Late last year, The New England Journal of Medicine published three studies showing that closing the patent foramen ovale (PFO) in patients who experienced a cryptogenic stroke with PFO reduces the likelihood of recurrent stroke.

The Patent Foramen Ovale Closure or Anticoagulants Versus Antiplatelet Therapy to Prevent Stroke Recurrence (CLOSE) trial evaluated PFO closure in patients who had a recent cryptogenic stroke attributed to PFO with an associated atrial septal aneurysm or large interatrial shunt. The researchers used a variety of closure devices. No strokes occurred in the PFO closure group more than 5 years after the procedure, whereas stroke occurred in 14 patients in the antiplatelet-only group, essentially a 100% reduction in stroke risk.

The Gore REDUCE Clinical Study (REDUCE) did not mandate a specific size or atrial septal aneurysm, but the trial showed that patients undergoing PFO closure with Cardioform Septal Occluders (Gore and Associates) plus antiplatelet therapy experienced a 77% lower relative incidence of clinical ischemic stroke and a 49% reduction in the composite of clinical and radiographic new brain infarctions as compared to patients receiving antiplatelet therapy.

The Randomized Evaluation of Recurrent Stroke Comparing PFO Closure to Established Current Standard of Care Treatment (RESPECT) trial had a similar approach and design, and investigators found that use of the Amplatzer PFO Occluder (Abbott Vascular, formerly St. Jude Medical) was associated with a 45% reduction in the risk of ischemic stroke and a 62% reduction in the rate of cryptogenic stroke.

The collective results of these trials offer a compelling case for PFO closure. However, the data also underscore the importance of determining whether a PFO is the cause of the stroke and anatomically suitable for closure. On this point, specifically, the collective publication of these findings serves as a call to action to neurologists to take a direct role in selecting patients for the procedure and participating in shared decision making.

Neurology and PFO

Although neurologists may order the echocardiogram that identifies the PFO in a patient with a recent stroke, in the past, we have typically deferred to cardiologists to decide on closure. However, given the changing landscape and the more sophisticated understanding of cryptogenic stroke, neurologists must take a more active role. In fact, the United States Food and Drug Administration label for the Amplatzer PFO Occluder specifically states that it be used for patients “who have had a cryptogenic stroke due to a presumed paradoxical embolism, as determined by a neurologist and cardiologist following an evaluation to exclude known cause of ischemic stroke.” Neurologists should only refer the patient to a cardiologist for consideration of closure after performing a thorough evaluation and accounting for a host of factors.

About 25% of the general population has a PFO, but it is only worth closing if it is the likely cause of the stroke. We as neurologists must confirm that the stroke has an embolic appearance and that there is no other probable cause. An embolic ischemic stroke typically involves the cerebral or cerebellar cortex, while small subcortical and pontine strokes are unlikely to be embolic in origin. At a minimum, ancillary testing is needed to ensure that there is no large artery stenosis, no high-risk source of cardioembolism such as atrial fibrillation or major structural cardiac anomaly, and no prothrombotic disorder. If any of these criteria are not met, neurologists should counsel the patient that the PFO is most likely an incidental finding and probably not the cause of his or her stroke and therefore should not be closed. If the stroke is embolic and cryptogenic, then PFO closure should be considered, as these were the selection criteria used for these trials.

One additional tool that may be helpful is the Risk of Paradoxical Embolism (RoPE) score, which aims to help determine whether a stroke is attributable to the PFO. Generally, young patients with few vascular risk factors (eg, hypertension, diabetes, smoking) have high RoPE scores and are most likely to have had their stroke as a consequence of the PFO. There is a plan to go back and look at RoPE across all of these trials and to see if it helps identify which patients are most likely to benefit.

The anatomical features of the PFO may be important, but this is less clear. Notably, the CLOSE trial only included patients with large PFOs or PFOs plus atrial septal aneurysm, and it seemed to have the greatest relative benefit of all the trials. However, REDUCE and RESPECT included a broader range of PFOs and still demonstrated benefit, so it remains unclear if these features truly matter. However, the cardiologists will be best able to determine if closure is technically feasible and which device is most appropriate for the given anatomy.

Finally, this multidisciplinary team must convey their findings to patients to inform their decision making. In most cases, the patient will rely on the neurologist to confirm that this procedure will treat the likely cause of the stroke.

Complications of Closure

PFO closure has a favorable safety profile, but there are risks related to the device and the procedure. Atrial fibrillation may occur and is generally transient and periprocedural, although there is about a 1% chance of late and prolonged atrial fibrillation. This is a topic of ongoing investigation.

The challenge in trying to compare the effects of occult atrial fibrillation across devices or studies is that when the first trials were done more than a decade ago, researchers were not as attentive to the idea of occult atrial fibrillation as we are now. With increased awareness of occult atrial fibrillation and improved methods to detect it, with each subsequent trial, more atrial fibrillation is found, perhaps because clinicians are looking harder for it. This is likely relevant both before and after PFO closure. Prior to closure, mobile cardiac telemetry monitoring should be considered to ensure that occult paroxysmal atrial fibrillation was not the likely cause of the stroke. This is probably uncommon in younger adults but becomes increasingly prevalent and worth investigating in patients over 50. Cardiac monitoring is yet another opportunity for us to collaborate with our cardiology colleagues to ensure the right diagnosis. After closure, patients should be monitored clinically for any signs or symptoms of arrhythmia.

Are there some patients who have anatomic or other clinical features that place them at higher risk for atrial fibrillation? This is unknown at present, but perhaps we will be able to learn more from pooling the results of all PFO trials.


The CLOSE, RESPECT, and REDUCE trials represent a transformative advance in the management of PFO and stroke. Their rigorous design and patient selection criteria present the most compelling case yet that closure is effective, but neurologists must lead the quest to determine the likely underlying cause of the stroke and whether closure is the best route.

Scott E. Kasner, MD
Professor of Neurology
Director of the Comprehensive Stroke Center
The University of Pennsylvania Medical Center
Philadelphia, Pennsylvania