The timing and relationship of atrial fibrillation with exercise in veteran male endurance athletes

Abstract Background Endurance athletes exhibit an increased prevalence of atrial fibrillation (AF) which may be up to five times greater than the general population. Whilst retrospective studies have sought to determine whether previous exercise activity increases the risk of developing AF, there have been no long-term studies which have simultaneously comprehensively tracked exercise activity using wearable technology in combination with continuous cardiac monitoring in veteran athletes. Purpose We prospectively correlated the specific exercise training habits of veteran endurance athletes using computerised exercise tracking devices (CETDs) with the incidence and timing of AF on long-term implantable loop recorder (ILR) monitoring. Methods 106 male healthy endurance athletes (cyclists/triathletes) aged ≥ 50y who undertook ≥ 10h/week of exercise for ≥ 15y were recruited. Exclusion criteria consisted of pre-existing cardiovascular disease. Athletes underwent clinical assessment, 12-lead ECG, exercise testing and cardiac magnetic resonance (CMR) before receiving an ILR to detect the primary endpoint of six minutes of AF. We prospectively tracked exercise activity performed using CETDs consisting of cycling computers, global positioning system smartwatches, cycling power meters and chest-worn heart rate monitors via an automated online training diary. Results Sixteen (15.1%) athletes developed AF (median follow-up; 3.4 years (interquartile range (IQR) 3.2 - 3.5 years)). Of those, 15/16 (93.8%) athletes experienced recurrent AF episodes and 8/16 (50%) eventually developed symptoms. One athlete with AF suffered a stroke. The mean heart rate (HR) during AF was 76 beats per minute (BPM). The first episode of AF occurred nocturnally in 6/16 (37.5%). One athlete developed AF within one hour of exercise whilst the median time from exercise to onset of AF was 19.5 hours (9.3 – 45.3 hours). Athletes with AF were significantly older than those without AF (63.5 ± 5.6 vs 58.5 ± 5.4 years, P=0.001) (Table 1). Athletes who developed AF also had significantly greater systolic blood pressure (BP) (127.3 ± 10.7 vs 118.9 ± 10.3mmHg, P=0.004) and diastolic BP (78.0 ± 7.8 vs 73.7 ± 6.9mmHg, P=0.03) than athletes without AF. There were no significant differences in prospective exercise type, volume or intensity between athletes with and without AF (Figure 1). There were also no training differences in the month prior to the onset of the first AF episode compared with average training patterns over 2 years (Table 2). Conclusion In veteran male endurance athletes, prospective exercise training habits did not influence the incidence of AF. Furthermore, AF onset was seemingly unrelated to recent exercise exposure. Our findings suggest that lifetime competitive exercise exposure rather than recent training patterns contribute to AF onset in veteran male athletesTables 1 and 2For image description, please refer to the figure legend and surrounding text. Power & HR Zones in AF vs no AFFor image description, please refer to the figure legend and surrounding text.