Training with supplemental oxygen - why it won't help at sea level

In the latest Ask a Cycling coach podcast the hosts discussed at some length the prospect of doing key sessions with supplemental oxygen to increase work capacity. Coach Jonathan (@Jonathan) seemed particularly excited by this, and concerned about potential muscular-skeletal damage from increased work capacity. Sorry to be the bearer of bad news, but I’m not sure this would help at sea level. Below is a theoretical arguement based on physiology, rather than a data based study, so feel free to prove me wrong with some solid data (@chad !)!

Haemoglobin (Hb) binds four oxygen molecules. The binding of one oxygen molecule increases Hb’s ability to bind further oxygen molecules, and so you end up with a sigmoidal oxygen (dis) association curve - see the curve here.

Supplemental oxygen shifts our position on the Hb-Oxygen association curve to the right by increasing partial pressure of Oxygen (ppO2). At low ppO2 (at elevation) this would increase the amount of oxygen in the blood. However, once all the Hb molecules are saturated with oxygen the blood stops being able to carry any more oxygen. At sea level the blood is already saturated with Oxygen (as explained in the podcast) and so any supplementary oxygen will not increase the amount of oxygen in the blood.

This would also explain why the studies that Chad cited used such high levels of supplemental oxygen - theoretically you would expect no change by getting the oxygen levels too high, but you would by getting the oxygen levels too low.

The one caveat with this argument is that some (very little) oxygen is dissolved in the blood plasma, and comes out of solution at the lower ppO2 found at the working muscles. This would continue to go up with increased lung ppO2, thus delivering slightly more oxygen to the muscles. However, this effect is pretty tiny compared to how much oxygen haemoglobin can deliver.

I hope this makes sense and clarifies things! Great podcast and product as always.


What causes oxygen toxicity?

Oxygen using social media too often


I’m with @wasket on this one.

Here’s an alternative explanation:

Max ppO2 is generally 1.4-1.6 ATM, although I’ve read that Navy divers go up 2.0.



Divers including myself can attest that when breathing at depth (ppO2 increases by 0.21 for every 33 ft), we feel zero aerobic benefit. It would be pretty cool if we became supermen/superwomen at 100 ft, but alas we feel exactly the same as we do at 10 ft.


Supplemental oxygen has no effect on the Hgb-ox dissociation curve. The only things that shift the curve are pCO2, pH, temp and a molecule called 2,3-DPG.

That being said, you cannot push oxygen saturation past 100%. If you have a good pair of lungs, that is most likely your saturation at room air. Supplemental oxygen will not make your Hgb carry more oxygen if it is already completely saturated. It will push your PO2 higher (dissolved oxygen in blood) but total dissolved oxygen is so little compared to oxygen bound to Hbg that in most practical situations one can completely ignore it.

If you have shit lungs due to smoking, are fighting an infection that impairs your oxygen saturation etc, supplemental oxygen may help. High concentration of oxygen for a long time has been shown to actually damage lungs.

This is so true and is a pet peeve of mine when I see other sports cough cough football have an O2 mask on the sideline. An extremely fit football player should have full saturation of oxygen, and even if he doesn’t (which would mean something’s very wrong) what happens a couple of breaths after they take the mask off and actually play?

1 Like

Yes. Sorry I wasn’t clear - increasing ppO2 changes the position along the x axis which is why I had said on the curve rather than moving the curve itself! :grinning:

Bringing me back to my undergrad physiology days…

That always baffled me… I am a bear of little brain, although I have spent too long on mole fractions and partial pressure calculations in my life: If we can squeeze more O2 into the bloodstream in such a manner, then how come our inhalation => exhalation O2 content goes from 21% to 16% and not closer to 0% implying a partial pressure deficit in the lungs? To me that relatively low O2 drop implies that in the gas exchange (O2hb <-> CO2Hb) CO2 removal is the limiting factor and not O2 uptake. :man_shrugging:t2:

The other element to that for me was I grew up playing rugby. My post match, mugged, deranged with exhaustion, ass thought that the episodic, body armored, discontinuous, unlimited substitution and tightly time constrained nature of Gridiron play actions may have had something to do it with! :stuck_out_tongue_winking_eye:

1 Like

I will say one thing in favor of supplement O2. In uni I was a combustion researcher and my lab had, naturally, bottled O2… CP not medical grade of course… That never stopped us using it as a hangover cure! It appeared to work… compressed the whole exercise. Never could work out why… perhaps it was just a collective panacea! The source of that method was old military pilot yarns.

1 Like

I am curious what happens to respiratory rate when exercising with supplemental oxygen. You are all correct that you cannot increase haemaglobin concentration past 100% and therefore supplemental oxygen cannot meaningfully increase blood oxygen carrying capacity, but perhaps oxygen could decrease work of breathing because each breath contains more oxygen. In that case, less energy is being used by respiratory muscles and there may be a performance benefit. Carbon dioxide clearance may be a limiting factor, I don’t know. I’m curious in theory only - I have no intention of trying it out myself and can see no real world benefit!

Good point - that might well be the case. I guess you would also expect stronger CO2 clearing as the concentration of CO2 in inspired air would be slightly lower, although I think this effect would be minimal given the very low concentration of CO2 in the atmosphere compared to expired air.

1 Like

Sorry, not picking on you, but either you are not being clear or I can’t understand what you are saying. Increasing PO2 will increase SpO2 upto an SpO2 of 100%. A PO2 of 60 mm Hg corresponds to SpO2 of 90%, barring any left or right shift in the curve. This curve is so well defined that when a blood sample is run, only the PO2 is measured, the SpO2 is actually a calculated value from the curve. So there is no shifting of the curve with an increasing PO2. We are probably just splitting hair.

TLDR; an increasing PO2 cause a rapid rise in SpO2 (that is the hemoglobin saturation) up to a PO2 of 60 mm Hg (leading to an SpO2 of 90%) After this an increasing PO2 causes a very slow increase in SpO2, and once the hemoglobin is 100% saturated, there can be no increase in SpO2. If anyone tells you otherwise, it is a marketing gimmick.

Hemoglobin affinity for CO2 actually changes with the change in blood O2 levels - it’s called the Haldane effect.

1 Like

One thought did occur to me. O2hb is not the only way O2 gets into the bloodstream: Good old Frick’s Law. It will be in equilibrium saturation in the plasma too. That’s not available for cellular respiration though is it? Might explain the hangover cure though: increased metabolite consumption.

It is available to diffuse down a concentration gradient, but the equation for blood oxygen carrying is: CaO2 = (1.34HbSats) + (0.003*PaO2) if measured in mmHg. The 0.003 means the amount is insignificant compared with bound oxygen.


There’s no scientific proof of that


Don’t forget that supplemental oxygen (in theory) decreases the metabolic costs of breathing heavily, which are significant.

Probably me not being clear, but we agree and I was trying to say what you are saying, but in a less clear way.

By saying a shift on the curve I meant by increasing ppO2 we shift how saturated our haemoglobin is at the lungs, which corresponds on the graph to a shift along the x axis and up the y axis. The amount that shift occurs is (as you say) described by the Hgb-O2 dissassociation curve. Hence, an increase of ppO2 at the lungs (e.g. by shifting from elevation to sea level, or by increasing %O2 in inspired air) can be visualised as a shift along the x axis, or a shift up the curve (from the point of view of the blood oxygen saturation returning to the heart from the lungs). I realise that wasn’t very clear when I first wrote it! Not sure this really clarifies, but we do agree :slight_smile: