Oxygen solubility and half times

[Piscean]

Gentlemen,

I was wondering about Dr. Deco's comment that it takes a while for the body to saturate with O2 and so I was wondering if we are not worring about something which won't happen. I assume it takes more than a few minutes even for our blood to saturate. It takes about a minute for our blood to go around our body once and of course during that time 1 to 1.5l of O2 are removed from it by our metabolism. So even if the spike in O2 concentration does occur (and I am fairly convinced by DPT's mathematics thus far) I suspect that it may never get to our brains. Sound plausible?



Piscean.

 

[devjr]

Piscean, Hi:

Yes, agreed. Dr Deco offered those numbers for us to chew on. However, Dr P and I got involved in some partial pressure arm wrestling. It was handicapped by conflicting premises and different units of measurement. If the initial conditions are loaded towards high O2, then there is agreement in theory. Of course, metabolism is significant, as you noted.

 

[Piscean]

Devjr,

Make no mistake, I enjoyed the arm wrestling. Actually I had written another reply but my session timed out and I lost it. The one thing I forgot to put in on the second time of writing is that I am still a bit bewildered by the fact that we don't see anything on the oxygen partial pressure monitor in the loop. Surely the meter can't be that slow can it? I didn't necessarily want to suggest that the high PO2 doesn't occur in the loop, only that it never gets to your brain. I guess how high it goes in the loop depends a lot on the original O2 percentage and how fast one descends.

Piscean.

 

[The Iceni]

Piscean wrote

"I was wondering about Dr. Deco's comment that it takes a while for the body to saturate with O2 and so I was wondering if we are not worring about something which won't happen. I assume it takes more than a few minutes even for our blood to saturate. It takes about a minute for our blood to go around our body once and of course during that time 1 to 1.5l of O2 are removed from it by our metabolism. So even if the spike in O2 concentration does occur (and I am fairly convinced by DPT's mathematics thus far) I suspect that it may never get to our brains. Sound plausible?"

I am not so sure, my fishy friend!

Dr deco confirmed that about 1 litre of oxygen could be dissolved in an average body for every increase in pressure of one atmosphere (22 mls/Kg/bar) but that this could take several hours. However, this represents an average for the whole body but blood and brain are both fast tissues and will rapidly reach saturation, possibly within a few seconds.

I surmise, and admittedly have no evidence to justify this assumption, that the amount of oxygen rapidly dissolved in these fast tissues to reach saturation (and the quantity already dissolved in the still unsaturated slower tissues) will be quite insufficient to reduce the pp O2 in any practical way simply because the dissolved quantities are so small. It seems likely therefore, that the partial pressures seen in the blood, and therefore the brain, will closely reflect any changes in the partial pressure of inspired oxygen, even if the slower tissues are far from the steady state. Remember the pp O2 is reduced by solution.

We have still not explained why the recorded descent spikes - of raised pp O2 - do not closely reflect the predicted increases. I do suspect that sensor delays must be implicated here. The same sensors are used in Nitrox oxygen meters and I often find it takes several seconds for the readings to stabilise when I check my own Nitrox cylinders.

We are, however, left with a conundrum!

Whether the recorded spikes are accurate or not it seems perfectly clear that if Heliair 12 were used as the diluent the actual spiking would be lower simply because Heliair 12 contains about half as much oxygen as air. (At first sight, Heliar 12 is a sensible choice for diluent simply because it can safely be used as a bail-out gas because it can support life at surface pressures; pp O2 0.12 bar).

However, does this make Heliair 12 a safer diluent than air?

My first reaction is probably not, as I suspect the added hazards that accrue when helium is used for diving, with respect to DCI, will far outweigh any added protection from the minimal but real risks of oxygen toxicity from spiking on any descent when oxygen-containing air is used as the diluent.

Kind regards,

Paul

 

[Piscean]

Doctor Paul,

I have been thinking some more about this problem. I found the evidence you were lacking that states that the blood in the lungs reaches equilibrium in 0.3 seconds (http://www.uel.ac.uk/life-sciences/resources/pph243-respiration.htm) so it seems that you are right, the ppO2 in the lungs will be reflected quickly in the brain. There is an interesting effect associated with the haemoglobin (sort of). In one of your other posts (I can’t find it at the moment), you gave a very nice description of the oxygen cascade as it enters our cells. I found other descriptions on the web which agreed with you. There you said that the ppO2 in our blood is about 100 Torr (0.13 bar), on the web I found that it is about 40 Torr (0.05 bar) in our veins (i.e. after the O2 required for metabolism has been delivered to its destination). At http://www.mtsinai.org/pulmonary/ABG/PO2.htm I found this formula for the O2 content of blood:

CaO2 = Hb (g/dl) x 1.34 ml O2/g Hb x SaO2 + PaO2 x (.003 ml O2/mm Hg/dl)

Or in metric
CaO2 = Hb (g/dl) x 1.34 ml O2/g Hb x SaO2 + PaO2 x (2.28 ml/dl/bar)

CaO2 is the O2 content of arterial blood; SaO2, the fractional saturation of the haemoglobin; PaO2 the partial pressure of O2 in the arterial blood. Hb is haemoglobin and Hg is mercury.

Reading off the graph found in the document above which shows blood oxygen content for haemoglobin contents of 10 and 15g/dl (http://www.mtsinai.org/pulmonary/ABG/O2curve-large.jpg) and assuming 15 g haemoglobin per 100 ml of blood (typical for a man), we can see that approximately 6 ml of O2 are delivered to the body by 100 ml of blood. I postulate that the ppO2 in our tissues must be less than or equal to that in our veins (otherwise it would never have got that low in our veins – the O2 has to have a partial pressure gradient to make it go anywhere). Now assuming the amount of O2 required by our bodies is constant over time (= nice relaxing dive), as we descend, and the ppO2 increases more and more of this can be delivered by simple solution in the plasma. I’m too lazy to work out a model for the PaO2 based on inspired ppO2 so let’s assume they are equal. If that is the case, in order to deliver 6 ml of O2 by simple solution we would have to inspire air with a ppO2 = 6/2.28 = 2.6 bar. (I have ignored the 3% unsaturated Hb in arterial blood at ppO2 100 Torr and the residual O2 in solution in the venous blood for simplicity and on the basis that they roughly cancel each other out.) Since we are delivering all of the O2 via the plasma and not via the haemoglobin, the haemoglobin remains saturated (97% at least) and therefore the ppO2 in the venous blood must be about 100 Torr (= 0.13 bar). So, the interesting effect I was talking about is that despite an increase of 2.4 bar in the ppO2 in the lungs, the ppO2 in the tissues has only increased by 0.08 bar, not much at all (albeit 160% higher than it had been at the surface when breathing ppO2 0.21 bar). Of course now if the ppO2 increases further, the ppO2 in our venous blood and I assume therefore tissues also, will increase at roughly the same rate as the inspired ppO2 because the haemoglobin is always saturated and can no longer absorb the extra O2, preventing the ppO2 increase. If we breathe ppO2 3 bar then I suggest the ppO2 in our tissues will be 3-2.6 = 0.4 bar. (The 2.6 difference comes from the 6 ml of O2 that gets metabolised.) So there you have it. I am not sure where this leaves us. I venture to suggest that it means a brief spike in ppO2 is not really dangerous unless it goes above 2.6 bar.

Any more thoughts?

I suppose I didn't really address the sensor thing, but I agree with you, it must be sensor delays - are the sensors "flow-through"? If they are not then I would expect delays of a few minutes perhaps due to diffusion. And I don't know anything about heliox diving (then again, a month ago I didn't know anything about haemoglobin either), I'll have to postpone that one for later, perhaps devjr could help us?

Piscean.

P.S. A joke for Dr. Paul:
Q:What did the rebreather diver say when his ppO2 dropped to 0?
A:I think I'm going to Suffolk....

(Suffolk, suffoc...ate get it? )

 

[The Iceni]

Hi Piscean

Lost Yooper asked about oxygen narcosis, to which I contributed a long dissertation on how I thought oxygen must behave in the body at high partial pressures.

It is at

http://www.scubaboard.com/showthread.php?postid=44833&t=5097#post44833


One of your fellow countymen, McGilvery has produced an excellent biochemistry text book, "Biochemistry a Functional Approach", in which he produces a very nice graph of the pressure changes in the tissues compartments, to which you refer.

Yes, I think you are right. Haemoglobin must be involved in oxygen tranport even at the high pp O2 seen in diving since it is only fully saturated at a pp O2 in excess of about 3 bar.

Thus the risks of oxtox are certain when inpiratory pp O2 approaches 3 bar. This is shown in graphical form in "Diving and Subaquatic Medicine", Edmunds, Lowry & Pennefather . third edition , p 242. (ISBN 7506 2131 1) - There now exists a fourth edition which I do not have, I'm afraid.

Why it is so variable and manifests itself at lower pressures than this is quite fascinating.

I'll take another look at your figures for haemoglobin carriage and get back to you if I think we are both on the wrong tack.

Kind regards,

Paul

By the way, Avoid Suffolk like the plague!

People only come to East Anglia if they are obliged to.
You do remember the film "Deliverance" don't you?

 

[Piscean]

Hello Paul,

Just me again, I wanted to check that we are on the same page so to speak. First, I think that haemoglobin plays less and less of a part in O2 transport as the ppO2 increases, once the ppO2 is up to 2.6 bar or so, I think the haemoglobin plays almost no part at all. The point I wanted to make is that in order to deliver enough O2 to the body in the plasma a large change in ppO2 is required because the amount of O2 dissolved per bar ppO2 is only 2.3 ml or so. Thus though the ppO2 in the lungs is very high, it drops very rapidly as O2 is delivered to the tissues (amount of O2 = change in pressure * solubility per unit pressure). At ppO2 below 2.6 bar in the lungs, not enough O2 is dissolved in the plasma and so the ppO2 near the tissues drops below 0.13 bar and the oxyhaemoglobin starts to release oxygen with only a very small additional partial pressure change. Once it is greater than 2.6 bar in the lungs the required O2 can be completely delivered by the plasma and so the ppO2 starts to increase rapidly because a small amount of extra O2 dissolved in the plasma means a large change in partial pressure (by the formula above).

I hope when you read that you thought "Yes, that's what I thought he mean't", the bit about haemoglobin being involved in O2 transport even at high ppO2 made me think that I might not have been clear enough.

Thanks for the references, I don't know where my nearest library is that could get them for me so it could be a while before I read them. I'd like to learn more about ox tox though so I will have to find them somehow. You are right about all this stuff being fascinating. My next crazy theory is that the CNS side of ox tox could be explained in some way by my sums above. For example, tolerance is higher at low activity rates, whereas excercise increases susceptibility. Since my basic physiology tells me that the brain and muscles have essentially separate blood supplies (not many big muscles in the head) you could think that the increased blood flow would mean that more blood flows past the brain, but doesn't have the O2 sucked out of it by any muscles. Since the brain still only requires the same amount of O2 it doesn't need to take 6 ml from every dl of blood, but less, as a consequence, the ppO2 increases rapidly and so you get ox tox earlier. I would be interested to hear what Dr. Deco has to say about that. I hope you will both excuse me for doing physiology like a chemical engineer - I didn't take much biology after O-level.

On a more personal note - I actually spent 3 years just over the county border at Churchill College (in a town with a bridge in it) when I was an undergrad and managed to avoid Suffolk the whole time I was there (limited range on my bicycle). Never seen "Deliverance" though. I assume also that McGilvery is American and thus not one of my countrymen, since I am a countryman of yours.ut: (Here's to the 20oz pint!)

Cheers,
Piscean

 

[The Iceni]

Hi again Piscean,

The links you gave were very interesting. If I may plagiarise what Lawrence Martin says, the delivery of oxygen to the tissues by the blood depends on

1) The haemoglobin (Hb) concentration (normally between 11 and 15 g/dl)

2) The % saturation of this Hb, which can carry 1.34 ml O2 per gram, when fully saturated. (14.7 - 20.1 mlO2/dl)

3) The amount of oxygen carried in simple solution @ 2.28 ml per dl per bar O2.

4) The inspiratory partial pressure of oxygen, which determines "2" and "3".

Thus; CaO2 ml/dl = (Hb x 1.34 ml x SaO2) + (PaO2 x 2.28)

As you say the arterial pp O2 is about 100 mmHg (0.13 bar) and the mixed venous blood about 40 mmHg (0.053 bar) - a difference of less than 0.1 bar. The amount of oxygen in simple solution at these pressures is 0.29 ml/dl in arterial blood and 0.12 ml in mixed venous blood. Thus only about 0.17 ml/dl is delivered to the tissues from simple solution on the surface breathing air, which is quite insufficient to support metabolism. As little as 1% of the total haemoglobin is capable of delivering about the same amount of oxygen from solution with a pressure differential of 0.1 bar between arterial and venous blood.
This is why haemoglobin is so important.

Haemoglobin acts like a liquid sponge soaking up and carrying much more oxygen than could ever be carried in solution, which it then released at the tissues. When breathing air on the surface, with the arterial haemoglobin at 97% saturation at 100 mmHg, arterial blood carries about 20 ml O2 per dl, while the haemoglobin in venous blood is about 70% saturated (at 40 mmHg) so it contains about 14.2 ml O2 per dl.

Complications to understanding arise because the amount of O2 required for metabolism by the various body's tissues is not the same nor is the blood supply. For example the brain has a metabolic rate about the same as, or higher than, that of active muscle and it also has a proportionately much higher blood supply than any other organ. In consequence, venous blood from an active limb will have a ppO2 much, much lower than the average of 40 mmHg seen in mixed venous blood returning to the heart and in the jugular veins it could be higher, but on average, as you say, the blood delivers about 6 ml of oxygen per decilitre.

As the inspiratory pp O2 increases to multiples of 1 bar, more and more can be delivered by simple solution but haemoglobin will still be involved. Because it carries so much more than is dissolved haemoglobin remains important right up to the point where it remains fully saturated in the venous blood.

You say " If we breathe ppO2 at 3 bar then I suggest the ppO2 in our tissues will be 3-2.6 = 0.4 bar. (The 2.6 difference comes from the 6 ml of O2 that gets metabolised.) " I am not sure this is strictly true. While this may reflect what is seen in the blood, all tissues have a lower oxygen tension than seen in venous blood. In addition there are the confounding variables of differing tissue metabolic rates, variable blood supplies to those tissues and in addition the auto-regulation of blood supply to the brain. (CNS oxygen toxicity is not seen outside the brain!)

Using the formula and an Hb of 15, I estimate that when the inspiratory ppO2 is 3 bar, arterial blood contains 27 ml O2 per dl - 20.1 mls combined with Hb (which is 100% saturated) plus 6.9 mls in solution. If an average of 6 ml of oxygen is consumed mixed venous blood will contain 21 ml O2 per dl (19.67 mls combined with Hb (98% saturated) plus 1.3 mls in solution). This is the same as arterial blood when the inspired pp O2 is 0.6 bar. Please see the attached Excel spreadsheet (If it works on this site! - still trying!)

Haemoglobin only remains fully saturated when the ppO2 exceeds 3 bar but this has major adverse consequences to the carbaminohaemoglobin buffer and to the acid-base balance at tissue level, particularly in metabolically active tissues, such as the brain. This is because it is no longer available to "mop up" excess oxygen to reduce the pp O2, but is also no longer available to absorb and deactivate acidic carbon dioxide molecules, generated by oxydative phosphorylation in metabolically active tissues such as the brain. Any increase in acidity shifts the haemoglobin dissociation curve to the right, releasing molecular oxygen and effectively increasing the local ppO2 even more. I believe this is why carbon dioxide retention is so heavily implicated in the genesis of oxygen toxicity.

I understand none of the sensors in current rebreather models are "flow-through", they simply protrude into the inspiratory side of the rebreather loop. I believe the electronics "average out" these readings before more oxygen can be injected into the loop.

I entirely agree that a brief spike in ppO2, whether recorded on the sensors or not, ought not to be really dangerous unless it goes above 2.6 bar. However, without reliable data on the actual changes seen in the brain in life this is entirely speculative since we know oxygen can be toxic at much lower levels than this and we also know that susceptibility is so variable.

The problem as I see it is, at depths close to the MOD of the diluent it may be all but impossible to perform an efficient diluent flush quickly to return the inspiratory pp O2 to recognised safe levels.

I am not a rebreather diver and this is for discussion purposes only.

Paul

PS Deliverance - dueling banjoes?

 

[The Iceni]

*****!!!!

 

[Dr Deco]

Paul:

I truth, I am putting my money on the idea that much of the problem lies in the time constant of the oxygen probe (sensor delay). You folks seemed to have touched on that also.

In my experience, it requires about one minute for an oxygen sensor to give a relatively good reading. The 50% value is reached in about 20 seconds.

Dr Deco
:doctor: