NYTimes article with a good summary of recent literature about prolonged endurance exercise and myocardial fibrosis (scarring/thickening = generally bad).
It is associated with number of marathon events, ultramarathon events, and amount of training. Pretty good variety of evidence in healthy humans and rats, which sheds light on mechanism and suggests (but doesn't prove) causation. One could make arguments that it's an adaptation that's not necessarily deleterious in this population, as it does not specifically look at outcomes (age at death). I'd also argue, weakly, that there might be some selection in the age-matched control population: healthy non-exercising folks survived to a healthy age with a larger genetic influence, whereas the exercising folks might have done more to help their own cause. And, I'd like to know what happens to young athletes that stop training regularly, and older athletes that picked it up later in life. Finally, you or I may or may not care ("I'd do it anyway"), but the interesting part of the rat study is that it was reversible within weeks. In that respect, I do care and think there will be some interesting lessons from future study.
My take? For any "Born to Run"-style oversimplification of regular distance running being "natural," I have to also believe humans regularly chilled out for a bit (see, e.g., cave drawings, brain and social development). Chasing down antelopes is good and all, but when a nomadic tribe finds a lush, fertile valley, only a fool would keep going. And remember, the notion that healthy humans regularly ran great distances completely ignores the ~50% of the adult population that happened to bear and nurse children for years at a time, *and* happen to be specifically good at endurance events!
So, if our goal is to regularly run distance events for the rest of our lives, and to maximize the amount of this, some evidence is pointing in the direction of taking some time off. How long, how frequent, and what the actual benefits might be, are still open questions.
The March 2011 issue of Trailrunner Mag has an interesting article: "Potential for atrial arrhythmias from long-term endurance training." This caused me to look further into current research on the subject. Atrial fibrillation has most certainly been associated with long-term endurance exercise in numerous studies; atrial fibrillation is most certainly associated with myocardial fibrosis in numerous studies (albeit most often in disease).
Based on some responses (including my own) here and mostly elsewhere (e.g. NYTimes comments), I wanted to make a stronger statement than my initial one: there is clear evidence that decades of endurance training most certainly can do damage to the heart. Period, end of sentence. As one paper put it:
A growing body of evidence reporting altered cardiac function and myocardial damage after arduous exercise, together with the increased prevalence of arrhythmias observed in highly trained athletes, suggests that repetitive bouts of prolonged, arduous exercise may be deleterious to long-term cardiac health. (Whyte, et al., Br J Sports Med 2008;42:304-305)
Results from studies like this are tough because the audience has a strong, vested opinion. The average person is much less likely to opine about studies involving mice or yeast, but dietary and exercise studies are much more accessible and applicable to everyone, yet complicated by individual genetic and epigenetic differences ("My grandpa smoked til he was 95").
With all that said, it may be prudent to remain aware of future developments and ways to mitigate these risks -- are there genetic susceptibilities? Dietary influences? Certain timing of training and de-training that can mitigate these risks? Can screening techniques catch abnormalities earlier, should different screening techniques be used for folks with different lifestyles?
Or it may be reasonable to accept the risks: we also know that a restricted calorie diet, with less exercise, is the best way to live the longest, but most of us are not interested in that, either. So the enjoyment may outweigh the risk, or the increased risk may be more than offset by decreased risk of other risk factors (in general) to an acceptable level.
But the main point is that there is a clear, growing, and damning body of evidence of heart damage in endurance athletes. Runners, cyclists, skiers. Post-mortem, and retrospective cohorts. Humans and rats. See for yourself if you agree or disagree.