Atrial fibrillation variability on long‐term monitoring of implantable cardiac rhythm management devices

Background Atrial fibrillation (AF) burden and duration are predictors of thromboembolic events. The random nature of these measures may affect clinical decision making. The objective of this study was to determine temporal changes in AF burden as detected by continuous monitoring. Hypothesis AF burden changes over time when detected by continuous monitoring. Methods A post hoc analysis of patients enrolled in the TRENDS (A Prospective Study of the Clinical Significance of Atrial Arrhythmias Detected by Implanted Device Diagnostics) study with ≥1 stroke risk factor(s) who were implanted with a dual‐chamber cardiac rhythm management device (CRMD) and had AF burden data available for ≥2 years was performed. AF burden was defined as no AF, low AF (<5.5 hours on any given day), or high AF burden (≥5.5 hours in a day), and was first assessed over the initial 30 days following enrollment and then reassessed at 6‐month intervals for 2 years. Results Among 394 patients included, the average age was 70.2 ± 10.9 years, 71% were male, and mean CHA 2 DS 2‐ VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke or TIA, vascular disease, age 65–74 years, sex category) score was 3.7 ± 1.6. In the 30‐day baseline period, 75.1% of patients had no AF, 11.2% had low AF, and 13.7% had high AF. Over the subsequent 2 years, 40.0% of patients initially classified as no AF or low AF experienced periods with high AF, whereas 59.3% of patients initially classified as high AF experienced ≥6 consecutive months with no AF or low AF. Advanced age was the sole predictor of AF progression. Conclusions Significant temporal variability in AF burden exists when measured continuously with an implantable CRMD.