Approach to Percutaneous Closure in Patients with Multiple Atrial Septal Defects

Tywaun Tillman, MD, Reynaldo Mulingtapang, MD, J. Thompson Sullebarger, MD
Tywaun Tillman, MD, Reynaldo Mulingtapang, MD, J. Thompson Sullebarger, MD


J INVASIVE CARDIOL 2008;20:E167-E170

Percutaneous closure of secundum atrial septal defect (ASD) and patent foramen ovale (PFO) has become a safe and effective alternative to surgery in appropriately selected patients. Closure of these defects is performed under fluoroscopic and echocardiographic (transesophageal or intracardiac) guidance. While numerous devices and techniques are currently in development, the most commonly used devices in the United States include the Amplatzer Septal Occluder and PFO Occluder (AGA Medical, Minneapolis, Minnesota), and the CardioSEAL® and STARflex® occluders (NMT Medical, Inc., Boston, Massachusetts). In most patients, a single defect of the interatrial septum may be closed with a single device. Less often, patients may present with multiple defects or complex, fenestrated defects. In some cases, a single device may be placed through one defect, and may overlap and close a second defect if it is close by. However, some patients require multiple closure devices, which may involve special challenges. The use of multiple closure devices in a single patient remains uncommon; fewer than 2% of reported septal occluder cases required multiple devices.1 We report the technical details of 5 cases requiring 2 closure devices with clinical and echocardiographic follow up ranging from 3 months to 4 years, demonstrating high success and low complication rates.

