by James AJ Heathers
Ph.D. Cand., M.Sc., B.Ec.
Short answer:
Because low frequency power isn’t a good measure of ‘sympathetic activity’, and low/high frequency (LF/HF) power isn’t a good measure of ‘sympathovagal balance’. What we’re interested in as an index of recovery is a controlled measure of high frequency (HF) power.
Long answer:
Frequently with my own research, and about heart rate variability in general, people often ask me: “Do you measure LF/HF balance?” I do not. Here’s why:
Over a long enough period of time, anywhere from a few minutes upwards, heart rate isn’t stable. It exists in an environment of continual modification due to the inter-connections between spontaneously generated autonomic rhythms, the vasculature (blood vessels), breathing and any kind of imposed stress – exercise, focused attention, etc. This provides us with two dominant frequencies (two speeds of regular fluctuation in beat-to-beat variability), and because one is faster than the other, they’re called high and low frequency.
High frequency
HF rhythms are mediated primarily by the vagus nerve’s innvervation of the sinoatrial node on the top of the heart. This reflects respiratory sinus arrhythmia, the process where breathing in increases heart rate, and breathing out decreases heart rate. And as the vagus nerve contains the primary parasympathetic outflow to the heart, this allows us to measure HF rhythm as a measure of parasympathetic dominance – the resting state of the autonomic nervous system.
Low frequency
LF rhythms are less well understood. Original research from the 80s proposed that that LF, proportional to HF, measured the state of autonomic balance in the heart. However, this research has been persistently misunderstood. The original papers used this in the context of a tilt-table test, not a normal standing or lying HR measurement. That is, to get their ‘sympathetic’ changes the researchers tilted the experimental participant on a flat table to various angles, and measured the relative HRV present.
This provides various degrees of postural stress, which is carefully compensated by the activity of the baroreflex – the body’s system for regulating blood pressure. Research since strongly supports the fact that LF power is best understood as measuring the outflow of the baroreflex response to regular changes in blood pressure. While this definitely involves the sympathetic nervous system – which controls both the size of the blood vessels and heart rate – it isn’t by any means a straightforward relationship where we can measure sympathetic activity at the heart by measuring LF HRV power.
Now, while LF power is useful for research and an interesting phenomenon, it isn’t much good for making determinations about your training state for several reasons.
Firstly, it’s poorly understood and there is a lot of debate surrounding the mechanisms involved. Also, at rest we are more concerned with the absence of parasympathetic activity than the presence of sympathetic activity. If you are overtrained, your primary concern is the absence of parasympathetic activity. While the autonomic nervous system is usually either dominated by one or the other, there are a variety of situations where this isn’t the case. Lastly, we can control the parasympathetic signal we measure by the simple task of breathing at a certain rate – this gives us a consistent situation in which to measure our HF activity over time.As a consequence, the most informative and replicable measure of HRV in the short term is a paced breathing HF task. And that is what we use.
About the Author
James is doing a Ph.D in cardiovascular psychophysiology, improving HRV methods and applications. In his spare time, he lifts things up and puts them down.
More of the science behind ithlete
To read more about the science behind the ithlete app check out the How it Works page.
I practice a number of different forms of meditations and breathing techniques. I have been interested to see very high LF readings particularly for primordial sound meditation. I have searched for explanations but cannot seem to find any. Can you give me angry answers as to why for instance LF shoots up massively when the sound ah is repeated?
Hi Nigel,
I think the most likely explanation for this is that your HRV has become ‘coherent’ ie most of the power from breathing rate HRV has been concentrated at the baroreflex damped oscillation frequency of approx 0.1 Hz. This is right in the middle of the LF band, which runs from 0.04 to 0.15 Hz. The separation of LF & HF bands was established empirically for spontaneous breathing and there are a few circumstances where the breathing ‘HF’ component moves into the LF band, leading to high powers in that band. Hope this helps!
Hi Simon, thanks for the reply, I have only just seen it as there does not seem to be a system that notifies us of replies. Yes what you say makes sense. I do practice achieving coherence also.
Hi Nigel
Just come across your query. Sounds to that you are breathing at or near your resonant frequency which is somewhere between 4.5 and 7 breaths per minute.
When breathing at your RF it should concentrate power in the LF range. This also gives maximum baroreflex gain.
I determine the RF of clients that suffer from CF and/or FM. Overtime , with practice, health improvements will take place and it will increase their SDNN. SDNN is an indicator of the body to self regulate also resilience.
Cheers
John Coates
What is RF? Have CFS. Thanks
Hi Angela,
RF = Resonant Frequency. This is the frequency at which your breathing rate equals your LF HRV peak. Between 7 and 4.5 breaths per minute (average 6) – lower for bigger people.
Hi John only just spotted your reply sorry for the late delay. Yes I am breathing about 6 breaths per minute. I shall check if my SDNN shows any noticeable effects, when I get some time.
Hi there, not sure if this is still a live chat but wanted to ask anyway – does breathing 6 breaths per minute increase SDNN, as well as LF? And if so why is this the case? Thanks, Alison
I’m not sure if this thread is still live but I wondered why the SDNN also increases as well as the LF? I am also doing coherent breathing – 6 breaths per minute – and my LF is high and my SDNN is super high. I need to experiment doing spontaneous breathing and see how this changes things. Thanks :)
Hi Alison. The thread is still going as long as it’s useful! We certainly still stand behind the conclusions of the article. You can best think of SDNN as an ‘all inclusive’ HRV measure, so it will increase when any HRV parameter increases. LF actually measures the resonance of the baroreflex loop which regulates systemic blood pressure – coherent breathing stimulates this resonance & produces large HRV waves. Not to be confused with the paced breathing measurement ithlete does, which is for consistency of measurement and does not stimulate the baroreflex measurement.
Hi Simon (and others who posted replies above),
Thanks for your responses to a post that was a good few years old, very diligent!
I noticed the big wave pattern whilst doing my Kriya meditation last night, saw that it correlated with a massive spike in LF power and wondered what the heck could be going on. I’ve not yet seen enough of my own data to know how to interpret the SDNN numbers, but will now keep an eye on them.
Thanks again
George.
Hi George
Thanks also for your replies! The big wave pattern coinciding with a big spike in LF is (fortunately!) quit easily explained. Slow deep breathing (6 breaths / min or slower) excites whats called the baroreflex resonance, a little like whistling at the pitch that makes a wine glass sing. This frequency is ~0.1 Hz, and is therefore in the LF band, rather than the HF band where normal spontaneous and ithlete paced breathing sit. Although it’s low frequency, the parasympathetic nerves are very active, and that’s one of the reasons that meditation makes you feel calm.
The Breathesync app https://www.dobreathe.com/ uses this exact principle to achieve a relaxed state in only a couple of mins.
Hope this helps,
Simon
Hi Simon,
Many thanks for the reply. It’s great that you still take the time to respond to old blog posts like this!
I’ll check out the app.
Thanks again
George.