Share this article on:

With Extended Monitoring, More Detection of Atrial Fibrillation

Fitzgerald, Susan

doi: 10.1097/01.NT.0000453254.38002.53
Back to Top | Article Outline


Two different studies found that extended electrocardiographic monitoring periods turned up more cases of atrial fibrillation, and patients who received extended monitoring were more likely to be placed on anticoagulation therapy, which is proven to prevent recurrent stroke.

Longer-term electrocardiographic (ECG) monitoring of patients who had a stroke of unknown origin is much better for detecting previously undiagnosed atrial fibrillation than the 24-hour cardiac monitoring that is typically done, according to two new randomized clinical trials.

In a Canadian multicenter study — known as EMBRACE for 30-Day Cardiac Event Monitor Belt for Recording Atrial Fibrillation After a Cerebral Ischemic Event — 572 patients aged 55 or older who had a cryptogenic stroke or transient ischemic attack (TIA) within the previous six months and no known atrial fibrillation (after a standard diagnostic work-up including at least 24-hour Holter monitoring) were randomly assigned to be further evaluated for 30 days with an external event-triggered loop monitor or undergo an additional 24-hours of Holter monitoring. The trial was funded by the Canadian Stroke Network and involved 16 sites within the Canadian Stroke Consortium.

“We found that the prolonged monitoring intervention led to a five-fold increase in the detection of atrial fibrillation compared with the control group and significantly increased anticoagulant treatment rates — a finding we consider to be a clinically meaningful change in patient management that has the potential to avert recurrent strokes,” the study's lead author, David Gladstone, MD, PhD, a neurologist and associate professor in the University of Toronto department of medicine told Neurology Today.

The second trial — Cryptogenic Stroke and underlying Atrial Fibrillation for CRYSTAL AF — involved 441 patients with cryptogenic stroke or TIA who were enrolled within 90 days at 55 centers in Europe, Canada, and the US. The patients, who were at least 40 years old and had no signs of atrial fibrillation during initial evaluation, were assigned to either an insertable cardiac monitor (ICM) or conventional follow-up. The implantable monitor was made by Medtronic, which funded the study. [See “FINDINGS FROM EMBRACE AND CRYSTAL AF” only on and on the iPad for this issue.]



Taken together, these studies have implications for secondary stroke prevention practice in regions where prolonged cardiac rhythm monitoring is available.

Most previous guidelines had not recommended more than 24 hours of ECG monitoring after an ischemic stroke, and the ideal duration and type of monitoring — external versus implantable — had not been established, according to background information in the second study. The lead researcher Tommaso Sanna, MD, a cardiologist and professor at the Catholic University of the Sacred Heart in Rome, Italy, told Neurology Today that after seeing the results of his study, “I would not feel reassured after a negative 24-hours ECG monitoring.”

“I believe that long-term ECG monitoring should be included in the workup of patients with cryptogenic stroke, and the longer we monitor, the greater atrial fibrillation we will find,” Dr. Sanna said in an email.

An editorial accompanying the two studies in the June 26 edition of The New England Journal of Medicine likewise suggested that the findings should lead to a change in clinical practice. “Prolonged monitoring of heart rhythm should now become part of the standard of care of patients with cryptogenic stroke,” wrote Hooman Kamel, MD, an assistant professor of neurology and neuroscience at Weill Cornell Medical College.

Between one-quarter and one-third of ischemic strokes are labeled cryptogenic because no cause is identified. In some cases doctors may suspect atrial fibrillation even though it does not show up in short-term monitoring.

However, documentation of atrial fibrillation is usually required before initiating anticoagulation therapy, otherwise antiplatelet therapy is the usual recommendation for secondary stroke prevention in the absence of atrial fibrillation, according to current guidelines. In the case of clear-cut atrial fibrillation, anticoagulation is more effective than antiplatelet therapy in reducing the risk of recurrent stroke. Dr. Gladstone pointed out that the minimum clinically significant amount of paroxysmal atrial fibrillation that warrants anticoagulant therapy remains uncertain and that these trials were not powered to assess the effect of monitoring on stroke or death rates.

