Endurance Training and Heart Health, Revisited

The perennial question of the benefits and risks of running has been back in the news in the past few weeks. First there was the recent publication of another paper adding to the previously reported findings from the Copenhagen heart study. The main conclusion from this long-term study of mortality among runners is that moderate amounts of running increase the probability of a longer life. However, the newspapers seized on the statement that large amounts of running were not statistically safer than a sedentary life-style. That in itself was a trivial conclusion despite its sensational appeal to newspaper editors. The number of people in the sample doing a large amount of exercise was too small to produce statistically robust evidence of either benefit or harm. While mortality rate was higher in those doing a lot of running compared with those doing a modest amount, it was nonetheless lower than in sedentary individuals, but the decrease was not statistically significant. I suspect that the somewhat sensational reporting was at least partly due to the fact that James O’Keefe joined the scientists who conducted the study, to write the paper. I have previously remarked that in my eyes, O’Keefe appears more like a snake-oil merchant than a scientist.

The other publication, a study of adverse cardiac events in over a million British women by Armstrong and colleagues from Oxford, is more measured in its reporting. It too shows that moderate amounts of exercise are beneficial, but those doing a larger amount of exercise had less good outcome than those doing a modest amount (e.g. half and hour three times per week). Nonetheless, even those exercising daily had a better outcome than sedentary individuals.

A third large epidemiological study, the Aerobic Longitudinal Study of 55,137 American adults also revealed that moderate exercise is associated with a substantial reduction in mortality, but yet again those doing a large amount of exercise tended to have higher mortality that those doing moderate exercise. Thus three large epidemiological studies have all demonstrated that moderate exercise is associated with major health benefits, but these benefits are reduced, though not entirely abolished, in those doing a large amount of exercise.  The evidence suggests that in at least some individuals a large amount of exercise is associated with harmful effects on health. This finding is not surprising in light of the very strong evidence from many studies that at least a minority of individuals who do very large amounts of exercise suffer heart damage.

Evidence of heart rhythm disturbances

The best documented adverse effect of extensive amount of endurance training and racing is disturbance of cardiac rhythm, especially atrial fibrillation. A review by Mont and colleagues revealed that long-term endurance athletes have a to 2-10 fold increase in risk of atrial fibrillation.   There is also an increased frequency of potentially more dangerous rhythm disturbances arising in the ventricles. Ventricular rhythm abnormalities are quite common in elderly endurance athletes, and also occur in a substantial minority of young athletes. For example, Verdile and colleagues observed ventricular rhythm abnormality in 367 (7.3%) of 5011 highly trained young athletes with average age 24 without other evidence of heart disease. Six of these individuals underwent successful surgical ablation of the aberrant heart tissue, while 7 with frequent or complex rhythm disturbances who declined surgery were prohibited from competitive sport.   However no adverse cardiac event occurred in any of the 367 young athletes during a follow-up period of average duration 7 years, indicating that at least in young athletes with no other evidence of cardiac abnormality, the arrhythmias are usually benign.

What are the implications for individuals who want to exercise vigorously?

Is there an upper limit to the amount of exercise that is healthy, and if so, what is it? Or can the likelihood of adverse effects on health be reduced by adjusting the way in which we train? There is a twist in the tail of the Oxford study of a million women that throws some light on this. A sub-group analysis revealed that among those who were obese, the women taking a large amount of exercise had a somewhat higher risk of cardiac events than those who exercised only three times a week. However, among those with BMI below 25, those who exercise frequently have a lower risk than those who exercise only three times week.   This suggests that it is not the amount of exercise in itself that does the damage, it is more likely that it is the amount of stress generated by the exercise that matters.    Emerging evidence about the mechanism by which excessive exercise might produce harmful health effects in runners throws a little more light on the issue.

What are the possible mechanisms of damage?

What determines who among endurance athletes is at greatest risk of damage? The mechanism of the damage remains uncertain, but a growing body of evidence provides some clues. Exercise remodels the heart.  The walls of the ventricles become thicker and the cavities become dilated. This is the typical athletes heart. Some athletes also exhibit fibrosis of the muscle. Fibrosis arises when damaged muscle is repaired with a scaffold of fibrous material.  This is a part of the normal mechanism by which inflammation repairs damaged body tissues but can become disruptive if the fibrous deposits become permanent. Fibrosis of heart muscle is likely to disrupt the normal conduction pathways via which electrical signals initiate heart muscle contraction. Although not directly proven, fibrous deposits are a prime suspect for rhythm disturbances.

