Article Text

55 The Prevalence of Undiagnosed Atrial Fibrillation in Non-anticoagulated Heart Failure Patients Using Remotely Collected Data
  1. Richard Till,
  2. Martin Cowie
  1. Imperial College London/Royal Brompton Hospital


Purpose Treatment with oral anticoagulant therapy (OAT) reduces the risk of stroke in patients with AF. Our aim was to determine the prevalence of undiagnosed paroxysmal AF using remotely collected data from non-anticoagulated HF patients with implanted cardiac devices, and to estimate the risk of stroke and bleeding in such individuals using standard scores.

Methods Medtronic CareLink was used to remotely collect data on heart rhythm from patients registered at a single UK centre with a suitable dual chamber HF implantable electronic device. Demographic and clinical datawere obtained from paper and electronic medical records, including previous documentation of AF, use of OAT, and the data required to calculate stroke (CHADS2 and CHADS2VASC) and bleeding risk (HASBLED) scores.

Results Data from 138 patients (80% male; mean age 67 (±12) years) were analysed. 5 patients were in permanent AF at implant. Mean follow-up was 34 (±19) months. 94 (68%) had AF during follow-up of whom 57 (61%) were receiving OAT (52 for AF, 5 for other indications). 37 patients with AF during follow-up were not on OAT, 3 of whom had a previous diagnosis of AF; OAT was considered contraindicated in 2 and no reason given for lack of OAT in 1. Thus 34 patients (25% of whole group) had previously unreported device detected AF but were not receiving OAT (Figure 1). The mean follow-up for this group was 32 (±17) months. The median total AF duration was only 24 seconds (IQR 5 seconds -2 mins 57 seconds). 7 (5% of all patients) had a total AF burden > 10 mins (Table 1). For the 34 patients with AF not on OAT, the median CHADS2 score was 2 (IQR 1–3) and CHADS2VASc 3 (IQR 2.25–5). The median HASBLED score was 2 (IQR 1–2).

Abstract 55 Table 1

Conclusions 25% of HF patients with cardiac implantable electronic devices have undiagnosed AF and risk factors that would warrant consideration of treatment with OAT with a low bleeding risk. However, most of these patients only had a very low AF burden. 5% of patients had more prolonged undiagnosed AF and were not receiving OAT. Further work is required to quantify the risk associated with very short duration AF in heart failure.

  • heart failure
  • atrial fibrillation
  • remote monitoring

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.