Download Predictors for Regression of Large Secundum Atrial Septal Defects

Original Article
Acta Cardiol Sin 2013;29:82-87
Pediatric Cardiology
Predictors for Regression of Large Secundum
Atrial Septal Defects Diagnosed in Infancy
Kuan-Miao Lin, Chi-Di Liang, Shao-Ju Chien, Ying-Jui Lin, I-Chun Lin, Mao-Hung Lo,
Ting-Hsin Wu and Chien-Fu Huang
Purpose: To determine predictive factors of spontaneous closure or size reduction in large secundum-type atrial
septal defects (ASD) diagnosed in infancy prior to catheterization or surgical intervention.
Methods: From June 2003 to October 2009, 59 infants diagnosed with secundum-type ASDs measuring ³ 8 mm in
the first year of life were retrospectively enrolled. We reviewed medical records, as well as electrocardiography
and echocardiography findings. Patients were divided into 2 groups according to the last ASD size: group A (n = 23),
ASD reduction in size to < 5 mm or spontaneous closure; group B (n = 36), size of ASD remained ³ 5 mm.
Results: The ASDs spontaneously closed in 10 (17%) patients at a median age of 26.0 ± 5.1 months (range, 15-58
months), or decreased to < 5 mm in 13 (22%) (range, 6-27 months) patients. There was a significant difference in
age at diagnosis between the 2 groups (p = 0.014). Patients in group A were younger than those in group B at the
time of diagnosis. Changes in ASD size (p < 0.001) and body weight percentile (p = 0.01) were also significantly
different from the 6-month follow-up. ASD diameter of ³ 10 mm was a negative predictive factor for size reduction
(p = 0.017).
Conclusions: Spontaneous closure or size reduction of large ASDs was not uncommon in patients diagnosed during
infancy. Patients with initial ASD sizes between 8 and 10 mm who were younger at the time of diagnosis and
showed better weight gain were more likely to have favorable outcomes.
Key Words:
Infancy · Large secundum atrial septal defect · Natural course · Spontaneous closure
INTRODUCTION
meters of > 8 mm. A series of studies showed that these
ASDs rarely decreased in diameter or closed spontaneously because of large left-to-right shunting.1,2 A
large ASD should be closed surgically or through interventional catheterization to prevent progression of pulmonary hypertension, atrial arrhythmia, or heart failure. It is recommended that the optimum period during
which to close ASDs through interventional catheterization is when certain signs and symptoms are observed, such as large ASDs with right heart volume overload or pulmonary-to-systemic flow ratio is above 1.5.
This window of opportunity typically arises when children are between two and four years of age, but they
may be even older.3
Before intervention, clinicians should evaluate ASD
size, patient growth, body weight gain, and signs of
heart failure. Echocardiography and electrocardiography
Most secundum atrial septal defects (ASDs) are
well-tolerated in infancy. Small ASDs, defined as those
with diameters of less than 5 mm, have stable hemodynamics and high rates of spontaneous closure during
infancy. 1 Large ASDs are defined as those with dia-
Received: December 30, 2011 Accepted: April 25, 2012
Department of Pediatric Cardiology, Kaohsiung Chang Gung Memorial
Hospital and Chang Gung University College of Medicine, Kaohsiung,
Taiwan.
Address correspondence and reprint requests to: Dr. Chien-Fu Huang,
Department of Pediatric Cardiology, Kaohsiung Chang Gung Memorial
Hospital and Chang Gung University College of Medicine, No. 123,
Ta-Pei Road, Niao Sung District, Kaohsiung 833, Taiwan. Tel:
886-7-731-7123 ext. 8259; Fax: 886-7-733-8009; E-mail: jeff4474
@cgmh.org.tw
Acta Cardiol Sin 2013;29:82-87
82
Regression of Large ASD Diagnosed in Infancy
measured in the subcostal coronal and sagittal views
during the cardiac cycle, to avoid the risk of overestimating the ASD diameter on color Doppler images. RVH
was diagnosed when we observed paradoxical motion
of the interventricular septum on the M-mode study
and/or the presence of larger right ventricular volume
than left ventricular volume on the apical four-chamber
view.6 Additionally, echocardiographic evaluations were
reviewed by 2 cardiologists.
Transcatheter closure of ASD with the Amplatzer
Septal Occluder (ASO; AGA Medical Corp., Golden Valley, MN, USA) was performed for patients who fulfilled
the closure criteria, including age of more than 2 years,
body weight over 10 kilograms, ratio of pulmonary
blood flow to systemic blood flow (Qp/Qs ratio) ³ 1.5,
and/or presence of RVH.7,8
Groups A and B were compared in terms of gender,
gestational age at birth, age at diagnosis, initial minimal
heart rate, presence of RVH, medical therapy, and associated cardiac anomalies. ASD diameter and weight percentile were recorded during the 3-, 6-, 12- and 24month follow-up visits. The timing of the last follow-up
visit varied with patient age and time of follow-up.
(ECG) should be performed regularly to monitor ASD
size and to detect evidence of right heart dilatation or
pulmonary hypertension. Information about the natural
course of large ASDs is useful for decisions regarding
clinical management. The aim of the present study was
to analyze predictive factors of spontaneous closure or
regression in size of large ASDs in infants.
MATERIALS AND METHODS
From June 2003 to October 2009, 59 infants (18
boys and 41 girls) less than 1 year of age, with large
secundum-type ASD (defined as ASDs with diameters of
³ 8 mm) were enrolled in our study. The nature of our
study was retrospective, and data collection was approved by the Institutional Review Board of Chang Gung
Memorial Hospital. Exclusion criteria included the presence of multiple ASDs, and then combined with other
congenital heart diseases, observation time less than 24
months before closure of ASDs, or failure to attend
follow-up visits.
We followed-up patients regularly, according to
their clinical indications. Reduction in ASD size was defined as a decrease to less than 5 mm, as measured on 2
separate occasions. All 59 patients were divided into 2
groups, as determined by the ASD size at the last patient
visit. Patients in group A showed significant reductions
in ASD size to < 5 mm or spontaneous closure. Among
group B patients, ASDs remained at ³ 5 mm in size.
Medical records were reviewed, including chest radiography, ECG, echocardiography, and cardiac catheterization. These medical records also included information
on age at diagnosis, gender, body weight, and weight
percentile according to the World Health Organization
(WHO) Child Growth Standards database,4 clinical symptoms, medication used (digoxin and/or furosemide), and
whether device closure was performed. ECG was recorded as rhythm, conduction, and right ventricular hypertrophy (RVH) by age-related interpretation.5
Complete echocardiography examination was performed using the Philips SONOS 7500 system (Philips,
Andover, MA, USA). Data were collected using 2-dimensional images, M-mode transducer, pulse wave/continuous wave, and color flow Doppler. The maximal diameter of ASDs obtained by 2-dimensional imaging was
Statistical analysis
The statistical analysis was performed with the
Statistical Package for the Social Sciences version 19
(SPSS, Chicago, IL, USA). The non-parametric data were
expressed as mean ± standard error or median (range).
Clinical parameters as categorical variables between 2
groups were compared by chi-square test or Fisher’s exact test, as appropriate. Quantitative variables as comparisons of serial ASD size and body weight percentile
were performed using the Mann-Whitney U test. A
Kaplan-Meier survival function was used to predict ASD
closure as a function of time, and the log-rank test was
used for comparison of two curves. A p value of < 0.05
was considered significant.
RESULTS
The median age at diagnosis was 4.0 months (range,
5 days to 11 months), and the median ASD diameter
upon diagnosis was 9.4 ± 0.3 mm (range, 8.1-19.0 mm).
The median duration of follow-up was 33.5 ± 2.1 months
83
Acta Cardiol Sin 2013;29:82-87
Kuan-Miao Lin et al.
tile, p = 0.005; at the 12th month: 55 ± 6 percentile vs.
35 ± 5 percentile, p = 0.012; Figure 2). It also happened
at the time of decreasing ASD size. Our finding suggests
that good body weight gain contributes to size reduction in ASD.
The likelihood of ASDs reaching sizes of < 5 mm is
described by a Kaplan-Meier survival curve (Figure 3).
Patients with ASDs of initial diameters 8-10 mm had a
significantly greater likelihood of reduction to < 5 mm
than those with ASD initial diameter of > 10 mm (p =
0.017). These data suggest the initial size is an important predictor of size reduction, even in large ASDs.
(range, 6-72 months). The clinical symptoms at the time
of diagnosis included heart murmur (90%), respiratory
distress (8%), and arrhythmia (2%).
There were 23 patients in group A and 36 patients in
group B; their clinical characteristics are shown in Table
1. There were no statistically significant differences between the groups with respect to gender, right heart
dilatation, initial weight percentile, or gestational age at
birth. Patients in group A were significantly younger at
the time of diagnosis than were patients in group B (3.0
± 0.6 months vs. 5.1 ± 0.5 months, respectively; p =
0.014).
Beginning at the 6-month follow-up visit, changes in
ASD diameter sizes in the 2 groups were significantly different (-2.83 ± 0.54 mm vs. 0.54 ± 0.45 mm, p = 0.001;
Figure 1). Overall, 13 patients (22%) showed reduced
ASD sizes of < 5 mm at the median age of 13.0 ± 1.6
months (range, 6-27 months). Ten patients (17%) had
spontaneously closed ASD at the median age of 26.0 ±
5.1 months (range, 15-58 months). Thirty patients (51%)
underwent transcatheter closure of ASD at a median
age of 40 months (range, 25-114 months). The median
diameter of ASDs treated with transcatheter closure was
13.2 ± 0.9 mm (range, 8.2-26 mm), as measured by
transthoracic echocardiography (TTE) before intervention.
Growth was evaluated by body weight gain. The
weight percentiles were not significantly different between group A and group B at the time of diagnosis (37
± 6 percentile vs. 31 ± 4 percentile; p = 0.448). During
follow-up at the 6 and 12 month visits, the weight percentile was significantly higher in group A than in group
B (at the 6th month: 54 ± 7 percentile vs. 30 ± 4 percen-
DISCUSSION
The natural courses of ASDs were predicted by ASD
Figure 1. Changes in diameter of atrial septal defects at 3rd, 6th, 12th
and 24th month follow-up. ** Significant difference was defined as p <
0.05. Group A, reductions in ASD size of < 5 mm or spontaneous closure;
Group B, ASDs remained at ³ 5 mm in size.
Table 1. Clinical characteristics of patients with large atrial septal defect
Gender (male/female)
†
Prematurity
Age at diagnosis (month)
th
Initial BW percentile ( )
ASD size (mm)
Presence of RVH
Medication (digoxin and/or furosemide)
Group A (N = 23)
Group B (N = 36)
p value
6/17
4
3.0 ± 0.6
37 ± 60
9.0 ± 0.3
16
6
12/24
2
05.1 ± 0.5
31 ± 4
10.4 ± 0.4
30
20
0.773
0.196
*0.014*
0.448
0.081
0.213
*0.026*
†
* Significant difference was defined as p < 0.05; Prematurity: defined as less than 37 weeks’ gestational age.
Group A, reductions in ASD size of < 5 mm or spontaneous closure; Group B, ASDs remained at ³ 5 mm in size.
ASD, atrial septal defect; BW percentile, body weight percentile; RVH, right ventricular hypertrophy.
Acta Cardiol Sin 2013;29:82-87
84
Regression of Large ASD Diagnosed in Infancy
Figure 2. Differences of weight percentile at diagnosis, 3rd, 6th, 12th
and 24th month. * Significant difference was defined as p < 0.05. Group
A, reductions in ASD size of < 5 mm or spontaneous closure; Group B,
ASDs remained at ³ 5 mm in size.
Figure 3. Proportion of ASD regressing to < 5 mm according to the
initial diameter. * Significant difference was defined as p < 0.05. ASD,
atrial septal defect.
cording to age at the time of diagnosis. Radzik et al.1 analyzed ASD patients who were diagnosed before 3
months of age. The overall rate of spontaneous closure
was 87%, but no patient had spontaneous closure in the
subgroup of large ASD (diameter ³ 8 mm). Hanslik et al.9
reviewed the natural course of ASDs in children, and the
patients were divided into groups according to ASD size
(4-5 mm, 6-7 mm, 8-10 mm or > 10 mm). Spontaneous
closure occurred in 12% of patients with ASDs of 8-10
mm in diameter, and in no patients with ASDs of >10
mm in diameter. Twenty-four percent of patients with
8-10 mm ASDs and 9% of those with ASDs > 10 mm experienced regression of ASDs to £ 3 mm. Furthermore,
children less than 1 year of age at the time of ASD diagnosis experienced higher rates of ASD closure. In this
study, we focused on patients with large defects (over 8
mm), who were diagnosed within the first year of life.
The overall rate of spontaneous closure was 17%, and
the rate of size regression to £ 5 mm was 22%. Both of
these ratios were higher than those reported in previous studies. For those with ASDs of > 10 mm, there
were higher rates of spontaneous closure (9%) and size
reduction (14%) among the 22 patients. However, there
was less likelihood of regression in size or spontaneous
closure among patients with larger ASDs with a size of
> 10 mm than those with 8-10 mm ASDs. An ASD diameter of > 10 mm was a negative predictor for spontaneous closure of ASD.
sizes and patient age in earlier studies.1,2,9,10 Small ASDs
were defined as those with a diameter of 3-5 mm because of stable hemodynamics with small left-to-right
shunting. Small ASDs had a high spontaneous rate of
closure (87%) in infancy.2 ASDs with diameters between
5 and 8 mm were defined as a moderate size. However,
medical follow-up is suggested because these ASDs may
decrease or increase in size. ASDs with diameters of > 8
mm were called large ASD. In previous studies, these
ASDs rarely show spontaneous decrease in diameter or
closure because of their large left-to-right shunting.1,2
Previous studies have shown that spontaneous closure of large ASDs occurs in 0-8% of cases among different age groups.1-3,9-12 Different sample sizes and variable
ages of diagnosis may account for the discrepancy of
this study and previous studies.1,12,13 It is of clinical importance to establish some guidelines about the incidence of spontaneous closure or size regression in ASDs
among infants aged less than 2 years. However, this
study cannot rule out the possibility that, before diagnosis, some patients with large ASDs had reduced their
sizes or even spontaneously closed prior to the age of 1
year.
Age at diagnosis and ASD diameter were commonly
associated with the incidence of spontaneous closure in
the natural course of ASD.1,2,9-12 For ASDs with the same
size, the incidence of spontaneous closures varied ac85
Acta Cardiol Sin 2013;29:82-87
Kuan-Miao Lin et al.
lution of the left-side anomaly could promote spontaneous closure or reduction of ASD, even in large defects.
Finally, measurement of the size of secundum ASDs
was a limitation of our study because measurement of
the maximal diameter may have been hampered by the
oval shape of defects and cardiac cycle. For the median
size of ASDs in the neonates or young infants, an ASD
diameter between 5-8 mm could be relatively large
when comparing this to intact interatrial septal length.
Besides the evaluation of an absolute diameter of ASDs,
we could record the area of ASDs, the ratio of ASD diameter to body surface area or the ratio of ASD diameter
to interatrial septal length to predict the defect size for
further studies. The factors influencing closure of ASDs
may be related to defect shape, ratio of pulmonary to
systemic blood flow, or right heart volume overload.
Given the limitations of transthoracic echocardiography
and the invasiveness of transesophageal echocardiography, transthoracic 3-dimensional echocardiography
may provide accurate measurements of the maximal
diameters and rims of ASDs.22,23
The mechanism responsible for spontaneous ASD
closure is unclear. Bostan et al.12 reported that the type
of interatrial communication was a predictive factor, and
that the incidence of spontaneous closure of valve-like
openings was higher. Another common theory involves
formation of an atrial septal aneurysm (ASA). Brand et
al. 13 suggested that an ASA may play a role in spontaneous closure of the associated ASD. Recently, Demir
et al.14 examined 9 patients with ASAs and ASDs measuring > 7 mm. All showed decreases in size over time. Defects spontaneously closed in 3 patients. In our group A
patients, valve-like openings were detected during serial
follow-up in 8 patients. These may be related to anatomic changes and pressure gradients between the 2
atria.11,15 Further evaluation was limited because our patients were followed-up by transthoracic echocardiography only, without direct measurement of atrial pressures.
Failure to thrive was common in patients with congenital heart disease.16-18 Growth was generally better
among patients in group A. Patient body weight percentiles were significantly higher in group A than in
group B, and the difference was noted at the time of
ASD diameter with rapid reduction at 6-12 months follow-up. There was a good correlation with the rate of
size reduction and weight gain in the first year of follow-up. It was further shown that better weight gain
was a good predictor of ASD size reduction in this study.
However, these data should be verified in larger
patient studies and over longer follow-up periods. In
addition, the change in ASD diameter was significantly
different between the 2 groups in our study since the
6-12 month follow-up. Therefore, close medical followup of infants with ASD is recommended from the time of
diagnosis.19
We excluded large ASDs associated with left-side
cardiac anomaly in this study. In our experience, of 5
patients with coarctation of the aorta (CoA) or patent
ductus arteriosus (PDA), ASDs spontaneously closed in 4
patients after resolution of these left-side problems.
Yeager et al. 20 reported 9 cases with ASD and coarctation, including 4 defects measuring > 8 mm; these
ASDs either closed or decreased in diameter after resolution of coarctation. According to Riggs et al.,21 large
PDA may cause elevated left atrial pressure, resulting in
dilation of the edges of the ASDs. We suggest that resoActa Cardiol Sin 2013;29:82-87
CONCLUSION
In our study, we focused on the early diagnosis of
large ASDs with diameters > 8 mm in infancy. Compared
with previous studies, the present study showed a elevated possibility of spontaneous reduction in size or closure for large ASDs diagnosed in infancy, especially
those measuring between 8 and 10 mm in diameter. Additionally, it was observed that younger age at diagnosis
and better weight gain were positive predictors. However, an ASD diameter of > 10 mm was still a negative
predictor.
REFERENCES
1. Radzik D, Davignon A, Van Doesburg N, et al. Predictive factors
for spontaneous closure of atrial septal defects diagnosed in the
first 3 months of life. J Am Coll Cardiol 1993;22:851-3.
2. Helgason H, Jonsdottir G. Spontaneous closure of atrial septal
defects. Pediatr Cardiol 1999; 20:195-9.
3. Hugh DA, David JD, Robert ES, et al. Moss and Adams’ heart disease in infants, children, and adolescents: including the fetus
and young adult. 7th ed. Philadelphia: Lippincott Williams &
86
Regression of Large ASD Diagnosed in Infancy
Wilkins, 2008:632-45.
4. WHO multicentre growth reference study group. WHO child
growth standards based on length/height, weight and age. Acta
Paediatr Suppl 2006;450:76-85.
5. O'Connor M, McDaniel N, Brady WJ. The pediatric electrocardiogram. Part I: age-related interpretation. Am J Emerg Med
2008;26:506-12.
6. Bommer W, Weinert L, Neumann A, et al. Determination of right
atrial and right ventricular size by two-dimensional echocardiography. Circulation 1979;60:91-100.
7. Christian S, Qi LC, Ziyad MH. Transcatheter closure of congenital
and acquired septal defects. Eur Heart J 2010;12:E24-34.
8. Huang TC, Lee CL, Lin CC, et al. Transcatheter closure of atrial
septal defects with the amplatzer septal occluder-clinical results.
Acta Cardiol Sin 2004;20:223-8.
9. Hanslik A, Pospisil U, Salzer-Muhar U, et al. Predictors of spontaneous closure of isolated secundum atrial septal defect in children: a longitudinal study. Pediatrics 2006;118:1560-5.
10. Senocak F, Karademir S, Cabuk F, et al. Spontaneous closure of
interatrial septal openings in infants: an echocardiographic study.
Int J Cardiol 1996;53:221-6.
11. Cockerham JT, Martin TC, Gutierrez FR, et al. Spontaneous closure of secundum atrial septal defect in infants and young children. Am J Cardiol 1983;52:1267-71.
12. Bostan OM, Cil E, Ercan I. The prospective follow-up of the
natural course of interatrial communications diagnosed in 847
newborns. Eur Heart J 2007;28:2001-5.
13. Brand A, Keren A, Branski D, et al. Natural course of atrial septal
aneurysm in children and the potential for spontaneous closure
Acta Cardiol Sin 2013;29:82-87
of associated septal defect. Am J Cardiol 1989;64:996-1001.
14. Demir T, Oztunc F, Eroglu AG, et al. Outcome for patients with
isolated atrial septal defects in the oval fossa diagnosed in infancy. Cardiol Young 2008;18:75-8.
15. Mody MR. Serial hemodynamic observations in secundum atrial
septal defect with special reference to spontaneous closure. Am
J Cardiol 1973;32:978-81.
16. Poskitt EM. Failure to thrive in congenital heart disease. Arch Dis
Child 1993;68:158-60.
17. da Silva VM, de Oliveira Lopes MV, de Araujo TL. Growth and
nutritional status of children with congenital heart disease. J
Cardiovasc Nurs 2007;22:390-6.
18. Chang HK, Wang JN, Hung WP, et al. Double-outlet left ventricle
with Ebstein anomaly in a neonate with VACTERL association.
Acta Cardiol Sin 2011;27:65-7.
19. Kharouf R, Luxenberg DM, Khalid O, et al. Atrial septal defect:
spectrum of care. Pediatr Cardiol 2008;29:271-80.
20. Yeager SB, Keane JF. Fate of moderate and large secundum type
atrial septal defect associated with isolated coarctation in infants. Am J Cardiol 1999;84:362-3,A9.
21. Riggs T, Sharp SE, Batton D, et al. Spontaneous closure of atrial
septal defects in premature vs. full-term neonates. Pediatr
Cardiol 2000;21:129-34.
22. Acar P, Roux D, Dulac Y, et al. Transthoracic three-dimensional
echocardiography prior to closure of atrial septal defects in children. Cardiol Young 2003;13:58-63.
23. Su CH, Weng KP, Chang JK, et al. Assessment of atrial septal defect - role of real-time 3D color Doppler echocardiography for
interventional catheterization. Acta Cardiol Sin 2005;21:146-52.
87