(Circulation. 1999;99:1209-1214.)
© 1999 American Heart Association, Inc.
Clinical Investigation and Reports |
From the Division of Cardiology, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada (Y.V.M., J.F.S., L.K.H.); the Department of Cardiology, Children's Hospital, Boston, Mass (S.A.K., M.E.v.d.V.); and Children's Hospital of Eastern Ontario, Ottawa, Ontario, Canada (B.S.).
Correspondence to Lisa K. Hornberger, Division of Cardiology, The Hospital for Sick Children, 555 University Ave, Toronto, Ontario, Canada, M5G 1X8. E-mail hornberg{at}sickkids.on.ca
| Abstract |
|---|
|
|
|---|
Methods and ResultsWe reviewed the prenatal and postnatal echocardiograms and outcomes of 16 fetuses with TGA/intact ventricular septum or small ventricular septal defect. Of the 16 fetuses, 6 prenatally had an abnormal FO (fixed position, flat, and/or redundant septum primum). Five of the 6 had restrictive FO at birth. Five fetuses had DA narrowing at the pulmonary artery end in utero, and 6 had a small DA (diameter z score of <-2.0). Of 4 fetuses with the most diminutive DA, 2 also had an abnormal appearance of the FO, and both died immediately after birth. One other fetus had persistent pulmonary hypertension. Eight fetuses had abnormal Doppler flow pattern in the DA (continuous high-velocity flow, n=1; retrograde diastolic flow, n=7).
ConclusionsAbnormal features of the FO, DA, or both are present in fetuses with TGA at high risk for postnatal hypoxemia. These features may result from the abnormal intrauterine hemodynamics in TGA. A combination of restrictive FO and DA constriction in TGA may be associated with early neonatal death.
Key Words: heart defects, congenital transposition of great vessels echocardiography
| Introduction |
|---|
|
|
|---|
| Methods |
|---|
|
|
|---|
Echocardiographic Examinations and
Measurements
The fetal echocardiographic examinations were
performed with a 3.5-, 5-, or 7.5-MHz transducer on an Acuson 128 or
Hewlett-Packard system, or a 7- to 4-MHz curved-array probe on
an Advanced Technology Laboratories high-density imaging system. All
images were recorded on videotape for offline analysis.
Based on the findings of Wilson et al,12 the prenatal atrial septal appearance was described as redundant if the aneurysmal septum primum bulged >50% of the way across to the left atrial free wall; flat, if the angle between the septum primum and the rest of the atrial septum was <30°; or fixed, if the septum primum did not have the typical swinging motion during the cardiac cycle.
The diameter of the ductus arteriosus was measured as described by Tan et al,13 and the measurements were converted to z scores by use of data from normal fetuses. When there was an anatomic narrowing within the vessel, typically at the pulmonary end, the diameter was measured at the narrowest portion. Spectral Doppler and color flow mapping were used to record flow in the ductus arteriosus, which was considered antegrade if directed from pulmonary artery to aorta and retrograde if the opposite.14 By Doppler, laminar antegrade flow throughout systole and part or all of diastole was considered to be normal ductus arteriosus flow.15
Postnatal Course
Postnatal history, arterial blood gas data,
postnatal echocardiographic reports, and autopsy
reports were collected from each center, and the postnatal course was
assessed. A restrictive foramen ovale after birth was confirmed by
postnatal echocardiography or at autopsy, both in
conjunction with a consistent clinical history defined as
severe hypoxemia (PaO2<25) and
acidosis (pH<7.30) after birth, necessitating urgent balloon atrial
septostomy. The foramen ovale was considered restrictive by
echocardiography if there was a diminutive
interatrial communication (<5 mm), with flow acceleration by
color Doppler suggesting a gradient.
Calculations and Statistical Analysis
The mean z scores for ductus arteriosus diameter in
all 16 fetuses were tested against the normal mean of 0 with the use of
a single-group t test.
| Results |
|---|
|
|
|---|
|
Foramen Ovale
Prenatal echocardiography demonstrated an
abnormal appearance of the foramen ovale in 6 of the 16 fetuses with
TGA (38%) (Figures 1
and 2
). Five of the 6 had postnatal evidence
of restrictive foramen ovale. In 2 with serial antenatal studies, the
septum primum was more mobile in the earlier studies performed at 27
weeks (patient 2) and 19 weeks (patient 3) than in the studies
performed later in gestation, at 32 weeks and 28 weeks, respectively.
