Background/aims:
Atrial fibrillation (AF) accounts for one-quarter of embolic ischaemic stroke. Use of long-term implantable cardiac monitors (ICM) can increase AF detection post-ischaemic stroke, however, the causality of brief episodes and management of AF detected remote to stroke remains controversial. This review aims to synthesise current evidence on patient selection, clinically relevant AF burden and anticoagulation decision-making.
Methods:
We conducted a targeted narrative review, due to heterogenous study designs and evolving evidence, using a comprehensive search strategy of MEDLINE (2020–2025). Controlled vocabulary (MeSH) and free-text keywords relevant to stroke, AF and ICMs were combined using Boolean operators. Records were exported into Covidence and reviewed independently by the authors; disagreements were resolved via consensus.
Results:
ICM increases detection of AF past 1-3 years of monitoring in cryptogenic stroke, but also in large artery occlusion and lacunar stroke. In non-stroke/TIA patients, screening for AF with ICM increases diagnosis however does not reduce stroke-risk. In high stroke-risk populations, anticoagulation of subclinical/device-detected AF with NOAC lowers stroke-risk while carrying an increased bleeding risk. Structural and serum cardiac biomarkers, including aggregates of cardiomyopathy and natriuretic peptides, may impact patient selection for ICM, although thresholds for clinical application are lacking.
Conclusion:
Judicious use of ICM is recommended, with patient-selection guided by neuroimaging and cardiac biomarkers, focusing on patients in whom detection would alter antithrombotic management. Consensus opinion on clinically relevant AF burden would assist decision-making. Dedicated randomised clinical trials evaluating stroke recurrence, bleeding-risk and overall clinical benefit, in stroke/TIA patients, are needed.