Transcatheter Closure of Complex Atrial Septal Defects in Children: A Single-Center Experience

Abstract

Objective

Although transcatheter atrial septal defect (ASD) closure has become the preferred treatment for ostium secundum ASD, closure of complex defects—characterized by large size, rim deficiencies, multiple fenestrations, aneurysmal or malaligned septum, or low body weight—remains technically challenging. This study evaluated the efficacy of transcatheter closure in children with complex ASDs.

Materials and Methods

This retrospective single-center study included 15 patients aged 0–18 years who underwent transcatheter ASD closure between 2023 and 2025. All procedures were performed using the Occlutech Figulla Flex II ASD occluder and were guided by transthoracic and transesophageal echocardiography. Procedural data, complications, and follow-up outcomes were recorded.

Results

The mean age was 7.0±4.4 years, and 60% of the patients weighed ≤15 kg. Complete occlusion was achieved in all children. Two patients (13.3%) experienced transient complications—one device embolization and one case of atrial premature beats—which resolved without permanent sequelae. No mortality or residual shunts were observed at 6-month follow-up.

Conclusion

Transcatheter closure of complex ASDs is safe when performed by experienced operators who conduct careful anatomical assessments and select appropriate devices. Alternative implantation techniques may enhance procedural success in challenging anatomical conditions. Transcatheter closure is a reliable, minimally invasive alternative for complex ASD cases.

Introduction

Atrial septal defect (ASD) is the most common congenital heart condition in children, accounting for approximately 10–15% of all cases. Ostium secundum ASD is the most suitable for transcatheter closure. Compared with surgical repair, transcatheter closure is preferred because it offers a shorter hospital stay, fewer complications, and cosmetic benefits.^1,2

Some ASDs are more complex because of their unique anatomical features, which can make procedures more challenging. Complex ASD is often characterized by factors such as a large defect size (greater than 20 mm), a defect diameter-to-body surface area ratio exceeding 20 mm/m², a device-to-body weight ratio over 1.5, multiple openings, an aneurysmal or malaligned septum, significant rim deficiencies, especially near the aorta, or other associated cardiac anomalies. These factors may occur together, and the presence of any one of them defines a complex ASD. Such characteristics can make it more challenging to secure the device properly and increase the risk of complications, such as embolization, residual shunt, or disturbances in cardiac conduction.^3,4

Some reports of serious complications, including device instability, embolization risk, deployment difficulties in rim-deficient cases, heart perforation, atrioventricular block, or device-induced aortic insufficiency, have raised concerns about the implantation of large devices in complex and large ASDs.^5–10

Recent studies have shown that, with careful anatomical assessment, proper device selection, and skilled operators, transcatheter closure of complex ASDs can be performed effectively and safely. Additionally, alternative methods such as pulmonary vein deployment, the left atrial roof maneuver, and the FAST (Fast Atrial Sheath Traction) technique have been documented to improve procedural success, especially in cases with rim deficiencies.^11–13

This study assessed the outcomes of transcatheter closure in children with complex ASDs at our center.

MATERIALS AND METHODS

Study Place and Design

This retrospective study included patients (n=15) aged 0–18 years who underwent transcatheter closure of complex ASDs at a city hospital in Türkiye between 2023 and 2025 and were identified as having complex or large ASDs.

Ethics Approval

This study was approved by the Izmir City Hospital Non-Interventional Ethics Committee (Approval Number: 2025/510, Date: 26.09.2025)

Patient Selection

The diagnostic criteria included patients who underwent transcatheter closure of complex ASDs with a device-to-body weight ratio above 1.5 or a body weight under 15 kg; a defect diameter measured by balloon sizing that was 50% or greater than the total interatrial septum (IAS); patients under one year of age regardless of defect size; a stretched balloon diameter-to-body surface area ratio (mm/m²) of 15 or greater; an aneurysmal IAS; rim deficiency other than isolated aortic rim deficiency; multiple or multifenestrated defects; isolated inferior vena cava, superior vena cava, posterior, or combined aortic and inferior vena cava/superior vena cava rim deficiency; malalignment between the rims in the IAS; a prominent Eustachian valve; or severe pulmonary stenosis.

Patients with other genetic anomalies, irreversible pulmonary hypertension, and ostium primum, venous/coronary sinus, or ideal ASDs (centered, normal anatomy, small-to-moderate defects in larger children, and no other congenital heart disease) were excluded.

Data from 15 patients who underwent closure of complex ASDs performed by the authors between 2023 and 2025 were evaluated. During the data collection period, four patients with completely inadequate inferior posterior rims were referred for surgery.

Clinical Investigations

Before the intervention, demographic data (sex, age, body weight/height), current symptoms, physical examination findings, electrocardiogram (ECG) recordings, and laboratory results were evaluated. Clinical data, including defect size, rim adequacy, and right heart chamber dimensions, were assessed using transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE).

The procedure was performed under general anesthesia. During cardiac catheterization, pulmonary artery pressure and the pulmonary-to-systemic flow (Qp/Qs) ratio were measured. The defect diameter was determined using the balloon stop-flow technique, and the device size was selected to be 1–2 mm larger than the balloon-measured diameter. The definition of complex ASD was based on defect size, rim adequacy, the presence of an aneurysmal or malaligned septum, and multiplicity of defects.

All patients underwent closure using the Occlutech Figulla Flex II ASD Occluder (Fig. 1). Figure 1 shows an image of a patient with an insufficient aortic rim who received an 18-mm device.

Figure 1.

Occlutech Figulla Flex II atrial septal defect (ASD) occluder image after release of the 18-mm device.

Pulmonary vein deployment or left atrial roof techniques were used as needed to facilitate device positioning. Unfractionated heparin (75 U/kg) was administered during the procedure to maintain an activated clotting time between 250 and 300 s. Cefazolin was administered for infective endocarditis prophylaxis. Before releasing the device, the Minnesota maneuver was performed to assess stability and rule out interference with adjacent cardiac structures (Fig. 2).

Figure 2.

Minnesota maneuver performed before release of the device.

After the procedure, all patients received acetylsalicylic acid (3–5 mg/kg/day) for six months and were advised to maintain infective endocarditis prophylaxis for one year. Patients underwent follow-up with TTE and ECG on post-procedural day 1, week 1, and months 1, 3, and 6. Both procedural complications and follow-up findings were recorded.

All procedures were performed according to relevant national ethical standards and the 2008 revised version of the 1975 Declaration of Helsinki. Written informed consent was obtained from the parents regarding the use of the procedures and clinical records.

Statistical Analysis

Statistical analyses were performed using SPSS version 27.0 (IBM Corp, Armonk, NY, USA). Continuous variables were presented as mean±standard deviation (minimum–maximum) values. Categorical variables were presented as numbers and percentages. Correlations between measured values were evaluated using Pearson’s correlation analysis. An independent-samples t-test was used to compare variables across device sizes. A significance level of p<0.05 was accepted.

RESULTS

Demographics

A total of 15 patients with complex ASDs were included in this retrospective, single-center, cohort study (Table 1). Of these, 66.7% (n=10) were female. The most common presenting symptom was failure to thrive, observed in ten patients. Reduced exercise tolerance was noted in three patients, and two were asymptomatic. The mean age was 7.0±4.4 years (range, 2–16). The mean body weight was 26.8±20.2 kg (range, 11–80). The mean height was 125.9±18.8 cm (range, 105–172).

Table 1.

Patient characteristics

Patient no.	Age	Weight	Defect size by TTE	Defect size by TEE	Defect size by balloon sizing	Device diameter	Complex ASD characteristics
1	4	11	19	20	20	21	Insufficient aortic rim; large ASD
2	8	40	22	26	28	30	Insufficient aortic rim; large ASD
3	3	14.5	16	16	17	18	Insufficient aortic rim
4	12	41	18	19	19	21	Aneurysmal septum
5	16	80	22	24	25	27	Insufficient aortic rim; large ASD
6	5	14	14	15	15.5	16	Two defects
7	11	38.5	17	21	22.5	24	Two defects
8	5	15	17	17.5	18	19	Large ASD
9	13	55	17	19.5	20	21	Prominent Eustachian valve
10	3.5	14	16	14.7	14.5	15	Malaligned septum
11	5	15	17	18.5	18.7	19	Large ASD
12	11	25	12	11	11.8	13.5	Pulmonary stenosis; complete AV block with paced rhythm
13	3	12	15	15	15	16	Insufficient inferior rim; large ASD
14	2	12	15	15	15	16	Insufficient aortic rim; large ASD
15	4	15	16	17	17	18	Insufficient aortic rim; large ASD

ASD: Atrial septal defect; AV: Atrioventricular; TTE: Transthoracic echocardiography; TEE: Transesophageal echocardiography; M: Mean; SD: Standard deviation.

ASD Characteristics

On echocardiography, the mean defect diameter was 16.8±2.6 mm (range, 12–22); on TEE, it was 17.9±3.8 mm (range, 11–26); and on balloon sizing, it was 18.4±4.2 mm (range, 11.8–28). The mean pulmonary artery pressure was 15.6±1.5 mmHg (range, 12–18). The Qp/Qs ratio was 1.8±0.1 (range, 1.6–2.0).

The mean balloon sizing defect diameter-to-IAS ratio was 0.46±0.09, the balloon sizing defect diameter-to-body surface area ratio was 23.2±7.7, and the balloon sizing defect diameter-to-weight ratio was 0.9±0.4. The mean right ventricular end-diastolic diameter was 24.8±3.1 mm (range: 21–32 mm). Nine patients (60%) weighed ≤15 kg.

Fourteen patients showed incomplete right bundle branch block, and one exhibited complete atrioventricular (AV) block with a paced rhythm and severe pulmonary stenosis (peak gradient: 80 mmHg). The patient with congenital complete AV block and a paced rhythm had a previously implanted epicardial pacemaker. The device had been in place for 11 years; the pacing mode was ventricular pacing and sensing (VVI), and transvenous pacemaker implantation was performed during the same session. The patient underwent transvenous lead placement, followed by balloon valvuloplasty for pulmonary stenosis, and finally transcatheter ASD closure.

Anatomically, one patient had a malalignment defect, two had multiple defects, one had an aneurysmal septum, and one had a prominent Eustachian valve. Six patients (40%) had short aortic rims (Fig. 3). In another patient, the inferior rim was short and floppy. In two patients with multiple defects, the distance between defects was less than 7 mm; therefore, a single device was used.

Figure 3.

Insufficiency of the aortic rim in a patient with a 22-mm large atrial septal defect (ASD) on transthoracic echocardiography (TTE).

In three of the 15 patients, device replacement with a different size was performed because the initially implanted device was unsuitable. The right upper pulmonary vein (RUPV) approach was used in 10 patients, and the left atrial roof technique was used in five patients.

Device Size Characteristics

The mean device size was 19.6±4.5 mm (range, 13.5–30), and the average fluoroscopy time was 7.7±2.2 min (range, 5–15). A highly significant positive correlation was found between device size and defect size measured by TTE, TEE, and balloon sizing (r=0.918–0.951; p=0.000) (Table 2). The mean device size was considered, and age, weight, and defect size measured by TTE, TEE, and balloon sizing were compared between patients with device sizes <20 mm and >20 mm. Differences were found in age, weight, and defect size measured by TTE, TEE, and balloon sizing (p<0.05). No difference was found in fluoroscopy duration (p>0.05) (Table 3).

Table 2.

Correlation between device size and defect size measured by transthoracic echocardiography, transesophageal echocardiography, and balloon sizing

	Defect size by TTE	Defect size by TEE	Defect size by balloon sizing	Device size
Defect size by TTE
Pearson correlation	1	0.951*	0.925*	0.918*
Sig. (2-tailed)		0.000	0.000	0.000
N	15	15	15	15
Defect size by TEE
Pearson correlation	0.951*	1	0.992*	0.983*
Sig. (2-tailed)	0.000		0.000	0.000
N	15	15	15	15
Defect size by balloon sizing
Pearson correlation	0.925*	0.992*	1	0.995*
Sig. (2-tailed)	0.000	0.000		0.000
N	15	15	15	15
Device size
Pearson correlation	0.918*	0.983*	0.995*	1
Sig. (2-tailed)	0.000	0.000	0.000
N	15	15	15	15

*: Correlation is significant at the 0.01 level (2-tailed). TTE: Transthoracic echocardiography; TEE: Transesophageal echocardiography.

Table 3.

Comparison of patients’ age, weight, defect size, and fluoroscopy duration with device size below and above 20 mm

	Device size		Test
	>20 mm Mean±SD	<20 mm Mean±SD
Age	10.6±4.1	4.6±2.6	t=3.474, p=0.004
Weight	44.2±22.6	15.1±3.8	t=3.843, p=0.002
Defect size by TTE	19.1±2.3	15.3±1.6	t=3.799, p=0.002
Defect size by TEE	21.5±2.8	15.5±2.1	t=4.746, p=0.000
Defect size by balloon sizing	22.4±3.4	15.8±2.0	t=4.587, p=0.001
Fluoroscopy duration	7.1±0.4	8.1±2.9	t=0.082, p=0.453

SD: Standard deviation; Test: Independent-samples t-test; TTE: Transthoracic echocardiography, TEE: Transesophageal echocardiography.

Complication Characteristics and Management

Complete occlusion was achieved in all children, resulting in a success rate of 100%. Complications occurred in two patients (13.3%). One patient experienced device embolization (the patient with AV block and pulmonary stenosis). The device was successfully retrieved from the ascending aorta using a snare, and the defect was subsequently closed with a larger device (13.5 mm) without further complications (Fig. 4). Transient atrial premature beats developed immediately after device release in another patient and resolved after a single dose of dexamethasone.

Figure 4.

Removal of the embolized device from the aorta using a snare.

In a five-year-old patient with multiple defects, a minimal residual shunt was observed in a smaller nearby defect on the first day after ASD closure, and complete closure was confirmed at the three-month follow-up. No residual shunts were detected in the remaining patients.

No patient experienced permanent complications or death. The patient with device embolization, who also underwent pacemaker implantation and balloon valvuloplasty, was monitored for five days and discharged without further issues. None of the patients required intensive care unit admission. All patients were asymptomatic, and echocardiographic evaluations showed complete closure in all cases at the 6-month follow-up visit.

DISCUSSION

This study demonstrated that transcatheter closure of ASDs with complex anatomical features is safe and effective. Complete occlusion was achieved in all patients, with only two instances of transient complications and no permanent adverse events noted. These results are consistent with those previously reported in the literature.^6–10 Closure of complex ASDs can be challenging because of factors such as large defect size, rim deficiencies, or an aneurysmal septum. These anatomical features hinder stable device placement and precise deployment.¹⁴ Mahmoud et al.¹⁵ noted that when multiple fenestrations or rim deficiencies are present, thorough echocardiographic assessment and proper device selection can help reduce complications. In this study, the mean device size was 19.6±4.5 mm (range, 13.5–30). Differences were found in age, weight, and defect diameter measured by TTE, TEE, and balloon sizing between patients with <20 mm and >20 mm device sizes. A highly significant positive correlation was found between device size and defect diameter measured using TTE, TEE, and balloon sizing. The combined use of TEE and balloon sizing was more effective in difficult ASD cases because of the low effectiveness of TEE in sizing.⁴

In this study, only three patients required device replacement with a different size. The high correlation between measurements demonstrates that these assessments were performed accurately and highlights the importance of precise measurements in predicting device size, even in complex ASDs. The RUPV approach was used in most cases in this study. Traditional techniques for the closure of complex ASDs include the left atrial ceiling, left upper pulmonary vein, RUPV, and balloon-assisted techniques.⁹ Naseem et al.¹⁰ reported that traditional techniques were used more frequently than the left upper pulmonary vein technique, whereas Pillai et al.⁴ frequently recommended the balloon-assisted technique for complex ASDs. Although the choice of technique depends entirely on operator preference and experience, the RUPV technique was used in this study based on the operator’s preference.