None of the fetuses with a normal appearance of the foramen ovale in
utero had evidence of a restrictive foramen ovale after birth.
|
|
Two of the 6 patients (patients 1 and 2) with autopsy confirmation of a
restrictive foramen ovale and a constricted ductus arteriosus died
shortly after birth. At 35 weeks of gestation, patient 1 had an atrial
septum that bulged tensely from the right atrium toward the left atrium
in a fixed position without the normal phasic movement and had only a
small gap (Figure 3
). In patient 2, at 32
weeks of gestation, the atrial septum was thick and flat without normal
movement, and there was only a small flow orifice through the septum
primum. Both fetuses had normal function of the left ventricle and no
pulmonary or mitral insufficiency. Their conditions immediately
after delivery were good, with good muscle tone and a strong cry.
Within minutes, however, both developed profound hypoxemia and poor
cardiac output unresponsive to aggressive resuscitation, which included
immediate institution of prostaglandin therapy in both and
an unsuccessful attempt at balloon atrial septostomy from both the
umbilical and femoral veins in patient 2. Autopsy in both neonates
revealed a nearly closed ductus arteriosus and closed foramen ovale.
Patient 2 also had an interrupted inferior vena cava with
azygous continuation, which was the cause of the failed balloon
septostomy. The lung pathology in patient 1 revealed intimal
proliferation (Heath-Edwards grade 2 changes) in the pulmonary
artery.
|
Ductus Arteriosus
For all 16 fetuses with TGA, the mean z score of the
ductus arteriosus diameter was significantly below the published normal
data (P<0.0001), ranging from -0.1 to -6.3, with a mean
of -1.8 (Figure 4
). None had maternal
indomethacin administration.
|
In 5 of the 16 fetuses (31%), the ductus arteriosus appeared
anatomically constricted or narrowed at the pulmonary artery
end (Figure 5
). In patient 2, the ductus
arteriosus appeared narrowed in an earlier study performed at 27 weeks.
However, in patients 8 and 9, the ductus arteriosus had a normal
appearance at 20 and 22 weeks, respectively, but appeared constricted
later in gestation. Ductal constriction was associated with a
restrictive foramen ovale in 2 fetuses with early neonatal death
(patients 1 and 2) (see above). Persistent pulmonary
hypertension developed postnatally in 1 patient with a severely
narrowed ductus arteriosus in utero, which remained patent at birth
(patient 6). The last 2 cases (patients 8 and 9) had an uneventful
postnatal course.
|
Eight of the 16 fetuses (50%) had an abnormal Doppler flow pattern
in the ductus arteriosus. Continuous, high-velocity antegrade flow
(Figure 3
) was observed in 1 fetus (patient 1). The other 7 had
bidirectional flow in the ductus arteriosus (Figure 6
). In 2, this pattern was observed
throughout the examination, and in 4 it was intermittent, with a normal
pattern observed at other times (Table
). Bidirectional flow was
identified as early as 27 weeks in patient 2, 24 weeks in patient 7
(Figure 6
), and 19 weeks in patient 13. However, in 2 others
with serial antenatal studies (patients 3 and 11), there was antegrade
flow in the ductus arteriosus throughout the cardiac cycle in the
earlier examinations (at 28 and 23 weeks, respectively). Three fetuses
with bidirectional ductal flow had prenatal narrowing of the ductus
arteriosus and/or restrictive foramen ovale (patients 2, 3, and 6).
|
| Discussion |
|---|
|
|
|---|
Prediction of Early Neonatal Death in TGA
We found the combination of foramen ovale restriction and ductus
arteriosus constriction to be fatal in 2 patients after discontinuation
of the placental circulation in TGA. With neither site available for
mixing, the systemic and pulmonary circulations were entirely
in parallel, resulting in severe hypoxemia. Retrospectively, the
prenatal echocardiograms had demonstrated abnormal features of both
structures.
