POSTGRAD. MED. J., (1966), 42, 378

INERT GAS NARCOSIS -AN INTRODUCTION1. P. UNSWORTH, B.M., B.Ch., L.R.C.P.,

Walton Hospital, Liverpool

INERT gas narcosis is the physical and mentaldisturbance, both sulbjective and objective, thatoccurs when breathing gas mixtures containingcertain members of the inert gases under pres-sure. These include krypton, argon, xenon, nitro-gen and possilbly neon. The phenomenon ofnarcosis associated with nitrogen has beenalternatively termed "nitrogen intoxication"though "narcosis" is the generally acceptedname. (Unsworth, 1960; Miles, 1962.)

It is more than a hundred years since thefirst report in print of the problem that has beenfacing divers ever since their descents wentdeeper than 100 feet or the ambient pressureincreased above 4 atmosiphere-nitrogen nar-cosis. For many years, it was only helmeteddivers and caisson workers who were 'sulbjectto this condition, but With the introduction dur-ing the last three decades of self-contalinedunderwater 'breathing apparatus, free divers areeven more liable to the serious and fatal sideeffects of nitrogen narcosis.The signiificance of inert gas narcosis has a

bearing not only on naval diving in warfare butalso undersea salvage and exploration. Withthe introduction of ihigh pressure chambers intothe field of anaesthesia and surgery, the in-herent properties of nitrogen narcosis may havewider implications, as, 'for example, the pos-si'ble use of nitrogen or other inert gas, underpressure as an anaesthetic agent, or as a side-effect when 'high pressures of compressed airare used.

HistoricalIn 1826 Colladon, a French physician, descriib-

ing a prolonged descent in a dliving bell, re-marked u'pon !his "state of excitement as if Ihad drunk some alcoholic liquor". It is doubt-ful if, 'in fact, Colladon did suffer from narcosisas the bell only descended to 20 metres althoughhis comparison with alcohol would suggest it.Another Frenchman, Junod, in 1835, noted thatunder pressure "the functions of the brain areactivated, imagination is lively, thoughts have a

peculiar charm and in some persons, symptomsof intoxication are present". It was not until1861 that a professional diver described hisown signs and sym,ptoms, and suggested a pre-cautionary measure. J. B. Green, an American,reported that on his deep dives of 150 feet ormore, 'he noticed a feeling of excitement fol-lowd by drowsiness, and he considered it im-portant that at this stage, the diver should bebrought u'p. From Green's full description itis clear that he noted hallucinatory changesand impairment of a diver's judgement. PaulBert (1878) remarked upon the objective signswhen nitrogen is 'breathed under pressure but atthat time did not pursue their cause though heestablished that dissolved nitrogen was theaetiological factor in decompression sickness orthe "bends".The next report of nitrogen narcosis came

from Damant (1930) d-uring the British Ad-miralty IDeep Sea diving trials to 320 feet.Damant described the findings of 'Hill and Selby,that although the dlivers were all 'picked menwho had 'been put through a specially searchingmedical examination, some of them becameabnormal mentally or emotionally. The effectwas attributed at first to the high ambient airpressure and oxygen poisoning, or to ilmpuritiessuch as carbon dioxide or monoxide. But care-ful work ruled these out, and though oxygenand carbon diox'ide were no 'longer held res-ponsilble, no incrimination of nitrogen was made.The first definiitive theory that nitrogen was

t-he aetiological factor in mental changes whileunder air pressure was put forward in 1935by Behnke, Thompson and Motley. They statedthat air at, and higher than, 3 atmospheres, pres-sure exerted a narcotic effect on man, witheuphoria, mental retardation and loss of neuro-muscular control, with coma intervening athigher 'pressures.

Since the end of the 'war, research on substi-tution diluents continued and in 1948 the oxy-hydrogen mix,ture was introduced (Zetterstrom:Burjstedt and Severin). Ten years earlier, how-

copyright. on 19 N

ovember 2018 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.42.488.378 on 1 June 1966. Dow

nloaded from

UNSWORTH: Inert Gas Narcosis

ever, Behnke and Yaribrough, and End (1938)had used helium as diluent in gas mixtures. Thiseliminated nitrogen narcosis from deep divesthough both hydrogen and helium had difficultfeatures, the former being explosive with ahigher oxygen concentration that 4 per cent,and the latter posing problems of decompres-sion, voice distortion and thermal conductivity.

