Download this article's Factoid (PDF & PPT for Gold Subscribers)

The presence of a patent foramen ovale (PFO) in patients with transvenous leads for pacemakers or defibrillators more than triples the long-term risk of stroke or transient ischemic attack (TIA), according to a registry study published online August 14, 2013, ahead of print in Circulation.

Investigators led by Samuel J. Asirvatham, MD, of the Mayo Clinic (Rochester, MN), retrospectively evaluated 6,075 consecutive patients at their institution who received a cardiac implantable electronic device with endocardial leads between January 2000 and October 2010. Echocardiography identified a PFO in 364 patients.

Stroke/TIA Risk Tripled

At a mean 4.7 ± 3.1 years after implantation, more than 3 times as many patients in the PFO group compared with the non-PFO group experienced a stroke or TIA consistent with a cardioembolic etiology (primary endpoint; 8.2% vs. 2.0%; P < 0.0001). About two-thirds of events in each group were classified as strokes, for an overall stroke rate of 5.2% vs. 1.4% (P < 0.0001). Kaplan-Meier estimates of stroke/TIA risk showed higher rates for PFO vs. non-PFO patients increasing over time (table 1).

Table 1. Kaplan-Meir Cumulative Incidence of Stroke/TIA Over 3 Years^a

^a P < 0.001 for overall comparison.

Even after adjustment for age, sex, and history of prior stroke/TIA, stroke/TIA risk remained strongly associated with the presence of a PFO (HR 3.30; 95% CI 2.19-4.96); additional adjustment for history of A-fib and aspirin and warfarin use at the time of implantation showed similar results (HR 3.36; 95% CI 2.23-5.07; both P < 0.0001).

In exploratory analyses, the link between PFO and stroke/TIA was observed in multiple subgroups based on age (< 65 vs. ≥ 65), sex, prior history of stroke/TIA, baseline history of A-fib, baseline aspirin and warfarin use, and thromboembolic risk as measured by the CHA₂DS₂-VASc score (0-2 vs. ≥ 3; all P < 0.001).

There was no difference in rates of all-cause mortality between patients with or without a PFO (46% vs. 39%; P = 0.25), a finding that persisted after adjustment for age, sex, and history of stroke/TIA (HR 0.90; 95% CI 0.76-1.06; P = 0.20) as well as after additional adjustment for history of A-fib and baseline aspirin and warfarin use (HR 0.89; 95% CI 0.75-1.05; P = 0.15).

According to the authors, the “fact that endovascular leads are a nidus for thrombus development and adopt an intracardiac course that frequently is in juxtaposition to the interatrial septum likely accounts for the strong association we observed between PFO and stroke/TIA in device recipients.”

Zeroing in on Patients Likely to Benefit from Closure

In a telephone interview with TCTMD, Robert J. Sommer, MD, of Columbia University Medical Center (New York, NY), said that it is common knowledge that patients can develop thrombus on pacemaker wires and common suspicion that the mechanism of stroke with PFO is paradoxical embolism. “Putting these 2 together has always been a source of concern, but this is the first time the relationship has been studied formally, and the results are pretty impressive,” he noted.

The findings are important, Dr. Sommer asserted, because “at a time when we’re struggling with indications for PFO closure and therapy with anticlotting agents, the study offers a little bit of clarity despite its retrospective nature. As we look for specific patients most likely to benefit from therapy for a PFO, this is one of the few indications for which we have good evidence that it works.”

Although many clinicians already screen device candidates for a PFO, the new data justify making that a strong recommendation, Dr. Sommer said. Use of transesophageal echocardiography (TEE) is not necessary, he added, noting that transthoracic echocardiography and intracranial Doppler, in conjunction with agitated saline injection, are commonly available noninvasive alternatives.

PFO Warrants Consideration of Strategy

The authors propose several possible strategies when a candidate for an implantable device is found to have a significant PFO:

• Anticoagulation
• Epicardial or subcutaneous device placement
• PFO closure

Dr. Sommer noted that there is as yet no proof that long-term anticoagulation for this purpose is safe or effective. As for epicardial placement of a device, the downside is that it requires open-chest surgery, both to implant the device initially and again to replace the device leads if they break or wear out. Moreover, he noted, some data suggest that epicardially placed pacemakers do not have the same durability as transvenous devices.

“I think [the study] is going to scare a lot of people who routinely put in pacemakers,” Dr. Sommer said. “Since most [arrhythmia patients] are already [receiving echocardiography] as part of their evaluation, I don’t think practice is going to need to change dramatically. But just the awareness that having a big right-to-left shunt in this population may be a significant risk factor for stroke should at least warrant discussion about what to do about a PFO in a patient being considered for a pacemaker.”

Like the authors, Dr. Sommer said the next step is to confirm the main findings in a cohort study.

Study Details

There were no notable differences in clinical characteristics between PFO and non-PFO patients. Patients with PFO were more likely to have a history of A-fib at baseline (P = 0.03). Aspirin was used less often at the time of device implantation among PFO patients than non-PFO patients (P = 0.026), although warfarin use was similar between the groups.

TEE reports were available for 12.6% of patients, and 29.7% of PFO diagnoses were based on TEE findings.

Related Stories:

• Meta-analysis Gives Edge to PFO Closure for Stroke Protection
• Final Results Provide CLOSURE on Device Therapy for PFOs, Cryptogenic Stroke
• Most Recurrent Ischemic Events After PFO Closure Related to Comorbidities