Cardiac implantable electronic devices (CIEDs) can be programmed to monitor and identify atrial high-rate episodes (AHREs), also described as “subclinical atrial fibrillation (SCAF),” which is defined as asymptomatic atrial fibrillation (AF), atrial tachycardia (AT), and atrial flutter (AFL) of short duration. Individuals may benefit from using CIEDs to detect AHREs as increasing evidence indicates that AHREs are associated with stroke risk. However, the expected stroke risk in patients with AHREs was much lower than that in clinically documented AF patients with similar risk profile.1)2)3)4)5) Most studies used an episode duration longer than 5–6 min to define AHREs,3)6)7)8)9)10) although a reported false-positive rate of 17.3%, making physician review more essential compared with AHREs lasting >6 hours, a false-positive rate of 3.3%.11) Current guidelines on non-valvular AF recommend anticoagulation therapy in patients with CHA₂DS₂-VASc scores ≥2 and consideration of anticoagulation therapy in patients with CHA₂DS₂-VASc scores of 1.12)13) Recently, the European Heart Rhythm Association suggested consideration of anticoagulation therapy for patients with AHREs >5.5 hours and with two additional stroke risk factors.14) However, the optimal primary stroke prevention indication and strategy remain uncertain.

Previous studies have reported various predictors of AHREs in patients with CIEDs. In the ASSERT study,1) lower resting heart rate and sinus node disease were associated with the prediction of AHREs. Other studies reported that a history of heart failure predicted AHREs,9) while others reported that a history of recurrent transient ischemic attack (TIA) and heart failure (HF),15) increased age, increased left atrium (LA) size, and lower systolic blood pressure predicted AHREs.16) A meta-analysis including 10 studies found that baseline characteristics; diabetes mellitus (DM); old age; thromboembolism events; congestive heart failure, hypertension (HTN), age ≥75 years, DM, stroke, CHADS₂ score; and coronary artery disease (CAD) were not associated with the prediction of AHREs.17) Therefore, AHREs predictors have not been well established and the optimal treatment for preventing stroke primarily in AHREs is uncertain. Thus, the aim of the present study was to prospectively investigate clinical predictors for AHREs and evaluate the clinical development of AF and cardiovascular events according to the presence of AHREs.

Between September 2017 and July 2020, this study enrolled 816 patients with a median follow-up of 18 months. The median patient age was 73 (65, 79) years and 60% were women. Atrioventricular block (58.1%) was the most common indication for CIEDs implantation, followed by sick sinus syndrome (41.9%). The median CHA₂DS₂-VASc score was 3 (2, 4); 677 (83.0%) of patients had scores ≥2 and 66.9% had hypertension. The median LA diameter was 40 mm (36, 45), the left ventricular ejection fraction (LVEF) was 65% (60, 70) by transthoracic echocardiography, and the PR interval was 184 ms (164, 204) by electrocardiography. Fifty-five (7.1%) showed atrial premature contraction (APC) at >1% per 24 hours. Overall, 358 (43.9%) patients were taking renin-angiotensin-aldosterone system inhibitors (Table 1).

This study observed a substantial incidence in AHREs among patients 19 years of age or older with a history of sick sinus syndrome or atrioventricular node disease who had undergone pacemaker implantation and were free of clinical AF. AHREs with a longest duration >6 minutes were detected in 12.5% of patients within 6 months after implantation and were detected at least once during a median follow-up period of 18 months in 13.7% of patients. Episodes of subclinical atrial tachyarrhythmias were almost five times as common as episodes of clinical AF. During the course of the study, AHREs detection was confirmed in all clinical AF patients. Clinical AF developed in 21.4% (24/112) of patients with longest AHREs duration >6 minutes and 48.9% (24/49) of patients with longest AHREs duration >6 hours, suggesting that there can be a gap between subclinical events and clinical detection. Larger LA size (>41 mm) and sick sinus syndrome for CIEDs implantation were significant predictors of longest AHREs duration >6 minutes. By means of longest AHREs duration >6 hours, prior history of stroke or TIA, larger LA size (>41 mm), and sick sinus syndrome were significant predictors. However, we did not demonstrate a meaningful relationship between AHREs and clinical events including CVA, bleeding, MI, and all-cause death.

