Silent Atrial Fibrillation Increases Stroke Risk

Susan Jeffrey

January 12, 2012

January 12, 2012 — Atrial tachyarrhythmias in the absence of clinical atrial fibrillation (AF) occur frequently among patients with pacemakers and are associated with a significant increase in the risk for stroke or systemic embolism over time.

Results of the Asymptomatic AF and Stroke Evaluation in Pacemaker Patients and the AF Reduction Atrial Pacing Trial (ASSERT) show that in this population of patients with pacemakers who have hypertension but no history of AF, episodes of device-detected atrial tachycardia greater than 6 minutes were seen in approximately one third of patients during almost 3 years of mean follow-up.

Further, these arrhythmias were associated with a 2.5-fold increase in the risk for ischemic stroke and systemic embolism. In a subgroup of patients with a CHADS2 score of 2 or higher, device-detected atrial tachyarrhythmias increased the absolute risk for stroke to 3.78% per year.

This study, "shows clearly for the first time that silent atrial fibrillation in patients with pacemakers is in fact associated with quite a large increased risk of stroke," Stuart J. Connolly, MD, director of the Division of Cardiology, Salim Yusuf Chair at McMaster University, Hamilton Health Sciences, Population Health Research Institute, Ontario, Canada, said in an interview with Medscape Medical News.

"These silent, short episodes of atrial fibrillation that we're seeing seem to behave like the more typical clinical atrial fibrillation that we're all familiar with," he added. What remains to be determined is whether this risk could be reduced with anticoagulation.

Their findings are published in the January 12 issue of the New England Journal of Medicine. They were first presented at the American Heart Association Scientific Sessions in November 2010, as reported by Medscape Medical News at that time.

Cryptogenic Stroke

A large number of strokes have no apparent cause, and about 25% of stroke patients have no advanced vascular disease, Dr. Connolly said. "Increasingly, in the last few years, people have suspected that these patients have atrial fibrillation, and some monitoring gets done to see if it can be diagnosed in these patients, but that has not really been very fruitful."

It is known that patients can have asymptomatic AF, and that pacemakers often detect these asymptomatic cases incidentally because the devices have a lead right in the left atrium, he said. The questions addressed in this study were how prevalent silent AF is, and whether it is actually causing strokes.

Patients were eligible for the study if they were 65 years of age or older, had a history of hypertension that required treatment, and had undergone first implantation either of a pacemaker of sinus-node or atrioventricular-node disease or of an implantable cardioverter defibrillator for any indication in the preceding 8 weeks. Exclusions included any history of AF or atrial flutter lasting more than 5 minutes, or treatment with a vitamin K antagonist for any reason.

A total of 2580 patients were enrolled and monitored for 3 months to detect subclinical atrial tachyarrhythmias, defined as episodes with an atrial rate of greater than 190 beats/minute for more than 6 minutes. They were then followed-up for a mean of 2.5 years for the primary outcome of ischemic stroke or systemic embolism. Patients with pacemakers were further randomly assigned to either receive continuous atrial overdrive pacing or not.

At 3 months, subclinical atrial tachyarrhythmias had been detected in 261 (10.1%) of the participants These episodes were associated with significantly increased risk for clinical AF and for ischemic stroke or systemic embolism, and remained predictive of stroke risk even after adjustment for stroke risk factors.

Table. Risk for Clinical AF and Ischemic Stroke With Subclinical Atrial Tachyarrhythmias

Outcome Hazard Ratio 95% Confidence Interval P Value
Clinical AF 5.56 3.78 - 8.17 < .001
Ischemic stroke or systemic embolism 2.49 1.28 - 4.85 .007
Ischemic stroke or systemic embolism after adjustment for stroke predictors 2.50 1.28 - 4.89 .008

"Of 51 patients who had a primary outcome event, 11 had had subclinical atrial tachyarrhythmias detected by 3 months," according to the authors, but none had had clinical AF during that period.

