In response to my recent post on over-training and Heart Rate Variability (HRV), Ewen asked if I had an opinion about which Heart Rate Monitor with capacity to measure HRV might be best. I have not yet purchased such a device. Before I offer my tentative thoughts on what might be the best device to buy, it is worth a brief deviation back to the Australia of my childhood in the 1950’s.
A trip back to Australia in the 1950’s
In those days, Australia dominated the world in several sports, but especially in swimming. The really memorable character was Dawn Fraser, who won gold in the 100m freestyle in Melbourne (1956), Rome (1960) and Tokyo (1964). Among the men, two of the greats were John Konrads and Murray Rose. A feature of the Rome Olympics was the battle between Konrads and Rose, with Rose winning gold in the 400m freestyle and Konrads in the 1500m. Konrads held the world 400m record at the time. During his career he broke multiple world records over distances from 220 yards to 1650 yards.
What has this got to do with measuring over-training? Following my recent posting on Heart Rate Variability, Mystery Coach sent me a very interesting report which Forbes Carlile and his wife Ursula presented to the Australian Sports Medicine Association in April 1959, entitled ‘T wave changes in strenuous exercise’. Forbes Carlile was in those days a giant figure in swimming coaching, in Australia and internationally. Carlile and his wife had recorded over 500 ECGs from swimmers, cyclists and oarsmen, in many cases performing recordings at different points in the season and relating these recordings to changes in performance
The main conclusion of the report was that stressful training or racing produces a decrease in amplitude or sometimes complete inversion of T waves in the ECG. Carlile and his wife reported: ‘In general the sportsmen with a relatively light training load gave a series of practically unchanged electrocardiograms whereas those who were training strenuously frequently showed T wave changes in all leads.’
The pictures of the ECG traces were dramatic. For me, one of the fascinating contrasts was between the recordings for John Konrads and those for several other top level swimmers. The three recordings for Konrads were done at the beginning of hard training at the end of November 1958, and then again immediately before and after a 440 yd race on 28th January 1959. Unlike the pattern seen in the other top level athletes, Konrads’ T wave amplitude increased during hard training, and remained unaffected by the race. However for several other top-level swimmers, their T waves showed quite perceptible flattening during periods of intense training. In these instances, the decrease in T wave amplitude was associated with deterioration in performance.
Carlile and his wife concluded: ‘we suggest that serial electrocardiograms offer a practical and scientific means of guiding the sportsman in his training.’ Examination of the ECG traces provided in their report made it very difficult to disagree. Though in light of the fact that Dawn Fraser bestrode Australian swimming like a colossal cheeky Amazon at the time, one wonders about the use of the word ‘sportsman’ – but the 1950’s were of course over before another famous but slightly cheeky Australian woman, Germaine Greer, turned not just our T waves, but our attitudes upside down with ‘The Female Eunuch’.
What has happened to T waves since 1959?
In fact we now know quite a lot more about the things that produce a change in T waves. T waves are the most labile feature of the ECG and can be affected by many stresses on the body. One unifying feature is that T wave amplitude is diminished when the sympathetic nervous system is overactive. By performing scans of the heart after administering a radioactive tracer substance called I123-MIBG , which competes with noradrenaline to bind to the receptors on the surface of cells in the heart that mediate the effects of the sympathetic nervous system, it is possible to show that over-activity of the sympathetic nervous system is associated with suppression of the T waves.
Thus, in principle, T wave suppression appears to be a good candidate to assess the form of over-training characterized by excess sympathetic activity. There is of course a problem that anxiety also causes over-activity of the sympathetic nervous system, and can cause suppression of T waves. Therefore assessment of T waves would only be useful if interpreted in light of other features affecting the physical and mental state of the athlete.
I do not know whether athletes at the Australian Institute of Sport still have serial ECG’s to assess training stress but I suspect this is unlikely. Since the 1990’s the emphasis has shifted from the shape of the ECG waveform to heart rate variability, but the fact that non-invasive assessment of the effects of the autonomic nervous system on the heart has been possible for half a century, yet there is no widespread accepted procedure, makes me cautious in offering any advice.
Back to the measurement of HRV
Despite promising findings regarding the use of HRV to adjust training schedules (as reported by Kiviniemi and colleagues in the study I described on my blog posting yesterday), the situation is complex, so I think that investment in a HRM capable of recording HRV is a speculative investment. They are not cheap, though if you can afford it and regard it as an interesting tool for investigation rather than a certain answer to the question of how to adjust training load, then I think it might be worthwhile.
The two companies that have invested extensively in HRV technology are Suunto and Polar. As a person with a wish to understand the underlying physiology, I find the websites of both companies very frustrating. Both companies have clearly recognized that there is no simple measurement that applies to all individuals under all circumstances and both have developed ways of calculating training stress that takes account of the characteristics and situation of the individual.
As mentioned in my blog recently, for assessment of over-training, Polar appear to place the main emphasis on 5 measurements performed on standing up from relaxed resting. This is a variant of the traditional orthostatic test, and involves assessment of changes in heart rate and heart rate variability. In my post yesterday I gave a link to the Polar website. It might also be useful to read the second part of this document prepared by the Heart Rate Monitor Shop in which they describe the OwnOptimizer test performed using Polar RS800.
My main concern about Polar’s OwnOptimizer is that it does not employ data based on the body’s response to a training session, and I am not sure how easy it is to derive estimates of autonomic function during training or during other activities of daily living from the Polar RS800.
Note added 30 June 09 (22:00): I have discovered that FirstBeat Technolgies software can read the data from a Polar RS800. Therefore, it appears that the various useful computations that I attribute to the Sunto T6 when used in conjuntion with the First Beat Technologies software might also be achieved using the Polar RS800. I am frustrated by the fact that neither the Polar website or Suunto website make it clear that the capability of their devices might be mproved by use of Firstbeat Technologies software.
The Suunto t6, when used in conjunction with software developed by Firstbeat Technologies appears to provide useful information about autonomic function at any time of the day or night. As far as I can see the recommended way to detect over-training is from overnight recordings. The software measures what are described as ‘stress reactions’ during sleep, and if these continue throughout the night, the athlete is at risk of over-training. The software also produces two potentially informative quantities related to stress during training: training effect (an estimate of the stress on the body arising from training session) and EPOC (an estimate of the body’s additional oxygen requirement post exercise, estimated from HRV measurements).
Useful information about the Suunto t6 is provide by Eddie Fletcher, a indoor rowing coach with international credentials and a clientele that includes international indoor rowing champions.
He has written some interesting articles for Peak Performance. The following article from PP 237 is available on his website:
Emma Snowsill (Triathlon gold medal winner in Beijing) uses Suunto t6c red arrow.
A tentative recommendation
If I had to choose between Polar and Suunto, I would choose Suunto t6 (with the FirstBeat Technolgies software – though at this stage I am uncertain whether or not the Polar RS800 might also yield similar information when used in conjuction with FirstBeat technolgies software ). However, because my own personal approach is to try to understand the physiology, for the time being, I am inclined to continue to use my own amateur set-up. With my set up I can also examine the waveform of the ECG. I am still inclined to think that the size and shape of the T wave might be quite informative (despite the lack of clearcut conclusions subsequent to the report by Forbes and Ursula Carlile fifty years ago). However, my set-up does not allow wireless recording, so it is only useful for resting and standing measures. I think that Suunto (when used in conjunction with Firstbeat software) is probably on the right path with assessment of autonomic nervous system function during sleep, every day activities and training. I aim to post some more information on these issues in my blog over the next few weeks.
Additional Edit (30 June): As I explore the Firstbeat Technologies website ( http://www.firstbeat.fi/ ) I am starting to get a clearer understanding of which devices can be used to perform the various measurements (all derived from HRV data) that have been developed by Firstbeat Technologies, which is a spin-off from the Research Institute for Olympic Sports, Jyväskylä, Finland
My current understanding is as follows:
Suunto t3 and t4 provide a ‘real time’ read out of Training Effect.
Suunto t6 with Suunto training manager software can provide more detailed analysis including Training Effect and EPOC
It appears that Firstbeat have provided Suunto with the relevant software for incorporation in the Suunto Products.
Furthermore, I understand that data from either Suunto t6 or Polar RS800 can be read directly by Firstbeat Technologies software and used to compute Training Effect, EPOC and several other physiological variables.
In my experience, the Firstbeat Technologies website is clear and informative, whereas I found it harder to glean the facts from either the Polar or Suunto websites.