Researchers from both studies acknowledged that it was impossible to say whether the atrial fibrillation detected post-stroke was the cause of the stroke. “Atrial fibrillation may be unrelated to the index stroke event or may develop afterwards,” explained Dr. Gladstone, “but finding atrial fibrillation of sufficiently long duration, even if incidental, provides opportunities to anticoagulate for maximum risk reduction.”

Dr. Gladstone's team is now working on an economic analysis to determine the most cost-effective protocols. He said that cost-effectiveness will depend on appropriate patient selection such as targeting older patients who have a high pre-test probability of atrial fibrillation, as well as on the cost of devices, monitoring duration, and patient adherence.

Back to Top | Article Outline


Steven Messé, MD, FAAN, an associate professor of neurology at the Perelman School of Medicine at the University of Pennsylvania, said that based on previous smaller studies his center already tends to recommend that patients with cryptogenic stroke or TIA undergo four weeks of outpatient monitoring. He said the center is “increasingly recommending the use of implantable monitors in select patients, although the overall number of implanted devices remains small at this point.”

He said the two new studies provide important additional data for practitioners to consider, though he said there still are unanswered questions.

“The rate of atrial fibrillation in a matched population of patients without stroke remains uncertain,” Dr Messé told Neurology Today. Also, unanswered, he said, is “how much atrial fibrillation is required before anticoagulation is worth the increased cost, potential inconvenience, and higher bleeding risk compared to aspirin.”

Antonio Culebras, MD, FAAN — a professor of neurology at the State University of New York Upstate Medical Center and lead author of the AAN's recent guideline on prevention of stroke in nonvalvular atrial fibrillation — said that the review of evidence used to compile that document helped establish the principle that more monitoring yields more.

“Stroke units should formally organize the process of monitoring patients following a cryptogenic stroke,” he said. “My inclination is to use implanted recorders because I abide by the principle that the rate of identification depends on the duration of monitoring, and implanted recorders can stay in situ for as long as necessary.”

David Thaler, MD, PhD, the interim chairman of neurology at Tufts University School of Medicine, said his practice has evolved over the past five years to typically recommend that cryptogenic stroke patients wear a Holter monitor for 30 days. Still, he said it would be helpful to know more precisely which subsets of patients might benefit most from prolonged monitoring.

“You don't want to burden someone with a year's worth of cardiac monitoring if they are not likely to have atrial fibrillation,” he said, noting that the goal should always be to conduct a thorough evaluation up front to identify the cause of stroke.

Dr. Thaler also cautioned that while the two new studies found that longer monitoring turns up more atrial fibrillation, they weren't designed to prove whether the incidence of another stroke was reduced as a result.

Mitchell Elkind, MD, FAAN, a professor of neurology and epidemiology at Columbia University, said that there is still more to be learned about atrial fibrillation and its connection to stroke. “You don't have to be in atrial fibrillation at the time of your stroke for that stroke to be related to your atrial fibrillation.”

Dr. Elkind said he expects that the findings from the two latest studies will lead more physicians to utilize long-term ECG monitoring for cryptogenic stroke patients. “These data confirm the practice that many of us have already begun to follow, but I think it will work its way into guidelines,” he said.

Dr. Sanna said his group is planning a cost-effectiveness analysis. “I would be interested in estimating the cost per diagnosis of AF and to compare it with the cost of not making the diagnosis in terms of preventable recurrent stroke,” he said.

The Canadian researchers are also working on a health economics analysis based on their study results to help guide practice decisions, Dr. Gladstone said.



Back to Top | Article Outline


•. Sanna T, Diener H-C, Passman RS, et al. for the CRYSTAL AF Investigators. Cryptogenic stroke and underlying atrial fibrillation. N Engl J Med 2014; 370:2478–2486.
    •. Kamel H. Heart-rhythm monitoring for evaluation. N Engl J Med 2014; 370: 2532–2533.
      •. Gladstone DJ, Spring M, Dorian P, et al. Atrial fibrillation in patients with cryptogenic stroke. N Engl J Med 2014; 370:2467–2477.
        •. Neurology Today archive on atrial fibrillation:
          •. Neurology archive on atrial fibrillation:
            © 2014 American Academy of Neurology