ORI GI NAL ARTI CLE

Arrhythmias after transcatheter closure of perimembranous

ventricular septal defects with a modied double-disk
occluder: early and long-term results
Pan Li

Xian-xian Zhao

Xing Zheng

Yong-wen Qin
Received: 13 January 2011 / Accepted: 6 May 2011 / Published online: 4 June 2011
Springer 2011
Abstract With the development of interventional tech-
niques and devices, transcatheter closure of perimembra-
nous ventricular septal defect has been widely performed.
However, there has been a lack of long-term follow-up
results about postoperative ECG changes of PmVSD
patients. We report our experience of early and late
arrhythmias after transcatheter closure of PmVSD with a
modied double-disk occluder (MDVO). We performed a
retrospective review of 79 patients (47 males, 32 females)
between September 2002 and May 2007 who underwent
transcatheter closure of perimembranous ventricular septal
defect. Symmetric and asymmetric PmVSD occluders were
used. The diameter of the evaluated defects ranged from 3
to 12 mm, as measured by TTE and 3 to 15 mm by left
ventriculography. Most cases of PmVSD were treated
successfully with a single procedure, resulting in a suc-
cessful closure rate of 97% (77/79 patients). There was no
death in any of the patients. After the operation, 79 patients
were followed-up for a range of 1076 months
(35.3 17.4 months). In this series, 11 cases of incom-
plete right bundle branch block and ve cases of complete
right bundle branch block occurred during the early period
after operation. During long-term follow-up, these issues
declined in prevalence to ve and four cases, respectively.
Moreover, reversible third-degree AVB occurred during
closure or after the procedure, and two of the three patients
with reversible AVB received a temporary heart pacemaker
implantation. These patients recovered 1 h, 6 days, and
9 days later, respectively. During 1076 months of follow-
up, no complications occurred in any of the patients,
including residual shunt, severe aortic valve, or tricuspid
valve regurgitation. Device closure of perimembranous
ventricular septal defects with a modied double-disk
occluder (MDVO) resulted in excellent closure rates and
acceptably low arrhythmia rates.
Keywords Transcatheter closure Congenital heart
disease Septal defects Arrhythmia
Introduction
Ventricular septal defects (VSD) are one of the most
common forms of congenital heart diseases (CHD),
accounting for 1030% of all cases. Approximately 75% of
VSD cases are of the perimembranous type [1]. There
currently exist two common therapies for treating a peri-
membranous ventricular septal defect (PmVSD): surgical
repair and interventional device closure. Compared to
surgery, the benets of interventional therapy include
minimal wound and early recovery, as well as no need for
use of a heartlung machine, thereby diminishing device-
associated complications [2, 3]. Since the development of
the Amplatzer Membranous VSD Occluder (AGA Medical
Corp, Plymouth, MN, USA), transcatheter device closure
of PmVSD has become the primary therapy at many
institutions [4]. The rate of successful closure using the
Amplatzer has been reported to be between 90 and 100%
[5, 6]. As the defect occurs adjacent to the cardiac con-
duction system, arrhythmia is a common complication after
closure in patients with PmVSD [7]. Serious arrhythmias,
especially those associated with complete or type II sec-
ond-degree AV block and complete left bundle branch
P. Li X. Zhao X. Zheng Y. Qin (&)
Department of Cardiology, Changhai Hospital,
Second Military Medical University,
168 Changhai Road, Shanghai 200433,
Peoples Republic of China
e-mail: ywqin1@yahoo.com.cn
1 3
Heart Vessels (2012) 27:405410
DOI 10.1007/s00380-011-0155-z
block (CLBBB) could affect patients ventricular function
and quality of life. Hence, long-term evaluation with ECGs
is needed to assess early and late arrhythmias. However,
there has been a lack of long-term follow-up of postoper-
ative ECG changes among PmVSD patients [8]. In the
current study, we report on 37 years of follow-up ECG
results of 79 patients who underwent transcatheter closure
of PmVSD using modied double-disk occluder (MDVO)
with symmetric and asymmetric left disks.
