The Haywire Heart – How too much exercise can kill you and what you can do to protect your heart – Review


There has been much written in the popular press over the past decade or so about the downsides of training for endurance sports particularly as it relates to the so-called “athletes heart”. In many of these articles (and as usual) the claims by medical “scientists” are often misrepresented, misinterpreted, and selectively presented to inflate any findings beyond even the typically overstated and often unsupported conclusions of the authors in contemporary medical journals including results from clinical trials. This represents a severe dis-service to the general population and specifically to those who choose to train for endurance sports.

The authors of “The Haywire Heart” dance the line between the constructive, by providing a useful source of guidance on the subject of arrhythmias and endurance sport, and the deconstructive, by adding further, in my opinion, to informational dis-service. I think they have stepped a bit too far on the side of overblown alarm and not presented a balanced, statistically framed, and critical context in which to absorb the significant body of research outlined and referenced in the book. The subtitle of the book is a particularly egregious alarmist and unsupported statement.

Of course, readers today gravitate toward sensationalism at the expense of accuracy and precision (it is not obvious why this is the case, other than the continued failure of our educational systems to address critical thinking skills, particularly in the US). As a result, if an author expects to reach any sizable non-fiction audience, nothing less than a muted sensationalism needs to be a central theme. In this, the authors of this book succeed. In providing a balanced, statistically accurate evaluation of the potential for the development of arrhythmias in endurance athletes, they fail.

The authors make a good start at a balanced approach in the introduction, but thereafter obvious bias and sensationalism creeps in and incrementally erodes the foundations of the “science” being presented. This process ends with a ridiculous chapter on “supplements”. Too bad because all of the elements of a very good guide to the subject are here and without the hype and misrepresentation it would represent a true contribution.

On the positive, the book will provide the reader with a reasonable description of the workings of the heart, the observed arrhythmias and various treatment options, and how the skimpy evidence on the effects of endurance sport training on the development of arrhythmias might be interpreted. The book is a worthwhile read provided one takes a critical stance on many of the conclusive statements throughout the book by pursuing investigation of appropriate referenced sources and applying logic, questioning “fact” , and using reason. Additionally, there is good coverage of what to be aware of as it concerns the symptoms that are typically observed in those endurance athletes who have developed arrhythmias- so if you are one of the very few who might be at risk you will have some idea of what sort of heart episodes to be sensitive to. In no way does the meager evidence for the incidence of arrhythmias that appear to be correlated with endurance sport exercise lead to any significant general concern for athletes training for endurance sports.

Although I am by no means an expert in the field of electrophysiology, cardiac arrythmias, or cardiac clinical research, I did spend 10 years of my scientific career on the device side developing new therapies and devices for the treatment arrhythmias. I have been intimately involved with development of state-of-the-art implantable pacemaker, defibrillator, and cardiac re-synchronization therapy (CRT) devices. In this work I interacted with Electrophysiologists, clinical researchers, and device experts to help provide the most efficacious treatments possible at the time. My background in the field is substantial, however I would not describe my level of understanding as “expert”.

As a research scientist I feel qualified to evaluate the scientific evidence presented and critically review any conclusive arguments and extensions to recommendations for endurance athletes. Additionally, I am a life-long endurance athlete who knows, trains with, and has raced against National-level and Olympic-level competitors for over 40 years- both when we were all “pink-lunged” youngsters and now as semi-decrepit Masters. In fact, many years back, as  budding road cyclist and graduate student doing research at Los Alamos, I would occasionally train with one of the authors (Lennard Zinn) up into the Jemez Mountains and around the Via Grande just above town. It was here that I first discovered my abilities as a climber and I am forever grateful for the “calibration” that Lennard supplied- he is a truly talented cyclist.

statistical context

Perhaps one the largest failings of this book is the complete absence of a summary of the observed incidence of arrhythmias in the general population and the common correlates and concomitant diseases that are associated with arrhythmias. As a result the reader is left to wonder what the expected occurrence rate of arrhythmias is in the general population and how this is related to age and other associated conditions.

Numerous epidemiological studies have been conducted on various arrythmias and particularly for atrial fibrillation since AF is the most common arrhythmia condition. The primary identified factor for the development of AF is age- the older you are the more likely you will develop AF. A recent study of AF in a European population showed an overall incidence of about 2% but this rate is highly dependent on age. The incidence at ages less than 49 years is about 0.14% but increases dramatically to about 4% at ages 60-70, and to about 14% at ages greater than 80. Additional studies of other populations typically yield slightly different incidence rates that are within error estimates. Other concomitant conditions include hypertension, obesity, diabetes, coronary artery disease, valvular heart disease, and cardiomyopathy.

With this type of statistical basis for the incidence, correlates, and concomitant conditions for arrhythmias, one is in a position to properly evaluate any increased incidence observed in endurance athlete populations. However, the book never reviews nor even supplies this information- which is very disappointing.

I have listened to a few of the podcasts that the authors have participated in and in those podcasts with Mandrola present it is stated by him, up front, that the condition is rare and any discussion should be cognizant of this fact. However, this never comes through in the book. Also in the podcasts, the significant uncertainty in making any concluding statements about the relationship between endurance exercise and the development of arrhythmias is highlighted. However, later in the podcasts, statements that are inconsistent with this are rampant.

the “athlete’s heart”

Fundamental to the all of the evidence and discussion on the development of arrhythmias in endurance athletes is a condition called the “Athletes Heart”. “Athlete’s Heart” is a syndrome described, in simple terms, by an enlarged heart and a low resting heart rate. Cardiac adaptation to endurance exercise and training leads to hypertrophy of (primarily) the left ventricle, dilated chambers of the heart, an associated increased stoke volume, and a generally attendant lower resting heart rate. These effects can be (and often are) misdiagnosed by uninformed (and often poorly educated) medical professionals as other conditions not typically found in the general population, e.g. cardiomeagly (enlarged heart), cardiac hypertrophy (thickening of the left ventricle), and bradycardia (low heart rate). Repeated loading of the heart muscle in ways typified by endurance sport training (e.g. long (>60 min) exercise  sessions and high intensity intervals) will naturally lead to cardiac remodeling that includes all of these conditions. One rare potential outcome of this remodeling is the (hypothesized) development of electrical malfunction within the heart which can be manifest as various types of arrhythmias and is the subject of the book.

It has been hypothesized that arrhythmias can be the result of endurance exercise via a mechanism that includes a process by which the heart muscle is taxed in such a way that it “stretches” and leads to small tears that are naturally healed by laying down collagen in the affected areas of the heart. This healing leads to something called fibrosis (aka “scaring”) and it is proposed that fibrosis can eventually lead to electrical discontinuities within the heart and manifest as an arrhythmia. There is no mechanistic understanding of the relation between cardiac fibrosis and electrical malfunction- it’s just an idea at this point. However, whilst reading this book you might be lead to believe that such a mechanism is fact.

alcohol use and atrial fibrillation

Although not central to the concerns about AF and endurance exercise, the following represents an example of selectively presented mis-information by the authors.

The authors present a graph (Fig 5.1) on page 130 that relates “chance” of atrial fibrillation with number of alcoholic drinks consumed per day, referencing a recent study  that has shown statistical evidence for increased risk of atrial fibrillation/atrial flutter for those who consume even moderate amounts (1-6 drinks per week!) of alcohol. Firstly, the graph is entirely misleading because the authors relate  a quantifiable variable (number of alcoholic drinks per day) with a qualitative variable they call “chance” of AF. Even though the referenced study presents a full analysis of the quantifiable data for the “chance” variable, the authors chose not to plot this quantitative information. This is a common method utilized by sensationalists to obsequiously over-state a relationship. This happens because a graph such as this shows, in this case, a doubling of the “chance” (or risk) for the ailment (in this case AF) per daily drink of alcohol but gives no information on the underlying risk magnitude. The underlying risk can be very low- as it is here. In addition the actual data on “chance” of AF used in the graphic is in units of risk ratio, i.e. total risk with alcohol consumption divided by total underlying risk. For there to be doubling of the risk this ratio would need to be equal to 2. It isn’t and this is detailed below.

If one analyzes the data presented in the referenced study it is found that the risk for the no alcoholic drink population is about 2% and that this risk is essentially unchanged for those who have about one drink per day, increases by about 10% (a risk ratio of 1.1) for those who have 2 drinks per day, and by about 20% (a risk ratio of 1.2) for those who drink 3 drinks per day. What this means is that the risk increases from 2% to 2.2% for those that drink 2 drinks per day. So although the obfuscating graph (and accompanying text) purports to show a doubling of the risk, it is not a doubling, and the magnitude of the increased risk, at least as far as I am concerned, is inconsequential. Make your own determination but, authors: please provide the data when it is available instead of confusing and obfuscating what is observed. The medical literature is rife with such drivel (see for example the papers by Ioannidis referenced at the beginning of this post).

The authors may have confused increased risk with increased incremental risk- which is a very different thing and one that the general reader would have limited interest in. Even if they meant to include the word “incremental” in front of “risk” (where appropriate) in this section, the data are inaccurately presented and is inexcusable in a book intended for a general audience. The relevant value is the increased risk- something that the reader can best understand and personalize.

For clarity a plot of the the risk ratio versus the number of alcoholic drinks per day based on data from the referenced study is presented below:

Risk ratio for the development of AF as a function of alcohol consumption in the general population based on data from a study referenced in the book “The Haywire Heart”.

A side note here that is good news to those who enjoy beer- the authors of the referenced study concluded the following as it concerns specific alcoholic drinks:

For specific alcoholic beverages, consumption of more than 14
drinks/week of liquor or wine was associated with
increased risk of AF (Table 1). There was no association
with beer.

It is not clear what to make of this observation other than this result highlights how little is known about the subject and any mechanistic cause that might increase the risk of developing AF. Since life stress is thought to be a contributing factor for AF, are wine drinkers more stressed out than laid back beer drinkers? Who knows… but the absorption of alcohol into the body seems to not be reliably associated with increased risk of AF, at least based on these data.

development of Arrhythmias in endurance athletes are rare

As noted, the incidence of arrhythmias correlated to endurance athletes is rare. In one long term study (reference 5 of Chapter 4) of about 52,000 cross country skiers who participated in the annual Vassaloppet 90km ski race in Sweden, an overall occurrence rate of less than 2% for all arrhythmias was found during the 10 year study period. Note that this incidence rate is similar to the results of the epidemiological study of just AF (not all arrythsmias) in a general population summarized above.

Apart from simple observations from incidence rates, it is important to point out that such studies are properly done by using exposure metrics rather than incidence rates. This is because the exposure (in this case, endurance exercise) is the parameter being studied to determine any correlated affects on the body. However, this exposure is confounded with age, a primary observed correlate with AF. Therefore the data are often presented in “person years” of exposure (in this case exposure to endurance exercise, not age, which is another type of exposure that confounds the data), as they are in this referenced study of Vasaloppet skiers. In addition, the models used to analyze the data employ techniques to correct the dataset for age and other concomitant conditions known to be present in individual subjects.  Using these metrics and modeling, the authors found an increased incidence of atrial fibrillation in the skier population that is functional with finishing time (a hazard ratio (instantaneous temporal risk) of about 1.3 among those with faster times) and number of races (a  hazard ratio of about 1.3 among those who participated in more races).

At this juncture, it is not, and cannot be, determined that AF and endurance exercise have any mechanistic connection. Any such connection is purely speculation. Additionally, there may be other, perhaps more important, factors (such as stress) not being measured that lead to the increased incidence of AF in this skier population. Since much of the data on health history and lifestyle factors (such as stress) is self-reported, significant errors are likely extant in the dataset. This is a fundamental issue with the results of all epidemiological observational studies (EOS) that attempt to link disease with exposure. This fact is never discussed in the book, yet much of the data presented derives from EOS.

Further, it can be pointed out that, even in this endurance athlete sub-population of cross country skiers, the incidence of AF is very low, however the authors of this book attempt to leave the reader (or at least this reader) with the impression that the condition is common among long term endurance athletes. The authors further reinforce this by using the anecdotal experience of one of the authors (Lennard Zinn) and making arguments that arrhythmias are much more common in endurance athletes- specifically, that an unexpected number of fellow long term endurance athletes known to Mr. Zinn came out of the woodwork after he was diagnosed and expressed that they too have various arrhythmias. Such expressed high incidence is just not supported by the (meager) data available, and, given the Boulder-centric observations, one might postulate that this supposed high incident rate of arrhythmias might best be called “Boulderythmia”- a disease that affects high strung, high stress, and “striver” male endurance athletes living in Boulder, Colorado.

