JACC Vol. 26, No. 6 1545 [606032]

JACC Vol. 26, No. 6 1545
November 15, 1995:1545-8
PEDIATRIC CARDIOLOGY
High Prevalence of Muscular Ventricular Septal Defect in Neonates
NATHAN ROGUIN, MD, ZHONG-DONG DU, MD, MILA BARAK, MD, NADIM NASSER, MD,
SYLVIA HERSHKOWITZ, MD, ELLIOT MILGRAM, MD
Nahariya, Israel
Objectives. This study sought to use echocardiography to eval-
uate the prevalence of muscular ventricular septal defect in
neonates.
Background. Ventricular septal defect is usually asymptomatic
and closes spontaneously. An increase in its prevalence has been
noted recently. One reason is the improved detection of small
defects, especially with the increased use of echocardiography.
Therefore, one would expect a higher prevalence in neonates on
the basis of echocardiographic screening.
Methods. Color Doppler echocardiography was performed in
1,053 consecutive neonates 6 to 170 h old at Western Galilee
Hospital, Israel. Data on the neonates, parents and family were
obtained to analyze the influencing factors. The identified patients
were followed up for 1 to 10 months or until ventricular septal
defect closure.
Results. Muscular ventricular septal defect was found in 56 (25
male, 31 female) of the 1,053 neonates, a prevalence of 53.2/1,000
live births. All neonates were asymptomatic. Six had a systolic
murmur. Electrocardiographic findings were normal in 44
(97.8%) of 45 neonates followed up, and left ventricular hypertro- phy occurred in 1 (2.2%). By echocardiography, 50 ventricular
septal defects (89.3%) were single and 6 (10.7%) were multiple.
The defects (range 1 to 5 mm in diameter, mean [-SD] 2.3 +- 0.8)
occurred anywhere along the muscular septum; 43 (76.8%) were
detectable only on color Doppler imaging. The left atrium and left
ventricle were mildly dilated. Of 45 neonates who were followed up
for 6 to 10 months or until closure of the defects, 40 (88.9%) had
defects that closed spontaneously. The risk of ventricular septal
defect was not significantly associated with gestational age, birth
weight, birth order, maternal age, diabetes, smoking, exposure to
drugs or infection, paternal age, familial congenital heart disease,
religion or consanguinity.
Conclusions. There is a prevalence of muscular ventricular
septal defect in neonates of 53.2/1,000 live births. The patients
were asymptomatic, and 88.9% had defects that closed spontane-
ously within 1 to 10 months. These defects may be caused by
environmental factors. In many cases, muscular ventricular septal
defect may also result from delayed physiologic development.
(J Am CoU Cardiol 1995;26"1545-8)
Ventricular septal defect is usually asymptomatic and often
closes spontaneously (1). The reported rates of spontaneous
closure vary between 50% and 75% in small defects (2-4).
Therefore, the prevalence of ventricular septal defect should
be higher in neonates. An increase in its prevalence, especially
that of muscular ventricular septal defect, has been reported in
some recent studies (5-7). The reason was ascribed to im-
proved detection of small isolated defects, especially with the
increased use of echocardiography (1,5,7). Color Doppler
echocardiography has proved to be a sensitive and reliable
method for identifying ventricular septal defect (8-12). We
therefore examined 1,053 consecutive neonates with color
Doppler echocardiography to evaluate the prevalence of iso-
lated muscular ventricular septal defect at birth and to analyze
its clinical and prognostic characteristics and influencing fac-
tors.
From the Heart Institute and Department of Neonatology, Western Galilee
Hospital-Nahariya, Technion Faculty of Medicine, Nahariya, Israel.
Manuscript received April 28, 1995; revised manuscript received June 27,
1995, accepted July 10, 1995.
Address for corresoondence: Dr. Nathan Roguin, Heart Institute, Western
Galilee Hospital-Nahariya, P.O. Box 21, 22100 Nahariya, Israel. Methods
Subjects and protocol. Color Doppler echocardiography
was performed in 1,053 consecutive neonates in the Western
Galilee Hospital-Nahariya, Israel, from April to September
1994. Data on the neonates, parents and family were obtained
by interviewing the parents on the day of examination. Mater-
nal exposure to medicines or infections meant that the mother
took medicines or had infection during the first trimester of
pregnancy. Consanguineous marriages among Arabs living in
the Galilee in northern Israel have been found to be high
(13,14). To analyze whether consanguinity contributes to the
prevalence of ventricular septal defect, family relationships
were assessed. The couples were defined as consanguineous if
they had a common grandparent or great-grandparent (i.e.,
first or second cousins) (14). At the first examination the
neonates were 6 to 170 h old (mean 37 h). Parental consent was
obtained for each neonate. The study had the approval of the
Hospital Human Investigation Committee.
The neonates with muscular ventricular septal defect were
followed up at intervals of 2 to 3 months or until spontaneous
closure of the ventricular septal defect was confirmed. At each
follow-up visit, clinical history, physical examination, 12-lead
©1995 by the American College of Cardiology 0735-1097/95/$9.50
0735-1097(95)00358-B

1546 ROGUIN ET AL. JACC Vol. 26, No. 6
MUSCULAR VENTRICULAR SEPTAL DEFECT November 15, 1995:1545-8
electrocardiography and color Doppler eehocardiography were
performed.
Echocardiography. The echocardiographic examination
was performed with either an Aloka Color Doppler SSD-870
System (Aloka Co.) or a Hewlett-Packard Sonos 1000 Flow-
Mapping System with a 5-MHz transducer at the lowest
available pulse repetition frequency and moderately high flow
gains. The muscular portion of the ventricular septum was
visualized using the parasternal left ventricular long-axis,
parasternal four-chamber, apical four-chamber, apical five-
chamber, parasternal aortic root and left ventricular short-axis
views. Careful modulation of the transducer was needed to
scan every part of the muscular septum and show the largest
flow mapping of the ventricular septal defect diameter (10).
Classification of defects was according to Soto et al. (15). The
diagnosis of muscular ventricular septal defect was established
on the basis of two criteria: 1) the presence of a mosaic image
passing through the muscular ventricular septum from the left
to the right ventricle; 2) and a turbulent systolic flow jet
recorded on the right surface of the ventricular septal defect by
pulsed or continuous wave Doppler (8,16). Care was taken to
avoid misdiagnosing the accelerated flow in the right ventric-
ular trabeculae or coronary artery as a ventricular septal
defect. The diameter of the ventricular septal defect on color
flow mapping was measured. All examinations were recorded
on videotape and reviewed by at least two pediatric cardiolo-
gists. Ventricular septal defect was identified only when the
diagnoses of the two pediatric cardiologists concurred.
Statistics. To analyze the factors during pregnancy influ-
encing the occurrence of ventricular septal defect and the
dimensional changes of the heart in neonates with muscular
ventricular septal defect, 975 neonates without any structural
cardiovascular disease served as the control group. The prev-
alence of ventricular septal defect was based on the number of
live births with ventricular septal defect divided by all 1,053
neonates. The chi-square test or Fisher exact test was used to
compare differences between the rates, and the Student t test
was used to compare the differences between the means at a
significance level of 5%.
Results
Prevalence of ventricular septal defect and influencing
factors. There were 56 isolated muscular ventricular septal
defects in 1,053 consecutive neonates, a prevalence of 53.2/
1,000 live births (25 male [44.6%], prevalence 45.2/1,000 live
births; 31 female [55.4%], prevalence 62/1,000 live births).