Materials and methods
Between September 2002 and May 2007, 79 patients with
PmVSD who accepted interventional therapy using modi-
ed double-disk occluders (MDVO) were enrolled in this
study. Clinical records, electrocardiograms, transthoracic
echocardiography (TTE) and angiograms were recorded
and reviewed. All patients were screened by TTE in the
long-axis parasternal view and the apical ve-chamber
view to measure the rim under the aortic valve, and also in
the short-axis parasternal view to measure the rim from the
tricuspid valve to the site of the defect. Left ventriculog-
raphy was used to evaluate the shape, size, and location of
the defect, as well as the distance from the aortic valve to
the defect. The Ethics Committee of our hospital approved
the study, and informed consent was obtained from each
patient.
Patient characteristics
Including criteria for closure were as follows: (1) age more
than 3 years old; (2) PmVSD diagnosed by TTE and the
diameter of defects between 3 and 16 mm; (3) left-to-right
shunt. Exclusion criteria included: (1) aortic valve pro-
lapse, (2) severe aortic or tricuspid regurgitation, (3) mean
pulmonary artery pressure C70 mmHg, (4) right-to-left
shunting through the defect, and (5) New York Heart
Association (NYHA) functional class IV. No patient with
postoperative residual PmVSD or postinfarction was
included. The characteristics of all patients and the pro-
cedural data are summarized in Table 1.
Occluder devices and occluder selection
Amplatzer Membranous VSD occluder is the only device
that is specically designed for PmVSD, and has been
widely used in the clinical setting since 2002 [9]. In the
current study, we used an MDVO (Shanghai Shape
Memory Alloy Ltd, China), a new device that was
designed based on the Amplatzer occluder [4]. Due to its
high efciency, exibility, and versatility, the modied
double-disk occluder has been widely used in treating
Chinese patients with VSD. The occluder was made of
exible nitinol wire (0.01 mm) covered in a polyester
fabric, which was sutured to the nitinol wire and xed
into a double round shape. As needed, several sizes of
occluder (420 mm) can be provided to operators. The
ends of the occluders are xed by 316L medical stainless
steel. The xation steel ring of one end has a screw-
thread connecting with the screw nail of the push cable
head. The occluders can be delivered by a 610 French
sheath [4]. There are three types of MDVOs in this study:
a double-disc symmetrical occluder, a small-waist double-
disc asymmetric occluder, and a zero eccentricity occlu-
der (Fig. 1). In the double-disc symmetrical occluder, the
diameter of the left disk is 4 mm larger than that of the
waist. The diameters of the small-waist double-disc
asymmetric occluder are different in the left- and right-
side disc, with the left disk being 6 or 8 mm larger in
diameter than the waist. The above two types of MDVOs
were used for defects with a rim C2 mm under the aortic
valve. On the other hand, the zero eccentricity occluder
was used when defects had a rim of \2 mm. In this last
Table 1 Clinical characteristic of patients
n Rate
(%)
Gender (n) 79
Men 47 59.5
Women 32 40.5
Age (years) 15.7 13.5
\15 (n) 45 57.9
C15 (n) 34 43.0
Procedure time (min) 51 26 (20146)
Fluoroscopic time (min) 12 10 (258)
Systolic PA pressure (mm Hg) 34 12 (1160)
Mean PA pressure (mm Hg) 22 9 (766)
Pulmonary to systemic ow ratio 2 0.5 (1.62.8)
VSD diameter on TTE (mm) 5.6 2.6 (312)
Mean size of the device used (mm) 7.1 3.0 (418)
Sub-aortic rim B2 mm (n) 17
Sub-aortic rim[2 mm (n) 62
Defect shape by ventriculography (n)
Infundibular 22 27.8
Aneurysmal 38 48.1
Tubular 16 20.2
Window-like 3 3.7
Type of device used (n)
Double-disc symmetric occluder 36
Small-waist double-disc asymmetric
occluder
19
Zero eccentricity occluder 24
Two occluders 8
406 Heart Vessels (2012) 27:405410
1 3
occluder, the diameter of the left disk is 6 mm larger
than that of the waist and the left disk extends towards
the apex while no superior margin extends towards the
aorta.
