures and Pulses, New York, 1956, Grune & Stratton, Inc., p. 101. Leatham, A. : Auscultation of the Heart, Pediat. Clin. North America 5:839, 1958. Wood, P.: McDonald, L., and Emanuel, R.: The Clinical Picture Correlated With PhysioIogrcal Observations in the Diagnosis of Congenital Heart Disease, Pediat. Clin. North America 5:981, 1958. Leatham, A., and Vogelpoel, L.: The Early Systolic Sound in Dilatation of Pulmonary Artery, Brit. Heart J. 16:21, 1954. Minhas, K., and Gasul, B. M.: Systolic Clicks: A Clinical, Phonocardiographic, and Hemodynamic Evaluation, AM. HEART J. 57:49, 1959. History of Interventricular Septal Defect, Brotmacher, L., and Campbell, M. : The Natural Brit. Heart J. 20:97, 1958.