Case 1. An 18-year-old male presented with multiple syncopal episodes since childhood, now increasing in frequency. An echocardiogram revealed the presence of a small ASD, with enlargement of the right atrium and ventricle. He was referred for percutaneous closure. Transesophageal echocardiography (TEE) showed 2 separate atrial septal defects, and right heart catheterization revealed a 3:1 left-to-right shunt. Balloon stretch-sizing of the larger posterior/ superior defect with a 24 mm AGA balloon (AGA Medical) showed a stretch diameter of 14 mm. A 14 mm Amplatzer Septal Occluder was deployed under fluoroscopic and TEE guidance. Subsequently, the stretch diameter of a smaller, more anterior defect was 9 mm. However, deployment of a 10 mm Amplatzer occluder was unsuccessful due to the close proximity to the first device and prolapse of the device through the defect. A 16 mm Amplatzer septal occluder was then successfully deployed with the larger size, allowing for more stability. The patient was placed on 3 months of dual antiplatelet therapy with aspirin and clopidogrel. Transthoracic echocardiography at 1 month confirmed both devices in stable position, without left-to-right shunting. He has not had any complications or recurrence of syncope over a period of 3 months since his procedure. Prolonged follow up will be needed to determine if his syncopal episodes have resolved.
Case 2. A 50-year-old female was previously found to have a cardiac murmur at age 39. TEE confirmed the presence of a single, small secundum ASD. Since the patient was asymptomatic, she was monitored without further treatment for several years. However, a routine follow-up echocardiogram showed right heart enlargement and reduced left ventricular ejection fraction of 35%. A repeat TEE several weeks later confirmed enlargement of the right heart, but the ejection fraction was noted to be 60%. There was an aneurysmal atrial septum and an ASD with significant left-to-right shunting. She denied symptoms of dyspnea, angina, stroke or transient ischemic attack, and had no history of coronary artery disease. Due to the presence of a shunt and right heart enlargement, she was referred for percutaneous closure. The stretch-sizing was 10 mm. After closing the defect with a 10 mm Amplatzer occluder, a second adjacent defect was observed by TEE, with continued left-to-right shunting. Stretch-sizing of the second defect was also 10 mm. A 12 mm Amplatzer septal occluder was deployed in this defect and oversized to slightly overlap the edge of the first occluder. Proper position of both occluders was confirmed by TEE and fluoroscopy, with no residual leftto- right shunting. Follow-up echocardiography at 7 months showed stable Amplatzer occluders, no interatrial shunting and an ejection fraction of 60%. She has done well 13 months post procedure without any further symptoms or change in right-heart dimensions.
Case 3. A 39-year-old female initially presented with recurrent transient ischemic attacks and strokes despite antiplatelet and coumadin therapy. There was no evidence of extracranial or intracranial cerebrovascular disease, and on TEE, the only significant finding was a large PFO with an atrial septal aneurysm and a positive bubble contrast study, suggesting paradoxical embolism. She was referred for closure of the defect. The PFO, which had a stretch diameter of 15 mm, was closed using a 33 mm CardioSEAL device. A small ASD was also noted anterior and superior to the PFO, with a stretch diameter of 8 mm. Due to unfavorable anatomy, an attempt to close this defect with a 23 mm CardioSEAL device was unsuccessful. The Amplatzer occluder was not available at our institution at that time, so the patient returned to the laboratory 4 months later, and the defect was closed with a 10 mm Amplatzer under intracardiac echocardiography and fluoroscopic guidance. Immediately after closure, a tiny shunt was noted at the rim of the Amplatzer occluder. Follow-up transthoracic echocardiograms at 6 months and 17 months showed stable position of both devices and no evidence of the residual shunt. At that point, she was taken off of all medications including aspirin, clopidogrel and coumadin. She has not experienced any thromboembolic events during the followup period greater than 4 years.
Case 4. A 37-year-old female presented complaining of atypical chest pain as well as exertional dyspnea. Echocardiography revealed a secundum ASD with left-to-right shunting (2.1:1), as well as enlargement of both atria and the right ventricle. The patient was scheduled for percutaneous closure. TEE performed during the procedure revealed two separate defects. The first defect had a stretch diameter of 14 mm. A 14 mm Amplatzer occluder was deployed successfully, with residual shunting noted through the second defect. This defect was located anterior and inferior to the first and had a stretch diameter of 14 mm measured with the first occluder in place. A second 14 mm Amplatzer occluder was deployed. Careful evaluation by echocardiography and fluoroscopy confirmed proper, stable placement of the devices and no residual shunting. She has continued without any complications over the past 3 years since her procedure.
Case 5. A 28-year-old female who presented with symptoms of heart failure and palpitations was found to have a large ASD and right heart enlargement while undergoing an evaluation for tachycardia. Her symptoms included dyspnea upon exertion and fatigue. After the initiation of medical treatment of her tachycardia, percutaneous ASD closure was planned. Right-heart catheterization revealed a shunt ratio of 10:1. TEE during the procedure revealed two separate ASD stretch-sizing of the first defect, measured approximately at 18 mm with suboptimal echocardiographic visualization. A 20 mm Amplatzer septal occluder was deployed after additional assessment. The second defect was sized at 25 mm after deploying the first occluder; however, attempts to deploy a 26 mm Amplatzer occluder were unsuccessful, due to prolapse of the device into the left atrium. A larger 28 mm device was attempted, but there was a tendency for the first occluder to interfere with the orientation of this device. Finally, a 32 mm occluder was successfully deployed with a stable and properly oriented position. Both defects were completely occluded by the two occluders. The devices were thoroughly evaluated for stability and position under echocardiographic and fluoroscopic guidance, particularly given the large size of the second device. The patient underwent an electrophysiologic study prior to discharge due to continued episodes of tachycardia and prolongation of her PR interval postprocedurally from 216 ms to 320 ms once restarting her beta-blocker. She was found to have multiple inducible tachycardias (atrial fibrillation, atrial flutter, and three different atrial tachycardias) and was consequently treated medically with sotalol. Her first-degree atrioventricular (AV) block improved. Upon follow up at 1 month, she reported improvement in her symptoms of exertional dyspnea and fatigue, which continued to improve to minimal limitation over the 3-year follow-up period. She did not experience any complications during the follow-up period and did not require pacemaker placement secondary to AV block. Follow-up echocardiograms showed continued stable position of the ASD occluders, with no residual shunting across the interatrial septum.

Discussion. The approach to closure of multiple ASDs presents unique challenges in each case. The location of secundum ASDs often varies.2 The approach to percutaneous closure of a defect depends upon its size and its relationship to adjacent structures including the aorta, the coronary sinus, the pulmonary veins, the mitral and tricuspid valves and the free walls of the right and left atria. The extent of tissue rim around the defect is important for successful deployment, as a poor rim may not be sufficient to support the device, and residual defects may be left at the edges. Oversizing of devices may compensate for this, but could also result in compromise of adjacent structures or late erosion.7 The decision to close small residual interatrial defects or fenestrations should be based on the initial clinical indications for closure and the size and hemodynamic significance of the shunt.