Procedure
Adult patients received local anesthesia with 2% Lido-
caine, while children (B10 years) received general anes-
thesia. Heparin (100 IU/kg) was injected intravenously
before the procedure. Initially, the right femoral artery and
vein were punctured. A pigtail catheter was placed into the
left ventricle through the femoral artery sheath. The size,
location, shape, and relationship between the aortic and
tricuspid valves were assessed by left ventricular angiog-
raphy at 4560 in the left anterior oblique and a 25
cranial tilt. A 79 F delivery sheath was placed into the left
ventricular apex. The guidewire and catheter were
advanced together to the left ventricle from the femoral
artery. The size of the device was at least 12 mm larger
than the maximum diameter of the defect according to
ventriculography. When using the asymmetric occluder,
the platinum marker on the left disk was positioned
towards the apex. The device was deployed only when its
proper position was obtained and interference with the
aortic and tricuspid valve had been excluded on the basis of
aortic angiography, ventriculography and TTE. The
patients were monitored with TTE and continuous ECG for
57 days before discharge. After the procedure, patients
were treated with daily oral aspirin (35 mg/kg) for 1 month
and thereafter with 100 mg/day (for children\10 years old:
35 mg/kg/day) for 5 months.
Follow-up
After discharge, all patients were followed-up in an out-
patient clinic. The shape and orientation of the device and
residual shunt were re-evaluated by TTE, chest radiogra-
phy and electrocardiography (ECG) at 1, 6, and 12 months
postoperatively and once a year thereafter.
Results
Between September 2002 and May 2007, 79 patients
(47 males, and 32 females; mean age SD = 15.7 13.5
years) underwent PmVSD occluder closure. The diameter
of the evaluated defects ranged from 3 to 12 mm as mea-
sured by TTE and 3 to 15 mm by left ventriculography.
The mean (SD) diameter of the device used was
7.1 3.0 mm. The length of the ventricular septal rim
below the aortic valve ranged from 0 to 5 mm. According
to ventriculography, the shape of defects were described as
infundibular, aneurysmal, tubular, and window-like types
in 27.8, 48.1, 20.2, and 3.7% of patients, respectively. VSD
coexisted with ASD in three patients and with PDA in ve
patients. The double-disc symmetric, the small-waist dou-
ble-disc asymmetric, and the zero eccentricity occluder
was used in 36, 19, and 24 patients, respectively. Among
the patients for whom small-waist double-disc asymmetric
and zero eccentricity occluders were used, 28 patients had
an aneurysmal type (Table 2).
All 79 patients were followed-up for a range of
1076 months (35.3 17.4 months). The follow-up rates
at 1, 3, and 5 years were 94, 85, and 73%, respectively.
Fig. 1 The three types of MDVO used to close PmVSD. a Double-disc symmetrical occluder, b small-waist double-disc asymmetrical occluder,
c zero eccentricity occluder
Table 2 Three kinds of
occluder used in different shape
of defects
Infundibular Aneurysmal Tubular Window-like
Double-disc symmetric 15 10 11 0
Small-waist double-disc 3 16 0 0
Zero eccentricity occluder 4 12 5 3
N 22 38 16 3
Rate 27.8% 48.1% 20.3% 3.8%
Heart Vessels (2012) 27:405410 407
1 3
Most cases of PmVSD were treated successfully with a
single procedure, resulting in a successful closure rate of
97% (77/79 patients). Fifteen minutes after the procedure,
62 patients had no residual shunt and 17 patients had a
trace amount of residual shunt veried by ventriculography
and TTE. TTE revealed that the residual shunt disappeared
in 12 patients within 57 days post-operatively, while the
residual shunt in ve patients continued. Before the pro-
cedure, tricuspid or aortic regurgitation was found in 18
and 6 patients, respectively. During the follow-up period,
among 61 preoperative patients with an absence of tricus-
pid regurgitation, seven new cases of mild tricuspid
regurgitation developed. Among 73 preoperative patients
without aortic regurgitation, three new cases of mild aortic
regurgitation developed. In contrast, 15 of 18 patients with
pre-existing moderate to severe tricuspid regurgitation
showed a decrease in its severity. There was no death in
any of the patients.
Pre- and postprocedure ECGs and 24-h dynamic ECGs
were reviewed to measure the heart rate, PR interval, QRS
duration, and QRS axis, along with presence and frequency
of arrhythmias. After MDVO placement, there was no
signicant change in QRS duration and QRS axis. How-
ever, the heart rate and PR interval had signicant differ-
ences between pre-procedure ECGs and the long-term
follow-up ECGs results (p \0.05; Table 3). ECGs at
57 days after operation and at long-term follow-up were
both available for all 79 patients. In the meantime, data of
the new occurring arrhythmia during follow-up were also
collected in all patients. The types of arrhythmia that were
observed and analyzed included blockage of heart con-
duction (atrioventricular block, right bundle branch block,
left bundle branch block), atrial brillation, and premature
contraction (ventricular premature beat, atrial premature
beat). Of all the patients in the study, 27 had an aneurysmal
PmVSD (group 1), while the remaining 52 patients had a
non-aneurysmal PmVSD (Group 2). In group 1, arrhythmia
developed within 57 days post-procedure in 12 cases
(44.4%), but this prevalence reduced signicantly to 11.1%
(three cases) over the long term (p = 0.006). In group 2,
six new cases (11.5%) of arrhythmia developed in the early
period after operation and this prevalence reduced to 3.8%
(two cases) over the long term. However, this change was
not statistically signicant (p = 0.141).
Conduction abnormalities were the main postoperative
complications. Specically, 11 cases of incomplete right
bundle branch block and ve cases of complete right
bundle branch block occurred during the early period after
operation. During long-term follow-up, these issues
declined in prevalence to ve and four cases, respectively.
While reversible AVB occurred during closure or after the
procedure (Fig. 2), none of the patients were symptom-
atic. However, two of the three patients with reversible
AVB received a temporary heart pacemaker implantation
(Table 4). These patients recovered 1 h, 6 days, and
9 days later, respectively. No other complications occur-
red in the three patients during long-term follow-up.
Furthermore, not a single case of severe arrhythmia
occurred in the 79 patients throughout long-term follow-
up for 35 years.
Discussion
Post-operative complications of PmVSD include arrhyth-
mia, hemolysis, aortic valve insufciency, tricuspid insuf-
ciency, and occluder displacement [10]. As the defect is
adjacent to the cardiac conduction system, arrhythmia is
the most common complication seen in the clinic. Many
studies have reported the short-term follow-up results of
PmVSD patients treated by interventional technique and
with the use of a device, nding high success rates and less
complications compared with operation. Unfortunately,
there exists little long-term PmVSD follow-up data, espe-
cially for complicated PmVSD, such as large, or multi-
shunt aneurysmal defects, or defects with small sub-aortic
or sub-tricuspid rims. In this study, we summarized our
long-term experience over 35 years of follow-up of 79
patients receiving transcatheter closure using modied
double-disk occluders. Most cases (97%) of PmVSD were
treated successfully with the initial procedure, and no
deaths occurred. These results from our department suggest
that transcatheter closure of PmVSD is a safe and effective
treatment.
Table 3 Electrocardiographic values before and after procedures
Preprocedure (mean SD) 57 days after operation At the end of the follow-up
Mean SD t value p value Mean SD t value p value
Heart rate (beats/min) 84 18 87 22 1.214 0.228 80 15 -2.077 0.041
PR interval (ms) 141 22 144 33 0.935 0.353 147 21 2.577 0.012
QRS duration (ms) 86 15 101 8 1.658 0.101 110 12 1.750 0.084
QRS axis () 23.5 40.3 21.7 37 -0.418 0.677 17.78 -1.304 0.196
408 Heart Vessels (2012) 27:405410
1 3
The successful closure of a PmVSD depends on the
diameter, anatomical location, as well as the shape of the
defect. According to ventriculography, the shape of defects
in the current series can be roughly classied into infun-
dibular, aneurysmal, tubular, and window-like. This clas-
sication was helpful in properly selecting the device. For
example, infundibular tubular and window-like defects are
easily treated with satisfactory results. On the other hand,
PmVSDs with aneurysm, especially for multi-hole VSDs,
are more troublesome and require treatment by a special
small-waist double-disc asymmetric occluder that can
cover all the left inlets of VSD completely with little
interference on tricuspid valve function. It is generally not
recommended to position the occluder inside the aneurysm,
because studies show that closing the outlet of the aneu-
rysm might induce further enlargement of the aneurysm
[11]. Previous studies have suggested that PmVSDs with a
sub-aortic rim of less than 2 mm should not undergo
transcatheter treatment [12]. However, in the current study,
17 PmVSD patients with sub-aortic rims of less than 2 mm
were closed successfully with MDVOs. The key to suc-
cessful closure is the left disk extension towards the apex
and no superior margin extension towards the aorta while
using the zero eccentricity occluder. Furthermore, in an
effort to avoid aortic regurgitation when using the asym-
metric device, left ventriculography must be performed
before deploying the occluder. In our experience, the zero
eccentricity occluder was suitable for those patients with a
B2 mm sub-aortic rim and the long-term follow-up results
were encouraging.