Considerable experimental work has beencarried out on the qualitative and quantitativemeasurement of mental disaibility of nitrogenunder pressure, with that of Bennett and hisco-workers outstanding. They used both electro-encephalographic changes and flicker-fusionfrequencies in estimating time of onset of nar-cosis. ('Bennett and Glass, 1957; Bennett, 1958,Bennett and Cross, 1960.)Other workers attempted to ;relate changes

in performance efficiency to pressure of inertgases. (S'hilling and Wil'lgrube, 1937; Kiesslingand Maag, 1960; Frankenhaeuser, Graff-Lon-nevig and Hesser, 1963; Poulton, Catton andCarpenter, 1964.) These latter workers, in theirinitial experiments, found evidence of perform-ance impairment in su'bjects at only 2 atmos-pheres absolute but more recent work has notrevealed any abnormality lbelow 4 atmospheresabsolute.

Clinical FeaturesThe signs and symptoms of inert gas narcosis

are varialble in 'tiime 'of onset, aind in regardto the pressure at which it appears, individualsshow some effedt 'soon after reaching 4 'atmos-pheres absoluite. Sensitivity differs among diversand some may have demonstralble signs at Jesspressure while 'others m'ay inot be affected fti,lthe pressure reaches 5 or 6 atmospheres. Manyfactors influence the pressure at which the onsetof symp:oms may occur and ;the severilty ofthe at,ack. Exertion, prevlious faibigue, alcoholicexcess prior to 'a dive, even apprehension, mayboth advance 'onset and increase severilty. It alsoappears thait carbon difoxide retention, througheither ilneffecltive apparatus or an increased pro-portion of carbon dioxide in ;bhe respired gas,ailso speeds 'the onset of narcosis.The symptoms which generaitly, though not

i'nevxKJalbly, appear first are lightheadedness,dizziness, euphoria, aind apparent mental stimu-laltion associaited with great self-confidence. Thecompletion of a difficult underwater task willseem easy ..ald straightforward, and safety pre-cautior, usualily metlicul'ou'sly observed, appearto 'become superfluous. The subject shows agrea)t itendency to become jovial and 'alkative,a,nd easily amused. Later, mental concentrattion

becomes dtifficult and 'the recollcstion of ideasrequires great effort. Other symptoms noticed'include perilpherall numbness, lip 'trebling notundlike that of oxygen poisoning, and hallucin-a-tions. Occasionally ithis iiniltial ph,ase of nibrogenstimulation may be passed very rapidly andalmost ,unnldiced, ,the diverithen being affectedvery profoundly by the phase of exitremelethargy and drowsiness.The signs of nitrogen narcosis are most con-

veniently observed in a pressure chamber. Theseinclude delayed responses to sensory stimuliparticularly visual, aud'itory and tactile. Mis-takes are made in mental arithmetic that wouldnot normally 'be made on the surface. A lossof fine neuro-muscular contrcl and co-ordina-tion renders a delicate manipulation impossible.Although all individuals are narcotised to someextent under pressure, an emotionally stableperson reacts to the stress by increased effortand may carry out his task quite well untilconsciousness is 'lost. The unstaJble individualis incapable of any further purposeful effort.As the pressure is increased, the signs and

symptoms 'become more severe, there is lossof memory that may last for many 'hours, andbetween 10 and 13 atmospheres loss of con-sciousness occurs. With a helmet diver, thisis not so dangerous as wit-h a free diver in'whom the loss of the mouthpiece would resultin death 'by drowning.The similarity between the presenting features

of nitrogen narcosis and acute alcoholic poison-ing is very close and hence the first descrip-tion of Junod is so apt and observant. TheFrench name for the condition, "l'Ivresse", isthat used for -the vice of drunkenness. Anotherinteresting parallel between nitrogen narcosisand alcoholic poisoning is the variability andindividual susceptibility that exist. Resistanceto nitrogen narcosis (and alcohol) can be builtup 'by practice and experience, and adaptationdoes occur.

TheoriesMany theories have'been proposed to explain

inert gas narcosis. The first, put forward in1835 by Junod and later in 1881 by Moxon,suggested increased pressure alone as the factorinvolved. J'unod considered that "the increaseddensity of the air lessened the calibre of thevenous vessels, resulting in greater blood flowin the arterial system and towards principalnerve centres especially the brain, protected 'byits bony case from direct pressure". He thoughtthat this increased blood 'flow stimulated thenerve centres and resulted in narcosis. Moxon,

June, 1966 379copyright.

on 19 Novem

ber 2018 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.42.488.378 on 1 June 1966. D

ownloaded from

POSTGRADUATE MEDICAL JOURNAL

on the other hand, regarded the increased pres-sure as driving the blood from the surface ofthe body into parts not accessible to respiratoryexchange and that this devitalised blood causedemotional changes. These theories iare patentlyincorrect. If pressure did affect superficial bloodflow, cutaneous 'blanching would Tbe obviousbut this does not occur. The physical propertiesof the body are such that it may be regardedas incompressitble, with pressure being equalisedthroughout the body bulk. A further argumentagainst pressure alone being responsible forinert gas narcosis is the later demonstrationthat ibreathing gas mixtures of different con-stituents at the same pressures induced differentlevels of mental and physical change.The psychological aspect of deep diving was

once blamed as the cause of narcosis (Hill andGreenwood, 1906). When psyehological testswere carried out on divers who had failed tocomplete tasks under pressure, Plhillips (1931)showed that many of these subjects were of thesuppressed nervous type who habitually exer-cised control and that they suffered from latentsuppressed claustrophobia. Phillips had littlehesitation in ascribing the objective mentalchanges of his deep diivers to mental instability.However, if divers, or subjects in a pressurechamber, are supplied with different gas mix-tures at the same 'pressure, the severity of signsand symptoms may be altered. It is unl'ikelythat claustrophobia or other psychologicaldeviation contrilbute directly to inert gas nar-cosis although they may provide a backgroundof instability against which early changes ofnarcosis become more apparent.

Breathing oxygen at atmospheric pressurecauses sutbjective and objective reactions insome individuals and Birch (1859), reportingthese, maintained that they were not of psycho-logical origin. Thus the stimulating effect ofincreased oxygen tension in compressed air waspostulated as a cause of the "compressed air"syndrome. In 1878 Bert made this assumptionthat high oxygen tensions produced such effects,and he utilised ,high air pressures in his experi-ments into oxygen toxicity. Other workers sincethen (have been olf the same opinion (Binger,Faulkner and Moore, 1927; Smith, Heim,Thomson and Drinker, 1932).Not until Donald (1947) did anyone clearly

define the salient features of oxygen poisoningand when compared with those of nitrogenunder pressure, there should 'be no doubt thatthere exist two separate conditions. Damant(1930) was convinced that the subjective andobjective changes associated with air under

pressure were not due to oxygen. Recently theinterest in h'igh partial pressures of oxygen asa cause of narcosis has been re-aroused byFenn (1965). 'His hypothesis stems from someinteresting work done on the fruit fly, Droso-phila, in atmospheres of oxygen and nitrogenunder pressure. From his experiments hedemonstrated that a correlation exists betweenthe proportions of oxygen and nitrogen, andthe survival rates of the flies while under pres-sure. Though he states that his results are notdirectly applicable to human diving, he doessuggest that it may be possiible by keeping theoxygen tension low or normal to avoid the nar-cotic effects of nitrogen at depth. This wouldcertainly substantiate the claims of Albert Buhl-man and Hans Keller (1962) that nitrogennarcosis does not exist, even with 90 per centnitrogen, at 600 feet. But, although Fenn sug-gests a lowering of -the oxygen partial pressure,this has been strongly criticised by manyworkers as dangerously incorrect. The effectof decreasing t'he Po2 without substituting athird gas, would 'be to increase the inert gastension, and thereby markedly potentiate nar-cosis.

Narcosis experienced under compressed airwas attributed to nitrogen by Behnke, Thomp-son and Motley (1935). The manner in whichnitrogen acted was considered to be due to itsoil/water solubility ratio, acting in the samemanner as the diphasic anaesthetic agents. Thecorrelation between solubility of anaestheticagents in water and lipid and its depressantaction had been described much earlier byMeyer (1899). Behnke and his co-workers foundthat if nitrogen is replaced in a gas mixture byhelium, the narcotic effect is lessened orabolished. Conversely if argon is sutbstituted,the narcosis is more profound and with earlieronset. Such gases as krypton and xenon havealso been used in mixtures at atmospheric pres-sure and have produced definite central nervoussystem depression and anaesthesia in bothanimals and man. ('Lawrence, Loomis, Tobiasand Turpin, 1946; Cullen and Gross, 1951; Car-penter, 1953). The ease with which a gas pro-motes narcosis under pressure is directlyproportional to its fat soluibility and oil/waterdistribution coeflicient (Table 1), and this ledearly workers to use hydrogen or helium asdiluent in gas mixtures to be used under pres-sure, as narcosis only occurs with these twounder extremely high pressures.A further agent once thought to 'be the cause

of the narcosis associated w'ith inert gases iscarbon dioxide. The narcotic and anaesthetic

380 June, 1966copyright.

on 19 Novem

ber 2018 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.42.488.378 on 1 June 1966. D

ownloaded from

June, 1966 UNSWORTH: Inert Gas Narcosis 381

TABLE 1GAS SOLUBILITIES

Solubilities in mg./ml. solvent at 370C.

Gas Water Fat Distribution coefficientArgon 0.027 0.140 5.2Nitrogen 0.013 0.067 5.1Hydrogen 0.017 0.036 2.1Heliam 0.009 0.015 1.66

TABLE 2GAS DENSITIES

Hydrogen- Helium- Nitrogen- Argon-oxygen oxygen oxygen (air) oxygen

Viscosity relativeto air 0.35 0.66 1 1.13

Wt. in lbs. of1 cu. ft. at 1 atmo. 0.022 0.026 0.081 0.193

effects of this gas were noted by many earlyworkers i(Bert, 1878; Hill and Flack, 1908).Case and Haldane (1941) found that carbondioxide added to compressed air en'hanced thenarcotic effect, and noted the increase in res-piration associated with a rise in partial pres-sure of CO2 in compressed air. Bean (1945;1950) has put forward the most detailed theoryand he considers CO2 an important contrilbutorto, if not chief cause of, those reactions attri-buted to nitrogen. Bean postulates that com-pression increases the partial pressure of carbondioxide by interfering with t'he dynamics of airflow within the respiratory tree, and leads tohypercapnia. As pressure increases, air densityrises and ventilatory efficiency is reduced. AsMiles (1957) has shown, if air is breathed at 200feet, the maximum breathing capacity is re-duced by 50 per cent, and at 600 feet, thereduction is 75 per cent.A study by Bean '(1950) on the rapid com-

pression of anaesthetised dogs showed a pro-nounced increase in alveolar Pco2. The experi-mental data'put forward, though limited, wereconsidered to provide evidence that changes inalveolar CO2, and alterations of blood andtissue CO2 caused, or contributed to, nitrogennarcosis. However, one point made by Beanto explain t'he lack of hyperventilation in res-ponse to high blood Pco2 was that the responseto a given C02 increase in compressed air maynot be the same as the response at normal pres-

sure. It is true that a large excess of CO2may diminish rather than augment ventilation,and that an increased partial pressure of oxygenmay modify the response to carbon dioxide(Dripps and Dumke, 1943) bu't the Pco2 increasemay not be large (Rashbass, 1955) or alterna-tively the Po2 may not ibe greatly raised. It ispossible that compression using nitrogen causesa desensitisation to carbon dioxide, and there-fore CO2 retention with no hyperventilation.The increased density and viscosity of air, andother gas mixtures i(Table 2), contribute to thediminution of effective pulmonary ventilation.When carbon dioxide is added to the inspiredmixture, this has an additive effect on nitrogenor inert gas narcosis. It 'has been suggested thatthe vasodilatation of increased partial pressuresof carbon dioxide, particularly on the cerebralcirculation, allows the effects of high oxygenpartial pressure to become apparent and thatthe picture under these circumstances is a com-bination of nitrogen narcosis and oxygentoxicity. Work of Hesser and his colleagues(1963) has shown that oxygen excess has apotentiating effect on nitrogen narcosis, andthat performance efficiency decreased with in-creasing oxygen pressure. It was suggested thatthis increase in narcotic effect was possijbly dueto interference with carbon dioxide eliminationfrom the tissues, one of the factors involvedin simple oxygen toxicity. Thus raised partialpressures of oxygen increases that of carbon

copyright. on 19 N

ovember 2018 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.42.488.378 on 1 June 1966. Dow

nloaded from

382 POSTGRADUATE MEDICAL JOURNAL

dioxide and nitrogen and carbon dioxide havea synergistic narcotic action.

Bii'hlman (1963) be!lieved he could explainthe sensations and difficulties attributed tonitrogen narcosis by altered respiratory physio-logy, mainly CO,2 retention due to hypoven-tilation. His protege Hans Keller has dived to600 feet using '90 per 'cent nitrogen withoutapparent signs of narcosis. However, several ofBiuhlman's divers other than Keliler, comiplainedof abnormal sensations under high pressures ofnitrogen and argon. Biuhlman suggested that,under pressure, some specific and reversible gasconfiguration formed in the brain, dependent,not on the nature, ibut on the density of thegas, a view not incompatible with the nitrogenhypothesis.Of the several agents proposed to explain

pressure narcosis, nitrogen, or other inert gas,has been most widely accepted. Carbon dioxideretention will contribute to the severity of signsand symptoms initiated by an inert gas but itis doubted if it alone produces the clinicalpicture.

Electrical StudiesIn an attempt to discover upon which part

of the nervous system nitrogen and the inertgases acted, Bennett and his co-workers (1957;1960) investigated t'he EEG patterns of humansubjects under pressure. Earlier experiments(Marshall, 1951; Jullien, Roger and Chatrian,1953) had shown that EEG waves were alteredby increased air pressure. The main and mostimportant finding by Bennett and Glass wasthe abolition of alpha-wave blocking. Thisoccurrence of alpha-blocking is found in 50 percent of normal subjects at atmospheric pressurewhen the subject concentrates on mental arith-metic or similar problem. It was found that ifthe subject was exposed to a high enough pres-sure for a long enough ti'me, the alpha-blockingwas abolished. The time from beginning ofexposure to pressure till abolition was foundto be inversely proportional to the square ofthe pressure. The relationship between time tothe abolition of desynchronisation of alpharhythm and pressure has suggested a nitrogenor inert gas diffusion gradient into some partof the central nervous system (Hempleman,1952) producing impairment of m e n t a 1efficiency. If after abolishing alpha-blocking, anoxy-helium mixture is substituted for the nitro-gen-oxygen mixture, alpha-blocking will re-appear. This would suggest nitrogen as theagent responsible, as any concurrent signs and

symptoms of narcosis would be alleviated atthe same time.An alternative method of determining the on-

set of changes in cortical activity in subjectsunder pressure has been used by Bennett andCross (1960). They showed that using fusionfrequency of a flickering neon light and electro-encephalography, the time to abolition of alpha-blocking and maintained fusion of flicker arethe same. This method of flicker fusion may beused in those individuals who do not showalpha-blocking even at atmospheric pressure.

Site of ActionThe site of action where inert gases under

pressure induce narcotic change is still notentirely clear. It seems probable that the cortexis not affected directly 'but via some part ofthe brain stem. One site that has been suggestedis the reticular formation of the mid-brain andhypothalamus, a neuronal system connectedwith consciousness (Magoun, 1952; French, Ver-zeano, Magoun, 1953). Magoun showed thatstimulation of t'his activating system rousedan animal from sleep, converting the EEG pat-tern from slow sleep waves to more rapid lowervoltage waves of activity. He also demonstratedthat the electrocorticogram was blocked uponstimulation of the reticular formation. It isthought that the action of high pressure nitro-gen and other inert gases is upon the reticularformation, and the time to abolish alpha-block-ing may be due to gaseous diffusion throughoutthe brain stem and thalamus.On a cellular level, it is less easy to define the

site of action of inert gases as narcotic agents.The reticular formation is an extremely complexsystem of neurones, short nerve fibres andsynapses, and inert gas molecules under pres-sure may affect any one or all of these con-stituents.

Marshall (1951), using nitrogen and argon upto 96 atmospheres, could not show any effecton 'isolated frog sciatic nerve, and no effecton frog nerve-muscle preparations with nitro-gen at 82 atmospheres. But she did find thatspinal synapses and therefore reflexes were sen-sitive to inert gas pressures. More recently,Gottlieb and Weatherley (1965) confirmed thathigh pressures of helium, neon, nitrogen andargon up to 15 atmospheres had no effect ontransmission across neuromuscular junctions oralong nerve fibres, in frog sciatic nerve-gastrocnem'ius muscle preparations. Work ofFrench and others (1953), points to the synapsesof the reticular formation as possible sites ofinterference with normal conduction, and has

382 June, 1966copyright.

on 19 Novem

ber 2018 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.42.488.378 on 1 June 1966. D

ownloaded from

UNSWORTH: Inert Gas Narcosis

shown that synaptic transmission in the centralnervous system is more susceptible to the effectsof narcotics than the conduction process innerve fibres, and some memlbers of the inertgases may be included among the narcotics.Bennett (1964) using auditory provoked stimulirecorded from the cortex of cats, has suggestedthat the most likely sites for blockage are thecentral synapses, and that the level of narcosisis related to a critical concentration of inertgas molecules at t'hese sites.A further theoretical site of action of nitro-

gen is within the neurone itself or within theneurone/glial unit which has been demonstratedso well by Hyden (1962). He describes how theglial supporting cell entirely covers the neuroneand acts as the barrier between the neurone andthe capillary, and the two constitute a bio-chemical and functional unit. Even the prota-gonists of the theory that the neurone is affectedby nitrogen, are divided as to the exact mechan-ism. There are those who support the lipidtheory originated iby H. H. iMeyer in 1899 andmodified by Clements and Wilson (1962) tostate that nitrogen under pressure acts at lipidinterfacial films within living cells. The an-tagonists of the lipid theory (Pauling, 1961;Miller, 1961) claim there is a protein-bindingaction with formation of microcrystals withinthe cytoplasm.Clements and Wilson believe that inert gases

can interact significantly with the interfaciallipoprotein of living cells altering both perme-ability and enzyme relationships, and thus inter-fering with oxidative p1hosphorylation andelectron 'transport whlich probably areassociated with the llipoproteins of themitochonidria. They suggest an histotoxicanoxia, a view supported by Russek(1962), Miles (1962), Bennett (1963), andearlier work of Ebert, Hornsey and Howard(1958) on the effect of irradiation on growthof bean shoots in inert gases. It was suggestedby Ebert and his colleagues that oxygen wasresponsible for radiosensitivity, and displace-ment of oxygen from sites within the cell bynitrogen reduced this sensitivity. Using an anti-psychotic agent Frenquel, Bennett (1963)showed a protective action in rats not onlyagainst inert gas narcosis but also oxygenpoisoning. Bennett suggests that nitrogen nar-cosis is an histotoxic hypoxia, increasing themetabolic work of the central nervous systemin the production of central inhibition.Hyden describes the biochermical and func-

tional interrelationship ibetween the neuroneand its supporting element, the glial cell. He

showed that the glia provides a maintenancesystem for the neurone and acts as the blood/brain barrier, the two forming a close unit,anatomically, histologically and metabolically.When he suibjected preparations to atmospheresof low oxygen tension, the glial cell convertedfrom aerobic to anaerobic production of energyresulting in a drop of efficiency from 55 percent to only 3 per cent. This permitted theneurone to utilise all available oxygen. Hydenpointed out that the glial cell was composedof 70 per cent lipoid material while the neuronewas only 5 per cent. Thus the glial cell mightbe expected to atbsorb more nitrogen than theneurone and be more affected by any deranginginfluence. Should this occur, oxygen consump-tion of the glia will drop as will its efficiency.This depletion of energy may cause delay inionic potassium and sodium exchange with theneurone, to delay in suibstrate transfer and ulti-mately to cessation of adequate function ofneurone and glia. This would support the his-totoxic hypoxia and lipid theories as proposedby Clements and Wilson, and other workers. Asimilar theory of decreased membrane perme-albility was proposed by Mullins (1954) althoughhe argued that it was inert gas molecules accu-mulating in the membrane pores that hinderedpermeability of ionic material. It may be thatthe point at which aerobic respiration withinthe glia gives way to anaerobic glycosis coin-cides with the abolition of alpha-blocking onthe EEG.The argument that the brain consists largely

of water and is, therefore, unlikely to be influ-enced by highly fat-soluble compounds is putforward by antagonists of the 'lipid theory ofnarcosis and anaesthesia, mainly Pauling (1961).He offers, as alternative, a theory of micro-crystal formation within cell cytoplasm quiteindependent of lipid cell membranes or otherstructures. Pauling suggests that nitrogen andother inert gases such as xenon, take part in theformation of clathrates, in which the gas atomsoccupy chambers in a framework of molecules.These molecules interact with one anotherthrough hydrogen bonding to give nitrogen orxenon hydrate. The clathrates to be present atbody temperature must be stalbilised by cyto-plasmic proteins, hence this theory is sometimesreferred to as "protein binding". Pauling pos-tulates that these microcrystals act in two waysby trapping electrically charged ions asso-

ciated with impulse conduction and dampingdown electrical circuits, and 'by preventing closeenough contact of enzyme/suibstrate configura-tions and thus decreasing the rate of chemical

lune, 1966 383copyright.

on 19 Novem

ber 2018 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.42.488.378 on 1 June 1966. D

ownloaded from

384 POSTGRADUATE MEDICAL JOURNAL June, 1966

reactions and, therefore, metabolic rate of cells.A somewhat similar theory has been proposedby Miller (1961) but Which does not involvethe formation of clathrates. Featherstone andSchoen'born (1964) have reviewed the bio-physical aspects of both lipid- and protein-binding theories proposed and they concludethat there is no clear evidence that either pre-ponderates but that probably a mutual relation-Thip exists.

ConclusionsAlthough the narcosis from inert gases under

pressure may be avoided by the use of sub-stitution mixtures, the academic investigationas to the exact cause of this phenomenon con-tinues. The use of hydrogen, helium and pos-sibly neon, as diluents, and of anti-narcoticdrugs, as yet theoretical, have been investigatedand help to reduce the incidence of narcosis.The elucidation of inert gas narcosis being his-totoxic anoxia remains as the latest of muchwork to be done on problems associated withnitrogen and other inert gases under pressure.

SummaryInert gas narcosis is a condition affecting

the physical and mental state of subjects breath-ing air or mixtures containing certain of theinert gases at pressures greater than 4 atmos-pheres. The signs and symptoms are describedand possible theories advanced, from Which themost tenable are nitrogen, or other inert gas,and carbon dioxide retention. The protein andlipid binding properties of inert gases underpressure are mentioned. Finally, several sitesof action are suggested, among which are thecentre synapses and the neuroglial cell.The author wishes to thank Professor T. C. Gray,

Surgeon-Calptain S. Miles, R.N., and especially Dr.P. B. Bennett, for their advice and help.

REFERENCESBEAN, J. W. (1945): Effects of Oxygen at Increased

Pressure, Physiol. Rev., 21, 1.BEAN, J. W. (1950): Tensional Changes of AlveolarGas in Reactions to Rapid Compression and De-compression and Question of Nitrogen Narcosis,Amer. J. Physiol., 161, 417.

