Atrial fibrillation in former world-class rowers: role of environmental and genetic factors

Abstract

Background and aims: Endurance sport has been associated with an increased risk of atrial fibrillation (AF). The aim of this study was to assess the extent to which this is due to exercise burden or genetic predisposition.

Methods: Former rowers aged 45-80 years who competed at international championships were compared with a control group extracted from the UK Biobank, matched (1:100) for age and sex. Evaluation included 12-lead and Holter electrocardiograms, cardiac magnetic resonance imaging, and genetic analyses including rare variant evaluation and derivation of a validated AF polygenic risk score (AF-PRS).

Results: Of 121 rowers [age 62 years (interquartile range 54-69), 74% male], 26 (21.5%) had AF as compared with 364 of 11 495 control subjects (3.2%), prevalence risk ratio 6.8 [95% confidence interval (CI) 4.7-9.8]. Incident AF over 4-year follow-up was also greater [6 of 95 rowers (6.3%) vs 252 of 11 131 controls (2.3%), hazard ratio 2.8 (95% CI 1.6-5.0)]. Compared with controls, athletes demonstrated greater structural and electrophysiological cardiac remodelling. Athletes had similar cardiovascular risk factor profiles but a higher stroke prevalence than controls [3.3% vs 1.1%, risk ratio 3.0 (95% CI 1.1-7.9)]. Rare pathogenic and likely pathogenic variants in cardiomyopathy genes had low prevalence in athletes (2.7%) and were not enriched in those with AF. In contrast, in those subjects with a high AF-PRS (defined by the upper quartile in a healthy reference population) the odds of having AF increased 3.7-fold in athletes (95% CI 1.5-9.4) and 2.0-fold in controls (95% CI 1.7-2.4; P = .37 for between-group comparisons).

Conclusions: Despite having a favourable cardiovascular risk factor profile compared with controls, elite endurance athletes had a markedly higher prevalence and incidence of AF. These data suggest that exercise-induced cardiac remodelling and genetic susceptibility contribute to AF in endurance athletes.

Keywords: Arrhythmia; Athlete; Atrial fibrillation; Genomics; Polygenic.

Structured Graphical Abstract

The study compared 121 former elite rowers with 11495 age- and gender-matched subjects from the community (UK biobank). Structural and electrical remodelling was greater in the athletes. Both athletes and controls with atrial fibrillation (AF) were more likely to have a high AF polygenic risk score than those without AF (middle panels). Remarkably, athletes were 7-times more likely to have AF (bottom left). Incident AF over 4-years follow-up was also increased in the athlete cohort as compared with controls (bottom right).

Figure 1

Atrial fibrillation in former elite rowers vs a matched general population. Pie charts demonstrating a marked increase in prevalence of AF (21.5% vs 3.2%, P < .0001) in former elite rowers than in the large community-based UK Biobank population

Figure 2

Atrial fibrillation incidence in former elite rowers vs a matched general population. Survival analysis demonstrating a 2.8-fold greater incidence in AF among former elite rowers as compared with the UK Biobank population [hazard ratio 2.8 (95% CI 1.6–5.0)]

Figure 3

AF polygenic risk score in former elite rowers vs a matched general population. Violin plots demonstrating the distribution of AF polygenic risk scores in former eliter rowers and the UK Biobank population comparing those with and without prevalent AF. In both populations, a high AF-PRS (defined as the upper quartile scores from the healthy ASPREE cohort) was associated with a greater likelihood of AF (2.0 and 3.7 times more likely in the UK Biobank and rowers, respectively)