ORIGINAL ARTICLE

Original Article

Transcatheter Closure of Secundum Atrial Septal

Defects with the Amplatzer Septal Occluder in
Adults and Children—Follow-up Closure Rates,
Degree of Mitral Regurgitation and Evolution
of Arrhythmias
Nigel J. Wilson, FRACP a,∗ , Joanna Smith, RN b , Bundit Prommete, MD c ,
Clare O’Donnell, FRACP a , Tom L. Gentles, FRACP a and Peter N. Ruygrok, FRACP d
a Green Lane Cardiovascular Service and Starship Children’s Hospital, Auckland, New Zealand
b Cardiac Investigation Unit, Starship Children’s Hospital, New Zealand
c Police Hospital, Bangkok, Thailand
d Green Lane Cardiovascular Service, Auckland City Hospital, New Zealand

Aim: To report the results of transcatheter atrial septal defect (ASD) closure with the Amplatzer septal occluder (ASO)
from the single centre providing interventions for congenital heart disease in New Zealand.
Methods: A single centre retrospective review of all patients 1997–2004 inclusive, undergoing planned transcatheter
ASD closure was undertaken. Implantation success, complications and latest patient follow-up are described.
Results: Percutaneous ASD closure was planned in 227 adults and children. (55% had additional medical co-
morbidities.) The mean ASD size was 22 ± 6 mm, and device size ranged from 5 to 40 mm. Closure was successful in
93%, unsuccessful in 5% and not attempted in 2%. There were two device embolisations, one immediately post-release
and one within 24 h. Minor procedural complications occurred in 5% of patients. There were no cases of suspected or
proven device erosion. Ninety-two percent had documented follow-up beyond 6 months. At latest follow-up the closure
rate was 98.5%. There was no significant change in the degree of mitral regurgitation. Sixteen of 24 with pre-closure
arrhythmias resolved while 10 remained in atrial fibrillation. Six patients developed new arrhythmias but these were
well controlled medically.
Conclusion: There is a high implantation and closure rate using transcatheter ASD closure with the Amplatzer septal
occluder in children and adults avoiding the need for cardiac surgery. Follow-up at one year shows there is no progression
of mitral regurgitation, and supraventricular arrhythmias have usually resolved.
(Heart, Lung and Circulation 2008;17:318–324)
© 2008 Published by Elsevier Inc on behalf of Australasian Society of Cardiac and Thoracic Surgeons and the Cardiac
Society of Australia and New Zealand.
Keywords. Amplatzer septal occluder; Transcatheter atrial septal defects

Introduction
A SD (atrial septal defect) closure is indicated for right
ventricular volume over-load with left to right flow
at any age beyond four years to prevent arrhythmias and
right heart failure in later decades.
Percutaneous transcatheter closure of ASDs was first
reported in 19741 but no devices were available until the
Received 13 February 2007; received in revised form 12 June
2007; accepted 30 October 2007; available online 14 April 2008
∗ The aim of the study was to report the results of
Corresponding author at: Starship Children’s Hospital, Private
Bag 92 024, Auckland, New Zealand. Tel.: +64 9 307 4949; fax: +64
9 631 0785.
E-mail address: nigelw@adhb.govt.nz (N.J. Wilson).
1990s. At Green Lane Hospital the Sideris device was ini-
tially used from 1995 to 1997 for transcatheter closure
but there was a high residual shunt rate and the device
was not suitable for any defect over 20 mm. From Decem-
ber 1997 we have used the Amplatzer septal occluder
(ASO), AGA Medical, Golden Valley, MN, USA. This is cur-
rently the most frequently used device for transcatheter
ASD closure worldwide with an estimate of 30,000 devices
implanted. It is capable of closing defects up to 40 mm and
is fully retrievable and redeployable, an advantage over
most other previous and current devices.
implantation and follow-up with emphasis on residual
shunt, evolution of arrhythmias and degree of mitral
regurgitation. All consecutive patients taken for cardiac

© 2008 Published by Elsevier Inc on behalf of Australasian Society of Cardiac and Thoracic 1443-9506/04/$30.00
Surgeons and the Cardiac Society of Australia and New Zealand. doi:10.1016/j.hlc.2007.10.013
Heart, Lung and Circulation Wilson et al. 319

ORIGINAL ARTICLE
2008;17:318–324 Transcatheter closure of secundum ASD

catheterisation from December 1997 to December 2004 Table 1. Comorbidities (n = 227)

were included with follow-up data until October 2005. Number of
Patients (%)
Methods and Patient Population Technique Total comorbidity 125 (55)

Patients were selected for cardiac catheterisation by Type of comorbiditya

echocardiography. Adult candidates had undergone Cardiac
transoesophageal echocardiography (TOE) and children Other cardiac anatomical defect 32 (14)
(e.g. mitral or pulmonary valve
transthoracic echocardiography (TTE) at their referral cen- abnormality)
tre. Those with partial anomalous pulmonary venous Documented tachyarrhythmia 23 (10)
drainage and/or deficient ASD rims by TOE or TTE did Acquired cardiac disease 14 (6)
not proceed to cardiac catheterisation. The technique of (hypertension, ischaemic heart
transcatheter ASD closure and the ASO device have been disease)
Pulmonary hypertension (mean 12 (5)
extensively described in previous reports.1–3 After general pulmonary artery pressure
anaesthesia, transoesophageal echocardiography (TOE) is >30 mm Hg)
performed to examine the site, size and suitability for Other medical comorbidity 46 (20)
device closure. The ASD should be <30 mm in adults Stroke or TIA 2 (1)
Asthma or other comorbidity 22 (10)
(smaller in children proportional to their weight) and there Chromosomal abnormality or 14 (6)
should be a distance >5 mm from the margins of the defect syndrome, e.g. down syndrome.
to the mitral valve, to the right upper pulmonary vein, a
Some patients had ≥1 comorbidity.
and to the inferior and superior vena-cavae. Use of the
Amplatzer septal occluder was approved by the multiple
regional ethics committees as structured in New Zealand Patient Population (n = 227)
in 1997. After an initial favourable experience and FDA Atrial septal defect closure was attempted in 164 (72%)
and CE approval of the ASO the procedure was considered females and 63 (28%) males (n = 227) of whom 78 (34%)
part of standard clinical practice and the hospital informed were aged between 0.9 and 15 years and 149 (66%) were
consent process only was used. Only femoral vein punc- adults aged 15–78 years. The mean weight at the time
ture is used for vascular access. Heparin and antibiotic of catheterisation was 53 ± 25.8 kg (range 7.4–110.7 kg).
prophylaxis were given routinely after initial haemody- Comorbidities were identified in 126 (55%) (Table 1.)
namic study to measure pulmonary artery pressure (PAP) In addition to documented tachyarrhythmia several
and pulmonary to systemic shunt (Qp/Qs) ratio. The patients had frequent ventricular or supraventricular
device size was selected after balloon stretched sizing ectopy prior to ASD closure which in general appeared
(BSS) of the ASD. We used a pull through technique to to resolve post-closure.
assess the BSS in our initial experience but changed to the
quicker static balloon technique when the Amplatzer siz-
ing balloon became available. Device placement was per-
Results
formed using fluoroscopic and TOE guidance. After device Patients underwent ASD closure if there was right ventric-
implantation all patients received aspirin at 3–5 mg per kg ular volume overload and/or a Qp/Qs ratio of ≥1.5:1. The
to a maximum of 150 mg for six months to prevent thrombi exception was one patient without a measurable L to R
formation while device endothelialisation occurs. Patients shunt, a 20-year-old man seeking a career as a navy diver.
were discharged the next day after TTE, chest radiography The mean balloon stretched size (BSS) was 22 ± 6.1 mm,
and ECG. After discharge patients were followed-up with range 8–36 mm. Mean ASO size deployed was 22 ± 6.4 mm,
clinical reviews, ECG and TTE at one month and one year range 5–40 mm. BSS was not always performed at the
post-implantation. The first 71 patients also had a three extreme ranges of sizes, in which case TOE measurements
month follow-up assessment which was subsequently dis- were used. Mean Qp/Qs ratio was 2.3 ± 1.2, range 1–9.3.
continued. Follow-up had been performed either at the Mean pulmonary artery pressure (PAP) was 18.7 ± 5.8 mm
tertiary centre or by cardiologists at the referring hospital. Hg, range 9–41 mm Hg.
Data were collected by chart review, with patients recalled Longer procedure times occurred in patients with pul-
if there was no follow-up greater than three months post- monary hypertension when haemodynamics were tested
ASD device implantation. We also recommend five year with vasodilators to assess suitability for ASD closure
follow-up for all patients, and follow-up to age 15 for chil-
dren. In light of recent reports of atrial wall erosion,4,5 we Table 2. Procedure and Fluoroscopy Times
recommend TTE 48 hours after discharge for patients with
a deficient aortic rim. All (n = 227) Time (min) Mean (S.D.)
Successful ASD closure was defined as no residual shunt Procedure time, all devices 71 (33), range (12–220)
or a trivial (<1 mm colour jet width) or small (colour jet Fluoroscopy time, all devices 17 (12), range (4–97)
width 1–2 mm) residual shunt. An unsuccessful proce- Procedure time, two devices 88 (36), range (66–130)
used (n = 4)
dure was defined by a moderate shunt (>2–4 mm colour jet Fluoroscopy time, two devices 26 (8), range (19–35)
width) or a large (>4 mm colour jet width) residual shunt at used (n = 4)
follow-up and persistence of right ventricular overload.6
 320 Wilson et al.
ORIGINAL ARTICLE
Transcatheter closure of secundum ASD
Figure 1. Mean (±S.D.) procedure time vs. year of procedure 1997–2004.
(Table 2). There was progressive reduction in procedure
time in the first five years of the program which then has
tended to plateau (Fig. 1).
ASO Implantation
Two hundred and twenty-seven patients were catheterised
for planned closure of their ASD and 211 (93%) had suc-
cessful device placement. In 16 patients device placement
was unsuccessful but of these, there was no attempt at
device implantation in 5 patients due to the ASD being
considered too small (n = 3) or too large (n = 2) after TOE
and catheter assessment. In 9 (4%) patients the device was
implanted but not released due to either insufficient rim
for placement with the device pulling through the ASD
(n = 5) or the device interfering with the mitral valve appa-
ratus (n = 4). One 32 mm device embolised immediately at
implantation requiring emergency surgical retrieval and
surgical closure of the ASD. The second device emboli-
sation was recognised by echocardiography the next day.
The 12 mm device had embolised to the aortic arch with-
out haemodynamic obstruction and was retrieved with a
snare catheter from the femoral artery without complica-
tion. The ASD was resized and an 18 mm device placed
accurately with no further complications or sequelae from
the embolised device. One 18 kg child underwent success-
Figure 2. Residual shunt size at latest follow-up (n = 202).
Heart, Lung and Circulation
2008;17:318–324
ful closure of his large ASD with a 28 mm Amplatzer but
on subsequent TTE it was judged the device was too large
for the atrium with the potential for mitral valve damage.
Two days later the device was electively removed surgically
and the ASD closed with a pericardial patch.
Pulmonary Hypertension
There were 12/211 (6%) patients with a mean PAP of
>30 mm Hg at the time of cardiac catheterisation. Ten
were female, the median age was 31 (range 5–70 years).
The mean PAP was 35 ± 6 mm Hg (range 30–49 mm Hg).
The mean Qp/Qs ratio was 2.1 ± 1.4, range 1–6.2. Pul-
monary vascular resistance ranged from 2.1 to 18 ␮m2 .
Comorbidities in these patients included four smokers,
two overweight, one trisomy 22 mosaic and one diabetic.
Procedural Complications
There were no deaths or cardiac perforations. There were
three serious complications including the two device
embolisations as described. One patient had sciatic nerve
palsy from a retroperitoneal haematoma with incomplete
recovery, a complication of femoral vein access, not related
to device placement per se.
Minor procedural complications occurred in 12 patients
(5%) and are listed in Table 3.
Heart, Lung and Circulation Wilson et al. 321

ORIGINAL ARTICLE
2008;17:318–324 Transcatheter closure of secundum ASD

Table 3. Procedure Complications During and Immediately

After Device Placement
All (n = 227)
Major 3 (1%)
Immediate surgical removal of device and 1
surgical closure
Device urgently retrieved next day via 1
catheter; new device placed
Sciatic nerve compression from 1
retroperitoneal haematoma
Minor 10 (5%)
Device surgically removed 2 days after 1
insertion, abutting MV
Initial device tore atrial septum, larger device 1
placed
Pleural and pericardial effusions, resolved 1
Cardiac arrhythmia requiring treatment 1
Temporary loss of voice, likely trauma from 1
endotracheal tube
Prolonged time for femoral vein haemostasis 2
Access site haematoma 1 Figure 3. Mitral regurgitation pre-ASD closure compared to mitral
AV fistula and haematoma 1 regurgitation at latest follow-up (n = 194).
Mild MR post-procedure, not present before 1
device placement
third month. This included an echocardiographic assess-
ment of shunt size and change in mitral regurgitation. This
three month follow-up is now omitted from the follow-up
Follow-up
protocol.
All 211 patients with an implanted device underwent TTE
the following day. Nine (4%) patients did not receive any
Arrhythmias
further follow-up; 5 of whom were overseas patients who
returned home and 4 were patients who did not attend out- Pre-existing arrhythmias were documented in 26 patients;
patient appointments or had moved with no contactable atrial fibrillation/atrial flutter (n = 15), SVT (2), atrial tachy-
forwarding address. Follow-up for the cohort ranged from cardia (2), long QT syndrome (1), frequent ectopy (3) and
1 day to 5.96 years, mean 1.8 years. Of the remaining 203 ablated arrhythmias (3 with AVNRT 2 and WPW 1). These
patients, 185 (91%) completed follow-up greater than 6 patients with arrhythmia were adults with a median age
months. 52 (range 22–79 years), except for a 5-year-old with long
QT syndrome.
There was no record of persistent arrhythmia follow-
Residual Shunt
ing ASD closure in 16/26 (62%) of these patients. In the
Of the 202 patients who had follow-up 183 (91%) had com- 10 patients with persistent arrhythmias, eight had atrial
plete closure. Two patients did not have an echo and in flutter/fibrillation and two had atrial tachycardias.
one patient the presence or absence of a residual shunt New arrhythmias occurred in six (3%) of patients and
was not stated. Six (3%) had a trivial shunt reported, 7 (4%) symptoms persisted beyond three months in four patients
had a small shunt reported on echo, 2 (1%) had a moderate (Table 4). Treatment was successful with anti-arrhythmic
shunt reported and one patient had a large shunt (5 mm)
reported on echo. Thus, by definition,6 197 of 200 (98.5%)
patients had successful ASD closure (Fig. 2). Table 4. New Symptoms and Events at Follow-up
All (n = 202)
Mitral Regurgitation Arrhythmias post-closure 16 (8%)
For the entire cohort the degree of MR pre-ASD closure New arrhythmias 6 (3%)
was compared to the degree of MR at latest follow-up, Present pre-and post-closure 10 (5%)
Palpitations 12 (6%)
at a mean 1.2 years later (range 1 day–6.0 years). There Chest pain 9 (4%)
were 202 patients who had follow-up but only 194 reported Epistaxis 4 (2%)
on MR grade. The degree of MR was unchanged in 160 Headaches and migraine 11* (6%)
patients (88%). In 2 (1%) patients MR increased by two Heart failure 1
grades from trivial to moderate post-closure (device was Nerve damage 2 (1%)
Pulmonary hypertension 2 (1%)
not touching the mitral valve). In 18 patients (9%) an Respiratory tract infections 4 (2%)
increase in MR of one grade was observed. Thirteen (7%) Non-specific shortness of breath 7 (3%)
had a decrease by 1 grade of MR post-device placement Transient ischaemic attack 1
including 1 patient who had severe MR which decreased ∗
One patient with pre-existing migraine had worsening symptoms
to trivial (Fig. 3). leading to a decision by the referring cardiac centre to have the device
A three month follow-up showed that no new echocar- surgically removed with surgical ASD closure. Migraines persisted
diographic events were occurring between the first and afterwards, although possibly at a lower frequency.
 322 Wilson et al.
ORIGINAL ARTICLE
Transcatheter closure of secundum ASD
medication alone (n = 1), radio frequency ablation (1), car-
dioversion (1) and one awaits treatment.
New symptoms and events at follow-up are summarised
in Table 4.
Pulmonary Hypertension
Two patients who had pulmonary hypertension at the
time of ASD closure subsequently died. A device was
implanted in each and pulmonary haemodynamics were
monitored for 24 hours prior to release of the device. One
28-year-old woman had a mean PAP pre-ASD closure of
41 mm Hg which had fallen to 36 mm Hg the following
day when the device was released. Symptoms initially
improved in the months following ASD closure but within
two years pulmonary hypertension persisted with mean
PAP at 40 mm Hg with severe right ventricular dysfunction.
LV function was also markedly impaired. Bosentan was
recommended to treat her pulmonary hypertension but
had not been approved by the funding authorities when
she was admitted with a respiratory infection. She became
rapidly hypoxaemic, hypotensive and despite inotropic
and ventilatory support, died of right-sided heart failure
two years post-ASD closure.
Another woman aged 67 died one day following coro-
nary artery bypass surgery, 1.5 years after ASD closure.
Her mean PA pressure was 32 mm Hg at the time of ASD
closure. Her right ventricle showed significant dysfunc-
tion weaning from cardiopulmonary bypass and she died
in low cardiac output. Pulmonary artery pressure had not
been remeasured prior to the cardiac surgery but it may
have contributed to the poor outcome.
Right Ventricle Size
Right ventricular size was recorded prior to closure in 180
patients by subjective TTE assessment. Three percent (6)
of the cohort had a normal RV size, 35% (63) had a mildly
dilated RV, 54% (98) had a moderately dilated RV and 7%
(13) had a severely dilated RV. At the last follow-up RV size
assessment was made in 197 patients. RV had returned
to normal in 75% (147), 19% (38) had persistently mildly
dilated RV, 5% (10) had moderately dilated RV and 1%
(2) had severely dilated RVs (Fig. 4). Both patients with
severely dilated RVs two years post-closure had signifi-
cant pulmonary hypertension, one of them as reported,
subsequently died from right-sided failure secondary to
pulmonary hypertension. All patients with persistently
dilated RV were adults, aged 26–78 years (median 46 years)
except for a 7-year-old who had undergone an outflow
tract repair for pulmonary valve atresia This patient had
pulmonary valve regurgitation.
There was no significant change in RV size beyond one
year follow-up. Most of the resolution of RV size occurred
within the first one to three months post-closure.
Discussion
Success of ASD Closure Programme
This study demonstrates a high chance of successful
closure of atrial septal defects using the ASO. Patients
Heart, Lung and Circulation
2008;17:318–324
Figure 4. RV dilatation—comparison of pre-closure RV size (left) and
post-closure RV size at latest follow-up (right).
experience minimal morbidity after a femoral venous
puncture and are discharged the day following the pro-
cedure. Cardiac surgery with cardiopulmonary bypass is
avoided.
Patients with tachyarrhythmia should be considered for
radiofrequency ablation (RFA) prior to transcatheter ASD
closure as a left atrial approach may be necessary for a suc-
cessful RFA. Atrial fibrillation/flutter resolved following
ASD closure in half of those with tachyarrhythmia sug-
gesting that atrial size may play a role in the mechanism
of the tachycardia. The incidence of new arrhythmias in
this and other series is small.7 A study on atrial arrhyth-
mias after surgical closure of atrial septal defects suggests
the risk of atrial flutter or atrial fibrillation in adults post-
surgical repair is related to age at the time of surgery and
the mean PAP. It remains to be seen whether transcatheter
ASD closure will be associated with late arrhythmia
or not.
Procedural Complications
Procedural complications were low with only two devices
embolising in this series spanning seven years. Experi-
ence in our unit and internationally has shown that after
embolisation the smaller devices are usually retrievable
with snares or other retrieval devices, permitting further
attempts at transcatheter closure. Larger devices are usu-
ally not retrievable with a catheter but require cardiac
surgery for retrieval with concomitant ASD closure.
Follow-up
Most children and many adults are asymptomatic when
their ASDs are closed, the indication for closure being
to prevent the complications of right heart failure and
arrhythmias in mid and late adulthood. Although the
study did not review late symptomatology, adult patients
with exertional dyspnoea usually report considerable
improvement in exercise tolerance.
Follow-up out to five years post-transcatheter ASD
closure has been characterised by an absence of new prob-
Heart, Lung and Circulation
2008;17:318–324
lems related to the device.8 The present study confirms
that the RV usually returns to normal size after ASD clo-
sure in children and most adults.9
Pulmonary Hypertension
Selection of patients with pulmonary hypertension for
ASD closure should be based on careful haemodynamic
assessment. Some have suggested that the level of pul-
monary hypertension for safe ASD closure can be higher
for transcatheter closure than traditional cardiac surgery.
Despite some early improvement in symptoms reported
by such patients, prospective studies are needed to deter-
mine the safe level of pulmonary hypertension and
pulmonary vascular resistance.
Transcatheter Versus Surgical Closure of ASD
There are now compelling reasons for offering tran-
scatheter closure as the procedure of choice over surgery
with cardiopulmonary bypass. RV function is better
preserved10 and there is less chance of minor or major
complications11 with transcatheter closure. Mortality is
close to zero for transcatheter ASD closure and although
rare with cardiac surgery in the current era, it can still occur
(Green Lane Hospital surgical data base). The cost of the
procedure in New Zealand is cheaper unless two devices
are used (catheter approximately $12K versus $20K for
surgery), hospital stay shorter and there are obvious cos-
metic advantages. Adults return to work and children
return to school earlier than with cardiac surgery. In chil-
dren there is evidence of better developmental outcome.12
However the anatomy and size of the ASD precludes tran-
scatheter closure in up to 40% of children and 20% of adults
in our experience.
One complication that did not occur in this series but
has been recently reported is a 0.1% incidence of erosion
of the atrial wall by the device with perforation, sometimes
with cardiac tamponade and even death.4,5 Analysis of the
28 cases reported worldwide has shown that this occurred
within three days in two thirds of cases. It has been associ-
ated with oversizing devices especially when the aortic rim
is deficient. Patients usually present with chest pain. The
presence of a pericardial effusion on echocardiography is
an indication for urgent referral to the cardiothoracic unit.
CT scanning will differentiate serous fluid from blood in
the pericardium.
Oversizing an ASD could occur by operator choice when
using a device larger than the measured balloon ASD size.
Our data showed we had not chosen larger devices with
the average balloon stretched size (BSS) 22 ± 6.1 mm and
the device size 22 ± 6.4 mm but an erosion would have
been unlikely in a series of 200 devices with erosions occur-
ring with an incidence of 1/1000.
AGA, the manufacturer of the device have now rec-
ommended use of the “stop-flow” rather than balloon
stretching of the atrial septum to size the device. It is
not possible to determine from the reports on erosions,
