Huston Embrace trial

Atrial Fibrillation in Patients with
Cryptogenic Stroke
Gladstone DJ et al. N Engl J Med 2014; 370:2467-2477
Presented by Kris Huston | July 21, 2014
30-Day Cardiac Event Monitor Belt for Recording Atrial Fibrillation
After a Cerebral Ischemic Event
Multicenter, 3-year study funded by the Canadian Stroke Network
Atrial fibrillation is a major cause of stroke worldwide.
Strokes due to atrial fibrillation are potentially preventable.
Paroxysmal atrial fibrillation often evades detection.
Many strokes have undetermined etiology (cryptogenic).
Cryptogenic stroke
Brain infarction that is not attributable to a source of definite
cardioembolism, large artery atherosclerosis, or small artery
Diagnosis of exclusion based on thorough investigation of other
potential etiologies.
Cryptogenic stroke accounts for 30 to 40 percent of ischemic
strokes (and half the TIAs).
Primary objective
To determine the diagnostic yield of a 30-day cardiac event
monitor compared to a repeat 24-hour Holter monitor for
detecting occult paroxysmal atrial fibrillation in patients with a
recent ischemic stroke or TIA of undetermined etiology.
Study population
• 55 years or older
• Experienced ischemic stroke or TIA within previous 6 months
• No cause identified by post-stroke/TIA diagnostic workup,
including ECG, 24-hr Holter, brain imaging, and echo.
• No history of atrial fibrillation or flutter.
Randomization, ECG Monitoring, and Follow-up
Baseline Characteristics of the Patients
Study design
Multicenter, two-arm, unblinded, randomized controlled trial
Sample size: N = 564
Participants were randomized (1:1) to receive either:
(1) a 30-day event-triggered loop recorder (intervention group) or
(2) a repeat 24-hour Holter monitor (control group)
Study design
Detection of one or more episodes of atrial fibrillation or atrial
flutter ≥30 seconds as assessed at the 90 day follow-up
1) Oral anticoagulation use at 90 days
2) Atrial fibrillation lasting 2.5 minutes or longer
Detection of Atrial Fibrillation in the Two Monitoring Groups
Incremental Yield of Prolonged ECG Monitoring for the Detection of Atrial Fibrillation
in Patients with Cryptogenic Stroke or TIA
Anticoagulant Therapy in the Two Monitoring Groups
Protocol Adherence
Intervention group: 233/284 (82%) completed 3 or more weeks of
Among patients in whom atrial fibrillation was not detected
before 30 days, 204/240 (85%) completed 3 or more weeks, and
148/240 (61.7%) completed 4 weeks.
• Among patients 55 years or older with a recent cryptogenic
stroke or TIA, paroxysmal atrial fibrillation was common.
• Noninvasive ambulatory ECG monitoring for a target of 30 days
significantly improved the detection of atrial fibrillation by a
factor of more than five and nearly doubled the rate of
anticoagulant treatment, as compared with the standard
practice of short-duration ECG monitoring.
Author-reported Limitations
• Total burden of atrial fibrillation per patient could not be
determined due to limited recording capacity of device
• Patients with large, severe strokes were underrepresented
• Monitoring started late (ave 75 days after index event),
reducing overall sensitivity of atrial fibrillation detection
• 30-day time limit and noninvasive monitoring diminished
diagnostic yield
• No comparison to age-matched controls without stroke or TIA
Critical Appraisal
Strengths of the study
• Large, multicenter, well-designed RCT
• Good adherence in intervention group
• Device manufacturers had no role in the trial
• Results have potentially important clinical implications
Critical Appraisal
Weaknesses of the study
• Lack of diversity in study population
• Did not address stroke recurrence and survival benefit
• Detection of atrial fibrillation is not the solution to cryptogenic
stroke; need to explore other stroke etiologies
• Benefit of anticoagulant therapy for subclinical AF not proven
• Optimal duration and type of cardiac monitoring still unclear
In this study, 441 patients with cryptogenic stroke and no
evidence of atrial fibrillation during at least 24 hours of ECG
monitoring were randomly assigned to prolonged ambulatory
cardiac monitoring with a subcutaneous implantable loop
recorder or to a control group with conventional follow-up.
At six months, atrial fibrillation detection was significantly higher
in the monitored group (8.9 percent, versus 1.4 percent in the
control group, hazard ratio 6.4, 95% CI 1.9-21.7).
Sanna T et al. Cryptogenic stroke and underlying atrial fibrillation. N Engl J Med 2014;
NEJM editorial
June 26, 2014
The results of two studies published in this issue of the Journal
indicate that prolonged monitoring of heart rhythm should now
become part of the standard care of patients with cryptogenic

similar documents