By: Richard Devanney

The original aim of these articles was to provide a brief history of the incremental developments made by notable figures in the field of decompression research and make them accessible via the gift of plain English and ridiculous analogies. The idea was also to outline current topics of interest in decompression research. The problem is, the last article didn’t even come close to achieving that.

Here’s an attempt to rectify that, but first, it’s probably worth putting in the disclaimer now. I recently read a layman definition of decompression. It said it was like measuring with a micrometre, marking with chalk, and cutting with an axe. I would add “whilst being repeatedly tasered” to that.

Decompression studies are increasingly focussing on how certain physiological and biochemical changes to the body at the sub-clinical DCS scale occur when diving. As with all science, the outcomes often raise more questions than they answer. This is largely due to the number of variables at play in terms of both complex individual biology and a wide variety of diving parameters such as:

  • Thermal changes
  • Dive profile
  • Ascent speed(s)
  • Dive duration
  • Depth
  • Time at depth
  • Gas choice
  • Switching between mixes
  • Ascent profile
  • It goes on.

Basically, there are no cut and dry answers and the search for a greater understanding continues. An excellent recent article by Mark Powell also outlined some of the research that is described below, but he is better than me at telling TDI he’s writing an article on decompression updates (sorry TDI!). Anyway, here’s my take on it all, within the context of the previous three articles in this series.

Reducing decompression stress

Rates of DCS are thought to be around 10-20 cases in every 100,000 no-stop exposures. That’s actually very low, so it seems we are doing something right. Current recommendations on safe recreational diving emphasise not only staying within the limits such as depth, time at depth and ascent rates, but to also add an extra level of conservatism on top of that. This is because staying within the limits is no guarantee of being DCS free.

As for decompression diving, the same advice applies- be conservative to reduce your risk. For technical diving there is less DCS and sub-clinical DCS data available, and it’s even more difficult to assess any trends in DCS causality given the increased choice of diving parameters available. The best advice is to really pad out that deco and incorporate diving practices that serve to keep you well within commonly accepted limits.

Yet emphasising the need to be conservative is one thing, but what does it really mean? Ask most divers and they will say:

  • stay well within your NDLs
  • don’t go over your depth
  • maybe do a minute or two longer on your safety stop
  • stay hydrated.

Technical divers will say plan your dive and dive your plan, meaning do what the software tells you to do. All sensible suggestions; but decompression research shows that there are many more factors to consider, and they all fall under the umbrella of decompression stress.

These risk factors for DCS often interact in complex and little-understood ways that affect your physiology, mostly with no manifestation of DCS. But that doesn’t mean you should ignore them until the science is in. They are not esoteric principles, they are practical considerations, applying established good practices can only be a good thing- as long as you consider them collectively. You might imagine them like pieces of a jigsaw puzzle- they fit together but are fuzzier around the edges. Just to clarify, I’m not saying you will get DCS from undertaking or failing to finish a jigsaw puzzle, even if it may feel like it.

Variables that can affect decompression stress include:

  • Dive parameters: Choice of dive planning software (and conservatism settings), interpretation of what dive the planning software is telling you, ability to stick to the plan- dive profile, depth, time at depth, stops, repetitive dives, gas choice and mix changes, ascent rate.
  • Exercise: Type, timing (pre-dive, post dive, at different stages of the dive), intensity.
  • Thermal status: Dive duration, diving conditions, thermal changes during different stages of a dive.
  • Predisposition: Age, sex, DCS history, PFO history, genetics, epigenetics, circulation, obesity, smoking, acclimatisation, effects of drugs, fitness level.
  • Hydration: Dehydration, hyper-hydration.
  • Environment: Currents, waves, surge, distance to water entry/exit, type of entry/exit (a long hike, or climbing into a RIB), sea sickness susceptibility.
  • Behaviour and attitude: Competency, regularity and effectiveness of skills practice, complacency, normalisation of deviance, many forms of cognitive bias, pushing the boundaries, not challenging bad practices/ideas, over-reliance on equipment/technology (blindly following your computer, not understanding your computer), peer pressure, buddy.
  • Communication: proper briefings AND debriefings, especially on fun dives, being honest with yourself.

Some of the above are direct forms of decompression stress, in that there is an immediacy during a dive that could affect your physiology. Others such as behaviour or communication are indirect- they can influence the direct forms. So, when the current thinking suggests considering all forms of decompression stress and taking active steps to reduce them, it’s not a bad idea to step back every once in a while and look at how you approach to the dives that you do.

Whilst a saw-tooth profile is an obvious example of decompression stress, walking up the ladder in your dive gear at the end of your deco dive is less obvious, as it’s something you do all the time- you’re not exactly going to levitate back onto the boat are you, but you could take a brief rest before climbing the ladder, take it a bit more steadily, or de-kit if possible. A good example of this is seeing many sidemount divers keep both tanks attached as they climb up after a deco dive, maybe out of the macho attitude thing, pure habit, or even convenience of getting it over and done with. I see cylinder removal before exiting the water as one of the advantages of sidemount. I get it though, there are many reasons why it’s not possible to do anything other than walk up that ladder in your full kit. It’s just one of many examples to have a think about, but a particularly salient one given that we now know that bubbles can be shunted into the arterial system through sudden exertion. Some things you cannot avoid, but how you do them can make a difference. It’s way beyond this article to go into these in any kind of detail. The point is, thinking about how they relate to you and your diving may give you pause to reflect on any changes You can implement to give yourself a buffer against decompression stress. For much more detail, have a look at this really interesting presentation by Neal Pollack.

As for more specific decompression research, here’s an overview of what those t̶a̶s̶e̶r̶ ̶v̶i̶c̶t̶i̶m̶s̶  researchers have been up to.

Deep stops

Many technical divers incorporate a so-called deep stop into their dive profile, either via ratio deco, or as added by dive software that uses a bubble model, commonly VPM or RGBM. Dr Richard Pyle is widely known to have been one of the first divers to systematically incorporate deep stops into his decompression dives. He apparently felt a little badgered by the diving community to come up with a methodology as to what depths to do these deep stops at, so he winged it a little given the knowledge of the time. However, research undertaken by the developers of VPM and RGBM at around the same time added some theoretical credence to the necessity of deep stops. They have subsequently become an integral part of any dive planning software that uses VPM or RGBM algorithms.

We know that silent bubbles (Venous/Vascular Gas Emboli or VGE) are produced during a dive, and we also know that bubbles that reach a certain size and frequency are thought to be the main contributor to DCS. The thinking behind the bubble models is that the initial ascent from depth allows too much tissue supersaturation in gas content models such as Buhlmann’s ZHL. This, it is argued, allows bubbles to form and grow. Therefore, the aim of a bubble model is to produce fewer bubbles that are also smaller. This is achieved by ensuring that tissue supersaturation is delayed, and only able to occur in small increments, i.e. stopping earlier and therefore deeper on the ascent. To determine when to stop, bubble model algorithms track a theoretical bubble for each tissue compartment (instead of M-values), to ensure that it stays within a “critical volume” during the ascent.

Note that I said “theoretical” (if you can sense a “but” at this stage, it’ll be along in the next sentence). But (told you), there are problems with bubble models, and this is not really breaking news. None of these models has ever measured an actual bubble during a dive; these bubbles are derived by hypothesis and in-vitro experiment in labs with gels. It has been suggested that when you are deep, there are no bubbles to try and keep small in the first place[i]. Three separate studies that counted VGE scores following dives planned and executed using gas content and bubble models, all noted that bubble counts from the bubble models were either the same or greater than those of the gas content model. I realise that I wrote the word bubble a lot there.

The third study is perhaps the best known[ii]. It was conducted by the US Navy Experimental Diving Unit (US NEDU) and published in 2011. The outcome measure for the study was actual DCS in some volunteers (who was on that ethics committee?). The study was terminated as soon as there were statistically significant differences in DCS cases between the two models, and it ceased when the bubble model divers had 10 cases of DCS out of 198 dives (5.6%). The gas content divers had 3 occurrences of DCS from 192 dives (2%). Following the study, the US Navy declined the use of bubble models for their diving operations. One of the other studies involved the French Military coming to the same conclusion for their diving operations.

The conclusion of the NEDU study was that deep stops do protect the fast tissues from supersaturation, but it comes at a cost, and the cost is increased loading of slow tissues during those deep stops. These tissues only begin supersaturation after surfacing. The late supersaturation of slow tissues was considered the most plausible cause of DCS for the NEDU study. Bubble model software developers considered the study controversial, because of the way it was run and what it was designed to measure. They argued that the bubble model the US Navy used was different to their own particular bubble model, yet dives planned using VPM of the same length had even more deep stops than the NEDU profiles. You can easily find the studies themselves online, along with detailed discussion on scubaboard- remember to leave your will to live on the bedside table before logging in.

Another study involving United Team Diving (UTD) divers was published in March 2017[iii]. The divers were using two algorithms:

  • Buhlmann with 30-85 Gradient Factors (GFs- see below)
  • UTD’s ratio deco, which advocates deep stops

The short summary of that study is that the ratio deco divers had more inflammatory markers than the GF divers post dive- enough to be considered statistically significant. The bubble scores of the ratio deco divers were also very slightly higher than the GF divers, but not enough to be considered statistically significant. As outlined further in this article, inflammation is one of the many complicated links in the DCS chain that researchers are working hard to more fully understand the mechanisms of. What would have been useful in the study would have been an additional group of divers running a higher GF low of 40 or 50. More on GFs below.

As the online silt of the armchair dive committee continues to settle, decompression research continues, and the technical diving community is slowly shifting away from bubble models.

Gradient factors

So what algorithms do people use other than bubble models? Many technical divers plan their decompression dives using a gas content model such as ZHL-16b or c, with GFs added in. GFs allow you to set the percentage from ambient pressure towards the M-value line for leading tissues for both the deep and shallow portions of the dive, and the software or your dive computer will plot that “gradient” to follow during the ascent. The deep portion is called the low GF and the shallow part is the high GF. A common GF of 30/85 means a line drawn between 30% of the M-value on the deeper part of the dive, and 85% on the shallower portion. The closeness of the leading tissue compartment towards its M-value during ascent is determined by the steepness of the line in between. You can set your gradient factors to go over the M-value if you are so inclined, and you can also set them to be pretty similar to a bubble model (the low gradient factor set at 20% will give you a deep stop so that your leading tissues stay within 20% of the M-value, meaning not much of an ascent from current ambient pressure). But you can also go the other way. Although 30/85 is a common setting, many people now set their low GF to 40 or even 50, so it brings you up much shallower before giving you a stop, with the high GF set to somewhere between 70 and 85. A high GF of 70 will extend your shallow stops compared with 85.

To be clear, no-one knows what the optimal ascent profile should be for decompression dives. If GFs are the best approach, we don’t know what the best settings should be. Research does suggest that starting your deco shallower than as dictated by bubble models may create lower VGE counts, but how much shallower is not an answer anyone currently has. All that can be said is if you have extra time to extend your shallow stops within the limits of your dive (CNS, available gas, conditions, thermal exposure, presence of sharks), then do it. Also ensure that you understand GFs and any implications of adjustment before diving with them. Your buddy’s GF may work for them; it may not work for you.

Also bear in mind that the studies outlined above were undertaken on decompression dives only, not no-stop dives. As of January 2018, the Diver’s Alert Network (DAN) states that deep stops have been shown to be beneficial for recreational dives, though other studies had conflicting outcomes.

Vascular function

If someone asked you to name one of your organs, you would likely say liver, heart, lungs, or if you’re a pianist, Yamaha. I doubt you would say vascular endothelium, which is probably easier to say if you’ve been stung repeatedly in the mouth by a wasp. Anyway, it’s not just one of your organs, it’s a single layer of cells that covers the inner surface of every blood vessel in your body, so you could say it’s a fairly important part of you. As well as governing the wellbeing of your circulation, it regulates your vascular tone, which determines the level of vasoconstriction and vasodilation of your blood vessels. These are terms technical divers should be familiar with; the former occurs when breathing hyperoxic mixtures, the latter arises from hypercapnia caused by increased CO2 inspiration. Normal functioning of the vascular endothelium requires relaxed vascular tone, so that the blood vessels are dilated to allow greater levels of perfusion. This is affected during dives by levels of nitric oxide, oxidative stress, microparticles, and inflammation. These change the permeability of the vessel wall and are thought to contribute to vascular dysfunction. To say their individual and combined effect is complex and poorly understood is quite the understatement. Reading summaries of the studies is exactly like trying to speed-read an upside-down, backwards, Mandarin braille bible. Vascular dysfunction is thought to play a role in the initiation and consequences of DCS, with the effects potentially beginning at depth during a dive.

In cases of DCS it has been shown that microvascular endothelial cells lose their attachment to the basement membrane, which is likely to cause the endothelium to leak fluid and impair vascular permeability. Yet diving has been shown to change vascular function without DCS occurring. Another link in the chain is that increased numbers of microparticles have been found in divers after both simulated and real dives. Microparticles are essentially fragments of broken cells, most of which are derived from platelets. Platelets activate white blood cells and increase vascular permeability. They are linked to oxidative stress and result from inflammation. So microparticles are good markers for DCS studies.

Oxidative stress is a state of imbalance between oxygen free radicals and antioxidant activity. Oxygen free radicals can damage cellular lipids, membranes, proteins and DNA. There is some evidence that eating dark chocolate before a dive helps to limit oxidative stress due to its anti-oxidant properties, which is probably the best news ever. It has been postulated that oxidative stress deactivates the production of nitric oxide, which is otherwise thought to reduce inflammation and bubble formation. But the relationship between oxidative stress and nitric oxide is not clear and so far, studies have not shown consistent results. In one study, blocking nitric oxide production in rats before simulated dives increased the probability of DCS. Administration of nitric oxide prevented it. But in another study, the quantity of microbubbles post (simulated) dive was not affected by pre-dive administration of drugs known to increase production of nitric oxide.

The only thing we can really say for certain, is that diving temporarily impairs vascular function due to the interaction between oxidative stress, nitric oxide, inflammation, microparticles, and bubble formation. Vascular dysfunction has a role to play in the initiation and consequences of DCS, but the exact mechanisms are not understood, and more studies are required.



Whilst it has always been thought that exercising before diving is something that should be avoided, there is now evidence that this advice is incorrect. I can hear the collective groans of dive master trainees all over the world as they load their dive boats with cylinders. Numerous studies on rats and humans (not at the same time) have shown that VGE counts post dive are significantly reduced if the diver undertakes some form of exercise between 24 hours and 1 hour before a dive. This includes both moderate aerobic exercise like jogging or cycling, and intense anaerobic exercise, like jogging or cycling harder, or an intense game of speed chess.

It’s not known exactly why this is, but numerous hypotheses have been put forward.

  • One theory is that exercise increases nitric oxide production, which, as previously outlined has a positive influence on vascular function.
  • Exercise also dilates blood vessels, and that could help to eliminate gas nuclei before they can act as bubble seeds.
  • Other theories explain that an increase in blood perfusion reduces blood volume, which decreases blood pressure and reduces blood flowing to the tissues. This may lower inert gas uptake.

Advice on exercising after diving remains unchanged, don’t do it, it’s really bad. Slow chess is fine.

What about during a dive? It’s drilled into divers to not exert themselves at any point during a dive, as this has consequences in terms of increased CO2 retention, CNS risk, increased gas consumption, and increased narcosis at depth, as well as looking like a wind-up toy has just been released. Similarly, for decompression diving historical good practice was to be as relaxed as possible on the deco ascent, and remain sedentary so you reduce the risk of CNS and don’t aggravate the off-gassing process. However, evidence now suggests that some movement during deco is going to promote greater off-gassing through increased perfusion. If you’re doing a little bit of swimming during your bottom time and then doing nothing on the ascent, all that gas in your tissues will come out at a slower rate through in an imbalance in perfusion rates between on and off-gassing. The ideal state would be minimal movement on the bottom phase and a little activity on the ascent. We’re not talking push ups on the trapeze, just some gentle regular finning or backfinning practice (now there’s an idea).


Hydration has also been shown to enhance the beneficial effects of exercise on VGE counts, both pre and post-dive. Dehydration has long been considered a risk factor of DCS, and this is still the case. Being dehydrated before a dive has the effect of thickening the blood. This leads to something called hemoconcentration, which is an increase in red blood cell concentration, caused when the level of blood fluid drops. Because of this, any bubbles in the bloodstream are retained more easily, which is believed to be one of the many factors that can instigate DCS.

Diving is always considered to be a relaxing and serene experience, but from your body’s point of view it’s not quite the yoga session you thought it might be. During a dive, something called central blood shift occurs because of the effect of hydrostatic pressure on your body.

For the average person, around 700ml of blood is moved to the body’s core from the limbs. This engorges the capillaries and raises cardiac filling pressures by 15-20 mm Hg, which in turn causes the heart to increase in size by about half a litre. Diuresis results as your body tries to respond to reduce the blood pressure, leading to a reduction in blood plasma volume, and potentially dehydration (and back we go to hemoconcentration).

Studies have shown that divers who have symptoms of DCS do not have hemoconcentration immediately after the dive, but by the time they get to a chamber, they do. It’s thought that this is due to fluid moving from the cells to the blood vessels during the dive, to compensate for the loss of fluid in the blood. Around 20 minutes after surfacing, the fluid moves back to the cells. As an aside, there is some evidence that dehydration before a dive reduces inert gas loading during the dive because the thicker blood can’t carry as much gas. But this is likely more than countered by the subsequent effects of hemoconcentration post dive.

Tests on military divers showed that drinking a saline-glucose drink before a dive significantly reduced VGE counts after the dive. On longer deco dives you should if possible drink fluids during the decompression phase. After any dive you should rehydrate. But remember, there is effective hydration and useless, waste of time hydration. Being dehydrated before a dive, then realising it and chugging a litre of water will not change anything, it will just cause diuresis, which makes the situation worse. Instead try and stay hydrated all the time by drinking small amounts of water regularly.

Oxygen pre-breathing

Of all the different factors outlined in this article, the breathing of oxygen before dives has produced the most consistent results that show it to be beneficial in reducing DCS risk. There may be multiple reasons why.

Ironically, hyperbaric oxygen has antioxidant properties, and also anti-inflammatory properties, and has been shown to reduce platelet activity post dive. But most people don’t have access to a hyperbaric chamber before a dive. It’s not hugely surprising that oxygen has this effect, given that it’s the primary treatment for DCS.

Pre-breathing oxygen in between dives also increases the rate of microbubble washout. However, dive computers have no way of knowing that you’ve been pre-breathing oxygen, so you should treat it in the same way you would if you were diving nitrox with your computer set to air- as a way of adding conservatism, instead of attempting to informally reduce surface interval times or extend repetitive dive bottom times.

Pre-dive heat exposure

Believe it or not, someone thought it would be a great idea to put someone in a sauna, then send them off on a simulated dive to see whether it affected their VGE count. Wouldn’t you know, there were significantly reduced VGE bubbles after the chamber dive. Other studies have shown similar results. But why would pretending to be in Finnish have this effect on a diver?

Researchers think it may be down to so-called heat-shock proteins. These are known to interact in some way with nitric oxide as it relates to endothelial function. But as with anything to do with saunas and Finns, more research is needed before every dive centre should install one (a sauna, not a Finn).

Pre-dive vibration

Is it me, or are these things getting weirder the further along this article we get? Although vibration does makes sense on some level. Combat divers apparently believed that if they drove fast to the dive site and slow on the return, it would help reduce their DCS risk. Whether that’s true or not I don’t know, but subjecting people to whole-body vibrations for 30 minutes before a dive did reduce their VGE scores following the dive. The cause could be related to changes in blood flow, that in turn induces changes to the endothelium.

Another idea is that friction detaches gas nuclei from the walls of blood vessels. Either way, I now have visions of people in saunas sitting in vibrating chairs, breathing oxygen whilst playing speed chess.

Immersion Pulmonary Oedema (IPO) (IPE in the US- Edema)

Although physiologically distinct from DCS, IPO is now considered to be a common cause of death during diving. The reason that it is not attributed as the cause of many diving fatalities is because it is often mistaken for drowning (also the end result of heart-related problems during a dive). As the name suggests, it is precipitated by immersion- swimming or diving. Two-thirds of triathletes that die during the event do so during the swimming section. Most of these deaths are believed to be due to IPO. It is thought that 1% of scuba divers will get IPO at some point when diving.

Symptoms of IPO include cyanosis, coughing and breathlessness (rapid breathing), and production of froth and blood from the lungs. Rapid breathing can reduce the PPO2 of oxygen getting to the brain, and this, combined with the drop in the PPO2 in the breathing gas of open circuit divers during ascent, mean that confusion and/or shallow water blackout is also a possibility, which may lead to death.

Factors that increase the risk of IPO include cold water (causing vasoconstriction), hyper-hydration (increased blood pressure), exertion and stress, high PPO2 (vasoconstriction), underlying heart disease and hypertension. But IPO can also occur in fit and healthy individuals, and even athletes (triathletes).

It occurs due to pressure changes between the intra-alveolar pressure and capillary pressure in the lungs. If blood pressure in alveolar capillaries increases and gas pressure in the alveolar sac is reduced (usually during inspiration), then there is a greater risk of fluid being pulled out of the alveolar capillary and into the alveolar sac via negative pressure. Under normal conditions, fluid does not enter the alveolar sac because the blood contains around 40g/L of albumen. Albumen is hydrophilic, and effectively imposes a suction pressure of around 23mm Hg.

As long as gas pressure in the alveolar does not exceed this, then IPO will not occur. However, immersion in water creates a hydrostatic pressure on the body, which can create a negative pressure during inhalation. In combination with the other outlined factors, negative inhalation pressures during diving can be enough to exceed 23mm Hg and cause IPO. Rebreather diving is a particular risk factor, due to the use and location of counterlungs, and being exposed to higher PPO2s for longer periods (vasoconstriction).

If someone is displaying symptoms of IPO, sit the casualty upright and provide 100% oxygen. Do not administer any fluids, and monitor them until handing them over to medical professionals. Once in hospital they may be given diuretic and vasodilator drugs, Continuous Positive Airway Pressure (CPAP), and they may need mechanical ventilation. IPO can be fatal, and it can recur. If someone has had a IPO and survived, it is generally recommended that they stop diving, unless there is an underlying cause that can be treated.

I’m sorry there is no concise conclusion to all this research that strongly suggests doing this or that to stay safer during your dives. I believe that current decompression research is opening up some very interesting avenues that may help to do just that in the future. But for now the best advice is to think about the wider picture in terms of decompression stresses, and see what you can do to reduce them collectively during your own diving.

Staying up to date with developments in decompression research is not easy, and often relies on the good work of the Rubicon Foundation, which releases diving-related scientific papers one year after publishing. This is fantastic of course, but the papers themselves can often be a daunting read. Thankfully, Simon Pridmore had the great idea of getting the decompression researchers themselves to summarise their studies in plain English. The outcome of that was the recently published book called Scuba Physiological. The sections of this article on vascular function and preconditioning were s̶t̶o̶l̶e̶n taken from this excellent read. If you want to get it from the horse’s mouth in much greater detail, I highly recommend getting yourself a copy.

[i] Neil Pollock 2016, BSAC presentation on decompression stress.

[ii] 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, 2011.

iii Spisni E et al. A comparative evaluation of two decompression procedures for technical diving using inflammatory responses: compartmental versus ratio deco. DHM 2017;47(1):9-16.

Leave a Reply