Deep Stops Increases DCS

[victorzamora]

Please... use the correct terms.. UWSojourner's heat graphs do NOT show supersaturation. They show tissue gas, which... is mostly useless information, and actual decompression stress and supersaturation cannot be seen in tissue gas graphs.


The most important part of a plan is the actual supersaturation, because that is where the extra-vascular tissue bubbles that are thought to make DCS, and will form and grow.



REAL supersaturation information:

Someone (Kevin) correct me if I'm wrong here....but I think Ross H is reading the heat maps wrong. The heat maps compare two profiles, compartment-by-compartment. You can't compare two compartments in the same profile on the one heatmap directly. So pointing out that "Cell 9" is the controlling one and that Kevin's heat map shows Cell11 as the highest isn't a correction. Kevin's heat map is showing that Cell 11 is more saturated in VPM-B than in Buhlmann....not that Cell 11 is more saturated than Cell 9.

 

[EFX]

DT is total decompression time neglecting the ascent time. DT is calculated by adding the times for all the stops. The significant thing about these profiles is the DT - it is the same. The Navy had to shorten the DT for the deep stop profile to make it a constant between the profiles. But in doing so they added a variability to the test in the off-gassing rate.

Why would you be so quick to agree that bubble models in general and VPM in particular are inferior to dissolved gas models in general and Buhlmann GF's in particular, when the NEDU study is testing something quite different from what [non-Navy] bubble models are doing?

This begs the question: how can we test the effectiveness of deep stops vs. shallow stops and how do we eliminate the variability in off-gassing rate? Here's how I would do it:

Have two groups of divers and two profiles to the same depth and time using the same decompression algorithm. Control the pre-dive variables: everyone gets enough sleep (no alarm clocks), same number of meals and snacks (no skipping or extra snacks), no alcohol or caffeine, same amount of exercise, no illnesses or medications. Controlling these parameters does not eliminate but reduces variability.

Select a profile using Buhlmann GF's (BGF) that will bend at least 5% of the divers. (Note: this will strike some readers as reckless but unless you bend some divers you have no way of determining the effectiveness of either method). GF low and high must be the same to eliminate the Buhlmann algorithm which is a dissolved gas model from incorporating deep stops. Using VPM-B with the same profile record the deep stops. Both groups use Buhlmann with the agreed GF's. One group dives the profile following the BGF exactly. The other group using BGF inserts stops according to the VPM plan. All participants dive in both groups.

At this point you might be asking, "How is this going to prove anything? The DC is adjusting the (shallow) stops and times to make the dives safe".

The key to this is removing the variability in off-gassing rate. By letting the BGF control the depth and length of the (shallow) stops you ensure the off-gassing rate will be the same between profiles. The BGF is a dissolved gas model and knows nothing about free gas mechanics; that is the purview of bubble or dual gas models. The variable you are testing is deep stops. If more of the shallow stop divers get bent then we know that the dissolved gas coming out at a safe rate is causing existing bubbles in the tissues to grow and become a problem. We also know that the deep stop divers, by stopping deep briefly, are limiting the quantity and growth of bubbles beyond the critical size to reduce the frequency of DCS using a relatively safe dissolved gas model.

Does this make sense?

 

[scubadada]

The NEDU study showed that divers decompressing according to the probabilistic BVM(3) bubble model had statistically more DCS than divers decompressing according to the deterministic, gas content VVAL18 Thalmann algorithm. The Navy decided not to adopt the bubble model.

Since 2011, people have been attempting to explain the results. People have also attempted to generalize the results to more commonly used, non-Navy decompression algorithms such as VPM and Buhlmann ZHL-16C. This has led to considerable, often acrimonious, debate.

So, here we are. Each of us must weigh the evidence and make our own decision, I have

I have found this thread interesting and thought provoking. For fun and education, I ran the 170 foot air dive for 30 minutes on Ross' MultiDeco using 6 different Buhlmann settings from 100/100 down to 30/70 and with VPM from +0 to +5.

None of the generated decompression profiles look similar to the NEDU shallow stops or the deep stops profile. The decompression times were also considerably shorter than in the NEDU study, varying from just under an hour to 136 minutes. Somehow, I just can't imagine choosing a reasonable Buhlmann or VPM profile and ending up with a DCS rate of 5%, even 1.6% seems high to me.

Rather than generalizing directly from the NEDU study, looking at information as shown in @UWSojourner post #681 seems more promising. As others have pointed out, it is unlikely that all the quality research on decompression we'd like to see will ever be carried out.

 

[UWSojourner]

So, here is a question I posted in Decompression Controversies:

Why would you be so quick to agree that bubble models in general and VPM in particular are inferior to dissolved gas models in general and Buhlmann GF's in particular, when the NEDU study is testing something quite different from what [non-Navy] bubble models are doing?

I'll answer your question. I hope you'll answer mine as well.

The NEDU concluded that the reason the deep stop model exhibited more risk than the shallow model was the greater supersaturation exposure in the deep stop profile (i.e. integral supersaturation). The deeper stops required by bubble model theory come at the cost of higher total supersaturation exposure.

We have several things that are converging to the same conclusion. These were outlined by Dr. Mitchell in his presentation here, but I list them below.

1. First, remember that there was never any evidence that bubble models were better. They were developed based on theory and experiments in gel, not by dive trials. The theory is very attractive -- "do smarter stops, get out quicker". They were never a physiological model.
2. A French study on 50-60m air dives showed deep stop protocols were not superior; one was inferior. They declined to switch to bubble models.
3. Another study showed high grade bubbles following a deep stop VPM profile in 60-70m trimix dives in virtually every diver. High grade VGE is associated with higher DCS.
4. NEDU study -- we all know that one. But these were live dives pushed to DCS. The deep stop profile exhibited over 3x the DCS risk. As stated above the only credible explanation is that the additional supersaturation exposure in the deep stop profile resulted in more DCS.

This pattern of deep stop models keeping the diver deeper, as required by bubble theory, comes at the cost of higher supersaturation exposure. What actual dive trials are telling us is that the tradeoff appears to be a poor exchange. We can demonstrate that exchange in dive profiles we are all familiar with as I did here. All the available research points in the direction that the exchange is a poor one.

While I wouldn't say that these conclusions have been "so quick", those are my reasons.

Now a question for you. In the post here I show a pretty common profile where, for the same decompression time, VPM-B results in 30% more supersaturation exposure once the diver surfaces.

In light of the studies cited, what benefit do you see in the VPM-B profile that comes at the cost of 30% more decompression stress? Why do you believe the exchange (deeper stops for more supersaturation exposure) is a good one?

 

[rossh]

I'll answer your question. I hope you'll answer mine as well.

The NEDU concluded that the reason the deep stop model exhibited more risk than the shallow model was the greater supersaturation exposure in the deep stop profile (i.e. integral supersaturation). The deeper stops required by bubble model theory come at the cost of higher total supersaturation exposure.

We have several things that are converging to the same conclusion. These were outlined by Dr. Mitchell in his presentation here, but I list them below.

1. First, remember that there was never any evidence that bubble models were better. They were developed based on theory and experiments in gel, not by dive trials. The theory is very attractive -- "do smarter stops, get out quicker". They were never a physiological model.
2. A French study on 50-60m air dives showed deep stop protocols were not superior; one was inferior. They declined to switch to bubble models.
3. Another study showed high grade bubbles following a deep stop VPM profile in 60-70m trimix dives in virtually every diver. High grade VGE is associated with higher DCS.
4. NEDU study -- we all know that one. But these were live dives pushed to DCS. The deep stop profile exhibited over 3x the DCS risk. As stated above the only credible explanation is that the additional supersaturation exposure in the deep stop profile resulted in more DCS.

This pattern of deep stop models keeping the diver deeper, as required by bubble theory, comes at the cost of higher supersaturation exposure. What actual dive trials are telling us is that the tradeoff appears to be a poor exchange. We can demonstrate that exchange in dive profiles we are all familiar with as I did here. All the available research points in the direction that the exchange is a poor one.

While I wouldn't say that these conclusions have been "so quick", those are my reasons.

Now a question for you. In the post here I show a pretty common profile where, for the same decompression time, VPM-B results in 30% more supersaturation exposure once the diver surfaces.

In light of the studies cited, what benefit do you see in the VPM-B profile that comes at the cost of 30% more decompression stress? Why do you believe the exchange (deeper stops for more supersaturation exposure) is a good one?

You keep making these false and misleading statements.

The nedu did not test deep stops, did not test VPM, did not test any connection or resemblance to tech dive practices - those are FACTS.

Just accept it. Stop trying to morph the nedu test into something it never was.

As to your points:

1/ Never any evidence that shallow stops are better either - so you can't assume it either way

2/ French navy test - Yes one air dive, using test divers with prior DCS history.

3/ VGE - so what - almost all dives, all models, all plans, all gas types make VGE. Even Breath hold divers can make grade 4 VGE ! Get used to it - VGE have been with us for 40+ years.

4/ Absolute rubbish - the wishful thinking continues. Nedu did not push the dives, there is almost NO supersaturation present in both profiles. Hard to make DCS with no profile stress.​

You say 30% more stress on the surface?? Obviously that is invalid measure. Or maybe its the because its has 30% less in the dive - where supersaturation limiting actually becomes important and counts.


.

 

[drmwc]

Just wondering, since the figures (5% vs 1.6% DCS risk) are less than 5% apart, are they considered to be statistically significant, especially with the small sample size of the study? Just recalling statistics classes where a difference of less than 5% is considered not statistically significant.

The American Statistical Society has warned against p-value testing to choose between hypotheses.

An Error Occurred Setting Your User Cookie

 

[EFX]

In light of the studies cited, what benefit do you see in the VPM-B profile that comes at the cost of 30% more decompression stress? Why do you believe the exchange (deeper stops for more supersaturation exposure) is a good one?

I haven't read the other studies you mention in any detail so I can't comment on whether those studies have any relevancy to sport diving algorithms. As to your heat map study you cite above, I agree with your conclusion: there is no benefit. That is only because you are comparing apples and oranges. Look at your heat maps. The VPM graph shows what looks like a 500% increase in (amount? rate?) of inert gas uptake over the BGF graph. So, the two profiles are not the same: VPM-B +3 is not equal to GF 60/75. I suggested an approach to testing deep vs. shallow stops that removes the variability in gas uptake by utilizing the same algorithm.

The question cannot be settled by arguing over total supersaturation exposure alone. This is because the problem that bubble (deep stop) models try to solve is dissolved gas infusing into bubbles which is not completely dependent on supersaturation (or pressure gradient) values. It's entirely possible that lower levels of supersaturation may be enough to enlarge bubbles beyond their critical size to cause DCS.

Maybe we should ask a different question. Is the use of blind dissolved gas models or higher GF lows worth the risk of possible DCS hits? The fact that you cite no compelling scientific evidence is not convincing. On the contrary there is plenty of anecdotal evidence that points to the reasonable effectiveness of deep stops. We need more studies conducted by non-military funded organizations that don't include hidden agendas or cross purposes with sport diving.

 

[PfcAJ]

4/ Absolute rubbish - the wishful thinking continues. Nedu did not push the dives, there is almost NO supersaturation present in both profiles. Hard to make DCS with no profile stress.​

.

And yet there was dcs and more of it on the profile with a deeper stop distribution.

 

[UWSojourner]

The fact that you cite no compelling scientific evidence is not convincing. On the contrary there is plenty of anecdotal evidence that points to the reasonable effectiveness of deep stops. We need more studies conducted by non-military funded organizations that don't include hidden agendas or cross purposes with sport diving.

I suspected that was the case. And that's ok.

This is just my opinion, but the fact that you can summarily dismiss several studies of actual dives by researchers, some tasked with keeping their country's divers safe, in preference for "anecdotal evidence" says a lot. Mostly it tells me that no matter what evidence is offered, some will not be convinced.

But your reference to "hidden agendas" is a little Rossonian and bizarre.

 

[scubadada]

And yet there was dcs and more of it on the profile with a deeper stop distribution.

But the stops were very different than a VPM profile, 29 minutes at 70 and 60 feet for the NEDU deep stop profile vs. 14 minutes at 90, 80, 70 and 60 feet for VPM+3, just as an example. Like I said, the NEDU profiles do not look similar to VPM or Buhlmann profiles for the 30 min air dive to 170 feet.