25th December, 2025, A/Prof Chee L Khoo
With increasing prevalence of obesity, type 2 diabetes, hypertension and obstructive sleep apnoea, we are encountering increasing prevalence of atrial fibrillation in our practice. These are the common risk factors which we associate with a higher risk of developing AF. Surprisingly, there is one group which stands out different from the expected groups listed above. This group is at the other end of the reverse J-curve – elite endurance athletes. Most endurance athletes do not have cardiometabolic risk factors. A recent Australian study explored the prevalence of AF amongst our elite rowers.
There has been previous reports of the association between endurance exercise and AF. Baldesberger et al. studied 62 former Tour De Suisse cyclists and compared them with 62 non-athletes (golfing enthusiasts) (1). At an average age of 66 years old, they found an AF prevalence of 10% in the former professional cyclists as compared with 0% among the golfers (P = .028).
In the recently published ProAFHeart study former elite rowers aged 45–80 years who competed at a national, world championship or Olympic level (2). A minimum of 10-year participation was required for enrolment. Between them, they won nine Olympic gold medals, ten silver medals and two bronze medals. A matched control group at a ratio of 1:100 was extracted from the UK Biobank (3). Participants were matched for gender, ethnicity, and an age within 5 years of the corresponding athlete age and randomly selected from a subset of 25,140 UK Biobank participants who had undertaken the imaging assessment which includes repeat medical history, electrocardiogram (ECG), dual-energy X-ray absorptiometry (DEXA), and cardiac magnetic resonance imaging (CMR) (4).
During follow-up, participants were considered to have
incident AF if it was self-reported, diagnosed as a hospital inpatient or
recorded in primary care databases with data censored at 4 years to match the
follow-up period of the athlete cohort. All participants completed a
questionnaire detailing the frequency, duration, and intensity of exercise
training during the years of international competitive rowing and after
retirement.
After a median follow up of 4.4 years, 121 former elite rowers were recruited for the trial. The mean age was 62 ± 9 years, 74% were male, and all were white ethnicity. Even in retirement, athletes performed more exercise, were taller, leaner, and had greater bone mineral density than controls. The athletes were less likely to have ever smoked compared with controls (25% vs. 38%), and no athletes were current smokers as compared with 4% of controls. Interestingly, athletes were more likely to drink at either extreme (seldom or frequently), whereas control subjects were more likely to drink intermittently. A lot of celebration was going on!
There were definite cardiac changes in the elite (or former) athletes. The cardiac volumes of the athletes as measured on CMR were significantly larger than those measured in the control group, but there was no difference in left ventricular ejection fraction. There were also significant differences in the ECG parameters between the two groups. The athletes had lower heart rates, longer PQ intervals, longer QRS durations, and longer QT intervals than the control participants.
Now, the results are not straightforward to dissect. Just as you may suspect, in the cohort of former world-class rowers with a median age of 62 years, the prevalence estimate of paroxysmal, sustained or chronic AF was between 8.1% – 21.5% compared with 3.2% in the control subjects (p< 0.001). This cohort included those who had AF at baseline. This prevalence is second only to mitral stenosis (5) and similar to that of hypertrophic cardiomyopathy (6), two conditions in which regular screening for AF is recommended. There was also a slight excess of strokes among the rowers as compared with the matched UK Biobank population. Perhaps, they miss some asymptomatic AF in the athletes.
Some rowers retired and performed less regular exercises after retirement while lifelong rowers continued regular exercises. Both groups performed similar amounts of exercises during their competitive years. Atrial fibrillation was more prevalent among retired athletes than lifelong athletes ((27.5% vs 9.8%; HR 2.8), whereas incident AF tended to be less [5.1% in retired athletes vs 8.1% in lifelong athletes; risk ratio 0.64). Obviously, there maybe a confounding effect here. Perhaps, retired athletes were forced into retirement due to the development of AF. Retired athletes in this study tended to be older and were males. Despite the confounders, it may be reasonable to conclude that the excess AF prevalence is not confined to those who continue to train and that a lifetime of intense exercise is not requisite to the development of AF. Just because you have retired before AF appeared, doesn’t mean AF will not develop when you retire.
The mechanisms underpinning the increase prevalence of AF is unclear. The usual risk factors of hypertension, smoking, and diabetes is less prevalent in this cohort of athletes. Greater alcohol intake has been associated with AF burden50 and interestingly athletes were more likely to drink most days than control subjects. Wow, there must be a lot of celebrations after events. On average, the cardiac volumes in the athletic cohort were 15%– 20% greater than controls. The athletes had lower heart rates and prolonged conduction and repolarization times. These are well described features of exercise-induced cardiac remodelling (‘athlete’s heart’),53
Is genetics part of the reason?
Polygenic risk score (PRS) is a means of quantifying the aggregated genetic risk derived from common genetic variants. By assigning a weighted score to single nucleotide polymorphisms associated with AF derived from whole genome studies in large populations, our study confirms that AF risk is partly determined by a complex interplay of genetic factors that are weak in isolation but moderately influential in combination. In this study, the overall yield of rare variants was low and did not differ between athletes with and without AF.
In summary, former elite endurance sport participation is associated with cardiac remodelling that can persist decades after retirement and is associated with a very high prevalence and incidence of AF. Apart from the traditional risk factors of hypertension, smoking and diabetes, a history of past elite and endurance sports participation is another risk factor we need to consider in screening for AF in our patients. We now have many options with the availability of modern personalised devices with efficacy in the detection of AF. Smart watches may be a prudent investment for older elite and endurance athletes.
References:
- Baldesberger S, Bauersfeld U, Candinas R, Seifert B, Zuber M, Ritter M, et al. Sinus node disease and arrhythmias in the long-term follow-up of former professional cyclists. Eur Heart J 2008;29:71–8
- Flannery MD, Canovas R, Janssens K, et al. Atrial fibrillation in former world-class rowers: role of environmental and genetic factors. Eur Heart J. 2025 Dec 15;46(47):5114-5125.
- Downey P, Peakman TC. Design and implementation of a high-throughput biological sample processing facility using modern manufacturing principles. Int J Epidemiol 2008; 37:i46–50
- Petersen SE, Matthews PM, Bamberg F, Bluemke DA, Francis JM, Friedrich MG, et al. Imaging in population science: cardiovascular magnetic resonance in 100,000 participants of UK Biobank—rationale, challenges and approaches. J Cardiovasc Magn Reson 2013;15:46
- Kim JY, Kim SH, Myong JP, Choi Y, Hwang YM, Kim TS, et al. Ten-year trends in the incidence, treatment and outcomes of patients with mitral stenosis in Korea. Heart 2020;106:746–50.
- Rowin EJ, Link MS, Maron MS, Maron BJ. Evolving contemporary management of atrial fibrillation in hypertrophic cardiomyopathy. Circulation 2023;148:1797–811
- Prior DL, La Gerche A. The athlete’s heart. Heart 2012;98:947–55