Migration of PFO closure device and entrapment within tricuspid valve leading to tricuspid regurgitation: surgical correction of an extremely rare complication after transcutaneous PFO closure
- Myat Soe Thet ,
- Amir Sepehripour ,
- Abed Elfattah F A Atieh and
- Carmelo Di Salvo
- Cardiothoracic Surgery, St Bartholomew's Hospital, Barts Health NHS Trust, London, UK
- Correspondence to Myat Soe Thet; myatsoe.thet@nhs.net
Abstract
A 41-year-old woman was referred to tertiary cardiothoracic surgery centre following embolisation of the Amplatzer patent foramen ovale (PFO) closure device to septal leaflet of tricuspid valve with reopening of PFO. Two years earlier, she presented with thalamic stroke, and she was found to have a PFO following investigations. The following year she underwent transcatheter closure. Six months after the percutaneous closure, she presented again with significant periods of shortness of breath. Imaging studies revealed the migration and embolisation of PFO closure device to the septal leaflet of tricuspid valve with reopening of the foramen and significant tricuspid valve regurgitation. She underwent open heart surgery using cardiopulmonary bypass for retrieval of the device, closure of the foramen and repair of the tricuspid valve. The patient recovered well without any significant issues following surgery.
Background
Foramen ovale is the communication between the right and left atrium of the heart during fetal life for the venous blood to shunt from right to left atrium bypassing the non-functional fetal lung. The foramen usually closes following delivery due to increased pressure in the left atrium compared with the right. If it remains open after the age of one, it is termed as patent foramen ovale (PFO), and it is found in up to 27% of the general population. PFO is usually the culprit in cryptogenic ischaemic stroke, especially in younger adults.1
Closure of PFO with transcatheter approach with antiplatelet therapy has become the gold standard in managing patients with cryptogenic stroke with PFO.2 Studies have shown that there is reduction in recurrent strokes in patients who underwent transcatheter PFO closure compared to those who were managed with medical treatment alone.3 However, transcatheter approach using PFO closure device is not without risks. Atrial fibrillation and other arrhythmias, device thrombosis and embolisation, residual shunt, access site haematoma, recurrent stroke and cardiac tamponade are reported complications associated with the procedure.4
We report an extremely rare case of the Amplatzer PFO closure device retrieval following device migration, embolisation with resultant reopening of foramen ovale and damage to tricuspid valve leading to tricuspid valve regurgitation.
Case presentation
A previously fit-and-well 41-year-old woman was referred by the cardiology team at a district general hospital in England to our tertiary cardiothoracic surgery centre due to a complication after PFO closure device. She presented to the cardiology team 2 years earlier with a thalamic cryptogenic stroke. After undergoing investigations, bubble echocardiogram confirmed the presence of a PFO. She underwent transcatheter closure PFO without any complications. The procedure itself was deemed challenging, however, postprocedural echocardiogram showed the device was well seated. She was started on antiplatelet and lipid lowering therapy for secondary prevention of cerebrovascular and cardiovascular disease.
Six months after the procedure, she presented to the cardiology team again with periods of significant shortness of breath, especially on exertion and usually accompanied by occasional heavy chest pain. She also developed paroxysmal nocturnal dyspnoea and had to use at least two pillows while sleeping at night. She could not lie flat completely. The repeated bubble echocardiogram showed PFO device migration onto the septal leaflet of the tricuspid valve (figure 1) with moderate to severe tricuspid regurgitation, device embolisation, reopening of the PFO (figure 2).
Transthoracic echocardiogram showing migration of PFO closure device onto the septal leaflet of the tricuspid valve. PFO, patent foramen ovale .

Transthoracic echocardiogram showing moderate to severe tricuspid valve regurgitation. 2D, two-dimensional imaging; CF, colour flow mapping; WF, wall filter; CW, continuous wave doppler; TR, tricuspid regurgitation.

On examination, the patient was haemodynamically stable. There were no signs of congestive heart failure. The rest of the physical examination was not remarkable. Following a multidisciplinary team discussion, surgical retrieval of the PFO closure device was deemed necessary to prevent further embolic events as well as to repair or replace the tricuspid valve to correct the regurgitation. The patient was urgently transferred to our tertiary cardiothoracic surgery centre for open heart surgery.
Intraoperative transoesophageal echocardiogram confirmed the transthoracic echocardiogram findings. A standard median sternotomy was performed to see a smaller-than-usual heart inside the chest for a 41-year-old woman. Ascending aorta and bicaval venous cannulation was performed, and cardiopulmonary bypass was initiated. Aortic cross clamp was applied and cold blood antegrade cardioplegia was used to achieve good diastolic arrest. Right atrium was obliquely opened.
The PFO closure device was found to be deeply endothelialised and embedded in the septal leaflet of the tricuspid valve apparatus. A combination of extensive sharp and blunt dissection was carried out to remove the device (figure 3) without damage to the leaflet to preserve the native tricuspid valve. It was difficult to assess the competency of the tricuspid valve intraoperatively due to distorted anatomy of the valvular apparatus, and it would not have been possible to perform tricuspid ring annuloplasty. Therefore, the septal and posterior leaflets of the tricuspid valve were bicuspidised to correct the regurgitation. On ventilation of the lungs, the PFO became apparent at the apex of the interatrial septum and was closed with a simple continuous suture without the need for a patch. The aortic cross clamp was removed, and the right atrium was closed followed by the weaning of the heart from cardiopulmonary bypass.
Deeply endothelialised PFO closure device following surgical extraction. PFO, patent foramen ovale.

Intraoperative transoesophageal echocardiogram was performed again and showed the persistence of severe tricuspid regurgitation. A brief intraoperative multidisciplinary team meeting, including the cardiac surgeon, anaesthetist and cardiologist, was held and the consensus was to correct the regurgitation since it would be clinically significant postoperatively. A further attempt to repair the tricuspid valve was made to avoid replacing the tricuspid valve, since the biological tricuspid valve would not last longer than a decade and a metallic tricuspid valve would render the relatively young patient to have life-long anticoagulation.
The heart was put on bypass, and the right atrium was opened again. The tricuspid valve was further bicuspidised and the right atrium was closed. The subsequent transoesophageal echocardiogram showed significant improvement in tricuspid regurgitation. Cardiopulmonary bypass was weaned off with a total aortic cross clamp time of 37 min and bypass time of 60 min. Sternum was closed using stainless steel wires in a routine manner. The patient was transferred to the intensive care unit in a stable condition.
Outcome and follow-up
Postoperatively, she was noted to have resolution of the presenting symptoms. The patient did not have any conduction issues, and did not require any epicardial pacing. She remained in sinus rhythm despite intraoperative surgical manipulation of the septal leaflet of the tricuspid valve considering the close proximity of the valve to the conduction system of the heart. The patient had uneventful recovery, and was discharged from the hospital 4 days after the procedure. Long-term follow-up will be at the local referring hospital by the cardiology team.
Discussion
The reported patient underwent open heart surgery for a rare complication of transcatheter PFO closure after 6 months of the initial procedure. The transcatheter PFO closure is a relatively low risk procedure with major complication rate of 1.5%, which includes death, fatal pulmonary embolism, cardiac tamponade, requirement of surgical intervention and minor complication rate of 7.9% including atrial fibrillation and other atrial arrhythmias, transient heart block, asymptomatic device thrombosis without requirement of surgery and vascular venous access site complications.5 Although some of the complications, including device retrievals, can be managed with transcatheter approach, others require open heart surgery.6
Up to 1.2% of the complications to transcatheter PFO closure may need subsequent cardiac surgical intervention.7 Previously reported cases include device erosion into left ventricular outflow tract, cardiac rupture or tamponade and fistula formation between aorta and right atrium.7–10 PFO closure device migration to the tricuspid valve is extremely rare, and there has been only one reported case in the current literature to our knowledge.11 These conditions usually require urgent surgical retrieval of the PFO closure device.
In summary, this case highlights a rare yet serious complication of transcatheter PFO closure using a device requiring subsequent open heart surgery. The extent of endothelialisation within 6 months after device insertion is extensive, and during the management of the complication, tricuspid valvular apparatus should be preserved as much as possible. Despite the close proximity of the involved septal leaflet of the tricuspid valve, it is possible to extract the device and repair the valve without any conduction abnormalities.
Learning points
-
Transcutaneous closure of patent foramen ovale is associated with rare, yet serious complications.
-
Timely referral to the cardiothoracic surgical team is important to provide safe and effective patient care.
-
Surgical retrieval of the migrated embolised device and repair of the tricuspid valve could be done safely.
Footnotes
-
Contributors MST: concept, data collection, interpretation of data, drafting and critical revision, final approval of the work, accountability of the work. AS: concept, interpretation of data, drafting and critical revision, final approval of the work, accountability of the work. AEFAA: concept, interpretation of data, critical revision, final approval of the work, accountability of the work. CDS: concept, interpretation of data, critical revision, final approval of the work, accountability of the work.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Patient consent for publication Obtained.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2021. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer