Efficacy and clinical implications of a stepwise screening strategy for atrial fibrillation after stroke: Insights from the SAFAS study.

Background: Although guidelines recommend screening patients for atrial fibrillation (AF) after stroke, the optimal timing and combination of screening tools remain unclear.

Aims: We evaluated the suitability of a sequential combination of screening techniques for AF detected after stroke (AFDAS). We also compared patient characteristics according to the timing of AFDAS.

Methods: Patients without previous AF admitted for acute ischaemic stroke were prospectively enrolled. After a stepwise screening approach for AFDAS based on electrocardiogram, telemetry monitoring and in-hospital long-term Holter, patients with cryptogenic stroke underwent implantation of an implantable cardiac monitor (ICM). Early AFDAS was defined as AF diagnosed during hospitalization and late AFDAS as AF diagnosed on an ICM.

Results: Of the 240 patients included, 104 (43.3%) had a documented cause of stroke not related to AF. Among the 136 remaining patients, AFDAS was detected in 82 (60.3%) during the acute screening phase or during the 3-year follow-up by ICM. Early AFDAS was diagnosed by ECG, telemetry and in-hospital long-term Holter monitoring in 17 (20.7%), 25 (30.5%) and 18 (22.0%) patients, respectively. Among 76 patients who had an ICM implanted for cryptogenic stroke, AF was detected in 22 patients (28.9%). Except for age and stroke location, patients with early AFDAS did not differ from those with late AFDAS, particularly with regard to the prevalence of atrial cardiomyopathy markers.

Conclusion: A stepwise approach to AFDAS screening allows early detection of AF in a considerable number of patients during their hospitalization. ICMs remain complementary to non-invasive screening tools for the detection of remote episodes of AF.