Rigorous and long-term cardiac monitoring strategy in cryptogenic stroke or TIA patients in a tertiary care centre results in a two-fold increase in AF detection

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): Special Research Fund (Bijzonder Onderzoeksfonds, BOF) Hasselt University. This study is part of Limburg Clinical Research Center, supported by the foundation Limburg Sterk Merk, province of Limburg, Flemish government, Hasselt University, Ziekenhuis Oost-Limburg and Jessa Hospital. Background Atrial fibrillation (AF) is a major cause of ischaemic stroke. Oral anticoagulation is recommended in stroke survivors with AF to prevent recurrence. Prolonged ECG monitoring using insertable cardiac monitors (ICMs) has been shown to increase the detection rate of AF compared to standard 24-hour ECG monitoring in cryptogenic stroke patients. However, prolonged ambulatory ECG monitoring is underutilized, likely contributing to an underdiagnosis of AF and missed anticoagulation treatment opportunities for secondary stroke prevention. Purpose This study aims to evaluate how our evolving cardiac monitoring strategy affected AF detection rates one year after cryptogenic ischaemic stroke or transient ischaemic attack (TIA) in a tertiary care centre. Methods We retrospectively identified all consecutive cryptogenic stroke or TIA patients admitted to our centre between 1/01/2017 – 1/01/2022. Patients with a pacemaker or implantable cardioverter-defibrillator were excluded from the analysis. Data were collected from the electronic medical record. Available cardiac monitoring modalities included 24-hour Holter monitoring, 7-day Holter monitoring, and insertable cardiac monitors (ICM). After October 2020 (i.e., period 2), the latter became part of our routine diagnostic workup for AF detection in case extended (7-day) Holter monitoring was negative. Results All 691 cryptogenic stroke or TIA patients admitted to our hospital during the inclusion period were considered. These were elderly patients (69.7 ± 13.2 years) with a CHA2DS2-VASc score of 3 [2 - 5]. Figure 1 shows evolving trends in the use of the different cardiac monitoring tools before and after October 2020. In particular, the use of 7-day Holter monitoring and long-term continuous monitoring with ICM increased after implementing our new diagnostic protocol compared to 24-hour Holter monitoring (p<.001). ICMs were inserted 95 [66 – 145] days after index stroke or TIA. The new diagnostic protocol detected more AF one year after the index stroke or TIA (3.7% vs. 10.1%, p<.001, Figure 2). In 95.6% of patients with AF detected within one year, oral anticoagulation was initiated. Future research should determine whether adopting a strategy of routine ICM following cryptogenic stroke and, thus, more appropriate anticoagulant therapy may improve clinical outcome in these patients. Conclusions The increased use of prolonged cardiac monitoring in cryptogenic stroke or TIA patients resulted in a two-fold increase in AF detection one year after stroke. Therefore, our results underscore the need to implement guideline-recommended prolonged rhythm monitoring using ICMs in addition to short-term ECG monitoring to ensure adequate secondary stroke prevention.