Holter-electrocardiogram-monitoring in patients with acute ischaemic stroke (Find-AF_RANDOMISED): an open-label randomised controlled trial

Background: Atrial fibrillation is a major risk factor for recurrent ischaemic stroke, but often remains undiagnosed in patients who have had an acute ischaemic stroke. Enhanced and prolonged Holter-electrocardiogram-monitoring might increase detection of atrial fibrillation. We therefore investigated whether enhanced and prolonged rhythm monitoring was better for detection of atrial fibrillation than standard care procedures in patients with acute ischaemic stroke.

Methods: Find-AF_randomised is an open-label randomised study done at four centres in Germany. We recruited patients with acute ischaemic stroke (symptoms for 7 days or less) aged 60 years or older presenting with sinus rhythm and without history of atrial fibrillation. Patients were included irrespective of the suspected cause of stroke, unless they had a severe ipsilateral carotid or intracranial artery stenosis, which were the exclusion criteria. We used a computer-generated allocation sequence to randomly assign patients in a 1:1 ratio with permuted block sizes of 2, 4, 6, and 8, stratified by centre, to enhanced and prolonged monitoring (ie, 10-day Holter-electrocardiogram [ECG]-monitoring at baseline, and at 3 months and 6 months of follow-up) or standard care procedures (ie, at least 24 h of rhythm monitoring). Participants and study physicians were not masked to group assignment, but the expert committees that adjudicated endpoints were. The primary endpoint was the occurrence of atrial fibrillation or atrial flutter (30 sec or longer) within 6 months after randomisation and before stroke recurrence. Because Holter ECG is a widely used procedure and not known to harm patients, we chose not to assess safety in detail. Analysis was by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT01855035.

Findings: Between May 8, 2013, and Aug 31, 2014, we recruited 398 patients. 200 patients were randomly assigned to the enhanced and prolonged monitoring group and 198 to the standard care group. After 6 months, we detected atrial fibrillation in 14% of 200 patients in the enhanced and prolonged monitoring group (27 patients) versus 5% in the control group (nine of 198 patients, absolute difference 9·0%; 95% CI 3·4-14·5, p=0·002; number needed to screen 11).

Interpretation: Enhanced and prolonged monitoring initiated early in patients with acute ischaemic stroke aged 60 years or older was better than standard care for the detection of atrial fibrillation. These findings support the consideration of all patients aged 60 years or older with stroke for prolonged monitoring if the detection of atrial fibrillation would result in a change in medical management (eg, initiation of anticoagulation).

Funding: Boehringer Ingelheim.