Archive for June, 2010

Is my Polar RS800cx a boon or a curse?

June 27, 2010

My Polar RS800cx heart rate monitor has created chaos. I purchased it last summer with the hope that measuring heart rate variability would help me answer the most challenging question facing any runner trying to maximize performance with limited training time: the question of finding the optimum balance between training load and recovery. Almost a year later, I do not know whether has been a great benefit to my health and wellbeing, or a curse that has blighted my running.

There is little doubt that the beneficial effects of training depend as much on the quality of recovery as on the quality of the training itself. One answer is to spend a lot of time sleeping. This is the answer adopted by many African athletes and also individuals such as Paul Radcliffe. Unfortunaltey, for many of us with commitments to jobs and family, this simple solution is not practical. So in an attempt to maintain the balance between stress and recovery we try to find some measurement that will allow us to push our bodies far enough to reap the benefits of training, but warn us when we are in danger of pushing beyond the point where additional training becomes counter-productive.

One of the most promising is measurement of the balance between the two arms of the autonomic nervous system. This system has two opposing components: the sympathetic component which exerts its effects via adrenaline, and mediates the response to stress; and the parasympathetic component which acts through the messenger molecule, acetylcholine, that mediates relaxation and recovery. Fortunately, we can monitor the balance between sympathetic and parasympathetic activity relatively easily by measuring the beat-by-beat fluctuations in heart rate. This requires a heat rate monitor with the capacity to measure and record R-R intervals – the intervals between successive R waves (in the ECG) generated by successive heart beats. High frequency beat-to beat fluctuations (at frequencies around respiratory frequency or higher) reflect parasympathetic activity. Lower frequency fluctuations represent sympathetic activity.

The principle is simple: by measuring heart rate variability (HRV) during a period of a few minutes under consistent circumstances each day (eg after getting up in the morning) we can determine whether the amount of high frequency fluctuation indicates an adequate level of recovery and hence indicates that we have recovered sufficiently to benefit from a hard training session that day. Not only is the principle is very simple, but the practice also appears beguilingly simple in an era when electronic devices that existed only in the dream-world of science fiction during my childhood, can now be purchased for a sum of money not much beyond what my teenage son would happily spend on tickets for a rock concert. Both Polar and Suunto produce heart rate monitors that will provide the required R-R data, which can be readily uploaded to a computer for analysis, while other devices, such as Ithlete, can be plugged into an ipod or an iphone to produce an apparently easily interpreted on-screen message.

The scientific evidence

However, despite the simplicity and accessibility of the data, there is remarkably little good scientific evidence that daily monitoring of the autonomic nervous system leads to optimal training. In a promising small study [1], Kiviniemi and colleagues from Oulu in Finland compared a group 9 runners randomly assigned to a 4 week training program adjusted according to daily measurement of HRV, with 9 runners assigned to a conventional fixed training plan comprising four high intensity and two low intensity sessions each week. The group whose training load was reduced on days when the HRV measurement indicated inadequate recovery showed a significant increase in VO2 max whereas the group following the fixed schedule did not. Furthermore the group whose training was adjusted according to HRV showed a significantly greater increase in maximum running velocity on a treadmill test. This study supports the hypothesis that adjusting training according to level of parasympathetic activity leads to better performance.

A much more detailed study of 8 runners preparing for a marathon [2] by Manzi and colleagues from Rome, also suggests that HRV can be informative, but leads to a more complex conclusion. I will review Manzi’s study in greater detail in my next post, but the main conclusion is that relatively greater ratio of low frequency variability to high frequency variability, indicating relatively more active sympathetic nervous system activity, predicts better marathon performance. Overall, the evidence suggests that HRV can be informative, but the interpretation of the data is complex.

My experience

Meanwhile, my personal experience with my Polar RS800cx has been even more perplexing. In February, I posted an R-R recording that appeared to indicate that I have atrial fibrillation (AF).  I have subsequently posted several discussions of the paradoxical evidence that endurance training is associated with increased risk of AF. John Bedson, who runs a support group for people with AF and has a wealth of information about the condition, is adamant that my trace shows AF. I am less sure, but I am nonetheless taking the evidence seriously. I have in fact been investigated fairly thoroughly by a cardiologist; so far without any clear evidence of heart abnormality emerging. An echocardiogram revealed a normal heart structure and ECG revealed only the mild hypertrophy expected in an endurance athlete.

Unfortunately, the interpretation of the evidence from my Polar is confounded by clear evidence that the device is unreliable in a number of respects. Despite appearing to function well on many occasions, there are times when the trace makes no sense at all (for example showing quite long periods of unbelievably low heart rate); sometimes the data is not recorded; sometimes pressing the various buttons does not achieve the intended effect, sometimes the transmission of data from the device to my computer does not work properly. I had sent it back to Polar (UK) a few months after I purchased it. They kept it for five weeks and after they returned it, I found that they had corrected a fault in the foot-pod, but did not appear to have done anything to deal with several of the other inconsistencies.

One of my reasons for deciding to purchase the RS800cx was my belief that Polar produce good quality equipment. Subsequent to my unfortunate experiences, several people have shared anecdotes about erratic behavior of their Polar HRM’s but in most instances, the company were prepared to replace defective equipment. I accept that the RS800cx is a sophisticated device and it is perhaps not surprising that sometimes things go wrong. However, it seems to me that a company that produces such products for the market at cost that is modest but not entirely trivial, should be prepared to offer good service. I am about to send my RS800cx back to Polar once again, in the hope that they can either rectify the problem or replace the device.

So at present I face a quandary.  If I really do have AF, I owe substantial gratitude to Polar for allowing me to discover it. AF occurring without overt symptoms is potentially quite dangerous, especially as AF begets more AF. It is important to minimize the duration of episodes of AF. On the other hand, I believe that complete cessation of exercise would be even more harmful to my long term health. Thus, learning how to adjust my level of exercise to minimize AF, if indeed I suffer from AF, is crucial. A reliable HRM offers me the best prospect of being able to continue to run in reasonable safety. However, a monitor that produces unreliable data merely generates uncertainty and anxiety.


[1] Antti Kiviniemi, Arto Hautala, hannu Kinnunen & Mikko Tulppo (2007) Endurance training guided individually by daily heart rate variability measurements. Eur J Appl Physiol. 101(6):743-751.

[2] Vincenzo Manzi, Carlo Castagna, Elvira Padua, Mauro Lombardo, Stefano D’Ottavio, Michele Massaro, Maurizio Volterrani, and Ferdinando Iellamo (2009) Dose-response relationship of autonomic nervous system responses to individualized training impulse in marathon runners Am J Physiol Heart Circ Physiol 296: H1733–H1740.

The enigma of the runner’s heart

June 5, 2010

The heart of the endurance runner is an enigma.  Typically the ventricles are enlarged and the muscular walls at least moderately thickened.  As discussed in my post of 20th June 2009, the muscle is well perfused, unlike the situation in ischemic heart disease, in which obstruction of the coronary arteries results in a restricted supply of blood to the hypertrophied heart muscle.   Athletes are less likely to die of heart disease and live longer than sedentary individuals [1, 2].  However, as discussed in my blog on 28th February 2010, disturbances of heart rhythm are common. 

Elderly athletes have a very high rate of peculiarities of heart rhythm  For example, Jensen-Urstad and colleagues [3] recorded the electrocardiogram (ECG) continuously for 48 hours in 11 elderly male athletes with a life-long history of strenuous exercise, and 11 age-matched control men who were either sedentary or had been only moderately physically active.  Disturbances of rhythm were more common in the athletes.   Nine of the 11 athletes compared with 4 of the 11 controls had complex abnormalities of ventricular rhythm. In general abnormalities of ventricular rhythm are worrying because such abnormalites might lead to chaotic ventricular contraction (ventricular fibrillation) which is fatal.  In addition 9 of the 11 athletes compared with 4 of the controls exhibited more than 100 premature atrial contractions per day.   Premature atrial contraction is less worrying, though in some circumstances it can precede atrial fibrillation, which is disruptive and can lead to fatal stroke.  The two athletes without complex ventricular rhythms were among the 9 with more than 100 premature atrial contractions per day, so all 11 athletes exhibited a noteworthy disturbance of rhythm, in contrast to 5 of the 11 controls.  Overall, the study suggests that that life-long strenuous athletic training increases the risk of disturbance of heart rhythm. At least some of these disturbances of rhythm are of a type associated with risk of sudden death.

On the other hand, the increased life expectancy of athletes suggests that the benefits of running compensate for the increased risks.  In a 5 year follow up study  of 122 middle-aged or elderly cross-country skiers by Lie & Erikssen [4], of the 23 who were found to have ECG abnormalities at initial screening, almost all continued regular training but none suffered from heart attacks, though two did develop angina.  Overall, this study suggests that progression to serious heart conditions is rare. 

Who is at greatest risk?

Nonetheless, this leaves the question of whether there might be identifiable factors that create greater risk in some athletes.  It is well recognized that a small number of runners collapse and die during races, especially during long races such as the marathon, but the risk of sudden death is trivial compared with the increased life expectancy.  It is widely believed that most of those who die unexpectedly had previously undetected  rare abnormalities such as congenital hypertrophic cardiomyopathy.  However, the very high prevalence of rhythm disturbances in those who have engaged in long-term strenuous training reported by Jensen-Urstad [3] raises the possibility that risk might build up slowly but inexorably during long-term heavy training, and furthermore that the risk might indeed be a direct consequence of the hypertrophy of cardiac muscle arising from training.

Atrial fibrillation (AF)

While I am aware of little evidence apart from that provided by the study of Jensen-Urstad and colleagues [3] to suggest that heavy training might play a causal role in ventricular rhythm disturbances, the evidence that training might contribute to atrial fibrillation (AF) is much stronger.  I reviewed some of this evidence in my posting on 28th February 2010.  Perhaps most compelling of all was the evidence from the recently reported study by Grimsmo and colleagues [5] demonstrating that 12.8 % of 122 long distance skiers competing in the annual 54 Km cross-county race from Rena to Lillehammer, in Norway developed ‘lone’ AF  (not accounted for by any underlying heart abnormality) during a follow-up period of 28-30 years.  The mean age of onset of the atrial fibrillation was 58 years.  Although AF is quite common among the elderly, occurring in about 5% of men over the age of 65, 12.8% is a very high rate to find among relatively young middle aged men.

AF has significant consequences.  It limits cardiac output and therefore limits performance; it quite often causes significant disruption of every-day activities; and,  more importantly from the perspective of health and longevity, it increases the risk of stroke.  Clots of blood form in the fibrillating left atrium and if dislodged, are prone to be borne via the aorta and the carotid arteries to the brain where they obstruct smaller blood vessels, resulting in a stroke.   The frequency of this in athletes with AF is unknown, but the one longitudinal study of athletes who continued to exercise after developing AF is salutary. In a 9 year follow-up study of 30 athletes with AF, Hoogsteen and colleagues [6] found that 3 (10%) had died; paroxysmal atrial fibrillation continued in 15 (50%) ; permanent atrial fibrillation emerged in 5 (17%) ; and 7 (23%) of showed no further atrial fibrillation.  Of three deaths during the follow-up, one was a sudden death during a race and is likely to have been due to a heart rhythm disturbance.  The other two were attributed to strokes.  Although Hoogsteen re-assuringly concludes that these deaths were not directly attributable to AF, the outcome is disconcerting.

My tachygrams

On 28th February, I had posted a heart rate record made in R-R mode (that is, a record that gives the beat by beat variation in heart rate) during one of my easy-paced runs that showed wild beat to beat fluctuations that looked like AF – though I had been entirely unaware of any symptoms during the run.   The definitive diagnosis of AF depends on a demonstration of absence of the P waves that reflect the orderly conduction of an electrical impulse from the sino-atrial node where it is generated, through the walls of the atria to the atrio-ventricular node, from which point it spreads through the  ventricles producing the powerful ventricular contraction that ejects blood into the arteries. The record of R-R intervals (the tachygram) is based on detection of the sharp R wave associated with ventricular contraction, but contains no information about the P wave.  Therefore, my wild tachygram is not absolute proof of AF, though the comments which John Bedson recently posted on my post of 28th February maintain that the evidence of AF is unequivocal and  further, he suggests that my running career is over.

John has extensive experience of AF and I respect his opinion.  However this leaves me in a quandary.  I had of course been to see a doctor after I had discovered the crazy heart rhythm, and have had some investigations.  So far the only feature the investigations have revealed is ventricular enlargement, but this is an expected consequence of my history of extensive running and other heavy exercise in young adult life, and my more modest but fairly regular running in recent years.  As far as my official medical record goes, no significant abnormality has yet been identified.

Nonetheless, I have scrutinized the tachygrams recorded in the majority of my training sessions over the past 8 months, and at this stage have concluded that the picture is one that warrants some caution.  I have had the wildly fluctuating rhythm on four occasions, lasting from a few minutes to half an hour, though never accompanied by any conscious awareness of symptoms.  On three of these occasions I had inhaled salbutamol before the run to relieve wheeziness due to asthma.  Since I have only used the salbutamol inhaler on 8 occasions over the eight month period, the data indicate that I have had the wild rhythm on 37% percent of the occasions when I have used salbutamol, and much fewer than 1% of occasions when I have not used salbutamol.  So a link between the wild rhythm and salbutamol is unlikely to be a chance occurrence.  Salbutamol relaxes the smooth muscle in the bronchi thereby relieving the wheeze of asthma, but can activate the adrenaline receptors in heart muscle, so it seems quite likely that salbutamol has caused accelerated and perhaps chaotic heart beat.   I have now abandoned use of salbutamol and have no intention of ever using it again, except in an emergency. At present after a run, my peak airways flow is around 200-250 liters per minute whereas it is in the range 500-600 litres per minute after use of salbutamol.  I will have to find some other way of dealing with the asthma that does not involve adrenaline-like medication. 

My future

Is my running career over, as John Bedson suggests? In general rhythm disturbances get worse as one gets older, and it is likely that my crazy heart rhythms will get worse, though it is noteworthy that in the 30 cases of AF followed by Hoogsteen [6], apparent recovery occurred in 23% while  17% progressed to persistent AF, and 50% continued to have episodes of AF.  

The ultimate decision might be forced upon me by the outcome of further cardiac investigations but if a reasonably thorough series of investigations reveals no significant abnormality, I will have to make a decision based on a careful appraisal of the evidence.  The full picture must take account of  both the risk and benefits.  While the study by Grimsmo provides compelling evidence that heavy endurance training can cause AF, the overall picture is more complex.

Beneficial effects of exercise

A study by Mozzafarian and colleagues [7] reveals that light to moderate exercise actually reduces the risk of AF.   Within a sample of over 5 thousand elderly individuals, those who exercised with moderate intensity had a 39 percent reduction in liklihood of suffering AF in comparison with sedentary individuals (that is, a relative risk of suffering AF of 61%). Even those reported to exercise at high intensity had a modest reduction compared to sedentary individuals, having a relative risk of 71%.  High intensity was not defined precisely but almost certainly was less intense than the exercise taken by the skiers studied by Grimsmo [5].  Although Mozzafarian and colleagues did not define high intensity clearly, they did provide an estimate of the reduction in risk produced by fairly vigorous walking.  Those in the highest quartile (i.e. the top 25% of the sample) for distance and pace of walking had only half the risk of developing AF compared with the lowest quartile.  Those who walked 60 or more blocks per week had a risk that was 56% of the risk among those who walked 4 or less blocks per week.  Similarly, those who walked at greater than 3 miles per hour had a risk that was only 59% the risk of those who walked more slowly than 2 miles per hour.  Thus, at least up to the upper end of the scale of recreational walking, the risk of AF decreases steadily with increasing volume and intensity of exercise.

It is also necessary to consider the other benefits of exercise.  There is good evidence that quite vigorous exercise increases life expectancy [1]; reduces risk of cancer [8]; lowers blood pressure and lowers risk of ischemic heart disease [2]; and there is rapidly accumulating evidence that it is associated with improved memory and other aspects of brain function [10].  Furthermore the overall evaluation of risks and benefits should take account of the fact that I simply enjoy running in the woods and meadows

These considerations lead me to conclude that if I do have AF, complete cessation of running would probably decrease my overall quality of life and longevity.  However, the challenge is to decide what is the optimum amount and type of exercise.

With regard to type of exercise, it is interesting to note that all instances of crazy rhythm have occurred when I was running in the low aerobic zone. It has never appeared during tempo runs, long interval sessions (1000m or 2000m repeats) or HIIT sessions.  On one occasion it had developed during a low aerobic run and disappeared when I increased HR into the mid-aerobic zone, as I climbed a steep hill.  On another occasion it occured as my heart rate decreased while jogging duirng the ‘cool down’ after a run.  It has never occurred during sessions of elliptical cross training, even though I generally push my cardiovascular system harder on the elliptical than when running.  Nonetheless, the study by Jensen-Urstad [3] suggests that the risk is associated with intensity, and the study by Mazzafarian [7] shows that the risk of AF is lowest for those who exercise at moderate intensity, so at least for the time being I will avoid high intensity sessions. 

The bliss of ignorance

I would never have known about the possibility of AF if I had not performed R-R recordings.  In light of the frequency of heart rhythm disturbances in elderly athletes reported by Jensen-Urstad [3], I wonder whether indeed a high proportion of elderly runners might actually have similar evidence of crazy rhythms but continue to run in blissful ignorance.  However I do not regard ignorance as blissful.  I would rather know as many facts as there are to know.  I am greatly reassured that whatever decisions I make regarding type and amount of running in the future, I can use my heart rate monitor to keep a close eye on whether or not my problems are deteriorating. This gives me the confidence to experiment cautiously. 

One crucial issue is whether there are any practical markers of deterioration that will provide advance warning of AF.   Examination of the tachygrams on those occasions when I have developed the crazy rhythm, reveals clear evidence of premature atrial contractions (PACs) – sharp spikes in which the rate of a singe beat is more than 30% higher than the surrounding beats, before the onset of the crazy rhythm.  As shown by Jensen-Urstad [3], PACs are common in elderly people and especially common in runners.  In most instances they are no related to AF.  Furthermore, as far as I can see I have a lower frequency of PACs than the elderly runner studied by Jensen-Urstad.   However in my case, there appears to be a clear relationship between PACs and the crazy rhythm and therefore, at least for the time being, I will regard frequency of PACs as a guide to either deterioration or improvement.  

An additional reason for regular monitoring it the possibility of seeking active treatment.  Medication can be effective in some cases, though several experts advocate a relatively minor surgical procedure in which a burst of radio frequency energy is employed the ablate the tissue that appears to be source of the chaotic rhythms, usually near the point where the pulmonary vein enters the left atrium.   However there is a small but appreciable mortality associated with this procedure, and despite the optimism of those who advocate the procedure, long term outcome remains uncertain.   A recent review carried out under the Health Technology Assessment programme [11] advised caution regarding this procedure.

As for my future racing career, I will simply have to see what story the tachygrams tell me.  For the near future, I am making no plans to race nor setting high performance goals for the next few months. 


[1] Paffenberger Jr RS, Hyde RT, Wing AL, et al. (1986) Physical activity, all-cause mortality, and longevity of college alumni. N Engl J Med.  314:605–14.

[2]  Berlin JA, Colditz GA. (1990) A meta-analysis of physical activity in the prevention of coronary heart disease. Am J Epidemiol. 132:612–28.

[3] K Jensen-Urstad, F Bouvier, B Saltin,M Jensen-Urstad (1998) High prevalence of arrhythmias in elderly male athletes with a lifelong history of regular strenuous exercise Heart 1998;79:161–164

[4] Lie H & Erikssen (198 ) Five-year Follow-up of ECG Aberrations, Latent Coronary Heart Disease and Cardiopulmonary Fitness in Various Age Groups of Norwegian Cross-country SkiersActa Medica Scandinavica 216 (4),  377 – 383

[5] Grimsmo J, Grundvold I, Maehlum S, Arnesen H, (2010) High prevalence of atrial fibrillation in long-term endurance cross-country skiers: echocardiographic findings and possible predictors–a 28-30 years follow-up study. Eur J Cardiovasc Prev Rehahil. 17(1): 100-105

[6] Hoogsteen J, Schep GGG, VanHemel NM, Van Der Wall EE (2004) Paroxysmal atrial fibrillation in male endurance athletes: a 9 year follow-up.  Europace 6(3): 222-8

[7] Mozaffarian D, Furberg CD, Psaty BM & Siscovick D. (2008)  Physical Activity and Incidence of Atrial Fibrillation in Older Adults: The Cardiovascular Health Study. Circulation. 118;800-807.

[8] Jones et al (2010) Effect of aerobic exercise on tumor physiology in an animal model of human breast cancer J Appl Physiol 108:343-348

[9] Smith, JP,  Blumenthal JA, Hoffman BM, Cooper H, Strauman TA, Welsh-Bohmer K, Browndyke JN, Sherwood A. (2010)  Aerobic exercise and neurocognitive performance: a meta-analytic review of randomized controlled trials.  Psychosomatic Medicine. 72(3):239-52.

[10] Rodgers M, McKenna C, Palmer S, Chambers D, Van Hout S, Golder S, Pepper C, Todd D, Woolacott N. (2008) Curative catheter ablation in atrial fibrillation and typical atrial flutter: systematic review and economic evaluation  Health technology assessment (Winchester, England), 12(34):iii-iv, xi-xiii, 1-198.