Decompression Sickness: AReviewdecompression and gave credit to Pol and 2 Watelle (I854) as the first authors who had attempted to explain decompression accidents. Pol and Watelle

Brit. J. industr. Med., I968, 25, 4.

Decompression Sickness: A ReviewR. I. McCALLUM

From the Nuffield Department of Industrial Health, University of Newcastle upon Tyne

Work in compressed air is a dangerous activity,whether it is carried on under water in a con-ventional diving suit, in self-contained breathingapparatus or in a diving bell, or in relatively dryconditions in a caisson or an underwater tunnel.Surprisingly large numbers of men have worked atone time or another in compressed air in tunnels orcaissons and it is the dangers which arise to thehealth of these men that are considered here.When a tunnel is driven through water-logged

strata or through porous ground under a river, theopen end of the tunnel must be sealed off and airpumped in to balance the hydrostatic pressure,should there be any danger of water flooding in andbringing with it unmanageable quantities of silt orsand. A pier of a bridge may be constructed indeep water by means of a caisson (Fr. caisse, a box),which is a compressed air chamber in which a gangof men excavate foundations. In both tunnels andcaissons men come and go through an air lock andmust be compressed on entering and decompressedat the end of the work period. Although workingconditions in these circumstances have much im-proved over the last 6o years, our understanding ofthe reactions of the human body to atmosphericpressures greater than the normal and of thepathogenesis of decompression sickness is stillinsufficient to prevent illness completely.

In engineering practice it is customary for work-ing pressure to be expressed as gauge pressure inpounds per square inch (p.s.i.g.) which is thepressure over and above the normal atmosphericpressure of I4-7 p.S.i. (I 03 kg./cm.2). For certaincalculations absolute pressure (gauge pressure plusatmospheric pressure) is used, and for clarity itmust always be stated whether gauge or absolutepressure is referred to. In naval practice it is usualto think in terms of feet of sea water, each 33 feetof depth corresponding to about one atmosphere orapproximately I5 p.s.i. Sometimes the unit ofpressure used is the atmosphere (atm.). On the

Received for publication May 8, I967.

European continent pressures are expressed inkilogrammes per square centimetre (kg. /cm.2).Although the use of these different units can beconfusing, it is not difficult to convert one intoanother.Decompression procedures used for deep sea

divers are generally held to be safer and to result inmuch less decompression sickness, particularly bonedisease, than those used for tunnel and caissonworkers. It is often asked why tables and proceduressimilar to those used in diving are not applied incivil engineering work, and to answer this a briefdescription of the particular circumstances affectingsuch work must be given.

Civil Engineering Projects

The practical and human problems in caissonand tunnel work are quite different from thoseencountered in deep sea diving. These civilengineering projects occur sporadically and irreg-ularly; they may be situated anywhere in the BritishIsles and may be reduced or may disappear duringtime of war or economic difficulty.A high turnover of labour is characteristic of

civil engineering compressed air work, and in acontract which at any one time employs two or threehundred men in compressed air, many more thanthis number will have been exposed to compressedair for greater or lesser periods by the time thecontract is completed (Table I). For example, atthe Dartford Tunnel (Golding, Griffiths, Hemple-man, Paton, and Walder, I960), over a period oftwo years during which between 250 and 320 menwere on the active compressed air list at any giventime, I,200 men actually worked in compressed airat one time or another. A very high proportion ofmen attracted to this type of work by the high rateofpay do not remain at it for more than a few weeks,probably because it is not only dangerous butphysically hard. The Work in Compressed AirSpecial Regulations, I958 (Ministry of Labour andNational Service, I958) ensure that fit young men

4

on July 30, 2021 by guest. Protected by copyright.

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.25.1.4 on 1 January 1968. D

ownloaded from

http://oem.bmj.com/

Decompression Sickness: A Review

are selected from those new to the work and thatolder experienced men are also in good health.Nevertheless the general standard of health must besomewhat lower than that of naval divers. Com-pared with naval divers, compressed air workers arepoorly trained in decompression procedures andare undisciplined; they are often itinerant, livingaway from home in lodgings or camps, and tend todrink heavily. If more than one contract is beingcarried out in the country at the same time, menmay move from one site to another at short noticeif they know that the pay is better. This mobility,variations in the level of working pressure, and theoccurrence of strikes may lead to loss of acclim-atization to work at high air pressure and con-sequently to a greater risk ofdecompression sickness.The difficulty of identifying individual workers whomove from contract to contract is well illustrated bythe files of the Central Registry of Compressed AirWorkers at the University of Newcastle upon Tynewhich contain records of 42 different men namedGallagher, I4 of whom have Patrick as their firstname and 38 of whom have at some time beenemployed at the same work site (Griffiths, personalcommunication).The work is heavy labouring, usually for an

eight-hour shift at pressures which may on occasionreach a legal maximum of 50 p.s.i.g., although themajority of work is carried out at much lowerpressures. Work in compressed air may go on formany months, especially if constructional difficultiesare encountered or the undertaking is large. In thetwo tunnels under the River Clyde, one ofwhich wasopened for use in July I963 and the other in MarchI964, work in compressed air went on for aboutthree and a half years between May I959 andJanuaryI963 (Haxton and Whyte, I965). In deep diving, onthe other hand, although pressures are very muchhigher than those used in caisson or tunnel work,the periods of exposure to pressure are com-paratively brief. The total period of time for whicha diver is engaged on a particular project is oftenquite short, perhaps a few days or weeks, and asmall number of highly trained and experiencedmen is involved.

Gas Bubbles as the Cause of DecompressionSickness

Men have worked in diving bells, diving suits,and caissons in pressures over 4 atmospheresabsolute since the sixteenth century, with manyserious accidents (Bert, I878). Bert made a cleardistinction between the effects of compression anddecompression and gave credit to Pol and2

Watelle (I854) as the first authors who hadattempted to explain decompression accidents.Pol and Watelle neatly summed up the mainproblem in compressed air work in the phrase 'onne paye qu'en sortant'. Bert showed that decom-pression sickness was related to the appearanceof nitrogen bubbles in the blood and tissues follow-ing rapid decompression, and that very slowdecompression would prevent symptoms occurring.He advised immediate recompression for treatmentof symptoms.

His observations were based largely on detailedand painstaking animal experiments, but his des-cription of decompression accidents in humans isaccurate and still relevant. He concluded thatpost-decompression phenomena depended for theirintensity on the height of pressure to which menhad been exposed and to the rapidity of decom-pression; that up to 2 atmospheres absolute thereappeared to be no ill effects but that above this levelcutaneous lesions and limb pains appeared moreand more frequently, and that it was not until over3 atmospheres absolute pressure had been reachedthat the very serious accidents occurred. Bert drewattention to the variation found between differentindividuals in the effects of decompression and alsoin the same persons in different and ill-understoodcircumstances. His animal experiments includedthe use of reduced pressures as well as increasedpressures, the toxic effects of oxygen at increasedpressures, and explosive decompression. Althoughit is now universally accepted that gas bubbles arethe cause of decompression sickness, the exactmechanism by which pains are produced and thesite of the lesion are still unknown, and the role ofthe bubbles in aseptic necrosis of bone has not beendemonstrated in man or in animals.

Decompression Procedure

Until the early part of this century men werehabitually decompressed very rapidly, taking onlya few minutes even after several hours at highpressure, and severe symptoms, gross disablement,and deaths were common (Hill, I9I2). Compressedair was first used in tunnelling at the Hudson River,New York in I879 where a maximum pressure of35 p.s.i.g. was reached, but no complete medicalrecord of this contract was made. Ryan (I929)quotes a figure of I2 deaths in a year in 50 workmendriving the Hudson River tunnel in I890, andBoycott (I906) gives a mortality rate of 25% perannum for the same undertaking. After a medicallock was introduced the death rate dropped toabout i%.

5


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

TABLE ICOMPARISON OF VARIOUS COMPRESSED AIR CONTRACTS

Total No.Period of Daily Work of Workers Maximum No. of

Contract Compressed Force in Pressure Shift Length DecompressionsAir (mths) (Compressed Compressed (p.s.i.g.)

Air) Air

East River Tunnel, New 84 - 48 Two shifts daily of Ij to 1,360,000York 4 hrs according to19I4-21 pressure

Silent Valley, Belfast 30 Two shifts of 3 hrs with 3,6001927-28 I-hr intervals

Hooghly River Bridge 24 - 45 Two daily ? length 44,5001930-32

Howrah Bridge, India 6 509 40 4 hrs 12,4001938

Feisal and Ghazi Bridges - 36 8 hrs 12,500I946

Tyne Pedestrian Tunnel I8 24 376 42 6 or 8 hrs (4 hrs for 3 40,0001948-50 mths)

Thames Caisson 3 I5 - 35 - 2,IOO1950

Lincoln Tunnel, New i8 300 704 34 Two shifts daily of ij to 138,000York 4 hrs according to

1955-56 pressure

Auckland Harbour Bridge 24 I5I 393 49 8 hrs, 6 hrs, and 4 hrs; 10,0261955-58 (one pier) 3 hrs for short period

Dartford Tunnel 24 250-320 1,200 28 8 hrs 122,0001957-59

Clyde Tunnels 6o Max. 200 1,362 34 8 hrs 240,2591958-63 Av. I50

Blackwall Tunnel 44 200 1,536 39 8 hrs 8i,ooo1960-64

Tyne Road Tunnel 38 90 650 42 8 hrs 44,8001963-66

The earliest book on decompression sicknessamong compressed air workers to be published inthe United Kingdom was by Snell (I896), who wasappointed by London County Council as a full-timemedical officer in charge of compressed air work onthe first Blackwall Tunnel. It was driven throughloose gravel for much of its course and up to about80 men were employed at any one time in com-pressed air. They worked three shifts of approx-imately eight hours and there were over 200 casesof decompression sickness recorded, of which Snell

discussed 50; they included two or three cases ofparalysis. The exact method of decompression andits duration are not stated, but Snell instancesrapid decompression as a minor cause of decom-pression sickness. At Blackwall each man-lock hadtwo pairs of air-cocks, the larger for decompressingmaterial and the smaller for men. With the smallercock fully open, air could escape in 4 minutes toatmospheric pressure whereas the larger one allowedair to escape in 30 seconds. Men were not allowedto use the larger one but this regulation was often

6 R. I. McCallum


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


TABLE I-Continued

7

COMPARISON OF VARIOUS COMPRESSED AIR CONTRACTS

Decompression Type IISickness as % ofI 1 Total Decompression Procedure Decanting

Type I Type II Type I Type II Decompression(Bends) Rate (%) Rate (%) Sickness

68o0 0°05 New York I92I Code. Up to No36 p.S.i.g., 3 lb. in 2 min. Over36 p.s.i.g., I lb./min.

27 0-76 Drop to half absolute pressure, then No7 min./lb.

I17 - 0-26 'One pound per minute from all Nopressures'

344 9 2-8 0-07 2.5 Drop to half absolute pressure in YesI0 min. then 7-8 min./lb.

99 I o-8 - I0 min. from max. pressure No

333 I7 o-87 over 0-04 4.2 Mainly British 1958 Noi8 p.s.i.g.

89 4-0 - Institution of Civil Engineers, 1936

42 2(?) 0-03 overall OOOI 4-5 New York State-three-stage No0-07 over15 p.s.i.g.

2I8 44 2-6 over 0-44 I7 Modified British I958 Yesi8 p.s.i.g.

650 35 o-56 overall 0-04 5.0 British 1958 and extra 5 min. at i No0-93 over p.S.i.g.i8 p.s.i.g.

398 71 0-2 overall 0-04 15-0 Modified British I958. Decompres- No0-29 over sion increased by average of 7 min.i8 p.s.i.g.

824 39 I-05 over 0-04 4.0 Modified British I958 (io min./lb. For pilot tunneli8 p.s.i.g. in slow phase) only

693 i8 I-7 over 0-04 2-5 British I958, and extra 5 miin. at i NoI8 p.s.i.g. p.s.i.g.

disobeyed. A lock made out of a boiler was used fortreatment. Snell emphasized a close relationshipbetween the amount of decompression sickness andthe ventilation of the compressed air space andthe CO2 level in the air. He was of the opinion thatless illness would have occurred if shorter shifts hadbeen worked at the high pressures.

In I904 the New High-Level Bridge was con-

structed at Newcastle upon Tyne with the help ofthree caissons. Here the shift length varied with thepressure, e.g., the shift was io4 hours at 25 p.s.i.g.

and 7 hours at 30 to 35 p.s.i.g., but it includedbreaks in free air of II hours for breakfast andi hour for dinner so that the longest period incompressed air at one time was 4 hours. Thedecompression time was i minute for every 3pounds of pressure so that the maximum decom-pression time was less than I2 minutes. A medicallock was also provided (Parkin, 1905; Boycott, I906).A major advance in the control and mitigation of

the hazards of work in compressed air was the workof Haldane (Report to Admiralty, 1907; Haldane


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

R. I. McCallum

and Priestley, I935). Haldane concluded fromanimal experiments and observations on divers thatbubbles of nitrogen do not appear in the body unlessthe amount of supersaturation is more than that ofa decompression from a total pressure of 21atmospheres. The absolute pressure could alwayssafely be halved, whether the pressure was high orlow, up to about 6 atmospheres, as the volume ofnitrogen released would be the same whether thepressure was reduced from 4 to 2 atmospheres orfrom 2 to i atmosphere. Nitrogen would beeliminated more rapidly than by reducing pressureat an even rate, and the time spent at high pressurewould be reduced. Thereafter decompression wascarried out to atmospheric pressure at a ratecalculated to avoid critical supersaturation withnitrogen in any part of the body. The safety ofrapidly halving the absolute pressure was confirmedin experiments first with goats and later with men.Haldane's hypothesis, on which the statutorydecompression table in force in Britain is stillbased, revolutionized the conduct of compressed airwork and provided a scientific basis for decom-pression. A disciplined procedure was encouragedand the mortality and morbidity of the previous erawere reduced. A supplement to the AdmiraltyReport of I907 contained a stage decompressiontable which was widely used for divers in civil work,and to some extent at tunnels and caissons.

Hill (19I2) commented, however, that the math-ematical calculations on which the Admiralty tableof stage decompression was based had never beenpublished and so could not be critically examined.He doubted the superiority of stage decompressionover a uniform rate of decompression on the basis ofexperimental work with pigs, and also because theAdmiralty table was based largely on theoreticaldata about the circulation of the blood which couldnot be regarded as fixed in the variable conditions ofactivity of the human body.

In 1935 the Institution of Civil Engineersappointed a committee to draw up Regulations forthe guidance of engineers and contractors under-taking work under compressed air. This Committeeincluded J. S. Haldane, Sir Robert Davis, andCaptain G. C. C. Damant. In their report(Institution of Civil Engineers, I936) the Com-mittee stated that practical experience in tunnellingwork had shown that for long exposures to pressuresup to 35 p.s.i.g. the decompression times could beless than those in the Admiralty tables, but that atpressures over 40 p.s.i.g. the times were too short.The Committee thought that, for exposuresgreater than 4 hours' duration, the liability tosymptoms of decompression sickness did notincrease, because by this time the body had become

more or less fully saturated with nitrogen, and thatthe working period could thus be safely prolongedprovided that adequate time was given for decom-pression. The assumption that saturation of bodytissues with nitrogen is complete in a period as shortas 4 hours is not now accepted.These Regulations contain a table of recom-

mended decompression times for varying periods incompressed air up to 50 p.s.i.g. using the principleofstage decompression. However, the recommendedtimes are qualified by allowing a progressive reduc-tion to two-thirds of their value for men withprevious experience in compressed air work 'with-out having shown serious symptoms'. Thus, anexperienced man could be decompressed after a6-hour shift at 28 to 30 p.s.i.g. in 32 minutes,compared with 65 minutes by the table in use today.In contrast, the length of working period wasprogressively reduced as the working pressureincreased, so that at 50 p.s.i.g. only a 4-hour periodof work was recommended.The Report also advised that new starters should

work only half a shift on first entering compressedair, and that recording gauges should be used inlocks. It appears to have been the practice whenworking at high pressure to have two shiftsseparated by a 3- or 4-hour interval, but it waspointed out by the Committee that this doubledthe number of decompressions for a full shift andtook more time, which had to be paid for. It wasconsidered safer and more economical to have asingle long shift in pressure and a prolongeddecompression following it. Nevertheless, theReport contains suggested times of working periodsfor a system with two periods per shift which hasnow been abandoned in the United Kingdom but isin use elsewhere. In the State of New York in theUnited States of America, regulations are based ona two-period per shift system, but even there thissystem has been criticized as inadequate and con-tributing nothing to the safety of the workers(Duffner, I962). In order to achieve satisfactorydecompression, a three-stage procedure has alsobeen used in New York, but it appears that the mainreason for using 'split shifts' was pressure from themen's trade union.The recommendations of the Institution of Civil

Engineers were widely used until a new decom-pression table was compiled by the CompressedAir Committee of the Institution of Civil Engineersand the Ministry of Labour. Discussions with theHome Office had in fact begun in I939 but wereinterrupted by the War. The new table was firstused shortly after work in compressed air had begunin a tunnel under the River Tyne in I948 (Patonand Walder, I954) and was adopted in the present

8


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

Decompression Sickness: AReview9

TABLE IITHEa BRITISH DECOMPRESSioN TABLE (RULE 8) FROM THE WoRxc IN CompREssED AIR SPECIAL REGULATIONS, i958.

Section I Section 2 Section 3

Fastest permissible reduction of pressure from figure in Section 2 to zero.Shortest permissible times (T) in minutes, and fastest permissible rates (R) in minutes Per lb. for the working periods in the

different coluwnns.Lowest

'Basic per-missible 'Working Period': More than Cc) but not more than (d) hourspressure' pressure in ._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

more than first two More than Cc) (d) Cc) Cd) (c) (d) (c) (d) Cc) (d) Cc) (a) Cc) (a) (c) (d)(a) to not min. after 4 hours 31-4 3-31 21-3 2-2i 11-2 I-il 1-' 04imore than starting de- ---(b) lb. per compression T .R T R T R T R T R T R T R T R T R

sq. in. Cib. per Mmn. Mmn. Mmn. Mmn. Mmn. Mmn. Mmn. Min. Mmn. Mmn. Min. Mmn. Min. Mmn. Mmn. MinM. Mmn.Ca) Cb) sq. in.) per lb. per lb. per lb. per lb. per lb. per lb. per lb. per l. per lb.

18-20 2 '3 61 6 3 5 21 4 2 4 2 3 it 2 I 2 I I20-22 3 24 8 14 41 II 31 10 31 9 3 7 21 6 2 4 I 2.22-24 4 35 9 22 51 I 8 41 x6 4 14 31 11 3 9 2 6 '1 3 I24-26 5 46 9 32 61 28 51I 23 41 19 4 x6 3 12 2 8 '1 4 I26-28 6 56 91 42 7 38 61 3'1 5 25 4 22 31 '5 21 II 2 5 I28-30 7 65 91 52 71 48 7 40 51 32 41 27 4 20 3 14 2 7 I30-32 8 74 91 6i 71 57 7 50 61 40 5 32 4 25 3 i6 2 8 I32-34 9 83 9 70 8 6:5 7 59 61 49 51 37 4 29 3 z8 2 110 I34-36 10 91 9 78 8 74 71 68 7 58 6 43 41 34 31 21 2 II I36-38 1i 98 9 87 8 82 71 76 7 67 6 53 5 39 31 26 21 13 I38-40 12 105 9 95 8 90 71 84 7 75 6 62 5 44 31 31 21 i6 it40-42 13 113 9 102 8 98 71 92 7 83 61 70 51 49 4 35 21 i8 if42-44 14 120 81 109 8 105 71 99 7 91 61 77 51 55 4 39 3 20 I144-46 15 127 81 ii6 8 112 71 107 7 99 61 85 51 63 4 44 3 22 if46-48 i6 '33 81 123 8 120 71 115 7 io6 61 93 6 72 41 48 3 24 it48-50 17 139 8 130 8 126 71 I22 7 114 61 ioi 6 80 41 52 3 26 if

Reproduced by permission of the Factory Inspectorate.

Work in Compressed Air Special Regulations of theMinistry of Labour which came into force in1958 (Table II).Compressed air illness was made a notifiable

condition under the Factories Act in I938 and aprescribed disease qualifying for benefit under theNational Insurance (Industrial Injuries) Act ofI946. Notified cases of decompression sickness area better indication of the amount of work incompressed air being carried out in a particularyear than they are of success in the control of thecondition. It is usually only the more severe formsof decompression sickness which are notified, whilechronic bone lesions have only recently beenrecognized widely and accurately recorded.

The Work in Compressed Air SpecialRegulations, 1958

Introduction of these Regulations had the effectof requiring every contractor employing men incompressed air above i8 p.s.i.g. to observe certainminimum standards of conduct and to follow adecompression procedure which, at the time, wasfelt to be the safest practicable method.The Regulations describe in detail the arrange-

ments required when work is being carried out incompressed air on a civil engineering contract.They specify the design and equipment necessary

for the man-lock in which workers will be com-pressed, and decompressed at the end of their shift.A competent lock attendant must be in charge andmust keep a register in which is entered the time atwhich a man goes into the lock, the pressures in theworking area at the beginning and end of his shift,and the time taken to decompress him. Compressionand decompression must be carried out accordingto rules set out in the Regulations. Medicalexamination must be carried out by an officially'Appointed Doctor' who must certify a man'sfitness for work in compressed air within the threedays before he starts work, unless he has beenemployed in the work within the previous threemonths. He must also be examined every fourweeks if the air pressure at work is above 1i8p.s.i.g., and after an upper respiratory tract in-fection. The employer must supply each man witha label to be worn next to the body showing that he isa compressed air worker and giving the address ofthe medical lock.Each man has a personal Compressed Air Health

Register in which a record of the examination isentered; this he should take to each employment incompressed air. The local hospital must beinformed that compressed air work is being under-taken and given the name of the Appointed Doctorand the date of completion of compressed airoperations.

9


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

R. L McCallum

Medical Selection and Supervision Indeciding on a man's fitness for work the doctorexcludes, among others, those with ear or sinustrouble and the obese. It is usual for a new workerto be taken into a medical lock by the attendantand instructed in clearing the pharyngo-tympanictubes when air pressure is increased. A radiographof the chest is not compulsory in civil engineering,although it is required by the diving regulations.

Decompression The procedure for decom-pression from pressures of i8 p.s.i.g. up to 50p.s.i.g. is set out in a table in the Regulations whichspecifies the minimum decompression times allowedfor various lengths of shift (Table II). At the endof a shift in compressed air the men enter the man-lock and the pressure is then lowered fairly rapidlyto half the absolute pressure. If, for example, aman has worked for more than 4 hours at a pressureof 25 p.s.i.g. then the pressure can be droppedrapidly to 5 p.s.i.g. This figure is arrived at byadding gauge pressure and atmospheric pressure toobtain the absolute pressure, 25 + I5 = 40p.s.i., halving the result and subtracting atmospheric

pressure to give gauge pressure again, - = 20,2

20 -I = 5 p.s.i.g. After this initial drop thepressure must be reduced to atmospheric pressure ata slow steady rate of 9 minutes per pound, and allthis takes nearly an hour to complete.

Since the construction of the Tyne PedestrianTunnel in I948 there have been several majorcompressed air undertakings covered by the I958Regulations, such as the Dartford Tunnel (I957-59)and the Blackwall Tunnel (I960-64) under theRiver Thames, the Clyde Tunnels (I958-63), theTyne Road Tunnel (I963-66), and a number ofpower station cooling water tunnels.The decompression table has also been used in

other countries, often with modifications (Rose,I962; Meesters, I967), and modifications havealready appeared in Britain (Whyte, I967).

Evaluation of the 1958 Decompression Table

Sufficient use of the table has now been made tohave produced a body of opinion critical of it andsome other aspects of the Regulations. The bendsand more serious forms of decompression sickness(Table III) can still occur much too frequently, andfrom time to time there are deaths following work incompressed air (Annual Report of Chief Inspectorof Factories for I963).

Paton and Walder (1954), in their investigation

AcuteType I

Type II

TABLE IIIFoRMs OF DECOMPRESSION SICKNESS

Mild limb pain ('the niggles')Severe limb pain ('the bends')Skin mottling and irritation ('the

itches')

Vomiting with or without epigastric painVertigo ('the staggers')Tingling and numbness of limbsParalysis or weakness of limbsDyspnoea ('the chokes')Severe headacheVisual defects, such as flashes of light or

scotomataAngina, symptoms suggesting coronary

dysfunction, irregular pulseCollapse, hypertensionComaDeath

Chronic Neurological and psychiatric forms, in-cluding paralysis

Aseptic necrosis of bone-juxta-articular, head, neck, and shaftlesions

of decompression sickness at the first Tyne Tunnel,in which the bends rate was o-87% of decompres-sions, did not comment critically on the decom-pression table itself. Duffner (1955) considered itfaulty on the grounds that no two-stage procedurebased on the usual calculations was adequate butthat a three-stage procedure was much better.Rose (I962) made a statistical survey of decom-

pression sickness in caisson workers employed onthe construction of Auckland Harbour Bridge inNew Zealand, in which pressures from 27 tO 49p.s.i.g. were used in six steel caissons. There wereno regulations governing work in compressed air inNew Zealand so that the tables of the Institution ofCivil Engineers (1936) and of the British I958Special Regulations were critically examined andcompared. The latter table was eventually adoptedfor use at the Auckland Harbour Bridge but withmodifications. Rose points out that the table is morestraightforward than that of the Institution of CivilEngineers, but, although it involves a much longerperiod of decompression at lower working pressures,this margin narrows as pressure rises, until at ornear the limit at the highest pressure of about 49p.s.i.g. the decompression time was shorter by 2minutes than the Institution of Civil Engineers'table. It was noticed that for shifts of over 4 hoursthe number of minutes per pound decreased as thepressure increased, e.g., at 30 to 32 p.s.i.g. workingpressure it was 91 minutes but at 48 to 59 p.s.i.g.

IO


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


it fell to 8 minutes. In contrast, the decompressiontimes for working periods of less than 4 hoursincreased with the rise in pressure. Rose concludedthat this apparent discrepancy arose from a desire toreduce the decompression time for shifts of morethan 4 hours' duration to a figure acceptable in anindustrial undertaking, and that presumably amargin of safety had been built into figures in therest of the table which had been removed or reducedin the upper range of working pressures.The decompression times for the Auckland

Harbour Bridge contract were therefore increasedin the 40 p.s.i.g. and over range to give a morelinear relationship between the total decompressiontime and the working pressure than with either thetable of the Institution of Civil Engineers or theBritish I958 Regulations. The effects of theseamendments introduced at Auckland were notremarkable since difficulties due to the variableswhich beset all investigations of this type, such asstrikes and stoppages with consequent acclimatiza-tion effects, made it difficult to draw any firmconclusion. Paton (I963) points out that, althoughthe I958 table is the mathematical consequence ofHaldane's theory, the experience at Auckland againraises doubts about its validity.

It is interesting that many contractors in Britainof their own initiative now routinely use decom-pression times longer than those required by theRegulations (Whyte, I967), and this trend can beseen in other countries. Limitation of time spent athigh pressures, as is enforced in some statutorydecompression tables, is not usual in Britain.The most striking change of opinion in Britain

has been in respect of the maximum tissue exitpressure. In early discussions on this, an exitpressure of i5 p.s.i.g. was considered but wasabandoned in favour of i8 p.s.i.g., which becamethe legal maximum (Paton, I967). Although it wasknown that bends could occur from time to time atpressures between i5 and i8 p.s.i.g., it is onlyrecently that civil engineers have felt that this wasa serious matter and have in many cases spontan-eously insisted on timed decompression for ex-posures in this range. The factor of acclimatization,to which Paton and Walder (1954) drew attentionand which Rose (I962) underlined, has led in somecontracts to the introduction of acclimatizationshifts of shorter duration than the usual period forthe first one or two shifts of new workers. There is,however, general dissatisfaction with the lack ofprogress in improving the safety of compressed airwork, in the occasional and apparently uncontrol-lable fatalities, and, most recently, in the recognitionof the frequent occurrence of aseptic necrosis ofbone in caisson and tunnel workers.