The incidence of this lethal combination remains unclear. Our study was probably biased because the experience with the 2 cases motivated our retrospective review of fetal TGA. Both a multi-institutional prospective study and an institutional retrospective study at The Hospital for Sick Children, Toronto, suggested that as many as 4% of neonates with TGA die before repair.6 11 One problem with ascertaining the number of infants with TGA and the lethal combination of restrictive foramen ovale and ductal constriction is that the neonate may die before a diagnosis of TGA is made and referral to a tertiary care center can be arranged. In fact, a further review of autopsy records at The Hospital for Sick Children in the past 10 years identified 2 more cases of very early death after delivery at outside hospitals with TGA, and autopsy confirmed the presence of ductus arteriosus and foramen ovale restriction. Because the surgical mortality for the arterial switch operation is as low as 2% in many institutions and <5% in most,16 the preoperative mortality is an important issue in the management and outcome of infants with TGA.
Nevertheless, the management of infants with this lethal condition is still problematic. Balloon atrial septostomy before interruption of the placental circulation may improve the outcome of these infants.
Prenatal Nature of the Foramen Ovale in TGA
Our study demonstrated that in the fetus or prenatally, TGA is
frequently associated with an abnormal foramen ovale with either an
abnormal septum primum angle, abnormal motility of the septum primum,
or an aneurysmal septum primum. The fixed position and the
angle of <30° may represent restrictive foramen ovale, with
a continuous pressure difference between the left and right
atria.17 Two explanations have been proposed for a
restrictive foramen ovale in utero: primary restriction and secondary
restriction, with the latter being the result of
hemodynamic alterations (eg, elevated left atrial
pressure).17 For the fetuses with TGA, we suspect that the
latter mechanism may explain the prenatal restriction. Three of the 4
fetuses with echocardiographic features of foramen
ovale restriction had abnormal flow in the ductus arteriosus, and the
other had large pulmonary veins with a prominent color flow
signal. These findings are consistent with increased
pulmonary blood flow, which might increase left atrial pressure
and lead to foramen restriction as normally occurs after birth. In
support of this, Rizzo et al18 found an abnormal
pulmonary venous flow pattern in a fetus with TGA
consistent with increased left atrial pressure.
One other mechanism of foramen restriction, as suggested by 1 of our 5 cases, was the presence of a redundant septum primum that nearly occluded a large foramen ovale orifice after birth. Although a redundant septum primum may be found in as many as 6% of normal fetuses12 and was found in our series in 1 fetus with TGA but no postnatal restriction, its presence should alert the fetal echocardiographer to the possibility of foramen restriction at birth. Therefore, our study provides evidence for at least 2 potential mechanisms of foramen ovale restriction in TGA: (1) elevated left atrial pressure with reduced foramen flow orifice in utero and altered atrial septal configuration and (2) a redundant septum primum occluding the foramen orifice after birth.
Prenatal Nature of the Ductus Arteriosus and Pulmonary
Vascular Bed in TGA
The common occurrence of both an abnormal foramen ovale and
abnormal structure and flow through the ductus arteriosus in fetal TGA
may be explained by the pathophysiology of TGA (Figure 7
). In the normal fetal circulation, more
highly oxygenated blood from the placenta crosses the
foramen ovale, resulting in an oxygen saturation in the left ventricle
of
10% higher than that in the right ventricle. In the fetus with
normally related great arteries, this blood is pumped to the ascending
aorta. In fetuses with TGA, this more highly oxygenated
blood in the left ventricle is pumped to the pulmonary artery.
This blood, in addition, returns to the pulmonary veins and may
make the oxygen saturation in the left ventricle even higher than in
the normal heart. Even in patient 2, with azygous continuation from the
inferior vena cava to the superior vena cava, more highly
oxygenated blood from the umbilical vein, typically
connected to the hepatic vein, should cross the foramen ovale as usual.
We speculate that the increased oxygen content in the pulmonary
artery may lead to ductal constriction and decreased pulmonary
vascular resistance. In the fetal lamb, Konduri et al19
found that a 13% increase in oxygen saturation increased
pulmonary blood flow 3-fold and increased the left atrial
pressure from 4 to 8 mm Hg. This lower pulmonary vascular
resistance may explain the retrograde diastolic flow in the
ductus arteriosus in many of the fetuses with TGA even as early as 19
weeks of gestation.
|
The smaller diameter of the ductus arteriosus that we observed in
fetuses with TGA may also be explained by the abnormal intrauterine
hemodynamics. In the normal fetal circulation, 30% to
40% of the total cardiac output passes through the ductus
arteriosus.20 In fetuses with TGA, however, the amount of
blood crossing the ductus arteriosus may be the same as the amount that
passes through the foramen ovale (Figure 7
) in the absence of
significant retrograde diastolic ductal flow. This amount
would represent
20% of the total cardiac output, according
to observations in normal fetuses.20 21 Both an elevated
left atrial pressure and prenatal restriction of the foramen ovale
probably contribute to a further reduction in ductus arteriosus flow
later in gestation. If the ductal diameter is a function of flow
volume, as is thought to be the case for other vascular structures, one
would expect the diameter to be smaller than in the normal fetus, even
in the absence of active constriction.
