Fluoroscopy-guided closure of patent foramen ovale for secondary prevention of paradoxical embolism

Our study is the world’s largest single-center series of percutaneous closures of the patent foramen ovale (PFO) without the use of intraprocedure echocardiography. The series has grown to include almost 1000 patients; however, this report pertains to the first 525 consecutive patients for which there is at least 1 year of follow-up.¹ Most patients were treated for secondary prevention after a presumably paradoxical stroke.

Subjects and methods

Eight different PFO closure devices were used in 525 patients. All of the devices allow deployment of the left part of the device deep in the left atrium. The device is then pulled back until it is stopped by the interatrial septum, while pulling shut the PFO valve. At this moment, the right part of the device is deployed, which, if placed correctly, saddles the muscular and wedge-like septum secundum cranially and the thin and often aneurysmatic septum primum caudally. Instead of using echocardiography (transesophageal or intracardiac) to assess correct placement of the device, a hand injection of contrast medium into the right atrium was used (Figure 1).

A projection is chosen that displays the device in perfect profile, with the interatrial septum separating the left and right parts of the device. Although the septum primum is a flimsy membrane that is easily contorted by the device, particularly before it is released from its pusher or cable, the fleshy septum secundum conspicuously separates the 2 parts of the device. In fluoroscopy, the septum secundum appears as a roundish white speck between the left (cranial) parts of the device, which seem to be biting into it; hence, the name “Pacman sign” (Figure 2). When the Pacman appearance is observed, the device can be released safely.

Results

Three hundred ninety-six patients received Amplatzer PFO Occluders and 129 were treated with a variety of other devices. This summary concentrates on the 396 receiving Amplatzer PFO Occluders, which provide the easiest implantation technique, the highest success rate, and the lowest thrombosis rate. Because of these characteristics, these devices have become the standard for PFO closure at our center and are the most commonly used devices worldwide.¹ Amplatzer PFO Occluders are made of Nitinol mesh formed into double disks connected with a thin, stretchable waist. Each disk contains a polyester fabric inlay that is fully absorbable.

Complete PFO closure as assessed by contrast transesophageal echocardiography performed at 6 months or later can be achieved in 91% of cases. Grade 1 shunts (1 or a few bubbles passing) are present in 17% before occlusions and 6% after occlusion; grade 2 shunts (cloud of bubbles passing) in 7% before and 2% after; and grade 3 shunts (wide open PFO) in 80% before and 1% after.

The 3 sizes of Amplatzer PFO Occluders used in this series, the 18-mm Occluder (18-mm disks on both sides), the 25-mm Occluder (18-mm disk on the left and 25-mm disk on the right), and the 35-mm Occluder (22-mm disk on the left and 35-mm on the right), were used for different indications and yielded different results. Three more sizes have been added since the inception of this study: 25/25 mm, 30/30 mm, and 35/35 mm. These devices are commonly referred to as Cribriform Occluders, alluding to the fact that they are particularly suitable for multiple, small atrial septal defects (Swiss cheese septum) with or without an associated PFO.

The complete closure rate with both the 18-mm and the 25-mm Occluders is 93%, whereas it is only 73% with the 35-mm Occluder. The 18-mm device, however, was used in only 8% of cases, usually for small PFOs without atrial septal aneurysm (ASA). The 25-mm Occluder was used in 79% of cases (routine device), and the 35-mm device was used in the remaining 13% of cases, usually for large PFOs with very mobile septum primum (also referred to as ASA). The larger device is used for these situations to avoid embolization, which notably never occurred in this series. Whether a smaller device would have been equally safe with a higher complete closure rate in these patients cannot be deduced from these data.

The clinical follow-up was remarkable, with a lack of recurrent ischemic stroke, transient ischemic attacks, other peripheral embolism, or device problems in 99% of subjects at 1 year and 97% of subjects at 5 years. This will have to be compared with the natural course in such patients, in which the incidence of such events is reported to be between 3% and 16%.² The best comparative data so far point to an advantage of device closure over drug prevention, at least when compared with aspirin rather than warfarin treatment.³

Complications

No cardiac complications occurred with the Amplatzer PFO Occluder. Arteriovenous fistulas occurred at the puncture site in 0.5% of cases. Some of them may have been caused by the incidental coronary angiography performed in about 70% of cases. Had intracardiac echocardiography been used, an additional equal-size or larger venous introducer would have been placed, potentiating the risk for arteriovenous fistulas. The introducer size for the Amplatzer PFO Occluder is 9 French; the one for echocardiography is 9-11 French.

At the follow-up transesophageal echocardiogram approximately 6 months after the intervention, thrombi were seen on devices in less than 0.5% of cases. These were all clinically silent, and most disappeared with a several-month regimen of oral anticoagulation therapy. This finding corroborates those observed in a similar German multicenter study (Figure 3).⁴ The only other complications encountered during follow-up were a few incidences of transient supraventricular arrhythmias reported as palpitations.

Discussion

This study shows that a PFO can be reliably closed with modern equipment during a 15-minute intervention that only requires local anesthesia of the groin. The procedure is performed using a venous puncture (unless additional arteriograms are performed), allowing patients to put their finger on the groin at the end of the procedure and walk out of the catheterization laboratory. A couple of hours later, patients can be discharged to resume full physical activity immediately.

Subjects in the study received aspirin for 5 to 6 months after the procedure and clopidogrel (Plavix) for 1 to 6 months, with the recommendation to take precautions against endocarditis for the first 2 to 3 months. Transesophageal echocardiography is commonly performed after 6 months. If complete closure, good device position, and absence of thrombosis are documented, no further follow-up measures or restrictions are required.

The results of this study put into perspective the current overly restrictive indications for percutaneous PFO closures accepting only patients with 2 events who have been receiving medical treatment in the United States and those with at least 1 event in the rest of the world. Dangerous-looking PFOs (wide gap, ASA, Eustachian valve leading the blood from the inferior vena cava directly onto the PFO) are prevalent in about 4% of the population and should be closed before they do serious harm, as exemplified in the accompanying case report. In addition, PFO closure should be made readily available to at-risk individuals, such as divers, brass musicians, glass blowers, supersonic jet pilots, and those working in a crouching position, as well as those with migraine headaches. The primary goal even in the last group is prevention of stroke or other paradoxical embolisms, such as myocardial infarction, splenic infarction, kidney infarction, or acute ischemia of the legs or arms; improvement of headache is merely a welcome bonus.

Conclusion

Several randomized trials of PFO closure for stroke prevention are under way, and some are about to complete enrollment. It is questionable, however, whether a follow-up period of a few years is sufficient to reveal the benefit of the procedure, as events often take decades to happen. Migraine trials may bring positive results earlier and may open the door more widely for PFO closure.

Disclosure

The author has no relationship with any commercial entity that might represent a conflict of interest with the content of this article.