# Atrial fibrillation detection in endurance athletes vs non-athletes: Positive predictive value and burden estimation using implantable cardiac monitors

## TL;DR

This study compares how well implantable cardiac monitors detect atrial fibrillation in athletes and non-athletes, finding similar accuracy and improved results with longer detected episodes.

## Contribution

The study provides new insights into the accuracy of ICMs for AF detection in athletes and how episode duration affects monitoring reliability.

## Key findings

- PPV of ICM-detected AF episodes was similar between athletes and non-athletes.
- Longer AF episodes significantly increased PPV, reaching 0.89 for episodes over 60 minutes.
- Automated AF burden estimation improved when excluding episodes shorter than 6 minutes.

## Abstract

Implantable cardiac monitors (ICMs) enable continuous arrhythmia detection and quantification of atrial fibrillation (AF) burden. However, the influence of an athletic lifestyle on the positive predictive value (PPV) of ICM-detected AF episodes, and the accuracy of automated AF burden estimation, remain uncertain.

This study aimed to compare the PPV of ICM-detected AF episodes between athletes and non-athletes with known AF, and to evaluate the accuracy of automatic AF burden estimation.

During a 4-week pre-randomization phase of 2 randomized trials, 3041 AF episodes lasting ≥30 seconds were detected by ICMs in 32 endurance athletes and 159 non-athletes with non-permanent AF, totaling 4202 monitored days. All episodes were manually adjudicated to determine true AF.

The patient-averaged PPV was 0.59 (95% confidence interval [CI] 0.42–0.74) for athletes and 0.51 (95% CI 0.45–0.58) for non-athletes (P = .24). PPV increased significantly with longer episode duration, reaching 0.89 (95% CI 0.83–0.94) for episodes >60 minutes vs 0.42 (95% CI 0.35–0.48) for episodes <6 minutes. Findings were consistent across groups. Correlation between automatic and validated AF burden was 0.67 (95% CI 0.56–0.77), improving to 0.85 (95% CI 0.80–0.92) after excluding episodes shorter than 6 minutes.

The PPV of ICM-detected AF episodes is similar in athletes and non-athletes. Longer AF episode duration enhances accuracy, and excluding short episodes improves automated AF burden estimation. ICMs may serve as good alternatives for AF monitoring also in athletic subpopulations.