In some cases, multiple small defects or fenestrations in close proximity may be closed with a single device, which may overlap several defects or fenestrations.3 Device designs that cover a large surface of the septum, such as the CardioSEAL, Starflex and Amplatzer fenestrated devices may be helpful in these cases. In the patient with multiple defects, the potential for interaction between the devices must be considered, and careful evaluation of the defects and their anatomic relationship to surrounding structures is of utmost importance — the mitral and tricuspid valves,coronary sinus ostium, aorta and vena cavae.4 The risk of impinging upon and disrupting these structures increases with very large or multiple devices. Cao et al suggest that a tissue rim > 7 mm should be present between defects to allow for two devices; however, it may be difficult to obtain these measurements given the frequent difficulty in viewing both defects simultaneously. Routinely measuring the distance between defects is not incorporated into our practice; however, an attempt is made to close multiple defects with a single device, oversized if necessary, if the defects are in close proximity and the anatomy is suitable.
Thorough evaluation by echocardiography, including TEE prior to the procedure, is helpful if there are multiple defects (Figure 1). In each of the cases presented here, only a single defect was detected by transthoracic echocardiography. A preprocedure TEE can aid in planning the procedure and device selection and avoiding unpleasant surprises. As threedimensional echocardiography becomes more available, it will likely have a role in defining anatomy preprocedure as well.5 Intracardiac echocardiography offers an alternative to TEE during the procedure and offers greater detail, particularly of the inferior atrial septum and the relationship of a closure device to the mitral valve.4,6 Despite careful planning before the procedure, closure of multiple defects prolongs procedure time, and patient cooperation is crucial to the accurate placement of the closure device; therefore, the form of sedation or anesthesia must be carefully considered.

In our experience, and as documented by others, closure of the larger defect should be performed first.7 Frequently, the devices will interfere with one another, in which case, the second device may be placed so as to slightly overlap the rim of the first device (Figures 2 and 3). In order for the devices to orient properly, the second may need to be relatively larger, despite the stretch diameter of the defect. It is unclear if staging deployment of the 2 devices in separate procedures confers any advantage. In theory, the first device could be displaced or even embolized during attempted placement of the second device, especially in patients with large defects. In Case 4, a CardioSEAL device was placed first, followed by an Amplatzer several weeks later (Figure 4). In the other cases, 2 devices were placed at the same time. All procedures were successful, and all patients were doing well at follow up.

The risk of serious complications with percutaneous closure of interatrial septal defects is low overall and is only slightly increased in multiple occluder cases. A series of 22 patients with 2 septal occluders were evaluated, and the initial success rate approached 98%. There was 1 device embolization and no significant complications otherwise. In our combined follow-up period totaling 135 months, there were no major events. One patient developed temporary worsening of an underlying first-degree AV block, which resolved spontaneously with no long-term sequelae. There were no embolic events, including the patients who underwent PFO/ASD closures for the indication of stroke/TIA secondary to paradoxical embolic phenomena, and no device embolizations, perforations or erosions.

Conclusions. Transcatheter closure of complex, interatrial septal defects can be performed safely with excellent success rates. Careful technique and imaging are crucial to success. Our experience suggests that multiple devices may be used with a high rate of success and a low complication rate. As new device designs continue to become available, the multiple device approach to these patients is likely to become more widely accepted.




1. Omeish A, Hijazi ZM. Transcatheter closure of atrial septal defects in children and adults using the amplatzer septal occluder. J Interv Cardiol 2001;14:37–44.
2. Podnar T, Martanovic P, Gavora P, Masura J. Morphological variations of secundum-type atrial septal defects: Feasibility for percutaneous closure using Amplatzer septal occluders. Catheter Cardiovasc Interv 2001;53:386–391.
3. Suarez J, Medina A, Pan M, et al. Transcatheter occlusion of complex atrial septal defects. Catheter Cardiovasc Interv 2000;51:33–41.
4. Mitchell A, Roberts P, Eichhofer J, et al. Echocardiographic assessment and percutaneous closure of multiple atrial septal defects. Cardiovasc Ultrasound 2004;2:9.
5. Cao Q, Radtke W, Berger F, et al. Transcatheter closure of multiple atrial septal defects. Initial results and value of two- and three-dimensional transesophageal echocardiography. Eur Heart J 2000;21:941–947.
6. Bartel T, Konorza T, Arjumand J, et al. Intracardiac echocardiography is superior to conventional monitoring for guiding device closure of interatrial communications. Circulation 2003;107:795–797.
7. Awad S, Garay F, Cao Q, Hijazi Z. Multiple amplatzer septal occluder devices for multiple atrial communications: Immediate and long-term Follow-up results. Catheter Cardiovasc Interv 2007;70:265–273.