Did the patent foramen ovale (PFO) cause the patient’s stroke? This, experts say, is the key question when deciding whether to recommend PFO closure. Heart–brain teams can help with the answer.   

Long path to approval

When the Food and Drug Administration (FDA) announced approval of the Amplatzer PFO occluder (St. Jude Medical, which is now owned by Abbott) for 18- to 60-year-old patients with cryptogenic stroke in October 2016, the agency’s press release highlighted the 10-year results of the RESPECT trial, which were presented at TCT.16 that week. The rate of new strokes was low in both patient groups—those who received the device plus medical therapy as well as those who received medical therapy alone—but the device group achieved a 50 percent rate reduction in recurrent cryptogenic stroke. The devices had “demonstrated a reasonable assurance of safety and effectiveness,” the FDA wrote.

The approval came a decade after the agency halted the humanitarian device exemption (HDE) that had previously made the Amplatzer (then owned by AGA Medical) and the CardioSEAL STARFlex (NMT Medical) available for patients with cryptogenic stroke who had experienced at least one additional stroke despite treatment. The companies voluntarily withdrew the devices because the HDE indication’s limit of 4,000 U.S. eligible patients per year had been exceeded.

In the period between the HDE withdrawal in 2006 and the FDA approval in 2016, debate over the value of PFO closure was heated, with results of some studies, including the findings of the original RESPECT trial (N Engl J Med 2013;368:1083-91), fueling disagreement between interventional cardiologists and neurologists. In July 2016, a few months before the FDA approved the Amplatzer PFO occluder, the American Academy of Neurology (AAN) published new recommendations discouraging use of PFO closure for cryptogenic stroke (Neurology 2016;87:1-7).

Both the AAN and the FDA have stressed that having a PFO is common while the need for treatment is rare. It’s estimated that about one-quarter of Americans have the hole between their heart’s atria, says the FDA, but it “typically causes no health problems and does not require treatment.”

Since the device was approved, the CLOSE and Gore-REDUCE trials, both presented this spring at the European Stroke Organization Conference in Prague, delivered positive news for PFO closure. Gore-REDUCE showed a 77 percent relative reduction in recurrent strokes with PFO closure, and CLOSE found a five-year absolute risk drop-off for recurrent stroke of 4.9 percent in PFO closure patients.

Patient selection is key

The key to both trials’ positive results may have been enrolling the right patients, meaning those whose earlier strokes were likely caused by a PFO, rather than other causes. Making this determination is one of the reasons that correct patient selection requires finesse and teamwork in practice, says Michael Savage, MD, professor of cardiology at Thomas Jefferson University's Sidney Kimmel Medical College in Philadelphia.

“Someone with vague symptoms, like dizziness or very vague numbness, might be labeled as having had a stroke or [transient ischemic attack] by their family physician, who then refers the patient to a cardiologist for the [PFO closure] procedure,” he explains. It could turn out the PFO had nothing to do with the symptoms, which means the patient didn’t require closure.”

The challenges of selecting the right patients and reducing the potential for unnecessary procedures are driving greater collaboration between cardiologists and neurologists. “Only a neurologist will let us know it’s a stroke of unknown cause,” says Naser Ammash, MD, interventional cardiologist at the Mayo Clinic in Rochester, Minn. He relies on his neurology colleagues to determine whether there’s a definable cause for a patient’s stroke, such as atherosclerotic plaque, a clotting disorder or an arrhythmia.

Bram Zuckerman, MD, director of the cardiovascular devices division at the FDA’s Center for Devices and Radiological Health, made a similar point in last fall’s press release. “As the device labeling clearly states,” he said. “Patients need to be evaluated carefully by a neurologist and cardiologist to rule out other known causes of stroke and help ensure that PFO closure with the device is likely to assist in reducing the risk of a recurrent stroke.”

“Therapeutic uncertainty is inherent in decisions in patients with PFO and crytogenic stroke,” wrote Savage and colleague David Fischman, MD, in an editorial (Interv Cardiol 2017;9[1]:53-55). “How do we achieve the appropriate clinical equipoise between benefit and risk, between device overutilization and underutilization?”

The answer, Savage suggests, is to mandate a “heart–brain team approach” modeled on the heart team concept used by cardiologists and cardiothoracic surgeons to determine which patients are appropriate for transcatheter aortic valve replacement and for bypass surgery vs. angioplasty. “Only in the case of a heart–brain team, it would be a cardiologist and a neurologist,” he wrote in an email.

Teams take different forms

At Thomas Jefferson University, where Savage is also director of the cardiac catheterization laboratory, patients who have suffered a stroke or transient ischemic attack typically are admitted to the hospital, where they are evaluated first by a neurologist and then a cardiologist. Elsewhere teams may work together differently, he says, but they almost always include an interventional cardiologist and a neurologist who review cases together and with patients. “It’s teamwork, but not as if, physically, specialists are in the same office.”

Today, the members of the Mayo Clinic’s heart–brain team do see patients together, says Ammash. Several years ago, he and colleagues created an informal team in response to what he perceived as a “ridiculous” number of unwarranted PFO closures being performed industrywide. In the past year, the Mayo Clinic has created a more formal process, with a new “Heart Brain Clinic” that provides multidisciplinary evaluations for patients with neurological symptoms that may be due to a cardiac etiology, such as atrial fibrillation or a PFO.

“We make decisions and see patients together, rather than one seeing them at one point and a neurologist seeing them two weeks later,” he says. “Otherwise, it’s too fragmented.”

At the University of Washington (UW) Medical Center in Seattle, cardiologists and neurologists try for joint visits. “Ideally, an interventional cardiologist and a stroke neurologist see a patient and, at some point, do a joint visit with a patient,” says Creighton W. Don, MD, PhD, associate professor of medicine and cardiologist at UW. “They should both have evaluated the patient and discussed the feasibility and indication for the PFO closure.”

Even if they don’t see the patient together, a neurologist does a complete workup and a cardiologist evaluates the patient to determine “among other things, whether he or she actually has a PFO,” says Don, adding that in some cases imaging has revealed there was no PFO.

Savage hopes that the heart–brain team will burgeon into a universal model for PFOs, but he’s not sure if it will be feasible at smaller, community hospital programs, where neurologists might not be available to consult on cases. “They might be a little overwhelmed, so cardiologists might be making more decisions unilaterally for this particular patient population.”