European Heart Journal (2011) 32, 553–560 CLINICAL RESEARCH

doi:10.1093/eurheartj/ehq352 Congenital heart disease

Benefit of atrial septal defect closure in adults:

impact of age
Michael Humenberger 1†, Raphael Rosenhek 1†, Harald Gabriel 1, Florian Rader 1,
Maria Heger 1, Ursula Klaar 1, Thomas Binder 1, Peter Probst 1, Georg Heinze 2,

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

Gerald Maurer 1, and Helmut Baumgartner 1,3*
1
Department of Cardiology, Medical University of Vienna, Vienna, Austria; 2Center for Medical Statistics, Informatics and Intelligent Systems, Medical University of Vienna, Vienna,
Austria; and 3Adult Congenital and Valvular Heart Disease Center, Department of Cardiology and Angiology, University of Muenster, Albert Schweitzer Str. 33, 48149 Muenster,
Germany

Received 14 February 2010; revised 12 July 2010; accepted 4 August 2010; online publish-ahead-of-print 12 October 2010

See page 531 for the editorial comment on this article (doi:10.1093/eurheartj/ehq377)

Aims To evaluate the effect of age on the clinical benefit of atrial septal defect (ASD) closure in adults.
.....................................................................................................................................................................................
Methods Functional status, the presence of arrhythmias, right ventricular (RV) remodelling, and pulmonary artery pressure
and results (PAP) were studied in 236 consecutive patients undergoing transcatheter ASD closure [164 females, mean age of
49 + 18 years, 78 younger than 40 years (Group A), 84 between 40 and 60 years (Group B) and 74 older than
60 years (Group C)]. Defect size [median 22 mm (inter-quartile range, 19, 26 mm)] and shunt ratio [Qp:Qs 2.2
(1.7, 2.9)] did not differ among age groups. Older patients had, however, more advanced symptoms and both,
PAP (r ¼ 0.65, P , 0.0001) and RV size (r ¼ 0.28, P , 0.0001), were significantly related to age. Post-interventionally,
RV size decreased from 41 + 7, 43 + 7, and 45 + 6 mm to 32 + 5, 34 + 5, and 37 + 5 mm for Groups A, B, and C,
respectively (P , 0.0001), and PAP decreased from 31 + 7, 37 + 10, and 53 + 17 mmHg to 26 + 5, 30 + 6, and
43 + 14 mmHg (P , 0.0001), respectively. Absolute changes in RV size (P ¼ 0.80) and PAP (P ¼ 0.24) did not signifi-
cantly differ among groups. Symptoms were present in 13, 49, and 83% of the patients before and in 3, 11, and 34%
after intervention in Groups A, B, and C. Functional status was related to PAP.
.....................................................................................................................................................................................
Conclusions At any age, ASD closure is followed by symptomatic improvement and regression of PAP and RV size. However, the
best outcome is achieved in patients with less functional impairment and less elevated PAP. Considering the continu-
ous increase in symptoms, RV remodelling, and PAP with age, ASD closure must be recommended irrespective of
symptoms early after diagnosis even in adults of advanced age.
-----------------------------------------------------------------------------------------------------------------------------------------------------------
Keywords Transcatheter closure † Age at intervention

years.4,5 In small series, surgical closure was associated with

Introduction symptom and possibly survival improvement even in patients
Atrial septal defect (ASD) not uncommonly remains undetected older than 60 years.6,7 The sole randomized trial performed in
until adulthood accounting for 25 –30% of newly diagnosed conge- patients older than 40 years, however, found reduced morbidity
nital heart defects.1 Atrial septal defect closure has become an but not mortality after surgical ASD closure.8 Thus, the benefits
established therapy that is performed in increasing numbers of of ASD closure in adults, particularly those of advanced age,
adult patients.2 Early surgical repair results in excellent long-term remained uncertain.9 – 11 More recently, transcatheter ASD
outcome, whereas results appear less favourable when interven- closure has been shown to be feasible and safe in children and
tion is delayed until adulthood.3 In particular, controversial findings adults.12 – 15 Being significantly less invasive and associated with
have been reported for surgery performed after the age of 40 fewer complications even in older adults,16,17 it became an
†
The first two authors contributed equally to this work and are listed in alphabetic order.
* Corresponding author. Tel: +49 251 8346110, Fax: +49 251 8346109, Email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2010. For permissions please email: [email protected]
554 M. Humenberger et al.

attractive therapy for the elderly. Patients with pulmonary hyper- Wood units either at baseline or after vasoreactivity testing with nitric
tension may also benefit.18,19 Furthermore, an improved exercise oxide were considered for ASD closure.
capacity was reported even for asymptomatic and mildly sympto- In patients who were found to have a left atrial pressure of
matic adult patients.20,21 Despite being a widely performed and .15 mmHg during intraprocedural invasive evaluation, balloon occlu-
sion of the ASD was performed and pressure measurement was
accepted treatment modality, data on age-dependent benefits of
repeated. In patients with a left atrial pressure increase of
this procedure are, however, limited, particularly in the elderly.
.10 mmHg, further heart failure treatment was requested before con-
The aim of the present study was, therefore, to evaluate the
sideration for defect closure to reduce the risk of left heart failure after
effects of transcatheter ASD closure on functional status, arrhyth- intervention.
mias, right ventricular (RV) size, and pulmonary arterial pressure
according to the age at intervention in a large unselected cohort Catheter intervention
of adults.
All procedures were carried out under general anaesthesia with endo-
tracheal intubation and guided by fluoroscopy and TEE. After haemo-

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

dynamic assessment, all patients underwent balloon sizing of the
Methods defect. The ASO was chosen 2 – 4 mm larger than the stretched
diameter.
Patient population Aspirin therapy (100 mg/day) was initiated at least 2 days prior to
The study population consists of 236 consecutive adults (mean age and maintained for at least 6 months after the intervention. Intrave-
49 + 18 years, 164 females) who underwent transcatheter ASD nous heparin was administered intraprocedurally.
closure with the Amplatzer septal occluder (ASO; AGA Medical Cor-
poration, Golden Valley, MN, USA). Of these, 78 were younger than Follow-up
40 years (Group A), 84 were between 40 and 60 years (Group B),
The patients underwent serial follow-up examinations 1 day, 1 week,
and 74 were older than 60 years (up to 82 years; Group C) at the
3 – 6 months, 12 months, and then yearly after the intervention includ-
time of intervention. Indication for closure was a significant left-to-right
ing clinical examination, TTE, and electrocardiography. Particular care
shunt (signs of RV volume overload), irrespective of the presence of
was taken to determine the functional status and to obtain information
symptoms. Patients with severe pulmonary vascular disease (.5
regarding symptom development or any complications. Transoesopha-
Wood units) even after vasoreactivity testing or after targeted treat-
geal echocardiography was only performed on indication (suspected
ment were not considered for ASD closure.
residual shunt 6 months post-interventionally, suspicion of embolism).
During the study period, transcatheter ASD closure was attempted
in 237 patients, but one patient required surgery because of significant
residual shunt (success rate 99.6%). Transcatheter ASD closure was Statistical analysis
not attempted in patients who had a balloon-stretched defect diameter The distribution of continuous variables within age groups was
of .36 mm or a native diameter of .25 mm (80% of the patients assessed by the Shapiro– Wilk tests for normal distribution. These
found to be ineligible for transcatheter closure) and in those with tests revealed significant deviations from the normal distribution in
inadequate defect morphology (insufficient rim, multiple defects, nearly all variables and subgroups. Thus, continuous variables were
complex aneurysm, and significant additional lesions). For these described by medians and inter-quartile ranges and compared
reasons, 16% of the patients referred to our institution had surgical between the three age groups by the Kruskal– Wallis tests. In the
ASD closure, whereas 84% had transcatheter closure during the case of a significant three-group comparison, pairwise Wilcoxon’s
study period. rank sum tests were performed. For dichotomous variables, these
tests were replaced by x2 tests. Changes in RV size and PAP from base-
Echocardiography line to subsequent follow-up assessments were evaluated by the analy-
sis of variance for repeated measurements (RM-ANOVA), assuming a
Transoesophageal echocardiography (TEE) was routinely performed in
covariance of ‘unstructured’ type between subsequent measurements
all patients screened for transcatheter closure to assess ASD mor-
on the same patients. For comparison of age groups, we used the
phology and exclude additional lesions such as an anomalous pulmon-
analysis of covariance (ANCOVA), using the change from baseline to
ary venous connection.
6 months follow-up value as a dependent variable, age group as a
A comprehensive transthoracic echocardiogram (TTE), including
nominal fixed factor, and the baseline value as a covariate. The
M-mode, two-dimensional, continuous-wave, pulsed-wave, and
baseline-adjusted group means were computed by the LSMEANS
colour Doppler echocardiography, was performed before intervention
method of SAS/PROC GLM; these baseline-adjusted group means
and at each follow-up visit.
refer to the expected values of the dependent variable in each
Right ventricular size was measured by taking the transverse diam-
group given an average baseline value. Comparisons between the
eter in the apical four-chamber view, and pulmonary artery pressure
three age groups were adjusted by the Tukey– Kramér method.
(PAP) was estimated from the tricuspid regurgitant velocity.22 The
Inspecting the distribution of residuals from RM-ANOVA and
shunt ratio (Qp:Qs) was obtained by the measurement of the velocity
ANCOVA with original and log-transformed RV size and PAP revealed
time integrals as well as the cross-sectional areas at the corresponding
a slightly closer agreement with the normal distribution after log-
sites in the pulmonary artery and the left ventricular outflow tract.23
transformation. However, there was no difference in the observed
pattern of significances between both types of analysis. Thus, for
Invasive evaluation better interpretability, only results on the original values of RV size
Invasive evaluation was performed prior to intervention when patients and PAP are reported here. We also supply means and standard devi-
presented with a non-invasively estimated systolic PAP of .50% of ations for these parameters, despite slight deviation from the normal
systemic pressure or .60 mmHg. In these patients, pulmonary vascu- distribution, as these are compatible to the results of RM-ANOVA
lar resistance (PVR) was carefully assessed. Only patients with PVR ≤5 and ANCOVA. A Spearman’s correlation was used to determine the
Atrial septal defect closure in elderly 555

relationship between age, RV size, and PAP. The SAS System 9.2 (SAS managed conservatively. Spontaneously resolving supraventricular
Institute, Inc., Cary, NC, USA, 2008) was used for statistical analysis. All arrhythmias were common during the procedure. At the end of
tests were two-sided, and P-values ,0.05 were considered as indicat- the procedure, four patients had new AFib or atrial flutter. Of
ing statistical significance. these, three converted spontaneously and one required medical
conversion. One patient developed transient complete atrioventri-
cular block not requiring permanent pacemaker implantation. One
Results patient with a history of recurrent transient ischaemic attacks had a
transient worsening of his neurological deficit without neuroradio-
Baseline characteristics
logical signs of new ischaemia. Intraprocedural pressure measure-
Baseline patient characteristics are shown in Table 1. Defect size,
ment revealed 15 patients with a left atrial mean pressure of
device size, and shunt ratio did not significantly differ among age
.15 mmHg and 3 patients .20 mmHg before ASD closure.
groups.
None of them increased by .10 mmHg during balloon occlusion,
Limited exercise capacity and shortness of breath were the most

