Cardiac monitoring for detection of atrial fibrillation after TIA: An updated systematic review and meta-analysis

Abstract

Background and aims:

The optimal monitoring strategy for atrial fibrillation (AF) detection after transient ischemic attacks (TIAs) is uncertain, as existing evidence is largely derived from ischemic stroke or mixed stroke-TIA populations. This systematic review and meta-analysis aimed to determine the rate of newly diagnosed AF in TIA patients using different electrocardiographic monitoring modalities and durations.

Methods:

A literature search from 1966 through 31 May 2025 was conducted according to Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines using a protocol registered on PROSPERO (CRD420251075716). Prospective observational studies and randomized controlled trials including TIA patients who underwent at least 12 h of cardiac monitoring were eligible. The primary outcome was newly detected AF lasting at least 30 s, with subgroup analyses by monitoring duration and modality.

Results:

Forty-two studies enrolling 3981 TIA patients were included. The pooled AF detection rate was 6.5% (95% confidence interval (CI): 4.5–9.3). AF detection rates increased with monitoring duration: 3.5% at 1 day, 6.3% at 7 days, 9.6% at 30 days, 13.1% at 90 days, and 19.1% at 12 months, respectively. Implantable cardiac monitors yielded statistically significantly higher AF detection rates than non-invasive cardiac monitoring (20.8% vs. 4.7%, p < 0.0001). AF detection appeared higher in selected cohorts (based on older age, undetermined cause, and more extensive cardiovascular work-up) than in unselected cohorts (9.0% vs. 4.5%, p = 0.031). However, stratified analysis of non-invasive monitoring showed no difference between selected and unselected TIA patients (5.0% (95% CI: 2.3–10.5) vs. 4.5% (95% CI: 3.1–6.5)). Substantial heterogeneity was observed among studies (I² = 81.9%).

Conclusion:

This meta-analysis found high rates of newly detected AF in TIA patients undergoing prolonged cardiac monitoring, which are similar to rates previously reported after ischemic stroke. Studies using implantable cardiac monitors specifically in TIA patients are scarce. Randomized trials are needed for a face-to-face comparison of different monitoring strategies and to determine whether prolonged cardiac monitoring improves clinical outcomes in TIA patients.

Graphical abstract