Apocalypse Rebreather

[Dr Simon Mitchell]

Hello Brad,

We could continue this point by point debate forever, so I'm just going to make a few key observations.

You place great faith in the CE / client testing process and seem to be confident that it covers the issue I am concerned about. As far as I can see it does not. You seem incapable of accepting that the CE testing process would not cover every conceivable issue of potential importance, but I suspect that those responsible for it did not even think of this problem. In saying this I mean no disrespect; this is a very new field for CE testing. There are probably many areas in which it could be improved for new monitoring technologies in rebreathers, and it will take them a long time to get it exactly right. I should also say that testimonials to apparent appropriate operation from you and the tiny number of other users currently in possession of these units (or the commercial ones) are completely irrelevant to this debate.

I have no problem accepting that the DL CO2 sensor has been found to accurately measure PCO2. In other words, if you expose it to gas containing 5kPa of CO2 then it will read 5kPa. For the reasons we have pointed out this is different to it accurately reflecting the CO2 levels in the diver. There is much discussion of a complicated processing algorithm that will allegedly achieve this. A simple experiment similar to ours with the addition of recording what the pod says too would tell us whether it actually works, and I cannot believe it has not been done. Maybe it has, but no data have been released. Why not, especially in the face of the current controversy? You seem comfortable to dismiss the problem because “the only controversy seems to be generated by you et al ”, but here is reality Brad: DL are about to release a life support system with a crucial innovative feature. Context area experts have identified potential limitations in this feature and have published their concerns in a peer reviewed paper in the scientific literature. That’s a heady mix. I hope DL have got it right.

Just to provide perspective on this for readers who are fond of quoting me as saying the Apoc CO2 measurement system does not work. This is not what we have said. We have tested their configuration using non-Apoc components and found, as predicted in a debate with Alex Deas almost 2 years ago, that there are potential inaccuracies when the diver is taking shallow breaths. Our work actually suggests that the system will work under most circumstances (when larger breaths are taken). Nevertheless, in a life support product, especially one that bases autobailout on the relevant data, it needs to be right under virtually all circumstances that could conceivably arise. At present, it is not clear to me that this is the case. I am looking forward to getting a chance to test it, and I can’t work out why Brad has been prevented from providing his unit so we can do this.

Simon M

 

[rjack321]

Wow is this 2008? RBW?

 

[gianaameri]

Brad,

Does the APOC warn the user of CO2 retention?

A. Yes
B. No

A plain and straight answer would be appreciated.

 

[Rainer]

Is the King of France bald?

Brad,

Does the APOC warn the user of CO2 retention?

A. Yes
B. No

A plain and straight answer would be appreciated.

 

[Dr Simon Mitchell]

Brad,

Does the APOC warn the user of CO2 retention?

A. Yes
B. No

A plain and straight answer would be appreciated.

Hello,

I can answer this for Brad.

The simple answer is yes.

The more descriptive answer is as follows.

It is the CO2 in the blood (the arterial blood to be specific) which defines whether there is a CO2 problem (high or low). We cannot directly measure this with ease under any circumstances, let alone underwater. However, there is a key fact that opens up another possibility for monitoring; specifically, the pressure of CO2 dissolved in the arterial blood is equilibrium with the pressure of CO2 in the alveoli in the lungs. Thus, if we can measure the CO2 in the gas coming out of the alveoli then it serves as a reasonable surrogate for the arterial CO2. For this purpose, we assume that the very last part of an exhalation is coming from the deepest part of the lung, that is, from the alveoli. We call this the "end tidal" gas, and the pressure of CO2 measured in this gas ("the end tidal CO2") is an adequate approximation of arterial CO2.

It follows that measuring end tidal CO2 in the expired gas will tell you if the diver has high CO2 from any cause. CO2 retention is no different in this regard. CO2 retention occurs when we do not breathe enough to eliminate all the CO2 we are producing. Although we are not breathing enough to get rid of enough CO2, the CO2 in each breath will still reflect the actual arterial value. This is what the Apoc is attempting to do and it is a legitimate strategy.

