Medicare Analysis Suggests Benefit of PFO Closure Beyond Age 60

BASEL, Switzerland—In a study of Medicare fee-for-service beneficiaries older than 60 who had had an ischemic stroke, percutaneous patent foramen ovale (PFO) closure was associated with improved clinical outcomes, researchers report.

The rate of recurrent ischemic stroke through 3 years of follow-up was a relative 38% lower among those who underwent the procedure than among those who did not, according to Jeffrey Saver, MD (David Geffen School of Medicine at UCLA, Los Angeles, CA), who presented the findings here at the European Stroke Organisation Conference (ESOC) 2024.

Overall, rates of safety endpoints were low in both groups, although some events—including atrial fibrillation (AF) or flutter and venous thromboembolism (VTE)—were significantly more frequent following PFO closure.

“This is, as far as we know, the largest comparative study [of PFO closure] performed in patients over age 60,” he said.

That’s noteworthy because the randomized trials supporting the use of the procedure were almost entirely restricted to participants ages 18 to 60, with guidelines largely confining their recommendations to this age group. New ESO guidelines on the diagnosis and management of PFO after stroke that were released Thursday and published in the European Stroke Journal, for instance, contain a strong recommendation for closure plus antiplatelet therapy in patients ages 18 to 60 who have no identified cause of stroke other than PFO, but state that the evidence is insufficient to make a recommendation regarding use in patients younger than 18 or older than 60.

Nonetheless, prior studies have shown that PFO closure is being performed in older patients despite the lack of randomized evidence to support the practice.

This new analysis suggests that “in properly selected patients over age 60, PFO closure may be beneficial in reducing recurrent ischemic stroke with a low incidence of adverse events,” Saver said, adding, however, that “this is an encouraging observation that needs to be validated by randomized trials.”

Commenting on the study for TCTMD, Peter Schellinger, MD, PhD (Johannes Wesling Klinikum Minden, Germany), indicated that the results should be interpreted carefully and that the decision to proceed with PFO closure in any patient should be made with appropriate caution.

“I’m not not doing it, but I’m critical, and now we are talking about the patients older than 60 where there are no randomized data of any quality that I’m aware of at all,” Schellinger said. “Considering that every fourth person on this planet has a PFO, for me sometimes this smells like expanding the market.”

Less Stroke but More AF, VTE

For the study, Saver and colleagues used two sources—administrative claims data from the Centers for Medicare & Medicaid Services and the Abbott patient device-tracking database—to identify Medicare fee-for-service beneficiaries older than 60 who had an ischemic stroke between 2016 and 2022 and subsequently underwent PFO closure with an Amplatzer or Amplatzer Talisman PFO Occluder (both Abbott). Their outcomes were compared with those in beneficiaries who had an ischemic stroke during the same time frame, were diagnosed with a PFO or atrial septal defect, and did not undergo PFO closure.

After excluding patients with a history of atrial fibrillation/flutter (a potential complication of PFO closure) and performing propensity-score matching, the investigators compared 1,132 patients who underwent the procedure with 4,376 who didn’t. Overall, mean age was 72 and 44% of patients were women.

The primary effectiveness endpoint was recurrent ischemic stroke through 3 years of follow-up, and there was a lower rate observed in the closure group (1.65 vs 2.66 per 100 patient-years); that difference was significant (HR 0.62; 95% CI 0.44-0.88).

This is an encouraging observation that needs to be validated by randomized trials. Jeffrey Saver

An analysis of 13 falsification endpoints designed to assess for residual bias showed no difference for 11 of those endpoints, with one favoring PFO closure and one favoring no procedure. That indicates “no evidence of . . . broad confounding across disease states, suggesting a specific effect on ischemic stroke,” Saver said.

He pointed out, however, that there was a lower risk of 3-year mortality following PFO closure, suggesting possible residual confounding. In a post hoc analysis adjusting for that difference in mortality, there remained a signal of lower ischemic stroke risk after the procedure, although the difference was no longer significant (HR 0.72; 95% CI 0.45-1.14).

The main safety endpoint was a composite of all-cause mortality, atrial fibrillation/flutter, venous thromboembolism, bleeding requiring transfusion, and vascular access-site complications requiring intervention at 30 days, and this was significantly higher after PFO closure (3.71% vs 1.74%; P < 0.001). That was driven by increases in atrial fibrillation/flutter (1.41% vs 0.64%; P = 0.01) and VTE (1.86% vs 0.37%; P < 0.001), with no significant difference in all-cause mortality (0.44% vs 0.66%; P = 0.40) and relatively few cases of major bleeding and access-site complications.

Lacking RCTs in Older Patients, ‘Follow the Guidelines’

Saver acknowledged that the study has several limitations, including the possibility of residual selection bias or confounding due to its nonrandomized design. In addition, generalizability outside of the Medicare fee-for-service setting is uncertain, there was limited information available on certain procedural and clinical characteristics, information on stroke severity was incomplete, and the databases did not indicate which antithrombotic regimens patients were taking.

Schellinger, who was one of the moderators of the session at which Saver presented the results, said he believes the investigators were appropriately cautious in their conclusions.

“I’m very critical with regard to implantable devices and the way the randomized trials are done,” he said. “This is derived from the [left atrial appendage] occluders, but also in the PFO trials, there is a lot of fancy statistics going on, and I fear that too many devices are implanted considering the number needed to treat for a benefit.”

We need to be careful not to treat every patient that has a PFO . . . with a procedure that has risks associated with it. Peter Schellinger

Even in younger patients, strokes are increasingly related to typical risk factors like diabetes, lack of physical activity, high cholesterol, and hypertension, “and in my opinion, frequently the PFO is accused of being the source of a stroke but it’s difficult to prove,” Schellinger said. Plus, he said, PFO closure comes with potential complications, including AF and pulmonary embolism, and occluders are difficult to remove if necessary.

His general message regarding use of PFO closure at this point was to “follow the guidelines,” which focus on use of the procedure in patients younger than 60. “I just urge my colleagues and peers, and also my fellows, attendings, and residents, to make that selection based on established scores, such as the RoPE score. Look very thoroughly for other stroke etiologies, especially occult malignancy, if you don’t have a typical explanation. . . . And if you are really sure at the end and you have a high-risk PFO, then recommend the procedure and recommend it in a center that has a higher case load.”

The bottom line, Schellinger stressed, is that “we need to be careful not to treat every patient that has a PFO . . . with a procedure that has risks associated with it.”