whether “oversizing” was as a result of operator choice of
device size or resulted from over stretching of the atrial
septum.4,5
Wilson et al. 323
ORIGINAL ARTICLE
Transcatheter closure of secundum ASD
We have modified our follow-up protocol since these
reports. All patients are warned to urgently contact their
local cardiology centre if they experience chest pain or
breathlessness and those with a deficient aortic rim have
another echocardiogram approximately 48 hours after
their discharge echocardiogram.
Study Limitations
Many follow-up echocardiograms were done outside the
referral centre and reports only were used. Some reports
described residual shunt size as tiny or small rather than
as a measurement in millimetre.
Conclusions
The results of transcatheter ASD closure with the ASO
in New Zealand shows a high implantation and closure
rate with few significant complications. Follow-up at one
year shows there is no progression of mitral regurgitation,
and supraventricular arrhythmias have usually resolved.
There are few new events after implantation, at one month,
one year or five years.
Acknowledgements
We thank the referring cardiologists and echocardio-
graphers who supplied clinical and echocardiographic
follow-up data. We are grateful to Barbara Semb for sec-
retarial assistance. Joanna Smith was funded by the Green
Lane Research and Education Fund in 2005.
References
1. Harper RW, Mottram P, McGaw DJ. Closure of secun-
dum atrial septal defects with the amplatzer septal occluder
device: techniques and problems. Catheter Cardiovasc Interv
2002;57:508–24.
2. Masura J, Gavora P, Formanek A, Hijazi ZM. Transcatheter clo-
sure of secundum atrial defects using the new self = centering
amplatzer septal occluder: initial human experience. Cathet
Cardiovasc Diagn 1997;42:388–93.
3. Thanopoulos BD, Laskari CV, Tsaousis GS, Zarayelyan A,
Vekiou A, Papadopoulos GS. Closure of atrial septal defects
with the amplatzer occlusion device: preliminary results. J Am
Coll Cardiol 1998;31:1110–6.
4. Divekar A, Gaamangwe T, Shaikh N, Raabe M, Ducas J.
Cardiac perforation after device closure of atrial septal
defects with the Amplatzer septal occluder. J Am Coll Cardiol
2005;45:1213–8.
5. Amin Z, Hijazi ZM, Bass JL, Cheatham JP, Hellenbrand
WE, Kleinman CS. Erosion of Amplatzer septal occluder
device after closure of secundum atrial septal defects:
review of registry of complications and recommendations
to minimize future risk. Catheter Cardiovasc Interv 2004;63:
496–502.
6. Du Z-D, Hijazi ZM, Kleinman CS, Silverman NH, Larntz
K, for the Amplatzer Investigators. Comparison between
transcatheter and surgical closure of secundum atrial sep-
tal defect in children and adults. J Am Coll Cardiol 2002;39:
1836–44.
7. Gatzoulis MA, Freeman MA, Siu SC, Webb GD, Harris L.
Atrial arrhythmia after surgical closure of atrial septal defects
in adults. N Engl J Med 1999;340:839–46.
 324 Wilson et al.
ORIGINAL ARTICLE
Transcatheter closure of secundum ASD
8. Yew G, Wilson NJ. Transcatheter atrial septal defect clo-
sure with the Amplatzer septal occluder: five-year follow-up.
Catheter Cardiovasc Interv 2005;64:193–6.
9. Veldtman GR, Razack V, Siu S, El Hajj H, Walker F, Webb
GD, Benson LN, McLaughlin PR. Right ventricular form and
function after percutaneous atrial septal defect device closure.
J Am Coll Cardiol 2001;37:2108–13.
10. Dhillon R, Josen M, Henein M, Redington A. Transcatheter
closure of atrial septal defect preserves right ventricular func-
tion. Heart 2002;87:461–5.
Heart, Lung and Circulation
2008;17:318–324
11. Berger F, Vogel M, Alexi-Meskishvili V, Lange PE. Compar-
ison of results and complications of surgical and Amplatzer
device closure of atrial septal defects. J Thorac Cardiovasc Surg
1999;118:674–8.
12. Visconti Kj, Bichell DP, Jonas RA, Newburger JW, Bellinger
DC. Developmental outcome after surgical versus interven-
tional closure of secundum atrial septal defect in children.
Circulation 1999;100(Suppl. II):II-145–50.