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

and all of them underwent successful closure without signs of left
frequently reported symptoms. Prior to intervention, 121 patients
heart failure.
were symptomatic. The presence of symptoms markedly increased
with age (Table 1). Although 75.3% of Group A patients were Early complications and residual shunts
asymptomatic, 51.9% of those in Group B and 87.1% in Group C
presented with symptoms [45.2% New York Heart Association
(first day to 3 months)
(NYHA) III]. Overall, there was a significant increase in PAP with For arrhythmias see below. Six patients developed a small pericar-
age, although even some elderly patients presented with still dial effusion, which disappeared spontaneously within 3 months.
normal PAP (Table 1 and Figure 1; r ¼ 0.65, P , 0.0001). Pulmonary One patient developed gastrointestinal bleeding on oral anticoagu-
artery pressure was clearly related to functional class with PAP lation given for AFib. No moderate or severe residual shunt was
33 + 9, 42 + 12, and 60 + 20 mmHg for patients in NYHA observed. Although a mild residual shunt was more common on
classes I, II, and III, respectively (P , 0.0001). There was a statisti- day 1, only seven patients (3%) had a definite and three (1.3%) a
cally significant relation but only weak correlation between age and questionable mild residual shunt 3 months post-interventionally.
RV size (r ¼ 0.28, P , 0.0001). Defect size correlated weakly with
RV size (r ¼ 0.41, P , 0.0001) and poorly with PAP; nevertheless,
Follow-up
the relation was statistically significant (r ¼ 0.21, P ¼ 0.007). Early follow-up data (3– 6 months) were available for all patients.
Moderate tricuspid regurgitation (TR) was present in 23 patients Three patients were lost to follow-up (2.3 + 1.6 years). None of
(9.7%), of whom 18 were older than 60 years. Only seven patients five observed deaths were related to ASD.
(3%) had severe TR. All of them were older than 60 years, had elev- The following late complications (after 3 months) were
ated systolic PAP (60.6 + 13.4 mmHg), and were mostly sympto- observed: 5 years after implantation, one patient developed a
matic requiring diuretic therapy (three patients in NYHA class III). large thrombus on the left atrial occluder disc, which embolized
The invasively measured systolic, diastolic, and mean PAP as well into all four extremities and the spleen requiring surgical embolect-
as transpulmonary gradient significantly increased with age omy. The patient underwent chemotherapy for haematological
(P,0.0001). Systolic PAP obtained by echocardiography in the disease before the event and eventually fully recovered.
awake patient was on average 10 mmHg higher than the invasively Three cerebral events were observed. Two patients with an
measured pressure during general anaesthesia. Right and left atrial ischaemic event had no residual shunt or thrombus on the occlu-
mean pressures were significantly higher in the oldest patient der. One patient receiving oral anticoagulation for AFib had minor
group. We found no significant difference in RV end-diastolic cerebellar bleeding.
pressures (P ¼ 0.21).
None of the patients in Group A presented with atrial fibrilla- Early and late arrhythmias
tion (AFib). Paroxysmal AFib was present in 9.5 and 18.9% of Atrial fibrillation was recorded in 11 patients (Group B, 7; Group
the patients in Groups B and C, and persistent AFib was present C, 4) within the first week. Four of these spontaneously converted
in 2.4 and 32.4%, respectively. to sinus rhythm, whereas four underwent successful electrical and
Co-morbidities such as arterial hypertension and vascular disease three medical cardioversion. Additionally, four patients developed
were mainly present in patients older than 40 years (Table 1). new AFib after 3– 6 months—two of these were medically and
two electrically cardioverted.
All 26 patients with pre-existing persistent AFib remained in
Catheter intervention and procedural
AFib. In addition, one patient who had paroxysmal AFib at entry
complications developed persistent AFib at follow-up. However, 10 of the 22
The median device size was 24 mm (22, 28 mm), median procedure patients (5 of 8 in Group B and 5 of 14 in Group C) with parox-
time 40 min (30, 55), and median fluoroscopy time 7.3 min (5.6, ysmal AFib remained in sinus rhythm during follow-up.
10.8) with no significant differences among age groups (Table 1).
No major procedural complications occurred. Transient Regression of right ventricular size and pulmonary artery
ST-elevation with complete resolution, probably due to air embo- pressure
lization, was observed in two patients, uncomplicated groin haema- Right ventricular diameter decreased from 43 + 7 mm at baseline
toma in five, and one developed a femoral artery pseudoaneurysm to 38 + 6 mm (P , 0.0001) on the first post-interventional day,
556 M. Humenberger et al.

Table 1 Patient characteristics at study entry

All patients Group A, Group B, Group C, P-value

(n 5 236) age <40 40– 60 years age >60 ...........................................................
A/B/C A/B A/C B/C
years (n 5 84) years
(n 5 78) (n 5 74)
...............................................................................................................................................................................
Age (years) 49 + 17.4 29 + 6.7 50 + 5.6 71 + 6.1 ,0.0001* ,0.0001* ,0.0001* ,0.0001*
Gender (female), n (%) 164 (69.5) 48 (61.5) 55 (65.5) 61 (82.4) 0.0160* 0.6468 0.0058* 0.0183*
...............................................................................................................................................................................
Echocardiographic data
RV size (mm) 43 (39, 48) 41 (37, 46) 43 (38, 41) 45 (42, 50) 0.0007* 0.2006 ,0.0001* 0.0141*
PA systolic pressure 37 (30, 48) 30 (26, 35) 35 (31, 41) 51 (42, 62) ,0.0001* 0.0005* ,0.0001* ,0.0001*
(mmHg)

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

Qp:Qs 2.21 (1.7, 2.9) 2.11 (1.7, 2.6) 2.15 (1.6, 2.6) 2.45 (2.75, 3.1) 0.0893 0.9452 0.1116 0.0263*
Defect size (mm) 22 (19, 26) 23 (19, 26) 22 (17.5, 26.5) 22.5 (20, 28) 0.2397 0.2736 0.5995 0.0954
Moderate TR, n (%) 23 (9.7) 1 (1.3) 4 (4.8) 18 (24.7) ,0.0001* 0.2007 ,0.0001* 0.003
Severe TR, n (%) 7 (3) 0 0 7 (9.6) 0.0003* NA 0.0054 0.0039
...............................................................................................................................................................................
Invasive data (mmHg)
RA mean pressure 7 (5, 9) 7 (5, 8.5) 7 (5, 9) 8 (6, 11) 0.0194* 0.6346 0.0111* 0.0236*
RV end-diastolic 9 (7, 11) 9 (7, 10) 9 (7, 11) 10 (8, 12) 0.2162 0.6998 0.0937 0.1911
pressure
PA systolic pressure 28 (23, 36) 24 (20, 27) 29 (23, 33) 36 (30, 50) ,0.0001* 0.0002* ,0.0001* ,0.0001*
PA diastolic pressure 11 (8, 14) 9 (8, 12) 10 (8, 14) 14 (9, 17) 0.0001* 0.0409* ,0.0001* 0.0048*
PA mean pressure 18 (15, 23) 15 (14, 18) 18 (16, 22) 23 (19, 30) ,0.0001* 0.0020* ,0.0001* 0.0001*
LA mean pressure 9 (7, 11) 9 (7, 10) 9 (7, 11) 10 (8, 14) 0.0259* 0.4481 0.0121* 0.0438*
Transpulmonary gradient 10 (7, 13) 7 (6, 9) 10 (7, 12) 13 (9, 17) ,0.0001* 0.0005* ,0.0001* 0.0042*
...............................................................................................................................................................................
Clinical data
NYHA I, n (%) 105 (46.5) 58 (75.3) 38 (48.1) 9 (12.9) ,0.0001* ,0.0001* 0.2203 ,0.0001*
NYHA II, n (%) 88 (38.9) 18 (23.4) 38 (48.1) 32 (45.7) 0.0063 0.003 0.0081 0.8011
NYHA III, n (%) 33(14.6) 1(1.3) 3(3.8) 29(41.4) ,0.0001* 0.3481 ,0.0001* ,0.0001*
Persistent atrial 26 (11) 0 2 (2.4) 24 (32.4) ,0.0001* 0.1703 ,0.0001* ,0.0001*
fibrillation, n (%)
Paroxysmal atrial 22 (9.3) 0 8 (9.5) 14 (18.9) 0.0003* 0.0051* ,0.0001* 0.0887*
fibrillation, n (%)
Arterial Hypertension, 39 (16.5) 0 19 (22.6) 20 (27) ,0.0001* ,0.0001* ,0.0001* 0.5214
n (%)
Diabetes mellitus, n (%) 9 (3.8) 0 4 (4.8) 5 (6.8) 0.08 0.051 0.0196* 0.5893
Hypercholesterolaemia, 18 (7.6) 1 (1.3) 3 (3.6) 29 (39.2) ,0.0001* 0.3481 ,0.0001* ,0.0001*
n (%)
Coronary artery disease, 11 (4.7) 0 4 (4.8) 7 (9.5) 0.083 0.051 0.00542 0.247
n (%)
Peripheral arterial 3 (1.3) 0 2 (2.4) 1 (1.4) 0.40 0.17 0.303 0.6361
disease, n (%)
Carotid artery disease, 4 (1.7) 0 1 (1.2) 3 (4.1) 0.1391 0.3337 0.0725 0.2529
n (%)
Chronic obstructive 8 (3.4) 0 2 (2.4) 6 (8.1) 0.018 0.1703 0.0101 0.1013
pulmonary disease,
n (%)
...............................................................................................................................................................................
Device size (mm) 24 (22, 28) 24 (22, 28) 24 (20, 29.5) 26 (22, 30) 0.3213 0.7207 0.1449 0.2679
Fluoroscopy time (min) 7.3 (5.6, 10.8) 7.15 (5.6, 10.4) 6.7 (5.5, 10.4) 8.55 (6.5, 12) 0.1471 0.8479 0.1090 0.0714
Procedure time (min) 40 (30, 55) 40 (30, 59) 40 (30, 56) 37 (30, 48.5) 0.8866 0.7233 0.6544 0.8456

Data are presented as mean + standard deviation or median (inter-quartile range) where appropriate.
LA, left atrial; NYHA, New York Heart Association functional class; Qp:Qs, pulmonary to systemic flow ratio; PA, pulmonary artery; RA, right atrial; RV, right ventricular; TR,
tricuspid regurgitation.
*Statistically significant difference.
Atrial septal defect closure in elderly 557

36 + 6 mm after 1 week, and 34 + 6 mm after 3–6 months. At comparison of Groups B and C, and P ¼ 0.0903 for comparison
last follow-up, RV size remained stable with 34 + 6 mm (Figure 2). of Groups A and B). The baseline-adjusted mean decreases in
Pulmonary artery pressure decreased from 41 + 16 to 35 + PAP in Groups A, B, and C were 11.6, 10.4, and 2.3 mmHg,
13 mmHg on day 1 and 34 + 12 mmHg 3 months after ASD respectively, and significantly differed between Groups A and C
closure (P , 0.0001) remaining stable thereafter (33 + 14 mmHg (P , 0.0001) and Groups B and C (P ¼ 0.0001) but not between
at last follow-up; Figure 3). Groups A and B (P ¼ 0.8210). A moderate correlation between
A decrease in RV size and PAP was observed in all age groups. age and PAP persisted post-interventionally (r ¼ 0.63, P ,
The absolute changes did not significantly differ among groups 0.0001) and patients older than 60 years were most likely to be
(decreases in RV size 9 + 7, 8 + 7, and 8 + 6 mm for Groups A, left with persistently elevated PAP. Although no patient in Group
B, and C, P ¼ 0.80; decreases in PAP 5 + 8, 8 + 9, and 9 + A and only five patients in Group B (6%) had a systolic PAP
14 mmHg, respectively, P ¼ 0.24). Consequently, older patients ≥40 mmHg, this was the case in 38 patients (51%) of Group C.
who had significantly larger RVs and higher PAPs before ASD

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

closure ended up with larger ventricles and higher PAPs after inter-
Tricuspid regurgitation
vention. The baseline-adjusted mean decreases in RV size for
After ASD closure, the degree of TR decreased. Only 2 of orig-
Groups A, B, and C were 10.1, 8.4, and 6.1 mm, respectively
inally 7 patients still had severe TR and 17 (originally 23) had mod-
(P , 0.0001 for comparison of Groups A and C, P ¼ 0.0163 for
erate TR. They also improved with regard to functional status. Of
the seven patients with originally severe TR, three were asympto-
matic and four in NYHA class II after closure.

Functional status
Symptomatic improvement was observed across all age groups
(Figure 4). After 3 –6 months, all but two Group A patients who
remained in NYHA class II were asymptomatic. In Group B, nine
patients remained in NYHA class II, whereas 89% were asympto-
matic. In Group C, however, 22 patients remained in NYHA
class II and 3 in NYHA class III. Two of them suffered from
marked persistent pulmonary hypertension, and one had advanced
obstructive pulmonary disease. Nevertheless, even Group C
patients improved markedly with 69% being asymptomatic post-
interventionally when compared with 16% before.
Functional status was closely related to PAP. At 3 –6 months,
patients in NYHA classes I, II, and III had a systolic PAP of 31 +
9, 47 + 15, and 67 + 3 mmHg, respectively. A PAP above
Figure 1 Correlation between systolic pulmonary artery 40 mmHg was present in 2 of 3 patients in NYHA class III, in 18
pressure and age (r ¼ 0.65, P , 0.0001).

Figure 2 Right ventricular (RV) size before, 1 day, 1 week, and Figure 3 Systolic pulmonary artery pressure (sPAP) before, 1
3 months after atrial septal defect closure for patients younger day, 1 week, and 3 months after atrial septal defect closure for
than 40 years (green line), patients aged 40 – 60 years (orange patients younger than 40 years (green line), patients aged 40 –
line), and patients older than 60 years (red line). 60 years (orange line), and patients older than 60 years (red line).
558 M. Humenberger et al.

analysis of the impact of age at intervention on the long-term

effect of ASD closure on functional status, arrhythmias, and
other adverse events as well as RV size and PAP.
The study confirms the feasibility and safety of transcatheter
ASD closure in all age groups. Although one occurrence of late
thrombus formation on the occluder with embolization in a
patient with haematological disease may raise concern, this
remained the only major complication in the long-term follow-up.
Krumsdorf et al.31 published a study with over 1000 patients, which
focused on thrombus formation on different devices and found
that in all occluders, with exception of the Amplatzer occluder
(no incidence), there was a low incidence of thrombosis.

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

Nevertheless, embolic events were extremely rare.
Figure 4 Symptomatic status before (pre) and after (post)
atrial septal defect closure for patients younger than 40 years, Arrhythmias
patients aged 40 – 60 years, and patients older than 60 years. The incidence of AFib—one of the major causes of morbidity in
patients with ASD—is closely related to age.32 After surgery,
AFib occurs more frequently in patients older than 40 years at
of 30 in NYHA class II, and in only 23 of 203 asymptomatic the time of intervention,32 and the rate of new-onset arrhythmias
patients. was reportedly similar to medically treated patients.4,8 The differ-
However, 180 of 193 patients (93%) with a PAP ,40 mmHg ent timing of arrhythmia onset, however, suggests different under-
were asymptomatic, whereas this was only the case for 23 of 43 lying mechanisms, scars playing an important role after surgery.
patients (54%) with PAP ≥40 mmHg. Both acute and long-term effects on the interatrial septum and
At late follow-up, a slight worsening of the symptomatic status atria may be different after transcatheter closure. New-onset
was observed, again being more likely in older patients. AFib early after transcatheter ASD closure has previously been
reported.33 In our series, 6% of the patients developed AFib
within 3 months after the intervention. Sinus rhythm was restored
Discussion in all of them emphasizing its transient nature potentially caused by
The benefit of ASD closure in adults, particularly those of early mechanical irritation of the septum. Of note, AFib has been
advanced age, remains a matter of debate.9 – 11 Transcatheter reported to occur both after percutaneous ASD and patent
ASD closure has the advantage of being significantly less invasive foramen ovale closure.34 Silversides et al.35 reported that the like-
and associated with shorter hospital stay than surgery.24,25 In cur- lihood of remaining free of arrhythmia after transcatheter ASD
rently published series including the present one, no closure was greatest in patients younger than 40 years without a
procedure-related death was observed.15 – 17 In contrast, surgical history of arrhythmia. A positive effect of transcatheter ASD
ASD closure has been found to be associated with significant mor- closure on the long-term incidence of arrhythmia was described.36
tality at advanced age.3,6,26,27 Although primarily low-risk patients In agreement with our observations, the incidence of arrhythmias
were referred to surgery in the past, data of larger unselected after closure was reduced in patients (particularly, the younger
series are now available including the elderly and patients with sig- ones) with paroxysmal AFib but not in those with persistent
nificant co-morbidities. Although some previous series documen- AFib.37 Thus, transcatheter ASD closure may trigger early transient
ted the feasibility and safety of the procedure in older adults, arrhythmias, persistent AFib is unlikely to be affected, but paroxys-
data on the actual benefit remained so far insufficient and inconclu- mal arrhythmias may improve although that likelihood decreases
sive: in the most recent and largest series including 144 patients with age.34
older than 60 years, Majunke et al.15 only report on the efficacy
and safety of the procedure but no details about clinical benefits Functional status
and changes in RV and PAP. Swan et al.28 retrospectively analysed Controversial data were reported regarding the effect of late sur-
the results of ASD closure in 185 patients, 50 older than 60 years. gical ASD closure on the functional status.4,5 The present study
They reported that RV size and PAP decreased in the older group demonstrates how the incidence and intensity of symptoms
but provide only short-term results, not including clinical outcome increase with age. Symptomatic improvement was observed
and effects of age. Similarly, Elshershari et al.29 only reported short- across all age groups but was most impressive in older patients.
term results for a group of 41 patients older than 60 years, allow- Nevertheless, a significant portion of older patients remained
ing no conclusions with regard to age effects or factors associated symptomatic, and some patients deteriorated again during late
with insufficient improvement. Yalonetsky and Lorber30 studied follow-up.
only small groups of patients aged 40 –60 years and above 60
years (23 patients in each group), reporting a greater reduction Pulmonary hypertension
in PAP in the older group again without detailing the outcome. The prevalence of pulmonary hypertension in patients with ASD
The present study is to our best knowledge the first report of a has been reported to be between 6 and 27%.18,38 – 41 Two previous
large cohort of consecutive adult patients with a comprehensive studies report successful transcatheter ASD closure in patients
Atrial septal defect closure in elderly 559

with pulmonary hypertension. Although PAP remained elevated in Study limitations

a significant portion, some decrease associated with symptomatic The major limitation of the present study is its non-randomized
improvement was reported.18,19 However, these previous studies nature and the lack of a control group, making it impossible to
did not allow conclusions on age-related effects, and it remains study effects of ASD closure on survival. However, it can no
unclear how many of these patients also had a severely increased longer be justified to perform a randomized trial on ASD
vascular resistance. closure even in the elderly. Exercise testing was not routinely per-
In the present study, patients with severe pulmonary vascular formed since it is poorly standardized in the elderly. However,
disease were excluded. Nevertheless, the study clearly demon- objective measures such as RV size and PAP were obtained. Func-
strates that PAP increases continuously with age and that a signifi- tional improvement indeed went along with the changes of these
cant portion of patients had severe pulmonary hypertension. A variables. Finally, patients with severe pulmonary vascular disease
significant and similar decrease in PAP after ASD closure was (.5 Wood units even after vasoreactivity testing or targeted
observed in all age groups. However, older patients who had treatment) were excluded, making it impossible to draw con-

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

higher pre-interventional PAPs still had significantly higher PAPs clusions for these patients.
after intervention and significantly less baseline-adjusted pressure
decrease. The increase in PAP in these patients can be assumed
to be partially due to a high pulmonary flow. However, although
patients with severe pulmonary vascular disease were excluded,
Conclusions
part of the patients had a more or less increased vascular resist- Transcatheter ASD closure can be safely and successfully per-
ance. All patients decreased with their PAP after ASD closure, at formed in adults at any age. Regression of RV size and PAP as
least partially due to the decrease in transpulmonary flow and well as symptomatic improvement can be expected across all
probably also due to some decrease in PVR. Although we age groups. However, the best outcome is achieved in patients
cannot provide invasive follow-up data, it is likely that persistent with less functional impairment and less elevated PAP. Considering
elevation of PVR must be assumed to be the reason for a still elev- the continuous increase in symptoms, RV remodelling, and PAP
ated PAP in part of the patients. Importantly, there was a close with increasing age, ASD closure must be recommended
relation between PAP and functional status. irrespective of symptoms early after diagnosis even in adults of
Similar to our results, Yong et al.42 recently reported that advanced age.
patients with advanced pulmonary arterial hypertension responded
with a significant decrease in PAP after interventional ASD closure Conflict of interest: none declared.
but were less likely to reach normal PAPs and become
asymptomatic.
References
1. Lindsey JB, Hillis LD. Clinical update: atrial septal defect in adults. Lancet 2007;369:
1244 –1246.
Right ventricular remodelling 2. Karamlou T, Diggs BS, Ungerleider RM, McCrindle BW, Welke KF. The rush to
Atrial septal defect closure has been shown to result in a significant atrial septal defect closure: is the introduction of percutaneous closure driving
reduction in RV and an increase in left ventricular volumes.43,44 A utilization? Ann Thorac Surg 2008;86:1584 –1590; discussion 1590 –1591.
3. Murphy JG, Gersh BJ, McGoon MD, Mair DD, Porter CJ, Ilstrup DM,
reduction in RV size occurs immediately after ASD closure with a McGoon DC, Puga FJ, Kirklin JW, Danielson GK. Long-term outcome after surgi-
further reduction within the following 3– 6 months.43,45 The cal repair of isolated atrial septal defect. Follow-up at 27 to 32 years. N Engl J Med
improvement of both RV and LV functions has been reported.46 1990;323:1645 – 1650.
4. Konstantinides S, Geibel A, Olschewski M, Gornandt L, Roskamm H, Spillner G,
Our findings demonstrate that a significant decrease in RV size Just H, Kasper W. A comparison of surgical and medical therapy for atrial septal
can be expected even in elderly patients, although the latter defect in adults. N Engl J Med 1995;333:469 –473.
present with larger ventricles before and after the intervention. 5. Shah D, Azhar M, Oakley CM, Cleland JG, Nihoyannopoulos P. Natural history of
secundum atrial septal defect in adults after medical or surgical treatment: a his-
torical prospective study. Br Heart J 1994;71:224–227; discussion 228.
6. John Sutton MG, Tajik AJ, McGoon DC. Atrial septal defect in patients ages 60
years or older: operative results and long-term postoperative follow-up. Circula-
Mortality tion 1981;64:402 – 409.
So far, the only randomized study of surgical ASD closure in adult 7. Nasrallah AT, Hall RJ, Garcia E, Leachman RD, Cooley DA. Surgical repair of atrial
septal defect in patients over 60 years of age. Long-term results. Circulation 1976;
patients did not find a survival benefit but was limited by the exclu- 53:329–331.
sion of markedly symptomatic patients and a short follow-up. 8. Attie F, Rosas M, Granados N, Zabal C, Buendia A, Calderon J. Surgical treatment
Nevertheless, observational studies suggest that surgical ASD for secundum atrial septal defects in patients .40 years old. A randomized clini-
cal trial. J Am Coll Cardiol 2001;38:2035 –2042.
closure confers a survival benefit: in one retrospective series6 84 9. Gatzoulis MA, Redington AN, Somerville J, Shore DF.Should atrial septal defects
patients with surgical ASD closure had an improved long-term sur- in adults be closed? Ann Thorac Surg 1996;61:657 –659.
vival in comparison with 95 medically treated patients.4 Since 10. Ward C. Secundum atrial septal defect: routine surgical treatment is not of
proven benefit. Br Heart J 1994;71:219 –223.
transcatheter ASD closure can be performed with almost no mor- 11. Webb G. Do patients over 40 years of age benefit from closure of an atrial septal
tality even in the elderly, an overall mortality benefit of the cath- defect? Heart 2001;85:249–250.
eter intervention is even more likely. However, this will never be 12. Chessa M, Carminati M, Butera G, Bini RM, Drago M, Rosti L, Giamberti A,
Pome G, Bossone E, Frigiola A. Early and late complications associated with trans-
proven by a randomized trial since it would be unethical to catheter occlusion of secundum atrial septal defect. J Am Coll Cardiol 2002;39:
perform such a study nowadays. 1061 –1065.
560 M. Humenberger et al.

13. Masura J, Gavora P, Podnar T. Long-term outcome of transcatheter secundum- 29. Elshershari H, Cao QL, Hijazi ZM. Transcatheter device closure of atrial septal
type atrial septal defect closure using Amplatzer septal occluders. J Am Coll defects in patients older than 60 years of age: immediate and follow-up results.
Cardiol 2005;45:505 –507. J Invasive Cardiol 2008;20:173 – 176.
14. Butera G, Romagnoli E, Carminati M, Chessa M, Piazza L, Negura D, Giamberti A, 30. Yalonetsky S, Lorber A. Comparative changes of pulmonary artery pressure
Abella R, Pome G, Condoluci C, Frigiola A. Treatment of isolated secundum atrial values and tricuspid valve regurgitation following transcatheter atrial septal
septal defects: impact of age and defect morphology in 1,013 consecutive patients. defect closure in adults and the elderly. Congenit Heart Dis 2009;4:17– 20.
Am Heart J 2008;156:706 –712. 31. Krumsdorf U, Ostermayer S, Billinger K, Trepels T, Zadan E, Horvath K, Sievert H.
15. Majunke N, Bialkowski J, Wilson N, Szkutnik M, Kusa J, Baranowski A, Heinisch C, Incidence and clinical course of thrombus formation on atrial septal defect and
Ostermayer S, Wunderlich N, Sievert H. Closure of atrial septal defect with the patient foramen ovale closure devices in 1,000 consecutive patients. J Am Coll
Amplatzer septal occluder in adults. Am J Cardiol 2009;103:550 –504. Cardiol 2004;43:302–309.
16. Butera G, Carminati M, Chessa M, Youssef R, Drago M, Giamberti A, Pome G, 32. Gatzoulis MA, Freeman MA, Siu SC, Webb GD, Harris L. Atrial arrhythmia after
Bossone E, Frigiola A. Percutaneous versus surgical closure of secundum atrial surgical closure of atrial septal defects in adults. N Engl J Med 1999;340:839 –846.
septal defect: comparison of early results and complications. Am Heart J 2006; 33. Szkutnik M, Lenarczyk A, Kusa J, Bialkowski J. Symptomatic tachy- and bradyar-
151:228–234. rhythmias after transcatheter closure of interatrial communications with Amplat-
17. Rosas M, Zabal C, Garcia-Montes J, Buendia A, Webb G, Attie F. Transcatheter zer devices. Cardiol J 2008;15:510 –516.
versus surgical closure of secundum atrial septal defect in adults: impact of age 34. Spies C, Khandelwal A, Timmermanns I, Schrader R. Incidence of atrial fibrillation

Downloaded from https://academic.oup.com/eurheartj/article/32/5/553/425441 by guest on 10 December 2020

at intervention. A concurrent matched comparative study. Congenit Heart Dis following transcatheter closure of atrial septal defects in adults. Am J Cardiol 2008;
2007;2:148 –155. 102:902 – 906.
18. Balint OH, Samman A, Haberer K, Tobe L, McLaughlin P, Siu SC, Horlick E, 35. Silversides CK, Haberer K, Siu SC, Webb GD, Benson LN, McLaughlin PR,
Granton J, Silversides CK. Outcomes in patients with pulmonary hypertension Harris L. Predictors of atrial arrhythmias after device closure of secundum type
undergoing percutaneous atrial septal defect closure. Heart 2008;94:1189 –1193. atrial septal defects in adults. Am J Cardiol 2008;101:683 –687.
36. Giardini A, Donti A, Sciarra F, Bronzetti G, Mariucci E, Picchio FM. Long-term
19. de Lezo JS, Medina A, Romero M, Pan M, Segura J, Caballero E, Pavlovic D,
incidence of atrial fibrillation and flutter after transcatheter atrial septal defect
Ortega JR, Franco M, Delgado A, Ojeda S, Mesa D, Lafuente M. Effectiveness
closure in adults. Int J Cardiol 2009;134:47 –51.
of percutaneous device occlusion for atrial septal defect in adult patients with pul-
37. Silversides CK, Siu SC, McLaughlin PR, Haberer KL, Webb GD, Benson L,
monary hypertension. Am Heart J 2002;144:877 –880.
Harris L. Symptomatic atrial arrhythmias and transcatheter closure of atrial
20. Brochu MC, Baril JF, Dore A, Juneau M, De Guise P, Mercier LA. Improvement in
septal defects in adult patients. Heart 2004;90:1194 –1198.
exercise capacity in asymptomatic and mildly symptomatic adults after atrial septal
38. Cherian G, Uthaman CB, Durairaj M, Sukumar IP, Krishnaswami S, Jairaj PS, John S,
defect percutaneous closure. Circulation 2002;106:1821 – 1826.
Krishnaswami H, Bhaktaviziam A. Pulmonary hypertension in isolated secundum
21. Giardini A, Donti A, Formigari R, Specchia S, Prandstraller D, Bronzetti G,
atrial septal defect: high frequency in young patients. Am Heart J 1983;105:
Bonvicini M, Picchio FM. Determinants of cardiopulmonary functional improve-
952 –957.
ment after transcatheter atrial septal defect closure in asymptomatic adults.
39. Craig RJ, Selzer A. Natural history and prognosis of atrial septal defect. Circulation
J Am Coll Cardiol 2004;43:1886 – 1891.
1968;37:805 –815.
22. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA,
40. Steele PM, Fuster V, Cohen M, Ritter DG, McGoon DC. Isolated atrial septal
Picard MH, Roman MJ, Seward J, Shanewise J, Solomon S, Spencer KT, St John
defect with pulmonary vascular obstructive disease—long-term follow-up and
Sutton M, Stewart W. Recommendations for chamber quantification. Eur J Echo- prediction of outcome after surgical correction. Circulation 1987;76:1037 –1042.
cardiogr 2006;7:79 –108. 41. Sachweh JS, Daebritz SH, Hermanns B, Fausten B, Jockenhoevel S, Handt S,
23. Dittmann H, Jacksch R, Voelker W, Karsch KR, Seipel L. Accuracy of Doppler Messmer BJ. Hypertensive pulmonary vascular disease in adults with secundum
echocardiography in quantification of left to right shunts in adult patients with or sinus venosus atrial septal defect. Ann Thorac Surg 2006;81:207–213.
atrial septal defect. J Am Coll Cardiol 1988;11:338 –342. 42. Yong G, Khairy P, De Guise P, Dore A, Marcotte F, Mercier LA, Noble S,
24. Du ZD, Hijazi ZM, Kleinman CS, Silverman NH, Larntz K. Comparison between Ibrahim R. Pulmonary arterial hypertension in patients with transcatheter
transcatheter and surgical closure of secundum atrial septal defect in children and closure of secundum atrial septal defects: a longitudinal study. Circ Cardiovasc
adults: results of a multicenter nonrandomized trial. J Am Coll Cardiol 2002;39: Interv 2009;2:455 –462.
1836 –1844. 43. Schussler JM, Anwar A, Phillips SD, Roberts BJ, Vallabhan RC, Grayburn PA. Effect
25. Thomson JD, Aburawi EH, Watterson KG, Van Doorn C, Gibbs JL. Surgical and on right ventricular volume of percutaneous Amplatzer closure of atrial septal
transcatheter (Amplatzer) closure of atrial septal defects: a prospective compari- defect in adults. Am J Cardiol 2005;95:993–995.
son of results and cost. Heart 2002;87:466 – 469. 44. Teo KS, Dundon BK, Molaee P, Williams KF, Carbone A, Brown MA,
26. Galal MO, Wobst A, Halees Z, Hatle L, Schmaltz AA, Khougeer F, De Vol E, Worthley MI, Disney PJ, Sanders P, Worthley SG. Percutaneous closure of
Fawzy ME, Abbag F, Fadley F, Duran CMG. Peri-operative complications following atrial septal defects leads to normalisation of atrial and ventricular volumes.
surgical closure of atrial septal defect type II in 232 patients—a baseline study. Eur J Cardiovasc Magn Reson 2008;10:55.
Heart J 1994;15:1381 – 1384. 45. Sun P, Wang ZB, Xu CJ, Yang SM, Jiang L. Echocardiographic and morphological
27. Horvath KA, Burke RP, Collins JJ Jr, Cohn LH. Surgical treatment of adult atrial evaluation of the right heart after closure of atrial septal defects. Cardiol Young
septal defect: early and long-term results. J Am Coll Cardiol 1992;20:1156 – 1159. 2008;18:593 –598.
28. Swan L, Varma C, Yip J, Warr M, Webb G, Benson L, McLaughlin P. Transcatheter 46. Salehian O, Horlick E, Schwerzmann M, Haberer K, McLaughlin P, Siu SC,
device closure of atrial septal defects in the elderly: technical considerations and Webb G, Therrien J. Improvements in cardiac form and function after transcath-
short-term outcomes. Int J Cardiol 2006;107:207 –210. eter closure of secundum atrial septal defects. J Am Coll Cardiol 2005;45:499 –504.