Smithsonian Institution Smithsonian Institution Sc holarly Pres s The FuTure oF Diving: 100 Years oF halDane anD BeYonD michael a. lang anD alF o. BruBakk eDiTors lang anD BruBakk Sm ithsonian Institution Scholarly Pres s The FuTure oF Diving 100 Years oF halDane anD BeYonD ISBN-13: 978-0-9788460-5-3 ISBN-10: 0-9788460-5-2 Cover photo By haNS WIld/tIme lIfe pICtureS/Getty ImaGeS. p.o. Box 37012 mrC 957 Washington, d.C. 20013-7012 www.scholarlypress.si.edu doWNload a pdf of thIS puBlICatIoN throuGh WWW.SCholarlypreSS.SI.edu. requeSt prINt CopIeS By emaIl: SChol_preSS@SI.edu The Future of Diving 100 Years of Haldane and Beyond Michael A. Lang and Alf O. Brubakk Editors A Smithsonian Contribution to Knowledge WasHington, D.C. 2009 This publication of The Future of Diving: 100 Years of Haldane and Beyond is co-sponsored by the smithsonian institution and the office of naval Research. The symposium was convened by the Baromedical and Environmental Physiology group of the norwegian University of science and technology in trondheim, norway, on 18?19 December 2008. Published by smitHsonian institUtion sCHolaRlY PREss Po Box 37012 mRC 957 Washington, D.C. 20013-7012 www.scholarlypress.si.edu Compilation copyright ? 2009 by the smithsonian institution text and images in this publication may be protected by copyright and other restrictions or owned by individuals and entities other than, and in addition to, the smithsonian institution. Permission must be obtained by the rights holders to reproduce these materials. Cover image: Photo by Hans Wild/time life Pictures/getty images. ISBN-13: 978-0-9788460-5-3 ISBN-10: 0-9788460-5-2 The paper used in this publication meets the minimum requirements of the american national standard for Permanence of Paper for Printed library materials Z39.48?1894. Contents FoREWoRD by David H. Elliott vii aCknoWlEDgmEnts by Michael A. Lang and Alf O. Brubakk ix ExECUtivE sUmmaRY xi INTRODUCTION introductory Remarks Alf O. Brubakk 1 Welcoming Remarks Stig A. Sl?rdahl 3 J. s. Haldane, the First Environmental Physiologist Alf O. Brubakk and Michael A. Lang 5 Environmental Physiology of the Future Richard E. Moon 11 DeCOmpReSSION phySIOlOgy an introduction to Clinical aspects of Decompression illness (DCi) Costantino Balestra 17 Haldane still Rules! David J. Doolette 29 Biochemical approach to Decompression Susan R. Kayar 33 Exercise and Decompression Russell S. Richardson 41 Diving medicine and the Cellular stress Response George A. Perdrizet 47 individual Risk of Decompression sickness: Possible Effects of genomic or Epigenomic variation altering gene Expression Andreas M?llerl?kken and Ingrid Eftedal 53 inducing HsP for Protection against DCs: Heat Exposure before Diving Reduces Bubble Formation in man Jean-Eric Blatteau, Emmanuel Gempp, Jean-Michel Pontier, Costantino Balestra, Tony Mets, and Peter Germonpr? 59 iv ? tHE FUtURE oF Diving lack of signs of Brain injury in Rats on mRi after Decompression Marianne B. Havnes, Marius Wider?e, Marte Thuen, Andreas M?llerl?kken, and Alf O. Brubakk 65 Exercise, Endothelium, and Diving Physiology ?eljko Duji? 71 animal Experiments for Evaluating Decompression: Development of the modern submarine Escape method Mikael Gennser 77 DeCOmpReSSION meThODOlOgy Ultrasound for Evaluating Decompression Olav S. Eftedal 83 Current trends in Ultrasound imaging technology, sURF imaging, and Decompression induced microbubbles Lasse L?vstakken, Andreas M?llerl?kken, and Svein-Erik M?s?y 87 The Future of Dive Computers Michael A. Lang and Sergio Angelini 91 Discussion: Decompression physiology and methodology 101 STRaTegIC appROaCheS Breeding Diving scientists: Cloning, spawning, or Cultivation? ?yvind Ellingsen 109 national Centre for Hyperbaric and Diving medicine Marit Gr?nning 111 Baromedical and Environmental Physiology group (BaREn) Alf O. Brubakk 113 Physiopathology of Decompression (PHYPoDE) network Francois Guerrero, Jacek Kot, Peter Germonpr?, Jean-Michel Pontier, Alessandro Marroni, Adel Taher, Frans Cronje, and Costantino Balestra 117 international Cooperation in Diving Research Matthew Swiergosz 119 Discussion: Diving Researcher Recruitment 123 eNvIRONmeNTal phySIOlOgy How Do marine mammals avoid DCs? Andreas Fahlman 129 Does Diving Destroy the Brain? Stephen Daniels 137 Effects of Diving on the lung Einar Thorsen 145 The limits of Breath-Hold Diving Peter Lindholm 147 ContEnts ? v Parameters of Extreme Environment Diving Michael A. Lang 153 Creativity and improvisation as Phenomena and acting Potential in Different Contexts Bj?rn Alterhaug 161 Discussion: environmental physiology and Strategies for the Next 100 years 171 aPPEnDix a: ?The Prevention of Compressed-air illness? (1908) 175 aPPEnDix B: symposium speakers 281 aPPEnDix C: symposium Participants 285 it was a disappointment for me not to be able to attend this conference, but it is an honour to write its foreword. While a postgraduate student in oxford, i had a period of using the lloyd-Haldane gas-analysis apparatus.1 my contribution to respiratory research was negligible,2 but some serious effects arose in a colleague from chronic exposure to the mercury that had spilt from the apparatus and was trapped be- tween the laboratory floorboards. mercury poisoning was notifiable even in Haldane?s time; but while this incident introduced me to inhalation toxicology and health protec- tion (another area in which Haldane excelled), it also shows that an important lesson does not necessarily remain in the spotlight for a hundred years. one hundred years ago John scott Haldane was already a Fellow of new College, oxford, the Reader in Physiology at oxford University, and a Fellow of the Royal soci- ety (FRs). in the opening review, michael lang and alf Brubakk illustrate the breadth of his scientific output and his philosophical criticisms. We should also remember that, like many other researchers, he was often his own subject, and once after breathing a carbon monoxide mixture for nearly a half hour, he felt ?distinctly abnormal.? This international meeting covers what has happened since Haldane?s publications and focuses on the hyperbaric aspect of his contributions to physiology. The soldiers exposed to poison gas in the trenches and the coal miners exposed to ?after-damp? in the mines may still out number divers as the beneficiaries of Haldane?s investigations into respiratory physiology and health protection, but these two major problems that Haldane solved, occur within an otherwise physiologically normal body. in contrast, there are additional factors in diving that arise from the effects of a unique physiological variable, that of environmental pressure. Haldane investigated the decompression problems of deep air diving for the admiralty and made immense progress in providing a practical outcome for the diver and a foundation for the fur- ther development of decompression tables. The additional complexity of the effects of raised environmental pressure on the whole body affects many cellular and physiologi- cal mechanisms, some of which in turn may influence the body?s response to the pres- ence of bubbles. all these developments and revelations have catalysed the researches reviewed here and have inspired the convenors of this meeting to reprint for us the original paper of 1908. This paper is the foundation of what follows in these proceed- ings, more than 100 years later. Professor David H. Elliott, O.B.E., D.Phil (Oxon), F.R.C.P., F.F.O.M. Foreword viii ? tHE FUtURE oF Diving NOTES 1lloyd, B.B. 1958. Journal of Physiology, 143:5P. 2Cunningham, D.J.C., D.H. Elliott, B.B. lloyd, J.P. miller, J.m. Young. 1966. a comparison of the effects of oscillating and steady alveolar partial pres- sures of oxygen and carbon dioxide on the pulmonary ventilation. Journal of Physiology, 179:498?508. in 1908, John scott Haldane and coworkers published ?The Prevention of Com-pressed-air illness,? which has formed the basis of modern decompression pro-cedures. one century later, it was fitting to try to look into the future to better understand the physiological effects of the underwater environment on man. to that end, a large number of friends and colleagues agreed to gather in trondheim, norway, just days before Christmas, 18?19 December 2008. all participants contributed significantly by agreeing to speak at this symposium that celebrated Haldane?s seminal achievements, by taking part in the discussions, and by contributing papers to this volume. as coeditors, we deemed the content of the Haldane symposium and the speakers to be of such quality to warrant publication of en- during materials in the form of a proceedings volume. Because this decision was made shortly before the event, we are grateful to all authors for their responsive collaboration on this project. We thank David Elliott in particular for his fitting foreword. The Faculty of medicine of the norwegian University of science and technol- ogy kindly offered conference facilities. The Baromedical and Environmental Physiol- ogy group at ntnU provided technical and organizational support. in particular, we thank astrid Hjelde, svein-Erik gaustad, marianne Havnes, andreas m?llerl?kken, and kim-vidar Rasdal for their assistance with symposium logistics. although he was not able to attend the symposium due to unforeseen last-minute circumstances, we give a special thank you to our friend, colleague, diver, physiologist, and nasa astronaut mike gernhardt for providing his special videotaped presentation ?Diving in space,? which was a symposium highlight. Financial support was provided by the imCa (international marine Contrac- tors association); aCERgY norway; tECHniP norway; statoilHydro norway; and HElPRo as, trondheim. This publication was supported in part by the U.s. navy, office of naval Research, Undersea medicine Research Program. We also thank ginger strader and Deborah stultz of the smithsonian institution scholarly Press for their expert assistance in the timely compilation and production of these proceedings. Michael A. Lang Alf O. Brubakk Washington, D.C., USA Trondheim, Norway acknowledgments the symposium results of ?The Future of Diving: 100 Years of Haldane and Be-yond,? convened by the Baromedical and Environmental Physiology group of the norwegian University of science and technology, 18?19 December 2008, in trondheim, norway, are reported in 28 papers and 3 discussion sessions in this vol- ume. These proceedings include a wide range of environmental research studies demon- strating the breadth and diversity of decompression physiology studies. The first section treats an overview of the norwegian University of science and technology, John scott Haldane?s life and work, and environmental physiology of the future. subsequent pro- ceedings papers represent findings by international workers on overarching topics of de- compression physiology, decompression methodology, strategic approaches to decom- pression research, and environmental physiology. The volume includes contributions on clinical aspects of decompression illness, biochemical approach to decompression, the cellular stress response, individual risk of decompression sickness, and the possible ef- fects of epigenomic variation altering gene expression; on inducing heat-shock proteins for protection against decompression sickness, lack of brain injury after decompression in rats; on exercise, endothelium and diving physiology, and animal experiments for evaluating decompression; on ultrasound decompression methodologies and the fu- ture of dive computers; on decompression research strategies, collaborative networks, and recruiting new researchers to the field; on marine mammal adaptations to avoid decompression sickness, whether diving destroys the brain, diving in space, and effects of diving on the lung; on the limits of breath-hold diving and parameters of extreme environment diving; and, on creativity in science. as we are preparing to face challenges of new ideas, methodologies, and interdisciplinary approaches to studies of decompres- sion sickness, we are reminded of the limited increase in knowledge of decompression sickness as a systemic disease since Haldane?s seminal work 100 years ago. Executive summary Welcome everyone. i am very pleased to see so many people fill up this audi-torium. i never expected that the idea of this Haldane symposium would catch on and encourage so many of you to attend. our thoughts were that organizing a symposium to honor Haldane one hundred years after his seminal publi- cation in 1908 would be a good idea, because since that time, we should have a better grasp of what has happened in the field of decompression and, more importantly, what should happen in the future. We will try to discuss some of the issues that we think are important for the future and how we might approach them. in particular, how can we get young people interested in this rather limited field of science? There are quite a lot of interesting biological and practical diving problems that need to be addressed. The interactions at this symposium are going to be important. Hopefully many people will take this opportunity to talk with each other and by making their contri- butions, lead to a useful result. We have asked all speakers to submit a manuscript by march 31, 2009, and we will hopefully not have to hound you about this since it is difficult to do after we have already paid for the airline tickets. speaker forms need to be filled out and submitted to us for reimbursement. The chairman of today?s session is otto molvaer of Deep-x whom i have chosen because of his excellent ability to keep speakers within their time limits. Even though speakers may have a lot to say, we ask you to stay within your time limit because of our full program schedule. The discussions will be recorded because our goal is to publish this volume. my good friend michael lang from the smithsonian has done a lot publishing, and he has convinced me that, although this will be much work, this Haldane symposium is such an event that we need to record it for posterity and disseminate the results. in order to do that, we have supplied two roving microphones. it is important that when you make a comment or ask a question you state your name in order to attribute the remark to the appropriate speaker. otherwise, we will not know who said what, which might be an advantage in some cases. in others, some speaker might say ?i didn?t say that? for which we will have a video tape back up to prove that they did! There are several ntnU people in the audience, and we will try to help you in any way we can. i now welcome stig sl?rdahl, Dean of the ntnU medical Faculty, to share some welcoming remarks. once again, welcome to the Haldane symposium. Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, N-7491, Trondheim, Norway. introductory Remarks Alf O. Brubakk ladies and gentlemen, it is an honor for me to welcome you to this symposium, to trondheim, and to the norwegian University of science and technology (ntnU) Faculty of medicine, your host. trondheim is the number one student city of norway. Every fifth inhabitant in trondheim is a student; in total there are about 30,000 students. about 20,000 are en- rolled in the university and about 10,000 in the county college. trondheim has a long tra- dition of being a student city. We can state that we have the oldest school in scandinavia, dating back to 1030, an optimistic figure that was related to the church and the cathedral. The norwegian University of science and technology, as a university, is a fairly new construction, dating back to 1996. several institutions were merged to establish the new university: the Engineering school of norway, the norwegian institute of technol- ogy, the medical school, and the University for social sciences and Humanities. We are proud to say that our university enrolls the best students in norway. of the top 1% of students graduating from high school in norway, ntnU enrolls more than 60%, and of the top 10% of high school graduates, ntnU enrolls around 50%. as you know, norway is a social democratic country, so i will not comment on this any further. in- ternationally, universities work really hard to attract the best students, providing the chance of producing the best research with those really good brains. The Faculty of medicine has 720 registered medical students, around 250 PhD stu- dents, and 250 students attending seven different master programs. altogether, around 900 people are employed by the faculty. alf Brubakk started working here at the very beginning of the medical school, which at the time constituted only a small part of the university, but which is now one of the largest faculties. The proximity between the Engineering school and the hospital has been very important for the medical school?s research activities. alf was actually one of the professors who initiated this collaboration. in 1970, he had lunch with a fellow from the Engineering school, and being a very creative individual, immediately after this lunch, alf started to collaborate with him on the cardiovascular system. in order to estimate what happens in the cardiovascular system, they agreed that more information was needed, and this triggered the first discussions on Doppler ultrasound and instru- mentation. This was the beginning of the establishment of a business called vingmed, later purchased by gE, which entered into the international market of echocardiogra- phy. since then, the collaborations between the medical school and the Engineering school have increased considerably and alf continues to be part of this. The Faculty has conducted a population-based health survey in a county near trondheim. For the third time, we have invited the entire population of the county to participate. The HUnt study (Helse-unders?kelsen i nord-tr?ndelag) started in 1984, when more than 90% of the population showed up. ten years later, 70% of the Dean, Faculty of Medicine, Norwegian University of Science and Technology, 7491, Trondheim, Norway. Welcoming Remarks Stig A. Sl?rdahl 4 ? tHE FUtURE oF Diving population participated, and in 2006, 56% of the population showed up to give blood samples, go through an examination, and provide other vital data on health and occupational back- ground. People working in the oil industry also participated. in total, over 100,000 individuals were invited to participate in the HUnt 3 study, one of the world?s largest health stud- ies. The Principal investigator of the HUnt projects has been Jostein Holmen, and the HUnt 3 Project Director was steinar krokstad. Research related to diving has been an ever-increasing un- dertaking within our faculty. i came here as a research fellow in 1987, working in the same Department as alf, and i have, since then, been following the development of the diving research activities. i am quite impressed with the scope of alf ?s work and his creativity and success in this process. again, i wish you a warm welcome and i expect you will have two interesting symposium days. i also hope you will have an opportunity to see more of trondheim. We are cur- rently building a new university hospital here on campus, a $2.5 billion dollar project. The Faculty of medicine will be housed inside the hospital buildings, and basic scientists will be moved to the academic floor within the new hospital with interconnecting skywalks between the buildings. our vision is that more frequent interaction between basic scientists, clinical colleagues, and engineers will stimulate exciting col- laborative research. The entire construction project should be finalized by 2013. Hopefully we can reconvene the 200-year Haldane symposium in the new venue. INTrOducTION John scott Haldane can be considered the first environmental physiologist. as such, he was interested in how the human body functioned during normal life. to study that, he took his physiological insight to various extreme environments like coal mines, high altitudes, and under the ocean. Thus, using man as his subject, he showed how theoretical and experimental physiology could have direct applied uses. in modern terms, it could be said that he was the first to show the value of translational research, demonstrating for instance that the sensitivity of canaries to carbon monoxide (Co) could be used to prevent Co toxicity in man. He was born in Edinburgh on 3 may, 1860, fourth son of Robert Haldane by his second wife mary Elizabeth and died march 16, 1936, of pneumonia. Few pictures of J. s. Haldane actually exist. Haldane believed that ?the aim of the science of physiology is to deliver general principles which shall enable us to predict behaviour of the living body under various physiological conditions.? He was also considered the father of oxygen therapy: ?The first step in good practice is to know what the oxygen is aiming at, where it is going, and in what quantities.? We know about Haldane from his work on developing decompression tables, but his contribution can only be understood if one considers his previous work and interests. First of all, Haldane was an observer and an experimentalist, who always pointed out that careful observation and experiments had to be the basis of any theoretical analysis. aBSTRaCT. John scott Haldane was born in 1860 in Edinburgh, scotland and can truly be called the first environmental physiologist. He studied poisonous gases occurring in coal mines and wells, sunstroke, the physiological action of carbon monoxide and the use of a caged canary for early Co detection, the regulation of lung ventilation (with J. g. Priestley, 1905), and devised the haemoglo- binometer, the apparatus for blood-gas analysis. He also described the effects of oxygen deficiency and exercise on breathing. During the First World War, he worked on effects of poisonous gases and designed a portable oxygen administration apparatus. His work on hypoxia and the acclimatization of the human body to high altitude revolutionized concepts in respiratory physiology. Haldane published some landmark books on his philosophical ideas about the true significance of biology. most importantly, however, Haldane investigated the problems of deep diving for the British admi- ralty, developing the stage decompression method, a lasting contribution to the diving world. This elaborate experimental investigation was conducted in part in a steel pressure chamber at the lister institute and with divers in scottish deep-water lochs. in 1908, J. s. Haldane published those results in his seminal paper ?Prevention of Compressed-air illness? in the Journal of Hygiene with Dr. a. E. Boycott and lieutenant g. C. C. Damant. stage decompression allowed divers to be safely brought to the surface and made it possible to conduct 120 fsw salvage operations on the laUREntiC to recover over ?5,000,000 of gold ingots without recordable incident. Alf O. Brubakk, Norwegian University of Science and Technology, Department of Circulation and Medical Imaging, 7491, Trondheim, Norway. Michael A. Lang, Smithsonian Institution, Office of the Under Secretary for Science, P.O. Box 37012, MRC 009, Washington, D.C. 20013-7012 USA. J. s. Haldane, the First Environmental Physiologist Alf O. Brubakk and Michael A. Lang 6 ? tHE FUtURE oF Diving ?Why think when you can experiment? and ?Exhaust experi- ments and then think.? His passion for obtaining data was dem- onstrated by the fact that during his medical studies in Jena, he carefully observed the amount of beer being drunk, noting that the students on the average drank about 20 pints per evening. Haldane received his education at the Edinburgh academy, University of Edinburgh (1884), and University of Jena in cen- tral germany, after which he was a Demonstrator at University College, Dundee. From 1907?1913 he was a Reader in Physiol- ogy at oxford University where his uncle, John Burdon-sander- son, was Waynflete Professor of Physiology. J. s. Haldane became member of the Royal society in 1897 but had issues such as ?The Royal society system of selecting papers and excluding ?specula- tive? ones makes the meetings, Proceedings and transactions as dull as ditch water and quite unrepresentative of the progress of British science? (goodman, 2007). Regardless, he was award- ed Royal medallist of the society in 1916, Copley medallist in 1934, and in 1928, he was appointed Companion of Honour for his scientific work in connection with industrial disease. J. s. Haldane was personally gifted with a unique power of encouraging the faculty for research, and his teaching was characterized by his efforts to make the students observe and think for themselves. He had the great ability to add both force and charm or character, the effect of which was securing the at- tachment of his pupils. Haldane had a profound sense of public service, and he believed passionately that the world could be made a better place through the application of science. From miners dying of carbon monoxide poisoning and soldiers being gased like rats in the trenches, to mountaineers and aviators coping with high altitudes, Haldane showed that science could bring light into the darkness. a friend described him as ?almost quixotically anxious to do good to all mankind ? and to teach them all a thing or two? (goodman, 2007). Haldane was also a philosopher of science and many of his lectures were published in books, such as The Sciences and Phi- losophy in 1929, The Philosophical Basis of Biology in 1931, and The Philosophy of a Biologist in 1935. ThE SElf-ExpErImENTEr Haldane was ?himself such a coalmine canary, putting his own health and life on the line to protect others? (goodman, 2007). Haldane?s own philosophy was ?all life is a physiological self experiment.? once, on his way home from his laboratory after such an experiment, he was stopped by an oxford po- liceman who had observed the scientist?s stumbling progress. Haldane explained that it was not due to alcohol, but gas. His housekeeper offered her sympathies to his wife, kathleen: ?i know how you feel, ma?am. my husband?s just the same on a Friday night.? Haldane liked nothing better than to explore dangerous mine shafts and sewers. But it was in the specially constructed, air-tight chamber in his lab that the effects of gases on people were revealed. in an age before risk assessments, institutional Review Boards and Human subjects Ethics Committees, Hal- dane was a serial self-experimenter. He also thought nothing of exposing his own son Jack to dangerous doses of chlorine and other noxious gases. His young daughter naomi once told a 6-year-old friend outside their house: ?You come in. my father wants your blood.? Her friend screamed and ran away. fIrST papEr Haldane?s first essay in 1883 with his brother lord Haldane contributed to ?Essays in Philosophical Criticism? examined the relationship of philosophy to science and attempted to answer the questions: ?What is man; discover man?s relationship to his environment; and, knowing man?s relationship to his environ- ment, determine his function, what is he most suited to do in the world?? His real interest was the study of the relationship between the organism and the environment. This, as well as his deep feeling for social issues, would determine the focus of his professional life. fIrST STudy Haldane as sherlock Holmes, environmental investigator, asked a) What is bad air? b) What makes air dangerous to breathe? c) How can its bad effects be prevented? He proceeded by study- ing the air in overcrowded Dundee slums, turning up without warning in the middle of the night to collect air in bedrooms where eight people were sleeping. His results indicated that rooms of 180 cubic feet had 65% more carbon dioxide, twice as many molds, 254% more organic matter, including hydrogen sulfide (0.07% can be fatal), and 1000% more bacteria than normal. SElf-ExpErImENTS some of Haldane?s rebreathing experiments revolved around the concept of the good air being used up. His findings included that oxygen was a gas that supported life longer than an equal amount of air; that carbon dioxide spoils pure air once it was breathed; that after seven hours, o 2 was down to 13% and Co 2 was up to 6.5% accompanied by symptoms of heavy panting, severe headache, and vomiting; and, that after breath- ing o 2 of 2%, he went unconscious after 40 seconds. Word of his propensity for experimentation got out prompting one neighbor to knock on his door asking ?my wife?s cat has been lost and we thought that possibly? it might be here.? Haldane invented the haemoglobinometer, the apparatus for blood-gas analysis and also designed an apparatus for the accurate and fast analysis of air or mixtures of gases (snyder, 1937). ThE STINk When a select Committee called upon him to ?delve inside the lower depths of government and analyse the stink that flowed beneath,? Haldane ventured into the sewers below Westminster BRUBakk anD lang: HalDanE, tHE FiRst EnviRonmEntal PHYsiologist ? 7 mechanisms (lack of sunlight, poor ventilation, ladder climb- ing, infections, smoke, high and low temperatures, unsanitary conditions, and breathing of stone dust - silicosis). acott (1999) referred to Haldane as ?the father of the salt tablet? for his rec- ommendation of salt replacement during excessive sweating. Haldane?s altitude studies consisted of work at Pikes Peak (1911) and mount Everest of which he said ?the highest points of the Earth could be reached without the help of oxygen, pro- viding they had the right men and the right weather.? He was further involved with the design of the first prototype space suit (1921) and balloon flights to 90,000 feet (27,430 m) in 1933. also, the identification of nitrite (no 2 ?) as the active ingre- dient in red meat curing procedures dates to the late 19th cen- tury. J. s. Haldane was the first to demonstrate that the addition of nitrite to hemoglobin produced a nitric oxide (no)-heme bond, called iron-nitrosyl-hemoglobin (HbFeiino). The reduc- tion of nitrite to no by bacteria or enzymatic reactions in the presence of muscle myoglobin formed iron-nitrosyl-myoglobin. it is nitrosylated myoglobin that gives cured meat, including hot dogs, their distinctive red color, and protects the meat from oxidation and spoiling (gladwin, 2004). dEcOmprESSION STudIES aNd dIvINg TablES Haldane found that it was not pressure that damaged the body, but differences in pressure. Paul Bert (1878) dived 24 dogs to 290 fsw with rapid decompression, resulting in death by nitrogen bubbles. in his studies using goats (Figs. 1a, b) at the lister institute of Preventive medicine, Haldane used slow ascent rates of 5 feet/min. The decompression studies results were published in the seminal paper by Boycott, Damant, and Haldane (1908), one hundred years ago. Haldane made the important observation that no diver had ?the bends? after rapid decompression from 42 feet to the sur- face leading to the general principle that a 2 to 1 pressure dif- ference could be tolerated. a staged-decompression technique was developed and tables describing uptake and elimination of nitrogen were developed by his son Jack Haldane at age 13. The body was divided into six compartments with different half times, and the deepest dive tested to 210 fsw (64 msw). Haldane also suggested that oxygen should be used to shorten decompression provided the pressure was kept less than 2 bar because of the fear of oxygen toxicity. However, Haldane made little contribution to the therapy of decompression sick- ness although he recognized that recompression was the treat- ment of choice. He had doubts about the safety and efficacy of uniform decompression practice. Haldane?s experiments were conducted at the lister institute of medicine in a recompres- sion chamber and he assumed the following: a) for bubbles to form, the pressure of gas in the tissues must exceed the external pressure; b) that body tissues will hold gas in a supersaturated state unless a certain limit is reached; c) that any decompression is free from risk only if the degree of supersaturation ?can be Palace. a born iconoclast, he successfully challenged the idea that ?sewer air? was a cause of typhoid and other diseases. cOal mINE caNarIES Serinus canaria is affected by gas 20 times faster than man, prompting Haldane to hold canaries in cages in the mines. Be- ing fundamentally a kind man and concerned about the animal?s well being, Haldane modified the cage so that only the front was open and could be o-ring sealed when closed. He had mounted a small oxygen bottle on top of the cage. When the canary fell off its perch from breathing toxic gas, he would shut the front door and open the oxygen bottle, ensuring the canary?s survival. mINE accIdENT ExpErImENTS in 1896, a mine explosion occurred in tylorstown Colliery, Rhondda valley, south Wales, with over 100 men below. Enter J. s. Haldane, ?medical detective.? The toxic gas after the ex- plosion (afterdamp) had not extinguished miners? lamps, lead- ing Haldane to believe that oxygen was present and suffocation needed to be ruled out. His diagnosis was accurate: 75% of the deaths were attributed not to blast injuries, but to Co poison- ing as evidenced by pink and red skin coloration and carmine- red blood samples. Further refinement of Haldane?s curiosity about why the miners died led to more self-experimentation. Breathing 0.2% Co for 71.5 minutes, Haldane?s vision became dim, his limbs weak, he had some difficulties in waking up or walking without assistance, and his movements were very uncertain. afterwards he confessed feeling confused, making spelling mistakes, experi- encing indistinct vision, and not recognizing what he saw. oxy- gen breathing made dramatic improvements. on another occa- sion, after 29 minutes of breathing Co, Haldane calmly noted that he felt distinctly abnormal: he was panting, breathing 18 times a minute, his limbs shook, and his pulse was racing. soon, he began to feel unsteady on his feet. rESpIraTION STudIES Haldane and Priestly (1905) showed that the regulation of breathing is normally determined by the tension of carbon di- oxide in the respiratory center in the brain and that this nervous center is sensitive to variations in the tension of carbon dioxide supplied in the arterial blood. since carbon dioxide is one of the principal products of metabolism in the tissues, an explanation was afforded of the automatic changes in breathing rates that occur with alteration in bodily activity. mOrE ENvIrONmENTal STudIES Haldane conducted other environmental studies on the ef- fects of hyperthermia, nutrition, and in particular lung disease 8 ? tHE FUtURE oF Diving borne with safety?; and d) that tissue perfusion was the limiting factor in inert gas uptake (Boycott et al., 1908). His decompression experiments examined the depth and pressure exposure, its duration, and the mode and decompres- sion rate. initially, a few experiments were conducted on rab- bits, guinea pigs, rats, and mice, but it was difficult to detect symptoms in these smaller animals. The goat (Figs. 1a, b) was chosen as the experimental model ?because they were the larg- est animals which could be conveniently dealt with? and ?those who are familiar with them can detect slight abnormalities with a fair degree of certainty.? The dog was rejected because they had noted that Heller et al. (1900) had previously used them to pro- duce ?safe? decompression profiles that had failed in humans. goats were excluded from the experiments if they were ill. only 5-8 goats were used per experiment. The chamber was not ventilated because they believed Co 2 to have a minimal effect on the susceptibility to decompression sickness. The chamber temperature was not controlled, and no allowance was made for any variation in atmospheric pressure. large pressure variations were used to produce minor to severe symptoms. The compres- sion time of 6 minutes was neglected in short exposures but included in longer, deeper exposures. at the time of the experiments, Haldane knew from naval diving data that decompression from 2 bar produced no symp- toms irrespective of the time of exposure. However, decompres- sion from 2.25 bar produced the ?occasional slight case,? hence Haldane?s assumption that halving the pressure would not pro- duce symptoms. He used a ?perfusion? mathematical model of gas uptake, the half lives of which were calculated from data available at that time (acott, 1999). The most common bends symptoms in goats were limb pain where the affected limb, often the foreleg, was raised (Fig. 1a); pain was detected by ?urgent bleating and continual restlessness? with the goat often gnawing at the affected area ?such as the testicles?; temporary paralysis noted about 15 min after decompression with improvement within 30 min and the animal being ?quite well? the next day; permanent post- decompression paralysis where the hind legs were noted to be paralyzed immediately with any spontaneous improvement fol- lowed by a permanent relapse (urinary retention and an acute gut distension were also noted); ?obviously ill? goats were noted to be apathetic and ill, refused ?to move or to be tempted with corn (of which goats are inordinately fond)?; dyspnoea, a sin- ister symptom usually occurring just before the animal died; and death. importantly, Haldane?s data showed that goats had an individual variability and susceptibility to decompression sickness. Diving tables were published in 1907, the Royal navy ad- opted them for military divers in 1908 and the U.s. navy in 1912. They became the Blue Book for civilian divers who felt no discomfort from working at 210 feet. Referring to applied physiology, Richard Haldane, John scott?s brother, stated ?Dr. Haldane has shown yet one more instance of the application of science to practical work.? Haldane continued to think about the decompression problem for the rest of his life, considering how to extend and extrapolate the tables. The s.s. laUREntiC (White star lines), built by Har- land & Wolff at their Belfast yard, went down the slipway on 29 april 1909 at the time that construction on the titaniC started. Captain Reginald norton was chosen to carry 43 tons of gold bullion from great Britain to Canada. on 25 January 1917, she struck a mine and sank within an hour in 40 m of water in lough swilly, Donegal, ireland, with a huge loss of life and all the gold bullion. in 1906, Commander guybon C.C. Damant had set a world diving record of 210 feet during naval endurance diving tests. His experiences as a salvage diver were FIgURe 1. Bends in the foreleg of a goat (a) and goat chamber dives (b). Both figures from Boycott et al. (1908) and reprinted with permis- sion from Cambridge University press. BRUBakk anD lang: HalDanE, tHE FiRst EnviRonmEntal PHYsiologist ? 9 well known to the admiralty. in 1917, the 36-year-old Damant included augustus Dent on his team. He was a diver aboard s.s. laUREntiC when she sank, and knew where the bullion room was. an incredible salvage feat between 1917 and 1924 recovered 3,186 of the missing 3,211 gold bars. a further?5 in- gots were recovered in 1932 by another salvage operation, leav- ing 20 gold bars still unaccounted for at the bottom of lough swilly worth some ?10 million at current prices. over 5,000 salvage dives were conducted using Haldane?s tables in a 200 square-yard working area at 120 feet with no loss of life or seri- ous problems. haldaNE?S ENdurINg lEgacy apart from his many important contributions to physiolo- gy, where perhaps the description of the principles of breathing control is the most significant, Haldane demonstrated that envi- ronmental physiology was a scientific discipline of considerable theoretical and practical value. By studying the intact organ- ism in various extreme environments, he was able to determine how the environment influenced normal physiological variables as well as disease processes. This approach is still controversial, even in our own field of diving, where there is considerable sup- port for focusing on ?what works, works? and where studies of basic mechanisms are considered more of academic rather than practical value. Haldane, with his strong focus on science as the basis for his practical approach, demonstrated that real progress for improving health can only be achieved by combin- ing the two. This is even clearer today as we increasingly are understanding how the environment may even influence gene expression. Acknowledgments We wish to thank the Haldanes and their enduring effect, the Baromedical and Environmental Physiology group at the norwegian University of science and technology in trond- heim, and the smithsonian office of the Under secretary for science. lITEraTurE cITEd acott, C. J., J. s. Haldane, J. B. s. Haldane, l. Hill, and a. siebe. 1999. a brief resume of their lives. South Pacific Undersea Medical Society Journal 29(3):161?165. Bert, P. 1878. Barometric pressure. translation by Hitchcock, m. a. and F. a. Hitchcock. Columbus, ohio: College Book Company, 1943. Republished Bethesda, maryland: Undersea medical society, 1978. Boycott, a. E., g. C. C. Damant, and J. s. Haldane. 1908. Prevention of com- pressed air illness. Journal of Hygiene, Cambridge 8:342?425. gladwin, m. t. 2004. Haldane, hot dogs, halitosis, and hypoxic vasodilation: the emerging biology of the nitrite anion. Journal of Clinical Investigations 113(1):19?21. goodman, m. 2007. Suffer and Survive. Gas Attacks, Miners? Canaries, Space- suits and the Bends: The Extreme Life of J. S. Haldane. london: simon and schuster. 320 pp. Haldane, J. s., and J. g. Priestley. 1905. The regulation of the lung-ventilation. Journal of Physiology 32:225?266. Heller, R., W. mager, and H. von schr?tter. 1900. Luftdruckerkrankungen. Mit Besonderer Ber?cksichtigung der Sogenannten Caissonkrankheit. vienna: H?lder. snyder, J. C. 1937. a note on the use of the Haldane apparatus for the analysis of gases containing ether vapor. Journal of Biological Chemistry 122:21?25. INTrOducTION Environmental physiology can be defined as the study of adaptation to environ-mental conditions, which are often adverse. These include the following condi-tions: altitude, hyperbaria, immersion, extremes of temperature, dehydration, and altered g force: microgravity or increased g. Environmental physiology is worthy of study for at least two reasons. The first is that because humans and other animals live in, or voluntarily expose themselves to, such adverse conditions, an understanding of the biological effects is of intrinsic inter- est and importance. The second is that for those interested in understanding human physiology, environmental physiology in the field consists of an experiment in which people willingly expose themselves for fun to the kind of condition that might, for clas- sic experimental purposes, be considered unethical. like most areas of science, environmental physiology received its major start in the 19th century. an example is the series of observations that were made during high altitude balloon flights. glaisher, in 1871, reported that ?the number of heart beats per minute increases with the altitude ? at 10,000 feet, certain persons are of a flaming aBSTRaCT. Environmental physiology consists of the varied responses and adaptations of organ- isms to environmental stresses. These include cold, heat, water immersion, high and low ambient pressure, and microgravity. Why should environmental physiology be studied? First of all, under- standing of the human response to the environment can lead to protective strategies in adverse conditions. a second reason is that humans often voluntarily expose themselves to conditions in the field that would never be allowed by any human ethics committee. studying people under such circumstances can elucidate and lead to understanding of basic mechanisms. systematic study of humans under different environmental conditions began in earnest in the 19th century, when ex- emplary experiments were conducted in hypo- and hyperbaria. Hypoxia, hyperoxia, high and low g forces, and high ambient pressure have been extensively studied. However, much remains to be learned. For instance, exposure to altitude hypoxia is known to cause several clinical syndromes. While strategies to prevent and treat these conditions are well known, their etiologies are obscure. Procedures for compressing and decompressing divers have also been worked out, at least to a de- gree of safety that is acceptable. We know that bubbles instigate decompression illness. However, we know little about the biochemical connections between bubbles and syndromes. The almost exclusive focus on bubble formation, growth, and resolution has blinkered investigators to molecu- lar pathways that bubbles may trigger, the understanding of which could lead to improved preven- tion and treatment. i propose two approaches that need greater emphasis. (1) most physiology is based upon studies of the steady state, but there is much to be learned by studying physiological transients. (2) While studies of macro-physiology are crucial, they alone will not lead to a complete understanding of environmental physiology, which can only be achieved by parallel study of cellular and molecular mechanisms. Center for Hyperbaric Medicine and Environmental Physiology, Duke University Medical Center, Durham, North Carolina, 27710, USA. Environmental Physiology of the Future Richard E. Moon 12 ? tHE FUtURE oF Diving purplish red, while others are hardly affected. at 17,000 feet my lips were blue? (Doherty, 2003). amongst other things, glaish- er?s observations pointed out the variability of the biological re- sponse to hypoxia. alTITudE phySIOlOgy: frOm ThE fIEld TO ThE lab an understanding of the effects of varying degrees of hy- poxia on oxidative capacity has been obtained from field mea- surements by Pugh et al. (1964). Field observations then led to laboratory studies of acclimatization in a hypobaric chamber, where sophisticated measurements can be obtained such as arte- rial and mixed venous blood gases, cardiac output, and ventila- tion-perfusion relationships using the multiple inert-gas elimi- nation technique (sutton et al., 1988). Exposure to extreme levels of hypoxia and hyperoxia in a laboratory setting has also given us an understanding of vasoregulation in the pulmonary circulation. While hypoxia causes an increase in pulmonary vas- cular resistance, hyperoxia at 3 atmospheres absolute (ata) ac- tually reduces pulmonary vascular resistance (mcmahon et al., 2002). it has been pointed out that the warm, comfortable envi- ronment in the laboratory setting does not accurately represent conditions in the field. This has led to application of monitor- ing techniques previously available only in the lab to the moun- tain environment with often spectacular results (scherrer et al., 1996; maggiorini, 2006). Ear oximetry under field conditions has revealed evidence of exceptionally low hemoglobin-oxygen saturations (Hackett and Roach, 1987) (see Fig. 1). End-tidal PCo 2 measurements atop mount Everest revealed values lower than expected from the laboratory environment (sutton et al., 1988; malconian et al., 1993). more recently, arterial blood gas analysis has been obtained from climbers very close to the sum- mit of mount Everest, with yet different values (grocott et al., 2009). of course, arterial and mixed venous blood gas analysis rep- resents only surrogate measurements of processes at the level of the cell (see Fig. 2). We have learned from near infrared absorp- tion that there is variability in the response of tissue to the type of changes that occur during altitude exposure. For example, Hampson and Piantadosi (1990) observed measurable changes in cytochrome a,a 3 redox level in the brain during manipula- tion of arterial PCo 2 . During hypocapnia cytochrome a,a 3 re- dox in the brain moves towards a reduced state, presumably due to a reduction in cerebral blood flow and oxygen delivery. at the same time, only minimal changes occur in muscle (see Fig. 3). This may explain how the profound hypocapnia that occurs during high altitude exposure may adversely affect brain oxygenation while muscle oxygenation is maintained within ac- ceptable limits. as yet, estimates of tissue oxygenation using needle electrodes or near infrared optical techniques have had limited field use. The clinical conditions associated with prolonged hypoxia have been well described from field observations. acute moun- tain sickness (ams) consists of nausea, headache, and malaise. While this disorder is poorly understood, the time course of its spontaneous resolution even when hypoxia is maintained (sampson and kobrick, 1980), and development of appropriate prophylactic measures have also been developed from observa- tions in the field. Headache and progressive encephalopathy are the hallmarks of high altitude cerebral edema (HaCE), in which, despite its name, there is scant evidence for significant cerebral edema (see below). High altitude pulmonary edema (HaPE) is a condition of generally healthy people, which became well-known during the india-China border conflict of 1962, where the incidence of HaPE was noted to be 13-15% among indian army recruits transported rapidly from sea level to altitudes above 3,350 m (menon, 1965; singh et al., 1965). The condition was initially believed to be due to acute left heart failure induced by hypoxia, a hypothesis that was quickly dismissed by right heart catheter- ization measurements. Pulmonary artery wedge pressure in pa- tients afflicted with HaPE was shown to be normal, although FIgURe 1. ear oximeter in a sleep study of a climber at 4,400 m on mt. mcKinley. When he attempted to climb to a higher altitude, he developed severe amS. The severe degree of his hypoxemia may have been explained by a negligible increase in ventilation when exposed to hypoxia (absent hypoxic ventilatory response). Re- drawn from hackett and Roach (1987), with permission. FIgURe 2. Oxygen transport cascade from atmosphere to mito- chondrion. From moon and Camporesi (2004), with permission. moon: EnviRonmEntal PHYsiologY oF tHE FUtURE ? 13 than low altitude U.s. residents, yet during hypoxia their fore- arm blood flow is twice as high (Erzurum et al., 2007). dIvINg phySIOlOgy aNd mEdIcINE as in altitude physiology, high pressure physiology began as a result of the adverse effects of large numbers of caisson and tunnel workers during and after exposure to compressed air. musculoskeletal and neurological syndromes occurring after decompression from high pressure were initially not understood and not treatable. in 1878, Bert (1978) provided the first clues as to pathophysiology by demonstrating that decompression of animals after a hyperbaric exposure produced bubbles in the blood and that administration of oxygen would resolve them. We now know that bubble formation in tissues can occur either via de novo formation of bubbles in situ due to supersaturation of inert gas (usually n 2 or He) in tissue or due to pulmonary barotrauma during decompression. treatment of so-called de- compression illness (DCi) was developed entirely empirically on the basis of men experiencing relief of symptoms upon re- entering the compressed air environment during a following shift. This was eventually systematized by Ernest moir, an en- gineer, who instituted regular recompression for afflicted men working in the Hudson River tunnel project in new York City. moir?s approach of recompression using air reduced the death rate from 25% per annum to 2 deaths of 120 men at work in 15 months (1.3%) (moir, 1896). This therapeutic success was fol- lowed by successful strategies to prevent compressed air illness, in which staged decompression was proposed by Boycott et al. (1908). While it is generally agreed that bubbles are the instigators of decompression illness, little is known about the downstream events that produce decompression sickness. one mechanism for which there is some evidence is related to bubble-induced damage to the endothelium. once believed to be nothing more than the lining of blood vessels, the endothelium is now under- stood to play a major role in maintenance of fluid homeostasis and regulation of vascular tone. in the endothelium, the en- zyme nitric-oxide synthase catalyzes the reaction of oxygen and l-arginine to produce nitric oxide (no) and citrulline. nitric oxide can then diffuse into the bloodstream where it can inhibit adhesion of platelets and leucocytes. it can also diffuse into the adjacent smooth muscle cell. There, it activates the enzyme gua- nylyl cyclase, which facilitates the conversion of gtP to cyclic gmP (cgmP). This initiates a series of reactions that result in relaxation of vascular smooth muscle. This mechanism underlies the effect of agonists that interact with receptors on the endo- thelium cell, such as catecholamines. it has been demonstrated that vascular bubbles can injure the endothelium and reduce the vasoactive effects of compounds such as substance P and ace- tylcholine (nossum et al., 1999). nitric oxide production can be up-regulated by exercise. indeed, studies in animals (Wisl?ff et al., 2004) and humans (Dujic et al., 2004) have provided preliminary evidence that appropriately timed exercise prior to pulmonary artery pressure was high (Fred et al., 1962; Hultgren et al., 1964). This provided a clue, since the previously described in- crease in pulmonary artery pressure with hypoxia (hypoxic pul- monary vasoconstriction, HPv) tends to be greater in people susceptible to HaPE (Hultgren et al., 1971). Furthermore, reducing Pa pressure with vasodilators can prevent and treat HaPE (scherrer et al., 1996; maggiorini et al., 2006; anand, 1998). observations using the balloon occlusion method to estimate pulmonary capillary pressure have further implicated high intravascular pressure as a cause of HaPE (maggiorini et al., 2001). a genetic predisposition to HaPE has been suggest- ed by evidence implicating polymorphisms of the enos gene (Dehnert et al., 2007). While the mechanisms of HPv are incompletely under- stood, recent advances have been based upon molecular evi- dence implicating nitric oxide as a mediator. observations by mcmahon et al. (2002) have demonstrated the relationship between blood Po 2 and s-nitrosohemoglobin. in low Po 2 situ- ations, no is released from its binding to reactive thiols (Cys b 93 ) of hemoglobin. a genetic factor is suggested by racial dif- ferences in the degree of HPv. The rise in pulmonary artery pressure in response to hypoxia is greatest in north americans, somewhat attenuated in south americans, and significantly less in tibetans, a population adapted to high altitude (groves et al., 1993). Recently it has been observed that tibetans have a significant higher concentration of nitroso proteins in the blood FIgURe 3. muscle and brain redox as a function of pCO 2 . During hypocapnia cytochrome a,a 3 redox in the brain becomes more re- duced, presumably due to reduced cerebral blood flow and oxygen delivery. hypocapnia induces only minimal changes in muscle. Redrawn from hampson and piantadosi (1990), with permission. 14 ? tHE FUtURE oF Diving a decompression stress may reduce the number of intravascular bubbles. similarly, administration of nitroglycerine to animals prior to decompression stress can reduce venous gas embolism (m?llerl?kken et al., 2006). Whether either of these interven- tions will reduce decompression sickness in humans remains to be investigated. frOm bOdy TO mOlEculE: OrThOSTaTIc INTOlEraNcE afTEr SpacE flIghT orthostatic intolerance is commonly experienced after spaceflight (Buckey et al., 1996). This was initially believed to be mostly caused by relative hypovolemia induced in space by diuresis due to central translocation of blood from the ex- tremities (Blomqvist and stone, 1983). if this were the sole mechanism the syndrome should be preventable by volume loading prior to re-entry and landing. However, evidence has accumulated suggesting impaired vasoconstrictive response to standing (Buckey et al., 1996). Using a rat model designed to simulate weightlessness (hind limb unloading with head down position) impairment of the arterial vasoconstrictor response to norepinephrine was observed (Delp et al., 1993; Purdy et al., 1998). another possible mechanism involves up-regulation of enos (nyhan et al., 2002). such observations not only pro- vide a more complete understanding of the problem, but will undoubtedly lead to more effective pharmacological or physi- ological countermeasures. STEady STaTE mEaSurEmENTS: blINdErS ON While advanced technology is facilitating increasingly so- phisticated measurements in environmental physiology, our understanding is thus far largely built on steady state measure- ments. However, significant transients can be missed by steady state techniques. For example, in primary pulmonary hyper- tension, indwelling pulmonary artery catheters have shown that mean pulmonary artery pressure and pulmonary vascular resistance may change substantially from hour to hour (Rich et al., 1985). similarly, during acute immersion in cold water, quite large transient changes occur in both systemic and pul- monary artery pressure (Wester et al., 2009). another example is ?normal pressure? hydrocephalus, a condition consisting of gait disturbance, cognitive decline, and incontinence. in this condition, enlarged ventricles suggest that intracranial pressure (iCP) is elevated. static measurements of iCP in this condition are usually normal, however long term observations of intra- cranial pressure in these patients reveal cyclical variations, with periodic elevations to high levels (mcgirt et al., 2005). The pathophysiology of high altitude headache is unknown, but the possibility has been raised that it may be due to increased intracranial pressure, although with some contrary evidence. For example, a study in which normal volunteers breathed 12% oxygen for 18 hours induced either severe headache or clinical acute mountain sickness (ams) (Bailey et al., 2006). However, there was neither mRi evidence of brain edema nor elevation of lumbar CsF (cerebrospinal fluid) pressure. The possibility that transients may be important has been suggested by a remark- able field study, in which iCP was directly measured in three subjects during an expedition to Hagshu Peak in the Himalayas (altitude 6,330 m) (Wilson and milledge, 2008). While under resting conditions CsF pressure was normal, intermittent pres- sure elevations occurred in two of three subjects studied. of these three subjects, the only one to develop headaches was a subject in whom elevated CsF pressure was observed. Summary measurements at the level of the whole organism provide fundamental observations and remain the bedrock of our de- scriptions of environmental physiology. Whole human or an- imal studies should provide the impetus to look beyond our original observations into the cellular and molecular mecha- nisms. By so doing, we may gain an understanding of not only phenomena of immediate interest, but of physiology in general. Further advances will require widening of the scope of research (traditionally based on steady state methods) to include tech- niques that can detect and measure transients. By implementing a wider array of investigative techniques to include transient phenomena and cellular and molecular methodologies, a clearer understanding might be obtained of hitherto elusive phenom- ena such as high altitude pulmonary edema and decompression sickness. lITEraTurE cITEd anand, i. s., B. a. Prasad, s. s. Chugh, k. R. Rao, D. n. Cornfield, C. E. milla, n. singh, s. singh, and W. selvamurthy. 1998. Effects of inhaled nitric oxide and oxygen in high-altitude pulmonary edema. Circulation 98:2441?2445. Bailey, D. m., R. Roukens, m. knauth, k. kallenberg, s. Christ, a. mohr, J. genius, B. storch-Hagenlocher, F. meisel, J. mcEneny, i. s. Young, t. steiner, k. Hess, and P. Bartsch. 2006. Free radical-mediated dam- age to barrier function is not associated with altered brain morphology in high-altitude headache. Journal of Cerebral Blood Flow and Metabolism 26:99?111. Bert, P. 1978. Barometric Pressure (la Pression Barom?trique). Bethesda, mary- land: Undersea medical society. Blomqvist, C. g., and H. l. stone. 1983. Cardiovascular adjustments to gravi- tational stress. in shepher, J. t., and F. m. abboud, eds. The Cardiovas- cular System. Bethesda, maryland: american Physiological society. pp. 1025?1063. Boycott, a .E., g. C. C. Damant, and J. s. Haldane. 1908. Prevention of com- pressed air illness. Journal of Hygiene (Cambridge) 8:342?443. Buckey, J. C., Jr., l. D. lane, B. D. levine, D. E. Watenpaugh, s. J. Wright, W. E. moore, F. a. gaffney, and C. g. Blomqvist. 1996. orthostatic intoler- ance after spaceflight. Journal of Applied Physiology 81:7?18. Dehnert, C., m. m. Berger, H. mairbaurl, and P. Bartsch. 2007. High altitude pulmonary edema: a pressure-induced leak. Respiratory Physiology and moon: EnviRonmEntal PHYsiologY oF tHE FUtURE ? 15 Neurobiology 158:266?273. Delp, m. D., t. Holder-Binkley, m. H. laughlin, and E. m. Hasser. 1993. va- soconstrictor properties of rat aorta are diminished by hindlimb unweight- ing. Journal of Applied Physiology 75:2620?2628. Doherty, m. J. 2003. James glaisher?s 1862 account of balloon sickness: alti- tude, decompression injury, and hypoxemia. Neurology 60:1016?1018. Duji?, ?., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n. m. Petri, U. Wisl?ff, and a. o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555(3):637?642. Erzurum, s. C., s. H. ghosh, a. J. Janocha, W. xu, s. Bauer, n. s. Bryan, J. tejero, C. Hemann, R. Hille, D. J. stuehr, m. Feelisch, and C. m. Beall. 2007. Higher blood flow and circulating no products offset high-altitude hypoxia among tibetans. Proceedings of the National Academy of Sciences 104:17593?17598. Fred, H. l., a. m. schmidt, t. Bates, and H. H. Hecht. 1962. acute pulmonary edema of altitude. Circulation 25:929?937. grocott, m. P., D. s. martin, D. Z. levett, R. mcmorrow, J. Windsor, and H. E. montgomery. 2009. arterial blood gases and oxygen content in climb- ers on mount Everest. New England Journal of Medicine 360:140?149. groves, B. m., t. Droma, J. R. sutton, R. g. mcCullough, R. E. mcCullough, J. Zhuang, g. Rapmund, s. sun, C. Janes, and l. g. moore. 1993. mini- mal hypoxic pulmonary hypertension in normal tibetans at 3,658 m. Journal of Applied Physiology 74:312?318. Hackett, P. H., and R. C. Roach. 1987. medical therapy of altitude illness. An- nals of Emergency Medicine 16:980?986. Hampson, n. B., and C. a. Piantadosi. 1990. near-infrared optical responses in feline brain and skeletal muscle tissues during respiratory acid-base imbal- ance. Brain Research 519:249?254. Hultgren H. n., C. E. lopez, E. lundberg, and H. miller. 1964. Physiologic studies of pulmonary edema at high altitude. Circulation 29:393?408. Hultgren, H. n., R. F. grover, and l. H. Hartley. 1971. abnormal circulatory responses to high altitude in subjects with a previous history of high-alti- tude pulmonary edema. Circulation 44:759?770. maggiorini, m., C. melot, s. Pierre, F. Pfeiffer, i. greve, C. sartori, m. lepori, m. Hauser, U. scherrer, and R. naeije. 2001. High-altitude pulmonary edema is initially caused by an increase in capillary pressure. Circulation 103:2078?2083. maggiorini, m., H. P. Brunner-la Rocca, s. Peth, m. Fischler, t. Bohm, a. Bernheim, s. kiencke, k. E. Bloch, C. Dehnert, R. naeije, t. lehmann, P. Bartsch, and H. mairbaurl. 2006. Both tadalafil and dexamethasone may reduce the incidence of high-altitude pulmonary edema: a random- ized trial. Annals of Internal Medicine 145:497?506. malconian, m.k., P. B. Rock, J. t. Reeves, a. Cymerman, and C. s. Houston. 1993. operation Everest ii: gas tensions in expired air and arterial blood at extreme altitude. Aviation, Space, and Environmental Medicine 64:37?42. mcgirt, m. J., g. Woodworth, a. l. Coon, g. Thomas, m. a. Williams, and D. Rigamonti. 2005. Diagnosis, treatment, and analysis of long-term outcomes in idiopathic normal-pressure hydrocephalus. Neurosurgery 57:699?705. mcmahon, t. J., R. E. moon, B. P. luschinger, m. s. Carraway, a. E. stone, B. W. stolp, a. J. gow, J. R. Pawloski, P. Watke, D. J. singel, C. a. Piantadosi, and J. s. stamler. 2002. nitric oxide in the human respiratory cycle. Nature Medicine 8:711?717. menon, n. D. 1965. High-altitude pulmonary edema: a clinical study. New Eng- land Journal of Medicine 273:66?73. moir, E. W. 1896. tunnelling by compressed air. Journal of the Society of Arts 44:567?585. m?llerl?kken, a., v. J. Berge, a. J?rgensen, U. Wisl?ff, and a. o. Brubakk. 2006. Effect of a short-acting no donor on bubble formation from a satu- ration dive in pigs. Journal of Applied Physiology 101:1541?1555. moon, R. E., and E. m. Camporesi. 2004. Respiratory monitoring. in miller, R.D., ed. Anesthesia. 6th ed. Philadelphia: Churchill livingstone. pp. 1437?1481. nossum, v., s. koteng, and a. o. Brubakk. 1999. Endothelial damage by bub- bles in the pulmonary artery of the pig. Journal of Undersea and Hyperbaric Medicine 26:1?8. nyhan, D., s. kim, s. Dunbar, D. li, a. shoukas, and D. E. Berkowitz. 2002. impaired pulmonary artery contractile responses in a rat model of micro- gravity: role of nitric oxide. Journal of Applied Physiology 92:33?40. Pugh, l. g., m. B. gill, s. lahiri, J. s. milledge, m. P. Ward, and J. B. West. 1964. muscular exercise at great altitudes. Journal of Applied Physiology 19:431?440. Purdy, R. E., s. P. Duckles, D. n. krause, k. m. Rubera, and D. sara. 1998. Ef- fect of simulated microgravity on vascular contractility. Journal of Applied Physiology 85:1307?1315. Rich, s., g. E. D?alonzo, D. R. Dantzker, and P. s. levy. 1985. magnitude and implications of spontaneous hemodynamic variability in primary pulmo- nary hypertension. American Journal of Cardiology 55:159?163. sampson, J. B., and J. l. kobrick. 1980. The environmental symptoms ques- tionnaire: revisions and new filed data. Aviation, Space, and Environmental Medicine 51:872?877. scherrer, U., l. vollenweider, a. Delabays, m. savcic, U. Eichenberger, g.R. kleger, a. Fikrle, P.E. Ballmer, P. nicod, and P. Bartsch. 1996. inhaled nitric oxide for high-altitude pulmonary edema. New England Journal of Medicine 334:624?629. singh, i., C. C. kapila, P. k. khanna, R. B. nanda, and B. D. Rao. 1965. High- altitude pulmonary oedema. Lancet 1:229?234. sutton, J. R., J. t. Reeves, P. D. Wagner, B. m. groves, a. Cymerman, m. k. malconian, P. B. Rock, P. m. Young, s. D. Walter, and C. s. Houston. 1988. operation Everest ii: oxygen transport during exercise at extreme simulated altitude. Journal of Applied Physiology 64:1309?1321. Wester, t. E., a. D. Cherry, n. W. Pollock, J. J. Freiberger, m. J. natoli, E. a. schinazi, P. o. Doar, a. E. Boso, E .l. alford, a. J. Walker, D. m. Uguc- cioni, D. kernagis, and R.E. moon. 2009. Effects of head and body cool- ing on hemodynamics during immersed prone exercise at 1 ata. Journal of Applied Physiology 106:691?700. Wilson, m. H., and J. milledge. 2008. Direct measurement of intracranial pres- sure at high altitude and correlation of ventricular size with acute moun- tain sickness: Brian Cummins? results from the 1985 kishtwar expedition. Neurosurgery 63:970?975. Wisl?ff, U., R. s. Richardson, and a. o. Brubakk. 2005. Exercise and nitric oxide prevent bubble formation: a novel approach to the prevention of de- compression sickness? Journal of Physiology 555:825?829. INTrOducTION Decompression illness (DCi) is a complex condition that can appear with a wide variety of signs and symptoms. any significant organic or functional decre-ment in individuals who have been exposed to a reduction in environmental pressure must be considered as possibly being DCi until proven otherwise. This applies to acute, sub-acute and chronic changes related to decompression and may be related to acute clinical symptoms or to situations that may develop subclinically and insidiously. it is in fact generally accepted that subclinical forms of DCi, with little or no reported symptoms, may cause changes in the bones, the central nervous system and the lungs (kelemen, 1983; shinoda et al., 1997; Wilmshurst and Ross, 1998). generally, a disorder is a physical derangement, frequently slight and transitory in nature. a disease is considered a condition of an organ, part, structure, or system of the body in which there is abnormal function resulting from genetic predisposition, diet, or environmental factors. a disease is typically a more serious, active, prolonged, and deep-rooted condition. DCi should be considered a disorder due to a physical primary cause that can transform into a disease unless adequate and timely action is undertaken to abort or to minimize the pathophysiological effects of bubbles on the body tissues. The predominant physical cause of DCi is the separation of gas in the body?s tis- sues, due to inadequate decompression, leading to an excessive degree of gas super- saturation (kumar et al., 1990). Rapid decompression (rate of ascent or omission of decompression stops) is a primary cause of gas separation in tissues (Figure 1). The most obvious prevention strategy for DCi is, therefore, determining and ob- serving appropriate ascent and decompression procedures (marroni and Zannini, 1981; aBSTRaCT. Decompression illness (DCi), Decompression sickness (DCs), Dysbaric illness (Di), disorder, syndrome are terms associated with the clinical signs or symptoms originally generated by a reduction of absolute pressure surrounding the patient. For 100 years, the definition of the ?dis- ease? is a matter of ?disputes? or ?consensi.? We understand nowadays that it is not enough to know how to cure evident clinical manifestations, but also to reduce or virtually eliminate the primary physical cause for the physiological damages: the gas separation phase from saturated tissues ? sta- tionary or circulating bubbles. to achieve this goal, research is more oriented on the decompression procedures or the diver preconditioning (heat exposure, physical activity, whole body vibration, an- tioxidant medication, oxygen breathing, hyperbaric oxygen therapy, hydration or dehydration) and postconditioning (different decompression procedures or models, deep stops, shallow stop followed by a deeper one, post exposure hydration, speed of ascent, exercise during decompression). some factors that were believed to be crucial, such as patency of the cardiac Foramen ovale or gender, are considered less important than modified decompression procedures that are studied today with sharper technology. Haute Ecole Paul Henri Spaak, Environmental and Occupational Physiology Laboratory, 91, Avenue C. Schaller, 1160 Auderghem, Belgium. an introduction to Clinical aspects of Decompression illness (DCi) Costantino Balestra 18 ? tHE FUtURE oF Diving marroni et al., 2001). Unfortunately, the recommendations for decompression are largely empirical and not always reliable. This is confirmed by the finding that more than half of the DCi cases managed by Divers alert network (Dan) worldwide over the past several years have not been associated with an obvious viola- tion of decompression procedures, dive table, or dive computer limits; they have been ?unpredictable.? This has led to a search for other contributing factors to the development of DCi, such as a Patent Foramen ovale, in an effort to explain the wide varia- tion in individual susceptibility to DCi. other factors include complement activation in the presence of gas bubbles as well as an uncertain relationship between gas bubbles, blood cells, and the capillary endothelial lining in response to bubble presence and development of DCi. The manifestations of DCi are sometimes trivial and subtle. These are likely to be ignored or denied by individual divers, training organizations, and emergency physicians unless they are made aware of them and offered specific information on the manifestations of DCi. However, there appears to be growing evidence that underreported, underestimated and undertreated signs and symptoms of DCi may result in permanent organic or functional damage so that raising the level of suspicion amongst divers and physicians alike becomes increasingly important. although the presence of Doppler-detectable gas bubbles in the blood is not necessarily predictive of clinically evident DCi, the appearance of DCi in the absence of detectable pul- monary artery and venous bubbles is rare. There is even grow- ing experimental and clinical evidence that suggests that asymp- tomatic ?silent? bubbles in the body may be causing cellular and biological reactions that release secondary potentially damaging biochemical substances in the blood. dEfININg dcI The previously adopted classification criteria, i.e., type i or type ii DCs (decompression sickness) and agE (arterial gas Embolism), are inadequate in that they presume that the underlying pathophysiology is fully appreciated. Unfortunately, there is great disparity in the application of this classification of DCi amongst specialists when asked to define similar cases of decompression disorders using this traditional classification. Consequently, a descriptive form of classification has appeared that uses the common term ?DCi,? followed by a description of the clinical signs and symptoms and their onset and develop- ment characteristics. The latter has been considered both more universally understandable and simpler to teach. it also shows a much higher degree of correlation among specialists describing the same DCi cases. For purposes of clarity and consistency, DCi refers to disorders of decompression that are clearly due to DCs or where the origin or embolized gas cannot be defini- tively attributed to pulmonary barotrauma. Where the cause of arterial embolization is the direct consequence of pulmonary overexpansion, the term agE is used. Epidemiologically, there is universal consensus among the international diving medical community that the incidence of DCi is generally very low and that there is no significant gender-related susceptibility. There is also consensus that neuro- logical manifestations are by far the most common form of DCi amongst recreational divers. many yet unknown aspects of DCi are the subject of on- going international studies. These include: the relationship be- tween gas separation and DCi injury; the relationship between clinical symptoms and the severity of the disease; the relation- ship between initial clinical onset, treatment results, and per- manent sequelae; the reason for the large variation in individual susceptibility to DCi; the lifespan of gas bubbles; and the true incidence of DCi. dcI hISTOry Boyle (1670) demonstrated that DCi could be produced in a reptile by a sudden lowering of atmospheric pressure. The first clinical recording of DCi was in compressed-air workers. triger (1845) reported that two men had suffered ?very sharp pain? in the left arm, and another had pain in the knees and left shoulder 30 minutes after emerging from a seven-hour exposure at pressure (the pressure could have ranged between 2.4 and 4.25 atm). although not knowing what it was, triger also reported the clinical treatment for DCi as ?rubbing with spirits of wine soon relieved this pain in both men and they kept working on the following days.? Pol and Watelle (1854) wrote that they were ?justified in hoping that a sure and prompt means of relief would be to recompress immediately, then decompress very carefully.? Yet it was only many years later that their advice was heeded. in 1878, Paul Bert demonstrated that the cause of DCi was dissolved nitrogen going into gas phase in body tissues and that this bubble formation was responsible for symptoms. Bert also highlighted the existence of ?silent bubbles? in the venous blood. He understood that recompression was the key treatment of value and that it should be applied promptly. He also used oxygen at one atmosphere following very rapid decompression FIgURe 1. Intravascular bubbles in rodent submaxillary capillar- ies (Courtesy of Divers alert Network). BalEstRa: CliniCal asPECts oF DCi ? 19 medical personnel), caisson work, and tunneling under pressure. PredisPosing FActors Whereas the primary factor causing DCi is undisputedly the reduction in ambient pressure causing a rapid inert gas de- saturation of tissues, several factors have been identified that can increase an individual?s susceptibility to DCi. exercise Exercise during exposure to increased ambient pressure (during the bottom phase of the dive) appears to increase the incidence of DCi. The probable explanation is that increased perfusion during exercise leads to a corresponding increase in inert gas uptake, which must be subsequently eliminated during decompression. Exercise during ascent has differential effects. During de- compression stops, mild exercise appears to be helpful. on the other hand, increased activity during pressure change appears to increase the DCi risk. at least 3 mechanisms may help to explain this effect: 1. The formation of gas micronuclei. Rapidly flowing blood, especially in the area of vessel bifurcation, may create foci of relative negative pressure through a venturi effect. mol- ecules of gas from the surrounding supersaturated blood may then diffuse into these foci down a partial-pressure gradient. The resulting localized collections of small num- bers of gas molecules called micronuclei are thought to act as a nidus for further bubble growth formation; 2. increased local Co 2 production by exercising muscle may play a role since Co 2 is a highly diffusible gas that could contribute to the formation of gas micronuclei. Even small increases in F i Co 2 seem to increase the incidence of DCi. The mechanism of this effect is not clearly understood; and 3. increases in core body temperature due to increased muscle activity may reduce the solubility of gas in body tissues leading to bubble formation. However, very recent research results are questioning some of these assumptions regarding exercise and diving, in particu- lar that of exercise prior to diving. The latter appears to lower DCi incidence depending on when the exercise is performed. The explanation of these findings is still hypothetical, although nitrous oxide seems to be protective when it is produced by an increase in physical exercise 20 hours before diving (Wisloff and Brubakk, 2001; Wisloff et al., 2003; Dujic et al., 2004; Wisloff et al., 2004). There is an association between recent lo- cal musculoskeletal injuries and an increased incidence of DCi at or near the site of the injury. The mechanism responsible for this phenomenon is unclear. Changes in local perfusion and increased gas micronuclei formation in injured tissue are postu- lated mechanisms. and observed that cardiopulmonary symptoms, but not spinal cord paralysis, could be relieved by normobaric oxygen breath- ing (Bert et al., 1943). moir (1896) published his work on the 1889 excavations of the Hudson River tunnel. Facing a tragic fatality rate of 25% of the employed workers due to DCi, he installed a recompres- sion chamber at the work site. Following this intervention there were only a further two deaths out of 120 men employed over the following 15 months. moir wrote: With a view to remedying the state of things an air compartment like a boiler was made in which the men could be treated homeo- pathically, or reimmersed in compressed air. it was erected near the top of the shaft, and when a man was overcome or paralyzed, as i have seen them often, completely unconscious and unable to use their limbs, they were carried into the compartment and the air pressure raised to about 1/2 or 2/3 of that in which they had been working, with immediate improvement. The pressure was then lowered at the very slow rate of one pound per minute or even less. The time allowed for equalization being from 25 to 30 minutes, and even in severe cases the men went away quite cured. Unknowingly, moir was recording both the means of pre- vention and treatment of DCi. variations of his techniques, now called surface decompression, are currently still used. Even though few subsequent publications appeared on re- compression treatment for the next 30 years, it was the widely accepted notion that, to be effective, recompression should commence promptly followed by slow decompression. These principles remain in effect to this day even though the pres- sures used, breathing gases applied, and rates of decompression observed, have undergone much modification. ETIOlOgy aNd paThOphySIOlOgy Of dcI DCi is a disease with protean clinical manifestations. it fol- lows the appearance of gas bubbles produced by the excessively rapid lowering of ambient pressure. This reduction enables inert gas dissolved in tissue to enter the gas phase causing the forma- tion of gas bubbles in tissues and body fluids. The clinical syndrome is known by a multitude of names including decompression sickness, DCi, decompression injury, caisson disease, bends, chokes, staggers, dysbarism and gas bub- ble injury. although arterial gas embolism is usually associated with pulmonary barotraumas, decompression bubbles can also lead to embolization if there is shunting between the venous drainage and systemic circulation (e.g., intracardiac and intra- pulmonary shunting). This blurs the boundaries between de- compression sickness and arterial gas embolism, which is why the term DCi was created. many languages do not differentiate between ?sickness? and ?illness? so that the terms ?dysbarism?, ?dysbaric illness,? or ?dysbaric injury? have become equivalent terms for DCi. Clinical settings of DCi include diving, aviation, hyperbaric oxygen therapy (i.e., nurses, chamber assistants, and 20 ? tHE FUtURE oF Diving cold wAter Diving in cold water tends to increase the incidence of DCi. inert gas uptake is generally not affected because the ex- ercising diver is usually warm and has increased tissue perfu- sion due to exercise (martini et al., 1989; gerriets et al., 2000). However, as the diver cools during the dive and at the safety or decompression stops, the diver?s tissues experience a reduction in blood flow due to the cold and an increase in solubility tends to retain more gas. as the diver rewarms after the dive, the ex- cess gas may be released as bubbles. Age advancing age increases the incidence of DCi for reasons that are not yet clearly known but may be related to the re- duction in pulmonary function or the reduction of tissue microvascularization. dehydrAtion Dehydration was reported as a factor that increases the risk of DCi during studies on aviators during World War ii. The mechanism is again unclear. Changes in the surface tension in serum favoring bubble formation have been postulated. anec- dotal reports suggest that prior alcohol ingestion increases the incidence of DCi, possibly through this mechanism. some re- cent papers add insight on the mechanism and advocate new ap- proaches, considering hydration of the tissues more important than plasmatic volume or surface tension (gempp et al., 2009). FAtigue as with alcohol, there is anecdotal evidence suggesting that significant fatigue preceding a dive increases the incidence of DCi. it is uncertain whether the fatigue is a subtle indicator of some unidentified biochemical factor or a nonspecific warning of general hemodynamic factors. paThOgENESIS Of dcI VAsculAr obstruction vascular obstruction by bubbles or bubble-formed com- plexes may occur in the systemic or pulmonary circulation, a most important element in the pathogenesis of DCi. vascular obstruction may occur as bubbles enter the circulation from su- persaturated tissues and slow down venous return or due to em- bolization of vascular beds by bubbles formed elsewhere. such disturbances may be clinically invisible in noncritical areas such as fatty tissue, but may be life threatening in critical organs such as the central nervous system and heart. Diffuse peripheral vascular obstruction and stasis with re- sultant tissue hypoxia or anoxia may lead to metabolic acidosis and hypovolemia due to increased capillary permeability. aci- dosis and hypovolemia may considerably impair cardiovascular function. vascular obstruction of pulmonary capillaries, sec- ondary to embolization of bubbles or bubble-formed complexes in venous blood, results in increased pulmonary vascular resis- tance, bronchiolar constriction and peribronchiolar oedema. These changes may lead to alterations in ventilation-perfusion ratios with resultant arterial hypoxemia, a condition called the chokes. blood-bubble interActions: coAgulAtion much attention has been devoted to the possible conse- quences of blood-bubble interaction. Bubbles are thought to be capable of activating Hageman Factor (Factor xii) with ac- tivation of coagulation, contributing to vascular obstruction. Bubbles constitute a foreign element in the blood and activate the complement and coagulation cascades. They may even cause ?denaturation? of lipoproteins with the release of large quantities of lipid. Electron-micrographic studies in animals have shown vascular obstruction by a complex that appears to be composed of a gas bubble surrounded by a layer of lipid, to which platelets are agglutinated. This and similar observa- tions have given rise to a variety of experimental work inves- tigating inter alia the possible usefulness of anticoagulants in DCi. to date there is no firm experimental evidence to indicate that disseminated intravascular coagulation occurs in DCi, nor that routine anticoagulation is therapeutically useful. Enhanced coagulation at local sites in tissue, however, may contribute to the pathogenesis of DCi. Coagulation Factor xiia is, however, capable of triggering the reaction of the complement system. The sequence of reactions of this system produces factors that increase capillary permeability and factors that are chemotac- tic to leukocytes. Factor xiia is also capable of activating the kinin-Bradykinin system with liberation of bradykinin and histamine. Bradykinin may cause local pain. Both are capable of increasing capillary permeability. locAl Versus VAsculAr bubbles There is little reason to doubt that the localized pain in a joint is the result of local gas formation. Webb et al. (1944a; 1944b) and Ferris and Engel (1951) showed that gas could be seen in periarticular and perivascular tissue spaces. They also demonstrated a correlation between the presence of gas and the occurrence of localized pain. The effectiveness of local pressure in relieving such pain, such as by inflating a blood-pressure cuff, adds legitimacy to the hypothesis. importantly, DCi often oc- curs simultaneously in several sites and limb pain may distract physicians from more sinister neurological abnormalities (Fig- ure 2). While bubbles within tissues are clearly a cause for con- cern, significant numbers of venous gas emboli may be recorded without any clinical manifestations. in fact, precordial Doppler BalEstRa: CliniCal asPECts oF DCi ? 21 detection of bubbles in the right ventricle or pulmonary arteries is not considered to have significant positive predictive value for DCi. However, high degrees of bubbling are associated with an increased risk of developing symptoms. on the other hand, Brubakk et al. (1984) have not observed DCi symptoms in in- dividuals with no bubbles in the pulmonary artery and in the muscles of the thigh. nevertheless, nishi (1993) has reported that DCi is always accompanied by bubbles, if all monitoring sites are considered. Brubakk et al. (1991) have argued that Doppler detection, which is performed at intervals, may miss occasional bubbles and that the exceptional cases that have pre- sented with DCi symptoms in the absence of Doppler bubble signals, may have fallen in this category (table 1). biochemicAl eFFect oF VAsculAr bubbles gas bubbles affect cells and disrupt biochemical process- es as has been demonstrated by in vitro studies. Thorsen et al. (1986) showed that gas bubbles are associated with the aggrega- tion of thrombocytes. The degree of aggregation seems to be in- dependent of the gas content of the bubble, but rather is related to its surface properties. independently, Ward (1967) and Bergh et al. (1993) have also reported that gas bubbles activate comple- ment in vitro and that the response is unrelated to the content of the bubble. This also supports the hypothesis that the bioactive properties of bubbles are related to their surface characteristics. Ward (1967) and Bergh et al. (1993) also differentiated between ?sensitive? and ?nonsensitive? individuals, depending on the de- gree of complement activation in response to bubbles. The latter was also related to clinical manifestations of DCi. individuals with low C5a levels before dives produced many gas bubbles and a single air dive seemed to reduce C5a levels suggesting that gas bubbles may activate both C5a and C5a receptors. This phe- nomenon has been confirmed by stevens et al. (1993) in divers up to 14 hours after treatment for DCi. Complement activation (kilgore et al., 1994) triggers the activation of neutrophils and the formation of multiple membrane-attack complexes (maC) that eventually lead to cel- lular destruction. This also causes the leukocytes to adhere to the endothelium as they circulate over damaged endothelium. such neutrophil activation has been demonstrated during de- compression (Benestad et al., 1990). C5a activation may be related to some of the skin changes seen in DCi: erythema, edema, and infiltration of inflamma- tory cells (swerlick et al., 1988). another important effect of C5a is vasoconstriction and blood flow reduction (martin et al., 1988). if circulation of blood is restricted during decom- pression, gas elimination is similarly reduced leading to possible critical supersaturation local bubble formation. Post ischemic hyperemia is not seen, possibly due to C5a activation, leukocyte adherence, or even persisting vascular or perivascular bubbles (Bergh et al., 1993). vik et al. (1990) also observed that pulmonary changes in pigs following decompression were similar to those observed after complement activation. lungs exposed to significant amounts of bubbles for approximately 100 minutes after de- compression developed considerable leukocyte invasion. Com- plement activation was therefore considered to be the most im- portant mechanism for acute lung injury (Ward, 1967; Ward et al., 1995). Certain pulmonary function changes have been observed in divers. These include a reduction in carbon monoxide dif- fusion capacity and compliance (Thorsen et al., 1986). They are believed to support the growing evidence that inflammatory processes may follow decompression. in fact, the reduction in carbon monoxide diffusion capacity is rapid and is associated with the development of bubbles (Dujic et al., 1992; 1993). vik et al. (1990) consider the lungs to be a primary target organ for gas bubbles that are probably exposed to their effects in all decompressions. indeed, the concept of the lungs serving as a ?bubble trap? has been purported for many years, but we are only now looking at the impact that this function has on the ?filter? itself. although removal of bubbles by the lungs prevents more harmful distribution of bubbles to the arterial system, the process of doing so also has consequences for the lung. in addi- tion, the filtering mechanism is not foolproof: if the gas load on FIgURe 2. huge bubble found in an aviator in 1945 (Courtesy of R. D. vann). TaBle 1. Doppler grades (precordial) in different studies and DCI incidence (modiefied after Brubakk et al., 1991). Bubble grade 0 i-iv spencer and Johanson (1974) n 110 64 DCi inc (%) 1.0 22 gotoh and nashimoto (1977) n 64 88 DCi inc (%) 0 19 marroni and Zannini (1981) n 64 33 DCi inc (%) 0 9 nishi (1993) n 1265 331 DCi inc (%) 0.6 8 Brubakk et al. (1991) n 68 40 DCi inc (%) 1.5 7.5 22 ? tHE FUtURE oF Diving the lungs is large, the filtering capabilities of the lungs will be exceeded and gas will enter the arterial circulation. an increase in pulmonary artery pressure of only about 30% is considered sufficient to cause arterialization of venous gas bubbles Central nervous system alterations in DCi are believed to result from multiple mechanisms, including intra- and extra- vascular (tissue) bubbles (Francis et al., 1990). vascular bubbles do not seem to be an important pathophysiological feature of spinal cord DCi. in a group of 10 amateur and 10 professional divers, five of whom had neurological DCi, no changes could be seen (morild and mork, 1994). However, the same authors re- ported changes in the endothelial layer of the brain ventricles in a group of divers (morild and mork, 1994; mork et al., 1994). Brubakk (1994) has entertained the possibility that this may not so much be evidence of intravascular gas bubbles in the brain as it may be indicative of an increase in venous pressure due to venous gas embolism of the lung interfering with venous return. another possible explanation may be gas bubbles in the spinal fluid adhering to the lining of the ventricles and caus- ing changes in the adjacent endothelium. Chryssanthou et al. (1977) has indeed shown that animals exposed to decompres- sion show changes in the integrity of the blood-brain barrier, and Broman et al. (1966) has also confirmed that even short contact between gas bubbles and endothelium (i.e., 1-2 mins) leads to such changes. Further studies in rabbits have shown that bubble-endothelium contact causes endothelial damage and progressive reduction of cerebral blood flow and function (Helps and gorman, 1991). clINIcal maNIfESTaTIONS Of dcI There is ongoing controversy about the best way to clas- sify decompression disorders. Until 1990, these disorders were divided into DCs and agE. DCs was then divided into two broad categories based on the severity of symptoms and the as- sociated treatment regimens. However, today we recognize that certain forms of DCs may be the result of paradoxical or even frank agE. Therefore, in the application of the traditional clas- sification that follows, a modifier (DCs or DCi) is added to in- dicate where such pathophysiological ambiguity exists. Certain manifestations of decompression disorders are known never to be associated with arterial gas embolism and therefore can con- fidently be classified as decompression sickness. These include limb ?bends? and lymphatic DCs. time of symptom onset in cases (all manifestations): ? 50% occur within 30 minutes of surfacing; ? 85% occur within 1 hour of surfacing; ? 95% occur within 3 hours of surfacing; ? 1% are delayed more than 12 hours; and ? some symptoms have been reported to begin as late as 24 hours and more after surfacing and even longer if there is subsequent altitude exposure (e.g., flying or mountaineering). 1. TypE I dcI PAin only - dcs in recreational compressed-air diving the upper extremities are involved three times more often than the lower limbs. This is reversed for caisson workers and in commercial saturation diving. The pain can range from slight discomfort to a dull, deep, boring, or even unbearable pain. it is usually unaffected by movement of the joint and there may be overlying edema and regional numbness. lymPhAtic mAniFestAtions - dcs The lymphatic manifestations of DCs presumably result from obstruction of lymphatics by bubbles. The manifestations can include pain and swelling of lymph nodes, with lymph- edema of the tissues drained by the obstructed lymph nodes. new data are suggesting that normobaric oxygen may improve the flow of lymph and may assist in resolving inert gas bubbles contained within the lymphatic system or even to remove some tiny micronuclei that can behave like proteins and thus be cap- tured by the lymphatics (Balestra et al., 2004). oxygen precon- ditioning is a known factor to reduce venous gas emboli post diving, which may be explained by micronuclei elimination by the lymphatic system. some recent results show a reduction in post-diving bubble grades after a known boosting of lymphatic captation by whole body vibration (Figure 3); more evidence is needed to asses the hypothesis cutAneous mAniFestAtions - dci itching is common following decompression from dry chamber dives where the skin is in direct contact with the cham- ber atmosphere rather than with water. This condition, some- times called ?diver?s lice,? is thought to be the result of gas dis- solving directly into the skin and causing cutaneous irritation and the release of histamines with subsequent itchiness upon decompression. This is not a true or systemic form of DCi and does not require recompression. on the other hand, itchiness of the skin following a dive in which the skin was wet, is more likely to be true cutaneous DCi. note that some in-water dives are performed in drysuits. Under these conditions the skin is in direct contact with compressed gas and ?diver?s lice? may ap- pear. However, this is usually accompanied by some degree of skin rash or visible skin change. Cutis marmorata is a form of DCi which is thought to re- sult from a complex interaction between bubbles, venous con- gestion and the immune system. it usually manifests itself as bluish-red ?blotches,? frequently affecting the upper back and chest. Prominent linear purple markings are also frequently ob- served. These manifestations are a systemic form of DCi and suggest significant bubble formation that may also be affecting other areas of the body. as a result, prompt recompression is BalEstRa: CliniCal asPECts oF DCi ? 23 recommended, which usually leads to prompt resolution. This sign frequently heralds more serious forms of DCi, and there is a statistical association with PFo even if the physiopathological link is currently not fully understood. 2. TypE II dcI PulmonAry dci This is a syndrome usually presenting with a triad of symptoms: a) substernal pain that usually burns and progressively in- creases. initially, the pain may be noted only when cough- ing or with deep inspiration. over time the pain may be- come constant; b) Cough that is initially intermittent and provoked by ciga- rette smoking (Behnke?s sign). Paroxysms of coughing may become intractable; and c) Progressive respiratory and dyspnea. The manifestations of pulmonary DCi are believed to re- sult from the combined effects of gas emboli in the pulmonary artery and obstruction of the vascular supply to the bronchial mucosa. Untreated pulmonary DCi may be fatal. neurologicAl dci although the exact mechanisms of neurological DCi are not fully understood, they are believed to include embolism, au- tochthonous (i.e., spontaneous interstitial bubble formation), venous stasis, and immunological phenomena. These mecha- nisms have different latencies and show different responses to recompression. The neurological manifestations of DCi are therefore unpredictable, and any focal neurological symptom or sign may be a manifestation of its presence. any neurological abnormality following a dive should always be assumed to be of central origin and treated accordingly. cerebrAl dci Brain involvement in DCi appears to be especially com- mon in high altitude aviators (i.e., flying in excess of 25,000 feet in unpressurized aircraft). in this group, pulmonary venous gas embolism is also common and hypoxia and positive-pressure breathing may facilitate the transfer of bubbles or immunologi- cal products into the systemic circulation. not surprisingly, a migraine-like headache accompanied by visual disturbances is a common manifestation of DCi. in divers, brain involvement usually presents more overtly with stroke-like symptoms. Col- lapse with unconsciousness is a rare presentation of DCs but common in agE. if it does occur, it represents a very grave form of DCi. sPinAl cord dci Paraplegia is a ?classic? symptom of DCi in divers and clearly represents spinal cord involvement. Bladder paralysis with urinary retention and fecal incontinence frequently ac- company such paraplegia. Recent years have seen a decline in both the proportion and absolute number of cases of serious paralysis in recreational divers (from 13.4 percent in 1987 to only 2.9 percent in 1997). similarly, loss of consciousness has dropped from 7.4 percent to 3.9 percent during the same peri- od. The incidence of loss of bladder function has dropped from 2.2 percent to 0.4 percent during this period (Dan Diving ac- cident Reports). interestingly, the reduction in severe neuro- logical symptoms has not been balanced by a proportional rise in pain-only or skin manifestations. Rather, there has been an unexpected appearance of mild, ambiguous neurological mani- festations, such as paresthesia or tingling, which appear to re- spond well to oxygen administration and recompression. inner eAr dci audiovestibular DCi is a not an uncommon manifestation of Cns involvement. Usually both the cochlea and vestibular apparatus are involved and the presenting symptoms include tinnitus, deafness, vertigo, nausea, vomiting, and ataxia. nys- tagmus may be present on physical examination. it is not clear whether the situation depends predominantly on bubble for- mation in the perilymph or is due to embolization of the audi- tory vestibular inner ear DCi is a serious medical emergency and must be treated immediately to avoid permanent damage. since the nutrient arteries of the inner ear are very small, rapid reduction in bubble diameter, with immediate 100% oxygen FIgURe 3. Bubble score after 30 min. of whole-body vibration preconditioning 2 hours before the dive (25 m/25 min), n = 6. 24 ? tHE FUtURE oF Diving administration and prompt recompression is essential. shock - dci shock occasionally occurs in DCi and is usually associ- ated with serious pulmonary manifestations indicating a hyper- acute form of DCi. multiple mechanisms may contribute to the patho genesis of shock in DCi, including loss of vascular tone from spinal cord involvement, myocordial depression from hypoxemia and acidosis, pulmonary embolization, and hypovo- lemia due to increased capillary permeability resulting in loss of plasma water and hemoconcentration. bAck, AbdominAl, or chest PAin - dci Pain in these areas, in contrast to limb pain, should be con- sidered with great attention, as it frequently represents spinal cord involvement. extreme FAtigue - dci Fatigue, out of proportion from that normally occurring after a dive, has long been regarded as a serious manifestation of DCi. However, the biochemical and pathophysiological mecha- nism of this symptom are unknown. NEW dcI claSSIfIcaTION The simple classification into type i and type ii DCi im- plies that the different categories are well defined disease entities and that there is reasonable agreement about the classification (Brubakk, 1994; Brubakk and Eftedal, 2001). However, two separate studies (kemper et al., 1992; smith et al., 1992), showed considerable uncertainty between experts using the classical classification system. For instance, cerebral DCi could not in many cases be distinguished from arterial gas embolism or vestibular barotrauma. other studies have shown that solely articular symptoms are rare, as they are usually ac- companied by central nervous system symptoms (vann et al., 1993; Freiberger et al., 2002). Extreme fatigue can be classi- fied as a minor symptom but could also be a sign of subclini- cal pulmonary embolism (Hallenbeck et al., 1975). Francis and smith (1991) therefore suggested the currently widely adopted term ?DCi,? to include the two previously used definitions of decompression sickness and arterial gas embolism. They also suggested avoidance of the classifications of type i and type ii, while adopting instead a descriptive classification method ac- cording to the clinical manifestations (signs and/or symptoms) and their evolution in time. Using this classification scheme, a very high degree of concordance between different specialists was possible (Pollard et al., 1995). Francis and smith (1991) proposed the following Classification table for Decompression injuries (table 2), which is a useful guideline to correctly de- scribe the various possible manifestations of a decompression disorder. dcI TrEaTmENT air was nearly always used as the breathing gas and oxygen treatment was not really explored until Yarbrough and Behnke?s (1939) preliminary experiments. The early treatment theory was conceptually homeopathic in trying to decide how deep to take the injured diver. The original depth of the dive was used as a guide. For example, if decompression from a depth of 40 meters caused the symptoms, recompression to the same pres- sure should alleviate them. However, there were controversies as others thought that the situation of the diver should lead the decision and that the depth of relief should mark the ini- tial treatment pressure. still others argued that bubbles may be compressed but never disappear and should always be assumed TaBle 2. Classification of DCI symptoms. DEFinition? aBRUPt Evolving statiC onsEt timE? immEDiatE FiRst DaY DaYs to YEaRs PainFUl limb bend: limb Bend: fluctuating osteonecrosis Periarticular pain pain after dive PaREstHEtiC tingling or numbness, tingling or numbness, Recurrent or episodic after may herald spinal DCi may be combined with pain treatment; probably benign asYmPtomatiC skin changes or painless skin changes or painless asymptomatic osteonecrosis swelling swelling CEREBRal loss of consciousness, Hemiparesis, delirium, Chronic neuropsychological hemiplegia, ?air embolism? brainstem signs, vertigo changes sPinal girdle pain, loss of leg Waxing and waning weakness, Chronic gait and bladder movement and bladder control bladder dysfunction, sensory disturbances levels CEREBRo-sPinal Unconscious diver with Combined spinal and Combined cerebral and spinal spinal findings cerebral signs, varying disability; spinal predominates sYstEmiC Chokes; acute systemic Fatigue, rare visceral DCi Rare cases of aRDs and lung respiratory collapse damage l o C a l iZ a t io n s o m a t iC BalEstRa: CliniCal asPECts oF DCi ? 25 to be present. For these reasons the recommendation was to compress the patient to the depth of relief plus at least one at- mosphere. The rationale was that if a bubble became extremely small, surface tension may cause it to collapse and disappear. it was generally realized that bubbles remaining in the tissues and circulation would continue to take up inert gas, as more nitro- gen was absorbed during the recompression treatment. in the 1924 edition of the U.s. navy Diving manual, a suggested recompression treatment procedure was first pub- lished, but more than 50% of the treatments were unsuccessful. The U.s. navy published treatment tables again in 1942 with- out much improvement in the results. van Der aue and Behnke (1945) experimented with better treatment methods that resulted in the publication of the U.s. navy air Recompression tables i to iv that became the univer- sal standard for the next 20 years The outcome improvement over the previous approach was dramatic and over 90%. The principles of these tables were: ? Recompression to depth of relief plus at least one atmo- sphere. in practice, this meant going to a minimal depth of 30 meters; ? a maximum treatment depth of 50 meters. This depth was considered a good compromise between optimal recom- pression of any bubble while minimizing nitrogen narcosis risk and subsequent decompression; ? The use of a 12-hour stop at 9 meters before surfacing. Theoretically, this ?overnight soak? was intended to allow all the tissues to saturate or desaturate to the 9-meter level, from where, according to Haldane, direct decompression to the surface would be safe; and, ? The use of intermittent oxygen breathing during the last hours of treatment. These tables were at first considered very successful, with a failure rate on the initial recompression of only 6% in 1946. Despite their great length, from 6 hours 20 minutes for table i to 38 hours 11 minutes for table iv, they represented the only available therapeutical solution at the time. in 1963, however, the observed failure rate for serious symptoms cases was 46% on the initial recompression, and it became 47.1% in 1964. The reason for the failure of these tables, which had initially been so successful, was that more and more civilian recreational scuba diving cases were being treated. The civilians had often dived in total ignorance of decompression requirements, not to speak about the pretreatment intervals, which were significantly longer than with the military divers. goodman and Workman (1965) started investigating the use of oxygen at moderate depths (2.8 ata) for the treatment of DCi. oxygen treatment had first been suggested by Behnke and shaw (1937). at that time, however, the U.s. navy Bureau of medicine and surgery was concerned that oxygen breathing in the chamber may be dangerous and not ?sailor proof ? and that the risks of oxygen toxicity and fire were too great (kindwall, 1998). For this reason Behnke?s excellent results were ignored. End (1937) had also noted that treating DCi with compressed air was inefficient and had introduced oxygen breathing with excellent results in over 250 cases of DCi in compressed air workers (kindwall, 1998). The first 52 cases treated with the ?new? goodman and Workman oxygen schemes showed that at least 30 minutes at the maximum depth of 18 meters and a total oxygen breathing treatment time of 90 minutes could be considered ?adequate? treatment, allowing for a 3.6% failure rate. treatment schedules were lengthened to two hours and four hours of oxygen breathing, with air breathing intervals of 5 to 15 minutes, to avoid oxygen toxicity. The two- and four-hour schemes were called U.s. navy table 5 and 6, respectively, and were officially published in 1967. The most significant concep- tual difference, with respect to the previous approach, was the importance given to the time of relief, instead of the depth of relief (kindwall, 1998). Compression ceased to be limited to the scope of reducing the bubbles until they could disappear and then decompressing the diver according to a safe profile. The concept of a real ?therapeutical? treatment scheme was introduced where compression is just the vehicle for a therapeutical drug (oxygen) that exerts its treating action dur- ing the entire treatment schedule while also offering a double tool for the reduction of the offending gas bubbles, both by pressure and by a favorable diffusion gradient. serial treatment of DCi had been advocated by many, until, at a meeting of the north sea offshore diving groups at the Royal society of medicine in london in 1976, a consensus was reached that if the diver had residual symptoms after the initial treatment, daily hyperbaric oxygen treatment should be continued for at least two weeks or until the patient?s signs and symptoms had plateaued (Elliott et al., 1974a, b; Elliott and moon, 1993; Eke et al., 2000). ancillary pharmacological treatment to recompression be- gan to be emphasized in the late 60s and 1970s. in 1979, the Undersea medical society organized a workshop on the manage- ment of severe and complicated cases of DCi, where the impor- tance of hydration, steroids, heparin, aspirin, and other agents were discussed. several modifications to treatment schemes were then introduced, such as the Comex table 30, using mixed gas at a maximum pressure of four atmospheres, and the concept of saturation treatment. This was first introduced by miller et al. (1978) with saturation treatment to start at four atmospheres while the patient breathed oxygen at 0.35 to 0.5 bar. Currently, many different treatment protocols are in use, while the U.s. navy table 6 probably enjoys the most universal use. available evidence indicates that this table is adequate in the majority of cases where treatment is initiated immediately (Dan Diving accident Reports). Unfortunately, there is often considerable delay in initiating treatment, and many of the sec- ondary effects of the bubbles on blood and tissues become im- portant. kelleher et al. (1996) has shown that initial treatment is effective in only about 66% of the cases. leitch and Barnard (1982) and leitch (1985) have demonstrated, however, that 26 ? tHE FUtURE oF Diving none of the alternative proposed protocols, including satura- tion decompression, are superior to U.s. navy table 6 The severity of symptoms should not be the only variable considered. Heliox and trimix dives may differ from air dives due to the differences in partition coefficients of helium and nitrogen in the tissues (Brubakk et al., 1986). The use of differ- ent gas mixes, particularly heliox for the treatment of DCi, also for compressed air diving, is controversial. some reports seem to indicate that helium is beneficial. air bubbles in tissue have been observed to disappear faster from the spinal cord if heliox is used instead of pure oxygen at 1 ata, but the reverse is true at the pressure of 2.8 ata (Hyldegaard and madsen, 1989; Hyl- degaard et al., 1991; Hyldegaard and madsen, 1994). Brubakk (1994) believes, having shown an increased shunt in the lung and a reduction of gas elimination at increased oxygen tensions, that there is actually little benefit in using high oxygen tensions and that lower tensions may be more advantageous. over the years, many attempts have been made to improve the treatment of DCi with certain drugs, generally, with little success. some of these possibilities have not been sufficiently studied and may deserve further attention, such as the use of fluorocarbons, which have a higher solubility for nitrogen than plasma. lutz and Herrmann (1984) were able to substantially reduce the mortality of rats undergoing rapid decompression from 8 ata when fluorocarbon was infused after decompres- sion. a further point that deserves attention and study is com- plement activation and its effect on the leukocyte-endothelium adhesion, which appears to have a certain role in DCi and for which drugs could have a role. in 1994, the European Commit- tee for Hyperbaric medicine organized its first European Con- sensus Conference, where DCi was one of the topics. in 1996 a second, more specific Consensus Conference was organized, the theme of which was ?The treatment of Decompression ac- cidents in Recreational Diving.? Following both Conferences, and after extensive presentations by leading international ex- perts, the two international Juries formulated recommenda- tions that have since been adopted as the current standards for the definition and treatment of DCi in Europe. cONcluSION DCi is generally considered a rather benign condition, if adequate treatment is promptly started, with a success rate in excess of 80%. There is universal consensus that 100% oxygen should be administered immediately as the single most impor- tant first aid treatment of any DCi case related to surface-ori- ented diving and that rehydration is a valuable adjunct during diving first aid treatment in the field. Hyperbaric treatment should be started within the short- est possible timeframe from surfacing or from the onset of the first DCi signs and symptoms. Hyperbaric treatment tables using 100% oxygen at environmental pressures not exceeding 2.8 bars, with various depth/time profiles, demonstrate very good results in more than 80% of the treated cases. There is no significant evidence that any other therapeutical schemes provide better results and would therefore be preferable as the hyperbaric treatment of first choice for DCi related to surface- oriented diving. it is accepted by many specialists that the use of high pres- sure (generally 4 bars maximum) treatment tables using a gas mixture of 50% Helium and 50% oxygen may prove highly ef- fective and provide good results in the cases that do not quickly and satisfactorily respond to the standard low pressure hyper- baric oxygen treatment tables. although conclusive scientific evidence suggesting the use of any pharmacological treatment other than oxygen is missing, the administration of adjunctive fluid therapy is considered very important and generally recommended by diving-hyperbaric medicine specialists, whereas the role of other drugs, such as steroids and anticoagulants, although widely used without any apparent adverse effect, is still controversial. The continuation of hyperbaric oxygen therapy, combined with a specific rehabilitation protocol in neurological cases when the initial DCi treatment tables are not totally successful, is considered important and there is growing scientific evidence that it can significantly contribute to eventually achieving a bet- ter functional recovery. lITEraTurE cITEd Balestra, C., P. germonpre, t. snoeck, m. Ezquer, a. leduc, o. leduc, F. Willeput, a. marroni, R. Cali Corleo, and R.D. vann. 2004. nor- mobaric oxygen can enhance protein captation by the lymphatic sys- tem in healthy humans. Journal of Undersea and Hyperbaric Medicine 31:59?62. Behnke, a. R., and a. shaw. 1937. the use of oxygen in the treatment of compressed-air illness United States Naval Medical Bulletin 35:1?12. Benestad, H. B., i. B. Hersleth, H. Hardersen, and o. i. molvaer. 1990. Functional capacity of neutrophil granulocytes in deep-sea divers. Scandinavian Journal of Clinical Laboratory Investigation 50:9?18. Bergh, k., a. Hjelde, o. J. iversen, and a. o. Brubakk. 1993. variability over time of complement activation induced by air bubbles in human and rabbit sera. Journal of Applied Physiology 74:1811?1815. Bert, P., m. a. Hitchcock, and F. a. Hitchcock. 1943. Barometric pressure: re- searches in experimental physiology (1878). College Book Co., Columbus, ohio. Boyle, R. 1670. new pneumatical experiments about respiration. Philosophical Transactions 5:2011?2058. Broman, t., P. l. Branemark, B. Johansson, and o. steinwall. 1966. intra- vital and postmortem studies on air embolism damage of the blood- brain barrier tested with trypan blue. Acta Neurologica Scandinavica 42:146?152. Brubakk, a. o. 1994. Decompression illness. What we do know, what we don?t know. in First European Consensus Conference on Hyperbaric Medicine, ECHM, Wattel, F. and D. mathieu, eds. Universit? de lille, CRam. Brubakk, a. o., and o. Eftedal. 2001. Comparison of three different ultrasonic methods for quantification of intravascular gas bubbles. Journal of Undersea and Hyperbaric Medicine 28:131?136. Brubakk, a. o., J. Davies, and s. Daniels. 1984. monitoring of saturation decompression by combined use of ultrasonic scanning and Doppler techniques. annual scientific meeting abstracts of the Undersea and Hyperbaric medical society, may 29?June 4, 1984, san antonio, texas. BalEstRa: CliniCal asPECts oF DCi ? 27 Brubakk, a. o., R. Peterson, a. grip, B. Holand, J. onarheim, k. segadal, t. D. kunkle, and s. tonjum. 1986. gas bubbles in the circulation of divers after ascending excursions from 300 to 250 msw. Journal of Applied Physi- ology 60:45?51. Brubakk, a. o., H. torp, and B.a. angelsen. 1991. noninvasive evaluation of flow changes and gas bubbles in the circulation by combined use of color- flow-imaging and computer postprocessing. Acta Astronautica 23:311?319. Duji?, ?., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n. m. Petri, U. Wisl?ff, and a. o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555:637?642. Duji?, ?., D. Eterovi?, P. Denoble, g. krstacic, and J. tocilj. 1992. lung dif- fusing capacity in a hyperbaric environment: assessment by a rebreathing technique. British Journal of Industrial Medicine 49:254?259. Duji?, ?., D. Eterovi?, P. Denoble, g. krstacic, J. tocilj, and s. gosovic. 1993. Effect of a single air dive on pulmonary diffusing capacity in professional divers. Journal of Applied Physiology 74:55?61. Eke, a., P. Herman, J. B. Bassingthwaighte, g. m. Raymond, D. B. Percival, m. Cannon, i. Balla, and C. ikrenyi. 2000. Physiological time series: distin- guishing fractal noises from motions. Pfl?gers Archiv: European Journal of Physiology 439:403?415. Elliott, D. H., J. m. Hallenbeck, and a. a. Bove. 1974a. acute decompression sickness. Lancet 2:1193?1199. ???. 1974b. venous infarction of the spinal cord in decompression sickness. Journal of the Royal Naval Medical Service 60:66?71. Elliott, D. H., and R. E. moon. 1993. manifestation of the decompression dis- orders. in The Physiology and Medicine of Diving, Bennett, P. B., and D. H. Elliott, eds. pp. 481?505. saunders Company ltd., london. End, E. 1937. Rapid decompression following inhalation of Helium-oxygen mixtures under pressure. American Journal of Physiology 120:712?718. Ferris, E., and g. Engel. 1951. The clinical nature of high altitude decompres- sion sickness. in Decompression Sickness, Fulton, J., ed.. saunders, london. Francis, t. J. R., J. l. griffin, l. D. Homer, g. H. Pezeshkpour, a. J. Dutka, and E. t. Flynn. 1990. Bubble-induced dysfunction in acute spinal cord decompression sickness. Journal of Applied Physiology 68:1368?1375. Francis, t. J. R., and D. J. smith, eds. 1991. Describing decompression illness. 42nd Undersea and Hyperbaric medical society Workshop. UHms Pub- lication number 79(DECo) 5-15-91. Bethesda: Undersea and Hyperbaric medical society. Freiberger, J. J., P. J. Denoble, C. F. Pieper, D. m. Uguccioni, n. W. Pollock, and R. D. vann. 2002. The relative risk of decompression sickness during and after air travel following diving. Aviation, Space, and Environmental Medicine 73:980?984. gempp, E., J. E. Blatteau, J. m. Pontier, C. Balestra, and P. louge. 2009. Pre- ventive Effect of Pre-Dive Hydration on Bubble Formation in Divers. British Journal of Sports Medicine 43:224?228. gerriets, t., k. tetzlaff, t. liceni, C. schafer, B. Rosengarten, g. kopiske, C. algermissen, n. struck, and m. kaps. 2000. arteriovenous bubbles fol- lowing cold water sport dives: Relation to right-to-left shunting. Neurology 55:1741?1743. goodman, m. W., and R. D. Workman. 1965. minimal Recompression, oxy- gen Breathing approach to treatment of Decompression sickness in Divers and aviators. U.s. navy Experimental Diving Unit, nEDU Report 5-65, Washington, D.C. 49 pp. gotoh, Y., and i. nashimoto. 1977. [Decompression bubbles in caisson workers (author?s transl)]. Nippon Eiseigaku Zasshi 32:529?533. Hallenbeck, J. m., a. a. Bove, and D. H. Elliott. 1975. mechanisms underly- ing spinal cord damage in decompression sickness. Journal of Neurology 25:308?316. Helps, s. C., and D. F. gorman. 1991. air embolism of the brain in rabbits pretreated with mechlorethamine. Stroke 22:351?354. Hyldegaard, o., and J. madsen. 1989. influence of heliox, oxygen, and n 2 o-o 2 breathing on n 2 bubbles in adipose tissue. Undersea Biomedical Research 16:185?193. ???. 1994. Effect of air, heliox, and oxygen breathing on air bubbles in aqueous tissues in the rat. Journal of Undersea and Hyperbaric Medicine 21:413?424. Hyldegaard, o., m. moller, and J. madsen. 1991. Effect of He-o 2 , o 2 , and n 2 o-o 2 breathing on injected bubbles in spinal white matter. Undersea Biomedical Research 18:361?371. kelemen, g. 1983. temporal bone findings in cases of salt water drowning. An- nals of Otology, Rhinology, and Laryngology 92:134?136. kelleher, P. C., R. J. Pethybridge, and t. J. R. Francis. 1996. outcome of neu- rological decompression illness: development of a manifestation-based model. Aviation, Space, and Environmental Medicine 67:654?658. kemper, g. B., B. J. stegman, and a. a. Pilmanis. 1992. inconsistent classifica- tion and treatment of type i / type ii decompression sickness. Aviation, Space, and Environmental Medicine 63(5): 386?410. kilgore, k. s., g. s. Friedrichs, J. W. Homeister, and B. R. lucchesi. 1994. The complement system in myocardial ischaemia/reperfusion injury. Cardio- vascular Research 28:437?444. kindwall, E. P. 1998. Compressed air work. in The Physiology and Medicine of Diving, Bennett, P. B., and D. s. Elliott, eds. pp. 1?18. W. B. saunders Company ltd., london. kumar, k. v., J. m. Waligora, and D. s. Calkins. 1990. Threshold altitude resulting in decompression sickness. Aviation, Space, and Environmental Medicine 61:685?689. leitch, D. R. 1985. Complications of saturation diving. Journal of the Royal So- ciety of Medicine 78:634?637. leitch, D. R., and E. E. Barnard. 1982. observations on no-stop and repetitive air and oxynitrogen diving. Undersea Biomedical Research 9:113?129. lutz, J., and g. Herrmann. 1984. Perfluorochemicals as a treatment of decom- pression sickness in rats. Pfl?gers Archiv: European Journal of Physiology 401:174?177. marroni, a., R. Cali-Corleo, C. Balestra, P. longobardi, P. germonpre, E. voel- lm, m. Pieri, and R. Pepoli. 2001. The speed of ascent Dilemma: ?instant speed of ascent? or ?time to surface? - Which one really matters? in 27th Annual Meeting of The European Underwater and Baromedical Society on Diving and Hyperbaric Medicine, van laak, U., ed. pp. 74?78. gesellschaft f?r tauch und ?berdruckmedizin, Frankfurt, Hamburg, germany. marroni, a., and D. Zannini. 1981. Effects of variations in the ascending speed on the production of circulating gas bubbles after compressed-air diving. Minerva Medica 72:3567?3572. martin, s. E., D. E. Chenoweth, R. l. Engler, D. m. Roth, and J. C. long- hurst. 1988. C5a decreases regional coronary blood flow and myocardial function in pigs: implications for a granulocyte mechanism. Circulation Research 63:483?491. martini, s., a. o. Brubakk, and C. Bech. 1989. Effect of high pressure on meta- bolic response to cold in rats. Acta Physiologica Scandinavica 135:359?366. miller, J. n., l. Fagraeus, P. B. Bennett, D. H. Elliott, t. g. shields, and J. grimstad. 1978. nitrogen-oxygen saturation therapy in serious cases of compressed-air decompression sickness. Lancet 2:169?171. moir, E. W. 1896. tunneling by compressed air. Journal of the Society of Arts 44:567. morild, i., and s. J. mork. 1994. a neuropathologic study of the ependymoven- tricular surface in diver brains. Journal of Undersea and Hyperbaric Medi- cine 21:43?51. mork, s. J., i. morild, a. o. Brubakk, s. Eidsvik, and H. nyland. 1994. a histopathologic and immunocytochemical study of the spinal cord in ama- teur and professional divers. Journal of Undersea and Hyperbaric Medicine 21:391?402. nishi, R. 1993. Doppler and ultrasonic bubble detection. in The Physiology and Medicine of Diving, Bennett, P.B., and D.H. Elliott, eds. pp. 433?453. W.B. saunders, london, Uk. Pol, m., and m. Wattelle. 1854. m?moire sur les effets de la compression de l?air appliqu?e au creusement des puits ? Houille. Annales d?Hygiene Publique et de M?decine L?gale second series, 241. Pollard, g. W., P. l. marsh, C. E. Fife, l. R. smith, and R. D. vann. 1995. as- cent rate, post-dive exercise, and decompression sickness in the rat. Journal of Undersea and Hyperbaric Medicine 22:367?376. 28 ? tHE FUtURE oF Diving shinoda, s., Y. Hasegawa, s. kawasaki, n. tagawa, and H. iwata. 1997. mag- netic resonance imaging of osteonecrosis in divers: comparison with plain radiographs. Skeletal Radiology 26:354?359. smith, D. J., t. J. R. Francis, R. J. tehybridge, J. m. Wright, and J. J. sykes. 1992. Concordance: a problem with the current classification of diving disorders. Undersea and Biomedical Research 19 (suppl.), 40. spencer, m. P., and D. C. Johanson. 1974. investigation of new principles for human decompression schedules using the Doppler ultrasonic blood bub- ble detector. institute for Environmental medicine and Physiology, seat- tle, Washington. Usa. stevens, D. m., s. l. gartner, R. R. Pearson, E. t. Flynn, R. B. mink, D. H. Robinson, and a. J. Dutka. 1993. Complement activation during satura- tion diving. Journal of Undersea and Hyperbaric Medicine 20:279?288. swerlick, R. a., k. B. Yancey, and t. J. lawley. 1988. a direct in vivo compari- son of the inflammatory properties of human C5a and C5a des arg in hu- man skin. Journal of Immunology 140:2376?2381. Thorsen, E., a. o. Brubakk, t. ovstedal, m. Farstad, and H. Holmsen. 1986. a method for production of n 2 microbubbles in platelet-rich plasma in an aggregometer-like apparatus, and effect on the platelet density in vitro. Undersea Biomedical Research 13:271?288. triger, m. 1845. letter to monsieur arago. in Comptes rendus de l?Acad?mie des Sciences, pp. 445?449. Paris. van Der aue, o. E., W. a. White, R. Hayter, F. s. Brinton, R. J. kellar, and a. R. Behnke. 1945. Physiological factors underlying the prevention and treatment of decompression sickness. U. s. navy Experimental Diving Unit, nEDU Report 4?45, Washington, D.C. vann, R. D., P. Denoble, m. n. Emmerman, and k. s. Corson. 1993. Flying af- ter diving and decompression sickness. Aviation, Space, and Environmental Medicine 64:801?807. vik, a., a. o. Brubakk, t. R. Hennessy, B. m. Jenssen, m. Ekker, and s. a. slordahl. 1990. venous air embolism in swine: transport of gas bub- bles through the pulmonary circulation. Journal of Applied Physiology 69:237?244. Ward, C. l. 1967. scuba diving - biologic and physical aspects. Aeromedical Reviews 1:1?25. Ward, R., D. Jones, and E. F. Haponik. 1995. Paradoxical embolism. an under recognized problem. Chest 108:549?558. Webb, J., g. Engel, and J. Romano. 1944a. The mechanism of pain in aviators: ?Bends.? Journal of Clinical Investigation 23:934?935. Webb, J., E. Ferris, and g. Engel. 1944b. Radiographic studies of the knee dur- ing bends, pp. 305. U.s. nRC Committee on aviation medicine report, Washington, D.C. Wilmshurst, P., and k. Ross. 1998. Dysbaric osteonecrosis of the shoulder in a sport scuba diver. British Journal of Sports Medicine 32:344?345. Wisl?ff, U., and a. o. Brubakk. 2001. aerobic endurance training reduces bub- ble formation and increases survival in rats exposed to hyperbaric pressure. Journal of Physiology 537:607?611. Wisl?ff, U., R. s. Richardson, and a.o. Brubakk. 2003. nos inhibition in- creases bubble formation and reduces survival in sedentary but not exer- cised rats. Journal of Physiology 546:577?582. ???. 2004. Exercise and nitric oxide prevent bubble formation: a novel ap- proach to the prevention of decompression sickness? Journal of Physiology 555:825?829. Yarbrough, o. D., and a. R. Behnke. 1939. The treatment of compressed air illness utilising oxygen. Journal of Industrial Hygiene and Toxicology 21:213?218. backgrOuNd Decompression sickness (DCs) is thought to arise from intracorporeal bubble formation during reduction in ambient pressure (decompression). During un-derwater compressed gas diving, alveolar inert gas partial pressures change as a result of the change in total pressure and blood and tissues take up inert gas in accord with Henry?s law. During subsequent ascent to sea level, the ambient pressure may drop below the sum of the partial pressures of all gases dissolved in tissue. This state, referred to as supersaturation, is a necessary condition for bubbles to form and grow from the excess dissolved gas. sufficient bubble formation may result in DCs from as yet unidentified pathophysiological mechanisms. to minimize the risk of DCs, dives are conducted according to depth/time/breathing gas decompression schedules derived from decompression algorithms that implicitly (gas content models) or explicitly (bub- ble models) minimize bubble formation through controlling tissue gas supersaturation. Haldane developed the first practical decompression algorithm (Boycott et al., 1908), and a century later this gas content approach remains the most prevalent form of de- compression algorithm. The key features of the Haldane decompression approach are illustrated in Figures 1 and 2. laNdScapE aT ThE TImE Of haldaNE DCs in humans was first described by triger in 1845, and by the time of Hal- dane the hypothesis was well established that DCs resulted from bubble formation from excess dissolved nitrogen taken up while breathing compressed air pressure (see Bert, 1878, for historical summary and relevant experiments). it was also known that the risk of DCs could be minimized by limiting time under pressure or using a slow decompression. Zuntz and von schrotter, applying the Fick principle, had developed an expression for nitrogen uptake and washout, in which body ? alveolar nitrogen aBSTRaCT. to minimize the risk of decompression sickness, dives are conducted according to depth/time/breathing gas decompression schedules derived from algorithms that minimize bubble formation through controlling tissue gas supersaturation. John scott Haldane developed the first practical decompression algorithm in 1908. Haldane?s decompression model featured multiple par- allel perfusion-limited compartments, limits of tolerable supersaturation using a 2:1 ratio, and a maximized tissue-alveolar gas partial pressure gradient for inert gas washout. more recent bubble models may skew decompression time towards deeper stops than gas content models. gas content models that are direct descendants of the original Haldane model remain the most prevalent ap- proach to decompression. Navy Experimental Diving Unit, 321 Bullfinch Road, Panama City, Florida, 32407-7015, USA. Haldane still Rules! David J. Doolette 30 ? tHE FUtURE oF Diving partial pressure difference declines exponentially with approxi- mately an 8-minute half-time (see kety, 1951, for review). on this basis, von schrotter and colleagues hypothesized that linear decompression at 20 minutes per atmosphere would allow suf- ficient washout of nitrogen to avoid DCs. ThrEE kEy fEaTurES Of haldaNE?S mOdEl Haldane?s decompression model provided the following three keys advanced in decompression theory: 1. multiPle PArAllel comPArtments Haldane modeled the uptake and washout of nitrogen in the body using a collection of parallel perfusion-limited com- partments in which gas uptake and washout are symmetrical and each compartment is characterized by a different half-time. Differing half-times were based on the premise that different body tissues receive variable blood flows and have differential solubilities for nitrogen. His slowest half-time was based on observations and experiments indicating that the risk of DCs increases with exposure duration for many hours, and his fastest half-time was in accord with experimental measurement of the rate of saturation of urine with nitrogen (Hill and greenwood, 1907). 2. tolerAble suPersAturAtion observations of DCs in caisson workers and Haldane and colleagues? experimental work with goats indicated that a com- pressed air exposure of greater than 2 atm abs was required to provoke DCs upon return to 1 atm abs ? suggesting that some degree of supersaturation can be tolerated without developing DCs. Without then available evidence to the contrary, Haldane hypothesized that bubbles did not form unless this tolerable su- persaturation was exceeded. Contradicting his own premise, he further hypothesized that a decompression from, for instance, 6 to 3 atm abs would be equally as safe as decompression from 2 to 1 atm abs because the different amounts of supersatura- tion represented the same potential volume of gas bubbles ac- cording to Boyle?s law. Haldane used a ?2:1 tissue ratio? (ratio of maximum allowed tissue-dissolved gas pressure to ambient pressure) to define safe ascent depths. This ?2:1 tissue ratio? allowed a greater tolerable supersaturation at depth (Fig. 2). FIgURe 1. ambient pressure (solid line) versus time and the cor- responding total gas pressures (?Ptisinert) in two compartments (half-times of 5 and 20 minutes) for an air-breathing dive. Classi- cal decompression is scheduled to keep dissolved gas pressure in k modeled compartments less than or equal to a maximum permissi- ble value, Ptis.maxk???akPamb?+?bk. For simplicity, the above figure shows only two compartments and sets a = 1 and b = 0, so that the ?safe ascent depth,? P amb.min = max((Ptisk ? b)/a)), is equal to max(Ptisk). In haldane and colleagues algorithm, compartment half-times were 5, 10, 20, 40, and 75. The haldane ?2:1 tissue ratio? arises from treating air as an inert gas and setting the safe ascent depth as P amb = max(Ptisk/2). By convention, decompression stops are taken at in- crements of 10 fsw (0.3 atm abs) deeper than sea level. FIgURe 2. Tolerable supersaturation. although decompression algorithms use a maximum permissible tissue tension (Ptis.max) to control ascent to each decompression stop, it is informative to examine the tolerable supersaturation (Ptis.max? -?Pamb.stop). For a 40 minute half-time compartment, the thick line shows the relation- ship for the haldane 2:1 tissue ratio; the thin, solid line shows the relationship for the Workman m-values (Workman, 1965); and the dashed line shows the relationship for the Thalmann vval-18 m-values (Thalmann, 1984). Note that gas content model develop- ment during the 20th century resulted in progressively less toler- able supersaturation at depth, which results in deeper and longer decompression stops. DoolEttE: HalDanE RUlEs! ? 31 can provide better description of time of symptom onset than gas content models (gerth and vann, 1997). also, outcome of some unconventional dive profiles may be better predicted by bubble models. For instance, intermittent recompression dur- ing decompression reduces bubble formation (mollerlokken et al., 2007), something that is predicted by bubble models (gern- hardt, 1991) but not gas content models. cONcluSIONS Despite some potential advantages of bubble models and some deficiencies in Haldane?s original model, gas content models that are direct descendants of the original Haldane model remain the most prevalent approach to decompression. in addition, the general shape of decompression stop distribu- tion promoted by Haldane is demonstrated to be the most ef- ficient for decompression from typical dives. Haldane still rules! lITEraTurE cITEd Bert, P. 1878 La Pression barom?trique, recherches de physiologie exp?rimentale. Paris: g. masson, ?diteur. 1168 pp. Boycott, a.E., g.C.C. Damant, and J.s. Haldane. 1908. The prevention of com- pressed-air illness. Journal of Hygiene (Cambridge) 8:342?443. gernhardt, m.l. 1991. Development and evaluation of a decompression stress index based on tissue bubble dynamics. Ph.D. Dissertation, University of Haldane?s experimental evidence supporting a greater tolerable supersaturation at depth was slim, consisting primarily of 10 goats that were decompressed on one occasion from 4.5 to 1 atm abs with 8/10 DCs and on another occasion decompressed from 6.1 to 2.6 atm abs with no DCs. However, in the later experiments goats were only held at 2.6 atm abs for one hour and 3/10 suffered DCs following subsequent decompression to 1 atm abs. Development of gas-content models in conjunction with man-trials subsequent to Haldane?s work has tended to limit the extent to which the tolerable supersaturation increases with depth (Fig. 2), which results in deeper and longer decom- pression stops, but all define a safe ascent depth based on some degree of tolerable supersaturation. 3. mAximize grAdient For wAshout ? the contentious issue The classical shape of decompression schedules - initial rapid decompression that slows as the surface is approached - follows from the previous two features of Haldane?s model. Un- der such models, the most efficient decompression is to effect the maximum tolerable decompression to maximize the tissue ? alveolar gas partial pressure gradient for inert gas washout. The decompression is slowed as successively slower half-time compartments govern the safe ascent depth (Fig. 1). This shape of decompression was shown to be more efficient than linear decompression by Haldane and colleagues. bubblE mOdElS Bubble decompression models (models that explicitly limit the formation and growth of bubbles) have been devel- oped since the time of Haldane (e.g., Hills, 1966; Yount and Hoffman, 1986; gernhardt, 1991; gerth and vann, 1997). Bubble models may prescribe decompression schedules that depart substantially from the classical shape described above. since the publication of Haldane and colleagues? classical work, it has been firmly established that bubbles often form even dur- ing safe decompression. a large decompression and accompa- nying supersaturation increases the probability/rate of bubble nucleation and bubble growth. Balancing these effects against maximing the tissue ? alveolar gas partial pressure gradient for inert gas washout, bubble models may prescribe deeper initial decompression stops or skew decompression time towards deep- er stops than gas content models. only recently (gerth et al., 2009) has such deep skew of decompression time been directly compared to classically shaped decompression and the latter found to be more efficient (Fig. 3). This is not to say that bubble models are without advan- tage. Bubble models provide better descriptions of certain as- pects of decompression than gas content models. For instance, because the peak supersaturation occurs immediately following ascent but peak bubble volume can be delayed, bubble models FIgURe 3. a recent manned evaluation of two decompression schedules that differed only in the depth distribution of decom- pression stop time (gerth et al., 2009). a gas content model decom- pression schedule (solid line) was compared to a schedule of equal length predicted by a bubble model (dashed line). Number of DCS / number of dives are indicated next to the schedules. The great- er incidence of DCS was observed on the bubble model schedule which had deeper initial decompression stops. The deeper initial decompression stops must result in less initial bubble formation, but to explain greater incidence of DCS, also in less efficient gas washout and subsequent bubble growth. 32 ? tHE FUtURE oF Diving Pennsylvania. 328 pp. gerth, W. a., and R. D. vann. 1997. Probabilistic gas and bubble dynamics models of decompression sickness occurrence in air and n2-o2 diving. Journal of Undersea and Hyperbaric Medicine 24:275?292. gerth, W. a., D. J. Doolette, and k. a. gault. 2009. Deep stops and their effica- cy in decompression: U.s. navy research. in Bennett, P. B., B. R. Wienke, and s. J. mitchell, eds. Decompression and the deep stop, 24-25 June 2008. Durham: Undersea and Hyperbaric medical society. pp. 165?185. Hill, l., and m. greenwood. 1907. The influence of increased barometric pres- sure on man.-ii. Proceedings of the Royal Society of London B - Biological Sciences 79:21-27. Hills, B. a. 1966. a thermodynamic and kinetic approach to decompression sickness. Doctoral Thesis, the University of adelaide, australia. 370 pp. kety, s. s. 1951. The theory and applications of the exchange of inert gas at the lungs and tissues. Pharmacological Reviews 3:1?40. m?llerl?kken, a., C. gutvik, v.J. Berge, a. Jorgensen, a. loset, and a. o. Brubakk. 2007. Recompression during decompression and effects on bubble formation in the pig. Aviation, Space, and Environmental Medicine 78:557?560. Thalmann, E. D. 1984. Phase ii testing of decompression algorithms for use in the U.s. navy underwater decompression computer. nEDU Report no.1- 84. Washington, DC: navy Experimental Diving Unit. 134 pp. Workman, R. D. 1965. Calculation of decompression schedules for nitrogen- oxygen and helium-oxygen dives. nEDU Report no. 6-65. Washington, DC: navy Experimental Diving Unit. Yount, D. E., and D. C. Hoffman. 1986. on the use of a bubble formation model to calculate diving tables. Aviation, Space, and Environmental Medi- cine 57:149?156. INTrOducTION Pills and capsules to solve health problems are ubiquitous. What about swallow-ing a capsule to reduce the risk of DCs? What indeed might we put into a cap-sule that would have this benefit? By now, it is generally agreed among DCs researchers that we have all been missing something critical in the sequence of events starting with gas supersaturation in blood or other tissues, progressing to bubbles, and culminating in manifestations of DCs. Haldane was of course correct that most of DCs risk can be eliminated by establishing and following rules for compression and decompression pressure and time sequences in order to manage inert gas loads and fluxes (Boycott et al., 1908). But there is a residual component of DCs risk that defies these time at depth rules (Weathersby et al., 1984), suggesting that there is more that we need to know than gas burdens alone. The same conclusion is drawn from the recogni- tion that almost all excursions in barometric pressure lead to some degree of bubbling, but there is a poor correlation between bubbling and subsequent DCs incidence and manifestations (Eftedal et al., 2007), until the bubbling reaches extraordinary levels. many lines of evidence suggest that there is a variable biological response to bubbles that may include a cascade of inflammatory activations (little and Butler, 2008; ny- quist et al., 2007). it may thus be possible to reduce the risk of DCs by oral treatments aBSTRaCT. Haldane envisioned that decompression sickness (DCs) risk is eliminated by obeying rules for time at depth in order to manage inert gas fluxes. But new millennium divers want more: longer dives, faster decompressions, shorter surface intervals between dives, and lower residual risk of DCs that resists management by time at depth. Can DCs risk be further reduced, or dive pro- files safely stretched, by something provided in an oral capsule? if we knew the chain of biological events starting with bubbles and leading to DCs, then we might find a pharmacological approach to blocking that chain. While there is hope for someone identifying such an approach, we currently do not have one. alternatively, we could place inside an oral capsule the enzymatic ability to remove some of the gas taken up by the body during the hyperbaric exposure. Biochemical decompression has been demonstrated in small and large animal models with H 2 as the diluent to o 2 , in a breath- ing mixture designed for ultra-deep (>350 m) diving. Hydrogen-metabolizing microbes (the archae Methanobrevibacter smithii) are placed in the large intestine where they create a chemical scrubber unit. The metabolism of M. smithii converts H 2 and Co 2 to methane and water. The release of one molecule of methane for every four molecules of H 2 consumed provides a non-invasive marker of the metabolic rate. By gas chromatographic sampling of the environment around the dive subject for methane, we have demonstrated that removing approximately 5-10% of the body burden of H 2 reduces the risk of DCs by approximately 50%. nitrogen biochemical decompression would be far more useful but far more difficult to implement, both biologically and energetically. We will need a new millennium Haldane to make the capsule that lowers DCs risk from air dives a reality, whatever it may contain. Susan R. Kayar, Associate, Booz Allen Hamilton, Consulting Senior Scientist, 5947 Valerian Lane, Rockville, MD 20852, USA. Biochemical approach to Decompression Susan R. Kayar 34 ? tHE FUtURE oF Diving of anti-inflammatory medications. However, highly successful pharmaceuticals of this nature have yet to be identified (little and Butler, 2008; montcalm-smith et al., 2007). There is another potential approach to lowering the risk of DCs by means of swallowing a capsule. This approach, which we have called biochemical decompression (kayar et al., 1998; Fahlman et al., 2001; kayar and Fahlman, 2001; kayar et al., 2001), involves placing in the capsule the enzymatic capacity, provided by live microbes, to remove some of the inert gas load in the body. The capsule travels to the large intestine where it releases the microbes. These microbes remove from their sur- rounding environment, and then from the blood passing through the walls of the intestines, molecules of gas that are inert to the diver but metabolites to the microbe. Removing a small but critical fraction of the inert gas load during the dive lowers the subsequent risk of DCs. The feasibility of biochemi- cal decompression has been demonstrated in a rodent and a swine model, using ultra-deep diving gas mixtures containing H 2 as the diluent to o 2 . We present here a short description of the demonstration of H 2 biochemical decompression, and a discussion of its place in DCs research and the future of div- ing. other pharmacological approaches may ultimately be more practical and far easier to attain in the quest for a capsule or pill against DCs. But the study of H 2 biochemical decompression, useful in its own right for ultra-deep H 2 dives, had important lessons to teach about DCs and inert gas loads, even if it did not lead to a bottle of capsules in a dive store or a navy dive locker. maTErIalS aNd mEThOdS microbes used in h2 biochemicAl decomPression The microbes used to eliminate H 2 from inside the rats and pigs were the methanogenic archae, Methanobrevibacter smithii, strain Ps. The M. smithii were obtained from the stock cul- ture collection of the Wadsworth Center for laboratories and Research, new York state Department of Health. Culture and handling of M. smithii are described in detail elsewhere (kayar et al., 1998). The metabolism of M. smithii converts: 4 H 2 + Co 2 ? CH 4 + 2 H 2 o in which the H 2 breathed by the experimental animal combines with carbon dioxide released by the aerobic metabolism of the animal to form water and methane; this process generates energy for the microbes. The methane released by M. smithii becomes the tracer gas for the metabolic rate of the microbes, which can be detected noninvasively by sampling the atmosphere around the experimental animal by gas chromatography. although we envision that M. smithii would be provided to people via oral capsules, for expedience in these experiments with animals, cultures of M. smithii were surgically injected into their intestines as described below. exPeriments with rAts These experiments have been described in detail elsewhere (kayar et al., 1998). male sprague-Dawley rats (Rattus norvegi- cus, body mass range 215-324 g) were used for all experiments. The rats were housed before experiments in an accredited ani- mal care facility with ad libitum access to food and water. all procedures were approved by an animal Care and Use Commit- tee and were conducted according to the principles presented in the Guide for the Care and Use of Laboratory Animals (national Research Council, 1996). two milliliters of the M. smithii cultures were injected directly into the upper end of the large intestine under sterile surgical conditions. a total of 63 rats received injections of M. smithii, 40 rats underwent the same surgical procedure but re- ceived 2-ml injections of 10% sodium bicarbonate in the intes- tine, and 60 rats served as untreated controls. For each of the dive experiments to test for M. smithii ac- tivity, four or five rats were placed together in a clear box inside a large hyperbaric chamber (5.6 m3 internal volume). a stream of chamber gas passed continuously through the animals? box and was sampled by a gas chromatograph every 12 min for He, H 2 , o 2 , and CH 4 content. The chamber was pressurized with helium to an absolute pressure of 11 atm for 1 h. (This initial pressurization with He, prior to introducing H 2 , is necessary to create a gas mix in the chamber that has the 0.2 atm Po 2 that is breathable to the animal, but below the explosivity limit of less than 5% o 2 in H 2 ). The helium in the chamber was then replaced with H 2 at constant total pressure to a final concentra- tion of 85 ? 92% H 2 , while adding o 2 as needed to maintain 2-4% o 2 ; this flushing procedure was completed in ~ 2 h. The chamber was further pressurized with H 2 to a total pressure of 23.7 atm (91-98% H 2 , 2% o 2 , balance He) and maintained at this pressure for 2.5 h. The chamber was then depressurized to 11 atm at 0.2 ? 0.45 atm ? min-1, which was intended to be slow enough to be unlikely to cause DCs in the rats (lillo et al., 1997). after 45 min at 11 atm, the H 2 in the chamber was replaced with He at constant total pressure. to reduce the H 2 concentration to < 4% (to remain below the explosivity limit for H 2 in combination with air) typically required 2 ? 2.5 h of He flushing. The chamber was then further depressurized at 0.1 ?0.25 atm ? min-1 until the chamber had returned to 1 atm. The animals were removed from the chamber, observed for signs of DCs (which were rare), and left in cages with food and water overnight for further observation after 12 h. a second dive procedure was used to test for DCs risk in rats. in this procedure, we used some animals that appeared in normal health after 12 h in air following the dive to measure M. smithii activity (19 rats with M. smithii and 18 surgical con- trols), and also animals that were either treated with M. smithii (n = 20), surgical controls (n = 20), or untreated controls (n = 50) that had not previously been compressed. For this dive, a small chamber (140 l internal volume) was pressurized with He to 10 atm. The chamber was flushed with a mixture of 2% kaYaR: BioCHEmiCal aPPRoaCH to DEComPREssion ? 35 on which the animal stood or reclined throughout the dive) at 5-min intervals, interspersed with 5-min rest intervals. The ani- mal was examined for its ability to stand and walk with normal gait for up to 1 h or until clear signs of DCs were identified by at least three observers (inability to stand, difficulty maintaining normal posture, failure to right after falling, labored breathing). animals were quickly euthanized in the chamber by asphyxi- ation with He as soon as DCs was diagnosed or the observa- tion hour had passed without DCs. The chamber was further flushed with He near 1 atm to remove remaining H 2 before opening the chamber door. rESulTS exPeriments with rAts The release rate of methane from rats treated with M. smithii is presented in Figure 1. methane was detected in the atmosphere around the treated rats from the beginning of the dive as animals breathed He, due to native intestinal microbes producing H 2 and supplying it as a metabolic substrate to the M. smithii, even before H 2 was introduced into the chamber. methane release rate increased as the chamber was pressurized with increasing amounts of H 2 , and decreased as H 2 pressure fell during chamber depressurization. notably, there was a spike in methane release rate at the beginning of depressurization, which may reflect a period of supersaturation and release of bubbles of H 2 in the intestine that gave more rapid access to H 2 for the metabolism of M. smithii. o 2 , 98% H 2 until He was less than 10%. The chamber was further pressurized to 23.7 atm and a final gas composition of 2% o 2 , 94% H 2 , 4% He. The rats remained at this pressure for 20 min, which was sufficient to saturate them with H 2 (lillo et al., 1997). The rats were then decompressed within 36-40 s to a pressure of 10.8 atm and observed for signs of DCs over the next 30 min. Throughout their stay in the chamber, the animals walked inside a slowly rotating drum mill of wire mesh; DCs was diagnosed as a change of gait or inability to walk inside the mill for at least 1 min, as agreed upon by two or three observers. at the end of the observation period, all animals were eutha- nized in the chamber by introducing 2 atm Co 2 . exPeriments with Pigs Pigs (Sus scrofa, male Yorkshires, 19-22 kg body mass) were housed before experiments in an accredited animal care facility, had ad libitum access to water, and were fed once daily (laboratory animal chow, 2% of body weight). all procedures were approved by an animal Care and Use Committee and conducted according to the principles presented in the Guide for the Care and Use of Laboratory Animals (national Research Council, 1996). Experiments with pigs were described in detail elsewhere (kayar et al., 2001). The treated pigs (n = 16) received injec- tions of M. smithii in the caecum and upper end of the large intestine under sterile surgery. The volume of M. smithii cul- ture injected in an animal varied with the volume and activ- ity of microbes that were successfully cultured that week. This volume ranged from 12 to 116 ml and represented an activity range from 200 ? 2,200 ?mol CH 4 ? min-1. surgical control animals (n = 10) underwent the same sterile surgical procedures as treated animals but received injections into the intestine of sterile saline solution. Untreated control animals (n = 10) were also studied. all animals were trained to walk on a treadmill and acclimated to being in the dive chamber at 1 atm prior to dive experiments. a single dive protocol was followed with pigs for measuring methane release rate and DCs incidence. For each dive, one pig was placed inside a large hyperbaric chamber (5.7 m3 internal volume). The pig could stand, walk a distance of several body lengths, or recline at will, and could access containers of pellet food and water. a stream of gas passed continuously from the chamber to a gas chromatograph for automatic sampling every 12 min for He, H 2 , o 2 , and CH 4 content. The chamber was pressurized with He to an absolute pressure of 11 atm, adding o 2 as necessary to maintain approximately 0.2 - 0.4 atm Po 2 . The chamber was then flushed with H 2 at constant total pressure to a concentration of 60-75% H 2 , 2 - 4% o 2 , balance He. The chamber was further pressurized with H 2 and o 2 to a final pres- sure of 24 atm absolute pressure (0.3 ? 0.5 atm Po 2 ), which was maintained for 3 h. The chamber was then depressurized at a rate of 0.9 atm.min-1 to 11 atm, while observing the animal closely. after arrival at 11 atm, the animal was made to stand and walk on a treadmill (which formed the floor of the chamber FIgURe 1. average methane release rate per rat (?) from 5 rats with Methanobrevibacter smithii (50 ?mol h 2 uptake ? min-1 ac- tivity) in their large intestines as they breathed h 2 (?) at various pressures (Kayar et al., 1998). 36 ? tHE FUtURE oF Diving Rats treated with M. smithii and tested immediately for DCs risk had a 25% incidence of DCs (Fig. 2a). This was sig- nificantly lower (P < 0.01, Chi-squared one-sided test) than the incidence of DCs for untreated rats of 56% and surgical con- trol rats of 65%. The incidence of DCs did not differ for these two control groups compared to each other (P = 0.34, Fisher?s exact test). a similar result was obtained for the rats that first underwent the dive exposure to measure methane release rate, and were then decompressed and allowed to stay at one atmo- sphere over night before their second dive to test for DCs inci- dence (Fig. 2b). animals treated with M. smithii had a 36.8% DCs inci- dence, which was significantly lower (P < 0.05 Chi squared one-sided test) than the 55.6% incidence of the surgical con- trol rats that also underwent the two dive procedures. exPeriments with Pigs Pigs, as in the experiments with rats, released methane from the beginning of the dive protocol, with methane release rate increasing as the chamber was pressurized with H 2 (Fig. 3). methane release rate continued to climb following the decom- pression to 11 atm, which we again speculate is due to bubbles of H 2 in the intestine allowing more rapid access of M. smithii to this metabolic substrate during tissue supersaturation with H 2 . since the dive with pigs was terminal at 11 atm (for reasons related to gas safety and economy), there was no opportunity to observe a subsequent fall in methane release rate from pigs as the chamber was depressurized to 1 atm, as there had been with rats (compare the later time periods of Fig. 3 to Fig. 1). methane release rate from pigs was significantly correlated with activity of M. smithii injected into them (P < 0.001, R = 0.70, slope = 0.046, least squares linear regression; Fig. 4). since pigs, unlike rats but similar to humans, have a native intestinal flora that includes M. smithii, there was variable but detectable methanogenic activity even in the surgical control and untreated control pigs (Fig. 4). addition of M. smithii to pigs significantly lowered their incidence of DCs during decompression. The treated animals had a DCs incidence of 44%, compared to 90% of untreated controls and 70% of surgical controls (P < 0.05, Chi-squared two-tailed test comparing treated animals to untreated control and surgical control groups combined; untreated controls and surgical controls not different from each other, P > 0.50, Chi- squared two-tailed test; Fig. 5). FIgURe 4. activity of Methanobrevibacter smithii injected into the intestines of pigs versus mean rate of release of methane from these animals during a 3-h exposure to hyperbaric h 2 (21.7 atm ph 2 ). De- compression sickness (DCS) outcome is indicated for each animal. The line represents least squares linear regression (y = 40.1 + 0.046X; R = 0.70, p < 0.001; Kayar et al., 2001). FIgURe 3. methane release rate data from one pig that had re- ceived surgical injections of Methanobrevibacter smithii (795 ?mol Ch 4 ? min-1 activity) 1?2 h prior to an exposure to hyperbaric h 2 (Kayar et al., 2001). FIgURe 2. Incidence of decompression sickness for rats with and without Methanobrevibacter smithii placed in their intestines, fol- lowing a standardized compression and decompression sequence (Kayar et al., 1998). (a) animals with no treatment (controls), and with a bicarbonate solution (surgical controls) or M. smithii (with methanogen) placed surgically in their intestines and tested within 1?3 h of surgery. (B) animals with surgically implanted bicarbon- ate solution or M. smithii tested for DCS ~24 h after the surgery and 12 h after completion of an exposure to hyperbaric h 2 to mea- sure methane release rate. error bars represent 95% binomial con- fidence limits. kaYaR: BioCHEmiCal aPPRoaCH to DEComPREssion ? 37 dIScuSSION The increasing evolution of methane from M. smithii? treated rats (Fig. 1) and pigs (Fig. 3) during the course of a hyperbaric H 2 exposure demonstrated that these methanogenic microbes did indeed extract H 2 from the blood of the animals while inside the intestines. We envision that gradients for H 2 in arterial and venous bloodstreams and across the lung can be modeled as illustrated in Figure 6 (Fahlman et al., 2001). a sink for H 2 anywhere in the animal, particularly in an organ with a large surface area and excellent perfusion such as the intestine, can extract H 2 down a pressure gradient from the arterial blood perfusing the intestine. This sink would result in lower mixed venous return of H 2 to the lungs. During compression, the sink would prevent the animal from reaching full saturation with H 2 . During decompression, the lower mixed venous PH 2 would create slightly lower alveolar PH 2 . since alveolar gas exchange with the environment is never 100% efficient, the slightly lower alveolar PH 2 would allow arterial PH 2 to be lower with ver- sus without M. smithii, thus accelerating the washout of H 2 throughout the animal. We believe this process of gas elimina- tion by M. smithii during biochemical decompression can be described in terms similar to those for a chemical scrubber unit in a mechanical system. The treatments with M. smithii unambiguously lowered the animals? risk of DCs for both rats and pigs (Figs. 2 and 5). in fact, we discovered serendipitously while establishing our basic dive protocol with pigs that their native gut flora of methanogens can provide enough H 2 scrubbing activity to reduce DCs risk (kayar and Fahlman, 2001). We made some simplifying assumptions regarding H 2 solubility in the tissues of animals and thus total H 2 content in rats and pigs of the body masses we used (kayar et al., 1998; kayar et al., 2001), and the estimated amount of H 2 eliminated by the M. smithii in each case (four times the amount of methane evolved; Equation 1). Based on these calculations, we estimated that the M. smithii removed approximately 5-10 % of the body burden of H 2 acquired during the hyperbaric H 2 exposure; this gas removal resulted in a roughly 50% decrease in DCs risk in both the rats and the pigs (kayar et al., 1998; kayar et al., 2001; Fahlman et al., 2001). it is important to note that although DCs risk was sig- nificantly negatively correlated with increasing methanogenic activity in our large animal model, eliminating more tissue H 2 was not completely deterministic of reduced DCs risk (Fig. 4). The two animals shown in this figure that released more than 130 ?mol CH 4 ? min-1 during the simulated dive both had sub- sequent DCs. This may well represent the same phenomenon that has puzzled diving physiologists for years; namely, that grades of vascular bubbles as measured by Doppler ultrasound and other means of bubble detection are not deterministic of DCs risk or severity (Eftedal et al., 2007). Thus, while we have definite evidence that increasing the elimination of tissue inert gas is significant in lowering DCs risk, there must be one or more other risk factors that we have yet to identify. it is feasible with existing technology to freeze-dry or oth- erwise concentrate and suspend the metabolism of M. smithii, then pack the inactive microbes into enteric-coated capsules. With customized coating, the capsules can be designed to break open only in the environment of the large intestine, allowing the microbes to be released and reactivated. since M. smithii is a normal constituent of the human intestinal flora for people on a western diet and is well characterized as nonpathogenic (miller, 1991), it would be a minor step to making such capsules ready for human safety testing. if the remaining steps to achieving safer and/or faster de- compression by taking a capsule are currently within our reach, why is it that a decade after the original demonstrations of H 2 biochemical decompression were published are we not all pop- ping capsules before our dives? The answer is of course because we are not all diving with H 2 . Hydrogen as a low narcotic potency diving gas was original- ly studied in sweden in response to being cut off from access to He during the second World War (Brauer and naquet, 1987). The inclusion of H 2 in an ultra-deep diving gas mixture reduces High-Pressure neurologic syndrome (HPns) and decreases breathing resistance, thus making possible the record dive to 700 m in marseille in 1991 (imbert, 2006). The U.s. navy?s motivation for investing in ultra-deep H 2 diving involved Cold War spying (sontag and Drew, 1998). However, the Herculean efforts needed for working safely with hyperbaric H 2 are barely hinted at by the descriptions above of the complex gas mix- ing and dive profiles used in these experiments, not to mention the added levels of complexity added when human subjects are FIgURe 5. Decompression sickness incidence (%) for pigs that were untreated controls, treated with intestinal injections of saline solution (surgical controls), or treated with intestinal injections of Methanobrevibacter smithii (methanogens). error bars represent 95% confidence limits on binomial distributions. *Significantly lower than in pooled control groups (p < 0.05, Chi-squared test; Kayar et al., 2001). 38 ? tHE FUtURE oF Diving used. nitrogen can also supplement heliox to reduce HPns, along with slower compression (imbert, 2006). There is thus little motivation for using H 2 as a diving gas when He is readily available or in operations in which submersibles can perform the tasks. The experimental H 2 diving in France of the 1980s and early 90s (e.g., abraini et al., 1994; Fontanari et al., 2000) is no longer taking place. Can the experience gained from these experiments with H 2 biochemical decompression be translated into something more practical, such as n 2 biochemical decompression? Theoretically, n 2 biochemical decompression should be possible (kayar et al., 1998). The processes of metabolizing n 2 and H 2 have similarities (lehninger, 1970). There are microbes that metabolize atmo- spheric n 2 to ammonia, nitrites or nitrates, a chemical process referred to as ?fixing? nitrogen, some of which are present in the native gut flora of humans and other mammals. However, there are multiple practical impediments to n 2 biochemical decom- pression. nitrogen fixation is a biochemically complex process that requires an input of energy, as compared to the metabolism of H 2 that liberates energy (lehninger, 1970). Consequently, most microbes will only fix significant amounts of atmospheric n 2 in the absence of fixed nitrogen compounds in their environ- ment. Feces are a major source of fixed nitrogen, hence their common use as plant fertilizer, giving the plants their needed precursors for making proteins. in order for a microbe to fix a significant amount of n 2 while inside the intestines of a diver, we would have to select or create a mutant microbial strain that is constitutive for fixing n 2 with high concentrations of fixed nitrogen compounds in its environment. nitrogen fixation is a relatively slow process, much slower than metabolizing H 2 ; identifying or engineering a mutant strain that is constitutive for fixing nitrogen at a rate that would be practical for diving, FIgURe 6. proposed physiological mechanism for h 2 biochemical decompression (Fahlman et al., 2001). In each figure, the number in parentheses represents the net result of the process of gas uptake (compression) or elimination (decompression) by the body at the end of a circulatory pass from the gas exchange surface throughout the body and returning to the gas exchange surface. Units in these figures are hypothetical and are used only to describe the current working hypothesis. See Fahlman et al., 2001 for a detailed description of this figure. kaYaR: BioCHEmiCal aPPRoaCH to DEComPREssion ? 39 and in a volume of microbes convenient to introduce into the intestines, will be a challenge. Then would come the challenge of tracking the nitrogen fixation reaction inside the diver. Unlike H 2 metabolism that conveniently allows stoichiometric quantities of methane to es- cape and be tested in the atmosphere around the diver, nitrogen compounds fixed by n 2 biochemical decompression processes would be in the aqueous phase and mixed with fixed nitro- gen compounds from multiple other biological processes. one would probably have to work with an isotopic n 2 and look for the label in feces and possibly the intestinal wall or other tissues in order to track this reaction. all these steps would be difficult, but not impossible. We believe that there are benefits to diving research from having completed the demonstrations of H 2 biochemical de- compression, despite the negligible role of H 2 diving today. it seems little likely that anyone would want to face all the obsta- cles to n 2 biochemical decompression without evidence of the feasibility of the concept of the intestine as a chemical scrubber unit for the inert gases acquired during diving. our analyses from H 2 biochemical decompression demonstrated that remov- ing only a few percent of the estimated total inert gas load of a diver can have a major impact on DCs risk. This finding not only made some of the obstacles to n 2 biochemical decompres- sion less daunting but also gave us a new way to examine our underlying assumptions about inert gas loads and DCs risk. The H 2 biochemical decompression research also reconfirmed our speculation that while inert gas load is the primary factor to DCs risk, it is not exclusive. it may indeed one day in the not-so-distant future be pos- sible to mitigate the risk of DCs before a dive by swallowing a capsule. The best current bet on that capsule?s content would be an anti-inflammatory drug. it remains to be seen if that capsule alone suffices for all our demands on diving and all our aver- sions to occupational hazards that we will want to place on it, or whether someone will have to rise to the challenges of n 2 biochemical decompression. Acknowledgments andreas Fahlman made amazing and enormous contribu- tions to this research, through his critical insights, his math- ematical and scientific gifts, his unflagging determination, and his congenial nature. Erich C. Parker made several of the fig- ures presented here, provided expert guidance in data analysis and DCs modeling, and gave the unconditional moral support of a loving and patient husband. The quality and quantity of work on this project, and the perfect safety record throughout hazardous duty, are also directly attributable to the dedication, professionalism, and friendship of our support team members, Richard ayres, Jerry morris, Roland Ramsey, and Chiefs an- thony Ruopoli and scott ario, U.s. navy. lITEraTurE cITEd abraini, J. H., m. C. gardette-Chauffour, E. martinez, J. C. Rostain, and C. lemaire. 1994. Psychophysiological reactions in humans during an open sea dive to 500 m with a hydrogen-helium-oxygen mixture. Journal of Ap- plied Physiology 76:1113?1118. Boycott, a. E., g. C. C. Damant, and J. s. Haldane. 1908. The prevention of compressed-air illness. Journal of Hygiene, Cambridge 8:342?443. Brauer, R. W., and R. naquet. 1987. Contributions of animal experimentation to the establishment of appropriate conditions for ethical use of hydrogen as a component of deep diving atmospheres. in Hydrogen as a Diving Gas, R. W. Brauer, ed. Bethesda. mD: 33rd Undersea and Hyperbaric Medical Society Workshop. pp. 3?12. Eftedal, o. s., s. lydersen, and a. o. Brubakk. 2007. The relationship between venous gas bubbles and adverse effects of decompression after air dives. Journal of Undersea and Hyperbaric Medicine 34:99?105. Fahlman, a., P. tikuisis, J. F. Himm, P. k. Weathersby, and s. R. kayar. 2001. on the likelihood of decompression sickness during H 2 biochemical de- compression in pigs. Journal of Applied Physiology 91:2720?2729. Fontanari, P., m. Badier, C. guillot, C. tomei, H. Burnet, B. gardette, and Y. Jammes. 2000. Changes in maximal performance of inspiratory and skel- etal muscles during and after the 7.1?mPa Hydra 10 record human dive. European Journal of Applied Physiology 81:325?328. imbert, J. P. 2006. Deep diving: the Comex experience. in Long-Term Health Effects of Diving, Hope, a., and J. Risberg, eds. Bodoni Press, Bergen, nor- way, pp. 155?164. kayar, s. R., and a. Fahlman. 2001. Decompression sickness risk reduced by native intestinal flora in pigs after H 2 dives. Journal of Undersea and Hyper- baric Medicine 28:89?97. kayar, s. R., t. l. miller, m. J. Wolin, E. o. aukhert, m. J. axley, and l. a. kiesow. 1998. Decompression sickness risk in rats by microbial removal of dissolved gas. American Journal of Physiology 275:R677?R682. kayar, s. R., a. Fahlman, W. C. lin, and W. B. Whitman. 2001. increasing activity of H 2 -metabolizing microbes lowers decompression sickness risk in pigs during H 2 dives. Journal of Applied Physiology 91:2713?2719. lehninger, a. l. 1970. Biochemistry. new York: Worth Publishers, pp. 559?562. lillo, R. s., E. C. Parker, and W. R. Porter. 1997. Decompression comparison of helium and hydrogen in rats. Journal of Applied Physiology 82:892?901. little, t., and B. D. Butler. 2008. Pharmacological intervention in the inflam- matory response from decompression sickness in rats. Aviation, Space, and Environmental Medicine 79:87?93. miller, t. l. 1991. Biogenic sources of methane. in Microbial Production and Consumption of Greenhouse Gases: Methane, Nitrogen Oxides, and Halo- methanes, Rogers, J. E. and W. B. Whitman, eds. Washington, DC: amer- ican society of microbiology, p. 298. montcalm-smith, E. a., a. Fahlman, and s. R. kayar. 2008. Pharmacological interventions to decompression sickness in rats: comparison of five agents. Aviation, Space, and Environmental Medicine 79:7?13. national Research Council. 1996. Guide for the Care and Use of Laboratory Ani- mals. Washington, D.C.: institute of laboratory animal Resources, na- tional academy Press. nyquist, P., R. Ball, and m. sheridan. 2007. Complement levels before and after dives with a high risk of DCs. Journal of Undersea and Hyperbaric Medicine 34:191?197. sontag, s., and C. Drew. 1998. Blind Man?s Bluff. new York: Perseus Books Publicaffairs, 349 pp. Weathersby, P. k., l. D. Homer, and E. t. Flynn. 1984. on the likelihood of decompression sickness. Journal of Applied Physiology 57:815?825. INTrOducTION this overview focuses upon the relationship between exercise and decompres-sion. i was fortunate enough to be involved in some of the first studies that Drs. alf Brubakk and Ulrik Wisloff engaged in to find the links between exercise and decompression, at least those of bubble formation and decompression sickness (DCs). There are multiple components that ultimately may yield DCs, starting with a reduction in pressure that leads to a gas phase separation; gas bubbles generated in the vascular circulation that lead to adverse reaction to these bubbles; and, a large array of other physiological responses that are quite variable including epidermal rash, seizures, and even death. Predominantly, these features of DCs manifest when the diver exits the water upon completion of the hyperbaric exposure. bubblE dETEcTION aNd cONSEQuENcES fOr ThE ENdOThElIum The use of ultrasound Doppler facilitates the detection and ultimately the quanti- fication of bubbles within the human vasculature, as can be seen in Figure 1, with the left panel as an example of cardiac image with a large number of venous gas bubbles (no pre-dive exercise) and the right panel with significantly less bubbles evident (pre-dive exercise) Duji? et al. (2004). With this ultrasound Doppler approach, allowing one to actually quantify the degree of bubbles formation it is possible to compare and contrast the prevalence of bubbles with differing interventions (e.g., a dive with and without prior exercise). systemically, there are significant deleterious consequences of bubbles passing aBSTRaCT. This overview focuses upon the relationship between exercise and decompression. specifically, studies that examine vascular bubble detection and the consequences for the endo- thelium, the relationship between exercise and decompression sickness (DCs), and finally, diving, vascular function, and the role of nitric oxide are discussed. it is concluded from both animal and human models that exercise performed 24 hours prior to decompression and potentially after a dive is not harmful as once thought but appears instead to protect from DCs. a reduction in the num- ber of venous gas bubbles may be related to the reduced risk of DCs. There is convincing evidence that the mechanism responsible for these interactions is related to no bioavailability and thus the administration of an no donor may be a reasonable pharmacologic alternative to exercise to protect against DCs. University of Utah, Department of Internal Medicine, Division of Geriatrics, Department of Exercise and Sport Science, Geriatric Research, Education and Clinical Center, VAMC SLC, Bldg. 2 Rm. 1D25, 500 Foothill Drive, Salt Lake City, Utah, 84112, USA. Exercise and Decompression Russell S. Richardson 42 ? tHE FUtURE oF Diving through the vasculature. Histological slides from the endo- thelium of the pulmonary artery (Fig. 2) illustrate the typical healthy condition (left) compared to the damaging effect of bubble laden blood passing through the vasculature due to rapid decompression. rElaTIONShIp bETWEEN ExErcISE aNd dcS The risk factors for decompression sickness are diver spe- cific (i.e., age, gender, body mass index, fitness, dehydration, patent foramen ovale, and previous DCs), dive related (i.e., duration, depth, ascent protocol, frequency, exercise, and tem- perature) or post-dive related (i.e., exercise, ascent to altitude, and surface environment). it is clear that there are several men- tions of exercise as a risk factor and depending on its timing and intensity could play a role in the DCs phenomenon. Carturan et al. (2002) detected a strong relationship be- tween the number of bubbles formed and the maximum capac- ity to consume oxygen (vo 2 max) invoked by an graded exercise test. after similar dive exposure, reasonably fit individuals with a vo 2 max of ~46 ml/kg/min exhibited only ~0.07 bubbles/cm2, whereas less fit individuals with an average vo 2 max of ~35 ml/ kg/min exhibited ~0.5 bubbles/cm2. Thus, ?fitness? appears to afford some sort of protection from post-dive bubble formation. Recreational divers include not only younger healthy individuals but also older people who tend to have a lower vo 2 max, likely due to both decreased activity and aging per se. indeed, there is a reasonably robust relationship between age and exercise ca- pacity, specifically, as one ages, approximately 8% of vo 2 max is lost per decade (Fig. 3). vo 2 max can be improved by exercise training notwithstanding significant genetic components of this variable. However, the somewhat disappointing observation has been made, that if one is well trained when young, although starting higher in terms of metabolic capacity, it is likely that the loss of this capacity still occurs at the same rate even in the face of continued efforts to remain physically active. Therefore, for all, ultimately the line of decaying vo 2 max intersects the x-axis and, at least conceptually, signals the end of life as we know it. However, the good news for the promotion of physi- cal activity is that this does suggest that the upward shifted and parallel line of decay for exercise trained individuals will inter- sect the x-axis at a greater age, suggestive of longevity associated FIgURe 1. evidence of bubbles in the heart after a dive without prior exercise (left) and with prior exercise (right), from Duji? et al. (2004). FIgURe 2. endothelium of the pulmonary artery in a healthy state (top) and damaged by bubbles (bottom). FIgURe 3. age and exercise capacity. RiCHaRDson: ExERCisE anD DEComPREssion ? 43 does not have to be to yield this reduction in bubble formation and a reduced incidence and severity of DCs (Wisl?ff et al., 2004). Duji? et al. (2004) took these experiments into the hu- man realm where a single exercise bout prior to a dive produced extremely clear, bubble free, ultrasound images of the ventricles compared to the bubble-laden images from dives that were pre- ceded by inactivity. Thus, there is clearly a protective effect of acute exercise, but what is the optimum timeframe for acute exercise before a dive? With our current knowledge it seems that there may be some species differences. specifically, in rats the optimum time for exercise prior to a dive to yield protection from bubble formation and improve survival is approximately 20 hours prior to submersion (Wisl?ff et al., 2004). interest- ingly, in the animals, survival time was elongated when the ex- ercise occurred 48 hours prior to the dive, but there was not a large reduction in bubble formation. in humans, on the other hand, the optimum time to exercise prior to a dive appears to be much shorter, with acute exercise just 2 hours prior to a dive providing a protective effect Duji? et al. (2004). to truly sepa- rate species-specific differences from varying methodologic ap- proaches, more work is needed in this area. dIvINg, vaScular fuNcTION, aNd NITrIc OxIdE Hyperbaric pressure, induced by diving puts a significant stress upon the cardiovascular system. Recently, Duji? et al. (2006) examined central cardiovascular responses, in terms of cardiac function, to a single air dive to 30 m for 30 min. Fol- lowing the dive there were significant increases in end diastolic and end systolic volumes and a reduction in ejection fraction. in combination, these variables provide evidence that the dive exposure resulted in greater total peripheral resistance, against which the heart has to work. Endothelial dysfunction, a major mediator of vascular resistance, could be the cause of this in- crease in cardiac work. an additional link between DCs and the endothelium is the recognition that the bubbles associated with this disorder may originate in or around the endothelium. of endothelial micronuclei, Harvey (1951) stated ?if bubbles come from nuclei they must arise from gas nuclei sticking to or formed on or within the endothelial linings of the vascular sys- tem or extravascular spaces, and only when they have enlarged to the point of instability do they pass into the blood stream.? Thus, although circumstantial at best, evidence is growing that the endothelium may play a role in the development of DCs. although, as discussed, exercise training results in signifi- cant adaptations, even a single bout of exercise is a powerful stimulus that will elicit a multitude of responses. With respect to the endothelium, nitric oxide production, free radical gener- ation, and antioxidants to protect from free radicals, are all the result of an acute exercise bout (kojda and Hambrecht, 2005). among many other ideas, this raised the question, does nitric oxide blockade ablate the effect of exercise on bubble formation and what is the impact of such a blockage on survival time? with regular exercise. as a brief aside, it has been reported that the English nobel laureate Professor a.v. Hill, who studied the limitations to maximal exerise at University College, london, plotted his vo 2 max many time throughout his life and was sub- sequently able to predict his own death to within several years. Returning to the link between exercise capacity and DCs, based upon the work of Carturan et al. (2002), the implication here is that other factors, such as age, may play into the equation that links physical activity with the susceptibility to dive induced DCs. in 2001, Drs. Wisl?ff and Brubakk were conducting de- compression studies in rats without any real thought to the role of exercise in this process. While in the midst of these typical terminal studies (due to DCs), they experienced an acute short- age of animals and acquired some rats which were by happen- chance exercise trained. When these animal were exposed to the same physiologically challenging decompression regimen that had killed the majority of other rats they found that intravas- cular bubble formation was dramatically reduced and all the animals survived. Ultimately, the take-home message from these somewhat serendipitous experiments was that aerobic endur- ance exercise training reduced bubble formation in rats exposed to hyperbaric pressure (Wisl?ff and Brubakk, 2001). it was here that my laboratory became involved in these studies, having an interest in exercise, endothelial function, and nitric oxide. Exercise training results in a myriad of adaptations, one such adaptation is an increase in vo 2 max. This is facilitated by adaptations that lead to greater capillary growth factors and elements that improve performance. With our laboratory?s in- terest in oxygen transport and utilization we have documented morphometric differences between trained and untrained sub- jects showing an increased capillarity and coupled these struc- tural findings with functional measurements that reveal elevated oxygen extraction across the muscle bed. This is the result of an increased diffusional conductance for oxygen and reflects a greater ease of passage for oxygen from blood to tissue. interest- ingly, as the result of exercise training and the aforementioned angiogenesis, the vascular endothelial growth factor signal in response to acute exercise is now attenuated. This makes sense teleologically, for as you repeatedly exercise in the form of train- ing, there needs to be some limit to the capillary growth, as ulti- mately there has to be balance achieved between contractile and vascular structures in skeletal muscle (Richardson et al. 1999). However, what exactly links these complex exercise-induced ad- aptations and the apparent protection from DCs afforded by physical activity are not yet completely understood. looking for the link between exercise adaptations and DCs, again utilizing the rat model, revealed that although exer- cise training and the subsequent increase in ?fitness? per se may be beneficial, it was recognized that an acute bout of exercise exhibited the same protective effect against DCs (Wisloff et al., 2004). indeed, it was determined that progressive long-term training is not necessary to elicit this phenomenon. The stimu- lus is simply a single exercise session, which can be repeated, but 44 ? tHE FUtURE oF Diving Wisloff et al. (2003) exercised two groups of rats and then ad- ministered an no blocker (l-namE) prior to the dive simula- tion (Fig. 4). The previously exercised control group demon- strated very little bubble formation after the dive because of the prior exercise, and after 60 minutes were sacrificed, as there was no sign that they were going to die from DCs. The other group infused with l-namE, the nitric oxide synthase inhibitor, ex- hibited far greater bubble formation and only survived, on aver- age, for approximately 25 minutes following the dive (Wisloff et al., 2003) (Fig. 4). This study provided substantial evidence that nitric oxide does play a role in both bubble formation and susceptibility to DCs. in a follow-up to this rodent work Duji? et al., (2006) turned their attention to humans. intravascular nitroglycerin that exogenously elevates nitric oxide bioavailabil- ity was used to assess the importance of no levels. in control dives, there was much greater bubble formation than when div- ers were provided with nitroglycerin (Duji? et al., 2006). Thus, providing additional support for the role of nitric oxide and subsequently endothelial function in the process of post-dive bubble formation and DCs. Flow-mediated dilation (FmD) is a common bioassay used to assess vascular function and has recently been employed to better understand the interaction between exercise, vascular function, and DCs. to study FmD, an area of ischemia is cre- ated, often in the lower arm, by clamping off the blood flow with a suprasystolic cuff for approximately 5 minutes. The cuff release causes a hyperemia, after which the dilation of the ves- sel under scrutiny is measured by ultrasound Doppler, yielding both vessel diameter and blood velocity. The latter variable is likely of importance in addition to the assessment of diameter, as it has recently been recognized that the shear stress expe- rienced by the vessel because of the hyperemia is perhaps the predominant determinant of the observed dilation and there- fore blood velocity should be measured and taken into account. Thus, taking shear into account, the vasodilation resulting from an FmD test is thought to reflect nitric oxide bioavailability. Utilizing this FmD approach, Duji? et al. (2006) revealed a significant attenuation in endothelial function following a dive with post-dive function starting to be restored after 24 hours. This is further evidence that nitric oxide may be a major player in the development of DCs. Within endothelial cells, arginine is converted to Citrul- line by the phosphorylation of endothelial nitric oxide syn- thase (enos), the production of no, and subsequent smooth muscle relaxation (vasodilation). Prior to this final step, no can be ?intercepted? from its target of the smooth muscle by reacting with superoxide, a form of oxidative stress. This is a very rapid reaction (proposed to be the fastest in the body) that results in peroxynitrite and reduces no bioavailability. The re- cent use of antioxidants to study the role of oxidative stress and no bioavailability in the post-dive scenario provides more evi- dence that nitric oxide is a major player. specifically, Duji? et al. (2006) revealed that FmD is attenuated as a consequence of a dive but the prophylactic treatment with antioxidants lessened this effect, presumably by increasing nitric oxide bioavailability, again supportive of an interaction between no, vascular func- tion and diving. cONcluSION Both animal models and studies in humans suggest that ex- ercise performed 24 hours prior to decompression, after a dive, and even during decompression stops is not harmful as once thought, but appears instead to protect from DCs. a reduc- tion in the number of venous gas bubbles may be related to the reduced risk of DCs. There is convincing evidence that the mechanism responsible for these interactions is related to no bioavailability and thus the administration of an no donor may be a reasonable pharmacologic alternative to exercise to protect against DCs. Acknowledgments Funding for Dr. Richardson and his laboratory was provid- ed by The national institute of Health, Heart Blood and lung (Po1 Hl 09183-01a1), The tobacco-Related Disease Research Program (15Rt-0100), and the Francis Family Foundation. lITEraTurE cITEd Carturan, D., a. Boussuges, P. vanuxem, a. Bar-Hen, H. Burnet, and B. gar- dette. 2002. ascent rate, age, maximal oxygen uptake, adiposity, and circulating venous bubbles after diving. Journal of Applied Physiology 93:1349?1356. Duji?, ?., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n.m. Petri, U. Wisloff, and a.o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555(3):637?642. Duji?, ?., a. obad, i. Palada, Z. valic, and a. o. Brubakk. 2006. a single open sea air dive increases pulmonary artery pressure and reduces right ventricu- lar function in professional divers. European Journal of Applied Physiology 97(4):478?485. Duji?, ?., Z. valic, and a. o. Brubakk. 2008. Beneficial role of exercise on scuba diving. Exercise and Sport Sciences Reviews 36:38?42. Harvey, E. n. 1951 Physical factors in bubble formation. in Fulton, J.F., ed. De- compression Sickness. Philadelphia and london: W.B. saunders. pp. 90?114. kojda, g., and R. Hambrecht. 2005. molecular mechanisms of vascular adap- tations to exercise. Physical activity as an effective antioxidant therapy? FIgURe 4. effect of NO on bubble formation and survival time. RiCHaRDson: ExERCisE anD DEComPREssion ? 45 Cardiovascular Research 67:187?97. Richardson, R. s., B. grassi, t. P. gavin, l. J. Haseler, k. tagore, J. Roca, and P. D. Wagner. 1999. Evidence of o 2 supply-dependent vo 2max in the exercise- trained human quadriceps. Journal of Applied Physiology 86:1048?1053. Wisl?ff, U., and a. o. Brubakk. 2001. aerobic endurance training reduced bub- ble formation and increases survival in rats exposed to hyperbaric pressure. Journal of Physiology 537:607?611. Wisl?ff, U., R. s. Richardson, and a. o. Brubakk. 2003. nos inhibition in- creases bubble formation and reduces survival in sedentary but not exer- cised rats. Journal of Physiology 546:577?582. ???. 2004. Exercise and nitric oxide prevent bubble formation: a novel ap- proach to the prevention of decompression sickness? Journal of Physiology 555:825?829. INTrOducTION the Cellular stress Response (CsR) has its early beginnings linked closely to a phenomenon known as the Heat shock Response (HsR). The initial description of the HsR was made, unexpectedly, by investigators using the polytene chromosome of the Drosophila fruit fly as a model for studying gene expres- sion (Ritossa, 1964). tremendous growth in the fields of molecular and cell biology have occurred since the initial description of the HsR by Feruccio Ritossa in 1962 (Welch, 1993). The discovery and characterization of the heat-shock genes initiated the rapidly expanding field of molecular medicine that we are experiencing today (schlesinger et al., 1982). Early on it became clear that the heat-shock (Hs) gene expression could alter cellular physiology in a predictable fashion. increased Hs gene expression conferred a state of protection to cells, permitting them to resist potential- ly lethal insults such as, hyperthermia and acute ischemia (Horowitz and Robinson, 2007; Rokutan et al., 1998). Furthermore, it quickly became apparent that there were many ?noxious insults? that could induce Hs gene expression besides the original example of hyperthermia. one of the original, nonthermal, biochemical inducers of Hs genes was 2,4-dinitrophenol a potent blocker of the electron transport chain and aerobic respiration (Ritossa, 1964). since that time, the list of known inducers of the Hs genes continues to grow and now includes exposure to diving environments and aBSTRaCT. Divers and their physicians are well acquainted with the physiologic and medical im- plications of transient excursions into extreme environments. Exertion, hypothermia, anxiety, and exposure to various breathing mixtures under pressure are all components of this complex equation. The physiologic responses that characterize these exposures have been extensively studied, including that of oxygen toxicity. The pathophysiologic changes associated with decompression illness have likewise been extensively studied but remain poorly understood. more recently, investigative atten- tion has begun to focus upon the roles played by the cellular and biochemical changes underlying these phenomena. The complex processes of inflammation, oxidant stress, and vascular endothelial cell activation are becoming recognized as playing a major, albeit obscure, role in the physiologic and pathologic responses of divers to their environment. all of cellular life has preprogrammed within its genome the ability to respond in an adaptive way to noxious stimuli. This complex cel- lular response is referred to as the Cellular stress Response (CsR) and is the focus of attention in a wide range of medical investigations, including cardiovascular disease, diabetes, cancer, and trauma. The activation of the genes controlling this CsR (heat shock and molecular chaperone genes) can dramatically alter the physiologic outcome following exposure to extreme cellular environments, such as acute inflammation, hypoxia, or ischemia. This paper aims to introduce and review the history and development of the CsR as we know it today, relate the CsR to diving and hyperbaric medicine, and introduce the potential of hyperbaric oxygen to trigger and control the CsR. Department of Surgery, Morristown Memorial Hospital, Box 88, 100 Madison Ave., Morristown, New Jersey, 07960, USA. Diving medicine and the Cellular stress Response George A. Perdrizet 48 ? tHE FUtURE oF Diving hyperbaric oxygen (HBo 2 ) (kregel, 2002). Current models suggest that the common denominator through which these seemingly unrelated agents trigger this protective response is via the appearance of denatured intracellular proteins (Ben- saude et al., 1996; kim et al., 2007). it is now clear that the CsR can be induced by a wide array of conditions and agents which, in turn, can provide protection from an equally wide array of harmful or even potentially lethal insults. a second major characteristic of the CsR is that it is a time- sensitive phenomenon and thus its onset, duration, and reso- lution are potentially predictable events. This predicable time course lends itself to useful integration into many medically rel- evant applications, including the fields of diving and hyperbaric medicine. The time course of this Hs-gene expression and as- sociated state of cytoprotection was initially mapped out within the context of hyperthermic treatment of cancer and became known as thermotolerance (Herman et al., 1982). The early as- sociation of new Hs-gene expression with the phenomenon of thermotolerance represents the first association of heat-shock proteins (HsPs) with a biological function, i.e., acquired resis- tance to lethal thermal injury (landry et al., 1982). since that time, understanding of the molecular and biological behaviors of the HsR has steadily progressed. it is now widely appreci- ated that the functional outcomes of Hs-gene expression are the induction of the transient states of cytoprotection and anti- inflammation. These altered physiologic states are temporally associated with the synthesis of HsPs, especially the 70kD pro- tein, HsP70. The HsPs belong to a larger class of proteins, the molecular chaperones, which have now been assigned basal as well as stress-related functions. since the initial application of the HsR to the problem of organ preservation in 1986 (Perdrizet et al., 1989) significant advances have been made in all levels of Hs-related research (Wheeler and Wong, 2007). The body of knowledge surround- ing the molecular chaperones and CsR has grown, now being represented by an international society (international Cell stress society), a dedicated journal (Cell Stress and Chaperones) and annual scientific meetings. a detailed description of the HsR, HsPs, and CsR is beyond the scope of this paper, but excellent reviews provided by Csermely, 1998; nover, 1991; maresca and lindquist, 1991; morimoto et al., 1994, 1997; Rokutan et al., 1998. For this review of CsR and HsPs with an emphasis on models relevant to the science and practice of diving and hyper- baric medicine, four questions are addressed. IS dIvINg STrESSful? The concept of stress was originally defined with great dif- ficulty by selye (1946) as ?stress is any stimulus to which the system is not adapted.? He further notes that not all stress is harmful and differentiated between good stress (?eustress?) and harmful stress (?distress?). This latter distinction clearly recog- nizes the existence of an intrinsic and adaptable state of resis- tance to stress. a stimulus that is harmful to one cell or organism may not be harmful to another and, furthermore, by manipulat- ing the state of resistance of the individual a potentially harm- ful stimulus can be made less harmful. Thus, taking a person unfamiliar with diving and asking them to dive to 200 feet in the cold north sea would be certainly distressful, perhaps even lethal. However, for the same individual with proper diver train- ing such excursion would not be stressful but exhilarating or ad- dictive! Diving is a clear psychological stressor. managed appro- priately, the response to stress is adaptive, but if overwhelmed, it can become lethal. Psychological stress has been shown to induce the CsR in humans, including induction of the HsP (Flerov et al., 2003). simply studying for, or taking, an important profes- sional examination can reach the threshold for the induction of the CsR (morita, 2005). Relevant to diving and decompres- sion illness, this stress-induced CsR is associated with a state of anti-inflammation, including down-regulation of inflammatory cytokine release and vascular endothelial cell activation (House et al., 2001). no one would argue that excursions to any extreme environment are not stressful on both psychological and physi- ological levels. But is ?routine? or recreational diving stressful? Psychological stress aside, uncomplicated, routine diving pres- ents at least three potential stressors to human biology: a) expo- sure to increased partial pressures of dissolved gases, especially nitrogen and oxygen, b) thermal stresses, both hot and cold, and c) weightlessness or microgravity. This last factor is counterintui- tive, weightlessness would appear to be the antithesis of stress). one may also add to this list sleep deprivation, dehydration, and the fasted state as variable but real contributors to the total stress related to routine diving activities. is there scientific evidence at the cellular level that diving is stressful? a brief literature review reveals that a myriad of examples exist in which diving is associated with cellular stress, as evi- denced by increased synthesis of HsPs. monocytes sampled from saturation divers on day 40 at 400 msw demonstrated a 175-fold increase in HsP70 and 210-fold increase in HsP27 protein expression (matsuo et al., 2000). lymphocytes taken from experienced sCUBa divers demonstrated elevated base- line expression of HsP70 protein and mRna and reduced CD18 mRna compared to nondiving volunteers (Cameron et al., 2008). lymphocytes taken from humans three hours fol- lowing a single dive (n = 7 male divers, 40msw, 25min, breath- ing air) demonstrated elevated heme-oxygenase-1 (Ho-1, also known as HsP32) and the neutrophils showed a reduction in myeloperoxidase activity (Ferrer et al., 2007). The authors at- tributed these changes in white blood cells to the presence of oxidant stress. They also noted elevated markers of damage to red blood cells (lDH-lactate dehydrogenase) and skeletal mus- cle cells (CPk-creatinine phosphokinase) consistent with cel- lular stress from the single dive. interestingly, the apparently stressless environment of weightlessness provided by neutral buoyancy may in fact be a significant cellular stressor. a review of the published literature demonstrates multiple studies that confirm the ability of mi- crogravity states to induce synthesis of stress proteins in both PERDRiZEt: Diving mEDiCinE anD tHE CEllUlaR stREss REsPonsE ? 49 in vitro (Dennog et al., 1999); HsP72 in human lymphocytes (shinkai et al., 2004); and, Ho-1 in human lung cell line in vitro (speit and Bonzheim, 2003). Despite these cellular and biochemical findings, patients exposed to HBo 2 generally do not experience any acute discomfort or harm. Contrast this to the experience of using classical whole-body hyperthermia to achieve the same end. Whole-body hyperthermia (core body temperature of 102 - 108?F) is extremely uncomfortable for hu- mans and posses a severe cardiovascular stress. Thus, the relative clinical ease by which HBo 2 t (whole-body hyperoxia) induces a CsR stands in stark contrast to whole-body hyperthermia. By virtue of its tolerability by humans it is a very attractive agent by which to manipulate the CsR in patients. The mechanism by which HBo 2 t induces the CsR is currently unknown but will likely involve the denaturation of intracellular proteins by the increased production of reactive oxygen species during exposure to hyperoxic stress. caN dIvINg Or hbO2T INducE prOTEcTION agaINST lEThal OxIdaNT INJury? a logical next question to ask is if HBo 2 can induce the CsR, can it also induce tissue-level protection? a state of cy- toprotection against numerous noxious insults is known to temporarily associate with the CsR and HsR gene expression. specific examples exist in a wide array of models and attests to the robustness of this phenomenon. The published literature contains models describing ischemic preconditioning, endotox- in tolerance, thermotolerance, and stress conditioning, demon- strating the fundamental similarity of these seemingly unrelated phenomena (Perdrizet, 1995). The unifying concept was first described by Hans selye as the triphasic nature of the stress response. selye?s observations were made on the organismal level and have now been confirmed to be active at the cellular level (Perdrizet, 1997). The phenomenon of ?acclimatization? described as an adaptive resistance to decompression illness in caisson workers would appear to represent yet one more exam- ple of the triphasic nature of the stress response (Walder, 1968). Furthermore, a prior sublethal episode of DCi appears to confer protection against a subsequent DCi episode in rabbits and is associated with increased synthesis of HsP70 (su et al., 2004). Exposure of rodents to whole-body hyperthermia in a model of DCi protected the lungs and was associated with increased pulmonary expression of HsP70 (Huang et al., 2003). HBo 2 exposure has been shown to confer protection against numerous noxious insults. This should come as no surprise given the consistent observation that HBo 2 exposure induces the CsR and HsR gene expression in animal and hu- man cells and tissues. The ability to protect tissues against acute ischemia would have wide potential application and benefit in medical and surgical therapies. There is extensive evidence in experimental animal models of acute ischemia that precondi- tioning with HBo 2 can increase ischemic tolerance of many animal and human tissues and cells. Zebra fish exposed to mi- crogravity during development demonstrated an increased ex- pression of HsP70 within lens tissue (shimada and moorman, 2006). adult zebra fish exposed to microgravity demonstrated elevated synthesis of HsP70 within splenic tissues (ohnishi et al., 1998). Human vascular endothelial cells, the monocytic cell line (U937) and human lymphocytic cell line (Jurkat), demon- strated an increased synthesis of HsP70 following exposure to microgravity in vitro (Cotrupi and maier, 2004; maier, 2006; Carlsson et al., 2003; Cubano and lewis, 2001). The stress re- sponse to microgravity appears to be highly conserved across species as E. coli up-regulates sigma factor in response to micro- gravity and is consistent with the universal nature of the CsR (lynch et al., 2004). Finally, the occurrence of a diving com- plication, like DCi, expectedly induced the synthesis of stress proteins in the rodent (montcalm-smith et al., 2007). IS hbO2 STrESSful? Hyperbaric oxygen therapy (HBo 2 t) is widely practiced throughout the world and has been for many years. HBo 2 t is provided by highly trained personnel in very controlled, clini- cal environments. it is without question that the practice of hyperbaric medicine is very safe, but is HBo 2 stressful? modern hyperbaric medicine programs routinely expose hundreds of very frail and debilitated patients to HBo 2 on a daily basis for the purposes of aiding the wound healing response and do so with very low complication rates. it is therefore, from a clinical perspective, an unexpected notion that HBo 2 t might be stress- ful to humans as it appears so very well tolerated by individuals whose physiologic reserve has been compromised by years of chronic disease (i.e., obesity, diabetes, cardiovascular disease). as with the submersion example in diving, entry into a hy- perbaric chamber can provoke acute anxiety and as such likely induces the synthesis of stress proteins. aside from this psy- chological stress, does the respiration of high partial pressures of oxygen also induce a CsR? Humans have clear limits to the amount of oxygen they can be exposed to before direct organ toxicity occurs (Clark et al., 2006). The doses of oxygen cur- rently used in the practice of hyperbaric medicine have been ad- justed to minimize gross end-organ toxicity. Current algorithms used in HBo 2 t rarely produce organ injury, but is it stressful at the cellular level? a review of the published literature reveals a consistent finding: mammalian tissues exposed to HBo 2 produce large amounts of stress proteins. This new protein synthesis indicates that HBo 2 t is triggering a CsR in animal and human cells and tissues for both in vitro and in vivo settings. several recent re- ports in the medical literature serve as examples. HBo 2 induces: expression of HsP70 in a murine neuroblastoma cell line in vi- tro (shyu et al., 2004); HsP90 and nos in rodent brain and is consistent with a response to oxidant stress (Thom et al., 2002); lens epithelial stress protein expression (Padgaonkar et al., 1997); Dna damage and stress proteins in human lymphocytes 50 ? tHE FUtURE oF Diving tissues. Prior exposure of rodents to clinically relevant HBo 2 confers a state of ischemic tolerance in brain (Wada, 2001), heart (kim, 2001), and spinal cord (Dong, 2002). The mecha- nism of this protection is under investigation and may involve HsP32 (a.k.a., heme-oxygenase-1) (li et al., 2007). HBo 2 pro- tects rodent lungs from lPs toxicity and increases Ho-1 ex- pression (Huang et al., 2005). HBo 2 protects rodent liver and is associated with increased expression of hepatic HsP70 (Yu et al., 2005). HBo 2 can protect rodent brain from acute ischemia and increases stress protein expression (konda et al., 1996). it is important to recognize that given the universal nature of the CsR and the highly conserved nature of its genetic appa- ratus, a phenomenon of cross-protection exists. Thus, the stress- or used to induce tissue protection need not be the same agent against which protection is desired. For example, heat acclima- tization can delay the onset of Cns oxygen toxicity in rodents and is associated with elevated HsP70 (arieli et al., 2003). Heat shock can protect against DCi, oxygen prebreathing can reduce the incidence of DCi, HBo 2 protects against lPs-induced lung injury (Huang et al., 2005). caN prE-ExpOSurE Of humaN cEllS, TISSuES Or OrgaNS TO hbO2 INcrEaSE rESISTaNcE TO acuTE ISchEmIa? HBo 2 protects human lymphocytes from Dna damage by oxidant injury (hydrogen peroxide). This protection was associ- ated with increased synthesis of HsP32 (gr?ger et al., 2005). to date, there is only one study that has tested the ability of HBo 2 pretreatment to protect humans from acute ischemia during coronary artery bypass grafting procedure employing cardiopulmonary bypass (alex et al., 2005; Yogaratnam et al., 2006). sixty-four patients were randomized to receive (n = 33) or not receive (n = 31) three pretreatments of HBo 2 t (100% oxygen, 2.4 ata, 90 min per treatment) at 24, 12, and 2 hours prior to the surgical bypass procedure. Patients receiving the HBo 2 t were found to have significantly less cognitive impair- ment when tested four months following surgery as compared to the controls (HBo 2 t: 30% versus Control: 55%, p < 0.05). The observed protection of neurocognitive function was asso- ciated with significantly less HsP70 in the serum from those patients treated with HBo 2 t as measured at 2 hours following completion of the surgical procedure. cONcluSION Humans exposed to the routine diving environment are ex- posed to significant biological stressors as evidenced by increased synthesis of stress proteins, the hallmark of a cellular response to stress. Clearly important factors such as the degree, dura- tion, and innate host resistance all contribute to determine the ultimate impact these environments have on diver health and disease. in short, diving is stressful. Humans exposed to HBo 2 t demonstrate an increased synthesis of stress proteins. The in- creased synthesis of stress proteins is the biochemical signature of the triggering of a CsR and as such may have wide-ranging potential for medical application. But are these changes in gene expression associated with cellular protection from oxidant in- jury such as that which occurs with acute ischemia? Exposure to both diving and hyperbaric oxygen environments results in the up-regulation of the CsR and HsR genes and these changes are predictably related to a transient period of cellular resistance to oxidant injury. Because HBo 2 is so well tolerated on a clinical level, it may represent the ideal pharmacologic agent to stress condition human beings prior to exposure to anticipated stress- ful events, such as invasive medical or surgical procedures. Can HBo 2 t precondition humans? The clinical need to provide robust protection from acute ischemic injuries associated with common medical and surgical procedures is great. additional interventional trials which can test the HBo 2 t-stress condition- ing hypothesis need to be performed and the nature of the po- tential benefit for future medical care determined. lITEraTurE cITEd alex, J., g. laden, a. R. Cale, s. Bennett, k. Flowers, l. madden, E. gardiner, P. t. mcCollum, and s.C. griffin. 2005. Pretreatment with hyperbaric oxygen and its effect on neuropsychometric dysfunction and systemic inflammatory response after cardiopulmonary bypass: a prospective ran- domized double-blind trial. Journal of Thoracic and Cardiovascular Surgery 130(6):1623?1630. arieli, Y., m. Eynan, H. gancz, R. arieli, and Y. kashi. 2003. Heat acclimation prolongs the time to central nervous system oxygen toxicity in the rat. Pos- sible involvement of HsP72. Brain Research 962:15?20. Bensaude, o., s. Bellier, m. F. Dubois, F. giannoni, and v. t. nguyen. 1996. Heat-shock induced protein modifications and modulation of enzyme ac- tivities. EXS 77:199?219. Cameron, B. a., t. m. mclellan, D. J. Eaton, and s. g. Rhind. 2008. inflam- matory gene expression in the leukocytes of na?ve subjects and experienced divers following acute hyperbaric stress. Journal of Undersea and Hyper- baric Medicine 35(4):243 abstract a6. Carlsson, s. i., m. t. Bertilaccio, E. Ballabio, and J. a. maier. 2003. Endothelial stress by gravitational unloading: effects on cell growth and cytoskeletal organization. Biochimica et Biophysica Acta 1642:173?179. Clark, J. m., C. J. lambertsen, R. gelfand, and a. B. troxel. 2006. optimiza- tion of oxygen tolerance extension in rats by intermittent exposure. Journal of Physiology 100:869?879. Cotrupi, s., and J. a. maier. 2004. is HsP70 up regulation crucial for cellular proliferative response in simulated microgravity? Journal of Gravitational Physiology 1:P173?176. Csermely, P. ed. 1998. stress of life from molecules to man. Annals of the New York Academy of Sciences 851:entire volume. Cubano, l. a., and m. l. lewis. 2001. Effect of vibrational stress and space- flight on regulation of heat shock proteins HsP70 and HsP27 in human lymphocytes (Jurkat). Journal of Leukocyte Biology 69:755?761. Dennog, C., P. Radermacher, Y. a. Barnett, and g. speit. 1999. antioxidant status in humans after exposure to hyperbaric oxygen. Mutation Research 428:83?89. Dong, H., l. xiong, Z. Zhu, s. Chen, l. Hou, and t. sakabe. 2002. Precon- ditioning with hyperbaric oxygen and hyperoxia induces tolerance against PERDRiZEt: Diving mEDiCinE anD tHE CEllUlaR stREss REsPonsE ? 51 spinal cord ischemia in rabbits. Anesthesiology 96:907?912. Ferrer, m. D., a. surenda, J. m. Batle, P. tauler, J. a. tur, and a. Pons. 2007. scuba diving enhances endogenous antioxidant defenses in lymphocytes and neutrophils. Free Radical Research 41:274?281. Flerov, m. a., n. E. ordyan, B. a. margulis, i. v. guzhova, a. v. v?yushina, s. g. Pivina, and i. a. gerasimova. 2003. The use of HsP70 for prevention of consequences of unavoidable stress in rats. Bulletin of Experimental Biology and Medicine 136:120?122. gr?ger, m., g. speit, P. Radermacher, and C. m. muth. 2005. interaction of hy- perbaric oxygen, nitric oxide, and heme oxygenase on Dna strand breaks in vivo. Mutation Research 572:167?172. Herman, t. s., C. C. sweets, D. m. White, and E. W. gerner. 1982. Effect of heating on lethality due to hyperthermia and selected chemotherapeutic drugs. Journal of the National Cancer Institute 68:487?491. Horowitz, m., and s. D. Robinson. 2007. Heat shock proteins and the heat shock response during hyperthermia and its modulation by altered physi- ological conditions. Progress in Brain Research 162:433?446. House, s. D., P. t. guidon, Jr., g. a. Perdrizet, m. Rewinski, R. kyriakos, R. s. Bockman, t. mistry, R. a. gallagher, and l. E. Hightower. 2001. Effects of Heat shock, stannous Chloride, and gallium nitrate on the Rat in- flammatory Response. Cell Stress and Chaperones International 6:164?171. Huang, k. l., C. P. Wu, Y. l. Chen, B. H. kang, and Y. C. lin. 2003. Heat stress attenuates air bubble-induced acute lung injury: a novel mechanism of diving acclimatization. Journal of Applied Physiology 94:1485?1490. Huang, t. Y., P. s. tsai, t. Y. Wang, C. l. Huang, and C. J. Huang. 2005. Hyperbaric oxygen attenuation of lipopolysaccharide-induced acute lung injury involves heme oxygenase-1. Acta Anaesthesiologica Scandinavica 49:1293?1301. kim, C. H. 2001. Hyperbaric oxygenation pretreatment induces catalase and reduces infarct size in ischemic rat myocardium. Pflugers Archiv, European Journal of Physiology 442:519. kim, H. J., n. R. Hwang, and k. J. lee. 2007. Heat shock responses for under- standing diseases of protein denaturation. Molecules and Cells 23:123?131. konda, a., s. Baba, t. iwaki, H. Harai, H. koga, t. kimura, and J. takamatsu. 1996. Hyperbaric oxygenation prevents delayed neuronal death following transient ischaemia in the gerbil hippocampus. Neuropathology and Applied Neurobiology 22:350?360. kregel, k. C. 2002. Heat shock proteins: modifying factors in physiological stress responses and acquired thermotolerance. Journal of Applied Physiol- ogy 92:2177?2186. landry, J., D. Bernier, P. Chr?tien, l. m. nicole, R. m. tanguay, and n. mar- ceau. 1982. synthesis and degradation of heat shock proteins during de- velopment and decay of thermotolerance. Cancer Research 42:2457?2461. li, Q., J. li, l. Zhang, B. Wang, and l. xiong. 2007. Preconditioning with hyperbaric oxygen induces tolerance against oxidative injury via increased expression of heme oxygenase-1 in primary cultured spinal cord neurons. Life Sciences 80:1087?1093. lynch, s. v., E. l. Brodie, and a. matin. 2004. Role and regulation of sigma s in general resistance conferred by low-shear simulated microgravity in Escherichia coli. Journal of Bacteriology 186:8207?8212. maier, J. a. 2006. impact of simulated microgravity on cell cycle control and cytokine release by U937 cells. International Journal of Immunopathology and Pharmacology 19:279?286. maresca, B., and s. lindquist, eds. 1991. Heat Shock. Berlin: springer-verlag. matsuo, H., n. shinomiya, and s. suzuki. 2000. Hyperbaric stress during satu- ration diving induces lymphocyte subset changes and heat shock protein expression. Journal of Undersea and Hyperbaric Medicine 27:37?41. montcalm-smith, E., J. Caviness, Y. Chen, and R. m. mcCarron. 2007. stress biomarkers in a rat model of decompression sickness. Aviation, Space, and Environmental Medicine 78(2):87?93. morimoto, R. i., a. tissi?res, and C. georgopoulos, eds. 1994. The Biology of Heat Shock Proteins and Molecular Chaperones. Cold spring Harbor: Cold spring Harbor laboratory Press. 610 pp. morimoto, R. i., m. P. kline, D. n. Bimston, and J. J. Cotto. 1997. The heat- shock response: regulation and function of heat-shock proteins and molecular chaperones. Essays in Biochemistry 32:17?29. morita, k. 2005. Expression analysis of psychological stress-associated genes in peripheral blood leukocytes. Neuroscience Letters 381:57?62. nover, l. 1991. Heat Shock Response. Boca Raton: CRC Press. ohnishi, t., k. tsuji, t. ohmura, H. matsumoto, x. Wang, a. takahashi, s. nagaoka, a. takabayashi, and a. takahahsi. 1998. accumulation of stress protein 72 (HsP72) in muscle and spleen of goldfish taken into space. Advances in Space Research 21:1077?1080. Padgaonkar, v. a., F. J. giblin, k. Fowler, v. R. leverenz, J. R. Reddan, and D. C. Dziedzic. 1997. Heme oxygenase synthesis is induced in cultured lens epithelium by hyperbaric oxygen or puromycin. Experimental Eye Research 65:435?443. Perdrizet, g. a. 1995. Heat shock and tissue Protection. New Horizons 3:312?320. ???-. 1997. Hans selye and Beyond: Responses to stress. Cell Stress and Chaperones 2:1?6. Perdrizet, g. a., t. g. Heffron, F. C. Buckingham, P. J. salciunas, a. o. gaber, F. P. stuart, and J. R. Thistlethwaite. 1989. stress conditioning: a novel approach to organ preservation. Current Surgery 46:23?25. Ritossa, F. 1964. Experimental activation of specific loci in polytene chromo- somes of Drosophila. Experimental Cell Research 35:601?607. Rokutan, k., t. Hirakawa, s. teshima, Y. nakano, m. miyoshi, t. kawai, E. konda, H. morinaga, t. nikawa, and k. kishi. 1998. implications of heat shock/stress proteins for medicine and disease. Journal of Medical Investiga- tion 44:137?147. schlesinger, m. J., J. ashburner, and a. tissi?res, eds. 1982. Heat Shock from Bacteria to Man. Cold spring Harbor: Cold spring Harbor laboratory Press. selye, H. 1946. The general adaptation syndrome and the diseases of adaptation. Journal of Clinical Endocrinology 6(2):117?230. shimada, n., and s. J. moorman. 2006. Zebra fish lens: changes in gravitational force cause changes in gene expression in the lens of developing zebra fish. Developmental Dynamics 235:2686?2694. shinkai, m., n. shinomiya, s. kanoh, k. motoyoshi, and H. kobayashi. 2004. oxygen stress effects on proliferation rates and heat shock proteins in lym- phocytes. Aviation, Space, and Environmental Medicine 75:109?113. shyu, W. C., s. Z. lin, k. saeki, a. kubosaki, Y. matsumoto, t. onodera, m. F. Chiang, P. Thajeb, and H. li. 2004. Hyperbaric oxygen enhances the expression of prion protein and heat shock protein 70 in a mouse neu- roblastoma cell line. Cellular and Molecular Neurobiology 24(2):257?268. speit, g., and i. Bonzheim. 2003. genotoxic and protective effects of hyperbaric oxygen in a549 lung cells. Mutagenesis 18:545?548. su, C. l., C. P. Wu, s. Y. Chen, B. H. kang, k. l. Huang, and Y. C. lin. 2004. acclimatization to neurological decompression sickness in rabbits. Ameri- can Journal of Physiology 287:R1214?1218. Thom, s. R., v. Bhopale, D. Fisher, Y. manevich, P. l. Huang, and D. g. Buerk. 2002. stimulation of nitric oxide synthase in cerebral cortex due to el- evated partial pressures of oxygen: an oxidative stress response. Journal of Neurobiology 51(2):85?100. Wada, k. 2001. Preferential conditions for and possible mechanisms of induc- tion of ischemic tolerance by repeated hyperbaric oxygenation in gerbil hippocampus. Neurosurgery 49:160. Walder, D. n. 1968. adaptation to decompression sickness in caisson work. Bio- meteorology 11:350?359. Welch, W. J. 1993. How cells respond to stress. Scientific American 268(may):56?64. Wheeler, D. s., and H. R. Wong. 2007. Heat shock response and acute lung injury. Free Radical Biology and Medicine 42:1?14. Yogaratnam, J. Z., g. laden, l. guvendik, m. Cowen, a. Cale, and s. griffin. 2006. Hyperbaric oxygen: a new drug in myocardial revascularization and protection? Cardiovascular Revascularization Medicine 7:146?154. Yu, s. Y., J. H. Chiu, s. D. Yang, H. Y. Yu, C. C. Hsieh, P. J. Chen, W. Y. lui, and C.W. Wu. 2005. Preconditioned hyperbaric oxygenation protects the liver against ischemia-reperfusion injury in rats. Journal of Surgical Re- search 128:28?36. INTrOducTION During recent years there has been considerable focus on the risk of injury and dysfunction to the central nervous system in professional diving, commonly known as decompression illness. Different models have been developed to prevent the adverse effects of decompression for more than 100 years, but still the mechanisms behind the clinical manifestations are not completely understood. What we do know is that the initial step in the cascade of pathophysiological events lead- ing to decompression sickness (DCs) is the formation of gas bubbles (Francis and mitchell, 2003) and that most health hazards experienced in diving are consequences of changes in gas volume and formation of gas bubbles due to reduction in the ambi- ent pressure during a diver?s ascent. The god?ysund conference in 1993 highlighted the possible detrimental long term health effects of diving. This consensus conference was followed by a second conference in 2005 in Bergen. Both concluded that long term detrimental health effects of diving may occur and can represent a threat to oc- cupational divers? quality of life (Hope and Risberg, 2006). Recently, two epidemiological studies from the north sea have verified the pres- ence of functional changes following dives. one showed that there was a relationship be- tween decompression and these changes (irgens et al., 2007) and the other also showed additional risk for welders who dived (Ross et al., 2006). in a previous study, we found aBSTRaCT. Changes in the physical or chemical surroundings of any organism, such as the pres- sure changes experienced on ascent from diving, are likely to trigger adaptive changes in gene expres- sion patterns. However, the molecular responses are not necessarily uniform. Considerable variation in both constitutive and inducible gene expression is seen between and even within individuals at different times. The expression of a specific gene is regulated both by the Dna-sequence of the gene itself and by epigenetic modifications that alter a gene?s availability for transcription without chang- ing the sequence of Dna. a diver?s capacity for adaptation to decompression may vary according to his or her individual genetic and epigenetic make up. some of the genes involved in nitric oxide homeostasis in the endothelium display considerable individual differences in activity, which is known to trigger susceptibility to a number of diseases. This could indicate a link between endothe- lial dysfunction and decompression sickness (DCs) making the endothelium a highly interesting organ to study. We plan to examine alterations in genetic expression profiles of Rna in vascular endothelium in situ following decompression using rats as experimental models. We further plan to look for epigenomic variations between individual animals by examining the methylation pattern of cytosine in genomic Dna. By doing so, we hope to gain knowledge of genetic responses to the physical and biochemical changes experienced in diving. Ultimately, the goal is to be able to better predict individual risk of developing DCs, and possibly also to prevent or relieve disease by means of preconditioning or pharmacological intervention. Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Olav Kyrres gt 9, N-7489, Trondheim, Norway. individual Risk of Decompression sickness: Possible Effects of genomic or Epigenomic variation altering gene Expression Andreas M?llerl?kken and Ingrid Eftedal 54 ? tHE FUtURE oF Diving a correlation between manifestation of DCs and the number of times the diver underwent decompressions (Jacobsen et al., 1997; Brubakk and Eftedal, 2000). We have further found that a significant number of experienced professional divers (70%) have had symptoms of DCs without ever reporting them (Bru- bakk and Eftedal, 2000) and that signs of minor brain dysfunc- tion were detectable in these divers. Even if clinical symptoms of DCs are relative rare in professional diving today, it seems well documented that decompression is still the major inducer of long-term effects on the central nervous system. ThE paThOphySIOlOgy Of gaS bubblES The symptomatology of DCs is heterogeneous. Peripheral symptoms are frequently seen, but several central organ systems, including the central nervous system (Cns) and the respiratory system, may also be affected. it is quite possible that the patho- genesis of DCs at least in part is of an inflammatory origin as there is significant interindividual susceptibility to decom- pression trauma. Furthermore, repetitive dives have resulted in greater tolerance to DCs due to acclimatization (Bergh et al., 1993; Hjelde et al., 1995; Ersson et al., 2002; Buttolph et al., 1998). Bubbles may further lead to a tissue response with activation of platelets, the coagulation cascade (Eckmann and Diamond, 2004; nyquist et al., 2004; Thorsen et al., 1993), and complement (Hjelde et al., 1995). The pathological effects of bubbles may originate from a mechanical disruption of the tissue concerned, for example, the endothelium (nossum et al., 1999), with compression of non- compliant tissue or blood vessels and lymphatic tissue, or from simply obstructing blood vessels. signs and symptoms of DCs also differ with the pressure profile and breathing gas. neuro- logical symptoms are most common after short deep dives or altitude exposures with little or no pre-oxygenation. it is gener- ally assumed that localized gas bubbles are responsible for all DCs incidents in the Cns. However, Wilmshurst and Bryson (2000) showed that a large PFo can be found in about 50% of divers having central nervous symptoms. They have also ob- served that large shunts correlate well with spinal cord DCs. Wilmshurst et al. (2001) demonstrated further that there is a relationship between right-to-left shunts and cutaneous DCs, which often is associated with more serious DCs involving the Cns and the lung (Conkin, 1999; Zwart, 2000). gas bubbles can cause changes in barrier permeability even in the absence of clinical manifestations of DCs. Breakdown of the blood-brain- barrier (BBB) and blood-lung-barrier (BlB) may allow proteins and leukocytes to move into the extravascular brain tissue, with subsequent formation of oedema (Hjelde, et al., 2002; Chrys- santhou et al., 1977; van Hulst et al., 2003). leukocytes have been implicated in the progressive fall in cerebral blood flow and decreased cerebral function in animal models of gas em- bolism (Dutka et al., 1989; Helps and gorman, 1991).various plasma proteins including the coagulation system, complement and kinins are also activated by bubbles (Ward et al., 1987; Pekna et al., 1993). While bubbles in the venous system can explain pulmonary symptoms of DCs, there are other manifestations of DCs that can only be explained by bubble formation within the tissue themselves (Francis and mitchell, 2003). Extravascular bubbles may form in tissue that is aqueous or lipid, and except for ex- treme decompression bubbles are seldom observed in heart, liver and skeletal muscle (Wienke, 1994). Daniels (1984) reported that the earliest bubbles detected were intravascular, but that sub- stantial accumulation of stationary bubbles would occur before any signs of DCs were seen. in the periphery of the body, small intravascular bubbles may grow to sufficient size to occlude small vessels and thereby give rise to stationary intravascular bubbles. Blocking of the microcirculation causes not only tissue ischemia but also retards the elimination of dissolved gas and produces local areas with gas tensions higher than the surrounding tissue (Davies, 1983). once formed, extravascular bubbles persist for long periods of time. gas bubbles have been shown to persist up to two days after the original decompression (Davies, 1983). ENdOThElIum The endothelium plays a key role in the short- and long- term regulation of the cardiovascular system and harbours many factors that influence blood flow, blood coagulation as well as angiogenesis (triggle et al., 2003). The vascular endothelium consists of a monolayer of cells lining the luminal surface of all blood vessels in the body (Fig. 1). The endothelium functions by sensing various physiologic stimuli and triggering release of multiple vasoactive substances, including nitric oxide (no). such physiologic stimuli can be for example substances pres- ent in the blood or the shear stress associated with the blood flow. a large number of vasoactive substances are produced and secreted from endothelial cells to act on the underlying vascu- lar smooth muscle cells. The balance between dilating and con- tracting factors is critical for maintaining vascular homeostasis (Cockcroft, 2005). During resting conditions, the endothelial lining of the blood vessels forms a relatively inert surface that regulates and se- cures unhindered flow of cellular elements through the capillary beds. in response to an inflammatory signal initiated by bubbles, endothelial cells may be converted from an inactivated to an ac- tivated state inducing cellular functional changes. These changes may in turn destabilize pre-existing nuclei and make them ex- pand into bubbles. activation of the endothelium generates en- dothelial microparticles (EmP), which are fragments of activated endothelial cells. These may in turn reduce the endothelial func- tion, possibly by increasing expression of the endothelial adhesion molecules such as vascular cell adhesion molecule-1 (vCam-1), intracellular adhesion molecule-1 (iCam-1) and E-selectin, and by influencing no production (Jimenez et al., 2003; Brodsky et al., 2004). Following a decompression, this activation could be caused by endothelial damage from gas bubbles. m?llERl?kkEn anD EFtEDal: Risk oF DEComPREssion siCknEss ? 55 decompression using rats as experimental models. By doing so, we aim to gain knowledge of how specific genes respond to the physical and biochemical changes experienced in diving. Ulti- mately, the goal is to be able to better predict individual risk of developing DCs and possibly also to prevent or relieve disease by means of preconditioning or pharmacological interventions. large-scale techniques developed for transcriptomics and systems biology are useful for examining how genetic variations in complex biological pathways contribute to variable pheno- types such as disease susceptibility. in this study, we plan to use micro arrays for cDna and global Dna methylation to study both genetic and epigenetic variation, since both may contrib- ute to phenotypic variability. mutations to the Dna sequence can cause changes in gene expression by altering the affinity for transcription factors, changing the catalytic properties of a translated protein, or abolishing expression altogether. Epi- genetic patterns such as methylation of cytosine to 5-methyl- cytosine by imprinting during embryogenesis are also involved in regulation of gene expression. These patterns may be altered later in life by environmental factors such as diet, infections, and age. Epigenetic changes have been shown to be involved in the development of malignant disease through altering the rate of transcription of tumor suppressor genes or proto-oncogenes (lopez et al., 2009) and similarly may be involved in suscepti- bility to other diseases (Jirtle and skinner, 2007). Considering the apparent increased susceptibility to DCs with the diver?s age, it is interesting to note that both genome-wide and gene- specific methylation patterns have been shown to change with age (Bjornsson et al., 2008; Richardson, 2003). ThE raT aS a mOdEl SySTEm fOr ENdOThElIal fuNcTION Rats are generally considered to be genetically closer to hu- mans than mice, and its genome was recently sequenced. global Rna expression arrays that are highly suited for studies of gene expression patterns in rats, mice, and humans are commercially available. By using full genome cDna expression arrays, global alterations in expression patterns that may be subscribed to al- tered Dna translation can be examined simultaneously. a key first step in any genomic study is precisely defining responders and nonresponders to any controlled influence. one approach is to identify candidate genes that may determine the action of the bubbles by screening for Rnas that are signifi- cantly differentially transcribed in animals that have undergone decompression. Then the candidate may be screened for genetic variants and altered biological activity. fuTurE rESEarch The first part of our study will be a pilot involving a rela- tively small number of animals. We will extract Rna from the endothelial lining of rat aortas following dives using well frOm bubblES TO alTErEd gENE ExprESSION it is well documented that gas bubbles occur both on the arterial and venous side of the circulation following decompres- sion from a dive, and we have increasing knowledge of the dam- aging effects these bubbles have on the endothelium (Brubakk et al., 2007). Changes in the physical or chemical surroundings of any organism, such as those experienced on ascent from diving, are likely to trigger adaptive changes in gene expression patterns. The observation that some of the genes known to be involved in the nitric oxide (no) homeostasis display considerable indi- vidual difference in activity (kojda et al., 2001), and that no is an important mechanism in gas bubble formation (Wisloff et al., 2004; m?llerl?kken et al., 2006) points towards possible genetic links between endothelial dysfunction and DCs. When exposing the body to extreme physical or chemical conditions, such as the rapid alterations in surrounding pressure experienced by divers during and immediately after compres- sion and decompression, what is usually sufficient in terms of protein expression and interactions may no longer be enough to ensure health. The amount and rate of changes in pressure experienced by divers is a challenge not normally experienced under basal life conditions, and the genes needed to cope with these alterations thus may not have been subject to strong se- lective forces to retain high activity of expression. Haploinsuf- ficiency is a condition where the capability to adjust the level of Rna-expression in response to a stimulus is lowered due to partial or complete inactivation of one copy of an otherwise diploid gene. since decompression is not a phenomenon nor- mally experienced by humans or most mammals, it may well be that individual variation in susceptibility to DCs is reflected in levels of Rna-expression. We propose to examine alterations in genetic expression profiles in the endothelium in situ following FIgURe 1. The structure of the blood vessel wall with the three layers intima, media, and adventitia. The figure is reproduced in agreement to conditions given by Stijn a. I. ghesquiere, http://creativecommons.org/licenses/by-sa/2.5/. 56 ? tHE FUtURE oF Diving established diving protocols. Rna extraction will take place im- mediately after decompression to best preserve the true expres- sion pattern of genes. genetically similar animals not subjected to decompression but otherwise identically treated animals will serve as controls, and Rna will be extracted and analysed using the same protocol. The second part of the study will take into account the results of the Rna expression studies, further examining genes and pathways found to be significantly altered in animals dur- ing decompression. We also have access to a rat strain that we have found to be more susceptible to decompression-related ad- verse effects than other strains and plan to compare their genetic profiles with the profiles of less susceptible animals. knock-out animals are now commercially available for thousands of mouse genes, and a limited number of rat knock-outs are also emerg- ing. They may supply models for further studies. Finally, we seek to examine genetic variation in humans that may explain variable susceptibility to DCs. a bio-bank consisting of blood samples draw from divers may form the ba- sis for studies in which individuals who present with specific adverse effects of decompression are examined retrospectively for the presence of genetic variants suspected to be involved in susceptibility to disease. if clinically relevant genetic alterations are found, we seek to prevent or treat disease by precondition- ing or pharmacological intervention, or to facilitate informa- tion in advance for those who may be at particularly high risk of DCs. lITEraTurE cITEd Bergh, k., a. Hjelde, o. J. iversen, and a .o. Brubakk. 1993. variability over time of complement activation induced by air bubbles in human and rabbit sera. Journal of Applied Physiology 74:1811?1815. Bjornsson, H. t., m. i. sigurdsson, m. D. Fallin, R. a. irizarry, t. as- pelund, H. Cui, W. Yu, m. a. Rongione, t. J. Ekstr?m, t. B. Harris, l. J. launer, g. Eiriksdottir, m. F. leppert, C. sapienza, v. gudnason, and a. P. Feinberg. 2008. intra-individual change over time in Dna methylation with familial clustering. Journal of the American Medical Association 299:2877?2883. Brodsky, s. v., F. Zhang, a. nasjletti, and m. s. goligorsky. 2004. En- dothelium-derived microparticles impair endothelial function in vitro. American Journal of Physiology 286:H1910?1915. Brubakk, a. o., and o. s. Eftedal. 2000. Evaluation of reverse dive pro- files. in lang, m. a., and C. E. lehner, eds. Proceedings of the Reverse Dive Profiles Workshop. Washington: smithsonian institution. pp. 111?121. Brubakk, a. o., o. s. Eftedal, and U. Wisl?ff. 2007. Endothelium and div- ing. in aird, W., ed. Endothelial Biomedicine. Cambridge University Press. pp. 497?505. Buttolph, t. B., E. J. Dick, Jr., C. B. toner, J. R. Broome, R. Williams, Y. H. kang, and n. l. Wilt. 1998. Cutaneous lesions in swine after de- compression: histopathology and ultrastructure. Journal of Undersea and Hyperbaric Medicine 25:115?121. Chryssanthou, C., m. springer, and s. lipschitz. 1977. Blood-brain and blood-lung barrier alteration by dysbaric exposure. Undersea Biomed- ical Research 4:117?129. Cockcroft, J. R. 2005. Exploring vascular benefits of endothelium-derived nitric oxide. American Journal of Hypertension 18:177s?183s. Conkin, J. 1999. How should cutis marmorata be treated? Aviation, Space, and Environmental Medicine 70:939?940. Daniels, s. 1984. Ultrasonic monitoring of decompression procedures. Philosophical Transactions of the Royal Society of London 304:153?175. Davies, J. m. 1983. studies on bubble formation after decompression with special reference to the development and testing of the integrating ultrasonic pulse-echo imaging system for bubble detection. oxford: University of oxford Press. Dutka, a. J., P. m. kochanek, and J. m. Hallenbeck. 1989. inf luence of granulocytopenia on canine cerebral ischemia induced by air embo- lism. Stroke: A Journal of Cerebral Circulation 20:390?395. Eckmann, D. m., and s. l. Diamond. 2004. surfactants attenuate gas embolism-induced thrombin production. Anesthesiology 100:77?84. Ersson, a., m. Walles, k. ohlsson, and a. Ekholm. 2002. Chronic hy- perbaric exposure activates proinf lammatory mediators in humans. Journal of Applied Physiology 92:2375?2380. Francis, t. J. R., and s. J. mitchell. 2003. manifestations of Decompres- sion Disorders. in Brubakk, a. o., and t. s. neuman, eds. Bennett and Elliott?s Physiology and Medicine of Diving, 5th ed: saunders. pp. 578?600. ???. 2003. Pathophysiology of Decompression sickness. in Brubakk, a.o., and t.s. neuman, eds. Bennett and Elliott?s Physiology and Medicine of Diving, 5th ed. saunders. pp. 530?557. Helps, s. C., and D. F. gorman. 1991. air embolism of the brain in rab- bits pretreated with mechlorethamine. Stroke: A Journal of Cerebral Circulation 22:351?354. Hjelde, a., g. Bolstad, and a. o. Brubakk. 2002. the effect of air bub- bles on rabbit blood-brain barrier. Journal of Undersea and Hyperbaric Medicine 29:31?38. Hjelde, a., k. Bergh, a. o. Brubakk, and o. J. iversen. 1995. Complement activation in divers after repeated air/heliox dives and its possible rel- evance to DCs. Journal of Applied Physiology 78:1140?1144. Hope, a., and J. Risberg. 2006. Long Term Health Effects of Diving. The God?ysund 1993 Consensus Conference Revisited. Bergen: BoDoni. 302 pp. irgens, a., m. gr?nning, k. troland, E. sundal, H. nyland, and E. thors- en. 2007. Reduced health-related quality of life in former north sea divers is associated with decompression sickness. Occupational Medi- cine (London) 57(5):349?354. Jacobsen, g., J. E. Jacobsen, R. E. Peterson, J. H. mclellan, s. t. Brooke, t. nome, and a. o. Brubakk. 1997. Decompression sickness from saturation diving: a case control study of some diving exposure char- acteristics. Journal of Undersea and Hyperbaric Medicine 24:73?80. Jimenez, J. J., W. Jy, l. m. mauro, C. soderland, l. l. Horstman, and Y. s. ahn. 2003. Endothelial cells release phenotypically and quantita- tively distinct microparticles in activation and apoptosis. Thrombosis Research 109:175?180. Jirtle, R. l., and m. k. skinner. 2007. Environmental epigenomics and disease susceptibility. Nature Reviews 8:253?262. kojda, g., Y. C. Cheng, J. Burchfield, and D. g. Harrison. 2001. Dys- functional regulation of endothelial nitric oxide synthase (enos) expression in response to exercise in mice lacking one enos gene. Circulation 103:2839?2844. lopez, J., m. Percharde, H. m. Coley, a. Webb, and t. Crook. 2009. the context and potential of epigenetics in oncology. British Journal of Cancer 100:571?577. m?llerl?kken, a., v. J. Berge, a. Jorgensen, U. Wisl?ff, and a. o. Brubakk. 2006. Effect of a short-acting no donor on bubble formation from a saturation dive in pigs. Journal of Applied Physiology 101:1541?1545. nossum, v., s. koteng, and a. o. Brubakk. 1999. Endothelial damage by bubbles in the pulmonary artery of the pig. Journal of Undersea and Hyperbaric Medicine 26:1?8. nyquist, P. a., E. J. Dick, Jr., and t. B. Buttolph. 2004. Detection of leukocyte activation in pigs with neurologic decompression sickness. Aviation, Space, and Environmental Medicine 75:211?214. Pekna, m., l. nilsson, k. nilsson-Ekdahl, U. R. nilsson, and B. nilsson. m?llERl?kkEn anD EFtEDal: Risk oF DEComPREssion siCknEss ? 57 1993. Evidence for iC3 generation during cardiopulmonary bypass as the result of blood-gas interaction. Clinical and Experimental Im- munology 91:404?409. Richardson, B. 2003. impact of aging on Dna methylation. Ageing Re- search Reviews 2:245?261. Ross, J. a. s., J. i. macdiarmid, C. l. taylor, l. m. osman, a. murray, g. Waiter, s. semple, s. J. Watt, D. J. godden, a. lawson, W. R. adie, and J. Crawford. 2006. long-term Health Effects in United king- dom Professional Divers. in Hope, a., and J. Risberg, eds. Long-term Health Effects of Diving: The God?ysund 1993 Consensus Conference Revisited. Bergen: BoDoni. pp. 26?45. thorsen, t., H. klausen, R. t. lie, and H. Holmsen. 1993. Bubble-induced aggregation of platelets: effects of gas species, proteins, and decom- pression. Journal of Undersea and Hyperbaric Medicine 20:101?119. triggle, C. R., m. Hollenberg, t. J. anderson, H. Ding, Y. Jiang, l. Cero- ni, W. B. Wiehler, E. s. ng, a. Ellis, k. andrews, J. J. mcguire, and m. Pannirselvam. 2003. the endothelium in health and disease?a target for therapeutic intervention. Journal of Smooth Muscle Research 39:249?267. van Hulst, R. a., J. klein, and B. lachmann. 2003. gas embolism: patho- physiology and treatment. Clinical Physiology and Functional Imaging 23:237?246. Ward, C. a., D. mcCullough, and W. D. Fraser. 1987. Relation between complement activation and susceptibility to decompression sickness. Journal of Applied Physiology 62:1160?1166. Wienke, B. R. 1994. Basic Diving Physics and Applications. Flagstaff: Best Publishing Company. Wilmshurst, P., and P. Bryson. 2000. Relationship between the clinical features of neurological decompression illness and its causes. Clinical Science (London) 99:65?75. Wilmshurst, P. t., m. J. Pearson, k. P. Walsh, W. l. morrison, and P. Bry- son. 2001. Relationship between right-to-left shunts and cutaneous decompression illness. Clinical Science (London) 100:539?542. Wisl?ff, U., R. s. Richardson, and a. o. Brubakk. 2004. Exercise and ni- tric oxide prevent bubble formation: a novel approach to the preven- tion of decompression sickness? Journal of Physiology 555:825?829. Zwart, B. P. 2000. treatment of cutis marmorata. Aviation, Space, and En- vironmental Medicine 71(4):457. INTrOducTION Preventive measures to reduce the risk of decompression sickness (DCs) can involve several pre-dive procedures reported in some experimental studies such as oxygen breathing (Butler et al., 2006), exercise before diving (Duji? et al., 2004; Blatteau et al., 2005), intake of exogenous nitric oxide (Duji? et al., 2006) or prehydration (gempp et al., 2009). activation of the stress protein response (heat-shock protein, HsP) by mild hy- perthermia allows cells to resist subsequent metabolic insults that would otherwise be lethal, a phenomenon referred to as?preconditioning. HsP induction can lead to protec- tion against various types of stress such as ischemia, hypoxia, hyperoxia, inflammation (kregel, 2002), and even bubble-induced injury from decompression (Huang et al., 2003; medby et al., 2008). HsP are involved in many regulatory pathways and have an important function in controlling the folding and structure of proteins acting as chaperones. The most abundantly and best studied HsP are the 70 kDa families. Hsp70 are the most tem- perature sensitive and are highly inducible. moderate increases in temperature (>38?C) produce elevated Hsp70 levels which are dependent on the duration of heat stress. The aBSTRaCT. Heat stress is a nonpharmacological preconditioning strategy which can lead to pro- tection against various types of subsequent insults, such as ischemia, hypoxia, inflammation, drugs, and even bubble-induced injury from decompression. it has been suggested that the protective effect of heat exposure against DCs in rats could be related to biochemical processes involving heat shock proteins (HsP) of the 70-kDa range. The purpose of this study was to determine the efficacy of sauna-induced heat exposure prior to a simulated dive on bubble reduction and to ex- amine the adjustments in Hsp70 concentration and hemodynamic parameters. sixteen divers were compressed in a hyperbaric chamber to 400 kPa for 25 min and decompressed at a rate of 100 kPa. min-1 with a 4 min stop at 130 kPa. Each diver performed two dives, one with and one without pre-dive FiR dry sauna session for 30 min at 65?C, ending 1 hour prior to the dive. Brachial artery flow mediated dilation (FmD), blood pressure, and bodyweight measurements were taken before and after the sauna session along with blood samples for analysis of plasma volume, protein concen- trations, plasma osmolality, and plasma Hsp70. Circulating venous bubbles were detected with a precordial Doppler after surfacing. This study shows that a single pre-dive sauna session significantly decreases circulating bubbles after a chamber dive. Plasma Hsp70 significantly increased 2 h after sauna completion. The sauna session led to an extracellular dehydration, resulting in hypovolemia and bodyweight loss. a significant rise in FmD and a reduction in systolic blood pressure and pulse pressure were observed. inducing HsP from heat stress could offer a new way of reducing decom- pression sickness risk. sweat dehydration and no pathway could be also involved in this protective effect. Further investigation is required to elucidate the preponderant mechanism underlying this heat exposure induced reduction in bubble formation. Jean-Eric Blatteau, Emmanuel Gempp, and Jean- Michel Pontier, D?partement de M?decine Hyperbare, H?pital d?Instruction des Arm?es Sainte-Anne, 83800, Toulon Arm?es, France. Costantino Balestra, Environmental and Occupational Physiology Laboratory, I.S.E.K. Haute Ecole Henri Spaak, 1160, Brussels, Belgium. Tony Mets, Akademisch Ziekenhuis, Vrije Universiteit, 1120, Brussels, Belgium. Peter Germonpr?, Centre d?Oxyg?noth?rapie Hyperbare, H?pital Militaire Reine Astrid, 1120, Brussels, Belgium. inducing HsP for Protection against DCs: Heat Exposure before Diving Reduces Bubble Formation in man Jean-Eric Blatteau, Emmanuel Gempp, Jean-Michel Pontier, Costantino Balestra, Tony Mets, and Peter Germonpr? 60 ? tHE FUtURE oF Diving protective effect of heat exposure against DCs in rats could be related to increased expression of Hsp70 (Huang et al., 2003; medby et al., 2008). Hsp70 induction could also be involved in the mechanisms responsible for diving acclimatization after repeated compression-decompression cycles (su et al., 2004). moreover, it has been demonstrated that heat-inducible proteins are also able to interact with endothelial nitric oxide (no) pathway (Harris et al., 2003) which may influence the degree of bubble formation in hyperbaric conditions (Wisloff et al., 2003, 2004). it is well recognized that high environmental temperatures lead to sweat response resulting in dehydration. in a previous work, we have also shown that moderate dehydration resulting in stroke volume (sv) reduction induced by a pre-dive exercise could decrease venous circulating bubbles in divers (Blatteau et al., 2007). The purpose of this study was to determine the effective- ness of sauna-induced heat exposure prior to a simulated dive on bubble formation and to examine the adjustments in Hsp70 concentrations and haemodynamic parameters. mEThOdS diVing Procedure 16 military divers (38.4 years avg. and Bmi of 25 kg.m-2) underwent a simulated dive in a dry hyperbaric chamber to 400 kPa (30 m) for 25 min and a stop at 3 m for 4 min. Each diver performed two dives five days apart, one dive without sauna and one preceded by a sauna session. The order of the two dives was randomly allocated. Divers were instructed to avoid physical exertion during the 2 days that preceded each trial. sAunA Procedure subjects were seated for 30 min at 65?C with the head out of the cabin in a Far infrared-Ray dry sauna system. in similar conditions, deep body temperature rises about 1?C during sau- na exposure (tei et al., 1995). The sauna session ended 1 hour before the simulated dive. The divers were not allowed to drink water during the entire protocol. bubble detection Circulating bubbles were detected using a pulsed Doppler on the precordial area at 20, 40 and 60 min after surfacing (at rest and after 2 flexions). The bubble signals were graded ac- cording to the spencer scale before conversion to the kissman integrated severity score (kiss). This score takes into account the kinetics of the bubbles at the different recording times and is assumed to be a meaningful linearised measure of postdecom- pression intravascular bubble activity status that may be treated statistically (nishi et al., 2003). PhysiologicAl PArAmeters Blood samples were collected by venipuncture for hema- tocrit (Hct), hemoglobin (Hb), plasma proteins, and osmolal- ity after resting in a supine position for 20 min; measurements were determined before and 30 min after the sauna session. Per- cent changes in plasma volume (Pv) were calculated using an Hb-Hct transformation equation (Harrison et al., 1982). Heart rate (HR) and blood pressure (BP) were obtained immediately before each venipunction by using a portable monitoring sys- tem. Pulse pressure and mean arterial BP were calculated from the systolic and diastolic arterial BP. Weight measurements have been achieved immediately before and 35 min after the end of sauna. additional measurements were taken to evaluate the sauna exposure (without diving) in 10 divers for 1) arte- rial response to reactive hyperaemia (flow-mediated dilatation- FmD); 2) plasma HsP70 levels. FmD was determined before and 60 minutes after the sauna session, at the level of the right brachial artery, according to a protocol previously described (Corretti et al., 2002). This indirect method evaluates nitric oxide-dependent endothelial function. HsP70 measurements were done at 30 min, 2 h, 8 h, and 24 h after sauna completion. venous blood samples were sepa- rated from blood cells and stored at -70?C. HsP70 in serum was detected using an in-house sandwich Elisa method previously described (njemini et al., 2005). stAtisticAl AnAlysis Data are presented as mean ? s.D. Data were analysed using nonparametric statistics because of the small sample size. Wil- coxon signed rank test was used for paired data, whereas com- parisons in different times for HsP70 kinetics were evaluated by Friedman test (repeated measures anova on ranks). Differ- ences between groups were considered significant at p < 0.05. rESulTS none of the divers suffered from DCs after the dives. The maximum bubble count was observed 40 minutes after surfac- ing following the respective protocol (with or without sauna). a single session of sauna ending 1 hour before the dive signifi- cantly reduced - 27.2% kiss bubble grades at rest (mean kiss 1.95 versus 7.17, p = 0.031) and - 35.4% after flexions (mean kiss 3.6 versus 10.18, p = 0.039). only one diver showed a slight increase in venous bubble grade after performing sauna (Fig. 1). subjects performing a single sauna session had significantly elevated plasma HsP70 concentrations from baseline values 2 hours after the sauna completion (mean HsP values 1939 ver- sus 729 ng.l-1, p = 0.005) with no significant differences at 30 min, 8h and 24 h (Fig. 2). We observed a significant reduction of body weight by 0.6% after sauna (-450 ? 18 g, mean ? s.D. p < 0.001). Plasma BlattEaU Et al. : HEat ExPosURE BEFoRE Diving ? 61 and/or alter the population of gaseous nuclei from which bub- bles form (Blatteau et al., 2006, 2007). dehydrAtion although dehydration is commonly proposed as a risk factor for DCs in divers, there are no data that support this assertion in man. animal studies are few and give contradic- tory results (Broome et al., 1995; leni et al., 2001; Fahlman and Dromsky, 2006; skogland et al., 2008). We have recently shown that moderate dehydration and hypovolemia induced by a pre-dive exercise might be involved in reducing post-dive circulating bubbles (Blatteau et al., 2007). From the above find- ings, we hypothesize that moderate dehydration could be ben- eficial on bubble formation, while severe dehydration appears to increase the risk of DCs. The uptake or release of gas by a particular tissue depends on both the rate of blood flow to the tissue and the rate of gas diffusion into the tissue from blood. it may be seen that if the blood flow is lower, the rate of inert gas uptake would be slower and consequently bubble formation would be reduced (Francis et al., 1990). Dehydration-induced volume decreased by 2.7 % after sauna (p < 0.001). Plasma proteins significantly increased by 3% after sauna (73.1 ? 2.9 versus 70.9 ? 2.8 mg.l-1, mean ? s.D. p < 0.001), whereas there were no significant differences in plasma osmolality before and after sauna (Fig. 3) systolic blood pressure and pulse pressure decreased signifi- cantly (112 ? 10 versus 119 ? 13 mmHg, p = 0.013 and 40 ? 17 versus 46 ? 19 mmHg, mean ? s.D. p = 0.005, respectively) whereas diastolic and mean blood pressure remained unchanged after sauna. HR was not modified by sauna. We found a significant increase of FmD (Fig. 4) observed 1 hour after the sauna completion from baseline values measured 15 min before the sauna exposure (13 ? 2.8% versus 7.7 ? 3 %, mean ? s.D. p = 0.002). dIScuSSION The main result of our study is that a single pre-dive sauna session significantly decreased circulating bubbles. our working hypothesis is that heat preconditioning could contribute to re- duce bubble burden by reducing nitrogen load during the dive - 5 0 5 10 15 20 25 30 35 40 45 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 divers b u b b le s c o re K IS S FIgURe 1. Individual KISS bubble scores (after flexions), follow- ing hyperbaric exposure to 400 kpa for 25 min. Black bars repre- sent controls and grey bars conditions with sauna ending 1h prior to the dives. 0 500 1000 1500 2000 2500 3000 3500 Pre s auna Pos t 30m in Pos t 2h Pos t 8h Pos t 24h HS P70 ng/L * FIgURe 2. expression of plasma hSp70 (mean ? sem) before and after FIR sauna exposure (65?/30min) in 10 subjects. * denotes p < 0.05. - 3 - 2 - 1 0 1 2 3 % Bo d y weigh t Plasma volu me Plasma proteins * * * - - - -- - -- - -- - -- -- - - -- - -- - -- - -- -- - - -- - -- - -- - -- -- - - -- - -- - -- - -- -- - - - - - -- - -- - - FIgURe 3. The sauna session led to an extracellular dehydration resulting in hypovolemia (- 2.7% of plasma volume) and body weight loss (- 0.6%). 80 90 100 110 120 Pr e sauna Pos t 1h * FMD % of control values FIgURe 4. Brachial artery Flow mediated Dilation (mean ? sem) before and 1h after FIR sauna exposure (65?/30min) in 10 subjects. * denotes p < 0.05. 62 ? tHE FUtURE oF Diving hypovolemia reduces stroke volume (Jimenez et al., 1999; Blat- teau et al., 2007). Because heart rate was unchanged after the sauna session in our study, we hypothesized that blood flow was reduced at the start and during the dive, thus limiting inert gas load and bubble formation afterwards. nitric-oxide (no) PAthwAy it has been demonstrated that endothelial no, an impor- tant vasodilator with anti-atherogenic properties, can attenu- ate bubble formation and DCs incidence probably by reducing gaseous nuclei from which bubbles form (Wisloff et al., 2003; 2004). We observed a 50% increase in FmD, suggesting an no-mediated effect on endothelial function after a single sauna session. This was partially reflected in the observed reduction of systolic blood pressure and pulse pressure observed 30 minutes after the sauna. it remains however unclear whether this no- mediated endothelial function improvement is directly related to the HsP increase, or whether a different process is involved. heAt-shock Proteins numerous regulatory pathways may be considered in the real bioprotective mechanisms of HsP (Fig. 5). it is assumed that endothelial nitric oxide synthase is regulated by inducible HsP. moderate heat shock (core temperature: 42?C, 15 min) in rats increases HsP90 and HsP70 protein levels and vascu- lar enos expression (Harris et al., 2003). However, this study also demonstrates that a longer heat exposure (42?C, 1 hour) of endothelial cells results in an increase in enos protein con- tent with no change in HsP70 and HsP90 expression, suggest- ing that the heat-induced change in enos expression could be induced by other mechanisms. Recent studies demonstrate in animals models of acute lung injury that the activation of HsP may attenuate the damage caused by oxidative stress in the lung by a direct inhibition of proinflammatory mediators (including adhesion molecules such as iCam-1), and also by an inhibi- tion in the sequestration of neutrophils in pulmonary capillaries (Pespeni et al., 2005). From the above findings, the role of HsP appears more related to the attenuation of tissue reaction to vascular bubbles, than in a direct bubble reduction process. cONcluSION This study shows that a single pre-dive sauna session signifi- cantly decreases circulating bubbles after a chamber dive, which may reduce the risk of decompression sickness. Heat exposure induced-dehydration and no pathway could be involved in this protective effect. it is unclear whether HsP act directly on the bubble formation process or influence the attenuation of tissue reaction to vascular bubbles (Fig. 5). in this way, a pre- dive aerobic exercise performed within 2 hours, including sweat dehydration, a potential increase in body temperature and HsP and a rise in endothelial no may therefore provide some of the observed benefits as identified in this paper. Acknowledgments We would like to thank the participating divers for their patience and sweat (special thanks to Daniel Caturan and Ju- lien Hugon); the students helping out by the experiments (amandine Caveau and Jennifer J?rome) and the team of the hyperbaric center sainte-anne for their excellent technical as- sistance during the course of the study. This study was previ- ously published in Aviation, Space, and Environmental Medicine 79(December 2008). (Blatteau J-?, gempp E, Balestra C, mets t, germonpre P. Predive sauna and venous gas bubbles upon decompression from 400 kPa. Aviat Space Environ Med 2008; 79:1100?1105). lITEraTurE cITEd Blatteau, J.-E., E. gempp, C. Balestra, t. mets, P. germonpre. 2008. Predive sauna and venous gas Bubbles upon Decompression from 400 kPa. Avia- tion, Space, and Environmental Medicine, 79:1100?1105. Blatteau, J. E., E. gempp, F. m. galland, J. m. Pontier, J. m. sainty, and C. Robinet. 2005. aerobic exercise 2 hours before a dive to 30msw decreases bubble formation after decompression. Aviation, Space, and Environmental Medicine 76:666?669. Blatteau, J. E., J. B. souraud, E. gempp, and a. Bousssuges. 2006. gas nuclei, their origin, and their role in bubble formation. Aviation, Space, and Envi- ronmental Medicine 77:1068?1076. Blatteau, J. E., a. Boussuges, E. gempp, J. m. Pontier, o. Castagna, C. Robinet, F.m. galland, and l. Bourdon. 2007. Haemodynamic changes induced by submaximal exercise before a dive and its consequences on bubble forma- tion. British Journal of Sports Medicine 41(6):375?379. Broome, J. R., C. l. kittel, and E. J. Dick. 1995. Failure of pre-dive hydration status to influence neurological DCi rate in pigs. Journal of Undersea and Hyperbaric Medicine 22:52. Butler, B. D., t. little, v. Cogan, and m. Powell. 2006. Hyperbaric oxygen pre-breathe modifies the outcome of decompression sickness. Journal of Undersea and Hyperbaric Medicine 33:407?417. Corretti, m. C., t. J. anderson, E. J. Benjamin, D. Celermajer, and F. Char- bonneau. 2002. guidelines for the ultrasound assessment of endothelial- dependent flow-mediated vasodilation of the brachial artery: a report of FIgURe 5. Regulatory pathways of bioprotective hSp mechanisms. Heat HS P N O d eh ydr atio n v aso dilation Elimin ation gas nuclei h ypo vo lemia in er t gas uptak e Bu bb le f or mation p o s t d i v ep red i v e b loo d flow ? ? BlattEaU Et al. : HEat ExPosURE BEFoRE Diving ? 63 the international Brachial artery Reactivity task Force. Journal of the American College of Cardiology 39:257?265. Duji?, ?., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n. m. Petri, U. Wisl?ff, and a. o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555:637?642. Duji?, ?., i. Palada, Z. valic, D. Duplan?i?, a. obad, U. Wisl?ff, and a. o. Brubakk. 2006. Exogenous nitric oxide and bubble formation in divers. Medicine and Science in Sports and Exercise 38:1432?1435. Fahlman a., and D. m. Dromsky. 2006. Dehydration effects on the risk of se- vere decompression sickness in a swine model. Aviation, Space, and Envi- ronmental Medicine?77(2):102?106. Francis, t. J. R., a. J. Dutka, and J. m. Hallenbeck. 1990. Pathophysiology of decompression sickness. in Bove, a. a., and J. C. Davis, eds. Diving Medi- cine. Philadelphia: saunders. Pp. 170?187. gempp, E., J. E. Blatteau, J. m. Pontier, C. Balestra, and P. louge. 2009. Pre- ventive effect of pre-dive hydration on bubble formation in divers. British Journal of Sports Medicine 43:224?228. Harris, m. B., m. a. Blackstone, H. Ju, v. J. venema, and R. C. venema. 2003. Heat-induced increases in endothelial no synthase expression and activ- ity and endothelial no release. American Journal of Physiology - Heart and Circulatory Physiology 285:333?340. Harrison, m. H., m. J. graveney, and l. a. Cochrane. 1982. some sources of error in the calculation of relative change in plasma volume. European Journal of Applied Physiology 50:13?21. Huang, k. l., C. P. Wu, and Y. l. Chen. 2003. Heat stress attenuates aire bub- ble-induced acute lung injury: a novel mechanism of diving acclimatiza- tion. Journal of Applied Physiology 94:1485?1490. Jimenez, C., B. melin, n. koulmann, a. m. allevard, J. C. launay, and g. savourey. 1999. Plasma volume changes during and after acute variations of body hydration level in humans. European Journal of Applied Physiology and Occupational Physiology 80:1?8. kregel, k. C. 2002. invited review: Heat shock proteins: modifying factors in physiological stress responses and acquired thermotolerance. Journal of Applied Physiology 92:2177?2186. leni, P., J. P. menu, F. laforest, B. Zouani, and J. l. m?liet. 2001. Extracellular dehydration induces functional motor spinal deficit after severe decom- pression in the rabbit. Proceedings of the XXVIIth Annual Meeting of the European Underwater and Baromedical Society, Hamburg, germany. pp. 13?17. medby, C., a. Bye, U. Wisl?ff, and a. o. Brubakk. 2008. Heat shock increases survival in rats exposed to hyperbaric pressure. Diving and Hyperbaric Medicine 38(4):189?193. nishi, R. Y., a. o. Brubakk, and o. s. Eftedal. 2003. Bubble detection. in Brubakk, a. o., and t.s. neuman, eds. Bennett and Elliot?s Physiology and Medicine of Diving, 5th ed. london: WB saunders. pp. 501?529. njemini, R., C. Demanet, and t. mets. 2005. Comparison of two Elisas for the determination of Hsp70 in serum. Journal of Immunological Methods 306:176?182. Pespeni, m., m. Hodnett, and J. F. Pittet. 2005. in vivo stress preconditioning. Methods 35:158?164. skogland, s., l. B. stuhr, H. sundland, R. E. olsen, and a. Hope. 2008. venous gas emboli in normal and dehydrated rats following decompression from a saturation dive. Aviation, Space, and Environmental Medicine 79:565?569. su, C. l., C. P. Wu, s. Y. Chen, B. H. kang, k. l. Huang, and Y. C. lin. 2004. acclimatization to neurological decompression sickness in rabbits. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 287:1214?1218. tei, C., Y. Horikiri, J. C. Park, J. W. Jeong, k. s. Chang, Y. toyama, and n. tanaka. 1995. acute hemodynamic improvement by thermal vasodilation in congestive heart failure. Circulation 91(10): 2582?2590. Wisl?ff, U., R. s. Richardson, and a. o. Brubakk. 2003. nos inhibition in- creases bubbles formation and reduces survival in sedentary but not exer- cised rats. Journal of Physiology 546(2):577?582. ???. 2004. Exercise and nitric oxide prevent bubble formation: a novel ap- proach to the prevention of decompression sickness? Journal of Physiology 555(3):825?829. INTrOducTION Decompression sickness is a clinical diagnosis associated with a number of different symptoms. However, in serious decompression sickness (DCs type ii), neurological symptoms dominate (moon and gorman, 2003). Francis and mitchell (2003) state that if emboli are responsible for the pathologic findings in DCs, the brain rather than the spinal cord should be embolized since it constitutes more than 98 % of the mass of human Cns and receives 75-85 times the blood flow of the spinal cord. However, the most common site for type ii DCs is in the spinal cord which is affected in as many as 20-50% of cases that develop type ii DCs (Jallul et al., 2007). The spinal cord white matter is especially vulnerable to bubble forma- tion because nitrogen is highly soluble in myelin in addition to the poor collateral blood supply of the spinal cord (Barratt et al., 2002). Even though damage is more commonly observed in the spinal cord, bubbles seen in the carotid artery in divers 4 hours after decompression raises concern (Brubakk et al., 1986). Reversibility of blood brain barrier (BBB), cerebral microcirculation, and histologi- cal changes after decompression were shown in a study using rats (nohara and Yusa, 1997). immediately after decompression there were changes in the brain. These find- ings were not seen after 1 and 3?6 hours but reappeared between 48?72 hours after decompression. Histological examinations have also been performed on goats, where Glial fibrillary acidic protein (gFaP) immunohistochemistry staining revealed that lesions following decompression sickness in Cns material are more likely to be found in animals exhib- iting clinical signs of decompression sickness than those that do not (Woodger et al., 2001). These results are, however, based on 3 out of 17 animals that had signs of DCs and the general conclusion from this study is that there are no significant correlations between DCs and neuropathological lesions in the brain or spinal cord (Blogg et al., 2004). The same conclusion was reached by m?rk et al. (1994), who were unable to detect any neuropathological changes in the spinal cord of 10 amateur divers and 10 experienced divers. aBSTRaCT. in serious decompression sickness (DCs type ii), neurological symptoms dominate. However, not much is known about brain injury after decompression. This pilot experiment was done to establish a method for evaluation of central nervous system (Cns) effects of decompression using magnetic resonance imaging (mRi) techniques. Rats were compressed on air in a hyperbaric chamber and decompressed following a decompression profile known to give considerable bubble production. The pulmonary arteries were monitored for bubbles by ultrasound and the brains were scanned by mRi. no signs of injury in the brain were seen on mRi after decompression. Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Olav Kyrres gt. 9, N-7489, Trondheim, Norway. lack of signs of Brain injury in Rats on mRi after Decompression Marianne B. Havnes, Marius Wider?e, Marte Thuen, Andreas M?llerl?kken, and Alf O. Brubakk 66 ? tHE FUtURE oF Diving Jallul et al. (2007) reported no signs of mRi abnormalities in the spinal cord or brain of a male human diver from mRi scans taken within 24 hours after injury (paraplegic after rec- ompression). in mRi scans taken three weeks after injury, areas of abnormal high intensity appeared both in the spinal cord and the brain. another diver with serious neurological DCs had ab- normally high signals in the white matter in the brain from mRi scans taken 3 days after the accident. This was not visible on the mRi scan performed 8 weeks after the injury. Based on the observation that mRis of divers with DCs show multiple cerebral infarctions in the terminal and border zones of the brain arteries, kohshi and Wong (2001) proposed that following decompression, arterialized bubbles pass the lungs and heart to the brain. Ultrasound has previously been applied in order to evaluate vascular gas bubble formation in the vascular system. However, the sensitivity of the method is limited by the resolution of the instrument and the amount and size of bubbles formed. The technology has been well suited for moving bubbles, but lately new ultrasound technology is capable of detecting stationary bubbles as well (masoy et al., 2008). Thus, mRi would appear to be a superior method. The presence of gas bubbles in the Cns may cause localized inhomogeneity of the magnetic field that can be detected on t 2 *-weighted mR images. t 2 -weighted images are commonly used to detect oedema, brain infarctions, and structural changes in the brain. manganese-enhanced mRi (mEmRi) has previously been used to detect ongoing ischemic neuronal death from stroke in adult rats (aoki et al., 2003; 2004) and recent studies sug- gest that mEmRi can be used to detect inflammatory processes with activated microglia/macrophages and reactive astrocytes in the rat brain (Haapanen et al., 2007; Wideroe et al., 2009; Yang and Wu, 2007). another mRi technique is Diffusion ten- sor imaging (Dti) where the diffusion of water is measured in several directions in 3 dimensions. This method can be used to differentiate between brain tissues and is particularly useful for visualization and evaluation of white brain matter. Diffu- sion along the white matter tracts has little restrictions, whereas diffusion perpendicular to the tracts is highly restricted by the myelin sheets and cell walls. This results in fast diffusion mainly along the tracts and can be calculated and expressed as a high Fractional anisotropy (Fa). a change or difference in Fa value in a white matter area indicates axonal damage or rearrange- ment (mori and Zhang, 2006). in the present preliminary study, injury to the brain af- ter decompression was investigated with mRi and ultrasound. our hypothesis was that neurological decompression sickness (type ii DCs) is caused by the organism?s response to vascu- lar gas bubbles. it follows that if vascular gas bubble formation can be reduced or eliminated, neurological DCs will be less likely to occur and also reduce the risk of long term sequelae. Previous studies have demonstrated that vascular bubble for- mation can be significantly reduced or eliminated by physi- cal exercise and nitric oxide in rat, pig, and man (Dujic et al., 2004; mollerlokken et al., 2006; Wisloff and Brubakk, 2001; and Wisloff et al., 2004). mEThOdS a total of 9 rats were compressed on air in a hyperbaric chamber and decompressed following a decompression profile known to give considerable bubble production. Ultrasound was used to monitor and grade bubbles in the pulmonary artery for 30 minutes after decompression, using the grading method of Eftedal and Brubakk (1997). Five animals served as controls on the mRi scan. after observation, the rats were transferred to the mR Fa- cility, ntnU. Before the mRi acquisition, the rats were anes- thetized (4% isoflurane induction) and a 25 Ch neoflon was inserted in the tail vein of the rats. mRi was performed on a 7 tesla magnet (Biospec 70/20 as, Bruker Biospin mRi, Ettlin- gen, germany) with water-cooled (Bga-12, 400 mt/m) gradi- ents. a 72-mm volume resonator was used for RF transmission and an actively decoupled rat head surface coil was used for RF reception (Bruker Biospin mRi). During scanning, the anaes- thetized (2% isoflurane in 30% o 2 , 70% n 2 ) rats lay prone in the dedicated water-heated rat bed and the head of every animal was fixed in the same position with inbuilt earplugs, tooth bar, and nose-mask. This assured the same placement of the head of the animals within the magnet from scan to scan. after a gradient echo FlasH pilot scan (acquisition time 1 min), a series of t2*-weighted images were obtained using a multi-gradient echo (mgE) sequence with the following pa- rameters: tE = 4/11/18/25/32/39/46/53/60/67/74/81 ms; tR = 1650 ms; 2 averages; acquisition time 7 min 55 sec. t 2 -maps were then obtained using a RaRE sequence with RaRE-factor = 4; tE = 25/50/75 ms; tR = 4000 ms; 3 averages; acquisition time 7 min 12 sec. apparent Diffusion Coefficient maps were obtained using a Echo Planar imaging (EPi) sequence with 3 directions and 6 b-values (100/200/400/600/800/1000 ms); tE = 53.47 ms; tR = 3000 ms; 6 averages; acquisition time 8 min 24 sec. For all these scans, 17 coronal slices were acquired with slice thickness of 1 mm. The field of view (Fov) was 40 ? 30 mm and the acquisition matrix (mtx) was 256 ? 144 giving an isotropic in-plane resolution of 156 ? 156 ?m2. The animals were then returned to the animal facility and closely observed for any signs of decompression sickness. The next day, animals were injected with a single dose of 40 mg mnCl 2 (# 7773-01-5, sigma-aldrich inc., st. louis, Usa) per kg bodyweight (~318 ?mol mn2+/kg) at a concentration of 100 mm intraperitoneally to serve as mRi contrast for later scans. surviving animals were returned to the mR Facility after 7 and 14 days for new mRis. at both time-points, t 2 -maps and t 2 *-weighted images were acquired with the same parameters as described above. to evaluate manganese-uptake and inflamma- tion, a 3D data set was obtained on day 7 using a t 1 -weighted gradient echo FlasH sequence with flip-angle = 30?, tR = 12 ms, tE = 3.25 ms. Fov = 30 ? 35 ? 20 mm and acquisition HavnEs Et al. : laCk oF BRain inJURY in Rats ? 67 changes, infarctions, or manganese uptake that could suggest ongoing inflammatory processes. The histological investigation of the brain has not yet been performed. dIScuSSION This pilot experiment was done to establish a method for evaluating Cns effects of decompression using mRi techniques. The fact that we could not see any oedema or damage on the mRi may suggest that the decompression experienced was in- sufficient to cause any damage to the brain. on the other hand, injury to the brain caused by decompression may have been too subtle to be detected using the current mRi methods. Fur- ther evaluation of the experiment using immunohistochemical matrix was 192 ? 168 ? 96 with zero-filling to 192 ? 224 ? 128, the interpolated resolution was 156 ?m isotropic. acquisition time was 52 min with 16 averages. The B1-field of the volume- coil was considered homogeneous within the field of view, while the spatially inhomogeneous sensitivity of the surface-coil used in the 3D t 1 -weighted FlasH acquisition was corrected for using two additional scans in coupled and single coil operation: 3D t 1 -weighted FlasH sequences: Flip-angle = 30?, tR = 12 ms, tE = 3.25 ms, matrix size 32 ? 32 ? 32, acquisition-time was 2 min for each scan. Fovs were the same as those used to obtain the 3D data set. on both 7 and 14 days after the decompression, Diffusion- tensor imaging (Dti) was performed to evaluate specific white matter injury and to look for changes in white matter. The Dti was acquired with an EPi sequence using 30 directions and b = 1000 ms; 5 b0 images and Fov = 40 ? 40 mm; mtx = 172 ? 172 giving a resolution of 233 ? 233 ?m2. on day 7 tE = 37.5; tR = 3000 ms; 17 slices at 1 mm were acquired with 2 averages giving an acquisition time of 14 min. on day 14 tE = 37.5; tR = 5000 ms; 33 slices at 0.5 mm were acquired with 6 averages giving an acquisition time of 1 hr 10 min. rESulTS bubble FormAtion And clinicAl signs a total of 9 rats were decompressed, of which 3 died and 6 had bubble grades from 2 to 4. Three rats died because of con- siderable amounts of bubbles, 2 rats died immediately after the dive, and a third rat had neurological symptoms of DCs in the form of paralysis of the hind legs and died after the first mRi scan. The remaining 6 animals had no clinical symptoms. mAgnetic resonAnce imAging localized reduction of signal intensity was seen as stripes through the cortex as well as in the areas corresponding to the main cerebral arteries on the t 2 * images (Fig. 1). These findings were more pronounced in the rats exposed to decompression. such signal reduction on t 2 *-weighted images are normally re- lated to increased amounts of deoxygenated haemoglobin in the blood. However, air bubbles in the blood may also cause the same effect due to a local disturbance of the magnetic field. neither aDC-maps nor t 2 -maps showed any signs of oede- ma immediately after decompression. There were also no signs of oedema, infarctions, or structural changes at later time points assessed by the t 2 images (Fig. 2). Diffusion-tensor images in some of the rats (in both groups) showed reduced fractional an- isotropy in the external capsule, indicative of injury to the white matter tracts (Fig. 3). However, the relative sensitivity of this particular area to imaging artefacts makes this a most uncertain finding. The high-resolution 3D t 1 -weighted images acquired on day 7 after decompression showed no sign of structural FIgURe 1. T 2 *-map of a decompressed rat. localized reduction of signal intensity was seen as stripes through the cortex as well in the areas corresponding to the main cerebral arteries on the images. FIgURe 2. T 2 -map of brain of a decompressed rat. No signs of damage or oedema can be seen. 68 ? tHE FUtURE oF Diving investigations might be able to provide us with more informa- tion on the effect of decompression on rat brain function and validate the mRi results. in spite of a considerable amount of vascular gas bubbles detected in the pulmonary artery and clinical signs of neuro- logical DCs in one rat, the preliminary results show no brain damage after decompression. However, there was a signal re- duction on t 2 *-weighted images. This is normally related to in- creased amounts of deoxygenated haemoglobin in the blood, but air bubbles in the blood may also cause the same effect, due to a local disturbance of the magnetic field. it has previ- ously been shown that if tissue oxygenation falls critically low, this can lead to blood-brain barrier dysfunction, inflammation, demyelination, and eventually, axonal damage (James, 2007). it is tempting to speculate whether repeated diving with low levels of oxygen in brain tissue would give the latter damages. The number of experimental animals is low. still, these re- sults raise questions whether the rats have a protective mecha- nism in the brain against bubbles or whether bubbles constitute any threat at all. it is possible that the major damage is in the spinal cord, as has been shown in some human studies. Hu- man divers have reported pains and problems caused by decom- pression, like loss of concentration, blurred vision, and other problems that are possibly caused by injuries to the brain. it is possible that these changes in the brain are too subtle to be observed in mRi scans or that develop over time with repeated diving. Evidence of brain damage caused by decompression is conflicting. Bubbles observed in the carotid artery may lead to injury (Brubakk et al., 1986), but a number of mRi stud- ies show no visible damage to the brain after decompression in divers and animals (Hutzelmann et al., 2000; Rinck et al., 1991; Yoshiyama et al., 2007). Erdem et al. (2009) proposed that higher incidence of lesions in the cerebral white matter of divers is evidence of cumulative, subclinical injury to the neu- rological system that may affect the long-term health of military and recreational divers. cONcluSION The number of animals in this study is too low to draw any conclusion whether the mRi is a useful method to study brain injury after decompression in rats. no histology of the brains of these animals has yet been performed. to properly investigate the role of vascular gas bubbles in neurological decompression sickness, it is necessary to perform further experiments with better control of the size and location of the possible damage. The mechanisms behind decompression sickness are com- plex, and the individual susceptibility for the disease is diverse. in order to study the effect of decompression on Cns we have to know that Cns is affected. By injecting air bubbles in ap- propriate amounts directly into the carotid artery, we can ensure that there are bubbles directed into the circulation of the brain. in order to visualize the acute effect of bubbles, one can perform the injection during the mRi scan. Histology should also be prepared to investigate more subtle changes. lITEraTurE cITEd aoki, i., s. naruse, and C. tanaka. 2004. manganese-enhanced magnetic reso- nance imaging (mEmRi) of brain activity and applications to early detec- tion of brain ischemia. NMR in Biomedicine 17(8):569?580. aoki, i., t. Ebisu, C. tanaka, k. katsuta, a. Fujikawa, m. Umeda, m. Fuku- naga, t. takegami, E.m. shapiro, and s. naruse. 2003. Detection of the anoxic depolarization of focal ischemia using manganese-enhanced mRi. Magnetic Resonance in Medicine 50(1):7?12. Barratt, D. m., P. g. Harch, and k. van meter. 2002. Decompression illness in FIgURe 3. Fractional anisotropy (Fa) maps of two rat brains. a. Top: Control animal shows intact white matter with high Fa values (bright areas). B. Bottom: 1 week post-decompression shows a reduced fractional anisotropy in the external capsule and corpus callosum, indicative of injury to the white matter tracts, but may also represent imaging artefacts and needs fur- ther investigation. HavnEs Et al. : laCk oF BRain inJURY in Rats ? 69 divers: a review of the literature. Neurologist 8(3):186?202. Blogg, s. l., g. a. loveman, F. m. seddon, n. Woodger, a. koch, m. Reuter, m. gennser, and m.g. White. 2004. magnetic resonance imaging and neuropathology findings in the goat nervous system following hyperbaric exposures. European Neurology 52(1):18?28. Brubakk, a. o., R. Peterson, a. grip, B. Holand, J. onarheim, k. segadal, t. D. kunkle, and s. tonjum. 1986. gas bubbles in the circulation of divers after ascending excursions from 300 to 250 msw. Journal of Applied Physi- ology 60(1):45?51. Duji?, Z., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n.m. Petri, U. Wisl?ff, and a.o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555(3):637?642. Eftedal, o., and a. o. Brubakk. 1997. agreement between trained and un- trained observers in grading intravascular bubble signals in ultrasonic im- ages. Journal of Undersea and Hyperbaric Medicine 24(4):293?299. Erdem, i., s. Yildiz, g. Uzun, g. sonmez, m. g. senol, m. mutluoglu, H. mutlu, and B. oner. 2009. Cerebral white-matter lesions in asymptomatic military divers. Aviation, Space, and Environmental Medicine 80(1):2?4. Francis, t. J. R., and s. J. mitchell. 2003. Pathophysiology of Decompression sickness. in: Brubakk, a.o., and t.s. neuman, eds. Bennett and Elliott? s Physiology and Medicine of Diving (5th ed) london: saunders. pp. 530?557. Haapanen, a., U. a. Ramadan, t. autti, R. Joensuu, and J. tyynela. 2007. In vivo mRi reveals the dynamics of pathological changes in the brains of cathepsin D-deficient mice and correlates changes in manganese-en- hanced mRi with microglial activation. Journal of Magnetic Resonance Imaging 25(7):1024?1031. Hutzelmann, a., k. tetzlaff, m. Reuter, s. muller-Hulsbeck, and m. Heller. 2000. Does diving damage the brain? mR control study of divers? central nervous system. Acta Radiologica 41(1):18?21. Jallul, s., a. osman, and W. El-masry. 2007. Cerebro-spinal decompression sickness: report of two cases. Spinal Cord 45(1):116?120. James, P. B. 2007. Hyperbaric oxygenation in fluid microembolism. Neurological Research 29(2):156?161. kohshi, k., and R. m. Wong. 2001. [Central nervous system involvement in decompression illness]. No To Shinkei [Brain and Nerve] 53(8):715?722. m?s?y, s. E., o. standal, P. nasholm, t. F. Johansen, and B. angelsen. 2008. sURF imaging: in vivo demonstration of an ultrasound contrast agent detection technique. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 55(5):1112?1121. m?llerl?kken, a., v. J. Berge, a. Jorgensen, U. Wisl?ff, and a. o. Brubakk. 2006. Effect of a short-acting no donor on bubble formation from a satu- ration dive in pigs. Journal of Applied Physiology 101(6):1541?1545. moon, R. E., and D. F. gorman. 2003. treatment of the Decompression Disor- ders. in Brubakk, a. o., and t.s. neuman, eds. Bennett and Elliott? s Physi- ology and Medicine of Diving (5th ed) london: saunders. pp. 600?651. mori, s., and J. Zhang. 2006. Principles of diffusion-tensor imaging and its ap- plications to basic neuroscience research. Neuron 51(5):527?539. mork, s. J., i. morild, a. o. Brubakk, s. Eidsvik, and H. nyland. 1994. a histopathologic and immunocytochemical study of the spinal cord in ama- teur and professional divers. Journal of Undersea and Hyperbaric Medicine 21(4):391?402. nohara, a., and t. Yusa. 1997. Reversibility in blood-brain barrier, microcircu- lation, and histology in rat brain after decompression. Journal of Undersea and Hyperbaric Medicine 24(1):15?21. Rinck, P. a., R. svihus, and P. de Francisco. 1991. mR imaging of the cen- tral nervous system in divers. Journal of Magnetic Resonance Imaging 1(3):293?299. Wideroe, m., o. olsen, t. B. Pedersen, P. E. goa, a. kavelaars, C. Heijnen, J. skranes, a. m. Brubakk, and C. Brekken. 2009. manganese-enhanced magnetic resonance imaging of hypoxic-ischemic brain injury in the neo- natal rat. Neuroimage 45(3):880?890. Wisl?ff, U., and a. o. Brubakk. 2001. aerobic endurance training reduces bub- ble formation and increases survival in rats exposed to hyperbaric pressure. Journal of Physiology 537(2):607?611. Wisl?ff, U., R. s. Richardson, and a. o. Brubakk. 2004. Exercise and nitric oxide prevent bubble formation: a novel approach to the prevention of de- compression sickness? Journal of Physiology 555(3):825?829. Woodger, n., a. C. Palmer, and W. F. Blakemore. 2001. a double-blind study of the neuropathology in goats exposed to hyperbaric conditions. off- shore technology Report 2001/090. suffolk: Health and safety Executive Books. 28 pp. Yang, J., and E. x. Wu. 2007. manganese-enhanced mRi detected the gray mat- ter lesions in the late phase of mild hypoxic-ischemic injury in neonatal rat. Conference Proceedings IEEE Engineering in Medicine and Biology Society. pp. 51?54. Yoshiyama, m., s. asamoto, n. kobayashi, H. sugiyama, H. Doi, H. sakagawa, and m. ida. 2007. spinal cord decompression sickness associated with scuba diving: correlation of immediate and delayed magnetic resonance imaging findings with severity of neurologic impairment - a report on 3 cases. Surgical Neurology 67(3):283?287. INTrOducTION ideas and views about the effects of diving on humans in relation to exercise, endothelial, and heart function were presented at the Haldane symposium in trondheim, norway. Recent information was reviewed on the effect of exercise before, during, and after diving on gas bubble formation that occurs after surfacing, as well as the acute changes in cardiovascular function after single, open-water, com- pressed air dives. our recent studies (Duji? et al., 2004) have shown that pre-dive exercise and exercise performed during a decompression stop (Duji? et al., 2005b) may significantly reduce bubble formation. These findings are important because ve- nous gas bubbles (vgB) can be found in the vasculature and the right heart after the majority of recreational and professional dives and are statistically related with central nervous system decompression sickness (DCs), especially if a large amount of bubbles is found in the pulmonary artery (nishi et al., 2003). vgB, when occurring without any acute clinical signs, have been termed ?silent bubbles? (Behnke, 1951) and have in general been assumed not to have any negative effects. However, vascular bubbles will lead to a reduction of endothelial function in the pulmonary artery of the pig (nossum et al., 1999) and the brachial artery in man (Brubakk et al., 2005), even without the passage of bubbles through the pulmonary circulation. Endothelial dysfunction represents one of the earliest events in the pathogenesis of cardiovascular disease and is observed in several clinical conditions. keeping in mind that scuba div- ing is performed regularly by millions of recreational and professional divers world- wide and that the diving population is often older and may include unfit individuals or even individuals with some diseases such as hypertension, the potential acute or aBSTRaCT. generally, the idea that exercising before, during, and after diving is not advised is based on the assumption that it would increase the incidence of decompression sickness. our data show that exercise performed before or during diving results in a reduction of venous gas bubble formation after diving. Pre-dive administration of an acute nitric oxide donor had the same effect. Exercise performed after diving did not increase the number of gas bubbles and there was no evi- dence for an increased risk for right-to-left intrapulmonary shunting. in addition, we have shown that open sea single air dives may be associated with arterial endothelial dysfunction and right heart overload due to increased pulmonary artery pressure. some of these asymptomatic cardiovascular changes may last for several days after the dive and may be attenuated by acute or chronic antioxi- dant administration (vitamin C and E). long-term consequences of acute subclinical cardiovascular alterations induced by diving are presently unknown and will be a goal for future research efforts. Understanding the mechanisms of acute adverse effects of diving on human physiology may also contribute to the therapeutic options aimed at the improvement of cardiovascular health in patients and in the general population. Department of Physiology, University of Split School of Medicine, Split, Croatia. Exercise, Endothelium, and Diving Physiology ?eljko Duji? 72 ? tHE FUtURE oF Diving long-term negative cardiovascular effects of diving on human health is disturbing and must be focused on by the medical and general communities. dETEcTION Of vENOuS gaS bubblES by ulTraSOuNd vgB in the pulmonary artery and the right heart can be monitored with the use of ultrasound, both by Doppler (ni- shi et al., 2003) and/or echocardiographic imaging (Eftedal and Brubakk, 1997). Both methods have been proven valid for in- dicating the risk of DCs after dives (sawatzky, 1991; Eftedal et al., 2007). Ultrasound images are graded according to a pre- viously reported method (Eftedal and Brubakk, 1997) as fol- lows: 0 ? no bubbles; 1 ? occasional bubbles; 2 ? at least one bubble/4th heart cycle; 3 ? at least one bubble/cycle; 4 ? con- tinuous bubbling, at least one bubble/cm2 in all frames; and 5 ? ?white-out,? individual bubbles cannot be seen (this grade had been previously observed only in animals). The grading system is nonlinear when compared to the actual number of bubbles in the pulmonary artery, and bubble grades can be converted into bubbles/cm2 as described by nishi et al. (2003). Bakovi? et al. (2008) were the first to report recently a case of a recreational diver who developed grade 5 (?white-out?) in the right heart with evidence of intrapulmonary (i-P) shunt re- sulting in grade 4 vgB in the left heart during resting condition, but still without any symptoms of DCs (Fig. 1). The diver had no evidence of patent foramen ovale as documented by contrast transesophageal echocardiography (Fig. 2). This case indicates that even extremely high bubble grade may not be related to an increased incidence of DCs. However, as stated above, the sta- tistical relationship between detectable bubbles and the risk of DCs is well documented. Based on the well-known and agreed upon mechanism for decompression problems (gas bubble for- mation), we believe that the absence of vgB is a good indicator of decompression safety and that it is important to work on the procedures that reduce bubble formation after diving. ExErcISE aNd dIvINg The effect of exercise before, during, and after diving on bubble formation is still controversial. Exercise, which is an in- herent part of professional diving, has long been considered to be an additional risk factor for DCs (Cook, 1951). Diving is as- sociated with exercise of variable intensity, depending on ocean currents, temperature, and stress. Therefore, divers are exposed from a mild to moderate level of exercise intensity, with possible short periods of submaximal or even maximal effort. intensive exercise immediately before diving may produce microscopic muscular injuries, which may promote bubble formation (vann et al., 1989). We recently found that a single bout of high-intensity ex- ercise interval training performed 24 hours before a dive signifi- cantly reduced the number of bubbles in the right heart (Duji? et al., 2004) (Fig. 3). Blatteau et al. (2005) reported a similar finding recently for exercise performed 2h before a dive. These results are in concordance with previous studies performed in rats that suggested that there appears to be ?good? and ?bad? exercise (Wisl?ff and Brubakk, 2001). according to this model, the timing of the pre-dive exercise is critical: if it is too close (0.5h, 5h and 10h) or too far (48h) from the dive, the benefi- cial effects will disappear (Wisl?ff et al., 2004). Furthermore, the beneficial effect of properly timed pre-dive exercise partially depends on nitric oxide (no) as shown by studies using some forms of no inhibitors. namely, despite no-blockade, exercise FIgURe 1. echocardiographic apical four-chamber view of the heart representing right ventricle (Rv) (grade 5) and left ventricle (lv) (grade 4) gas bubbles at 40 min after surfacing. Ra right atri- um; la left atrium. FIgURe 2. Transesophageal echocardiographic view of the heart using a contrast agent. Figure reveals intraatrial septum (arrow) without visible defects and presence of contrast agent solely in the right atrium (Ra) during the valsalva maneuver. la left atrium; SvC superior vena cava. DUJi? : ExERCisE, EnDotHEliUm, anD Diving ? 73 pulmonary circulation acts as a filter to prevent spillover of venous bubbles to the arterial side, the next question was: is there a right-to-left shunting of gas bubbles through pulmo- nary circulation during exercise after the dive? We have repeated the study of Eldridge et al. (2004) with the same high-intensity cycle exercise after diving and found no evidence of right-to- left shunting of ?endogenous? bubbles versus contrast bubbles. We speculated that a bolus application of agitated contrast may acutely overwhelm the filtering capacity of pulmonary micro- circulation and open the i-P shunts. Thus, we suggest that the paradoxical results observed in our study are related to lower bubble load. Duji? et al. (2008) recently reviewed more detailed information on the relationship between exercise and diving. in summary, the true cause of the beneficial effect of pre-, during, and post-dive exercise remains elusive and very poorly understood. other effects of exercise must be considered as well in future studies (e.g., endothelial function, no, and heat shock proteins). cardIOvaScular fuNcTION afTEr dIvINg This overview of the current knowledge of the effects of diving on the cardiac function and endothelium of the conduit arteries focuses on the findings from recent studies performed in the field and simulated (hyperbaric chamber) diving conditions. We believe that diving research should be performed, whenever possible, under open water conditions in order to include the additional stressors such as immersion, exercise, cold, weight of the breathing apparatus, hemoconcentration, and psycho- logical factors, which are usually present during routine diving activities. The data on cardiovascular effects of simulated versus open water diving in humans is controversial. Diesel et al. (2002) showed a lack of increase in systolic pulmonary artery pressure (sPaP) after simulated altitude decompressions induced high- grade venous gas bubbles. valic et al. (2005) have confirmed this finding after a simulated chamber dive. The results of hu- man studies seem to contradict the animal data that reported increased PaP (vik et al., 1993). in addition, we have recently found that a single open water air dive is associated with acute, post-dive depressions in lung respiratory function and cardiac output (Duji? et al., 2005a). marabotti et al. (1999) have also found that a recreational scuba dive impairs the cardiac per- formance within two hours after surfacing, suggesting a right ventricular overload and an impairment of both right and left ventricular diastolic function. in contrast to our findings ob- tained during a dry dive, we observed an increase in the PaP and pulmonary vascular resistance (PvR) after a single open water air dive to 30 m (Duji? et al., 2006) (Fig. 4). The increased PaP and PvR were not expected to cause any clinical symptoms, bearing in mind the large functional reserve in the pulmonary circulation. The exact mechanism(s) of increased PvR after diving is still unknown. vgB lodged still provided some protection, although to a reduced degree, indicating no-dependent and no-independent components of that mechanism. Furthermore, both acute and chronic no- donors protected against bubble formation and reduced the death rate in rats (Wisloff et al., 2003; Wisloff et al., 2004), without performing exercise. This was subsequently confirmed in man (Duji? et al., 2005b). The uptake and elimination of inert gas depends on local blood flow. Exercise performed during the bottom phase of a dive will increase blood flow and consequently nitrogen uptake because of active hyperemia in skeletal muscles (Flook, 1997). With more nitrogen dissolved in the tissues, the time for its elimination should be extended. Dick et al. (1984) have shown that more nitrogen is indeed eliminated after a dive with exer- cise during the bottom phase. However, when mild exercise was performed during a decompression stop, the number of vgB detected after diving was reduced, likely due to increased gas elimination via augmented alveolar ventilation caused by the physical activity (Jankowski et al., 1997; Duji? et al., 2005b). The positive effects of a rather short, 3-minute exercise period during a decompression stop suggests that the nitrogen pool for bubble formation is coming from the ?faster? tissues. Furthermore, in a subsequent study, we have shown that post-dive, high-intensity exercise did not increase the risk for DCs but rather caused an 8-fold reduction in bubble forma- tion. This was somewhat surprising since we hypothesized that it would promote venous bubble formation, based on the pre- vious reports of increased nitrogen elimination with post-dive exercise after simulated dives to 20 msw (muth et al., 1994). These inconsistencies may potentially be explained with the dif- ferences in type of diving (open water versus simulated) and the exercise intensity (high versus moderate). The most likely cause of reduced bubble formation observed during post-dive exercise is the increased blood flow induced by exercise that may cause a depletion of bubble nuclei at the surface of endothelial cells, without which bubbles cannot grow. Eldridge et al. (2004) have shown that exercise-induced in- trapulmonary a-v shunting occurs in healthy humans. Because FIgURe 3. The figure depicts change in venous gas bubbles within the right heart and pulmonary artery following a dive without (a) and with (B) a previous bout of strenuous exercise in one diver. In (a), there are numerous, clearly visible venous gas bubbles. after performing the exercise, bubbles are completely absent (B). 74 ? tHE FUtURE oF Diving in the pulmonary circulation may cause mechanical, humoral, and biochemical effects. Humoral factors, like tromboxane a 2 , histamine, endothelin-1, and serotonin have been implicated in pulmonary microembolization by bubbles (malik, 1983). nos- sum et al. (2002) have shown that venous bubbles may damage the pulmonary endothelium, which may also result in pulmo- nary vasoconstriction. nossum et al. (1999) also reported that vascular bubbles led to a reduction in endothelial function in pig pulmonary artery and that even a low bubble load was associated with a delayed endothelial dysfunction in rabbit 1-6 hours after expo- sure (nossum et al., 2002). since venous bubbles are trapped in the pulmonary circulation, they were assumed to have no further effects on the arterial side. it was also assumed that a right-to-left shunt (R-to-l), which occurs due to a persistent foramen ovale (PFo), was necessary for the bubbles to affect the systemic arteries. in that regard, we found a post-dive (both dry and field) increase in the brachial artery diameter and a reduction in the arterial endothelial function (assessed by flow- mediated dilation). although no large PFo was detected, the possibility of a significant number of gas bubbles in the arterial circulation was not very likely (Brubakk et al., 2005; obad et al., 2007) (Fig. 5). one explanation for this effect is a diving-induced activa- tion of the endothelium which may, as shown by other inves- tigators, lead to a production of endothelial microparticles and further endothelial damage and dysfunction, even at remote sites (Brodsky et al., 2004). Endothelial microparticles have a size of a few microns and could possibly pass the lung filter and enter the arterial system. Thus, it is possible that changes in the endothelial function can occur without the arteries being in direct contact with the gas bubbles. next, we analyzed the effects of pre-dive administration of antioxidant agents since it was shown that hyperoxia per se (without diving) caused FmD reduction (obad et al., 2007). Pre-dive ingestion of high doses of vitamins C and E (2 grams and 400 iU, respectively) in a randomized placebo-controlled study design partially reversed the endothelial dysfunction after a dive (Fig. 6), while changes in pulmonary artery and heart function were unaffected. it was also found that FmD, previously shown to be reduced up to 48 hours after a single open water air dive, was better preserved in this case and recovered faster. Summary aNd IdEaS fOr fuTurE STudIES The ideas presented above suggest the following: 1. another means for reducing the number of venous gas bubbles after air diving may include exercising before, dur- ing decompression and after diving, and administering acute no donors before the dive. additional mechanistic studies on large numbers of divers should be performed in the future in order to complement current findings; 2. asymptomatic cardiovascular and arterial endothelial FIgURe 4. Changes in systolic pulmonary arterial pressure (Spap). (a) acct/RveT ratio; (B) pulmonary vascular resistance (pvR); and, (C) at 20 and 40 min after dive. values are mean ? SD; *, p < 0.05 compared to baseline. FIgURe 5. (a) Difference in brachial artery diameter and flow- mediated dilatation (FmD) before and after the dive. (B) Differ- ence in brachial artery diameter and FmD before and after breath- ing 60% O2. The data are the mean and 95% confidence intervals. DUJi? : ExERCisE, EnDotHEliUm, anD Diving ? 75 dysfunction detected after diving lasts for several days after a single open water dive. long-term effects are presently unknown; and 3. acute and long-term use of antioxidants may contribute to prevention of some of these effects (endothelial function). Based on the abovementioned findings, it seems that high- intensity, pre-dive exercise and moderate exercise performed during decompression stops would appear to be a wise pre- scription for reducing the number of venous gas bubbles after air dives. However, before these procedures can be widely ad- opted as a predictable safeguard against DCs, we need further standardization related to the exercise?s duration and intensity. Pharmacological intervention with an acute no donor such as nitroglycerine seems to complement this procedure. Despite a wealth of information about the physiological consequences of scuba diving gathered during the last several years, many more questions need to be answered in the future such as: 1. is pre-dive exercise beneficial under field conditions for endothelial function, cardiovascular function, or preven- tion of bubble formation? What mechanisms are involved in this phenomenon (no, heat shock proteins, vasoactive substances such as endothelin-1, angiotensin ii, etc.)? 2. What are the cumulative effects of multiday air or trimix diving? 3. What are the long-term effects of diving on the cardiovas- cular system? 4. Does age and gender (especially female) make any difference? 5. What about genetics of DCs susceptibility and resistance (significance of genetic studies and analyses, detection of candidate genes? polymorphism - like enos)? Understanding the involved mechanisms is a next critical step that may ultimately allow us to prevent DCs by various pharmacological means. last, but not least, after completion of these planned studies we have to ask ourselves ?what do the ideas and data presented above fail to explain?? Acknowledgments The author?s work on this topic was supported by numerous grants awarded by the Croatian ministry of science, Education and sports (currently with 216-2169133-0130)and the Unity through knowledge Fund project 33/08. The author would like to thank many collaborators over the years, and especially Dr. alf o. Brubakk, who in 2001 visited split and first suggested to test the physiological effects of pre-dive, high-intensity exercise on humans after diving. lITEraTurE cITEd Bakovi?, D., D. glavas, i. Palada, t. Breskovi?, D. Fabijanic, a. obad, Z. valic, a. o. Brubakk, and ?. Duji?. 2008. High-grade bubbles in left and right heart in an asymptomatic diver at rest after surfacing. Aviation, Space, and Environmental Medicine 79:626?628. Behnke, a.R. 1951. Decompression sickness following exposure to high pres- sures. in Fulton, J.F. (ed.). Decompression Sickness. saunders: Philadelphia. pp. 53?89. Blatteau, J. E., E. gempp, F. m. galland, J. m. Pontier, J. m. sainty, and C. Robinet. 2005. aerobic exercise 2 hours before a dive to 30 msw decreases bubble formation after decompression. Aviation, Space, and Environmental Medicine 76:666?669. Brodsky, s. v., F. Zhang, a. nasjletti, m. s. goligorsky. 2004. Endothelium-de- rived microparticles impair endothelial function in vitro, American Journal of Physiology: Heart and Circulatory Physiology 286:H1910?H1915. Brubakk, a. o., D. Duplan?i?, Z. valic, i. Palada, a. obad, D. Bakovi?, U. Wisl?ff, and ?. Duji?. 2005. a single air dive reduces arterial endothelial function in man. Journal of Physiology 566:901?906. Cook, s. F. 1951. Role of exercise, temperature, drugs and water balance in de- compression sickness. in Fulton, J.F. (ed.). Decompression Sickness. saun- ders: Philadelphia. pp. 223?241. Dick, a. P., R. D. vann, g. Y. mebane, and m. D. Feezor. 1984. Decompression induced nitrogen elimination. Undersea Biomedical Research 11:369?380. Diesel, D. a., m. t. Ryles, a. a. Pilmanis, and U. i. Balldin. 2002. non-inva- sive measurement of pulmonary artery pressure in humans with simulated altitude-induced venous gas emboli. Aviation, Space, and Environmental Medicine 73:128?133. Duji?, ?., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n.m. Petri, U. Wisl?ff, and a.o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555(3):637?642. FIgURe 6. Flow-mediated dilatation (FmD) at pre-dive and post- dive (30 min, 24 h, 48 h and 72 h) periods for control dive (a) and dive after application of antioxidants (B) in protocol 2 values are represented as individual responses (thin lines) and mean response (thick line) for both dives; *, significant difference (p < 0.05) be- tween pre-dive and post-dive values; ?, significant difference (p < 0.05) between dives with and without antioxidant treatment. 76 ? tHE FUtURE oF Diving Duji?, ?., D. Bakovic, i. marinovic-terzic, and D. Eterovic. 2005a. acute effects of a single open sea air dive and post-dive posture on cardiac output and pulmonary gas exchange in recreational divers. British Journal of Sports Medicine 39:e24. Duji?, ?., i. Palada, a. obad, D. Duplancic, D. Bakovic, and Z. valic. 2005b. Exercise during a 3-min decompression stop reduces postdive venous gas bubbles. Medicine and Science in Sports and Exercise 37:1319?1323. Duji?, ?., a. obad, i. Palada, Z. valic, and a. o. Brubakk. 2006. a single open sea air dive increases pulmonary artery pressure and reduces right ventricu- lar function in professional divers. European Journal of Applied Physiology 97(4):478?485. Duji?, ?., Z. valic, and a. o. Brubakk. 2008. Beneficial role of exercise on scuba diving. Exercise and Sport Sciences Reviews 36:38?42. Eftedal, o. s. and a. o. Brubakk. 1997. agreement between trained and un- trained observers in grading intravascular bubble signals in ultrasonic im- ages, Journal of Undersea and Hyperbaric Medicine 24:293?299. Eftedal, o. s., s. lydersen, and a. o. Brubakk. 2007. The relationship between venous gas bubbles and adverse effects of decompression after air dives. Journal of Undersea and Hyperbaric Medicine 34:99?105. Eldridge, m. W., J. a. Dempsey, H. C. Haverkamp, a. t. lovering, and J. s. Hokanson. 2004. Exercise-induced intrapulmonary arteriovenous shunt- ing in healthy humans. Journal of Applied Physiology 97:797?805. Flook, v. 1997. The effect of exercise on decompression bubbles. in mekjavic, i.B., and m.J. tipton, eds. Proceedings of the XXIII Annual Scientific Meet- ing of the European Underwater and Baromedical Society, Bled, slovenia. Pp. 55?61. Jankowski, l. W., R. Y. nishi, D. J. Eaton, and a. P. griffin.1997. Exercise dur- ing decompression reduces the amount of venous gas emboli. Journal of Undersea and Hyperbaric Medicine 24:59?65. malik, a. B. 1983. Pulmonary microembolism. Physiological Reviews 63:1114?1207. marabotti, C., F. Chiesa, a. scalzini, F. antonelli, R. lari, C. Franchini, and P.g. Data. 1999. Cardiac and humoral changes induced by recreational scuba diving. Journal of Undersea and Hyperbaric Medicine 26:151?158. muth, C. m., C. m. staschen, v. Warninghoff, U. van laak, and P. Rader- macher. 1994. Exercise effects on central venous nitrogen tensions after simulated non-decompression dives. Journal of Undersea and Hyperbaric Medicine 21:297?303. nishi, R.Y., a. o. Brubakk, and o. s. Eftedal. 2003. Bubble detection. in Ben- nett and Elliott?s The Physiology and Medicine of Diving, Brubakk, a. o. and t. s. neuman, eds. saunders: london. Pp. 501?529. nossum, v., s. koteng, and a. o. Brubakk. 1999. Endothelial damage by bub- bles in the pulmonary artery of the pig. Journal of Undersea and Hyperbaric Medicine 26:1?8. nossum, v., a. Hjelde, and a. o. Brubakk. 2002. small amounts of venous gas embolism cause delayed impairment of endothelial function and increase polymorphonuclear neutrophil infiltration. European Journal of Applied Physiology 86:209?214. obad, a., i. Palada, Z. valic, v. ivan?ev, D. Bakovi?, U. Wisl?ff, a. o. Brubakk, and ?. Duji?. 2007. The effects of acute oral antioxidants on diving-in- duced alterations in human cardiovascular function. Journal of Physiology 578:859?870. sawatzky, k. D. 1991. The relationship between intravascular Doppler-detected gas bubbles and decompression sickness after bounce diving in humans. toronto, ontario: York University. valic, Z., D. Duplan?ic, D. Bakovic, v. ivan?ev, D. Eterovi?, U. Wisl?ff, a. o. Brubakk, and ?. Duji?. 2005. Diving-induced venous gas emboli do not increase pulmonary artery pressure. International Journal of Sports Medi- cine 26:626?631. vann, R. D., W. a. gerth, and n. E. leatherman. 1989. Exercise and decom- pression sickness. in vann, R. D., ed. The Physiological Basis of Decompres- sion. Proceedings of the 38th UHms Workshop. Bethesda. Pp. 119?145. vik, a., B. m. Jenssen, o. s. Eftedal, and a. o. Brubakk. 1993. Relationship between venous bubbles and hemodynamic responses after decompression in pigs. Journal of Undersea and Hyperbaric Medicine 20:233?248. Wisl?ff, U., and a. o. Brubakk. 2001. aerobic endurance training reduces bub- ble formation and increases survival in rats exposed to hyperbaric pressure. Journal of Physiology 537:607?611. Wisl?ff, U., R. s. Richardson, and a. o. Brubakk. 2003. nos inhibition in- creases bubble formation and reduces survival in sedentary but not exer- cised rats. Journal of Physiology 546:577?582. ???. 2004. Exercise and nitric oxide prevent bubble formation: a novel ap- proach to the prevention of decompression sickness? Journal of Physiology 555:825?829. INTrOducTION it is common knowledge that animal experimentation played an essential part in laying the foundations of our understanding of decompression physiology. in 1670 Robert Boyle made the initial observation of a bubble in the eye of a viper exposed to low ambient pressure (Boyle, 1670) and he put forward the hypothesis that decom- pression would cause bubbles to form in all parts of the body of the animal, thus causing the severe symptoms he described. two hundred years later Paul Bert (Bert, 1878) in a series of experiments on several species of animals was able to link the cause of de- compression sickness to the formation of nitrogen bubbles. Thirty years later, Boycott, Damant and Haldane (Boycott et al., 1908) used goats in their seminal experiments working out the method of staged decompression to prevent compressed-air illness, or as we now call it, decompression sickness. During WW ii newton Harvey and his group (Harvey, 1945) used physical, cellular, and animal models to investigate the con- ditions necessary for bubble formation in the body. nevertheless, it is hard to fault alf Brubakk (1996) when he writes that ?animal studies have had relatively little impact on both the development of [diving] procedures and methods for treatment of accidents.? Clearly, animal experimentation has played a much smaller role than human experimentation in the development of new diving techniques during the latter half of the 20th century. one of the reasons for this may be that the main drivers of the development of modern diving have been the military and commercial diving companies and the academic community has only played a second- ary part. The need for better tables or less narcotic or dense breathing gases has often oc- curred as a result of technological improvements, allowing the divers to dive deeper and/or stay at depth for longer periods. Problems have usually occurred while trying to extrapolate existing decompression tables or methods. By changing the methods incre- mentally it has been possible to use human divers to test new techniques because the complications have been considered to be of limited severity and any resulting decom- pression sickness treatable. aBSTRaCT. although animal experiments helped to lay the foundations of our understanding of decompression physiology and to develop the first methods to avoid decompression sickness, new diving methods have relied only sparingly on animal experiments. a short review is provided of the development of the modern submarine escape method: a rescue method from sunken sub- marines. The development of this method has depended to a very large extent on the use of animal experiments. Department of Defence Physiology, Swedish Defence Research Agency, SE-164 90 Stockholm, Sweden. animal Experiments for Evaluating Decompression: Development of the modern submarine Escape method Mikael Gennser 78 ? tHE FUtURE oF Diving But in certain cases, where the danger posed to the divers either was much greater or was unknown, animal experiments have been extremely useful. Ralph Brauer when reviewing the work leading to the deep hydrox and hydroheliox dives felt that it was ?difficult to escape the conclusion that here [was] an in- stance where animal experimentation furnished the basic data upon which others found it possible to initiate work directly aimed at testing the feasibility of exposing humans to this gas? (Brauer, 1987). The reason animal experiments were used ini- tially to test hydrogen as a diving gas was, however, only partly due to the unknown decompression properties of the gas. The main reason the first exposures were made with animals was to make sure the gas was inert and excluded any risks of toxicity (?rnhagen, 1987). ThE NEEd fOr ThE SubmarINE EScapE mEThOd There is another diving method that is less well known that has relied almost exclusively on animal experiments for its development, and where the main risk apart from barotrauma is severe central nervous system decompression sickness (Cns DCs), i.e., the modern submarine escape method. after WW ii, as a response to the grievous losses experi- enced by the submarine fleet, the British admiralty instigated research and development to improve the possibilities for sub- mariners to escape from sunken submarines. There had been several successful escapes from submarines reported during the war. However, to be able to escape the pressure inside the sub- marine had to be equalized with the external pressure, and in single compartment submarines this took a long time. in the mean time, the survivors would be exposed to the increased pressure, and to increased partial pressures of nitrogen, carbon dioxide, and oxygen. it was considered that the chances of a successful escape from depths deeper than 100 ft (30 msw) were dismal. to improve the chances of a good outcome and increase the safe escape depth, special escape chambers were devised that would equalize much quicker to the surrounding pressure. The submariner would be breathing clean air via a dedicated air sup- ply that inflated the buoyancy stole and the hood that he wore. since the stole was connected to the hood, once the escape hatch had opened and the submariner would start to ascend through the water column, the expanding air in the stole would fill the hood and thus maintain a cushion of air in front of his face during the whole ascent. hISTOry Of TESTINg prOcEdurES very early on it was decided that air would be used in the gas supply to reduce possible complications with oxygen toxic- ity and to also have the possibility of using other sources of compressed air in case the dedicated air supply was unavailable. The questions that now had to be answered were how fast a compression one could withstand and how fast and from what depth could the ascent be without causing decompression illness. Before any wet dives were tested experiments were carried out in dry pressure chambers. it was obvious that animals would have to be used in all but the most conservative exposures until a safe method had been devised. The animal that was chosen was the goat. Boycott et al. (1908) had chosen the goat mainly for its ease of handling. However, there were a number of other factors that had to be considered in the submarine escape trials. Preferably the animals should be awake and non-sedated during the exposures so that even mild symptoms could be noted: the animals should be able to equilibrate pressure during the rapid compressions and, most importantly, the animals should not tend to hold their breath during the ascent, but exhale as the pulmonary air expanded. it turned out that the goats could be trained to accept both the high noise volumes during compres- sion and decompression and because of their strong Hering- Breuer reflex the animals would exhale in most cases during the ascents. During a 25-year period between 1945 and 1970 the Royal navy carried out animal experiments (and some human expo- sures) to develop a method that would allow submariners to es- cape from a disabled submarine from depths close to 200 msw. This work has been reviewed in two articles by one of the main investigators (Donald, 1979; 1991). The final escape ?dive pro- file? turned out to be a pseudo-exponential compression rate with a doubling of pressure every 4th to 5th second to 180 msw depth (19 atm), a four second pause at depth, and then a linear ascent with 2.75 ? 3 msw/s (Fig. 1). Thus, a ?dive? to 180 msw and back to the surface would be completed in less than 90 seconds. in 1970, the method was tested successfully by subma- riners escaping from 180 msw depth off malta (Upshot v). There followed a pause in the research and development of submarine escapes until the mid-1980s. By that time one had come to the conclusion that in almost all conceivable submarine accidents there would be a pressure increase inside the subma- rine and it was therefore necessary to find out from what depth one could ascend safely given the internal pressure inside the submarine and the time spent at that increased pressure. The FIgURe 1. Redrawn time-depth profile from submarine escape procedure carried out from submarine from a depth of 150 msw. (Copy of original data given to the author by Cdr. Bo persson, Royal Swedish Navy.) gEnnsER: animal ExPERimEnts FoR DEComPREssion ? 79 Blogg, 2008). However, the use of oxygen-enriched gas instead of air during the ascent did not reduce the amount of initial bubbling (Blogg et al., 2003). in all likelihood the very high partial pressures of oxygen made oxygen behave as an inert gas, even in a tissue with high metabolic rate like the central nervous system, and nitrogen/oxygen bubbles were formed. This had in fact been predicted by earlier investigators (Donald, 1991). apart from the first paper on ?oxygen bends? (Donald, 1955), these experiments are probably the first instances with clear evi- dence of oxygen bubbles. aNImal mOdEl advaNTagES The advantages of using an animal model include con- trolled conditions, avoiding the exposure of humans to un- necessary risk, achieving objective end points, and allowing for more extensive measurements. another advantage of using animals in this particular setting, where the end-points were relatively severe decompression sickness, can be seen in Figure 3. The figure shows the timing of first DCs symptoms in goats after submarine escape (with or without preceding saturation at increased air pressure). also shown in the graph is a line in- dicating the time when symptoms (Cns symptoms and joint pains) started to appear in escapers from the bottomed Peruvian submarine Pacocha. The internal pressure in the submarine had increased to 2.7 atm and the crew had waited far too long be- fore deciding to escape towards the surface. The timing of the symptoms in the submariners fits rather well with the goat data. ?safe-to-escape? curve had to be determined. also, for this work goats were used, although one series of man trials eventually was carried out. The safe-to-escape curve was in the end calculated from data from a total of 568 goat dives that resulted in 45 cases of DCs and 72 man dives with 2 cases of DCs (White et al., 2004). The resulting safe-to-escape curve is shown in Figure 2 with human and goat data marked respectively. The remaining data has been extrapolated using an exponential/linear decompres- sion model with Hill?s risk model added for saturation dives (loveman, 2004). to sum up, goats were used to develop a rela- tively safe escape method prior to human testing. phySIOlOgIcal rISkS The reason these profiles are dangerous is the ever present risk of pulmonary barotrauma with arterial gas embolism and the fact that these profiles will preferentially load the fast tissues (e.g., the central nervous system) with inert gas. Donald used classical Haldanean theories to calculate that after an escape from 300 msw a very fast tissue (half-time of 1.3 min) would yield a Haldane ratio of more than 11 upon surfacing and a fast tissue (half-time of 5 min) would yield a Haldane ratio of almost 6 (Donald, 1991). Consideration of results of experi- ments with broken ascent profiles (i.e., stops at shallow depths) indicated that the critical tissues were in the ?5-minute? range (Donald, 1991). Flook (1997), in a theoretical consideration of the free ascent profile, showed that fast tissues such as the cen- tral nervous system, would be the preferential target for bubble formation. later research used this information showing that even short periods of oxygen breathing prior to deep escapes reduced the amount of circulating bubbles post ascent (gennser and FIgURe 2. UK safe-to-escape curve. Redrawn from loveman et al., 2004. Solid dots show limiting manned exposures; squares show the limits of animal exposures. FIgURe 3. Time of occurrence of DCS symptoms in goats after surfacing from submarine escape profiles. Triangles: limb bends; circles: CNS symptoms; + dead on arrival at surface (probably ar- terial gas embolism), squares: CNS symptoms. Solid line indicates timing of DCS symptoms in survivors from the peruvian subma- rine pacocha. Stars show time of symptoms in two divers taking part in submarine escape experiments (CNS symptoms and mar- bling). Data from Blogg et al. 2003; gennser and Blogg, 2008; Seddon, 1997; Benton, 2002; and harvey and Carson, 1989. 80 ? tHE FUtURE oF Diving Conversely, the symptoms that appeared in two subjects taking part in a trial with deep submarine escapes after shallow air sat- uration occurred much later. it is hard to avoid the suggestion that the DCs found after the experimental trial is of a less severe nature and the symptoms may well have subsided without any treatment. it must of course be noted that all symptoms occur- ring in a volunteer in an experimental study must be taken very seriously. However, in a mass casualty situation it is more than likely that these divers would not have merited any treatment in the first instance. a definite advantage with the animal model used in these trials, was that the animals were awake and not exposed to anes- thesia. in one trial where goats were lightly anesthetised to allow drawing arterial blood from carotid loops, rapid compressions to 50 msw caused the animals to become apneic. interestingly, the resulting lung squeeze delayed the uptake of nitrogen in the arterial circulation by 30 secs to a minute, something that would have markedly reduced the risk of decompression sick- ness during an escape (Flook et al., 1998). lONg-TErm cNS EffEcTS? The fact that goats were used in repeated dives, but other- wise kept in a stable, controlled environment for their whole life, made it of interest to investigate whether the diving activity appeared to have had any long-term effect on their central ner- vous system. Blogg et al. (2004) conducted experiments with 30 goats (13 of which had never shown symptoms of DCs and 17 had been treated successfully for DCs). magnetic resonance imaging (mRi) of the brains showed ?white spots? in 3 animals. There was no correlation of ?white-spots? with either diving ac- tivity or DCs. Following up with a histopathological analysis of the brains, the ?spots? were determined to be functional, not anatomical, lesions and there were no histopathological lesions in any of the brains, indicating no long-term effect of the diving activity in these animals. However, goats (like all Artiodactyla) have a carotid rete, which might provide protection against ce- rebral embolization. no correlation was found between spinal degeneration and DCs or diving, but spinal scarring in two ani- mals treated for DCs was discovered. although this study was too small to yield any clear-cut evidence one way or another, clearly it did show the possibility of using animal experimenta- tion to attack one of the more important health questions con- cerning diving: whether diving as such gives rise to long-term effects on the central nervous system. cONcluSION This short review of the development of the submarine es- cape method has tried to show the importance of animal experi- ments in a situation where it would clearly have been unethical to use human subjects. in the future there may be other such instances where animal experiments will have to precede human trials in order to not expose human volunteers to unnecessary risks. However, the most likely future use of animal experi- ments, and where animals have previously been used most fre- quently, is in the investigation of underlying pathophysiological mechanisms. modern molecular biological techniques, such as knock- out mice, will yield many new tools and avenues for investi- gating the interaction between bubbles and tissues. However, one should not forget the experiments that nature itself has provided for us with natural divers such as diving mammals. although both animal rights issues and the size of the animals make it almost impossible to carry out laboratory research on all but the smallest species, some remarkable field studies have been performed (kooyman, 1981). With the use of microcom- puters it has been possible to follow the natural diving behavior of these animals in the wild, and to even carry out invasive mea- surements such as mapping blood gases during actual deep dives (Zapol et al., 1989). in the future, improved miniaturization of ultrasound imaging devices may yield very interesting results, both in regard to blood-flow studies and possible post-dive bubble generation. Diving mammals may also yield biochemi- cal clues on the pathophysiology of decompression sickness. it has for instance been shown that many diving mammals appear to lack certain factors in the coagulation cascade (Robinson et al., 1969). Experiments with knock-out mice or inbred mice that are poor in the same factors may yield information on the importance of an intact coagulation system in the pathological sequence of decompression sickness. in short, although animal experiments will probably only be used sparingly in testing and developing actual diving meth- ods, as an instrument to dissect and clear up the mechanisms behind decompression sickness animal experiments will in all likelihood still have a big part to play. lITEraTurE cITEd Benton, P. 2002. submarine escape trials 1999?2001?provisions of medical support. Journal of Royal Navy Medical Services 88(3):108?115. Bert, P. 1878. La Presion Barometrique. translated by Hitchcock, m. a. and F. a. Hitchcock (1943) Columbus: College Book Company. Blogg, s. l., m. gennser, g. a. m. loveman, F. m. seddon, J. C. Thacker, and m. g. White. 2003. The effect of breathing hyperoxic gas during simu- lated submarine escape on venous gas emboli and decompression sickness. Journal of Undersea and Hyperbaric Medicine 3(3):163?174. Blogg, s. l., g. a. m. loveman, F. m. seddon, n. Woodger, a. koch, m. Re- uter, m. gennser, and m. g. White. 2004. magnetic resonance imaging and neuropathology findings in the goat nervous system following hyper- baric exposures. Journal of European Neurology 52:18?28. Boycott, a. E., g. C. C. Damant, and J. s. Haldane. 1908. Prevention of com- pressed-air illness. Journal of Hygiene, Cambridge 8:342?443. Boyle, R. 1670. new pneumatical observations about respiration. Philosophical Transactions of the Royal Society 5:2011?2031. Brauer, R. W., and R. naquet. 1987. Contributions of animal experimentation to the establishment of appropriate conditions for ethical use of hydrogen as a component of deep diving atmospheres. in Brauer, R.W. (ed.) Hydro- gen as a diving gas. Thirty-third Undersea and Hyperbaric medical society Workshop, UHms publication no. 69. Pp 3?12. gEnnsER: animal ExPERimEnts FoR DEComPREssion ? 81 Brubakk, a. o. 1996. Which animal models are to be used to study decompres- sion accidents? in Proceedings of the Second European Consensus Conference on Treatment of Decompression Accidents in Recreational Diving. Wattel, F., and D. matthieu (eds.) marseille, France. pp 85?101. Donald, k. W. 1955. oxygen bends. Journal of Applied Physiology 7(6):639?644. ???. 1979. submarine escape breathing air. a review and analysis of animal and human experiments by the Royal navy. Bulletin of European Physiopa- thology and Respiration 15(5):739?745. ???. 1991. a review of submarine escape trials from 1945 to 1970 with a par- ticular emphasis on decompression sickness. Journal of Royal Navy Medical Services 77:171?200. Flook, v. 1997. a theoretical study of the extent and duration of decompression bubbles following submarine escape procedures. in mekjavic, i.B., and o. Eiken, eds. Diving and Hyperbaric Medicine, Proceedings of the 23rd European Underwater Baromedical society Congress, Bled, slovenia. Pp 62?67. Flook, v., and a. l. Ustad. 1998. Blood nitrogen concentrations in goats fol- lowing rapid compression to 6 bar. sintEF report no. stF78, F98112, sintEF Unimed, Uk. gennser, m., and s. l. Blogg. 2008. oxygen and carbogen breathing before simulated submarine escape. Journal of Applied Physiology 104:50?56. Harvey, C., and J. Carson. 1989. The B.a.P. Pacocha (ss-48) collision: The es- cape and medical recompression treatment of survivors. naval submarine medical Research laboratory, nsmRl special Report sP89-1, pp. 50. Harvey, n. E. 1945. Decompression sickness and bubble formation in blood and tissues. Bulletin of the New York Academy of Medicine 21(10):505?536. kooyman, g. l. 1981. Weddell Seal: Consummate Diver. Cambridge University Press. Pp. 176. loveman, g. a. m. 2004. iso-risk curves for escape from saturation in a dis- tressed submarine. in ?rnhagen, H. ed. Humans in Submarines. Royal swedish navy, Pp. 181?185. ?rnhagen, H. 1987. swedish animal experiments using hydrogen-oxygen mix- tures at pressures up to 6.0 mPa. in Brauer, R.W., ed. Hydrogen as a diving gas. Thirty-third Undersea and Hyperbaric medical society Workshop, UHms publication no. 69. Pp 25?30. Robinson, a. J., m. kropatkin, and P. m. aggeler. 1969. Hageman factor (factor xii) deficiency in marine mammals. Science 166:1420?1422. seddon, F. m. 1997. safe to escape curve. animal studies: august 1993 to Febru- ary 1997. Defence Research agency Report DERa/ssEs/CR971023/1.0, london, Uk. White, m. g., F. m. seddon, k. m. Jurd, J. C. Thacker, m. stansfield, and g. a. m. loveman. 2004. Uk safe-to-escape curve research. in Ed: ?rnhagen, H., ed. Humans in Submarines. Royal swedish navy, Pp. 175?180. Zapol, W. m., R. D. Hill, J. Qvist, k. Falke, R. C. schneider, g. C. liggins, and P. W. Hochachka. 1989. arterial gas tensions and hemoglobin con- centration of the freely diving Weddell seal. Undersea Biomedical Research 16:363?373. INTrOducTION For 350 years we have known that decompression, the process of coming up from a dive, can induce formation of gas bubbles in the body (Boyle, 1670). We also have considerable empirical evidence that these gas bubbles can cause decompression sickness. still, our knowledge of how bubbles are formed and how they cause decompression sickness is very limited. to study these processes, we need some means of detecting the gas bubbles. gas bubbles in liquid are strong reflectors of sound, and ultrasound techniques are well suited for detection of gas bubbles in the vascular system. The two ultrasound techniques most widely used for bubble detec- tion are Doppler and imaging. The Doppler technique gives audible sound signals (echoes) from the blood flow. any bubbles present in the flowing blood will give sharp clicking sounds that can be discriminated from the smooth flow signal (Eftedal and Brubakk, 1997). Rather than counting single bubbles, the amount of bubbles is usually estimated according to some classification scheme. The most widely used grading schemes are the spencer code (spencer and Johansen, 1974) and the kisman-masurel (k-m) code (kisman and ma- surel, 1983). Both schemes go from grade 0 (no bubbles) to grade 4 (many bubbles), and the difference between the schemes lies in how the measurements are performed rather than the number of bubbles assigned to each grade. The k-m code includes mea- surements at rest and after movement, both in central and peripheral vessels, whereas the spencer code is designed for measurements in the pulmonary artery at rest. Doppler systems are compact, inexpensive, and well suited for field studies. Ultrasound imaging can also be used to detect bubbles in the body, and the heart is the preferred location for detection. Bubbles can be seen as bright spots in the image, aBSTRaCT. studies show that the occurrence of venous gas bubbles is a highly sensitive, although not specific, predictor of decompression stress, at least for sub-saturation dives. Consequently, the absence of detectable bubbles is a strong indicator of decompression safety. traditionally, new de- compression procedures have been evaluated by recording the incidence of decompression-related symptoms. The number of test dives required for a reliable assessment of a procedure based on this traditional evaluation is high, making validation of decompression procedures extremely resource demanding. We propose an alternative method for validation of decompression procedures based on detection of vascular gas bubbles. The method makes use of Bayesian statistics to estimate the risk of decompression sickness for the tested procedure based on previous knowledge about the corre- spondence between vascular gas bubbles and decompression sickness. The new method will greatly reduce the number of dives required to approve or reject a procedure and can facilitate introduction of new and safer decompression tables. StatoilHydro ASA, BE-SA D1 31, Box 7200, 5020, Bergen, Norway. Ultrasound for Evaluating Decompression Olav S. Eftedal 84 ? tHE FUtURE oF Diving and the number of bubbles is again usually estimated by scor- ing. We have previously suggested a grading scheme for ultra- sound imaging with 6 levels from 0 to 5 (Eftedal and Brubakk, 1997). This scheme is in most cases directly comparable to the Doppler schemes (Brubakk and Eftedal, 2001). imaging sys- tems are generally more complex and expensive than Doppler systems, but portable instruments suitable for field studies are now available. cOrrESpONdENcE bETWEEN vaScular gaS bubblES aNd dcS in the literature, there are several published studies that address the issue of correspondence between detectable venous gas bubbles and the occurrence of decompression sickness, most of them on Doppler-detected bubbles. The studies use different procedures for bubble detection and the output parameters vary. some use maximum observed bubble grade, some use time inte- grals. some studies also combine data from various types of ex- posures. The different studies are therefore not easy to compare. The most extensive study on bubble/DCs correspondence in sub-saturation diving was presented by sawatzky (1991) in his master?s thesis where he systematically analyzed maximum Doppler scores and DCs incidence in 1,726 nitrox (nitrogen and oxygen) dives and 1,508 heliox (helium and oxygen) dives. The data were obtained from chamber decompression trials at the Defence and Civil institute of Environmental medicine (DCiEm) in toronto, Canada. The uniqueness of these data lie both in their quality and quantity. it is the most compre- hensive collection of data published, collected according to the recommended standards for Doppler monitoring (Eatock and nishi, 1986), and Doppler grading is performed and verified by competent personnel. DCs diagnoses were made according to specific criteria (temple et al., 1999). The most important of sawatzky?s data are reproduced in table 1. sawatzky (1991) concluded that the risk of developing DCs increases with increasing Doppler grades and the best in- dicator of that risk is the single highest Doppler grade regard- less of location. He explained this with the greater ease with which bubbles can be detected in the subclavian veins where the signal to noise ratio is higher compared to precordial mea- surements. This is supported by the findings of Eckenhoff et al. (1990) who found an increased sensitivity with subclavian as compared to precordial monitoring. in sawatzky?s data there is only one case of DCs not accompanied by detectable bubbles. The negative predictive value is 0.999, and the sensitivity of detectable bubbles as a test for DCs is 0.99. on the other hand, there are 1,720 exposures where bubbles were detected without any symptoms of DCs, and the positive predictive value is only 0.04, so detectable bubbles only yields a 4% chance of having DCs. Even when using bubble grades iii and iv as test criteria, the positive predictive value is as low as 0.07. Therefore, the data strongly suggest that for sub-saturation diving the absence of detectable bubbles is a good indicator of decompression safe- ty, but the occurrence of bubbles, even high grades, is a poor predictor of decompression sickness. There are few studies on bubbles and DCs for saturation diving, and most existing studies are performed on small pop- ulations. in a study of 125 exposures in 29 heliox saturation dives, gardette (1979) found 23 cases of muscle or joint pain. Five of these cases were not accompanied by detectable gas bub- bles, yielding an incidence of 10% for bubble grade 0. Doppler- bubble detection was performed precordially at rest and after TaBle 1. DCIem data on Doppler detected bubbles and DCS incidence in bounce dives. ?Overall maximum grade? is the highest obtained score regardless of position (precordial or subclavian veins) or rest/movement. air dives Heliox dives Condition km grade Exposures DCs % DCs Exposures DCs % DCs Chest, rest 0 1264 7 0.6 945 6 0.6 i 131 0 0.0 105 2 1.9 ii 137 8 5.8 184 1 0.5 iii 191 25 13.1 272 22 8.1 iv 3 1 33.3 2 1 50.0 Chest, movement 0 1164 3 0.3 879 7 0.8 i 109 2 1.8 70 0 0.0 ii 111 3 2.7 114 1 0.9 iii 305 26 8.5 313 11 3.5 iv 37 5 13.5 132 13 9.8 overall maximum grade 0 819 0 0.0 623 1 0.2 i 287 3 1.0 214 1 0.5 ii 183 2 1.1 187 0 0.0 iii 365 27 7.4 347 15 4.3 iv 72 9 12.5 137 15 10.9 EFtEDal: UltRasoUnD FoR EvalUating DEComPREssion ? 85 Hamilton and schreiner (1989) concluded in a UHms work- shop that new procedures should be validated primarily by ex- tensive, dedicated laboratory testing before being put into the field for operational evaluation. traditionally the criterion of success for such testing has been the observed incidence of DCs compared to a certain predefined maximum level. The main problem with evaluating decompression procedures this way is the large number of dives required to determine DCs risk with any degree of certainty. From the binomial distribution, we find that more than 300 exposures with no incidents are needed to confirm a DCs incidence below 1% with a 95% confidence interval. For recreational diving, the incidence should be con- siderably lower than this, so the expenses and time required for validation can actually prevent new procedures from being im- plemented. also, this method for validation of procedures will inevitably lead to decompression injury among the test subjects. Furthermore, the DCs diagnosis is based on symptoms and is therefore subjective. Divers may choose not to report symptoms or may report symptoms due to apprehension or from other causes not related to the decompression (nishi et al., 2003). occurrence of vascular bubbles is a more objective observation that cannot be hidden by the diver. at Defence Research and Development Canada (DRDC, formerly DCiEm), bubble detection has been used for table validation for many years (nishi and Eatock, 1989). an em- pirically selected limit of km grade ii or greater in 50% of the subjects has been used to discriminate between stressful and acceptable procedures. simple criteria such as this will reduce the number of dives required for validation compared with ob- servation of DCs alone, but they do not make full use of the information obtained by bubble detection. With an established bubble/DCs correlation, bubble de- tection can be used to estimate the actual risk of having de- compression sickness when testing a decompression procedure. We have designed a method based on Bayesian statistics where each bubble grade is assigned a DCs risk based on established data. in our method we used sawatzky?s nitrox data from table 1. This set of risks is combined with observed bubble grades in a series of test dives for the new procedure. in this way, we can estimate the DCs risk and, more importantly, calculate a 95% credible interval for the estimate. a credible interval is a poste- rior probability interval, used in Bayesian statistics for purposes similar to those of confidence intervals in frequentist statistics. For illustration we have applied the method to bubble data from field tests of no-decompression limits for altitude diving. The experiments were carried out in the mountainous region of eastern turkey in a small lake at 3,500 meters above sea lev- el. six different combinations of depth and bottom time were tested, with 10 dives for each dive profile. Bubble detection was performed with Doppler and bubbles were scored according to the k-m scheme. maximum bubble grades for all dive profiles tested are shown in table 2. There were no cases of DCs or ad- verse effects of decompression. Based on a ?traditional? evalua- tion, we find that the DCs risk is 0-31% (binomial distribution, movement, and bubbles were graded according to the spencer scale. The study showed an increasing risk of muscle and joint pain with increasing bubble grade, with a 34% incidence for grade iii bubbles after movement. There were no cases of grade iv bubbles and no cases of serious decompression sickness. Hypobaric exposures are in principle similar to decompres- sion from saturation in that the subjects start the decompres- sion from saturation. in contrast to saturation diving, there are several large studies addressing the issue of bubbles and DCs for hypobaric exposures. one study by Conkin et al. (1998) re- ported on 1,322 exposures, mainly simulations of extravehicu- lar activities (Evas) in space. There were 480 exposures with de- tectable bubbles and 168 cases of DCs. The authors found that the absence of detectable bubbles was highly correlated with the absence of symptoms, with a negative predictive value of 0.98. For grades iii and iv, they found DCs incidences of 19% and 48%, respectively, and a corresponding positive predictive value of 0.39. The authors concluded that vascular gas bubbles is a necessary, but not sufficient, condition for DCs and that infor- mation about gas bubbles is useful to assess the risk of DCs in hypobaric decompressions. in another review study of 2,044 hypobaric exposures with 819 cases of DCs from the air Force Research laboratory DCs Research Database, Pilmanis et al. (1999) found a consider- ably higher DCs risk with lower bubble grades (grade 0-19%, i-26%, ii-50%, iii-52%, iv-62%). The studies contained in the database are all designed to provoke DCs in a substantial portion of subjects, and with an overall incidence of over 40% they have certainly succeeded. These figures question the useful- ness of Doppler detected bubbles in excluding DCs. in a study of various clinical manifestations of DCs in the same database, Balldin et al. (2004) found that only 27 out of 49 cases (55%) of central nervous system (Cns) manifestations were accompa- nied by detectable bubbles. They concluded that echo imaging has limited application for use as a predictor of altitude-Cns DCs. The bubble/DCs correlation seems to depend on the type of exposure. in general, the body can tolerate a substantial amount of gas bubbles without any signs or symptoms of de- compression sickness, and the presence of gas bubbles therefore has little diagnostic value. However, at least for sub-saturation diving, the absence of gas bubbles is an excellent indicator of decompression safety. bubblE dETEcTION fOr valIdaTION Of dEcOmprESSION prOcEdurES in a clinical setting, the absence of detectable vascular gas bubbles could theoretically be used as a tool for excluding the DCs diagnosis. in practice, however, this would call for surveil- lance by highly trained personnel at 20 - 30 minute intervals for at least two hours directly after surfacing, which is clearly not an option. a more feasible application for bubble detection is in the assessment of the safety of decompression procedures. 86 ? tHE FUtURE oF Diving 95% confidence interval) for each dive profile. This is obviously of little value for assessment of the procedures. if we apply the Bayesian method for estimating DCs risk based on detected bubbles, we find estimates for DCs risk and 95% credible in- tervals as shown in table 3. cONcluSION We conclude that the suggested method will greatly reduce the number of dives needed to validate or reject new decom- pression procedures. The use of bubble detection has other benefits, and there is evidence indicating that gas bubbles may cause subclinical damage in the absence of DCs, e.g., to the endothelium of the lung (Brubakk et al., 2005). Whether such subclinical damage may accumulate to cause irreversible damage is not known, but should be a point of concern (Eckenhoff, 1985). limiting the number of so-called silent bubbles (i.e., bubbles not causing clinical DCs) is therefore beneficial. another important point in favor of monitoring vgE is that the risk of inflicting DCs to test divers is reduced, as a procedure can be rejected before a single case of DCs is encountered. There is no standard criterion for acceptance or rejection of new decompression procedures with respect to DCs inci- dence. on-site recompression possibilities may justify use of procedures with higher DCs risk having longer bottom time or shorter decompression. For recreational diving, the DCs risk should be as close to zero as possible. our proposed valida- tion method evaluates the DCs risk and allows comparison of the risk for simultaneously tested procedures. The method can greatly reduce the cost of testing of new decompression proce- dures both for commercial and recreational diving. lITEraTurE cITEd Balldin, U. i., a. a. Pilmanis, and J. t. Webb. 2004. Central nervous system decompression sickness and venous gas emboli in hypobaric conditions. Aviation, Space, and Environmental Medicine 75(11):969?972. Boyle, R. 1670. new Pneumatical Experiments about Respiration. Philosophical Transactions 5:2011?2058. Brubakk, a. o., and o. s. Eftedal. 2001. Comparison of three different ultra- sonic methods for quantification of intravascular gas bubbles. Journal of Undersea and Hyperbaric Medicine 28(3):131?136. Brubakk, a. o., D. Duplan?i?, Z. valic, i. Palada, a. obad, D. Bakovi?, U. Wisl?ff, and ?. Duji?. 2005. a single air dive reduces arterial endothelial function in man. Journal of Physiology 566(3):901?906. Conkin, J., m. R. Powell, P. P. Foster, and J. m. Waligora. 1998. information about venous gas emboli improves prediction of hypobaric decompression sickness. Aviation, Space, and Environmental Medicine 69(1):8?16. Eatock, B. C., and R. Y. nishi. 1986. Procedures for Doppler ultrasonic moni- toring of divers for intravascular bubbles. Downsview: Defence and Civil institute of Environmental medicine. Report no. 86-C-25. Eckenhoff, R. g. 1985. The role of monitoring venous gas emboli after decom- pression. Undersea Biomedical Research 12(4):485?486. Eckenhoff, R. g., C. s. olstad, and g. Carrod. 1990. Human dose-response relationship for decompression and endogenous bubble formation. Journal of Applied Physiology 69(3):914?918. Eftedal, o., and a. o. Brubakk. 1997. agreement between trained and un- trained observers in grading intravascular bubble signals in ultrasonic im- ages. Journal of Undersea and Hyperbaric Medicine 24(4):293?299. gardette, B. 1979. Correlation between decompression sickness and circulating bubbles in 232 divers. Undersea Biomedical Research 6(1):99?107. Hamilton, R. W., and H. R. schreiner. 1989. Editorial summary: validation of decompression tables. in schreiner, H. R., and R. W. Hamilton, eds. Validation of Decompression Tables. The 37th Undersea and Hyperbaric medical society Workshop. Bethesda: Undersea and Hyperbaric medical society. Pp. 163?167. kisman, k., and g. masurel. 1983. method for evaluating circulating bubbles detected by means of the Doppler ultrasonic method using the ?k.m. code.? (English translation of 283 CERtsm 1983). toulon: Centre d?Etudes et de Recherches techniques sous-marines. nishi, R. Y., a. o. Brubakk, and o. s. Eftedal. 2003. Bubble detection. in Bru- bakk, a. o., and t. s. neuman, eds. Bennett and Elliott?s Physiology and Medicine of Diving. 5th ed. london: saunders. Pp. 501?529. nishi, R. Y., and B. C. Eatock. 1989. The role of bubble detection in table valida- tion. in schreiner, H. R., and R. W. Hamilton, eds. validation of Decom- pression Tables. The 37th Undersea and Hyperbaric medical society Work- shop. Bethesda: Undersea and Hyperbaric medical society. Pp.133?138. Pilmanis, a. a., n. kannan, k. m. krause, and J. t. Webb. 1999. Relating venous gas emboli (vgE) scores to altitude decompression sickness (DCs) symptoms. Aviation, Space, and Environmental Medicine 70(4):364. sawatzky, k. D. 1991. The relationship between intravascular Doppler-detected gas bubbles and decompression sickness after bounce diving in humans. msc Thesis. toronto: York University. spencer, m. P., and D. C. Johanson. 1974. investigation of new principles for human decompression schedules using the Doppler ultrasonic blood de- tector. onR technical Report (Contract n00014-73-C-0094). seattle: institute for Environmental medicine and Physiology. temple, D. J., R. Ball, P. k. Weathersby, E. C. Parker, and s. s. survanshi. 1999. The dive profiles and manifestations of decompression sickness cases after air and nitrogen-oxygen dives. technical Report no. 99-03, 2nd vol. Bethesda: naval medical Research Center. TaBle 2. Results from field tests of no-decompression bottom times for altitude diving. maximum bubble score regardless of location, graded according to the Kisman-masurel code. There were no cases of grade Iv bubbles. Depth [msw] 15 18 21 24 27 30 Bottom time [min] 38 29 24 21 17 14 grade 0 6 7 7 4 5 3 grade i 4 2 0 3 2 3 grade ii 0 0 1 1 1 3 grade iii 0 1 2 2 2 1 n 10 10 10 10 10 10 TaBle 3. estimated DCS risks and 95% CI for the no-decompres- sion limits as calculated by the Bayesian method for validation of decompression procedures. Depth [msw] 15 18 21 24 27 30 Bottom time [min] 38 29 24 21 17 14 P(DCs) [%] 1.7 1.9 2.2 2.5 2.4 2.3 95% Ci 0.7-3.1 0.8-3.5 1.0-3.9 1.3-4.1 1.2-4.0 1.2-3.8 INTrOducTION miniaturization of ultrasound imaging technology is at a point where small pocket-sized devices can be made with close to high-end image quality. For measurements of decompression stress as vascular gas embolisms, the technol- ogy enables field monitoring and research. in addition to miniaturization, the ultrasound systems are getting more sophisticated, and of particular interest is the increased use of ultrasound contrast agents. The technological development opens up new possibilities for monitoring and research in the diving and hyperbaric environments. Recently, a new method for detecting contrast agents has been developed at ntnU, which utilizes the nonlinear nature of the contrast agents (gas bubbles are easier to expand than to com- press). The method, called sURF (second order UltRasound Field), makes it possible to filter out specific parts of the reflected signal in order to detect the agent of interest. The sURF technology is as such well suited for decompression-induced microbubbles, since it can be designed to detect specific size gas bubbles and filter out all other disturbing signals. backgrOuNd The purpose of all decompression procedures is to prevent injury to the diver, and it is generally agreed that these injuries are caused by the formation of gas bubbles in the body. gas bubbles form in nearly all decompressions, and the risk of developing decompression sickness (DCs) increases with the number of gas bubbles (nishi et al., 2003). Paul Bert demonstrated the relationship over 100 years ago (vann, 2004), and his hypothesis was later central to Haldane?s theory (Boycott et al., 1908). today we know that bubble formation during decompression is not simply a consequence of inert gas aBSTRaCT. Central to Haldane?s theory, the risk of developing decompression sickness (DCs) increases with the number of gas bubbles. The introduction of ultrasound to detect vascular gas bubbles generated during and after decompressions has made it possible to compare different de- compression situations and models without going to the binominal endpoint of DCs or no-DCs. miniaturization of ultrasound imaging technology is progressing to small pocket-sized devices with close to high-end image quality. technological achievements are due to vast developments that have taken place in engineering fields such as acoustics, electronics, material physics, and signal pro- cessing, but some challenges such as power consumption and heat generation remain unresolved. Recently, a new method, second order UltRasound Field (sURF) for detecting contrast agents has been developed at ntnU, which utilizes the nonlinear nature of the contrast agents and filters out specific parts of the reflected signal in order to detect the agent of interest. Potential future uses in baromedicine include field monitoring, continuous monitoring, and development of implantable devices. Lasse L?vstakken and Svein Erik M?s?y, Ultrasound Research Group, Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Olav Kyrres gt 9, N-7489, Trondheim, Norway. Andreas M?llerl?kken, Baromedical and Environmental Physiology Group, Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Olav Kyrres gt 9, N-7489, Trondheim, Norway. Current trends in Ultrasound imaging technology, sURF imaging, and Decompression induced microbubbles Lasse L?vstakken, Andreas M?llerl?kken, and Svein-Erik M?s?y 88 ? tHE FUtURE oF Diving supersaturation, because numerous experiments indicate that bubbles originate as preexisting gas nuclei (Blatteau et al., 2006). a systematic study of the phenomena accompanying de- compression is complex and difficult, as practical measurement methods to monitor the processes taking place in body tissues are lacking. However, the introduction of ultrasound to detect vas- cular gas bubbles generated during and after decompressions has made it possible to compare different decompression situations and models without going to the binominal endpoint DCs or no-DCs. Recently, a Bayesian approach to validate decompres- sion procedures has also been developed, an approach which is based on detection of vascular gas bubbles (Eftedal et al., 2007). gas-filled microbubbles are not just interesting to research- ers and clinicians working with hyperbaric problems. The intro- duction of contrast agents for ultrasonic imaging, especially for cardiology and abdominal applications, are also a highly special- ized application for working with gas filled bubbles. Ultrasound contrast agents are generally of a size of 1?4 ?m, they are trans- pulmonary and provide strong reflections that can be detected. modern contrast agents are usually filled with perfluorocarbon gas, stabilized in a lipid monolayer shell. currENT TrENdS IN ThE mINIaTurIZaTION Of ulTraSOuNd ImagINg TEchNOlOgy since the advent of medical ultrasound imaging in the late fifties, a continuous improvement in terms of image quality and real-time operation has taken place, and an increasing number of clinical applications and image modalities have appeared. it is also important that ultrasound systems have decreased in size and are now available from portable high-end systems to hand-held low-end systems. The technological achievements are due to vast developments that have taken place in engineering fields such as acoustics, electronics, material physics, and signal processing. The motivation to develop ultrasound system hardware compo- nents of smaller size is driven by different factors, for example: 1. to develop small instruments with good image quality that can be operated anywhere in the field, driven by batteries. 2. in the development of real-time 3-D ultrasound imaging systems using 2-D phased-arrays, the amount of system channels is substantially increased, requiring too much space and dissipating too much heat unless electronic com- ponents can be made very small. 3. intra-cardiac and intra-vascular imaging using tiny trans- ducer arrays inside 1 mm catheters. Why arE TOday?S hIgh-ENd ulTraSOuNd SySTEmS NOT alrEady haNd-hEld? Fundamental technical challenges are present for transducer design and electronics, and trade-offs are needed at the expense of image quality. one of the main technological challenges in the miniaturization of ultrasound system components is making the high-voltage transmission electronics small. Further, the minia- turization of reception electronics and signal processing boards is progressing steadily, but power consumption and heat genera- tion is of major concern and needs to be reduced. one potential solution, given attention especially in high-frequency imaging today, is to integrate future transducer technology and electron- ics directly on silicon wafers for miniaturized production. WhaT cOuld ThE NExT-gENEraTION dIagNOSTIc ulTraSOuNd ImagINg SySTEm lOOk lIkE? Envision a PDa-sized instrument with extreme portability (pocket-sized, lightweight, thin cable), with an image quality close to high-end systems, a lower cost compared to high-end systems, an extended battery life for more than one hour, easy to use with a simple user interface, semiautomatic image optimiza- tion, and limited to specific clinical application functionality. indeed, instruments close to these specifications have already been launched on the market and in continuous smaller ver- sions from 2000 until today. a true hand-held instrument was recently released on the market from siemens (siemens P10), and more are sure to follow from the major manufacturers and smaller niche companies. pOTENTIal fuTurE uSES IN barOmEdIcINE some visions of the potential use of miniaturized ultra- sound instruments in baromedicine can be offered: ? Field monitoring: Hand-held systems with Doppler mea- surements and B-mode imaging can more easily be brought to the field for monitoring and research. ? Continuous monitoring: Extremely small battery operated devices the size of a diver watch could be strapped around the arm or around the neck for continuous monitoring during diving and decompression. indeed, small systems already exist for instance in fetal heart beat monitoring and for detecting pulse / perfusion during resuscitation. ? implantable devices: simple transducer devices for Doppler imaging can be made small enough to fit on implantable de- vices powered by an outside source. For example, embedded in stents or as a clamp around a peripheral vessel of interest. many more applications can surely be envisioned. as some additional pointers, in addition to miniaturization, higher fre- quency imaging will make it possible to detect smaller bubbles, and more advanced signal and image processing may improve methods for bubble detection. We leave this subject open for discussion! l?vstakkEn Et al. : tREnDs in UltRasoUnD imaging ? 89 Blatteau, J. E., J. B. souraud, E. gempp, and a. Boussuges. 2006. gas nuclei, their origin, and their role in bubble formation. Aviation, Space, and Envi- ronmental Medicine 77:1068?1076. Eftedal, o. s., H. tjelmeland, and a. o. Brubakk. 2007. validation of decom- pression procedures based on detection of venous gas bubbles: a Bayesian approach. Aviation, Space, and Environmental Medicine 78:94?99. m?s?y, s. E., ?. standal, P. n?sholm, F. Johansen, B. angelsen, and R. Hansen. 2008. sURF imaging: In vivo demonstration of an ultrasound contrast agent detection technique. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control. 55(5):1112?1121. nishi, R. Y., a. o. Brubakk, and o. s. Eftedal. 2003. Bubble Detection. in Bru- bakk, a. o., and t. s. neuman. eds. Bennett and Elliott? s Physiology and Medicine of Diving, 5th ed. london: saunders. Pp.501?530. vann, R. D. 2004. inert gas Exchange and Bubbles. in Bove, a. a., ed. Bove and Davis? Diving Medicine, 4th ed. Philadelphia: saunders. Pp. 53?77. The ultrasound research group at the Department of Circu- lation and medical imaging at ntnU in trondheim has devel- oped a new method for detecting contrast agents that gives better resolution and improves the sensitivity compared with traditional techniques. The method is called sURF (second order UltRa- sound Field) imaging, and instead of one pulse being transmit- ted as in traditional ultrasound transducers, the sURF method transmits two pulses simultaneously (Fig. 1). one pulse is used to manipulate the size of the microbubbles, and the other is used for detection of the change in size. By utilizing the nonlinear nature of the contrast agents (gas bubbles are easier to expand than to compress), one can filter out specific parts of the reflected signal in order to detect the agent of interest (m?s?y et al., 2008). in relation to decompression-induced microbubbles, it is accepted that the process of bubble formation starts with microscopic bubble nuclei. De novo formation of gas bubbles requires an unrealistic pressure gradient in relation to normal decompression activity. But, by definition, no one has actually seen these bubbles in nondiving subjects. They are believed to be on the order of 1 ?m, and there are no detection systems available today that can highlight such small bubbles. However, the sURF technology is well suited for decompression induced microbubbles, since it can be designed to detect specific size gas bubbles and filter out all other disturbing signals. The result is that in a standard ultrasonic picture where you can not see any bubbles at all, the sURF can visualize the presence of bubbles if they are of the same size as ultrasound contrast agents. and in theory, one can build transducers based on the sURF technol- ogy that can detect microbubbles down to 2 ?m. lITEraTurE cITEd Boycott, a. E., g. C. C. Damant, and J. s. Haldane. 1908. The prevention of compressed air illness. Journal of Hygiene, Cambridge 8:342?443. FIgURe 1. SURF-pulse complexes for contrast agent detection and imaging. For additional information, please visit: http://www. ntnu.no/surf/. INTrOducTION Historically, the diving community has depended predominantly on the United states navy air Decompression tables, a direct descendant of Haldane?s work, which has served divers well for over five decades. Dive computers, utilizing mathematical models of human tissue compartments and gas exchange, allow the con- stant computation of the diver?s decompression status during the dive. They vary in the assumptions incorporated in their models and in their capabilities. as predicted by lang and Hamilton (1989) these real-time tools now enjoy widespread use in the recreational, scientific and military diving sectors. logically, dive computer evolution was a natural progression from decompression tables and they have experienced several generations of development. Computers replaced the diver?s watch and depth gauge, provided greater accuracy and computerized, real-time, at-depth, continuous dive profile data, eliminating the need for the diver to remember tables and make decompression decisions while under water and while multilevel diving, allowed for longer bottom times than permitted by tables. many divers are highly motivated in their activities and interested in maximizing underwater time and efficiency. They view decompression requirements as a hindrance and distraction from their dive objectives, yet are generally concerned about safety. Evaluations of the available databases on pressure-related injuries to examine the effectiveness of dive computers showed that these devices had demonstrable advantages over dive tables. it remains clear that neither tables nor dive computers can eliminate all decompression problems, which have a probabilistic component to their occurrence. However, the current generation of dive computer technology represents an important tool for further improving diver safety. Divers alert network has managed to collect 172,000 dive profiles from 1999 to 2009 through its Project Dive Exploration (PDE), a aBSTRaCT. The age of electronic diving has arrived with the development of the modern elec- tronic dive computer as the most significant advancement in self-contained diving since the inven- tion of the aqualung by Jacques Cousteau. twenty-five years after modern day dive computer introduction, several key questions remain surrounding the decompression models used, vali- dation and human testing, acceptable risk, limitations, failures, and operational reliability. a brief history of analog dive computers and electronic digital computers and their function is discussed. Existing decompression models incorporated into dive computers are discussed with comments on the variety of approaches since Haldane. Educated predictions are offered on the functionality, features, and configurations of future dive computer evolution based on benefits from advances in consumer electronics technology, and monitoring technology integrated into the dive computer algorithm that allows for a closer approximation of physiological parameters. Final remarks conclude with how advances in diving physiology research based mainly on Haldane?s original work in 1908 will shape the dive computer landscape of the future. Michael A. Lang, Smithsonian Institution, Office of the Under Secretary for Science, PO Box 37012, MRC 009, Washington, DC 20013-7012 USA. Sergio Angelini, Angelini Consulting, Egg 28, 5444 K?nten, Switzerland. The Future of Dive Computers Michael A. Lang and Sergio Angelini 92 ? tHE FUtURE oF Diving worldwide study of recreational diving to record more than one million dives to produce statistically accurate analyses of depth profiles, diver characteristics, and diver behavior. This collec- tion of depth/time profiles for statistical analysis and modeling will assist in characterizing the relationship between diving and health effects, developing flexible, low-risk decompression pro- cedures for multilevel, multiday repetitive diving, and studying the effects of flying after diving. since the appearance of the first commercially mass-pro- duced electronic dive computer, the 1983 oRCa industries? EDgE model, the operational experience with dive computers is enormous, yet some key considerations remain: 1. Decompression models: What models are dive computers programmed with? Does it matter? should the manufac- turers specify the model in their brochures? What are the primary criteria for model effectiveness and ?acceptability?? 2. validation and human testing: What comprises an ac- ceptable validation protocol? should all computers be tested on human subjects with Doppler monitoring? if so, what type of dive profiles should be used, and what does this really prove? and the rejection criteria would be what exactly? Comparisons with existing decompression tables demonstrate the range of no-decompression limits (nDls) for tables and computers. For square-wave dive profiles, nDls of dive computers are generally more conservative. multilevel profile comparisons are more tenuous because of mechanical constraints of the organization of dive table limits versus real-time interval updates of dive computers. should the manufacturer publish validation data or divulge their modifications or adjustments of published algorithms? should they be evaluated by an independent agency? 3. acceptable risk: it is generally recognized that zero bends is unachievable and that for operational reasons sectors of the diving community accept different DCs rates. What levels of ?bends? risk are acceptable? 4. limitations: should depth and time limitations be im- posed on dive computers? if so, how are these determined? specifically, what is the applicability of dive computers with regard to long shallow dives, short deep ?bounce? dives, stage-decompression dives, repetitive multiday, multilevel dives, reverse dive profiles, variable ascent rates, diving at altitude, and desaturation levels for flying after diving. 5. Dive computer failure: What is a diver to do regarding decompression during or after a dive should the computer fail? are there standardized contingency plans to continue diving after a computer failure, or a requirement for a back- up dive computer? 6. Operational reliability: The operational experience has been generally good. are there specific dive computer com- ponent or battery failures? should the manufacturer pro- vide reliability data? The incidence of decompression sickness would appear to be an appropriate metric to evaluate the efficiency of dive comput- ers. assuming that the diver wore the computer, actually looked at it during the dive, and the computer can be interrogated by the hyperbaric chamber operator, useful dive profile information can be retrieved and used in treatment decision-making protocols. hISTOry The introduction of scuba in the mid-1940s changed div- ing operations that were carried out by hardhat divers using surface-supplied air for dives at single depths for as long as they needed to complete the mission while decompression status was monitored by surface tenders. scuba divers without surface con- tact now had to be responsible for their own decompression status under water. Without an unlimited air supply from the surface the repetitive dive concept became an actuality with the exchange of full scuba cylinders. Three-dimensional freedom of movement during a dive led to multilevel dive profiles. various mechanical and electrical analog and microproces- sor-based digital dive computers to determine a diver?s decom- pression status in real time have been produced since the in- crease of scuba in the 1950s. Current computers only use depth and time as variables to compute decompression status. Future computers should incorporate individual and environmental variations and additional variables that play a role in decom- pression sickness susceptibility, and perhaps ultimately monitor actual inert gas levels in the diver. The U.s. navy Committee for Undersea Warfare and Underwater swimmers met in 1951 at scripps institution of oceanography to identify improvements required in scuba diving equipment and how to control the decompression of a non-tethered, free swimming scuba diver. groves and monk?s (1953) report established the foundation for most of the early designs for decompression devices and presented a preliminary design for a diver-carried pneumatic analog computer which simulated nitrogen uptake and elimination in two theoretical tissue groups and summarized its benefits: The gauge automatically takes into account the depth-time his- tory of the entire dive. The resulting continuous ?optimum as- cent? should be somewhat more efficient than the usual step-wise ascent, the latter being used only because of its greater simplicity of presentation in tabular form. There are two other situations for which the gauge is conceivably an improvement over the table. For repeated dives the gauge automatically takes into account the residual elevation of nitrogen pressure in the body from the pre- ceding dives. (Divers are known to be more subject to bends on subsequent dives.) in the case of an emergency ascent, such as may be required by an exhaustion of breathing air, the gauge gives some indication of the desirable recompression procedure. This report also included a basic design for the ?Ultimate gauge,? an electrical analog computer that would show both decompression and air consumption status so that the diver would know if the remaining air supply would be sufficient to lang anD angElini: tHE FUtURE oF DivE ComPUtERs ? 93 ladder network of series resistors and parallel capacitors to simu- late nitrogen diffusion within the body. ambient pressure mea- surement was supplied by a depth sensor that varied the voltage supplied to the network. two 1/2D alkaline cells powered an oven that housed the electronics and kept them at a constant 90 oF. Four small mercury batteries were used as the computer network power source. The display was a micro-ammeter cali- brated in fsw displaying how many feet the diver could safely ascend. Workman (1963) found that minimal decompression requirements were adequately predicted for schedules through- out the depth range tested (40?190 fsw) for ascent rates of 20 and 60 fpm. longer and deeper exposures were not provided adequate depth and total decompression time at stops com- pared to the U.s. navy air decompression tables. Continuous ascent decompression predicted by the tRaCoR computer was inadequate both in depth and duration of total decompression time. temperature dependency of the instrument was excessive, particularly for cold exposures, and resulted in widely varying decompression requirements for the same dive schedule. Work- man (1963) further suggested that a mechanical analog com- puter could be used to avoid the instability and breakdowns that occurred in the electrical circuitry. 1962: DCIEM Analog Computer Series (Nishi, 1989) Developed by D.J. kidd and R.a. stubbs at the Defence and Civil institute for Environmental medicine (DCiEm) with four compartments to simulate the nitrogen absorption and elimination in the diver. initial versions? compartments were ar- ranged in parallel, the final design?s arranged in series, resulting in the kidd-stubbs decompression model (kidd and stubbs, 1966). The maRk v s was the first thoroughly tested, suc- cessful decompression computer. The four serial compartments gave effective half-times of 5 to over 300 mins (nishi, 1978). The display consisted of a depth gauge face with two needles: one to indicate the diver?s present depth, and the other to indi- cate the depth to which the diver could safely ascend (Huggins, 1989). The unit was small enough to fit into a housing 9 cm in diameter and 18 cm long, which could be easily carried by a scuba diver. another version of the device, called the maRk vi s, was designed utilizing the same algorithm for hyperbaric chamber use. The maRk v s was produced by spar aerospace in the late 1960s for sale to industrial and military agencies with operational depth limits to 60 msw. in 1970, spar developed a smaller and lighter version operational to 90 msw. Due to the complexity of construction, high manufacturing costs, and ex- tensive maintenance and calibration requirements, the maRk vs computer was not a commercially viable product for recre- ational divers. 1973: GE Decompression Meter Designed by Borom and Johnson (1973) utilizing semiper- meable silicon membranes to simulate nitrogen diffusion. These perform the required decompression schedule. searle (1956) indicated in a navy Experimental Diving Unit report the need for some type of decompression device because of the ever-widening fields of both civilian and military free-swimming diving using self-contained breathing apparatus. Particularly when scuba diving was untended from the surface, there arose a very pressing need for a small portable apparatus to be used by the diver to indicate proper decompression and ascent. Huggins (1989) thoroughly reviewed the history of dive computer evolution through 1988. AnAlog comPuters 1955: Foxboro Decomputer Mark I Designed by Hugh Bradner and mead Bradner, manu- factured by the Foxboro Company with 40- and 75-minute halftime compartments (both with 1.75:1 surfacing ratios), a pneumatic design, and 5 bellows (Fredrickson, 1956). nitrogen absorption and elimination from the compartments was simu- lated by the flow of gas through porous resistors between bellows, which were exposed to the ambient pressure, and bellows sealed in a vacuum, kept under a constant pressure by a spring. searle?s (1956) evaluation reported the actual compartment half-times simulated by the Foxboro Decomputer mark i as 27.7 and 52 minutes, causing deviations from U.s. navy table decompres- sion ranges for some dives. no further development occurred because the U.s. navy published new air no-decompression/de- compression tables and repetitive dive tables in 1957. The navy apparently rejected the idea of a decompression computer and accepted option ?a? of the groves and munk report, i.e., depth gauge, watch, tables, and diver wits (Huggins, 1989). 1959: SOS Decompression Meter Designed by Carlo alinari, manufactured by sos Diving Equipment limited as a one-compartment pneumatic comput- er with half-time variations with the pressure differential across the ceramic resistor. The ambient pressure increased on the flex- ible bag, forcing gas through the ceramic resistor (simulating nitrogen uptake and elimination in the body) into the constant volume chamber. The pressure increase was measured by the bourdon tube gauge, indicating the safe ascent depth to the div- er. on ascent, the gas pressure in the constant volume chamber became greater than the external pressure and the gas flow re- versed (Huggins, 1989). Howard and schmitt (1975) evaluated ten sos meters and determined their no-decompression limits to be more conservative than the U.s. navy limits at depths shallower than 20 msw, but less conservative at deeper depths. 1963: TRACOR Electrical Analog Computer Developed by texas Research associates inc. as the first elec- trical analog decompression computer, employing a 10-section 94 ? tHE FUtURE oF Diving membranes operated better than porous resistors because the simulated half-time of a compartment did not vary with depth (as in the sos meter). a four-chamber device was built to simu- late the U.s. navy air Decompression tables using compart- ment half-times of 24, 39, 90 and 144 mins. initial evaluations by gE showed that the membrane-based decompression meter concept was sound. The size of the unit could be reduced and temperature dependence was ?well within satisfactory limits.? However, no information on any subsequent development and testing was available (Huggins, 1989). 1975: Farallon Decomputer manufactured by Farallon industries, the device was a pneumatic analog computer utilizing four semipermeable membranes (two for gas uptake, two for elimination) that sim- ulated two theoretical tissue groups. air from the collapsible gas chamber flowed through the ?fast tissue? (large) and ?slow tissue? (small) membranes when exposed to elevated pressures. The increased pressure within the mechanism caused the pistons to move along the display color-coded green, yellow, and red, indicating the diver?s decompression status. When the ambient pressure was reduced to a lower pressure than inside the tis- sue simulator, the air flowed out through the ?repetitive dive membrane.? Both compartments had offgasing membranes that simulated a slow offgasing rate. testing at scripps institution of oceanography determined that the Farallon Decomputer failed to ?approximate? the U.s. navy air Decompression limits and repetitive dives proved even less acceptable, was too permissive, and developed too much mechanical deterioration with use (Flynn, 1978). digitAl comPuters The dive computer consists of a watertight housing with a through-hull pressure transducer that transforms pressure sensed through an analog-digital converter to the microproces- sor, powered by a battery. Read-only memory, random-access memory and a clock feed into the microprocessor, which out- puts information to the diver via the computer?s display (Fig. 1). Huggins (1989) outlined the evolution of a series of digital dive computers once the microprocessor revolution was underway in the mid 1970s. DCiEm unveiled the xDC Digital Decompres- sion Computer series using the kidd-stubbs model. The xDC3 Cyberdiver was actually the first diver-carried microprocessor- based underwater decompression computer. like the Cyberdiv- er, the DaCoR Dive Computer suffered from very high power consumption and was a Us navy dive table reader. Thalmann (et al., 1980; 1983; 1984) and Presswood et al. (1986) worked on developing an E-l (exponential linear) decompression model and algorithm to program into an Underwater Decom- pression Computer to be used with the Usn constant partial pressure of oxygen closed-circuit mixed gas system. This model assumed that nitrogen absorbed by tissues at an exponential rate (as in Haldanean models), discharged at a slower linear rate. in 1996, Thalmann?s vval 18 model was tested in the Usn?s Cochran navy dive computer. oRCa industries, inc. released the EDgE (Electronic Dive guide) in 1983, the world?s first commercially successful, mass produced electronic dive computer that paved the way into this new approach to decompression status monitoring. The oRCa 12-compartment model (half times from 5 to 480 mins) was based on no-decompression limits (to 130fsw) determined, in part, by Doppler ultrasonic bubble detection (spencer, 1976). The EDgE display was perhaps one of the most innovative to date, divided into graphical and digital information split into two sections by a curve (limit-line) representing the maximum pressure (m-values) allowed in the twelve compartments. one glance by the diver established whether all compartment bars were above the limit-line, indicating a no-decompression dive. The skinnyDipper (also distributed as a private labeling, sig- matech, by sherwood scuba) from oRCa industries utilized the same decompression model as the EDgE, but its simpler display scheme consisted of three numerical segments and no graphics. The smE-ml, a nine-compartment Haldanean mod- el with half-times ranging from 2.5 to 480 mins, is also based on Doppler research and was manufactured by sUUnto of Finland. it stored ten hours of dive information that could be recalled at any time after the dive. The Datamaster ii (also distributed as the Datascan ii by U.s. Divers Co.) was manu- factured by oceanic using a pseudo-Haldanean decompression model of six compartments with half-times of 5 to 120 mins. This model allowed no off-gassing from the compartments until reaching the surface. The Datamaster ii lead the way in calculat- ing air consumption, tank pressure and air time remaining. in 1979, the Hans Hass Decobrain i was a swiss table- based computer for high-altitude diving that could perform multi-level computations using the table?s repetitive group des- ignators but only by using the 80-minute half-time compart- ment, which easily put it ?out of range? as a decompression FIgURe 1. Dive computer schematic. lang anD angElini: tHE FUtURE oF DivE ComPUtERs ? 95 device. in 1985, the Decobrain ii by Divetronic was based on B?hlmann?s 16-compartment swiss model (ZHl-16) developed with compartment half-times ranging from 4 to 635 mins and designed for altitude diving up to 4500 meters above sea level. time to fly information was first introduced. The DaCoR microbrain (also manufactured by Divetronics) used six com- partments (4.5 to 395 minute half-times) that corresponded to the 16-compartment swiss model. The aladin (Uwatec), g.U.i.D.E. (Beuchat) and the Black Fox (Parkway) is the same unit manufactured by Uwatec with a 6-compartment version of the swiss decompression model utilizing four sets of m-values based on the altitude ranges of the dive. The Uwatec computer could be interrogated and the log entries for the last five dives recalled by activating two wet switches. The aladin Pro Plus in 1987 was likely the second commercially successful mass- produced dive computer. Huggins (1989) aptly concludes it is interesting to speculate about the present state of scuba diving if the Foxboro Decomputer mark i had performed properly and had been adopted for U.s. navy use in 1956. if so, the present U.s. navy air decompression tables might not have been comput- ed and the standard tool used to determine decompression status might have been a dive computer. Dive computer technology would be far more advanced, and more information and studies about the effects of multi-level diving would be available today. dEcOmprESSION mOdElS in 1908, John scott Haldane published a paper (Boycott et al., 1908) that to date represents the most significant milestone in decompression physiology. a multitude of researchers (Hills, 1966; Workman, 1963, 1965; B?hlmann, 1990) and many oth- ers over the years have published numerous versions of decom- pression models which, by and large, are all intrinsically linked to this century-old publication. as a diver descends in the water column and is exposed to increased ambient pressure, the partial pressure of the inhaled inert gas is higher than that of the dissolved inert gas in the various bodily tissues. This imbalance leads to inert gas travel- ling from the lungs via the blood stream throughout the body, where it is absorbed in the various tissues at a rate which is a function of the tissue itself (e.g., muscle tissue will ?load? up with inert gas faster than fat tissue). The characteristic by which a tissue loads with inert gas is defined by the term ?half time?, an artificial parameter that defines the time required for a tissue to equilibrate to within 50% of the imposed external pressure. similarly, as the diver ascends at the end of the dive and is exposed to a diminishing ambient pressure, the partial pressure of inert gas in a tissue will become higher than the partial pres- sure of the inhaled inert gas (supersaturation), and hence the inert gas transfer process is inverted. Excess inert gas is returned from the tissues via the blood stream to the lungs, from where it is eliminated by exhalation. The key concept in every form of Haldanean implementation is that decompression sickness is preceded by inert gas bubbles forming due to excessive supersat- uration. Therefore, a successful decompression strategy involves controlling the supersaturation in each tissue within defined values. The various versions of Haldanean models differ primar- ily in the number of tissues considered, their half times and their tolerance to supersaturation (up to the tipping point of bubble formation) and mathematical tricks are applied to cover a variety of influencing factors (e.g., cold, workload, repetitive diving). The primary reason for the success of Haldanean mod- els is that, in spite of their simplistic approach, a vast amount of data exists to which the models have been fitted. similar to the flower-like trajectory of mars around the Earth in a Ptolemaic view, enough empirical observation and data fitting can make any model yield excellent results within its tested range. During the 1980s the prevailing opinion was that bubbles formed during almost all dives, even those not producing any sign or symptom of decompression sickness. This prompted a new wave in decompression modeling that implicitly included bubble formation and growth, and its consequences to the div- er. as a main departure from the Haldanean model, inert gas was not only present in dissolved form, but also in free form as a bubble. David Yount proposed a free-phase decompression model, the variable Permeability model (Yount and Hoffman, 1986), michael gernhardt the tissue Bubble Dynamics model (gernhardt, 1991), and Wayne gerth and Richard vann (gerth and vann, 1997) the Probabilistic gas and Bubble Dynamics model. The most widely implemented model in a simplified version in a variety of dive computers is the Reduced gradient Bubble model (Wienke, 1990). gutvik and Brubakk (2009) are the proponents of Copernicus, and lewis and Crow (2008) pre- sented an introduction to their gas Formation model (gFm). Whereas Yount and Hoffman, and Wienke consider supersatu- ration as a mechanism to begin bubble formation, gernhardt, gerth and vann, and gutvik and Brubakk track bubbles from their initial form as microscopic nuclei and follow their evolu- tion and growth as the dive progresses. These latter models are of considerable higher mathematical complexity and cannot be solved within the realm of a modern microprocessor. The overarching goal of future dive computer models should be to more closely reflect the individual physiology of the diver, evolving as a true electronic instrument designed to solve a physiological problem. moon et al. (1995) reinforced that the probabilistic models on which tables and computers are based should reflect the individual reality of the divers, to enable them to conduct their dives in accordance with their individual characteristics. aScENT raTES, rEpETITIvE dIvINg, TImE TO fly, aNd mIxEd gaS fuNcTIONS Divers must adhere to the manufacturer?s recommended ascent rate, whether variable or uniform, which is an integral 96 ? tHE FUtURE oF Diving component of the algorithm?s tissue tension calculations. train- ing in, and understanding of, proper ascent techniques is fun- damental to safe diving practice, including mastering proper buoyancy control, weighting and a controlled ascent with a ?hovering? safety stop in the 10?30 fsw zone for 3?5 min (lang and Egstrom, 1990). it is in the ascent phase of the dive that computers reveal one of their strengths. Existing computers have maximum ascent rates that do not exceed 60 fsw/min from depth and many are limited to 30 fsw/min in shallower wa- ter. Future dive computer models may favor slower rates but we make the observation that, operationally, the 30 fsw/min is achievable and effective, while slower rates most likely are not. multilevel, multiday repetitive computer diving within the tested envelope is the mainstream practice today, and it appears to be less stressful than square wave profile diving. Deep re- petitive dives with short surface intervals should nevertheless be given special consideration. Because of limited analysis of the existing profile databases, no firm conclusions have been reached regarding repetitive diving limits to date (lang and vann, 1992). The maximum depth sequence of repetitive dive profiles is not restricted by dive computers. lang and lehner (2000) found that there was no physiological reason for prohib- iting reverse dive profiles for no-decompression dives less than 40 msw (130 fsw) and depth differentials less than 12 msw (40 fsw) because this was never a rule in either U.s. navy or com- mercial diving, but more of an operational constraint of the or- ganization of depth/time profiles in a square-wave table format. There exists no dependable distinction between ?safe to fly? and ?not safe to fly? in dive computers. There is a gradual reduction of risk for which the diver needs to choose an accept- able degree (e.g., wait at least 24 hours, the longer the wait, the further the reduction in probability of decompression sickness). lang and Hamilton (1989) provide examples of dive computer computations for ?time to fly? that include offgasing to 1?2 fsw (2?4 psi) over ambient pressure, waiting until 12 hrs have elapsed after the last dive, or not exceeding 0.58 bar as maxi- mum ceiling setting (against a minimum aircraft cabin pressure of 0.75 bar). adjusting oxygen fractions in dive computer software from 0.21 to standard oxygen-enriched air (nitrox) of 0.32 or 0.36 is simple and an available function of most computers. Huggins (2006) evaluated several dive computers capable of calculat- ing heliox and trimix dive profiles (the EmC-20H by Cochran Undersea technology, the Hs Explorer by Hydrospace Engi- neering, the nitek He by Dive Rite, and the vR3 by Delta P technology). The decompression software that purportedly emulated these four dive computers was used to calculate the response to specific 300 fsw/20 min total bottom time (tBt) dive scenarios, including decompression gas switches. Huggins opined that in surface-supplied mixed-gas operations diver-car- ried dive computers are best used as a backup and that the ma- jor control of decompression should be assigned to the surface- support personnel using a preplanned set of heliox or trimix tables that the dive computer emulates. ThE fuTurE: fuNcTIONalITy, fEaTurES aNd cONfIguraTIONS The dive computer of the future will benefit from advances in science and technology. These can be grouped into three dis- tinct categories: benefits from advances in consumer electronics technology, monitoring technology integrated in the algorithm, and advances in decompression physiology research. beneFits From AdVAnces in consumer electronics technology The combined worldwide sales of dive computers from all manufacturers does not exceed 500,000 units per year, while apple alone sold over 30 million iPhones in the first 12 months. it becomes obvious then that dive computers do not drive new technologies, but rather benefit from a trickledown effect. in a world dominated by PDas, smartphones and iPods, not only is the technological development unbridled, but the cost of these new technologies keeps declining and becoming more affordable. Hence, in spite of the relatively small volumes of dive computers produced, we can expect to start seeing more and more advanced embedded technologies. other outdoor activities, such as hik- ing, climbing and camping, are also promoters of new technolo- gies that can find an application within a dive computer. High Resolution Color Display Barring a few exceptions, dive computers today utilize a segment display. in these types of displays, information is pre- sented by ?turning on? certain segments within a large array. Due to the constraints of fitting a wide variety of informa- tion on a small display, segment displays typically present only numbers and symbols. advantages of this technology are low energy consumption and very sharp representation. The main disadvantage, however, is the inability to show anything other than what is ?preprogrammed? into the display. This means that any interaction between the diver and the computer takes place through a display of numbers and symbols. in an emergency situation, the diver sees blinking symbols and/or numbers and from this has to infer the nature of the emergency and take ap- propriate action. The possibility exists that, if the diver does not recognize or otherwise understand the meaning of the blinking symbol, this can lead to an increase in stress in the diver and could potentially precipitate a risky situation. The switch to a high resolution color display is the most obvious consequence of the proliferation of PDas and smart- phones. Color dot-matrix displays can play an important role in enhancing the safety of the dive in many ways: a. Before the dive: menu navigation via text in a language of choice means simplicity and clarity in setting up the com- puter for the dive; b. During the dive: one obvious advantage is the clear lang anD angElini: tHE FUtURE oF DivE ComPUtERs ? 97 representation of all relevant information, possibly with a choice of font size and in a pattern customized by the user. in addition, the combination of text and color can be tremendously helpful in alerting the diver of a potentially risky situation by describing the exact nature of the prob- lem and recommending a course of action. For instance, a diver on nitrox exceeding the maximum operating depth of the breathing gas would see a clear text message such as max oPERating DEPtH ExCEEDED (the nature of the problem) followed by a clear text message such as as- CEnD to 40 msW (the recommended course of action). a dive computer with a standard segment display cannot do more than beep madly and show blinking symbols; and, c. after the dive: logbook viewing function with several pages of information, including a graphic representation of the dive. Rechargeable Battery today?s computers function well with replaceable batteries, allowing between 100 and 800 dives before the battery runs out. in most cases replacing the battery is a very simple process which, combined with a battery price of a few dollars/euros, makes this an attractive solution. Reliability, an important fac- tor in a life support system, is also very high in this configura- tion. Color displays, however, require higher energy consump- tion and thus the switch to a rechargeable battery becomes necessary. With a typical lifetime of 500 charge/discharge cycles and assuming 5 to 10 dives on each full charge, this would al- low 2500 to 5000 dives before the battery needs to be replaced. Charging of the battery can take place via a UsB cable con- nected to a PC or directly to a power outlet, or, as in the case of the UEmis scuba Diver assistant, via solar cells. GPS Receiver gPs receiver use has become widespread in outdoor instru- ments and the automotive industry, where its role is of much higher importance and obvious benefit than in a dive computer. gPs works only through air, hence on the surface, and therefore an application in a dive computer might seem inappropriate. However, it would allow divers to locate dive sites simply by recording their gPs coordinates. additionally, at the end of the dive, the emerging diver would be able to estimate the distance and direction from the point of entry (boat or shore), which could be useful in a situation of low visibility. Underwater Communication and Navigation Communicating underwater with the dive buddy or even all the way to the dive vessel would represent an enormous step forward in diver safety (but perhaps not necessarily in dive en- joyment, because many divers love diving for the peace and quiet provided by the silent world). Furthermore, with the proliferation of navigation systems in automotive technology, it seems only logical to have similar gadgets guiding us through a dive. Data transmission underwater over a certain distance requires the use of ultrasound technology. Radio frequency, as utilized for instance for the transmission of tank information from a sensor on the first stage regulator to the dive computer, is strongly attenuated by water and thus would require too much power to be useful over a longer distance. Ultrasound, on the other hand, can travel very far underwater with relatively little power. Unfortunately, ultrasound is not necessarily a universal technology in consumer electronics, hence its integration in a dive computer may not be in the near future. attempts have been made though, and for professional use there are voice communication systems which, though bulky, function rather well. gPs-like underwater navigation would require the repro- duction of a satellite system for triangulation (set of buoys that translate the gPs signal from the surface to an ultrasound sig- nal underwater) and, as such, would be costly and cumbersome. simpler devices, which only show the direction and distance to the boat, have been introduced several years ago (Uwatec nEv- ERlost, Desert star systems DivEtRaCkER) but have not enjoyed extensive market penetration. EPIRB Emergency Position indicating Radio Beacon (EPiRB) is a distress signal technology utilized in the maritime industry. EPiRBs are tracking transmitters that aid in the detection and location of boats, aircraft, and people in distress. strictly speak- ing, they are radio beacons that interface with Cospas-sarsat, the international satellite system for search and rescue (saR). When activated, such beacons send out a distress signal that, when detected by nongeostationary satellites, can be located through triangulation. often using the initial position provided via the satellite system, the distress signals from the beacons can be homed in on by saR aircraft and ground search parties who can in turn come to the aid of the concerned boat, aircraft, or people. For instance, should a diver get carried away by a cur- rent during a drift dive, an EPiRB built into the dive computer would allow for a relatively quick location and rescue. The re- lated technology is unfortunately rather costly and most divers may never need to be rescued at sea. beneFits From monitoring technology integrAted into the Algorithm The principal objective of a dive computer is to recom- mend an ascent schedule as a result of the diver?s exposure to a specific depth/time profile. The depth defines the inert gas partial pressure in the inhaled breathing gas which, combined with the length of the exposure (time at depth), drives the in- ert gas uptake into the diver?s tissues. Clearly, perfusion (blood circulation through the body) plays a significant role in that it transports the inert gas through the body from and to the lungs. Consequently, a change in perfusion during the dive, as may be 98 ? tHE FUtURE oF Diving induced by exercise (increased perfusion) or exposure to cold (vasoconstriction in the arms or legs, hence a reduced perfu- sion), is expected to play a role in the ongasing and offgasing of inert gas. in particular, if the perfusion was increased during the deeper parts of the dive when much inert gas uptake is oc- curring, and/or the perfusion were reduced during the shallower parts towards the end of the dive, when inert gas elimination is occurring, the simplistic approach of considering only inert gas partial pressures may not be sufficient. in today?s dive com- puters evidently enough conservatism is built in to cover these effects, as evidenced by the relatively low incidence rates of de- compression sickness. There are attempts to account for changes in perfusion. one approach is to lump any deviation from a ?normal? expo- sure into additional conservatism in the model (?personal fac- tors?). The clear disadvantage of this method is that the diver needs to define and predict before the dive whether strenuous exercise or chilling is expected to occur. The other approach, followed at the moment only by UWatEC and UEmis, is to evaluate changes in perfusion based on actual measurements during the dive. an increase in workload is measured either by heart rate monitoring (UWatEC) or by a change in breath- ing pattern (UWatEC and UEmis). Cold water effects, which theoretically could lead to a reduction in perfusion during the decompression phase of the dive, are based on the concept that the colder the water, the more vasoconstriction plays a role (angelini, 2007). a thermally insulated diver, however, may be warmer in 4 oC water than a poorly protected diver in the Caribbean, and here a pre-dive set cold factor could be more practical. Regardless, in spite of the theoretical validity of the effect of changes in perfusion during the dive, the actual implementa- tion within a decompression model has not been experimen- tally validated or clinically proven. a thorough review of cold as decompression sickness stress factor was performed by mueller (2007). one can argue that diving is a reasonably safe activity and that therefore these model complications are uncalled for. another point of view is that this is an indication of excessive conservatism in today?s models so that, with proper implemen- tation of these phenomena, a diver could enjoy more freedom. However, as advances in science and decompression physi- ology are made, we propose the continued development of the following technologies: 1. heart rate monitoring. There is a proliferation of heart- rate monitoring devices in most outdoor and fitness activi- ties. as people become more aware of the importance of exercise to their well being, they also discover heart rate monitoring as an excellent tool for fitness evaluation. Re- cording the heart rate during a dive can be useful, besides from an implementation of workload-related nitrogen cal- culations, to become aware of how the body responds to the environment, leading to either increased comfort and enjoyment (recorded heart rate is low and consistent) or the avoidance of certain types of stressful dives (high and/ or erratic heart rate). 2. Skin temperature measurements. vasoconstriction is the result of the brain?s recognition that the core body tempera- ture is diminishing. in order to maintain the function of critical body parts, the brain reduces blood circulation to the limbs (arms and legs) with their large surface to volume ratio to reduce heat loss and protect the heart, lungs, and brain. skin temperature measurements transmitted to the dive computer would allow for a quantification of the cool- ing. in addition to an implementation of vasoconstriction in the decompression model, this could be very important as an alarm trigger for approaching hypothermia. Hypo- thermia is an acute danger when pain and feeling of cold disappears once the body gives up on shivering as a mecha- nism of generating warmth. 3. Oxygen saturation measurement. This is of primary in- terest to free divers because the risk of oxygen depletion and consequent shallow-water blackout is high. neverthe- less, this and other blood monitoring technologies could find applications in scuba or rebreather diving. 4. Inert gas bubble detection. inert gas uptake and conse- quent offgasing is in and of itself not the cause of decom- pression sickness. Problems can occur when the combina- tion of excessive amounts of inert gas dissolved in the body and a diminishing ambient pressure lead to the gas coming out of solution and forming free gas bubbles in the body. some decompression models attempt to describe this free gas formation, with all the complexity that follows from the physics associated with such an event. it would be very use- ful if it were possible to detect bubble formation during the dive, integrated into a feedback loop into the decompres- sion algorithm (regardless of the nature of the algorithm itself ). There exist, however, two rather large obstacles to this. First, the bubble detection technology existing today is based on ultrasound or Doppler monitoring, both re- quiring rather cumbersome equipment that could hardly be placed on a diver during the dive. The second problem is that bubbles really do not grow to a discernible level un- til 20 to 40 minutes after the dive, so that in-water detec- tion might only be useful in extreme dives in which some- thing went seriously wrong. on the other hand, this line of thinking could lead to the development of a similar or new kind of technology aimed at detecting a physiologically vi- able parameter that gives an indication of decompression stress in the body. any parameter that gives online feedback into a decompression model as to the state of the diver with respect to potential DCs would be a tremendous benefit. beneFits From AdVAnces in decomPression Physiology reseArch as described above, decompression models in existence to- day are, aside from a few mathematical manipulations, almost lang anD angElini: tHE FUtURE oF DivE ComPUtERs ? 99 entirely based on the ideas of John scott Haldane presented in the historic 1908 paper (Doolette, 2009). actual bubble models that carry out the pertinent bubble-dynamics calcula- tions (and the related nonlinear differential equations) are too complex to be managed by the limited processing power of a dive computer microprocessor. Even if this were to be solved, what remains is the need to validate such a radically different approach to decompression. There is an attempt by the norwe- gian University of science and technology to build a complete bubble model under the project name Copernicus. as much as the earth-centered planetary model was intrinsically flawed yet allowed reasonable ocean navigation via immense empirical ob- servations of the movements of the stars (made to fit this wrong model), a sun-centered planetary system yielded much better and accurate results allowing for significant broadening of the range of validity of the model itself. Haldanean theory, which does not consider inert gas in free form, and consequently its effects on the human body, has been refined over a century with the input of Workman (1963), B?hlmann (1990), Thalmann et al. (1980), and Thalmann (1983; 1984) to name a few, and provides us today with an extremely valuable and powerful tool in spite of its underlying wrong assumption. Copernicus, the decompression model, is the attempt to find the sun-centered model for decompression physiology, yet at the moment the wealth of data with which the existing models have been refined gives the Haldanean-based models a clear advantage. science and its related research should nevertheless persist in the pursuit of the ?truth.? a full bubble model availability within a few years would be very welcome. such a model should incorporate those aspects of relevance in approximating the human body such as body fat, age, gender, and fitness level. cONcluSION Electronic dive computers have for all practical purposes replaced dive tables in recreational and scientific diving and are increasingly implemented in particular segments of the military diving community. For the commercial diving industry and its standard operating methods of surface-supplied/controlled div- ing or saturation diving, a dive computer?s advantages in moni- toring decompression status appear to be minimal. it would not be unreasonable to state that regardless of the number of algo- rithm variations incorporated in modern dive computers, they all appear to fall within an acceptable window of effectiveness based on available databases of pressure-related injuries. it is also clear that neither tables nor dive computers can eliminate all decompression problems, but if utilized conservatively, com- puters have emerged as an important tool for the improvement of diver safety. all things considered, the dive computer?s functions of as- cent rate monitoring, real-time computation of nitrogen bal- ances, air consumption monitoring and profile downloading capability form a solid, reliable basis for advancements that will emerge in the future. Benefits from advances in consumer electronics technology will undoubtedly incorporate features such as high resolution color display, rechargeable battery, gPs receiver, underwater communication and navigation, and EPiRB. Further, benefits from monitoring technology integrat- ed in the dive computer algorithm will surely include heart rate monitoring, skin temperature measurements, oxygen satura- tion monitoring, and perhaps even inert gas bubble detection. We can only imagine the progress that John scott Haldane?s brilliant decompression insight would have made had the dive computer tools available to us now and in the future been avail- able to him 100 years ago. lITEraTurE cITEd angelini, s. 2007. Cold water decompression. in lang, m. a. and m. D. J. sayer (eds.) Proceedings of the International Polar Diving Workshop. sval- bard, march 15?21, 2007. Washington, DC: smithsonian institution. pp. 55?62. Borom, m. P., and l. a. Johnson. 1973. a Decompression meter for scuba Div- ing Utilizing semipermeable membranes. Report x73CRD129. general Electric Company Corporate Research and Development. B?hlmann, a. a. 1990. tauchmedizin: Barotrauma, gasembolie, Dekompres- sion, Dekompressionskrankheit, Dekompressionscomputer (german Edi- tion). Berlin: springer verlag. 148 pp. Doolette, D. 2009. Haldane still rules! in lang, m.a. and a.o. Brubakk, eds. The Future of Diving: 100 Years of Haldane and Beyond. Washington, DC: smithsonian institution scholarly Press. Pp. 29?32. Fredrickson, P. 1956. The analysis and testing Decomputer (mark i), Report mR-138, Foxboro, mass.: The Foxboro Company. gernhardt, m. l. 1991. Development and evaluation of a decompression stress index based on tissue bubble dynamics. Ph.D. Dissertation, University of Pennsylvania. 328 pp. gerth, W. a., and R. D. vann. 1997. Probabilistic gas and bubble dynamics models of decompression sickness occurrence in air and n2-o2 diving. Journal of Undersea and Hyperbaric Medicine 24:275?292. groves, g., and W. munk. 1953. a decompression gauge for divers. sio Refer- ence Report 53?64. san Diego: University of California, scripps institu- tion of oceanography. gutvik, C. R., and a. o. Brubakk. 2009. a Dynamic two-Phase model for vas- cular Bubble Formation During Decompression of Divers. IEEE Transac- tions on Biomedical Engineering 56(3): 884?889. Hills, B. a. 1966. a thermodynamic and kinetic approach to decompression sickness. Doctoral Thesis, the University of adelaide, australia. 370 pp. Howard, R., and k. schmitt. 1975. The Decompression meter - another look. The Undersea Journal 3:9?13. Huggins, k. 1989. The history of underwater decompression devices and com- puters. in lang, m.a. and R.W. Hamilton. (eds.) Proceedings of the AAUS Dive Computer Workshop, UsC Catalina marine science Center. Costa mesa: american academy of Underwater sciences. Pp. 7?29. ???. 2006. Evaluation of dive computer options for potential use in 300 fsw heliox/trimix surface-supplied scientific diving. in lang, m.a., and n.E. smith (eds.) Proceedings of the Advanced Scientific Diving Workshop. Washington, DC: smithsonian institution. pp. 47?64. kidd, D. J., and R. a. stubbs. 1966. analog Computer solution of Decompres- sion. Defense Research medical laboratories. toronto: Defence and Civil institute for Environmental medicine. lang, m. a., and R. W. Hamilton (eds.) 1989. Proceedings of the AAUS Dive Computer Workshop, UsC Catalina marine science Center. Costa mesa: american academy of Underwater sciences. 231 pp. lang, m. a., and g. H. Egstrom (eds.) 1990. Proceedings of the Biomechanics of Safe Ascents Workshop, Woods Hole, massachusetts. Costa mesa: ameri- can academy of Underwater sciences. 220 pp. 100 ? tHE FUtURE oF Diving lang, m. a., and R. D. vann (eds.) 1992. Proceedings of the Repetitive Diving Workshop, Duke University medical Center. Costa mesa: american acad- emy of Underwater sciences. 339 pp. lang, m. a., and C. E. lehner (eds.) 2000. Proceedings of the Reverse Dive Pro- files Workshop. Washington, DC: smithsonian institution. 295 pp. lewis, J. E., and s. Crow. 2008. introducing the gas Formation model. Under- sea Journal 3rd quarter. Rancho santa margarita: Professional association of Diving instructors. Pp. 57?59. moon, R. E., R. D. vann, and P. B. Bennett. 1995. The Physiology of Decom- pression illness.?Scientific American. august issue. Pp. 54?64. mueller, P. H. J. 2007. Cold stress as DCs factor. in lang, m. a. and m. D. J. sayer (eds.) Proceedings of the International Polar Diving Workshop. svalbard, march 15?21, 2007. Washington, DC: smithsonian institution. pp. 63?72 nishi, R. Y. 1978. Real time Decompression monitoring by Computers. in Hyperbaric Diving Systems and Thermal Protection, oED vol. 6, Johnson, C.E., m.l. nuckols and P.a. Clow (eds.). new York: The american soci- ety of mechanical Engineers. Pp. 25?38. ???. 1989. Dive Computer Experience in Canada. in lang, m. a. and R. W. Hamilton (eds.) USC Catalina Marine Science Center. Costa mesa: american academy of Underwater sciences. Pp. 69?79. Presswood, C., E. Thalmann, and m. Curley. 1986. test and Evaluation of two Prototype model Underwater Decompression monitors. nEDU techni- cal Report no. 10?86. Washington DC: navy Experimental Diving Unit. searle, W. F. 1956. Foxboro Decomputer mark i. navy Experimental Diving Unit technical Report no. 7?57. spencer, m. P. 1976. Decompression limits for Compressed air Determined by Ultrasonically Detected Blood Bubbles. Journal of Applied Physiology 40 (2): 229?235. Thalmann, E., i. Buckingham and W. H. spaur. 1980. testing of Decompres- sion algorithms for use in the U.s. navy Underwater Decompression Computer, Report 11?80, navy Experimental Diving Unit. Thalmann, E. 1983. Computer algorithms used in Computing the mk 15/16 Constant 0.7 ata oxygen Partial Pressure Decompression tables, Report 1?83, navy Experimental Diving Unit. ???. 1984. Phase ii testing of decompression algorithms for use in the U.s. navy Underwater Decompression Computer, Report 1?84, navy Experi- mental Diving Unit. Wienke, B. R. 1990. Reduced gradient Bubble model. International Journal of Biomedical Computing?26: 237?256. Workman, R. D. 1963. Evaluation of a Decompression Computer Developed for Divers. nEDU technical Report no. 1?63. Washington DC: navy Experimental Diving Unit. ???. 1965. Calculation of Decompression schedules for?nitrogen-oxygen and Helium-oxygen Dives. nEDU technical Report no. 6?65. Wash- ington DC: navy Experimental Diving?Unit. Yount, D. E., and D. C. Hoffman. 1986. on the use of a bubble formation model to calculate diving tables. Aviation, Space, and Environmental Medi- cine 57(2):149?156. m. gennser: The idea of self-experimentation was passed on by J. s. Haldane to his son J. B. s. Haldane who during the war spent many hours in chambers experiencing oxygen convulsions with kenneth Donald and others. That became the ethics of environmental physiology back then, performing experiments on oneself. another interesting aspect of his decompression sickness work, notwithstanding today?s competition between gas-loading models and bubble models, is in his second conclu- sion of his 1908 paper where he states that the gas wash-out will follow the same kinetics as the gas wash-in, unless there is bubble formation. a. Brubakk: Right, during World War i, because of the gas attacks, he also played an enormous role by studying the physi- ological effects of different gases. He performed a number of self-experiments breathing cyanide in order to develop the best gas masks. Haldane was in just about every field of environmen- tal science at the time. O. ellingsen: Can you elaborate on your statement that Hal- dane was never a professor and there were probably political rea- sons? also, Boycott?s name was first author on the 1908 paper. a. Brubakk: Boycott was not a scientist, he was a navy man. Haldane?s uncle, John Burdon-sanderson, held the Waynflete Chair in Physiology at oxford where he stayed all his life. it is a complicated story, but it is clear that Haldane stepped back and did not apply for a professorship when he could have done so because they thought it would be politically better to have other people in that position. as far as we know, he never got a pro- fessorship. That was maybe good because he had other things to do. B. gardiner: i came close to Haldane in that i actually had tea with his daughter naomi mary margaret mitchison while sit- ting at his house in scotland. naomi, 97 at the time, talked to me about daddy, which was quite profound. a. hope: Does bubble formation have anything to do with high-altitude sickness? Discussion: Decompression Physiology and methodology Chairman: Otto I. Molvaer R. moon: at the rate at which mountaineers can climb, proba- bly not. The rate of decompression is so slow that at least in situ bubble formation due to supersaturation is very unlikely. if you read some of the accounts of balloon flights, the participants often describe inability to move during their high-altitude ex- posure. one wonders whether they were perhaps suffering not only from hypoxia but also from decompression sickness. one other hypothesis was raised a few years ago: in the presence of hypoxia, there could conceivably be acidosis, particularly within the brain, that could cause bubbling due to Co 2 . The rate at which the acidosis occurs, if it even exists, given the high buffer- ing capacity of the blood, makes that scenario unlikely. Bubble formation is therefore probably not important, except in situa- tions where you can increase altitude very rapidly as in balloon flights or military fighter aircraft. a. Brubakk: one revealing fact that may point in the direction that bubbles may have an important role is nitric oxide. if you can give no, you can more or less totally eliminate the amount of gas bubbles. We also know that in hypoxia, no production is reduced. m. gennser: Has there been any follow-up to groves? experi- ments where he exercised tibetans and high-altitude living south american indians and measured pulmonary artery pres- sure? The indians do not seem to have adapted in the same way after living 1,000 years at altitude. R. moon: as i recall the numbers, the tibetans have been liv- ing at altitude for 20,000 years, perhaps 5,000 generations that seems to have resulted in this adaptation. The slide i showed you from Beall is the most recent study i know of suggesting that the mechanism of the tibetan adaptation may be related to no, which they have more of in their blood. The results show they had higher blood flow and oxygen delivery to the forearm and other areas as well. R. Richardson: myoglobin spectroscopy is a good mechanism to measure intercellular Po 2 . R. moon: i believe we are getting closer to where we want to go with that. 102 ? tHE FUtURE oF Diving ?. Duji?: on the high-altitude pulmonary oedema, there are some publications related to polymorphism where subjects are more susceptible to increasing pulmonary artery pressure be- cause of less nitrates in the blood, so it might be that there is some genetic predisposition to being more sensitive or resistant to the same level of hypoxia. R. moon: That disease seems to be an individual susceptibility issue due to genetic predisposition. D. Doolette: in C. Balestra?s model diagram of decompression sickness, the focus in the beginning was almost entirely on vas- cular bubbles, but there is the possibility that bubbles forming in the tissue are also involved. a. hope: on the supersaturation issue, can you say anything about this area under the tissue curves? You can have fast tissues that are saturated when you stop. What about the slow tissues in air-diving decompression? D. Doolette: That is probably the reason why when you look at the no-stop data, they came up with much greater allowed supersaturation for fast tissues than for slow tissues. Fast tissues will have a greater supersaturation and a greater driving force for bubble formation, but that only lasts a short amount of time. in our probabilistic models, where we are fitting back to data, the fastest compartments do not contribute equal amounts, be- cause we have scaling factors that allow a fast compartment to be more or less important than a slow compartment as far as how much it contributes to the risk. i do not remember what those values are so i cannot tell you what the relative contribu- tion is. We find that the medium and slow compartments con- tribute most to risk, while the fast compartments you only need to describe in submarine escape data. a. Brubakk: You have decompression models that have super- saturation, and they are theoretical concepts. Then you evalu- ate how these theoretical models fit with your model by doing experiments where the end result is clinical decompression sick- ness. This is a very vague endpoint. a very well-known decom- pression researcher said the symptoms of decompression sick- ness are very different, with more symptoms than syphilis and diabetes combined. The problem is in the use of an endpoint that is poorly defined. D. Doolette: There is indeed difficulty with symptom defini- tion, particularly when looking back at historical data. one of the great utilities of this technique for us is that we have data back to the 1940s. it is a way for us to use this huge body of information that we have, about 8,000 dives with that as end- point, and driven modeling in that direction. it is difficult when doing a study on how to classify the very vague symptoms that people get. We do it purely by whether the diver was diagnosed with DCs by a Diving medical officer. Divers tend to know when they get bent. a. Brubakk: it is not a problem when you have serious symp- toms and the diver keels over and cannot walk. The issue is the subtle symptoms because we do not know their pathophysiology or importance. What does it mean to grade symptoms, because this will have an effect on the endpoint? D. Doolette: laurent at Duke University has been looking into marginal symptoms which are not quite decompression sick- ness, but the diver is not quite well. He has cast some doubt on whether we should be counting those at all in our models. g. perdrizet: You are really fine tuning the supersaturation to the tolerance of the individual. other examples of tolerance to extreme environments of the individual may vary. D. Doolette: That is an interesting point. it is actually an un- avoidable part of the way that we study that we have to ignore that issue to an extent. our modeling assumes that every dive is independent and yet it is not because we use the same divers over and over again. in that man-trial where we had just under 400 dives, there were 81 or 84 divers, i cannot remember ex- actly because there were two trials going on at the same time. some divers dived between 1 and 12 times on that study. only one person only dived once, but everyone else in that study dived both profiles at least once. There is a problem with the population we can draw from. From the historical data that we have, it would be very hard to reconstruct how many times div- ers dived. There are techniques for looking at that; you can use mixed effect modeling and look at the individual as a random variable as well. p. lindholm: Do you have that data now for those dives? D. Doolette: Yes, we could reconstruct it. if we go back to our data starting in the 1940s, we could probably find that data by going back to the original logs, but we do not have some of the old material on hand at the moment. ?. Duji?: These military divers exercise regularly, which of course makes a difference with respect to bubble formation and their effects on the body. Do you have information of what they do regularly besides diving? D. Doolette: That information is not very well characterized. We had done vo 2 measurements on a few divers so we know what sort of vo 2 max values a typical diver has. it is a huge range of divers we have from very young guys to senior people. They all have to pass a physical exam annually, and they all do physical training at least three times per week, but i do not have actual values. ?. Duji?: Do you take the pre-dive exercise regimen into account? D. Doolette: all we have them do is fill out a questionnaire about what they have done, so we have that information. Be- cause we are interested in operational procedures, we do not restrict the divers much to not make it too artificial. When we are doing a purely scientific experiment, we may put more re- strictions on the divers than normal. otherwise, they do what they normally do, but let us know what that is. The pre-dive exercise can be quite variable. J. Ross: mild decompression illness is much more common than serious life-threatening decompression illness. are you perhaps skewing your table to one event, mild decompression illness and not necessarily to the other severe decompression illness? molvaER: DisCUssion ? 103 are not in our tissues. if it works in the gut, why would you want to put it anywhere else? Biochemists suggested extracting the isolated enzyme and putting it anywhere in the body. But it is a foreign, bacterial protein that will produce an inflamma- tory response. You can do liquid breathing now if you do not mind never being able to come out of having your lungs filled with perfluorocarbons, but you will not have to worry about bending. C. gutvik: How do you diagnose DCs in your animal models? S. Kayar: When you are dealing with a 60-70% hit rate, it is usually clear. it is a severe hit, the animals obviously have dif- ficulty walking. There are also dubious ?pig-headed? cases where the pig is fine, but just does not want to walk on the treadmill. it is on the bottom of the chamber and is turned on and off at a comfortable walking pace at 5-minute intervals. R. Richardson: is there a comment on the differences between wet and dry dives in the chamber? S. gaustad: immersion prior to the dive adapted rats for swim- ming because immersion effects last for a couple of days. From the dry dives we got lots of bubbles. There is also a diuretic ef- fect from immersion resulting in increased urine secretion. R. Richardson: true, there is a lot of evidence that water im- mersion causes diuresis. D. Doolette: The dry versus wet dives you showed were two very different dives in profile (depth and time) and dry versus wet, so the effect might not be as large as you imagine. in one very large meta-analysis of dry versus wet diving, they really wanted to be able to detect a large difference in decompression sickness in humans between the two, but physiologically, the effect was not as large as imagined. R. Richardson: it may turn out that the no specific role may be involved in the wet-dry business. e. Thorsen: You demonstrated the interaction between oxygen radicals and nitric oxide production. Diving is associated with hyperoxia. How do different depths, with increased partial pres- sures of oxygen, affect the protective effect of nitric oxide? R. Richardson: i am not sure about the actual effects, but in terms of the physiology of oxidative stress there are some para- doxes of free radical generation. Certainly, hyperoxia elevates oxidative stress and you are likely to have greater free radical stress. on the paradoxical side of that, if you are in hypoxia, you tend to have greater oxidative stress. The answer is hyperoxia will give you a greater chance for oxidative stress and therefore a greater chance for no bioavailability. The protective role of antioxidants would be offensive in that situation. m. gennser: This spring we carried out a trial looking at exer- cise 24 hours before diving, compared to 2 hours before diving. When you have three conditions and you expect to see the low- est bubble scores in one of two conditions, it occurs in the third one. We found less bubbles in the control situation where they had not conducted any exercise prior to the trial. our methods D. Doolette: That is quite possible because we do not distin- guish between mild and serious decompression sickness and try to come up with a model for that. We try to design our dives to avoid serious decompression sickness for ethical reasons. it could possibly be the case that we are skewing our procedures, i do not know for sure but do not think so. We have come across that problem when we recently tested new no-stop limits. our estimates of decompression sickness were about 2% incidence. We got less than that, only 1%, but they were all serious cases. That is a limitation of that technique. Central nervous system decompression sickness is not part of our model. O. eftedal: Ed Thalmann once commented that all symptoms, no matter how minor or trivial, must be considered because they impinge on decompression stress, which means that the definition of decompression sickness in your technique is very different from an operational sense. D. Doolette: Ed said that in respect to testing. in our modeling we do count minor symptoms as well and assign a value of 1/10 of a decompression sickness incident. all minor symptoms are noted and incorporated into the model, although there is some debate of late as to whether we should be doing that. S. angelini: in the diagram of activity versus release, there was zero activity that was showing an order of magnitude higher release than in some cases with activity. S. Kayar: That is because pigs, like humans, have Methanobrevi- bacter smithii in their natural intestinal flora. They also have mi- croorganisms, as do we, that produce hydrogen (H 2 ). M. smithii turns H 2 into methane (CH 4 ). S. angelini: How can you quantify the effectiveness of the bac- teria with all the background noise? S. Kayar: You first start up the dive profile with control animals with a 60% hit rate. How do you decide on that initial profile that produces enough hits? You do not want 100% hits because you would be so far over the limit that you could remove tons of gas, and it would still not help. You do not want a 10% hit rate in your control group because then you have nowhere to go down to. The dive profile choice is not a sensible approach, it is a guess with a convenient depth, a convenient dive time, and decompression rate to see how many animals bend. The dive profile can then be adjusted up or down. if you are at the 50% bends rate, you are on the steep part of the curve, so a bit deeper profile should produce more bends but the pigs were not bend- ing. We were continuously measuring the animals? methane production rate. The animals? native methanogens were battling us by doing biochemical decompression before we put more in. a. hope: is there any chance that it is possible to consume the hydrogen in other tissues than the intestine? Can you also com- ment on the ultimate solution to the decompression problem, liquid breathing? S. Kayar: Hydrogen consumption is limited to the intestines because there are no other areas in the mammalian body where you can put live bacteria that will not trigger an inflammatory response. Even though they are not pathogenic, these bacteria 104 ? tHE FUtURE oF Diving were different from the teams of Duji? and Blatteau. We no- ticed that our subjects, although they had some hyperbaric ex- perience, were not divers, whereas the other groups tested were divers. This is something we will test further. is there anything in divers that would affect the no metabolism? R. Richardson: There are a number of things that can occur, but i would immediately ask what the exact exercise paradigm was? m. gennser: it was the exact same intensity and duration as Duji??s trial. R. Richardson: a lipid load or high-fat meal can increase the no bioavailability. You can reduce FmD (flow-mediated dila- tion) by 50% by having a mcDonald?s meal. in your case, it does not sound like you were measuring FmD per se. Clearly, we are measuring FmD to look at no, and if the bubbles are involved you have to be very cautious while investigating endo- thelial function and relating it to diving. People must be fasted for 12 hours, no caffeine, no smoking, to make sure we are mea- suring endothelial function without being skewed by something they did previously. We typically require 24 hours with no exer- cise for these FmD studies. C. Balestra: i agree totally with your approach. We had more or less the same results looking at antioxidants and FmD and with the same precautions such as no smoking, but we gave them 30 g of Belgian chocolates and could not find a reduc- tion of FmD after the dive. it is also an antioxidant system, a cathecin. microvascularization can be increased with exercise. venous gas bubbles increase is dependent on delta Po 2 and not on the steady state. We are giving oxygen like exercise, which just increases blood flow into the tissues and in turn increases oxygen, so delta Po 2 will be positive. R. Richardson: That is interesting, you are using hyperoxia to drive oxygen into the cell by increasing gradient, as opposed to dropping the gradient, inside the cell. D. Doolette: How universal is your calibration data to test all sorts of different dives? are we going to have to do 1,000 more calibration dives with DCs if the dives we are interested in looking at are sufficiently different from the calibration data? O. eftedal: These data are for air bounce dives. if you want to use these for other types of dives, you have to gather new data, which will obviously create a lot of decompression sickness. J.e. Jacobsen: This is very interesting. are these dives you make reference to going down, staying on the bottom, and then checking the decompression schedule coming up? Do you take into consideration whether there is any decompression stress during the bottom phase? O. eftedal: These were just square profiles. We used bottom times just to confirm that they stayed exactly on the profile that was prescribed. if they did not, we did not use the data. J.e. Jacobsen: Did you measure bubble score at the start of decompression and follow it up? O. eftedal: no, they were actually not decompression dives. The dives were performed in a lake, and we started monitoring once the divers were out of the water. The bubble numbers you see are the maximum numbers. We know from experience that the maximum numbers are usually reached witihin 40 minutes. J.e. Jacobsen: Can we transfer this information into saturation diving? O. eftedal: That is the problem with the lack of basic informa- tion for saturation diving. The bubble-decompression sickness relationship for saturation diving has not been well described. J. pontier: Have you done experiments with animal models that are applicable? O. eftedal: it is difficult to use data from animal experiments in this setting for new operational procedures. What we can do with animals is compare different procedures. With our pig model we have a more sophisticated method for detecting bub- bles and can actually count single bubbles. This is a better way of doing it than crude grading of bubbles. a. Fahlman: are you using inbred rats or regular rats? a. m?llerl?kken: We are using regular sprague Dawley rats. a. Fahlman: With inbred rats, you can use genetic variation and may be able to see environmental variation, because you can get a lot of variation in the investigative environment. a. m?llerl?kken: We have used two different strains that are very different when it comes to bubble production or reactions to decompression. a. Brubakk: The next step is to try to use rats that are geneti- cally modified, most specifically by exercise where you can see they get different responses to exercise models to look at genetic mechanisms that might play a role. We believe that particular adaptation will protect them from decompression sickness and bubble formation as far as our present data indicate. D. Doolette: to what extent is individual variability, which you are talking about here with genetics, and day-to-day variability important? We have divers who do identical profiles day after day and then get bent maybe 1 out of 8 times. a. Brubakk: it seems clear that there is variation, and it could be that you need a decompression table with a lower limit. it is obvious that individual variation plays a significant role. We think that epigenetics is a very interesting term, which is a ge- netic response of your genes to the environment that has genetic effects. But, we do not know the answer. C. Balestra: We did a study of 5 divers who accepted to do 2 dives per day for 5 days and being scanned for bubbles every day. There was high variability in divers where we found bubbles present. The divers who had no bubbles continued to have no bubbles. We do not have sufficient data for a clear answer either. adaptation may be one indicator. a. Fahlman: We looked at the dive response in rats, and there is a huge variability difference between strains, but not between days. You have to be able to reduce the variability to get a han- dle on decompression sickness mechanisms. D. Doolette: You found very little day-to-day variability of sus- ceptibility to decompression sickness with inbred strains of rats? molvaER: DisCUssion ? 105 a. Fahlman: no, not for decompression sickness, we were look- ing at the dive response, distinguishing between good and bad divers and how long they could hold their breath. You reduce the genetic variability of phenotypes and see little difference within strains or between days. D. Doolette: Right, it is interesting that everyone thinks they are immune to decompression sickness until they get bent. l. Skjerven: one observation is of air divers working every day, 1 to 2 dives, same decompression and same conditions, yet on Friday they get the bends. a. Brubakk: many have data like that and yet other data show that repetitive diving reduces the risk, but we still do not know. O. molvaer: Can ultrasound somehow trigger decompression sickness? S. m?s?y: a known fact is that ultrasound can cause cavitation. For imaging, you have to be able to not cause cavitation. as far as causing decompression sickness, i would say no, not during imaging. O. molvaer: Can it be deleterious or damaging in pregnancies using ultrasonic diagnostics? S. m?s?y: This is a highly controversial issue, but i would say no. i have seen studies done on mice, and although not compa- rable to humans, much higher frequencies of ultrasound were used, which causes energy absorption in the tissues. a. hope: Could it be possible with this new technique to detect or measure stationary bubbles or is it only Doppler? S. m?s?y: Yes, they can move very slowly. We saw on the B- mode images of pigs? stationary bubbles. For large bubbles, i suggest going to contrast agents which add a few microns. D. Knaus: Can you elaborate on what the actual signal is that you are analyzing and how it is processed? S. m?s?y: The signal is not the frequency difference or the fre- quency mixing that you use, because you would have to use a very long pulse. We use very short pulses and are actually using the fundamental signal from high frequency itself. This signal would be different, it is compressed and expanded, resonance based. The low resonance of the bubble you can compress and expand. J. Buckey: You mentioned that the nitric oxide produced vaso- dilatation and reduced pre-existing micronuclei. What do you propose as mechanism for that? J. Blatteau: i do not know exactly. it is possible that vasodilata- tion accelerates the attenuation of nitric oxide. C. Balestra: another possible mechanism is related to caveolae changing due to production of no. e. Thorsen: is it possible that you are studying the effect of heat exposure and vibration and you have no control over either of those? J. Blatteau: of course. a. arnfinsen: i assume you said the stroke volume might be re- duced because of dehydration so the gas uptake was decreased. Could it not be a similar problem that the offgasing during de- compression would be reduced if the stroke volume was back to normal during decompression? J. Blatteau: in fact in our study we did not measure stroke vol- ume, it is just a hypothesis. g. perdrizet: Did you measure the core body temperature in your model? J. Blatteau: We did not measure the core temperature. g. perdrizet: What cell type did you measure the HsP70 stress protein in, white blood cells? J. Blatteau: i do not remember exactly. C. Balestra: The plasma came from white blood cells through a ?sandwich method.? a. Brubakk: Can i make a suggestion? i do not think it has anything to do with gas uptake and gas elimination at all. i am more and more convinced that supersaturation is a red her- ring when it comes to bubble formation. Data to support that is from the studies we have done, where we can eliminate gas bubbles totally at the same level of supersaturation. This does not fit with the prevailing model that supersaturation creates a significant problem. With our animal experiments the bubble formation was the same but the effect of the bubbles was re- duced, so this is still not solved at all. O. molvaer: Did you examine the spinal cord in the paralyzed rats? m. havnes: no, we did not. e. Thorsen: Would it be possible to monitor for gas bubbles on the arterial side? m. havnes: i do not know. We can actually see the bubbles, but in the brain it is very difficult. a. Brubakk: From the studies we have done with exercise and nitric oxide, we seem to be able to totally eliminate gas bubble formation in these animals. The question now becomes is that a safe procedure is when you cannot see any bubbles anymore? We were a little surprised that we did not see anything, but the investigation still goes on, and we are learning a lot about tech- nique. it may be that there are a lot more bubbles in our data than we know of. our studies also show we can have endothelial dysfunction even if we cannot measure any gas bubbles using ultrasound. Endothelial effects on the venous side also have an effect on the arterial side even if bubbles are not transported over. J. Ross: You are using manganese contrast but because you have not seen it does that mean that there is no endothelial damage? m. havnes: The uptake of the manganese in the brain is there. manganese behaves like calcium, it flows into the cells. J. Ross: it is not encapsulated, it will not interact with another process and make it worse, because it is a transition metal. J. pontier: Do you recall the chronology of signs of decompres- sion sickness? m. havnes: one of the rats was paralyzed in the hind leg for a day, and then it cleared up. i have not done it systematically, but i watch them and try to get them to run around and see if 106 ? tHE FUtURE oF Diving they behave strangely. g. perdrizet: You have a very sophisticated model, and i do not know much about decompression sickness in the brains of rats. Can you use light microscopy or postmortem studies that give you a positive control here? m. havnes: i am not sure we would get feedback, because the rats were so damaged, but earlier studies were mostly on the spinal cord, and i have not seen many studies on the brain. a. Brubakk: We had hoped we had a model that could give us a positive result, that was the intention at least, but it turned out much less than expected. We did a study many years ago looking at spinal cords in diseased divers who had died from dif- ferent reasons. a number of those divers (10 professional and 10 amateur divers) had experienced decompression sickness. With the most sophisticated techniques we had available 10 years ago, we could not find anything in the spinal cord. We do not have enough data, in part because we do not know exactly what we are looking for. J.e. Jacobsen: somehow we need to find a better link to the operational world besides just basic research. in the operational world, we are focusing very much on saturation diving where the concern is long-term effects and a study of operational expo- sure data. We should also take more into consideration the bot- tom phase of the dives. There could well be some decompression stress accumulating before the final decompression takes place. many years ago, there was a project looking for critical exposure factors and we actually did find some too. We need the link between the profile, or elements of the profile, and probability of DCs as an outcome of the total saturation period. today we are easily capable of producing time/depth profiles and study them. Within a short time, we will have even more informa- tion available when the automatic monitoring systems come on line aboard the vessels. But we have not yet developed a tool to describe the character of the exposure. This is important for the doctor who sits in front of the diver and has the health record on the left side and the exposure information on the right side. However, the exposure information is not summarized in any way and cannot currently present a clear picture to the medi- cal doctor, who therefore cannot link the two in practice. The project also had an important strategy: first, it was to connect a lot of exposure data and search for critical exposure factors, then test this exposure data out on animals. That clear link has not been made yet. an example of an accumulative factor on the bottom phase from the analysis we have done was when the relative pressure change between the shallowest and deepest stores in saturation increased by 1%, the probability of produc- ing DCs as an outcome of total exposure increased by 5%. That was a qualitative measure of DCs. There were other examples, but we need to look for more factors. all of the presentations at this symposium have been interesting, and we should not forget that there is an industry out there that needs practical tools to monitor the long-term effects and that should be taken into consideration. a. Brubakk: i agree that it is important to try to find practical applications of the research here. it is also important to have an understanding of what we are actually looking for and what we are doing. one of the reasons we are doing the kind of research presented at this meeting is trying to understand some of the mechanisms that can help us explain and give you the tools that you need. We have had a number of presentations in many different fields that show there are various methods that are nor- mally not considered part of decompression research but are quite useful. Emerging techniques such as the new ultrasound equipment clearly could, if we have the right methods and un- derstanding of the theory, provide us with a lot of information to develop better procedures. J. Ross: Understanding fundamental physiology of responsive people put under pressure and then decompressed is very im- portant as you are considering the long-term health effects of an industry. in saturation diving, you use pressure chambers to commute to work. What you actually do at work may be more important than any change in pressure. This industrial work force is working at pressure and is exposed to accidents, illness, industrial contamination and petrochemicals in particu- lar, hand-arm vibration syndrome, cold-induced events, excess noise, etc. The responses to those exposures may also be funda- mental physiological changes induced by the environment. to measure long-term health effects, you need to take all factors into account, not just pressure changes, but exposure in the ho- listic sense. J.e. Jacobsen: Thank you alf, we agree of course, it is just that when we do the basic research, we still need to remember the link as to why we do it and follow up on the operational side as well as pointed out by Dr. John Ross. B. gardiner: i would like to share an anecdote on John scott Haldane. last week on the news in the U.k., a ten-year-old boy wanted to cycle to school, but the school management was so concerned about his health and safety and protection that they demanded his mother drive behind him in the car. The thought of John scott Haldane turning his child into experimentation must have shocked people like that. i do know however that John scott Haldane was a very kind man. i visited the mines Rescue Center in the north of England and came across the canary cage that he used to investigate mine disasters. many people thought it was the explosions that killed the miners, but Haldane found it was the toxic gases that asphyxiated most of those people. He took his canary cage with him down into the mines when he did experiments, and when the bird fell off its perch, he would put his mask on. The reason that the canary cage is important to see John scott Haldane, is that it was made of a complete box with glass sides and a front door that hung down. When the canary fell off his perch, Haldane would close the o-ring sealed cage door. on top of the cage?s carrying handle was a small bottle of oxygen, which he would open, the bird lived and everybody was happy. molvaER: DisCUssion ? 107 D. Doolette: There have been several talks on using vgE as a measure of decompression stress and how that could work. There are at least two ways why vgE might be a measure of decompression stress. some people believe the gas emboli in cells may cause symptoms, others believe it may just be a reflec- tion of what is going on extravascularly. There was also a talk that seemed to suggest that the use of nitric oxide and exercise affects micronuclei inside the vasculature, then using vgE as a measure of stress after that. are we measuring the right thing anymore with extravascular vgE effects by not looking at DCs as an endpoint? We are looking at bubbles or changing how they behave. Can we still say without some validation that they are still a measure of decompression stress? a. Brubakk: We might be measuring a totally wrong thing. The reason we at least are measuring the wrong thing is that it is the only thing we can measure. i would love to have a method for instance to know changes that happen in some tis- sues and get some data on that. The only things i really believe in, and fully agree with Haldane, is that we need data from something we can measure. Decompression sickness is a very slippery beast that we cannot really define. my group is going in a totally different direction. We use something we can measure and then use occam?s Razor principle and try to explain it with the simplest mechanism we can think of and see where that gets us. maybe over time or with better methods we will have to acknowledge that we were totally wrong, but can at least have some fun trying to get it right. a. m?llerl?kken: it is important for the future to pinpoint the exact research questions. as far as i know, most questions being asked at different laboratories do not give us straight yes or no answers, but they open up all different kinds of opportunities that lead to disagreements on the importance of bubbles, stress, or other markers. The ability to state more concise questions on the decompression syndrome is a major challenge for all of us. C. Balestra: my answer to this is collaboration, which is the only thing we can do together to find the right answers. g. perdrizet: You cannot manage what you cannot measure. lyme disease may be an analogy. in the literature, there is continuous argument and controversy over that disease because we have no way to measure it. The decompression theory will remain a theory until you find an objective way to measure it. R. moon: alf Brubakk made the point that there was no rela- tionship between bubbles and supersaturation, but surely did not mean that bubbles are caused by something other than supersaturation. a. Brubakk: You need supersaturation, but that is not enough. my point was that you can have tremendous supersaturation without bubble formation. R. moon: Probably a caution, but you suggested quite correctly, i think, that venous gas embolism is a more objective measure of decompression stress than symptoms. However, i think you intended to calibrate your decompression stress by using symp- toms. one of the presentations had some nice graphs looking at the correlation between venous gas embolism and decompres- sion sickness. after all, is it not the sickness we are trying to prevent? We know that bubbles are almost ubiquitous among divers, particularly saturation divers, and yet very few divers ac- tually get bent. a. Brubakk: The point here, as was said earlier, is that what we are looking for is a bubble-free decompression. We want to use gas bubbles in a way that can produce a procedure that has as few bubbles as possible. That is our current thinking. it is clear that there is a statistical relationship between many vascular gas bubbles and the increased risk of decompression sickness. The argument i made was that using decompression sickness as end- point is extremely difficult in developing models or ideas about what is really going on because there is no clear definition of the clinical symptoms. i agree with what Thalmann suggested that any symptom after decompression could be related to decom- pression sickness and use that as endpoint. R. moon: i would agree with that, but it is not unique in medi- cine. There are many other clinical syndromes for which there are no objective measures, in fact, perhaps even most. a. Brubakk: true, and that is why we are in trouble. good morning ladies and gentlemen. i am probably one of the few audience members who does not dive and therefore does not know much about diving, which is probably the reason i am only going to talk for about five minutes. as you know, we are building a new hospital. in that respect, i am asked to share how we see the future for diving physiology and diving medicine in our faculty. We have plans for our labs, and we believe that there is no future for science without scien- tists, so how are we going to plan for that? i do not think we have to spend much time describing our enthusiasm for translational research, a necessity for the future of any physiological science and probably also true for diving. How are we going to breed young scientists? There are several methods for produc- ing offspring and cloning is one of them. We have had reasonably successful stories for the last fifteen years, and some of these success factors give room for reflection. alf Brubakk and others here have had a very solid basis in medicine and physiology with grounding in clinical and occupational consulting. i was struck by ?eljko Duji??s pre- sentation that showed that almost all researchers involved in science at the University of split were mDs with PhDs. We need to have many mDs on board in order to be successful. They need to be involved in clinical and occupational consulting in order to experience the problem base for the experimental work that is going to be the transla- tional part. several speakers have mentioned that the translational approach combining experi- mental and clinical work is important to figure out the cellular and molecular mecha- nisms but also for the evaluation of the outcome of physiological experiments. in my opinion, it is also important to have close industry and national collaborations where the actual professional diving takes place. That is the cloning part that we believe has contributed to our success, and we want to copy that model for the future. no one knows how and what the new discoveries are going to be, but for certain you need fresh minds and much enthusiasm to accomplish that. We are fortunate to have many young students who we engage quite early in medical school to work on their student theses. it is important to involve people in research from the very start. it is much easier to engage 20-year olds in diving research and environmental physiology than in geriatric research. in that regard, diving science is very fortunate. spawning involves spreading out the enthusiasm by seeding all over the place, but you also need to do some cultivation, the money-consuming and difficult part. again, solid roots are required in clinical and occupational consulting. in the cultivation pro- cess of new researchers, competition and selection are important, and career grants must be available for young, accomplished postdocs. The spawning of trying to reach everybody stops at some point. The funding then needs to be based on the people who Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, NO-7491, Trondheim, Norway. Breeding Diving scientists: Cloning, spawning, or Cultivation? ?yvind Ellingsen 110 ? tHE FUtURE oF Diving are contributing with large scientific output. stable fertilization is required in order for this to happen, which means a good basis of industry and government funding. How are we going to do that, through direct funding by the industry to the vari- ous research groups? if you really want to do a good selection of young researchers, a large chunk of the funding should be channeled through established funding agencies for the peer- review part and to ensure that the resources are spent on the best candidates. INTrOducTION the national Centre for Hyperbaric and Diving medicine was established in January 2008 at the request of the central authorities of norway in stortings-melding 12, 2005-2006, on health, environment, and safety in the petroleum industry, Chapter 7 - Diving (stortingsmelding, 2005-2006). This document acknowl- edges the consensus from the god?ysund ii meeting (Hope and Risberg, 2005) that ?there is evidence that changes in lung function, Cns (central nervous system), bone, and cochleo-vestibular system can be demonstrated in some occupational divers. The magnitude of these changes is highly variable and has the potential to influence divers? quality of life. The knowledge about the precise mechanism is still limited and calls for further research on preventive measures, including health surveillance.? ESTablIShmENT Of ThE NaTIONal cENTrE The norwegian labour Union highlighted the need for a national Centre for div- ing medicine, independent of industry, with a high level of clinical expertise emphasiz- ing the principles of evidence-based medicine. The Centre is located at the Department of occupational medicine, Haukeland University Hospital. This department is responsible for elective and acute hyperbaric therapy on a regional and national level and medical assessment of occupational disease in all norwegian divers. since 2000, about 200 out of a total of 375 former north sea divers and about 40 inshore divers have been examined with collaboration from several hospital departments (e.g., divers with decompression sickness are inpatients at the Department of neurology). NaTIONal cENTrE gOalS in accordance with the official definition of norwegian national Centres, our Cen- tre will promote research, education, and public information in the field of hyperbaric and diving medicine (HoD, 2003). The Centre is expected to advise central authorities on developing a national plan for hyperbaric medicine and in all matters related to div- ers? health, environment, and safety. The Centre will also advise health professionals, develop national standards and a registry for decompression sickness and occupational diseases in divers. Health promotion is the main goal in this endeavor. Helse Bergen Haukeland University Hospital, Jonas Liesvei 65, N-5021, Bergen, Norway. national Centre for Hyperbaric and Diving medicine Marit Gr?nning 112 ? tHE FUtURE oF Diving rESEarch The Centre will focus on epidemiological and multidisci- plinary clinical research and also conduct basic and translational research. We are involved in a multicenter study of hyperbaric treatment of radiation-induced cystitis. in another ongoing project we are studying possible Cns effects of hyperoxia in patients undergoing hyperbaric therapy. a prospective longitu- dinal study of occupational divers and long-term health effects in retired divers is being conducted. a follow-up study on divers referred for decompression sickness is being conducted. studies of the effects of diving and hyperoxia on pulmonary mechanical and gas exchange functions are being continued with emphasis on nitric oxide as mediator. to further multicenter studies and to accomplish other goals, we are establishing closer contact and cooperation with other national and international centers. lITEraTurE cITEd HoD i-19/2003 - om h?yspesialiserte tjenester ? landsfunksjoner og nasjonale medisinske kompetansesentra. Hope, a., and J. Risberg, eds. 2005. long-term health effects of diving. The god?ysund 1993 consensus revisited. Bergen: BoDoni. 302 pp. stortingsmelding 12, 2005?2006. Health, environment and safety in the petro- leum industry, pp 65?69. INTrOducTION the start of the program known today as the ntnU Baromedical and Envi-ronmental Physiology group (BaREn) was in 1970, when as a young medi-cal doctor just finishing my internship, i was offered a grant to do research in trondheim. at that time there was no medical Faculty in trondheim, only in Bergen and oslo, but the county wanted to support research as part of an effort to get recog- nition as the next medical faculty in norway. trondheim had the only technical university in the country, and my job was to see if there was a possibility to establish research in the area of medical technology. together with a young engineer, Rune aaslid, we developed a computer model (JEnnY - named after the first patient) of the human cardiovascular system that we wanted to use for clinical work. This model consisted of over 50 nonlinear equations, but no digital com- puter existed at the time able to handle that. so Rune built an analogue computer to do this. We had hoped that this model could be used clinically, but we were probably way ahead of our time! Based on this work, we saw the need early on for better methods to evaluate vascu- lar function. We therefore decided to develop an ultrasonic system for measuring blood- flow velocity in the aorta and large vessels in the thorax (PEDoF). This was the start of a considerable and long-standing development of ultrasonic equipment for cardiovas- cular measurements that is now part of the instruments being sold as gE vingmed. EvOluTION When we first got interested in diving research, we saw the potential of ultrasonic equipment for detecting gas-phase formation after decompression. Bubble detection us- ing ultrasound had been around for a considerable time but, apart from a few enthusi- asts, the establishment saw little value in this. ?Bubbles are not decompression sickness? we were told by the experts. However, as we could demonstrate that gas bubbles could be detected in the carotid artery of nearly all divers following a deep experimental dive (Brubakk et al., 1986), we realized that these methods had considerable value. Financed by mobil oil, a program was started in trondheim in 1985 to look at various aspects of the underwater work environment, of which decompression was just one aspect. The program looked at decompression, bacteriology, ergonomics, and toxi- cology, trying to understand how this environment could influence man?s health. later, BaREn was established as a group within the Department of Physiology and Biomedical Engineering, and then in the Department of Circulation and medical imag- ing at the medical Faculty. as can be seen in the appendix, the group has over the years Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, N-7491, Trondheim, Norway. Baromedical and Environmental Physiology group (BaREn) Alf O. Brubakk 114 ? tHE FUtURE oF Diving published a significant number of papers. BaREn?s work has led to over 25 master?s theses and 5 PhD dissertations in the field of diving and decompression physiology. The main aim of the group is to train scientists in the field of underwater medicine and physiology. in more general terms, the program wants to provide students with an understanding of how the environment influences human physiology. The approach is multidisciplinary and focused on understanding the basic mechanisms involved. dEcOmprESSION rESEarch During later years, the group focused on decompression and in particular on how the endothelium is involved. This has led us into studies of endothelial dysfunction in general and how it can be influenced by environmental factors like exer- cise. Through collaboration with scientists interested in exercise physiology we could show that gas-bubble formation can be sig- nificantly influenced by exercise. originally, we predicted that exercise would increase gas elimination, thus reducing the risk of DCs. The results obtained experimentally were quite sur- prising?gas elimination was not increased, but a single bout of high-intensity exercise 24 hours before a dive that killed all our control animals, prevented all vascular gas bubble forma- tion (Wisl?ff and Brubakk, 2001). This result, originally first shown in rats, has since then been demonstrated in man (Duji? et al., 2004). The mechanism is related to the production of no, shown both in rats and pigs as well as in man (Wisloff et al., 2003; Brubakk and Wisl?ff, 2005). This led us to focus on vascular gas-bubble formation as the most important factor for evaluating the risk of neurological DCs. By focusing on the endothelium, we realized that endothe- lial dysfunction was an important factor involved in decompres- sion and that hyperoxia and vascular gas bubbles significantly in- fluenced the outcome of a dive. We also concluded that stress was an important factor in diving and that hyperoxia and gas-bubble formation are examples of a stress response. stress in the right dose and at the right time has a significant health-promoting ef- fect, thus showing that diving elicits a general response to stress. a number of decompression models exist. our effort, called Copernicus, is a model based on physiological principles that try to take individual factors of the diver and his environment into account. in a way, the circle is closed because BaREn is a result of research started over 30 years ago as JEnnY at the start of my scientific career. vISION fOr ThE fuTurE Diving research is a very small field reflected by the fact that there are probably less than 50 full-time scientists engaged in its study around the world. University departments with a diving research focus are even rarer. it is our belief that this field can only survive and grow if collaboration with other fields of physiology and medicine can be established. We are experiencing a world where sea levels are rising, needs for new energy sources independent of fossil fuels are required, and fish farming will have increasing importance. a man-in-the-sea capability is needed as will a better understand- ing of how to live and work under the ocean. i personally am about to retire in a couple of years. Consid- erable effort is under way to ensure that our work can continue. We hope that trondheim will also be a Centre of Excellence in diving research in the future. When planning for a month, sow rice, when planning for a year, plant trees, when planning for a decade, train and educate men. ?old Chinese proverb lITEraTurE cITEd Brubakk, a .o., o. Eftedal, and U. Wisl?ff. 2005. Endothelium and diving. in aird, W.C., ed. The endothelium: A comprehensive reference. Cambridge University Press. Brubakk, a. o., R. Peterson, a. grip, and B. Holand. 1986. gas bubbles in the circulation of divers after ascending excursions from 300 to 250 msw. Journal of Applied Physiology 60:45?51. Duji?, ?., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n. m. Petri, U. Wisloff, and a. o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555(3):637?642. Wisl?ff, U., and a. o. Brubakk. 2001. aerobic endurance training reduces bub- ble formation and increases survival in rats exposed to hyperbaric pressure. Journal of Physiology 537(2):607?611. Wisl?ff, U., R. s. Richardson, and a. o. Brubakk. 2003. nos inhibition in- creases bubble formation and reduces survival in sedentary but not exer- cised rats. Journal of Physiology 546:577?582. appENdIx PublicAtions 1999?2009 2009 glavas, D., D. Bakovi?, a. obad, i. Palada, t. Breskovi?, Z. valic, a. o. Bru- bakk, and ?. Duji?. 2009. Effects of tetrahydrobiopterin on venous bubble grade and acute diving-induced changes in cardiovascular function. Clini- cal Physiology and Functional Imaging 29(2):100?107. gutvik, C. R., and a. o. Brubakk. 2009. a dynamic two-phase model for vas- cular bubble formation during decompression in divers. IEEE Transactions in Biomedical Engineering 56:884?889. 2008 Bakovi?, D., D. glavas, i. Palada, t. Breskovi?, D. Fabijanic, a. obad, Z. valic, a. o. Brubakk, and ?. Duji?. 2008. High-grade bubbles in left and right heart in an asymptomatic diver at rest after surfacing. Aviation, Space, and Environmental Medicine 79:626?628. Duji?, ?., Z. valic, and a. o. Brubakk. 2008. Beneficial role of exercise on scuba diving. Exercise and Sport Sciences Reviews 36:38?42. glavas, D., a markotic, Z. valic, n. kova?ic, i. Palada, R. martinic, t. Breskovi?, D. Bakovi?, a. o. Brubakk, and ?. Duji?. 2008. Expression of endothelial selectin ligands on human leukocytes following dives. Experi- mental Biology and Medicine 233(9):1181?1188. Epub 2008 Jun 5. medby, C. a. Bye, U. Wisl?ff, and a. o. Brubakk. 2008. Heat shock increases survival in rats exposed to hyperbaric pressure. Diving and Hyperbaric Medicine 38(4):189?193. BRUBakk: BaREn gRoUP ? 115 Exercise during a 3-min decompression stop reduces postdive venous gas bubbles. Medicine and Science in Sports and Exercise 37:1319?1323. Hjelde, a., o. a. gederaas, H. E. krokan, and a. o. Brubakk. 2005. lack of effect of hyperoxia on photodynamic therapy in three different cell lines. Medical Science Monitor 11:351?356. m?llerl?kken, a., a. o. Brubakk, and U. Wisl?ff. 2005. The effect of hyperbaric oxygen on aerobic capacity in rat. European Journal of Underwater and Hy- perbaric Medicine 6:5?9. valic, Z., D. Duplan?i?, D. Bakovi?, v. ivan?ev, D. Eterovi?, U. Wisl?ff, a. o. Brubakk, and ?. Duji?. 2005. Diving-induced venous gas emboli do not increase pulmonary artery pressure. International Journal of Sports Medi- cine 26:626?631. Westin, a. a., J. asvall, g. idrovo, P. Denoble, and a. o. Brubakk. 2005. Div- ing behaviour and decompression sickness among galapagos underwater harvesters. Journal of Undersea and Hyperbaric Medicine 32:175?184. 2004 Brubakk, a. o. 2004. Hyperbaric oxygen therapy: oxygen and bubbles. Journal of Undersea and Hyperbaric Medicine 31(1):73?79. Duji?, ?., D. Duplan?i?, i. marinovi?-terzi?, D. Bakovi?, v. ivan?ev, Z. valic, D. Eterovi?, n.m. Petri, U. Wisloff, and a.o. Brubakk. 2004. aerobic exercise before diving reduces venous gas bubble formation in humans. Journal of Physiology 555(3):637?642. Eftedal, o. s., s. lydersen, g. Helde, l. White, a. o. Brubakk, and l. J. stovn- er. 2004. a randomized, double blind study of the prophylactic effect of hyperbaric oxygen therapy on migraine. Cephalalgia 24(8):639?644. Wisloff, U., R. s. Richardson, and a. o. Brubakk. 2004. Exercise and nitric oxide prevent bubble formation: a novel approach to the prevention of de- compression sickness? Journal of Physiology 555:825?829. 2003 Brubakk, a. o., and t. s. neuman, eds. 2003. Bennett and Elliott?s The Physiol- ogy and Medicine of Diving. london: saunders, 5th edition. Brubakk, a. o., B. krossnes, a. Hjelde, s. koteng, s. J. m?rk, and H. ?rnha- gen. 2003. The effect of 100 kPa oxygen and recompression to 200 kPa on air on organ injury following decompression in the pig. European Journal of Underwater and Hyperbaric Medicine 4:5?12. Brubakk, a. o., B. Wienke, and s. koteng. 2003. Decompression profile and bubble formation after dives with surface decompression. Experimental support for a dual phase model of decompression. Journal of Undersea and Hyperbaric Medicine 30:181?193. m?rland, t., l. lang?, and a. o. Brubakk. 2003. inert gas exchange from ni- trogen bubbles in diffusion limited tissue: Relevance for the use of helium in the treatment of decompression sickness. European Journal of Underwa- ter and Hyperbaric Medicine 4:13?19. nishi, R. Y., a. o. Brubakk, and o. s. Eftedal. 2003. Bubble detection. in Ben- nett and Elliott?s The Physiology and Medicine of Diving. Brubakk, a .o. and t. s. neuman, eds. saunders: london. Pp. 501?529. nossum, v., a. Hjelde, k. Bergh, a.l. Ustad, and a. o. Brubakk. 2003. anti- C5a monoclonal antibodies and pulmonary polymorphonuclear leukocyte infiltration ? endothelial dysfunction by gas bubbles. European Journal of Applied Physiology 89:243?248. Wisloff, U., R. s. Richardson, and a. o. Brubakk. 2003. nos inhibition in- creases bubble formation and reduces survival in sedentary but not exer- cised rats. Journal of Physiology 546:577?582. 2002 Hjelde, a., g. Bolstad, and a. o. Brubakk. 2002. The effect of air bubbles on rabbit blood-brain barrier. Journal of Undersea and Hyperbaric Medicine 29:31?38. Hjelde, a., m. Hjelstuen, o. Haraldseth, J. D. martin, s. R. Thom, and a. o. Brubakk. 2002. Hyperbaric oxygen and neutrophil accumulation/tissue damage during permanent focal cerebral ischaemia in rats. European Jour- nal of Applied Physiology 86:401?405. Hjelde, a., v. nossum, m. steinsvik, J. i. Bagstevold, and a. o. Brubakk. 2002. 2007 Brubakk, a. o., o. s. Eftedal, and U. Wisl?ff. 2007. Endothelium and diving. in aird, W., ed. Endothelial Biomedicine. Cambridge University Press. pp. 497?505. Eftedal, o. s., H. tjelmeland, and a. o. Brubakk. 2007. validation of decom- pression procedures based on detection of venous gas bubbles: a Bayesian approach. Aviation, Space, and Environmental Medicine 78(2):94?99. Eftedal, o. s., s. lydersen, and a. o. Brubakk. 2007. The relationship between venous gas bubbles and adverse effects of decompression after air dives. Journal of Undersea and Hyperbaric Medicine 34:99?105. m?llerl?kken, a., C. gutvik, v. J. Berge, a. J?rgensen, a. l?set, and. a.o. Brubakk. 2007. Recompression during decompression and effects on bubble formation in the pigs. Aviation, Space, and Environmental Medicine 78(6):557?560. obad, a., i Palada, v. ivan?ev, Z. valic, D. Fabijanic, a. o.Brubakk, and ?. Duji?. 2007. sonographic detection of intrapulmonary shunting of venous gas bubbles during exercise after diving in a professional diver. Journal of Clinical Ultrasound 35(8):473?476. obad, a., i. Palada, Z. valic, v. ivan?ev, D. Bakovi?, U. Wisl?ff, a. o. Brubakk, and ?. Duji?. 2007. The effects of acute oral antioxidants on diving-in- duced alterations in human cardiovascular function. Journal of Physiology 578:859?870. obad, a., Z. valic, i. Palada, a. o. Brubakk, D. modun, and ?. Duji?. 2007. antioxidant pretreatment and reduced arterial endothelial dysfunction af- ter diving. Aviation, Space, and Environmental Medicine 78(12):1114?1120. 2006 Brubakk, a. o., and C. gutvik. 2006. optimal decompression from 90 msw. in lang, m.a. and n.E. smith (eds). Proceedings of the Advanced Scientific Diving Workshop. Washington, DC: smithsonian institution. Pp. 39?46. Brubakk, a. o., o. s. Eftedal, and U. Wisl?ff. 2006. Endothelium and div- ing. in aird, W. (ed). Handbook of the Endothelium. Cambridge University Press. Duji?, ?., a. obad, i. Palada, Z. valic, and a. o. Brubakk. 2006. a single open sea air dive increases pulmonary artery pressure and reduces right ventricu- lar function in professional divers. European Journal of Applied Physiology 97(4):478?485. Duji?, ?., Z. valic, and a. o. Brubakk. 2006. Exogenous nitric oxide and bubble formation in divers. Medicine and Science in Sports and Exercise 38(8):1432?1435. l?set, a., a. m?llerl?kken, v. Berge, U. Wisl?ff, and a. o. Brubakk. 2006. Post?dive bubble formation in rats: Effects of exercise 24 h ahead repeat- ed 30 min before the dive. Aviation, Space, and Environmental Medicine 77(9):905?908. m?llerl?kken, a., v. J. Berge, a. J?rgensen, U. Wisl?ff, and a. o. Brubakk. 2006. Effect of a short?acting no donor on bubble formation from a satu- ration dive in pigs. Journal of Applied Physiology 101(6):1541?1545. Wisl?ff, U., P. m. Haram, and a. o. Brubakk. 2006. Exercise and the endothe- lium. in aird, W., ed. Handbook of the Endothelium. Cambridge University Press. 2005 Berge, v. J., a. J?rgensen, a. l?set, U. Wisl?ff, and a. o. Brubakk. 2005. Exer- cise ending 30 min pre?dive has no effect on bubble formation in the rat. Aviation, Space, and Environmental Medicine 76(4):326?328. Brubakk, a. o. 2005. Decompression illness. What we do know, what we don?t know. in marroni, a., D. mathieu, and F. Wattel (eds). The ECHM Collec- tion. volume 1. pp 11?34. Brubakk, a. o. 2005. Which animal models are to be used to study decom- pression accidents? in: marroni, a., D. mathieu, and F. Wattel, eds. The ECHM Collection. volume 1. pp 201?211. Brubakk, a. o., D. Duplan?i?, Z. valic, i. Palada, a. obad, D. Bakovi?, U. Wisloff, and ?. Duji?. 2005. a single air dive reduces arterial endothelial function in man. Journal of Physiology 566:901?906. Duji?, ?., i. Palada, a. obad, D. Duplan?i?, D. Bakovi?, and Z. valic. 2005. 116 ? tHE FUtURE oF Diving Evaluation of cerebral gas retention and oedema formation in decom- pressed rats by using a simple gravimetric method. Scandinavian Journal of Clinical Laboratory Investigations 62:263?270. medby, C., H. R?e, and a. o. Brubakk. 2002. microdialysis in cisterna magna during cerebral air embolism in swine. Journal of Undersea and Hyperbaric Medicine 29:226?234. nossum, v., a. Hjelde, and a. o. Brubakk. 2002. small amounts of venous gas embolism cause delayed impairment of endothelial function and increase polymorphonuclear neutrophil infiltration. European Journal of Applied Physiology 86:209?214. 2001 Brubakk, a. o., and o. Eftedal. 2001. Comparison of three different ultrasonic methods for quantification of intravascular gas bubbles. Journal of Under- sea and Hyperbaric Medicine 28:131?136. Juvik, E., E. solig?rd, a. o. Brubakk, and a. m. Brubakk. 2001. arterial air embolism after venous air infusion in newborn pigs. Acta Pediatrica 90:786?782. medby, C., a. o. Brubakk, and H. E. myrvold. 2001. iatrogen gassemboli. Tidsskrift for Den norske legeforening 121:2604?2606. Wisl?ff, U., and a. o. Brubakk. 2001. aerobic endurance training reduces bub- ble formation and increases survival in rats exposed to hyperbaric pressure. Journal of Physiology 537:607?611. 2000 Brubakk, a. o. 2000. man in extreme environments. Aviation, Space, and Environmental Medicine 71(suppl):a126?130. Brubakk, a. o. 2000. surface oxygen is an acceptable definite treatment for DCi. Journal of the South Pacific Undersea Medicine Society 30:1955?1961. Brubakk, a. o. 2000. What is the optimal treatment ?dose? for decompression illness. Journal of the South Pacific Undersea Medicine Society 30:92?95. Brubakk, a. o. 2000.on-site recompression treatment is acceptable for DCi. Journal of the South Pacific Undersea Medicine Society 30:166?173. Brubakk, a. o., and o. s. Eftedal. 2000. Evaluation of reverse dive profiles. in lang, m. a., and C. E. lehner, eds. Proceedings of the Reverse Dive Profiles Workshop. Washington DC: smithsonian institution. pp. 111?121. Hjelde, a., and a. o. Brubakk.2000. variability in serum surface tension in man. Applied Cardiopulmonary Pathophysiology 9:9?12. Hjelde, a., s. koteng, o. s. Eftedal, and a. o. Brubakk. 2000. surface tension and bubble formation after decompression in the pig. Applied Cardiopul- monary Pathophysiology 9:47?52. nossum, v., a. Hjelde, k. Bergh, and a. o. Brubakk. 2000. lack of effect of anti-C5a monoclonal antibody on endothelial injury by gas bubbles in the rabbit after decompression. Journal of Undersea and Hyperbaric Medicine 27:27?35. 1999 Hjelde, a., a. o. Brubakk, k. Bergh, v. videm, and a.l. Ustad. 1999. Effect of anti-C5a antibody on blood-lung and blood-brain barrier in rabbits after decompression. Journal of Undersea and Hyperbaric Medicine 26:249?256. nossum, v., s. koteng, and a. o. Brubakk. 1999. Endothelial damage by bub- bles in the pulmonary artery of the pig. Journal of Undersea and Hyperbaric Medicine 26:1?8. the recent changes in diving habits and population characteristics increase the risk of decompression sickness among recreational divers. moreover, it is also expected that diving companies will increase their activity because oil companies are eager to increase offshore excavations. Those dives should be done with increased level of decompression safety, since in the last years, a lot of former commercial divers have been financed by large amends for long-term deleterious effects of inappropriate desaturation of inert gas during decompressions conducted according to currently used decompression tables. Because of the important and yet unexplained inter- and intra- individual variability of bubble formation and bubble sensitivity, there is a need for new approaches of decompression management and better diver education. Finally, no actual university course is given in the specific field of research applied to decompres- sion after diving or working under pressure. not even formal medical information is provided through European universities during the standard medical education curricu- lum. Therefore, a proposal for a FP7-people marie Curie initial training network has been submitted to the European Commission in september 2008. This itn proposal, entitled Physiopathology of decompression: Risk factors for the formation of intravascular bubbles during decompression (PHYPoDE), aims to: 1. develop a coherent and integrated research training program on decompression phenomena with specific attention to the formation of intravascular bubbles with their pathophysiological and clinical consequences, including endothelial function and decompression sickness. 2. develop an exchange of knowledge that will be structured by the initial training network activities, giving young researchers opportunities to share research tech- niques and resources, and jointly participate in courses, seminars, workshops, and congresses in order to benefit from the best international scientists? knowledge in this field, and get in touch with the scientific community from private and public institutions; and 3. widen career prospects of young researchers by embracing the whole chain of re- search applied to decompression: ? fundamental research for pathophysiological understanding of decompression; ? applied research in the industry; ? clinical research on decompression sickness; and ? ground experimentation with professional divers. The PHYPoDE network unites 9 teams, situated in France, Poland, Belgium, italy, Egypt, and south africa. among these, one partner is a nonprofit association, Francois Guerrero, UFR Sport et Education Physique, Universit? de bretagne Occidentale, Brest Cedex, F-29285, France. Jacek Kot, National Center for Hyperbaric Medicine, Medical University of Gdansk, Gdansk, Poland. Peter Germonpr?, Centre d?Oxyg?noth?rapie Hyperbare, H?pital Militaire Reine Astrid, 1120, Brussels, Belgium. Jean-Michel Pontier, D?partement de M?decine Hyperbare, H?pital d?Instruction des Arm?es Sainte-Anne, 83800, Toulon Arm?es, France. Allesandro Marroni, P.O. Box DAN Europe, 64026, Roseto, Italy. Adel Taher, Hyperbaric Medical Center, Sharm el Sheikh, Egypt. Frans Cronje, Hyperbaric Medicine and Research, University of Stellenbosch, Stellenbosch, South Africa. Costantino Balestra, P.O. Box DAN Europe, 64026, Roseto, Italy; and Environmental and Occupational Physiology Laboratory, I.S.E.K. Haute Ecole Henri Spaak, 1160, Brussels, Belgium. Physiopathology of Decompression (PHYPoDE) network Francois Guerrero, Jacek Kot, Peter Germonpr?, Jean-Michel Pontier, Alessandro Marroni, Adel Taher, Frans Cronje, and Costantino Balestra 118 ? tHE FUtURE oF Diving with worldwide activities (Dan Europe), four are hyperbaric medical centres and national reference centres for treatment and evaluation of injuries related to exposure to hyperbaric en- vironments (nCHm, HBoC, HmC, HmR). one industrial partner is a world leader for professional deep diving (ComEx s.a.), the other is specialized in the field of high-technology diving breathing apparatus (HBt). Finally, four network part- ners are higher education institutions with research activities in complementary fields of human physiology (UBo, nCHm, isEk, HmR). The PHYPoDE network will conduct the five subpro- grams below, which combine epidemiological study, fundamen- tal experimental and integrated approaches of the decompres- sion phenomena in both human and animal models, and the following prospective studies: 1. investigation of the environmental and behavioural deter- minants of decompression bubble formation; 2. determination of individual physiological and anthropo- metric determinants of decompression bubble formation; 3. research on secondary prevention of decompression sickness; 4. training of young researchers; and 5. dissemination of knowledge. although scientific and training collaborations already ex- ist between partners involved in the PHYPoDE network, they will be reinforced through this project. such constitution of a formal network will primarily benefit from the mutual recogni- tion of reciprocal quality of training that will increase the col- laboration and exchangeability of ?know-how? and knowledge at all levels. Host organizations will benefit from an integrated approach, gaining knowledge and time towards developing new decompression schedules. Furthermore, mutual recognition of the training program by the network institutions will lead to the possibility of a European degree in this field. Young researchers, the scientific community, and indus- tries will interact through planned conferences and workshops, with relationships and job opportunities emerging from the PHYPoDE project. The PHYPoDE network will meet indus- trial needs for knowledge and provide experts in decompres- sion. Beyond academic collaborations, the interdisciplinary programmes will simplify the connections with industry and facilitate the validation of research output by the practical ap- plication of scientific results. indeed, industries will look for sci- entific output in order to develop safer diving tools and update actual decompression software. This network will reinforce the partners? collaborations with both industrial entities from with- in and outside the network via communications in the planned conferences. The network will share knowledge with the internation- al scientific community and industries of the diving sector through international conferences. other training events such as network workshops will be open to external participation in specific cases. visiting scientists will be invited to participate in PHYPoDE workshops in order to ensure the highest scientific quality of the projects. They will be chosen at the international level to complete the network range of competencies or to ex- tend the network knowledge to different sectors such as space flights. visiting scientists will also be provided the opportunity to interact with external experts from Europe and beyond. Acknowledgments We acknowledge the participation of Bernard gardette and nicola Donda in support of the goals of the PHYPoDE network. the office of naval Research (onR) has funded science and technology re-search in Undersea medicine (Um) since the early 1950s. The Um Program is a part of onR?s Discovery and invention (D&i) portfolio, which makes broad investments in basic (6.1) and applied (6.2) research that will increase fundamental knowledge, foster opportunities for breakthroughs, and provide options for advanced technology research (6.3) through the Future naval Capabilities (FnC) Program. D&i programs such as Um nurture creativity and seek a balance between risk, opportunity, and potential naval impact. Figure 1 illustrates the Um program?s interactions and con- tributions to the U.s. navy and worldwide undersea community. uNdErSEa mEdIcINE rESEarch prOgram in order to leverage contemporary biomedical and electrophysiological technolo- gies, more recent Um Program efforts have shifted from defining safety/performance windows for undersea operations (e.g., dive table development) to investigating the mechanisms underlying health and performance challenges associated with manned un- dersea operations. Current investments are focused on, but not limited to, the following focus areas and sub-topics: ? Decompression illnesses (DCi) ? Basic and applied research in mechanisms of DCi ? advanced applications in DCi mitigation ? nonrecompression prevention and treatment strategies ? oxygen (o 2 ) toxicity ? Basic and applied research in mechanisms of o 2 toxicity ? advanced applications in o 2 toxicity mitigation ? short- and long-term Health and Performance Threats ? Basic and applied research in diver/submariner health and performance risks ? advanced applications in health and performance risk management for the diver/submariner During the 2007 fiscal year, the Um Program was designated as a national naval Responsibility (nnR). This designation was a major achievement because it established Um as a research program that is critically important to naval operations. While the in- struction that defines the nature of nnRs may be modified at the request and approval of the Chief of naval Research (CnR), the historical purpose of a nnR is to sustain (a) a robust research capability to work on long-term science and technology (s&t) challenges of interest to the Department of the navy, (b) a pipeline of new scientists and engineers Office of Naval Research, One Liberty Center, Suite 1425, 875 North Randolph Street, Arlington, Virgina, 22203-1995, USA. international Cooperation in Diving Research Matthew Swiergosz 120 ? tHE FUtURE oF Diving in disciplines of unique naval importance, and (c) s&t products necessary to ensure future superiority in integrated naval warfare. INTErNaTIONal cOOpEraTION aNd prOgram OppOrTuNITIES We assume that all diving and submarine communities share s&t challenges similar to those that were outlined in the previous section. Therefore, we should leverage the entire international community of scientists and engineers who have established expertise in diving and submarine research to sat- isfy Um Program goals and objectives. This does not equate to onR being the sole source for funding worldwide research in this area, but rather establishes itself as a leader to promote in- ternational cooperation and optimization of current and future investments. in some cases, this will involve direct funding of work efforts, and in other cases, it may take the form of align- ing international research efforts in order to more efficiently and effectively address common research goals and objectives. The Um Program seeks preproposals (white papers) and pro- posals for basic and applied research that address either DCi, o 2 toxicity, or other short- and long-term health/performance threats associated with manned undersea deployments. Work efforts nor- mally run for two to three years on fiscal year cycles (1 october - 30 september). Program opportunities are outlined below. ? Preproposal ? Format: www.onr.navy.mil/sci_tech/34/prop_prepro- posal.asp ? submission: email attachment to the Um Program of- ficer no later than 1 march for potential funding begin- ning 1 october ? Proposal ? Format: www.onr.navy.mil/sci_tech/34/prop_format_ instructions.asp ? submission: ? Universities, organizations, non-profits: www.- grants.gov per onR Broad agency announcements (Baa; www.onr.navy.mil/02/baa/) no later than 1 June for potential funding beginning 1 october ? Commercial: www.grants.gov (must establish as a commercial/profit entity) per onR Broad agency announcements (Baa; www.onr.navy.mil/02/baa/) no later than 1 June for potential funding begin- ning 1 october. Email attachment per onR Broad agency announcements (Baa; www.onr.navy.mil- /02/baa/) no later than 1 June for potential fund- ing beginning 1 october ? animal and Recombinant Dna Use: www.onr.navy. mil/sci_tech/ahd_usage.asp ? Human subjects Research sponsored by the Depart- ment of the navy: www.onr.navy.mil/sci_tech/34/- documentation_requirements.asp onR global (onRg) is the international presence for onR, which actively seeks opportunities to promote science and technology collaboration of mutual benefit between the United states and researchers around the globe. ? visiting scientist Program (vsP; www.onrglobal.navy.mil/ vsp.asp) ? Conference support Program (CsP; www.onrglobal.navy. mil/csp.asp) ? naval international Cooperative opportunities in science and technology Program (niCoP; www.onrglobal.navy. mil/nicop.asp) buIldINg aN agENda fOr ThE fuTurE an international meeting focused on the science of under- sea medicine does not currently exist. The Undersea and Hyper- baric medical society (UHms) hosts an annual meeting that is normally held in the United states. more recent UHms annual meetings have been dominated by plenary and continuing edu- cation sessions that more directly support clinical practice and recreational diving. The European Underwater and Baromedical society (EUBs) also hosts an annual meeting. its content has also been dominated by issues related to clinical practice and recreational diving. other professional organizations that hold annual meetings have a sparse representation of research focused on undersea medicine. While all of these meetings have merit, and we hope that our sponsored scientists and engineers con- tribute to the dissemination of professional information at these meetings, there is no indication that their agendas will consis- tently focus on the advancement of basic and applied research in undersea medicine and its previously discussed s&t challenges. more importantly, none of these meetings encourage young sci- entists to pursue a career in undersea medicine research. FIgURe 1. Office of Naval Research (ONR) interactions and con- tributions to the undersea community. The Discovery and Inven- tion portfolio outlined in bold within the ONR block represents the Undersea medicine program. 6.1 6.2 6.3 D & I FNC Bureau of Medicine and Surgery 6.4 Medical Applications Program R equirements /C apability Gaps Pr oduc ts Diving Community SUBL ANT SUBP AC N AVSEA 00C Tes t & Evaluation sWiERgosZ: CooPERation in Diving REsEaRCH ? 121 roadmap for undersea medicine research and encourage our brightest and most promising young scientists to dedicate their professional endeavors to our community. Future s&t capabil- ity in undersea medicine is contingent on our current ability to foster and develop new scientists and engineers. This objective may be achieved by establishing global collaborations that sup- port graduate student and postdoctoral awards. The Undersea medicine Program Review is an annual re- view of s&t research efforts funded by the Um Program at onR, Biomedical Deep submergence Program at naval sea systems Command (navsEa 00Cm), and related advanced efforts within the 6.4 medical applications Program at the Bu- reau of medicine and surgery (BUmED). The agenda for this meeting could be modified to include other sponsored work in undersea medicine. together, we could build an international R. moon: Does your funding by the government include only support for facilities, or does it include disbursement to other researchers in the form of grants? m. gr?nning: We do not issue grants; you have to apply for them elsewhere. our government funding only covers the daily activi- ties. But that is one of the big things we have to work towards. m. lang: You did not provide any numbers, what is your bud- get for international cooperation in diving research? m. Swiergosz: i am not at liberty to divulge that information. The budget that i manage is modest. at the last program review, we received high marks because my performers do good work in what is considered one of the better, relevant programs at the office of naval Research. We have been able to grow the budget and acquire some supplemental funding. This is the first year we have really engaged onRg heavily, and we have a commitment from them for support of the program. it is enough to support research and we will keep on going. m. lang: How many postdoctoral fellows, for example, did your program fund last fiscal year? m. Swiergosz: six, and in terms of graduate students, about thirty. There are also some predoctoral students who have re- ceived grant funding for good proposals. a. Brubakk: Perhaps one of the main points you are making is that international cooperation is very important. We have to establish a mechanism to make a decision on which important questions we want to answer. Diving research is a small com- munity with only a few laboratories around the world and a handful of people interested in this particular problem. i believe the time has come, one hundred years after Haldane, to decide on at least five very important questions. Then we would try to establish collaborations for the next two years to focus our ef- forts on those questions. The way to handle that is to establish an organization and state that if you want money, come in and join this effort. This effort has to be funded and organized with scientifically credentialed people with publication records in high-impact journals. after two years, we then review what we have done and decide which problem to address next. We need Discussion: Diving Researcher Recruitment Chairman: Richard E. Moon the type of cooperation we have never had but which is crucial to move this effort forward. m. Swiergosz: i wholeheartedly concur. to revisit the question about funding, the donor does not have enough money to fund all diving medical research. Furthering alf ?s thoughts, we need to build a roadmap, find out what resources we have, and le- verage them to the best of our ability. in order to argue for an increase in budget, notwithstanding the great science already being produced, there is a need for a cohesive strategy. This will allow us to identify and fill in the gaps in our knowledge. a. Brubakk: i am convinced that we need to get rid of the people with gray hair and the way to do that is to attract new people with high scientific expectations with less gray hair to become interested. We need to get these people to enter the top programs in genomics or whatever field it is and decide that these environmental physiology questions are so exciting that this is what they want to pursue for a long period of time. How do we accomplish that? Funding availability is only part of the story. money is a part of the story and will also follow perhaps the ability to generate excitement about this. This is also a pub- lic relations question: how do we sell ourselves in order to get it done? olav Eftedal suggested that we try to establish something similar to the human genome project. We have to make a com- mon effort and decide that this is important to do and convince the rest of the world of the same. R. moon: We need to hear from our consumers about their ideas but perhaps we could start by articulating the issues and challenges and then how to attract the funding. S. lothe: our challenge in the saturation diving community is that the gray hairs have become long. We need to take a look at the long-term health effects of diving. Entrance and exit se- lection criteria for the new generation of saturation divers also need to be established to ensure they can continue in this indus- try for a long period of time. should there be age-based exclu- sion criteria? The fastest growing segment of the diving industry is the recreational diver. This diver is no longer happy to dive to 124 ? tHE FUtURE oF Diving 20 or 30 m, they wish to dive to 100 m with rebreathers. This will be a regulatory challenge for all countries. ?. Duji?: The key issue remains the money and how we might influence our national and international funding instruments to support this rather small niche of science, which is really diving/ environmental physiology. in Europe, we have to really lobby in Brussels to change the priorities. There are preproposal steps we could take based on these discussions to initiate larger, multi- center collaborative studies. in our country, people with good research tracks are working as mDs because there is not much funding in diving physiology. We need to sell this idea through PR efforts and nice web pages on diving projects to influence the decision makers who are distributing the money. Without this effort, we cannot sustain long-term multicenter efforts as they are occurring in hypertension and diabetes for example. The exercise symposium had some excellent papers on coop- erative projects that serve as examples on how to collaborate. Those multicenter trials of high intensity exercise have a project coordinator, partners, and agreements on methodologies. These types of large-scale projects must have involvement from the national institutes of Health, the national science Foundation, and the European science Foundation, in addition to other na- tional science foundations. R. moon: The U.s. national funding agencies? take on this type of research support is that there are only a small number of div- ers with a small number of morbidities, unlike Hiv or diabetes. The funding priorities are fairly low for diving, except when there is a national priority, for example protecting divers in the military or the commercial field where there is a national inter- est. Before we arrive at the funding strategy, we need to identify the diving problems worthy of being solved. ?. Duji?: long-term health effects and safety of the millions of recreational divers diving deeper and deeper is a research possibility. a. Fahlman: Environmental physiology should be used as a tool to investigate clinical problems. Bubbles are a huge problem in cardiopulmonary bypass surgery. inert gas exchange every time we go under anaesthesia also needs to be better understood. m. lang: The common thread in the presentations by s. slor- dahl and o. Ellingsen centered on the issue of attracting new young researchers to the field for pure science. solving specific operational diving issues may be viewed as an applied science and is perhaps best addressed through a contract with a spe- cific outcome or solution. David Doolette said it best, ?Haldane still rules.? in 1991, we held a large, interdisciplinary repeti- tive diving workshop at Duke University medical Center. The international DCs incident rates, as presented by the diving communities at that time, were 1:1,000 dives for commercial diving, 3.2:10,000 dives for recreational diving, and 1:100,000 dives for scientific diving. at those levels of DCs incident rates, deemed ?acceptable? by those three diving communities (recognizing that zero incidents is desirable but unachievable), it would appear that even 18 years ago, decompression sick- ness was not a significant problem. going forward, focusing solely on decompression sickness may not be the path to fol- low. For the advancement of science, not solely the solution of operational diving problems, the probability of attracting bright young researchers would more likely follow a redirected research emphasis that focuses on big scientific questions. a. Brubakk: i quite agree. We need to look at scientific prob- lems in a way that has practical applications, hopefully for mil- lions of people. as an example, some work we have done here was actually by chance, looking into diving, bubble formation, and endothelial function, has changed the whole field of exer- cise physiology. Everyone believed until then that the effect of exercise came from doing it for months at a time until some- thing changed. First, exercise is becoming more important as a drug, and second, it is the single bout of exercise that is im- portant. This result came from a diving experiment and is the type of research that will excite people outside of the diving community and attract funding. C. Balestra: We have to go outside our research niche. another idea is to understand how a healthy human coping with an ex- treme environment compares to an older human coping with a normal environment. I. eftedal: i found it very interesting from a medical genetic point of view to determine how the genome or proteome is challenged by a changing physiology. Health is to a very large extent defined by how much you can express your proteomes in response to a challenge you have. People will be attracted to study the effects of the environment on the genetics, genomics, and proteomics. ?. Duji?: sleep apnea may be a healthy research model to exam- ine as there are millions of suffering people. Hypertension, pul- monary arterial disease, lipid disorders are others. nonspecific effects of stress response during diving are another topic worthy of investigation and how we can have a wider impact outside of our community. R. moon: another way of looking at it is that there are people working on some of these phenomena in other fields that could be attracted to work on some of our diving problems. For exam- ple, there are many investigators working on spinal cord injuries and biochemistry of heat-shock proteins who could apply their techniques to diving. S. lothe: We have extreme exposure divers in their 50s and 60s who have for the last 30 years spent 180 plus days in saturation at 100-200 m depth every year. These healthy individuals may be of interest as to what makes them tick. These divers are fit and interested and do not want to stop saturating. They have experienced the most extreme pressure changes of any popula- tion. What makes them fit? Why are they healthy? moon: DisCUssion ? 125 anaesthesia; the review response is ?what problem?? We anaes- thetize you, you wake up, and you survive, no problem. in order to secure funding, we need to articulate what the new challenge is beyond what we can do today. J. Jakobsen: That is exactly the point when we approach diving management, what problems, we do not have any problems. one way to tackle this might be to implement a more compre- hensive monitoring of the exposure that takes place during div- ing. There exists no clear definition of what exposure is or the analytical tools to describe the exposure, which could be petro- chemical contamination, thermal, oxygen, or other parameters. The automatic monitoring system is mapping just about every- thing so we know an awful lot about operational dives, but what are going to do with all that operational data? We need to find a proper way of storing it, reduce it, and analyze it by developing tools to determine if there is a long-term health effect. R. moon: That is a very good point but how does one get around the issue that you just raised that the diving doctor community is trying to get people to stop doing what is currently seemingly ok? m. gennser: integrative physiology has gone down a lot in the past decades, displaced by genomics, biochemistry, proteomics. Progress made in those fields now needs to be tested on themes exposed to human environments. one of our selling points is actually that our exposures to extreme environments being de- compression or hypoxia or large accelerations can expose physi- ological, biochemical mechanisms, but this is not immediately evident. one colleague has been working on arm pain experi- enced by pilots of high performance aircraft. The research there has provided insight into causal mechanisms. J. Jakobsen: in the 80s and 90s, there was a situation where most of the divers were permanently employed, and there was good feedback from the divers on their problems. today the divers may not tell the medical doctor of their problems for fear of losing their job, so there needs to be a security system to share that information. a. Brubakk: We cannot solve all diving problems at once and need to come back and decide which problems we think are the most important. D. Doolette: Cross-fertilization of diving problems that an- dreas Fahlman mentioned to solve other problems is a very good idea. i did that for ten years until the funding agency realized there was an easier way of doing that. and as michael lang said there are operational procedures to be developed as well such as diving to 100 m for science purposes. Decompression is a very obscure field, and it is really operational issues that are going to drive it; if they are not there, we are not going to exist. We can solve the cross-fertilization problems in another way. our responsibility is to recruit scientists into this field to keep it go- ing. it will not support very many people, but it needs to be a continuous program. The navy keeps funding some university programs even though there are no current problems because they know there will be problems in the future. Collaboration J. Jakobsen: The key word is natural selection. Why are some selected to continue and why do others drop out? J. Ross: This kind of question inevitably leads to psychological issues and people?s perception of the risk of the job they are in and how they should be responded to. Why do people go to see their doctor? is it because they perceive minor symptoms of everyday illness, or are these people who do something very different with high profile health effects? some people will be vulnerable to that concept and some people will not. These are not just diver issues. it goes back to the impact of technology on human populations. Repetitive strain injury is a phenomenon where people react adversely to a change in their work circum- stances. This is a psychological issue with potential particularly in the oilfield diving industry. R. moon: articulation of the problems and availability of mon- ey will lead to young and senior researchers? willingness to work on these questions. twenty-five years ago Hiv research did not exist, and it is now a multibillion dollar enterprise. Where did those researchers come from? They came because there was a problem, money to solve the problem, a complicated issue that was interesting to work on. There is no shortage of interesting issues to work on in diving. J. Jakobsen: The question is strategy for recruiting. The div- ing contractor needs to feel supported by the research commu- nity, not through investigations of all kinds of reasons why he should stop diving. support would come by finding methods of adapting to problems and solving his potential trouble so that he can continue diving. The latter approach is very important. management of many diving companies is reluctant to allow diving researchers into their ranks because it is a potential show stopper. R. moon: That is an important point because the diving com- munity?s traditional approach has often been to find reasons why people should not dive, which is easy to accomplish with- out very much data. m. Swiergosz: The quality of life issue for many U.s. navy divers is their continued ability to dive. We should be try- ing to enable human beings to go into this potential adverse environment. S. Daniels: The last time there was any significant investment in this field was when the U.s. navy developed a strategy to dive beyond 500 m. That coincided with the explosion of explora- tion for oil and gas at depths beyond which diving technology could provide for. There is a common perception these days that those problems have been solved. There is no clear statement by any of our community about what the problem is. Until we can articulate that, we will not attract people into this field. There exists exactly the same issue in anaesthesia, which is a field with scientific terms that has been running for approxi- mately the same amount of time as diving. nowadays, certainly in the U.k., under appropriate national government funding organizations, there has to be clinical representation. assume an application for support of a program to look at mechanisms of 126 ? tHE FUtURE oF Diving is the key and in the civilian world there needs to be a group of people who get together and make a home for those scientists. a permanent ongoing presence will attract people, a short-term funded program will not. Creation of a virtual international in- stitute of diving with some key members, and a common set of problems to be addressed will feed that longevity. R. moon: an important aspect of collaboration is meetings like this. matt swiergosz runs a meeting every year in the U.s. along similar lines which creates a major portion of the atmosphere of excitement that draws young people into the field. ?. Duji?: We can sell environmental physiology as human inte- grative physiology, a field that is coming back. looking only at molecules will not solve complex situations. at the University of Wisconsin in milwaukee, there was a departmental fusion of anesthesiology with Physiology 30 years ago because a fac- ulty member in anesthesiology, John kampine, was really a big name in Physiology as well. i completed my work in anesthe- siology but actually got my PhD in physiology. By being con- nected with another field is how anaesthesia survived, and it is now considered one of the best basic experimental anaesthesia departments in the U.s. p. lindholm: many of the commercial operational dives are not a big problem. today the recreational divers go beyond the lim- its and do not have a safety net. The technical divers push the limits and present with strange decompression issues. Breath- hold divers are drowning in swimming pools, mostly due to lack of training. Recreational divers on beta-blockers for hyper- tension come to the diving doctor and ask if they can still go diving. Who is going to do that research? There have been some advances in asthma and diabetes in diving. in the recreational diving community, there are problems from a medical point of view. it remains however a small group and small problem with respect to other societal medical problems. J. gangenes: north sea diving operations are saturation diving almost constantly from 60 m down to 300 m. The discussions have been regarding surface-oriented decompression for normal air or nitrox dives. The saturation divers do not complain be- cause of their employment. Part of their living is diving. The challenge we have is those divers who have been under pres- sure for 30 years now do not tell us their problems, and we do not know what the maximum amount of years under pressure should be. J. Buckey: The perception of this field in general does not come so much from diving as it does from our larger clinical cousin, hyperbaric oxygen therapy. There is much gray research in that area that is not being done, and there are problems with how hyperbaric oxygen therapy is being applied. That may be where much of the doubt is coming from in getting more people into this area. more funding in the HBot area would likely bring in additional new researchers who would then also be interested in diving. m. lang: Directors of scientific program use metrics to evaluate the effectiveness and success of the research funding provided. one such metric is quantity and quality of publications in high- impact journals such as Science, Nature, or Proceedings of the National Academy of Sciences. Positive results were shown here in this regard, and perhaps those studies that were successfully published in such journals are areas that should be further pur- sued as they were successful in breaching the wall and acquiring continued funding. J. Ross: We are talking about decompression sickness here, and the problem as i see it is getting all the people to examine who actually have the problem. That way you can apply advanced techniques and investigations of the people who actually have the disease. to that end, there is major room for international collaboration for studies of DCs. in scotland, we have four chambers that treat between 100-150 cases per year, and there are other chambers around the world with similar numbers. This is where you can also pick up cases of severe decompression sickness that will illustrate what the major issues are. Why do divers die from decompression illness? once you get a hold of that kind of question and the real cases, you can do the research. Decompression illness is less common as time goes by with the development of decompression technologies. We are developing an international database that will inform us of what is actually going on with people with decompression illness. S. gaustad: to help provide some funding perhaps a small an- nual fee to an organization that does diving research may be the way to go, similar to Dan member fees. C. Balestra: For reference, Dan Europe tries to get 1 Euro from each member for diving research. a. Brubakk: We tried to establish some years ago the notion of having every recreational diver pay 5 dollars, pounds, Euros, whatever denomination, in exchange for which they get 5 extra minutes of bottom time. This is not an impossible idea. Every diver would pay to be member of this group with a community feeling that they are supporting diving research and getting five extra minutes of bottom time. That?s PR, the idea is to not get too specific. a. arnfinsen it seems the oilfield diving companies have been challenged several times about providing the money for diving research. What about Cmas and PaDi who are providing the recreational divers, where are they? a. Brubakk: We challenge everybody. J. Jacobsen: There is a negative focus of much of the research that has been going on historically. it is focused on the accidents and outcomes of what we are doing. This is overdocumented compared to the success stories. We annually conduct thousands of saturation hours worldwide without incidents. Divers get in alive and get out alive. We should also focus on how we as com- panies and the scientific community document operational re- search. This is scientifically very low on the scale of what people publish. if you review the literature, operational research results are almost nonexistent. That will lead to a documentation of moon: DisCUssion ? 127 best practice, which is what we live off. it is also something we can easily sell internally in our industry and to those who fund us, the oil companies. That is how we review what we are do- ing without jeopardizing statoil or the norwegian government. my second comment is about terminology and the concept of bounce diving, which is very different in the commercial diving community, a historical concept that was used offshore and is not used anymore. The conceptualization of saturation diving and what is best practice is different around the world and even within large companies. saturation diving has also changed from being bottom-focused work in very deep water to shal- low-water saturation because the oil industry has changed. We now do shallow sats in the air range of 15-20 m with a narrow excursion window and problems or possibilities that we know very little about. The tables were developed many years ago and are now carved in stone and almost impossible to change be- cause of regulations. my final point is that of the transition of the knowledge gained within these various fields in a feedback loop to where we can utilize it. We are now doing extensive de- velopment of personal diving equipment: rebreathers, drysuits, electrically heated undergarments, etc., which definitely have applications in military diving, where these developments are also taking place. We are reinventing the same wheel in many instances for different reasons. There is no obvious feedback and lack of knowledge among diving communities of what each is doing in this small community around the world. O. eftedal: The problem of petrochemical exposure under pres- sure is of interest and concern to occupational medicine. m. lang: Does anyone have the decompression sickness inci- dent rate for saturation diving, or are there no problems when divers come out of sat? O. eftedal: very little, there have been only two cases in nor- way in 15 years. J. Ross: i recently did an assessment of DCs rates among div- ers in the north sea, and the incident rate has not change in many years. There is not much literature on diving from a pub- lic health or preventive medicine perspective. i do population- based medical research, and there are not many studies of large populations of divers. i reiterate the need for generating data to establish what the real problems are in populations of divers. g. perdrizet: i have installed a large multiplace chamber in the new England region of the United states. i have seen two re- quests for recompression treatment for decompression illness. However, in a normal week, i will see multiple head injuries, crush injuries, and spinal cord injuries. in terms of the hyper- baric medicine issue, we should focus on that. Right now we are linked in a seven center trial where we are trying to get money. We have been turned down once by the Department of Defense but are resubmitting. it is very difficult with a high-profile type diagnosis, but there needs to be a perception by the funding agencies that you have a problem. Hiv shows up on time with affected people, but you do not have that here. i recently sub- mitted a grant proposal to the niH to use hyperbaric medicine in a major emerging infectious disease in the world right now that is causing deaths in our hospital. it is interesting that a pall sits over hyperbaric medicine right now, which is in its infancy. This is a big opportunity but a big uphill struggle which we all suffer from. R. moon: We have a problem otherwise many of us would not be sitting here. The main discussion points are summarized below. 1. phySIOlOgy/BIOChemISTRy ? immersion ? inert gas exchange ? Pulmonary gas exchange ? Breath hold diving ? narcosis ? HPns ? genomics/Proteomics ? Exercise ? stress 2. ChalleNgeS ? Publication of nonaccident experience ? Change approach to extend, not exclude, diving careers ? shallow saturation tables are antiquated ? DCs ? pathophysiology ? treatment ? saturation diving ? Decompression ? infections ? Emergency care ? long-standing sat divers ? trace gases ? What is ?best practice?? ? agEing DivERs ? Exclusion criteria (age, co-morbid conditions) ? DivER sElECtion ? sat divers ? oxYgEn EFFECts ? toxiC EFFECts (petrochemical) enhanced by increased pressure? ? Thermal issues ? long-term effects ? Psychological issues ? Population studies in divers needed (vulnerability of people to work-related environments) 3. meThODOlOgy ? mechanisms of HBo ? Cross-fertilization ? Cardiopulmonary bypass ? anaesthesia? ? military-commercial diving cross-fertilization 4. mONey ? multicenter collaboration: DCs cases pulmONary ShuNT aNd alvEOlar cOllapSE in 1940, Per scholander published his seminal work on diving mammals. in one section of this opus, he discussed the implications of the anatomy of the respiratory system of marine mammals. He noted that marine mammals appeared to have a very compliant rib cage and stiffened upper airways. He suggested that the increasing pressure with depth would compress the chest and push all the air into the upper air- ways. This would prevent gas exchange at depth and thereby reduce n 2 uptake during breath-hold dives. Using measured volumes of the upper and lower respiratory system in whales and seals (Hyperoodon ampullatus, Cystophora cristata, Halichoerus grypus, Ba- laenoptera physalus, Phocaena communis), alveolar collapse and termination of gas ex- change was estimated to occur at depths ranging from 30 m to 210 m depending on the diving lung volume (scholander, 1940). Following the work of scholander, a number of studies have attempted to answer the question of when and where the lungs may collapse in different species. Biological tissues have limited scope to resist pressure differences and negative transthoracic pres- sures exceeding 100 kPa will damage tissues (Brown, 2000). it was therefore concluded that the chest must compress and eventually collapse to prevent damage. Both direct and indirect evidence of chest collapse has been reported in the diving dolphin (kooyman, 1973; leith, 1989; Ridgway et al., 1969). However, compression of the rib cage does aBSTRaCT. Ever since the pioneering work by scholander in 1940, it has been generally accepted that alveoli collapse at shallow depths in breath-hold diving marine mammals, reducing the amount of n 2 taken up. This has been described as the major adaptation to prevent decompression sickness (DCs) in mammals. However, more recent experimental and theoretical work suggests that gas ex- change may continue to depths much deeper than previously believed. Foraging marine mammals perform bouts of repeated long and deep dives interspersed by short surface intervals. With con- tinued gas exchange, this results in accumulation of blood and tissue n 2 to levels that would cause a high incidence of DCs in terrestrial mammals. it has been suggested that a coupling between physiology and dive behaviour may be important in decreasing end-dive n 2 levels. For example, mammals may adjust their diving lung volume during dives to different depths, allowing for behav- ioural control of the depth of alveolar collapse. a slower ascent rate close to the surface coupled with a presurface tachycardia may reduce the end-dive venous n 2 tension (Pn 2 ) by as much as 45%. The short and shallow dives that commonly occur between dive bouts may moderate n 2 supersaturation and thereby DCs risk. adipose tissue, with its higher n 2 solubility, may lessen DCs risk during the first few dives in a bout by functioning as a sink to buffer extreme levels of n 2 , but may eventually become a liability and force animals to end a dive bout. it is also possible that marine mammals evolved adaptations making them less susceptible to bubbles. Understanding these successful adap- tations may provide novel ways of preventing or treating DCs in human divers. Biology Department, Mailstop 50, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, 02543, USA. How Do marine mammals avoid DCs? Andreas Fahlman 130 ? tHE FUtURE oF Diving not prove that the alveoli have collapsed and that gas exchange has stopped. Collapse depth has been estimated by assuming a rigid trachea and highly compliant lung (Denison and kooy- man, 1973; stephenson, 2005). This assumption is questionable as the trachea in Weddell and elephant seals showed significant compressions at depth of only 54 m. Direct measurement of inert gas exchange during diving has suggested that alveolar col- lapse, and a concomitant cessation of gas exchange, occurs at 30 m depth in the Weddell seal (Falke et al., 1985) and 70 m in the dolphin (Ridgway and Howard, 1979). These depths are relatively shallow and could possibly prevent significant levels of n 2 uptake and minimize DCs risk during deep diving (Falke et al., 1985; Ridgway and Howard, 1979; scholander, 1940). an experimental study, on the other hand, suggested that gas exchange may persist until a depth of 170 m in seals and sea lions diving with lung volumes ranging between 18 to 67 % of total lung capacity (tlC) (kooyman and sinnett, 1982). in ad- dition, recent modelling attempts provide alternate explanations for the uptake and removal of n 2 in the dolphin and Weddell seal (Bostrom et al., 2008; Fahlman et al., 2009; Fahlman et al., 2006), consistent with the idea postulated by scholander (1940) and experimentally shown by kooyman and sinnett (1982), that alveolar compression results in an increasing pulmonary shunt. ESTImaTINg luNg cOllapSE dEpTh uSINg N2 upTakE aNd rEmOval the dolPhin lung collapse in the dolphin was estimated at 70 m in a study by Ridgway and Howard (1979). They studied freely div- ing dolphins trained to dive repeatedly to 100 m with average dive duration of 1.5 min and surface interval of 1 min. The two dolphins dived repeatedly 23 times for an hour. at the end of the bout, each dolphin jumped onto a work platform and a catheter was inserted into the dorsal epaxial muscle tissue. The muscle n 2 washout rate was converted to a tissue half-time (? tiss1/2 ), which determines the time to equilibrium after a change in pressure. The time constant is physiologically relevant and related to the solubil- ity of the gas, the blood flow, and the volume of blood that passes by the tissue in a given time (Fahlman et al., 2009; Fahlman et al., 2006; Fahlman et al., 2007). The data from the dolphins sug- gested that muscle ? tiss? were 6.6 min and 5.2 min for each of the two animals. The observed values of Pn 2 could be explained us- ing a standard exponential gas exchange model that assumes sym- metrical gas uptake and removal and assumes that gas exchange was terminated at 70 m. The authors suggested that this provided evidence for lung collapse (Ridgway and Howard, 1979). How- ever, most, if not all, breath-hold diving mammals show a higher heart rate while breathing at the surface as compared to diving, i.e., the dive response (Butler and Jones, 1997). since blood flow affects inert gas uptake and removal in breath-hold diving mam- mals (Fahlman et al., 2009; Fahlman et al., 2006; Fahlman et al., 2007), the assumption of symmetrical gas uptake and removal has to be reconsidered. in fact, a recent study showed that a 50% re- duction in blood flow during diving as compared with the surface value (Fahlman et al., 2006) provides estimated values that agreed with the observed data in the dolphin. the weddell seAl in a study on freely diving Weddell seals (Falke et al., 1985), arterial blood was sampled during the descent phase every 30 sec. The results showed that, independent of the maxi- mum dive depth (80 m to 200 m), arterial Pn 2 continued to increase during descent to a depth of 30 m. This was followed by a continual decline in arterial Pn 2 as the animal continued the descent (Falke et al., 1985). it was concluded that this was evidence of alveolar collapse and termination of gas exchange at a depth of 30 m, and the gradual decline in arterial Pn 2 caused by tissue absorption of the available blood Pn 2 . However, this conclusion is incompatible with cardiopulmonary physiology and not evidence of complete alveolar collapse, but rather agrees with the suggestion of a pulmonary shunt that increases with depth (Bostrom et al., 2008; scholander, 1940). if the data were evidence of alveolar collapse, arterial Pn 2 should immediately have declined close to ambient levels. The reason for this is the circulatory transit time, which has been shown to be between 2-3 min in Weddell seals during forced chamber dives (kooy- man et al., 1972). While the dive response differs in forced and freely diving animals, it being more severe in the former, the forced dives resulted in a 87% decrease in the heart rate dur- ing diving in the elephant seal (kooyman, pers. comm.) and by 75% in the freely diving Weddell seal (Falke et al., 1985). if heart rate is directly related to cardiac output, the reduction in blood flow is similar in these deep diving species. in other words, as the animal descends, it takes 2-3 min for the first volume of blood to go from lung to tissues and back to the lung. Thus, a dive-related increase in mixed venous Pn 2 lags behind arterial Pn 2 by 2-3 min (kooyman et al., 1972). The seals reached 30 m during the 80 m or 200 m dive after ~ 2 min of submergence (see Fig. 2 in Falke et al., 1985). Therefore, if the alveoli collapse within 2-3 minutes of submergence, arterial Pn 2 should drop instantaneously close to the value at the sur- face because any n 2 taken up before alveolar collapse would not have made it around the circulatory system. The antagonistic effect of compression on diffusion rate is an alternative explanation that explains the data in the Wed- dell seal (Bostrom et al., 2008; Fahlman et al., 2009; scholan- der, 1940). Compression of the respiratory system will, on the one hand, result in an increasing partial pressure gradient and thereby an increase in the diffusion rate. on the other hand, compression will also result in a reduction in the diffusion rate caused by the reduced gas exchange surface area and increas- ing diffusion distance. Thus, compression will initially increase the diffusion rate and inert gas uptake but as depth increases a pulmonary shunt will develop that decreases the uptake rate (Bostrom et al., 2008; Fahlman et al., 2009; scholander, 1940). FaHlman: HoW Do maRinE mammals avoiD DCs? ? 131 affects gas exchange, termination of gas exchange is routinely quoted in animal physiology textbooks as the primary adapta- tion that protects marine mammals from elevated n 2 levels and DCs. scholander did indeed suggest that cessation of gas ex- change could protect against DCs but indicated that the actual diving lung volume would determine the depth at which this oc- curred, and there is theoretical (Bostrom et al., 2008; Fahlman et al., 2009) and experimental evidence (kooyman and sinnett, 1982) that supports this idea. it is less well known that scholan- der also reported two possible cases of DCs in a fin whale and hooded seal during a single dive. if mammals adapted for pro- longed deep diving can experience DCs during a single dive, DCs may also be observed in breath-hold diving humans. While marine mammals perform single deep and long dives without apparent DCs symptoms, more remarkable are the extensive foraging bouts carried out by many diving mam- mals and birds. such dive behaviour should result in accumula- tion of n 2 in tissues causing a higher risk of DCs. scholander concluded that ?by repeated dives, conditions as regards diving meAsuring chAnges in gAs exchAnge At Pressure While the studies on dolphins and Weddell seals implicitly assumed that termination of gas exchange occurred instantane- ously, others have shown experimental and theoretical evidence that compression results in a shunt that increases with pressure (Bostrom et al., 2008; Fahlman et al., 2009; kooyman and sin- nett, 1982; scholander, 1940). in one study, measuring pulmo- nary shunt in California sea lions and harbour seals at pressures equivalent to a depth of 70 m to 90 m, respectively, showed a reduction in gas exchange that correlated with depth and diving lung volume (kooyman and sinnett, 1982). at a depth of 90 m (10 ata), the shunt exceeded 70% in the harbour seal and complete alveolar collapse and termination of gas exchange was estimated to occur between 160 m and 170 m (17 ata to 18 ata, kooyman and sinnett, 1982). The species used for this study were chosen as they show the most divergent airway struc- ture of those measured in pinnipeds (Denison and kooyman, 1973). Despite this, the compression shunt at pressures below 8 ata (70 m) were not remarkably different from each other (see Figs. 2 and 3 in kooyman and sinnett, 1982). a recent mathematical model, describing compression of the upper and lower respiratory tract (Bostrom et al., 2008), showed that graded alveolar collapse and its effect on gas ex- change produced results that agreed with observed data in the dolphin (Ridgway and Howard, 1979), California sea lion (kooyman and sinnett, 1982), Weddell (Falke et al., 1985) and harbour seal (kooyman and sinnett, 1982). in addition, the model also predicted compression of the upper respiratory tract that agreed well with the measured data in the Weddell seal. Furthermore, it was predicted that complete collapse would not occur until a depth > 150 m, similar to the predictions made by scholander (1940) and kooyman and sinnett (1982). if gas exchange does not cease at shallow depths, is it pos- sible that some species of mammals live with elevated n 2 levels that could cause bubble formation with alterations in the dive behaviour? in fact, mathematical modelling estimates suggest that deep diving beaked whales may experience end-dive mixed venous Pn 2 values that result in 50% DCs in similar sized ter- restrial animals (Fig. 1, Hooker et al., 2009). if so, how do they avoid DCs when foraging for food and could climate change impose behavioural changes that increase risk? OThEr mEchaNISmS TO avOId dcS only one study has measured the effect of compression on gas exchange (kooyman and sinnett, 1982) and no study has directly measured cessation of gas exchange. Theoretical mod- els and experimental data propose that the depth for complete alveolar collapse ranges between 30 m to 300 m depending on the structural properties of the respiratory system and the diving lung volume (Bostrom et al., 2008; Fahlman et al., 2009). While little information exists how the respiratory system in marine mammals compresses during breath-hold diving and how this FIgURe 1. estimated N 2 saturation pressure (aTa) that would re- sult in 50% decompression sickness (DCS) in a range of terrestrial animals after a rapid decompression (Flynn et al., 1971; Berghage et al., 1979). Black circles are tissue saturation P N2 for terrestrial animals. The solid line indicates the best fit regression logeD50 = 0.730 ? 0.205 ? logMb. Open and grey symbols are average mixed venous inert gas tension (Pv N2 ) for Blainville's beaked whale (Meso- plodon 78 ), Cuvier's beaked whale (Ziphius 34 ) and northern bottle- nose whale (Hyperodon 28 ) using model a. The figure was previ- ously published by hooker et al. (2009). 132 ? tHE FUtURE oF Diving disease would certainly tend to be worse on account of an ac- cumulation of invaded nitrogen. There is every reason to believe that this risk exists unless there is sufficient ventilation between dives (scholander, 1940). This agrees with a recent study by Zimmer and tyack (2007) and Fahlman et al. (2009) in which it was concluded that repetitive diving to shallow depths may cause elevated tissue Pn 2 and bubble formation. necropsy re- sults in stranded beaked whales and dolphins (Cox et al., 2006; Jepson et al., 2003) were suggestive of DCs-like symptoms. These mass stranding events correlated with naval exercises us- ing mid-frequency sonar. it was suggested that the sonar activ- ity may have led to disturbances of the natural dive behaviour resulting in dive profiles causing bubble formation. only a few alternative explanations have been proposed to explain how marine mammals avoid elevated inert gas up- take during breath-hold diving. kooyman (1973) summarized most of these in a review on the respiratory adaptations in ma- rine mammals: 1) increased tissue and blood n 2 solubility, 2) a special n 2 absorbing tissue, 3) changes in cardiac output and varying blood-flow distribution are some possible physiological adaptations that may help prevent excessive inert gas uptake in addition to pulmonary shunt and alveolar collapse. it is also possible that a coupling between physiology and behaviour is important to moderate tissue and blood Pn 2 levels and this may have a significant impact on diving animals with increasing an- thropogenic disturbances and global environmental change. increAsed blood n2 solubility There are no published studies that have measured the solubility of n 2 in tissues of diving mammals or birds, but the solubility of blood is similar in the seal and human (kooyman, 1973). The foam normally found in the upper respiratory tract of marine mammals has been suggested to be a potential n 2 absorbing agent but there is no experimental support for this supposition (kooyman, 1973). animal research has shown that inert gas removal can be accelerated by intestinal microbes that metabolize a small portion of the inert gas burden (Fahlman and kayar, 2003; kayar et al., 1997). For example, a 5% re- duction in the inert gas burden reduced DCs incidence by as much as 50% (Fahlman et al., 2001). nitrogen-fixing microbes are found in the gut of animals and if present in diving mam- mals would provide one additional avenue for inert gas removal. interestingly, a similar suggestion was described by scholander where n 2 from blood in vitro disappeared in the presence of o 2 (scholander, 1940). it was suggested that this observation was caused by n 2 fixation of a microbe called organism-x but this was dismissed by others (scholander, 1940). sPeciAl n2 Absorbing tissues Diving mammals perform extended dive bouts consisting of repeated dives interspersed by surface intervals that com- monly are shorter than each dive. The Pn 2 level of each tissue throughout a dive and a bout depends on the specific ? tiss? . Be- cause this variable is governed by local blood flow, ? tiss? changes for most tissues. most diving mammals have large amounts of subcutaneous fat that reduces heat loss and acts as an energy res- ervoir during extended periods without food. The 5-fold higher n 2 solubility in fat as compared with lean tissue, combined with the reduction in cardiac output and redistribution of blood flow that represents the dive response, results in a high ? tiss? for adi- pose tissue (Fahlman et al., 2009; Fahlman et al., 2006). These properties have led researchers to suggest that adipose tissues could act as a n 2 absorbent and reduce bubble formation during deep and short duration dives (Behnke et al., 1935; Fahlman et al., 2007; kooyman, 1973). During the first few dives of a dive bout, tissues with a low ? tiss? (Cns and muscle) experience high Pn 2 during the dive, but much of the accumulated n 2 is removed during the ascent and only low levels remain as the animal surfaces (Fahlman et al., 2007). The high ? tiss? of subcu- taneous fat, on the other hand, leads to a slow but continuous increase in Pn 2 (Fahlman et al., 2007). During the ascent, the presurface tachycardia reported in both diving mammals and birds (andrews et al., 1997; Elsner, 1965; Froget et al., 2001; kooyman and Campbell, 1972) and increased perfusion to adi- pose tissue allows a portion of the n 2 in the fast tissues to be taken up by the fat without any dramatic increase in Pn 2 . This could help reduce overall mixed-venous Pn 2 and thereby de- crease the likelihood of bubble formation (Fahlman et al., 2009; Fahlman et al., 2007; kooyman, 1973). Thus, fat Pn 2 is neg- ligible at the beginning of the bout, but slowly increases dur- ing each dive even during most of the ascent. This continuous increase in Pn 2 could eventually result in elevated adipose Pn 2 that could force the animal to undertake a long surface inter- val (Fahlman et al., 2007). Consequently, adipose tissue could help buffer Pn 2 at the beginning of a dive bout but become a liability after a long bout (Fig. 2). This could have significant consequences if the food source available for the animal moved deeper due to global warming or became scarcer as a result of over fishing possibly leading to reduced foraging efficiency. chAnges in cArdiAc outPut And blood-Flow distribution The dive response has been suggested as a useful physiologi- cal mechanism to reduce inert gas uptake (Fahlman et al., 2007; Ponganis et al., 1999; scholander, 1940). While this suggestion makes intuitive sense, a previous study only analyzed a 1-hour dive bout consisting of 23 dives (Fahlman et al., 2006). a more recent theoretical study, estimating tissue and blood Pn 2 levels in deep-diving king penguins during a foraging trip, showed that an increase in blood flow during diving led to an increased Pn 2 at the end of an extended dive bout in some tissues but a decrease in Pn 2 in other tissues (Fahlman et al., 2007). For example, diving bradycardia caused a substantial reduction in brain and central circulation Pn 2 , but an increase in muscle and fat Pn 2 . These surprising results suggest that the diving-related FaHlman: HoW Do maRinE mammals avoiD DCs? ? 133 reduction in blood flow does not always reduce n 2 levels during repeated diving. interestingly, each tissue had a specific blood- flow rate that resulted in maximum end bout Pn 2 (Fig. 3). a ? tiss? was computed for each tissue, and it was shown that the ? tiss? resulting in maximum end bout Pn 2 was the same for the different tissues and similar to the average dive duration of 1 to 1.5 min (Fahlman et al., 2007). it will be interesting to investigate if the ? tiss? that results in maximum end bout Pn 2 corresponds to average dive duration in different species and if so, this could be an inherent property of inert gas flux in deep diving animals. if that is the case, one would predict that diving animals would avoid tissue perfusion rates that result in tissue ? tiss? close to the average dive duration. However, as the circulatory system is also responsible for remov- ing Co 2 and supplying o 2 , blood flow changes to each tissue is a trade-off between the need to exchange metabolic gases and the need to reduce DCs risk. Thus, the blood Pn 2 at the end of a dive or an extended bout is a complex function of the need to supply o 2 to, and re- move Co 2 from, central organs while simultaneously reducing uptake of n 2 . The question is to what extent blood-flow changes are used as a means to reduce extreme Pn 2 without ischemic injury; this will be an interesting area of research. couPling behAViour And Physiology to reduce tissue And blood n2 leVels Diving mammals and birds may also use behavioural means coupled with physiology to reduce the inert gas burden. it has been shown that the presurface tachycardia (andrews et al., 1997; Froget et al., 2001) and reduction in ascent rate close to the surface seen in some species (Banish and gilmartin, 1992; Hooker and Baird, 1999; sato et al., 2004; tyack et al., 2006) may reduce the inert gas burden by up to 45% before reaching FIgURe 2. (a) ambient pressure (p amb ), and (B?D) estimated N 2 tensions (pn 2 ) for a tissue (B, heart) an intermediate (C, muscle) and a slow tissue (D) fat in a 12 kg king penguin. Data shows how the fast and intermediate tissues rapidly reach equilibrium while the slow tissue continuously increases during the entire diving bout. Figure is modified from Fahlman et al. (2007). FIgURe 3. model sensitivity analysis comparing (a) central circu- lation, (B) muscle, (C) brain and (D) fat pn 2 levels against diving tissue time constant (?). Results shown are predicted compartment values at the time the bird reaches the surface after the last dive in a dive bout (black circles), or after deep (>50 m, light gray triangles) or shallow (?50 m, dark gray squares) dives. Figure is modified from Fahlman et al. (2007). 134 ? tHE FUtURE oF Diving the surface (Fahlman et al., 2006). However, another study was unable to show any changes in predicted end-dive Pn 2 when the ascent rate was varied from 1 m sec-1 to 20 m sec-1 in beaked whales (Zimmer and tyack, 2007). a reduction in the ascent rate is only useful when the tissue Pn 2 > ambient Pn 2 and at any other time a diving animal should attempt to return to the surface as fast as possible as this will reduce additional n 2 up- take. in other words, the effect of ascent rate is case specific and depends on the actual blood and tissue Pn 2 level. The short and shallow surface dives that are observed be- tween deep dives or at the end of extended dive bouts could be a behavioural phenomenon that helps reduce supersaturation and bubble formation while gas exchange and inert gas removal continues (Fahlman et al., 2007). it must be pointed out that to be protective, these decompression dives have to be to a depth that allows removal of n 2 and therefore not deeper than the current tissue and mixed venous Pn 2 . it is therefore interesting to note than in some diving animals these decompression dives are deepest at the end of a dive bout and subsequently become more shallow (Fig. 4, Fahlman et al., 2007). cONcluSION aNd fuTurE rESEarch if diving animals use either behavioural and/or physiologi- cal means to reduce the inert gas burden, how do they know that they are at risk? Can they sense low levels of bubbles and does this affect physiology and behaviour? to better understand how diving mammals avoid elevated n 2 levels, research efforts must improve our understanding of gas exchange during breath-hold diving. This is not only an interesting physiological problem but an important question in clinical pulmonary medicine, because recruiting a collapsed human lung may represent a severe clini- cal problem. Thus, clinical medicine may greatly benefit if we can understand how marine mammals are able to repeatedly collapse and recruit their alveoli during each deep dive. in ad- dition, if marine mammals live with elevated blood and tissue n 2 levels, do they have any specialized adaptations that reduce DCs risk? such information may result in novel methods to reduce DCs risk in humans. Acknowledgments support for this work was made by a grant to aF from the of- fice of naval Research (onR award no. n00014-07-1-1098). lITEraTurE cITEd andrews, R. D., D. R. Jones, J. D. Williams, P. H. Thorson, g. W. oliver, D. P. Costa, and B. J. le Boeuf. 1997. Heart rates of northern elephant seals diving at sea and resting on the beach. Journal of Experimental Biology 200:2083?2095. Banish, l. D., and W. g. gilmartin. 1992. Pathological findings in the Hawai- ian monk seal. Journal of Wildlife Disease 28:428?434. Behnke, a. R., R. m. Thomson, and l. a. shaw. 1935. The rate of elimination of dissolved nitrogen in man in relation to the fat and water content of the body. American Journal of Physiology 114:137?146. Berghage, t. E., t. D. David, and C. v. Dyson. 1979. species differences in decompression. Undersea Biomedical Research 6:1?13. Bostrom, B. l., a. Fahlman, and D. R. Jones. 2008. tracheal compression de- lays alveolar collapse during deep diving in marine mammals. Respiratory Physiology and Neurobiology 161: 298?305. Brown, R. W., and J. P. Butler. 2000. The absolute necessity of chest-wall col- lapse during diving in breath-hold diving mammals. Aquatic Mammals 26:26?32. Butler, P. J., and D. R. Jones. 1997. Physiology of diving birds and mammals. Physiological Reviews 77:837?899. Cox, t. m., t. Ragen, a. Read, E. vos, R. Baird, k. Balcomb, J. Barlow, J. Caldwell, t. Cranford, l. Crum, l. amico, a. D. D?spain, g. Fernandez, a., Finneran, J. gentry, R. gerth, W. gulland, F. Hildebrand, J. Houser, D. Hullar, t. Jepson, P. D. ketten, D. macleod, C. D. miller, P. moore, s. mountain, D. C. Palka, D. Ponganis, P. Rommel, s. Rowles, s. taylor, B. tyack, P. Wartzok, D. gisiner, R. mead, J. Benner. 2006. Understand- ing the impacts of anthropogenic sound on Beaked Whales. Journal of Cetacean Research and Management 7(3):177?187. Denison, D. m., and g. l. kooyman. 1973. The structure and function of the small airways in pinniped and sea otter lungs. Respiration Physiology 17:1?10. Elsner, R. 1965. Heart rate response in forced versus trained experimen- tal dives in pinnipeds. in Essays in Marine Physiology, pp. 24?29. oslo: Universitetsforlaget. Fahlman, a., s. k. Hooker, a. olszowka, B. l. Bostrom, and D. R. Jones. 2009. Estimating the effect of lung collapse and pulmonary shunt on gas ex- change during breath-hold diving: the scholander and kooyman legacy. Respiratory Physiology and Neurobiology 165:28?39. Fahlman, a. and s. R. kayar. 2003. Probabilistic modeling for estimating gas kinetics and decompression sickness risk in pigs during H 2 biochemical decompression. Bulletin of Mathematical Biology 65:747?66. Fahlman, a., a. olszowka, B. Bostrom, and D. R. Jones. 2006. Deep diving mammals: Dive behavior and circulatory adjustments contribute to bends avoidance. Respiration Physiology and Neurobiology 153:66?77. Fahlman, a., a. schmidt, D. R. Jones, B. l. Bostrom, and Y. Handrich. 2007. to what extent does n 2 limit dive performance in king penguins? Journal of Experimental Biology 210:3344-3355. Fahlman, a., P. tikuisis, J. F. Himm, P. k. Weathersby, and s. R. kayar. 2001. FIgURe 4. ambient pressure (p amb , aTa) and estimated mixed venous supersaturation ({pn 2venous - pn 2ambient }?Pn 2ambient -1 for a king penguin performing short and shallow dives (black dots) or resting at the surface (grey dots) during an interbout interval. Figure is modified from Fahlman et al. (2007). FaHlman: HoW Do maRinE mammals avoiD DCs? ? 135 on the likelihood of decompression sickness during H 2 biochemical de- compression in pigs. Journal of Applied Physiology 91:2720?2729. Falke, k. J., R .D. Hill, J. Qvist, R. C. schneider, m. guppy, g. C. liggins, P. W. Hochachka, R. E. Elliot, and W. m. Zapol. 1985. seal lung col- lapse during free diving: evidence from arterial nitrogen tensions. Science 229:556?557. Flynn, E. t. J., and C. J. lambertsen. 1971. Calibration of inert gas exchange in the mouse. in lambertsen, C. J., ed. Underwater Physiology. academic Press, new York, pp. 179?191. Froget, g., P. J. Butler, Y. Handrich, and a. J. Woakes. 2001. Heart rate as an indicator of oxygen consumption: influence of body condition in the king penguin. Journal of Experimental Biology 204:2133?2144. Hooker, s. k., and R. W. Baird. 1999. Deep-diving behaviour of the northern bottlenose whale, Hyperoodon ampullatus (Cetacea: Ziphiidae). Proceed- ings of the Royal Society of London B 266:671?676. Hooker, s. k, R. W. Baird, and a. Fahlman. 2009. Could beaked whales get the bends?: Effect of diving behavior and physiology on modelled gas exchange for three species: Ziphius cavirostris, Mesoplodon densirostris and Hyper- oodon ampullatus. Respiratory Physiology and Neurobiology. 167(3):235? 246. http://dx.doi.org/10.1016/j.resp.2009.04.023 Jepson, P. D., m. arbelo, R. Deaville, i. a. P. Patterson, P. Castro, J. R. Baker, E. Degollada, H. m. Ross, P. Herraez, a. m. Pocknell, Rodriguez, F. Howie, F. E. Espinosa, a. Reid, R. J. Jaber, J. R. martin, v. Cunningham, a. a. Fernandez. 2003. gas-bubble lesions in stranded cetaceans. Nature 425: 575?576. kayar, s. R., E. o. aukhert, m. J. axley, l. D. Homer, and a. l. Harabin. 1997. lower decompression sickness risk in rats by intravenous injection of foreign protein. Journal of Undersea and Hyperbaric Medicine 24:329?335. kooyman, g. l. 1973. Respiratory adaptations in marine mammals. American Zoologist 13:457?468. kooyman, g. l., and W. B. Campbell. 1972. Heart rates in freely diving Weddell seals, Leptonychotes weddelli. Comparative Journal of Biochemistry and Physiology 43a:31?36. kooyman, g. l., J. P. schroeder, D. m. Denison, D. D. Hammond, J. J. Wright, and W. P. Bergman. 1972. Blood nitrogen tensions of seals during simu- lated deep dives. American Journal of Physiology 223:1016?1020. kooyman, g. l., and E. E. sinnett. 1982. Pulmonary shunts in harbor seals and sea lions during simulated dives to depth. Physiological Zoology 55:105?111. leith, D. E. 1989. adaptations to deep breath-hold diving: respiratory and circu- latory mechanics. Undersea Biomedical Research 16:345?354. Ponganis, P. J., g. l. kooyman, R. vanDam, and Y. le maho. 1999. Physiologi- cal responses of king penguins during simulated diving to 136 m depth. Journal of Experimental Biology 202:2819?2822. Ridgway, s. H., and R. Howard. 1979). Dolphin lung collapse and intramuscu- lar circulation during free diving: evidence from nitrogen washout. Science 206:1182?1183. Ridgway, s. H., B. l. scronce, and J. kanwisher. 1969. Respiration and deep diving in the bottlenose porpoise. Science 166:1651?1654. sato, k., J. B. Charrassin, C. a. Bost, and Y. naito. 2004. Why do macaroni penguins choose shallow body angles that result in longer descent and as- cent durations? Journal of Experimental Biology 207:4057?4065. scholander, P. F. 1940. Experimental investigations on the respiratory function in diving mammals and birds. Hvalr?dets Skrifter 22:1?131. stephenson, R. 2005. a theoretical analysis of diving performance in the Wed- dell seal (Leptonychotes weddelli). Physiological and Biochemical Zoology 78:782?800. tyack, P. l., m. Johnson, n. a. soto, a. sturlese, and P. t. madsen. 2006. Extreme diving of beaked whales. Journal of Experimental Biology 209:4238?4253. Zimmer, W. m. x., and P. l. tyack. 2007. Repetitive shallow dives pose de- compression risk in deep-diving beaked whales. Marine Mammal Science 23:888?925. INTrOducTION inadequate decompression can cause acute neurological decompression sickness that may involve speech, hearing and visual deficits, motor dysfunction, and paralysis (Francis and mitchell, 2003a; 2003b). similarly, exposure to pressures greater than 2 mPa (200 msw equivalent) leads to acute High Pressure neurological syndrome (HPns) dysfunctions, which include sensory and cognitive impairment, epileptiform electrical discharges, and motor dysfunctions (tremor and convulsions) (Bennett and Rostain, 2003). since decompression and exposure to pressure can both cause significant acute neurological dysfunction, the question has long been posed as to whether long-term effects may arise (Dutka, 2003). it is clear from a number of studies that serious, acute neurological decompression sickness, especially that involv- ing paraplegias of spinal origin, often respond poorly to treatment and leave signifi- cant long-term sequelae. Cerebral dysfunction, in contrast, more often responds to treatment and as an isolated incident does not appear to have the same propensity aBSTRaCT. inadequate decompression can cause acute neurological symptoms including speech, hearing and visual deficits, motor dysfunction, and paralysis. similarly, exposure to high pressure, greater than 250 msw equivalent (2.5 mPa), also leads to a variety of central nervous system dys- functions including sensory and cognitive impairment, epileptiform electrical discharges, and mo- tor dysfunctions (tremor and convulsions). The severity and onset pressure of symptoms is related to both the magnitude of the pressure and the speed of compression. The appearance of these acute manifestations of neurological damage has sustained the impetus for a search for long-term neuro- logical damage related to diving activity. The neurological symptoms related to decompression can easily be understood as arising through hypoxic mechanisms following intravascular bubble forma- tion occluding cerebral blood vessels. The severity and precise neurological manifestations would then be related to the extent and location of the bubble formation. long-term changes have been postulated to occur either as a result of irreversible acute damage or an accumulation of damage related to multiple, asymptomatic hypoxic events, in a manner similar to that associated, clinically, with transient ischaemic attacks. mechanism(s) that might lead to long-term neurological deficit arising as a result of exposure to pressure are more difficult to establish, largely because pressure- induced changes would be difficult to distinguish from decompression-induced changes. There is a theoretical possibility that pressure-induced potentiation of nmDa-sensitive glutamate receptor activity, which has been observed experimentally, might lead to cell death through calcium-induced excitotoxic mechanisms. However, it has not been established that long-term neurological deficits are caused by diving, through any mechanism. more light may be shed on this question by mod- ern molecular imaging technologies, high-field magnetic resonance imaging (mRi) coupled with functional-mRi or positron emission tomography. nevertheless, there is no clear evidence, to date, that diving, in the absence of serious acute damage, leads to long-term neurological dysfunction. Cardiff University, Wales Research and Diagnostic Positron Emission Tomography Imaging Centre, School of Medicine, Heath Park, Cardiff, CF14 4XN, Wales, UK. Does Diving Destroy the Brain? Stephen Daniels 138 ? tHE FUtURE oF Diving for long-term effects. However, with both spinal and cerebral effects multiple instances of damage are frequently associated with residual and long-term damage (Dutka, 2003). an im- portant question is whether long-term neurological deficits accrue in the absence of acute decompression sickness, neu- rological or other. The question as to whether exposure to pressure is able to cause long-term neurological deficits is much more difficult to resolve. all exposures to raised ambient pressure will at some point involve decompression back to atmospheric pressure. Furthermore, since practical decompression will always involve some degree of bubble formation it will be extremely difficult to resolve whether any residual central nervous system dysfunc- tion arose from the exposure to pressure or the subsequent decompression. to attempt an answer to the questions as to whether de- compression, in the absence of acute decompression sickness, or exposure to pressure per se may lead to long-term neurological dysfunction, first the mechanisms via which such damage may occur and second the epidemiological evidence for damage will be considered. mEchaNISmS fOr lONg-TErm cNS dySfuNcTION neurologicAl decomPression sickness Decompression schedules that are acceptable in practice are invariably associated with some degree of bubble formation (Daniels, 1984; Daniels et al., 1989). in general, bubbles give rise to pathophysiological effects either indirectly, through acti- vation of biochemical pathways, directly as a result of hypoxic mechanisms (because blood vessel blockage gives rise to areas of ischaemic tissue) or through direct mechanical stimulation, particularly of nerve pathways (Francis and mitchell, 2003a). The biochemical pathways stimulated are largely pro-inflamma- tory (Francis and mitchell, 2003a) and as such may exacerbate, prolong, and intensify pain sensations from direct mechanical stimulation of nerve fibres via release of mediators of nocicep- tive neurotransmission such as bradykinin, 5-hydroxytrypta- mine, prostaglandins, and nerve growth factors. However, it is the activation of hypoxic mechanisms (Fig. 1) that have the potential to cause permanent loss of neurons, with an associated accumulation of neurological deficit. it is an interesting feature of neurological decompression sickness that it is most commonly the spinal cord that is af- fected rather than the brain. This is in contrast to other embolic causes of central nervous system injury (e.g., cardiac surgical patients) where it is overwhelmingly the brain that is injured, not the spinal cord (Francis and mitchell, 2003a). Furthermore, pathologically, the damage seen in spinal decompression sick- ness comprises punctate white matter (myelinated nerve fibres) haemorrhages and necrosis whereas ischaemia is associated with grey matter (nerve cell bodies) damage. a number of hypotheses have been proposed to account for these finding. First there is the embolic-ischaemic hypothesis, which appears to be an ac- ceptable explanation for arterial gas embolism but bubble for- mation in arterial blood is extremely unlikely and, therefore, emboli would have to appear in the arterial circulation follow- ing passage through a patent foramen ovale or through pulmo- nary shunting. although both mechanisms are known to occur, they are relatively rare and are normally thought most signifi- cant with severe decompression stress, usually associated with accidental or emergency decompressions (Francis and mitchell, 2003a). nongaseous emboli, arising from vascular agglutina- tion or coagulation initiated via biochemical cascades induced by bubbles, do occur, but since recompression therapy is nor- mally effective at relieving symptoms, they would seem to be an unlikely direct cause of much damage. However, such emboli may well contribute to areas of ischaemia and to the initiation of hypoxic mechanisms outlined in Figure 1. such mechanisms may underlie those cases of neurological decompression sick- ness that are unresponsive to recompression therapy. Finally, the autochthonous bubble hypothesis has been proposed that links central nervous system damage to extravascular bubble forma- tion within the white matter, which is rich in lipid and with a low vascularisation and, therefore, would be an ideal tissue for bubble formation. such bubbles might cause permanent neuro- nal loss through axonal destruction, which leads to cell death, or alternatively, stimulate biochemical processes that lead to necrosis. although the above mechanisms have usually been invoked to explain the acute symptoms of neurological decompression sickness, it is clear that by linking bubble formation to hypoxia, either directly or indirectly, these mechanisms may lead to per- manent neuronal cell death that may be acutely asymptomatic. FIgURe 1. pathophysiology of bubbles. The biochemical process- es that may be activated by in vivo bubbles acting either directly or indirectly and that may lead to cell death. any one stress induced through bubble-induced processes may lead to cell death from any of the three fundamental mechanisms: apoptosis, autophagic, or necrotic cell death. (adapted from galluzzi et al., 2007) DaniEls: DoEs Diving DEstRoY tHE BRain? ? 139 48 matched controls. They used both t1 and t2 weighted im- ages and fluid-attenuated, inversion-recovery weighting, which is particularly suitable for revealing cerebral lesions. They found no difference in the distribution of lesions between the divers and the controls. slosman et al. (2004) examined 215 healthy recreational divers measuring global cerebral blood flow (CBF), using 133xe single photon emission computed tomography (sPECt), a bat- tery of psychometric and neuropsychological tests, and evaluat- ing diving activity by questionnaire. Their conclusions were that diving may have long-term effects on cognitive function when carried out in cold water, with more than 100 dives per year, and to depths below 40 m. taylor et al. (2006) studied the neuropsychological test performance of 102 divers with a complaint of ?forgetfulness or loss of concentration,? 100 nonforgetful divers, and 100 non- forgetful nondivers. verbal memory, current intelligence, and sustained attention were worse for the divers with a complaint than for the nonforgetful divers or the nonforgetful nondivers. tests of memory but not of executive function were different for the divers with complaints compared to the two control groups. in particular, mixed-gas bounce diving and surface-oxygen de- compression diving, but not other techniques, were associated with memory deficit. However, their overall conclusion was that the relationships between diving experience and neuropsycho- logical test performance were small and associated only with diving techniques used in the offshore oil and gas industry. irgens et al. (2007) investigated by questionnaire the im- pact of decompression sickness and diving exposure on health- related quality of life (HRQol) in 375 norwegian north sea divers registered before 1990. They recorded demographic data, relevant health data and data on diving education, history of decompression sickness, and medical outcomes study short- Form general Health survey (sF-36; Ware et al., 1994) in 230 divers. They concluded that HRQol was reduced in these However, since neurons are not readily regenerated, accumula- tion of such damage over time may lead to a long-term neuro- logical deficit. ePidemiology a variety of studies, in the main cross-sectional epidemio- logical studies, have been conducted in an attempt to identify a link between diving and long-term health consequences, par- ticularly neurological deficits. six such studies are summarised in table 1 where the number of divers, their experience, the essential methodology, and main findings are summarised. The methodologies range from use of magnetic resonance imaging (mRi) in relatively small numbers of subjects, looking spe- cifically for evidence of diving related white matter lesions, to health-related quality of life questionnaires issued to much larg- er groups with comparisons made either to population?s norms or to matched controls. The earliest study of knauth et al. (1997) revealed white matter lesions in 11 out of a group of 87 recreational divers, each with a minimum of 160 dives experience. seven of the af- fected divers showed no evidence of left-to-right cardiac shunt and only showed a few lesions. in contrast, 4 of the 11 divers had evidence of left-to-right cardiac shunt and exhibited mul- tiple lesions. However, a further 21 divers, who also showed evi- dence of left-to-right cardiac shunt, had no detectable lesions. although suggesting a case for neuronal damage in the absence of symptoms and evidence that left-to-right cardiac shunting may predispose towards neurological damage, these results should be interpreted with caution. a cross-sectional study can- not identify cause and these changes may have arisen from non- diving related events. There are also no matched controls with a similar proportion of cardiac shunt for comparison. Hutzelmann et al. (2000) examined 59 experienced, elderly but healthy divers and compared their mRi scans to those of TaBle 1. Outcome from 6 cross-sectional epidemiological studies investigating possible long-term neurological sequelae from decompres- sion. SCUBa: self-contained underwater breathing apparatus; mRI: magnetic resonance imaging; pFO: patent foramen ovale; DCS: decom- pression sickness; T1, T2 & FlaIR: alternative means of weighting an mRI image to reveal structural details; Fluid attenuated Inversion Recovery is particularly useful for revealing brain and spinal cord lesions; CBF: cerebral blood flow; lTNe: long term neurological effects; NpT: neuropsychological test score; hRQol: health-related quality of life score; lThe: long-term health effect; pRO: professional diver; OSW: offshore worker. Divers/Control technique Result Reference 87 (sCUBa) mRi & Doppler for PFo White matter lesions in absence knauth et al., 1997 of DCs, more with PFo 59 (experienced)/48 mRi: t1 & t2 & FlaiR no increase in white matter changes Hutzelmann et al., 2000 215 (sCUBa) global-CBF and >40 m diving and cold water slosman et al., 2004 cognitive performance may predispose ltnE 202/100 Case-controlled, Weak correlation between nPt taylor et al., 2006 neuropsychological tests and diving; only in PRo divers 375 (PRo)/norwegian norms HRQol Questionnaire significant correlation between irgens et al., 2007 neurological DCs & reduced HRQol 2958 (PRo)/2708 (osW) Postal questionnaire no evidence for ltHE Ross et al., 2007 140 ? tHE FUtURE oF Diving divers (compared to norwegian norms), having had decompres- sion sickness contributed significantly to a reduction in all sF- 36 scales, with neurological decompression sickness having the greatest impact, and cumulative diving exposure contributed to a reduced HRQol. in contrast to the study of irgens et al. (2007), Ross et al. (2007) found no evidence to suggest any major impact of div- ing on long-term health of Uk divers who started their career before 1991. They sent a questionnaire that addressed lifestyle, occupation and health status to 2,958 male professional div- ers, registered with the Uk Health and safety Executive before 1991, and to 2,708 nondiving men who had worked in the off- shore oil industry in 1990?1992. it is possible that the study of irgens et al. (2007) was influenced by the highly publicised cam- paign for the norwegian government to award compensation for work-related illness to all norwegian north sea professional divers. There may be associations between HRQol and diving activity but these studies cannot prove a causal relationship. in summary, the evidence that diving causes long-term neurological damage is very weak. in order to pursue this, a longitudinal study would afford the best hope of establishing cause and effect. However, given the very low incidence of de- compression sickness amongst professional divers in the north sea these days (1 or 2 cases per year), it is unlikely that such a study would be successful. high Pressure neurologicAl syndrome The central nervous system excitability associated with ex- posure to elevated pressures (above 2.5 mPa) is not related to epilepsy, despite the many similarities in the outward expression of symptoms, nor is it an expression of a failure in the motor control originating in the basal ganglia, cf Parkinson?s Disease. at a molecular level pressure selectively affects two ionotropic neurotransmitter receptors, one that mediates fast inhibitory neurotransmission and the other that mediates fast excitatory neurotransmission. The sensitivity of postsynaptic glycine recep- tors is progressively decreased by increasing pressure. This would lead to a reduction in inhibitory control, most pronounced in the spinal cord, but expressed throughout the central nervous system. as an illustration that effects on a single postsynaptic receptor can have profound consequences, there is a hereditary disease, hyperexplexia, in which there is a point mutation in the glycine receptor giving rise to a reduction in its sensitivity to the neurotransmitter glycine, which is expressed clinically as an increase in excitability (Eulenburg et al., 2006). in addition, the sensitivity of specific glunmDa receptors, which mediate excita- tory neurotransmission, is increased, leading to an increase in excitability. interestingly, the specific glunmDa receptor affected is that composed of R1 and R2C subunits, which are localised to the cerebellum, a region of the brain known to be important for motor control (Daniels and grossman, 2003). in addition to the effects of pressure on the glycine and glunmDa receptors, pressure also selectively affects an ion channel species, n-type Ca2+ channels that regulate calcium en- try into cells, 5Ht 2C receptors that mediate anxiety, and the adenylyl cyclase activated intracellular signalling pathway. The regulation of calcium entry into excitable cells is of crucial im- portance to excocytosis, the process of neurotransmitter release, but also if too much calcium enters the cell and cannot be se- questered into intracellular organelles then excitotoxic mecha- nisms (see below) may be initiated that will lead to cell death. The 5Ht 2C receptors are inhibited by pressure which would be expected to induce a state of anxiety that may explain some of the neurological responses divers report at pressure (Heisler et al., 2007). The intracellular signalling pathway mediated by ad- enylyl cyclase is responsible for regulating the activity of protein kinase a, which in turn regulates the activity of many neuroac- tive proteins by phosphorylation (ahn and Choe, 2009). Pres- sure increases the activity of this intracellular signalling pathway (Daniels and grossman, 2003). at the cellular level pressure depresses synaptic transmis- sion via presynaptic mechanisms, which exacerbates the effect of a reduction in postsynaptic sensitivity. to some extent the depression in synaptic transmission is partially compensated by frequency-dependent facilitation of transmission. in general, under normal circumstances a balance between excitation and inhibition exists. Pressure appears to manifest its excitable ac- tion via complex effects at both molecular and cellular levels that disturb that balance. excitotoxicity The mechanisms described above go some way towards ex- plaining the ability of pressure to cause acute hyperexcitability in the central nervous system. However, they do not provide a mechanism by which long-term neurological deficit could accu- mulate. The key to potential long-term effects lie in the increase in activity of the glunmDa receptor and the voltage sensitive n-type calcium channel that is brought about by exposure to pressure. The glunmDa receptor is permeable to Ca2+, as well as na+ and k+, and thus activation of these two species will cause an increase in intracellular Ca2+. normally, intracellular calcium concentrations are kept extremely low (< nm) because Ca2+ is actively taken up into intracellular organelles, particularly the endoplasmic reticulum, and stored. Calcium is released from these stores by the activation of inositoyl triphosphate receptors following postsynaptic metabotropic glutamate receptor activa- tion and helps initiate neurotransmitter release by exocytosis. However, if the intracellular concentration of Ca2+ exceeds the capacity of the storage system because of the action of pressure, then a positive reinforcement is initiated because Ca2+ activates transient Receptor Potential channels that are themselves per- meable to Ca2+ (Fig. 2). The high calcium concentration then leads to activation of neuronal nitric oxide synthase, the pro- duction of nitric oxide and peroxynitite radicals that cause cell death via necrotic mechanisms. in addition, calcium stimulates the release of capsases and apoptotic peptidase activating factor DaniEls: DoEs Diving DEstRoY tHE BRain? ? 141 1, which are responsible for initiating apoptotic cell death (Be- sancon et al., 2008). Exposure to pressure is, therefore, capable of stimulating processes that may lead to cell death through both necrotic and apoptotic mechanisms. an accumulation of such damage over time and with repeated exposure to pressure might be sufficient to give rise to symptoms of neurological deficit. The question is whether there is any evidence for such effects. ePidemiology aarli et al. (1985) investigated 23 professional divers before and after dives to 300 and 350 msw. twelve divers were studied during pressure exposure with neurophysiological and neuro- psychological tests. all twelve showed neurological impairment at pressure. transient neurological changes appeared in 4 divers post exposure. more persistent changes were reported in 6 div- ers who had been exposed to 2 dives within 3 months. However, the difficulty in separating the effects of pressure from those of decompression remains. ThE fuTurE ? mOlEcular ImagINg? as suggested above, the likelihood of longitudinal question- naire and/or neuropsychological testing being able to resolve the question of long-term neurological deficit related to diving is low because the incidence of decompression sickness is so low. However, it is possible to envisage longitudinal studies that em- ploy modern molecular imaging techniques (mRi, functional mRi (fmRi), Positron Emission tomography (PEt), and sin- gle Photon Emission Computed tomography (sPECt)) being used to probe changes in cerebral and spinal function during the course of a diving career. indeed, it might be possible to separate the effects of decompression from those of pressure by comparing three groups: a group of divers who predominantly dive to depths greater than 250 m, a group of divers who do not dive to depths greater than 75 m, and a matched group of nondiving controls. it might be expected that pressure effects would appear mainly in the brain whereas those associated with decompression would appear in the spinal cord. magnetic resonance imaging has, until relatively recently, been used to examine anatomical features at very high resolution (< 1 mm) using a variety of image weighting techniques such as t1, t2, and fluid attenuation inversion recovery (FlaiR). more recently, fmRi techniques have been developed to probe function in addition to structure. in the context of potential hypoxic damage, two techniques have shown great promise: blood-oxygen level dependent (BolD) fmRi (murata et al., 2006) and fluctuation-imaging using gradient-recalled echo- echo-planar imaging (gRE-EPi) (liu et al., 2007). The areas of lesion and the functional significance of the lesions can be revealed with great precision using either technique. Positron Emission tomography (PEt) uses radiotracers that incorporate an atom that spontaneously decays by emitting a positron. When this positron annihilates on contact with an electron in the surrounding medium, two antiparallel 511kev gamma rays are emitted. it is the simultaneous detection of these antiparallel gamma rays that forms the basis of PEt imaging. Positron emitting isotopes include 11C, 15o, and 18F, which have half-lives of 20 min, 2 min, and 110 min, respectively. These atoms are ones found commonly in biological molecules and therefore PEt radiotracers can be made that closely resemble endogenous compounds or small molecular weight ligands that bind to endogenous compounds, such as neurotransmit- ter receptors or cellular transporter proteins. The function of the physiological processes probed by PEt will, therefore, be minimally perturbed by these tracer molecules. as an example of the type of PEt studies that can be applied to ischaemia, nariai et al. (2003) looked at regional cerebral blood flow, using H 2 o15, before, during, and after a period of ischaemia. They also looked postischaemia at the function of adenosine a 1 receptors, using the 11C-labelled a 1 receptor antagonist [1-methyl-11C] FIgURe 2. excitotoxicity. postulated mechanisms that may be in- duced by exposure to high pressure (> 2.5 mpa): pressure induced stimulation of NmDa-sensitive ionotropic glutamate receptors (gluNmDa) leads to excessive influx of Ca2+ into the post-synaptic cell. This is augmented by the pressure-induced stimulation of voltage-sensitive Ca2+ channels (vSCC), which, in turn, leads to the activation of Ca2+-sensitive Transient Receptor potential chan- nels (TRp/SOCC Store Operated Calcium Channels), which pro- duces a further increase in Ca2+ influx. In addition, Ca2+ is released from internal stores via activation of post-synaptic metabotropic glutamate receptors (mglu 1/5 ) that activate phospholipase C (plC) leading to the production of inositoyl triphosphate (Ip 3 ), which acts on receptors on the endoplasmic reticulum (eR) to release Ca2+ from store. The excessive intracellular Ca2+ induces neuronal nitric oxide synthase (nNOS), leading to the production of nitric oxide (NO) and subsequently peroxynitrite radicals that cause necrotic cell death. excessive intracellular Ca2+ also induces apop- totic peptidase activating factor 1 (apaf-1) and capsases, which initiate apoptotic cell death. eaaT1: excitatory amino acid trans- porter 1; aa: arachadonic acid; pKC protein kinase C; gq guanine nucleotide binding protein, class q. 142 ? tHE FUtURE oF Diving 8-dicyclopropyl-methyl-1-methyl-3-propylxanthine, gaBa a re- ceptors, using the 11C-labelled benzodiazepine binding agent 11C-flumazenil, and cellular metabolic activity, using 18F-labelled 18fluorodeoxyglucose. The advantage of PEt over mRi is the sensitivity of detection of the tracers, with nm concentrations being sufficient for detection. The spatial resolution of PEt im- aging is much lower than that of mRi, and so conventionally PEt is combined with x-ray Ct, and more recently mRi, which provides anatomical location for the detected PEt activity. single photon emission computer tomography (sPECt) uses gamma-emitting isotopes such as 99mtc, 111in, and 123i with half-lives of 6 hr, 2.805 day and 13.3 hr, respectively. although the half-lives are considerable longer than those of PEt iso- topes 99mtc, and 111in cannot be incorporated into conventional biological molecules, macromolecular complexes have been constructed using peptide nucleic acid sequences or antibodies to target the tracers at the physiological processes of interest (Heckl, 2007). sPECt with 123i incorporated into an amphet- amine analogue has been used to study cardiovascular reactivity following ischaemic injury (aso et al., 2009). like PEt, the spatial resolution of sPECt is relatively poor and increasingly it is being combined with x-ray Ct or mRi to provide ana- tomical location. one potential advantage of sPECt over PEt is that unlike PEt where the gamma-ray energy is always 511 kev, with sPECt the gamma-emitting isotopes can be distin- guished on the basis of their energy. Thus, two (or more) iso- topes could be imaged, simultaneously, to reveal multiple func- tional responses. cONcluSION at present, there is no clear evidence that diving, in the absence of serious acute neurological decompression sickness, causes long-term neurological deficit. Cross-sectional popula- tion studies, whether questionnaire, neuropsychological, or functional, cannot prove a causal relationship. Furthermore, since the operational incidence of decompression sickness in well-regulated diving activity is extremely low, it is unlikely that longitudinal studies based on questionnaire or neuropsycho- logical tests can be successful either. to date it has not proved possible to separate the effects of pressure per se from those of decompression. However, it is possible that in the future, using appropriate molecular imaging techniques and well-matched groups of deep (> 250 m) divers, shallow (< 75 m) divers and non-divers, functional longitudinal studies may be able to re- veal whether repeated exposure to pressure and/or decompres- sion causes neurological deficits beyond what would be expect- ed from normal aging processes. lITEraTurE cITEd aarli, J. a, R. vaernes, a. o. Brubakk, H. nyland, H. skeidsvoll, and s. t?n- jum. 1985. Central nervous dysfunction associated with deep-sea diving. Acta Neurologica Scandinavica 71(1):2?10. ahn, s. m., and E. s. Choe. 2009. activation of group i metabotropic gluta- mate Receptors increases serine Phosphorylation of gluR1 ?-amino- 3-Hydroxy-5-methylisoxazole-4-Propionic acid Receptors in the Rat Dorsal striatum. Journal of Pharmacology and Experimental Therapeutics Doi:10.1124/jpet.108.149542. aso, k., k. ogasawara, m. sasaki, m. kobayashi, Y. suga, k. Chida, Y. ota- wara, and a. ogawa. 2009. Preoperative cerebrovascular reactivity to ac- etazolamide measured by brain perfusion sPECt predicts development of cerebral ischemic lesions caused by microemboli during carotid endar- terectomy. European Journal of Nuclear Medicine and Molecular Imaging 36(2):294?301 Bennett, P. B., and J. C. Rostain. 2003. High Pressure nervous syndrome. in Bennett and Elliott?s Physiology and Medicine of Diving. Brubakk, a.o. and t.s. neuman, eds. Pp. 323?358. Elsevier science: Edinburgh, london, new York, oxford, Philadelphia, st. louis, sydney, toronto. Besancon, E., s. guo1, J. lok, m. tymianski and E. H. lo. 2008. Beyond nmDa and amPa glutamate receptors: emerging mechanisms for ionic imbalance and cell death in stroke. Trends in Pharmacological Sciences 29(5):268?275. Daniels, s. 1984. Ultrasound monitoring of decompression procedures. Philo- sophical Transactions of the Royal Society of London B 304:153?175. Daniels, s., F. Bowser-Riley, and i. g. vlachonikolis. 1989. The relationship be- tween gas bubbles and decompression sickness. in Gas supersaturation and bubble formation. Brubakk, a. o., Hemmingsen, B. B. and g. sundnes, eds. Pp. 387?416. norway: tapir. Daniels, s., and Y. grossman. 2003. Biological effects of pressure. in Bennett and Elliott?s Physiology and Medicine of Diving. Brubakk, a. o. and t. s. neuman, eds. Pp. 265?299. Elsevier science: Edinburgh, london, new York, oxford, Philadelphia, st. louis, sydney, toronto. Dutka, a. J. 2003. long-term effects on the central nervous system. in Bennett and Elliott?s Physiology and Medicine of Diving. Brubakk, a.o. and t.s. neuman, eds. Pp. 680?699. Elsevier science: Edinburgh, london, new York, oxford, Philadelphia, st. louis, sydney, toronto. Eulenburg, v., k. Becker, J. gomeza, B. schmitt, C.m. Becker, and H. Betz. 2006. mutations within the human glYt2 (slC6a5) gene associated with hyperexplexia. Biochemical and Biophysical Research Communications 348:400?405. Francis, t. J. R., and s. J. mitchell. 2003a. Pathophysiology of decompres- sion sickness. in Bennett and Elliott?s Physiology and Medicine of Diving. Brubakk, a. o. and t. s. neuman, eds. Pp. 530?556. Elsevier science: Edinburgh, london, new York, oxford, Philadelphia, st. louis, sydney, toronto. ???. 2003b. manifestations of Decompression Disorders. in Bennett and Elliott?s Physiology and Medicine of Diving. Brubakk, a.o. and t.s. neu- man, eds. Pp. 578?599. Elsevier science: Edinburgh, london, new York, oxford, Philadelphia, st. louis, sydney, toronto. galluzzi, l., m. C. maiuri, i. vitale, H. Zischka, m. Castedo, l. Zitvogel, and g. kroemer. 2007. Cell Death and Differentiation 14:1237?1266. Heckl, s. 2007. Future contrast agents for molecular imaging in stroke. Current Medicinal Chemistry 14:1713?1728. Heisler, l. k., l. Zhou, P. Bajwa, J. Hsu and l. H. tecott. 2007. serotonin 5-Ht2C receptors regulate anxiety-like. Behaviour Genes, Brain and Be- havior, 6:491?496. Hutzelmann, a., k. tetzlaff, m. Reuter, s. m?ller-H?lsbeck, and m. Heller. 2000. Does diving damage the brain? mR control study of divers? central nervous system. Acta Radiologica 41:18?21. irgens, a., m. gr?nning, k. troland, E. sundal, H. nyland, and E. Thors- en. 2007. Reduced health-related quality of life in former north sea divers is associated with decompression sickness. Occupational Medicine 57:349?354. knauth, m., s. Ries, s. Pohimann, t. kerby, m. Forsting, m. Daffertshofer, m. Hennerici, and k. sartor. 1997. Cohort study of multiple brain lesions in sport divers: role of a patent foramen ovale. British Medical Journal 314:701?705. liu, Y., H. D?arceuil, J. He, m. Duggan, g. gonzalez, J. Pryor and a. de DaniEls: DoEs Diving DEstRoY tHE BRain? ? 143 Crespigny. 2007. mRi of spontaneous Fluctuations after acute Cerebral ischemia in nonhuman Primates. Journal of Magnetic Resonance Imaging 26:1112?1116. murata, Y., k. sakatani, t. Hoshino, n. Fujiwara, t. kano, s. nakamura, and Y. katayama. Effects of Cerebral ischemia on Evoked Cerebral Blood oxygenation Responses and BolD Contrast Functional mRi in stroke Patients. Stroke 37:2514?2520. nariai, t., Y. shimada, k. ishiwata, t. nagaoka, J. shimada, t. kuroiwa, k.i. ono, k. ohno, k. Hirakawa, and m. senda. 2003. PEt imaging of ad- enosine a 1 Receptors with 11C-mPDx as an indicator of severe Cerebral ischemic insult. Journal of Nuclear Medicine 44:1839?1844. Ross, J. a. s., J. i. macdiarmid, l. m. osman, s. J. Watt, D. J. godden, and a. lawson. 2007. Health status of professional divers and offshore oil indus- try workers. Occupational Medicine 57:254?261. slosman, D. o., s. de Ribaupierre, C. Chicherio, C. ludwig, m. l. montan- don, m. allaoua, l. genton, C. Pichard, a. grousset, E. mayer, J. m. annoni, and a. de Ribaupierre. 2004. negative neurofunctional effects of frequency, depth and environment in recreational scuba diving: the ge- neva ?memory dive? study. British Journal of Sports Medicine 38:108?114. taylor, C. l., J. i. macdiarmid, J. a. s. Ross, l. m. osman, s. J. Watt, W. adie, J. R. Crawford, and a. lawson. 2006. objective neuropsychological test performance of professional divers reporting a subjective complaint of ?forgetfulness or loss of concentration. Scandinavian Journal of Work, Environment, and Health 32(4):310?317. Ware, J. J., k. snow and m. kosinski. 1994. sF-36 Health survey: manual and interpretation guide. The Health institute, new England medical Center: Boston, mass. dIvINg ExpOSurE the diving exposure is multifactorial with several exposure factors known to have effects on the lung, the central nervous system (Cns), and the musculoskeletal system (mss). some effects are a physiological adaptation to the environment, but some effects are harmful and mediated by inflammatory responses causing diving- related disease and having a potential for long-term residual effects. The basic physical components of the diving exposure are pressure, time, and gas mixture. The cumulative hyperoxic exposure is a function of the partial pressure of oxy- gen of the gas mixture and time. The decompression stress is a function of the partial pressure of the inert gases of the gas mixture, time, and the first derivative of pressure with respect to time. Hyperoxia and decompression stress with formation of free gas have well-known effects on the lung and on the Cns and mss causing decompression sickness and dysbaric osteonecrosis. in some way, any diving-related exposure factor will be related to these three basic physical components. There are also other factors associated with diving such as pollutants of the breath- ing gas mixture, the diving habitat, or the water of chemical, bacteriological, or even radioactive nature. The components of the diving exposure can be easily monitored and controlled in experimental diving and the effects of each component can be studied when controlling for the others. in that way hyperoxia, decompression stress as evaluated by the forma- tion of free gas, submersion, and breathing resistance have all been shown to have acute and subacute effects on the lung. When it comes to long-term effects of diving, assessment of cumulative diving expo- sure over a diving career of many years is much more difficult. all necessary information Institute of Medicine, University of Bergen, Haukeland University Hospital, N-5021 Bergen, Norway. Effects of Diving on the lung Einar Thorsen aBSTRaCT. Hyperoxia and decompression stress with formation of free gas have well-known ef- fects on the lung. other factors associated with diving are pollutants of the breathing gas mixture, the diving habitat, or the water of chemical, bacteriological, or even radioactive nature. Hyperoxia, decompression stress as evaluated by the formation of free gas, submersion, and breathing resistance have all been shown to have acute and subacute effects on the lung. in analyses of long-term term effects of diving over an entire career, the inclusion of different factors or estimates of cumulative hyperoxic and hyperbaric exposure and decompression stress in the models can result in significant effects being missed or, if there is any effect, it may be ascribed to a factor that is associated with the causal factor and not to the causal factor itself. Five independent cross-sectional studies have confirmed the finding of a reduction in small airways conductance and its relation to diving and exposure to hyperoxia. several studies of the effects of single dives have shown changes in lung func- tion immediately after a dive. at least for the effect of hyperoxia, a longitudinal follow-up for three years has demonstrated a persisting residual effect on lung function. 146 ? tHE FUtURE oF Diving is expected to be in the divers? professional logbooks. However, the logbooks are not easily retrievable and recreational diving is usually not registered. Even in the prospective longitudinal fol- low-up studies of apprentice divers? logbook data were retriev- able in only about 25% of the divers (skogstad et al., 2002). in the study of Uk divers (HsE, 2004), the quality of ques- tionnaire data was compared with logbook data in the fraction of subjects having a complete log. There was good agreement concerning the number of years of professional diving and the number of days in saturation, but less agreement concerning the number of air and mixed-gas bounce dives. The recall bias can be considerable and the estimate of number of dives and diving-related events may be influenced by the purpose of the study, the selection criteria for inclusion in the study, cultural, and geographical factors. However, by having an estimate of the number of dives, this number includes a measure of the cumu- lative exposure to diving because the exposure factors in some way are related to time. in analyses of long-term term effects of diving, the inclu- sion of different factors or estimates of cumulative hyperoxic and hyperbaric exposure and decompression stress in the mod- els introduces a problem of co-linearity between the exposure factors derived from the basic physical factors. The result can be that significant effects are missed or, if there is any effect, it may be ascribed to a factor that is associated with the causal factor and not to the causal factor itself. For epidemiological studies however, it may be sufficient to relate any effects to cumulative diving exposure and leave the causal relationships to studies of experimental dives. EffEcTS ON ThE luNg at the international consensus conference on long-term ef- fects of diving at god?ysund in 1993, three studies were pre- sented showing a reduction in small airways conductance in divers that was related to cumulative diving exposure (Hope, 1994). The three studies had been conducted independently of each other in the United kingdom, in the former soviet Repub- lic, and in norway. a later german study (tetzlaff et al., 1998) and more recently an israeli study (adir et al., 2005) have con- firmed the finding of a reduction in small airways conductance and its relation to diving and exposure to hyperoxia. all these studies were cross-sectional in their design. Prospective longitudinal follow-up studies were initiated in norway after the god?ysund conference. a cohort of ap- prentice divers was followed up for 6 years from the start of their professional diving career at the basic training course at the norwegian Professional Diving school in oslo (skogstad et al., 2002). more than 80% of the subjects completed the 6-year follow-up examination and the results confirmed the findings of the cross-sectional studies. a group of apprentice policemen followed up for 6 years served as control. several studies of the effects of single dives, including deep and shallow saturation dives and shallow, air bounce dives, have shown that exposure to hyperoxia and decompression stress as evaluated by the incidence of venous gas microembolisms, con- tribute to changes in lung function immediately after a dive. at least for the effect of hyperoxia, a longitudinal follow-up for 3 years has demonstrated a persisting residual effect on lung func- tion (Thorsen and kambestad, 1995). cONcluSION The above-mentioned effects have been demonstrated in active divers. Whether the effects persist into retirement is still not resolved. in the Haukeland University Hospital study the prevalence of spirometric airway obstruction in retired divers was significantly higher than in the general population, but in the study of Uk divers there was no difference between the div- ers and a control group of offshore workers (HsE, 2004). a higher prevalence means that a larger fraction of the population has, or is at risk of, clinical chronic obstructive lung disease. Acknowledgments This paper is a compilation of two previously published papers: ?long-term Health Effects on the lung? and ?assess- ment of Diving Exposure,? both published in Long-Term Health Effects of Diving, ed. a. Hope and J. Risberg, the god?ysund 1993 Consensus Conference Revisited (Bergen, norway: nUi, 2006). Reprinted with permission. lITEraTurE cITEd adir, Y., a. shupak, a. laor, and D. Weiler-Ravell. 2005. large lungs in divers: natural selection or a training effect? Chest 128:224?228. Health and safety Executive. 2004. Coordinated investigation into the possible long-term health effects of diving. Research Report 230. aberdeen: Uni- versity of aberdeen for the Health and safety Executive. Hope, a., t. lund, D. H. Elliott, m. Halsey, H. Wiig. 1994. long-term health effects of diving. an international consensus conference. god?ysund 6-10 June, 1993. John grieg forlag a/s, Bergen, pp. 387?391. skogstad, m., E. Thorsen, t. Haldorsen, and H. kjuus. 2002. lung function over six years among professional divers. Occupational and Environmental Medicine 59:629?633. tetzlaff, k., l. Friege, m. Reuter, J. Haber, t. mutzbauer, and B. neubauer. 1998. Expiratory flow limitation in compressed air divers and oxygen div- ers. European Respiratory Journal 12:895?899. Thorsen, E., and B. k. kambestad. 1995. Persistent small-airways dysfunction after exposure to hyperoxia. Journal of Applied Physiology 78(4):1421?1424. INTrOducTION Breath-hold diving has been utilized as a means of harvesting food from the sea floor for many years, with descriptions dating back thousands of years (nukada, 1965). one form of food harvesting is by spear fishing and the first official world record in deep diving on one breath was set (with spear in hand) by diving to 30 m in 1949 (although descriptions exist of deeper dives predating this event). For many decades a few divers competed with each other setting deeper and deeper depths, registering them as world records in deep breath-hold diving. This ?style? of div- ing was later characterized as ?no limits,? i.e., the diver can use a weight to descend faster and a lift bag to ascend faster. some examples of the most recognized deep divers are Enzo maiorca (first to 50 m), Jacques mayol (first to 100 m), and Umberto Pelizzari (first to 150 m). The current no-limits record is 214m (Fig. 1) that was recently set by Herbert nitsch (who was also the first diver to reach 200 m). During the last decade, the sport has evolved to the point where athletes can compete in different forms of deep diving apart from the no-limits dives; current disciplines for competition include swimming down and up with or without fins, swimming as far as possible in a pool while holding one?s breath, and floating motionless in a pool, again for as long as possible on one breath. two of the disciplines, ?no limits? and ?variable weight,? are usually held as special events organized for individual record attempts rather than competitions. The association inter- nationale pour le D?veloppement de l?apn?e (aiDa) is the current governing body that aBSTRaCT. The limits of breath-hold diving are discussed regarding both duration and depth. For duration, the calculations presented explain a possible, theoretical breath-hold duration (according to known physiology) of 10 minutes 24 seconds for comparison with the current world record of 10 minutes 12 seconds. one technique that makes the current record possible is glossopharyngeal insufflation, by which the diver may fill the lungs with an extra 3-4 liters of air, increasing oxygen stores. in terms of depth, the record has stretched from 30 m in 1949, to the current record of 214 m diving with ?no limits.? The depth limit is not yet known (the diver surfaced from 214 m in full health) but various factors limiting deep diving are discussed. The compression of gas, according to Boyle?s law, may cause a range of barotraumas associated with the descent, including problems with equalization of middle ear, pulmonary oedema, and haemoptysis. The increase in partial pressure of nitrogen may cause nitrogen narcosis with a risk of the diver becoming incapacitated at depth, as well as gas supersaturation in tissues leading to decompression sickness on, or after, the ascent. While divers have reached depths of 113 m by physically swimming down and up, the ?no limits? divers use assisted descent and ascent. to reduce problems with compression, pulmonary gas stores can be increased by glossopharyngeal insufflation, and sinuses may be filled with (sea) water as a means of equalization. For the elite divers, the amount of oxygen they can store seems to be the limit for duration, while the problems of compression and nitrogen partial pressure are the most likely limiting factors for depth. Karolinska Institutet, Department of Physiology and Pharmacology, Berzelius vag 13, 17177 Stockholm, Sweden. The limits of Breath-Hold Diving Peter Lindholm 148 ? tHE FUtURE oF Diving verifies records and organizes international competitions (www. aida-international.org). What are the limits of breath holding? The world records give an indication of the limits that may be achieved by se- lect (perhaps genetically predisposed) individuals who devote themselves to extensive training in order to push their physi- ological boundaries. From the physiologist?s point of view, two important limits can be distinguished: the first is the limit set by the onset of unconsciousness from the gradually develop- ing hypoxia during breath holding (lindholm, 2007; lindholm and lundgren, 2006). The other is the voluntary breakpoint of a breath hold in a subject. The breakpoint is highly variable due to motivation, tolerance, and conditions (e.g., hyperventilation, exercise, rest) (lindholm and lundgren, 2009). There is a large discrepancy between the limit of consciousness and the volun- tary breakpoint, which may vary from the edge of unconscious- ness to a breath-hold duration of a few seconds where changes in gas stores can hardly be detected in arterial blood. This factor should be considered when evaluating the literature on breath- hold diving, as many studies have focused on voluntary breath holding without concurrent measurements of end-tidal or arte- rial gas partial pressures. it is possible to hold one?s breath until unconsciousness, even though most individuals would not do so voluntarily and most would only reach this limit under duress, for example, in a drowning scenario. The point of unconsciousness is consider- ably easier to accomplish with prior hyperventilation and exer- cise during the breath hold, as evidenced by deaths in swimming pools (Craig, 1961). When diving at great depth, the changes in pressure further aggravate the risk of unconsciousness during ascent where the partial pressure of oxygen falls rapidly (ascent blackout) (lindholm, 2007; lindholm and lundgren, 2009). The current ?limits? of breath-hold diving in the form of world records are listed in table 1. dIvE duraTION How is a dive duration of 10 minutes and 12 seconds pos- sible? Can it be explained by current physiological knowledge? For elite competitors, the limit of their breath hold is loss of con- sciousness (loC) or loss of motor control (lmC), i.e., severe hypoxia. We can calculate the oxygen stores at the start of the breath hold and divide them by the metabolism to come up with a reasonable suggestion for the maximum duration approachable by man (lindholm, 2002; lindholm and lundgren, 2006). if we assume that the limit for lmC is Pao 2 < 3-3.5 kPa (20-21 mmHg) (lindholm and lundgren, 2006), that sao 2 < 50% (~40% depending on duration and PCo 2 ), and that svo 2 would be 10-30% in different compartments, then what would the starting oxygen stores be? Consider a male diver of 70 kg body weight, with a low basal metabolic rate equivalent to that during sleep of 240 ml o 2 /min (greger and Bleich, 1996). His residual volume is two liters, vital capacity eight liters, and he has the ability to in- sufflate another three liters by glossopharyngeal insufflation. Hyperventilating prior to the dive will reduce PaCo 2 to < 3 kPa (20 mmHg), and increase Pao 2 to 18-19 kPa (135 mmHg) (Pao 2 98% to 99%), with a Pvo 2 increase ~ from 75% to circa 85-88%. This would give a total oxygen store that can be used for breath holding of: ? Desaturation of arterial blood 100 - 40 = 60 ? Desaturation of venous blood 88 - (10*1/3 + 30*2/3) = 88 - 70/3 = 88 ? 23 = 65 ? Usable Blood stores: ((0.25*60) + (0.75*65))/100*70*0.08*182*1.36 = 884 ml o 2 ? Usable lung stores: 13*0.8259*(135 - 21)/760 = 1610 ml o 2 total usable oxygen stores: 2494 ml o 2 . 2494 ml / 240 ml/min = 10 minutes and 24 seconds. it is possible that the duration of breath holding can be improved further, partly by the selection of genetically gifted in- dividuals with, for example, large lungs; the ability to insufflate large amounts of extra air and still hold their breath without in- creasing metabolism; large blood volume; and, a high hemoglo- bin concentration. other means to extend breath-hold duration would be to consume less oxygen, i.e., reduce the metabolism during breath holding or extend the limit of consciousness by increasing tolerance to hypoxia. dISTaNcE When aiming to swim as far as possible on one breath, the limiting factors are oxygen stores, the efficiency of the swim- ming technique and also the individual?s anaerobic capacity (i.e., the muscles? ability to produce energy without consuming oxygen during the breath hold). The distance will also depend on technology such as swim suits that reduce drag, and the FIgURe 1. herbert Nitsch?s 214 m record ?no limits? dive (with permission from aIDa international). linDHolm: limits oF BREatH-HolD Diving ? 149 the limit for depth in the assisted category ?no limit? is un- known, as the deepest dive made was successful and people may yet go deeper. While there are no reports of serious illnesses or mortality in any competitive events, the assisted categories have claimed both lives and caused decompression illness. The ?no limits? fatalities reported have been due to technical malfunc- tion and entanglement at depth leaving the diver at depth for too long. The extreme environment will put the diver at risk of barotrauma of descent, decompression illness, decompression sickness, nitrogen narcosis, and barotrauma of ascent. bArotrAumA oF descent When a breath-hold diver descends deeper and deeper, the gas in the body will be compressed according to Boyle?s law. This means that the rigid spaces such as the sinuses and the middle ear will require additional gas for equalization. normally, this is achieved using air from the lungs. at some depths, the diver will have descended to a point where the residual gas volume in the lungs equals the diver?s residual volume (schaefer et al., 1968). What happens then if the descent is continued? Primarily, one needs to account for the shift of blood into the thorax, which is due to immersion but further enhanced by compression. This effect will lower the residual volume of the lungs at depth in comparison to a measurement made on dry land (Craig, 1968; schaefer et al., 1968). The rigid air spaces will require gas, and the technical air space created by a facemask can be countered by not using a mask, but lenses or fluid-filled goggles with special lenses enabling enough vision for guidance. other divers counter this problem by filling the sinuses prior to diving (germonpr? et al., 2008) or during the actual dive by letting seawater passively flood sinuses and the middle ear (pers. comm. with martin stepanek and kirk krack). secondly, if the diver descends beyond the point of com- pression of the residual volume, what will then happen? When gas is compressed, atelectasis, i.e., collapse of parts of the lungs, may occur. another possibility is that the high pressure difference over the pulmonary capillaries could cause fluid filtration into the interstitial space and then into the alveoli (lindholm et al., 2008). This would result in pulmonary oedema, which may be aggravated by pulmonary hemorrhage, where erythrocytes leak into either the alveoli or the airways (Boussuges et al., 1999). bArotrAumA oF Ascent There are reports of divers surfacing from single deep dives with neurological symptoms, thereafter being treated with re- compression in a hyperbaric chamber (mcCrory et al. 2004; Thorsen et al., 2007). This is most likely due to decompression sickness (see below) but there is another speculation: compres- sion of the lungs, in combination with a blood shift to the tho- rax could possibly cause air to be trapped; upon ascent the same air could rapidly expand resulting in pulmonary barotrauma and arterial gas embolism. another factor that could aggravate material and shape of the fin(s) when used. dIvE dEpTh how deeP cAn mAn go And whAt Are the limits? During a deep dive, there is of course the limit of dura- tion, i.e., the diver has to maintain consciousness until he/she can resume breathing at the surface. The duration of the breath hold will be affected by various degrees of exercise dependent on the technique of descending to depth. The technique can vary between swimming with legs (constant weight), legs and arms (constant weight no fins), and pulling oneself with the arms along a rope (free immersion). Even the assisted disciplines require exercise in comparison to a static breath hold. in addi- tion, the changing gas pressures may affect metabolism and the distribution of gas stores during a dive. Divers reach greater depths each year. as noted above, the limit of deep diving by means of swimming will most likely be limited by duration and exercise capacity, but it may also be affected by the individual?s ability to tolerate pressure at depth. apart from the effects of changing pressure on gas stores, the effect of pressure on buoyancy (gas compression makes the div- er heaviest at depth) makes the exercise effort expended when depth diving different from a distance swim in a pool. Currently TaBle 1. World records (December 2008). DURaTION Static apnea (STa) 10 min 12 sec: tom sietas 8 min 0 sec: natalia molchanova DISTaNCe Dynamic without Fins (DNF) 213m: tom sietas/ 151 m: kathryn mcPhee Dave mullins Dynamic with Fins (DyN) 250m: alexey molchanov 214 m: natalia molchanova DepTh Constant Weight without Fins (CNF) 86m: William trubridge 60m: natalia molchanova Constant Weight (CWT) 113m: guillaume n?ry 95m: natalia molchanova Free Immersion (FIm) 108m: William trubridge 85m: natalia molchanova DepTh with assistance not competitive disciplines variable Weight (vWT) 140m: Carlos Coste 122m: tanya streeter No limit (NlT) 214m: Herbert nitsch 160m: tanya streeter 150 ? tHE FUtURE oF Diving this condition is the ability of some divers to use glossopharyn- geal insufflation to overfill their lungs with 3-4 liters of air prior to diving (lindholm and nyr?n, 2005; loring et al., 2007). The high pressure, together with a shift of perhaps a liter of blood into the thorax, could cause a relative overfilling of the lungs with air during the ascent unless the diver exhales during the last 5-10 meters to the surface. decomPression sickness Calculations based on Haldanean decompression theory show that repetitive breath-hold diving may cause supersatura- tion of tissues and put the breath-hold diver at risk of decom- pression sickness (Thorsen et al., 2007). in the case of a single deep dive, this would also be the case if gas exchange remains unaffected by compression of the lung gas stores. it is possible that compression will limit nitrogen uptake from the lung at depth, but to what extent gas exchange is affected during such a dive is not known and outside the scope of this paper. nitrogen nArcosis The diving depths reached by breath-hold divers would in- capacitate a scuba diver breathing air. The absence of reports of nitrogen narcosis may be due to the short duration at depth, as well as the repetitive nature of the dive training, i.e., the diver descends and returns according to a procedure repeated numer- ous times during training to gradually deeper and deeper depths imbuing an element of acclimatization to functioning during narcosis. There is one published account of nitrogen narcosis during a dive to 160 m (streeter, 2006). The dynamics of gas exchange would give the diver the highest Pn 2 in the nervous system during the early parts of ascent. nitrogen narcosis is also known to cause amnesia, which may complicate the description of such events. as noted above, the extent that nitrogen uptake is affected by lung compression during deep breath-hold diving is not known, but it seems plausible that nitrogen narcosis as well as decompression sickness may be limited by collapse of gas exchange areas in the lungs at great depths. The current world record dive to 214 m included a 30 sec- ond long ?decompression stop? or slow ascent during the last 10 m, included in the 4 min 24 seconds of total dive duration. it is possible that the slow ascent reduces the risk of DCs (the ascent speed equals 3 - 4 m/s during the deeper parts of the ascent to be compared with the recommended 10 m/min during regular scuba diving). The slow ascent could also give time for blood and gas redistribution in the lungs, reducing the theoretical risk of barotrauma on ascent. cONcluSION it is possible for (some) humans to hold their breath for more than 10 minutes or to dive to more than 200 m. These durations and depths can be explained: the compression of gas in the body requires that the diver has an elastic chest (muscles and ribcage) and a trachea and lungs that can collapse to a re- sidual volume of less than 0.5 liters. special lenses in fluid-filled goggles can be used instead of a mask to reduce the noncollaps- ible air spaces and glossopharyngeal insufflation may be used to increase pulmonary gas stores to over 12 liters. Water equaliza- tion of the sinuses and middle ear, while painful, is a useful adjunct. Pulmonary oedema and partial lung collapse (atelecta- ses) will likely occur. Whether a decompression stop is needed or effective to reduce the risk of decompression sickness is not known. Currently, the limits of deep breath-hold diving seem to be pulmonary barotrauma of descent causing pulmonary oede- ma and the effects of nitrogen: nitrogen narcosis incapacitating the diver at depth or causing decompression sickness on ascent. lITEraTurE cITEd Boussuges, a., C. Pinet, P. Thomas, E. Bergmann, J. m. sainty, and D. vervloet. 1999. Haemoptysis after breath-hold diving. European Respiratory Journal 13:697?699. Craig, a .B., Jr. 1961.Causes of loss of consciousness during underwater swim- ming. Journal of Applied Physiology 16:583?586. ???. 1968. Depth limits of breath hold diving (an example of Fennology). Respiratory Physiology 5:14?22. germonpr?, P., C. Balestra, and P. musimu. 2008. Passive flooding of paranasal sinuses and middle ears as a method of equalization in extreme breath-hold diving. British Journal of Sports Medicine doi:10.1136/bjsm.2007.043679. greger, R., and m. Bleich. 1996. normal values for Physiological Parameters. in greger, R., and U. Windhorst (eds.) Comprehensive Human Physiology. Berlin, Heidelberg, new York: springer. pp. 2427?2449. lindholm, P. 2002. severe hypoxemia during apnea in humans: influence of cardiovascular responses. karolinska institutet. ???. 2007. loss of motor control and/or loss of consciousness during breath- hold competitions. International Journal of Sports Medicine 28:295?299. lindholm, P., a. Ekborn, D. ?berg, and m. gennser. 2008. Pulmonary edema and hemoptysis after breath-hold diving at residual volume. Journal of Ap- plied Physiology 104:912?917. lindholm, P., and C. E. lundgren. 2006. alveolar gas composition before and after maximal breath-holds in competitive divers. Journal of Undersea and Hyperbaric Medicine 33(6):463?467. ???. 2009. The physiology and pathophysiology of human breath-hold div- ing. Journal of Applied Physiology 106:284?292. lindholm, P., and s. nyr?n. 2005. studies on inspiratory and expiratory glos- sopharyngeal breathing in breath-hold divers employing magnetic reso- nance imaging and spirometry. European Journal of Applied Physiology 94(5/6):646?651. loring, s. H., C. R. o?Donnell, J. P. Butler, P. lindholm, F. Jacobson, and m. Ferrigno. 2007. transpulmonary pressures and lung mechanics with glos- sopharyngeal insufflation and exsufflation beyond normal lung volumes in competitive breath-hold divers. Journal of Applied Physiology 102:841?846. mcCrory, P., E. matser, R. Cantu, and m. Ferrigno. 2004. sports neurology. Lancet Neurology 3:435?440. nukada, m. 1965. Historical development of the ama?s diving activities. in Rahn, H. (ed.) Physiology of Breath-Hold Diving and the Ama of Japan. Washington, DC: national academy of sciences, national Research Council. p. 25?40. schaefer, k. E., R. D. allison, J. H. Dougherty, Jr., C. R. Carey, R. Walker, F. Yost, and D. Parker. 1968. Pulmonary and circulatory adjustments deter- mining the limits of depths in breathhold diving. Science 162:1020?1023. streeter, t. 2006. nitrogen narcosis during no limits freediving world record to 160 m (525 ft). in lindholm, P., n. Pollock, and C. E. g. lundgren, linDHolm: limits oF BREatH-HolD Diving ? 151 eds. Breath-hold diving: Proceedings of the Undersea and Hyperbaric Medical Society/Divers Alert Network Workshop, Divers alert network, Durham, north Carolina. pp. 17?25. Thorsen, H. C., g. Zubieta-Calleja, and P. E. Paulev. 2007. Decompression sickness following seawater hunting using underwater scooters. Research in Sports Medicine 15:225?239. INTrOducTION the commonality between poles and tropics is their shared remoteness, re-moved from the thoughts of society that sponsors our diving research. ad-ditionally, the global economic downturn engenders an ?extreme? research fund-raising environment. This 4th international Polar Year (iPY) 2007?2009 is a global research effort to better understand the polar regions and their climatic effect on the Earth. The research completed during iPY will provide a baseline for understanding future environmental change. Previous iPYs were held in 1882?1883 and 1932?1933, and the international geophysical Year in 1957?1958. a ?vision for international Polar Year 2007?2009? was published by the national academy of sciences? Polar Research Board. as the first major iPY scientific symposium, ?smithsonian at the Poles? was con- vened may 3?4, 2007, at the smithsonian institution. Highlights included the na- tional museum of natural History?s U.s. antarctic meteorite Program, U.s. antarctic Program invertebrate Collection, and arctic Cultural studies; the office of the Under secretary for science?s U.s. antarctic Program scientific Diving Program; the smithso- nian Environmental Research Center?s antarctic Photobiology and Polar invasions Biol- ogy Programs; the national Zoological Park?s Weddell seal Energetics Project; and, the smithsonian astrophysical observatory?s south Pole antarctic submillimeter telescope/ Remote observatory Program. a second smithsonian iPY symposium was convened at the national air and space museum, november 30?December 1, 2007: ?Making aBSTRaCT. The 2007?2009 international Polar Year (iPY) is a global research effort to better un- derstand the polar regions and their climatic effect on the Earth. iPYs in 1882?1883, 1932?1933, and the international geophysical Year of 1957?1958 were the precursors to this 4th iPY, the first to use extensive scientific diving techniques for polar underwater research. The science completed above and below the ice during iPY will provide a baseline for understanding future environmental change. our research methodologies must also advance in the next 50 years, requiring some new ap- proaches to diving physiology, equipment and procedures. Polar diving is one of the more extreme- environment diving modes where parameters such as thermal protection, cold stress as DCs risk, and regulator function require special consideration. Contaminated water diving is another extreme environment of microbiological and chemical nature that requires contaminant and risk assessment. Three illustrative cases review extreme environment diving parameters: the past (smithsonian tropi- cal Research institute oil spill Project), present (U.s. antarctic Program benthic pollution studies), and future (smithsonian Environmental Research Center invasive species Project). a final extreme environment is diving at depth. an advanced scientific diving workshop recently examined diving modes and operational parameters of open-circuit mixed gas scuba, closed-circuit rebreathers, un- derwater habitats, saturation systems, surface-supplied, and bell diving. Smithsonian Institution, Office of the Under Secretary for Science, P.O. Box 37012, MRC 009, Washington, DC, 20013-7012, USA. Parameters of Extreme Environment Diving Michael A. Lang 154 ? tHE FUtURE oF Diving Science Global: Reconsidering the Social and Intellectual Implica- tions of the International Polar and Geophysical Years.? The at50 symposium, celebrating the 50th anniversary of the signing of the antarctic treaty will convene november 30?December 3, 2009, at the smithsonian national museum of natural History. Polar environmental factors that make these ecosystems unique are biological, geological, physical, and sociological in nature. studies of arctic-antarctic marine faunal comparisons and adaptations are receiving much interest. Research on car- bon dioxide levels, temperatures, greenhouse gases, and climate change centers on the polar regions because of their global importance. Human impacts and their ethical and economic considerations are integral to the conservation efforts of these pristine ecosystems. ocean circulation and heat release, sea level rise, and the extent of sea ice cover are topics that currently re- ceive much attention due to their direct impacts on the climatic effects on where human populations reside. scuba diving conducted by scientists places the trained sci- entific eye under water and provides research value and flexibil- ity that unmanned systems often do not. scientific diving is a valuable research tool that has become an integral methodology in the pursuit of scientific questions in extreme environments in polar regions, in contaminated waters, and at depth. ExTrEmE ENvIrONmENTS: pOlar dIvINg The purpose of the project using scientific diving is the ad- vancement of science. The tasks of a scientific diver are those of an observer and data gatherer. scientific divers, based on the nature of their activities, must use scientific expertise in study- ing the underwater environment, and, therefore are scientists, or scientists-in-training. The establishment of the first scientific diving safety pro- gram occurred at scripps institution of oceanography in 1951 with a two-fold purpose: a) a research support function that as- sists the diving scientist with specialized underwater equipment, advice, and diver support, and b) a risk management function that protects the safety and health of the individual scientist, and the employing organization from excessive liability expo- sure. The following risk parameters are generally accepted for scientific diving in extreme environments (adapted from low- rance, 1976). The ultimate responsibility for safety rests with the individual diver. safety is the judgment of acceptability of risk. Risk is a compound measure of probability and severity of harm to human health. There are degrees of risk, therefore, degrees of safety. Estimating risk is a scientific event (an objec- tive and probabilistic pursuit) and accepting risk is a political activity (a personal or social judgment). nothing is absolutely free of risk. The international Polar Diving Workshop (lang and sayer, 2007) was convened march 15?21, 2007 at the arctic marine laboratory in ny-?lesund, svalbard, resulting in recommenda- tions on ice-diving equipment, cold water decompression, and scientific ice diving operational procedures. The U.s. ice diving experience is derived mainly from the national science Founda- tion?s U.s. antarctic Program scientific Diving Program (lang and Robbins, 2007) whose activities since 1989 have averaged annually 700 ice dives, 30 scientific divers, 41-minute dive times, and depths shallower than 40 m. thermAl Protection The shell drysuit is conceptually like a raincoat, it keeps the diver dry and the undergarments provide warmth. The level of insulation (unaffected by depth) can be adjusted for different diving requirements and individual needs. The primary means of passive thermal protection is by trapping and stabilizing a low conductivity gas. The active heating options that exist are not a replacement for a highly effective passive system for remote po- lar scientific diving operations because an infinite energy source is not available on site that can deliver between 250 and 1000 W to the diver. Free-flooding hot water suits require an infinite power source and, in general, water is not an optimal medium to work with in - 40oC polar environments. Electrically heated and liquid-heat transport garments also require energy sources that are difficult to provide in extreme diving environments. a good compromise may be the hybrid system, a high-efficiency passive system with supplemental heating of hands and feet. The drysuit is configured to accept power penetrations and attached gloves without wrist seals. Highly efficient passive insulation, e.g., Thinsulate, incorporates the wiring harness. Electrically heated glove and sock liners (50 W system with ground-fault interrupter, screening, and temperature control) are powered by a compact lithium battery. inadequate thermal protection in extreme environments leads to progressive safety concerns including the distraction ef- fect of cold, loss of dexterity, loss of cognitive functions, non- freezing cold injuries, and ultimately life-threatening hypother- mia. moreover, respiratory heat loss is approximately 10% of the diver?s metabolic rate, influenced by atmospheric composi- tion, temperature, water vapor content, and ambient pressure. Further heat loss comes from warming the gas and the addition of water vapor. The inability to complete underwater and post- dive tasks impacts upon scientific productivity. stinton (2006) considered matrix (undergarment) composition and selection for polar diving, including low conductivity under hydrostatic loads of 35 to 210 cmH 2 o (0.5 to 3.0 psi), effectiveness when wet (hydrophobic characteristics) or immersed (flooded), con- formity to the diver?s body, low conductivity to bulk ratio, ma- terials, and construction costs. The diver?s hands and feet are the factors limiting exposure time and efficiency in passive sys- tems. Circulation to the hands can be improved by attaching Zipgloves? by DUi, inc. without a wrist seal to the drysuit. medical examination gloves worn under insulation create a va- por barrier that keeps moisture out of the insulative matrix. va- por barrier socks worn on the feet perform the same function. The use of a gas interlayer to warm the hands by occasionally lang: PaRamEtERs oF ExtREmE EnviRonmEnt Diving ? 155 flow, or massive free flow rapidly expending a diver?s air sup- ply (Clarke, 2007). Both conditions are referred to as regulator ?freeze-up.? most operational freeze-up problems in scientific diving operations result in the second stage of regulators where divers? warm, moist exhalant breath induces ice crystal forma- tion on the second stage air delivery valve. First stage regulator problems are minimal until there is enough ice build-up that mechanically interferes with the compression of an exposed spring. most first stages used for ice diving are environmentally sealed (dry bleed or silicon sealed) or of a diaphragm versus piston type, negating the freeze-up problem. once the freeze-up occurs, the air flow is shut off by closing one of the two redun- dant cylinder valves or sliding the isolation valve on the second stage shut, and the dive is aborted. Factors influencing regulator freeze-up are design (deter- mined by the manufacturer), quality control (individual regula- tor), depth (due to increased gas density), mass flow (depth and respiratory minute volume), time, and temperature. Configu- ration control is influenced by pressures of the market place, resulting in scuba regulator models typically experiencing a short half-life. Designing a regulator tolerant to freeze-up is a black art for most manufacturers without field testing sites. Even minor ?cosmetic? model changes can affect freeze-up risk. The UsaP scientific diving program initially used double-hose regulator until 1990. in 1991, single-hose sherwood maximus sRB3600 regulators were introduced to the program with the second stage modified with a heat retention plate and interme- diate pressure lowered from 145 to 125 psi. Full-face masks have been used successfully by the norwe- gian Polar institute (Hop and Pavlova, 2006) and the British antarctic survey. sayer et al. (2006) describes operation and maintenance of full-face aga masks with manual bail-out side valve connected to a rear-mount pony cylinder. Robbins (2006) cautions about the higher failure rate of full-face masks (aga, Heliox 18, and superlite 17 helmets) in the -1.86oC waters of mcmurdo sound, approximately 2oC colder than those of the arctic, antarctic Peninsula, and perennially ice-covered lakes of the antarctic Dry valleys. ExTrEmE ENvIrONmENTS: cONTamINaTEd WaTEr dIvINg scientiFic diVing: risk Versus science beneFit minimization of risk to all team members is a priority con- sideration. Contaminated water is defined as water that contains any chemical, biological, or radioactive substance that poses a chronic or acute health risk to exposed personnel. The contami- nants can be of the following nature: chemical (hydrocarbons, solvents, PCBs, heavy metals, tributyl-tin fluorides, pesticides, oxidants); biological (sewage, bacterial pathogens, viruses, pro- tozoans, microorganism toxins); or radiological. Contamination assessment is generally performed by local positioning them high inflates the gloves and reduces the hy- drostatic pressure on the hands. twice the amount of insulation covering the legs should be used for insulating the feet, without impairing their circulation, and avoiding the use of tight fins. air has been the traditional drysuit inflation gas. The use of argon, with lower conductivity than air, as a suit inflation gas has become very common in the technical diving commu- nity. However, its value has been debated following tests done by Risberg and Hope (2001) whose results from skin and core temperature monitoring during dives did not show a significant difference between air and argon. Weinberger (1989) reported a 19% improvement in suit insulation using Co 2 . argon and Co 2 have conductivities that are comparatively close relative to air (stinton, 2007). The difference in test results can possibly be explained by Risberg and Hope?s utilization of a 6-mm foam neoprene drysuit with wooly bears with a large fraction of the total insulation coming from the foam neoprene. The addition of argon into the drysuit would not change the intrinsic insula- tion of the foam, and at 9 m of depth the foam still contributes a major portion of the total suit/system insulation. Weinberg used shell drysuits that had relatively little intrinsic insulation with the majority provided by the undergarments. argon or possibly Co 2 are not a fix for a poor insulation package. They are a means to gain additional performance when there is no more room in the drysuit for additional thermal layers. The ef- fect of Co 2 use in suits at deeper depths is not known. aerogel is perhaps the most promising matrix for future thermal insulation. it is silica-based and the world?s lowest den- sity solid (99.8% air, density 3 g/l, average particle size 2-5 nm). Current aerogel undergarment production challenges are its encapsulated construction to control silica dust and the 5-fold increased cost over traditional insulation materials. cold stress And dcs risk The effect of cold on DCs risk is not fully understood. However, the diver should be kept warm throughout the dive and during the immediate post-dive period external heat ap- plication and heavy lifting should be avoided (mueller, 2007). Being only slightly cold may have the same effect on bubble grades as being severely hypothermic. long-term health effects for divers with a high proportion of cold water dives should be considered in the future. The effect and timing of exercise has recently been studied with implications for cold water diving as a stress factor (Dujic et al., 2005, 2006; gerriets et al., 2000; Jankowski et al., 2004; mekjavic et al., 2003; Ruterbusch et al., 2004, 2005; tetzlaff et al., 2001). regulAtors When scuba regulators are dived in very cold water, as un- der sea ice in polar regions, there is a chance that first or second stage regulators will malfunction due to the accumulation of ice in or around the regulator, yielding complete occlusion of air 156 ? tHE FUtURE oF Diving health offices, Environmental Protection agency, or state agen- cies that can usually provide information on potential contami- nants in local bodies of water or beaches. Bacterial contamina- tion and other common contaminants are routinely tested for in any body of water in which people swim or drink from. is the contaminant in the water column, in the sediment, or on the surface? is the contaminant water soluble? is the contaminant toxic on contact, on ingestion, on inhalation, or by proximity? are the potential effects acute, chronic, or long term? The identification of hazards should be based upon existing scientific evidence, which can show a cause and effect relation- ship between making dives in contaminated water and serious injury to the diver. The dose/response relationship would re- quire that an objective decision be made as to the degree to which different contaminants cause an observed effect. This would normally involve a study of a population of scientific divers and its known response to various exposures. We need to know the likelihood of increased injury in the diving popula- tion that is produced by the hazard. The type of contaminants, length of exposure and equipment used are all part of the ?dose? that must be evaluated. Risk is relative to a specific set of con- ditions. The analysis of potential public exposure will depend, in part, on the potential damage or benefits of the practice of contaminated water diving. it may also depend upon the effect of a variety of intervening variables such as physiological fitness, age, equipment, proper post-dive decontamination procedures and many others which may have an effect on susceptibility to contaminants. What is the specific nature of the calculated risk that divers must accept if they choose to dive in contaminated water? The description of the risk is then based upon the objec- tive evaluation of the likelihood of the occurrence of undesir- able side effects following a given ?dose? of contaminated water diving exposure. Usually, risks of 1 in 1,000,000 are consid- ered acceptable for virtually any risk. Risks at the level of 1 in 100,000 are minimal, but the severity of the injury becomes an issue. Contaminated water diving risk assessment needs a review of the actual number of exposures or the accurate size of the population as well as the number and severity of the injuries to make a reasonable assignment of risk. Without the denomi- nator, the attempts at assigning risk are speculative. u.s. nAVy wAter QuAlity cAtegories Category One Highest Contamination (level a protection): grossly contaminated with concentrated chemical or microbiological contamination. Examples include heavy fuel slicks and sewage operations. Divers should use full diving helmets with surface- supplied air and communications, vulcanized rubber suits with integrated helmet-mating collar and dry gloves with rings. The helmet should be equipped with the double exhaust-valve as- sembly design for use in contaminated water. The helmet must be used in the free-flow mode. Category Two moderate Contamination (level B protection): increased levels of both chemical and microbiological contamination are expected. Divers may use a positive pressure full-face mask and use it in the positive pressure mode. a block should be used for emergency gas switching to bail out gas in the advent of primary supply failure. a drysuit is required. Category Three Baseline Contamination (level C protection): no expecta- tion of contamination above baseline that is normal for human habitation. This category represents what most dive teams and other research divers will face during the normal course of events. Divers should wear a positive pressure full-face mask to avoid water contact with mucous membranes and mouth (unless water analysis shows contact with the mouth is an acceptable risk) and thermal protection appropriate for the diving conditions. Category Four no contamination (level D protection): This includes situ- ations where no contaminated sources are known or expected such as offshore oceanic locations or drinking water reservoirs, recreational swimming areas or areas where water quality is rou- tinely checked and no contaminants are reported. Potable water supplies or perennially ice-covered antarctic lakes can have other challenges in a reverse mode. special train- ing and equipment is often needed to safeguard these bodies of water from diver- or equipment-introduced contaminants. contAminAted wAter scientiFic diVing Procedures Dive plan considerations for risk management purposes in- clude consideration of the following: contaminant information; diver equipment suitability, monitoring, minimization of con- tact, method of decontamination, and emergency plan; tenders? handling and containment of contaminated items, method of decontamination, personal protection equipment, and emer- gency plan; decontamination and quarantine; and drysuit per- meability, tensile strength and penetration resistance, ease of donning and doffing, fit, thermal balance and heat stress (dehy- dration, body temperature, heart rate). Decontamination proce- dures are widely available from various sources, for example, the U.s. navy, Diving Unlimited international, national oceanic and atmospheric administration, Environmental Protection agency, or Federal Emergency management administration. special circumstances include rainfall runoff (increased pollutants from farming; known point and nonpoint source polluters should be checked); sediments (many lakes and com- mercial harbors have sediments with significantly higher levels of contamination than the water column, e.g., polychlorinated biphenyls and heavy metals); and hazardous materials (areas lang: PaRamEtERs oF ExtREmE EnviRonmEnt Diving ? 157 with gross fuel contamination, e.g., leaking ships, storage tanks or aircraft recovery, or in areas with a high concentration of creosote-soaked wood or antifouling paint). select contAminAted wAter diVing exAmPles Case 1 - Past: Smithsonian Tropical Research Institute Oil Spill Project marine environments are subject to man-made disasters. The escape of 100 000 barrels of oil into the mangroves and reefs of Bahia las minas (Caribbean) has had unexpectedly prolonged effects (Jackson et al, 1989). oil seeps into the sedi- ments around mangroves and returns to coat the coral reefs year after year as heavy rainfalls (exacerbated by the effects of de- forestation) slowly wash it out. injury, postimpact regeneration and growth of corals have been assessed by scientific divers at this oil spill site (guzman et al, 1994). The skeletons of corals record the history of acute disasters as well as chronic stresses. x-ray analyses of corals done in response to the oil spill docu- ment a worrying decline in coral growth over the past century. oil pollution, nutrient pollution and sedimentation due to ex- cessive runoff from deforested areas are extreme coral stressors. multiple stressors have a combinatorial effect on an otherwise resilient system, e.g., galeta Point, Panama (Jackson, et al., 1989; keller and Jackson, 1993; guzman et al., 1994). Case 2 - Present: U.S. Antarctic Program (USAP) Benthic Pollution Studies several benthic pollution studies were initiated in 1992 in the contaminated water of Winter Quarters Bay and around the mcmurdo outfall (Conlan et al., 2004). initial sampling was done by UsaP commercial divers who were working at mcmurdo station. scientific divers, anxious to see the study site and use the diving mode began diving with the commer- cial divers to complete the sampling requirements. all diving was done with program-supplied commercial diving equipment (Robbins, 2006). The 2-day training program includes a mini- mum of 2 familiarization dives under direct supervision of the UsaP Dive supervisor. Hands-on briefings with the scientific Diving Coordinator cover system set-up, introduction/familiar- ization with bandmask/helmet, out-of-air emergencies, tether management, line pull signals, free-flow procedures, equaliza- tion, de-fogging, decompression requirements, and tending the surface-supplied diver. a 3-person crew is the minimum personnel requirement for UsaP surface-supplied diving. The positions include a su- pervisor/tender, a diver, and a suited standby diver. The standby diver can use either scuba or surface supply. in addition to contaminated water operations, surface-sup- plied diving has found other niches in the UsaP diving reper- toire. surface-supplied diving is now the exclusive mode used by UsaP divers operating in the Dry valley lakes. Environmental protocols mandate the use of solo divers to minimize disrup- tion of lake haloclines. safety concerns demand that solo divers using relatively unreliable band- or full-face masks be provided with a large supply of breathing gas. UsaP experience with Exo-26 masks has been 11 free-flows in 106 dives (10.4% fail- ure rate). aga masks have had 2 free-flows in 26 dives (7.7% failure rate). These data come from dives in the Dry valley lakes where water temperatures range between 0?C and 2?C. it is as- sumed that failure rate would be even higher in -1.8?C water of mcmurdo sound. specific failure rates for either the Heliox-18 or superlite-17 helmets cannot be extracted from the UsaP da- tabase, although it is felt to be similar to the full-face masks. surface-supplied diving is used when workloads demand higher respiration rates than can be supplied by scuba. another current benthic pollution study requires coring in an area of frozen sediment. at other sites for this study, a large number of cores are required at 40 m. The ability to work fast, with inher- ently high respiration rates, results in a reduction of the number of dives required to complete this sampling. since 2001, 459 surface-supplied dives have been logged by 32 surface-supplied divers. The DUi tls350 or Rs1050 drysuits or trelleborg viking Pro or HD vulcanized rubber dry- suits work better in extremely cold environments because they do not freeze in the air. Case 3 - Future: Smithsonian Environmental Research Center Invasive Species Project Established in 1997 pursuant to the national invasive spe- cies act of 1996, the national Ballast information Clearing- house (nBiC) is a joint program of sERC and the U.s. Coast guard that collects, analyzes, and interprets data on the ballast water management practices of commercial ships that operate in U.s. waters. The principal goals are to quantify the amounts and origins of ballast water discharged in U.s. coastal systems and to determine the degree to which such water has under- gone open-ocean exchange or alternative treatments designed to reduce the likelihood of ballast-mediated invasions by exotic species. The World Conservation Union (iUCn) established gener- ally accepted distinct bioregions of the world. organisms have been transferred among bioregion boundaries with economic damage and environmental impacts as a result. Hull fouling and ballast water act as invasive species vectors requiring sampling in select harbors and ports at locations such as oakland-san Francisco Bay, Prince William sound, Chesapeake Bay, and tampa Bay by scientific divers, commercial divers, and remotely operated vehicles (Rovs). ExTrEmE ENvIrONmENTS: dEpTh a majority of the marine ecosystem habitats of interest are located in waters of up to 130 m depth, which roughly cor- responds to the continental shelf and the photic zone. The 158 ? tHE FUtURE oF Diving advanced scientific Diving Workshop (lang and smith, 2006) examined the existing technologies (open-circuit mixed gas, closed- and semi-closed circuit rebreathers, habitats and fly-away saturation systems, bell diving, and surface-supplied mixed gas) with the goal of establishing a robust capability for government- agency sponsored (nonmilitary) scientific divers to once again work at depths of up to 90 m through a phased approach. safety concerns have gradually eroded this depth limit to what has be- come a 60 m operational compressed air window for scientific diving, as published by the Department of labor occupational safety and Health administration (osHa) in 1982 and 1985 (29 CFR 1910). osHa does not regulate the scientific diving community with regard to technology, which provides us with the operational flexibility to employ mixed gases in our research methodology to meet the nation?s marine science needs. Reliable commercial diving technology exists to reach these depths, but is not routinely employed by the scientific commu- nity. training a competent scientific diver in surface-supplied mixed-gas diving, where the dive, gas mix, depths, bottom times, voice communication, and decompression are controlled from the surface would appear to be the simplest method of choice. The limitations of diving on a hose under these constraints do not seem to outweigh the advantages of immediate access to 90 m depth. The sparsely explored depths between 60 and 100 m are reflected in our incomplete knowledge of contributions of deep coral reefs to shallow-reef systems and the infrequent dis- covery of cryptic and deep-water species that are new to science. attempts at introducing rebreathers into mainstream scien- tific diving programs have met with inertia and significant safety concerns due to issues of reliability, availability, time investment in training, and proficiency requirements. The workshop focus was therefore not on rebreathers per se, as they have received much attention through numerous venues over the last 15 years and should continue to evolve on a parallel track. Rather, de- liberations emphasized a scientific diving capability to conduct research between 60 and 90 m by evaluating a re-expansion of the scientific diving envelope through mixed gas and surface- supplied techniques. Acknowledgments i wish to thank John scott Haldane, alf Brubakk of the norwegian University of science and technology, the smith- sonian office of the Under secretary for science, Bob stinton and Dick long of Diving Unlimited international, inc., John Clarke of the U.s. navy Experimental Diving Unit, Rob Rob- bins of Raytheon Polar services, nEDU, the national science Foundation office of Polar Programs, and martin sayer of the nERC national Facility for scientific Diving. lITEraTurE cITEd Clarke, J.R. 2007. scuba regulators for use in cold water: the U.s. navy perspective. in lang, m. a. and m. D. J. sayer (eds.) 2007. Proceedings of the International Polar Diving Workshop. svalbard, march 15?21, 2007. Washington, DC: smithsonian institution, pp. 35?44. Conlan, k. E., s. l. kim, H. s. lenihan, and J. s. oliver. 2004. Benthic changes during 10 years of organic enrichment by mcmurdo station, antarctica. Marine Pollution Bulletin 49:43?60. Duji?, ?., i. Palada, a. obad, D. Duplan?i?, D. Bakovic, and Z. valic. 2005. Exercise during a 3-min decompression stop reduces post- dive venous gas bubbles. Medicine and Science in Sports and Exercise 37(8):1319?1323. Duji?, ?., a. obad, i. Palada, v. ivan?ev and Z. valic. 2006. venous bub- ble count declines during strenuous exercise after an open sea dive to 30 m. Aviation, Space, and Environmental Medicine 77(6):592?596. gerriets, t., k. tetzlaff, t. liceni, C. schafer, B. Rosengarten, g. kopiske, C. algermissen, n. struck, and m. kaps. 2000. arteriovenous bub- bles following cold water sport dives: relation to right-to-left shunt- ing. Neurology 55(11):1741?1743. guzman, H. m., k. a. Burns, and J. B. C. Jackson. 1994. injury, regen- eration and growth of Caribbean reef corals after a major oil spill in Panama. Marine Ecology Progress Series 105:231?241. Hop, H., and o. Pavlova. 2006. ice amphipods in drifting sea ice around svalbard: quantitative collections by scuba divers. in lang, m.a. and m.D.J. sayer (eds.) 2007. Proceedings of the International Polar Diving Workshop. svalbard, march 15?21, 2007. Washington, DC: smithso- nian institution, pp. 181?186. Jackson, J. B. C., J. D. Cubit, B. C. keller, v. Batista, k. Burns, H. m. Caffey, R. l. Caldwell, s. D. garrity, C. D. getter, C. gonzalez, H. m. guzman, k. W. kaufmann, a. H. knap, s. C. levings, m. F. marshal, R. steger, R. C. thompson, and E. Weil. 1989. Ecological effects of a major oil spill on Panamanian coastal marine communi- ties. Science 243:37?44. Jankowski, l. W., P. tikuisis, and R. Y. nishi. 2004. Exercise effects dur- ing diving and decompression on postdive venous gas emboli. Avia- tion, Space, and Environmental Medicine 75(6):489?495. keller, B. D., and J. B. C. Jackson (eds.) 1993. long-term assessment of the oil spill at Bah?a las minas, Panama synthesis Report. new orleans: oCs study mms 93-0047. U.s. Department of the inte- rior, minerals management service, gulf of mexico oCs Regional office. lang, m. a., and n. E. smith (eds.) 2006. Proceedings of the Advanced Sci- entific Diving Workshop. Washington, DC: smithsonian institution. lang, m. a., and m. D. J. sayer (eds.) 2007. Proceedings of the International Polar Diving Workshop. svalbard, march 15?21, 2007. Washington, DC: smithsonian institution. lang, m. a. and R. Robbins. 2007. UsaP scientific Diving Program. in lang, m. a. and m. D. J. sayer (eds.) 2007. Proceedings of the Inter- national Polar Diving Workshop. svalbard, march 15?21, 2007. Wash- ington, DC: smithsonian institution. pp. 133?155. lowrance, W. W. 1976. Of Acceptable Risk: Science and the Determination of Safety. los altos, California: W. kaufmann. mekjavic, i. B., m. J. tipton, and o. Eiken. 2003. thermal Consider- ations in Diving. in Brubakk, a. o. and t. s. neuman, eds., Bennett and Elliott?s Physiology and Medicine of Diving, 5th Edition. london: Elsevier science ltd. mueller, P. H. J. 2007. Cold stress as DCs factor. in lang, m. a. and m. D. J. sayer (eds). Proceedings of the International Polar Diving Work- shop. svalbard, march 15?21, 2007. Washington, DC: smithsonian institution. pp. 63?72. osHa, U.s. Department of labor. 1982. Commercial Diving standards. 29 Code of Federal Regulations Ch. xvii subpart t 1910.401 and appendix B to subpart t - guidelines for scientific Diving (1985), Washington, DC. Risberg, J., and a. Hope. 2001. thermal insulation properties of argon used as a dry suit inf lation gas. Journal of Undersea and Hyperbaric Medicine 28(3):137?143. Robbins, R. 2006. UsaP surface-supplied diving. in lang, m. a. and n. lang: PaRamEtERs oF ExtREmE EnviRonmEnt Diving ? 159 E. smith (eds.) Proceedings of the Advanced Scientific Diving Work- shop. Washington, DC: smithsonian institution. pp. 187?191. Ruterbusch, v. l., W. a. gerth, and E. t. long. 2004. Diver thermal status as a risk factor for decompression sickness (DCs). Journal of Undersea and Hyperbaric Medicine 31(3):95. Ruterbusch, v. l., W. a. gerth, and E. t. long. 2005. Benefits of warm decompression can be realized with hot water suits. Journal of Under- sea and Hyperbaric Medicine 32(4):33. sayer, m. D. J., k. P. Fraser, t. m. shimmield and J. P. Wilkinson. 2006. Under sea ice diving operations carried out by Uk organizations in support of scientific research. in lang, m. a. and m. D. J. sayer (eds.) 2007. Proceedings of the International Polar Diving Workshop. svalbard, march 15?21, 2007. Washington, DC: smithsonian institution. pp. 95?110. stinton, R. t. 2007. a review of diver thermal protection strategies for polar diving: present and future. in lang, m. a. and m. D. J. sayer (eds). Proceedings of the International Polar Diving Workshop. sval- bard, march 15?21, 2007. Washington, DC: smithsonian institu- tion. pp. 13?34. tetzlaff, k., l. Friege, a. koch, l. Heine, B. neubauer, n. struck, and t.s. mutzbauer. 2001. Effects of ambient cold and depth on lung function in humans after a single scuba dive. European Journal of Ap- plied Physiology 85(1?2):125?129. Weinberg, R. P. 1992. alternate drysuit inf lation gas for improved thermal insulation. in R. nishi (ed.) Proceedings of the DCIEM Diver Thermal Protection Workshop. toronto: DCiEm. pp. 41?43. prEludE at the Haldane symposium, my colleague and jazz musician John P?l inderberg and i were asked to play and give a talk on jazz improvisation. in 1908, J. s. Hal-dane and coworkers published their paper ?The prevention of compressed-air- illness? that has formed the basis of modern decompression procedures. one century later it was time for a symposium in trondheim, norway, December 18-19, 2008. The title of our presentation was: ?Improvisation: Between Panic and Boredom.? For us to talk to international researchers on diving seemed a little strange, what has jazz improvisa- tion to do with diving? However, reactions to our performance and presentation, which mainly consisted of jazz music sounding from a baritone sax and a double bass supplemented with com- ments, were very positive. at certain points, it seemed that there might be links between these two seemingly rather separate areas, diving and improvisation. For example, blow- ing and breathing through a baritone sax has certain similarities with breathing through a diving apparatus. kinra and okasha (1999) published on the risk for health damage and illness due to exertion of blowing a jazz saxophone titled Unsafe sax: cohort study of the impact of too much sax on the mortality of famous jazz musicians. a depressing title for my fellow sax playing musician! our hypothesis that improvisation has a constructive and fruitful acting potential on all kinds of human behavior was still encouraged by this thin thread between these areas. Per request of the symposium Convener, alf o. Brubakk, to write about creativity and improvisation as part of the Haldane symposium proceedings, i accepted to give a try to his request. Parts of this writing are based on my article: Improvisation on a Triple Theme (alterhaug, 2004). at the symposium, we started our presentation by playing a standard jazz tune, ti- tled ?Whisper not? by Benny golson. We then commented on how we, in the course of performance, created new melodies, rhythms, and variants of harmonies on this tune?s structure: the balance between structure and flexibility. Related to this explanation, we aBSTRaCT. This paper focuses on creativity and improvisation; how these phenomena and acting strategies in real time seem to create optimal conditions for innovative activity. as such, they rep- resent supplementary and alternative ways of acting, learning, and organising in different contexts. That?s my way of preparation - to not be prepared. and that takes a lot of preparation! ?lee konitz Department of Music, Norwegian University of Science and Technology, N-7491 Trondheim, Norway. Creativity and improvisation as Phenomena and acting Potential in Different Contexts Bj?rn Alterhaug 162 ? tHE FUtURE oF Diving told the following story, which was told to us June 19, 2003, at a seminar in trondheim arranged by sintef industrial manage- ment, safety and Reliability, where saxophonist John P?l inder- berg and i presented a paper. a manager of the deep-sea diving section on a north sea oil platform spoke about his experiences with training deep-sea div- ers in emergency situations. The managers tried to give the divers highly detailed and exact written instructions for how to react in dangerous situations. However, when the divers strictly followed the instructions in dangerous situations, they concentrated on the prescribed rules so much that they forgot to use their own imagination and creativity to solve this complex situation. The result was critically dangerous and they needed outside help to save their lives. The instructions were then reformulated to just a few fundamental points, which were easy to remember, and which gave a better balance between structure and flexibility. in other words, the divers were given the opportunity to improvise on internalised knowledge, trusting their inborn capacities. This meant that they could keep calm and did not panic; accordingly, they handled the most complex situation in an adequate way through improvising. This short story brings to the fore the importance of cre- ativity and improvisation as an important acting potential in real time situations. in the following, text examples are present- ed and an argument made for a better understanding and signif- icance of these latin-derived words that are often interpreted in many ways, often far from their etymological meaning. Before approaching these words more in detail, i will briefly mention some personal experiences, which refer to a more or less confus- ing understanding of improvisation. The word improvisation is generally used with a negative connotation: a kind of accidental acting without planning, which often leads to incomplete and bad solutions. in the musi- cal contexts that jazz improvisation has, improvisation has of- ten been characterised by the uninitiated as chaotic, without structure, meaningless. as a young jazz musician i was asked by an elder relative of mine during a concert: ?Bj?rn, couldn?t your band please play a tune that you all know.? The improvised melodies sounded that unfamiliar and strange to my relative, that he thought we just played occasional, strange melodies at the same time, which we were actually doing. But, these melody lines were interconnected through an underlying structure. a structure organised by a common-rhythm feel that gave the in- dividual musician freedom, space and time to express himself in the collective, where the interactional conditions ?on the spot? enhanced the creative and improvisational forces. During many years as a jazz musician and academic, i have met a lot of strange and confusing questions, especially on the meaning of the word ?improvisation.? Undoubtedly, there is need for some clarification. This paper aims to bring to light some aspects connected to the word and phenomenon improvi- sation in music and other contexts. i will also argue that insight and knowledge on creativity and improvisation might open up alternative ways of acting, learning, and organising. apprOachINg crEaTIvITy aNd ImprOvISaTION ?Creativity? is derived from the latin word creare, ?to cre- ate,? and as such it is closely related to improvisation. Even if these two words seem to represent the same meaning, and often are used by chance, there is a distinction that may be useful to figure out (J?rgensen, 2009). Creativity is a solution-oriented and innovative development process where there is an exchange between exploration, reflection, and new exploration. The ex- ploration part does not necessarily have to be of a practical kind, and might be a combination of practice and reconsiderations; a circular circuit of idea-acting consequence. Creativity is un- derstood as a slow-floating improvisation, a process based on knowledge, proficiency, and reflection over time. This creates the platform for the ?in-the moment-innovation,? what appar- ently happens in a ?magical? way by creative persons. improvisation may be described as immediate action, where there is no time for reflections in the moment, yet action based on deep internalized knowledge, proficiency, and creativ- ity. improvisation is the creativity manifested in the moment?s invention. in interactive acting situations, improvisation has to, in order to benefit a common goal, unfold in a proficient collective that shares the creative capacity; the members of the spirit of the community have to be on the same wavelength in order to interact-improvise together on the spot. trust, security, tolerance, and humor are important keywords in this kind of interaction. an example of positive effects due to creativity and improvi- sation is the U.s. airways Captain and his crew who in January 2009 were forced to land an airplane on the Hudson River be- cause of a bird strike that damaged the engines. This emergency situation was solved elegantly due to a high state of readiness, internalised skills, and practice. The pilot and the crew had been trained and prepared for different emergency situations, but likely not for this very special situation. However, all kinds of emergency actions and procedures had been rehearsed repeatedly in simulators and in other contexts. When this dangerous situ- ation occurred, the crew was mentally and physically prepared; the procedures and necessary actions for this situation were per- formed automatically. Creativity and improvisation were coor- dinated on the spur of the moment, a superb combination for solving unexpected situations. There was obviously no time to look in the manual and find written procedures to solve this un- expected and dramatic situation, which was totally unforeseen. is it possible to be prepared for the unexpected, the unforeseen, and if so, how? These will be the main questions to be consid- ered. improvisation - and jazz improvisation especially - will be used as a metaphor for a further exploration of this topic. mEaNINgS aNd ETymOlOgy The different usages of improvisation in daily speech of- ten cause confusion when improvisation as a concept and altERHaUg: CREativitY anD imPRovisation ? 163 for many days. The greeks grew up with the epic poems such as the Iliad and the Odyssey. However, they did not read them, they listened to them. This is an illustration of the importance of the aural aspect and recitation/learning by heart as crucial elements of improvisation, irrespective of historical period. Plato calls the art of poetry ?the art which the ear can perceive,? and uses the expression ?hearing something from a book.? The Roman Quintillian (35?100 aD) says the following about making a good speech: ?to be free to improvise, you have to know the speech so well that you do not feel restricted by it? (andersen, 1995). Proficiency as an improviser in rhetoric contexts, according to the old sources, is due to two factors: natural predisposition and practice, i.e., talent and training. The former is regarded as the most important, but the ability to improvise is also a mat- ter of practising, and one can find a lot of explicit advice and instructions for the improviser concerning memory techniques and elements that influence the performance, including voice, face, body, ear, and rhythm (andersen, 1995). Even pressure to deliver and nervousness are treated in this connection. For in- stance, Crassus said: ?? if not even the most proficient and re- laxed speakers go to work with a certain anxiety and are moved by nervousness when they start to speak, well, then they are lost to all sense of shame.? This focuses the very important role of emotion, state of mind, and mood in performance situations. From ancient times until today, we find improvisation in every known musical culture, even if there is a great variation in how it is handled in the different traditions. JaZZ aNd ?afrIcaN WayS? in the beginning of the 20th century, new orleans became the center for the development of a new musical expression: jazz. improvisation reappeared in this music in a way that proved to be of vital importance and influence for all kinds of music through- out the century. ?of all the musical forms to emerge during the twentieth century, jazz was by far the most significant? (shipton, 2001). Jazz has a relatively short history. The first references to it come from the West Coast of america, where the San Francisco Bulletin of march 1913 used the term to describe a dance mu- sic full of vigor and ?pep? (shipton, 2001). a definition of this expression has since caused many discussions, and even today there is no final conclusion, which should be looked upon as a healthy state and evidence for a vital cultural expression. The New Grove Dictionary of Jazz definition is as follows: ?a music created mainly by black americans in the early 20th century through the amalgamation of elements drawn from European-american and tribal african music.? This rather open definition recognises that??on the whole, ethnicity provides a core, a centre of gravity for the narrative of jazz, and is one ele- ment that unites the several different kinds of narratives in use today? (Deveaux, 1991). With the emergence of jazz in the first part of the 20th century, improvisation gradually found its place on the musical phenomenon is discussed. among very different meanings, there seem to be two main meanings of the word: 1) improvi- sation as emergency actions (e.g., the plans failed so i had to improvise?). This statement presupposes that human action is normally based on following rules and instructions; and, 2) improvisation based on a high state of readiness, internalised skills and practice, a first-rate way of acting. This meaning is based on another important concept, tacitly knowing. During our adolescence, we develop a kind of readiness for ?whatever may happen.? This readiness is acquired mainly through direct involvement in various problem-solving situations, and this type of knowledge is mediated via bodily experiences and alert awareness. Readiness is crucial when acting in real-time turbu- lent and chaotic surroundings. in this paper, improvisation is understood in accordance with the second meaning. all human beings, whatever their background, have cre- ative potential. it is therefore necessary to emphasize that cre- ativity and improvisation are human features of a general char- acter, which we may recognise as easily in a masterpiece as in everyday activities. Creativity has through history been of deci- sive importance for man?s ability to survive, e.g., when invent- ing new tools and adopting new approaches and perspectives. Even though the word ?improvisation? was not used in a musi- cal context until approximately 1850, improvisation has a long history. in antiquity, we find it as an important part of rhetoric training and in the field of music. in Western music history; we know that medieval music was mostly improvisational and that leading composers such as Bach, mozart, Beethoven, and Chopin were great improvisers. The latin improvisus means ?unforeseen? or ?on the spur of the moment.? The root of the word improvisation is the latin word for ?see.? Visus is something which has been seen, while pro means before, in advance, etc. Provisus does not exist as a separate word in latin, but it would have had the mean- ing ?something which has been seen in advance.? The prefix im- is negative, yielding the meaning ?something which has not been seen in advance.? From improvisus, which gradually got the meaning ?unforeseen? or ?unexpected,? ?surprising,? italian has formed a verb improvisare, to do something without prepara- tion, solve an unexpected situation, and the noun improvisation is derived from this verb (gunhild vid?n, pers. comm.). ImprOvISaTION IN aNTIQuITy another latin term was extempore actio, which originally meant ?acting outside time? ? outside the normal flow of time, which could be related to the term ?experienced? time (Bergson, 1990). other definitions of improvisation are discussed below. ancient greek and Roman performance culture was mainly oral, and we see parallels to many music cultures of today, e.g., folk music, jazz, and popular music. This ancient literacy was both a tool and basis for oral communication. in athens, approxi- mately 500 BC, there were annual organised recitals that lasted 164 ? tHE FUtURE oF Diving scene and in recent years seems to generate interest within dif- ferent fields of research. Jazz has spread all over the world, and today we can describe jazz as global music. one of the main causes for this relatively fast dispersion is to be found in the african elements of jazz: african ways of behavior with respect to life and music are closely interwoven. The expression ?african ways? should not be conceived as if the african is something un- ambiguous or homogenous. The african refers to heterogeneous practices, but fundamental to these ways of communicating is the inclusive, interactive, dialogic aspect; where the actor or musician is the message (sidran, 1981). in the combination of singing, dancing, and drumming there is a ?celebration of life? which has many layers of tacit knowledge, experience, religion, and joy. in every such performance and meeting it seems very important to bring forth personal, individual voices; inviting others to make their contribution and to be open to all direc- tions and possibilities such dialogues and interactions may take. Behind these actions and activities, communities of practice full of joy and sorrow, there is a deep feeling of belonging to a col- lective of local identity where nature plays a decisive part. This rather generalised description and essentialism need some remarks. in the postcolonial period, parts of africa have been saddled with poverty and deep conflicts between tribes and regions, a gloomy fact that strongly contradicts my idealisation of african ways of acting. These human, political, social, and global urgent and complex questions are part of this paper; however, these questions are not discussed here. The main focus of this pa- per is particular aspects of african ways of acting that i have ex- perienced in different musical settings, e.g., in West african tradi- tions and in the interplay with african-american jazz musicians, these aspects provide a valuable contribution to the discussion and discourses on interaction and communication in a global world. ubuNTu in his book ?no Future without Forgiveness,? the Reverend Desmond tutu talks about a south-african term ubuntu, which means, ?to be a human being.? all human relations that are im- printed of mutual respect, dignity, and community are terms or expressions for ubuntu. Ubuntu-Botho means ?the art or virtue of being a human creature.? We are born to live in a subtle net- work of mutual dependence. ?i am only a human being through another you.? We say ?a person is a person through other per- sons.? it is not, ?i think, therefore i am.? it says rather: ?i am human because i belong. i participate, i share? (tutu, 1999). This view of the world or humanism represents a universal ubuntu that includes all people irrespective of race, sex, ethnic- ity, political, cultural, and/or religious belonging. as the Rev- erend tutu said: ?We are all born different, like a rainbow. a rainbow is a rainbow just because the colours are different. The different colours come together, and then we get a rainbow.? The influence of ?african ways? then seems to represent a po- tential for action that may be looked upon as ?an alternative modernism?: celebration of the unpredictable, improvisational, ?unforeseen,? rooted in tradition and internalised knowledge and experience combined with training to act ?on the spot.? The risk of a romanticisation of ?african ways? is clearly pres- ent in such pompous pronouncements, but as an alternative to over 200 years of Western rationality, time has come to develop conceptual frameworks grounded on alternatives: interaction, emotions, social creativity and internalised, tacit knowledge, and africa offers a rich source for such a framework. ThE ImmOvablE aNd ThE STaTIc The african-american art form jazz, jazz improvisation may be placed philosophically in a tradition that radically con- tradicts central parts of Western rationalist thinking: a static, mechanistic worldview rooted in the immovable world of ideas of Plato. in this respect, jazz may be linked to the thinking of the French philosopher Henri Bergson and the american prag- matists of the early 1900s. Flexibility and process are focused, life is considered as a stream of movements and changes, pro- cesses and tendencies that cross and intermingle, the complex. Consciousness and rationality are in this thinking constituted through creative action, often compared to an ecosystem. The norwegian ecophilosopher sigmund kval?y setreng has, with reference to Bergson?s process philosophy, developed a pedagogic model to elucidate the difference between European traditional system thinking and a philosophy grounded in pro- cess thinking. setreng (2001) uses the concepts ?complicated? and ?complex.? The complicated may be compared with a ma- chine based on reversibility, a closed system controlled from a center. The symphony orchestra?s hierarchical structure is used as a metaphor for this form of organising, the complicated. as a contrast, he uses the complex, which is open and irreversible self- organising grounded in interaction, comparable to an ecosys- tem. Here the jazz combo, a small group of two to six members, is used as an illustration of complexity. of course one can also find hierarchical structures in strictly organised jazz big bands with 18-20 members and more. setreng underlines that his models are simplified ideal types, used to illustrate a basic differ- ence in organisational thinking, representative of two different cultures (setreng, 2002). setreng?s thinking seems to have paral- lels to Edgar morin?s concept: self-eco-re-organisation, suggest- ing that the nature of the organisation changes as well and that it is an ongoing process of self-renewal that always happens in a context, in an environment, never in isolation or abstraction. no self without an eco. no text without context (montuori, 2003). dIffErENT pErSpEcTIvES, ObJEcTIvE TO SubJEcTIvE academic research on creativity during the 20th century has mainly concentrated on the individual, and it has often been based on a Western, modernistic understanding with em- phasis on objectivity, rationality, order, and theory. The tradi- tional goals of Western universities have been on a theoretical altERHaUg: CREativitY anD imPRovisation ? 165 understanding of the world. ?Practical? knowledge traditions, with their background in what people do in the world, have received little attention, as practice is regarded as less important than theoretical issues. nearly all knowledge and searching for the truth is grounded in traditional systems thinking: logical- rational thinking expressed in a technical language. The subjective, emotional, accidental, and tacit knowledge have had almost no space within this hierarchic, hegemonic par- adigm, which in a context of art often is grounded on a ?genius- aesthetic? alien to most cultures outside the Western world. (see e.g., Encyclopaedia of Psychology, 2000, vol. 2, p. 339??The musical idea has to be written down and orchestrated??). This way of conceiving the world is mainly rooted in the metaphysics of antiquity: inflexibility regarded as synonymous with abso- luteness and perfection, a world view that has contributed to a static understanding of reality in which control and prediction, in human interrelationships too, have had the highest priority, expressed in quite simple, mechanistic models. Within musicology, one would expect that studies of ethnic music cultures should lead to new and extensive research on cre- ativity, but research on this topic appears to be marginal. at the end of the last millennium, certain endeavours to do research within this field could be seen. However, it is remarkable that new, innovative research, based on impulses from, in this case jazz improvisation, occurs quite often within disciplines such as organisational theory, economy, business, social science and psychology, and now also diving, but only haphazardly within musicology. ThE JaZZ mETaphOr Berliner?s (1994) book, ?Thinking in Jazz,? has been of vital importance for understanding the jazz culture and has inspired different writers to use jazz improvisation as a metaphor for knowledge, innovation, and creativity in organisational con- nections (Barrett, 2002; Hatch, 2002; Weick, 2002). However, when using the jazz metaphor, one has to be aware of the fact that jazz has changed through the years and that the metaphor has to be used carefully as the difference between for example swing Jazz from the 1930s and Free Jazz from the 1960s is es- sential. However, in spite of differences, one element has been central through the whole jazz tradition, i.e., improvisation. WOrdS, myThS, aNd dEfINITIONS Bailey (1980) stated: The word improvisation is actually very little used by improvising musicians. idiomatic improvisers, in describing what they do, use the name of the idiom. They ?play flamenco? or ?play jazz?; some refer to what they do as just ?playing.? There is a noticeable reluc- tance to use the word and some improvisers express a positive dis- like for it. i think this is due to its widely accepted connotations, which imply that improvisation is something without preparation and without consideration, a completely ad hoc activity, frivolous and inconsequential, lacking in design and method. and they ob- ject to that implication because they know from their own experi- ence that it is untrue. They know that there is no musical activity that requires greater skills and devotion, preparation, training and commitment. and so they resent the word, which in some places has become almost a term of abuse. They recognise that, as it is generally understood, it completely misrepresents the depth and complexity of their work. But i have chosen to retain that term throughout this book; firstly because i don?t know of any other which could effectively replace it, and secondly because i hope that we, the other contributors and myself, might be able to re- define it. one of the most common myths concerning improvisation is the romantic one. The improvising jazz musician is the incar- nation of god; improvisation is a supernatural phenomenon, a privileged gift of the chosen few. When we talk about impro- visation on a high artistic level, it may be correct to speak of a special talent, which not all of us are in possession of. But there is one thing that both the specially gifted people and others that deal with jazz improvisation have in common: hard work, the long and demanding way to achieve an instrumental standard and personal strength that justifies the power of intuition. There are various definitions of improvisation. Dictionaries give us such definitions as ?a performance according to the in- ventive whim of the moment? or ?to perform without prepara- tion.? ?improvisation? is here confined to the first usage of the term mentioned at the beginning: improvisation seen almost as an inferior, second-hand solution. But the second mentioned usage, the one relevant for jazz, is something quite different. That activity requires thorough preparation, both mentally and with respect to skills that have to be internalized, which means that the performer is prepared to handle the unexpected, to handle an error as a new creative challenge, breaking with ha- bitual patterns. all professional jazz musicians know that ?play- ing jazz,? improvising, means to strive hard to obtain the ideal state, the ?golden moments,? these ecstatic heights in musical interaction that are the main reason why we play. But as we all know, unfortunately these great moments really happen all too seldom. However, behind this motivation and intentionality to reach this ?peak performance,? there has to be an existential urge. Besides these elements, it is crucial that there is a good bal- ance between challenges and skills. if this balance is not optimal, the musicians will either feel bored or anxious, the musicians? potential for interaction weakened, and this will have a negative impact on the ensemble play. When this balance is optimal, the musicians will feel good and in this ?aesthetics of presence,? they are in a state often referred to as ?flow.? musicians describe this state as ?being played,? only observing that the fingers are playing their instrument. as such the performer?s condition will be a kind of constructive uncertainty and confusion, being in a transcendental state. This seems close to social psychologist mihaly Csikscentmihalyi?s description of the ?flow? concept: 166 ? tHE FUtURE oF Diving Flow denotes the holistic sensation present when we act with total involvement.... it is the state in which action follows upon action according to an internal logic, which seems to need no conscious intervention on our part. We experience it as a unified flowing from one moment to the next, in which we feel in control of our actions, and in which there is little distinction between self and environment, between stimulus and response, or between past, present and future (Csikscentmihalyi, 1975). When jazz musicians are asked to define improvisation, they often answer as Bailey mentioned, ?i am just playing,? or they try to find metaphoric terms, as did composer, arranger and trumpeter Thad Jones in an interview: Here the aesthetics of presence hold unrestrictedly. You give your- self up, surrender without ulterior motives; egoism and spirit of competition yield for generosity, presence and interdependence. one develops a presence that is like telepathic intuition ? during such moments, improvisation is like the language that develops between two loving partners and that usually is called eroticism (oversand, 1987). it is tempting to link Jones? poetic description of improvisation to the old sanskrit word l?la, meaning ?divine play,? the play of creation, destruction, and recreation, the folding and unfolding of the cosmos. L?la, free and deep, is both the delight and enjoy- ment of this moment, and the play of god. it also means love (nachmanovitch, 1990). as we see, to find adequate descriptions or definitions of this transient phenomenon in scientific linguistic terms is dif- ficult, which underlines the complexity of interactions ?in the course of performance.? However, among the many different definitions, i find Paul Berliner?s definitions on jazz improvisa- tion particularly fruitful in this context: improvisation involves reworking pre-composed material and de- signs in relation to unanticipated ideas conceived, shaped, and transformed under the special conditions of performance, thereby adding unique features to every creation (Berliner, 1994). This definition recognises that successful, creative improvisation is dependent on preparation and training as fundamental fac- tors to spontaneous and intuitive action. ImprOvISaTION IN OThEr cONTExTS Even if we in the last decade can find quite a lot of litera- ture on improvisation outside the field of music, the necessity of improvisation seems rarely recognised in different organisa- tional contexts: industry, management, education, etc., prob- ably because these activities are mainly grounded on a reliance of linearity and traditional systems thinking. When regular planning breaks down, and solutions on the spot are needed, this way of acting (improvising) is often regarded as a sign of failure, called ?emergency solutions? or ?fire fighting,? followed by a cry for new and more detailed plans which often results in a lot of newly written, detailed, complicated instructions and bureaucratic procedures. The human potentiality for spontane- ous creativity under such circumstances is seldom fully put into action because we, in most Western societies, are trained to fol- low rules and prescriptions for most of our actions, and there- fore our natural creative capacity for finding the best solution in complex situations remains unused. The ability to find local solutions ?on the spot? is, as men- tioned, officially rarely accepted in work life. on the contrary, there is a tendency to regard improvisational activities as ?un- fair? and as a rupture of the activities? canonised, official admin- istration system. indeed, in socialist countries jazz was often suppressed because it pointed to individualism and freedom. to solve problems and local challenges that continually emerge during a working day, the employees that are closest to the problems then have to improvise ?off the cuff? by making use of their tacit knowledge. Thus, improvisation is regarded as a temporary expedient, something risky and dangerous in human activity and accordingly a practice and activity that should be avoided. The status of improvisation is evaluated to be very low and this outlook on improvisation obviously has not inspired research activities related to this vital topic. among the primary goals of this paper is to acknowledge prepared improvisational activity as a crucial and inevitable ac- tivity in different contexts that demand training and practice, for example in a small group of musicians. The aspects of in- teraction we can observe in a jazz group seem to be fruitful for finding metaphors that can be used to illustrate ways of impro- visation in other contexts. OrgaNISaTIONal ImprOvISaTION The relatively new field of organisational improvisation is concerned with the pressures on organisations to react con- tinually to today?s ever-changing environment. We often find questions as ?What can organisations learn from jazz?? Barrett (2002) offered seven features that could have implications for other contexts: 1. Provocative competence: Deliberate efforts to interrupt habit patterns. 2. Embracing errors as a source of learning. 3. shared orientation toward minimal structures that allow maximum flexibility. 4. Distributing task: continual negotiation and dialogue to- ward dynamic synchronisation. 5. Reliance on retrospective sense making. 6. ?Hanging out?: members in a community of practice. 7. taking turns in soloing and supporting/ accompanying. all of these characteristics are essential to the understand- ing of this musical expression. The third feature, ?shared ori- entation toward minimal structures that allows maximal flex- ibility,? what setreng (2002) calls ?light bridges? best captures altERHaUg: CREativitY anD imPRovisation ? 167 the core of interaction and creativity in a jazz group. The rela- tionship between structure and flexibility is an essential point concerning the best balance between order and chaos and in the space where creativity is given optimal conditions. STrucTurE aNd flExIbIlITy Flexibility is a term used in various contexts, but the con- cept is seldom defined. in the early 1970s, gregory Bateson defined flexibility as ?unused potential for change.? Flexibility consists of the possibilities one has not yet utilised, within de- fined frames. When the frames become narrower, the structure is filled with detailed prescriptions, and flexibility and creativity are reduced. When the structure in jazz and other human ac- tions is too complicated or rigid, flexibility, improvisation, and creativity do not have the chance to unfold. The example from the offshore industry, cited in the begin- ning of this paper, might be an example on how important a balanced weight between structure and flexibility is to get the optimal conditions for creativity and improvisation to unfold. Even if the differences between diving and jazz are considerable, for example, the consequences for making errors in diving are naturally far more serious than making errors in jazz improvisa- tion, there are many parallels concerning preparation, training, and alertness to act on the spot, using all senses, body and soul. Jazz is in daily speech often described as ?unstructured,? or ?lack of structure.? as we shall see, jazz is based on structure but not as distinctly as other forms of music. The structure in jazz is minimal and serves as a basis for the development of creative ideas, the patterns on which the musicians improvise. in this re- spect jazz differs from other musical expressions (e.g., pop, rock, art music) in how the structure is used as a frame for creative interaction. OrdEr ? chaOS Personally, i do not like the word ?structure? in musical surroundings very much. structure is often described nega- tively, with connotations to limits and restrictions. instead of structure, i find the words ?agreement? or ?mutual consent? to be more fruitful descriptions. ?mutual consent? gives an under- standing about the frames the musicians have agreed upon and that this agreement is not a diktat from outside, but a mutual consent that serves as a point of departure for creative activ- ity and display. in this sense, this agreement will be flexible; it may be enlarged and changed depending on the interaction in the collective dialogues. But if the mutual consents become too loose, the participants will not recognise the fundamental pattern, and they will lose contact with the agreement and with each other. The interplay breaks down and the result may be chaos. Chaotic performances may of course, for a short period, be interesting and even exciting, but the collective creative forc- es are badly off where chaos rules. STaNdardS The agreements within jazz are very often rooted in what we call the ?standards.? american film and musical tunes that through the years have become renowned are for example ?i got Rhythm,? ?summertime,? which are the basis, the mutual consent, for the creative interaction in a jazz group. standards are often built on musical forms of 32, 16, 12 bars, which can be said to function as minimal structures. many other references for improvisation are of course also used as mutual consents: melodic fragments, rhythmic ostinatos, melodic modes, bass fig- ures, etc. These agreements (structures) are the tools whereupon improvisation unfolds. The structure is internalised and experi- enced implicitly: i.e., it is not necessary to accentuate ?one? in every beat. ?one? is in this context implicit and does not have to be articulated (Hatch, 2002) Further, structure is not sacred to the jazz musician, it serves its own alteration. The freedom im- parted by not having to play structural features means that the musician can play around them, and this encourages creativity. i attended a concert with the Herbie Hancock Quartet in trond- heim, norway, in July 2003. The first tune they played was the well-known standard melody ?i love You,? nicely reharmonised using alternative chords by Hancock. i knew the tune very well, but during the performance i had a lot of trouble recognising the 32-bar structure of the tune. i was lost by every repetition of the form, and i was quite frustrated by my inability to recog- nise the well-known structure. talking to the musicians after the concert i got the answer. at a concert in France, the musical form, the agreement, was accidentally prolonged by one bar: ?? and you see, it was so much fun playing 33 bars, instead of the ordinary 32, that we still are playing this new structure!? dIalEcTIc paradOxES This little alteration, one bar, was for the musician a fresh inspiration and gave new challenges for the creative interplay. as shown in this case, the head arrangement can be changed in the course of performance, a way of acting and communicating that demands a lot of experience, training, and practice. This inter- action implies confidence in tradition, alertness, listening and responding, taking risks, but to a great extent it also involves confidence and sense of security. an existential act is difficult to find adequate linguistic terms for and suggests a series of ?dia- lectic paradoxes.? The musicians stress the importance of being able to do solos, but they must also be capable of accompanying and supporting other people. They know and respect the jazz tradition and its legacy carriers, but at the same time they want to challenge and push the tradition forward. These ?dialectical paradoxes? between immersion in the tradition and taking risks, between standing out as an individual voice and being support- ers of others? voices, beg for new and creative ways of develop- ing a deeper understanding of creativity and improvisation. improvising in jazz means to create a situation where change, transformation, and process are focused and where even 168 ? tHE FUtURE oF Diving the structure, the referential foundation of improvisation, may be part of the alterations. among the most important elements in this activity is changing the relation towards oneself, as this will be a fruitful point of departure for change in other rela- tions. Compare this with the saying of the norwegian drummer Jon Christensen in relation to his attitude and preparation be- fore a concert performance: ?at least, you have to try to surprise yourself ? (alterhaug, et al., 2002). The communicational aspect to oneself and the dialogic aspect to the others are central to all improvisational activity, which should be a good basis for a ?well-functioning democ- racy.? in all kinds of dialogues, there are always certain elements of uncertainty present because one does not know the ?result? until the participation in the dialogue has come to an end. This fact is nicely expressed by asplund: ?i don?t know what i have said until you have answered and you don?t know what you have said until i have answered. You show me what i have said and i show you what you have said? (molander, 1993). prOcESS aNd cOmplExITy IN grOupS groups in different fields face the paradox of planning (try- ing to create its future) contrasted to their inability to foresee what the future entails. to cope with this paradox, the leader function is often strengthened in order to get a better and more effective organisation. This type of organising is based on a hierarchical and pyramidal thinking that seldom gives the ex- pected results, because it neither brings out all the potential in each individual, nor the complementary forces in the group. another, contrasting way of organising that converges with re- search in complexity theory is that which we can find in the jazz group. The complex responsive processes (CRP) perspective is a transformative process view, where human phenomena in organizations are a result of people interacting with each other and the environment. By the responding processes they create, they transform the reality of both themselves and their environ- ment (stacey, 2001). in small jazz groups, distributed tasks, ro- tation of soloing (leadership), and continual negotiation toward a dynamic synchronisation are elements that create a flow of ongoing invention and mutual support within the group. an example of this type of self-organising, flat structure, ?moveable hierarchies,? in a jazz group, is from a personal experience from my career as a jazz bass player. around 1970, the great tenor saxophonist Ben Webster was invited to give concerts in trondheim, norway, and i was asked to play bass in the quartet. i was not very experienced and al- most scared to death playing with such a jazz giant. There was not much time for rehearsals; we hardly met before we were on stage to play a concert of two sets. During the first set, i tried hard to follow Webster?s playing, very respectfully and intently, listening to every musical move he made, my sole concern be- ing to do my very best for the ?star.? in the intermission, i was very eager to hear what he thought about our support, and of course, my playing. Without asking, and with a humble look, i sat down close to Ben waiting for something to be said. after a while he said: ?Yeah Bj?rn, you?re doing fine. But, you shouldn?t listen that much, then you lose yourself. You know, i need your initiative to play my best and then our best!? What a lesson from a master: to get the best music, you have to play in your own manner and of course be extremely alert to everything that happens during the performance, in that way the individual and collective forces in a team will have the best possibilities to un- fold, and this of course gives the best music, through a collec- tive, nonhierarchical approach. in jazz improvisation, the group has the ideas, not the individual musicians. Communicating in an improvisational way gives joy, re- leases energy, and activates knowledge and reflection. The basis for this form of interaction and interplay is trust and freedom, two elements that secure the social dimension. in this perspec- tive, knowledge is continually reproduced and transformed as interaction processes amongst people. Being involved in impro- visation has a positive effect on the learning environment, social competence, and development of a person?s creative abilities, and this activity underlines that knowledge should not be stored in individual heads or in libraries, but activated and negotiated as complex, responsive processes in relations between people. Being involved in creativity and improvisation in one or another context should lead to a kind of meta learning, a meta doing that transfers across styles and forms (nachmanovitch, 1990). The directness and dialogic nature of improvisatory prac- tice, which is something that happens ?face to face? in real time situations, makes it particularly relevant and interesting in rela- tion to communicational aspects in a globalised reality. Fisclin and Heble (2004) argue that music, specifically creative impro- vised music and free jazz, can reinvigorate our understanding of the social function of humanities research within the broader context of how that research plays a role in shaping notions of community and ?new forms? of social organization. rESEarch ON muSIcal pErfOrmaNcE aNd ExpErT pErfOrmaNcE Recent research in a new discipline called the learning sci- ences shows musical improvisation as a very good pedagogical tool, and with general knowledge features related to expert per- formance. sawyer (2008) argues that improvisation should be at the core of the music curriculum, and he also refers to cogni- tive science research from 1970s until now. This research shows the mental structures and processes underlying expert perfor- mance. Examples include doctors diagnosing a patient, lawyers analysing a case, architects designing a building, and scientists interpreting the result of an experiment. Expert performance has four features of knowledge that have much in common with musical improvisation (sawyer, 2008): 1. Deep conceptual understanding. Experts have not simply memorized a large repertory of facts. of course they know altERHaUg: CREativitY anD imPRovisation ? 169 a lot of facts, but in the expert?s mind, those facts are em- bedded in complex conceptual frameworks. Experts under- stand the mechanisms underlying phenomena and are able to explain surface features in terms of underlying mecha- nisms and conceptual structures. 2. integrated knowledge. Each piece of knowledge is highly interconnected with all of the other pieces of knowledge. Expertise does not result from possessing distinct compart- mentalized knowledge; everything known is related in an integrated framework. 3. adaptive expertise. Experts have mastered a large range of standard procedures and solutions. When first encounter- ing a new problem, they typically will quickly recall a vari- ety of similar problems they?ve encountered in the past, and they will begin by considering one of the solutions that has worked in the past. But experts do not simply apply these memorized procedures in rote fashion; they are able to flex- ibly modify the routines they?ve mastered or to combine elements of distinct routines as is appropriate to the new problem. 4. Collaborative skills. Experts work together with other ex- perts in teams and in complex organizational structures. Unlike the hierarchical corporation of old, where everyone?s job description was quite specific, the boundaries between each team member are fluid, and many tasks require the simultaneous and joint contributions of multiple experts to be successfully accomplished. sawyer (2002) ends his paper this way: ??my hope is that music education will lead the way in moving to a new type of pedagogical practice, one that is grounded in the learning sci- ences, one that is based in improvisation.? pOSTludE Fifteen years ago, i introduced improvisation as a discipline for university music students, most of whom had no previous training in improvising. The musical practice and improvisa- tional interaction have of course been the most important activ- ity in this improvisational discipline, but reflections through dialogues, as well as rather provoking and open questions about the students? everyday experience through the process have been of vital significance. Throughout the years, the students? evaluation of the im- provisational practice have been unambiguous when it comes to the following question: ?Do you think the improvisation practice may have some effects on your way of acting in every- day life, like feeling more secure and being more self-confident when talking in front of an audience?? almost all answers to this question have been positive, i.e., that the students through the improvisation practice have felt a better mastering of their own tensions, insecurity, and fear and that the fear of making mistakes had decreased radically. to me this means a reinforced belief in improvisation as a kind of meta-learning; a pedagogical strategy that has been strongly underestimated in all kinds of organised learning, from kindergarten to academia. Even if this cannot be regarded as valid research, the interaction with the students and their responses have, to me at least, been of great inspiration, an encouragement to continue this work and it has also served to reinforce the processes that had a creative begin- ning through musical interaction. improvisation becomes, in this context, an important fac- tor in the development of the personality. The pur pose is to give support to a creative process of consciousness: a process where individuals are made aware of their potentials and see these in relation to the relations and contexts they take part in. These studies should bring new angles to the understanding of the dialogic aspects of human communication and also with regard to introspection, alertness and social competence. Finally, these studies should lead to research and production of pedagogical material on different levels in the educational system to find new ways of learning and perceiving. cOda one question often arises when dealing with improvisa- tion: ?to what extent can one learn improvisation?? to me, the answer is easy: as human beings, we already are part of a great improvisation, and we are all more or less unconsciously im- provising most of the day. We must simply become aware of this fact, ?our unused potential for change,? and then find the open spaces and the creative atmosphere to release our hidden, inborn improvisational capacities in different contexts, for the best for ourselves and the world we are part of. lITEraTurE cITEd alterhaug, B. 2004. improvisation on a triple Theme: Creativity, Jazz impro- visation and Communication. in Studia Musicologica Norwegica. oslo: Universitetsforlaget. alterhaug, B., s. H. J?rgensen, and s. setreng. 2002. Improvisare necesse est, in: Forskning p? tvers, tverrfaglige forskningsprosjekter ved ntnU, tapir akademiske forlag. andersen, ?. 1995. I retorikkens hage, (in the garden of Rhetoric). oslo: Universitetsforlaget. Bailey, D. 1980. Improvisation. london: moorland Publishing Co. ltd. Barrett, F. J. 2002 Creativity and improvisation in jazz and organizations. in Or- ganisational Improvisation. kamoche, k., m. Pina E Cunha, and J. vieira da Cunha, eds. london and new York: Routledge. pp. 135?162. Bateson, g. 1972. Steps to an ecology of mind, new York: Ballantine Books. Bergson, H. 1990. Tiden og den frie vilje (original title: Essai sur les donn?es imm?diates de la conscience) oslo: aschehoug. Berliner, P. 1994. Thinking in Jazz. Chicago: University of Chicago Press. Csikszentmihalyi, m. 1975. Play and intrinsic reward. Journal of Humanistic Psychology 15: 41?63. Deveaux, s. 1991. Constructing the Jazz tradition: Jazz Historiography. in Keeping Time, Readings in Jazz History. R. Walser, ed. oxford: oxford University Press. Fisclin, D., and a. Heble. 2004. The Other Side of Nowhere. middletown: Wes- leyan University Press Hatch, m. 2002. Exploring the empty spaces of organizing. in Organiza- tional Improvisation. kamoche, k., m. Pina, E Cunha, and J. vieira 170 ? tHE FUtURE oF Diving da Cunha, eds. london and new York: Routledge. pp. 71?92. J?rgensen, s. H. 2009. unpublished manuscript. ?komplekse bevegelser og fotballfaglige utfordringer.? kinra, s., and m. okasha. 1999. Unsafe sax: cohort study of the impact of too much sax on the mortality of famous jazz musicians. British Medical Journal 319(7225):1612?1613. molander, B. 1993. Kunnskap i handling, (knowledge in action) Daidalos. morin, E. 1994. La complexite humaine, (Human complexity) Paris: Flammarion. montouri, a. 1992. Chaos, creativity and self-renewal. World Futures. the Journal of general Evolution. ???. 2003. the complexity of improvisation and the improvisation of complexity: social science, art and creativity. Human Relations 56(2):237?255. london: sagE Publications. nachmanovitch, s. 1990. Free Play, Published by J.P. tarcher, inc. Distrib- uted by Putnam-Berkley. nettl, B. 1998. In the Course of Performance, Chicago: University of Chi- cago Press. The New Grove Dictionary of American Music. 1986. london: macmillan. oversand, k. 1987. Improvisasjon og tilstedev?relsens estetikk, (improvi- sation and the aesthetics of presence) musikklidenskapelig, o.k. ledang, ed. sawyer, k. 2008. improvisation and teaching. Critical Studies in Improvi- sation vol. 3, no. 2. setreng, s. 2001. Mang fold og tid, (Complexity and time) skrifter fra musikkvitenskapelig institutt. trondheim: ntnU ???. 2002. improvisation and musical Quality ? a Question of Com- plexity. in Challenges in Norwegian Jazz Research. trondheim: ntnU institutt for musikk. sidran, B. 1981. Black Talk. new York: Da Capo Press. shipton, a. 2001. A New History of Jazz. london and new York: Continuum. stacey, R. D. 2001. Complex Responsive Processes in Organizations: Learning and Knowledge Creation. london: Routledge. treitler, l. 1991. medieval improvisation. in The World of Music 33 (3). tutu, D. 1999. No Future without Forgiveness. an image Book. new York: Doubleday, Random House. Weick, k. 1998. Creativity and the aesthetics of imperfection. Ford, C. gioia, eds. Creative Action in Organizations. thousands lakes: sage Press. ???. 2002. the aesthetic of imperfection in orchestras and organi- zations. in Organisational Improvisation. kamoche, k., m. Pina E Cunha, and J. vieira da Cunha, eds. london and new York: Routledge. pp. 49?70. a. Brubakk: The heart rate effect was very interesting. This fits well with our own modeling work that shows that exercise ac- tivity is important in order to judge decompression procedures. There are no dive computers that take that into account. The second interesting point is what is down there that these ani- mals stop diving. The whales use sound detection in order to locate their prey. it has been suggested by sam Ridgway, be- cause air is a bad conductor, that whales stop diving because the sound reflections become less distinct and they therefore termi- nate hunting dives. They have to wait until the gas bubbles have gone away, which may be because of the sonic detection they have in their brain. is the lung collapse related to a surfactant problem? is the surfactant different in these animals? some hu- man physiology researchers talked about the possibility of get- ting whales or seals and developing an animal research facility to look at decompression. a. Fahlman: We have no idea. The last research done on lung collapse was by gerry kooyman. it is extremely difficult to do this type of work with a seal in a hyperbaric chamber. We are now trying to devise ways to visualize animals under pressure. if we can do that, we can get a sense of how the lungs col- lapse and proceed with comparative studies of terrestrial and marine mammals. The compliances of the respiratory system could be first looked at from physics, then histology, genetics, and proteomics. a. Brubakk: Can you acquire the surfactant? a. Fahlman: some studies have been done on that in dead, diseased animals. h. Janssen: Do these animals change behavior and ability to do this with age? also, do their neurological systems anatomically change with age? a. Fahlman: That has not been studied. That would be an excel- lent long-term study. a. Brubakk: They also do not appear to get HPns. S. gaustad: How do these animals tolerate these high blood exposures with lactic acid not showing up during hypoxia? Discussion: Environmental Physiology and strategies for the next 100 Years Chairman: Richard E. Moon a. Fahlman: They usually dive until the point that lactic acid starts increasing, called anaerobic diving, where they return to the surface before they run out of oxygen. There is no large build up of lactic acid, and they have an immense vital capacity. if they have to dive longer and deeper, they build up Co 2 and then have to stop. m. gennser: You showed lung collapse in harbor seals and sea lions, but these are not deep diving animals, are they? lung collapse is a mechanism to protect against high oxygen and ni- trogen pressure during deep diving. a. Fahlman: The compliance of their respiratory system is the most divergent, the most different. You do not see that in ex- perimental situations. no one has ever shown lung collapse in Weddell seals either. The best and most convincing study is by kooyman and sinnett (1982) where they actually measured the pulmonary shunt. all the other studies are estimated values like i have shown here. The nitrogen tensions are fitted with curves. in Ridgway?s graph, there is no doubt that the respiratory sys- tem compressed and the lungs collapse eventually, but i highly disagree with their 70 m estimated collapse depth, i believe it to be much deeper. one animal diving with 80% of lung capacity had a shunt of only 70% at 90 m. m. gennser: The fact that the animals reduce their speed close to the surface is interesting. There were some submarine escape ex- periments conducted where the speed was reduced from 30 m to the surface and that increased the final depth to about 50?60 m. a. Fahlman: That is what they do between 15 and 30 m, and that is also around the estimated nitrogen tensions they have when they come back to the surface. There is much guess work here, but at least with these models, we can define what we need to study next, i.e., blood flow distribution and gas exchange, which are also important for human divers. h. Janssen: When other mammals can do this continuously without sustaining damage, the question is the mechanisms of exposures. There are things besides pressure we are exposed to as human beings, especially what we are breathing through. We 172 ? tHE FUtURE oF Diving know very little about how we send the gas down to the diver through the umbilical and what this consists of and how this might impact on the exposure. We are exposed to fumes and other particles and occupational hazards like petrochemicals. We have divers who have spent most of their careers, 1000s of days in saturation, over tens of years. is there damage we should see or might this be an adaptation? S. Daniels: in essence, i completely agree with you. John Ross? earlier comment was absolutely right. The difficulty associated with trying to understand what might have caused any potential damage is that there are too many changes, too many potentially adverse circumstances that the diver will have been exposed to in their work environment. The only way we are going to under- stand this is to dissect it out in controlled situations where we can actually measure specific changes as a result of a particular manipulation of the animal. trying to do cross-sectional stud- ies is fraught with enormous difficulties. Regarding deep diving mammals, what i think i have identified in a particular sample i have received is simply a mechanism that might explain how they could be acutely sensitive, it says nothing about potential long-term effects. it is highly likely that those animals have de- veloped adaptation mechanisms that compensate for their expo- sure, but we have no idea what those might be. The complexities of the regulations protecting marine mammals and acquiring samples are ferocious, which is why so little study has occurred, but it would be fascinating to do so. a. Brubakk: michael gernhardt was previously a saturation diver, went on to get his PhD in decompression physiology, and then became a nasa astronaut. Regrettably, he could not be here in person due to his testing schedule in the arizona desert of the prototype moon buggy, but he did send us his presenta- tion on video. R. moon: We do have in attendance an original astronaut, Dr. Jay Buckey. J. Buckey: That was a great talk mike just gave. it is beautiful, actually stunning, to see the Earth from space. R. moon: Referring to the tail of the distribution, the 5th per- centile, you concluded that the prevalence of disease would be twice as high in divers. is that a theoretical calculation or have you actually observed people in that category? e. Thorsen: We have not. This study was of 135 retired divers. We calculated the prevalence and it was at least 5.1% divers com- pared to the age that has a prevalence of 1%. There has been some discussion about calculating prevalence, so it may be anything from 5% to 1% and is dependent on the denominator you use. B. gardiner: Were all those subjects ex-north sea professional sat divers? if so, i have always been a bit skeptical because satu- ration divers know how to fiddle that to keep getting their med- ical year after year. Do you do the straightforward spirometry they are used to offshore? e. Thorsen: We do spirometry properly and stick to exhalation criteria concerning the solidity of the curves. We look at FEv1 in absolute numbers, not the FEv1 ratio which is dependent on the effort or the point at which you would like to stop the exha- lation. The problem with spirometry is that it is often not done according to the standard and the maintenance of the equip- ment can be questioned at many levels. m. Swiergosz: Can you comment on the clinical relevance and why it matters? e. Thorsen: Five percent of retired divers with median age 50 have obstructive lung disease influencing their quality of life. The gold criterion is 1 or lower. This study says nothing about the severity of CoPD, but they are clinically classified as having CoPD. m. Swiergosz: it is an interesting question to follow up. What kinds of functions are they losing? What is the definition of quality of life other than our scientific one? e. Thorsen: Concerning lung function and reduction in FEv1, it is a reduction of their physical working capacity. The most important thing probably is that their CoPD is classified as an occupational exposure that induces some rights in the social security system that you will not get if there is some other cause. R. moon: How do you manage your saturation crew? J. gangenes: These vessels support continuous saturation div- ing. We have 12 men on the seabed all the time and normally six in decompression on a rotational basis. twelve divers are kept on the bottom 365 days, unless we are in refit. two div- ers are working all the time, which means the second bell goes down before the first one comes up and you have an on-bottom handle on these guys with a lot of production. m. gennser: What sort of scrubber material are you using, is it still soda lime? m. lothe: We have a couple of unique features and are bought off the shelf. Fortunately, the Us navy put money into develop- ment of the micropore stirrers which is soda lime based, packed in plastic with three channel gas lines that has a very high efficien- cy for Co 2 scrubbing while at the same time keeping the work of breathing very low. The rebreather is fitted with two of those which gives it an extremely low time constant work rate and we can use it as a passive rebreather instead of a push-pull system. m. gennser: Has the capacity of the micropore been tested down to 400 m? m. lothe: no, it has not yet been tested to 400 m as far as i know. We are not going to use it the rest of the way if we find it deficient at 225. O. molvaer: The hyperbaric rescue vessels seem very big, how heavy are they? J. gangenes: Perhaps 1900 kg. O. eftedal: are the displays outside or inside? i ask because this might be used for bubble monitoring through Doppler ultrasound. J. gangenes: inside, each diver has an individual unit, and we have used these screens for many years now. They can be used for whatever you like. R. moon: Can you actually calculate what the Pn 2 would be, is moon: DisCUssion ? 173 for the next 100 years. We should really try to outline what problems are possible to solve in the short term. my proposal is that we are going to try to repeat this event in three years, which should be a useful time span that, if this will lead to anything, will allow us to see a clear structure. The special focus and chal- lenge will be to review what we have done during that three year period. We think that we can establish a real cooperation and solve a problem that we think is important and produce something that is of value, scientifically and practically. That is what we had planned as an outcome off this particular meeting. C. Balestra: We were already planning to share a database with you to examine how to go forward with the recreational divers. is there something to be shared with the military diving community? m. Swiergosz: notwithstanding having a ?program,? our lead- ers often thrust an emerging requirement on us in a short time frame for which we must improvise. it might be good to work in small groups first to devise a strategy on solving the impor- tant problems. next month, i am bringing together some of the people we fund and some outside people to think out of the box to refresh us to devise a programmatic strategy to attack the scientific inquiry of oxygen toxicity. This could be done with any of the listed problems. R. moon: We have highlighted four of the best questions. We heard from our commercial diving colleagues about the impor- tance of ageing, diver selection, and trace gas toxicity. There is almost nothing written in the literature on those three. oxygen toxicity is a long-standing, extremely important problem and will probably remain so. m. lothe: one of the interesting parts of this event is that we have many distinguished members of the diving community from different countries around the world. too much time has been spent in the past on pushing your own agenda and believ- ing that what is going on in your own community is the only thing that matters. This has created unhealthy competition in who can be the most acceptable. From my point of view as a responsible manager for many different people in their everyday job, to go into these areas as a common effort to establish some worldwide acceptance of what is and what is not important in this community would be tremendously beneficial for every- body. From a commercial diving point of view, adhering to one set of standards versus over one hundred in the countries we operate in would give us the possibilities to change the costs of keeping up with all these national legislations and maybe use that in research. B. gardiner: in the future i would suggest a focus on physiol- ogy and biochemistry and perhaps genomics and psychology. What are the effects on people who spend long-duration time in isolation and perhaps join discussions with nasa on their ideas of people in space. a. Brubakk: Barring further comments or suggestions, i will close the Haldane symposium here and again thank the speakers for their contributed lectures on which they spent a lot of time and the audience for coming here to participate. there enough nitrogen in the lung and would it be sufficient to produce narcosis? Could it be that PCo 2 is part of the narcotic process? p. lindholm: The Co 2 could of course add to the narcosis. if you calculate the dive without lung collapse, yes, in those times you will reach high Pn 2 to get narcosis. it would not be possible however to really separate the o 2 toxicity from the Co 2 narcosis and the n 2 narcosis in this mix. C. Balestra: i wanted to add two things about Herbert nitsch?s way of making a record dive. What he is doing when he puts some air into the plastic Coca-Cola bottle, is he is sipping a bit of it to keep enough volume to be able to inflate his ears. no one can say if he is not rebreathing a bit. During rebreathing of even a few breaths, it is quite a long time of apnea you can do because you are not in apnea anymore. When he does his safety stop, i can accept that as a redistribution of blood coming from the blood shift, but not exactly for decompression. Claus muth pointed out that what he does is come out of the water and says everything is ok and then goes back down with oxygen. p. lindholm: We do not know if he needs that for decompres- sion. i also do not think he needs to rebreathe to keep his breath hold because it is just four and a half minutes. There are plenty of divers who actively swim down (the record being 130 m) with dive times also around four and a half minutes. C. Balestra: There is one paper published in the British Medical Journal on saxophone playing and the increased risk of mortal- ity. i have never seen in the medical literature a paper criticized as heavily as this one. The conclusion was that circular respira- tion is bad, and that was why saxophone players died before the others. There were 35 or so reactions to this paper with everyone arguing against it. J. Jakobsen: in norway, by law, we are required to monitor our diving activity. as was shown in an earlier presentation, we are shortly to be drowned in dive data. We must also save occupa- tional information about deep divers. Therefore, we must estab- lish an international exposure register. This is not to demonstrate that there are dangerous aspects to diving, but it is to be prepared to answer possible long-term effects of diving in the future. The exposure needs to be defined and analytical tools developed. a. Brubakk: our earlier discussions were not really about strat- egy and objectives for the future. as our list demonstrates, there are many things we can do. We need to ask which of these prob- lems are the most important and whether we have the tools and methodology to solve them. Personally, i would think that the problems of ageing, selection, oxygen toxicity, and toxicity in general are four areas where there are crucial questions, and there are methods that enable us to solve them. one important aspect about doing science is that there be a solvable problem and an available method. Establishing large databases without knowing what you are going to do with them is perhaps less im- portant than devising a strategy we can use in order to prepare appendix a: ?The Prevention of Compressed-air illness? (1908) John scott Haldane?s co-authorship of ?The Prevention of Compressed air illness? in 1908 was a major mid-career accomplishment among many of his other lifetime works that illustrate the breadth of his scientific output and his philosophical out- looks. The papers in this volume covered what has happened since Haldane?s publica- tion, appended herein, which focuses on the hyperbaric aspect of his contributions to environmental physiology. Haldane provided the baseline, a practical outcome for the diver, and a basis for the further development of decompression tables. all these developments have inspired the convenors of this celebratory symposium to reprint the original paper of 1908, the foundation of what followed in these proceedings more than a hundred years later. ?The Prevention of Compressed air illness,? by a. E. Boycott, g. C. C. Dam- ant, and J. s. Haldane, was originally published in the Journal of Hygiene 8, 3(June 1908):342?443. 176 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 177 178 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 179 180 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 181 182 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 183 184 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 185 186 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 187 188 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 189 190 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 191 192 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 193 194 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 195 196 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 197 198 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 199 200 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 201 202 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 203 204 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 205 206 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 207 208 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 209 210 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 211 212 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 213 214 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 215 216 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 217 218 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 219 220 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 221 222 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 223 224 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 225 226 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 227 228 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 229 230 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 231 232 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 233 234 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 235 236 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 237 238 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 239 240 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 241 242 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 243 244 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 245 246 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 247 248 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 249 250 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 251 252 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 253 254 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 255 256 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 257 258 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 259 260 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 261 262 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 263 264 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 265 266 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 267 268 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 269 270 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 271 272 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 273 274 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 275 276 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 277 278 ? tHE FUtURE oF Diving aPPEnDix a: ComPREssED-aiR illnEss (1908) ? 279 280 ? tHE FUtURE oF Diving appendix B: syposium speakers Bj?rn Alterhaug ntnU Costantino Balestra Haute Ecole Paul Henri spaak Jean-Eric Blatteau toulon arm?es, France Alf O. Brubakk ntnU Stephen Daniels Cardiff University David J. Doolette navy Experimental Diving Unit ?eljko Duji? University of split, Croatia Olav S. Eftedal statoilHydro asa ?yvind Ellingsen ntnU 282 ? tHE FUtURE oF Diving Andreas Fahlman naval medical Research Center Joar Gangenes acergy Mikael Gennser swedish Defence Research agency Michael L. Gernhardt nasa Marit Gr?nning Haukeland University Hospital Christian Gutvik ntnU Marianne B. Havnes ntnU John P?l Inderberg ntnU Susan R. Kayar Booz allen Hamilton Michael A. Lang smithsonian institution Peter Lindholm karolinska institutet Svein-Erik M?s?y ntnU aPPEnDix B: sYPosiUm sPEakERs ? 283 Einar Thorsen University of Bergen Andreas M?llerl?kken ntnU Richard E. Moon Duke University medical Center George A. Perdrizet morristown memorial Hospital Russell S. Richardson University of Utah Stig A. Sl?rdahl ntnU Matthew Swiergosz office of naval Research appendix C: symposium Participants SpEakErS Name affiliation email Bj?rn alterhaug ntnU bjorn.alterhaug@ntnu.no Costantino Balestra Haute Ecole Paul Henri spaak balestra@daneurope.org Jean-Eric Blatteau French navy je.blatteau@infonie.fr alf o. Brubakk ntnU alf.o.brubakk@ntnu.no stephen Daniels Cardiff University danielss@cardiff.ac.uk David J. Doolette navy Experimental Diving Unit (nEDU) david.doolette.as@navy.mil ?eljko Duji? University of split, Croatia zeljko.dujic@mefst.hr olav s. Eftedal statoilHydro asa oEFt@statoilhydro.com ?yvind Ellingsen ntnU oyvind.ellingsen@ntnu.no andreas Fahlman Woods Hole oceanographic institution fahlman@whoi.edu Joar gangenes acergy joar.gangenes@acergy-group.com mikael gennser swedish Defence Research agency mikaelge@foi.se michael l. gernhardt nasa michael.l.gernhardt@nasa.gov marit gr?nning Haukeland University Hospital marit.gronning@helse-bergen.no Christian gutvik ntnU christian.gutvik@ntnu.no marianne B. Havnes ntnU marianne.b.havnes@ntnu.no John P?l inderberg ntnU john.pal.inderberg@ntnu.no susan R. kayar Booz allen Hamilton susan.kayar.ctr@navy.mil michael a. lang smithsonian institution langm@si.edu Peter lindholm karolinska institutet peter.lindholm@ki.se svein-Erik m?s?y ntnU svein-erik.masoy@ntnu.no andreas m?llerl?kken ntnU andreas.mollerlokken@ntnu.no Richard E. moon Duke University medical Center moon0002@mc.duke.edu george a. Perdrizet morristown memorial Hospital george.perdrizet@athlantichealth.org Russell s. Richardson University of Utah rrichardson@ucsd.edu stig a. sl?rdahl ntnU stig.slordahl@ntnu.no matthew swiergosz office of naval Research matthew.swiergosz@navy.mil Einar Thorsen University of Bergen einar.thorsen@helse-bergen.no 286 ? tHE FUtURE oF Diving parTIcIpaNTS Name affiliation email arnfinn arnfinsen subsea 7 arnfinn.arnfinsen@subsea7.com John arne ask Ptil john.ask@ptil.no Jay Buckey Dartmouth College jay.buckey@dartmounth.edu Rune Djurhus nUi rdj@nui.no ingrid Eftedal st. olav?s Hospital, trondheim ieftedal@gmail.com Harald Engan mittuniversitetet, sweden harald.engen@miun.se lise Fismen nUi lfi@nui.no Bob gardiner marathonoil bgardiner@marathonoil.com svein Erik gaustad ntnU svein.e.gaustad@ntnu.no martin andre Heer acergy martin@heer.com astrid Hjelde ntnU astrid.hjelde@ntnu.no arvid Hope nUi aho@nui.no Thomas Huerligmann UWatEC thomas.huerligmann@uwatec.ch Jan Erik Jacobsen Deep-x jjacobsen@deep-x.com Dawn kernagis Duke University medical Center dawn.kernagis@duke.edu Darin knaus Creare inc dak@creare.com ingrid kruger ntnU ingridk@stud.ntnu.no Jason legore UWatEC jason.legore@uwatec.ch Berit linge subsea 7 berit.linge@subsea7.com Jasna ljubkovic medical school of split, Croatia jasna.ljubkovic@mefst.hr marko ljubkovic medical school of split, Croatia marko.ljubkovic@mefst.hr angelica lodin mittuniversitetet, sweden angelica.lodin@miun.se sjur lothe technip mlothe@technip.com Christian medby st. olav?s Hospital, trondheim christian.medby@stolav.no otto molv?r norwegian navy (ret.) marinmed@frisurf.no arvid P?sche Ergopro arp@ergopro.no scott Phillips Creare inc sxp@creare.com Jean michel Pontier French navy jm.pontier@free.fr kim vidar Rasdal ntnU kim-vidar.rasdal@ntnu.no matt Richardson mittuniversitetet, sweden matt.richardson@miun.se John Ross University of aberdeen j.a.ross@abdn.ac.uk Bjarne sandvik Ptil bjarne.sandvik@ptil.no martin D. J. sayer nERC scientific Diving Facility martin.sayer@sams.ac.uk k?re segadal nUi ksEg@statoilhydro.com kaj magnar sj?lie Havsport Dykkersenter as kaj@havsprot.no leif tore skjerven technip norway lskjerven@gmail.com martin skrove ntnU skrove@stud.ntnu.no Einar svendsen Hydro noryway (ret.) wosvends@online.no Ulrik Wisl?ff ntnU ulrik.wisloff@ntnu.no Smithsonian Institution Smithsonian Institution Sc holarly Pres s The FuTure oF Diving: 100 Years oF halDane anD BeYonD michael a. lang anD alF o. BruBakk eDiTors lang anD BruBakk Sm ithsonian Institution Scholarly Pres s The FuTure oF Diving 100 Years oF halDane anD BeYonD ISBN-13: 978-0-9788460-5-3 ISBN-10: 0-9788460-5-2 Cover photo By haNS WIld/tIme lIfe pICtureS/Getty ImaGeS. p.o. Box 37012 mrC 957 Washington, d.C. 20013-7012 www.scholarlypress.si.edu doWNload a pdf of thIS puBlICatIoN throuGh WWW.SCholarlypreSS.SI.edu. requeSt prINt CopIeS By emaIl: SChol_preSS@SI.edu