In the medical news last week came another interesting report on the relationship between physical activity and atrial fibrillation (AF). Reported in the medical journal, Heart, a team of investigators led by Nikola Drca from the Karolinska Institute in Stockholm gave us a study entitled “Atrial fibrillation is associated with different levels of physical activity levels at different ages in men.” This is the largest-ever study of the relationship between physical activity and AF, so the findings and conclusions deserve our attention here in the athlete community.
To set the stage for our discussion, recall that AF is a fairly common arrhythmia that is generated in the upper chambers of the heart–the left and right atrium. I’e written about AF here at the blog previously, in general terms for athletes. We’ve known for a long time that AF is associated with some other types of heart disease, particularly heart valve disease, and increases in prevalence as we age. The problem of AF is not benign. For athletes, the arrhythmia disrupts training or competition, but over the long term there is a small but real risk of stroke and also the chance of harm to the heart itself. Moreover, the available treatments–medications or ablation procedures–carry risk as well.
We also know that AF is associated with exercise. We know from prospective, longitudinal population studies that long-time, regular exercisers are more likely to develop AF over the years. And we know from targeted studies that athletes in the endurance sports are particularly at risk for developing AF.
Healthy endurance athletes should be asking the questions like:
- What are my chances of developing AF?
- How can I exercise safely and avoid developing AF?
Questions like these motivated the current study. Let’s take a look….
The Study
The purpose of the study was to determine if there was an association between physical activity level and the development of new AF in middle-aged men.
Back in 1997-1998, the investigators contacted all of the men aged 45-70 years who were residing in 2 counties in central Sweden and asked them to participate. About half (48,850 men, or 49%) responded by completing a questionnaire. From these, several thousand were excluded from further study because of: missing data on the questionnaire, death before the study’s follow-up period began, a current diagnosis of AF (1,496 men, or about 3% of the entire group), and the current diagnosis of some forms of cancer. This left 44,410 men to be part of the study. The average age of the participants at the time of enrollment was 60 years.
The questionnaire asked 2 questions about the amount of physical activity that the men engaged in. They were asked to think back and to recall their activity levels at age 15, at age 30, at age 50, and at the current (“baseline”) time. The participants younger than 50 years only answered for age 15 and for age 30 and “baseline.” The questions were:
“How many hours per week do you engage in leisure-time exercise (such as running, soccer, bicycling, swimming, floorball, gymnastics, cross-country skiing, etc.)?”
The respondents had to choose between:
- <1 hour
- 1 hour
- 2-3 hours
- 4-5 hours
- >5 hours.
“How much time each day do you spend walking or bicycling for everyday transportation purposes?”
The respondents had to choose between:
- <20 min per day
- 20-40 min per day
- 40-60 min per day
- >1 hour per day
In addition to the questions about physical activity, the respondents provided information about their medical history, smoking history, family medical history, medications, alcohol consumption, and level of education.
The respondents were then followed for 12 years to see what happened to them….and, in particular, to see if they developed AF.
The Results
At the end of the 12 years, there was an accumulated experience of 476,112 person-years! And during that time, there were 4,568 new cases of AF. Doing the math, that works out to 9.6 cases of AF per 1,000 person-years. Put another way, of the 44,410 participants, about 10.3% developed AF.
But the goal of the study was to determine if the individuals’ activity level was related to their chance of developing AF.
The primary finding of the study was that the self-reported leisure-time exercise at age 30 was indeed associated with the risk of developing AF. The investigators performed a statistical analysis designed to isolate the effect of the number of hours of exercise per week (and eliminating, as best possible, the effects of other influences on the risk of AF). They showed that the risk of developing AF was 19% greater among the individual who exercised >5 hours per week (at age 30) compared to those who exercised <1 hour per week. Remember that overall, 10.3% developed AF….so in terms of absolute risk difference between the extremes of exercise level reported at age 30, we’re talking about a couple percentage points different in the chances of developing AF. It’s because of such a large number of participants that differences this small can be detected.
Another important finding of the study was that the self-reported leisure-time exercise at baseline (when participants enrolled in the study, at mean age 60 years) was not related to the development of AF. And furthermore, the investigators found no relationship between the amount of self-reported leisure-time exercise at age 15 or at age 50 and the development of AF. The only such relationship was for the activity level at age 30.
The last important finding of the study came from a subgroup analysis. The participants who were at greatest risk of developing AF were those who reported >5 hours per week of leisure-time exercise at age 30….and <1 hour per week of leisure-time exercise at the time of enrollment in the study. These individuals were 49% more likely to develop AF during the 12-year follow-up period than individuals who exercised <1 hour per week at both age 30 and at the time of enrollment in the study.
There was no relationship between the amount of walking/cycling for transportation purposes at any age and the development of AF.
The Take-home Messages
- The current study adds to our knowledge about exercise and AF. I hope that more studies are to come.
- The relationship between exercise and AF is not completely understood and, indeed, the relationship may not be straightforward. Studies like this one ask participants to recall and note their activity levels at just a few moments during their lifetime. That can be hard to do, memory being what it is. Undoubtedly, the risk for developing AF must be related to some sort of dose of exercise over time. In retrospect, that’s hard to quantify. People exercise more some years than others. Some start exercising and some stop. They start and stop different types of exercise. I’ll bet that the intensity is important, too, yet how do we quantify this (in some straightforward way) so that analyses can be performed?
- Previous studies, both in large longitudinal cohorts as well smaller invetigations of particular endurance athletes, have shown an increased risk of AF over the long term among individuals who exercised a lot. This phenomenon has been shown best for individuals in young to middle age, and for men more so than women. This relationship has been shown in enough different populations that we should accept it as fact. The current study points out that the increased relative risk compared to non-exercisers may reflect a rather modest increase in absolute risk. But that’s at odds with previous studies that have shown as much as a several-fold increased risk for AF among heavy exercisers. But whatever the magnitude, that risk must be considered together with the other, well-established benefits of exercise over the long term when athletes are making decisions about their activity level. Don’t forget that exercisers live longer. That’s an important endpoint to keep in mind.
- The current study is curious in at least one respect. At first glance, it seems a bit odd that a relationship between exercise and development of AF could only be established for the amount of exercise reported at age 30…..and not at age 15, 50, or at the time of enrollment. Why is that? Could it be possible that we should exercise freely, with no worry about AF, while we’re young….and then again when we’re old? Maybe the 30-year-olds exercise with greater intensity. Maybe they accumulate more hours of exercise over more years. Maybe they engage in different forms of exercise that carry more risk. Perhaps “>5 hours” for the 30-year-olds was actually much more than 5 hours. The current study doesn’t provide answers. This needs to get sorted out with future studies. We need to better define the safe dose of exercise with respect to AF.
- Finally, why is exercise associated with AF? In truth, we don’t know in any detail. It seems that it must have something to do with the structure of the atrium that changes over years’ time with exercise. The investigators note several of the potential reasons: enlargement of the atrium, inflammatory changes in the atrium, and overdevelopment of the parasympathetic portion of the autonomic nervous system. Perhaps all of these play a role. I like Dr. John Mandrola’s blog post this week about this issue. I like his take.
Related Posts
1. Cyclist’s Account of Atrial Fibrillation