Observations in human and animal fetuses with normal cardiovascular anatomy have shown that prenatal constriction of the ductus arteriosus produces anatomic changes in small pulmonary arteries,22 23 similar to neonates with idiopathic persistent pulmonary hypertension.24 This may be related to an increase in pulmonary blood flow and pressure. One could speculate that the common occurrence of persistent pulmonary hypertension and early progression of pulmonary vascular disease in the infants with TGA and intact ventricular septum may be related to the prenatal increase in pulmonary blood flow and prenatal alteration of the pulmonary vascular bed.7 8 9 25 26 27 In our study, in fact, patient 1, with prenatal constriction of the ductus arteriosus, had such pulmonary vascular changes as previously described in TGA.10 Furthermore, on review of the autopsy records at the Hospital for Sick Children, in the past 10 years, we found that 3 of 7 newborns with TGA/intact ventricular septum and preoperative death had established pulmonary vascular changes. All 3 had refractory hypoxemia and acidosis after birth despite balloon atrial septostomy in 2, with death occurring within 5 days of life. An early arterial switch operation even before complete stabilization,8 9 with postoperative use of extracorporeal membrane oxygenation, if necessary, may improve the mortality of neonates with TGA and significant pulmonary hypertension.
Limitations
Because of the retrospective nature of our study, several of the
antenatal cases were examined only once early in gestation, hence it is
difficult to extrapolate the status of the ductus arteriosus and the
foramen ovale later in gestation. From those cases with serial studies,
however, it would appear that the abnormalities of the foramen ovale
and ductus arteriosus become more obvious later in gestation.
Conclusions
Fetuses with TGA and functionally intact
ventricular septum often have abnormal features of the
foramen ovale and the ductus arteriosus, which may be related to the
pathophysiology of TGA in utero. A fixed septum primum with little
mobility or a redundant septum primum prenatally may suggest the
potential for restriction of the foramen ovale at birth. A restrictive
foramen ovale in combination with ductal constriction, which can also
be identified prenatally, may be associated with very early neonatal
death despite aggressive resuscitation. We recommend that follow-up
fetal echocardiography be performed near term in
fetuses with TGA to reassess the ductus arteriosus and the foramen
ovale before delivery. Prenatal detection of restriction of the foramen
ovale with or without constriction of the ductus arteriosus with
better-planned perinatal management should improve the neonatal outcome
of infants prenatally diagnosed with TGA.
| Acknowledgments |
|---|
Received July 2, 1998; revision received October 27, 1998; accepted November 23, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. M. Gardiner, C. Belmar, G. Tulzer, A. Barlow, L. Pasquini, J. S. Carvalho, P. E.F. Daubeney, M. L. Rigby, F. Gordon, E. Kulinskaya, et al. Morphologic and Functional Predictors of Eventual Circulation in the Fetus With Pulmonary Atresia or Critical Pulmonary Stenosis With Intact Septum J. Am. Coll. Cardiol., April 1, 2008; 51(13): 1299 - 1308. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-M. Jouannic, L. Gavard, L. Fermont, J. Le Bidois, S. Parat, P. R. Vouhe, Y. Dumez, D. Sidi, and D. Bonnet Sensitivity and Specificity of Prenatal Features of Physiological Shunts to Predict Neonatal Clinical Status in Transposition of the Great Arteries Circulation, September 28, 2004; 110(13): 1743 - 1746. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. F. Goncalves, J. Espinoza, R. Romero, W. Lee, B. Beyer, M. C. Treadwell, and R. Humes A Systematic Approach to Prenatal Diagnosis of Transposition of the Great Arteries Using 4-Dimensional Ultrasonography With Spatiotemporal Image Correlation J. Ultrasound Med., September 1, 2004; 23(9): 1225 - 1231. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1999 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |