Deep Stops Increases DCS

[Dr Simon Mitchell]

Simon. As a leader in your field, answer me this one simple question.

Is the increased occurrence of DCS a direct reflection of doing extremely prolonged deco stops between 40' and 70'?

Hello Tom,

Thank you for engaging in this discussion constructively.

There is one important point to clear up. The NEDU test profiles were not "manipulated" or arbitrarily stretched etc etc as has been implied several times. They were the outputs of two models (one gas content model and one bubble model) developed by the USN. The gas content model is actually in use in the USN in certain applications, and one aim of the study was to help decide whether the bubble model should be used instead. The decompression times seem long in relation to what you / we are accustomed to but remember these were air dives with no oxygen decompression. Also, the USN models have to some extent been calibrated against navy databases of dives of known outcome and these longish durations are what is required (according to their data) to decompress from typical navy dives with an acceptable risk of DCS. Remember too that a typical navy dive will usually involve working harder and maybe getting colder than many of us are accustomed to. Just performing significant work during the bottom phase of the dive markedly increases the decompression time requirement for safe decompression and that is why the NEDU study replicated typical navy dive conditions by imposing significant work throughout the bottom time. Finally, the NEDU bubble model profile is not markedly different from one prescribed by VPM running on a high conservatism level around +7 (which falls within the original model parameters) or from one that was actually recommended for the same dive by Bruce based on RGBM at the 2008 Deep Stops Workshop in Salt Lake City. The latter was discussed in Wayne Gerth's paper in the Proceedings of that workshop. I don't know whether you have the document, but if you PM me your email address I would be happy to send you the pdf.

Thus, based on the above, I don't think it is possible to say that the outcome would be highly predictable.

However, to truly appreciate the significance of the NEDU study outcome you have to think a bit more deeply than mere consideration of what the profiles "look like" or how they compare to the profiles we typically use. The "truth in the universe" that we are seeking in such a study is whether distributing stop time deeper is a good strategy. We all know that you can do a decompression with lots of deep stops safely, especially if there are also lots of shallow stops too. But the crucial question in respect of the efficacy of deep stops is that if you have a fixed amount of decompression time, is it better to spend more or less of that time deep. This was the case in the NEDU study where there was a fixed amount of decompression time, and the bubble model distributed more of the stop time deep in comparison to the gas content model. Obviously you know the result.

That result occurred despite the fact that the NEDU bubble model profile did what bubble models are supposed to do: it protected the fast tissues from supersaturation early in the ascent, but at the expense of increased supersaturation late in the ascent (which is virtually inevitable if you distribute a fixed amount of decompression time deeper). This has been depicted in various ways, but it can be clearly seen in the heat map diagrams created by Kevin Watts which I linked to in my previous post. This suggests that the fundamental premise of bubble models (that controlling supersaturation and therefore bubble excitation early is protective) may be flawed. And do remember as you criticise this recent data, that this premise was completely untested; there was NEVER ANY clinical outcome data that supported bubble models. We all just jumped on board with the idea (yes, including me) because it was so theoretically attractive. To be explicit, the NEDU study appears to tell us that protecting fast tissues early at the expense of greater supersaturation in faster tissues later may be a flawed concept. The study is suggesting that the clincal problems are arising from tissues that we are allowing to supersaturate to a larger degree as a result of having done deep stops.

Is all this relevant to the profiles we dive?? I think the answer is a qualified yes. The same difference in the pattern of supersaturation in faster and slower tissues that appeared disadvantageous in the NEDU study can be seen in real world bubble model profiles. This is readily apparent if you look at those heat map diagrams I linked to earlier which compare CCR constant PO2 mixed gas profiles generated by VPM-B+4 and GF 40-74. However, it is probably true to say that the use of oxygen decompression, and the various modifications that seem increasingly common these days (like padding the final oxygen stop to some extent) probably diminish any difference in risk between profiles.... possibly substantially so. Thus, the real difference in risk between profiles with different emphasis on deep and shallow stops as dived in the real world may be significantly less than implied by the NEDU study..... BUT if you want the "truth in the universe" then I think the NEDU study is correct.

For the reasons above, and as I have said many times previously, I am NOT saying don't use bubble models, or don't do deep stops. I just think that discourse on this subject is too heavily tainted with emotion and 'religeon' and the emerging signals (of which the NEDU study is only one, and which divers are entitled to understand) are being obfuscated. The optimal decompression paradigm is far from established but I do think that it is fair to suggest that we are in a phase of reconsidering how much emphasis should be placed on deep stops.

Simon M

---------- Post added December 24th, 2014 at 10:39 AM ----------

Thank you again Simon, for the practical real world "caveat" to take away from the NEDU study, and your lecture/recitation during our 29June -11July Bikini Atoll Expedition 2013 -and most of all gratitude for the post-IWR follow-up treatment as my attending physician during that trip (I owe you a bag of IV Plasmalyte and Caldolor, but will you take a pint of beer next time we meet?)

Hello Kev,

Sorry... I have only just noted your PM. I was not ignoring you. It was a pleasure to help you out. A look forward to that beer some time.

And now we come back to what I've been saying since Post #2:

Look at the Bottom Mix Gas used in the NEDU Study link above (essentially Deep Air):
This is the simple main practical point IMO/IME, to take away from the study:

Of course you're going to have significant residual inert Nitrogen and potentially on-gas N2 at your deep stop & perhaps even possibly at intermediate deco stops on Eanx50 which may encroach on critical slow tissue M-values as well --if you were using a working bottom mix with a high fractional N2 content to begin with like Air. Plan accordingly, use a computer to track your inert tissue loading (i.g. Shearwater Petrel) and be prepared to extend your 6m depth 100% Oxygen deco profile along with a stand-by IWR contingency protocol.

I have tried to articulate my thoughts on the real world signficance of the NEDU study in my reply to Tom above. You actually reflect some of them in your commentary, eg extending the shallow oxygen stop. I am less convinced that the air vs helium thing is significant. As I said above, the apparently disadvantageous pattern of supersaturation distribution between faster and slower tissues in profiles that emphasise deep stops to a greater or lesser degree (see the heat map diagrams i linked to previously) also occurs in real world mixed gas dives with oxygen decompression.

I had an entire month at Bikini earlier this year.... pete and I did some fantastic exploration in the Saratoga and we are getting better at using remote strobes. I won't go to Bikini in 2015 but am the doctor on his Truk trip in November.

Hope you are well and happy and enjoying your diving.

Simon M

 

[Kevrumbo]

Regards & Happy Holidays Simon. . .!

I think the "DCS Marker" that the UTD Ratio Deco/Buhlmann GF 30/80 Study is looking for in the Bloodwork & Urinalysis post-dive are Heat Shock Proteins:

. . .Production of high levels of heat shock proteins can also be triggered by exposure to different kinds of environmental stress conditions, such as infection, inflammation, exercise, exposure of the cell to toxins (ethanol, arsenic, trace metals, and ultraviolet light, among many others), starvation, hypoxia (oxygen deprivation), nitrogen deficiency (in plants), or water deprivation. As a consequence, the heat shock proteins are also referred to as stress proteins and their upregulation is sometimes described more generally as part of the stress response.^[8]

Heat shock protein - Wikipedia, the free encyclopedia

Much more preferable as a pathological marker (ethically as well), than having frank acute DCS symptoms as an experimental parameter/criterion, and a hopefully a more consistent scalable indicator of decompression stress than Doppler/VGE scores. . .

 

[Divetech99]

Dr. Mitchell,

I've been reading your posts (at RB and SB) with great interest and I admire your commitment and dedication to your work and to the diving community at large. It is very unusual for a scientists of your stature to share and help other divers, technical divers particularly. You have my thanks.

Like many divers in this forum, I have been practicing deep stops. I'm using a 2007 Suunto D9 computer for both deco and NDL dives. The Suunto D9 uses RGBM and has a deep stop setting. Your work, and that of Dr Doolette have made me rethink if I'm doing the right thing with deep stops. I hope you can help answer the following questions:

1. Would you recommend turning off the Deep Stop feature on the Suunto D9 for deco dives? How about for NDL dives? I believe the D9 deep stops are built into Suunto's RGBM algorithm and therefore not added arbitrarily.

2. I currently practice an ascent rate of 33 feet per minute up to 30 feet. Then 15 - 20 feet per minute from 30 feet to the surface, with a safety stop at 15 feet for 3 minutes (if NDL or what the D9 dictates if on a Deco dive). Would you recommend keeping or changing this ascent rate practice?

3. On all dives (deco and non deco) in either air or nitrox, is it generally wise to extend time spent at 10 - 15 feet beyond what the tables or computer prescribes? In other words, logistics and temperature permitting of course, will we lower our DCS risk if we spend more time at 10-15 feet before surfacing even if the computer says it's ok to surface?

Many thanks.

 

[Rich Keller]

Rich, for gods sake...the navy doesn't dive this model in the way it was done in the study. Did they do it for the sake of experimentation?? Of course.

Y'all are some dense characters. You act like the profiles they dive here are in any way normal....for anyone.

Now you are starting to get it, they did this to test the limits and in order to do that it had to be taken to an extreme so they know what that limit is in the real world not just on paper. The conclusion they got from doing the testing is that deep stops are not as safe. Your theory might work on paper but in the real world I would go with the people who tested the theory rather then the guy who sold you the paper. This is not something new, it has always been known that the deeper you go and/or the longer you stay the less accurate the tables are. This just further proves that position. Dense? Sounds like the pot calling the kettle black to me.

 

[tomfcrist]

Rich,

Their conclusion that deep stops are not as safe was based on the testing of ONE model that is strictly a bubble volume model....it ignores gas diffusion. I'm not defending their profile(obviously).

What I am saying is that Deep stops can be done in such a way that changes the outcome of the study. In order to do this you have to recognize not only bubble volume, but dissolved gas within the same model and balance it.

You cannot invalidate an ideaology based on one model(which may or may not be a good model to begin with).

Bruce and Tim brought this up when designing this study, and as far as I can tell was ignored. They acknowledged it, but never tested it. They chose to test a model that provides deep stops of incredible length, which means even if the stop is controlling the fast compartments, the slow compartments will be racking up dissolved gas(which is not accounted for).

I just don't understand how research scientists can honestly not recognize even the possibility that a model that accounts for both bubble size and dissolved gas and balances the two can make deep stops either present either the same amount of risk or less than a dissolved gas model alone. If they do recognize the possibility, why did they not test all 3 profiles at the same time? From the transcripts of the deep stop workshop that Simon sent me( which I appreciate greatly), the best reason I can assume is that Wayne has issues with Bruce, and vice versa. I sense a lot of contempt between those two.

 

[Divetech99]

Tom, you bring up very thought provoking insights and they all contribute to a very meaningful discussion. However, Dr. Doolette answered points you raised in the RB thread. I'm quoting Dr. Doolette below:

Dr. Doolette :
"Clearly they do not look like technical diving schedules - they are not, they are deep air decompression schedules. In selecting the test pair of schedules, there were two principal criteria. First, they had to result in some DCS so there was something to compare. Second, they had to be long, so that they could have substantially different stop depth distribution, i.e. the deep stops schedule should require a substantial amount of time at deep stops, so any deep stops effect (good or bad) can manifest. There is no point in testing, for instance, two 90-minute decompression schedules where one has five or ten minutes of time spent at deeper stops – I would happily move five or ten minutes around in a 90-minute schedule and not expect it to make a any detectable change in my risk of DCS. Remember that the purpose of a decompression stop (deep or shallow): we stop to limit gas supersaturation and thereby limit bubble growth, and we stay to washout inert prior to moving to the next stop. The staying is important, the amount of gas washout that occurs in the course of one, two, or five minutes is relatively inconsequential. "

 

[tomfcrist]

We are going to have to agree to disagree here.

The B5 profile that Bruce suggested met those criteria, had a substantially different curve and was disregarded based on similarities in supersaturation of the slow tissues. But like I said, supersaturation is only part of the equation. What they are not, and have not acknowledged is the differance in how much gas is diffused into the bubble when staying at the deep stops for 20 minutes vice 3 or 4. They say it's inconsequential, I disagree. Bruce disagrees. They never bothered to test the theory, so they really have no basis to refute it.

 

[Dr Simon Mitchell]

Hi Tom,

A couple of points that arise out of your recent posts.

Their conclusion that deep stops are not as safe was based on the testing of ONE model that is strictly a bubble volume model....it ignores gas diffusion.

That is not correct. The basic structure of the model is depicted in Figure 9 of the Gerth / Doolette paper in the deep stop workshop proceedings I sent you. It clearly accounts for bubble / tissue gas exchange via diffusion.

The B5 profile that Bruce suggested met those criteria, had a substantially different curve and was disregarded based on similarities in supersaturation of the slow tissues. But like I said, supersaturation is only part of the equation.

Supersaturation drives everything in this process, and the reason that Bruce's suggested profile almost certainly would have resulted in the same DCS rate as the NEDU deep stops profile is that it results in a virtually identical integral of supersaturation and time in the slower tissues as the NEDU profile. This is nicely illustrated in Figure 19 of the Gerth paper in the deep stops proceedings. Thus, when you say the "curve is different" you can only be referring to the fact that Bruce's decompression begins with some very short deeper stops missing from the NEDU bubble model profile. This is true, but the notion that these stops are the magical key to reversing the result of the NEDU study is implausible. Although these stops would have protected the fast tissues from supersaturation even more than achieved by the NEDU bubble model, the fact remains that the NEDU bubble model did provide substantial fast tissue protection compared to the gas content model (see Figure 18 in the Gerth paper) and yet despite this, the gas content model had a lower incidence of DCS. Put another way, if protection of fast tissues from supersaturation early in the ascent was the key to reducing risk, you would have expected the opposite result in the NEDU study. I doubt that I will change your mind, but I hope this makes sense.

You cannot invalidate an ideaology based on one model(which may or may not be a good model to begin with).

I fundamentally agree with this. Everything must be done with caution, and contrary to the way my message is sometimes portrayed, I am mainly trying to keep divers abreast of the current state of knowledge rather than to turn them on or off one particular ideology or another. In the current state of our knowledge there is a signal that we have swung too far towards the deep stops concept. There is certainly a shift from the view that deep stop approaches are superior and safer; a perspective that took a strong hold among our craft group in the early 2000s. It is germane, once again, to reiterate that there have never been any comparative outcome data published to indicate value in the deep stop approach, and the current weight of evidence is raising some serious questions.

Simon M

---------- Post added December 26th, 2014 at 01:50 PM ----------

Dr. Mitchell,

I've been reading your posts (at RB and SB) with great interest and I admire your commitment and dedication to your work and to the diving community at large. It is very unusual for a scientists of your stature to share and help other divers, technical divers particularly. You have my thanks.

Hello,

Thank you for that comment. I guess it is mainly because I am "one of you" (as are David, Neal Pollock and a few of the others do a lot of work in diver education). Also, the research we discuss is done for a reason... to improve diving safety. But if the knowledge remains locked up in scientific journals / technical reports and does not filter out to mainstream divers it is all a waste of time really.

Like many divers in this forum, I have been practicing deep stops. I'm using a 2007 Suunto D9 computer for both deco and NDL dives. The Suunto D9 uses RGBM and has a deep stop setting. Your work, and that of Dr Doolette have made me rethink if I'm doing the right thing with deep stops. I hope you can help answer the following questions:

Just a minor technical point.... almost none of this decompression research is my work. David and his NEDU colleagues deserve the credit for it.

1. Would you recommend turning off the Deep Stop feature on the Suunto D9 for deco dives? How about for NDL dives? I believe the D9 deep stops are built into Suunto's RGBM algorithm and therefore not added arbitrarily.

This is a difficult question. My own response to the state of our current knowledge has been to "de-emphasise" deep stops. I use gradient factors on shearwater computers and whereas I started out on GFs with something like 10:90 I am now more like 40 or 50:70. Thus, less deep stops and more shallow stops. The 70 is a bit conservative perhaps, but I am often the only physician on a diving trip! This seems to me to be a pragmatic response to the current state of our knowledge but I make no claim that it is optimal.

Returning specifically to your question..... Do remember that the NEDU study was not an investigation of the safety of RGBM (or VPM). Bubble model algorithms have been used to plan and execute countless dives safely. The question raised by the NEDU study is whether bubble models prescribe optimal decompression. As I pointed out in an earlier post, the actual differences in risk in real world diving between bubble model profiles or alternatives you might choose may be very small. To borrow from George Orwell, all deco models are safe, but some may be safer than others. In addition, many factors other than the decompression algorithm you choose are influential in determining risk on any one dive. For these reasons, if you have a model that appears to have worked for you over some time I would be reluctant to recommend that you change it. You could easily make such a change, go diving, get bent, and naturally blame it on the change when the incident may have had little to do with the deco algorithm. On the other hand, I would not be critical of someone who chose (for themselves) to dive in a way that de-emphasised deep stops (as I have done). Maybe one pragmatic approach is to continue diving as you currently do, but definitely don't pad the deep stops, and if anything, maybe pad your last couple of shallow stops when opportune to do so.

2. I currently practice an ascent rate of 33 feet per minute up to 30 feet. Then 15 - 20 feet per minute from 30 feet to the surface, with a safety stop at 15 feet for 3 minutes (if NDL or what the D9 dictates if on a Deco dive). Would you recommend keeping or changing this ascent rate practice?

No, that sounds very sensible.

3. On all dives (deco and non deco) in either air or nitrox, is it generally wise to extend time spent at 10 - 15 feet beyond what the tables or computer prescribes? In other words, logistics and temperature permitting of course, will we lower our DCS risk if we spend more time at 10-15 feet before surfacing even if the computer says it's ok to surface?

Generally yes, and especially if you are using increased fractions of inspired oxygen for your decompression. A caveat though: in extending decompression one needs to be wary of substituting the risk of DCS for an increased risk of oxygen toxicity (and both are very poorly predicted outcomes using our usual guidelines). In general, in benign environments and with favourable logistics, I am sympathetic to the idea of padding the final stop, but it is impossible to be certain about the degree to which risk is reduced. One day when we have a better understanding of the influence of factors like temperature and exercise at depth, we may see a trend to tailoring decompression to the relevant circumstances of the dive, and not just depth and bottom time.

Simon M

 

[ajduplessis]

So when will we see a more effective model being introduced based on this vast and highly scientifically tested data?

 

[rjack321]

This is a difficult question. My own response to the state of our current knowledge has been to "de-emphasise" deep stops. I use gradient factors on shearwater computers and whereas I started out on GFs with something like 10:90 I am now more like 40 or 50:70. Thus, less deep stops and more shallow stops. The 70 is a bit conservative perhaps, but I am often the only physician on a diving trip! This seems to me to be a pragmatic response to the current state of our knowledge but I make no claim that it is optimal.

You not stopping or slowing down at all below low GF 40 or 50?

At least anecdotally, "pyle stops" and similar somewhat arbitrary additions have been de rigueur for over a decade with people claiming to "feel better" as a result. Personally I am fine adding time deep, after all while the NEDU study may have demonstrated that shifting time shallow improved efficiency for a given total deco time - there's no obligation not to lengthen the deco on both the deep end (by a few minutes through deep stops on back gas) and then compensate on the shallow end with a modest GF high of 70 or even 60.

This issue I struggle with is for non-backgas deco, non-CCR deco. Many of us coming up from a 30min 200ft open circuit dive will extend the time after the first gas switch at 70ft. Depending on the gases selected, this switch may have an elevated O2 content but gas laws would suggest there's N2 on-gassing. E.g. from 18/45 to EAN50. Even though the EAD of EAN50 is "only 32 feet" is this relevant one way or another? 70ft is incidentally right about where the NEDU study bubble model started adding time too. So should the diver move past the switch depth to the GF low ceiling asap? Should the switch be done later in the water column or a different mix selected (ideally) to correspond to the low GF depth ceiling?

I guess my question is how does the NEDU study inform practices like switching to EAN50 before the first mandatory low GF 40 stop? Or the shape of the deco once on a deco gas, which might be am 18/45 to EAN50 switch or a 12/65 to 21/35 switch. (these are just examples of current open circuit practices). The NEDU study's air backgas only and the Heat maps are hard to translate into open circuit diving with gas switches (not the O2 switch).

Thanks for your time here, I didn't think to ask this question while the RBW was active.