Most studies of patients with CIEDs have shown a high incidence of CIEDs-detected AHREs.1)3)4)6)8)9)11)18)19) A growing body of clinical data supports the hypothesis that AHREs are associated with an elevated risk of developing further clinical AF and stroke; however, the stroke risk is relatively lower than that in patients with clinical AF due to the short and rare nature of AHREs.14) Data from ASSERT trial1) showed that AHREs lasting more than 6 min in the first 3 months after CIED implantation were associated with a 5.6-fold increased hazard of clinical AF during a mean 2.5-year follow-up. Similar results were reported in the MOST trial,3) in which AHREs lasting at least 5 minutes were associated with a 5.9-fold increased hazard of clinical AF during median 2.3-year follow-up. Clinical atrial tachyarrhythmias on surface ECG occurred in 25.7% of patients in the MOST trial, 15.8% of patients in the ASSERT trial, and 28.9% of patients in the ASSERT II trial, which investigated patients ≥65 years of age.1)3)20) In subsets of ASSERT patients, older age, higher BMI, and longer AHRE episodes were associated with AF progression.21) In our study, 21.4% and 48.9% of patients with AHREs >6 minutes and >6 hours were confirmed to have clinically documented AF during follow-up. As the number of patients was small, we could not proceed with further analysis. Although AHREs are more likely to allow detection of ECG-documented AF, about 80% of patients with AHREs >6 minutes never develop ECG-documented AF in the subsequent years, probably due to the infrequent and short nature of AHREs in most patients and missing link that we have not proven yet.

We detected four CVA events during the median 18-month follow-up. Our findings did not demonstrate an increased risk of CVA in patients with CIEDs developing AHREs. However, a meta-analysis of seven studies22) showed that AHREs were associated with a 2.4-fold increased risk of stroke. Although the definitions of AHREs duration varied significantly between studies, subsequent stratified analyses showed that most events occurred in patients with >24-hours AHREs.23) Therefore, the predominant evidence suggests a dose-response relationship between longest AHREs duration and stroke risk. Long-term follow-up is necessary to determine the relationship between AHREs and stroke risk in this registry.

The increased risk associated with developing AHREs may be similar to that for AF, reinforcing the concept that AHREs and AF are likely to represent a clinical continuum in the spectrum of atrial tachyarrhythmias. Some predictors for AHREs have been identified in patients with CIEDs. In the ASSERT trial,1) sinus node dysfunction and decreased resting heart rate were associated with the prediction of AHREs. Gonzales et al.9) reported that previous HF predicted AHREs, while Cheung et al.24) found that sinus node disease and a higher proportion of ventricular pacing were associated with AHREs. A meta-analysis of 28 dual-chamber device studies showed that the clinical variables related to AF including increasing age, HTN, DM, CAD, embolic events, and CHADS₂ were not associated with the prediction of AHREs. In our study, the independent predictors of AHREs with longest duration >6 minutes were larger LA size (>41 mm), and sick sinus syndrome before CIED implantation. Between patients who developed AHREs and those who did not, an LA size of 41 mm was a useful threshold to identify patients who would develop AHREs. Although total APC >1% on Holter monitoring showed predictive value for AHREs development in univariate analysis, this variable was not associated in the multivariate analysis. Prior stroke/TIA, LA diameter >41 mm, and sick sinus syndrome were the predictors for longest AHREs duration >6 hours. These differences may be explained in part by differences in the definitions of AHREs duration. Therefore, our data indicate new predictors including prior stroke/TIA and LA diameter >41 mm before CIEDs implantation.

This study has several limitations. First, although this was a prospective and multicenter study, the study design was non-randomized and observational. The study duration was relatively short compared to that in previous studies, the enrolled population was small, and additional potential confounding factors may have significantly affected the results. Second, the study population was limited to patients with pacemakers with generally normal LV systolic function. Therefore, the results cannot be directly applied to implantable cardioverter-defibrillator or cardiac resynchronization therapy populations, whose prevalence of CIEDs-detected AF may be higher. Third, although AHREs lasting >6 hours can improve the positive predictive value,11) current guidelines12) consider AHREs lasting >5 minutes as clinically relevant, and the majority of studies3)6)7)8)9)10)used an episode duration of AHRE longer than 5–6 minutes. Therefore, it may be reasonable to analyze the results based on AHREs lasting >6 minutes. Finally, very few clinical events were observed during the follow-up period, including four cases of stroke. Patients are being enrolled and the follow-up period is relatively short; therefore, analysis of long-term data is required.

AHREs occurred frequently in patients with pacemakers who had a history of prior stroke/TIA, LA diameter >41 mm, and sick sinus syndrome before CIEDs implantation but no prior diagnosis of clinical AF. However, there was a lack of data regarding whether AHREs increased the stroke risk. More long-term follow-up data are required to determine the clinical impact of AHREs.