BEHNKE, A. R., THoMPSON, R. M., MOTLEY, E. P.(1935): Psychological Effects from Breathing Air atFour Atmospheres Pressure, Amer. J. Physiol.,112, 554.

BEHNKE, A. R., and YARBROUGH, 0. D. (1938):Physiologic Studies of Helium, U.S. nav. med. Bull.,36, 542.

BENNETT, P. B., and GLASS, A. ((1957): High PartialPressures of Nitrogen and Abolition of Blockingof the Occipital Alpha Rhythm, J. Physiol., 13S8,18P.

BENNETT, P. B. (1958): Flicker Fusion Frequencyand Nitrogen Narcosis. A comparison with E.E.G.Changes and the Narcotic Effect of Argon Mix-tures, M.R.C. (RNPiRC) Report U.P.S. 176.

BENNETT, P. B., and CRoss, A. V. E(1960): Alterationsin the Fusion Frequency of Flicker Correlatedwith Electroencephalograph Changes at IncreasedPartial Pressures of Nitrogen, J. Physiol. (Lond.),151, 28P.

BENNETT, P. B. ((1963): Prevention in Rats of Nar-cosis Produced by Inert Gases at High Pressures,Amer. J. Physiol., 205, 1013.

BENNETT, P. B. (1964): The Effects of High Pressuresof Inert Gases on Auditory Evoked Potentials inCat Cortex and Reticular Formation, Electroenceph.clin. Neurophysiol., 17, 388.

BERT, P. (1878): La Pression Barometrique. Paris:Masson.

BINGER, C. A. L., FAULKNER, J. M., and MOORE,R. L. (1927): Oxygen Poisoning in Mammals, J.exp. Med., 45, 849.

BIRCH, S. B. (1859): On Oxygen as a TherapeuticAgent, Brit. med. J., ii, 1033.

BUHLMAN, A. (1963): "Deep Diving", p. 52, Proc.Second Congress Underwater Activites London.

BURJSTEDT, H., and SEVERIN, G. (1948): The Preven-tion of Decompression Sickness and Nitrogen Nar-cosis by Use of Hydrogen as a Nitrogen Substitute,Milit. Surg., 103, 107.

CARPENTER, F. G. (1953): Depressant Action of InertGases on the Central Nervous System in Mice,Amer. J. Physiol., 172, 471.

CASE, E. M., and HALDANE, J. B. S. i(1941): HumanPihysiology under High Pressure, J. Hyg. (Lond.),41, 225.

CLEMENTS, J. A., and WILSON, K. M. (1962): TheAffinity of Narcotic Agents for Interfacial Films,Proc. nat. Acad. Sci. (Wash.), 48, 1008.

COLLADON, M. (1826): Relation d'une Descqnte enMer dans la Cloche du Plongeur. Paris.

COUSTEAU, J. Y. (1954): The Silent World. London:Hamish Hamilton.

CULLEN, S. C., and GROSS, E. G. (1951): The Anes-thetic Properties of Xenon in Animals and HumanBeings with Additional Observations on Krypton,Science, 113, 580.

DAMANT, G. C. C. (1930): Physiological Effects ofWork in Compressed Air, Nature (Lond.), 126, 606.

DONALD, K. W. (1947): Oxygen Poisoning in ManBrit. med. 1., i, 172.

DRIPPS, R. D., and DUMKE, P. R. (1943): Effects ofNarcotics on Balance between Central and Chemo-receptor Control of Respiration, J. Pharmacol. exp.Ther., 77, 290.

EBERT, M., HORNSEY, S., and HOWARD, A. (1958):Effects on Radiosensitivity of Inert Gases, Nature(Lond.), 181, 613.

END, E. (1938): Use of New Equipment and HeliumGas in World Record Dive, J. Indust. Hyg., 20, 51 1.

FEATHERSTONE, R. M., and SCHOENBORN, B. P. (1964):Protein and Lipid Binding of Volatile Anaesthetics,Brit. J. Anaesth., 36, 105.

FENN, W. 0. (1965): Inert Gas Narcosis, Ann. N.Y.Acad. Sci., 117, 760.

FRANKENHAEUSER, M., GRAFF-LONNEVIG, V., andHESSER, C. M. (1963): Effects on PsychomotorFunctions of different Nitrogen-oxygen Gas Mix-tures at Increased Ambient Pressures, Acta. physiol.Scand., 59, 400.

FRENCH, J. D., VERZEANO, M., and MAGOUN, H. W.;(1953): Neural Basis of the Anesthetic State, Arch.

copyright. on 19 N

ovember 2018 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.42.488.378 on 1 June 1966. Dow

nloaded from

Junte, 1966 UNSWORTH: Inert Gas Narcosis 385

Neurol. Psychiat. (Chic.), 69, 519.GOTrLIEB, S. F., and WEATHERLY, J. M. (1965):

PIhysiological Effects of the Noble Gases on FrogSciatic Nerve and Gastrocnemius Muscle, Amer.J. Physiol., 208, 407.

GREEN, J. B. (1861): Diving With and WithoutArmour. Buffalo: Leavitt.

HEMPLEMAN, H. V. (1952): A New Theoretical Basisfor the Calculation *of Decompression Tables,M.R.C. (RNPRC) Report U.P.S., 131, A.

HILL, L., and FLACK, M. (1908): The Effect of Ex-cess of Carbon Dioxide and of Want of Oxygenupon the Resipiration and the Circulation, J.Physiol. (Lond.), 37, 77.

HILL, L., and GREENWOOD, M. (1906): The Influenceof Increased Barometric Pressure on Man, Proc.roy. Soc., B., 77, 442.

HYDEN, H. (1962): The Neurone and its Glia-aBiochemical and Functional Unit, Endeavour, 20,144.

JULLIEN, G., ROGER, A., and CHATRIAN, G. E. (1953):Preliminary Report on Variations of the E.E.G.of the Cat at Various Air Pressures, Riv. Neurol.,23, 357.

JUNOD, T. (1835): Recherches sur les Effets Pihysiolo-giques et Therapeutiques de la Compression et dela Rarefaction de I'air, taut sur le Corps que lesMembres Isoles, Ann. gen. de Med., 2nd Series, 9,157.

KIESSLING, R. J., and MAAG, C. H. (1960): Per-formance Impairment as a Function of NitrogenNarcosis, U.S. Navy Experimental Diving UnitResearch Report, 3-60, 1.

LAWRENCE, J. H., LOOMIS, W. F., TOBIAS, C. A., andTURPIN, F. H. (1946): Narcotic Effect of Xenon.Preliminary Observations on the Narcotic Effect ofXenon with a Review of Values for Solubilities ofGases in Water and Oil, J. Physiol (Lond.), 105,197.

MAGOUN, H. W. (1952): The Ascending ReticularActivating System, Res. Publ. Ass. nerv. ment. Dis.,30, 480.

MARSHALL, J. M. (1951): Nitrogen Narcosis in Frogsand Mice, Amer. J. Physiol., 166, 699.

MEYER, H. H. (1899): Zur Theorie der Alkoholnar-kose, Arch. exp. Path. Pharmak., 42, 109.

MILES, S. (1957): The Effect of Changes in Baro-metric Pressures on Maximum Breathing Capacity,J. Physiol. (Lond.), 137, 85P.

MILES, S. (1962): Underwater Medicine. London:Staples Press.

MILLER, S. L. (1961): A Theory of Gaseous Anes-thetics, Proc. nat. Acad. Sci. '(Wash.), 47, 1515.

MoxoN, W. (1881): Croonian Lectures on the Influ-ence of the Circulation on the Nervous System,Lancet, i, 565.

MULLINS, L. J. (1954): Some Physical Mechanismsin Narcosis, Chem. Rev., 54, 289.

PAULING, L. (1961): A Molecular Theory of GeneralAnesthesia, Science, 134, 15.

PHILLIPS, A. E. (1931): Recent Research Work inDeep Sea Diving, Proc. roy. Soc. Med., 25, 693.

POULTON, E. C., CARPENTER, A., and CATIN, K. J.(1963): Mild Nitrogen Narcosis?, Brit. med. J.,ii, 1450.

POULTON, E. C., CATroN, M. J., and CARPENTER, A.(1964): Efficiency at Sorting Cards in CompressedAir, Brit. J. Industr. Mcd., 21, 242.

RASHBASS, C. (1955): The Unimportance of CarbonDioxide as a Cause of Nitrogen Narcosis, M.R.C.(RNPRC) Report U.P.S., 153.

RUSSEK, M. (1962): International Congress. Series47, p.l87, London: Excerpta Medca Founda:tion.

SHILLING, C. W., and WILLGRUBE, W. W. (1937):Quantitative Study of Mental and NeuromuscularReactions as Influenced by Increased Air Pressure,U.S. nav. med. Bull., 35, 373.

SMITH, F. J. C., HEIM, J. W., THOMPSON, R. M., andDRINKER, C. K. (1932): 'Bodily Changes and Devel-opment of Pulmonary Resistance in Rats Livingunder Compressed Air Conditions, J. exp. Med.,56, 63.

UNSWORTH, I. P. (1960): Nitrogen Narcosis, SomeAspects, St. Mary's Hosp. Gazette, 66, 272.

ZETrERSTROM, A. (1948): Deep Sea Diving with Syn-thetic Gas Mixtures, Milit. Surg., 103, 104.

copyright. on 19 N

ovember 2018 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.42.488.378 on 1 June 1966. Dow

nloaded from