1. Pathophysiology of Congenital Heart Disease in the Adult
Part III: Complex Congenital Heart Disease
Robert J. Sommer, MD; Ziyad M. Hijazi, MD, MPH; John F. Rhodes, MD
With the successes in cardiothoracic surgery and pediat-
ric cardiology over the past 3 decades, for the first
time, adults with congenital heart disease (CHD) outnumber
their pediatric counterparts.1,2 As a result, adult patients with
CHD are beginning to appear more frequently in the practices
of adult cardiologists. The present series is designed to
provide a review of the pathophysiology and natural history
of common congenital heart problems that are now being
seen by adult cardiologists. In the first 2 parts of the series,
simple shunts and congenital obstructive lesions were re-
viewed. This final chapter will examine the physiology and
natural history of and the indications for intervention in
common complex congenital cardiac malformations seen in
adult patients. These patients include those with single-
ventricle physiology who are reaching adulthood in signifi-
cant numbers for the first time, the early survivors of
innovative surgical techniques developed a generation ago.
The combination of shunts, obstructive lesions, chamber
hypoplasia, and abnormal arterial and venous connections
that is seen in this group of patients creates some of the most
interesting and complex changes in the normal physiology of
the heart. Each patient must be considered individually,
because small differences in septal defect size or pathway
obstruction may have an enormous impact on the overall
effectiveness of the circulation and on management of the
patient. For the purposes of this article, only the most
common types of complex CHD will be considered (those
that are most likely to be seen in the adult cardiologist’s
office). However, the principles of flow and resistance, as
reviewed in the prior 2 portions of this series and as related
below, are easily generalized to give the practicing physician
insight into the physiological consequences of a myriad of
rare cardiac malformations.
Adults with complex CHD can be divided into 2 groups:
those who have not previously had an intervention and those
who have. The former will present with new-onset symptoms
in adulthood or may be identified as a result of a physical
examination with a new physician, an abnormal ECG, or
before they begin a new job. Children in the latter group who
have had good outcomes from surgical interventions are
frequently lost to follow-up during adolescence. They may
return for care in early adulthood with ongoing or new
symptoms or for clearance for employment, higher education,
insurance, or sports. These patients are often unfamiliar with
their diagnoses and their surgical history, have long since
stopped taking medication, and have adapted to a lifestyle
that fits their level of cardiac function.
Adults Presenting With No Prior Surgery
Like many patients with simple congenital lesions, some
adults with complex CHD may reach adulthood without
having had a prior intervention, or occasionally without even
having had a diagnosis. These patients generally have re-
mained undiagnosed because of the absence of significant
clinical symptoms, although the lack of diagnosis may also
result from the limited availability of medical care in some
parts of the world.
Ebstein’s Anomaly of the Tricuspid Valve
Ebstein’s anomaly of the tricuspid valve (TV) is a well-
described congenital malformation3 in which the septal leaflet
of the TV is conjoined to the septal surface well below the
valve annulus into the body of the right ventricle (RV; Figure
1). The other 2 leaflets of the valve elongate to coapt with the
abnormal septal leaflet, displacing the resulting coaptation
point into the RV outflow tract.
Pathophysiology
There are 2 significant hemodynamic outcomes of the mal-
formation. First, coaptation is rarely adequate, so most
patients have at least moderate TV regurgitation. Second, the
effective RV chamber size is reduced (and the right atrial
[RA] size increased) by a magnitude that corresponds to the
displacement of the TV. The small size of the RV diminishes
its filling capacity, which results in increased impedance to
filling. RA filling pressures are elevated, exacerbated by the
tricuspid regurgitation. In more severe cases, venous conges-
tion may result. The frequent association of an atrial septal
From the Center for Interventional Vascular Therapy (R.J.S.), Cardiovascular Research Foundation, Columbia University Medical Center, New York,
NY; Department of Pediatrics and Medicine (Z.M.H.), Rush Center for Congenital and Structural Heart Disease, Rush University Medical Center,
Chicago, Ill; and Department of Pediatrics (J.F.R.), Division of Pediatric Cardiology, Duke University Medical Center, Durham, NC.
This article is Part III in a 3-part series. Part I appeared in the February 26, 2008, issue (Circulation. 2008;117:1090–1099) and Part II appeared in
the March 4, 2008, issue (Circulation. 2008;117:1228–1237).
Correspondence to Robert J. Sommer, MD, Director, Invasive Adult Congenital Heart Disease, Assistant Professor of Clinical Medicine and Pediatrics,
Cardiovascular Research Foundation, Columbia University Medical Center, New York, NY 10032. E-mail CHDDoctor@aol.com
(Circulation. 2008;117:1340-1350.)
© 2008 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.107.714428
1340
Special Report

2. communication, either an atrial septal defect (ASD) or a
patent foramen ovale (PFO), will allow right-to-left shunting
with RA pressure elevation, particularly with exertion (when
cardiac output is increased and the RV is asked to accept
more volume).
Natural History
Clinical symptoms are highly variable and are related prin-
cipally to the degree of TV displacement. With mild disease,
the patient may be asymptomatic, and the diagnosis may be
discovered incidentally during an echocardiogram performed
for other reasons. The patient may present with a murmur of
tricuspid regurgitation or with supraventricular tachycardia
(accessory bypass tracts are a common association4). With
more severe valve displacement and an intact atrial septum,
the patient may present with exercise intolerance. Although
the small RV may be able to produce a normal resting output,
it is preload limited, which limits its stroke volume. As
demands increase on the “maxed out” RV, left ventricular
(LV) preload requirements will not be met, which limits
systemic cardiac output. Signs of systemic venous congestion
may be present with exertion.
The more common presentation is that of a patient who
also has an ASD or PFO. These patients may also present
with exercise intolerance. However, with an atrial communi-
cation, when RV capacitance becomes inadequate and RA
pressure rises with exertion, the patient will develop a
right-to-left shunt at the atrial level. In contrast to the patient
with an intact septum, when RV output cannot meet the
preload requirements of the LV, the right-to-left shunt aug-
ments/normalizes LV preload and output, at the expense of
systemic arterial hypoxemia. In this population, even though
resting oxygen saturation may be normal at rest, exercise
intolerance is usually due to falling saturations, not to low
cardiac output. This can easily be proven during exercise
stress testing. With an atrial communication and right-to-left
shunting, these patients may also present with paradoxical
thrombotic embolization.
Indications for Intervention
The management of the patient with Ebstein’s anomaly
depends entirely on the degree of symptoms. In asymptomatic
patients, no intervention is required. In patients with arrhyth-
mia who are otherwise symptom-free, medical management
or radiofrequency ablation5 may be their only need. For
patients who present with exercise-induced cyanosis who are
normally or nearly normally oxygenated at rest, transcatheter
closure of the PFO/ASD may be an excellent option.