Although muscular ventricular septal defect seemed to be
present more frequently in female neonates, the difference
between prevalence in male and female neonates was not
significant (p > 0.05). Gestational age ranged from 36 to 42
weeks and birth weight from 2.42 to 4.42 kg. Apgar scores at
the first minute were 7, 8 and 9 in 1, 2 and 53 neonates,
respectively, and all were normal at the fifth minute. The birth
order was as follows: first born in 19 neonates (33.9%), second
in 14 (25%), third in 11 (19.6%), fourth in 7 (12.5%), and fifth Table 1. Data for Neonates and Their Families
VSD Control Group p
(n = 56) (n = 975) Value
Gestational age (wk) 39.1 +_ 1.2 39.3 + 1.6 0,84*
Cesarean delivery 9 (16.1%) 142 (14.6%) >OAt
Birth weight (kg) 3.40 _+ 0.47 3.32 +_ 0.49 0.12'
Maternal age (yr) 26.9 _+ 5.6 27.0 +_ 5.4 0.81"
Diabetes 1 (1.8%) 5 (0.5%) 0.25§
Hypothyroidism 1 (1.8%) 6 (0.6%) 0.24§
Exposure to antibiotics:~ 3 (5.4%) 81 (8.3%) 0.31§
Exposure to methyldopa~t 1 (1.8%) 4 (0.4%) 0.22§
Exposure to thyroxine~ 1 (1.8%) 6 (0.6%) 0.24§
Respiratory infection~ 1 (1.8%) 38 (3.9%) 0.25§
Urinary tract infeetion~: 2 (3.6%) 43 (4.4%) 0.28§
Hypertension 1 (1.8%) 5 (0.5%) 0.25§
Paternal age (yr) 30.8 _ 4,8 31.1 +- 5.9 0.47*
Siblings with CHD 2 (3.6%) 14 (1.4%) 0.16§
Family with CHD 2 (3.6%) 17 (1.7%) 0.20§
Family with diabetes 15 (26.8%) 251 (25.7%) >OAt
Family with G-6-PD 1 (1.8%) 4 (0.4%) 0.22§
Consanguinity
1st degree 9 (16.1%) 137 (14.1%) >0.1t
2nd degree 5 (8.9%) 68 (6.9%) >OAt
*Student t tests, tChi-square test. :~Mothers took medicines or had an
infection during the first trimester. §Fisher exact tests. Data presented are mean
value +- SD. CHD = congenital heart disease; G-6-PD = glucose-6 phosphate
dehydrogenase deficiency.
or later in 5 (8.9%). Compared with the control group, there
were no significant differences in gestational age, birth weight,
type of delivery, Apgar score or birth order (all p > 0.05)
(Table 1).
The ages of the parents of neonates with muscular ventric-
ular septal defect were 19 to 39 years for mothers and 23 to 42
years for fathers and were not significantly different from
parents of the control group (Table 1). Maternal diabetes,
hypothyroidism, hypertension, maternal exposures to medi-
cines (antibiotic agents, methyldopa, thyroxine) or infection
(respiratory or urinary tract infection), familial congenital
heart disease, familial diabetes and other genetic diseases were
not significantly more prevalent among the neonates with
ventricular septal defect. The risk of ventricular septal defect
was also not associated with consanguinity (Table 1). There
was no maternal alcohol consumption or exposure to chemicals
at the workplace during the first trimester of gestation in the
neonates with ventricular septal defect.
The neonates were the offspring of four religious denomi-
nations living in this area: Jews, Moslems, Druze and Chris-
tians. Of the 1,053 neonates, there were 316 Jews (30.0%), 441
Moslems (41.9%), 218 Druze (20.7%) and 78 Christians
(7.4%). Of the 56 neonates with muscular ventricular septal
defect, 13 were Jews, 26 Moslems, 15 Druze and 2 Christians.
The prevalence of muscular ventricular septal defect was 41.4,
59, 68.8 and 25.6/1,000 live births in Jews, Moslems, Druze and
Christians, respectively. There were no significant differences
among the four religious groups (p > 0.05).