What causes the damage that sets the scene for fibrosis? Some thought provoking clues come for studies of the effects of strenuous exercise on the right ventricle. The right ventricle has to pump blood through the lungs, and the capillaries in the lungs do not open up as much during exercise as the capillaries in the muscles. Hence, the right ventricle faces a relatively harder task than the left in having to push the increased volume of blood required with less benefit from an accommodating vascular system.  As a result there is demonstrable weakening of the right ventricle that persist for several days after very strenuous exercise. This weakening is associated with markers of transient heart muscle damage, such as increase in levels of cardiac enzymes in the blood stream. For a well trained athlete, the weakening is only appreciable after extremely strenuous exercise. For example, the weakening is only slight after a marathon, though more marked after an ironman, and still detectable a week later. For recreational runners, the damage can be appreciable after a marathon. But perhaps the crucial observation is that the amount of weakening appears to depends on how thoroughly the runners prepared for the marathon. In a study of runners in the Boston marathon in 2004 and 2005, Neilan and colleagues found that appreciable weakening of the right ventricle in those who had done less than 35 miles per week in the preceding moths, but no appreciable weakening in those who had done more than 45 miles per week.

What might convert transient damage into long term damage? In general when body tissues suffer transient damage the repair process includes the construction of a temporary framework of collagen fibres. If there is repeated trauma before full recovery there is greater risk that the temporary fibrous framework will be become permanent. Although there is little direct evidence that this happens in the heart, perhaps the most plausible explanation for the fibrosis observed in the heart muscle of endurance athletes is repeated trauma without opportunity for adequate recovery. Overall the evidence suggests simply that moderate exercise is beneficial for virtually everybody, but if you want to do a lot of exercise you need to build up gradually to avoid over-stressing the body.  Furthermore, it is plausible that demanding training or racing while incompletely recovered from previous strenuous training or racing creates an especially high risk of converting transient damage into long term fibrosis that might act as a precipitant of disturbance of heart rhythm.


Moderate exercise has major health benefits for virtually every one, but a large volume of endurance training diminishes the benefit for some individuals. There is no evidence indicating a fixed upper limit on the amount of exercise that is healthy.   On the other hand, the evidence suggests, but does not prove, that the risks of extensive training are likely to be low if you if you increase training load gradually and avoid demanding training or racing when inadequately recovered.

It is probably no coincidence that what appears to be the safest strategy for avoiding long term damage is similar to the widely accepted recommendation for training to improve performance: increase training load gradually and recover well after strenuous training or racing.

6 Responses to “Endurance Training and Heart Health, Revisited”

  1. Ewen Says:

    Thanks Canute – I’ll go with your conclusions. I’ve just read an article (referred to by Alex Hutchinson) about ‘more is better’! A study of 100,000 runners over 20 years – http://www.wsj.com/articles/SB10001424052748704350304574638550059084962

    I’m not sure that I’m doing “more” in comparison to Steven Abraham, who is cycling over 300k per day (at amazingly low HRs – but good speed!) trying to break the world record for distance cycled in a year. I wonder how healthy or unhealthy his heart will be after that?

    • canute1 Says:

      Thanks for the link to the WSJ article about the National Runners Health Study. On the whole the findings of that study are very encouraging. Unfortunately the various biases in this study of moderate committed runners who were prepared to compete the relevant questionnaires is likely to be even more serious than in other relevant epidemiological studies of exercise and health, but nonetheless, the finding of greater benefits form a greater amount of running (at least up to 80 Km/week on a wide range of health outcomes is re-assuring

  2. Robert Osfield Says:

    Great post Canute. The link between BMI and mortality in those who exercise are great deal is a particular stand out. I wonder if those with such a high BMI have an already stressed and unhealthy body that results in a compromised immune system that is less able to repair the heart.

    I suspect eating healthfully and sleeping well, and living without chronic stress could all be factors in avoiding the possible risks with high volume of exercise. I certainly notice slower recovery from training when my sleep is compromised and wouldn’t be surprised if my heart also was compromised just as much as my legs. One can easily tell when your legs are aching and needing more recovery time, while the heart just pretty well ticks along without complaining, but perhaps silently it’s also stressed and needing more recovery.

    • canute1 Says:


      Thanks for your comment. I agree that the observed decrease in benefit of a large amount of exercise (relative to moderate exercise) in obese women might not be due only to the greater stress resulting from the increase in energy-cost of running with increasing BMR, but possible reflects other aspects of life-style that affect health, including diet . Perhaps the conclusion should be extended to include the recommendation to combine ‘sensible’ training with a ‘sensible;’ lifestyle.

  3. The longevity of the long distance runner, part 3: Cardiac Outcomes | Canute's Efficient Running Site Says:

    […] ‘The athletes heart’; ‘Inflammation, heart-rhythms, training-effects and overtraining’; ‘Endurance training and heart health, revisited’) and will present a brief overview […]

  4. A contrast of two extraordinary marathon runners: Ed Whitlock and Gene Dykes | Canute's Efficient Running Site Says:

    […] there is some evidence that the damage can be alleviated by a careful build up.  Weakening of the right side of the heart after a marathon is less likely in those who have trained adequately.    That evidence is […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: