Device Closure of Fenestrated Atrial Septal Aneurysm:
Difficulties and Complications with Implantation of Two Devices

Atrial septal aneurysm is a cardiac malformation defined as a localized thinning and bulging of the atrial septum into either atrium. It appears to be acquired as the result of abnormal atrial hemodynamics with resultant pressure deformity and mobility of the more pliable, thin, and redundant septum primum. The prevalence of ASA associated with an atrial defect has been reported to range from 1.2% in adults up to 4.9% in the pediatric population. The incidence is reported as high as 10% of adults studied by transesophageal echocardiography. In neonates it will involute or completely resolve with age and cardiac growth once normal atrial hemodynamics have been established1. A persistent ASA can be isolated finding but generally is associated with other atrial septal defects or PFO. Fenestrated ASA may present with the clinical findings of an ASD with a significant left to right shunt, especially if there are multiple defects. When the shunt is small, the clinical presentation may be recurrent systemic thromboembolic events most likely as a result of paradoxical embolization. In some patients devoid of clinical symptoms, the aneurysm may be detected incidentally during echocardiographic examination. Though there are no universally accepted treatment strategies, recommendations for surgical closure have depended on clinical symptoms and the cardiac evaluation. Recent improvements in transcatheter devices for ASD closure have led to efforts to use such techniques to also eliminate the ASA if significant left to right shunt is present. Not all fenestrated ASA may be suitable for transcatheter device closure. We describe a patient with a fenestrated ASA associated with PFO and two ostium secundum ASDs, which echocardiographically seemed to be suitable for transcatheter device closure. Because of the remote locations of the defects complete closure required the implantation of two devices. Oversizing of one device later resulted in the development of atrioventricular valve insufficiency due to device interference with mitral valve leaflets and required surgical removal of the implanted devices. Case Report. A 12-year-old girl was referred to our institution for an evaluation of an ASD. An ASD and ASA were diagnosed at 1 month of age in another institution when a systolic murmur had been detected. At the time of her evaluation at age 12, she was asymptomatic. She had always been in good health and participated in sports without difficulty. There was no history of other illness or symptoms. The ECG performed at that time revealed a wandering atrial pacemaker and a non specific intraventricular conduction delay. There was a 2/6 systolic murmur at the upper left sternal border and fixed widely split second sound. There was no diastolic murmur. Transthoracic echocardiography and Doppler color flow imaging demonstrated an ostium secundum atrial septal defect with a small to moderate left to right shunt and a large aneurysm of atrial septum (Figure 1A). The atrioventricular valves were competent and the ventricular function was normal. She was reevaluated with annual clinical examination and two-dimensional echocardiography. She remained well until this year when the echocardiogram showed a mobile bulging ASA with predominant left to right shunt and some right to left shunt through an ASD during some cardiac cycles. Here was now mild right ventricular enlargement. Cardiac catheterization was performed to assess the shunt size and to determine if the ASA and ASD could both be treated with a transcatheter device. We had hoped that we could dilate one of the defects and use a device large enough to cover all defects and imbricate the ASA. During transcatheter closure of ASD, a transesophageal echocardiography was performed to assess the size of the aneurysm and to determine the presence and location of any additional atrial septal defects by color flow mapping. The pulmonary (Qp) to systemic (Qs) flow ratio as measured by oximetry was 1.6/1. TEE (Figure 1B) and right upper pulmonary vein angiography (Figure 2) demonstrated a PFO and large ASA with two perforations in the inferior margin of the fossa ovalis. Because the PFO appeared small and the ASA was large we did not balloon size the PFO but elected to place a 35 mm Amplatz PFO device (AGA Medical Corp, Golden Valley, Minn.) in the PFO. After placement of the PFO device, we observed by TEE that the device did not completely cover the ASA or two lower atrial septal defects. We attempted to balloon size the lower defects with two 24 mm AGA sizing balloon (Figure 3). One more posterior defect was clearly extremely small and was only 2–3 mm in size. We attempted to balloon size a more anterior defect and with this balloon inflation there seemed to be a significantly larger defect which had a maximum stretched diameter of 24 mm. The balloon appeared to be appropriately positioned in a more anterior and inferior defect by TEE and fluoroscopy. In retrospect, we believe that the balloon sizing had been in error. However, we selected a 24 mm Amplatzer ASD device and successfully implanted the device in the lower, apparently large defect anticipating that this large device would cover both of the lower atrial septal defects. Both the PFO and ASD devices seemed to be appropriately positioned but the ASD device was obviously constricted in its central portion and was bulbous in appearance. The ASD device overlapped with the PFO device but there was no evidence of deformity of the PFO device (Figure 4). A pulmonary artery angiogram appeared to show a satisfactory result with no significant residual shunt. The ASD device was close to the mitral valve leaflets but did not seem to interfere with mitral valve function. While the ASD device appeared to be oversized, we had hoped to close both lower defects with an oversized device and it was not clear why the device appeared to be significantly constricted by the apparent large defect. We elected to release the device anticipating that the device would later more effectively conform to the ASD size and become flattened onto the atrial septum. Post-cardiac catheterization evaluation included cardiac examination, ECG and TTE performed at one and four months. At the one month examination, the echocardiogram demonstrated that the devices were stable but there was still a small residual shunt between the devices whereas the clinical cardiac examination and ECG were normal. The oversized ASD device remained constricted centrally and the right disk remained bulbous in appearance. The ASD device did not contact the atrial septum appropriately and a small shunt persisted through the central portion of the device. (Figure 5 A and B) TTE also revealed the presence of mild mitral valve insufficiency not present in previous echocardiographic evaluations. Three months later the patient complained of frequent episodes of chest pain lasting five minutes. TTE now demonstrated moderate mitral insufficiency, apparently caused by left side of ASD occluder abutting against the mitral valve anterior leaflet (Figure 6 A,B). There was also mild enlargement of the left atrium. One month later, she underwent cardiac surgery to remove the devices and close the ASD with a patch. At operation the mitral valve was inspected and found to be intact. No intervention on the mitral valve was necessary. The immediate post surgery course was characterized by post pericardiotomy syndrome with pericardial effusion on TTE. The ASD was closed and there was now mild mitral insufficiency. At the last echocardiogram performed 4 months post-surgery, she was doing well. There was no shunt or ASA, and only trivial mitral insufficiency was present. There was no pericardial effusion. Discussion. In this case, there were two small defects close to each other in the lower portion of the ASA and a PFO at the upper margin of the fossa ovalis. We thought that we had enlarged one of the lower defects which could have been closed with a single device that would also covered the other small more posterior defect. Initial inspection of the atrial septum had suggested that the PFO was small but the ASA was large. We had decided to use a 35 mm Amplatzer PFO Occluder, with large retention disks which could have been large enough to cover all the defects and the ASA. However the ASA was too large and the lower defects too remote from the PFO. We were impressed that the 35 mm PFO device was very effective in closing the PFO and obliterating a portion of the ASA. We then hope to close the lower defects with a single separate device which would also further encompass the remaining portions of the ASA. We misinterpreted the size of the lower defect. We suspect that the balloon used to size the more inferior defect had slipped out of the defect during inflation, but because of the elastic nature of the ASA, the balloon protruded into and appeared to be appropriately placed in the left atrium. The ASA simply expanded over the sizing balloon during inflation. Unfortunately, because of the very thin nature of the stretched ASA, the TEE did not allow us to appreciate that this phenomena had occurred. There were two balloon catheters and the PFO device in the atrium so that the ASA was not apparent by TEE as it was closely applied over the sizing balloon. We had thought that the defect had simply dilated to a larger size with the sizing balloon. For this reason, we then elected to use the oversized 24 mm ASD device to close both of the lower defects. Complex ASA with multiple perforations present significant challenges to the interventionist. Accurate delineation of all defects, their precise location and size is difficult. It is possible that high resolution imaging with intracardiac echocardiography may have provided the correct information. The significant over sizing of the ASD device recognized by the constricted center, the bulbous appearance, and the failure to contact the atrial septum should have resulted in device removal. Failure to appreciate the elastic nature and mobility of the ASA resulted in an overestimation of the ASD size. In this situation a much smaller second ASD device or more likely an 18 or 25 mm PFO device would have been a more effective choice to close the lower defects. The small PFO device would not have impinged on the mitral valve, and with proper collapse of the device on the atrial septum, the right or left atrial disc would have covered the second defect. The option of surgical resection of the ASA and ASD closure should be offered for the patient’s consideration. In this case, the patient had opted for attempted device closure first. Once mitral valve incompetence had been recognized after device placement surgical removal of the devices and ASD patch closure was the appropriate solution resulting in recovery of mitral valve function without surgical intervention on the valve itself.

1. Berger F, Vogel M, Meskishvili VA, et al. Comparison of results and complications of surgical and Amplatzer device closure of atrial septal defects. J Thorac Cardiovasc Surg 1999;118:674–680. 2. Brand A, Keren A, Branski D, et al. Natural course of atrial septal aneurysm in children and the potential for spontaneous closure of associated septal defect. Am J Cardiol 1989;64:996–1001. 3. Cao QL, Radtke W, Berger F, et al. Transcatheter closure of multiple atrial septal defects. Initial results and value of two and three dimensional transoesophageal echocardiography. Eur Heart J 2000;21:941–947. 4. Carano N, Hagler DJ, Agnetti A, et al. Device closure of fenestrated atrial septal defects: use of a single Amplatz Septal Occluder after balloon atrial septostomy to create a single defect. Cathet Cardiovasc Intervent 2001;52:203–207. 5. Ewert P, Berger F, Kretschmar O, et al. Feasibility of transcatheter closure of multiple defects within the oval fossa. Cardiol Young 2001;11:314–319. 6. Ewert P, Berger F, Vogel M, et al. Morphology of perforated atrial septal aneurysm suitable for closure by transcatheter device placement. Heart 2000;84:327–331. 7. Gondi B and Nanda NC. Two-dimensional echocardiographic features of atrial septal aneurysm. Circulation 1981;63:452–457. 8. Hanley PC, Tajik AJ, Hynes JK, et al. Diagnosis and classification of atrial septal aneurysms by two-dimensional echocardiography: report of 80 consecutives cases. J Am Coll Cardiol 1985;6:1370 9. Mas JL Patent foramen ovale, atrial septal aneurysm and ischemic stroke in young adults. Eur Heart J 1994;15:446-449 10. Peuster M, Kaulitz R, and Hausdorf G. A novel method for transcatheter closure of atrial septal defect within an aneurysm of the fossa ovalis: double sheath technique. Heart 2000;84:E14. 11. Rosenfeld HM, van der Velde ME, Sanders SP, et al.Echocardiographic predictors of candidacy for successful transcatheter atrial septal defect closure. Cath Cardiovasc Diagn 1995;34:29–34. 12. Walsh KP, Wilmshurst PT, Morrison WL. Transcatheter closure of patent foramen ovale using the Amplatzer septal occluder to prevent recurrence of neurological decompression illness in divers. Heart 1999;81:257–261. 13. Wolf WJ, Casta A, Spire DW. Atrial septal aneurysm in infants and children. Am Heart J 1987;113:1149–1153.