High fitness might be associated with the development of new-onset atrial fibrillation in obese non-athletic adults

Background: Data on the association between AF and fitness are conflicting. Objectives: The aim of this analysis was to investigate the association between fitness, obesity and incidence of atrial fibrillation (AF) among apparently healthy non-athlete adults. Methods: We investigated 20 410 self-referred subjects who were annually screened in a tertiary medical centre. All subjects were free of AF and completed maximal exercise stress test according to the Bruce protocol at baseline. Fitness was categorised into age- and sex-specific quintiles (Q) according to the treadmill time. Subjects were categorised to low (Q1-Q2) and high fitness (Q3-5) groups. The primary end point was new-onset AF during follow-up. Results: Mean age was 48 ± 10 years and 72% were men. A total of 463 (2.3%) events occurred during an average follow-up of 8 ± 5 years corresponding to an AF event rate of 0.3% per person year. Univariate and multivariate models showed that AF risk was similar in both fitness groups. However, AF event rate was 0.55% per person year among high fitness obese subjects, compared with 0.31% for low fitness obese subjects (P <.01). Subgroup interaction analysis showed that AF risk is obesity-dependent, such that in the obese group (≥30 kg/m²) high fitness was independently associated with a significant 79% increased AF risk (95% CI 1.15-2.78; P =.01), whereas among non-obese subjects the rate of events was similar between both fitness groups (P for interaction = (.02)). Conclusions: Our findings suggest that high fitness might be associated with increased AF risk among obese subjects.