Clin Res Cardiol (2022). https://doi.org/10.1007/s00392-022-02002-5
Pathological ECG findings in 1000 24-h-Holter-ECGs in patients with ischemic stroke A subanalysis of the Find-AF 2 trial
T. Uhe1, K. Wasser2, W.-R. Schäbitz3, M. Köhrmann4, J. Brachmann5, U. Laufs1, M. Dichgans6, G. Gelbrich7, K. Gröschel8, R. Wachter1, für die Studiengruppe: Find-AF 2 study group
1Klinik und Poliklinik für Kardiologie, Universitätsklinikum Leipzig, Leipzig; 2Universitätsmedizin Göttingen, Göttingen; 3Universitätsklinik für Neurologie, Evangelisches Klinikum Bethel, Bielefeld; 4Klinik für Neurologie, Universitätsklinikum Essen, Essen; 5Medical School / Regiomed GmbH, Coburg; 6Institute for Stroke and Dementia Research, München; 7Institut für Klinische Epidemiologie und Biometrie, Universitätsklinikum Würzburg, Würzburg; 8Klinik und Poliklinik für Neurologie, Universitätsmedizin Mainz, Mainz;

Introduction
Multi-day electrocardiogram (ECG)-monitoring in stroke patients is recommended by current guidelines to detect atrial fibrillation (AF) and other arrhythmias. The extent of rhythm monitoring varies in the daily clinical practice. Here, we assessed the rate of atrial fibrillation and other arrhythmias in the 24-hour screening- Holter- ECGs of the ongoing randomized multicenter trial Find-AF 2.

Methods and Results
Find-AF 2 (NCT04371055) is a randomized and controlled open-label parallel multicenter trial with blinded endpoint assessment. Patients ≥ 60 years with recent (≤ 30 days) ischemic strokes according to the AHA/ASA definition of any aetiology are screened for eligibility. All eligible patients receive a 24-hour Holter-ECG prior to randomization. Holter ECG data are analyzed by the ECG core laboratory using a dedicated analysis software (CardioDayTM, GETEMED, Teltow, Germany) and following a predefined standard operation procedure. In this analysis, we included the first consecutive 1000 24-h-Holter-ECGs.

We found new arrhythmias in 4 patients (4%). AF was detected in 15 patients (1.5 %) with a median duration of the longest episode of 702 minutes [interquartile range (IQR) 394; 1233] (see Figure 1). This led to the initiation of anticoagulation in 15 patients (100%). In 16 patients (1.6%), pauses > 2.5 s (mean 3.3s ± 0.6s; longest pause 4.8s) were diagnosed – seven due to higher-degree AV-blocks and nine due to sinus node arrest or sino-atrial block. Furthermore, we reported episodes with relevant bradycardias due to higher-degree AV-blocks in two patients (0.2%). This resulted in the implantation of pacemakers in five patients (28%). Non-sustaining ventricular arrhythmias (duration 14.6 s and 12.3 s) occurred in two patients (0.2%). Finally, we found regular SV-tachycardias in five patients (median duration 22 min [IQR 2; 102]) – four of them had focal atrial tachycardias and one an AV-node-Reentry-tachycardia. 

Conclusion
The rate of patients with relevant ECG pathologies in the first 1000 patients meeting the inclusion criteria of Find-AF 2 was 4 % and led to a therapeutic change in 50 % of patients. 

 
Figure 1: Duration of the longest atrial fibrillation (AF) episode in patients with newly diagnosed AF 
https://dgk.org/kongress_programme/jt2022/aP1926.html