AN OXYGEN TOXICITY COMPUTER. (U)JUL 80 T E BERSHAGE. F R SORKAT

LNCLASSIFIED NAVHLTHRSCHC-80-28 N

-7-81

AN OXYGEN TOXICITY COMPUTER

T. E. BERGHAGE & F. R. BORKAT

REPORT NO. 80-28 /

S.JUN 15 19810

A

NAVAL HEALTH RESEARCH CENTERP. 0. BOX 85122

SAN DIEGO, CALIFORNIA 92138

NAVAL MEDICAL RESEARCH AND DEVELOPMENT COMMAND

BETHESDA, MARYLAND

AI

VAn Oxygen Toxicity Computer.

T. E. Berghage

Naval Health Research CenterP. 0. Box 85122

San Diego, California 92138

F. R. Borkat

Naval Ocean Systems CenterSan Diego, California 92152

1==4

Report No. 80-281 supported by the Naval Medical Research _n4 I1 4Ippmnt Command, Bethesda,

Maryland, Department of the Navy, under research Work Unitq M0099-PN:00111157. The views pre-

sented in this paper are those of the authors. No endorsement by the Department of the Navy

has been given or should be inferred.

Environmental Physiology Department

SUM~R

Pional Problem.

Clinical and operational use of hyperbaric oxygen is current ly I inmited by fixed exposure

values published in the U. S. Navy Diving Manual. 'liese published limits do not take into

consideration the priDr exposure history of an individual, do not allow for fluctuating

oxygen partial pressuies, and do not provide the operational personnel the flexibility

that is physiologically possible.

2eaicaL Ojective.

-4-This investigative effort was undertaken to explore the feasibility of developing an

ox',,en exposure computer for tracking an individual's hypIerbaric oxygen exposure and esti-

mating the toxicity risk at any point in tbae.

A2proach,

- A survey of the literature was performed to locate experimental data on time limits

for human exposures to various oxygen partial pressures. Animal experiments were performed

to evaluate a time weighted average metric for quantifying oxygen exposure. Finally mathe-

matical algorithms were developed for estimating human oxygen toxicity risk based upon the

best information presently available.

Results.

* Aiimal experiments seem to iindicate that a time weighted average can be

used for evaluating an accumulative oxygen exposure.

* Data on human exposure limits published in the scientific literature is

extremely variable and is subject to influences, both physiological and

environmental, that are hard to control in operational settings.

* ,lgorithms were developed based upon the best scientific information presently

available and were used in a computer model designed to aid clinicians in

assessing the risk of oxygen toxicity during hyperbaric oxygen exposures.

onciuzions.

-)'he development of an oxygen exposure computer seems to be very feasible and desirable

for extending operational flexibility. Critical elements needed in this development aie:

improved human data on oxygen exposure limits and validation of the time weighted average

as a metric for evaluatin, hunan oxygen exposure history.

BACKGROUND

It is well recognized that elevated oxygen partial pressures can be used advantageously

in a number of operational and clinical settings. Operationally oxygen has been used in closed

circuit breathing apparatus to ensure that underwater swimmers are undetected during clandestine

operations. It has also been used to reduce the risk of decompression sickness and decrease

the time required for decompression. Clinically hyperoxia has been used in the treatment of

decompression sickness, air emoblism, carbon monoxide and carbon dioxide poisoning, gas gan-

grene, and osteomyelitis. There are probably additional conditions in which oxygen can be

advantageously used, but supportive research is lacking.

The beneficial use of high partial pressures of oxygen is limited both operationally

and clinically by its acute and chronic toxicity. As Clark (3) has pointed out, the poison-

ing effects of oxygen have an impact on every living cell. "Given sufficient duration of

exposure, a toxic inspired PO2 will cause functional disruption and cellular damage in any

organ system of the body. The specific manifestations of 02 poisoning that limit the safe

duration of hyperoxia vary with the level of PO2 and the presence of other factors ... "

The two most widely recognized effects of oxygen poisoning occur in the pulmonary and

central nervous systems. Paul Bert (2) first observed symptoms of oxygen poisoning in the

nervous system of lower animals. He reported that at high oxygen partial pressures convul-

sions occurred with little warning after only a very short exposure. J. Lorrain Smith (14)

followed with the observation that animals exposed to moderately high tensions of oxygen

over prolonged periods suffered severe and finally fatal pulmonary damage. Numerous studies

followed in which both animals and humans exposed to hyperoxia produced similar or analogous

results.

An active part of the oxygen research program has been devoted to identifying ways of

reducing, delaying and eliminating its toxic effects. A number of drugs have been evaluated

and found unsatisfactory (6). The one generally accepted and now routinely applied technique

for postponing oxygen toxicity is the systematic alteration of the PO2 level. This inter-

mittency technique was first described by Lambertsen (10) and has subsequently been researched

by Penrod (13), Clark (3), Hall (7), Paegle et al. (12), and others. The general conclusion

from all of this work has been that intermittency of high and low oxygen partial pressures

greatly extends oxygen tolerance. The underlying mechanism at work during the intermittent

oxygen exposure has not been identified and the utilization of intermittency, although ex-

tending the acceptable dose range, seems to have complicated our understanding of acceptable

exposure limits. Previously accepted limits for continuous exposures similar to those shown

in Figs. 1 and 2 now have to be recalculated to accommodate fluctuating oxygen exposures.

Assessment of OxIYgen Risk.

The risk of oxygen toxicity during operational use varies greatly depending on the extent

of the exposure and a number of potentiating factors such as elevated alveolar and tissue

carbon dioxide, exercise, immersion, gas density, and temperature just to mention a few. A

further complication in the use of oxygen is that tolerance both between and within indivi-

duals, is not presently predictable. Attempts to correlate oxygen tolerance with age, height,

weight, physical fitness, athleticism, smoking, ingestion of alcohol, psychological health

and personality traits have failed (5). The best we can do at present is describe the oxygen

3t

-- & A

toxic ity phenoimenai and estimiate the risl :is sociated wivth icon uiiu exIx>iire. I l

restricted knot~ledge based onl coilt inuoLIUS CXx)iireSIV 1 J1 1 iiit iil i itot ill the A1,111 l i;It lon)

of this inforim~tioin and has forced W: tO adIopt I is idk I iin: I OPi thoie -,(' (0, xei and r- *t riC

ted the flexibility of fielid persomlu

~4o'zr'aw-tn c eo L:UKtj

'lhe Ut. S. Nv' perat jona~l use ot f oxviio I I iIi I t ed to\ I i xcd exposure va Ilime juibl i shed

ill tile U . Pk', LlVviu Muiluai. No attempt is ma1:de to, provilde operitm.rs or c ni si

thle field ki th thle flexib ili t to assess tile risk of x ~Ii tio ' *t 5, the requirements

of- their operation. For exam11ple, anll aiilerlwat er demolit ionl tearil 1D )t strl Puiue jlak~inl_ a clIan-

deCst iMc SWIM into an1 un1"frliendly harbor is ing a cloised ,ci rcut ox\geui i-cireat her, I f confronted

hea mine defense net, would no0t be author i zeCd to ditxe deeperl tli W feet to i ( Uyi nder it.

flowever, a short, deeper dive may do) little to iiAlter ti is, oxvteii tox icitle risk.

A cl inical example, often enicouintered inl tile tre:1tiiient ofi CN- lois ii s icklii-<

inolves the patient who upon reachin2 the end of thle 00i tot o i hr'itliii period during

a Navy oxygen treatment table 0, St ill has residual S vmp tos Te pIvs i cian ll I li ke to

extend the treatment anid give the pat ient thle beiiefit i)f aiddit i011 naIOXN. eu t ime , bult thle5. ic.& uny ~ana. nialIlows for one ait jina ox\Lgen pet too. flw; does hevalut

thle irisk associjated with such a dec is ion It would be helpfu1,1 llii hoth IOf thseij)crIIi ioal

s it uat ions if thle field personnel had aivailable tii them a cailculaition device for- est imatinug

tle risk iSSOClIatetd With their- dec is ions, rather tlu a fixed restrict iyei limiit.

'a- : tr~ ?_ ' Ux:,?en Rjosure.

In1 1108 Hi Its; and lbssett (9) suggestedl what the\, termed the princ ipie of' sulperpos itioni

fori inltegrating, the effects of fitictriat iii- ixgeti levls With thist j~'lin Ie tile\ pr-Iosed

Iilit i1ciite oxygeOn toxicitv is aiddit ive with respect to o\:\- en parltialI pies ,ilre thut not with

i>itto time. T1hey produced a demonst rat ion of' their superpos it in priniicle uising sixii;(Fi g. 3). Th'le res ult p rovidedl impressive support for tile conicept aiiil st mii ted lIi Its

ft(, devejlp anl oxyen dlose calclator. 'le cal CULlati MRS Used to taJbuilate an OXygenl

e q~osure )I I stoiry in terms of a Ciuuuiat i Ve tOxygen lox ic i tv I ndex warti i erhanps tile higgest

di .x utgeof' the Hills approach is t hat thle idex is not directly interpretable in termis

(d (x\ genl partil pressures and the existing t ime Ilimiits for continuous exposrtes . The

ma ;oil cont ri but ion of this work lies in its demonstration that the toxic effects; of vardims

s e rliallv p resenlted oxygen partial pressures are relIa ted in a li nea r fash iout. 'Ibis latter

fidrig led BerghIge (1) to explore the use of a time-weighted average :is an oxvge epsure

Ilie aulthor. expos;ei over 200~ rats to various inteniti ttencv schedules (Table I .

I flue ti -.gI te ciiu e iles il re Calcila ted 10l r eah flet ilit ing Cxp)oSIC s Ri eatidCompa red wi1thI

rt ij t s obta1 i neilt I-oil) emit iiitiuois oxygeut explosuIle.S. 'Pie resuilts ate sbimii iii Fi g. ..Al thluuigb

here I- a rotdeal Imiorie vaIr i abiii t in) thle tilutiiatiiig oxygenl exposure drita Compared with

the cont I nuiouis exposures, the mnatliemlut ica I descri pt "s of the tisi) relit ioush ilsa e ver much

aIIke.

Relationship Between Oxygen Pavtial Pressure and Time of

(Oiset of Convulsions for 500 of the Exposed Animals

(but inuous Oxygen Time-Weighted Average ofExllosures Fluctuating Oxygen Exposure

1122 1141y -183.34 + X y = -189.9b +

r = .98 r = .84

3io., SE = 48.b

I- = 157.12 F = 53.7

N = lime to convulsions for 5O' of the animials

X Partial pressure of oxygen

measured for t.,o continuous exposures)

(time-weighted average P0, for the fluctuating exposures)

r = (brrelation coefficient (Pearson product moment)

SE Standard error of the estimateF: = easure of' statistical significance

Similar results have been ebtaiad i% our analysis of the data presented by Clark (3), Fall

(7), and hills (8). The correlations between the time-weighted average oxygen partial. pres-

sure and time of onset of .yge toxicity symptoims for these three studies are well above

that obtained in the Berghage (1) study. The respective correlation coefficients are .98,

.98, and .99. The results of all of these studies seem to sup-ort the idea that a time-

weighted average oxyger partial pressure can be utilized as a means of assessing the magni-

tude of in intermittent ox)vzen exposi. t.

The animal -tudies described !_,bove may have provided the bridge necessary for the

it iii :.'a tion -)I ;l l - ,vCi toxicity data. Being able to calculate an oxygen dose

,ir-(>, .I Iluctitlnog ox 'gin exposure allows one to take existing scientific literature

i t oti ti_. it in 111 opieratiunal setting. Having the calculated dose in a form that does

it? l ltm ire a tal tl~lllslat ton or conversion for interpretation is an important feature

,11 till- alproich. Finally, it is essential that the informatioa on oxygen toxicity be pre-

co, ii a ri', probabil ity font iv opposed to a fixed limit, thus giving the individual

at thu -;cullo the flexibility ic/-he neceds to carry out the operation. A flow diagram forau1'!l ;J c;alcIulating s)stelll is showil in Fig. .