We typically assess the physiological impact of PFO/ASD
closure in the catheterization laboratory with temporary
balloon occlusion of the septal communication, which elim-
inates the shunt to the left heart. We can measure the
corresponding rise in RA pressure and the associated drop in
cardiac output (Fick technique) as the right-to-left flow is
eliminated. In patients with low resting filling pressures and
normal or nearly normal arterial saturations at rest, these
changes should be minimal, and closure of the defect may
proceed. In patients in whom RA pressure exceeds 15 mm Hg
with balloon occlusion, those in whom there is a significant
jump in RA pressure from a low baseline pressure to
Ͼ15 mm Hg with balloon occlusion, and those in whom there
is a substantial fall in cardiac output with balloon occlusion,
the atrial communication should likely be left open. The
tradeoff of low cardiac output for full oxygen saturation is
often not beneficial for the patient, because tissue oxygen
delivery (the product of oxygen content and cardiac output) is
usually less after elimination of the shunt, as has been demon-
strated with a number of other congenital malformations.6
In the unusual case of an adult patient presenting with a
small RV volume, tricuspid regurgitation, and true right-sided
heart failure symptoms, surgical reconstruction or replace-
ment of the TV in the annular position is probably the best
option.7 An alternative surgical approach, which has been
used primarily in children, is the “11/2 ventricle repair,” in
which a bidirectional Glenn shunt is performed (superior
vena cava disconnected from the RA and anastomosed
directly to the right pulmonary artery). This allows the blood
returning from the upper body to drain directly to the lungs
and back to the LV, whereas the small RV is unloaded and
needs to handle only return from the inferior vena cava.8 As
a rule, most patients with Ebstein’s anomaly and significant
symptoms due to hypoplasia of the RV will have undergone
prior surgery as children, receiving either restorative valve
surgery or single-ventricle repairs.
L-Transposition of the Great Arteries
(Physiologically “Corrected” Transposition)
L-transposition of the great arteries (L-TGA) is the result of
abnormal ventricular looping of the developing fetal heart, in
which the primitive ventricle moves to the right and the
bulbus cordis moves to the left.9 Because these structures are
the precursors of the LV and RV, respectively, the morpho-
Figure 1. Ebstein’s anomaly of the TV. The septal leaflet of
the TV (white arrowhead) is displaced apically into the body
of the RV, which diminishes the effective diastolic capacity of
the chamber. Blood may flow across the PFO (black arrow)
from RA to LA as RA pressures rise.
Sommer et al Complex Congenital Heart Disease 1341

3. logical RV and TV end up on the left side of the heart,
receiving blood from the left atrium (LA) and pumping blood
to the aorta, whereas the morphological LV and mitral valve
receive inflow from the RA and deliver deoxygenated blood
to the lungs (Figure 2). The aortic root, arising from the RV,
is anterior and leftward of the pulmonary artery root (“L”
transposed). The inverted atrioventricular (AV) connections,
combined with the reversed ventriculoarterial connections,
“correct” the flow pattern such that deoxygenated systemic
venous blood flows to the lungs, and oxygenated pulmonary
venous blood is pumped to the systemic circulation. Although
ventricular septal defects (VSDs) and pulmonary outflow
obstructions are frequently associated with L-TGA, patients
without associated lesions have nearly normal cardiac phys-
iology and may easily be undiagnosed until adulthood.
Natural History
Because the initial physiology of isolated L-TGA is normal,
children rarely have cardiac symptoms or even heart mur-
murs. It is the potential late failure of the RV and TV, which
both face the higher-resistance systemic arterial circuit, that
most frequently brings these patients to attention as young
adults. In this group, progressive dilation of the RV, as the
myocardium fails, typically leads to enlargement of the TV
annulus and worsening of the tricuspid regurgitation. The
volume load of the tricuspid regurgitation, in turn, worsens
RV chamber dilation, which further stretches the tricuspid
annulus. This progressive process has comparable physiolog-
ical consequences to those of a patient with a failing LV who
develops mitral regurgitation. Longitudinal studies have de-
scribed outcomes in patients with and without surgical
intervention.10,11
Resting ECGs typically have a q wave in the left precordial
leads (due to anatomic reversal of the septum itself), which
may also bring these patients to the attention of the adult
cardiologist. Complete heart block is a frequent accompany-
ing or presenting symptom in this population owing to the
associated abnormal development of the conduction system.12
Other associations include an Ebstein-like displacement of
the left-sided TV (common), which may contribute to TV
dysfunction, and ventricular noncompaction (rare), which
may contribute to ventricular dysfunction.13
Indications for Intervention
Management of patients with isolated L-TGA is symptom-
dependent. Most often, patients present with signs of conges-
tive heart failure, and management may be attempted with the
usual array of medications that would be considered in a
patient with a failing LV and a regurgitant mitral valve
(digoxin, diuretics, and afterload-reducing agents). There is
limited evidence that these therapies are as effective in the
L-TGA population. Care must be taken in patients with
underlying A-V conduction issues when drugs such as
␤-blockers or calcium channel blockers are used. TV replace-
ment therapy is reserved for those patients with severe tricuspid
regurgitation who have well-preserved RV function.14
For patients with symptomatic heart block, pacemakers
may be placed transvenously, as in any other patient, al-
though the ventricular lead will end up in the right-sided
morphological LV. Cardiac resynchronization therapy has
been used in small series for heart failure symptoms, with
some apparent benefit in this population.15 Heart transplan-
tation is a final common pathway for patients with severe
heart failure symptoms. The unusual relationship between the
aortic and pulmonary trunks is the unique complexity facing
the transplant surgeon in fitting the new organ into the
patient.
Unrepaired Tetralogy of Fallot
Tetralogy of Fallot is the most common complex cyanotic
congenital heart lesion. It also has the longest surgical history
and the most intensively studied outcomes and follow-up data
of all congenital cardiac anomalies. Adult patients with
unrepaired tetralogy of Fallot are extremely rare; however, in
areas where patients have no access to health care, particu-
larly in developing countries, some of these patients may
survive to adulthood.
The 4 distinct components of the tetralogy, as described in
1888 by Fallot, include VSD, subpulmonary stenosis, over-
riding aorta, and RV hypertrophy (Figure 3). However,
embryologically, the defect appears to be a single develop-
mental error8 that involves the terminal portion of the spiral
septum, which divides the primitive truncus arteriosus from
the pulmonary artery. In tetralogy of Fallot, this muscle
bundle is displaced rightward and anteriorly, which precludes
correct fusion with the growing muscular ventricular septum,
narrowing the pathway from RV to pulmonary artery and
Figure 2. L-TGA. The morphological LV and RV are inverted.
Deoxygenated systemic venous blood (black arrow) is pumped
to the pulmonary artery by the LV, whereas oxygenated pulmo-
nary venous return (white arrow) is pumped to the aorta by the
RV. Ao indicates aorta; MPA, main pulmonary artery.
1342 Circulation March 11, 2008

4. enlarging the aortic root such that it extends over the RV
outflow tract. The RV hypertrophy is a secondary response to
increased afterload. Obstructions in the branch pulmonary
arteries, coronary anomalies, a right-sided aortic arch, and
additional VSDs are common additional associations.
In tetralogy of Fallot, systemic and pulmonary venous
return, the RA and LA, the atrial septum, and the AV valves
are typically normal. Blood ejected from the ventricles has 2
possible pathways: through the anatomically aligned outflow
tract or through the VSD to the opposite outflow. The
direction and magnitude of blood flow from each ventricle
will be determined by the relative resistance of each pathway.
With subpulmonary valvar obstruction, hypoplasia of the
pulmonary valve, narrowing of the main pulmonary artery
trunk, and the frequently associated branch pulmonary artery
stenoses, the pathway from the RV to the lungs in tetralogy of
Fallot is typically a high-resistance pathway (the net resis-
tance of obstructions in series is the sum of the individual
resistors). The combination of a high-resistance pulmonary
pathway and a nonrestrictive VSD creates the hemodynamic
conditions for significant right-to-left shunting at the ventric-
ular level, which results in systemic arterial desaturation.
Natural History
The resistance of the pulmonary path is extremely variable
from 1 patient to the next. If the obstructive lesions are mild,
the net resistance to pulmonary flow will be low compared
with the resistance to systemic arterial flow. Blood from the
RV will flow predominantly to the lungs, with only a small
percentage crossing to the aorta. These patients will be
neither significantly cyanotic nor symptomatic, because the
systemic and pulmonary blood flows will be nearly equal.
The narrowing of the RV outflow tract causes flow turbu-
lence, which results in the systolic ejection murmur that
usually brings these patients to attention.
With more severe obstructions in the pulmonary pathway,
pulmonary flow (Qp) will be significantly less than systemic
flow (Qs) as more systemic venous blood crosses to the aorta
through the VSD. The patient will present with cyanosis and
exertional dyspnea due to poor tissue oxygen delivery. With
exercise and increasing myocardial contractility, which can
narrow the subpulmonary area further, or with a rise in
pulmonary vascular resistance, the net resistance of the
pulmonary pathway may increase acutely, augmenting the
right-to-left shunt. This results in a precipitous drop in
pulmonary blood flow and profound episodic cyanosis, the
so-called “tet spell.” Patients with long-standing unrepaired
tetralogy of Fallot make behavioral adaptations to respond to
these physiological crises. With falling oxygen saturation,
they learn to squat, compressing the arteries in the lower
extremities, which raises the resistance of the aortic pathway
and restores the baseline level of pulmonary blood flow.
Indications for Intervention
Any patient with unrepaired tetralogy of Fallot should be
considered for intervention. By adulthood, they are typically
cyanotic, profoundly exercise intolerant, and polycythemic16
and are at risk of arterial occlusion, hemoptysis, cerebral
abscess, and other life-threatening issues. When a complete
repair is anatomically and physiologically possible, correc-
tion of the defect provides the patient with a vastly superior
quality of life, although the surgery carries somewhat higher
mortality and morbidity in the adult than in the child.17
Palliative procedures such as aorta-to–pulmonary artery
shunts and transcatheter balloon dilation of the RV outflow
tract may improve arterial saturation and reduce symptoms as
a result; however, experience with these palliations is princi-
pally in children. Acute increases in pulmonary blood flow
correspondingly increase pulmonary venous return, which
places an acute volume load on the LV. The less compliant
LV of the adult may not handle the volume load as easily as
the pediatric ventricle, and new symptoms of congestive heart
failure are possible.
Tetralogy of Fallot With Pulmonary Atresia
In the most severe cases of spiral septum malalignment, the
pulmonary valve may be atretic, or the main pulmonary artery
trunk may not form at all.8 In some of these newborns, the
ductus arteriosus may be the only source of pulmonary blood
flow (blood flows from the high-resistance aorta to the
lower-resistance pulmonary arteries). In the first few days of
life, as the ductus begins to close, these patients will become
profoundly cyanotic and will require medical intervention
(prostaglandin E1) to maintain ductal patency, followed by a
surgical reconstruction of the RV outflow tract.
In some patients with tetralogy of Fallot with pulmonary
atresia, there may be no true (or visible) central pulmonary
arteries. Major aortopulmonary (bronchial) arterial collaterals
Figure 3. Tetralogy of Fallot. Malalignment of conal septum
results in overriding aorta, subvalvar pulmonary stenosis, and
VSD. With narrowing of RV outflow, desaturated blood flow will
be divided (black arrows), a portion to the lungs and a portion
across the VSD to the aorta, depending on the relative resis-
tance of each pathway. A right aortic arch is shown. Ao indi-
cates aorta; MPA, main pulmonary artery.
Sommer et al Complex Congenital Heart Disease 1343

5. develop in utero to supply blood to the developing lungs.
Frequently, a normal or nearly normal pulmonary blood flow
is maintained. As a result, the diagnosis can be missed in the
newborn period, because the infant has few symptoms. They
are usually diagnosed later, when crying exacerbates their
baseline hypoxemia or when continuous murmurs in the lung
fields are recognized. Because they are not ductal dependent,
these patients may reach adulthood without intervention;
however, in most cases, the collateral vessels fail to grow
proportionally to the rest of the child.18 The vessels will
become stenotic and create more resistance to pulmonary
flow. Surgical or transcatheter augmentation of pulmonary
blood flow may be required.
When collaterals are numerous and large enough, exces-
sive pulmonary blood flow can lead to symptoms of conges-
tive heart failure in infancy, potentially requiring medical,
surgical, or transcatheter embolic therapy. With the natural
tendency of the vessels to become stenotic, however, the
symptoms often resolve as the child grows. In cases in which
the individual collaterals are very large and short and do not
become stenotic, the resistance of that pathway will remain
very small. That segment of lung will be overperfused at high
pressure relative to other segments. This can result in the
unique physiology of segmental pulmonary vascular disease
(Eisenmenger syndrome). Bull and associates19 in Great
Britain have suggested that these patients may do equally
well if left alone rather than undergoing a complex series of
operations to reconstruct the pulmonary arteries and then
close the VSD.
Patients With Prior Surgery in Childhood
Most adult patients with complex CHD will have undergone
a prior (or multiple prior) intervention in early childhood and
may present again as adults with new or persistent symptoms.
This section will be limited to those patients with common
lesions who have undergone restorative or definitive pallia-
tive procedures.
D-Transposition of the Great Arteries
D-transposition of the great arteries (D-TGA), another early
embryological malformation, results in the embryological
inversion of the great arteries in an otherwise normally
developing heart.8 The aortic root is positioned anterior and to
the right of the pulmonary artery (“D” transposed) and
becomes the outlet for the RV. The pulmonary artery arises
from the LV (Figure 4).
Unlike L-TGA, D-TGA most often occurs without associ-
ated cardiac anomalies and is incompatible with postnatal
life. In patients with D-TGA, systemic and pulmonary venous
return is normal, with normal atrial and ventricular connec-
tions; however, because the aorta arises from the RV,
deoxygenated, systemic venous blood is returned directly to
the aorta and to the body without passing through the lungs to
become oxygenated. Similarly, because the pulmonary artery
is the outlet for the left ventricle, pulmonary venous blood
loops from the left heart to the lungs. Rather than 2 arterial
circuits in series, the neonate with D-TGA has 2 separate
independent circulations in parallel. Systemic oxygen satura-
tion falls to very low levels within a few cardiac cycles,
because it is not replenished. Anaerobic respiration ensues at
the cellular level, which results in a profound systemic
arterial acidosis. Without emergent intervention to improve
oxygenation of the systemic arterial blood, the baby will die.
Because of the natural communications present at birth that
allow the normal transition from the fetal to the postnatal
circulation, the vast majority of infants with D-TGA will
survive for hours or days. The ductus arteriosus allows for an
aorta-to–pulmonary artery shunt with falling pulmonary vas-
cular resistance. This additional pulmonary flow augments
pulmonary venous return to the LA. With LA pressure higher
than RA pressure, an LA-to-RA shunt can develop through the
patent foramen ovale. This left-to-right flow brings oxygenated
blood to the RA, which raises the oxygen content of the
mixed blood that ultimately reaches the aorta. As long as the
communication at the atrial level is adequate, and as long as
LA pressure exceeds RA pressure, the baby can survive.
In some neonates, when the ductus arteriosus begins to
close and pulmonary venous return falls, the favorable
hemodynamics will change, and systemic oxygen levels will
fall sharply. Reopening the ductus arteriosus with prostaglan-
din E1 will improve systemic oxygenation in this patient. In
others, the flap valve of the PFO is more competent and
severely limits left-to-right flow at the atrial level immedi-
ately after delivery. In these patients, even with a widely
patent ductus arteriosus and a large shunt at that level, there
will be little shunting at the atrial level, and systemic
oxygenation will be inadequate. The first transcatheter inter-
vention for CHD, the balloon atrial septostomy, was devel-
oped specifically for these patients by Dr William Rashkind
in 196520 and is still in use today. This palliation, now
Figure 4. D-TGA. The aorta and main pulmonary artery are
reversed, which creates 2 parallel circulations. Deoxygenated
systemic venous blood (black arrow) is pumped back to the
aorta, whereas the oxygenated pulmonary venous return (white
arrow) is cycled back and forth to the lungs. Ao indicates aorta;
MPA, main pulmonary artery.
1344 Circulation March 11, 2008

6. frequently performed under echocardiographic guidance at
the bedside in the nursery rather than in the catheterization
laboratory, enlarges the atrial communication, eliminating it
as a resistor to left-to-right flow. Systemic oxygen levels
increase dramatically and immediately with a successful
septostomy, which allows the infant to survive to undergo
more definitive surgical therapy.
Infant surgery for D-TGA has enabled these children to
reach adulthood in large numbers. However, surgical tech-
niques for repairing D-TGA have changed over the past 5
decades, and the clinical issues that present to cardiologists in
adulthood will depend on which surgical approach was used.
The first successful surgical corrections for D-TGA, the
Senning21 (1958) and Mustard22 (1964) procedures, both
involve a redirection of venous flow at the atrial level. In each
of these “atrial switch” procedures, pulmonary venous blood
is “baffled” to the TV and flows into the RV, which pumps it
to the aorta, supplying oxygenated blood to the systemic
circulation (Figure 5). As a result of the baffle, systemic
venous blood is excluded from the TV and flows instead
through the mitral valve to the LV and to the pulmonary
artery for reoxygenation. These operations enabled the long-
term survival of babies with D-TGA, many of whom have
now reached their fourth and fifth decades of life.
In 1976, Jatene et al23 first described a successful technique
for “anatomic” correction of D-TGA. In that approach, the
transposed aorta and pulmonary arteries are transected imme-
diately above their respective valves and reattached to the
opposite semilunar valve (arterial switch operation). A “but-
ton” of tissue surrounding each coronary ostium is excised
from the old aortic root and is transplanted with the coronar-
ies into the newly created aorta root. This operation restores
the appropriate atrioventricular and ventriculoarterial con-
nections. Since the mid-1980s, the arterial switch opera-
tion has evolved as the surgical treatment of choice, replacing
atrial switch operations. Physiologically, these children are
quite normal and are now reaching adulthood for the first
time.
“Natural History” After an Atrial
Switch Operation
Although long-term outcomes with atrial switch procedures
to date have been quite encouraging (actuarial survival from
the Mustard procedure was 80% at 28 years, with 76% of
survivors functioning at New York Heart Association class
I24), the fundamental problem with the atrial switch approach
is the surgical creation of a circulation close to that of
L-TGA, with the RV and TV facing the systemic circulation.
Late RV failure and tricuspid regurgitation are common
causes of morbidity and mortality in this population. Surgical
replacement of the TV may be helpful in a minority of
patients who present with tricuspid regurgitation and pre-
served RV function. As in the L-TGA population, resynchro-
nization therapy has been attempted15; however, as with
patients with L-TGA, many may ultimately require transplan-
tation25 for symptoms of congestive heart failure.
In addition, with extensive surgery at the atrial level,
particularly near the junction of the superior vena cava and
RA, these patients have a high incidence of bradycardia and
atrial tachyarrhythmias.26,27 In a Canadian series, at follow-up
20 years after surgery, only 40% of patients with an atrial
switch were in sinus rhythm.28 The presence of atrial
tachycardia has been cited as a predictor of sudden death in
this population.29 Medical management of the tachyarr-
hythmias carries similar risks to that of other patients with
sinus node dysfunction. Pacemaker and radiofrequency abla-
tive therapy can be quite difficult due to the complexity of the
atrial anatomy after this surgery.
The “baffles” created within the pediatric atria may be-
come obstructed with patient growth and may require surgical
or interventional therapy30,31 to relieve symptoms of systemic
or pulmonary venous obstruction. In intervening, care must
be taken that enlargement of 1 pathway does not impinge on
the pathway from the opposite set of veins. Small leaks in the
baffles may also be present but are generally hemodynam-
ically unimportant and are physiologically identical to small
atrium-level shunts in other types of patients. Larger defects
that allow significant shunting can be closed with transcath-
eter devices.
“Natural History” After an Arterial
Switch Operation
In light of the restoration of normal anatomic connections,
most children with the arterial switch operation have normal
cardiac function and physiology. In a German series of 188
patients who underwent an arterial switch, survival was 91%
at 10 years, with 96% of patients unlimited in their physical
Figure 5. Atrial switch operation (Mustard/Senning) for D-TGA.
A “chamber” is created along the back wall of the atria, such
that pulmonary venous return is redirected to the TV, to flow to
the RV and aorta (white arrow). Systemic venous blood from
superior and inferior vena cavae (black arrows) is excluded from
the TV and must flow over the baffle to the mitral valve, then to
the LV and lungs. PV indicates pulmonary vein; IVC, inferior
vena cava; and SVC, superior vena cava.
Sommer et al Complex Congenital Heart Disease 1345

7. activity, and 99% taking no medication.32 Most clinical issues
after the operation are the result of surgical residua from the
initial operation in infancy.
Coronary lesions are rare, with myocardial ischemia very
uncommon after reimplantation of the vessels. In a series of
755 arterial switch procedures, coronary perfusion issues
were detected in 34 children, with documented ischemia in
only 19.33 There are no data available yet on the risk of
coronary complications later in adulthood.
The other major clinical issue after the arterial switch is
related to anatomic distortion of the great vessels. Residual
supravalvar stenosis at the aortic and pulmonary reanastomo-
sis sites may be seen,34 but clinically significant obstructions
are most often resolved in the pediatric age group with
transcatheter or surgical therapy. Progressive dilatation and
valvar insufficiency of the reconstructed aortic root may be
problematic and may necessitate valve replacement later in
life.35 Branch pulmonary artery stenosis may result from
pulling the main pulmonary artery trunk forward to anasto-
mose it to the former aortic root.36 This may be a clinically
important afterload for the RV or may significantly redistrib-
ute pulmonary blood flow from 1 lung to the other if the
degree of obstruction of the branches is unequal. These
obstructive lesions have been treated successfully in the
pediatric population with balloon and stent angioplasty.37
Because arterial switch patients are now reaching adulthood
in growing numbers, the long-term outcome of Jatene’s
operation continues to be unveiled.
“Natural History” After Surgical Intervention for
Tetralogy of Fallot
In the late 1940s, before the era of open heart surgical repairs,
palliative shunts were developed to carry blood directly from
the aorta to the pulmonary artery, bypassing the obstruction in
the RV outflow tract.38 The Blalock-Taussig shunt (subcla-
vian artery to pulmonary artery), the Waterston shunt (direct
anastomosis of ascending aorta to right pulmonary artery),
and the Potts shunt (direct anastomosis of descending aorta to
left pulmonary artery) were common in the 1950s and 1960s
and enabled survival to adolescence and adulthood. Shunts
were later used in the short-term palliation of neonates with
severe cyanosis, so that complete open heart repairs could be
performed more safely in an older/larger child.
Complete surgical repair of tetralogy of Fallot separates
systemic and pulmonary venous return and creates unob-
structed flow to both great arteries. Closure of the VSD is
accomplished with an angled patch that directs LV flow to the
aorta. Repair of the RV outflow tract and pulmonary arteries
may entail subvalvar muscle resection, pulmonary valvot-
omy, transannular patching to enlarge a hypoplastic valve
annulus, and patch angioplasty of the main or branch pulmo-
nary arteries. Currently, corrective surgery is typically per-
formed in the first year of life.39 With good surgical results,
survival and quality of life have been excellent. In a retro-
spective analysis from the Mayo Clinic, 86% of patients were
alive at 32 years,40 and in a series of 151 patients from
Toronto, Canada,41 94% were in New York Heart Association
functional class I. Patients who have had only palliative
shunts remain cyanotic and have many of the long-term
issues of their counterparts who have not undergone repair
(see above); however, in patients who have never undergone
complete repair, longstanding or multiple shunt procedures
may severely distort or interrupt the branch pulmonary
arteries and may preclude more definitive repairs.
With a completed tetralogy of Fallot repair, normal cardiac
physiology is restored. However, a number of hemodynamic
issues may be seen in adults who have undergone surgical
correction as children, including residual RV outflow ob-
struction, residual obstruction in the branch pulmonary arter-
ies, residual shunts (at atrial, ventricular, or arterial levels),
significant right-sided valvar insufficiency, aortic insuffi-
ciency, dilated and poorly contractile ventricles,42–45 heart
block, and reentrant ventricular arrhythmia due to scarring in
the ventricles.46 Sudden cardiac death may occur in patients
with residual disease and may be predictable with electro-
physiological evaluation.47 Antitachycardia pacing may ben-
efit some patients who are established to be at greatest risk for
life-threatening arrhythmia.48
Most often, adult tetralogy of Fallot patients present with
new-onset or increasing exercise intolerance and must be
evaluated to rule out both potentially important surgical
residua and ventricular myocardial issues. RV dilatation and
dysfunction are common in this group and are most often
related to long-standing pulmonary regurgitation in patients
who have had surgical manipulation of the valve. With
dilatation of the RV and the tricuspid annulus, tricuspid
regurgitation may be secondary but certainly compounds the
issue of RV volume overload. The RV with preserved systolic
function typically responds well to pulmonary valve replace-
ment in many patients,49 with or without concurrent TV
repair. LV dysfunction is common in patients who underwent
surgery before 1970 and is most often related to suboptimal
myocardial protection during surgery or to injury of unsus-
pected anomalous coronary branches. MRI has proven ex-
tremely valuable in the quantitative assessment of ventricular
and valvar function in this population of patients who are
typically difficult to image with transthoracic echocardiogra-
phy50 due to surgical scarring and anatomic distortion. Pa-
tients who are free of clinical symptoms are frequently lost to
follow-up, but ongoing care and screening are essential
because the population remains at risk for ventricular dys-
function, ventricular arrhythmia, and, in some cases, sudden
death.
The Adult Patient With Single-Ventricle
Physiology (Fontan Operation)
Patients with functional single ventricles, those with absent or
hypoplastic pumping chambers, absent or hypoplastic AV
valves, or complex VSDs that cannot be closed, are reaching
adulthood in large numbers for the first time. Simple pallia-
tions such as shunts or pulmonary artery banding were
initially performed to maintain quality of life for a few years
in patients with appropriate anatomy. Long-term survival was
not seen until the introduction in 1971 of the Fontan opera-
tion,51 a more definitive palliative approach in which the
systemic venous return is rerouted to completely bypass the
right heart. Initially used for patients with tricuspid atresia,
the concept has now been applied to congenital lesions of all
1346 Circulation March 11, 2008

8. types in which the normal 2-ventricle circulation cannot be
fully restored.
The current surgical approach to creating the Fontan
physiology involves a staged rerouting of the systemic
venous return. In the newborn period, a lesion-specific initial
palliation will be performed, which, regardless of the initial
anatomy, will leave the patient with a common physiology:
complete mixing of systemic and pulmonary venous return,
unobstructed pulmonary venous return to the ventricle, an
unobstructed outflow tract to the systemic arterial circulation,
and a limited but reliable source of pulmonary blood flow.
These patients are able to survive to later palliation but face
the physiological consequences of ongoing cyanosis (com-
plete mixing of systemic and pulmonary venous blood) and a
volume-loaded ventricle (which must have enough volume to
pump blood to both the systemic and pulmonary circuits).
The second stage of palliation typically occurs between 3
and 6 months of life. A bidirectional Glenn shunt or hemi-
Fontan procedure52 interrupts flow between the superior vena
cava and the atrium and redirects superior vena caval flow
directly to the pulmonary arteries. The creation of the
cavopulmonary shunt improves the physiological conditions
of the infant. Because blood from the inferior vena cava
continues to mix with pulmonary venous return in the single
ventricle, the patient remains desaturated. However, oxygen-
ation of the deoxygenated systemic venous return is more
efficient than when mixed arterial blood is delivered to the
pulmonary artery, and the fraction of “blue” blood in the
ventricle is smaller, which results in higher systemic oxygen
saturations after the Glenn shunt (85% to 90%) than after the
initial palliation. Finally, volume loading of the ventricle is
eliminated because it pumps blood only to the systemic
circulation.
The final stage of the palliation, the Fontan operation,
redirects inferior vena caval flow away from the atrium to the
pulmonary arteries,53 leaving only pulmonary venous blood
to be pumped by the ventricle to the aorta (Figure 6). These
patients become fully saturated after the surgery. Despite the
complete separation of systemic and pulmonary venous
blood, Fontan physiology is distinct from that of the normal
heart.
In contrast to a 2-ventricle circulation, in which blood is
pumped through 2 parallel circuits by separate pumps, the
single functional ventricle in a Fontan circulation must
generate enough energy to push blood through both the
systemic and pulmonary vascular beds consecutively. There
is no energy added by a second ventricle to help push blood
through the lungs. Flow from the systemic veins, through the
lungs and back to the ventricle, is entirely passive (Figure 7).