Clinical features. All the neonates appeared healthy at the
first examination. No neonate had cyanosis or tachypnea. By

JACC Vol. 26, No. 6 ROGUIN ET AL. 1547
November 15, 1995:1545-8 MUSCULAR VENTRICULAR SEPTAL DEFECT
Table 2. Left Ventricular and Atrial Changes in Neonates With
Ventricular Septal Defects
VSD Control Group p
Index (n = 56) (n = 975) Value
LVEDD 18.3 _+ 1.7 16.4 +_ 1.8 <0.001
LA 13.3 +_ 1.7 11.2 + 1.7 <0.001
LA/AO ratio 1.39 + 0.21 1.17 _+ 0.20 <0.001
Data presented are mean value – SD. AO = aortic diameter; LA = left
atrial diameter; LVEDD = left ventricular end-diastolic diameter.
palpation, one had increased precordial activity, and none had
a palpable thrill. On auscultation, $1 and $2 were normal in all
the neonates; one neonate had a grade 3/6 harsh holosystolic
murmur heard in the third and fourth intercostal spaces, and
five had a grade 1 to 2/6 early short systolic murmur.
Echoeardiographic features. Of 56 neonates with muscular
ventricular septal defect, 50 had a single ventricular septal
defect (89.3%), and 6 had multiple ventricular septal defects
(10.7%) (5 with two, 1 with at least three); 59 defects with
measurable data were detected. The ventricular septal defect
diameter measured 1 to 1.9, 2 to 2.9, 3 to 3.9 and to 4 to 5 mm
in 9 (15.3%), 29 (49.2%), 18 (30.5%) and 3 (5.1%) neonates,
respectively (mean [_SD] 2.3 _+ 0.8). Ventricular septal de-
fects occurred anywhere along the muscular septum, with most
in the midtrabecular septum. The jet direction on color flow
mapping varied greatly, and there were no particular charac-
teristics.
Of 59 muscular ventricular septal defects with measurable
data, 13 (22%) were detectable by two-dimensional echocar-
diography with small dropouts in the muscular septum. The
left ventricle and left atrium were dilated mildly compared with
that in the control group (p < 0.001) (Table 2).
Follow-up of neonates with muscular ventricular septai
defect. Of 56 neonates with muscular ventricular septal defect,
45 (80.4%) were followed up for 6 to 10 months or until the
closure of the defect, 3 for 1 month; 8 (14.3%) did not return
for examination. The rates of spontaneous closure for different
ages were calculated on the basis of the 45 neonates who were
followed up for 6 to 10 months. Of these neonates, the
ventricular septal defect closed in 7 (25%) of 28 who returned
for examination within 1 month, 15 (37.5%) of 40 within 2
months, 19 (44.2%) of 43 within 3 months, 28 (62.2%) of 45
within 4 months, 32 (71.1%) of 45 within 5 months, 34 (75.6%)
of 45 within 6 months, 37 (82.2%) of 45 within 7 months and 40
(88.9%) of 45 within 8 to 10 months. In the 9, 20, 15 and 2
neonates with ventricular septal defect diameters 1 to 1.9, 2 to
2.9, 3 to 3.9 and 4 to 5 mm, 8 (88.9%), 17 (85%), 9 (60%) and
1 (50%) defect closed within 6 to 10 months, respectively.
There were no significant differences in the rates of closure in
ventricular septal defects of different sizes (p > 0.05). After
spontaneous closure of the ventricular septal defect, the atrial
and ventricular dimensions and ventricular function were all
normal on echocardiography.
All infants who were followed up had normal growth and
development. None had any clinical or echocardiographic signs of infectious endocarditis. In three of six neonates with a
systolic murmur heard at the initial examination, the murmur
disappeared at defect closure within 4, 4.5 and 8 months,
respectively. No pericordial overactivity or thrill was found.
The ECG findings were completely normal in 37 infants
(82.2%) and showed a polyphasic rsr' pattern in lead V1 in 7
(15.6%); that is essentially a normal variant. Left ventricular
hypertrophy was seen in one (2.2%) infant who had a defect
4 mm in diameter that did not close until she was 7.5 months
old. Infants with and without rsr' patterns in lead V1 did not
differ in prognosis (p > 0.05). No neonate had left deviation of
the QRS axis in the frontal plane.
Discussion
Prevalence of ventricular septal defect. In the present
study, the prevalence (53.2/1,000 live births) greatly exceeded
the expected rates. The reported prevalence of ventricular
septal defect in previous studies varied from 0.3 to 3.3/1,000
live births (17-20). Although an increase in its prevalence has
been noted in some recent reports, it has never exceeded
5/1,000 live births (2). Thus, prevalence in our study is >10
times greater than that in the previous reports.
Possible reasons for high prevalence of ventricular septal
defect. Two possibilities may account for these results.
1. Screening method. We used color Doppler echocardio-
graphic screening to identify neonates with muscular ventric-
ular septal defects. Most of the neonates in this study were
clinically healthy. When a murmur was heard by the pediatri-
cian, it was usually an early short systolic murmur, difficult to
differentiate from an "innocent murmur." In addition, most of
the ventricular septal defects closed spontaneously within 6 to
10 months. Therefore, neonates with small muscular ventricu-
lar septal defects were most easily missed without echocardio-
graphic screening. Even when our patients were examined by
echocardiography, those with small muscular ventricular septal
defects could also be missed if color Doppler flow mapping was
not used or if the muscular septum was not completely
scanned. Most of the ventricular septal defects were undetect-
able with two-dimensional echocardiography and occurred at
any point along the muscular septum. In previous studies
(17-19), the diagnosis of ventricular septal defect was made on
the basis of clinical diagnosis by a pediatric cardiologist in
patients usually referred by their pediatricians. To our knowl-
edge, no previous reports on the prevalence of ventricular
septal defect used echocardiographic screening in consecutive
neonates. Therefore, most of our cases would have been
missed in previous studies, which may explain the unusually
high prevalence in the present work.
2. Teratogens. There might have been some potential ter-
atogens that caused an epidemic of small muscular ventricular
septal defects in our region during the period from April to
September 1989. Of 56 neonates, 50 had no signs of ventricular
septal defect on physical examination. If these "silent" cases
were excluded, then the prevalence would be 5.7/1,000 live
births, which is still greater than that in previous reports

1548 ROGUIN ET AL. JACC Vol. 26, No. 6
MUSCULAR VENTRICULAR SEPTAL DEFECT November 15, 1995:1545-8
(2,17-20). Although the present work could not establish any
relation between the high prevalence of muscular ventricular
septal defect and known teratogens, we could not exclude the
possibility that there were some environmental factors or
unknown teratogens that resulted in this high prevalence of
muscular ventrieular septal defect.
Postnatal closure of the muscular septum. On the basis of
fact that there are an excess number of isolated ventricular
septal defects among premature infants, a lack of any abnor-
mal autopsy findings in patients with spontaneously closed
ventricular septal defects and a disproportionate lack of ven-
tricular septal defect among the autopsied adult population
compared with the living pediatric clinical population, Mitchell
et al. (21) proposed that the normal time of ventricular septal
closure may not be limited to the fourth and fifth postconcep-
tive weeks; rather, it may extend, in a minority of infants,
throughout pregnancy and into the postpartum period. The
high prevalence in our work may also be explained by the
postnatal closure of the muscular septum. Thus, in many cases,
these ventricular septal defects may result from a delayed
normal process rather than from disease. Therefore, to avoid
unnecessary anxiety, the parents should be informed of this
benign muscular ventricular septal defect whether it is identi-
fied by echocardiography intentionally or accidentally. It is not
necessary to look for this ventricular septal defect in asymp-
tomatic neonates.
Conclusions. There is a high prevalence (53.2/1,000 live
births) of muscular ventricular septal defect in neonates. All
the neonates were asymptomatic; 10.7% had a systolic murmur
on physical examination. The ECG findings were normal in
97.8% of neonates and showed left ventricular hypertrophy in
2.2%. Spontaneous closure occurred in 88.9% of neonates
within 1 to 10 months. These ventricular septal defects may be
caused by some environmental factors, and in many cases, they
may represent delayed physiologic development.
References
1. Newman TB. Etiology of ventricular septal defects: an epidemiologic
approach. Pediatrics 1985;76:741-9. 2. Moe DG, Guntheroth WG. Spontaneous closure of uncomplicated ventric-
ular septal defect. Am J Cardiol 1987;60:674-8.
3. Hoffman JIE. Natural history of congenital heart disease: problems in its
assessment with special reference to ventricular septal defects. Circulation
1968;37:97-125.
4. Alpert BS, Cook DH, Varghese J, Rowe RD. Spontaneous closure of small
ventricular septal defects: ten years follow-up. Pediatrics 1979;63:204-6.
5. Martin GR, Perry LW, Ferencz C. Increased prevalence of ventricular septal
defect: epidemic or improved diagnosis. Pediatrics 1989;83:200-3.
6. Layde PM, Dooley K, Erickson JD, Edmonds LD. Is there an epidemic of
ventricular septal defects in the U.S.A.? Lancet 1980;1:407-8.
7. Fixler DE, Pastor P, Chambedin M, Sigman E, Eifler CW. Trends in
congenital heart disease in Dallas county births, 1971-1984. Circulation
1990;81:137-42.
8. Snider AR, Bengnr AR. Two-dimensional and Doppler echocardiography in
the evaluation of congenital heart disease. In: Marcus ML, Schelbert HR,
Skorton D J, Wolf GL, Braunwald E, editors. Cardiac Imaging, a Companion
to Braunwald's Heart Disease. Philadelphia: Saunders, 1991:479-510.
9. Helmcke F, Souza A, Nanda NC, Villacosta I, Gatewood R Jr, Colvin E,
Soto B. Two-dimensional and color Doppler assessment of ventricular septal
defect of congenital origin. Am J Cardiol 1989;63:1112-6.
10. Hornberger LK, Sahn DJ, Krabill KA, et al. Elucidation of the natural
history of ventricular septal defects by serial Doppler color flow mapping
studies. J Am Coil Cardiol 1989;13:1111-8.
11. Johnson TB, Fyfe DA, Thompson RP, Kline CH, Swindle MM, Anderson
RH. Echocardiographic and anatomic correlation of ventricular septal defect
morphology in newborn Yucatan pigs. Am Heart J 1993;125:1067-72.
12. Ludomirsky A, Huhta JC, Vick GW, Murphy DJ, Danford DA, Morrow
WR. Color Doppler detection of multiple ventricular septal defects. Circu-
lation 1986;74:1317-22.
13. Freundlich E, Hino N. Consanguineous marriages among the rural popula-
tion of Arabs in Israel. Israel J Med Sci 1984;20:1035-7.
14. Gev D, Rognin N, Freundlich E. Consanguinity and congenital heart disease
in the rural Arab population in northern Israel. Hum Hered 1986;36:213-7.
15. Soto B, Becker AE, Moulaert AJ, Lie JT, Anderson RH. Classification of
ventricular septal defects. Br Heart J 1980;43:332-43.
16. Stevenson JG, Kawabori I, Dooley T, Guntherroth WG. Diagnosis of
ventricular septal defect by pulsed Doppler echocardiography. Circulation
1978;58:322-6.
17. Ferencz C, Rubin JD, Brenner JJ, et al. Congenital heart disease: prevalence
at live birth. Am J Epidemiol 1985;121:31-6.
18. Fyler DC. Report of the New England regional infant cardiac program.
Pediatrics 1980;65 Suppl:375-403.
19. Hoffman JJE, Christianson R. Congenital heart disease in a cohort of 19502
births with long-term follow-up. Am J Cardiol 1978;42:641-7.
20. Saab NG, Burchell HB, DuShane JW, Titus JL. Muscular ventricular septal
defects. Am J Cardiol 1966;18:713-23.
21. Mitchell SC, Berendes HW, Clark WM Jr. The normal closure of the
ventricular septum. Am Heart J 1970;80:432-3.