The membranous portion of the interventricular septum
is proximal to the heart conduction system. Thus, com-
pression of double round discs and the waist of occluder
will lead to edema and inammatory exudation of sur-
rounding tissues, resulting in occurrence of new heart block
or deterioration of previous conduction block. Among
them, the cAVB and bundle branch block are the most
common, occurring often within 13 days after operation.
Nevertheless, cases of late arrhythmias occurring at
12 weeks after operation or longer are not rare.
Fig. 2 a The preprocedure electrocardiographic results of a 25-year-
old female were basically normal. b Type II second-degree AVB was
seen 4 days after the operation, which resolved during long-term
follow-up (c)
Table 4 Patients with AV block after occluder implantation
Age
(years)
Procedure
time (min)
Sub-aortic
rim (mm)
Device
used
ECG pre-
procedure
Time of AVB
occurrence
Duration of
AVB (h)
ECG before
discharge
Treatment ECG at
35 years
Case 1 30 90 1 7 mm
zero
LAH 5 days after
implantation
1 LAH Corticosteroid LAH
Case 2 25 100 2 12 mm
zero
Normal During
procedure
144 Accidental
APB
TP and
corticosteroid
CRBBB
Case 3 19 50 0 10 mm
zero
Normal 4 days after
implantation
216 Incomplete
RBBB
TP and
corticosteroid
Normal
Zero zero eccentricity occluder; LAH left atrial hypertrophy, CRBBB complete right bundle branch block, APB atrial premature beats, RBBB right
bundle branch block, TP temporary pacemaker
Heart Vessels (2012) 27:405410 409
1 3
However, due to a lack of long-term PmVSD follow-
up data, delayed ECG changes were rarely considered by
physicians. Butera et al. [12] reported longer-term fol-
low-up in 104 patients with the Amplatzer Membranous
VSD Occluder at a median of 38.5 months. The inci-
dence of CHB was 8.7%, and a permanent pacemaker
was required in six (5.7%) patients; two in the early
phase and four during late follow-up. However, in our 79
cases, no cases of permanent AVB occurred meanwhile
transient third-degree AVB occurred in three patients
(3.8%) within the rst week after the procedure. This
rate is also slightly below the 5% (13/250) incidence of
cAVB that has been reported by the unofcial European
registry using Amplatzer occluder [13]. There are two
possible explanations for the discrepant ndings. Firstly,
most of our patients were adults or older children
whereas the majority of patients in several other studies
were predominantly children [12, 13]. The younger the
children are, the larger the ratio of device area/ventric-
ular septum area. Accordingly, for the same size of
device, the odds of affecting the conduction tissue is
higher. Secondly, the waist is longer in the MDVO than
in the Amplatzer occluder (22.5 vs. 1.5 mm). Thus, the
strain of the septum might be smaller in our patients,
which might play a role [14]. Effectiveness of cortico-
steroid therapy suggests that the edema and inammation
of the conduction tissue compressed by the occluder play
important roles in the mechanism of conduction block.
The physical injury of the conduction tissue caused by
the guide wire and catheter is another mechanism for
conduction block. In addition, chronic inammation or
brosis may possibly contribute to the late arrhythmias
[15].
One primary limitation of this report is the lack of
transesophageal echocardiography use, which may have
limited our ability to clearly visualize some defects before
closure, especially for patients with complicated defects.
Furthermore, the results of the current report should be
considered preliminary and must be conrmed in a larger
trial with longer-term follow-up.
In current clinical practice, PmVSD percutaneous clo-
sure is a valuable alternative to surgery. According to early
postoperative and long-term follow-up results in our
department, the transcatheter closure is effective with
limited residual shunt and no persistent serious arrhythmias
or other major complications. Our ndings suggest that
closure of PmVSD with a MDVO is both safe and
effective.
References
1. Mavroudis C, Backer CL, Jacobs JP (2003) Ventricular septal
defect. In: Pediatric cardiac surgery, 3rd edn. Mosby, Philadel-
phia, pp 298320
2. Masutani S, Taketazu M, Ishido H, Iwamoto Y, Yoshiba S,
Matsunaga T, Kobayashi T, Senzaki H (2011) Effects of age on
hemodynamic changes after transcatheter closure of atrial septal
defect: importance of ventricular diastolic function. Heart Ves-
sels. doi:10.1007/s00380-011-0122-8
3. Oses P, Hugues N, Dahdah N, Vobecky SJ, Miro J, Pellerin M,
Poirier NC (2010) Treatment of isolated ventricular septal defects
in children: Amplatzer versus surgical closure. Ann Thorac Surg
90:15931598
4. Gu MB, You XH, Zhao XX, Zheng X, Qin YW (2011) Trans-
catheter device closure of intracristal ventricular septal defects.
Am J Cardiol 107:110113
5. Holzer R, de Giovanni J, Walsh KP, Tometzki A, Goh T, Hakim
F, Zabal C, de Lezo JS, Cao QL, Hijazi ZM (2006) Transcatheter
closure of perimembranous ventricular septal defects using the
Amplatzer membranous VSD occluder: immediate and mid-term
results of an international registry. Catheter Cardiovasc Interv
68:620628
6. Shih CH, Liang PC, Chiang FT, Tseng CD, Tseng YZ (2010)
Transcatheter embolization of a huge renal arteriovenous stula
with Amplatzer Vascular Plug. Heart Vessels 25:356358
7. Butera G, Chessa M, Carminati M (2007) Percutaneous closure of
ventricular septal defects. Cardiol Young 17:243253
8. Masura J, Gao W, Gavora P, Sun K, Zhou AQ, Jiang S, Ting-
Liang L, Wang Y (2005) Percutaneous closure of perimembra-
nous ventricular septal defects with the eccentric Amplatzer
device: multicenter follow-up study. Pediatr Cardiol 26:216219
9. Holzer RJ, Balzer D, Cao QL, Lock K, Hijazi ZM (2004) Device
closure of muscular ventricular septal defects using the Amplat-
zer muscular VSD occluder: immediate and midterm results of a
US registry. J Am Coll Cardiol 43:12571263
10. Jian-Jun G, Xue-Gong S, Ru-Yuan Z, Min L, Sheng-Lin G, Shi-
Bing Z, Qing-Yun G (2006) Ventricular septal defect closure in
right coronary cusp prolapse and aortic regurgitation complicat-
ing VSD in the outlet septum: which treatment is most appro-
priate? Heart Lung Circ 15:168171
11. Wu H, Qin YW, Zhao XX, Hu JQ, Zheng X, Wang ES, Li WP,
Wang YY, Liu B, Liu J, Wang SQ (2009) Transcatheter closure
of multi-hole perimembranous VSD with aneurysm: 3-year fol-
low-up study. Clin Res Cardiol 98:563569
12. Butera G, Chessa M, Carminati M (2007) Transcatheter closure
of perimembranous ventricular septal defects: early and long-
term results. J Am Coll Cardiol 50:11891195
13. Pinto RJ, Dalvi BV, Sharma S (2006) Transcatheter closure of peri-
membranous ventricular septal defects using the Amplatzer asym-
metric ventricular septal defect occluder: preliminary experience
with 18-month follow-up. Catheter Cardiovasc Interv 68:145152
14. Qin YW, Chen JM, Zhao XX, Liao D, Mu R, Wang S, Wu H,
Guo H (2008) Transcatheter closure of perimembranous ven-
tricular septal defect using a modied double-disk occluder. Am J
Cardiol 101:17811786
15. Predescu D, Chaturvedi RR, Friedberg MK (2008) Complete heart
block associated with device closure of perimembranous ventric-
ular septal defects. J Thorac Cardiovasc Surg 136:12231229
410 Heart Vessels (2012) 27:405410
1 3
Copyright of Heart & Vessels is the property of Springer Science & Business Media B.V. and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.