The controversy around the Apoc has arisen as follows. The first part of the exhaled gas comes from the airways (excluding the alveoli) and the mouthpiece, and is often referred to as dead space gas. It contains little or no CO2 because gas exchange only occurs in the alveoli. Obviously, we do not want this dead space gas to contaminate the end tidal (alveolar) gas because it would dilute the CO2 and cause a falsely low reading. In anaesthesia we solve this problem but sampling the end tidal gas as soon as it comes out of the mouth. By sampling at the mouth at the end of the exhalation we can be confident that the dead space gas has disappeared down the breathing circuit and what we are sampling at the mouth is alveolar gas. We thus get a good estimate of the arterial CO2 as described above. Our concern is that in the Apoc the end tidal gas is not sampled at the mouth, but rather at the end of the exhale hose. This hose provides an opportunity for the expired dead space gas to mix with the alveolar gas and raises the possibility that the CO2 measured at the end for hose is falsely lowered. This would be more likely if the diver takes small breaths because the dead space volume does not change. Put it like this: if the dead space (exhaled first) is 200ml and the total breath size(tidal volume) is 2000ml, then virtually all of the dead space gas will be washed out of the breathing hose before the CO2 monitor makes the measurement at the end of the hose. The system will work under these circumstances. However, if the dead space is 200ml but the tidal volume is only 500ml, then it is likely that there will be significant mixing in the exhale hose and the measured CO2 will be lowered.

The latter is what our study confirmed. We found that with tidal volumes down to 1000ml the CO2 measured at the end of the hose was very close to the true end tidal CO2 measured at the mouth. Lower than 1000ml and the end of hose measurement becomes inaccurate. The significance of this is uncertain because we don't know a lot about the size of breaths typical of normal diving. It is plausible, however, that small breaths may be a symptom of CO2 toxicity and so at the very time you want the monitor to be accurate, there is a risk that its accuracy will decline. This, of course, takes no account of the possibility that DL have managed to design an algorithm that compensates for this problem. They claim to have done this, but I would like to see proof that it works.

I hope this all makes sense.

Simon

 

[gianaameri]

For this purpose, we assume that the very last part of an exhalation is coming from the deepest part of the lung, that is, from the alveoli. We call this the "end tidal" gas, and the pressure of CO2 measured in this gas ("the end tidal CO2") is an adequate approximation of arterial CO2.

Thank-you.

Now I understand how a single CO2 sensor correctly located... could in theory provide an indirect measure of arterial CO2 subject to the above assumption being true.

Is there any study which has verified the assumption quoted above (i.e. is it an hypothesis scientifically proven to be true)?

As a layman, I would have thought tthat o measure any amount being "retained" one would need to know reliably the amount going in, and the amount going out, the difference being the retained amount (hence the need of a CO2 sensor on the inhale, a CO2 sensor on the exhale, an accurate measurement of O2 metabolised, and an accurate measurement of CO2 produced).

 

[Brad_Horn]

We could continue this point by point debate forever, so I'm just going to make a few key observations.

Simon, It is kind of going around in circles as neither of us can currently predict the end result of your testing of the actual OR CO2 monitor in the iCCR, Incursion or Umbilical.

You place great faith in the CE / client testing process and seem to be confident that it covers the issue I am concerned about. As far as I can see it does not.

Done properly, the testing required for the CE certification does appear quite strict and in depth. I think the testing covers the validation that the CO2 monitor works as required. I agree with you that it does not provide a specific test that answers your concern.

You seem incapable of accepting that the CE testing process would not cover every conceivable issue of potential importance, but I suspect that those responsible for it did not even think of this problem. In saying this I mean no disrespect; this is a very new field for CE testing. There are probably many areas in which it could be improved for new monitoring technologies in rebreathers, and it will take them a long time to get it exactly right.

Agreed, which si why I have asked how you are going submitting this testing requirement to the SC7 et al committee's involved in the drafting of the minimum standards for rebreather performance.

I should also say that testimonials to apparent appropriate operation from you and the tiny number of other users currently in possession of these units (or the commercial ones) are completely irrelevant to this debate.

Agreed and any testing not witnessed by a notified body done in a certified, calibrated and fully equipped hyperbaric rebreather testing facility capable of testing down to 350m* would also appear pretty irrelevant for the same reasons.

*350m being the rated limit of the DL CO2 monitoring as I understand it except on the Apoc when its 100m due to available power.

A simple experiment similar to ours with the addition of recording what the pod says too would tell us whether it actually works, and I cannot believe it has not been done. Maybe it has, but no data have been released.

As I read the DL reports it has been and the recorded information compared against their mass spec reading. This testing and published data appears to be reflected in the flapper valve failure report just on a breathing simulator at 100m as opposed to a human in a dry lab. Looking at the respiratory rate report, their testing was done on the surface.

Context area experts have identified potential limitations in this feature and have published their concerns in a peer reviewed paper in the scientific literature. That’s a heady mix. I hope DL have got it right.

I also hope DL have it right. What I am interested in is if the potential limitations you et al have identified are actual or a not a factor.
By not a factor, what is missing from your report is consideration of the conditions when the diver may not actually still be on the Apoc on closed circuit. Such as when the diver is taking shallow breaths does this cause the ppo2 to raise to an alarm condition bailing the diver to OC.

I can answer this for Brad.

The simple answer is yes.

Thank you Simon, cause I sure as heck couldn't have answered him.

We found that with tidal volumes down to 1000ml the CO2 measured at the end of the hose was very close to the true end tidal CO2 measured at the mouth. Lower than 1000ml and the end of hose measurement becomes inaccurate. The significance of this is uncertain because we don't know a lot about the size of breaths typical of normal diving.

How are you et al resolving the unknown of not having a sample of the size of breaths typical in normal diving to know if a tidal volume of less then 1000ml is in fact an issue or just a hypothetical?
If divers don't typically have a tidal volume of less then 1000ml on rebreathers this whole issue would just seem to be a lot of hot air over nothing.

Kind regards
Brad

 

[gianaameri]

Thank you Simon, cause I sure as heck couldn't have answered him.

Brad,

can we please have an official answer from Dr. Deas on this critical question (Dr. Mitchell can only express an opinion on the technology (i.e. Capnography) as a concept, but not on the APOC CO2 monitor itself as he has never seen one, nor anybody else has, let alone tried and tested one).

Does Dr. Deas represent and warrant under his personal responsability (he is the signatory of the EN61508 APOC Certification) that the APOC End-tidal CO2 Monitor reliably and accurately warns the diver of CO2 Retention?

 

[Oskar Frånberg]

I was reading through the manual for the Apocalypse published on the open safety web page and came across two puzzling statements.

Under 5.6 CO2 monitoring limits
“The CO2 monitor is calibrated on your exhale breath during the pre-dive
checks. The monitor looks for a normal oxygen consumption rate, steady
and normal respiratory rate, and then looks at what the CO2 sensor output.
It allows a calibration only over a small range, to prevent a faulty sensor
being calibrated.”

And under 9.7.2 Calibration

”When your CO2 monitor is calibrating, the Diver Display will tell you
“Carbon dioxide cal underway”. By the time you react, the calibration is
finished. If your buddy looks at your CO2 display now, it shows 4.00%.
What this means is the end of breathe exhaled CO2 is 4kPa: the percentage
is a reference to the amount of CO2 in your breath at normal atmospheric
pressure (Surface Equivalent Value, SEV).”

I do not know but to me it sounds like the sensor is being calibrated using the end tidal CO2 level of the diver during prebreathing the unit. Looking through some old UBR I found this, Kerem D, Melamed Y, Moran A Alveolar Pco2 during rest and exercise in divers and non-divers breathing O2 at 1 ATA. Undersea Biomed Res. 1980 Mar;7(1):17-26. The measured end tidal CO2 in rest for divers and non-divers was 4.7 ± 0.4 kPa and 3.9 ± 0.7 kPa respectively.

If all this is considered to be 4 kPa then the accuracy of this is so low that it doesn’t matter if the method for determining end tidal co2 is correct or not.

Oskar

 

[berk]

Yes,

Sitting down; standing up? Perhaps this person calibrating just took a quick cool down in their drysuit with the vent open and now it's stuck to them like cling wrap.

The reservations about the CE test being valid as outlined in Simon's syphilis analogy is really the tip of the iceberg and first step.

As a hunter, yoga practioner and lover of skirt steak, the Thoracic cavity and lungs seem way too dynamic for what they are trying to accomplish. The basic outline I can't debate since end tidal co2 seems to be an accepted proxy for blood levels. But what does end tidal mean in the real world? I can see false positives and negatives despite all their assertions being valid.

I got some evil stares from my wife as I was pouring 333ml beer glasses of water into my counter lungs.