Venous gas bubbles in breath hold divers

Venous gas bubbles in breath hold divers remained a focus of researchers this year, with a notable presentation coming from Danilo Cialoni and his EDAN team1.  At EUBS 2017 they presented the extension of study previously reported and described in this blog. After discovering post-dive VGE in one breath hold diver, they studied VGE in 37 elite breath hold divers during their training in 42 meter deep pool with water temperature  of 32 oC.

Divers underwent echocardiographic Going downmonitoring before the series of dives, after a number of training dives, and every 15 minutes for up to 90 minutes after the last dive. Bubbles were detected in 39% of divers (28% low VGE grade and 11%  high VGE grade). Bubblers did significantly longer and deeper dives with shorter surface intervals. The data from this study will be used to correct the decompression algorithms for breath hold divers, which primarily means extending the time between the dives to prevent carrying over dissolved gas from one dive to another. Four divers did develop neurological symptoms of taravana during the study. All symptoms were mild and divers recovered after breathing oxygen at surface. Most notably, in one diver with taravana, bubbles were not discovered.

Another taravana case unrelated to this study was presented by another group2. A 39 year old diver performed about 30 dives over the course of 5 hours to depths between 29 and 32 meters, with dive times between 2 and 2.5 minutes each. A few minutes after his last dive the individual developed expressive aphasia (difficulty speaking and expressing thoughts) and a headache. The aphasia resolved shortly but the headache persisted and diver was admitted to an emergency department 48 hours post dive. The diagnostic workup included the brain MRI which revealed a brain injury. The patient was treated with one Table 6 and five HBO treatments at 2.5 ATA on the following days. His conditions significantly improved after the treatment and at 2 months follow up he was completely recovered.

Competitive breath hold divers should be aware that post-dive symptoms may be caused by brain injury and regardless of assumed cause (decompression or hypoxic) they need neurological examination and treatment in case of confirmed injury.


  1. Cialoni,D. et al. Prevalence of venous gas emboli in repetitive breath hold diving. Proposal for a new decompression algorithm. P 17
  2. Guerreiro F, et al. Decompression illness in extreme breath hold dive (Taravana syndrome) – A case report. P 47

What’s Left to Learn about Bubbles?



EUBS 2017 has left us with more questions than answers, on the topic of post-dive bubbles.

Ballestra presented the preliminary results of an exploratory study of the effects of sonic vibrations on post-dive venous gas emboli detected by transthoracic echocardiography1. Six divers performed dives to a depth of 33 meters depth for 20 minutes, in a fresh, warm water pool. Bubbles were detected in a standard way, with and without exposure to a sonic vibrations of 20 to 30 Hz. The amount of detected bubbles nearly doubled after sonic vibrations. If these preliminary result get confirmed, we will have to be concerned with post-dive exposure to a sonic noise from various sources, like music, helicopter vibrations and similar, and it is possible that we could find some convenient and fun ways to pre-condition our bodies before dives. This should be also considered in DCS cases in aircraft pilots.

In another study, the presenting author used contrast echocardiography, a standard clinical method, to monitor divers post-dive2. Unlike the more commonly used B-mode echocardiography,  which can detect circulating bubbles greater than 35 microns, the contrast echocardiography can detect much smaller bubbles (< 10 microns). Post-dive contrast echocardiography in seven divers did indeed show the presence of small bubbles in the right and left heart, even in absence of large bubbles detectable by standard B-mode echocardiography. Of particular note is the fact that the presence of small bubbles did not correlate with the amount of large bubbles detected.

The final study was a classical bubble study done by scientists from the Swedish Navy to evaluate the safety of the US Navy Diving Manual Revision 6 air decompression tables. Twenty-eight divers did 72 dives in controlled conditions with three different dive profiles at the no-D limit, or with one required decompression stop. Most dives resulted with VGE Spencer grade III or higher. Two divers were treated for limb DCS and four divers with high bubble load were given surface oxygen. This study confirms that high VGE grade correlates with the risk of DCS.

While the value of VGE monitoring for evaluation of decompression safety at the population level is not questionable, it does have clear limitations that are primarily reflected in great inter- and intra-individual variability. New technologies may help us to learn more about post-decompression bubbles dynamics and get closer to the personalized approach in prevention of DCS.


  1. Ballestra C., et al. Can sonic vibrations increase the number of decompression vascular gas emboli? P 12.
  2. Papadopoulou V., et al. Can current contrast mode echocardiography help estimate bubble opulation dynamics post-dive? P 18.
  3. Genser M., et al. Incdence of ost-dive bubbles and DCS usingthe US Navy Revison 6 ait tables. P 34


Outcomes of Decompression Illness

Recompression treatment and hyperbaric oxygen (HBOT) are standard treatment for decompression illness. While it is generally accepted that sooner recompression is associated with better outcomes, the urgency of treatment may not be same for all cases. Looking for practical guidelines we regularly consult published case series. Three case series presented at EUBS 2017 may be used to illustrate problems with such approach.

In the first paper1, authors compare outcomes in 24 mild cases of DCS treated in an on-site facility with an average delay to treatment 7.8 hours, to outcomes of 29 mild cases treated at an off-site facility with an averaged delay of 42 hours. Cases treated on-site almost all resolved completely after a single recompression (US Navy Table 6 or US Navy Table 5) and only one case needed an additional treatment. Of 29 cases treated off-site, only 17 resolved after first recompression, eight resolved after additional 1 to 5 tailing treatments, and four were left with some residual symptoms after tailing treatments. The authors suggest that this data supports rapid on-site treatment for all cases. In my opinion, there are two issues with this data.


First, the two datasets may not be comparable. Diagnostic criteria and outcome evaluation methods were not explicitly reported and may have been different in two facilities. Cases treated off-site may have left the site before symptom onset and travel may have contributed to the DCS. Second, the sample size is small enough that apparent differences may have been effected by chance.

The second reported case series2 includes 31 divers treated for DCS or AGE. Patients were thoroughly evaluated after the treatment and two to three months later. The evaluation included explicit inquiry about 20 separate symptoms and overall quality of life (VAS scale 1 – 100). At discharge and follow up 45% and 46% of patients respectively were free of symptoms. The most frequent persisting symptoms were tiredness, tingling, difficulty concentrating, and ear ringing.

The third presented series included 12 cases of inner ear DCS3. All patient were harvester divers and most dives included some omitted decompression time. The average time to treatment was 11 hours (5 – 72). Some cases received an on-site in-water recompression but still needed repeated HBOT. Patients received between 1 and 25 HBOT sessions. Eight cases recovered completely, three had residual symptoms and one was lost to follow up. Interestingly, the outcome was not apparently affected by the time to treatment.

These three case series with different case mixes could not be compared directly to each other, but they are a good reminder that the ability to generalize findings from small case series is questionable, particularly when it comes to the question of how the delay to treatment affects the outcome. This research also supports the call for a more standard and thorough description of case series, so we can better compare case studies, and evaluate the data.


  1. Wang Z. Efficacy of early treatment of decompression sickness in an on-site facility vs. delayed treatment of decompression sickness in an off-site facility. P 49.
  2. Johnsson J, et al. Recompression treated decompression illness signs and symptoms – initial findings and 2-3 months follow-up. P 72.
  3. Calderon J. et all. 12 cases of vestibular decompression sickness with clinical monitoring and recording of video. Hospital Ancud, Chile.

New Decompression Model Based on Occurrence of Gas Bubbles in Small Arteries

Decompression sickness is caused by gas bubbles that form in the body during and after decompression. The current thought is that gas bubbles originate on the venous side and pass to the arterial side either through intra-cardiac (PFO) or intra-pulmonary shunt (arteriovenous anastomoses). A group of scientists proposed recently a third mechanisms: the evolution of bubbles in the distal arteries, independent of venous gas bubbles.(1) They presented their work at the EUBS 2017 meeting (2) in Ravenna.

They base their theoretical work on previous experimental studies which identified so Blood cells backgroundcalled “active hydrophobic spots” (AHS) on inner surface of blood vessels.(3) Atomic force microscopy showed that on these spots tiny formations of gas bubbles, between 5 and 30 nm in diameter, were forming spontaneously. These formations are thousands of times smaller than venous gas bubbles, and if released into circulation they would probably be crushed immediately. However, in some specific conditions they could grow and reach the size of viable gas bubbles. This may happen in two stages.

The first stage of this condition occurs during decompression, when nano bubbles increase to the size of micro bubbles and are released into circulation. In the following second stage, bubbles grow due to simple diffusion of gas from the blood.  Favorable conditions for this may occur in small arteries with thin walls, through which inert gas from surrounding tissues may diffuse into the bloodstream. The smaller the artery, the more gas diffuses into it. These smaller arteries also have significantly slower blood flow, which enables an increase in the partial pressure of the inert gases. In normal conditions this contributes to a 1% increase of inert gas partial pressure in small cerebral arteries which does not cause much trouble. However, if the circulation slows down another 10%, the partial pressure increases by 44%, and this increase can contribute to a significant growth of microbubbles, occlusion of terminal arteries, and damage of the tissues manifesting as decompression sickness.

In the view of the authors, this proposed mechanism could explain some of the characteristics of DCS, like predominance of spinal cord DCS, effects of repetitive diving, variability of individual sensitivity to DCS, effects of aging and acclimation.

This is an interesting approach and we can expect significant results in this veign of research in the future. For now, those interested in decompression modelling should read the original papers, which are available online for free.


  1. Arieli R, Marmur A. A biophysical vascular bubble model for devising decompression procedures. Physiol Rep, 5 (6), 2017, e13191, doi: 10.14814/phy2.13191
  2. Arieli R. A new model of arterial decompression bubble development and spinal DCI. Abstract and Conference Book, EUBS 43 Annual Scientific Meeting, Ravenna (Italy), 12-18 September 2017, p 30.
  3. Arieli R. Nanobubbles Form at Active Hydrophobic Spots on the Luminal Aspect of Blood Vessels: Consequences for Decompression Illness in Diving and Possible Implications for Autoimmune Disease—An Overview. Front. Physiol. 8:591. doi: 10.3389/fphys.2017.00591

Does Diving Damage the Brain?

It is well known that compressed gas diving may result in acute decompression sickness and cause permanent injury to the brain and spinal cord. However, the risk of possible injury to the brain in the absence of acute decompression illness is less clear. Because of the controversy over the subject, and the lack of definitive evidence, DAN recently enlisted the help of a group of industry respected experts to provide their insight into the subject and published the results in Alert Diver (1).

The agents of neurologic decompression injuries are gas bubbles (emboli) that occur in tissue, travel with venous blood and may pass from venous circulation into the arterial system. Detectable venous gas emboli are often present after a dive, but they are usually removed through pulmonary capillary filtration. When the emboli pass to the arterial side, they may block arterial flow, causing tissue hypoxia in watershed areas and sometimes damage. The risk of arterialization increases in divers with a large PFO, but it can also occur through pulmonary arteriovenous shunts when there is high load of VGE. For decades this has been raising concern that brain injuries in divers may be more prevalent than previously thought and could potentially occur without a manifestation of acute decompression illness.

bubblesA recent paper published by our colleagues Balestra and Germonpre (2) seems to provide a quite clear answer to the question. The two researchers recruited 200 recreational divers who had never had DCS, and then randomly selected from among them 50 divers for further studies. In addition, they maintained a control group of subjects who had never been diving, and another control group of subjects who had been exposed to neurotoxic solvents. The aim of the study was to establish whether divers have more asymptomatic brain injuries than non-divers, review how divers perform on psychometric tests in comparison to non-divers, and research the possible effect of the presence of a PFO.

Balestra and Germonpre(2) used magnetic resonance imaging (MRI) to evaluate subjects for signs of asymptomatic brain injuries (unidentified bright objects – UBOs), performed echocardiographic tests for PFOs, and gave the subjects a battery of four neuro-psychometric tests. Divers who did not complete all studies were excluded, but 42 of the initial 50 remained in the study.

A significant PFO was detected in 38% of divers. UBOs were detected in 5 (12%) divers. Importantly, there was no correlation between the presence of a PFO and the ending or extent of UBO’s. That is the good news: diving without acute decompression illness does not cause UBOs, which were of concern to many divers and researchers.

Neuro-psychometric testing, however, produced inferior results for divers in two tests in comparison to non-divers, and similar results in comparison to the group exposed to neurotoxic solvents. On two other tests, divers did significantly better than the solvent group. This was not correlated with the presence of PFO. In summary, it appears that divers with five or more years of experience and at least 200 dives, have decreased short term memory and visual-motor performance, which could be a bad news if further studies confirm it.

Another interesting point from this study is that the prevalence of PFOs among study subjects was higher than in general population. The authors hypothesize that this may be due to strenuous intra-thoracic pressure changing activities, such as those encountered in diving, which may “open-up” previously sealed or microscopically small PFO. However, there are many other everyday life situations that raise intrathoracic pressure in similar manner as some dive maneuvers. In our opinion, this finding is of concern when discussing the prevalence of PFO in DCS case series. Even divers without a history of DCS may have greater prevalence of PFO than the general population.

This paper is worth reading and is available for free online at:http://journal.frontiersin.org/article/10.3389/fpsyg.2016.00696/full

  1. Willey J. Effects of diving on brain. Alertdiveronline. http://www.alertdiver.com/Brain
  2. Balestra C and Germonpré P (2016) Correlation between Patent Foramen Ovale, Cerebral “Lesions” and Neuropsychometric Testing in Experienced Sports Divers: Does Diving Damage the Brain? Front. Psychol. 7:696. doi: 10.3389/fpsyg.2016.00696

Skin Mottling after Diving May Be Result of Brain Lesions Caused by Gas Bubbles

Cutaneous decompression sickness (DCS), or “skin bends,” most often manifests as skin mottling on the torso, upper arms and buttocks to various degrees. An associated marbled look to the skin is sometimes referred to as cutis marmorata. While cutaneous DCS is most likely related to gas occurring in body — after decompression or due to lung barotrauma or some medical procedures — there generally is no accepted explanation how the free gas is related to skin changes.

Possible explanations include the occurrence of gas bubbles in subcutaneous tissues, occlusion of subcutaneous arteries with circulating bubbles bypassing the lung filter (as with a patent foramen ovale), inflammatory reaction bubbles present locally or bubbles causing endothelial injury at remote locations.


Bubble Production in Divers Who Have Had DCS

Venous gas embolism (VGE), or bubbles, in divers postdive indicates that their decompression was too fast, their bodies became supersaturated and free gas emerged from solution in tissues. The occurrence of free gas is considered a necessary condition for decompression sickness (DCS), which can happen even without VGE. However, the presence of VGE increases the number and types of possible harms to the body and thus the probability of DCS.

A number of studies indicate variability in proneness to DCS among divers; however, the question of whether divers who have suffered DCS produce bubbles more readily in general has not been answered yet. To answer this question, researchers would need to identify “bubblers” and “nonbubblers” and observe the outcomes of their dives over some period of time, which would require a lot of resources and time.