Is there data on how close to NDL undeserved hits occur?

[FishWatcher747]

My personal belief is that, absent medical factors which change the DCS probability (like PFO, since we all bubble), it's ascent rate, particularly in the last 30 feet, which plays a much more significant yet underappreciated role.

Well this is another factor that we could get data on.

I'm thinking of a retrospective case study of chamber patients with Type 2 DCS doing recreational NDL dives where we can get the graph of depth, time, and ascent rate from their computers. Include only those where there was no previous recent dive (apparently this is the most common scenario). Include only those who then get evaluated for PFO and separate the data for those with PFOs and ones without.

If one could get enough data then reasonable conclusions might be drawn about the relative role of depth, time and ascent rate in Recreational NDL Type 2 DCS undeserved hits.

 

[Johnoly]

If one could get enough data .......

Data costs money to collect. If you have enough money, you can get all the data you need. Data, money, world peace, 3 things we'll never have enough of.

 

[Wibble]

Aren't Non-Decompression Limits statistically calculated such that there's a certain number of standard deviations of people getting DCS? Even if doing deep NDL dives with the mandatory decompression safety stop the inert gas loading hasn't built up beyond an arbitrary 'safe' limit.

Given the millions of NDL dives that happen throughout the year where the absolute vast majority of them have no signs of DCS, what are you worrying about?

If it was a significant risk, why would you allow diving in the USA where lawyers would prevent you doing it, suing all and sundry for allowing this heinous crime of negligent diving to take place!

Also, diving to NDLs AND using recommended ascent rates, etc., any DCS is likely to be minor. One would expect that chamber rides are more often than not arse covering; showing that something's been done.


In my limited experience PFOs and fast ascents are the two biggest causes of DCS.
Diving with a friend on a 35m/115ft CCR dive; really gentle, lovely dive, we extended the decompression and had a very slow ascent to the surface. He felt really tired on the boat and on the drive home developed a skin rash which grew. Called the dive doc and went to the chamber and received 5 days of chamber treatments in 6 sessions. Subsequently diagnosed with a PFO and now not diving until he gets it fixed. Prior to that he'd had no DCS despite lots of diving.
(My profile, but we were diving together.)

Admittedly this isn't diving to NDLs, but was a gentle dive.

(Edit: added the GFs for comparison with post #16)

 

[Wibble]

Data costs money to collect. If you have enough money, you can get all the data you need. Data, money, world peace, 3 things we'll never have enough of.

Might be possible to collect ANONYMISED data from tools such as Shearwater Cloud. The Cloud application and database would need some form of flag and field for DCS symptoms.

There would also need some way of validating reports to weed out crap from the data (an insect bite isn't DCS, neither is heatstroke, etc.).

 

[L13]

I am curious to know if an analysis has ever been made on how close to NDL does the average diver get to then suffers DCS on an "undeserved" hit dive? If so what is the range and what is the shape of the curve?

Edit: Let us assume there was no rapid ascent contributing to the DCS, just a normal recreational dive following all usual training guidance.

How do you measure "close to NDL"?

With a GF of 95% a dive to 60ft has an NDL of ~30 min.
With a GF of 65% a dive to 60ft has an NDL of ~25 min.
Is a dive to 60ft for 25 min at NDL or 5 min less than NDL or 5/6 of NDL?

With a GF of 95% a dive to 90ft has an NDL of ~20 min.
With a GF of 65% a dive to 90ft has an NDL of ~10 min.
Is a dive to 90ft for 10 min at NDL or 10 min less than NDL or 1/2 of NDL?

As you can see from those examples, measuring "close to NDL" in either minutes or fraction of the total is not a consistent measure across depths and conservative settings. What is a consistent measure is GF (or the equivalent for other algorithms), both GF 65% dives are equally distant from NDL in an "avoid DCS" sense and works all the way into deco. Research has been done on DCS vs GF. DCS is low at 95%, drops rapidly by 85%, and by 65% there is diminishing returns for lowering GF further.

For other algorithms without GF, their conservative setting is the equivalent.

 

[scubadada]

My personal belief is that, absent medical factors which change the DCS probability (like PFO, since we all bubble), it's ascent rate, particularly in the last 30 feet, which plays a much more significant yet underappreciated role.
In a way, that's what safety stops do: they significantly drop your mean ascent rate from 30 feet. Unnecessary? Not in my book. It's the right thing to do for the wrong reason. It's not just safety padding. I believe that the incidence of "undeserved hits" would have been much greater in the last 30 years had we not incorporated this practice. It was not just a "good idea"; it was a great idea! Just hard to prove before ultrasound evaluation of venous bubbling...

What @rsingler said can be easily seen on the Shearwater Cloud dive graph. This was a 2nd solo drift dive of the day in Boynton Beach, FL. It started on the outside reef at Sharkie's Ledge and crossed over the reef to the inside. It was done on 34% nitrox. The GF at the start of ascent is 1%. Arriving at the safety stop, the GF is 20%. At the end of the safety stop, the GF is 13%. The final ascent was made over 1:10, less than 15 ft/min. The surfacing GF was 77%. This was about what I expected from following my SurfGF. After 10 min at the surface, the GF was 68%.

Seeing graphs like this, since purchasing my Teric in 2019, has reinforced the importance of the ascent, particularly, the final ascent.

 

[scubadada]

Risk of DCS close to the NDL?

The following comes from the SAUL Recreational Dive Planner SAUL Recreational Dive Planner | Modern Decompression and the blog regarding multiple dives with the same risk The Significance of DCS Probability on a Single Dive Profile | Modern Decompression that I referenced in an earllier post Is there data on how close to NDL undeserved hits occur?. The SAUL planner includes a 3 min safety stop at 15 ft and uses an ascent rate of 60 ft/min.

The following all pertain to diving 32% nitrox:

At a GF high of 95, the highest risk of DCS is diving the NDL of 34 min at 90 ft. For a single dive, the risk is 0.092% or 1/1,087.

At a GF high of 85, the highest risk of DCS is diving the NDL of 51 min at 70 ft. For a single dive, the risk is 0.013% or 1/7,692.

At a GF high of 75, all the dives at NDL have a risk of <0.001% or 1/100,000.

For the dive at a GF high of 95 for 34 min at 90 ft, you would expect a DCS hit at somewhere between 1,000 and 10,000 dives, closer to the latter.

So, regardless of the exact numbers, it would appear that the risk of an unexpected DCS hit when diving near the NDL, even with a liberal computer, is small. I expect that many episodes of unexpected DCS have additional risk factors other than the exposure at depth. These may include characteristics of the dive itself or personal risk factors; ascent profile, dehydration, patent foramen ovale, previous injury, cold ambient temperature, high body fat content, alcohol consumption... Decompression Sickness - StatPearls - NCBI Bookshelf

 

[FishWatcher747]

Thanks to all for contributing to this discussion.

Personally, I don't fear a case of Type 1 DCS. I do however have some concern over a Type 2 hit as it sometimes can be life altering, disabiling, and even deadly depending on how remote one is if it happens.

The Saul statistics listed above appear not to make a distinction between Type 1 and Type 2.

 

[scubadada]

Hi @Wibble

Thanks for adding the GFs to your graph. Looks like you went from essentially zero to 66% with a 4 or 5 minute final ascent.

 

[Wibble]

Thanks to all for contributing to this discussion.

Personally, I don't fear a case of Type 1 DCS. I do however have some concern over a Type 2 hit as it sometimes can be life altering, disabiling, and even deadly depending on how remote one is if it happens.

The Saul statistics listed above appear not to make a distinction between Type 1 and Type 2.

By its own definition, an undeserved bend has no specific cause as the human body is infinitely more complex than a simple mathematical model. You can mitigate the risk by diving to a far more conservative profile, extending stops, using nitrox, diving shallower, etc.

Ultimately diving has risk. DCS is just one of many many risks in diving.

The only sure fire way to completely mitigate these risks is not diving.