Transcatheter closure of ostium secundum atrial septal defects (osASD) has been the first-line treatment for most children and adults since the early 2000s. The use of multiple occluder devices during a single procedure is uncommon, and evidence on outcomes in this population remains scarce. This study aimed to evaluate the safety and feasibility of transcatheter osASD closure using multiple devices compared with single-device closure.

This prospective, single-center cohort study included consecutive children and adults referred for transcatheter osASD closure with an Amplatzer Septal Occluder (ASO) between May 1998 and December 2021 at Marie Lannelongue Hospital.1 The study protocol has previously been described. The study was approved by an independent institutional review board (IRB SI20.02.26.53909), and informed consent for procedures and data use was obtained from all patients.

Defect sizing was conducted by the fluoroscopic balloon pull-through technique using a 27-mm or 33-mm Equalizer balloon (Boston Scientific, Marlborough, MA, USA) or static balloon sizing with an AMPLATZER sizing balloon. For additional osASD closure, the same balloon sizing technique was assessed, prioritizing static balloon sizing.

All patients received antiplatelet monotherapy for 6 months, except those receiving long-term antiplatelet or anticoagulation therapy. Periprocedural data and major adverse events (MAEs) were retrospectively reported.1,2

Outcomes during the first months were compared between patients treated with a single device vs those treated with multiple devices. Long-term outcomes of patients with multiple-devices were also assessed. All patients underwent standardized follow-up including clinical examination, electrocardiogram, and transthoracic echocardiography at 1 week, 1 month, 6 months, 1 year, 3 years, 5 years, and at least every 5 years after the procedure. Additional cardiac magnetic resonance imaging was performed in cases of suspected significant residual shunt.

The characteristics of the study population, with a comparison between multiple devices vs a single device implanted for periprocedural data and MAEs during the first month, are presented in Table 1. Among 2253 patients, 216 (9.6%) had multiple osASDs: 132 (5.9%) double defects and 81 (3.6%) multifenestrated septa. Of the 216 patients, only 20 (9.3%) required implantation of more than 1 device; for the remaining patients, a single device was sufficient for occluding residual osASDs. In the multiple-device group, the first implant was performed in 2008. Thirteen procedures were guided by transthoracic echocardiography alone and 7 by transesophageal echocardiography. Four patients were children weighing ≥ 20kg; 3 patients had previously undergone percutaneous osASD closure and were found to have significant residual shunting during follow-up, which was closed by additional devices. All procedures used at least 2 ASOs, and 2 patients received an additional cribriform ASO. Three adults required 3 devices, the first implantation being performed in 2020. Sequential implantation was performed during the same procedure in 17 patients. Among the multiple-device group, 9 patients (45.0%) had multifenestrated septa.

The only echocardiographic feature significantly associated with multiple-device use was interatrial septal aneurysm (odds ratio [OR], 3.54; 95% confidence interval [95%%CI], 1.38-9.06; P=.008). There were no significant differences in procedural results between groups. Procedural success in the multiple-device group was 100%. No MAEs occurred in these 20 patients. Minor complications were infrequent and similar between groups (OR, 0.68; 95%CI, 0.09-5.12; P=.707). Early residual shunts, atrial arrhythmias, and small pericardial effusions occurred at comparable rates. Follow-up was available for all multiple-device patients, with a mean duration of 6.2 [range 1.2-16.4] years. During follow-up, no deaths, reinterventions, device embolizations, late erosions, thromboembolic events, or clinically significant residual shunts were observed. All 3 patients who received 3 devices remained free of adverse events (figure 1).

Figure 1.

Transcatheter closure of ostium secundum atrial septal defects: multiple vs single-device. ASO, Amplatzer Septal Occluder; MAE, major adverse events; OR, odds ratio; osASD, ostium secondum atrial septal defect.

(0.57MB).

While transcatheter closure is widely accepted as the standard of care for most osASDs with excellent safety results,1 the management of multiple defects remains debated, and some clinicians still favor surgical closure. Our study represents one of the largest cohorts to directly compare outcomes between single- and multiple-device transcatheter osASD closure. Multiple osASDs were identified in almost 10% of patients, consistent with previous literature.3,4 However, the need for more than 1 device was rare (<1%).

Our strategy in patients with multiple ASDs was primarily to close the dominant defect whenever possible and to implant additional devices only in the presence of a persistent hemodynamically significant residual shunt after closure of the main ASD. This conservative approach likely explains the very low proportion of patients ultimately requiring multiple devices in our cohort. However, when multiple devices were required, outcomes in this small subgroup were encouraging, with a 100% acute success rate, no periprocedural MAEs, and reassuring long-term safety.

As previously described in the literature,5 interatrial septal aneurysm was the echocardiographic feature significantly associated with the multiple-device group, emphasizing the need to track cases of multiple osASD with this finding. Concerns raised in earlier reports, including a higher risk of late erosion,3 significant residual shunts,6 arrhythmias, or thromboembolic events, were not confirmed in this cohort. Importantly, long-term follow-up demonstrated reassuring outcomes, with no MAEs, even in patients with 3 devices.

These findings provide reassurance that transcatheter closure with multiple devices can be safely considered in selected patients, particularly when anatomical complexity precludes closure with a single device. While the risk of residual shunts has been highlighted in previous series, our experience did not confirm a significant difference compared with single-device closure, likely reflecting careful case selection and procedural expertise. Despite receiving the same 6-month antiplatelet monotherapy treatment, no thromboembolic or significant atrial arrhythmia events were reported in the multiple-device group, thus confirming the safety and efficacy of this therapeutic strategy.

Nevertheless, the study has some limitations. Despite prospective data collection, the analysis of periprocedural outcomes was retrospective. The number of patients treated with multiple devices remains relatively small, limiting statistical power to detect rare major adverse events such as erosion or device embolization. In addition, as this was a single-center experience, outcomes may not be generalizable to all settings.

In this large cohort, multiple-device closure of osASDs was rare but feasible, safe, and associated with reassuring short- and long-term outcomes. When carefully selected, patients with complex osASDs can benefit from a percutaneous approach using more than 1 device, avoiding the need for surgery.