Heart rate variability: is the debate between Noakes and Ekblom about the wrong question?

In my recent post about training the heart I listed three aspects of heart structure and function that respond to training: blood supply; muscle hypertrophy; and fuel metabolism. There is a fourth which is less easy to understand, but might actually be the most important. So bear with me.  We will get back the the reality of every-day training before too long.

The fourth trainable aspect is heart rate variability (HRV); the variability in heart rate that occurs on a time scale of 2 to 20 seconds. It is well established that the heart rate varies breath by breath, inceasing during inspiration and decreasing during expiration, mainly due to input from the parasympathetic nervous system during expiration. Furthermore it is established that loss of this heart rate variability is a predictor of risk of sudden cardiac death. Finally, many stressors, including hard training or racing cause a reduction in heart rate variability and the degree of this reduction is correlated with the length of time it takes for the body to recover.  It should be noted that many different types of stress, possibly involving signals from different systems within the body, can produce a reduction in HRV.  It appears that a healthy heart requires resilience

Back to the central governor

For the past decade Tim Noakes in Capetown (J Appl Physiol 106: 347, 2009 and 106:341, 2009.) and Bjorn Ekblom from the Karolinska Institute in Stockholm (J Appl Physiol 106:339-341, 2009) have debated the issue of the central governor – a proposed governor in the brain that acts to limit power output so as to protect the heart. The debate between Noakes and Ekblom centres on the fact that even at a power output around VO2 max, there is good evidence that under many circumstances there is substantial coronary reserve (ability to increase coronary blood flow) and also ability to increase cardiac work. Ekblom has invoked this evidence to oppose Noakes theory of the central governor [Ekblom, Scand J Med Sci Sports 2000: 10: 119–122; J Appl Physiol 2007; 102:781-786 ] As far as I can see, Ekblom has got the best of the argument, at least in the eyes of many professional physiologists, though many athletes are more sympathetic to Noakes.

In my view, both Ekblom and Noakes might be asking the wrong question. If indeed it is loss of heart rate variability rather than simple deprivation of oxygen that puts the heart at risk, an effective governor will be geared up to identify serious decrease in HRV and at that point, limit the body’s power output.

Why might HRV be the key?

This seems plausible for several reasons. First of all, most deaths during or after a marathon do not occur when the heart is likely to be acutely deprived of oxygen, but rather following cumulative stress (Journal of the American College of Cardiology, vol 28, pp 428-431, 1996).  Cumulative stress decreases HRV. Secondly, the fact that decrease of HRV predicts recovery time from a race or training suggests that the brain not only keeps track of HRV but sets this in the context of how much longer the race will be. This would explain the observation that it is possible to sprint at the end of a marathon even when it was impossible even to increase pace slightly after hitting the wall a few miles earlier.

Why is it hard to improve max HR by training ?

This proposal provides a simple answer to the question of why peak heart rate is virtually non-responsive to training and tends to decrease with age, thereby resulting in decreasing VO2max and decreasing aerobic performance. A rise in HR is associated with an increase in sympathetic input; a decrease in parasympathetic input; and a decrease in HRV. At a certain heart rate, it would be dangerous to decrease parasympathetic input any further. As the tissues of the heart become stiffer with age, this might be expected to occur at a lower HR. This raises the crucial issue of whether training can prevent this deterioration. I will return to that issue later.

Nerves supplying the heart

To propose that Noakes and Ekblom have been arguing about an irrelevant measurement is a rather a bold claim to make, so you might reasonably expect me to provide some better evidence. The first body of evidence concerns the nerve supply to the heart. There is an internal nerve supply in the heart (Arora et al , Am J Physiol Regul Integr Comp Physiol 285: R1212–R1223, 2003), and also an external input from the two main divisions of the autonomic nervous systems: the sympathetic system which makes the heart beat faster and the parasympathetic system that slows the heart down

Sympathetic input rises under stressful circumstances while parasympathetic input increases during relaxation or recovery. Sympathetic input to the heart causes slow oscillations in heart rate (on a time scale of 7-20 second2); parasympathetic input causes more rapid oscillations on a time scale of 2-7 seconds. There is a great deal of evidence that a preponderance of slower parasympathetic oscillations predicts a longer life. Spinal cord stimulation which is likely to send parasympathetic signals to the heart tends to overcome the dangerous effects of oxygen deprivation in a diseased heart, and can be an effective treatment for angina (Foreman et al Cardiovascular Research 47;367–3750: 2000).

The heart talks to the brain

Furthermore the passage of nerve messages between heart and brain is not one way. The nerves in the heart send messages back to the brain, providing a means of informing the brain about the current variability and stress level. However the internal cardiac nerves cannot by themselves produce the heart rate variability that appears to be a feature of health. In a dog with the sympathetic and parasympathetic input removed surgically, coordinated rapid contraction of the heart occurs but HRV is abolished (Murphy et al Am J Physiol Regulatory Integrative Comp Physiol 266:1127-1135, 1994). There is a large body of evidence indicating that HRV predicts the amount of extra oxygen that will be required after a hard training run – and this does not appear to be simply a matter of the brain estimating the accumulation of lactate, at least when the heart has been under substantial stress (Dixon, Cardiovasc Res. 26(7):713-9,2000).  Suunto, who manufacture a heart rate monitor that can measure HRV, claim that loss of HRV might be a useful indicator of over-training. I think the evidence regarding detection of over-training is complex, but there appears to be a degree of truth in Suunto’s claim.

What does this mean for planning a training program?

If some or all of these speculations are correct, the crucial issue for the athlete is whether HRV can be improved by training. The good news is yes (eg Dixon, Cardiovasc Res. 26(7):713-9, 1992). In the near future I will return to the discussion of how we can best train the heart, and at that point I hope to provide a good answer to this question.

Meanwhile, with regard to Ewen’s comment after my post on Saturday suggesting that he is limited by his muscles, not his heart, I agree, but I would propose that it might be that his current strenuous training and relativley unremitting program might have led to a reduction in HRV. Hence, his central governor is acting to stop him pushing himself too near to VO2max. Ewen, do you have a HRM that will provide you with HRV values? I apologize in advance if I am shooting off with an unfounded hypothesis.


7 Responses to “Heart rate variability: is the debate between Noakes and Ekblom about the wrong question?”

  1. Ewen Says:

    Canute, I don’t know that my current situation can assist with your hypothesis. What you propose does make sense.

    Unfortunately, my HRM doesn’t measure HRV values – it just measure’s average and maximum for the distance, or for splits if I ask for them.

    I have ‘some’ HRV at rest, as I’ve noticed when wearing the HRM whilst driving, my HR might be 70 and then with relaxation I can get it down to 64 or 65. In recent training runs over my 7k undulating course, I’ve noticed the HR hasn’t been dropping as much on the downhills as it did 6 weeks or so ago. In the couple of ‘tired’ races, my HR has stayed pretty much on 150 (after the first km or so). It only rose to 157 on the uphill finish (which I was pushing hard). During the track season, I could pretty much average 155-7 in 3k races, and be in the low-mid 160s at the finish. I think my max is around 167.

    There was more variation in my training during the track season, but overall it wasn’t as demanding as it is now. 120k a week is pretty demanding for me, especially as I haven’t increased the sleep! I’m thinking of varying this ‘base’ plan after Gold Coast to allow more variety and recovery. I’m thinking of maybe 7-9 days of ‘base’, then 6-7 days of higher intensity, then 3-4 days of ‘light intensity’ (sprints etc). Not sure if that will help the HRV. Your thoughts?

  2. canute1 Says:

    Ewen, Thanks for that response. I am pleased to hear that you ‘still have some HRV’ and my current hypothesis is far too speculative to justify rushing out to buy a Suunto HRM.
    With regard to your proposal to include more variety in your training after Gold Coast HM, I think that is probably the safest course of action, but the fact that Hosaka does the same demanding sessions every day has led me to a lot of thinking but not yet led me to many conclusions. My overall understanding is that Hosaka built up his marathon fitness over decades. His current training volume is amazing for a man of his age, but maybe not for a man with his recent marathon performance, and I would be very cautious about copying him. I would love to have HRV data from Hosaka. After I have thought about it a bit more I will probably do a post on Hosaka.

    Meanwhile, I think you are likely to have a fairly robust heart after years of aerobic training, so provided you keep an eye on your overall physical and mental state (eg increasing resting HR, lethargy, low mood or motivation, are all pointers towards overtraining) I suspect you current training is not too risky. However, if I were you I would taper before the Gold Coast HM so that you are fit to give it your best effort.

  3. Ewen Says:

    Thanks Canute. I’ll be most interested in what you come up with about Hosaka’s training. I doubt I’m capable of copying his actual volume/intensity. I’m interested in following that model of training for a while (same daily session/s), at least for 4 or 5 days of the week. Maybe I’ll try his ‘moderate-paced’ progressive 1k repeats at some stage.

    I vaguely remember reading somewhere that his training didn’t always consistently have such volume due to business.

    I’ve hit a bit of a problem with a sore shin, so my taper for Gold Coast will be rather extended!

  4. Helen Says:

    Canute, this is all new knowledge to me..but again, you explain it so well. Does the training mileage correlate with risk of (race related) heart attack? Is it the v high mileage runners that are most at risk? If it’s to do with hard training and reduced HRV, wouldn’t you expect that?

  5. canute1 Says:

    Helen, thanks for your comment. Fortunately, heart attack during races is so rare that it is not easy to draw conclusions about who is at appreciable risk (apart from those with pre-existing heart disease). As I will discuss in more detail in a future post, training can increase HRV, so very high mileage might be fine if it is based on a good foundation. However, in the future post I will say why I think there might be appreciable risk for a marathon runner who commences a marathon while in an over-trained state, and without an adequate taper.

  6. Ewen Says:

    Canute, interestingly, after my 4 days off running this past week, I ran a short race (2.5k) on Sunday afternoon and was able to average a HR of 158 (compared to the 150 of recent races).

    I’m unsure of what that means. Did the fresh legs enable me to push the HR higher, or did the heart just work harder of its own accord because I’ve lost some aerobic condition?

  7. canute1 Says:

    Ewen, The question of whether your demanding training schedule caused you putative ‘central governor’ to limit your heart rate during the race a few weeks ago is intriguing. On balance I consider it is likely that this was the case.

    A year or so ago I made a misguided attempt to estimate my maximum heart rate by running two peak-effort 1km runs separated by a brief rest, expecting I would reach max HR during the second run. In fact, despite a subjective peak effort, when I glanced at my HRM halfway through the second 1Km I realized that my heart rate was already falling, despite not even reaching the peak value achieved on the first run. Disappointed with my poor showing, I decided to make a really determined attempt at the same 2x1Km session the next day, but couldn’t even achieve the peak HR I had reached in the first 1Km the previous day. This might be described as a ‘psychological’ block, but I do not think it should be dismissed as ‘merely psychological,’ as if I could have simply tried a bit harder. I think that there are times that our non-conscious mind acts to preserve us from our own over-enthusiasm. I continue to think that HRV is likely to be a useful indicator of when we are over-doing things. I will post some further evidence on that, though I probably wont be able to fit all these thoughts into my next blog.

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 )

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

%d bloggers like this: