Evolving Thought on Deep Decompression Stops

Written by: John Adsit, PADI Master Scuba Diver Trainer and Tec Trimix Instructor

Reviewed by: Simon Mitchell, MB ChB, PhD, DipOccMed, DipAdvDHM (ANZCA), FUHM, FANZCA, School of Medicina, University of Auckland

The decision to no longer require teaching deep stop strategies in the Tec Trimix Diver course (and all other PADI courses) may have surprised some instructors who didn’t know that the use of deep stops was even in question. Although at one time it may have appeared that deep stops had total acceptance, research has prompted frequent, sometimes heated, debates virtually since their introduction. (See below.) These not only occur online among divers, but significantly, within the hyperbaric medical community and the most credible names in decompression theory. Complicating the issue, part of the debate has been simply agreeing on what constitutes a “deep stop.”

tec_photo-1-3

In a 2008 UHMS deep stops conference, the assembled panel not only could not agree on a recommendation for deep stop use, but could not agree on a definition(1). It may be that a uniform definition for a complex question is part of the problem. A decade later, the emerging consensus seems to be that what we should be seeking is the deepest stop depth that produces optimally efficient decompression where “efficient” means least risk for a given decompression duration. Let’s review the history of this issue, look at the recent research, and summarize current thinking.

History of the Deep Stop

Many people credit Richard Pyle and the “Pyle stop” for the movement to add deep stops in decompression, but the famed article in which that self-proclaimed “fish nerd” suggested a deep stop approach(2) gives credit to earlier theorists, including David Yount, the creator of the of the Varying-Permeability Model (VPM), and Bruce Weinke, the creator of the Reduced Gradient Bubble Model (RGBM). These “bubble models” used a different approach to decompression than the standard gas content models that have been the norm since Haldane. Gas content models like Bühlmann ZHL-16C treat gases as almost entirely in solution (from a modeling perspective), and require a deepest stop based on empirically derived (i.e., observed results in actual dives) maximum safe tissue supersaturation. The bubble models, by contrast, purport to model the presence of bubbles within the tissues. As a consequence, these models require a first stop much deeper than do gas content models, both to protect fast tissues from too much supersaturation, and to control the growth of theorized existing bubbles.

Pyle advocated adding a deep stop half way between the depth at which you begin your ascent and the first stop depth your gas content algorithm identified. In most cases, this creates a first stop at about the same depth as a VPM or RGBM schedule. The rest of Pyle’s strategy, however, was more consistent with traditional gas content models than bubble models in that you went on to complete the decompression as a stipulated by your gas content algorithm.

Soon many others were advocating deeper stops. Erik Baker created gradient factors (GF) as a way to introduce deep stops into the Bühlmann algorithm. In a GF 20/80, for example, the first gradient factor (20) indicates the percentage of the supersaturation allowed by Bühlmann at the first stop. The lower this number, the less supersaturation allowed and the deeper the first required stop. The second gradient factor (80) indicates how much supersaturation is allowed relative to the Bühlmann limits upon surfacing. This means the lower the number, the longer the shallow stops must be to allow more outgassing before surfacing. As it happened, the 20/80 GF soon became a popular “norm,” typically giving a deep stop similar to Pyle’s and the bubble models.

The Wakulla Karst Plains Project advocated making a first stop at 80% of the diver’s average depth(3). That was later changed to 75% for the Ratio Deco approach used by two training organizations.

As an example of how these deep stop strategies differed from the conventional gas content models, for a 20-minute dive to 60 metres/200 feet using TMx 18/45:

Model                                                                       First Stop Depth

Bühlmann                                                               18 m/60 ft

Pyle, RGBM, VPM, Bühlmann w 20/80 GF        36-40 m/120 -130 ft

Ratio Deco approach                                            50 m/150 ft

Deep stops quickly became the norm in technical diving, despite the fact that it was all theoretical—no studies had been done on their effectiveness. This had changed by the UHMS workshop in 2008, however, in which the first studies were presented. These studies surprisingly did not support the use of deep stops, as will be described later. As a consequence, in the final workshop discussion session, the experts were unable to reach a consensus regarding the use of deep stops.

Since then, further studies have cast doubt on the effectiveness of deep stops, with most of the acknowledged experts in decompression theory now not recommending the originally proposed deep stop strategies. Despite this, deep stops are still commonly used. While some of this is a lack of awareness, a closer look shows that much of it is semantic, for it has evolved beyond a simple “deep stops good/deep stops bad” debate. Most of those opposed to deep stops of the sort typically prescribed by bubble models do not advocate using, for example, an unmodified Bühlmann model. Many are themselves using Bühlmann, but with the low GF value of 40 or 50 instead of 20. Using the previous 60-metre/200-foot example, the first stop would be in 27-30 metre/ 90-100 feet range – still deeper than the pure Bühlmann 18-metre/60-foot first stop.

In short, it appears the real opposition is to “too deep” stops, with the real question being how much shallower the first deep stops should be. Most experts suspect that the optimal first stop depth is deeper than the pure Bühlmann algorithm, so a simple yes or no debate on deep stops is meaningless. As tec diving author Mark Powell recently said, “’deep stops’ is a misleading term that doesn’t always help the discussion. Perhaps it’s time to move away from using this term in favour of being more specific as to exactly what sort of stops we’re talking about”(4).

To find answers about how deep the first stops should (or shouldn’t) be, we need to look more closely at the research.

tec_rh1_7300_029_edited

Looking at the Research

French Navy Study (2005)(5)
Perhaps the first study of deep stops profile was done by the French Navy while considering whether to move to a bubble model from their gas content model. They compared their model to three different deep stop protocols by measuring post dive venous gas bubbles. They discovered that none of the deep stops models were superior to the gas content model, and one of them was inferior.

US Navy NEDU Study (2008)(1)(6)
The US Navy was also considering changing to a bubble model, and so conducted a well-controlled experiment through its Experimental Diving Unit (NEDU). As discussed at the UHMS deep stops workshop in 2008, a deep stop schedule prescribed by a bubble model was compared to a conventional ascent schedule prescribed by a gas content model. Based on safety protocols, the study had to be cut short because the divers doing the deep stop schedule had DCS more than three times as often as the divers following the conventional, shallow stop schedule. The study concluded: “The deep stops schedule had a greater risk of DCS than the matched conventional schedule. Slower gas washout or continued gas uptake offset benefits of reduced bubble growth at deep stops.”

In other words, if the deep stops had been controlling bubble growth and allowed more off-gassing in the faster tissues, at the same time they slowed the off-gassing in intermediate tissues and allowed the slowest tissues to continue on-gassing so the overall nitrogen supersaturation ended up being higher.

Ljubkovic Study (2010)(7)
A study published in 2010 examined a specific bubble model (VPM) to see to what degree it controlled bubble formation. The research found that it did not control bubbles effectively in that all divers tested completed dives with a high incidence both arterial and venous gas emboli (bubbles). This study was not comparative, so although all divers tested showed bubbling, there is no comparison with the amount of bubbling that may have been present with divers using other ascent strategies.

Spisni Study (2017)(8)
A study published in Diving Hyperbaric Medicine in March 2017 showed the results of research that compared a deep stop profile to a profile with shallower first stops. The study examined divers using the UTD agency’s version of the Ratio Deco profile, which has its first stop at 75% of the maximum depth, with divers using a Bühlmann schedule with GFs of 30/85. All divers completed a dive with the same depth and bottom time but the 30/85 GF schedule prescribed a shorter overall decompression time than the Ratio Deco algorithm, which implied a significant advantage for the Ratio Deco group.

Despite this, the Spisni study results did not favor the Ratio Deco schedule. It found no differences in detectable bubbles 30 minutes post dive, but it found that the divers using Ratio Deco showed a worsening of the post dive inflammatory profile, with the 30/85 GF dives showing no increase. The study’s conclusion was “Overall, our findings contradict the idea that adding longer and/or deeper stops is useful to achieve a more effective decompression.”

Fraedrich Study (2018)(9)
A study published in December 2018 Diving and Hyperbaric Medicine examined whether computers using four algorithms (Suunto RGBM, VPM-B, EMC-20H, and Bühlmann ZHL-16C) conformed with what the US Navy had determined were the limits of safe ascent protocols in experimental dives with known decompression outcomes determined in the 2008 NEDU study. For the purposes of this study, the no stop time, first stop depth and total decompression time were examined and compared. According to the report, none of the computer algorithms passed all the test requirements with factory-default conservatism, but the Bühlmann ZHL-16C and Suunto RGB could with user-defined settings. Here are the relevant findings:

  • For Bühlmann ZHL-16C, GF low settings lower than 55 led to first stops that were considered to be too deep. The study mentioned that a GF Low of 70 was also acceptable, so it explicitly approved of a range of 55-70 but did not specify an upper limit. It also said a GF high of 70 or below was within the preferred US Navy limits.
  • With the deep stops setting on, the Suunto RGBM’s first stop was deeper than suggested by the NEDU study. When the deep stops setting was off, the first stop was within the safe range.
  • For VPM-B, all of the five conservativism settings gave a first stop that was considered to be unacceptably deep.
  • For EMC-20H (used by Cochran computers), the first stop was within limits when set to <5%, but in that setting it could not be tuned to have an acceptable total decompression time. It could alternatively be set to have an acceptable total decompression time, but in that setting it was deeper than suggested by the NEDU study.

What’s Missing
What is missing in this research is any study showing a benefit for the deeper ranges of first stops, including Ratio Deco and the bubble models. No experimental research with human divers has shown any benefit to such profiles.

Deep Stop Debates

Much of the debate about deep stops has taken place in online forums, with noted experts participating. These debates can be very informative, and a review of them shows an overwhelming consensus that current thinking is favoring first stops shallower than is typical of deep stop profiles prescribed by bubble models. Here are some of those debates.

RebreatherWorld (2013-2014)

Rebreather World (2014)

ScubaBoard (2014-2016)

CCR Explorers (2018)

The Dive Forum (2017-2018)

tec_sdmnt0613_0398adobe

Summary and Conclusions

As a result of the research done in the last decade, the leading experts in decompression theory are nearly universal in their opinion that some of the theories associated with the deep stop movement made the first stop too deep. Research Chair in Hyperbaric and Diving Medicine, Laval University, Quebec and former DAN Research Director Dr. Neal Pollock summed this up in a presentation on controlling decompression stress: “No deep stops! By slowing down deep, all you are doing is adding more time that you’re on-gassing into your slow and intermediate tissues that aren’t saturated. So deep stops have not been shown through any credible research to be in your best interest”(10).

Although there is essentially unanimity that first stops prescribed by bubble models are too deep, there is as yet no consensus on where that ideal first stop should be, with some prominent names in decompression diving now expressing personal preferences for shallower stops represented by low GF values of 40 and 50(11, 12).

According to Pollock, the ideal decompression strategy likely varies by the individual, and probably also by the dive. Still, there is certainly a decompression strategy that is most likely to be optimal for the general population, and perhaps someday we will have enough research to state with confidence what that might be. We are not there yet. The most recent research and the conclusions of the leading minds in decompression theory, though, indicate that the deeper stops prescribed by bubble models are typically too deep.

  • Doolette DJ, Gerth WA, Gault KA. Redistribution of decompression stop time from shallow to deep stops increases incidence of decompression sickness in air decompression dives. Technical Report. Panama City (FL): Navy Experimental Diving Unit; 2011 Jul. Report No.: 11-06. Durham (NC): Undersea and Hyperbaric Medical Society; 2009. p. 195-206. [cited 2013 Mar 3] Available here. 
  • Pyle RL. The importance of deep safety stops: rethinking ascent patterns from decompression dives. SPUMS Journal. 1997;27:112-5. Available here. 
  • Irvine, G. The First stop/Deep Stops. GUE. Available here. 
  • Powell M. Delving Deeper into Deep stops. In: Diver; July 2018. Available here.
  • Blatteau J-E, Hugon M, Gardette B. Deeps stops during decompression from 50 to 100 msw didn’t reduce bubble formation in man. In: Bennett PB, Wienke BR, Mitchell SJ, editors. Decompression and the deep stop. Undersea and Hyperbaric Medical Society Workshop; 2008 Jun 24-25; Salt Lake City (UT).
  • Gerth WA, Doolette DJ, Gault KA. Deep stops and their efficacy in decompression. In: Vann RD, Mitchell SJ, Denoble PJ, Anthony TG, editors. Technical diving conference proceedings 2008. Durham (NC): Divers Alert Network; 2009. p. 138-56. [cited 2012 Mar 3 login required] Available here. 
  • Ljubkovic M, Marinovic J, Obad A, Breskovic T, Gaustad SE, Dujic Z. High incidence of venous and arterial gas emboli at rest after trimix diving without protocol violations. J Appl 2010;109:1670-4.
  • Spisni E, Marabotti C, De Fazio L, Chiara Valerii M, Cavazza E, Brambilla S, Hoxha K, L’Abbate A, Longobardi P. A comparative evaluation of two decompression procedures for technical diving using inflammatory responses: compartmental versus ratio deco. Diving Hyperb Med. 2017 Mar; 47(1): 9–16. Published online 2017 Mar 31. doi: 28920/dhm47.1.9-16 . See more here.
  • Fraedrich D. Validation of algorithms used in commercial off-the-shelf dive computers. Diving and Hyperbaric Medicine. 2018 December 24;48(4):252–258. doi: 10.28920/dhm48.4.252-258. PMID: 30517958.)
  • Pollock N. Thoughtful Management of Decompression Stress. British Sub-Aqua Club. November 23, 2016. Available here. 
  • Mitchell S. Decompression Controversies. DAN South Africa. May 9, 2016. Available here. 
  • Partridge B. Review of Deep Stops in Technical Diving. November 9, 2015. Available here.