Deep Stops Increases DCS

[PfcAJ]

No. Ross isn't arguing what you say above. He and I are arguing that the deep stops the NEDU study used are fundamentally different from sport bubble models. So different, in fact, that from the perspective of VPM or RGBM they should be called long shallow stops.

So the deeper shallow stops cause more dcs.

Okay...

 

[shoredivr]

Why Simon Mitchell, decompression researcher, thinks the NEDU study is relevant to sport technical divers.

Deep stops debate (split from ascent rate thread) - Page 108

"I have said it before..... this is why the NEDU study is relevant to us. It legitimately challenges the philosophy that protecting the fast tissues from supersaturation early in the ascent is the defining component of efficient decompression. The most plausible explanation for the NEDU result is the differences in integral supersaturation in fast and slow tissues revealed in the beautifully depicted analyses of A1 and A2 on this thread, and as has also been demonstrated, it cannot be argued that these profiles are irrelevant to the deep stop profiles used by technical divers. "

 

[Dr Simon Mitchell]

Hello,

I have dedicated enough time to arguing with Ross Hemingway previously. That is not to say that I will not do so here if necessary, or in future on new matters of importance, but everything one needs to know on the “deep stops” subject (to date) is already clearly articulated on the internet. My advice is to go to the rebreatherworld threads which others have linked to if you really want a comprehensive appreciation of the subject. I have little appetite (or time) for getting into debates about the details again but I will make some relatively general comments.

To those who suggest that decompression is an inexact science, you are right. But it is nonetheless a science, and this particular area of scientific focus is slowly producing real data that inform the debate taking place here. Thus far, every published study that has objectively investigated the use of deep stops in decompression diving has produced a signal that their hypothetical advantages do not play out in reality. No one pretends to know what optimal decompression is, but we are becoming increasingly aware of what it isn’t. There is a strong signal in the available data that over-emphasising deep stops (which almost certainly includes the level of “deep stopping” prescribed by bubble models) is disadvantageous. I am aware (through attending presentations and collegial discussions) of two further studies involving tech diving-relevant profiles with differing emphasis on deep stops that are being prepared for publication. These are not my studies and I will say nothing further about them, but Ross will find them “interesting”.

Against the emerging evidence which indicates that we need to rethink the deep stop mantra there is NOTHING. There has NEVER been evidence that emphasising deep stops is a good idea. It was never more than a theoretically attractive concept that gained wide traction in the technical diving community for a variety of reasons. In this regard I am intrigued by comments like:

On the contrary there is plenty of anecdotal evidence that points to the reasonable effectiveness of deep stops

If this means that a lot of dives incorporating deep stops have been done with no DCS then yes, of course. But this in no way informs the question about the relative risk of profiles which emphasise deep stops to differing degrees. I want to be unequivocally clear on this. There is an expanding body of objective studies producing a confluent signal: over-emphasising deep stops increases risk of DCS, against which (to date) there is NOTHING.

My own position on this matter is a simple reflection of this situation and that is also true of David Doolette and other scientists involved in the area. Ross’s portrayal of us as anti-VPM conspirators is simply insane; we are just scientists following the evidence as it evolves. If the evidence was favouring deep stops then we would be leaning that way, and if the evidence were to swing back toward deep stops then I would swing back with it.

Ross, of course, argues that the most influential study to date (the NEDU study) was deeply flawed and repeatedly states that it was so badly designed that it did not even test deep stops despite its stated intention to do so. There is reason enough to challenge the face validity of that position, held as it is by a helicopter mechanic / deep stops salesman, whereas those who designed the study are arguably the world’s most respected decompression scientists, none with a commercial agenda.

But leaving the “eminence based medicine” and conflict-of-interest arguments aside, it is an aspect of his argument that resonates with some readers (including several on this thread) who struggle to understand the relevance of the NEDU study to typical technical profiles. This is something that uwsojourner has toiled hard to overcome, and his post 681 is highly recommended.

Deep Stops Increases DCS

My quick summary of the key concepts goes like this:

Even though the NEDU study profiles don’t seem “typical” to the average technical diver, they do the things that we expect of the respective profile types. The deep stops profile protects the faster tissues from supersaturation early in the ascent (which is one of the fundamental planks of deep stop theory), but at the cost of greater supersaturation in the slower tissues later in the ascent. The shallower stops profile results in more supersaturation in the faster tissues early in the ascent, but has less supersaturation in the slower tissues later in the ascent and at the surface. These patterns of supersaturation distribution are an inevitable consequence of having a finite amount of decompression time and choosing to emphasise either deep stops or shallow stops.

If deep stop theory was correct, then the protection of fast tissues early should have produced some degree of protective effect, but the opposite was seen. The profile with the higher supersaturation early was actually the safest. Moreover, when the respective profiles were evaluated for overall exposure to supersaturation (using the integral supersaturation approach explained by uwsojourner) the deeper stop profiles resulted in greater supersaturation exposure. This passes the face validity test. Supersaturation is what drives bubble formation, and bubbles are the vectors of primary injury in DCS. More supersaturation overall is therefore likely to be bad.

The relevance to “real” tech diving profiles is that the same patterns of supersaturation distribution and overall supersaturation can be seen when profiles of equal length but emphasising deep stops to a greater or lesser degree (eg as in VPM vs GF approaches – see uwsojourner’s post which I have linked to above) are compared. Why would we expect the outcomes to be any different in this setting where the supersaturation patterns are the same? The difference might not be as exaggerated as in the NEDU study (which was designed to be intentionally provocative), but the “truth in the universe” if we could do a big enough study is highly unlikely to be any different.

Can I be clear that my argument is not that endlessly shallow stops are better. Someone earlier made the reductio ad absurdum point that the ultimate shallow stop profile is to simply come shooting straight to the surface and not make any stops. Clearly you can’t do that! So, one could conclude that all decompressions are “deeper” than the ultimate shallow stops profile. It is the degree of “deep stopping” that is under question, and in particular, whether the approach prescribed by bubble models (which certainly emphasise deep stops) is optimal. The best evidence we have at this point in time suggests that bubble models over-emphasise deep stops. Exactly how far we should back away from deep stops is presently not clear. That work has not been done.

Simon M

 

[EFX]

I'm trying to understand your position. I think the arument you are making is that if they had done less deco and done it deeper they would have bent less divers? Does doing overly long deco typically bend divers? Do you expect to get a higher percentage of bent divers if you did 30 minutes at 20 feet vs 10 minutes at 20 feet that the dive table called for?

Nope. What I'm saying is if you redistribute the total deco time so that you do more time deep but less time shallow (as compared to the shallow profile) not only will you bend divers, you will bend rebar. Look at the two profiles from my post #698. If you enter the shallow stop profile in whatever dive planning software you care to use and then add stops at 70, 60, and 50 feet, what changes will the program do to the original shallow stop profile? It's going to add time to those shallow stops. But look at the NEDU profile. It reduced the shallow stop time. They did this to keep the overall deco time constant.

I'll have more to say on this tomorrow. I'm too tired to continue.

 

[Ayisha]

Thanks Dr Mitchell for taking the time again to explain everything so clearly!

 

[rossh]

...... that if they had done less deco and done it deeper they would have bent less divers? Does doing overly long deco typically bend divers? Do you expect to get a higher percentage of bent divers if you did 30 minutes at 20 feet vs 10 minutes at 20 feet that the dive table called for?

I think that is true. Real world divers do this already. Every planning tool and method we use today, or before, makes somewhere about half the time of the Nedu test. The current published USN rev 6 table time is about half the nedu test. Add to that the existing test data from olden days, that is far shorter than this. Add anecdotal - thousands of dive examples in all kinds of models with half the deco time.

So absolutely "....if they had done less deco and done it deeper they would have bent less divers."

"Does doing overly long deco typically bend divers?"

Yes it can. Stay longer, get colder, slow down circulation, and create a DCS where none should normally exist. Called thermal stress. These nedu test divers were deliberately made cold. Tech divers often encounter this condition with a dry suit flood in cold water. The affected body part afterwards will exhibit pain and extra stress or maybe a mild DCS from the cold.

Thermal stress was tested and reported here: The Influence of Thermal Exposure on Diver Susceptibility to Decompression Sickness.

.

 

[rossh]

Hello,

I have dedicated enough time to arguing with Ross Hemingway previously. That is not to say that I will not do so here if necessary, or in future on new matters of importance, but everything one needs to know on the “deep stops” subject (to date) is already clearly articulated on the internet. My advice is to go to the rebreatherworld threads which others have linked to if you really want a comprehensive appreciation of the subject. I have little appetite (or time) for getting into debates about the details again but I will make some relatively general comments.

To those who suggest that decompression is an inexact science, you are right. But it is nonetheless a science, and this particular area of scientific focus is slowly producing real data that inform the debate taking place here. Thus far, every published study that has objectively investigated the use of deep stops in decompression diving has produced a signal that their hypothetical advantages do not play out in reality. No one pretends to know what optimal decompression is, but we are becoming increasingly aware of what it isn’t. There is a strong signal in the available data that over-emphasising deep stops (which almost certainly includes the level of “deep stopping” prescribed by bubble models) is disadvantageous. I am aware (through attending presentations and collegial discussions) of two further studies involving tech diving-relevant profiles with differing emphasis on deep stops that are being prepared for publication. These are not my studies and I will say nothing further about them, but Ross will find them “interesting”.

Against the emerging evidence which indicates that we need to rethink the deep stop mantra there is NOTHING. There has NEVER been evidence that emphasising deep stops is a good idea. It was never more than a theoretically attractive concept that gained wide traction in the technical diving community for a variety of reasons. In this regard I am intrigued by comments like:



If this means that a lot of dives incorporating deep stops have been done with no DCS then yes, of course. But this in no way informs the question about the relative risk of profiles which emphasise deep stops to differing degrees. I want to be unequivocally clear on this. There is an expanding body of objective studies producing a confluent signal: over-emphasising deep stops increases risk of DCS, against which (to date) there is NOTHING.

My own position on this matter is a simple reflection of this situation and that is also true of David Doolette and other scientists involved in the area. Ross’s portrayal of us as anti-VPM conspirators is simply insane; we are just scientists following the evidence as it evolves. If the evidence was favouring deep stops then we would be leaning that way, and if the evidence were to swing back toward deep stops then I would swing back with it.

Ross, of course, argues that the most influential study to date (the NEDU study) was deeply flawed and repeatedly states that it was so badly designed that it did not even test deep stops despite its stated intention to do so. There is reason enough to challenge the face validity of that position, held as it is by a helicopter mechanic / deep stops salesman, whereas those who designed the study are arguably the world’s most respected decompression scientists, none with a commercial agenda.

But leaving the “eminence based medicine” and conflict-of-interest arguments aside, it is an aspect of his argument that resonates with some readers (including several on this thread) who struggle to understand the relevance of the NEDU study to typical technical profiles. This is something that uwsojourner has toiled hard to overcome, and his post 681 is highly recommended.

Deep Stops Increases DCS

My quick summary of the key concepts goes like this:

Even though the NEDU study profiles don’t seem “typical” to the average technical diver, they do the things that we expect of the respective profile types. The deep stops profile protects the faster tissues from supersaturation early in the ascent (which is one of the fundamental planks of deep stop theory), but at the cost of greater supersaturation in the slower tissues later in the ascent. The shallower stops profile results in more supersaturation in the faster tissues early in the ascent, but has less supersaturation in the slower tissues later in the ascent and at the surface. These patterns of supersaturation distribution are an inevitable consequence of having a finite amount of decompression time and choosing to emphasise either deep stops or shallow stops.

If deep stop theory was correct, then the protection of fast tissues early should have produced some degree of protective effect, but the opposite was seen. The profile with the higher supersaturation early was actually the safest. Moreover, when the respective profiles were evaluated for overall exposure to supersaturation (using the integral supersaturation approach explained by uwsojourner) the deeper stop profiles resulted in greater supersaturation exposure. This passes the face validity test. Supersaturation is what drives bubble formation, and bubbles are the vectors of primary injury in DCS. More supersaturation overall is therefore likely to be bad.

The relevance to “real” tech diving profiles is that the same patterns of supersaturation distribution and overall supersaturation can be seen when profiles of equal length but emphasising deep stops to a greater or lesser degree (eg as in VPM vs GF approaches – see uwsojourner’s post which I have linked to above) are compared. Why would we expect the outcomes to be any different in this setting where the supersaturation patterns are the same? The difference might not be as exaggerated as in the NEDU study (which was designed to be intentionally provocative), but the “truth in the universe” if we could do a big enough study is highly unlikely to be any different.

Can I be clear that my argument is not that endlessly shallow stops are better. Someone earlier made the reductio ad absurdum point that the ultimate shallow stop profile is to simply come shooting straight to the surface and not make any stops. Clearly you can’t do that! So, one could conclude that all decompressions are “deeper” than the ultimate shallow stops profile. It is the degree of “deep stopping” that is under question, and in particular, whether the approach prescribed by bubble models (which certainly emphasise deep stops) is optimal. The best evidence we have at this point in time suggests that bubble models over-emphasise deep stops. Exactly how far we should back away from deep stops is presently not clear. That work has not been done.

Simon M

Simon,

These are the central pieces of your Nedu argument.

You state:

1/ Fast tissues were protected,
2/ The deeper stop profiles resulted in greater supersaturation exposure
3/ Nedu test profile saturation patterns were the same as deep stops / tech dives
4/ Supersaturation is what drives bubble formation,

You imply;

5/ Deep stop theory was tested,

****************

Lets us examine those very carefully.

5/ Deep stop theory was tested,


Do you mean the fake one that you made up? "...it eliminates bubbles..."

Or real VPM theory:

VPM theory:

"The VPM presumes that microscopic voids, cavities, nuclei exist in water, and tissues that contain water before the start of a dive. Any nuclei larger than a specific "critical" size, which is related to the maximum dive depth (exposure pressure), will grow upon decompression. The VPM aims to minimize the total volume of these growing bubbles by keeping the external pressure large, and the inspired inert gas partial pressures low during decompression."



Summary

Simon, clearly your descriptions and associations you claim between deep stops like VPM-B and the NEDU, do not exist, or need a lot of work. If you can rebut the above, I'd like to here it. BUT ... faked up profiles, heat graphs, ISS and other non scientific junk methods are not acceptable.

If a non scientist like me can find the see the plain obvious faults and errors in your position, then you need to fix up your assumptions.


Note: there are many points in your post on which you and I do agree Simon. like .. over doing deep stops can be bad - like the RD adhoc method. But real model have real gas tracking, so real models do things properly and do not suffer from the issue you raise.

Simon, if you want to attack those adhoc methods, then go ahead. But please leave alone the peer reviewed, scientifically valid, published models like VPM, like RGBM. They don't need your fixing and fiddling.




.

 

[igor p]

To understand why the NEDU A2 bent divers read BRW papers presented before in this thread. Same you will find info about algorithms that corelates best to LANL data bank of profiles with known outcome - It consists of big number of real dive profiles with known outcome.

 

[Dr Simon Mitchell]

Were the fast tissues protected? No

Deep stops debate (split from ascent rate thread) - Page 108

Does deep stop (sic) increase supersaturation exposure? No

Deep Stops Increases DCS

 

[Dr Simon Mitchell]

Same you will find info about algorithms that corelates best to LANL data bank of profiles with known outcome - It consists of big number of real dive profiles with known outcome.

Igor, can I ask exactly what you mean by "real dive profiles with known outcome"?

Simon M