Atrial Fibrillation and Stroke May Be Temporally Related

CHICAGO—One-third of a large cohort of patients with an implantable cardiac device in place at the time of an ischemic stroke had one or more episodes of atrial fibrillation within the previous 30 days, Rhea C. Pimentel, MD, said at the 65th Annual Meeting of the American College of Cardiology.

The in-hospital mortality rate of these atrial fibrillation–related strokes was high: 11 of 42 (26%) patients with this event died during their stroke hospitalization, compared with six of 83 (7%) patients whose strokes were not temporally related to atrial fibrillation, said Dr. Pimentel, an electrophysiologist at the University of Kansas Medical Center in Kansas City.

Data from the Framingham Heart Study and other sources suggest that stroke in patients with atrial fibrillation entails about double the mortality rate of strokes in patients without atrial fibrillation. Mortality associated with atrial fibrillation–related stroke in the study was probably much higher because the hospital serves as a comprehensive stroke center and admits patients from across the Midwest, she said.

Dr. Pimentel reported data on 125 patients who presented with an ischemic stroke when a cardiac monitoring device was in place. This study is described as the largest patient series ever reported. Patients’ mean age was 73, and 41% were women. The mean CHADS2 score was 3.96 and the mean CHA2DS2-VASc score was 5.28. Of the patients, 62% had a pacemaker; the rest had an implantable cardioverter-defibrillator or cardiac resynchronization device. One-quarter of the group had a prior history of atrial fibrillation, and a fifth were on an oral anticoagulant—warfarin, in 70% of cases—at the time of their stroke.

Investigators defined a stroke-related atrial fibrillation episode as a total of at least one hour spent in atrial fibrillation at 30 days preceding the stroke. Eighty percent of affected patients had paroxysmal atrial fibrillation. They typically fulfilled the one-hour atrial fibrillation requirement with multiple short, self-terminated episodes rather than with an hour-long episode.

Being on an oral anticoagulant had no impact on in-hospital mortality rate, which was 14.2% in patients on warfarin or a newer anticoagulant and 14.3% in those who were not. Dr. Pimentel presented the results of the investigators’ initial look at the data. They are in the process of obtaining the patients’ international normalized ratio data, which “should be enlightening,” she said.

She and her coinvestigators also plan to subdivide their 30-day study period into five-day segments to learn how soon after an atrial fibrillation episode the strokes occurred. Researchers at Stanford University have reported that the greatest stroke risk in patients with atrial fibrillation occurs during the first five days after an atrial fibrillation episode. Dr. Pimentel’s group would like to confirm that observation.

In addition, because it remains an unresolved question whether any amount of atrial fibrillation is safe, Dr. Pimentel and her coworkers are considering reanalyzing their data using a cutoff of six minutes of atrial fibrillation rather than one hour during the 30 days prior to stroke.

—Bruce Jancin