Low resistance in this pathway is therefore critical. Anatomic
obstructions in the pulmonary vasculature, elevation of pul-
monary vascular resistance, competitive arterial flow into the
lungs, AV valve stenosis or regurgitation, elevation of ven-
tricular end-diastolic pressure, or rhythm disturbances with
loss of AV synchrony may create critical impedance to flow.
“Natural History” of a Fontan Circulation
Today, children with Fontan physiology are reaching adult-
hood in unprecedented numbers. Initially after the surgery,
Ͼ90% of patients are in New York Heart Association class I
or II,54 with marked improvements in exercise capacity
(compared with their preoperative state) due to improved
tissue oxygen delivery. Unlike patients with repairs of simple
defects such as patent ductus arteriosus or ASD, Fontan
patients experience ongoing attrition even after decades of
symptom-free survival. This is due to ventricular failure;
breakdown and growth-related senescence of constructed
pathways; the development of atrial arrhythmia; fluid reten-
tion and electrolyte disturbances; low cardiac output; and
thrombotic and thromboembolic issues. The classic paper by
Fontan et al55 in the early 1990s proved the palliative, rather
than reparative, nature of the procedure. In that series, after
12 years, only 70% of “perfect” Fontan patients were alive,
and fewer than half of the survivors remained in New York
Heart Association class I or II.
Advances in the pre-Fontan management of these children,
including staging of surgical reconstruction and improved
surgical techniques, are likely to have resulted in better late
outcomes since the Fontan paper was published, but it is clear
that the procedure will never be curative in nature.56 Although
the acute surgical management of these patients remains a
pediatric cardiology issue, the long-term effects of this
unusual cardiac physiology are increasingly confronting adult
cardiologists.
Figure 6. External conduit-type Fontan operation in a patient
with pulmonary atresia and hypoplastic RV. Superior vena cava
(SVC) is disconnected from RA and connected directly to pul-
monary artery (bidirectional Glenn shunt), and inferior vena cava
(IVC) flow is rerouted through external conduit directly to pulmo-
nary artery. Systemic venous blood (black arrows) flows to
lungs; only pulmonary venous blood is pumped to aorta (Ao;
white arrow). RPA indicates right pulmonary artery; LPA, left
pulmonary artery.
Sommer et al Complex Congenital Heart Disease 1347

9. Indications for Intervention
As described above, any circulatory issue that increases the
pathway resistance from the systemic veins back to the
ventricle will impede the passive flow of the Fontan, leading
to higher systemic venous pressures. Most often, this results
in a spectrum of symptoms typical of a patient with classic
“right heart failure”: systemic venous congestion, fluid reten-
tion (particularly peripheral edema, ascites, and pleural and
pericardial effusions), and low cardiac output due to reduced
systemic ventricular preload.
Alternatively, patients with increased pathway resistance
may decompress the Fontan circuit through the development
of venous collaterals from the higher-pressure Fontan path-
way to the lower-pressure pulmonary veins or LA. This
right-to-left shunt will lower the Fontan pathway pressure and
maintain ventricular preload but will result in increasing
cyanosis. Rising pathway pressures may also create increas-
ing right-to-left flow through existing residual defects in the
surgically constructed pathway inside the atria.
When clinical symptoms bring Fontan patients to attention
in an adult cardiology setting, skilled assessment is required
in the catheterization laboratory. In congested low-output
patients, small pressure gradients across venous connections,
not readily detectable by noninvasive techniques, may repre-
sent a significant resistance to flow. Such obstructions in the
branch pulmonary arteries and at suture lines may be balloon
dilated or stented to minimize resistance to forward flow.57,58
Coil embolization of aorta to pulmonary artery collaterals
should be performed to eliminate high-pressure competitive
flow, when identified. AV synchrony should be restored with
the use of medications or a pacemaker for patients who are
bradycardic or in junctional rhythm. In cyanotic patients,
careful angiographic evaluation of the Fontan pathway is
critical. Embolization of venous collaterals or device closure
of intracardiac defects may be beneficial,59 but elimination of
the right-to-left shunt may exacerbate symptoms of low
output and venous congestion by eliminating the “pop-off”
valve from the high-resistance pulmonary circuit.
When no treatable anatomic issues are identified, medical
management is extremely limited for the patient with a failing
Fontan circulation. In some cases, normal changes in ventric-
ular compliance that occur with aging may increase end-di-
astolic pressures sufficiently to cause venous pathway con-
gestion. The creation of a Fontan fenestration (a shunt at the
atrial level) may be performed surgically or in the catheter-
ization laboratory59,60 as a further palliation, often as a bridge
to transplantation. Although some degree of cyanosis is
created, this intervention will increase both cardiac output
and tissue oxygen delivery. Physiologically, this mimics the
creation of a small ASD in patients with end-stage pulmonary
hypertension.61
Cardiac transplantation is a last option for the failing
Fontan. In general, transplantation is extremely well tolerated
physiologically in this population. Pulmonary hypertension is
never an issue as it is in cardiomyopathy patients, because
Fontan patients have survived with only passive pulmonary
flow. However, the complexity of reconstructing the systemic
and venous connections and the pulmonary arterial pathways,
along with the scarring from multiple prior cardiac surgeries,
makes the transplant operation itself extraordinarily
challenging.
Conclusions
CHD in the adult population, more than any other type of
malformation, is taxing the current medical system, because
the number and complexity of patient issues are outgrowing
the limited facilities established to care for them. With better
survival from childhood, it is evident that few of our surgical
and catheter-based interventions are curative and that a
lifetime of care will be required. The education of both
internal medicine/cardiology trainees and those already in
practice, as well as the development of national databases for
tracking patients and their responses to therapy are critical to
the future care of this complex and diverse population.
Clinical systems designed for the transition of these patients
out of the pediatric clinics where many continue to receive
their care and into organized, regional centers of excellence
for adult patients with CHD will facilitate this process.
Disclosures
None.
References
1. Care of the adult with congenital heart disease: presented at the 32nd
Bethesda Conference, Bethesda, Maryland, October 2–3, 2000. J Am Coll
Cardiol. 2001;37:1161–1198.
2. Hoffman JI, Kaplan S, Liberthson RR. Prevalence of congenital heart
disease. Am Heart J. 2004;147:425–439.
Figure 7. Flow through Fontan pathway is entirely passive, run-
ning “downhill” to the ventricle from the systemic veins. At each
step, potential impediments to flow may raise the resistance of
the pathway. If resistance is excessive, venous stasis will occur,
resulting in ventricular preload reduction and low output. PA
indicates pulmonary artery; PVR, pulmonary vascular resistance.
1348 Circulation March 11, 2008

10. 3. Bialostozky D, Horwitz S, Espino-Vela J. Ebstein’s malformation of the
tricuspid valve: a review of 65 cases. Am J Cardiol. 1972;29:826–836.
4. Cappato R, Schluter M, Weiss C, Antz M, Koschyk DH, Hofmann T,
Kuck KH. Radiofrequency current catheter ablation of accessory atrio-
ventricular pathways in Ebstein’s anomaly. Circulation. 1996;94:
376–383.
5. Reich JD, Auld D, Hulse E, Sullivan K, Campbell R. The Pediatric
Radiofrequency Ablation Registry’s experience with Ebstein’s anomaly.
J Cardiovasc Electrophysiol. 1998;9:1370–1377.
6. Bridges ND, Lock JE, Mayer JE Jr, Burnett J, Castaneda AR. Cardiac
catheterization and test occlusion of the interatrial communication after
the fenestrated Fontan operation. J Am Coll Cardiol. 1995;25:1712–1717.
7. Dearani JA, Danielson GK. Surgical management of Ebstein’s anomaly
in the adult. Semin Thorac Cardiovasc Surg. 2005;17:148–154.
8. Sarris GE, Giannopoulos NM, Tsoutsinos AJ, Chatzis AK, Kirvassilis G,
Brawn WJ, Comas JV, Corno AF, Di Carlo D, Fragata J, Hraska V,
Jacobs JP, Krupianko S, Sairanen H, Stellin G, Urban A, Ziemer G;
European Congenital Heart Surgeons Association. Results of surgery for
Ebstein anomaly: a multicenter study from the European Congenital
Heart Surgeons Association. J Thorac Cardiovasc Surg. 2006;132:
50–57.
9. Moore KL, Persaud TVN. The Developing Human: Clinically Oriented
Embryology. 7th ed. Philadelphia, Pa: Saunders; 2003:329–380.
10. Prieto LR, Hordof AJ, Secic M, Rosenbaum MS, Gersony WM. Pro-
gressive tricuspid valve disease in patients with congenitally corrected
transposition of the great arteries. Circulation. 1998;98:997–1005.
11. Graham TP Jr, Bernard YD, Mellen BG, Celermajer D, Baumgartner H,
Cetta F, Connolly HM, Davidson WR, Dellborg M, Foster E, Gersony
WM, Gessner IH, Hurwitz RA, Kaemmerer H, Kugler JD, Murphy DJ,
Noonan JA, Morris C, Perloff JK, Sanders SP, Sutherland JL. Long-term
outcome in congenitally corrected transposition of the great arteries: a
multiinstitutional study. J Am Coll Cardiol. 2000;36:255–261.
12. Daliento L, Corrado D, Buja G, John N, Nava A, Thiene G. Rhythm and
conduction disturbances in isolated, congenitally corrected transposition
of the great arteries. Am J Cardiol. 1986;58:314–318.
13. Sharland G, Tingay R, Jones A, Simpson J. Atrioventricular and ven-
triculoarterial discordance (congenitally corrected transposition of the
great arteries): echocardiographic features, associations, and outcome in
34 fetuses. Heart. 2005;91:1453–1458.
14. van Son JA, Danielson GK, Huhta JC, Warnes CA, Edwards WD, Schaff
HV, Puga FJ, Ilstrup DM. Late results of systemic atrioventricular valve
replacement in corrected transposition. J Thorac Cardiovasc Surg. 1995;
109:642–652.
15. Diller GP, Okonko D, Uebing A, Ho SY, Gatzoulis MA. Cardiac resyn-
chronization therapy for adult congenital heart disease patients with a
systemic right ventricle: analysis of feasibility and review of early expe-
rience. Europace. 2006;8:267–272.
16. DeFilippis AP, Law K, Curtin S, Eckman JR. Blood is thicker than water:
the management of hyperviscosity in adults with cyanotic heart disease.
Cardiol Rev. 2007;15:31–34.
17. Atik FA, Atik E, da Cunha CR, Caneo LF, Assad RS, Jatene MB, Riso A,
Barbero-Marcial M. Long-term results of correction of tetralogy of Fallot
in adulthood. Eur J Cardiothorac Surg. 2004;25:250–255.
18. Brown SC, Eyskens B, Mertens L, Dumoulin M, Gewillig M. Percuta-
neous treatment of stenosed major aortopulmonary collaterals with
balloon dilatation and stenting: what can be achieved? Heart. 1998;79:
24–28.
19. Bull K, Somerville J, Ty E, Spiegelhalter D. Presentation and attrition in
complex pulmonary atresia. J Am Coll Cardiol. 1995;25:491–499.
20. Rashkind WJ, Miller WW. Transposition of the great arteries: results of
palliation by balloon atrioseptostomy in thirty-one infants. Circulation.
1968;38:453–462.
21. Senning A. Surgical correction of transposition of the great arteries.
Surgery. 1959;45:966–980.
22. Mustard WT. Successful two-stage correction of transposition of the great
vessels. Surgery. 1964;55:469–472.
23. Jatene AD, Fontes VF, Paulista PP, de Souza LC, Neger F, Galantier M,
Souza JE. Successful anatomic correction of transposition of the great
vessels: a preliminary report. Arq Bras Cardiol. 1975;28:461–464.
24. Wilson NJ, Clarkson PM, Barratt-Boyes BG, Calder AL, Whitlock RM,
Easthope RN, Neutze JM. Long-term outcome after the Mustard repair
for simple transposition of the great arteries: 28-year follow-up. J Am Coll
Cardiol. 1998;32:758–765.
25. Roos-Hesselink JW, Meijboom FJ, Spitaels SE, van Domburg R, van
Rijen EH, Utens EM, McGhie J, Bos E, Bogers AJ, Simoons ML. Decline
in ventricular function and clinical condition after Mustard repair for
transposition of the great arteries (a prospective study of 22–29 years).
Eur Heart J. 2004;25:1264–1270.
26. Dos L, Teruel L, Ferreira IJ, Rodriguez-Larrea J, Miro L, Girona J, Albert
DC, Goncalves A, Murtra M, Casaldaliga J. Late outcome of Senning and
Mustard procedures for correction of transposition of the great arteries.
Heart. 2005;91:652–656.
27. Khairy P, Landzberg MJ, Lambert J, O’Donnell CP. Long-term outcomes
after the atrial switch for surgical correction of transposition: a meta-anal-
ysis comparing the Mustard and Senning procedures. Cardiol Young.
2004;14:284–292.
28. Gelatt M, Hamilton RM, McCrindle BW, Connelly M, Davis A, Harris L,
Gow RM, Williams WG, Trusler GA, Freedom RM. Arrhythmia and
mortality after the Mustard procedure: a 30-year single-center experience.
J Am Coll Cardiol. 1997;29:194–201.
29. Kammeraad JA, van Deurzen CH, Sreeram N, Bink-Boelkens MT,
Ottenkamp J, Helbing WA, Lam J, Sobotka-Plojhar MA, Daniels O,
Balaji S. Predictors of sudden cardiac death after Mustard or Senning
repair for transposition of the great arteries. J Am Coll Cardiol. 2004;44:
1095–1102.
30. Daehnert I, Hennig B, Wiener M, Rotzsch C. Interventions in leaks and
obstructions of the interatrial baffle late after Mustard and Senning
correction for transposition of the great arteries. Catheter Cardiovasc
Interv. 2005;66:400–407.
31. Ward CJ, Mullins CE, Nihill MR, Grifka RG, Vick GW III. Use of
intravascular stents in systemic venous and systemic venous baffle
obstructions: short-term follow-up results. Circulation. 1995;91:
2948–2954.
32. von Bernuth G. 25 Years after the first arterial switch procedure:
mid-term results. Thorac Cardiovasc Surg. 2000;48:228–232.
33. Raisky O, Bergoend E, Agnoletti G, Ou P, Bonnet D, Sidi D, Vouhe PR.
Late coronary artery lesions after neonatal arterial switch operation:
results of surgical coronary revascularization. Eur J Cardiothorac Surg.
2007;31:895–899.
34. Prifti E, Crucean A, Bonacchi M, Bernabei M, Murzi B, Luisi SV, Vanini
V. Early and long term outcome of the arterial switch operation for
transposition of the great arteries: predictors and functional evaluation.
Eur J Cardiothorac Surg. 2002;22:864–873.
35. McMahon CJ, Ravekes WJ, Smith EO, Denfield SW, Pignatelli RH,
Altman CA, Ayres NA. Risk factors for neo-aortic root enlargement and
aortic regurgitation following arterial switch operation. Pediatr Cardiol.
2004;25:329–335.
36. Williams WG, Quaegebeur JM, Kirklin JW, Blackstone EH. Outflow
obstruction after the arterial switch operation: a multiinstitutional study:
Congenital Heart Surgeons Society. J Thorac Cardiovasc Surg. 1997;
114:975–987.
37. Gandhi SK, Pigula FA, Siewers RD. Successful late reintervention after
the arterial switch procedure. Ann Thorac Surg. 2002;73:88–93.
38. Williams WG, Rubis L, Trulser GA, Mustard WT. Palliation of tricuspid
atresia: Potts-Smith, Glenn, and Blalock-Taussig shunts. Arch Surg.
1975;110:1383–1386.
39. Hirsch JC, Mosca RS, Bove EL. Complete repair of tetralogy of Fallot in
the neonate: results in the modern era. Ann Surg. 2000;232:508–514.
40. Murphy JG, Gersh BJ, Mair DD, Fuster V, McGoon MD, Ilstrup DM,
McGoon DC, Kirklin JW, Danielson GK. Long-term outcome in patients
undergoing surgical repair of tetralogy of Fallot. N Engl J Med. 1993;
329:593–599.
41. Waien SA, Liu PP, Ross BL, Williams WG, Webb GD, McLaughlin PR.
Serial follow-up of adults with repaired tetralogy of Fallot. J Am Coll
Cardiol. 1992;20:295–300.
42. Child JS. Echocardiographic evaluation of the adult with postoperative
congenital heart disease. In: Otto CM, ed. The Practice of Clinical
Echocardiography. 2nd ed. Philadelphia, Pa: Saunders; 2002:901–921.
43. Vogel M, Sponring J, Cullen S, Deanfield JE, Redington AN. Regional
wall motion and abnormalities of electrical depolarization and repolar-
ization in patients after surgical repair of tetralogy of Fallot. Circulation.
2001;103:1669–1673.
44. Ghai A, Silversides C, Harris L, Webb GD, Siu SC, Therrien J. Left
ventricular dysfunction is a risk factor for sudden cardiac death in adults
late after repair of tetralogy of Fallot. J Am Coll Cardiol. 2002;40:
1675–1680.
45. Geva T, Sandweiss BM, Gauvreau K, Lock JE, Powell AJ. Factors
associated with impaired clinical status in long-term survivors of
tetralogy of Fallot repair evaluated by magnetic resonance imaging. J Am
Coll Cardiol. 2004;43:1068–1074.
Sommer et al Complex Congenital Heart Disease 1349

11. 46. Gatzoulis MA, Balaji S, Webber SA, Siu SC, Hokanson JS, Poile C,
Rosenthal M, Nakazawa M, Moller JH, Gillette PC, Webb GD, Redington
AN. Risk factors for arrhythmia and sudden cardiac death late after repair
of tetralogy of Fallot: a multicentre study. Lancet. 2000;356:975–981.
47. Khairy P, Landzberg MJ, Gatzoulis MA, Lucron H, Lambert J, Marçon F,
Alexander ME, Walsh EP. Value of programmed ventricular stimulation
after tetralogy of Fallot repair: a multicenter study. Circulation. 2004;
109:1994–2000.
48. Gatzoulis K, Frogoudaki A, Brili S, Stefanadis C. Implantable defibril-
lators: from the adult cardiac to the grown up congenital heart disease
patient. Int J Cardiol. 2004;97(suppl 1):117–122.
49. Therrien J, Siu SC, McLaughlin PR, Liu PP, Williams WG, Webb GD.
Pulmonary valve replacement in adults late after repair of tetralogy of
Fallot: are we operating too late? J Am Coll Cardiol. 2000;36:1670–1675.
50. Geva T, Sandweiss BM, Gauvreau K, Lock JE, Powell AJ. Factors
associated with impaired clinical status in long-term survivors of
tetralogy of Fallot repair evaluated by magnetic resonance imaging. J Am
Coll Cardiol. 2004;43:1068–1074.
51. Fontan F, Baudet E. Surgical repair of tricuspid atresia. Thorax. 1971;
26:240–248.
52. Jacobs ML, Rychik J, Rome JJ, Apostolopoulou S, Pizarro C, Murphy JD,
Norwood WI Jr. Early reduction of the volume work of the single
ventricle: the hemi-Fontan operation. Ann Thorac Surg. 1996;62:
456–461.
53. Woods RK, Dyamenahalli U, Duncan BW, Rosenthal GL, Lupinetti FM.
Comparison of extracardiac Fontan techniques: pedicled pericardial
tunnel versus conduit reconstruction. J Thorac Cardiovasc Surg. 2003;
125:465–471.
54. Gentles TL, Gauvreau K, Mayer JE Jr, Fishberger SB, Burnett J, Colan
SD, Newburger JW, Wernovsky G. Functional outcome after the Fontan
operation: factors influencing late morbidity. J Thorac Cardiovasc Surg.
1997;114:392–403.
55. Fontan F, Kirklin JW, Fernandez G, Costa F, Naftel DC, Tritto F,
Blackstone EH. Outcome after a “perfect” Fontan operation. Circulation.
1990;81:1520–1536.
56. de Leval MR. The Fontan circulation: What have we learned? What to
expect? Pediatr Cardiol. 1998;19:316–320.
57. McMahon CJ, El Said HG, Vincent JA, Grifka RG, Nihill MR, Ing FF,
Fraley JK, Mullins CE. Refinements in the implantation of pulmonary
arterial stents: impact on morbidity and mortality of the procedure over
the last two decades. Cardiol Young. 2002;12:445–452.
58. Sugiyama H, Yoo SJ, Williams W, Benson LN. Characterization and
treatment of systemic venous to pulmonary venous collaterals seen after
the Fontan operation. Cardiol Young. 2003;13:424–430.
59. Kreutzer J, Lock JE, Jonas RA, Keane JF. Transcatheter fenestration
dilation and/or creation in postoperative Fontan patients. Am J Cardiol.
1997;79:228–232.
60. Rychik J, Rome JJ, Jacobs ML. Late surgical fenestration for compli-
cations after the Fontan operation. Circulation. 1997;96:33–36.
61. Kerstein D, Levy PS, Hsu DT, Hordof AJ, Gersony WM, Barst RJ. Blade
balloon atrial septostomy in patients with severe primary pulmonary
hypertension. Circulation. 1995;91:2028–2035.
KEY WORDS: transposition of great vessels Ⅲ tetralogy of Fallot Ⅲ Fontan
procedure Ⅲ heart defects, congenital
1350 Circulation March 11, 2008