Connect with us

Diving Medicine

Is Oxygen Narcosis A Thing?

There’s been a long running debate as to whether oxygen is a narcotic diving gas. Training agencies CMAS, GUE, NAUI, PADI and PSAI include O2 in their equivalent narcotic depth (END) calculations. Others like BSAC, IANTD and TDI do not. The problem has been reliably measuring gas narcosis. Enter medical researcher and tech instructor Xavier Vrijdag, who has developed a novel EEG algorithm that can detect the subtle effects of gas narcosis on the brain. His results promise to lead to a deeper understanding of gas narcosis. Hint: Nitrox divers can breathe easy; it’s less narcotic than air!



By Xavier Vrijdag PhD. Header image: Drs. Vrijdag and Hanna van Wart preparing participant in front of the hyperbaric chamber. The EEG electrodes are placed inside the cap and two more electrodes are placed under the eyes. Photo by Payal Razdan.

🎶🎶 Predive Clicklist: Love Is Like Oxygen by Sweet (1978)

🎶 Love is Like Oxygen,
You get too much, you get too high,
Not enough and you’re gonna die.
Love gets you high 🎶

To sustain life, every breathing gas mixture for divers must include oxygen. As we descend, invariably, we will breathe oxygen at higher partial pressures than air at the surface. Oxygen is mixed with other gasses, like nitrogen and helium, to create suitable mixes for specific depths. One of the boundaries of going deeper is the experience of gas narcosis. 

The effects of gas narcosis increase with a diver’s depth because the ambient pressure and the partial pressure of the individual gasses in the breathing mix increase. Gas narcosis has traditionally been linked to nitrogen when breathing air. Narcosis causes behavioral changes and impaired cognitive abilities. Substituting helium wholly or partly for nitrogen ameliorates the narcosis effects. Similarly, it has been argued that replacing some nitrogen with oxygen reduces a gas mixture’s narcotic potency.

One tool to quantify gas narcosis in different breathing mixtures is equivalent narcotic depth (END). In this method, the narcotic effect of a certain breathing gas mixture, like trimix, is compared to breathing air. With the partial pressures for oxygen, nitrogen, and helium in the gas mixture and the intended dive depth, one can calculate the depth at which air would produce a similar narcotic effect. A long-standing debate within the dive industry is whether oxygen should be included in the ‘is narcotic’ or ‘is not narcotic’ portion of this calculation (See: Calculated Confusion: Can O2 Get You High? by Reilly Fogarty)

One of the most used approximations of the narcotic potency of gasses is the Meyer-Overton correlation. The Meyer-Overton correlation describes a relationship between the narcotic potency of a gas and its solubility in oil; the more soluble, the more narcotic. Oxygen has a higher solubility in oil than nitrogen. Therefore one could argue it is more narcotic than nitrogen. However, oxygen is special since we metabolize it.

Rebreather Forum 4

Nitrogen is a non-metabolic gas, meaning that the inhaled partial pressure of nitrogen equilibrates with the partial pressure of nitrogen in brain tissue after a short wash-in period. In contrast, oxygen is metabolically consumed, thus lowering the partial pressure of oxygen considerably below the inspired pressure. Therefore, despite its greater solubility in oil, one could argue it might be less narcotic than nitrogen. 

In addition, exceptions to the Meyer-Overton correlation exist; some gases that are very soluble in oil should be highly narcotic but turn out not to be. In anesthetic research, we have since learned that for a gas to be narcotic, its molecule needs to be able to bind to receptors on the neurons to cause its narcotic effect. Certain molecules have a shape that does not allow them to bind to these receptors. For oxygen, it is unknown if and how it binds to neuronal receptors. The point here is that just because oxygen is more oil-soluble than nitrogen, that does not automatically mean it is more narcotic than nitrogen. 

Participant seated inside the hyperbaric chamber, wearing the EEG cap. The breathing system is hanging above the participant, which is used in combination with a nose clip to administer the breathing gases. Photo courtesy of X. Vrijdag

Measurement Methods

Measuring the cognitive effects of nitrogen narcosis is most often done with psychometric tests, like math questions, motor tasks or memory queries. Each of these skills is affected at different partial pressures of nitrogen and is impaired at different depths breathing air. The book, Psychological and Behavioral Aspects of Diving by Nevo and Breitstein summarizes the scientific literature quite well. Higher cognitive functions, like reasoning and memory are impaired in the shallow range (~ 5% at 3 ATA (20 m/66 ft), 30% at 6 ATA (50m/164 ft)).

In contrast, more basic skills like gross motor tasks are only impaired at far higher pressure. Quantification of the subtle effects of inert gas narcosis is challenging. Psychometric tests can be affected by learning effects, participant motivation or boredom. Objective neurophysiological measurements like the critical flicker fusion frequency (CFFF) and quantitative electroencephalogram (qEEG) analysis could, in principle, be used to overcome these issues.

During the CFFF test, the participant looks at a small light oscillating rapidly on and off (flickering). The fusion frequency is obtained by increasing the flicker frequency until the participant perceives a change from flicker to fusion (no further flickering) or the other way around. An increase in CFFF equates to an increase in alertness, whereas a decrease in CFFF is associated with a decrease in alertness, for instance, caused by gas narcosis. Some studies have shown a decrease in CFFF during air-breathing dives at 4 ATA.

  • Rebreather Forum 4

Hyperbaric inert gas narcosis is generally considered a manifestation of depressed brain activity when measured with EEG at large depths. Continuous measurement of the EEG results in a considerable amount of data, and analysis is usually performed afterward. This requires much training and time. qEEG has the potential for real-time analysis of the continuous EEG, giving the user an easy-to-interpret result about the narcotic effect. The subtle changes of gas narcosis at typical sport diving depths would otherwise be hard to detect in the EEG. Such analyses would allow comparison between divers and successive dives.

Measuring hyperbaric oxygen and nitrogen’s comparative effects on the brain is difficult. Ideally, the comparison should occur at equivalent inspired pressures of the two gasses. However, nitrogen narcosis symptoms typically present at depths greater than 30m/100 ft (4 ATA); a pressure at which oxygen-breathing would carry the risk of cerebral oxygen toxicity. A pressure of 18 m/60 ft (2.8 ATA) is considered the maximum safe inspired oxygen pressure in hyperbaric oxygen treatments with resting patients in a dry hyperbaric chamber. [WARNING: A PO2 of 2.8 ATA is incompatible with diving!] It follows that the measurement method to detect gas narcosis needs to be sensitive enough to catch the subtle effects of gas narcosis in the more limited and safe inspired pressure range of oxygen. 

In a research program funded by the US Navy Office of Naval Research, I studied the subtle narcotic effects of nitrogen and oxygen together with Dr. Hanna van Waart, Professor Jamie Sleigh, and Professor Simon Mitchell. 

Measuring The Subtle Effects Of Nitrogen Narcosis

In the nitrogen narcosis study, twelve technical divers breathed air and 21% / 79% oxygen-helium (‘heliox’) at 18 m/60 ft and 50 m/164 ft (2.8 and 6 ATA, respectively) inside a hyperbaric chamber while we recorded the electrical signals of the brain (EEG), performance on a mathematics test and the CFFF. Breathing air at 50m/164 ft (6 ATA) is typically cited as causing sleepiness, euphoria, overconfidence, idea fixation, and impaired reasoning, memory, calculus, and judgment. However, these symptoms may be alleviated by enhanced concentration in motivated divers.

The CFFF did not change during the air-breathing hyperbaric exposure to 6 ATA compared to surface recordings. Hence, the CFFF did not detect the nitrogen narcosis effect at this depth. We also reviewed the CFFF diving literature and found a complicated and often contradictory picture. Various studies during air breathing conducted at 4 ATA have suggested that nitrogen narcosis is the cause of a reduction in CFFF at this depth. However, this result does not extrapolate to greater depth, even though it is known that cognitive performance is further reduced with increased depth. Several studies—including our experiments—performed at 6 ATA while breathing air did not show impairment, but instead showed either no change or an increase in CFFF. Hence, in our study, CFFF failed to detect or quantify a narcotic effect known to be present at 6 ATA.

On the contrary, we successfully developed a quantitative EEG metric to measure narcosis produced by nitrogen at hyperbaric pressures. The functional connectivity metric is based on the so-called mutual information analysis, and was summarized using the global efficiency network measure. Mutual information measures the information that is similar between two signals, calculated between all EEG channels. Higher mutual information means that the signals are more alike. 

Head plots of EEG functional connectivity for air-breathing exposures. Plots are showing increased connections between electrodes with increased pressure, while breathing air.

Global efficiency measures how well connected a network of channels is. High global efficiency based on mutual information means that more signals are similar. Normal cognitive function requires regional specificity, which might be lost (that is, signals become more similar) during impairment.

The novel EEG method successfully differentiated between breathing air at the surface and air at 50m/164 ft (6 ATA) with a 35% increase in the functional connectivity metric. To our surprise, we could even detect a trend of a 19% increase at the much lower pressure of 18 m/60 ft (2.8 ATA). This indicates a dose-response between EEG functional connectivity and nitrogen narcosis. No significant change in the metric was found during the hyperbaric heliox exposures, which was expected because helium is known to be a non-narcotic gas. This lack of an effect of heliox confirmed that nitrogen exposure and not the environmental pressure change was responsible for the changes we saw in EEG functional connectivity. 

  • Rebreather Forum 4

Another implication of the increased functional connectivity metric during air breathing at 18 m/60 ft (2.8 ATA) is that it should be sensitive enough to measure the narcotic effects of oxygen at 18m/60 ft inside a hyperbaric chamber, if such effects exist. As stated above, we can safely breathe oxygen at 18 m/60 ft inside the safety of a hyperbaric chamber, where there is no risk of drowning, compared to being submerged. If oxygen were to have similar narcotic effects as nitrogen, we expect that functional connectivity would change in a similar way during oxygen exposure as it did in the air-breathing experiment at 18 m/60 ft.

Measuring Oxygen Narcosis

In the oxygen narcosis study, twelve technical divers breathed 100% oxygen at the surface, 4 m/13 ft and 18 m/60 ft (1, 1.4, and 2.8 ATA, respectively) inside a hyperbaric chamber while we recorded the EEG. Four meters/13 ft (1.4 ATA) was chosen as it approximates the upper limit of oxygen exposures accepted in diving, while 18m/60 ft (2.8 ATA) is the maximum exposure to oxygen inside a hyperbaric chamber with a minimal risk of oxygen toxicity.

The EEG functional connectivity metric did not increase while breathing hyperbaric oxygen. This contrasts with the 19% increase we reported above in participants breathing air at 18m/60 ft (2.8 ATA). This suggests that oxygen is not producing the same changes in brain electrical activity seen during hyperbaric air breathing.

It is only now, with the development of this novel EEG analysis algorithm, that we have been able to investigate the subtle effects of hyperbaric oxygen in more detail, and based on our results we hypothesize that oxygen probably does not bind to the same neuronal receptors and hence does not cause similar narcotic effects to those induced by nitrogen. For nitrox divers, this probably means that the increased oxygen slightly reduces the narcotic potency of the breathing gas.

Dive Deeper

OzTek presentation: Measuring gas narcosis in divers 

British Journal of Sports Medicine: Towards gas narcosis monitoring in compressed gas diving (PhD Academy Award). (2022) 

Scientific Reports: EEG functional connectivity is sensitive for nitrogen narcosis at 608 kPa. (2022) 

Physiological Reports: Does hyperbaric oxygen cause narcosis or hyperexcitability? A quantitative EEG analysis. (2022) 

Diving and Hyperbaric Medicine Journal: Investigating critical flicker fusion frequency for monitoring gas narcosis in divers. (2020)  

InDEPTH: Calculated Confusion: Can O2 Get You High? by Reilly Fogarty (2019)

Alert Diver.Eu: Rapture of the Tech: Depth, Narcosis and Training Agencies by Michael Menduno (2020)

Alert Diver.Eu: Measuring Inert Gas Narcosis by Michael Menduno (2020)

Xavier Vrijdag is a diving medical researcher at the University of Auckland investigating the effects of gas narcosis in divers on the brain. He has a master’s degree in Technical Medicine for the University of Twente, the Netherlands, where he developed an algorithm to quantify cerebral arterial gas embolism in the hyperbaric environment. In 2022, he completed his doctoral thesis on the effects of nitrogen, oxygen, helium and carbon dioxide under pressure.

Xavier has worked as a researcher and technical physician at the department of Diving and Hyperbaric Medicine of the Academic Medical Centre, Amsterdam, the Netherlands. Before coming to New Zealand, he worked as a hyperbaric technician and researcher at Deep Dive Dubai. He is an instructor for medical courses of DAN Europe, holds certifications from seven training agencies, and is an SSI divemaster instructor, normoxic trimix instructor, cave instructor and freediving level 2 instructor, and is a GUE fundamentals instructor. 

Besides work, Xavier is an avid cave and wreck diver and photographer, as well as being part of a CaveSAR/LandSAR cave rescue team in New Zealand.

Subscribe for free

Diving Medicine

Do Freedivers Get Bent?

Decompression sickness (DCS) is a known risk for compressed gas divers, but until recently it wasn’t on the radar for their breath-hold counterparts. However, as a result perhaps of the growing numbers of freedivers and competitions, and the fact that elite divers now consistently push 100 m/326 ft depths, the number of possible DCS-related incidents is on the rise. Here science writer Reilly Fogarty explains what we know and what we don’t know about freediving DCS, and where further research may lead us.




By Reilly Fogarty. Header image and photos by Lorenzo Mittiga unless noted.

Decompression sickness (DCS) is old hat for divers; breathe inert gas under pressure, eliminate it too quickly on ascent, and Voilà! The changes in ambient pressure, the potential for tissue saturation, and the constant supply of inert gas make decompression sickness first and foremost on a diver’s list of concerns. 

Until recently, breath hold divers weren’t even on the radar of academics researching DCS—their dives were too shallow and too short in duration, the supply of inert gas held in a single breath was considered inconsequential, and injuries went unreported or improperly diagnosed. In the span of a few short years, a number of seemingly unrelated events occurred that changed the way we understand freediving physiology. 

First and most importantly, the sport itself exploded in popularity. Uninitiated outsiders watched freediving organizations, shops, and competitions pop up seemingly overnight. Suddenly, millions of people across the globe were learning to dive without SCUBA gear, and the safety and medical communities had to play catch-up. Not only were freedivers more numerous, but they were also becoming more advanced and aggressively pushing the limits of their sport. Soon divers weren’t just exploring the shallows, but reaching 152, 183, and 214 meters (500, 600, 702 feet) on a single breath (No Limits Apnea)—far deeper than all but the most extreme SCUBA divers. 

At the same time, researchers were noticing incidents of decompression sickness involving marine mammals. Whales, dolphins, and seals became the notable subjects of research involving decompression sickness and involvement of DCS in unexplained behaviors, injuries, and fatalities. While the physiology of these animals differs from our own, this research runs parallel to the work of academics who are working to to understand how breath hold divers could be experiencing DCS symptoms.   

While we were learning about how our marine counterparts experienced DCS, the incidents of DCS among freedivers increased—although it’s unknown whether this is more accurately attributed to the growth in participation or to deeper dives by athletes. Here’s what we know now:

Delivering oxygen therapy to an athlete. Photo by Jeroen Van Haudt

Freediving Physiology

It would be easy to assume that the physiological mechanisms at play in breath hold divers are identical to their better equipped SCUBA counterparts, but research has shown that this is not the case at all. While divers of both types share common fundamental responses to immersion, cold, and depth, SCUBA divers primarily have to contend with gas loading from a constantly replenished supply of inert gases. Breath hold divers, however, have the opposite problem and must perpetually race a depleting supply of oxygen in their lungs. The unique adaptations that make elite free divers capable of reaching extreme depths also correlate with specific changes in skeletal and cardiac muscle responses, respiratory symptoms, and neurological symptoms that resemble DCS. 

Among cardiac and skeletal muscles, some research has indicated that the combination of factors unique to breath hold diving may exacerbate exercise-induced muscle fatigue. The combination of extreme environmental conditions “prolonged physical activity and complex adaptation mechanisms” led to a notable increase in several markers of muscle stress following breath hold dives among participants. Whether this is a result of a specific facet of the breath hold dive—be that exposure, hypoxia or some other factor —remains to be determined, but the data indicates that breath hold diving can cause significant muscle stress. Tying these markers into modern decompression sickness theories involving systemic stress and endothelial involvement brings to light a number of fascinating possibilities for the development of DCS among freedivers. 

Photo by Lorenzo Mittiga

The extreme stress that athletes put on their lungs during a breath hold dive also contributes to the physiological differences among apnea divers. Repeated breath hold diving has been associated with pulmonary edema, and symptoms of acute respiratory distress syndrome, which are seen as lung comets (a measure of extravascular water in the lungs). Just as with muscle fatigue, it is difficult to specifically identify which facet of breath hold diving causes the greatest number of—or the most symptoms of—these conditions, but the correlation has been validated by several studies. Just as with SCUBA diving, it’s easy to see how an impairment in the pulmonary system could alter inert gas elimination and lead to DCS symptoms. 

Taravana—neurological symptoms that occur after breath hold diving—are both frequently reported and now identified as a type of DCS. While easily identifiable gas embolisms are rare among freedivers, these neurological symptoms have been noted for some time, but are just now being understood as indicative of decompression sickness. Most of these symptoms occur in relatively benign situations, with exposure profiles that would seem low-risk, but their evolution has been correlated with the types of bubble counts that one would expect to see among SCUBA divers.

By the published data alone, the risk of DCS for freedivers above 100 meters is essentially zero, and it reaches a maximum of 5 to 7 percent near 230 m/755 ft. The increase in the number of divers worldwide, and the depths those divers reach, has led many to believe that prior data may underestimate DCS risk, and has brought closer attention to many previously overlooked post-dive symptoms and incident reports. This research is ongoing, so the actual risk of DCS is virtually impossible to predict, but it seems likely that greater attention will result in an increase in the published occurrences of DCS among freedivers. 

Photo by Lorenzo Mittiga

New Decompression Models? Don’t Hold Your Breath

The combination of these factors, extreme stress to the muscular, cardiac, and pulmonary systems, common neurological symptoms, and pulmonary fluid concerns leads to an academically fascinating place—if one with few concrete lessons to take home. What we are left with are a number of factors that we know affect DCS risk among SCUBA divers, combined with significant neurological symptoms and exacerbated by pulmonary concerns. 

It’s a tough nut to crack—with known inputs and outputs, but very little understanding about the chemical pathways—and our current tissue supersaturation models just aren’t well equipped to estimate the risks involved. Without knowing more about how and why DCS symptoms evolve among freedivers, it is difficult to know exactly how to keep divers safe. 

Photo by Lorenzo Mittiga

Several training agencies have developed rudimentary protocols to limit dive depth and increase time between dives, and these are absolutely worth considering. Most agencies recommend the following as a general protocol:

  1. Limit dive sessions to under 2 hours, and terminate a session if a diver becomes cold.
  2. Calculate surface interval times based on depth and dive time to minimize gas loading. This is done for dives under 30 m/100 ft by doubling total dive time and taking an equivalent surface interval. For deeper dives up to 60 m/200 ft, surface interval can be calculated by dividing maximum depth by 5. 
  3. Limit divers diving deeper than 55 m/180 ft to one per day, and recommend breathing oxygen for at least 10 minutes following a dive past this depth.
  4. Ascend at a rate no greater than 1 meter per second.  
  5. Follow standard SCUBA diving recommendations regarding diving and hydration, flying, and post-dive exertion.

It is important to understand that these recommendations are empirical and require scientific research to become more convincing.

  • Rebreather Forum 4

Working divers and those reaching extreme depths more frequently may need greater conservatism to minimize their risks of DCS, which is why Performance Freediving International founder Kirk Krack has begun to develop mixed gas freediving protocols and training courses. Krack’s protocols involve promoting the use of nitrox pre-breathing before a dive to minimize both the risk of DCS and hypoxia. 

Photo by Lorenzo Mittiga

The use of pre-breathing gasses in freediving is controversial, but divers like Krack and world record holder Eric Fattah believe the benefits are significant. “Nitrox adds a dramatic safety factor,” Fattah claims. “CO2 becomes the limiting factor when diving with nitrox. It’s almost impossible to blackout.” Fattah’s claims have not been verified, particularly the statement that pre-dive nitrox breathing decreases the probability of blackout. It’s easy to see how there may be some notable benefits for freedivers. However, although there is interest in mixed gas use, the International Association for the Development of Apnea (AIDA), has not adopted the use of nitrox for safety divers. Based on their safety diver protocols, the organization does not see a need. However, breathing of pure O2 and O2-enriched gas mixes is not allowed within one hour before the start of performance in freediving competitions.

Shy of providing the conclusive mechanism of injury yourself, the best course of action for the most common types of freediving appears to be conservatively planning dive times and surface intervals, considering mixed gas training, and learning how to treat DCS should you begin to show signs and symptoms. It is unlikely that decompression algorithms will be published—let alone validated for freedivers—in the next few years, but it is possible to prevent the vast majority of injuries with only a conservative approach and a solid understanding of gas dynamics in the body. With any luck, the convergence of research into deep-diving marine mammals and into the physiology of freedivers will shed some light on how the stresses of apnea diving can increase DCS risk, and we may all be able to learn something from hooded seals during our next freediving course. 


Gregory S Schorr, Erin A Falcone, David J Moretti, Russel D Andrews

First long-term behavioral records from Cuvier’s beaked whales (Ziphius cavirostris) reveal record-breaking dives. PLoS One 2014; 9(3)

Danilo Cialoni et al, Serum Cardiac and Skeletal Muscle Marker Changes in Repetitive Breath-hold Diving. Published online 2021 Aug 21

Kate Lambrechts 1, Peter Germonpré, Brian Charbel, Danilo Cialoni, Patrick Musimu, Nicola Sponsiello, Alessandro Marroni, Frédéric Pastouret, Costantino Balestra Ultrasound lung “comets” increase after breath-hold diving. Eur J Appl Physiol 2011 Apr;111(4):707-13

D Cialoni 1 2, M Pieri 1, G Giunchi 1, N Sponsiello 1 2, A M Lanzone 3, L Torcello 4, G Boaretto 5, A Marroni 1 Detection of venous gas emboli after repetitive breath-hold dives: case report. Undersea Hyperb Med  Jul-Aug 2016;43(4):449-455.

J.R. Fitz Risk of decompression sickness in extreme human breath-hold diving. UHM 2009, Vol. 36, No. 2

J. D. Schipke and K. Tetzlaff. Why predominantly neurological decompression sickness in breath-hold divers? J Appl Physiol V120 no.12 June 2016 Journal of Applied PhysiologyVol. 120, No. 12

T Andrews Freediving And DCS? SDI|TDI|ERDI|PFI Blog Aug 2020

M. Menduno. Technical Freediving: Are Breathhold Divers Ready To Mix It Up? July 26, 2019

Dive Deeper

Shearwater Blog: Do Whales and Sea Turtles Do Decompression? by Guy & Anita Chaumette

YouTube: -214m (702ft) DEEPEST FREEDIVING EVER – Herbert Nitsch World Record No limits Extreme Diving Apnea

YouTube: Freediver Herbert Nitsch – WR#20 2007 No Limit 702 ft (214 m)

DeeperBlue: Technical Freediving: Are Breathhold Divers Ready To Mix It Up? by Michael Menduno

DeeperBlue: Technical Freediving In Hollywood Revealed

To explore additional stories, videos and webinars on freediving click: FREE

Reilly Fogarty is an expert in diving safety, hyperbaric research, and risk management. Recent work has included research at the Duke Center for Hyperbaric Medicine and Environmental Physiology, risk management program creation at Divers Alert Network, and emergency simulation training for Harvard Medical School. A USCG licensed captain, he can most often be found running technical charters and teaching rebreather diving in Gloucester, Massachusetts.

Category: latest features, diving medicine, freediving, diving safety, 

Continue Reading

Thank You to Our Sponsors

  • Fathom


Education, Conservation, and Exploration articles for the diving obsessed. Subscribe to our monthly blog and get our latest stories and content delivered to your inbox every Thursday.

Latest Features