Rccent advances in microprocessors have made possible the accomplishment oi all of the

i etivc., de.scribed above. The critical element in the development and acceptance of anis the acceptance of the time-weighted average as a measure of oxygen exposure. Existing

alimal stndies support this concept, but human data is needed for verification. The on!y way

tn obtain the nteded himnin data with the current prohibition on humn experimentation is to

build the OTC and put it into the field as a monitoring devicc. Toward this end we have

conducted an extensive review of the scientific literature on human hyperbaric oxygen expo-

,fin c , II I, I 0. he data from the scientfic literature were analyzed using the prin-

tiple (it the time-teighted average P0,. Frum this analysis two sets of equations were

dcir ied: one or the clinical setting in which a resting subject i, in a dry chamber en-

i ron ent and the other for the operational setting in which the subject is irniersed and

e'xerc II II.4 lor t li' cli ical set t ing the following infonat ion is est imated

I the probability of central nervous system (UNS} toxicity

occirrinig g vv cn the present exposore histor\ ;

2- tile t ime remaiiniug before a major increase in the pfobabil ittv

of CNS symptomu;

S thie Ii kel reduct ion in vital capacity that the pat i ent is

cxpL'.'lelc l1W with a ziven exposure history, and

Ithe tine remaining before the patient may r._ach a debilitatin-"

level of pulmonairy oxygen toxicity.

'11 c t lot ils tor the operat iounll setting are lijmited to CNS symptomsuL (itens (11 id (2)

hotet' li>e o1 the operational dive profiles involved.

Pli ,i' )Jproessor device shown in Fig. () was selected for use as the OTC because it i>snlll II redi I'I portb. ,and proc ides the progrxini ng f'lex hi I i ty ib ich wi II a lo, updating

of t ilc c.1lcuition algorithiuns as more ini-orliation becomes available. X we learn more ahoit

the WlLIderI . il', ox gell toxicity nechanisms, we will be able to improve the estimates provided

!I\ thet OT' md sharpen our" use of the hyperbaric oxygen tool

it th this tvpe of infomation avai lab le, tile o)perational I personnel in tile field ynl

e-a Iltate tile risks associated with their current situation and alter their pressure. time,

and ,-as mixtuic profile to sta' within acceptable risk limits.

Bergluae, '. E. L979. Yutes on three recompless it lreatlncrt research pro)ects. In

.1. '. lavis (lid.), eiccat''eu " o; ,tc Je.":ro.jou eiekncn, andi ,tc ai 'sm.'::n,.

I IndersUa %l(dica 1 ,Soc ictv KI'k>lol), ,'J0U1 rV,

. Bert , 1'. 19.13. oo:, tro. I enr j.. : N',;oj'ch'€ in 'ki 1%,:, t o 1',. '0 !OT, ... 1. A.

litchcock, Trans., fihio: (bllege Book Co., original lv published, 18781.

*. Clark, J1. M. 1974. "he toxicitv of oxygen. A!-. .7ax,. .. ' . 2i., 11 1:.11 -5i.

4. Clark, J. I. and C. .. l.Litbertsen. 1971. pluhtnnlary\ oxygen toxicity: A revim,.Ph.wma'I. ; 'e. , :.3:37- 133.

lnald, K. It. 197. Oxygen poison in in man. Uv- t..,od.. ., 1:(17-

1). I (.'rs'himalli, R., II. I.. tilbert, and 1). Cass;aimli (-. H9S8. Elffect of' variois sbstalnces (lM

surviVal tiorY's Of mIice exposed to different high oxvgen tensions. ... ;.h.'h '. .Y:

I laIl , l). A. 1907. 'le influence of the svstematic fluctuation of- Io, up)n the 1 na:tire ;11d

rate of the development of oxygen toxicity in gui nea pigs. ,Nt tlt's is, (;raduiate School

of Arts and Sciences, University of Pennsylvania.

8. Hills, B. A. 197(. A ctmi, itive oy.,..n toxicity index allowing for the regression of

effects at low inspired oxygen partial pressures . M. Report 4/76, Minist 0i Iki-fnce,

rlidon.

9. tlills, B. A. and A. N. tIossett. 1908. Predicting the occiurrence oi oxgell convutlsi ow.

l .A tes;t o) the principle o)f supl~erposition. Rl)rt II% 2'7(, NIRC, III.K.I.,0

10. Lambertsen, C. J. 1955. Respirator) and circulator) actions of high oxygen pressure.

In: Proceedings of the Underwater Physiology Symposiwn (Pub1. 277). Natl. Acad.

Sci-Natl. Res. Council, Washington, D.C.

11. Lambertsen, C. J. 1965. Effects of oxygen at high partial pressure. In: Handbook

of Physiologfy; Respiration, Chapter 39, Washington: Am. Physiol. Soc.

12. Paegle, R. D., W. N. Bernhard, and H. Turndorf. 1977. Intermittent exposure to 40

percent oxygen prolongs rat survival in 100 percent oxygen. Anesth. Analg., 56:

847-851.

13. Penrod, K. E. 1956. Effect of intermittent nitrogen exposures on tolerance to oxygen

at high pressures. Am. J. Physiol., 186:149-I51.

14. Smith, J. L. 1899. The pathological effects due to increased oxygen tension in the air

breathed. J. Physiol. (London), 24:19-35.

I5. Yarbrough, 0. D., W. Welham, E. S. Brinton, A. R. Behnke. 1947. Symptoms of oxygen

poisoning and limits of tolerance at rest and at work. U. S. Navy Experimental Diving

Unit Report 1-47, January.

7

Time To C2onvuls ions fo(Ir)) )I- ft .\n eisjj

L-xpo.sed to Var ious Intcrmj t teucv Sclhudu1 u';

Tim (11oHfigh Time on [k)W Oxvgenl ' line toOxygen HIigh Oxygen *ar Oi iiISPart iii Ipart mal Pressure for 50)PICs SUre Pres sure IN 1\Ithe An i mlalIsYETA S Ui Inue 0luu s O nit u

3.2Cont inuous 0 1410

3.2 is S) 333 . 24 )

3. 2 40) 1' 25

3. 2 lo) 33.2 8)) It)2

3.9 5

3.9 qU5

3.9 ~80 1 3

I.1 Son Iun S 81.2 is 1474.2 15 )127

4 .2I 20.2)4240 1 04

41.2 40 s 41. 40) 10)11

BREATHING 100% 02 DEPTH/TIME LIMITS10

20 10Runs

2511111111 5 Runslo

630. 11 Runsu.

• • 5 Runs-ConvulsionNausea

40 • • 13 Runs-Convulsion2 Nausea

5 Runs-Ringing Ear

50 • 2 Runs-Near Convulsion

", "f, A,- I0 30 60 90 120 150 10 2;0 240

Minutes

- Special Operations* Normal Operations*4,11 Depth/Time Limits Depth/Time Umits

Effects Noted Beyond Limits

*Considered safe for dives Involving moderate work withminimal C02 Inspired gas.

Fig. 1. - Central Nervous System toxicity limits for continuousexposure to hyperbaric oxygen while imersed and exer-cising. (t. S. Navy Diving MIanual)

k0

7

-PULMONARY LIMITS

x 30

2i CNS ASYMPTOTEIr 2 LIMITS - 2 0 ATM

z

.2'J0AM

1 2 3 4 5 6 7 8 9 1j1TIME (HOURS)

I i. 2. -Ceutr~i1 Nervou-; Svs-ten 11i pilimoiry tox -itv I irix

for COflt iflIUoS eXpu0I1ro to hovuiCOX%\SCfl for- !I-%resting subject s. (Clairk and ILirnhert son

9-x1000

8 1 P2 - - - - - - - - - -

IPlo

2

5-P2 0

Inmi Hg4

3.

2.t, 6 mins=5 minst

1 2 3 4 5 6 7 810

P1 in mm Hq

Ci e.. - 1c re Ihm i ousmip h'tL'en Lcm,,cuit I\(, expo umv t, pure,

oxyteen (f m i mul . ifl I m.( .1 \ Tjinmutt Z 1 1'+,for equal unset I ;lIt, .,i 10!W1 jL~ n i w III I rlt

00O

I-i

900 X CONTINUOUS OXYGEN EXPOSURE

!t So FLUCTUATING OXYGEN EXPOSURE

<Z700 kit

Uit

0600o

0

-500a.WIAJ

"M 400

0300I-

CL

cn 200Ii.I

S100-

0 i." 2b S.0 4.0 5:0

OXYGEN PARTIAL PRESSURE (ATA)

Fig. 4. The relationship between a time-weighted oxygen exposurecalculation and the onset time for 50% of the exposedanimals.

D EPT H (O sw I D E PT H (m sw l D AY DE N P E RC L N I

D M p

L -____--E7__ ---- - 2

M+I P A Po +33 AM 10

P

tt

S ELAPSE TIME 44 t

Po

CNS TOXICITY

I----------------' _

"'P P', ,

TIME TO SYW ', !i ?, ' L

PULMONARY TOXICITY

PERCENT REDUCTION IN |VITAL CAPACITYI j P

TIME TO SYM. 1(: '13,97 (P ,

:~.5. liloci. diagr m t101f C-I11CII itI g ti(, IIsi. o lX(CTI

toxicity occurring dcring a hvperox ic &lh-, l' (-lt:I.

12

Iq0

lig h.- Conceptual design for an oxygen toxicity computer.

UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE (When Date Entered)

REPORT DOCUMENTATION PAGE READ INSTRUCTIONSREPORT__ DOCUMENTATIONPAGE__ BEFORE COMPLETING FORM

I. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER

4. TITLE (and Subtirle) S. iYPE OF REPORT & PERIOD COVERED

(U) An Oxygen Toxicity Computer Interim6. PERFORMING ORG. REPORT NUMBER

7. AU THOR(s) 1. CONTRACT OR GRANT NUMBER(a)

T. E. Berghage & F. R. Borkat

9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT, PROJECT, TASKAREA & WORK UNIT NUMBERS

Naval Health Research CenterP. 0. Box 85122 M0099-PN.001-11S7S, le Lo CA ()2"R

II. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

Naval Medical Research & Development Command July 1980Bethesda, MD 20014 14

14. MONITORING AGENCY NAME & ADDRESS(II dlfferent from Controlling Office) 1S. SECURITY CLASS. (of this report)

Bureau of Medicine & Surgery UnclassifiedDepartment of the Navy IS,. DECLASSIFICATION/oOWmR.ADIMG

Washington, DC 20372 SCHEDULE

16. DISTRIBUTION STATEMENT (of this Report)

Approved for public release; distribution unlimited.

17. DISTRIBUTION STATEMENT (of the abstract entered In Block 20, It different from Report)

18. SUPPLEMENTARY NOTES

19. KEY WORDS (Continue on reverse side If ncecsary And Identify by block number)

Hyperbaric Oxygen Exposure LimitsOxygen Toxicity Recompression Therapy

20. ABSTRACT (Continue on reverse aide If necesear, and Identify by block number)

(U) Despite the availability of oxygen toxicity data in the scientificliterature clinicians in the field employing hyperbaric oxygen have onlya vague idea of the time-pressure limits for oxygen exposures. A jointdevelopment program between the Naval Health Research Center and the NavalOcean Systems Center has produced a prototype Oxygen Toxicity Computer (OTC).This small hand-held device can be used to track the course of a hyperbaricoxygen treatment and provide the attending physician with information on

DD I, 1473 EDITION OF I NOV Os IS OBSOLETE UNCLASSIFIEDS/N 0102,014,6601

SECURITY CLASSIFICATION OF THIS PAGE (Mee DMS Xwed)

- ON

UNCLASS IF lED- >,q!y CLASSIFICATION OF THIS PAGE(When Data Entered)

(a) the probability of central nervous system toxicity, (b) the timeremaining before a major increase in the probability of central nervoussystem symptoms, (c) the likely reduction in vital capacity, and (d)the time remaining before reaching a debilitating level of pulmonaryoxygen toxicity. With this information available the attending physiciancan alter his pressure-time-breathing medium profile to stay withinacceptable risk limits.

UINCLASS IF1IEDSECURITY CLASSIFICATION OF THIS PAGE(WIhsn Data Entered)