Monday, 7 July 2008

Monday Heart Rate Variability Measurements


Before the start of my (re)training programme I thought I would take a HRV measurement to see how things compare with a short time ago when my problems hit and use them as an ongoing baseline to see how things are going over the next few weeks.

The first graph is simply an orthostatic HR trace having spent the first 5 minutes in a recumbent position, I then stood up and remained still for the next 5 minute period. As you can see my HR responds quickly when I stand (good sympathetic tone) and then slows quickly down and remains steady at about 10BPM higher than when recumbent indicating that my parasympathetic nervous system is also performing as it should. The graph below indicates good recovery and is a very positive indicator.

The next graph is the same data but this time looking at the R-R intervals, or the time between individual heart beats. A high level of HR variability is a good indicator, low variability generally indicates fatigue, illness, or some other physiological pressure. As you can see the trace is highly variable indicating a good and healthy sympathetic/parasympathetic balance, again a positive indicator of readiness to train.

The final graph (Poincare plot or scattergram) of today's series below is a further visual representation of HRV and indicates a healthy and rested picture, when overtrained, fatigued, ill or whatever the dots tend to bunch together as HRV decreases in these states.

By way of comparison below are three graphs showing data obtained 5 days ago, the differences are clear to see and show a poorly recovered/unwell picture, a picture which has prevailed for some time. The training on the day immediately prior to both sets of measurements was identical and I had been doing far less training before the earlier data (below) was collected, if anything I should have been more recovered.

The first graph below shows my HR trace, there is the expected HR spike on standing but a much poorer recovery after standing suggesting less than optimal parasympathetic nervous system function. It is interesting to note that my recumbent HR was almost identical for both tests but far higher when standing for the earlier test below. Resting HR data alone would have shown no appreciable difference in my recovery state!

The second comparison graph again illustrates the HRV data, ie. the time periods between individual beats. You can clearly see that although there is reasonable HR variabilty whilst at rest (recumbent) as soon as my system is stressed (by standing) the HR variability deteriorates sharply indicating poor recovery, fatigue, or incomplete recovery from illness.

The final comparison graph again illustrates the HRV data as a Poincare plot and clearly shows the bunching up of individual points indicating poor HR variability caused by one or several of the previously mentioned factors.

So, with all this in mind it looks like things are looking a lot more promising than they have been, I'm glad now that I've taken some time out and tried to give myself a chance to recover properly, onwards and upwards!


  1. Looks interesting - what hardware/software are you using to analyse HRV?

  2. Hi Anthony

    I'm using the Polar RS800 HRM but there is also a Polar cycling specific computer which will do the job also, I think Suunto make one too. The ability to record R-R intervals is on the Polar model specifications page on their website.

    The bottom line is that the unit needs to be capable of being set to record R-R intervals and for that data to be downloadable. The graphs are produced with the Polar software that came with the watch which is Polar ProTrainer 5 version 5.20.131, there may be a more recent version as yet not downloaded by me!

    Hope this helps and thanks very much for your interest.