Recently, novel implantation strategies have addressed these anatomical challenges. The FAST technique, as described by Haddad et al.,¹¹ has demonstrated shorter procedure times and improved device stability, particularly in cases of aortic rim deficiency. Additionally, pulmonary vein deployment (Fig. 5) and the left atrial roof technique helped achieve optimal device positioning toward the septum, thereby enhancing procedural outcomes. At our center, we employed these alternative strategies in anatomically challenging cases and achieved favorable results. Although earlier large studies have shown higher complication rates among patients weighing under 15 kg, these risks can be reduced through careful patient selection, appropriate balloon sizing, and skilled operators. Notably, no major complications were observed in our cohort, in which 50% of the patients weighed ≤15 kg.

Figure 5.

Opening phase of the left disc of the device in the right upper pulmonary vein.

The success rate was >90% in studies by Houeijeh et al.,⁹ Baruteau et al.,⁵ and Aktas et al.¹⁴ involving large and complex ASDs. Our results demonstrate that transcatheter ASD closure is an effective and safe alternative to surgery for patients with complex ASDs, with a 100% success rate, a low risk of complications, and a short hospital stay. Houeijeh et al.⁹ reported inferior vena cava (IVC) thrombosis and complete AV cases but did not observe device embolization. In our study, no major complications were observed except for two cases (device embolization and transient atrial premature beats). Larger multicenter prospective studies are needed to compare different devices and techniques.

Limitations

This study had several limitations. First, the relatively small sample size restricted the generalizability of the findings and prevented subgroup analyses with sufficient statistical power. Second, the single-center, retrospective nature of the study may have introduced selection bias and limited observational accuracy. Third, the follow-up periods were relatively short for some participants, hindering a full assessment of late complications, such as erosion or delayed embolization. Finally, only the Occlutech Figulla ASD occluder was utilized; therefore, no direct comparison with alternative devices or implantation techniques was performed. Despite these limitations, this study offers valuable insights into the feasibility of transcatheter closure in patients with complex ASDs.

CONCLUSION

With careful anatomical assessment, appropriate device selection, and experienced operators, transcatheter closure of complex ASDs can be performed effectively and safely, with a 100% success rate and low complication rates. Even in cases involving large defects, rim deficiencies, or low body weight, alternative implantation techniques can improve success rates.

The RUPV approach was used in ten of the 15 patients, whereas the left atrial roof technique was used in the remaining patients. The mean device size was 19.6±4.5 mm. A highly significant positive correlation was found between device size and defect size measured by TTE, TEE, and balloon sizing. Differences were found in age, weight, and defect diameter measured by TTE, TEE, and balloon sizing between patients with <20 and >20 mm device sizes.

The success rate was 100%. Complications occurred in two patients (13.3%). One patient experienced device embolization (the patient with AV block and pulmonary stenosis), and transient atrial premature beats developed in another. Our results indicate that transcatheter closure is a robust, minimally invasive alternative to surgery. However, large-scale, long-term studies are necessary to compare the effectiveness of different devices and implantation methods.

Ethics Committee Approval

Ethics committee approval was obtained from Izmir City Hospital Non-Interventional Ethics Committee (Approval Number: 2025/510, Date: 26.09.2025)

Informed Consent

The ethics committee waived the requirement for consent.

Conflict of Interest

The authors have no conflicts of interest to declare.

Funding

The authors declared that this study received no financial support.

Use of AI for Writing Assistance

No use of AI-assisted technologies was declared by the authors.

Author Contributions

Concept – EG, KY; Design – EG, KY; Supervision – EG, KY; Resource – EG, KY; Materials – EG, KY; Data Collection and/or Processing – EG, KY; Analysis and/or Interpretation – EG, Literature Review – EG, Writing – EG, KY; Critical Review – EG, KY.

Peer-review

Externally peer-reviewed.