The stimulus of the I939-45 wartime experienceof diving and high altitude flying (Fulton, I948)(although the decompression of high altitude flyingdiffers in important respects from decompressionafter work at high atmospheric pressure (Gribble,I960)) brought the present post-Haldane era inwhich the validity of his 2: I ratio has been seriouslyquestioned over the whole range of workingpressures in diving and civil engineering. It is nolonger accepted that full saturation is completeafter 5 hours' exposure, and it has furthermore beensuggested that gas bubbles are always formed duringdecompression although they may be microscopicin size.

Acclimatization

It has been observed that in a group of mennewly introduced to work in compressed air attacksof bends are very frequent, and in contracts witha high bends rate much of this may be due to a highlabour turnover. New starters tend to leave thework after one or more attacks of bends so that aself-selected group is left. If these men areemployed for a period in stable conditions ofpressure, the bends rate may fall to a low level.This situation may be upset by various factors butmainly by alteration in pressure and by interrup-tion of the work. A rise in pressure of a few poundsor stoppage of work for a week or two due toholidays or a strike is usually followed by a briskcrop of attacks of bends which keeps the medicallocks full. In time, if the new pressure remainsstable and work is regular, the number of attacksfalls off again and may become negligible. Thisphenomenon of acclimatization was studied byPaton and Walder (I954) using the average dailybends rate per man in groups of men whoseemployment in compressed air at the first TyneTunnel could be followed for continuous periods oftime. The longer the period of time chosen thesmaller was the number of men available. It wasfound that there was a steady decline in bends ratefrom the first decompression up to the tenth tofifteenth day, the rate halving every 5 days, afterwhich the variations were no more than random.Acclimatization appeared to be acquired rapidlyand lost equally rapidly.The mechanism underlying acclimatization is of

the highest importance but its nature is completelyobscure. Paton and Walder (1954) suggested thatits time course resembles that in training peopleunaccustomed to physical exercise, in which stiffnessof muscles occurs at first but later wears off withcontinued exercise. They postulated that unaccus-tomed exercise causes minor foci of muscle damage

II


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

R. I. McCallum

and that these damaged areas may result in thesensation of stiffness and provide nuclei suitablefor bubble formation. However, the cause ofstiffness after exercise is still hypothetical and,although their theory may account for acclimatiza-tion, it does not explain the rapid loss of acclim-atization when the working pressure rises abruptly.

The Use of Oxygen in Decompression

Inhalation of oxygen as a means to preventdecompression sickness or in therapeutic recom-pression has been used successfully (Jones, Crosson,Griffith, Sayers, Schrenk, and Levy, I940) So thatit is surprising that in civil engineering work it hasrarely been exploited. Lack of experimentalevidence as to its value and questions of cost havebeen given as the explanation for this (Behnke andShaw, I937). Oxygen has been used extensively indeep diving by highly trained individuals undercarefully controlled conditions, but in civil engineer-ing work contractors have been reluctant to use itbecause of the fire risk. Above about 25 p.s.i.g.there is a risk of oxygen poisoning so that inpractice it has been restricted to the second half ofdecompression following the first drop of pressureto half the absolute pressure. Jones and others(1940) gave oxygen during the last 20 minutes ofeach regular decompression when the pressure wasnever more than I7 p.s.i.g. and usually belowI5 p.s.i.g. Over three months, 3,884 decompres-sions were carried out with oxygen inhalationthrough individual face masks from a speciallydeveloped system. No bends occurred in any of themen on oxygen although the pressure range was34 to 371 p.s.i.g. and the total decompression timewas only 30 to 48 minutes. In another 15,904ordinary decompressions without oxygen there were21 cases of decompression sickness (O-I3%). InII,I96 decompressions with oxygen, but underless stringently controlled conditions than the firstexperiment, there were 23 cases of decompressionsickness (0-2I %), three of which were ratherdubious and were therefore discounted. In 9,462decompressions without oxygen there were 12 casesof decompression sickness (O'I2%), all of whichwere moderate or mild whereas among the men notgiven oxygen five cases of decompression sicknesswere severe. It was concluded that with efficientoxygen administration, and proper supervision andeducation of workers in its use, the incidence ofdecompression sickness could be reduced andsevere cases eliminated. However, the data showthat the bends rate during this work was low evenwithout oxygen inhalation. When men were in the

same decompression chamber as those using oxygenmasks their bends rate tended to be low because ofthe relatively high oxygen content in the expiredair of the men having oxygen. In Japan, Kita (I964)has reported a reduced prevalence of bends and ashortening of the decompression times by two-thirds to half in 87 decompressions from caissonwork at 30 to 34 p.s.i.g. for periods of 31 to 6 hours.Seventy men were given oxygen during the slowphase of decompression, when the lock pressure haddropped to under 25 p.s.i.g., to avoid poisoning.But the use of oxygen during the later stages ofdecompression to facilitate release of nitrogen maynot be as effective as theoretical considerationssuggest because of the associated circulatory changes(Hempleman, I967). In an oxygen decompressionexperiment in Japan a man struck a match to smoke,which was forbidden. In the fire which followed,six men died and two others were seriously burnt.New oxygen breathing apparatus was devised whichautomatically releases expired air out of the lock(Nashimoto, I967). It is clear that, if oxygen is tobe used, discipline and safeguards will have to beof a high order to avoid serious fire accidents.

Respirators in Compressed Air

Fumes from an outbreak of fire in a pressurizedtunnel may compel fire fighters to wear respirators.The detection of harmful gases and the use ofbreathing apparatus in compressed air has beenstudied recently by the American Bureau of Mines(Berger, Curry, Watson, and Pearce, I964). Carbonmonoxide is toxic only if the ratio of partialpressures of CO and 02 is altered, but the effects ofCO may become prominent on decompression sothat it is necessary to treat poisoning with un-contaminated air at pressure. The effect ofhydrogensulphide on the central nervous system is enhancedas its partial pressure rises, and nitrogen dioxidealso becomes more toxic.Oxygen is unsafe above 26 p.s.i.g., and in a

self-contained breathing apparatus it becomesdangerous after 30 minutes at 36 p.s.i.g. Gas masksmay be difficult to use in pressure because of theincrease in air resistance, and canister heating dueto oxidation of carbon monoxide may be intolerable.Two types of breathing apparatus were tested. Thefirst was the self-contained demand type using com-pressed air in which the rate of air withdrawal riseswith increasing atmospheric pressure. Thus theoperational life of the apparatus is inversely pro-portional to the pressure at which it is being used,and the fitting of oversize cylinders in order toprolong the time in an hyperbaric environment has

12


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


the disadvantage of increasing both bulk and weight.As the result of these factors the use of this apparatusis limited to work at about 15 p.s.i.g., and then foronly a relatively short period. The second type wasthe self-contained recirculating oxygen breathingapparatus with C02 absorber (McCaa type). Thisnormally supplies pure oxygen but its use was

modified by the American Bureau of Mines tointroduce and maintain enough nitrogen to lowerthe partial pressure of oxygen to a safe level. Itwas claimed that, with the modified technique, thisapparatus could be used safely at pressures up to45 p.s.i.g. for one hour.

Effects of Pressure on Response toGas-Detecting Instruments

H2S, CO, and NO2 detectors are usually color-imetric tubes calibrated in terms of the volumeconcentration at atmospheric pressure. The maxi-mum allowable concentrations (M.A.C.) are alsoexpressed as volume concentrations, but both thedetector responses and physiological effects dependdirectly on the weight of the substance. The weightper volume concentration increases with pressure,although the volume per volume is unchanged, sothat for H2S and N02 the instrument can be usedwithout correction. For CO the M.A.C. applicableat pressure is obtained by dividing the reading bythe absolute pressure in atmospheres. As H2S inair interferes with the colour response for CO theH2S must be absorbed first (Berger and others,I964).

The Bends Rate

In a situation influenced by so many variables,such as length of shift, height of pressure, highlabour turnover, acclimatization, physical andmental characteristics of the workers, differences intemperature and humidity, and disciplinaryproblems, it is difficult to find a suitable index bywhich to judge the success of a decompressionprocedure or to compare one contract with another.The bends rate, which is the number of attacks ofbends treated by recompression as a percentage ofthe number of decompressions, is commonly used.While a bends rate of up to 2% has been considered

acceptable in Britain, it has varied from o-87% atthe Tyne Pedestrian Tunnel (Paton and Walder,I954) to 4% at a caisson in the River Thames(Lewis and Paton, I957). The fallacies in using thismeasure make it of very doubtful value unless anumber of factors can be allowed for. If the workhas been carried out at relatively low pressure and

with a low labour turnover, the bends rate is likelyto be low compared with a contract where thepressure is high and unacclimatized men areconstantly being recruited. The basic informationis suspect, as mild bends pain, called the niggles, iscustomarily accepted by the men as trivial, and evenquite severe pain may be self-treated with aspirinand alcohol. Records of attacks of bends dependentirely on the men reporting them, on their beingaccepted as such by the medical lock attendant ordoctor, and on being treated by recompression. Aman with severe decompression sickness will ofcourse almost always be recompressed.A comparison of some of the published data for

a number of compressed air contracts in differentparts of the world over the last 50 years (Table I)illustrates many of the difficulties in drawingconclusions from such information. Importantdetails, such as the total number of men employed,may not be given so that there is no indication oflabour turnover, and the duration of maximumpressure is often not given. In earlier contracts inwhich decompression was frequently much morerapid than in later ones, and in which the shiftlengths tended to be shorter, the bends rate was asgood or better. Decanting appears to be associatedwith a high bends rate but the maximum pressurewas also high.The exceptionally low bends rate of o o3 % was

reported from the New York Lincoln Tunnel (thirdtube) in I955 (Kooperstein and Schuman, I957).In 138,034 decompressions there were only 44 casesof decompression sickness. The maximum pressureused was 34 p.s.i.g., but for i i months of the workthe pressure was only I5 p.s.i.g. At this pressureone case of decompression sickness occurred. Asplit shift system was used and a three-stagedecompression. No detailed information is availableabout bone lesions in the men employed in thiscontract but large sums in compensation were paidin connexion with some 6o or so cases. Althoughmany of these lesions may have related to work inprevious contracts (Behnke, I967), it is not clearwhether any of the men employed had boneradiographs prior to being employed or periodicradiographic examinations during and after thework. No simple relationship between bone necrosisand bends has been shown so far (R6zsahegyi andFried, I963; Decompression Sickness Panel Report,I966).

Aseptic Necrosis of Bone in CompressedAir Workers

Until recently, aseptic necrosis of bone in com-

pressed air workers has been reported mainly in

I3


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

single patients or in very small series, and usually inmen presenting with symptoms affecting a majorjoint. It is now clear that if these lesions are

actively sought by systematic radiological examin-ations a high prevalence of symptomless lesions isfound (Fournier, Jullien, and Leandri, I965; De-compression Sickness Panel Report, I966). At theClyde Tunnels, about 20% of compressed airworkers remaining at the end of the contract had a

bone lesion or lesions, although only 4% of theselesions were juxta-articular and likely to cause

disability.This investigation followed a less extensive one

at the Dartford Tunnel (Golding and others, I960)where symptomless bone lesions were also found.The proportion of men with bone lesions at theClyde Tunnels was a conservative estimate, as onlyi8% of the men at risk throughout the wholecontract could be radiographed. Some of the lesionscalled doubtful would now, in the light of furtherexperience of early radiological changes and ofhindsight, be regarded as areas of aseptic necrosis.Former compressed air workers from the ClydeTunnels, who were not included in the survey, haveattended Glasgow hospitals with symptoms fromjuxta-articular bone lesions (Davidson, I964) andsome have been detected subsequently at othercontracts. At the Clyde it was found that thepresence of a bone lesion could be related to thelength of time at compressed air work and toexposure to pressures higher than 30 p.s.i.g., butthese two factors could not be separated. As atDartford, there was no close link between attacksof bends and subsequent bone necrosis.Some of the most interesting findings in the

Decompression Sickness Panel's report arose fromthe histological changes in the left humerus andright femur of a compressed air worker who haddied during treatment for decompression sickness(Bennison, Catton, and Fryer, I965). An area ofnecrosis can revascularize extensively, and in a

juxta-articular lesion it is when this process isincomplete that collapse of the articular surfacemay occur and cause disability. It is possible thatmany bone lesions heal spontaneously before thedeposition of new bone on dead trabeculae makesthem radiologically detectable.

The Cause of Bone Necrosis It is almostalways assumed that bone necrosis has the same

pathogenesis as the bends, namely bubbles ofnitrogen, and that these cause infarction by formingintravascularly or extravascularly and interruptingthe blood supply. A further assumption appears tobe that bone has a poor blood supply and, because

of this and the presence of fat in marrow, it isparticularly susceptible to damage from gas bubbles.

It is difficult to find in textbooks or journalsdescriptions of infarction of bone in the sense ofvisible obstruction to a vessel or end artery, and infact the blood supply to bone appears to be good.Although an area such as the head of the femurappears to be exceptionally vulnerable to ischaemicdamage, one of the striking features of asepticnecrosis of bone in compressed air workers is thatit may also arise in the shafts of long bones. Further,aseptic necrosis shows a marked tendency tosymmetry so that both shoulders or both hips, orall four joints, are often affected, and lesions roundthe knee joints are frequently symmetrical. Boneinfarction with necrotic areas comparable to thosefound in compressed air workers has not beenproduced in experimental animals even though thepressures used and the decompression times havebeen grossly dangerous by comparison with theconditions in tunnels or caissons.There is therefore no proof that gas bubbles cause

the bone lesions, and other possible mechanismsmust be considered. It is worth considering howbone might differ from other tissues on compressionand decompression, because of its rigid structure.When the body is compressed the change in pressureis immediately transmitted equally throughout thesoft tissues. Is it possible that in certain areas oflong bones, because of its combination of com-paratively rigid tissue and softer tissue, transientpressure differences occur which produce necrosisby a direct process ?The whole emphasis now in the study of decom-

pression in civil engineering has shifted to theelucidation of the cause of bone necrosis and itsprevention, and it is felt that the solution of thisproblem may also contribute to or solve that ofother forms of decompression sickness.

Lung Cysts and Type II DecompressionSickness

It has been suggested that type II decompressionsickness (Table III) is caused by gas embolism dueto air entering the circulation from without ratherthan to the formation of gas bubbles by their releasefrom supersaturated blood or tissues. At present,clinical distinction between type II decompressionsickness and air embolism is often difficult orimpossible in life but, as treatment is similar for thetwo conditions, the concept still remains a usefulone, certainly in the civil engineering context.During the construction of the Dartford Tunnel,

two men had attacks of severe decompression

R. I. McCallumI4


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


sickness (type II) and both men were found to havecysts in the lungs (Golding and others, I960). Afurther example occurred in a man working on theClyde Tunnel (Walder, I966). Other cases havebeen reported by Liebow, Stark, Vogel,and Schaefer(I959) and by Collins (I962), who described multiplethin-walled lung cysts which appeared in thepreviously normal radiograph of a 20-year-old manfollowing exposure to a pressure of 50 p.s.i.g.There were gross neurological symptoms afterdecompression.Walder (I963; I966) suggests that the blocking

of a bronchus by a mucosal plug or by oedema whilea man was at pressure might result in air-trapping inpart of a lung so that on decompression the trappedair would expand to form a cyst, or might rupture ablood vessel and cause air embolism. Walderinduced histamine bronchospasm in guinea-pigsexposed to compressed air and showed that duringdecompression bubbles appeared in the circulationand many of the animals died. Further studies onpig lungs, which are similar in structure to humanlungs, show that cyst formation and air embolismare possible. If there is gas supersaturation airembolism could also initiate further bubbling.

Routine lung radiography of compressed airworkers has unfortunately yielded little furtherevidence. At the Tyne Tunnel, for example, onlyone lung cyst was discovered in I83 radiographs ofthe chest. Even this cyst was overlooked in theinitial film until it was seen in a film of the shoulderjoint taken much later. The man had meanwhileundergone regular compression and decompressionas a tunnel worker without any serious con-sequences. It is possible that the mechanism whichWalder postulates does in fact occur and is thebasis of type II decompression sickness but thatactual lung cyst formation is not essential or thatthe cysts so formed are small and do not show ona radiograph.

Neurological Complications ofDecompression

The study and recording of central nervoussystem lesions, other than gross paralysis, and ofpsychiatric illness as long-term chronic effects ofwork in compressed air has been almost entirelyneglected. R6zsahegyi (i959) has described andclassified neurological forms of decompression sick-ness in men who worked on the building of theBudapest underground railway. He distinguishesfour types of neurological damage: multiple lesionsin the whole central nervous system, beginningwith acute collapse or a Meniere's syndrome, but

rarely recovering completely; multiple lesions inthe cerebrum and upper brain stem which maypresent acutely with coma but are often followed bya vegetative neurosis; lesions in the medulla oblong-ata, pons, and cerebellum, also producing a type ofMeniere's syndrome; and spinal lesions leading totetraplegia, impotence, vegetative neurosis, andpersonality changes. Latent damage to the cerebrummay be demonstrable by electroencephalography(Rozsahegyi, I967). It is to be hoped thatR6zsahegyi's work will stimulate further detailedstudies of central nervous and psychiatric symptomsin compressed air workers.

Effects on Cerebral Function

Some engineers who carry out measurements andcalculations in compressed air have concluded thataccuracy and ability are impaired even at moderatepressure. Experimental card sorting at normalatmospheric pressure and at pressures of 2 to 34atmospheres absolute was carried out by a group ofcompressed air workers at Tilbury under theauspices of the Decompression Sickness Panel(Poulton, Carpenter, and Catton, 1963). Theresults suggested that performance was affected bycompressed air at 2 atmospheres absolute, but onlywhile the task was being learnt. Compressed air hadlittle effect if the subjects had practised beforehand.If this is so, then it has serious implications for anunfamiliar emergency in compressed air, and thepossibility of impaired judgment of surgeons work-ing in hyperbaric chambers. However, when thesame experiment was repeated at the Royal NavalPhysiological Laboratory no evidence of a narcoticeffect at 2 atmospheres absolute was found, and itis suggested that the findings of Poulton and hiscolleagues were due to factors other than nitrogenor inert gas narcosis (Bennett, I966; Bennett,Poulton, Carpenter, and Catton, I967). Furtherexperimental work on the problem is obviouslynecessary to settle this point.

Cardiac Effects

Observations on electrocardiographic (E.C.G.)changes in compressed air workers do not appearto have been made in the United Kingdom, andeven on the continent of Europe they are notnumerous. In a review of the continental literature,including observations of his own, Zannini (I967)describes E.C.G. abnormalities due to compressedair work. These are pulmonary P waves, notchedP waves, depression of the S-T tract, flattening orinversion of T waves in leads 2 and 3, and lengthen-

IS


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

R. I. McCallum

ing of the PQ tract. These, he thinks, may be dueto the effect of bubbles in the pulmonary circulation.He also mentions E.C.G. changes which occur

only at high pressure and regress on decompression.They may be due to factors such as changes inautonomic tone, increased oxygen tension, variationin diaphragm level, and transitory changes dueto the work load. He considers that electro-cardiography should be a routine part of thepre-employment and periodic examinations ofcompressed air workers and of all cases of decom-pression sickness.

Blood Changes

Haemoconcentration has been found as a compli-cation of severe decompression sickness, particularlyfollowing exposure to altitude (Behnke, I967), butit may also occur in tunnel or caisson workers(Bennison and others, I965; Catton, I967). Bloodchanges can be anticipated when post-decom-pression shock occurs and may be associated withair embolism and perhaps also with fat and marrowembolism (Behnke, I967). Catton (I967) studiedvarious blood samples from compressed air workersemployed at 35 to 43 p.s.i.g. but found no diagnosticchange associated with the bends and concludedthat clinical examination was more informative.One man with type II decompression sickness(paralysis) had no unusual blood changes duringrecompression treatment.The relation of the surface tension of human

serum and susceptibility to decompression sicknessdetermined by exposing subjects to a standard testof 37,000 feet for 2 hours was studied by Walder(I948). There was a significant difference betweenthe mean static surface tension of susceptiblesubjects and non-susceptible subjects who hada high serum surface tension. This suggestedthat raising the surface tension might affordsome protection against decompression sickness,and preliminary experiments on susceptible sub-jects after ingestion of normal saline was followedby a temporary rise in static surface tension andfreedom from decompression sickness. This briefreport suggests a line of enquiry which might bepursued further in compressed air workers.

Treatment of Decompression Sickness

The Work in Compressed Air Special Regula-tions, 1958 do not refer specifically to the treatmentof decompression sickness other than to make theprovision of a medical lock a requirement. There isno statutory method of treatment, so it is

carried out according to the judgment of the lockattendant and the doctor, and can be suited to theneeds of the individual patient and his symptoms.The empirical procedures of Griffiths (I960; I967),which have circulated widely, have generally beenthe basis of treatment of decompression sickness inBritain. Griffiths believes that, by using the lowestpressure which will completely relieve the symp-toms, further absorption of gas is kept to a minimum.He classifies decompression sickness into two types-type I, or the bends, in which there are limb painswithout other symptoms, and type II, in which theremay be a great variety of symptoms and signsvarying from neurological or cardiovascular syn-dromes to collapse, shock, or coma (Table III).

For type I decompression sickness Griffithsoriginally recommended recompression either to3 lb. (0o2 kg.) above the maximum working pressureor more slowly until the pain had gone. Afterwaiting for I5 to 30 minutes decompression iscarried out by a modified regulation schedule inwhich the pressure is dropped rapidly to half theabsolute pressure plus 2 lb. and the slow phase isprolonged by taking I5 minutes for each remainingpound of pressure.Type. II decompression sickness necessitates

immediate recompression to working pressure orhigher if symptoms are not relieved, even up to theworking limit of the medical lock, which is usually50 p.s.i.g. The man is kept at the effective pressurefor half an hour after signs and symptoms havedisappeared, after which pressure is reduced by onepound every I5 minutes to I5 p.s.i.g., at which itis maintained for 4 hours. From this point there isa gradual reduction of pressure at the rate of i lb.every half hour with 'soaks' of an hour each at 8,4, and 2 p.s.i.g. Treatment can last for 24 hours ormore but is usually completely successful.

Similar methods were used at the Tyne Tunnel(Paton and Walder, I954), where men with bendswere recompressed to working pressure plus 3 lb.and then held for either I0 minutes or 20 minutes,after which they were decompressed as for an ex-posure to a working pressure of over 4 hoursaccording to the Regulation table. In severe casesmen were recompressed to working pressure andheld until I0 minutes after they had recovered fromsymptoms. After this the pressure was lowered ata decreasing rate from i p.s.i. in 3 minutes toi p.s.i. in 8 minutes, according to the height ofpressure, and provision was made for slower ratesif symptoms returned. Duffner (I955) comparedthese methods unfavourably with the United StatesNavy procedure for divers after which less than 2%of patients treated have a recurrence of theirsymptoms.

I6


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


In treating type I decompression sickness, about20% of patients may require further recompressionand this has led Griffiths recently to suggest mod-ifying the treatment by taking the initial pressuredrop to half gauge pressure rather than to halfabsolute pressure. In this way the number of menrequiring further recompression after treatment hasbeen much reduced.American practice (Lanphier, I966) has favoured

the use of the United States Navy TreatmentTables (Diving Manual, I963) which take no accountof the pressure at which the patient has worked.These tables are, of course, designed for the com-plications of deep diving and may involve takingthe patient to a pressure of 6 atmospheres absolute(go p.s.i.) but Lanphier recognizes the need forprocedures suited to compressed air workers asopposed to divers. He quotes Griffiths' methods infull and concludes that they can be used 'with goodexpectations of success'. With the U.S. NavyTreatment Tables Lanphier states that oxygen isnow considered obligatory during the final stages,but, although he recommends the use of oxygen inlow pressure recompression, it is still consideredexperimental.Some of the theoretical considerations involved

in the treatment of decompression sickness arediscussed by Hempleman (I967), who points outthat although it is rare for a tunnel worker whodoes not often work at over 40 p.s.i.g. to have to berecompressed beyond 45 p.s.i.g. to relieve hissymptoms, divers commonly require recompressionto pressures higher than 45 p.s.i.g. after dives todepths at a pressure greater than this. Thus thereis sometimes a connexion between the originalexposure pressure and the therapeutic pressure.

It has been suggested that there is a plasmadeficit in animals and humans with decompressionsickness and that this may be sufficiently large tocause death (Cockett, Nakamura, and Kado, I965;Cockett and Nakamura, I964). Experiments ondogs by these workers indicated that plasmareplacement with dextran but without recom-pression was effective in treating moderate decom-pression sickness and shock. They also usedhypothermia to extend the time interval between theattack and the application of recompression. Theyconclude, however, that recompression is still thetreatment of choice but that infusion of dextranwill correct the plasma deficit prior to recom-pression.

Occasionally in deep sea divers there is aparadoxical response to recompression in whichdeterioration in the patient's condition occursinstead of relief (Barnard and Elliott, I966). Thishas also been seen in a compressed air worker who

had pain in the tibia (Griffiths, personal com-munication).

Post-Mortem Examination of Fatalities

There is still much to be learnt about thepathogenesis of the occasional fatalities whichfollow decompression. Pathologists without specialknowledge of the problem could contribute sig-nificantly by recording the presence or absence of anumber of important features. A detailed descrip-tion of the points to be looked for has been drawnup for the M.R.C. Decompression Sickness Panel(Fryer, D. I., personal communication). Radio-graphy of the thorax should be carried out beforethe necropsy is begun to detect a lung cyst whichmight otherwise be damaged or destroyed before itwas recognized, and to show whether there is gasin the heart or great vessels. Removal of the humeriand femora may help to elucidate the cause andnatural history of aseptic necrosis of bone. Someareas of recent bone damage may not be demon-strable by radiography (Decompression SicknessPanel Report, I966), and the histopathology ofthese is important.The brain and spinal cord should be displayed

and examined carefully for intravascular bubblesand removed whole for fixation in formalin andsubsequent examination. The heart should beopened under water to show any gas in thechambers. Gas bubbles should also be soughtthroughout the body and their presence recordedseparately. Patent atrial septum should also benoted, and haemoconcentration looked for.

The Decompression Sickness Panel

High altitude flying and deep sea diving are thesubject of intensive study by full-time experts atthe research establishments of the Royal Air ForceInstitute of Aviation Medicine at Farnborough andthe Royal Naval Physiological Laboratory atAlverstoke.The study of industrial decompression sickness

has been almost entirely neglected in the UnitedKingdom since Haldane's original work (Patonand Walder, I954). There is no group of full-timeresearch workers primarily concerned with civilengineering decompression problems and able tocarry out a long-term programme of researchcomparable to institutions which exist in somecontinental countries. The only official bodyseriously engaged in research in this field is theDecompression Sickness Panel of the MedicalResearch Council's Occupational Health Com-mittee. Although nearly all the members of the

I7


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

R. I. McCallum

Panel are concerned for only part of their profession-al time with the problems of work in compressedair, the presence on it of academic, naval, air force,and clinical members, and of many general prac-

titioners acting as physicians to civil engineering

contracts, enables the Panel to collate informationand stimulate research into compressed air problems.Over a period of nearly 20 years members of the

Panel have been responsible for investigations intoa number of compressed air projects, includingimportant studies of decompression sickness at thefirst Tyne Tunnel (Paton and Walder, I954) and atthe Dartford Tunnel (Golding and others, I960);investigations into mental skill in compressed air(Poulton, Carpenter, and Catton, I963); andinvestigations into aseptic necrosis of bone intunnellers working in compressed air (Bennisonand others, I965; M.R.C. Decompression SicknessPanel, I966). A Registry of Compressed AirWorkers was set up at the University of Newcastleupon Tyne in I964, and a trial ofnew decompressiontables is currently being undertaken at two com-

pressed air contracts. These tables (BlackpoolTrial Tables, I966) are based on a decreasing criticalratio as the pressure to which the men are exposedrises, instead of Haldane's fixed 2:I ratio. For an8-hour shift the decompression times are substan-tially longer than the Regulation times but are stillshorter than in some schedules in use abroad(Table IV). This trial is primarily an attempt toreduce or control bone necrosis, and the long-termradiographic follow-up of compressed air workers isan essential part of the experiment. The civilengineering contractors concerned, the FactoryInspectorate, and the Civil Engineering ResearchAssociation are all co-operating with the Panel inthis trial, and use of the tables in some othercountries may contribute additional data.

International Co-operation

Information on the success of decompressionschedules in other countries is often difficult tofind or to assess. A small international workingparty held at the Ciba Foundation in London inOctober I965 has helped to clarify this situation.The proceedings of this working party have now

been published (Decompression of Compressed AirWorkers in Civil Engineering, Oriel Press, Newcastleupon Tyne).

Future Development

Although the Work in Compressed Air SpecialRegulations, I958 were a major advance in thecontrol of compressed air work they have proveddisappointing in practice. Unfortunately, regula-tions, although necessary, may have the effect ofinhibiting any experimentation, and this can deterattempts at improvement where the procedure isnot wholly effective. A contractor is understandablyunwilling to vary a procedure laid down by law,and particularly to shorten decompression timesshould this be required by a scientific investigation,in case this should lay the firm open to civil actionshould anything go wrong. However, the Regula-tions allow the Chief Inspector of Factories togrant a certificate of exemption from their require-ments in certain circumstances, and this procedureis being applied in the use of the Blackpool TrialTables.Many people concerned with compressed air

work feel that all the provisions of the Regulationsshould apply to pressures of I5 p.s.i.g. upwards,instead of i8 p.s.i.g. as at present, that acclimatiza-tion shifts should be a requirement, that the timesspent at high pressures should be specifically

TABLE IVDECOMPRESSION PROCEDURES CURRENTLY IN USE IN GREAT BRITAIN AND THE UNITED STATES OF AMERICA

Pressure Shift Total Timep.s.i.g. Length Decompression Procedure (min.)

(hours)

Great Britainnot exceeding

1958 Regulations 24 5 Quickly to 4 p.s.i.g., then to o p.s.i.g. in 35 min. about 37Royal Navy Diving Tables 24 5 I9 min. at I3 p.s.i.g., 40 min. at 8-7 p.s.i.g., about I07

45 min. at 4-3 p.s.i.g.Blackpool Trial Tables 24 5 30 min. at 8 p.s.i.g.; 60 min. at 4 p.s.i.g. about 94

United States of AmericaWashington D.C., I966 24 5 To 8 p.s.i.g. in 3 min.; to 4 p.s.i.g. in 4 min.; to II7

O p.s.i.g. in I IO min.Seattle, Washington 25 6 Intermediate details not known | 125

I8


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

Decompression Sickness: A Reviewz

limited, and that compulsory radiological examin-ation of major joints before and during employmentshould be introduced.Some improvements in the general requirements

in the Regulations could also be made to bring thegeneral level ofconduct of compressed air operationsby contractors up to that now observed by the mostexperienced firms. In particular, recording baro-graphs should be required on all manlocks.

Medical Supervision Under the I958 Reg-ulations there must be an Appointed Doctor tocarry out the required medical examinations. In anemergency any duly qualified medical practitionercan be called upon, but it is not stated whatqualifications or training he should have for thiswork. The doctor's function is not related to theuse of the medical lock, for which the only require-ment is that it shall be constantly under the chargeof a person trained in its use. Supervision ofrecompression and other treatment, and respon-sibility for the medical attendants and first-aidservices are not at present included as part of thedoctor's duties. In practice, contractors usuallyarrange with the Factory Inspectorate that theAppointed Doctor for the contract will also act asits medical officer. Such appointments are almostalways given to local general practitioners as apart-time post.This system can work very well but it has

unsatisfactory aspects. The great majority ofdoctors have no experience of compressed airproblems and are ill-equipped to deal with them.No formal training is available, even if the prac-titioner had time to take it. The contrast betweenthis situation and the specially trained medicalstaff of the Navy is very great. In other countries,full-time doctors are frequently appointed to com-pressed airwork contracts, and future Regulations inBritain might well specify trained medical officersat least for large contracts employing say 20 men ormore in compressed air at any one time. Forsmaller numbers of men, a short period of formaltraining for the doctor should be required. Themain problem of a full-time medical appointment isthe short-term projects and sporadic nature ofcompressed airwork, and the change in location fromcontract to contract which would tend to determost doctors from making a career of this work.A solution might be to arrange secondment of navalmedical staff to such contracts at cost to thecontractor. A start might be made in ensuring fullmedical supervision of compressed air work ifprovision for a full-time doctor, fit to enter com-pressed air, were included in the specification forthe contract when it is put out to tender. Formal

training for lock keepers and medical attendantsshould also be included as a requirement in theRegulations.

Developments which may markedly affect futurecompressed air work are the use of analoguecomputers to control decompression (Stubbs andKidd, I967), decompression with oxygen, the useof ultrasonics to detect the formation of bubbles intissues during decompression (Walder, I967), andinvestigations into the rate of inert gas uptake andelimination in tissues and the aetiology of inert gasnarcosis which are being investigated at the RoyalNaval Physiological Laboratory.The tendency to use longer decompression times

should lead to the provision of more comfortableand better equipped locks to which men can berapidly transferred from the man-lock by decanting,that is, by a fast decompression and recompressionwithin a few minutes.

In contrast to this trend, Hills (I966) has pro-posed a new hypothesis, based on the empiricalschedules of pearl divers off the Australian coast,which allows a saving of about one third of thedecompression time for a dive of 40 minutes'duration in I50 feet of water. Hills, from exper-imental evidence, postulates an inherent un-saturation of normal tissues, random nucleationfor gas phase separation, and random cavitations atliquid-liquid interfaces. Further experimentalwork based on his theories is being carried out, butat present the implications for deep diving and, inparticular, for work in tunnels and caissons remainto be assessed.One desirable change which might follow these

developments is the replacement of the presentcasual and unstable labour force by a relativelysmall but highly trained group of specialist minerssimilar to the professional divers, capable of usingmechanized equipment rather than muscular force,and employed under close medical and engineeringsupervision.

I am indebted to members of the Medical ResearchCouncil Decompression Sickness Panel and its Chair-man, Professor D. N. Walder, for the stimulus of manyinteresting discussions on compressed air work. Dr. P. D.Griffiths, of the Central Registry of Compressed AirWorkers, has helped in preparing the tables and readingthe manuscript.

REFERENCES

Annual Report of Chief Inspector of Factories on IndustrialHealth I963. (I964). H.M.S.O., London.

Barnard, E. E. P., and Elliott, D. H. (I966). Decompressionsickness: paradoxical response to recompressiontherapy. Brit. med. J., 2, 809-8 IO.

I9


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

Behnke, A. R. (I967). Blood changes in decompression sick-ness. 2. In Decompression of Compressed Air Workersin Civil Engineering: Proc. International WorkingParty held at the Ciba Foundation, London, I965.Ed. McCallum, R. I., pp. 276-280. Oriel Press,Newcastle upon Tyne.and Shaw, L. A. (I937). The use of oxygen in thetreatment of compressed-air illness. Nav. med. Bull.Wash., 35, 6I-73.

Bennison, W. H., Catton, M. J., and Fryer, D. I. (I965).Fatal decompression sickness in a compressed-airworker. J. Path. Bact., 89, 3I9-329.

Bennett, P. B. (I966). The Aetiology of Compressed AirIntoxication and Inert Gas Narcosis. Pergamon Press,Oxford.

Poulton, E. C., Carpenter, A., and Catton, M. J. (I967).Efficiency at sorting cards in air and a 20 per centoxygen-helium mixture at depths down to I00 feet andin enriched air. Ergonomics, 10, 53-62.

Berger, L. B., Curry, T. F., Watson, H. A., and Pearce, S. J.(I964). Safe Use of Respiratory Protective Equipmentin Work in Compressed Air: Detection and PhysiologicalEffects of Gases Encountered. U.S. Dept. Interior,Bureau of Mines, Washington, U.S.A. (R.I. Bureau ofMines, Report of Investigations 6540).

Bert, P. (I878). La Pression Barometrique, Recherches dePhysiologie Expirimentale. Masson, Paris.

Boycott, G. W. M. (I906). Caisson disease at the NewHigh-Level Bridge, Newcastle upon Tyne. Proc.Instn. Civil Engrs, I65, 23I-237.

Catton, M. J. (I967). Blood changes in decompression sick-ness. i. In Decompression of Compressed Air Workersin Civil Engineering: Proc. International WorkingParty held at the Ciba Foundation, London, I965.Ed. McCallum, R. I., pp. 268-275. Oriel Press,Newcastle upon Tyne.

Cockett, A. T. K., and Nakamura, R. M. (I964). A newconcept in the treatment of decompression sickness.Lancet, I, II02.Nakamura, R. M., and Kado, R. T. (I965). Physio-logical factors in decompression sickness. Arch.environm. Hlth, II, 760-764.

Collins, J. J. (I962). An unusual case of air embolismprecipitated by decompression. New Engl. J. Med.,266, 595-598.

Davidson, J. K. (I964). Radiology in decompression sickness:The Clyde Tunnel. Scot. med. J., 9, I-9.

Decompression Sickness Panel Report (Medical ResearchCouncil). (I966). Bone lesions in compressed airworkers. J. Bone Jt. Surg., 48B, 207-235.

Duffner, G. J. (I955). Review of Paton and Walder; Com-pressed Air Illness (I954), vide infra. Arch. industr.Hlth, iI, 87-88.

- (I962). Decompression sickness and its preventionamong compressed air workers. Metropolitan Engin-eers, Seattle, Washington, U.S.A.

Fournier, A. M., Jullien, G., and Leandri, M. (I965). LaMaladie Osteo-articulaire des Caissons. Masson, Paris.

Fulton, J. F. (1948). Aviation Medicine in its PreventiveAspects. Oxford University Press, London.

Golding, F. C., Griffiths, P., Hempleman, H. V., Paton,W. D. M., and Walder, D. N. (I960). Decompressionsickness during construction of the Dartford Tunnel.Brit. J. industr. Med., I7, I67-I80.

Gribble, M. de G. (I960). A comparison of the 'highaltitude' and 'high-pressure' syndromes of decompres-sion sickness. Ibid., 17, I8I-I86.

Griffiths, P. D. (I960). Compressed Air Disease. M.D.Thesis, University of Cambridge.

- (I967). A standard therapeutic recompression procedure.In Decompression of Compressed Air Workers in CivilEngineering: Proc. International Working Party heldat the Ciba Foundation, London, 1965. Ed. McCallum,R. I., pp. 229-23I. Oriel Press, Newcastle upon Tyne.

Haldane, J. S., and Priestley, J. G. (1935). Respiration (NewEdition). Oxford University Press, London.

Haxton, A. F., and Whyte, H. E. (I965). Clyde Tunnel:Constructional problems. Proc. Inst. Civil Engrs, 30,323-346.

Hempleman, H. V. (I967). Concepts underlying Britishnaval recompression procedure. In Decompression ofCompressed Air Workers in Civil Engineering: Proc.International Working Party held at the Ciba Found-ation, London, I965. Ed. McCallum, R. I., pp. 232-240. Oriel Press, Newcastle upon Tyne.

Hill, L. (I912). Caisson Sickness. Arnold, London.Hills, B. A. (I966). A Thermodynamic and Kinetic Approach

to Decompression Sickness. Libraries Board of SouthAustralia, Adelaide.

Institution of Civil Engineers. (1936). Report of the Com-mittee on Regulations for the Guidance of Engineersand Contractors for work carried out under Com-pressed Air. London.

Jones, R. R., Crosson, J. W., Griffith, F. E., Sayers, R. R.,Schrenk, H. H., and Levy, E. (I940). Administrationof pure oxygen to compressed air workers duringdecompression: prevention of the occurrence of severecompressed air illness. J. industr. Hyg., 22, 427-444.

Kita, H. (I964). Effect of oxygen breathing for the preventionof caisson disease. Proc. XIVth int. Congr. occup.Hlth, Madrid, I963. Excerpta med. Foundn., Int.Congr. Series, No. 62, pp. 771-772.

Kooperstein, S. I., and Schuman, B. J. (1957). Acutedecompression illness. A report of 44 cases. Industr.Med. Surg., 26, 492-496.

Lanphier, E. H. (1966). In Fundamentals of HyperbaricMedicine, Committee on Hyperbaric Oxygenation,No. I298. Nat. Acad. Sciences, Nat. Res. Coun.,Washington, U.S.A.

Lewis, H. E., and Paton, W. D. M. (I957). Decompressionsickness during the sinking of a caisson. A study ofsome of the factors in the pathogenesis of caissondisease. Brit. J. industr. Med., I4, 5-I2.

Liebow, A. A., Stark, J. E., Vogel, J., and Schaefer, K. E.(1959). Intrapulmonary air trapping in submarineescape training casualties. U.S. armed Forces med. J.,IO, 265-289.

Meesters, J. N. (I967). Decompression procedures used inHolland. In Decompression of Compressed Air Workersin Civil Engineering: Proc. International Working Partyheld at the Ciba Foundation, London, I965. Ed.McCallum, R. I., pp. 22-27. Oriel Press, Newcastleupon Tyne.

Ministry of Labour and National Service (1958). StatutoryInstruments. I958. No. 6i. Factories. The Work inCompressed Air Special Regulations, 1958. H.M.S.O.,London.

Nashimoto, I. (I967). The use of oxygen during decom-pression of caisson workers. In Decompression ofCompressed Air Workers in Civil Engineering: Proc.International Working Party held at the Ciba Found-ation, London, I965. Ed. McCallum, R. I., pp. 217-227. Oriel Press, Newcastle upon Tyne.

Parkin, A. (1905). Caisson disease, including the physiologicaland pathological effects of compressed air. Nthld. andDurham med. J., 13, 96-I32.

Paton, W. D. M. (I963). Review of Rose, R. J. (I962), videinfra. Brit. J. industr., Med., 20, 253-254.

20 R. I. McCallum


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


- (I967). Physiological basis of decompression tables incivil engineering. In Decompression of Compressed AirWorkers in Civil Engineering: Proc. International Work-ing Party held at the Ciba Foundation, London, I965.Ed. McCallum, R. I., pp. 2-13. Oriel Press, Newcastleupon Tyne.and Walder, D. N. (I954). Compressed air illness.Spec. Rep. Ser. med. Res. Coun. (London), No. 28i.H.M.S.O., London.

Pol, B., and Watelle, T. J. J. (i854). Memoire sur les effets dela compression de l'air appliquee au creusement despuits a houille. Ann. Hyg. publ. (Paris), 2nd Series,I, 241-279, quoted by Bert (vide supra.), p. 379.

Poulton, E. C., Carpenter, A., and Catton, M. J. (I963).Mild nitrogen narcosis, Brit. med. J., 2, 1450-I45 I.

Report of a Committee appointed by the Lords Commissionersof the Admiralty to Consider and Report upon theConditions of Deep-Water Diving (1907). H.M.S.O.,London.

Rose, R. J. (I962). Survey of Work in Compressed Airduring the Construction of the Auckland HarbourBridge. Special Report No. 6, Department of Health,Wellington, New Zealand.

R6zsahegyi, I. (I959). Late consequences of the neurologicalforms of decompression sickness. Brit.J. industr. Med.,I6, 3II-3I7.

(I967). Neurological damage following decompression.In Decompression of Compressed Air Workers in CivilEngineering: Proc. International Working Party held atthe Ciba Foundation, London, I965. Ed. McCallum,R. I., pp. 127-I37. Oriel Press, Newcastle upon Tyne.

, and Fried, L. (I963). Untersuchungen uber dieEntstehung und Dynamik der chronischen Osteoar-thropathie der Caissonarbeiter. Z. ges. Hyg., 12, 915-924.

Ryan, P. (1929). Reminiscences of a retired sand hog,Compressed Air Magazine, 34, 2647-2652.

Snell, E. H. (I896). Compressed Air Illness. Lewis, London.Stubbs, R. A., and Kidd, D. J. (I967). Pneumatic analogue

computer control of decompression. In Decompressionof Compressed Air Workers in Civil Engineering:Proc. International Working Party held at the CibaFoundation, London, I965. Ed. McCallum, R. I.,pp. 255-267. Oriel Press, Newcastle upon Tyne.

United States Navy Diving Manual (I963). Navy Depart-ment, Washington.

Walder, D. N. (I948). Serum surface tension and its relationto the decompression sickness of aviators. J. Physiol.(Lond.), 107, 43-44 P.

(I963). A possible explanation for some cases of severedecompression sickness in compressed air workers.In The Regulation ofHuman Respiration. Ed. Cunning-ham, D. J. C., and Lloyd B. B., pp. 570-572.Blackwell, Oxford.

(I966). Some problems of working in an hyperbaricEnvironment. Ann. roy. Coll. Surg. Engl., 38, 288-307.

(I967). In Report of Annual Meeting of AmericanSociety of University Surgeons. Brit. med. J., i,559-560.

Whyte, H. E. (I967). A modification of the British decom-pression table used at the Clyde Tunnels. InDecompression of Compressed Air Workers in CivilEngineering: Proc. International Working Party held atthe Ciba Foundation, London, I965. Ed. McCallum,R. I., pp. I70-I76. Oriel Press, Newcastle upon Tyne.

Zannini, D. (I967). Decompression sickness. Electrocardio-graphic changes in caisson workers. In Decompressionof Compressed Air Workers in Civil Engineering: Proc.International Working Party held at the Ciba Found-ation, London, I965. Ed. McCallum, R. I., pp. ii-i6o. Oriel Press, Newcastle upon Tyne.

21


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

Documents

Decompression Sickness: AReviewdecompression and gave credit to Pol and 2 Watelle (I854) as the first authors who had attempted to explain decompression accidents. Pol and Watelle