The population-attributable risk for stroke or systemic embolism with subclinical atrial tachyarrhythmias was 13%, the authors write. "Continuous atrial overdrive pacing did not prevent atrial fibrillation," they conclude.

Their findings are important for several reasons, Dr. Connolly said. "One is [that] there are a lot of patients with pacemakers out there, and we're detecting silent atrial fibrillation every day," he points out. There are approximately a million new devices implanted every year, and a prevalent pacemaker population of about 5 or 6 million people at any given time.

It has not been clear, however, what to do about this silent AF, because unlike clinical AF, the episodes are usually much shorter in duration and asymptomatic, and are only detectable by the device, he said. These new results, though, "don't mean that we should jump in and just start treating everybody," Dr. Connolly cautions. "Perhaps we should treat some patients, but probably we need to have more information, and perhaps even a clinical trial."

The other question the findings raise is how common silent AF is in the population of patients without pacemakers, and whether it may account for some proportion, at least, of cryptogenic strokes. Studies using implantable loop recorders in patients with risk factors for stroke, but no clinical AF, may help to answer that question.

"Pacemaker patients are not that different; they tend to share characteristics with a huge population of patients with vascular risk factors," Dr. Connolly notes. "They're elderly, hypertensive, diabetic, and the pacemaker is only incidental to their other disease."

One strategy already being undertaken is using implantable loop recorders in patients with a history of cryptogenic stroke to look for subclinical atrial tachyarrhythmias, he added. One such study, sponsored by Medtronic Inc and called the Study of Continuous Cardiac Monitoring to Assess Atrial Fibrillation After Cryptogenic Stroke (CRYSTAL-AF), began enrolment in 2009.

"So those are all important avenues that need to be explored, and will get explored," Dr. Connolly said. "I personally think that silent atrial fibrillation is probably fairly common in elderly patients with risk factors for stroke, and is probably responsible for some substantial burden of stroke. Exactly how much, that remains to be seen."

A Clinical Judgment

In an editorial accompanying the publication, Gervasio Lamas, MD, from the Columbia University Division of Cardiology at Mount Sinai in Miami Beach, Florida, calls a possible association between silent AF and risk stroke and systemic embolism an "intriguing" question, and writes that "[t]his robust, prospective, observational study leads the clinician to accept the association as true. Questions remain, however, about cause and effect, as well as about clinical significance."

Although it "strains logic" that a 6-minute episode of AF would cause a cardioembolic stroke, patients with brief episodes of AF are likely to have longer episodes that lead to cardioembolic events, he writes. To support causation, the AF should precede the stroke, and there should also be a "time-threshold effect, whereby a greater burden of atrial fibrillation or longer episodes of atrial fibrillation should confer a greater risk of stroke."

Alternatively, these short episodes may simply be a marker of stroke risk, perhaps signaling structural heart disease such as myocardial fibrosis or mitral valve disease, rather than causing thrombus in the left atrial appendage, he speculates. A proinflammatory state such as diabetes or metabolic syndrome could also be associated with brief AF episodes and with stroke; in this study, patients with CHADS2 scores of 2 or higher had a stroke event rate of almost 4% per year, he points out.

"The real question, of course, is to define a take-home message for the nonelectrophysiologist, like myself, who has a patient with an implanted pacemaker or defibrillator," Dr. Lamas writes. "But now we reach the limits of the present data."

Until clinical trials targeting this population with short, asymptomatic episodes of high atrial rate are done, there is no evidence for treatment with warfarin or other anticoagulants, he writes.

"For now, therefore, I will continue to turn to the now-venerable CHADS2 score, consider applying it to patients with asymptomatic episodes of atrial fibrillation lasting for hours, and make a clinical judgment about the need for anticoagulation," he concludes. "I will also wait for definitive studies to be performed in this interesting, at-risk population."

The study was supported by St. Jude Medical. Dr. Connolly reports receiving grant support and lecture fees from St. Jude Medical. Disclosures for Dr. Connolly's coauthors and for Dr. Lamas appear in the paper and editorial, respectively.

N Engl J Med. 2012;366:120-129, 178-180. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: