Diving Deeper into SCUBA... Science E-Book

Costantino Balestra

• Herausgeber: Acrodacrolivres
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 347

Veröffentlichungsjahr: 2017

Beschreibung

You will find in this book some valuable and reliable lessons about safe diving

The editors of and authors of this book are a cadre of scientists and physicians with broad experience and knowledge of diving physiology and decompression theory. As is often the case, it requires a group effort to succeed in advancing practical knowledge. The colloquialism "the whole is greater than the sum of its parts" is often true and the PHYPODE Reasearch Group epitomizes this concept. By logically grouping the various elements of diving science and medicine with provocative "food for thought" sections, the text offers valuable lessons to those interested in the current state of diving. Despite nearly 170 years of reasearch, the fundamenal nature of decompression stress remains elusive. As is well outlined in this book, great advances have been made to the practical elements allowing for safe diving. Nonetheless, there are glaring voids of knowledge related to the nature of bubble nucleation, its consequences and methods to ameliorate risk. The synergy exhibited in this text not only provides a foundation for what is known, it offers a glimpse of where research is taking us. - Professor Stephen R. Thom, Dept. of Emergency Medicine, University of Maryland School of Medicine

This is a book for all diving fans who want to discover their passion through a scientific approach.

EXCERPT

Decompression illnesses (DCI), or as they are called more scientifically: dysbaric disorders, represent a complex spectrum of pathophysiological conditions with a wide variety of signs and symptoms related to dissolved gas and its subsequent phase change.1, 2 Any significant organic or functional dysfunction in individuals who have recently been exposed to a reduction in environmental pressure (i.e., decompression) must be considered as possibly being caused by DCI until proven otherwise. However, apart from the more obvious acute manifestations of a single, sudden decompression, individuals who have experienced repetitive exposures (e.g. commercial or professional divers and active recreational divers) may also develop sub-acute or chronic manifestations, even if subtle and almost symptomless.

ABOUT THE AUTHORS

Dr. Costantino Balestra started to study neurophysiology of fatigue then started studies on environmental physiology issues. He teaches physiology, biostatistics, research methodology, as well as other subjects. He Is the Director of the Integrative Physiology Laboratory and a full time professor at the Haute Ecole Bruxelles-Brabant (Brussels). He is VP of DAN Europe for research and education, Immediate past President of the European Underwater and Baromedical Society.

Peter Germonpré is the Medical Director of the Centre for Hyperbaric Oxygen Therapy of the Military Hospital Brussels, Belgium).

Leseprobe

Introduction

Decompression illnesses (DCI), or as they are called more scientifically: dysbaric disorders, represent a complex spectrum of pathophysiological conditions with a wide variety of signs and symptoms related to dissolved gas and its subsequent phase change.1, 2 Any significant organic or functional dysfunction in individuals who have recently been exposed to a reduction in environmental pressure (i.e., decompression) must be considered as possibly being caused by DCI until proven otherwise. However, apart from the more obvious acute manifestations of a single, sudden decompression, individuals who have experienced repetitive exposures (e.g. commercial or professional divers and active recreational divers) may also develop sub-acute or chronic manifestations, even if subtle and almost symptomless.3

It is generally accepted that there exist sub-clinical forms of decompression sickness (DCS), with little or no reported symptoms, and that these may cause changes in the bones, the central nervous system and the lungs. When studying the physiology of decompression, the presence of symptoms (or not) may not be the most sensitive or reliable marker. In recent years, analysing “decompression stress” has taken more and more importance in the research of understanding decompression. Current research into DCS is focused on biological markers that can be detected in the blood.

Investigators are exploring the potential association between decompression stress and the presence of membrane microparticles (vesicles shed from a variety of cell types) in the blood.4–6 Microparticle levels increase in association with many physiological disease states as well as with the shearing stress caused by bubbles in the blood. The working hypothesis is that certain microparticles (possibly induced by inert gas bubbles) may initiate, be a marker of or contribute to the inflammatory response that leads to DCS. This investigation goes beyond the pure bubble model. While bubbles in the blood certainly play a key role in the development of DCS, their presence or absence does not reliably predict DCS symptom onset. Investigating this process at the molecular level may teach us a great deal more about DCS, providing insights that we hope will improve the effectiveness of both prevention and treatment.

Modern approaches to evaluating decompression stress have considered a wide range of other markers and influencing factors: physiological changes during and after the dive (reduction of flow mediated dilatation, dehydration, changes in blood pressure), physiological factors of personal susceptibility (age, sex, VO2max), environmental factors (temperature, altitude), as well as bubble counts. All this shows how much today’s approach to decompression is far removed from “traditional” concepts of saturation and desaturation.

In 2009, a European Commission project was initiated, providing the opportunity for education and tutoring of a number of young, inexperienced researchers in the field of decompression research. Baptised the “PHYPODE” project (Physiology of Decompression) these ESR (Early Stage Researchers) and ER (Experienced Researchers) had the opportunity to gain formal training and experience in various leading research institutions all over Europe. Nearing the end stages of the project, the 14 researchers who have been working for three years in PHYPODE have summarized current concepts and ideas, as well as some results of their cutting-edge research projects into a book, this book “The Science of Diving”. It is not only written in such a way that it should allow divers to learn more about the modern approaches to understanding decompression and its problems, but also, contribute to expanding the diving decompression knowledge of physiologists, medical personnel and basically anyone with an interest in “the heart of the matter”. Almost every young scientist participating in the PHYPODE project had the responsibility of writing a chapter. This was by no means a simple job, considering the different linguistic origins of this group of young researchers, many of whom had their own doctoral theses or research programmes to complete in parallel. Authors include renowned and established scientists and diving medicine specialists from the tutoring PHYPODE partner centres.

The Internet contains already a huge amount of information available on the Internet on such a topic. Why then, is this book necessary?

Let us illustrate our motivation by means of a story from Japan, where, quite recently, one major cosmetics company received a customer complaint because he received an empty soapbox. The company launched a huge investigation into the matter and discovered that the defect arose in the packaging department. They decided to develop a robust and reliable system ensuring zero defects in the process of product packaging and the company invested heavily in the design and implementation of a solution. A few weeks later, a similar problem occurred in a small soap-manufacturing company in India. This time the approach was very different: the manufacturer bought a big industrial fan and placed it facing the soapbox chain. Any boxes that were empty simply blew off the chain and the rest moved ahead to the storage house!

Our aim was to keep the concepts as clear as possible but maintain the scientific integrity of the subject. References are limited and proposed as further reading. Having as authors many of those conceiving some of the new approaches provided the opportunity of being the “fan that blows empty boxes”.

While the first goal of this book is to provide valuable insight and new ideas about diving physiology and medicine, there is a more direct way in which you, who bought this book, will contribute to the advancement of diving medicine: all royalties from the sales of this work will be donated to the European Underwater and Baromedical Society (EUBS – http://www.eubs.org), the European scientific society for diving and hyperbaric medicine.

The Editors

Prof. Costantino BALESTRA, PhD     Dr. Peter GERMONPRE, MD

The Co-editors

Miroslav Rozloznik, PhD     Peter Buzzacott, PhD     Dennis Madden, MSc

All authors (see author list at end of this book)

References

1 Papadopoulou V, Eckersley RJ, Balestra C, Karapantsios TD, Tang MX. A critical review of physiological bubble formation in hyperbaric decompression. Adv Colloid Interface Sci. 2013;191-192:22-30.

2 Papadopoulou V, Tang MX, Balestra C, Eckersley RJ, Karapantsios TD. Circulatory bubble dynamics: from physical to biological aspects. Adv Colloid Interface Sci. 2014;206:239-49.

3 Hemelryck W, Germonpré P, Papadopoulou V, Rozloznik M, Balestra C. Long term effects of recreational SCUBA diving on higher cognitive function. Scand J Med Sci Spor. 2013

4 Madden D, Thom SR, Yang M, Bhopale VM, Ljubkovic M, Dujic Z. High intensity cycling before SCUBA diving reduces post-decompression microparticle production and neutrophil activation. Eur J Appl Physiol. 2014

5 Culic VC, Van Craenenbroeck E, Muzinic NR, Ljubkovic M, Marinovic J, Conraads V, et al. Effects of scuba diving on vascular repair mechanisms. Undersea Hyperb Med. 2014;41:97-104.

6 Zanchi J, Ljubkovic M, Denoble PJ, Dujic Z, Ranapurwala S, Pollock NW. Influence of repeated daily diving on decompression stress. Int J Sports Med. 2014;35:465-8.

Tables

Table 1: No-decompression time in minutes for a given depth and given computer set to the standard settings (air, sea level, fresh water)

Table 2: Common levels of technical diving

Table 3: Major differences between recreational and commercial diving

Table 4: Pneumogauge corrections (originally published by United States Dept. of Navy in 1985)

Table 5: Decompression schedules for 30 minutes at 54 meters on air and heliox using different procedures

Table 6: Early Comex heliox saturation decompression procedures

Table 7: The average daily ascent rate during final decompression from saturation

Table 8: Bubble grading with the Spencer scale (first established in 1974)

Table 9: Kisman-Masurel code (originally published in 1983)

Table 10: Brubakk & Eftedal scale (2007)

Table 11: 2D Echocardiographic and PW Doppler Grade (Boussuges 1998)

Table 12: Physical properties and narcotic potency of different gases

Table 13: Clinical signs observed in man during nitrogen narcosis

Table 14: Seven cases with two data variables

Table 15: Squared Euclidian distances between observations inTable 14

Table 16: Divers Alert Network DCI clustering results

Figures

Figure 1: Types of dive computer used in the 39944 dives recorded in the DAN Europe DSL Database

Figure 2: Gradient Factors for the 12.5 minutes compartment with respect to M-Values, as reached by DSL divers during 14000 analysed dives (graph courtesy of Corrado Bonuccelli).

Figure 3: Exposure Factor (EF) distribution over 14000 analysed dives (graph courtesy of Corrado Bonuccelli)

Figure 4: Depth/time distribution of 39944 dives recorded in the DSL database

Figure 5: Declared health problems in divers from the DAN Europe DSL database

Figure 6: Type of algorithm (bubble or compartimental) used in dive computers over 10738 recorded dives and 165 DCS cases related to the use of diving computers.

Figure 7: Box-plot of calculated tissue supersaturation levels on 165 cases of DCS recorded over 10738 DSL dives.

Figure 8: Distribution of VGE Doppler recordings in the DAN DSL database.

Figure 9: Median bubble score post-dive with respect to hydration level

Figure 10: Physiological variations after a dive (n=90).

Figure 11: Surface tension of some body fluids at body temperature

Figure 12: Technical divers with underwater scooters and semi-closed re-breathers (Photo courtesy of Miroslav Potas)

Figure 13: Cavern; a starting point for cave diving, usually defined as a zone of a cave with direct visual contact and access to water surface illuminated with natural daylight.

Figure 14: Probability of death in case of equipment failure in series or parallel (adapted from Piantadosi, 2003; see related sources)

Figure 15: The supervisor carrying out a checklist on the diver.

Figure 16: Portable gas control panel

Figure 17: The umbilical consists at least of 3 hoses; the yellow hose (brightest in the picture) is the pneumogauge hose.

Figure 18: Pneumogauge schematic (source: US Navy Diving Manual rev 6)

Figure 19: The control panel of an offshore commercial operation

Figure 20: Typical saturation dive profile.

Figure 21: A typical saturation chamber system located below deck in a North Sea diving support vessel.

Figure 22: The classic perfusion gas exchange model.

Figure 23: The critical volume assumption scenario applied to saturation

Figure 24: Relationship between ascent rates and risk of DCI, for a given chamber ppO2.

Figure 25: The various saturation decompression rates from the main diving contractors plotted according to depth.

Figure 26: Richard Pyle puncturing the bladder of a fish while ascending from the depth.

Figure 27: Super-saturation of fast and slow compartments for a shallow and deep stop profile.

Figure 28: “Dive phone” iPhone App showing two different no decompression times (29 min. vs. 11min) (Divephone photo courtesy of Innovasub)

Figure 29: A horizontal diver during deco (photo courtesy of Peter Vaverka)

Figure 30: Without a stabilisation mechanism to counteract the microbubble’s surface tension, bubbles below a critical radius will tend to dissolve at atmospheric pressure

Figure 31: Four chamber view echocardiography of a diver at flight cruise altitude

Figure 32: The effect of breathing 100% oxygen on the oxygen window

Figure 33: Structure of the human artery and interactions between endothelial cells and smooth muscle cells.

Figure 34: FMD assessment by the measurement of brachial artery diameter with high-resolution ultrasound

Figure 35: Post-occlusive Reactive Hyperaemia (PORH) measurement technique.

Figure 36: Effect of SCUBA diving on circulating NO and FMD.

Figure 37: Apical four-chamber view: circulating bubbles in the right cavities of the heart

Figure 38: Apical four-chamber view: circulating bubbles in the right cardiac cavities with passage in the left cavities.

Figure 39: Frame-based bubble counting

Figure 40: Parasternal short axis view: study of pulmonary blood flow by Pulsed-Wave Doppler.

Figure 41: Normal pulmonary blood flow as shown by PW Doppler

Figure 42: Circulating bubbles detected in the pulmonary blood flow: the arrows indicate bubble signals in each cycle overriding the pulmonary blood flow

Figure 43: Possible positioning of the transducer for TTE: apical view (A) and parasternal view (B); transoesophageal echocardiography (C)

Figure 44: Proposed mechanisms underlying the protective effect of exercise on vascular gas bubble formation

Figure 45: Beneficial effect of large-volume oral hydration on vascular gas production monitored by Doppler after a dive

Figure 46: Speed of absorption of marked proteins (angular variation) in the axillary region after subcutaneous injection

Figure 47: Dynamic changes in HSP concentration after a single sauna exposure.

Figure 48: Effect of a pre-dive sauna exposure on post-dive endothelial dysfunction (measured by FMD) and bubble count.

Figure 49: Effect of pre-dive whole body vibration on post-dive endothelial dysfunction (measured by FMD) and bubble count.

Figure 50: Accelerated peripheral elimination of radioactive tracer after vibration in axillary lymph nodes.

Figure 51: Effect of dark chocolate on post-dive endothelial dysfunction and bubble count.

Figure 52: A proposed molecular mechanism of chocolate contribution on reducing endothelial dysfunction associated with breath-hold and SCUBA diving.

Figure 53: Effectiveness of different means of preconditioning on endothelial function preservation as measured by FMD and post-dive VGE production

Figure 54: Schematic overview of the circulatory system

Figure 55: Circulatory system before birth.

Figure 56: Circulatory system after birth.

Figure 57: Correlation and linear regression of the magnitude of CFFF change and time to complete in both Math Processing task and Trail-Making Task under normobaric oxygen (A), Hyperbaric Air (B) and Hyperbaric EANx (C)

Figure 58: Waterproof CFFF device specially developed for research purposes (Human Breathing Technology, Italy)

Figure 59: The Math Processing Task, one of the tests from the Psychology Experiment Building Language battery (PEBL) in a diving computer (Courtesy of Mares S.p.A.).

Figure 60: Percentage variation of CFFF during and after a 25 min air dive to 30 meters/110 feet.

Figure 61: NIRS measurements of HbO2 and HHb during and after a 25 min air or EANx dry chamber dive to 30 meters.

Figure 62: CFFF and computer-based behavioral testing measurements during and after a 25 min air or EANx dive to 30 meters/110 feet.

Figure 63: Clustering steps

Chapter 1: Recreational diving today: decompression habits, DAN Europe database insights

Authors:Costantino Balestra, Danilo Cialoni, Peter Buzzacott, Walter Hemelryck, Virginie Papadopoulou, Massimo Pieri and Alessandro Marroni

Take-home messages

SCUBA diving is a relatively safe activity

Recreational dives are routinely carried out to approximately 80% of the M supersaturation value

Computers are all similar in DCS incidence in theory but validation is difficult for typical recreational multilevel repetitive and multiday profiles

Databases are useful to collect supplemental data from diving, because dive profile analysis alone is not sufficient to accurately predict DCS risk.

Abstract

Compared with other sports, SCUBA diving remains a relatively safe activity but precisely defining risk is important. Diving databases such as the Diving Safety Laboratory (DSL) collection by Divers Alert Network (DAN) Europe can provide new insights into the causes of diving accidents, including decompression sickness (DCS) incidence with respect to the dive profile. Data from the DSL shows that in the recreational setting diving with a dive computer may be used by as many as 95% of divers. This points to the need of validating these tools with respect to DCS incidence, a difficult task.

The most widely used computers/algorithms in Europe are nowadays, irrespective of brand, the Bühlmann ZHL and the Wienke RGBM based ones, with a roughly 50/50 distribution of each within the DAN Europe DSL diver population. Analysis of the DSL database shows that the vast majority of all recorded DCS cases occurred without any violation of the respective algorithms, in other words, with compartment inert gas pressures well below the maxima allowed..

In addition, the DSL database and field research also show that many other physiological variables may be involved in the pathogenesis of DCS, even within computed “safe” limits.

The current dive computer validation procedures, although important and most useful as a first benchmark, still allow for a probability of DCS beyond ideal acceptability in a recreational setting. A more aggressive “physiological” approach to testing and validation of decompression algorithms should be implemented, as the recreational diving population nowadays is far from the fit 18-22year old military diver which constitutes most of the validation dataset from the US Navy. Such an approach needs to be able to identify and control the most significant physiological variables involved in the pathogenesis of DCS together with the inert gas supersaturation values, and relate both to the decompression algorithms.

1. Development of recreational dive limits

Recreational dives are dives limited in depth and time such that the diver may ascend to the surface at any time with an acceptably low probability of suffering decompression sickness (DCS). Diving beyond these limits requires the diver to stop en route to the surface to decompress. Hence, they are known as “no-stop” limits. A variety of decompression models are available for recreational divers to predict their no-stop limits (for details see Chapter 4Decompression theory).

The first experimental attempt to reduce the incidence of DCS was conducted for the Royal Navy by physiologist J.S. Haldane using a goat model and a set of diving depth/time tables were drafted by extrapolating the animal-derived results to suit a human circulatory system. The original tables were validated during seven dry hyperbaric chamber “dives” and 19 man-dives in deep open water using teams of attendants operating two surface supply pumps. They were approved for use by the Navy, published in 1908, and have formed the basis of diving decompression since.

Haldane’s tables were based on a gas-content model, whereby five theoretical compartments of varying (parallel) blood perfusion and inert gas solubility were each defined by the time it would take that compartment to half-fill with nitrogen, assuming exponential gas uptake and release. Each “half-time” compartment was then ascribed a maximum limit of tolerable “supersaturation” or over-pressurisation before the gas cannot be carried solely in a dissolved form anymore and starts forming bubbles (free phase gas). Initially, Haldane assigned a single maximum ratio (2:1) for all five theoretical tissue compartments, but in subsequent years, individual ratios have been adapted empirically by among others the US Navy, allowing larger supersaturation ratios for “fast” compartments, and limiting to lower supersaturation ratios for slow compartments, before making the allowed supersaturation ratio depth-dependent: this is the concept of the M-value line which is nothing more than the intercept and slope of that pressure(depth)-dependent allowed supersaturation before theoretical bubble formation.

These “neo-haldanian” models work well for short, shallow dives and ascents to altitude. Other models were subsequently developed including some based on explicit modelling of bubble behaviour (instead of compartment pressures) and an overview of all of these can be found in Chapter 4Decompression theory.

In a review article “Gas-content versus bubble decompression models” David Doolette commented that by 2005: “…all diver-carried electronic decompression computers (dive computers) use a real-time gas-content model.”.

Today some newer dive computers with branding implying the incorporation of a bubble model appear nonetheless to use a gas-content model and bubble models are still generally limited to home-computer/laptop software used primarily by technical divers. As recently as 2007, bubble models were still to be formally validated with human trials.

Modern commercially available dive computers for the recreational and more avid sports diver contain often a combination of different models (although in most cases, the exact algorithm used is not made public by the manufacturer). This results in a wide variability in allowable “no-stop” times at different depths (Table 1).

Which one of these computers is “correct” can never be determined, which may present a practical problem when a group of divers is diving with different computers. However, because there is a psychological barrier to surfacing when the computer is not yet “clean”, in practice, the most conservative profile is followed.

Table 1: No-decompression time in minutes for a given depth and given computer set to the standard settings (air, sea level, fresh water).

2. Dive computer validation

This section is based on the 2011 “Validation of Dive Computers Workshop” Proceedings (see reference list) and in particular the contribution entitled “Dive Computers: The Need for Validation and Standards” by Arne Sieber, Milena Stoianova, Ewald Jöbstl, Elaine Azzopardi, Martin D.J. Sayer and Matthias F. Wagner.

2.1. The problem

Dive computers have been used extensively in recreational diving for the last 25 years with a low incidence of DCS. It could therefore be argued that their use was “successful” in some respect. However there have been reported cases of DCS, even neurological ones, where recreational divers followed their dive computer on no-decompression dives. The most recent DAN Europe number suggests that around 80% of neurological DCS cases did not violate their diving computer recommendation.

Not many divers realize that at the moment there is no uniform procedure for testing and validating dive computers. They are not even listed under the European Union directive for personal protective equipment (PPE Directive 89/686/EEC). The norm usually applied during the CE certification of dive computers is the EN13319 which addresses only accuracy and precision of the depth sensor and timer. At the moment no dive computer manufacturer provides any details as to the models they use or the implementation of those models and none have ever performed any substantial human validation.

2.2. Defining what we mean by validation

To develop a validation procedure and guidelines one must first clearly define what the purpose of a dive computer is. In addition to acting like a timer and depth/pressure sensor in real time, divers rely on dive computers for their decompression calculation. That is, to calculate remaining no decompression time at current depth and, in the particular case of either dedicated dive computers or emergency decompression displays, decompression stops during decompression diving. The hidden assumption behind the “trust” each diver assigns to those decompression calculations is that by following the dive computer decompression stops or remaining no decompression time, the diver’s probability of developing decompression sickness (pDCS) is acceptable to him. We therefore all acknowledge in SCUBA diving, quite explicitly when we accept warnings in dive computer manuals or when signing liability release forms to go diving, that pDCS is never zero and that we aim to keep it below a threshold that is personally acceptable to us. However there is no dive computer that will display a pDCS for a specific dive plan in dive plan mode for instance, so the user has to implicitly trust that the recommendation he/she sees on their dive computer display in the form of a no decompression limit or decompression stop has been somehow validated as acceptable for the same type of diving.

The first step in devising validation procedures is therefore to define the “range of applicability” of the dive computer, which will obviously differ tremendously from commercial or military diving in freezing waters at night to recreational no decompression diving in warm waters with good visibility, for example. By clearly defining the range in which a dive computer will be used, precise validation procedures detailing operational needs (display readability in low visibility conditions, temperature sensing and operation, air integration, etc.) and decompression calculations (depth limit with nitrox, or Trimix, etc.) can be outlined.

2.3. Operational considerations

Operational considerations for the dive computer also form part of the validation process as they need to dictate whether the tool (dive computer) is adequate for use in the predetermined context safely. These include ease of operation of the dive computer, readability of the display in the worst visibility conditions encompassed in the range of expected diving, clarity and unambiguity of the information displayed, obvious failure mode, battery life and ease of displaying and downloading profile data after each dive.

2.4. Decompression calculations

A method for validating the decompression calculations produced by the dive computer needs to be developed. In this respect, two strategies can be envisaged, depending on whether the dive computer manufacturer clearly states which published and publicly disclosed diving algorithm he is implementing in his dive computer, or not.

In the first case, where the model is published, the validity of the model is not to be proved by the manufacturer which therefore only needs to show that his implementation of said model (in terms of hardware and software design) is faithful to the algorithm published. This would be the most straightforward case since the validity of the model, ie the probability of DCS (pDCS) that the predictions of this model give, are relegated to the developer of the model itself who needs to show how these are acceptable for a specific type of diving range.

In the second case, where the model is unknown, the predictions of the dive computer should be tested against profiles of known probability of DCS (using for instance the US Navy manned dives database) that would be typical for the expected use of the dive computer.

In both cases we come back to the important point of clearly defining the range of applicability of the dive computer to select adequate dives with known pDCS to compare. However it should be noted that since dive computers allow for a “real-time” calculation of the decompression limits, a perfect validation would require an infinite number of profiles to be tested as the combinations are endless. This is obviously impossible and even testing many different profiles is time-consuming and expensive. The other issue is the need to compare to known pDCS dives, which are usually square, table-like dives, not the typical recreational profile. Repetitive, multiday, altitude, gas switches and other common recreational practices are not taken into account using these dives.

Finally, we have been primarily focused on pDCS as the endpoint or comparison point between dive computers’ predictions and known outcomes. The obvious advantage in using pDCS is that we have data to compare to, especially from the US Navy dive computer validation which remains the most comprehensive database in this regard. However using pDCS is not without problems, one of which is whether to include marginal DCS events and those for which diagnosis was uncertain, not to mention that DCS symptoms have a wide range and clustering them altogether might hide different mechanisms at play. In addition, there is the ethical issue with testing pDCS on humans as this is basically inducing DCS in a small fraction of the test-subjects. Scientific ethical approval is difficult to obtain for this - the incidence of DCS among recreational divers is so low that exposing test subjects to a higher risk for validation purposes is considered non-ethical. Using the data already available to estimate pDCS from dives previously made poses a supplementary problem because these come mainly from military test subjects, ie young, male, fit, well-trained, healthy adults which may not be representative of the recreational diving population. This is why the detection of Venous Gas Emboli or VGE to validate diving algorithms has been proposed as an additional endpoint (in terms of reducing decompression stress). Even though it seems intuitively obvious that the more VGE present during the decompression after the dive, the higher the risk for DCS, the presence of VGE does not seem to be a very accurate predictor of the risk of DCS; however, the absence of VGE does positively correlate with a very low to non-existent risk of clinical manifestations of DCS.

2.5. Proposed lifecycle for dive computers (development, testing and validation)

The Validation of Dive Computers Workshop (2011) proposed the following sequence be adopted for validating all dive computers:

Overall scope definition: specify the principal functions of the dive computer, i.e. parameters to be displayed including display requirements, mechanical design, performance and operational range

Hazard and risk analysis: description of potential risks by fault of diver (exceeding depth limit or no deco time, etc.) or dive computer malfunction (hardware, software, etc.), including an estimate of severity and probability of risk/hazard occurring.

Safety requirements allocation: to limit probability of occurrence and consequences in cases of occurrence of the listed hazards and risks above (clear step with strategies to minimize problems, e.g. clear failure mode display in case of malfunction, etc.)

Design and implementation phase: designing the hardware and software for the dive computer and establishing verification and validation plans

Validation phase: check final product against complete list of requirements, including safety-specific requirements

3. Insights from DAN Europe database

“Validation of decompression safety is complicated and expensive. Thus, in most cases manufacturers do not have the data necessary to support claims of risk control or risk reduction — an important issue for divers.”Petar Denoble (Denoble, 2010)

Recreational Diving today is mostly done with the use of dive computers, which divers tend to trust with absolute “faith”. Not many divers realize that the validation protocols underlying the marketing of such computers and the algorithms they use are far from perfect and that even the most reliable computers still accept a probability of DCS ranging from 2 to 5%, with a probability of neurological DCS in the range of 0.2 - 0.5%. We believe the typical recreational diver is generally unaware of this fact and tends to believe that their dive computer is simply infallible and that nothing will happen to them if they follow the indications given them. Those who actively work in this medical and technical field know that this is not the case and that DCS remains a possibility, although rare.

The DAN Europe Diving Safety Laboratory (DSL) database is a comprehensive epidemiological database aimed at recording information about divers and dives with the scope to increase the safety of divers. Information on anthropometric data, breathing gas used, equipment malfunctions and medical history is recorded using a specific questionnaire. In addition, certain dive profiles are completed with the downloaded profile and even, in the case of dives recorded during “DSL Field Research Trips” precordial Doppler VGE assessment.

On a sample of 39944 dives, mostly dived according to Bühlmann ZHL or Wienke RGBM algorithms, 181 DCS cases were recorded. Figure 1 shows the proportion of dive planning methods employed by the divers.

Figure 1: Types of dive computer used in the 39944 dives recorded in the DAN Europe DSL Database.

The 9% figure refers to divers who either used their computer in “gauge” mode, or referred to decompression calculation software or Dive Tables.

3.1. Does Math rule?

Gradient Factors (GF), quoted in percentages, are simply an added safety factor which redefines the M-Value line gradient to be more conservative, in other words they set the limit of tolerated over-pressurization for the compartment lower by a certain percentage than the original model.

If we focus for a moment on the computed Gradient Factor for the hypothetical 12.5 minutes half-time compartment, we can see that on 14000 (out of the recorded 39944) dives so analysed, 95% were well below 80% of the maximum allowed supersaturation, with only a minor portion getting close to this value (see Figure 2).

Figure 2: Gradient Factors for the 12.5 minutes compartment with respect to M-Values, as reached by DSL divers during 14000 analysed dives (graph courtesy of Corrado Bonuccelli).

As a guide for comparative dive safety, Hempleman’s formula can be used to calculate an “Exposure Factor” (EF):

Where the dive depth (D) is in ATA (absolute pressure) and time (t) is in minutes. So, for example, looking at the DSAT Recreational Dive Planner and staying within the no-stop limits a dive to 2.8 ATA (18m) for 55 minutes gives an EF of 21, 25m for 29 minutes gives an EF of 19 and 35m for 13 minutes gives an EF of 16. Here is already a discrepancy: it looks like the deeper you dive the lower the EF, i.e. the safer is the dive. In addition, the Exposure Factor does not account for repetitive diving.

Although an outdated measure, discussion Hempleman’s Exposure Factor serves the illustrative role of showing the limits of any calculation model: it is only as good as its calibration dataset, and using it to extrapolate outside that range often fails dramatically. As a very rough index, an EF of 20 is considered an acceptable personal safety limit, scores up to 25 would be approaching the limits of what is considered safe and scores higher than 25 are to be considered exceptional exposures.

When calculating the Exposure Factor of the above 14000 dives we observed that about 60% of the dives were within the 20 EF mark, another 18% reached the 25 mark, and 22% of dives showed higher exposure factors (Figure 3). Yet, only 181 cases of DCS resulted. This highlights how safe a conservative EF limit of 20 or 25 could be for non-repetitive recreational dives. However, one needs to remember that EF does not take into account any other factors than depth and time – ascent rate, deep stops, and repetitive dives are left out of the equation.

Figure 3: Exposure Factor (EF) distribution over 14000 analysed dives (graph courtesy of Corrado Bonuccelli)

3.2. DAN Europe Database Dive Profile Collection

Data gathering to draw useful conclusions aimed at fostering diving safety is a must nowadays, and especially with the technology available should be attempted as much as possible. “In the field” SCUBA diving data collection however has only been marginally done by some commercial companies or military sections in recent years, but recreational diving data have been collected for several years by DAN, both in Europe and the USA. Soon, DAN Europe and DAN America will even combine their databases to create the world’s largest dataset of actual dive profiles.

The DAN Europe DSL database included, at the end of 2013, 2615 divers (2176 male and 439 female, mean age 42.54 ± 8.82 years) who completed 39944 dives (32890 performed by males and 7054 performed by females).

Anthropometric data were: mean height 174.5 +/- 8.21 centimetres (176.6 for men and 164.1 for women), weight 77.40 +/-12.73 kilograms (80.95 for men and 61.26 for women), BMI 25.34 +/-3.09 (25.91 for men and 22.75 for women). Precordial Doppler recordings were obtained for 5970 out of 39944 dives. Of the dives in the database, 91.30% were done breathing compressed air, 5.14% breathed nitrox, 0.48% trimix, while for 3.08% this information is missing. Depth/time distribution of the dives is shown in Figure 4.

As mentioned above, within this database 181 cases of DCS were recorded, giving an overall DCS prevalence of 0.45%. The “true” prevalence in the recreational diving population is likely (much) lower, as many of these DCS profiles were collected when divers presented for treatment; also, these dive profiles were often contributed by enthusiast, “hardened” frequent divers, who may not always accurately represent the occasional “vacation” no-stop diver. As would be expected, 91% of these dives were performed using a diving computer.

The incidence of other declared problems during the dive is also very low, with the majority of problem related to equalization (1.24%) buoyancy (0.50%), ascent speed (0.47%) and out of air situations (0.45%).

Figure 4: Depth/time distribution of 39944 dives recorded in the DSL database.

Most dives are recreational in nature (mean depth 28.03 ± 13.75m; min 5 m; max 192 m; mean dive time 46.02 ± 4.6 min). The curve represents the no-deco limits from the US Navy table.

The percentage of divers who declared having suffered from previous diseases or being aware of some medical conditions or anatomical peculiarities is shown on Figure 5.

Figure 5: Declared health problems in divers from the DAN Europe DSL database

3.3. Does the algorithm matter?

Original software was developed for the analysis of the Gradient Factor (GF) values for the dives collected in the DAN Europe database.

Initially, the DSL system was only compatible with compartmental model dive computers; therefore a direct comparison of the incidence of DCS between compartmental and so-called “bubble” models was not possible. Sometime after the start of the dive data collection program, DAN Europe was able to collect data from virtually all models of recreational dive computers on the market and a direct comparison of DCS incidence between compartmental and bubble models became possible (see Figure 1). It is worth remembering however that even “bubble” models (e.g. Wienke RGBM) are implemented mathematically as compartmental models on dive computers with correction factors to account for the full bubble model behaviour.

In 10738 dives where the full dive profile was available, dived with Bühlmann ZHL16 or Wienke RGBM algorithms, we recorded 165 DCS cases, almost equally distributed between the two types of model (1.35% vs 1.75%, Figure 6).

This incidence is higher than the overall incidence of DCS within the entire sample of dives collected but, this is probably an overestimation due to the data collection bias.

Figure 6: Type of algorithm (bubble or compartimental) used in dive computers over 10738 recorded dives and 165 DCS cases related to the use of diving computers.

The availability of full dive profiles allowed us to calculate inert gas supersaturation levels for different tissues at different time points (M values). It is interesting to observe that only 10% of these cases approached the maximum recommended inert gas supersaturation level according to the selected algorithm (90% or more of the M-values). Only another 10% of the recorded DCS cases occurred with supersaturation levels between 80 and 90% of the M-value. The remaining 80% of these DCS cases occurred with supersaturation levels lower than 80% of the maximum allowed by the specific algorithm, with a general average supersaturation level of 75% of the M-value (see Figure 7).

Figure 7: Box-plot of calculated tissue supersaturation levels on 165 cases of DCS recorded over 10738 DSL dives.

It shows a lower average supersaturation level than the maximum allowed according to the referral algorithm (minimum 0.41, first quartile 0.72, median 0.80, third quartile 0.84, maximum 1.10, average 0.75, SD 0.25). This suggests there is more to decompression algorithm validation than compartment supersaturation estimates based on depth-time profiles alone. Ascent rate may have been the deciding factor in many of these cases, in which case moment-to-moment relative supersaturation (relative to the ambient pressure) may come into play.

The DAN Europe DSL database also captures other parameters that may or may not have an importance in the occurrence of DCS. With time, these data may help improve decompression practice.

3.4. VGE measurements

“Validation of decompression safety may be sped up by using venous gas emboli (VGE) as the outcome of interest in place of DCS. Venous gas bubbles may be detected in divers without DCS symptoms. When bubbles become abundant (high VGE grade), the risk of DCS increases. Decompression trials that use VGE grade as the outcome do not provide the same insights as studies that include actual cases of DCS.”Petar Denoble (Denoble, 2010)

A total of 1181 Doppler measurements have been performed, with a further still 2100 waiting to be evaluated. For the dives where the complete dive profile was available, Gradient Factor calculations have been made and showed that a large part (95%) of the documented dives are below maximal saturation of medium half time tissues. This reflects in the Doppler scores, which show a low occurrence of High Bubble Grades (Figure 8).

Figure 8: Distribution of VGE Doppler recordings in the DAN DSL database.

3.5. (De)hydration

Analysis of recorded DCS incidents has led us to consider parameters other than just Doppler Scores and dive profile. We will take hydration state as an example. Recent data, measured in collaboration with technical divers as well as the French Navy, showed the importance of a normal hydration level. Each group of divers performed the same diving procedure, once in normo-hydrated condition and once being hyper-hydrated. The bubble count, measured with echography, showed a significant difference (see Figure 9).

Figure 9: Median bubble score post-dive with respect to hydration level

Other data measured on technical divers reported that in divers performing even very deep dives (up to 130 m) no haemoconcentration post-dive was shown (see the reduction of haematocrit in Figure 10). This was unusual since it is known that divers presenting to hyperbaric recompression centres for DCS treatment are usually haemoconcentrated.

A new model has to be proposed to understand such a discrepancy. We can hypothesize that during the dive a fluid shift occurs to compensate the immersion diuresis. This water would be like a “loan” to the vascular compartment.

A decrease of the extracellular water and a concomitant increase of the intracellular water can be measured by means of multi-frequency bio-impedancemetry immediately after the dive (Figure 10). One can imagine that the following sequence is involved: the progressive vascular compartment concentration, induced by immersion diuresis is compensated - during a limited time - by water shifting towards the vascular compartment from the extracellular compartment. This is advantageous during the off-gassing period since at that time the nitrogen gradient goes from the tissues to the vascular bed and the vascular dilution is useful to accept the increased nitrogen load (supersaturation).

However, this is a temporary situation and data are available to show that there is a reversal of this fluid shift at some later point after the dive. If this reversal of the fluid shift occurs when the supersaturation has not yet been sufficiently reduced by the pulmonary wash-out of nitrogen, bubbles might occur or be trapped in the “dehydrated” microcirculation, causing damage and clinical symptoms of DCS would occur. Perhaps this might explain why divers presenting with DCS are found to be haemoconcentrated.

Figure 10: Physiological variations after a dive (n=90).

On the contrary, a minor haemoconcentration occurring in particular circumstances may not always be deleterious. A French Navy group, considering plasma volume as a vector for nitrogen to reach the tissues, hypothesized that a minor reduction of the hydration state at the start of the dive might lead to a reduction of the saturation rate, and thus offer a degree of protection, not the opposite. This has been examined using sauna exposure before dive (see Chapter 7Preconditioning as a tool to improve diving safety).

Of course, sauna leads not only to minor dehydration but also to other interesting preconditioning effects such as increased cardiovascular activity (similar to mild physical effort), increased Heat Shock Protein production (HSP 70 and HSP 90), and increased Flow Mediated Dilation. More research continues into this paradox.

3.6. Surface Tension changes

Looking closer at the basic bubble dynamics, the classical equation describing Venous Gas Emboli radius change in supersaturated tissues shows:

where:

ℜ= The perfect gas constant