Improving AF Detection in Cryptogenic Stroke (CSAF-AW)

A Non-invasive, Prolonged Surveillance Strategy Using Apple Watch for Atrial Fibrillation Detection in Cryptogenic Stroke

No cause for stroke is found in up to 30% of cases despite extensive investigations. These are called cryptogenic strokes (CS). 1 in 4 stroke survivors will suffer another in 5 years and this is a leading cause of fear and anxiety. A common reason for CS is an undetected heart rhythm disorder called atrial fibrillation (AF). AF occurs intermittently, so it may not be detected during the mandated 24-96 hours of rhythm monitoring that is performed as part of the standard post-stroke investigation strategy. A randomised controlled study in 2014 showed that whereas this standard monitoring strategy picks up AF in 2% of CS patients, longer-term, continuous monitoring for 12 months can pick up AF in 13% of patients. This suggests the standard strategy may miss AF in a proportion of CS patients and thus also the opportunity to mitigate against further strokes with anticoagulation therapy. Prolonged monitoring has traditionally required a minimally-invasive surgical procedure to implant a recording device under the skin at a specialist centre. A specifically trained team is also required to interpret the large number of recordings this strategy yields. The Apple Watch (AW) is a wristwatch able to monitor a wearer's heart rate and rhythm regularity and facilitates real- time, single-lead ECG recordings. This over-the-counter, non-invasive device has demonstrated feasibility and has a Certification (CE) Mark for detecting AF. It may offer a potential non-invasive alternative long-term rhythm surveillance strategy to diagnose AF in these patients. The investigators propose a study in which CS patients will be randomised in a 1:1 ratio to receive additional AW monitoring on top of standard care versus standard care alone. The investigators shall then explore the incidence of AF in the two groups at 1 year and how it impacts clinical outcomes too.

The investigators propose a prospective, randomised controlled study to determine our hypothesis. This single-centre, investigator-initiated trial will be performed at Barts Health NHS Trust which provides tertiary cardiac electrophysiology services to 4 stroke units part of the London Stroke Care Network (SCN). Participants will be randomised on day 0 to either the active or control study arm. Participants randomised to the active arm will be provided with an AW Series 4 on Day 0 for the duration of the study. All participants will be followed up for 12 months after enrolment.

Participants randomised to the active arm will be provided with an AW Series 4 on Day 0 for the duration of the study. They shall undergo an education and training session to ensure technical competency of heart rhythm recording and familiarity with the recommended recording schedule for the duration of the study. Participants will also be given the contact details for a dedicated email mailbox for the duration of the study through which they can submit remote transmissions of ECG data.

Participants in the control arm will be advised to continue with the standard of care and advised to contact their direct clinical team or primary care physician should they experience any symptoms of concern (palpitations, dizziness, collapse). They will be contacted by the study team at 6 and 12 months for clinical assessment (symptoms, hospitalisation data, further stroke events, mortality).

For the duration of the study, participants in the active arm will be asked to make ECG recordings in three circumstances: Routinely once per day from the day of randomisation till the 12-month visit after AF Catheter Ablation (they will be encouraged to do so at the same time each day with the option to set a reminder on the watch). If the AW requests them to do so (the automatic AF-notification algorithm will be active throughout the study period and if AF is suspected based on the tachogram, it will trigger notification for the participant to make an ECG recording). This will be till the 12-month time point. If the patient has symptoms of palpitations, pre-syncope or syncope. This will be till the 12-month time point. Monitoring should continue even if AF has been diagnosed to characterise arrhythmia burden and profile. Participants will be given the contact details of a dedicated email inbox to which to send ECGs recorded through the AW.

Inclusion Criteria: Recent stroke or TIA event (<90 days) that is supported by consistency between symptoms and findings on brain magnetic resonance imaging (MRI) or computed tomography (CT). Patients with TIA will be enrolled only if symptoms at presentation were speech problems, limb weakness, or hemianopsia. Diagnosis shall be made by a Consultant Stroke physician. Stroke classified as cryptogenic following extensive investigations to rule out a cause as per local protocol, but to include at least: Trans-thoracic echocardiogram A period of at least 24-hour rhythm monitoring with no evidence of AF. MRI or CT-angiography of the brain Sufficient mobility and dexterity to perform an ECG recording on the AW by touching the dial on the watch when worn with the contralateral hand for 30 seconds. Access to a smartphone with Apple-operating system (OS) within their household for syncing with their watch. Participants must be able and willing to provide written informed consent to participate in the study. Exclusion Criteria: Documented history of AF or atrial flutter. Expected to undergo heart surgery in the 1 year following the stroke event. Any episode of myocardial infarction, coronary intervention or coronary artery bypass grafting <1 month prior to the stroke event Any other indication for oral anti-coagulation. Patient has a cardiac implantable electronic device in-situ already or is expected to undergo implantation in the 1 year following the stroke event (excluding implantable loop recorder for the investigation of the CS). Previous left atrial (LA) ablation or LA surgery Life expectancy less than 1 year