This is not to say that an endurance exercise correlated arrhythmias is not a serious condition- just that it’s a rare one and one that is seen almost exclusively in older athletes just as it is (and at a similar rate) in the general population. The message here should be that arrhythmias are more common as one ages not that endurance exercise causes arrhythmias.

A critical editorial comment on the above referenced study of cross country skiers serves as a very well stated and concise summary of the state of understanding of any proposed relationship of exercise-dose with AF. The authors assembled data from numerous studies to come up with a exercise dose to risk of AF functionality. A “strawman” U-shaped graphic is provided but much caution is expressed as to the power of any of the current studies to allow firm lines to be drawn or even to establish guidelines with respect to exercise dose and the potential for development of arrhythmias.

Another recent review article by one of the authors of the comment is also a very good read (and where I learned a new (to me) word- pleomorphic – a word I am already itching to use!). Using a combination of these authors opinions, a review of the referenced work in the comment, and a critical review of the article will allow one to develop a substantiated position on the subject. Have at it!

Also recommended is a recent (2015) PhD thesis (an expanded version of the type of study in the article on cross country skiers referenced above) that has much more information and data. This study included 200,000 Vasaloppet skiers and a time frame spanning the years 1989-2010. Here is how the author summarized his work:

We evaluated risk of death during the race in two papers (I,II). During 
90 years of annual races, cardiac arrest occurred in 20 skiers, of which
five survived. The death rate is in average two per 100 000 skiers.

We also studied the association with cancer incidence (paper III). The
overall reduction of cancer was modest among skiers compared with the 
general population, but for cancers related to lifestyle the risks were
markedly lower.

We investigated the risk for recurrent myocardial infarction and found
a 30% reduction among skiers (paper IV). In paper V we showed that 
skiers with a first stroke have a lower incidence of all-cause death. 
The skiers had a higher frequency of atrial fibrillation but had less 
severe stroke and no increased risk of recurrent stroke. Thus our data 
suggest that a lifestyle with a high level of physical activity may 
work as a protection after a cardiovascular event.

Summary: The short excess mortality in endurance physical activity is
by far outweighed by the long term protective effect of exercise in 
cardiovascular diseases and cancer.

An overview of the thesis and an interview with the author is provided here and confirms the overall benefits of endurance exercise and the observed slight increase in AF. It seems obvious, based on the data available at this juncture, that the benefits of endurance exercise far outweigh any risk and this, combined with awareness of the symptoms of arrhythmias, should allow one to fully embrace an endurance sport lifestyle without any significant concerns about development of heart arrhythmias due to endurance exercise.


With the sensationalist approach of The Haywire Heart, it will take a bit of extra work and pursuit of clarification and calibration from the literature to fully grasp the current understanding of the relationship of endurance exercise to the development of arrhythmias. This is unfortunate because with some minor editing the book could have been a “one stop shop” on the subject for the interested endurance athlete and general reader alike. But the sensationalist vibe and biased writing undermine the good work that is presented.

The book does serve as a good source for basic information on the heart and how it works, review of some interesting cases of endurance athletes who have developed arrhythmias, and as a source for references to background material. I suggest you read the thesis referenced above and the other relevant work (including the recent review article) and make your own conclusions, but suffice it to say that, in my opinion, the alarm that is the apparent primary focus of the book “The Haywire Heart” is misplaced and substantially detracts from a book that provides an otherwise good source of information on the subject.



4 thoughts on “The Haywire Heart – How too much exercise can kill you and what you can do to protect your heart – Review

  1. Very interestinf summary, thanks. I have previously only listened to the podcasts, but have not yet read the book.

    • Hi Snemmel,

      Other than the for the sections on symptoms and possible treatments, I cannot recommend the book. I think one will get those aspects of interest to an endurance athlete from the recent review articles and possibly listening to a podcast with a critical ear.

  2. Thank you. I think the book reflects a cycling industry tendency to attempt to make insufficient data into an absolute conclusion (usually to sell something).

    • Hi George,

      Thanks for taking the time to comment. Yes, that “desire to sell” may have affected the final edit of the book.

      Hope you are doing well… you were a big inspiration for a budding cyclist like myself and many others in the late 70’s-early 80’s!

Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s