A randomized patient-oriented trial to investigate whether continual symptomatic use of the Kardia Mobile monitoring device is a superior strategy for diagnosing Atrial Fibrillation (AF) in comparison to normal ambulatory monitoring with a 24-hour Holter monitor. Patients with referrals to an Electrophysiologist, symptoms of AF but no documentation of AF, and with at least one stroke risk factor based on the CHADS2 scoring system are randomized 1:1 to Kardia Mobile monitoring versus Holter monitoring alone.

Wait-times from referral by a family practitioner to seeing a specialist have increased greatly over the past 10 years in Canada. The Fraser institute reports an 11 week wait-time in BC in 20161. Our experience is that this is representative of the wait-time to see a cardiologist at our institute.

Family doctors can diagnose some cases of atrial fibrillation (AF) with a pulse check and 12-lead electrocardiogram (ECG). However this is the minority of patients and most diagnoses are made on the basis of more prolonged ambulatory ECG recordings as part of the investigation of paroxysmal symptoms by a specialist2.

There is an increasing prevalence of AF3 and the consequences of late diagnosis can be dire. For example a 76 year old lady with treated hypertension and yet to be diagnosed paroxysmal AF has a risk of 6.7% per year of suffering stroke, TIA or systemic thromboembolus4. During the ~11 week wait for an appointment plus a further ~8 week wait for thorough diagnostic testing and further follow up she therefore has a 2.5% risk of having an event. If this is multiplied across the thousands of patients waiting for appointments across Canada over the course of a year it is evident that numerous thromboembolic complications could be prevented through reduction of the time to diagnosis.

Furthermore the first-line investigation for diagnosis of paroxysmal arrhythmia is usually an ambulatory ECG monitor (Holter/Event monitor)5,6. However, patients frequently do not have symptoms during the period of monitoring and the diagnostic yield is between 15% and 39%7-10.

Implantable loop recorders (ILRs) can provide accurate and thorough ECG monitoring continuously for up to 3 years. However, they are used late in the diagnostic process and are too expensive for widespread use for AF screening and so can only have a limited impact on overall time to diagnosis.

We therefore have two main problems in early AF diagnosis with paroxysmal symptoms. First that patients are waiting a long time for diagnostic tests and second that the diagnostic tools usually used by clinicians are not particularly good at picking up AF. We believe both issues can be tackled through use of the Karda Mobile in the wait-time prior to an appointment with a specialist.

The Kardia Mobile (AliveCor, San Francisco) is a validated, CE marked, single channel ECG recorder that can be paired with a smartphone. It has a high sensitivity (98%) and specificity (97%) for AF diagnosis11. This device allows intermittent patient-driven monitoring based on symptoms. Since smartphones are now owned by 68% of Canadians, most patients can now access this tool12.

We hypothesize that the Kardia Mobile could be an excellent tool for the diagnosis of AF in the wait-time prior to seeing a cardiologist. We aim to expand the role of our AF clinic at the Royal Jubilee Hospital, Victoria, BC by introducing enhanced wait- time arrhythmia diagnostics using the Kardia in this pilot study.

We will enrol patients at high risk for thromboembolus (score≥1 as per CHADS-65 CCS algorithm13) who are awaiting an appointment with a cardiologist in our locality for investigation of paroxysmal symptoms that may be caused by arrhythmia. These patients will either have standard care with a 24h Holter monitor or enhanced monitoring using the Kardia Mobile in addition to standard Holter monitoring.

We aim to test whether using the Kardia Mobile during the wait-time can reduce the time to diagnosis of AF

Inclusion Criteria:

• Age >18
• Referral for episodic symptoms that may be due to arrhythmia (e.g. palpitations, dyspnea, or pre-syncope)
• At least one risk factor from CHADS-65 CCS Algorithm

Exclusion Criteria:

• Previous diagnosis of atrial fibrillation
• Already anticoagulated for another diagnosis (e.g. metallic heart valve or pulmonary embolism)
• Symptoms typical of non-arrhythmic cause (e.g. exertional chest pain)