Physical activity across the spectrum: protective and proarrhythmic effects on atrial fibrillation.
C L Davies, S Pallikadavath, D S March, A Puranik, A Singh, G A Ng, G P MccannAbstract
Background
Regular physical activity (PA) reduces the risk of atrial fibrillation (AF) in the general population, however, endurance athletes appear to have an increased risk, termed the extreme exercise hypothesis. The dose at which the protective effect of exercise reverses remains unclear. Defining this threshold could guide personalised exercise recommendations, improve prevention and detection, and inform future sports cardiology guidelines. Previous meta-analyses have typically examined PA within the general population and athletes separately, preventing clear comparisons across varying levels of PA exposure. Additionally, several relevant studies have been published in recent years, underscoring the need for an updated meta-analysis.
Purpose
The primary aim was to assess the association between PA and AF in both the general population and athletes. The secondary aim was to quantify this relationship and identify a threshold at which AF risk increases.
Methods
Following PRISMA 2020 guidelines, four databases were systematically searched from January 1990 to June 2025. We included studies reporting the exposure as any form of PA from within the general population or athletes, with comparators being sedentary or non-athletic controls, and the outcome being the incidence or prevalence of AF. Two reviewers independently screened and assessed quality using a validated study quality assessment tool, with disagreements resolved by a third reviewer. Where possible, PA exposures were standardised to metabolic equivalent of task hours per week (MET-hr-wk) using the 2024 Compendium of Physical Activities. Pooled odds ratios (ORs) were calculated using a weighted random-effects model. A weighted restricted cubic spline model of log-odds was applied to produce a dose-response curve.
Results
37 studies (1,298,291 individuals) were eligible: 22 involving general population PA and 15 involving athletes. PA in the general population was associated with a 22% reduction in odds of AF, compared to sedentary behaviour (OR 0.78; 95% CI 0.68, 0.89; Figure 1.A). Athletes were associated with a 2.74-fold increase in odds of AF, compared to non-athletes (OR 2.74; 95% CI 1.72, 4.39; Figure 1.B). Dose-response analysis (26 studies; 1,022,790 individuals) showed PA in the range of 20-37 MET-hr-wk was associated with the greatest reduction in odds of AF. PA beyond 73 MET-hr-wk was significantly associated with an increase in odds of AF, the equivalent of 6-7 hours per week of high-intensity sport (Figure 2). Heterogeneity was high, and athlete studies were of poorer quality than general population PA studies.
Conclusion
This meta-analysis is the first to identify a potential quantitative threshold for increased risk of AF (73 MET-hr-wk, 6-7 hr-wk of sport) at which the protective effect of exercise may reverse. Higher quality studies, particularly in athletes, are required to confirm this finding and explore potential mechanisms.Forest plots of odds ratios for AFDose-response curve and odds ratio of AF