Timing of Atrial Fibrillation Can Help Predict Stroke Risk

Daniel Singer, MD

New data from a case-crossover study of patients with atrial fibrillation (AF) and cardiac implantable electronic devices (CIEDs) is providing further insight into the associations between AF episodes and risk of stroke.

Among more than 450,000 patients included in the study, results suggest patients experiencing an arrhythmic state for 5.5 hours was associated with a more than 3-fold increase in odds of a stroke, with a 5-fold increase for stroke observed within the next 1-5 days after experiencing AF.

"Overall, these findings significantly underscore traditional thinking that AF is likely a causal risk factor for ischemic stroke as opposed to just a risk marker," said Daniel Singer, MD, professor of medicine at Harvard Medical School, professor in the Department of Epidemiology at the Harvard T.H. Chan School of Public Health, and Division of General Internal Medicine at Massachusetts General Hospital, in a statement. "The results indicate that prolonged episodes of AF increase stroke risk, but this risk decreases rapidly following the end of the episode. These findings raise the possibility that time-limited anticoagulation for infrequent episodes of AF may be an effective stroke prevention strategy."

With CIEDs improving monitoring abilities, Singer and a team of colleagues sought to assess temporal associations between episodes of AF and ischemic stroke. To do so, investigators designed their study as a case-crossover study leveraging data from within the Medtronic CareLink database from January 2007 to March 2017, which provided data related to 466,635 individuals with 120 days of continuous heart rhythm recording.

Overall, 891 patients identified for potential inclusion in the study experienced an ischemic stroke. This cohort had a median age of 76 (IQR, 67-82) years, 64.5% were male, 52.1% had a prior stroke or transient ischemic attack, and 47.8% had a clinical diagnosis of AF.

The exposure of interest for the study was defined as AF lasting 5.5 hours or more during days 1-30 vs days 91-120 prestroke. When assessing incidence of arrhythmic states, investigators found 92.6% had noninformative AF rhythm records. Of the 891, 682 (76.5%) had no periods of AF meeting the 5.5-hour threshold in either period, including 621 with no AF lasting 6 minutes or longer.

Among the 66 patients with informative, discordant arrhythmic states, 52 recorded AF of 5.5 hours or more in the case period compared to 14 in the control period (OR, 3.71 [95% CI, 2.06-6.70]). The greatest increase in stroke risk Amon these patients was observed 1-5 days following an AF episode. Investigators also pointed out AF episode greater than 23 hours on any given day was associated with the most apparent increase in stroke risk OR, 5.00 [95% CI, 2.08-12.01]).

"As of late, there has been significant discussion surrounding the temporal relationship of AF episodes to stroke, as well as the use of continuous cardiac monitoring, specifically with ICMs, to detect clinically meaningful AF episodes that may lead to a stroke," said Rob Kowal, MD, PhD, chief medical officer of the Cardiovascular Diagnostics and Services business, which is part of the Cardiovascular Portfolio at Medtronic, in the aforementioned statement. "This evidence highlights that patients are at highest risk for a stroke within five days of an AF episode that lasts five and a half hours or more. In addition to supporting the use of OACs in this patient population, this study also suggests the value of long-term cardiac monitoring to accurately detect these critical AF episodes."

This study, “Temporal Association Between Episodes of Atrial Fibrillation and Risk of Ischemic Stroke,” was published in JAMA Cardiology.