\■■iv-T A ■. \ : v . 11-. v v- . v. v v.-«i i~- w .. : . : -v ■ :-s i :; • «.».«* ‘ - - i> •*. s r ■: "s - -N ■ ', • ,\ ' . . •, £ I* • t f\ ' V7 <> ; ; \J [ J. I. . \7 *-• Av;S::Cn :.isd;di,). Oxygen ard air pressure at various altitudes as they influence the efficient functioning of the aviator.- By W. M. Boothty, August 19U2. C.A.M, Report No. 3Ul. Alveolar respiratory quotients; An experimental study *f the difference between true and alveolar respiratory quotients, with a discussion of the assumptions involved in the calculation of alveolar respiratory quotients and a brief review of experimental evidence relating to these assumptions* By J, B. Bateman and W. M, Boothby, lU June 19UU. C.A.M. Report No. 360. The effects of altitude anoxia on the respiratory processes. By H, F, Helmholz, Jr*, J* B, Bateman and W, M, B»othby, August 19hlu c C.A.m. Report No. 36U, Susceptibility to decompression sickness: The effects of prolonged inhalation of certain nitrogen-oxygen mixtures compared with those of exposure to pure oxygen* By J* Bi Bateman, September 19Ulu C.A.M. Report No. 381* The reduction of alveolar carbon dioxide pressure during pressure breathing and its relation to hyperventilation, together with a new method of representing the effects of hyperventilation. By J, B. Bateman, September 19hh* C.A.M. Report No. U28. The effect of pressure breathing upon the skin temperatures of the extremities. By J, B. Batenan and C. Sheard, 2 May 19U$* NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE RESTRICTED Report No, 163 Date June 1943 COMPARISON OF ALVEOLAR OXYGEN PRESSURES, OXIMETER READINGS AND PERCENTAGE OF SATURA- TION OF HEMOGLOBIN. From the Mayo Aero Medical Unit, Rochester, Minnesota, by Walter M. Boothby, Responsible Investigator, and F, J. Robinson, First assistant SUMMARY used I, The oximeter/at the Mayo Aero Medical Unit was a modification by E. D. Coleman of the Millikan oximeter known as Coleman Model 17. The manufacturer’s calibration was checked by (l) reading the per cent saturation of the hemoglobin directly from the oximeter scale and (2) practically simultaneoualy obtaining a sample of arterial blood for analysis of its oxygen content and after equilibration for its oxygen capacity in the manometrio apparatus of Van Slyke and Neill. The correlation is shown in the first chart. II, The special oximeter thus calibrated was used to study the relationship between the oximeter readings of the percentage saturation of the hemoglobin and the partial pressure of oxygen in the alveolar air, A total of 572 observations were obtained and the data is presented in a series of seven charts. The data shows that on the average the percentage saturation of hemoglobin is in excellent agreement with the partial pressure of oxygen in the alveolar air whether the subject is at rest or at work or whether the alveolar air was obtained by the Haldane-Priostley or bag—rebreathing method. The plots of the individual observations show a consxderabl scatter but a variability after all surprisingly small considering the possibilities of error both in the oximeter and in obtaining a corresponding alveolar cxygen pressure NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE RESTRICTED Report No, 163 Date June 1943 COMPARISON OF ALVEOLAR OXYGEN PRESSURES, OXIMETER READINGS AND PERCENTAGE OF SATURA- TION OF HEMOGLOBIN, From the Mayo Aero Medical Unit, Rochester, Minnesota, by Walter M, Boothby, Responsible Investigator, and F, J, Robinscn, First Assistant, REPORT I, In order to check the manufacturer’s calibration of a Millikan sximeter modified by E, D. Coleman, known as the Coleman Model 17 (no. 5769), Code, Power, Sturm and Wood of the Mayo Aero Medical Unit carried out in October 1942 a series of fifteen experiments in which (l) they read the per cent saturation of the hemoglobin directly from the oximeter scale and (2) practically simultaneously obtained a sample of arterial blood from the anticubital artery. The blood was collected in a 20 o,c. syringe containing heparin and a minimum amount of mineral oil, Duplicate samples were immediately analyzed for their oxygen content in the manomotrio apparatus of Van Slyke and Neill, The remainder of the blood was kept (not exceeding two hours) at low temperature in ice water until time was available for determination of its oxygen capacity. This data is reproduced in chart III-8A and shows that there is excellent correlation between the percentage saturation of the hemoglobin read »ff directly on this particular oximeter (manipulated under ideal conditions by trained personnel) and as determined by careful chemical analyses. Only one of the determinations falls just outside the range of + 5 percentage points and ten out of the fifteen observations are within + 2 percentage points. II, The present investigation in which 572 observations were made have been confined to determing whether this same oximeter which showed such close correlations with actual blrod gas determinations would show a correlation with the partial pressure of oxygen in the alveolar air. The averages of the percentage saturations of the hemoglobin obtained at the different elevations as read from the oximeter fall for the most part within 1 or 2 percentage points of Dill’s oxygen dissociation curve (pH 7,4) when plotted against the average of the partial pressure of oxygen in the alveoli whether at rest or at work and whether the alveolar air was obtained by the Haldane or by the bag-rebreathing method, ‘However, the individual observations of the different series show a considerable and apparently accidental scatter, This scatter is sufficiently great to render an individual observation rather doubtful as a criterion of a subject’s safety at simulated high altitudes in chamber work. On the other hand, the oximeter is a valuable instrument for determining a large number of observations which can be averaged to give a reliable mean. The vagaries of the cximeter itself and the difficulty of obtaining alveolar- air samples that truly represent a mean value of the oxygen partial pressure arj complicating factors. At low altitudes there is the added difficulty due to the 2 fact that the dissociation curve is approaching its asymptote and that therefore accidental high readings are less likely to occur and compensate for the accidental low readings due in part to blood being exposed to oxygen in some of the deeper alveoli which is considerably below the average that would be obtained in an alveolar air sample. Finally the oximeter does not reflect the influence of alterations in pH on the dissociation curve of hemoglobin; however, it is unlikely that in the tissue capillaries of the ear the change in pH will be sufficient to affect to more than a slight degree the dissociation curve, at least for two or three hours of moderate anoxia,, There are two ways of setting the oximeter. The method probably most generally used is to assume that approximately the oxyhemoglobin is 95 per cent satura- ted when breathing room air. This assumption is not necessarily true; therefore it is better to have the subject breathe oxygen and set the oximeter at 100 per cent. In our early experiments we used the first method as we hardly anticipated that the averages would be sufficiently consistent to indicate that as a rule if the oximeter was sat at 100 per cent on oxygen that when the subject subsequently breathed air it would indicate 96 to 98 per cent saturation with an average of 97 per cent, (See chart III-10A), In consequence we now always set the instrument on 100 per cent with the subject breathing oxygen. As there are at times fluctuations in the oximeter readings of 2 to 4 percentage points we finally adopted the routine of having the technician record the exact reading of the oximeter either at 10 or 15 second intervals, thus approximating a recording Instrument, All results obtained were included except occasionally when there were technical difficulties in the oximeter, especially when subject was exercising, as evidenced by extreme fluctuations from static electricity (in spite of attempted grounding) or a consistent drift in readings so that a control observation with the subject breathing oxygen was definitely off. The disturbing influence of static was minimized and usually eliminated by liberally sprinkling the carpet-rug on the floor of the chamber with water to increase the humidity when the room air bleeding into the chamber was very dry as in cold winter weather. While an indirect method like the oximeter does not justify the conclusion that as a rule in young individuals the arterial blood is normally saturated to 97 per cent when breathing air, yet it is sufficiently consistent to indicate that direct methods for determination of both the percentage saturation and also the oxygen tension in arterial blood should be used to carefully determine this point. Ill, Two distinct methods of obtaining alveolar air samples in the low pressure chamber at ground level (1,000 feet) and at different altitudes have been used, both at sitting rest and at light work. A, The standard method of obtaining an alveolar air sample is that commonly known as the Haldane and Priestley methodt the alveolar air is obtained by giving a quick, deep expiration into a 3/4 inch hose (with simple mouth piece) about 4 feet long; after the expiration the subject closes the mouth end of the tube with his tongue and the sample of the last part of breath drawn out into a mercury sampling tube. To render the method slightly simpler we use a special valve which is snapped at end of expiration to close off the sampling tube and so constructed that during the expiration the entire tube on the inside is smooth and contains no pockets. B, To obtain what might be a better average alveolar air a new method of collection was devised; a 5 liter rubber bag was attached to the same valve menbionce above instead of the long hose tube; the subject expired quickly and deeply into tie 3 completely empty bag, exactly as in the Haldane-Priestley method and then inhaled two or three times followed by complete expiration. In one series this was done for three expirations into bag and two inhalations from bag, closing the valve at end of third expiration, and in another series there were three inhalations and four expirations. No significant difference was found {chart III-lOC) between 3 and 4 expirations} therefore we will in the future standardize on 3 expirations. IV, The data thus obtained is presented in a series of charts which are in the main self-explanatory and therefore need little additional description. A. In chart III—10A are shown 259 observations of essentially simultaneous oximeter readings and alveolar oxygen pressures averaged for altitude. The oximeter was read several times during the minute before collecting the alveolar air samples by the Haldane—Priestley method. In 202 of these observations the oximeter was arbitrarily set at 95 per cent saturation with the subject breathing air at ground level (1,000 feet) while in 57 observations the oximeter was set at 100 per cant with the subject breathing oxygen, so the oximeter in the latter series read on the average 97 per cent when the subject returned to breathing air at ground level. It is to be noted that the average figures plot very satisfactorily along the oxygen dissociation curve of Dill for a pH of 7*4 and it appears that in normal subject the hemoglobin is on the average 97 per cent and not 95 per cent saturated when the subject is breathing air at or near sea level. B, Chart III-lOB shows the values for 259 individual observations which were expressed as averages in chart III-IOA. There is, as can be seen, a very definite scatter of the individual observations which is smoothed out in the averages of the preceding chart. Chart III-lOBa gives possibly a clearer picture of the correlation of those observations of chart III-lOB which were made with the oximeter set on oxygen. C, Chart III-10C shows the average values of 106 observations. Of these, 58 observations were obtained by the technic of exhaling deeply, three times into (inhaling twice from) a 5 liter bag and 48 observations by exhaling deeply four times into (inhaling three times from) the bag. Note that values for both the three breath and the four breath series are practically identical and when plotted show very excellent correlation with Dill?s curve for dissociation of oxyhemoglobin at pH of 7,4 although there is a slight indica- tion that there is a tendency for the data to fall somewhat to the left of the disso- ciation curves at 12,000 and 15,000 feet, possibly from the effects of the increased ventilation caused by the anoxia which always occurs at these altitudes. D. Chart III-lOB shows the individual observations comprising the average values given in chart III-10C. This figure shows a slightly lesser spread of the individual observations obtained by the bag—rebreathing method than those obtained by the Haldane-Priestley method shown in chart III-lOB, Chart III-lODa gives possibly a clearer picture of the degree of correlation of the data shown in chart III-10D. E« In chart XII-10E are shown the average values for altitude of 207 determinations of which at sitting rest 40 were made by the Haldane-Priestley method and 40 by the bag-rebreathing method (using three breaths); at work 63 were made by the Haldane-Prie stley method, and 64 by the bag-re bre a thing method. The work war. accomplished by having the subject step up on a 5 inch step 16 times per minute time with a metronome set at 80 per minute to obtain 5 beats to alternate legs unr* for elevation by introducing an extra non-elevating step on the floor* 4 The results show a very good correlation both at rest and at work by both methods of collecting alveolar air samples with Dill's dissociation curve (pH 7,4), The experiments at work showed a lower degree of percentage saturation of the hemo- globin by the oximeter than did those at rest but there was a corresponding decrease in the partial pressure of oxygen in tho alveolar air by both alveolar air methods so that the experiments at work as well as those at rest agree closely with Dill's dissociation curve at pH 7,4, ALVEOLAR OXYGEN PRESSURE v. OXIMETER READING la thla ahart ara plottad tha Individual abaarvatlona aa averaged far altltnda In praoeeding chart III-10A. 0* Oxlmatar aat at 95* with aubjaet breathing air at ground laval X • Oxlmatar aat at 100* nth aubjeot breathing oxygen at ground 1®1 a All sub.laots at sitting rsst. The oxyhemoglobin ourvoa for whole blood ara those reported by Major Dill and have boon redrawn from PHYSIOLOGY OP PLIGHT, Wright Plaid, 1940-42, paga 15. Boothby and Robinson Maroh 1943 Thla chart ihova tha correlation of tha oximeter re ,41cja (oxlmeta • aat at 100* aubjeota treat ilng oxygen at [round larel) and the per oe it aaturation o ’ hemoglobin oalou.ated from the ilTeoIar oxygen praaaura, HaUa ■^r?fe~;ny met fST Thla tata aama aa thit ahown by oro ilea In chart IIL-1 >B. All anb j eo ,a at alttlng r int. Boothby and Roblnaon March 1943. OXIMETER READINGS IH PER CENT OXYBSdOGLOBIN Calculations it par oant eat iratlon from al solar oxygen otm of UaJ.- 0111 far «kal i blood, pi 7.4 PITS1CL0CY Of fLIGHT. »rl|M flail, 1*40-42 page IS ALVEOLAR OXYGEN PRESSURE (axa. Hg. ) Mayo Aero Medical Unit COMPARISON OF OXIMETER READINGS AND ALVEOLAR OXYGEN PRESSURES 111—lOBa IX1—10B PER CUT OXYHEMOGLOBIN CALCULATED PROM ALVEOLAR OXYCEI PRESSURES *y»rw« Her. Oximeter No. of Oximeter A1t.02 Toot wt on Obtery. Reading* Pr.i*. 1,000 Air 21 95 99 ItroendlOxveea 12 97 no 3,000 Air 34 94 89 Oxygen e 95 90 6,000 Air 33 91 74 Oxruon 10 93 75 9,000 Air 32 88 61 Oxyi.n 8 90 64 12,000 Air 36 83 50 Oxrt Ko. 1 □ Sub j* it bo. 2 A _ lt_Klu. i-CL OXIMETER READINGS OXIMETER READING IN PER CENT OXYHEMOGLOBIN PER CENT OXYHEMOGLOBIN IN ARTERIAL BLOOD (MAMOMETRIC) (1) Rost - sitting in chair. Work - stepping onto 5 inch stool 16 times per minute with metronome set at 80 to obtain 5 beats to alternate legs used for elevation. (2) Alveolar air methods! (a) Haldane-Priestly. (b) Rebreathing method with single expiration rebreethed 3 times. Oximeter set at 100* with subject breathing oxygen at ground level. The oxygen dissociation curves for whole blood are those reported by Major Dill and have been redrawn from PHYSIOLOGY OF FLIGHT, Wright Field, 1940-42, page 15. Boothby and Robinson March 1943 The observations are averaged according to method of obtaining alveolar air at both rest and at work for each altitude. / IlOT. Act It- Air. Air Ho. of Oziieeter Air. 0? Foe t ltr(l) Method(21 Obserr. Press. 1,000 Rest 0 Haldane 20 96 105 [ground) ».it A Rebreath. 20 96 10Z Work x Haldane 27 95 97 Work + Rebreatb. 30 96 89 10,000 Rost O Haldane 6 86 60 Root £ Rebreatb. 8 89 59 Work * Haldane 11 85 57 Work + Rebreatb. 12 83 51 12,000 Rost O Haldane 8 86 54 Rost & Rebreatb 6 86 53 Work * Haldane 15 82 50 Work ♦ Rebreatb. 11 82 45 14,000 Root O Haldane 6 85 50 Rest A Rebreatb. 6 84 47 Work * Haldane 10 00 49 Work Rebreatb- 11 78 42 Hub.r ef oboerrotiono * 2Q7. This ch .rt sbo rs the correla- tion of oxlnet r readings (oxir.etex set at 100*/ %i lh subject, breathing oxygen at grou»d lercl) and jthe per cent eituretlor of ben>globln calcu- |lated from alr« olar oxygen pre seures. Boojbbp and R.blnaA *areh 1943 Alveolar air t; rebreatMEg MetWTOl 0 - maximum exiiraticn rebrea*'ied 3 tinea. X - maximum on iraticn rebreatked 4 times. This data saije as that showil by circles andl crosses in chart IH-lod. All subjeotj at sitting rest. Calculations of per oest sat\ration from alveolar oxygen pressure were >ased on the oa yhomogloblc association curve of Major «lll 'or whole bloo« , pH 7.4, PHYS;OLOGY OF FLIGHT, Aright Field, 1940-4J page 15. OXIMETER READINGS IV PER CENT OXYHEMOGLOBIN ALTtOLAR OXTCIR PRISSUHI (ms. H£. ) Ill-10n. Mayo Aero Medical Unit COMPARISON OF OXIMETER READINGS AND ALVEOLAR OXYGEN PRESSURES PSR CENT OXYHEMOGLOBIN CALCULATEP FROM ALVEOLAR OXYGEN PRESSURE OXIVITER READINGS IN PER CENT OXYHEMOGLOBIN III-10X Bootbby and Roblnaon Marob 1943 Oximeter set at 100* with subject breathing oxygen at ground level. The Individual observations comprising the averages are shown In chart III-10D. All subjects at sitting rest. The oxyhemoglobin dissociation curves for whole blood are those reported by Major Dill and have been redrawn from PHYSIOLOGY OP PLIGHT, Wright Pleld, 1940-42, page 15. In thl■ chart ar. plotted the Individual observations which w.r. averaged for altitude In proceeding chart III-IOC. Alveolar air by rabrantblng natbodi O - maximum expiration rahreathad 3 tlmai. X - maximum expiration rahronthad 4 tinea. Oximeter eat at 100* with eubjeot breathing oxygen at ground lorol. All subjecta at sitting raat. The oxyhemoglobin dlasoolatlon ourvoa for whole blood are thoao roportad by Major Dill and have been redrawn from PHYSIOLOGY OP PLIGHT, Wright Plold, 1940-42, page 15. Ho. of Ho. of Oxiaotor A1t.02 1,000 3 10 98 96 [ground 4 • »« ioo. 3,000 3 10 95 85 4 8 « 85 . 6,000 3 10 92 72 4 8 K 71 9,000 3 10 88 59 ■* e 89 59 12,000 3 10 83 47 4 8 83 47 15,000 3 8 77 40 4 B 77 40 Observations averaged for each altitude. Hunber of observations • 10«. Bootbby and Robinson March 1943 Alveolar air by rebreathing methodi A - maximum expiration rebreathed 3 times. O • maximum expiration rebreathed 4 times. ALVEOLAR OXYGEN PRESSURE (_. Eg.) ALVEOLAR OXYGEN PRESSURES (an. Eg.) III-10C XII-10D OXIMETER READING IN PER CENT OXYHEMOGLOBIN OXIMETER READING IN PER CENT OXYHEMOGLOBIN NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH • f the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE Report No, 222 Date | December 1943 RESTRICTED "TRACHEAL” VERSUS "ALVEOLAR” AIRi A REVIEW OF THE METHODS OF SELECTING CERTAIN PHYSIOLOGICAL DATA BEARING ON THE DESIGN OF OXYGEN SUPPLY SYSTEM FOR AVIATORS, From the MAYO AERO MEDICAL UNIT, Rochester, Minnesota, by Dr. J, B, Bateman; Walter M. Boothby, Responsible Investigator, Prepared at the request of Dr, Louis B. Flexner, National Research Council, SUMMARY 1, One point cf view favors the tracheal and the other stresses the advantages of the alveolar reference points; the limits and advantages of each are discussed. 2, Attention is drawn to a possible objection to the "alveolar air" equation. It is contended that the role of a relatively stagnant gas layer separating the tidal .air from the pulmonary epithelium has never been adequately discussed, although calculations of the composition of alveolar air depend for their validity upon the absence of any such unmixed fraction. For this reason it is unwise to place too much reliance upon theoretical calculations of the alveolar oxygen pressure under different environmental conditions. 3. The alveolar air equation is further examined from the following points of views (a) In relation■te experimental data on subjects breathing air or pure oxygen at simulated high altitudes, and air-nitrogen mixtures at sea level« (b) In its bearing on the calculation of specifications for the gas mixtures, to be supplied in order to maintain a normal physiological condition in persons at high altitudes. These are compared with the specifica- tions resulting from use of the "tracheal* air standard# (o) In its application to the calculation of "equivalent altitudes” under conditions involving unavoidable anoxia, 4, The practical suggestion arising from this discussion is that formal adop- tion of one "reference point" or the other is unnecessary. As a rule it will make little difference which is used; in the case of the anoxic subject, it may make a good deal of difference but recommendations in such oases should be made after due consideration of all available data. The results of theoretical calculations with dubious numerical factors and an equation of questionable validity should not be too widely applied. 5, A simple algebraic nomenclature is proposed for data on the respiratory gases. This has the advantage that it can be typewritten without use of special adjustments or unusual characters. NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of tho Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE RESTRICTED Report No . 222 Date; December 1943 "TRACHEAL" VERSUS "ALVEOLAR" AIR* A REVIEW OF THE METHODS OF SELECTING CERTAIN PHYSIOLOGICAL DATA BEARING ON THE DESIGN OF OXYGEN SUPPLY SYSTEM FOR AVIATORS. From the MAYO AERO MEDICAL UNIT, Rochester, Minnesota, by Dr. J. B, Bateman; Walter M. Boothby, Responsible Investigator. Prepared at the request of Dr. Louis B, Flexner, National Research Council. FART I. THE COMPOSITION OF ALVEOLAR AIR. a, The purpose of an oxygen supply system is to prevent or reduce that impairment of human physiological function which would occur if the subject were exposed t» the unmodified atmosphere at certain distances from sea level. The final criterion of performance of anch a system is therefore its success, when operating with reasonable economy of materials, in enabling persons using it to carry out any necessary tasks and!, to a smaller degree, in preventing any detrimental after-effects. Performance tests are, however, so unreliable and subject to such considerable personal variation that this criterion can hardly bo used as a basis on which to draw up the specifications of an oxygen supply system. Consequently it i 4 necessary to have recourse to reference points which can be given precise physical definition, with the obvious qualification that apparatus constructed in the light of such secondary criteria must always be subject to any modification suggested by efficiency tests in che field. Some dispute has arisen concerning the choice of reference point. It is prabably agreed that the percentage arterial saturation with oxygen is closely related to efficiency, although the relationship is certainly liable to be obscured by the effects of other physiological variables such as cranial blood flow and by aany sources of individual variation including the hyperventilation of anaxia, As a practical reference point this quantity is not to be considered, because of the lack, on the one hand, of sufficiently consistent and extensive experimental data, and on the other hand, of any generally applicable relationship between arterial saturation und the composition of inspired air. The next reference point which suggests itself is the alveolar oxygen pressure since this must be directly related to the pressure of dissolved oxygen in equilibrium with arterial blood. There has been some academic dispute between those who advocate the use of this quantity and those who prefer simply to use the partial pressure of oxygen in the inspired gas. The contentions of the former group are based implicitly upon two assumptions* (l) That in two given environmental situations a subject is in comparable physiological conditions if the alveolar partial pressures of oxygen and carbon dioxide are unchanged. (2) That there exists a simple algebraical relationship between the respiratory quotient, the composition of inspired air, and the composition of alveolar air, and that this relationship may be legitimately used fer the purpose of extrapolation or interpolation to conditions not adequately covered by experimental data. It is furthermore assumed that this equation shows the alveolar oxygen pressure to depend upon other variables in addition to the oxygen pressure in inspired air, making the use of the latter quantity inadmissible as an index of physiologically equivalent states. The other group, if their position has been properly understood, accept the theoretical validity of the relationship referred to, and therefore do not wish to assert that the partial pressure of oxygen in inspired air is strictly an index of physiological equivalence. They do, however, consider that the use of the alveolar standard and of the theoretical equation connected with it has only the effect of introducing changes which are too small to be detected with certainty in the available experimental data, and which therefore lend to the resulting specifica- tions a misleading and experimentally unwarranted air of precision* This attitude will be examined in a later section of this paper* the next paragraphs, however, will present a somewhat different point of view. It will be argued that it is at present impossible to derive a useful equation relating the composition of alveolar air - as it will be defined below - t* that of inspired air, and that the existing equations are not strictly valid. This argument, if correct, inevitably makes the choice of the inspired oxygen standard the only one that can be defended upon physiological grounds. b, The metabolic respiratory quotient can be calculated, as is well known, from the composition of the inspired and expired’ air; an equation exactly similar in form, but containing alveolar partial pressures instead of expired air partial pressure, has been derived in several different ways. These derivations have in common the underlying assumption, explicit or implicit, that the respiratory exchange can be expressed as the conversion of a certain volume of inspired air, by removal of oxygen and addition of carbon dioxide, into a certain new volume of alveolar air. The assumption is implicit in the derivation given by T. Benzinger {Erg, Physiol,, 40, 1, 1938) and explicit, notably, in that of J. S, Gray (School of Aviation Medicine, Randolph Field, Texas, Research Report No, 1, 12 April 1943), while F, Brink ("Calculations Relating to the Composition of alveolar Gas") substitutes alveolar values in his expired air equation without attempting to Justify this procedure. Within the limits of correctness of the above assumption, the equation is undoubtedly valid * Now it would seem probable that a certain fraction of what is collected as alveolar air, together with a residue that is never expelled even in the sharpest expiration, is not thoroughly mixed at each inspiration with fresh tidal air, but may be considered as a stagnant layer which acts as the vehicle, so to speak, of respiratory exchange, without actually itself undergoing periodic changes in compo- sition, This stagnant or partly stagnant layer in intimate contact with the pulmonary epithelium is in a steady state, in which it passes on to the pulmonary capillaries exactly as much oxygen as it receives from the mixture of inspired and alveolar air in contact with it, while transferring to the latter exactly the amount of carbon dioxide that it receives from the blood. Functioning as a buffer in such a way as to reduce the amplitude of fluctuations in the percentage saturation of pulmonary blood, the stagnant layer may be imagined to have the properties common to any system existing in a ture steady state or "dynamic equilibrium," The most significant and familiar of these properties from the point of view of the present discussion is the fact that the composition of the system in the steady state cannot be related in any simple manner to the rate at which matter or energy are passing 3 through it. Stated explicitly for the present case, it nay be said that the partial pressures of oxygen and carbon dioxide in the stagnant layer depend not only upon the rate at which oxygen is being removed and carbon dioxide produced, but also upon the various diffusion coefficients at the two boundaries of the layer. A complete calcu- lation of the composition of the stagnant layer would thus be difficult even for the over-simplified case here visualized; it becomes Immeasurably more difficult when we realize that the simple idea of a sharply defined stagnant layer is certain ly an abstraction, the real counterpart of which must be highly diffuse. The composition ‘ in the region nearest to the pulmonary epithelium will be regulated predominantly by diffusion, while as we move aoutward the effects of convection and mechanical mixing will become increasingly important. The effective thickness of the layer, if any single magnitude can be assigned to it under given conditions, will moveover be ex- pected to vary with the depth and rate of breathing in much the same manner as the dead space is known to depend upon these variables# Restating the thesis algebraically for a sharply defined stagnant layer, -we may write the number of molecules of oxygen transferred to the lung during a short tine as n(02) and the corresponding amount of carbon dioxide produced as 5 n(C02), In terns of the corresponding volumes S' v(02) and R v(C02) at the prevailing partial pressures. £ n(02) « p(02).s v(02)/RT £Nn(CC2) « p(C02). § v(C02) /RT The metabolic respiratory quotient is H «* o n( C02)/ £ n(02) - p(002). R v(C02) p(02 }. & v (02~) In the steady state the layer will maintain its constant composition by the uptake of b n(02) molecules of oxygen by diffusion from the urrounding alveolar air and loss of 8 n(C02) molecules of carbon dioxide. Without a knowledge of both the dynamics of these processes and of the magnitude of the respiratory exchanges R n, it is not possible to calculate the composition of the stagnant layer even in terms of that of the rest of the alveolar air. In view of the fact that the variable which really determines the oxygen tension of the pulmonary blood, and thus to a first approximation the physiological state of the individual, is the partial pressure of oxygen in the stagnant layer, the above conclusion is of some significance# It throws * doubt not only upon the valid- ity of calculations purporting to give the composition of alveolar air in terms of J that of inspired air; it also implies that even the measured partial pressures of oxygen in alveolar air nay not represent quite exactly the steady state values which actually determine the percentage arterial saturation. If the above arguments, which must be correct in principle, really apply to a significant fraction of the true alveolar air, why has the alveolar gas equation had such apparent success in the interpretation of experimental data? A two-fold answer may be given. In the first place, two circumstances - the relative constancy of alveolar carbon dioxide pressure and the condition that all the constituent partial pressures must 2 are, on account of its greater solubility, more favorable than for the absorption of oxygen when these parts are filled with pure air at each inspiration," (C, G. Douglas, J. S. Haldane, Y, Henderson and E, C, Schneider, Phil, Trans, R.S., 203B, 185, 1913; sf J, S, Haldane and J. G, Priestley, Respiration, New Haven, 1935, p, 16 ff,). The importance of the effects here described will not be amenable to final assessment until it becomes feasible to make exploratory measurements of leoal oxygen and carbon dioxide pressures in different parts of the respiratory passages. PART II. ALVEOLAR VERSUS INSPIRED OXYGEN PRESSURE AS A PRACTICAL REFERENCE POINT FOR THE CALCULATION OF EQUIVALENT ALTITUDES WHILE BREATHING GASES OTHER THAN AIR. It is improbable that any decision will be reached immediately concerning the justice of the above arguments, which attack the premiss rather than the manner of application of the alveolar air equation. We wish now to ignore these criticisms and to examine the arguments for the inspired air and alveolar air reference points solely on the basis of the opposing points of view expressed on page 2# It will be necessary to show how much difference there is between specifications based upon the two standards, and to find whether these differences have any practical significance that can be demonstrated unequivocally by existing experimental data. A simplified nomenclature may be appropriately introduced at this point. We shall consider that air breathed changes in composition in five stages indicated by the number of dashes 1 dry inspired air 1 moist inspired ( "tracheal•) air * meist alveolar air "* moist expired air *" dry expired air "W| Partial pressure of a component will always be p ; fraction of a component, f ; t*tal or barometric pressure, P.; respiratory quotient, Q. ; oxygen, 0 ; carban dioxide, C ; nitrogen, N . The partial pressure of saturated water vapor at body temperature will be given always as the numerical value 47. Thus: f* (0) = fraction of oxygen in dry inspired air. p" (0} = partial pressure of oxygen in moist inspired air, p".(C) = partial pressure of carbon dioxide in moist alvaelar air. 5 The alveolar air equation may be written as follows* p*»(o) = p"(o) - p#»(c)/a + f'Cohp-^cMx-.a)/* (1) whenever it may bo assumed that p'(C) is zero* With the equation in this form it is immediately apparent that the alveolar oxygon pressure is given by two terms. The first involves the inspired or "tracheal'’ oxygn pressure as such; the second involves the fraction of oxygen in the inspired air or, in other words, the fraction of nitrogen, Under certain circumstances the second term is zero* namely, if G is unity, or (to take a case which cannot be realized in practice) if f'{0) is zero and pure nitrogen is being inhaled. In this case, provided the alveolar oarben dioxide pressure is constant, the partial pressure of oxygen in the alveoli is strictly proportional to that in the inspired air, and the inspired air standard is Just as accurate as the alveolar air standard. Under all other circumstance the presence of the second term makes it clear that it is by no means "a matter of indifference whether oxygen lack is induced by decreased pressure or by breathing nitrcg Assuming an alveolar carbon dioxide pressure or 40 mm, Hg and a respiratory quotient of 0,8, the reduction of oxygen pressure in the alveoli produced by metabolic exchange would be 40 mm, Hg if pure exygen were breathed, and 50 mm, Hg in tha limiting case of inhalation of pure nitrogen," (T, Benzinger, oit,, p.4l), Whether this extreme difference of 10 mm, Hg in the alveolar oxygen pressure which results from removal of nitrogen has been demonstrated in practice can only be decided by an appeal to experi- mental data. The most extensive are those obtained by Boothby and his colleagues for subjects breathing pure oxygen at 35,000, 40,000 and 42,000 feet, and the alveolar pressures thus determined may be compared with those found in subjects breathing air at altitudes corresponding to the same inspired oxygen pressures as those prevailing during tha breathing of pure oxygen. The sensitiveness of the comparison is somewhat reduced by tha presence of nitrogen of unknown origin in the alveolar samples when r . oxygen was being inhaled; this represents, however, an uncertainty that is pr«bably inevitable with present mentods. In the data as here quoted, the inspired oxygen pressures have been calculated both with and without correction for the contaminating nitrogen, which has been assumed in the one case to have originated in a leak of air during inspiration and in the other in some unspecified manner, such as nitrogen elimination from the body. In Table 1, which is explained more fully in the legend, columns (4) and (8) represent directly comparable values of the alveolar oxygen pressure, at equal '•'tracheal" oxygen pressures, with and without the presence of nitrogen. It is evident that although the uncertainty in the values of p"(0) tends to obscure the effect, there is nevertheless a distinct difference in the sense predicted by the formula, hut somewhat smaller in magnitude. At the other end of the scale, where the effect of altitude is produced adding nitrogen to the inspired air, the decrease in alveolar oxygen due to added nitrogen should be smaller than in the case already considered. This is obvious from equation (l), since the effect produced is proportional to f*(0), and the range tf values from air to pure nitrogen is only one quarter of the range from air to pure oxygen. This is in agreement with the fact that Boothby*s mean curve for the change of p"'(0) for subjects breathing air at decreased pressure was found also to fit the corresponding set of values for subjects breathing air-nitrogen mixtures at ground level (1,000 ft.) Turning now to the practical implications of Table 1, wo see that if a regulator is so designed as to provide a gas mixture with a partial pressure of oxygen which remains constant as the altitude is increased, a certain amount of unnecessary 6 waste is incurred by reason of the fact that when the regulator is providing oxygen at such a rate, for example, as to give an inhaled mixture containing about 50 per cent oxygen, the alveolar oxygen pressure may be raised about 2,6 mm, Hg above its value whan air is breathed; if 75 per cent oxygen is inhaled, the increase would bo 4,9 mm,, and 7,1 mm, for pure oxygen. Thus it would seem reasonable to permit a certain gradual linear decrease of oxygen pressure in the inspired gas during the transition from air to pure oxygon. Exactly how much economy could be achieved in this way can best be seen by a calculation of the kind represented by Brink's chart A3. Rearranging equation (l) and substituting p*(0) = f»(0).{P -47) (2) we find f»(o) . p*,(o) + pw,(c)/a (p - p"‘(c) - 47) + p"*(c)/a (3) which enables us to calculate the fraction of oxygen in air which at any altitude P will correspond to a given value of the alveolar txygen p"‘(0), If the inspired gas xs prepared by mixing air with a fraction F*{0) of pure oxygen, then F»(0) = 1.26 f*(0) - 0.264 (4) On the constant "tracheal* oxygen criterion, on the other hand, the values jf f*{0) can be obtained from equation(2). The discrepancy between results based on the two criteria being greatest when pure oxygen is being breathed, it suffices to compare values of f’(0) or calculated for a value of P at which a very high proportion of oxygen is needed in order to keep the alveolar, or the "tracheal" oxygen pressures, as the case may be, at sea level values. Let P =• 200 ram, A convenient value for p*{0) is 100 mm. In Table 2 figures for f * (0) and F’(0)', calculated from equation (3), are given for the fallowing cases* p#,(C) =36 a = 0,82 p"’(C) =45 G = 0.82 p■»{C) « 36 G » 1,00 These are compared with the valves obtained from equation (2) for the case p"(0) = 142 mm. The greatest possible saving of oxygen as a result of using the alveolar reference point is represented by a difference of 0,905 - 0,062, or 0,043, or about 5 per cent, in the amounts of oxygen needed. Probable physiological variations among personnel tend to reduce this difference, so that it may be doubted whether anything is gained in this case by insisting on the alveolar criterion. The same argument holds if it be agreed to save oxygen by supplying only enough gas tc maintain the subject at, say, a 10,000 foat level. At altitudes ■where the sea level values cannot be maintained even by breathing pure oxygen, it is desirable to be able to estimate the probable effects of a given altitude by reference to the physiologically equivalent altitude when air is being breathed. Similar data are '‘Iso desirable for the effect of breathing various mixtures of air and oxygen at high altitudes,. In constructing charts of equivalent altitudes. It is of ooursa possible again to.use either the "tracheal" or the alveolar air reference point. Here, however, the matter is often compiler ted by the effects of anoxia, and in applying the alveolar air formula there arises the question of choice of appropriate values for CL, This will be illustrated here by some calculations for air and pure oxygen. All quantities referring to air breathing will be denoted by an asterisk and the formulas are the same as used by Brink for the same purpose* Combining equations (l) and (2) we may write for a subject breathing airi pw,(0)* * 0,209 (P* - 47) - pH'(C)*/G + 0,209 p"*(C)* (1-Ql)/CL (5) and for subject breathing oxygon* p"'(c) = P - 47 - p"»(C) (6) For equivalent altitudes p*‘(0)• » p*1(0) P-.(C). . p’*(c) (?) Combining (5) and (6) and retaining the symbol p"'(C)*, for which there are /xpe rimental values, we get F * 0.209 (P* - 47) - 0,791 p"»(C)* (l~a)/a + 47 (8) In applying equation (0, Brink used values of pWf(C)* taken from Boothbys data, together with the probable "true" or steady state respiratory quotient a, 0.82, In this way he obtained equivalent altitudes for a hypothetical steady state which in many oases can never be attained because of the accumulating effects of anoxia. T* this extent his results are dubious, and it might be considered preferable to use in equation (8) those non-stationary values of 0 which are in fact essential to the mutual consistency of the experimental data and equation (l). It is true that results obtained in this way should be treated with reserve, but this is no reason for retreat- ing to an entirely artificial position. Lutz and Schneider (B, R. Lutz and E. C, Schneider, Am. J. Physiol.; 50, 280, 1919), to be sure, succeeded in establishing the stationary values of Q after a 40 minute stay at 18,000 feet, but at this and still more at higher altitudes, the important information from a practical point of view would seem to be that defining the condition of the subject as it is established temporarily by the expedient of hyperventilation. Interesting remarks on this ques- tion are given in the paper of J, S, Gray (loc-cit). In Table 3 we give, for what they may be worth, values of the equivalent altitude calculated in three ways* (1) Brink’s method, using Boothby's p"'(C)* values and 9 = 0.82. Numerical values as follows (altitudes are in thousands of feet): Altitude 0 5 10 12 14 16 18 20 22 24 P 750.• 632,3 522.6 483.3 446.4 411.8 379.4 349.1 320,8 294,4 p-’(C). 40 38 35,5 35 34 33 32 30 28 25 (2) Applying equation (8) to experimental values of 0, the apparent respiratory quotient in a non-stationary sta-te. Numerical values as folxowr calcu- lated, from Booth/.y’s smoothed curves for p**(0)* and P**(C)*, 8 p 700 650 600 550 500 450 400 350 300 p” Co) - 94,2 83.5 73,3 63,6 54,6 46.2 39,2 34,0 30*5 P"’(c)* 36,7 36,6 36,2 35,8 35,2 34,2 32.6 30,0 26,0 a 0,839 0.830 0,825 0,832 0,851 0,876 0,929 1,029 1,214 (3) Equal inspired oxygen pressures* P « 0*209 (P*_47)/f»(0) + 47 To these are added* (4) Experimental equivalents for air-breathing and oxygen-breathing, taken from Boothby’s data on a basis of equal observed alveolar oxygen pressures. It will be observed from Table 3 and from Fig. 1, where the data are plotted, that the equal inspired oxygen figures provide a somewhat more .cautious estimate of the equivalent altitudes than do Brink's data. This is merely the result of the nitrogen effect already discussed. It amounts to a possible overestimate of the aquivalent altitudes by about 1,000 feet at 33,000 feet, increasing to nearly 2*000 feet at 45,000 feet. On the other hand, the calculation using non-stationary-atate values of '1 leads to a curve which approaches and even crosses the Inspired oxygen curve in the region of anoxia. The experimental figures tend to follow the equal alveolar oxygen curves, but they are subject to the same uncertainty as that pointed out in Table 1, The suggestion seems -warranted that charts of equivalent altitudes should follow curves (l) and (2) in the lower altitude range and should at higher altitudes deviate increasingly from (l) in the sense indicated by curve (2), with perhaps an increased thickness of the line to indicate the added ..pos sibilities of error in any data given for anoxic subjects (Fig* 2), Such a curve for 100 per cent oxygen *ill be nearly identical with that given by Gray (l,o,). It is scarcely possible to pursue this discussion further without detailed knowledge of the amount of practical importance which attaches to accurate statement of equivalent altitudes. It is clear that present knowledge is sufficient for the construction of oxygen supply and equivalent altitude charts which do not dangerously distort the physiological picture. Certain ill-controlled variables, notably leakage of air into the inspired gas, have been disregarded, and must be taken into account before the specifications are finally turned over to the engineers. We are not competent to express any opinion on this point. Concerning the argument over reference points, we contend that the foregoing discussion shows it to have been largely superfluous. Wo are not scholastics, com- mitted irrevocably to one view or the other. Our business is only to provide speci- fications for engineers who are quite unconcerned about the manner in which they were drawn up. If it is convenient at one point to work on the basis of inspired air and at another to make calculations involving alveolar air, this double procedure should be judged solely by the practical effectiveness of the resulting statement of physiological requirements. Tab la 1. Breathing Oxygen Breathing Air Effect of Nitrogen Found Predi ote d U) (zf (3) (4) Mean P*(0) P - (0) h p**(0) ~W) Ts) (7) (B) P"(0) h P"'(0) p"'(C) extremes corres- ponding to mean in Col. (1) (9) (10) 131.7 128,7 35,000 88,7 125.7 131.7 3208 88,7 85,7 125.7 4373 82.8 85,7 3,0 7.1 1 i ; i ’ 93.8 91,9 40,000 56.0 ! 90.0 j 93.8 11394 53,6 52.1 90,0 12341 50.5 4.9 < 7.1 1 ! 80.9 79,0 42,000 45.7 77,0 80.9 14740 43,7 42.3 77.0 15806 41,0 3.4 The data for breathing oxygen give experimental average values of alveolar oxygen pressures at the specified altitudes. The "tracheal* oxygen pressures in the first column are presented in pairs which represent the uncertainty caused by the presence of measurable amounts of nitrogen in the alveolar air. In obtaining the figures for subjects breathing air use was made of the smoothed curves of alveolar oxygen pressures given in Boothby's charts of 1313 observptions. The values of p"f(0) read from this curve correspond to the same p"(0) values that prevailed in the experiments breathing oxygen. Table 2, P =* 200 nun, Hg throughout. Altitude = 32,610 feet p"'(C) a £'(0) r*{ o) Constant alveolar oxygen. 36 0,82 0.894 0,862 p * ' (O) = 100 tnn. 45 0,82 0.957 0.942 36 1.00 0.882 0.082 Constant inspired and "tracheal" — 0,928 0.905 oxygen, p"(o) = 142 tnnu Table 3, Equivalent A1 broathi ng i h* ti tuda > air P* Equal Oxygen (i) Stationary State * Alveolar Pre s sure s (2) Non-station ary State h Equal Inspired Oxygen Pressures (3) h r Experime ntal Values (4) h P 0 760,0 33,8 189,1 33.7 190,1 33,0 196.0 •MB 3.25 674,8 35,8 172,0 35,7 172.8 35,0 170,7 T35.0 (35,7 10.7 508,6 40,4 138,1 40,6 136.8 39,6 143,5 [A0,0 \40.6 14,0 446,4 42.6 124.6 42.2 126.7 41,6 130,5 (A2,Q l 142.7 j H CD • O 379,4 45,0 110,9 44.2 115.2 44.0 116.5 - i i f\) . o • o 349.1 46.2 104.9 45.1 110.3 45.1 110.1 22.0 320.8 47.3 99,4 45.8 106,7 46,3 104,2 - 24,0 294,4 48.4 94,4 46,3 104,2 47.4 98.7 , . —,i - 12 LEGEM* TO FIGURE I. Effective or equivalent altitudes (ordinate) of subjects breathing pure oxygen at various actual altitudes (abscissa)# Curve Ij calculated on a basis of equal partial pressures of oxygen in the inspired or "tracheal" air. Curve;2i calculated from the alveolar air equation (in the form of equation 8), using experimental values of the alveolar carbon dioxide pressure and an assumed "steady state" respiratory quotient, Q, of 0,82, Curve 3t calculated from the alveolar air equation (equation 8), using experimental values of the alveolar carbon dioxide pressure and the apparent or "non-stationary" values of the respiratory quotient calculated from Boothby*s data using equation 1, LEGEND TO FIGURE 2, Equivalent altitude chart suggested by the curves given in Figure 1 (see text), FIGURE I c.wuii Ai_ i i i uul f'(0) = 1.0 J.B.BATEMAN DECEMBER 1943 HART X- 8 a ALTITUDE AYO AERO MEDICAL UNIT FIGURE 2 f'(0) = 1.0 ALTITUDE CHART 1-8 MAYO AERO MEDICAL UNIT J.B. BATEMAN DECEMBER 1943 NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE OPEN Special Report No.5 Report No* 3U0 August 19h2 OXXGEN AND AIR PRESSURE AT VARIOUS ALTITUDES AS THE! INFLUENCE THE EFFICIENT FUNCTIONING OF THE AVIATOR.* By Walter M* Boothby, from the Mayo Aero Medical Unit, Rochester, Minnesota. Note: This paper was written and submitted in its present form to the Committee on Medical Research, Office of Scientific Research and Develop- ment, for publication as a C.M.R* report on August 28, 19U2. Through oversight the paper became ’’lost*1 and is being issued now, October 19UU, but as of the original date and without change although subsequent studies by Wildhack, Brink, Gray, Bateman* Fenn and others have perfected the "alveolar air formula1’ and elucidated its applicability as well as its limitations. The air formulation," however, remains the point of reference for calculating specifications for designing oxygen equipment for aviation. ABSTRACT Change of altitude will affect oxygen and air pressure in relation to body needs through various factors, some physical and some physiologic. In the following sections an attempt has been made to trace separately the effect of these factors one at a time successively. In fact, however, they will act simul- taneously to produce a very predictable end result* This is especially true as will be seen in the examples given in Part II in which we trace the effect on the atmospheric air of known composition as it is inhaled and comes into final equi- librium in the pulmonary alveoli with the gases of the arterial blood. In Part I the presentation will be limited mainly to the effect of water vapor* This analysis has been carried out in considerable detail t# help those investigators, many of them engineers new to the subject, who now find themselves assisting in the development, testing or comparing, of the various types of oxygen equipment to be used by aviators. -* We are indebted to Major Joseph Berkson, (MC) of the Air Surgeon’s Office for assistance in the elucidation of this problem. Before going on active duty Major Berkson helped in the calculations and in the construction of the charts so that the physiological factors and physical factors involved could be presented diagraim- matically as an aid to engineers. The calculations have also been checked by E, J, Baldes, Ph.D. RESTRICTED MAYO AERO MEDICAL UNIT SPECIAL REPORT NO, 5 CONTRACT NO, OEMomr-129 w535 ac-25829 DATE 28 August 1942 OXYGEN AND AIR PRESSURE AT VARIOUS ALTITUDES AS THEY INFLUENCE THE EFFICIENT FUNCTIONING OF THE AVIATOR* ’’Tracheal1’ air T (B—47) f0ol ' a' Walter M, Boothby, M, D,, Chairman Fart I, Effect of Water Vapor Change of altitude will affect oxygen and air pressure in relation to body needs through various factors, some physical and some physiologic. In the following sections an attempt has been made to trace separately the effect of these factors one at a time successively. In fact, however, they will act simultaneously to produce a very predictable end result. This is especially true as will be seen in the exampZ.e given in Part II in which we trace the effect on the atmospheric air of known composition as it is inhaled and comes into final equilibrium in the pulmonary alveoli with the gases of the arterial blood. In Part I the presentation will be limited mainly to the effect of water vapor. This analysis has been carried out in considerable detail to help those investigators, many of them engineers new to the subject, who now find themselves assisting in the development, testing or comparing, of the various types of oxygen equipment to be used by aviators. 1, Effect of altitude on total atmospheric pressure. Basically the atmospheric pressure is the -weight, per unit area, of the column of air above the position at -which the pressure is measured. Therefore, if the density of the air were constant, the pressure would decrease uniformly with increase of altitude. But, because of a number of factors, the density is not constant. The weight of the air exerts a pressure on the air below it and the density at any point therefore depends on the pressure. Mathematically it would follow from this effect alone that the decrease of pressure with increase of altitude would be logarithmic. But there are other factors, chief among them being that the temperature is lower at the higher altitudes, owing In part to the cooling of air when it expands as it rises to a position of lower pressure. The cooling at higher altitudes in turn has the effect of increasing the density and accordingly increasing the pressure at any point below, .an equation has been set up for the altitude in terms of the measured air temperature and the air pressures. The decrease of tempera- ture with increase of altitude on the average is fairly uniform (2« C, per 1,000 feet). • We are indebted to Major Joseph Berkson, (MG) of the Air Surgeon's Office for assistance in the elucidation of this problem. Before going on active duty Major Berkson helped in the calculations and in the construction of the charts ss that the physiological factors and physical factors involved could be presented diagrammaticall as an aid to engineers. The calculations have also been checked by J*. Baldes, .'hr,Iu and is used to set up an empirical linear equation of the temperature at various altitude By combining this equation of the cooling effect and the basic logarithmic relation a standard atmosphere for reference is defied in -which the pressure at any altitude can be computed. At atmosphere assumed to be/15° C. and a pressure of 760 mm, (mercury barometer) at sea level and containing no moisture has been accepted as the U. S, Standard Atmosphere for aeronautic purposes. For this standard atmosphere the pressure at various altitudes has been calculated and tabled,* This data is charted in curve 1 of figure 1, In this country actual altitude accords fairly well with the atmospheric pressure values so calculated except for extreme seasonal changes which are not to be considered here. 2, Oxygen pressure in the air of the atmosphere. The atmospheric air is composed largely of oxygen (20,93 per cent) and nitrogen (79.04 per cent) with a small amount of carbon dioxide (0,03 per cent). It also contains minute quantities of other rare gases which are, like nitrogen, inert and can therefore, for present purposes be treated as a part of the nitrogen fraction. The percentage of all these gases is very closely constant at all altitudes so far studied. Measured relatively in terms of the volume that would be occupied by each J 73 . J gas, the fractions of oxygen and nitrogen in dry air are close to 0,2903 (20,93 per cent) and 0,7904 (79,04 per cent) respectively, and the fraction of carbon dioxide is 0,0003 (0,03 per cent) in the mixture of gases. The pressure exerted by each constituent gas is proportional to its percentage volume. Thus, if the total pressure of atmospheric air is B, the partial pressure of oxygen is 0,2093 x B and the partial pressure of oxygen in the dry atmospheric air is 20,93 per cent of the total atmospheric pressure of dry air at that elevation. The volumetric fraction of oxygen in dry atmospheric air usually used by physiologists is either 20,93 per cent or 20,94 per cent although the latest values given by Humphreys quoting from Hann and Suring and from Paneth is 20,95 per cent (W. J, Humphreys, Physios of the Air, McGraw-Hill Book C#., Inc., New York and London, 1940, P, 68 and 8l), Fer most practical purposes the value for oxygen can we rounded off to 21 per cent. Physiologists always use the percent of oxygen in dry air as the basis of their calculation because ? definite allowance for the pressure of water vapor in the lungs and body cavities must always be made. This allowance is very easy to compute as will be described later as it always corresponds to the body temperature with which the gas is in equilibrium; this body temperature is usually assumed to be 37* C, and at this temperature the pressure of water vapor in saturated air is 47 mm. It is possible that the vapor pressure of body fluids, especially on curved surfaces, departs somewhat from that of pure distilled water. Also the temperature will vary considerably under different conditions. However, for physiologic purposes the value of 47 mm, has been accepted as a fair average value. The value for oxygen of 20,75 per cent frequently used by aeronautical engineers is a percentage value based upon air assumed to contain an average amount of water v?por at 15° Cj Humphrey, quoting from Hann and Suring, states that this average value for oxygen will vary depending on the amount of moisture present, from 20,44 per cent at the equator to 20.94 per cent at 70° N. * National Advisory Committee for Aeronautics, Report No, 538; Altitude-Pressure Tables, W, B, Brombacher, 1935* 3 In the next Section (3) ‘the method of making allowances for water vapor is described in detail when the volumetric fractions of dry atmospheric air are taken as 20*93 per cent for oxygen, 79,04 per cent for nitrogen (including other inert rare gases) and 0,03 per cent for carbon dioxide. The above values are used here because they are those obtained by carefully calibrated volumetric gas analyzers like the Haldane which gives the results in the terms of the dry gases; however, a small amount of liquid water must be present so the gas being analyzed is completely saturated with water vapor at all buret readings in order to give an accurate per cent on the basis of dry air by the cancellation out of the water vapor when the volume contracts as the result of the absorption of the ideal gases. Curve II of figure 1 gives the partial pressure of oxygen in the atmosphere calculated as 0.2093 times the values given by curve I which is the total standard atmospheric pressure for various altitudes. In figures 1, 2 and 3 the altitude is scaled on the abscissa in feet because we are interested there in the physical factor of atmospheric pressure in relation to altitude. In other places where we are interested in graphic representations of physiology figures as in figure 4, the altitude will be given in its equivalent pressure because the physiologic factors are related to pressure expressed directly and not to elevation as such, as the latter is a modified logarithmic function of density. 3. Effect of saturation of air or gas mixture with water vapor. In Sections 1 and 2 we have considered the air as though it were dry, that is, free of water vapor. As a matter of fact, owing to the ubiquitous presence of water, the atmosoherio air contains water vapor in various amounts. Water vapor evaporates from liquid water into the atmosphere surrounding it until the pressure of the vapor reaches a certain value, determined by its temperature, at which point the atmosphere is said to be saturated. At 37° C., the average temperature of the lody, the pressure of water vapor in a saturated atmosphere is 47 mm. The presence of water vapor in the air will therefore change the partial pressures of the ideal ,.ases in the air, namely, oxygen and nitrogen (the latter including the rare gases). The atmosphere gases, oxygen, nitrogen and carbon dioxide within negligible Jimita are "ideal" or "true" gases and follow very closely the three gas laws. (l) Beyle’s Law; at any stated temperature a given mass of gas varies in volume inversely as the pressure, (2) Charles* Laws the volume of a gas at a constant pressure varies directly with the absolute temperature, (3) Avogadro’s Laws equal volumes of gases with the same temperature and pressure contain an equal number of molecules* Water vapor is not considered as a "true" or "actual" gas because under the temperature conditions being discussed it does not follow the three laws of ideal or true gases. The pressure of saturated water vapor in the presence of liquid water is dependent only on temperature and, unlike true gases at this temperature, entirely independent of the size of the available space. Water vapor is not compressible with increase in pressure as the size of the space in which it is present is decreased as happens with true gases. Water vapor as such no longer exists as the size of the decreases because it condenses back into liquid water which relatively speaking occupies no space. Example 1. Consider a liter of dry air at 37° C, and suppose there to be a diaphragm on the top of it that is freely movable with expansion of the air. Introduce some water at 37° C. Water vapor will arise from the liquid water into the space occupied by the gases, exert an additional pressure’; on the diaphragm, and the total volume will increase as illustrated on the right-hand side of figure 2„ As the volume increases, the pressure exerted by nitrogen and oxygen which together in the first place exerted a pressure equal to the atmospheric pressure, will decrease according to Boyle's law, that is. the partial pressure of these two gases will decrease in inverse proportion to the increased volume. The expansion of the gases will continue until the total pressure exerted by the water vapor and the nitrogen and oxygen togethei will be equal to the atmospheric pressure B. That is, when equilibrium is reached the pressure of water vapor (47 mm. at 37° C,) plus the pressure exerted by the oxygen and nitrogen will be equal to the atmospheric pressure. Therefore, at equilibrium the pressure of oxygen and nitrogen together but without the water vapor will have decreased by 47 mm. Since according to Boyle's law the volume will he inversely proportional to the pressure, the volume occupied by the oxygen and nitrogen after saturation will be to the volume before saturation as B is to (B -47). The effect of saturating dry air at 37° C, and at itmospherio pressure B. therefore, is to increase the volume in the proportion B/(B - 47) and to decrease the partial pressure of each constituent gas in the proportion (B-47)/B, Curve III of figure 2 gives for various altitudes the value of this propor- tionality factor by which the partial pressure of all the gases or of each const! tuor4? gas in dry atmospheric air at 37° C, is decreased when the previously dry air, after contact with liquid water, becomes saturated with water vapor* According to what has been outlined, if in figure 2 we multiply the partial pressure of oxygen in the atmospheric air dry given in Curve II by the factors given :.n Curve III, we obtain the partial pressure of oxygen in the atmospheric air satura- te I with water vapor at 37° C, as shown in Curve IV, It is convenient to remember that this factor makes the oxygon pressure in Curve IV of atmospheric air saturated r,t 37° C, wi th water vapor 10 mm. less (9,8 mm, to be more exact) than the corres- ponding value of Curve II which represents the oxygen pressure in dry atmospheric air. Inspired air before gaseous exchange takes place but after it has reached the body temperature of 37° C, and become completely saturated with water vapor as it passes -;hrough the warm moist nasal, tracheal and bronchial passages, may be conveniently designated by the arbitrary term "tracheal air.* Another common way of obtaining the water vapor correction for the pressure cf oxygen in tracheal air (air saturated with water vapor at 37° C.) is as follows} PT02 * (B - 47) 0.209 Where * pressure of oxygen in tracheal air. A point which at first sight is confusing is that the varying amount of water vapor or humidity present in the atmospheric air before inhalation can be omitted from the calculation} this can be done because under ordinary conditions the temperature and therefore the water vapor is less than that in the alveolar air. The reason in more detail is that the body is always dealing with the air in the lungs or other body cavities completely saturated with water vapor at 37° C, (body temperature;: it makes no difference, therefore, whether the air before entering the body is oomp-'-etely dry, partially saturated or completely saturated as long as the temperature of tue 5 gas is belaw or equal to body temperature. The final partial pressure of water vapor in the body with which the blood is in essential equilibrium is 47 mm,* which is that corresponding to the average body temperature of 37° C, As mentioned before, volumetric methods of gas analysis like the Haldane and other similar methods express the percentage of carbon dioxide, oxygen and nitrogen in terms of dry air because the water vapor in each fraction of gas absorbed by the reagent is condensed out although the proportionality is the same if both be considered saturated at the same temperature. The easiest way to allow for water vapor in most physiological problems is to consider the proportion of true gases present in a mixture, the volume of which has been reduced to the dry basis, and to allow for the presence of water vapor by deducting the partial pressure of the water vapor from the barometric or total pressure of the gases with the water vapor present. However, whenever measur- ing the volume of gases, care must be taken to make the physical conditions such that at least a small amount of liquid water is present in the container holding the gases which are being measured so that the entire gas mixture is completely saturated with water vapor. The temperature at which the gases are measured as well as the barometric pressure must be accurately known, otherwise considerable error may be caused. Compar- ison of gas volume must always be made at the same temperature and barometric pressure either absolutely dry or completely saturated. Example 2, Consider a liter of dry air at 37° C, in a confined space with an immovable diaphragm corresponding to a closed body cavity instead of a movable diaphragm corresponding to the alveolar spaces just discussed. Introduce some water at 37° C, Water vapor will arise from liquid water into the space occupied by the gases and exert an additional pressure on the diaphragm inside equal to 47 mm., the vapor pressure of water. The volume has been unable to increase, therefore, the total pressure (P-fc) has increased 47 mm, and therefore! B + 47 = Obviously the partial pressure of oxygen and of nitrogen is unchanged because these gases on the dry basis occupy the same volume as before and therefore their respective partial pressures are unchanged, = ,209 + 47) - = ,209 (P-fc - 47) = ,209 B Although the oxygen partial pressure (and similarly the nitrogen) in this instance was unchanged, the total pressure was changed by the introduction of water into dry air in a confined space. At 37c C, this resulted in an increased pressure in the ratio of £JL which is the same as ILl—lZ, Now if we remove the extra pressure B B of 47 on the diaphragm and allow the gases to expand until equilibrium with B is reached to occupy an enlarged volume in the proportion B/(B - 47), then our example in the previous section returns and Pq£ =» ,209 (B — 47) because now the same amount of oxygen occupies a greater space and therefore the partial pressure of the oxygen is correspondingly decreased. Example 3. Consider a closed gas cavity like the stomach, intestines or pneumothorax and assume for simplicity that this cavity is freely expansible the diaphragm in the first example), At body temperature of 37° C, any air in this cavity will be saturated with water vapor because the walls of the cavity are moist with body fluids. Therefore, the pressure of the gases, Pq, nitrogen, oxygen and • N-) te - The value 47 mm* is generally accepted by physiologists as representing the vapor pressure of tissue and body fluids. Under certain conditions the true value may be lower than this by 2 or 3 mm.; however, this problem, while needing investiga- tion, is ignored here; in any case the order of magnitude of the error is probably insignificant when calculating for average conditions. 6 carbon dioxide at sea level will be equivalent to the barometric pressure less that due to the 47 mm, of water vapor, therefore at sea leveli PG = B1 - 47 = 760 - 47 If we now ascend to 40,000 feet where the barometric pressure is 141 mm., the pressure of the gases similarly will be: Pg2 = B2 - 47 = 141 - 47, The expansion of the gases in this cavity will be increased therefore in the ratio of B-. - 47 760 - 47 713 ■L i 7 6 B2 - 47 141 - 47 94 “ and not simply in accordance with the ratio of -.54 B2 141 However, in a pneumothorax there is not as a rule, room for free expansion of the gases by collapse of the lung. More often there are adhesions so that the gas cannot freely dilate the cavity and as a result the pressure within the pneumo- thorax increases. This increase in pressure will be inversely proportional to the freedom with which the cavity expands, therefore, in most instances it is the increase in pressure accompanying the increase in volume that produces the harmful clinical effects. Example 4, An even more complicated condition exists, as pointed out by Bohnke and by Dill, in the growth of air bubbles in tissue fluids on rapid ascent to high altitudes. Air bubbles as shown by Boothby and Yfalsh first become visible to the eye at or around 12,000 feet (483 mm,) in the spinal fluid in the outer limb of a manometer connected to a needle which is introduced into the spinal canal of a human subject. Therefore, it can be presumed that bubbles begin to form in the fluid in the spinal canal and in other body fluids at about the same barometric pressure. On formation the bubble is composed not only of nitrogen but will also be in equilibrium with the partial pressure of carbon dioxide, oxygen and water vapor in the tissues. The partial pressure of oxygen and carbon dioxide are maintained essentially constant in the tissues as is the pressure of water vapor. At = 483 mm. (12,000 feet) let us assume with the subject breathing oxygen the tissue Pco2 ** 50, the Pq2 = 25 and the Ph20 * ln consequence, the pN2 = - (50 + 25 + 47) = - 122 = 483 - 122 = 361, On proceeding to higher elevations this nitrogen part of the gas bubble will enlarge as the barometric pressure decreases; and into this expanding bubble not only water vapor but also carbon dioxide and oxygen of the body fluids will diffuse because unlike the nitrogen in the tissues, the pressure of carbon dioxide and oxygen remain essentially constant with the subject breathing oxygen as the altitude increases up to 35,000 feet, B2 = 179 mm. Therefore carbon dioxide and oxygen will, like water vapor, diffuse into the air bubble and maintain approximate equilibrium. The total volume of the gas bubble at 35,000 feet may increase therefore after formation at 12,000 feet in the ratio of the respective barometric pressure less approximately 122 mm, as follows: Bx - (50 + 25 + 47) Bx - 122 483 - 122 361 B2 - (50 + 25 + 4?) = B2 - 122 = 179 - 122 " ~57 “ That is, a bubble which "has a volume sufficient to become visible at 12,000 feet will without the migration into it of any more nitrogen molecules be 6#3 times larger when the aviator reaches 35,000 feet; at 40,000 feet with a barometer of 141 mm# the bubble will hare made a total increase in size after its formation corresponding to the ratio 3fl . m In other words, going an additional 5,000 feet from 3 ? • -U- 141 - l22 35,000 ft# to 40,000 ft# triples the volume of the air bubble from the diffusion into it of carbon dioxide, oxygen and water vapor# As mentioned, the bubble first became vifible at 12,000 ft. and the carbon dioxide and oxygen, especially the carbon di- oxide, probably plays a large part in the birth of the bubble and will enter into the probability calculation of Piccard who for simplicity limited his exposition to nitrogen; this is likely because of the relatively large number of carbon dioxide molecules, loosely combined with base, present in tissue fluids as compared with nitrogen molecules free in solution equilibrium# However, as the subject is breathing oxygen and in consequence the nitrogen in the blood stream will be quickly washed out, there will be a progressive decrease of the nitrogen not only in the tissues but possibly also in the bubble* At the present time there is no means of determining the rate at which nitrogen will dif- fuse in or out of the bubble after it is once formed and therefore the actual in- crease in size of the bubble may be slightly different from that given in the example by an unknown amount as the result of the Influence of unknown factors. Example 5# The measurement of the vital capacity is another example which is often erroneously expressed# The vital capacity is actually the change in volume from the expansion of the chest by the greatest possible inhalation to the contrac- tion of the chest by the greatest possible expiration# This volume to represent accurately the movements of the chest must be expressed, therefore in the gas volume corresponding to the body temperature and to the existing barometric pressure with the gas saturated with water vapor at 37° C« The vital capacity has, however, frequently been erroneously expressed variously either at standard temperature and pressure dry or at the volume as measured either by a wet or dry spirometer at laboratory temperature* Numerically the volume should be expressed in liters to tfcree significant figures (two decimal points) for example 4*83 liters and not 4830 cubic centimeters as the latter indicates an accuracy of four significant figures which is impossible to obtain* To prevent errors of considerable magnitude, especially at high altitudes, it is necessary to know accurately the temperature and barometric pressure at which the vital capacity is measured# Also it is necessary to have the gases completely saturated with water vapor at the measuring temperature, therefore, the measurement of the volume must be over water by means of a wet meter or water float gasometer* a dry meter or dry bellows type of spirometer is used it is most difficult to determine the degree to which the gases are saturated as obviously they cannot be considered completely saturated unless liquid water is present#* As it is usually desirable to have both the S.T.P.3. value and the ambient alveolar value, the observed volumes should first be reduced to 0° C*, 760 mm* barometric pressure and dry; this S«.T.P*D. volume is then brought up to the volume, which the gases would occupy in the lungs at 27° G# and at the ambient barometric • The CO2 should not he absorbed# If It is absorbed an error of about —2 % is pro- duced at ground level -which increases to about - 21 % at 40,000 feet. pressure less 47 mm, for water vapor. The calculation* should be carried out in two stages as there is less likelihood of careless errors creeping ini (1) v . V„ , !I1_ , ?p.r.pwo, stpa o 273 + To 760 ( ?\ V. —v t 273 + 37 „ 760 V £) VL “ vs tpd X — x r 7T y 273 B0 - 47 If desired, however, the two equations can be combined so as to omit the intermediate calculation of the S.T.P.D. value as follows* (3) V, = V0 x x Bo ~ ?7ro 273 + 37 >6^ 273 + T0 '‘27-3 B0 - 47 (4) VL y0 x ilLli! x h-Z-Ill 273 + T0 B0 - 47 V0 = the volume as observed in water gasometer at TQ T0 = the temperature of the gas in the water gasometer at time of reading VQ B0 = the barometric pressure at time of reading VQ Vstpd ~ v°iutne V0 after correcting it to 760 mm, 0° C, dry Vjj = the volume V taken to ambient barometer, B0, 37° C, and saturated and therefore represents the true expansion of the lungs from complete expiration to complete inspiration, that is, the vital capacity, 47 = the value accepted by physiologists as representing the vapor pressure of body fluids at bedy temperature. Its true value may vary slightly from this under various body conditions. Eokman and Baraoh (Jour, Aviation Mod. 13:37, 1942, March) call attention to the importance of making allowance for water vapor, temperature and barometric pressure in calculating the vital capacity of the lungs. Unfortunately, they are in error in . the last factor of their equation when they subtract the water vapor corresponding to T0 from the vaper pressure of water corresponding to 37° C, Their formula VL = vo x — x 1— r is not correct and causes an error of nearly 273 + T0 B0 - (47 - P.J 3 per cent at 40,000 feet. Example 6, Addition of-nitrogen to the air in a room, chamber or gasometer is often used to simulate altitude instead of low pressure chambers. Sometimes results hare been obtained when using nitrogen that seam to be different from those obtained in the low pressure chamber. Usually, ..however, the simulated altitude is improperly calculated due usually to failure to handle the water vapor correctly. The following illustrates in three stages how the simulated altitude should be calculated. All pressures are expressed in millimeters of mercury. The term "tracheal air" is used arbitrarily to indicate atmospheric air saturated with moisture at body temperature which is the actual condition of the air as it enters the alveoli before any exchange with Mood gases has occurred. This is, of course, an arbitrary division because gas exchange proceeds more or less simultaneously with saturation. The word "trachea" does not have an anatomical limitation but, as mentioned above, is used arbt tr arilyo • A convenient sat of factors for making these calculations is given in Table I. A, The partial pressure of oxygen in the tracheal air at any altitude is obtained from the equation* (PIo2)a ■ - 47) 1 0-2093 Wheret (PTo2)a = partial pres sure of oxygen in the traohe al air at any altitude♦ Ba = total barometri o pressure at the altitu de. 47 = water vapor pressure of saturated air at 37° C, 0,209 * volumetric fraction of oxygen in atmospheric air (dry), B, The partial pressure of oxygen in the tracheal air when using nitrogen to simulate altitude is obtained from the equations (PT0Z)g ' (»g - 47> * *02 = partial pressure of oxygen in tracheal air obtained at ground level by simulating altitude by addition of nitrogen* Bg = total baro metric pressure at ground level, 47 » water vapor pressure of saturated air at 37° C. f(>2 = volumetric fraction of oxygen in the chamber air (dry) after nitrogen has been added. Wherei C, In order to compare the results obtained between an altitude simulated by nitrogen with those actually obtained by altitude or by utilizing a negative pressure chamber, the two expressions may be equated and then solved for Ba which would be the actual barometric pressure for an altitude corresponding to the nitrogen added* Equating the tiro equations! (Ba - 47) x 0,209 = (Bg - 47) x fo2 Solving for Ba B + 47 or B = (B* - 47) — P-i + 47 a 0.209 a ' g 1 0.209 It is to be noted specifically that this method in both instances deals properly and simply with the partial pressure of water vapor which is constant at 47 mm, of Hg in the lungs under all conditions. From the barometric pressure thus obtained one looks up in the "Altitude- Pressure Tables Based on the United States Standard Atmosphere" the corresponding altitude in feet. CHANGE OF ALVEOLAR OXYGEN PRESSURE WITH ALTITUDE - 1 XII-4 SATURATED V B-47 X V PG= x PG III - FACTOR FOR SATURATION DRY CHANGE OF ALVEOLAR OXYGEN PRESSURE WITH ALTITUDE - 2 XII—5 Ill-PER CENT O2 IV, V-LITERS OF O2 ADDED LITERS OF OXYGEN ADDED PER LITER OF VENTILATION (TRACHEAL AIR) XII-6 I, II - PRESSURE - MM. HG. Part II. The Role Played by the Combustion or Respiratory Quotient, Hyperventilation and Diffusion of Gases in the Final Gaseous Squillbrium in the Pulmonary Alveoli Resulting in the Alveolar Ratio If we know the partial pressure of the carbon dioxide in the alveoli, we may calculate at least approximately the uncompensated partial pressure of the oxygen in the alveoli for any altitude if the person is breathing air provided we make certain assumptions in regard to the alveolar CO2 and the respiratory quotient, even if we neglect certain other rather important factors. Wq shall suppose first that the "respiratory quotient" is unity. The oxygen which is removed from the alveoli into the bl*od is used eventually by various working tissues for combustion, that is oxidation, to supply the necessary energy for these working tissues. The oxygen combines with substances that function as fuel for the body, and there is produced as a result carbon dioxide by combination of the oxygen with the carbon of these substances. The "respirat ory quotient" (R.Gl.) is the ratio of the volume of carbon dioxide produced to the volume of oxygen consumed. If the substance which is burned as fuel is carbohydrate, there will be exactly as much carbon dioxide produced in volume as there is oxygen consumed because carbohydrate is composed of a certain number of carbon atoms each combined with two attftns of hydrogen and one of oxygen* During combustion the two atoms of hydrogen combine with the one of oxygen to form water, and there is left over one atom of carbon which combines with the one molecule, composed of two atoms, of oxygen obtained from respiration, to form one molecule of carbon dioxide. Thus, for every molecule, O2, of oxygen utilized, one molecule, of carbon dioxide is produced. Since in a gas equal numbers of molecules occupy the same volume at any particular temperature and pressure, the volume of carbon dioxide produced is the same as the volume of oxygen consumed, and therefore, when carbohydrate (sugar, starch) is the substance which is utilized for combustion the respiratory quotient approximates unity. Hence the alveolar air after exchange will have had added to it as much carbon dioxide as oxygen has been removed. Since there has been no change in the total volume, this means that the partial pressure of the oxygen will be decreased during exchange by the same amount as the pressure of the carbon dioxide is increased. Knowing this we may calculate easily what will be the uncompensated alveolar oxygen pressure at any altitude under these conditions. We have seen at an altitude at which the atmospheric pressure is B, the partial pressure of the oxygen, Pg^f in the tracheal air is (B - 47) 0.2093, During exchange, carbon dioxide will accumulate in the alveoli to a pressure of say 40 mm, anrWnp!^^e prreeas'IS?ie &/?3^Ue!sH0nn til amV'VxAW®*' tension p £^-ven under these circumstances when the respiratory quotient is unity, if certain other factors are neglected, as follows* P’o2 - rc2 - P'co2 = LLb - 47> n-2C93J - 4« The respiratory quotient will be less than unity if protein (R.O, - 0,82) or fat (R.ft* » 0,71) rather than carbohydrate is utilized for combustion. The hydrogen and oxygen contained in these substances exist in a proportion of more than two atoms of hydrogen to one of oxygen; that is, there is an excess of hydrogen atoms compared with water {H2O), During combustion some of the oxygen that Is obtained from respiration is utilized to combine with those extra atoms of hydrogen to form water, and the rest of the oxygen that is utilized combines with onrbon to form carbon dioxide as before. Thus, because some of the oxygen is used for the fsr"" <1 c... of water and does not appear as carbon dioxide, there will be formed less carbon 2 dioxide than there will be oxygen used and the respiratory quotient will be less than unity. Since more oxygen in volume is utilized than carbon dioxide produced, the total volume of the gases after exchange has taken place is less than the volume before exchange; the size of the difference will depend on how much fat is burned as this forms less carbon dioxide than oxygen utilized, therefore, the respiratory quotient approaches 0,71, The respiratory quotient will fluctuate ordinarily between 1 and 0,7,but is usually about 0,82 though in special conditions it will go outside of these ranges. To calculate the ratio of the carbon dioxide produced to the oxygen consumed, wo need to know the ratio of the volume of the respired air after exchange to the volume of the same air before exchange because technically pne collects, measures and analyzes the expired air and not the inspired air. This ratio of volumes can be calculated from a knowledge of the relative amount of nitrogen present. Since nitrogen does not take part in the metabolic exchange, the amount {number of molecules) of nitrogen is the same before and after exchange. Therefore, if the total volume of inspired air decreases, the fraction of nitrogen contained will increase and in the same proportion. If v is the volume of inspired air before exchange, V* the volume after exchange, fn, frn , f0 the fractions respectively of nitrogen, carbon dioxide and „ iLv«« corresponding fractions after exchange, then oxygen before exchange,! qu V f* ft a, 3r v - V* _ n , vt f f (VaU, A ftr steady state only *n n This correction factor must be used to calculate correctly the carbon dioxide produced, the oxygen consumed and the true respiratory or combustion quotient. The details of this calculation follow at end of article. The alveolar oxygen and carbon dioxide pressure and the alveolar ratio. However, in aviation it is not entirely the combustion or respiratory quotient that influences the partial pressure of oxygen in the alveoli, because the pressure of carbon dioxide and oxygen in the alveoli as well as the relationship between them - the alveolar ratio — depends on the following factors* (l) the rate and depth of the respiration, that is, the intensity and duration of hyperventilation. (2) on the amount of carbon dioxide produced in relation to the amount of oxygen absorbed, that is, on the true respiratory or combustion quotient, which produces a larger or smaller decrease in the volume of the expired air from that of the inspired air, and (3) the difference in the rate of diffusion of carbon diexide outward and cf oxygen inward. Variations in the predominance of these three factors will tend to make any calculations based on the analysis of a single alveolar air sample differ somewhat from the calculations based on the analysis of expired air collected over a considerable period of time. In aviation we would whenever possible determine the oxygen and carbon dioxide pressures in alveolar air directly but this is not always possible. However, during the last four years we have made many such analyses with the subject breathing air while gradually ascending from ground level (1000 feet at Rochester) to 20,000 feet and a few experiments up to 22,000 feet in a carefully oontrolied^low pressure The results thus obtained are very consistent and therefore instructive. charts The data shewn in/I-1, 1-2, 1-3, 1-4, 1-6 and XII-7 indicated that there was a definite, even progressive, decrease in the alveolar oxygen pressure up to a barometric pressure equivalent of about 12,000 feet and that above this level there 3 was a definite change in that the data no longer oould be represented by a straight line but instead gradually curved to the right; there was a corresponding fall in the carbon dioxide curve beginning at the same point indicating that at this level a sufficient degree of anoxia developed to cause hyperventilation* The first or straight part of this curve was particularly interesting and instructive because it oould be calculated with considerable accuracy by the following simple formulas Alv. = (B - 47) *2093 - {O2 absorbed in mm,) However, as the oxygen absorbed was unknown, Boothby and ®enson suggested that it oould be estimated by taking an average value for carbon dioxide and dividing this by an average H.&, .Doing this we found that a curve constructed by assuming a carbon dioxide of 40 mm, and an R.Q., of 1,0 or a carbon dioxide of 37 mm, and an R,9l, cf 0«87 would fit the data very closely in various series cf experiments between ground level (1000 feet) and an elevation of about 13,000 feet, but above this level the data oould no longer be represented by a straight line. Boycott and Haldane (Jr, Physiol., 35*355-377, 1908) had shown that en a range of four atmospheres froD about 3 3/4 atmospheres to 3/4 of an atmosphere, both the alveolar carbon dioxide and the alveolar oxygon could be represented by a straight lino; the alveolar carbon dioxide remained constant around 39 mm, and the alveolar oxygen progressively and evenly decreased. The fact that the alveolar oxygen can be predicted with considerable accuracy if we make a few assumptions based on average findings has recently been confirmed in a series of 240 experiments made at the Mayo Aero Medical Unit at barometric pressure varying from 3|- atmospheres down to £ an atmosphere* This data is shown in Chart XII-7, From our data, if all experimental values obtained at barometric pressures in excess of 500 mm, are used, it is found that the average alveolar oxygen pressure of the data can be predicted with greater accuracy from the average alveolar carbon dioxide pressure and the average alveolar ratio instead of the average re spiratorx quo ti ent as follows* / \ Mean p’rno Mean p* , « (B - 47) 0.2093 - Mean A,H, Mean PrcOo Where Mean A,R« » ~ ■ which represents what is frequently .2093 (B - 47) = p' 02 referred to by Haldane as the unoorreoted respiratory quotient. It is to be noted that Boothby, Lovelace and Benson erred slightly in using the true combustion or respiratory quotient instead of the unoorreoted respiratory quotient or, as we prefer to call it, the alveolar ratio. Let us discuss the data summarized in figure 4 in some detail at a few si gnif pressures. Example 1, At a pressure of 3,5 atmospheres where B * 2660 mm., the alveolar carbon dioxide Pco = 37 mm. and the average alveolar ratio is 0*8 (the average of all experiments above 500 mm, was 0r778), the amount of decrease in the oxygen pressure can be calculated by dividing 22 « 46. Therefore, we deduct the water vapor and proceed as fellowsj B =s 2660 r'H2fl “ JW3 x *2093 “ 557 - <»6 = 511 = p02 37 - P30z 2005 = pjj 2613 + 47 = 2660 = B The figure of 511 mm, for p02 corresponds to that actually found by alveolar air samples. Example 2, At sea level the calculation is as follows: A.R, = 0.8 = 37. The decrease in oxygen pressure « 2Z = 46 2 * 0 B = 76C PH,0 = — x .2093 = 149 - AC = 103 pGo * 713 * 37 p co2 573 pN« 713 + 47 = 760 = B Example 3, Next let us examine by three different assumptions the results obtained at 15,000 feet, B = 429 which is an important level because it is only 2000 to 3000 feet above the point where anoxia begins to make the aviator hyperventilel A, First we calculate on the assumption that the main factor influencing the A.R, is the combustion factor and that there is no hyperventilation and that therefore the A.R, is still 0.8 and the alveolar pc~ is stiH 37 mm, and therefore the oxygen decrease = = 46, .8 B = 429 PH 0 = -il 382 x ,2093 = 80 - 46 = 34 p0 ] 37 P ) ®21 311 PN 1 No hyperventilation 382 + 47 = 429 ~ 3 However, a pj. of 34 mm, does net agree with the actual observation in this series of experiments fror in other series of a similar nature. What is the difficulty It lies in the fact that the subject is hyperventilating and this reduces tbi alveolar to 32 mm, and the alveolar ratio is necessarily definitely elevated because this is an acute rapid elevation in an airplane and the subject is still blowing off carbon dioxide as will be illustrated in the example. 5 B* Therefore, with the alveolar p decreased to 32 mm, and with the A.R, CO 2 increased to 1,0 the oxygen decrease » 22 = 32 mm. In consequence I B = 429 47 Prr a » x ,2093 = 80 — 32 = 48 p« h2° 382 0 2 32 pco2 302 pn2 382 + 47 = 429 = B There is, as a result of the hyperventilation- an elovaticn of the A.R, and a lowering of the alveolar p therefore the combustion factor no longer is predominant; this produces a very large increase in the alveolar po2 from 34 ram, to 48 mm, which agrees with the observed data in figure 4, This however is not entirely beneficial and, as we will see, is in part only temporary. If instead of being in an airplane and therefore at the elevation for only a short time., the subject has gone up on a mountain to live, beneficial compensatory effects will gradually develop. These effects are known as acclimatization. Part of this process is that finally the hyperventilation produces a state of equilibrium. The alveclar C02 remains decreased at about 32 mm, but the j*.R. returns from 1,0 to normal of 0,8 as no more carbon dioxide is being blown off and the main factor influencing the A,R, is again the combustion factor = R.O, The results of this compensation are shown in the next example. C* The new acclimatized condition can be represented in part as follows; ACR. =* 0.8 and is close to and mainly dependent upon the combustion or true respiratory quotient# = which has been decreased by hyperventilation long enough for the body fluids to reach equilibrium# The oxygen decrease is ss 40 mm. Therefore* .8 B = 429 PH 0 = -il 2 392 x .2093 = 80 - 40 = 40 p* u 2 32 Pco2 310 PN2 382 + 47 = 429 * B The factors involved in acclimatization are many and sundry and their discussion cannot be taken up here. However, the first phase of acclimatization includes rather extensive and acute hyperventilation whereby the alveolar vC02 is lowered; carbon dioxide is washed out and because of the extensive hyperventila- tion the alveolar pis greatly increased as shown in example B, The second phase of acollnati zatlon as shown in Cboflns whon equilibrium of the >o4y to the * . . • ‘ ■* new alveolar pCQ is reached; the hyperventilation decreases and equilibrium is finally reached with a decreased alveolar C02 and normal A.R. and there is therefore partial recession in the alveolar p0 . In the first stage the alveclar oxygen rises from 34 mm. (example A) to 48 mm. (example B) and in the second stage merely as part of a physical phenomenon recedes about one-half its advance and tends to stabilize around 40 mm, (example C), This final decrease in alveolar is nore than compensated for by other important compensating factors. To summarize! (l) In example A where it is assumed the aviator is breathing normally without hyperventilating in an attempt to compensate for the lower partial pressure of oxygen in the inspired air, the alveolar oxygen pressure must necessarily be very low, 34 mm, (2) In example B when there is marked hyperventilation with a decrease in alveolar C02 pressure and the A,R. is therefore high, there is a marked elevation in the alveolar pressure. However, this is accompanied by all the ill effects of an uncompensated lowering of the alveolar C02 including a considerable shift in the hemoglobin dissociation curve to the left, (3) In example C, there is a recession in the partial pressure of oxygen and as a result a decrease in the intensity of hyperventilation; although the carbon dioxide remains lowered, the A,R, returns to a normal level corresponding to its normal relation to the respiratory quotient. This is merely one phase of the complicated mechanism of acclimatization to high altitude and does not occur in aviators under ordinary circumstances. Bill, Christensen and Edwards (Gas Equilibrium in Lungs at High Altitudes: Am. J, Physiol,, 1936, 115{537) make the following statement: Another observation was made which indicates that the occurraiDe of mountain sickness does not depend closely on the oxygen saturation of arterial blood. In several individuals arterial blood was drawn soon after arrival at a station and again a week or ten days later# Usually little change was found to occur in saturation although acclimatization had been going on in the meanwhile. In the case of Dill at Montt the saturation on the day of arrival was 73,7 per cent and 9 days later 71,6 per cent. On the second day after arrival he had had typical mountain sickness, but from that time on remained free of symptoms despite the fact that he had had the lowest saturation recorded at the station,1* Similar calculations for the data in figure 4 can be made for 18,000 and 20,000 feet. At those elevations the hyperventilation is so great thst few individuals can withstand the combined effects of acapnia superimposed on anoxia for more than a short period. The main value of the data and calculations presented in this paper rests on the fact that it forms a basis from which to calculate the amount of oxygen needed to maintain the alveolar oxygen pressure of the aviator at any desired level by the proper administration of oxygen. The average alveolar oxygen for any altitude up to 12,000 or 13,000 feet has been well established and this will not vary in normal young men up to these altitudes any more than it will at sea level. In fact, its regression with increasing altitude is ne constant and predictable over a total of 4 atmospheres as is the constancy of the alveolar PCO2 'kkroughout a similar range of barometric pressure. Between elevations of 12,000 and 20,000 feet by making allowance for hyperventilation the alveolar oxygen pressure as well as that of carbon dioxide can be predicted within 3 or 4 tame INSPIRED VOLUME If V Is the volume of respired air before exchange, V’ the volume after exchange, ffl, , fractions respectively of nitrogen, carbon dioxide and oxygen before exchange, f’n, **02 oorrespending fractions after exchange, then; V f« ft —— =3 —2. or V = D V* V* f p n The volume of carbon dioxide produced is C02 * v,f'co2 - v'co2 “ v,f’co2 - V pi fCOz - V. (f.COz - fC02 - £^J n fn The volume of oxygen consumed is °2 = ™ . V.f. . V. f\ n - Vf> - T.(f0 f- Boothbu - Benson (calculated R.Q.1.0) C - Normal respiration Barometric pressure - cmJ of Ho. Altitude -7 thousands of feet Oa curves COo curves i BoOTHBy AND DUBLIN Alveolar* pressure — of Mg. Alveolar 02 and COa Pressures and Alveolar Ratio at Various Altitudes while Breathing Air - Rapid Ascent All Subjects. Acclimatized to an Altitude of lOOO felt Number of Observations Legend • Individual Observations O Mean of Individual Observations ( Total Number of Observations - Zc?5) I Alvloi air Q, Pressure I Calculated Alveolar Oa Pressure pCfe-02093 (BP-T7J- Theoretical Alveolar Oe Pressure^ (WITHOUT HvPERVENTIC ATIOm) I Auveolar COa Pressure (Mean: Ground to 12,000 ft. = 37 mm Ho) Theoretical /Alveolar CO£ Pressure (without Hyperventilation) Alveolar C02 Pressure Alveolar Ratio (Mean: Ground to 12,000 ft. =0.892) Alveolar Ratio Theoretical. /A. R, J (without H VPERVEAiTILATlON ) 1 BP-245 Barometric Pressure - mm. of Hg i i Al TITUDE — THOUSANDS OF FEET MayO AeRO~ MEDICAL UnIT Pressure - mm. Hg. Alveolar pressures for venous total atmospheric pressures » Mayo Aero Medical Unit 1 O2 Pressure-mm. Hg. Alveolar ratio CO2 Pressure 1 1 _ NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE RESTRICTED Report No. 341 Date 14 June 1944 ALVEOLAR RESPIRATORY QUOTIENTS I AN EXPERIMENTAL STUDY OF THE DIFFERENCE BETWEEN TRUE AND ALVEOLAR RESPIRATORY QUOTIENTS, WITH A DISCUSSION OF THE ASSUMPTIONS INVOLVED IN THE CALCULATION OF ALVEOLAR RESPIRATORY OUOTIENTS AND A BRIEF REVIEW OF EXPERIMENTAL EVIDENCE RELATING TO THESE ASSUMPTIONS• From the Mayo Aero Medical Unit, Rochester, Minnesota by J, B. Bateman and Walter M, Boothby, SUMMARY Presented at the meeting of the Subcommittee on Oxygen and Anoxia, N,R,C,,June 16, 19-^4 1, Measurements have been made of the resting metabolic rate and the composition of Haldane-Priostley alveolar air from four subjects before and after a meal of rice Experiments were carried out both at ground level and ?t 12,000 feet simulated alti tude. 2. Comparison of "true" and "alveolar" respiratory quotients reveals the following f ct a. Ground level 12,000 feet s,a„ Range of true R.9, values* Alveolar R.O., is on the average 0,704 - 0.930 0.782 - 0,953 less than true R.Q., by 0.0155 0.0236 3ame in per cent 2 3 •, The small average differences mask considerably larger systematic personal differences and a pronounced tendency for the composition of Haldane-Priostley samples to show a smaller response to changes in true respiratory quotient than would be anticipated from the simple theory. 3, Examination of the tacit assumptions involved in the use of the usual alveolar R„3 formula suggests the following conclusions to be drawn from our experimental data* a. As an approximate procedure for the establishment of oxygen standards for aviators, the use of the alveolar air equation is justified. This does not imply, however, that for practical purposes the use of the "tracheal" air as a point of reference may not be more convenient. b. Systematic discrepancies which tend to disappear from averaged data demonstrate some failure in the assumptions underlying the alveolar air equation. 4, Existing information which might throw light upon the sources of failure of the equation is briefly reviewed. The review focuses attention upon the need for more refined techniques in the study of alveolar gases, and particularly in the inves- tigation of practical procedures - such as intermittent pressure breathing - which probably involve radical changes in the conditions under which pulmonary exchange takes place. NATIONAL RESEARCH COUNCIL* DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE Report No, 341 Date 14 June 1944 RESTRICTED ALVEOLAR RESPIRATORY QUOTIENTS* AN EXPERIMENTAL STUDY OF THE DIFFERENCE BETWEEN TRUE AMD ALVEOLAR RESPIRATORY QUOTIENTS, WITH A DISCUSSION OF THE ASSUMPTIONS INVOLVED IN THE CALCUL. TION OF ALVEOLAR RESPIRATORY QUOTIENTS -uND A BRIEF REVIEW OF EXPERIMENTAL EVIDENCE RELATING TO THESE ASSUMPTIONS, From the Mayo Aero MeAloal Unit. Rochester, Minnesota by J * B, Bateman and Walter M, Boothby, 1, Introduction, Through the publication of the recent "Handbook of Respiratory Data in Avia tlon wide currency has been given to the idea that changes in the partial pressure of oxygen in the alveoli resulting solely from changes in the fraction of inert gas in the inspired gas mixutre, the other pertinent quantities being kept oonstnat, can be calculated with the aid of a single additional variable, the metabolic respiratory quotient, Q"' , It is the purpose of the present paper to record new experimental data concerning the validity of this calculation, to examine the assumptions underlying the practical application of the equation involved, and finally to review briefly existing evidence concerning several questions that arise from our statement of the'premises. It will be suggested that, althoug these experiments may not greatly affect specifica- tions based upon the calculations presented in the "Handbook", they are not without a certain practical bearing upon the physiology of flight at extreme altitudes. 2, Nomenclature. Nomenclature slightly simplified from Bateman (1943)s P as barometric pressure P = P - 47 pO, pC, pN = partial pressures of oxygen, carbon dioxide and nitrogen in dry inspired gas at pressure P, The same symbols with one prime (e.g, pO*) refer to the partial pressures in the moist inspired gas mixture at body temperature (37° C) ("tracheal" gas). Two primes refer to "alveolar" partial pressures (thus* pC"), three to moist mixed expired air at body temperature (pN"'), and four to dry expired gas at pressure P, fO, fC, fN are the corresponding fractions of oxygen, carbon dioxide and nitrogen, related to the partial pressures as follows* fO = pO/P fN = pN/P = f N* = p H*/P fC* = PC-/P etc. 2 3. The alveolar air equation. The assumptions underlying the alveolar air equation used in the "Handbook* become obvious if the equation is written in symmetrical form as follows* f0"" + fC""/a) = j~ (fo + fc/a) = — (fo- + tc*/a) The left hand equation is the fundamental relation between the composition of inspire! and of expired air, and its well-founded derivation need not be repeated here. The right hand equation represents the extension of that on the left to any fraction what- soever of the expired air. For a given value of 0. it is clear that it can apply to any gas that, when mixed with a suitable proportion of fresh inspired gas, will give a mixture identical in composition with the total expired gas. It will not apply, conversely, to any mixture that does not satisfy this condition. This restriction thus limits the validity of the equation to portions of the inspired gas which have exchanged oxygen for carbon dioxide in the proportion 1 /GL and have then undergone, to a greater or lesser extent, mechanical admixture with unmodified inspired air. The limitation is important for several reasons. In the first place, it requires gas exchange in the lungs to occur, at each instant throughout the whole respiratory cycle, in more or less strict accordance with the metabolic respiratory quotient CL, This is in sharp contrast to the original equation, which makes allowance for any variations in the relative rates of transfer of oxygen and carbon dioxide that may occur during the respiratory cycle and demands only that a constant overall value of CL be established by the final composition of the total expired air. In the second place, it is neces- sary that modification of the alveolar gas during its removal from the lung shall occur solely by mechanical admixture and not, to any considerable extent, by diffusion. It is not difficult to imagine that those conditions could be violated very readily if. for example, uptake of oxygen in the lung is much slower and anatomically more circumscribed than loss of carbon dioxide; or if the ratio of blood flow to ventilation should be significantly different in different parts of the lung or should vary significantly during the respiratory cycle. These premises have not to our knowledge been clearly stated, although they have doubtless been apparent to the several workers who have presented derivations of ''alveolar air" equations; nor have they been subjected to any careful experimental scrutiny, A clear vindication would be provided by agreement between values of the respiratory quotient calculated from analyses of total expired air, on the one hand, and from various fractions of so-called "alveolar air", on the other, by means of the two parts of equation (l). Dr. Brink has verbally assured us that the approximate agreement between average respiratory quotients as given in metabolism tables, and as calculated from Boothby’s numerous analyses of alveolar air provided the justifica- tion for the semi-official adoption of an alveolar air equation for the purposes of aviation physiology. Such approximate agreement of averages for different group* of teat subjects is not altogether satisfactory, and indeed, in view of HaldaneTs state- ment (Haldane and Priestley 1935, p. 39) that "the true respiratory quotient is about a sixth higher than the alveolar respiratory quotient", is rather unexpected. It seemed desirable, therefore, to put the matter to direct test by determining to what extent changes in respiratory quotient, induced by eating suitable food, and measured by the standard methods of the metabolism laboratory, would be reflected in the composition of the alveolar air as determined by the Haldane-Priestley technique. The experimental part of this paper deals with the results of such experiments. It should be added that a satisfactory demonstration of the value of equation (1) would not carry with it any implication concerning the gas pressures which actually determine the composition of arterial blood. The generality of the equation is in this 3 respect its weakness, although in such applications as are propounded in the "Handbook" this is immaterial, since the calculations have reference to the particular mixture of "alveolar" gases that is obtained by a particular technique of collection. It is reasonable to assume that such a mixture will stand in a constant relationship to the "true" alveolar air, but it is not necessary for the two to be identical. This very elementary fact should be kept in mind, because it is always possible that under special conditions of collection of alveolar air — in pressure breathing, for example - the constant relationship may be lost, and the significance of such alveolar analyses may then require reconsideration. 4, Experimental comparison of respiratory quotients. The experiments consisted of a series of collections of expired air, each being followed immediately by alveolar sampling by the Haldane-Priestley method. The first two determinations were made with the subject in the basal state. The subject then consumed as much boiled rice as possible, and the collection of samples was resumed after an interval of twenty minutes. Two experiments were performed on each of four resting subjects, one at ground level (l,000 feet) and one in the decompres :ion chamber at a simulated altitude of 12,000 feet, breathing air. The following time table illustrates the procedure in a typical experiment! 0 minutes subject seated quitely in decompression chamber minutes collection of expired air 28 minutes collection of alveolar air 33 minutes collection of alveolar air 34-36 minutes decompress to 12,000 feet simulated altitude 46-54 minutes collection of expired air 54 minutes collection of alveolar air 59 minutes collection of alveolar air 60-66 minutes rice with sugar and milk being eaten 86-94 minutes collection of «ipir«d air 94 minutes collection of alveolar air 97-104 minutes collection of expired air 104 minutes collection of alveolar air 126-133 minutes collection of expired air 133 minutes collection of alveolar air 134 minutes return to atmospheric pressure The results are presented in Tables 1 and 2, from which it is apparent that at ground level, over a range of respiratory quotients, Q.w * , of 0.704 to 0,930, the value ft" calculated from the composition of alveolar air is on the average 0.0155, or roughly 2 per cent, less than the true value, 0W|, At a simulated altitude of 12,000 feet, with Cl"1 ranging from 0,782 to 0.953, the average difference is or about 3 per cent, in the same sense. Thus stated, the results throw a very light upon the use of the equation (l) aid the metabolic respiratory quotient in calculations of the composition of alveolar air. The small average differences, although in the same direction, appear to discredit the discrepancy of 17 per cent noted by Haldane. Closer inspection shows, however, that the low average difference tends to mask the consistent occurence of larger discrepancies in individuals; thus the average value of 0." * - ft" for the one male subject, MC, is + 0.092 at ground level and +0,075 at 12,000 feet, while for the subject, HC, the corresponding mean values are -0,019 and -0,064, These idiosynoraoies and their obliteration by the pooling of data for several individuals are also illustrated in Fig, 1, in /h ich 0" is plotted against a"1, For comparison, Haldanefs data (Haldane and Priestley, 1935, p, 39) are also plotted in Fig. 1* Aside from these individual differences the points that, when averaged, give such close agreement between 0.** and ft" are rather widely scattered. T his is •bvious from inspection of Tables 1 and 2 and can also be seen in Fig* 2 and Fig, 3, where ft"1 and ft" are plotted against time. Although these diagrams sh«w the general trend toward parallel changes in the two quotients, they also show that the uncertainty of measurement is so great that even quite considerable changes in ft"' may be accom- panied by opposing changes in ft". Because the chief present interest in the use of equation (l) lies in the calculation of alveolar partial pressures, it is instructive to express this scatter of experimental points quantitatively by comparing the changes in the experimental alveolar partial pre s sure s ,Po" and pC", with these calculated fro*. the accompanying changes in ft"*. This procedure has the advantage that of all the measurements made, ft"' is the least subject to error, so that a consistent discrepancy between calculated and observed values of pO" and pC" would serve to throw doubt upon the assumptions underlying the "alveolar" part of equation (l), while random discrepan- cies would give a measure of the variability of alveolar samples collected by the Haldane—Priestley technique. For the purposes of our comparison, equation (l) may be used in the following equivalent forms, in which pC is assumed to be zeros pO" = pO* (1 - pC"/P) - (1 - p0»/p).Pc"/a (2' pcw * ap (Po» - po") / (a.po* + p - Po») (3) The changes in alveolar partial pressures resulting from changes in ft con then be calculated for two extreme cases* (a) Suppose the change to occur wholly by change in the alveolar oxygen pressure, the ventilation rate being adjusted to keep the carbon dioxide pressure constant, then *(pO")/*a = (1 - pO*/F).pCVa2 (4) (b) Suppose the alveolar carbon dioxide pressure to be affected by the change of ft while the oxygen remains constant; then Xpc*) = I(po’ - po*).(r - p°‘) *a Q? _ po* (1 - a)^2 (5) These two extreme cases are illustrated numerically in Fig, 4, in which the ordinate of any point on the appropriate curve represents the increase in alveolar oxygen or carbon dioxide pressure that should accompany an increase of respiratory quotient by 0,1 unit when ft has a mean value given by the abscissa. The curves are approximate only, and are applicable to the condition of subjects breathing air at altitudes of 0 to 12,000 feet. The fact that the oxygen und carbon dioxide curves do not intersect precisely when 0. is unity, as they should according to equations (l), (4) and (5), is due to the fact that the values of pO'' and pCw used in the calculation, derived from Boothby’s data (reference (l). Chart A-l), are not mutually compatible with equation (l) for all values of d, The error involved is however not significant for present purposes. Using Fig, 4, we have calculated for all experiments the changes in alveolar /oa?$on Sfoxide that would be expected according to the change in ft"* which occurred from one set of measurements to the next. The adjustment does not in general occur by a change only in pC" and pO" but by a change in each of these quantities; the sum of the changes in pC" and pO" should however lie somewhere within the range calculated by equations (4) and (5) for the two gases separately. This forms the basis for the oomparison given in Tables 3 and 4 and in Fig* 5; the final column in the tables gives the differences between the observed change,, + and the mean of the changes calculated for oxygen and carbon dioxide separately# As Fig. 5 illustrates clearly, the uncertainty of measurement is such that in the majority of oases the alveolar partial pressures decrease when an increase would be expected, and vice versa# Expressed numerically, the average discrepancy, regard- less of sign, between the observed and calculated alveolar pressures is 3,35 mm#, for subjects at ground level (2,59 mm, if the one obviously discordant measurement is omitted), and 2,67 mm, at 12,000 feet simulated altitude. The observed change has a strong tendency to be less than that anticipated; on the average, 2,14 mm, less (1,29 mm, omitting the discordant figure) at ground level and 1.88 mm. less at 12,000 feet. The first pair of figures, 2,59 and 2,67 mm#., represents a somewhat smaller scatter than that found in Boothby’s 186 observations on a considerable number of subjects at ground level, as would indeed be expected of data obtained by repeated measurements on a small number of individuals#. . xv The second pair, 1,29 and 1,88 mm,, are sufficiently large in comparison with the first to carry a strong suggestion of a systematic tendency for the composition of the alveolar air to change less than would be expected in response to changes in respiratory quotient; the magnitude of the effect is however so close to that of the random error of sampling of the Haldane—Priestiey method that further work would be necessary to establish it with certainty. 5# Discussion, The close agreement between average respiratory quotients obtained from total expired air and from alveolar air must be regarded as a vindication of the use of equation (l) in drawing up average requirements for maintenance of alveolar partial pressures within the physiologically permissible range under conditions not too far removed from normal. The consistently low values of the alveolar respi- ratory quotient reported by Haldane have not been verified in our four subjects taken together but have been observed in one subject, while another subject exhibits the reverse tendency for the alveolar quotient to be too high# Such individual differences point to some lack of uniformity in the character of the alveolar air collected from different persons by the Haldane-Priestley technique, and therefore to some -failure, highly variab-le in degree from person to person, of the assumptions underlying equation (l). The tendency for changes in the composition of Haldane- Priestley alveolar samples to be less than those predicted in response to comparatively large changes in respiratory quotient, with their concomitant changes in ventilation rate and oxygen consumption (of. Fig# 2 and 3), points to the same conclusion, and it is of interest in this connection to consider existing data on the nature of the changes in alveolar air during the respiratory cycle and to review the ideas which have been advanced by way of explanation# It should be mentioned in passing that although few direct comparisons of alveolar and true respiratory quotients are to be f«und in the literature, extensive data on the apparent magnitude of the respiratory dead space calculated by Bohr’s formula testify to the fact that the two quotients are rarely in perfect agreement, since the dead space calculated from carbon dioxide pressure is systematically smaller than that from oxygen pressures for persons breathing air. It is evident that such a discrepancy implies a difference between 9* and 9"•, From equation (l) it follows that if a" = a*» pC" _ pc»* p0» (1 - pC W/P) - po* pO1 (1 - pC"’/?) - pO * * (6) On the other hand, the calculation of dead space is done with the equations V - Do « V,pC"’/pC" (7) V ~ D0 « V.(pO* - pO"1)/(p0 * - pO”) (8) (Y « total -volume expir ed, D * dead space), so that if Dc “ Do P«" pC"♦ — pO • - pOw pO • - po** * (9) and it can be seen by inspection that equation (9) represents a necessary condition for the equality of dn and Q," ' expressed by equation (6), Evidence concerning the origin of what we nay call diffusion processes in determining the composition of Haldane-Priestley samples, and of implied variations in apparent respiratory quotient, is derived almost exclusively from the results of experiments in which two or more samples of alveolar air are collected du»ing a single expiration* The experiments may involve either the normal alveolar air or the distribution of a foreign gas inhaled prior to the measured expiration* Despite considerable disagreement in points of detail, the information gained is fairly definite in outline, although still ambiguous as to implied interpretation. The simplest conditions are of course provided by the use of an inert foreign gas which plays no part in respiratory exchange. The early experiments of Siebeok (1910), which were made without fractionation of the expired air, appeared to show an equal distribution of inspired hydrogen in the entire residual air, al- though according to Siebeok, Grehant (l864) had already shown that of two alveolar samples taken during expiration the first always contained mere hydrogen than the second. This has since been repeatedly confirmed both by fractionation procedures (Krogh and Lindhard, 1913; Roelsen 1934, 1938, 1939) and use of a hot wire method (Aschoff et al. 1940) for continuous recording of expired hydrogen (Mundt, Schoedel and Schwarz, 1940); the rapid fall of hydrogen concentration during washing out of the dead space air is succeeded by a slow continuous decrease throughout expiration, Sonne (1934) has shown a similar decrease to occur after several deep rapid breaths of hydrogen. These results point conclusively to the imperfections of the lung as a nixing chamber, and because of the high rate of diffusion of hydrogen they must rep- resent a conservative estimate. Qualitatively concordant conclusions have been reach- ed by measurements of the rate at which the nitrogen of the residual air is washed out during breathing of pure oxygen (Ehgelhardt, 1939, Darling, Cournand, and Richards,1944) When we consider changes in the composition of expired alveolar air the situation is greatly complicated by continued respiratory exchange during the period of expiration. Measurements of this kind have significance therefore only if allow- ance can be made for this circumstance. The simplest assumptions that can be made are that respiratory exchange continues throughout the respiratory cycle at the uniform rate established by ordinary determinations of oxygen uptake and carbon dioxide production, and that the composition of expired alveolar air, as found from a samole collected at a given instant, corresponds exactly to the instantaneous composition of the entire residual air* these assumptions are valid, the observed rate of decrease of alveolar oxygen should agree with the rate calculated from the average rate of oxygen uptake and the total volume of residual and reserve air at the instant of measurement, and the value of Q." should remain constant, and equal to Q"* , for all samples collected. The results of several such comparisons have been cwllcotod in Table 5, which contains data referring to the middle range of either normal expirations or of deep (•alveolar'*) expiration proceeded by normal inspiration, and is confined to subjects at rest. The results listed under Krogh and Lindhard have been recalculated to conform to this manner of presentation. Attention is drawn to the following points* (l) Normal expiration; the data, representing average values for the smoothest part of a normal expiration, show, for two experiments on •ne individual, approximate agreement between the observed and calculated gradients; for seven experiments, on the other hand, the average observed gradient is considerably greater than that calculated. (2) Deep expiration* all data show, in the first stages, a consistently higher average rate of apparent oxygen absorption than that calculated, and a fall of alveolar respiratory quotient at a rate of about 0,02 per second. This behavior is succeeded, however, by a region in which, if we leave out of consideration the single experiment of Krogh and Lindhard, the apparent rate of oxygen absorption becomes considerably less than calculated, and may even change in sign, indicating a terminal rise in pO" and fall in pCH (see also Mackay 1940). This is accompanied by a relative constancy, or a slight increase, of Qi", The effect appears to be fairly well established, and the wide personal variations recorded would seem to reduce the significance of the single aberrant result obtained from Krogh and Lindard's data. These data show clearly that the processes underlying the changes in composition of expired alveolar air are somewhat complex. Interpretations vary, Krogh and Lindard, in a somewhat neglected paper, assumed that the apparent rates of oxygen consumption and carbon dioxide production calculated from the changes in alvetlar partial pressures were the true instantaneous rates, and that the former values were proportional to the pulmonary bleed flow, which was thus shown te be subject to enormous respiratory variations, the pattern of variation being totally different for different types of breathing. On the other hand, Krogh and Lindhard (1917) attributed their results with hydrogen, referred to above, to imperfect mixing in the alvoo lar sacs, which was supposed to occur chiefly by diffusion and to take an appreoialbe time. The existence of such gradients, which would surely introduce inconsistencies into Krogh and Lindhard,s calculations of pulmonary blood flow, is however doubtful, Mundt, Schoedel and Schwarz (1940) found that a pause between inspiration and expiration had no apparent equalizing effect on the hydrogen distribution; furthermore, modern views as to the mechanism of ventilation by simul- taneous increase in length and cross section ef alveolar ducts and air sacs, as well a_i of the larger air passages (of, Maoklin; 1929), would seem to make it probable that regions of imperfect mixing, within a given lebule, would be confined to the layer of stagnant gas in the alveelar depressions, and even these are presumed to become shallower as a result of distention of the air sac during inspiration. If the air sacs be thus regarded as perfect mixing chambers, we must seek to explain manifest irregularities as a result of differences in the ventilation of morpho- logically distinct regions of the lung. In accepting such a view, which has been particularly strongly urged by Sonne (1934, 1936, 1937, 1940; Sonne and Nielsen, 1932; Nielsen and Sonne, 1932 a, b) and Roelsen (1934, 1938, 1939), we alse admit that Krogh and Lindhard’s deductions concerning pulmonary circulation must be considerably in error, Sonne and his collaborators attempt to explain the exces- sively rapid fall in pOw and the rise in pC" as the result of the gradual admixture of air from relatively hyperventilated regions, which is expelled in the earliest stages of expiration, with air from poorly ventilated parts. Since this process is accompanied by a fall in alveolar respiratory quotient, it is clear that the ventilation and the blood flow in different regions of the lung are not so precisely matched as to give the same relative rates of transfer of oxygen and carbon dioxide; the relationship varies indeed to such an extent that terminal samples may even show a rise in p®** and a rise in CL*’• This cannot be explained on Krogh and Lindhard’s assumptions, and unless due to technical error can only mean that the terminal sample comes from a region which is both poorly ventilated and poorly perfused with blood (Sonne et al). It is our opinion that the probability, thus established, of regions of unequal ventilation and of unequal perfusion constitutes a modification of Krogh and Lindhard’s view rather than a refutation of it* The existence of local differences in blood flow - which have also been directly observed by Wearn and his collaborators (1926, 1934) - does not preclude respiratory changes in average blood flow; these can indeed be inferred from a whole series of phenomena. Much of the immediate value of Krogh and Lindhard's work lies in their emphasis of the changes in alveolar respira- tory quotient which inevitably occur during the respiratory cycle as a result of the comparatively steady uptake of oxygen and the gradual decrease in rate of carbon dioxide output as the gradient between the tension of carbon dioxide in the alveoli and in the blood becomes reduced toward the end of expiration* Since this can lead to changes in the instantaneous values of CL* f from about 4,0 to 0.16 during the course of expiration, it is not surprising that certain discrepancies appear in values of Q.1 * calculated from Haldane-Priestley samples, and that personal idiosyn- oraoy is evident in such data* It is equally clear that uneven ventilation of the lungs would also be expected to lead to discrepancies of the same kind unless there *.r were an extremely fine matching of ventilation with local blood flow. Our own ex- perimental contribution, besides its vindication of the alveolar air formula for limited practical use, has forced our attention upon the inadequacies of the Priestley method as a weapon for the further detailed study of gas exchange in the lungs. Having reviewed briefly the available data elucidating some of the factors which are ignored in the Haldane—Priestley technique, we wish to emphasize the urgent importance of the application and extension of this knowledge. On the academic side, the question of the equilibrium between alveolar gases and arterial blood cannot otherwise be carried beyond its present rough and ready state, while the practical value of different forms of intermittent pressure breathing can only be assessed with the aid of an intimate knowledge of the effect of such procedures upon the course of gas exchange in the lungs* Reference s Aschoff, J,, Mundt, E,, Schoedel, W, and Schwarz, H, 1940, Pflueg, Aroh. ges. Physiol,, 244, 87, Bateman, J, B, 1943, O.S.R.D., Committee on Aviation Medicine, Report No, 222, Barling, R, C,, Cournand, A, and Richards, D, W, 1944, J, Clin. Invest,, ,23, 55, Engolhardt, A. 1939. Z. Biol,, 99, 596. Grehant, N, 1064, J, de l*anat, et de la physiol., £, 523, Cited by Siebeok, 1910c Grosse-Brookhoff, F. and Schoedel, W, 1937, Unpublished data cited by Sohoedel, 1937* Haldane, J, S, and Priestley, J, G, 1935, Respiration, Yale University Press, New Have n. Handbook of Respiratory Data in Aviation, O.S.R.D,, C.M.R,, Washington, D, C,, 1944, Krogh, A, and Lindhard, J. 1913, J, Physiol,, £7, 30, Krogh, A. and Lindhard, J, 1914, Bioohem, Z,, 5J9, 260# Krogh, A. and Lindhard, J, 1917, J. Physiol., 5£, 59, Mundt, E,, Schoedel, W, and Schwarz, H, 1940, Pflueg, Arch, ges. Physiol,, 244, 99, Mackay, I, F, S,, 1940, J, Physiol,, 9J8, 73, Macklin, C, C,, 1929, Physiol. Rev., £, 1, Nielsen, E. and Sonne, C, 1932 a, Z, ges, exp, Med,, 85_, 46, Nielsen, E, and Sonne, C, 1932 b, Acta medica Scand,, Suppl, 50, 33, Roelsen, E, 1934, Acta medica Scand,, Suppl, 5£, 356, Roelsen, E. 1938, Acta medica Scand,, 95, 452, Roelsen, E, 1939* Acta medica Scand,, 98i, 141, Schoedel, W, 1937, Ergebn, Physiol,, 3J9, 450, Siebeck, R, 1910. Z. Biol,, 55, 267, Sonne, C. and Nielsen, E. 1932, Acta medica Scand,, Suppl, 50, 26, Sonne, C, 1934, Z. ges. exp, Med,, 9£, 13, Sonne, C, 1936, Acta medica Scand,, 90, 315, Sonne, C, 1937, Sotnd, Arch, f. Physiol., 75, 127, Sonne, C, 1940, Acta medica Scand,, 105, 313, Wearn, J. T,, Barr, J, S, and German, W, J, 1926-27, Proc, soo, exp, biol, tried,, 24, 114. Wearn, J, T,, Ernstene, A, C., Bromer, A. W,, Barr, J, S., German, W, J, and Zsohiesche, L, J,, Am, J. Physiol,, 109, 236, (1) (2) (3) (U) (5) (6) (7) (8) (9) . do) (id (12) Time V Expired Air Alveolar Air Date Sul 3. (min.) P L/min. pO"* pCn* pC" Qm Q" Q»" - Q" 12/22/U2 HC(f ) 52-1*1 71*0.1* 6.32 121.6 19.0 106.9 32.5 0.759 0.809 -0.050 37-25 6.5$ 121.2 18.2 106.8 33 a 0.701* 0.823 -0,119 20-29 8.97 121.8 20.1 95.1 35.8 0.822 0,662 +0.160 33-1*0 10. oh 125.2 18.8 108.5 33.7 0.918 0.898 +0.020 81-91 8.36 122.6 19.1* 109.0 31*.0 0,820 0.925 -0.105 1/13/1*3 HC(f) 71*3.2 7.13 123.2 18 a 107.6 30.8 0.757 0.761 -o.ool* — — — 105.3 30.9 — 0.712 — 12/22/U1 LC(f) 1*7-37 71*0.1* 7.08 117.7 21.8 102.8 31*.3 0.71*2 0.762 -0,020 33-22 7.88 119.8 21.0 103.1 31*.0 C.7C3 0.761* +0.019 20-30 8.71* 119.5 22.5 106.2 35.7 0.838 0,890 -0.052 35-U5 9.55 121.1* 21.1* 100.3 36.5 O.86U 0.769 +0.095 81j-93 8.90 H9.3 22.5 103.5 36.5 0.830 0.81*1* -O.Oll* 1/1*/1*3 I£(f) 733.3 7.81* 119.6 21.5 105.1* 33.1 0.851* 0.829 +0.025 — — — 10U.6 31.1 — 0.71*8 — 12/28/1*2 RS(f) 1*7-35 738.6 7.77 H9.3 22.1* 111.2 31.1* 0.81*0 C.911 -0.071 29-19 7.78 117.8 22.9 107.8 3 2.6 0.801* 0.8U7 -0.01*3 22-31 9.1*0 119.6 23.5 111.5 31.7 0,908 0.933 -0.025 37-1*5 9.50 120.3 23.3 110.9 33.5 0.930 0.971* -0.01*1* 8h 9.20 119.0 23.7 — — 0.888 —— — 1/12/1*3 HS(f) — 71*2.7 6.89 121.2 22.5 108.0 33.9 0.889 0.871 +0,018 — — — — 102.8 35.6 — 0.790 12/28/1*2 MC(ra) 37-29 738.6 11.1*5 125.2 16.3 99.5 31*.9 0.781* 0.721 +0.063 23-16 12.27 125.9 16.3 99.8 35.8 0,8lh 0.750 +0.C6U 2l*-30 il*.5o 125.1 17.8 100. U 36.2 0.867 0.772 +0.095 36-1*3 13.51 12U.8 18.9 101.7 35.7 0.922 0.786 +0.136 83-89 13.31* 12l*.l 18.8 99.9 36.5 0.873 0.770 +0,103 12/31/1*2 MC(m) 733.9 10.93 122.2 18.5 98.0 31*.8 0.811* 0,708 +0.1C6 — — 98.6 3h.2 0.705 Mean +6.0155 Table I EXPERIMENTS AT GROUPS) LEVEL Table 1 Cont'd, Explanation* Column (2)* f * female m = male Column (3)* Times are reckoned from time at -which subject finished eating rice. Negative sign is omitted from times preceding rice The second figure in each horizontal pair gives the time of completion of collection of expired air and also the time at which alveolar sample was taken. Column (5): V = ventilation rate in liters per minute at ambient pressure, body temperature* and saturated with water vapor* Columns (10) and (11) = respiratory quotients calculated from analyses of expired air (&"’) and of alveolar air (CL**) by equation (1), suitably rearranged. (1) ( 2) (3) (4) (5) (6) (7) (&) (9) (!•) (ii) (12) Dat 0 Sub ject Tine P V pO "f pC*» pO • pc- a* 13 Jan, HC (f) 22-12 481,6 7.80 70.5 16.9 59,8 28,1 0.782 0,872 -0,090 -7 — — — 52.6 30,6 — 0,750 — 20-28 10,85 71,2 18.3 62,1 28,9 0,897 0.991 -0,094 30-38 11.05 71.6 18,3 62.1 27.7 0,916 0,943 -0,027 60-67 11.65 72.1 17.6 62,4 28,1 0,912 0,977 -0,065 4 Jan, LC (f) 20-12 -7 477,© 8.63 68,0 19.9 54,0 30,6 0,859 0,806 +0,053 20-28 31-38 60-67 9,97 10.67 11.01 67.4 68.5 68.5 19.2 19.2 20.0 54.1 55,9 53,8 49.1 28.7 29,9 31,0 33,4 0,795 C , 850 0,892 0,750 0,834 0,812 0.770 -0,039 +0,038 +0,122 12 J an. RS (f) 21-13 -8 477,7 7.95 66,5 21.3 53.2 32,6 0.871 0,849 +0,022 21-28 32-38 60-68 9 3 10,13 f , 49 66.4 68,0 67,2 21,7 21.9 22.2 49,6 56,0 53.5 54.5 32,1 31.7 33,3 32.7 0,865 0,953 0,924 0,746 0.889 0,869 0,878 -0,024 +0,084 +0,046 31 Dec, MC (m) 34-28 481.6 12,46 71.4 1 7 . A 23-17 12.20 70.7 17 c J j § 4 28.1 0,tt66 0.745 +0.121 20-26 14.64 71.2 ifi n t) 3 0 5 30,7 0,843 0,777 +0,066 32-37 16.27 71,9 1ft R 5 X • 0 32,4 C.890 0,774 +0,016 76-81 17.00 73,4 1 ft n 51,8 31,6 0,866 0.768 +0.092 o , u 54,6 30,6 0,082 0,802 +0.080 Explanatio n t See Table 1 Moan +0.0236 TABES 2 Experiments at 12,000 Feet U) (2} (3) (4) (5) (6) (7) (8) (9) (10) HiT (12) (13) Observe d Calculated from G** Bisorepancy Bate Subject Q"« &Qim' pC" ApC* pO" ApO" £pCM+ jftpC • *apc" pO" Average (12)—(13) 22 Be c , HC 0^759 32.5 ' 106.9 -0.055 +0-.* - 0.1 + 0.5 -2.1 -2*9 -2.5 - 3,0 0.704 33„1 106.8 +0.118 + 2.7 -11,7 - 9.0 + 4*4 + 5,7 + 5,* -14,0 0*822 35.C 95.1 0.918 +0.096 33.7 -2.1 108,5 + 13.4 + 11.3 + 3*4 + 3*5 + 3,5 + 7,8 0.620 —0.098 34.0 +0,3 109.0 + 0,5 + 0.8 -3,4 -3,6 -3.5 — 4*3 22 Bee# LC 0.742 34,3 102.8 +0,041 —0,. 3 +0.3 0 + 1,5 + 2,0 + 1,8 -1.8 0.783 34,0 103.1 +0,055 + 1^7 -3.1 - 1.4 + 2.0 + 2,3 + 2,2 -3,6 0,838 35.7 106,2 +0.026 +0.. 8 -5,9 -5,1 +0,9 + 1,0 +1,0 -6,1 0,864 36,5 100.3 —0,034 0 + 3.2 + 3,2 -1.2 -1,3 -1.3 -4,5 0.830 36,5 103.5 28 Bee, RS 0.840 31.4 111.2 -0.036 + 1.2 -3.4 -2.2 -1.3 -1,5 -1.4 + 0.8 0,804 32.6 107.8 +0.104 -0.9 + 3.7 + 2,8 + 3.7 + 4.1 + 3,9 -1,1 0.908 31,7 111.5 +0.022 + 1.8 -0.6 + 1.2 + 0.8 +0,7 +0.7 + 0,5 0,930 33.5 110.9 28 Be <3, MC 0.784 34.9 99.5 +0.030 +0.9 +0 ■ 3 + 1.2 + 1*1 + 1,3 + 1.2 0 - 0.814 35.8 99.8 +0.053 +0,4 +0.6 + 1.0 + 1.9 + 2.1 + 2,0 -i.a 0,867 36,2 100,4 * +0.055 -0.5 + 1.3 +0.8 + 1*9 + 1.9 + 1.9 -i.i 0.922 35.7 101.7 -0.049 +0.8 -1.8 -1.1 -1.7 -1.7 -1.7 -0,7 0,873 36.5 99,9 Means t 3*35 mru'j TABLE 3. Comparison of Observed and Calculated Changes of Alveolar Partial Pressure3 Experiments at Ground Level Table 3- C»nt*d. Columns (10) anc (ll) represent the possible changes in pC* at constant pO", or in pO" at constant pC", corresponding to the changes in Q.W; recorded in column (4), according to Fig, 3, In column (13) the average of these values is compared, by subtraction, with the sum of the observed changes in pO" and pCw# Negative sign before a value in column (13) indicates that the observed change was less than that calculated,, including those cases in which the observed and calculated values changed in opposite directions# (1) (2) (3) (*) (5) (6) (7) (8) (»T~ (io) “(ii) (12) (13) Bi screpancy % - Qbae.rved Calculated from 9** Bate Subject a"* pC" ApC* pO" ApO" £-pC" pO " Average (12)—>(13) X 3 J « HC 0.782 30.6 52.6 +0.115 -1.7 +9.5 *7.8 + 4,1 + 4.5 + 4,3 + 3.5 0.897 28,9 62.1 * +0,019 -1.2 0 -1.2 +0*7 +0,7 +0.7 -1.9 0.916 27.7 62.1 -0 *004 +0*4 +0.3 +0.7 -0.1 —C . 1 —C . 1 -0.8 0.912 28.1 62,4 4 Jan* LC 0.859 28,7 54.1 -0.064 + 1.2 +1.8 + 3,0 -2.3 -2*6 -2.5 -5.5 0.795 29.9 55.9 +0.055 + 1*1 -2.1 -1.0 + 2.0 + 2.3 + 2.2 -3.2 0,850 +0.042 31.0 + 2,4 53.8 -4.7 -2.3 + 1.5 + 1,6 + 1.6 -3.9 0.892 33,4 49.1 12 Jan* RS 0.871 32.1 49,6 -0.006 -0,4 + 5.4 + 5.0 -0.2 -0,2 -0.2 -5.2 0.865 31.7 56.0 +0.088 + 1.6 -2.5 -0.9 + 3.1 + 3,1 + 3.1 -4.0 0.953 33.3 53.5 -0.029 -0.6 -1.0 -1.6 -1,0 -0.9 -1.0 4-0,6 0.924 32.7 54.5 31 Bee* MC 0.866 28.1 55,4 -0.023 + 2.6 -1.9 +0.7 -0,8 -0,3 -0,8 -1.5 0.843 30,7 53.5 +0.047 + 1.7 -2,5 -0.8 + 1.6 + 1,7 +1*7 -2,5 0.890 32.4 51.0 -0.030 -o.e +0.8 0 -1.0 -1*1 -1.1 -1.1 0,860 31.6 51.8 0.382 +0.022 30.6 -1.0 54.6 + 2.8 + 1.8 +0,3 +0.8 +0.8 + 1.0 • Mean: 2,67 -1,88 Explanation! See legend to Table 3. TA3LE 4 Comparison of Observed and Calculated Changes of Alveolar Partial Pres sure s Experiments at 12,000 feet I H M Number of Type of a(po")/dt l d(P0")/dt Experiments Expiration Observed Calculated dQ"/dt Observed Calculated dO"/dt ICrogh and Lindhard, 1914 2 No rmal -0.9 -1.0 —.■*- - Grosse-Brookhoff and Schoedel, 1937 7 Normal -2.2 -1.5 _ Krogh and Lindhard, 1914 1 Beep -2.2 -1,1 — 3.4 —1, 4 —i Nielsen and Sonne, 1932 Sonne, 1934 29 Deep -3.5 -1.4 -0,022 ■ +0.14- .-2,1 +0.007 Roelsen, 1939 a 26 Deep -2.9 -1.3 -0,025 -0.9 -1,7 -0,003 Note* Values of d (pO")/dt are in mm. Hg per second. Calculated values are obtained from average oxygen consumption and volume assuming uniform blood flow throughout the respiratory cycle* of air in lungs. TABLE 5 Rate of Change of Partial Pressure of Oxygen in Alveolar Air During Expirat ion COMPARISON OF RESPIRATORY QUOTIENTS CALCULATED FROM ANALYSES OF ALVEOLAR AND TOTAL EXPIRED AI R HC LC RS MC Alveolar Air V HALDANE TOTAL J.B.Bateman and W.M.Boothby June 1944 Mayo Aero Medical Unit Total Expired Air Chart X- 11 a O GROUND # 12,000 FEET Fig. I TIME COURSE OF CHANGE OF TRUE RESPIRATORY QUOTIENT AND ALVEOLAR RESPIRATORY QUOTIENT AFTER A MEAL OF RICE DATA OBTAINED AT GROUND LEVEL (1,000 FEET) Upper section of each quadrant contains points for true respiratory quotient Q'" • and for alveolar respiratory quotient Q“ 0 Lower sections show ventilation rate in liters per minute (atmospheric pressure, 37° C, 47mm. water vapor), • and oxygen consumption incc. per minute at 760 mm., 0° C, dry Abscissa : time in minutes Zero is time at which meal of rice was finished RICE 517 g. RICE 949 g. RICE 675 g. RICE 619 g. Mayo Aero Medical Unit Chart I-lib June 1944 TIME J. B.Bateman W.M. Boothby Fig. 2 TIME COURSE OF CHANGE OF TRUE RESPIRATORY QUOTIENT AND ALVEOLAR RESPIRATORY QUOTIENT AFTER A MEAL OF RICE DATA OBTAINED AT 12,000 FEET SIMULATED ALTITUDE Upper section of each quadrant contains points for true respiratory quotient, O'", and for alveolar respiratory quotient,Q", Lower sections show ventilation rate in lifers per minute ( ambient pressure, 37° C, 47mm. water vapor), and oxygen consumption in cc. per minute at 760 mm., 0°C, dry Abscissa: Time in minutes. Zero is time at which meal of rice was finished. Dotted line on left of each qradront shows point of ascent to 12,000 feet. 12,000 FEET RICE 552 g. RICE 326 g GROUND GROUND 12,000 FEET 12,000 FEET RICE 720 g. 12,000 FEET RICE 916 g GROUND GROUND Moyo Aero Medical Unit Chart X-lie June 1944 TIME J.B. Bateman WMBoothby Fig. 3 VARIATION OF ALVEOLAR PARTIAL PRESSURES WITH RESPIRATORY QUOTIENT Oz Curves; change in pO" for 0.1 unit change in Q at constant pC" COz Curves: change in pC" for 0.1 unit change in Q at constant pO" Mayo Aero Medical Unit June 1944 Chart I- I I d J.B. Bateman W.M.Boothby COMPARISON OF OBSERVED CHANGES IN PARTIAL PRESSURES WITH THOSE CALCULATED FROM CHANGES IN RESPIRATORY QUOTIENT OCCURRING AFTER A MEAL OF RICE Abscisso: Calculated change, ApC" + ApO" . Ordinate: Measured change, ApC" + Ap0" Units : Millimeters of mercury. 0 Ground level, 1,000 feet. • 12,000 feet. Points representing measured changes smaller than those calculated must all fall within sectors AOB and COD. The pairs of lines parallel to AOD represent the average discrepancy between calculated and measured values. Mayo Aero Medical Unit ChartX-lle June 1944 '®* ® J.B. Bate man W.M.Boothby NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE OPEN Report No. 360 August 1944 THE EFFECTS OF ALTITUDE ANOXIA ON THE RESPIRATORY PROCESSES. From the Mayo Aero Medical Unit, Rochester. Minnesota, by H. F, Hclmholz, Jr*, J, B, Batenan and W, M* Boothby, SUMMARY This paper is a sunnary of sonc results of analysis of alveolar gas fron persons breathing air at simulated altitudes up to 22,000 feet or pure oxygen up to 42,000 feet. The data are anonable to representation in sinplo diagrams which can be made the basis for discussion of several respiratory phenomena. This graphic approach has the advantage that while an alveolar air equation covers a whole scries of physically possible states, the experimental data show which of these arc actually mot with physiologically. At tho sane time they can be shown to be in harmony with the more general physical statement. The p»esentation given here is used to show in simple terms the slight difference in specifications for oxygon supply systems that arise from tho use of different reference points — on the one hand the "tracheal* reference point adopted by the Amy and Navy and on the other the "alveolar" reference point advocated by a number of physiologists, Th® essential part played by the alveolar respiratory quotient in the production of these differences is emphasized. An analysis of the mechanism of the adaptation to mild anoxia is given with reference to the following points* 1, (a) The distinction between the "steady state* of respiratory exchange which occurs under conditions within the limited range of normal adjustment mechanisms, the "semi-steady state* established as a temporary adjustment to respiratory stresses (such as anoxia) which are outside the permanently allowable range, the "non-steady state" of rapid transition. 0>) The time needed to establish a new steady state under mildly anoxic conditions. New data arc presented in illustration of this point. 2, The probable increases in circulation rate which are necessary at various altitudes in order to prevent undue decrease of venous oxygon pressure. The results of an estimate of relative cardiac output at different altitudes, based upon the alveolar air data, arc presented in a new diagram where they arc compared directly with the available experimental determinations of cardiac output of anoxic subjects. The discussion which follows is directed toward the use of present knowledge as a pointer for further experimental work. z NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE OPEN Report No, August 1944 THE EFFECTS OF ALTITUDE ANOXIA ON THE RESPIRATORY PROCESSES, From the Mayo Aero Medic.a‘ Unit, Rochester, Ml nno so by H, F, Eclnholz, Jr,,, J, B , Eatcnan and W0 m, lioothby. In the study of respiration (including circulation) at varying atmospheric pressure, the fundamental changes encountered must necessarily involve differences in the conditions of transfer of gases across the alveolar membrane* In the past years this laboratory has had vd.de experience with alveolar air analysis, and a presentation of data in a compact form and to serve as some sort of exposition of the implications scorns in order. In addition, many gaps in our knowledge can bo emphasised by such a recapitulation. The Haldano-Priostley method for collecting alveolar air samples as used at the Mayo Aero Medical Unit requires collection of the last fraction of a complete, vigorous exhalation started at the normal beginning of exhalation. This fraction is analysed in a modified Haldane apparatus, by trained technicians. The results are reported as partial pressures in millimeters of mercury calculated from the baro- metric pressure (less 47 nm, for water vapor) at the time the sample was taken. These samples have been shown, by comparisons with direct (Van Slyke) and indirect (oximeter) methods, to have partial pressures"which correspond to those in the blood when the means of many observations arc considered (16), (18), (8); (17), (15/.? (5),? (9), (:io) and (22)* This as the best available evidence that such samples indicate the actual composition of the gas mixture in contact, through the alveolar membrane, with the circulating blood, Discrepancies which have boon found at sea level atmospheric pressures between alveolar oxygen pressures and hemoglobin saturations calculated from the oxygen content cf arterial blood have been analyzed in a recent paper by Houghton, Darling and Root, who attribute them to experimental error. wTraoheal” air^forcelation. The continued study of alveolar air is primarily of interest to the physiologist, From the practical point of view, especially for purposes of designing oxygen equipment., it has been decided by the Army and the Navy to consider only the partial pressures in the ”tracheal" air, i,o, the inhaled gas mixture saturated with water vapor at 37° C* Thus, in oxygen regulator design, the specifications call for provision of oxycn«>a.V” mixtures which will maintain a given oxygen partial pressure in the "tracheal* air with changing barometric pressure* This has proved to bo a aatisf standard* In ao sense is the continued presentation of alveolar air data a suggestion that - latter shrnld. hr considered in the design cf ordinary oxygen equipment* As Bateman reiterated in Report 222 to the Committee on Aviation Medicine on "Tracheal" versus "Alveolar" Air, the use of the "tracheal" air formulation in calculating equivalent altitudes will on certain theoretical assumptions require the use of slightly more oxygen than that needed if the alveolar air is the reference point, especially at higher altitudes. However, it is well to remember that this amount is insignificant in relation to the variation inherent in any hitherto designed automatic mixing devices in which variation in demand is put upon the oxygon delivery valve. Moreover, different methods of using the alveolar air formulation give discordant estimates of equivalent altitude under conditions involving mild anoxia, especially when the phenomena of acclimatization to high ground elevations arc considered. Those differences, undoubtedly of scientific interest, would make the picture unnecessarily complicated as a basis for engineering recommendations. It is also known, as Bateman remarks, that the variation between individual alveolar air values forming the basic data is greater than the difference between equivalent altitudes derived by the two methods. These facts all clearly vindicate the official adherence to a "tracheal" air formulation* Alveolar air formulation. By definition, the tracheal air formulation makes no reference to the gas exchange in the lungs. It therefore fails to describe certain effects arising from the fact that during this process the amount of carbon dioxide received by the alveolar gases is seldom equal to the amount of oxygen rof» moved in a given time. Such effects arc implicit in any good formulation of the composition of alveolar air and its dependence upon the alveolar respiratory quotient (4), (10a), (23), (12), (7) but it has sometimes boon difficult for those concerned with aviation physiology to visualize the effects implied algebraically in these formulations. Because of their essential role as a basis for the physiology of respiration, repeated attempts have been made to show graphically how experimental data on the alveolar gases conform to the predictions of the formulae. Such data were presented in 1940 by Boothby, Lovelace and Benson for subjects breathing air, oxygon-air mixtures, and pure oxygen at different simulated altitudes up to 40,000 foot. During the next throe years, many additional observations wore made, and it was shown that other methods of collecting alveolar air gave essentially the same values as that of Haldane and Priestley, All observations on subjects breathing air at simulated altitudes from the ground (l,000 feet) to 20,000 feet were presented on charts submitted to the Committee on Aviation Medicine by Boothby and included in the Handbook (22) published under the auspices of the Subcommittee on Oxygen and Anoxia, The alveolar air equation in its several equivalent forms represents a relationship which must bo valid within the limits of validity of the several assumptions tacitly involved (3) if gas mixtures containing oxygen, carbon dioxide and other inert gases supply the oxygen needed and remove the carbon dioxide produced during any period of bodily function. The assumptions are on the whole iustified provided the body is in a steady or semi-steady state •• by which we mean for the present only that the respiratory cycle (in which wo include the circulating blood) shall maintain its identity over a number of respirations. Any sensible drift will upset the relationships established by the more or loss exact repetition of a cycle and will render the equation invalid. Apart from this limitation, the experimental Study of the remaining assumptions, implicit in some early observations tf J, S, Haldane (13) and others, has boon resumed by Bateman and Boothby (3), whose data justify Brink*s use of a combustion quotient when using the equation in calculations of average equivalent altitudes for large groups of people. Such a step implies that the relationships of the equation apply equally to inspired, alveolar, and total expired air, or to any mixture of these. In attempting to give a verbal explanation and to correlate the graphic with the algebraic statements, we shall present a simplified equation relating only to alveolar and oar£on-dioxide-free inspired air* Inspired Air rk* w°f> Alveolar Air 1 ~ -^co2 "—X .% 02 + +47,, ,,,,,(2) Where I and A represent inspired and alveolar air respectively and pN2, PO2 and PCO2 represent the partial pressure of nitrogen, oxygen and carbon diexide respec- tively and 47 mm, the average pressure of water at body temperature of 37* C, The essential process occurring in the alvesli is the removal of oxygen from the alveelar gas and the simultaneous addition of carbon dioxide. If the body is in a steady state, the ratio of the number of molecules of carbon dioxide produced to tho numbejf ff molecules of oxygen absorbed in a time covering any whole number of identical respirations will correctly represent tho ratio of the rates of turnover of these substances resulting from the oxidative processes of the body. In other w»rds, the alveolar respiratory quotient will bo identical with the combus- tion quotient. When the body is in a soml-steady state, this is not tho case, because wo have superimposed upon tho strictly metabolic exchanges between blood and tissues an abnormal pulmonary exchange of carbon dioxide, usually the consequence of some reflex or voluntary change in the character of the breathing. This abnormality must of course be reflected in the exchange between blood and tissues and tissue fluids, including the production of uuine, and must ultimately react in a deleterious manner upon tissue metabolism and the mechanism for the maintenance of a constant internal environment. Strictly speaking, whenever tho gas exchange in the lungs is disturbed in such a manner as to introduce a discrepancy between the apparent respiratory quotient and the true combustion quotient, it is impossible for any one breath to be precisely identical with its predecessor; tho fact that wc can sometimes speak of virtual identity and a "semi-steady state" chiefly arises from the circum- stance that the buffering power of the tissues permits a relatively large amount of carbon dioxide t» be removed without producing a significant change in pH, Thus it is often possible to establish, and to characterize by moans of apparent respira- tory quotients, scmi-stcady states which persist long enough to be of considerable importance in some of the emergency situations encountered in high altitude flying, Tho transition period between a steady state and a semi-steady state or a new steady state is one of rapid change, often of very brief duration, in which alveolar air analyses are difficult to reproduce and impossible to interpret reliably. During this period ff rapid change the subject is in a non-steady state. It i 8 this potentially variable relationship betvroon the oxygon and carbon dioxide transfer across the alveolar membrane that has caused some confusion in the comparison of the effects of breathing air and oxygen under various conditions met with in aviation. When pure oxygen la being breathed, there is no particular the number of carbon dioxide molecules produced Is smaller than the number of oxygon nolooules absorbed, more pure oxygen ■'.rill novo in to maintain the total pressure equal to that of the atmoxohere, and it will be impossible to dis- tinguish by means of alveolar gas analyses, between the oxygen which was originally present and that which has moved in to make up for a deficit of carbon dioxide* In other words, wo can learn nothing about the alveolar respiratory quotient from measurements of this type* Now when air is being breathed the situation is quite different* In-this case a deficit in the total number of oxygon and carbon dioxide molecules (H.-i. less than unity) causes an increase in the fraction of nitrogen pre- sent; the air of the atmosphere, acting as aipiston, will compress this mixture de- ficient in molecules as a result of pxilncnary exchange, to maintain atmospheric pressure* The result must bo an increase in the partial pressure of nitrogen in the alveoli — an increase that will depend upon the respiratory quotient in the manner indicated in equation (l). It is this change in nitrogen pressure that has caused some of the confusion; but to the sane phenomenon wo owe the possibility of neasur* ing respiratory quotients by analyses of e pired air or alveolar air. J*S.Haldane realized this many years ago. The nitrogen acts as an inert indicator, so tc speak, of changes ih the total number of "active" molecules In the luhg. -hen the respir- atory quotient is less than unity, we can detect the fact by an increase in the partial pressure of nitrogen; when, in a scmi-stcady state, it is greater than unity, the partial pressure of nitrogen decreases, and the alveolar gas contains relatively m more oxygen than it did before. The same effect is responsible for some of the differences between equivalent altitudes calculated according to the "tracheal" and "alveolar" reference points. The Increase in nitrogen pressure depends upon the fraction of nitrogen in the in • snired air; the more oxygen and the less nitrogen there is, the less significant is the effect of the respiratory quotient in increasing or decreasing the partial pressure of oxygen, and the closer the parallelism between the "tracheal" and the alveolar oxygen pressures. Another consequence Is that the temporary establishment of a higher alveolar oxygen pressure by hyperventilation, ©onetimes a desirable thing for anoxic persons, involves slightly different mechanisms according to the presence or nbionot of nitrogen in the inspired gas. ■hen pure oxygon is being breathed, the effect of hyperventilation is solely that of increasing the partial pressure of oxygen at the expense of carbon dioxide, ’./hen air is being breathed, we have in addition the effect of an increased apparent respiratory quotient in causing an increase in oxygon pressure at the expense of nitrogen. These are some of the phenomena that can only be discussed tfhen wo have an alveolar air equation or some good way of representing graphically the consequences of gas exchange in the lu gs, the fundamental and inescapable basis of which is the inequality of the rates of transfer of oxygen p.nd carbon dioxide across the alveolar membrane* Since this gas exchange is not mentioned cr even implied in the definition of tracheal air, a tracheal air formulation cannot tell us anything about its con- sequences. On the other hand, the effects involved are us«s iiy smaller than the variation between individuals, and are for the present of scientific rather than practical interest. Charts illustrating the composition of alveolar air at different altitudes In chart I the partial pressures of gases are plotted together in such a manner that in every case the relationship (2) is indicated. Thus the tracheal nitrogen pressure plus 47 plus tracheal oxygen pressure add together to give an ordinate equal to the atmospheric pressure, In the sane way, the experimental averages obtained from analysis of alveolar air at the corresponding atmospheric pressures are plotted. From those curves nay bo soon the effect of the alveolar exchange in increasing the pressure of nitrogen in the alveolar air and decreasing the pressure of'oxygen at moderate altitudes, when anoxia is not a factor. At higher altitudes a semi-stoady state supervenes; the nitrogen pressure change tends first to disappear and is ultimately reversed when the stimulus of anoxia produces a sufficient increase in ventilation for carbon dioxide to be added to the alveolar air in excess'of the oxygen absorbed. Charts 2 and 3 relate to the tine it takes to establish a now steady state and its constancy when once established. Chart 2 provides a control, representing the behavior of the alveolar gases at an altitude - 10,000 feet - too low for the increased ventilation associated with anoxia to become important. The data given in Chart 3 wore obtained for a similar group of subjects in a stay of 90 minutes at 15,000 foot. One notes the initial doorca.se of the nitrogen pressure to the tracheal value, a sign of increased loss of carbon dioxide, and the gradual re-establishment of a steady •tato resembling that found in the preliminary measurements at ground level. With increasing anoxic drive, it would require a longer time to reach a new steady state and the measurements normally made arc likely to refer to a semi-stoady state tending toward a final steady state which may, however, be impossible to attain and incompatible with furtinli This is already the case at 15,000 foot for most of the individuals upon whom the observations averaged in Chart 1 were made, as the rising value of tho alveolar respiratory quotient clearly indicates. Wo may draw from tho literature an extreme example of the effect described (ll). The alveolar air of Douglas immediately following a period of hard work on Pike’s Peak, where his alveolar oxygen in a control observation had been found to bo 55 mm,, had a partial pressure of oxygen of 67 mm, (sec Table l), returning in 27 minutes to 56 mm. According to Chart 1, with an atmospheric pressure of 460 mm,, this must have involved an alveolar nitrogen change from the control value, 377 mm,, to 368 mm. The apparent respiratory quotient at this time had become 1,35 compared to a control value of 0,86, This illustrates how the nitrogen effect predicted by the alveolar air equation contributed to tho high alveolar partial pressure of oxygen observed at tho moment of greatest respiratory stress. In Chart 4 the results of alveolar air studies made on subjects breathing essentially pure oxygen at pressures equivalent to altitudes up to 42,000 feet arc recorded in addition to those on Chart 1, As wo mentioned above, the only effect of increased ventilation is a fall in the partial pressure of carbon dioxide which is directly reflected in an increase in tho partial pressure of oxygen, A further use for this chart is in showing graphically the effect these changes in oxygon pressure during alveolar exchange have upon the idea of tho equivalence of certain pairs of altitudes when the atmosphere consists of air or of pure oxygen. To begin with, let us see hew the tracheal reference point can be used by inspection of this chart. Any point on tho one tracheal oxygen curve can of course be compared with the corresponding point on tho other, and perpendiculars from these points will give equivalent altitudes. For example, points A3, B3, or C3 may be compared with points A2, B2, or C2 respectively. It can bo soon, however, by inspecting tho points on tho alveolar oxygon curves, that somewhat different pairs of equivalents can be obtained on the basis of equality of alveolar oxygen pressure. Points A4, B4, and C4 do indeed represent tho altitudes equivalent to Al, Bl, and Cl if tho respiratory quotient is unity, which it might very well be if the subject had been eating candy or a moal rich in carbohydrate; but if it is much loss than this, the equivalent altitudes will be represented by points A5, B5 and C5, Considering now the A points alone, the ourres indicate that there may be a difference of 1,100 feet between the results ob- tained by use of the two reference points* The point A6 indicates the alveolar oxygen pressure to be expected if the tracheal air is used as a reference point* We may now reverse the procedure just described in order to illustrate how those con- siderations may be used in determining the altitude at which a nitrogen-oxygen mixture of given composition will keep a flyer in a physiological state equivalent to that which would be expected if he were breathing air at a specified lower altitude. Let us choose 4,300 feet as our standard altitude, at which the alveolar oxygen will be given by point A6 if the R.Q., is less than 1, and the tracheal oxygen by point A3. Further let us suppose, for the sake of argument, that the curves on the right refer not to pure oxygen but to a particular mixture very rich in oxygen, and that we wish to determine how high an aviator can fly on this mixture in order to remain "physiologically" at 4,300 feet. The "tracheal" oxygen criterion gives us 35,700 feet as our desired equivalent; the alveolar criterion gives about 36,300 feet. Thus the use of the latter tends to out the oxygen supply recommended for fulfill- ing a particular requirement in high altitude flying; the tracheal oxygen criterion "plays safe," But as we have already remarked, variation in individual requirements is really of greater importance than such relatively small theoretical differences as those considered here. In concluding this section let us express our awareness that no known standard can give us tables of equivalent altitudes that are strictly valid from a physiological point of view. In the graphic construction just outlined, we define equivalence on the basis of equal alveolar partial pressures of oxygen; but clearly we ought also to include the carbon dioxide, the blood flow, and probably innumerable other factors. The calculations given by Brink and others do define equivalent altitudes as those at which both oxygen and carbon dioxide partial pressures are equal; and this does indeed provide us with calculated tables of equivalents that satisfy the theoretical conditions imposed. xamination of experimental data shows clearly, however, that at those supposedly equivalent altitudes the alveolar carbon dioxide pressures rarely are in fact equal, so that our theoretical tables confront us with a physically possible situation which does not however appear to correspond to any actual physiological one. The reason for this is not easy to find, but further research will perhaps give us the explanation. In any event, we are in- clined to believe that application of the idea of equivalent altitudes should be limited to the attempt to give an aviator expecting to fly at very high altitudes some idea of the conditions he will encounter in terms of conditions already familiar to him. Attempts to refine, beyond this point, an essentially obscure idea seem hardly worth while. Anoxia and cardiac output. From the results of alveolar gas analysis, applying the premiss (already discussed in relation to experimental evidence on page l) that the partial pressures are reflected nearly identically in the blood assing through the lungs, we may draw certain conclusions concerning the venous oxygen pressures that are to be expected for various degrees of anoxia and various types of circulatory adjustment. For the purposes of calcmlation we need to take into account the following factors, and have done so in the manner indicated in each case: (l) The normal oxygen dissociation curre of human blood and its rariation with carbon dioxide pressure. This has been done by assuming an oxygen capacity of 20 volumes per cent and the dissociation curve given in L,J#Henderson*s nomogram for the blood of A.T.B. (Fig* 157,ref. 14) ■a (2) A normal oxygon consumption, which must be maintained undef* all circumstancesf This has been assumed equivalent to the loss of 6 volumes per cont oxygen, or 30 percentage points oxygon saturation, by blood circulating at a normal rate0 (3) A relationship between arterial and venous oxygen pressures implied by (a) loss of 6 volumes per cont oxygen for normal blood flow or 6/N volumes por cent when the blood flow is N times normal* (b) uptake by the circulating blood of Q. times as much carbon dlorJdo as the oxygen lost, where 0. i s the alveolar respiratory quotient. This has been done by introducing into Henderson’s nomogram respiratory quotient linos in such a manner that a straight edge pivoted about a given point on one such line traverses the "total CO?* scale and the r'Eb02t? scale at rates whose ratio is 0„ Those lines are somewhat similar to the respiratory quotient linos in Henderson's Fig, 41* Using these assumptions wo have made calculations applicable to two extreme hypothetical conditions,, First we have supposed the cardiac output to remain unchanged under conditions cf anoxia# In this case, tbs venous oxygra pressure would havo to follow approximately the course indicated by the solid lino in Chart c# V/c noto that at about 22,0(;0 foot- it falls as low as 20 mm- Hg, a value that we are inclined to consider rather improbable and highly arranging* It is true that oxygen tensions as low as 12 ran, have boon recorded in venous blood during heavy work >19) but Bainbridge (l) seta the lowest value at 20 nm* Since wo are hero dealing with on average that must apply to the blood supply of the nervous system, which will probably not tolerate as great a fall of venous oxygen pressure as other tissues, it is fairly safe to assume that at some point during the progressive removal of oxygen from the inspired air circulatory compensation occurs* For purposes cf calculation wo have assumed this to take place whenever the venous oxygon pressure falls below a certain critical value* If the critical level is set pore or leas arbitrarily at 28 mm* wa obtain the upper relative cardiac output curve shown in Chart 5, while the lower curve applies to a critical level of 26 mm. Such estimates are necessarily subject to Coraiderable error, but it is of interest that the soansy existing measurements of cardiac output at simulated altitudes (21), (20) fall within the range covered by these two estimates# Tho general but very rough plausibility of the picture may serve to indicate the need for well organized experiments which will clarify tho whole situation and obviate tho use of pu p h devices as Henderson's nomogram under somewhat extreme conditions to which they do not apply with any particular accuracy* The time has cone when the most urgent need is the more extended study of the character of physiological adaptation to anoxia, an end result of which is so clearly indicated by changes in the composition of alveolar air. TABLE 1 Changes in c o v.p o ? j1 j.j-> n of. a. d_v jfoJJj?y5 nr? nu?cnlar •vrork at, high &A$k.V&ht ( *rjTn .Oor.glun , Haidano., Hjndarson And Sevan e i d e v, Phil,- i’rana•, of Hcyal Sooiwty of London,, Sorics B, ToX0 203, pp, 105--318* Tahlo VII, p, 222it 1913). Measureneats on Douglas on top of P:*,kjps Peak. Colorado, Work consisted of naxinal exertion for 45 seconds on 25$ grade, £aroma trio prossv.ro 460 nn® co2 .**!?'■ Ml °2 [ nnn Eg] Alveolar i a. lo Nornal none inspiration and expiration 28 53 Or 36 2. Inncd:.atoly after ctopping cxcroiso 25 o7 le 35 3, 10 Dilutes after stopping 24 59 fuS2 4, IQ ninuton after stopping 2 3 58 oca9 5, 27 nr, nut -> s after stopping 25 50 0o78 6 o 38 mini.tos after stopping 23 53 Or 7n 7 » GO nluv.toe after etopping 28 53 0 079 REFERENCES (1) Balnbrldge, F.A., Book, A.V, and Dill, MB. The phy*« ciogy of wuscula- ©xcroite, Hovr York, 1931, pagff 141. (?) B*t«man, J.B, O.S.H.t), Committee on AvidHon liodioi ac , Report No. 225;, Doc, 1943, , 3 > Bateman, J.B, and Bcothby, W.M, O.S.R.J), Committee on Aviation Me,dicing* Report forthcoming, 19^4, (t) Rer-tfi nger, 7, Ergebn, Physiol, 40, 1 (1938), (5) Bock, A.V, and Field, H, J, Biol* Cham,, 62, £69, (l92 l), (o' Bcotboy, W.M, and Robinson, F.J, O.S.R.D, Committee on Aviation Ked.lc5.ne, Roper* Ho. 103, Juno 1943* (7) Brink, F. Essay A. in ref. (22), (1944), (3) Campbell, and Pculton, £,P. J, Physiol,, 54, xlU, (19??:, {9} Mil, FurJtthal, L.M, Van Canlaert, C,, Fool ling. A, an* Book, A.V* J. Biol, Chon,, 74j> 303 (1927), (10- Dill, D,E., Lawrence, J.S,, Hurxthal, L0M, and Book, AfV. J, Biol, Cnotn,, 74, ?13 (1927), (30c/ Dili, D.B, Vrai' Department Report Nc, Exp, M 653~iC3f* (11) Douglas, C,G,, Haldane, J.S,, Kendorson, V, and Schneider, F.C, Phil, Trans, Roy, Soo, London, 185,.2036 (1913), (12) Gray, J,S, School of Aviation Medicine, Randolph Field, Texas* Hecearoh Report No, 1, April 12, 1944. (13) Haldane, J.S, and Priestley, J.G, Respiration, Hy* Haven, 1935. (14) Eondoraon, L.J, Blood. 1 A study.In general physiology, New Karor*, 1928, v \ (15) Henderson, L.J., Book, A.V., Field, H, and Stoddard, J.L. J, Biol, Choa,, 59, 379 (1924). (16) Krogh, A, and ICrogh, M, Slcand, Arch., 23, 179 (1910)* (17) Konkins, J*C,, Dautrobando, L. and Potter, W,J, Heart, 10, 153 (1923), (IS Peters, J,P,, Barr, D.P* and Rule, F.D, J, Biol, Chom*, 45, >89 (1921), {19) Schneider, E,0. Physiology of muscular activity* Philaielpbia, 1941, page 156, (20) Starr, I, and MoMlchael, M, 0,S,ReD. Committee on Aviation Medicine, Report No e, 165? August 1, 1943c (21) Staelo? J0M,, OtS*R„D, Committee on Aviation Medicine, Monthly Progress Report No., 5 on Contract No„ OEMctr.r-120, September Gy 1942, (22) Euboommittea on Oxygen and Anoxia* Handbook of respiratory data in aviation* Washington, D..C,, 1944, (23) Wildhaok, W,A0 J„ Aeronautical Sci0, 543 (1942)« MAYO AERO MEDICAL UNIT ATMOSPHERIC TRIANGLE SCALES Brgqthing Omen EXPLANATION OF UPPER SERIES OF SCALES X Altitude 1,000 feet 31 Barometric Pressure 33T Partial Pressure of Nitrogen in Dry Air; (B x 0.7903) 32 Partial Pressure of Nitrogen in " Tracheal" Air after Saturation with 47 mm. Water Vapor, Body Temperature 37°C (BT PS) • (B-47) 0.7903 3C Partial Pressure of Nitrogen in Alveolar Air: (B-47)AfNz TZT Partial Pressure of Carbon Dioxide in Alveolar Air: (B—47) AfCOz All Pressures in mm. Hg. Air Saturated with Water Vapor 37* C Dry Dry PARTIAL PRESSURE OF GASES The Partial Pressure of Alveolar Gases must equal Barometric Pressure Unless A.R.Q. is Unity there is a change in the Lung of the Volume of the Respiratory Gases As many Molecules Nitrogen ore exhaled as are inhaled Therefore a change in Nitrogen Pressure can be used to calculate the change in Volume and to obtain the A.R.Q p ‘partial pressure of gas ' J At.“•‘Atmospheric" air dry T •" Tracheal" air (BTPS) A 'Alveolar air (BTPS) NITROGEN OXYGEN BREATHING AIR BRE ATM! NG OXYGEN ALVEOLAR RESPIRATORY QUOTIENT Nb Nitrogen available so A R.Q cannot be calculated Bar. Press. Alt. 1,000 Ft. Chart X- 6d- e J.B.Bateman and W.M.BocWiby Aug. 1944 Fig. 1 a. Mayo Aero Medical Unit EFFECT OF ANOXIA ON ALVEOLAR AIR PRESSURES AVIATOR BREATHING AIR AT VARIOUS ALTITUDES The pressures of 02, CO2 and N2 in the inspired tracheal air are characteristically altered during the respiratory cycle. In the "Steady State “these respiratory changes are based upon the character of food eaten which alters only the partial pressures but also the total volume of the alveolar air from the inspired air. Altitude anoxia, dependent upon its intensity and duration, superimposes in the " Semi - Steady State" definite additional changes in the alveolar nitrogen, oxygen and carbon dioxide pressures and consequently upon the various ratios or quotients that can be calculated therefrom. The alveolar oxygen pressure can be calculated by the following formula; pO" = pO' — [pC" + ( pN" - p N ' jj One prime' • Tracheal air;Tvto primes” 'Alveolar air Curve*based on the averages obtained from over 1,000 HALDANE - PRIESTLEY alveolar air determinations most of which were obtained on 10 males and 5 females : See chart X-6 b TRACHEAL AIR (BTPS) DRY ( AIR -TRACHEAL NITROGEN -ALVEOLAR NITROGEN WATER VAPOR PARTIAL PRESSURE -TRACHEAL OXYGEN AVAILABLE OXYGEN ALVEOLAR C02 -ALVEOLAR OXYGEN I Alveolar Respiratory Quotient: ARQ 'Alveolar Pressure Ratio; APR - Combustion Quot ient: RQ BAROMETRIC PRESSURE mm. Hg. Chart X- 6 dc ALTITUDE - THOUSANDS OF FEET Boothby, Helmholz and Bateman June 1944 W.M.Boothby and Fred Helmholz July 1943 .Mayo Aero Medical Unit Alveolar O2 and CO2 Pressures and Alveolar Pressure Ratios as affected by Duration of Stay at 10,000 feet Five subjects were taken to 10,000 feet without oxygen. Alveolar airs were obtained at intervals up to 120 minutes Time in Minutes after Reaching 10,000 feet Chart X— 10 b Mayo Aero Medico/ Unit Alveolar Og and CO2 Pressures and Alveolar Pressure Ratios as affected by Duration of Stay at 15,000 feet Six subjects went to 15,000 feet without Oxygen. Alveolar airs were obtained at intervals vp to 90 minutes W.M. Boothby J. B. Bateman Chart X -10 c June 1944 Time in Minutes after Reaching 15,000 feet MAYO AERO'MEDICAL UNIT Alveolar Da and C02 Pressures and Alveolar Pressure Ratios as affected by Duration of Stay at 15,000 feet Five subject were taken to 15,000 feet on “normal " oxygen, about 10 minutes at altitude mask was removed and alveolar airs obtained at intervals up to 90 minutes W.M. Booth by Mar.1942 Chart! - 10a Time in Minutes after Reaching 15,000 feet W M Boolhby Apr 1944 As the inspired tracheal air contains only oxygen and water vopor any decrease in volume of the alveolar air cannot decrease the alveolar oxygen pressure. The APR is always 1.0. The alveolar oxygen pressure can be determined directly by subtracting the alveolar CO* pressure from the tracheal oxygen pressure. PARTIAL PRESSURE OF INSPIRED, TRACHEAL AND ALVEOLAR AIR ALTITUDE C0f 0, Nt HfcO Vop. feet mm mm mm. mm mm A ! Chamber True 35,000 35,71 6 178.9 172 9 37.0 88 7 6 0 47.0 B ! Chamber True 40,000 40,580 140 7 136 9 339 56 0 3.8 470 C ! Chamber True 42,000 42,652 12 7 9 124 0 31 3 45 7 3.S 47.0 rif :n I I s ill ils Si. o c - M i . •s | s. s I I = - e « i»J: - 5 - III I MAYO AERO MEDICAL UNIT COMPARISON BETWEEN LOW ALTITUDES BREATHING AIR AND HIGH ALTITUDES BREATHING OXYGEN based on over 1400 determinations of the atveolar ofr by the Holdone - Priestley method on subjects occlimatiied to ground level of OOO feet AVIATOR BREATHING OXYGEN Holdang- Priestley Alveolar Airs al High Altitudes ALTITUDE- THOUSANDS OF FEET Al « Eipenmental alveolar 0*on o*ygen ot 35,716 feet A2 • Inured tracheal 0* on oiygen A3 • Inspired tracheal Ot on air equal to A 2 A4 * Alveolar Ozon air equal to Al assuming APR *1.0 A5 » Alveolar Ot on air equal to AI but veth actual eipenmental APR of 0.89 Illustrating the effect of Na of air In louring the ceiling whenever the APR is less than i 0 A6 * Alveolar Ot on air is 5mm lower than A4 due to increase of Na in alveolar air when breathing air, when APR »O.09 for the some altitude os A 3 A2 to Al * Alveolar C0») £»*ent»olly identical ot equivalent altitude* A3 to A4 > Alveolar CO|j A3 to A6 ‘Alveolar C0*+ increase in alveolar N2 A4 to AS ‘Loss in altitude due to increase in alveolar N2 A4, A5, A6 ; the trionglulor area indicates the probable error due to variations in APR when the onoiio of the aviator ot high altitude breathing oxygen is com- pared to a comparable degree of anoxia at altitudes breathing air Alv. pO« » Troch pOi — (aIv. pCOt — (Alv. pN* - Troch pNt)} „ p0‘* pO' -(pC — (pN‘ - pN' j) «**r« o»t prim» ' • Irotkmol air ood hro prim** " • otmotor Olr As the inspired tracheal air contains nitrogen in addition to oxygen and water vapor any decrease in volume of the alveolar air increases the pressure of nitrogen because there is no net change in the number of nitrogen molecules. Thus,there is less available oxygen and, there- fore. a decrease in the alveolar oxygen pressure In the absence of hyperventilation the APR is decreased. Both the alveolar C0« and the increase in nitrogen pressure must be sub- tracted from the tracheal oxygen to obtain the alveolar oxygen pressure. ILLUSTRATED BY THE A SERIES COMPARISON AVIATOR BREATHING AIR Low Altitude Breathing Air See Chart I-6b for complete data Ground 1 evel»1,000 ft.» 733 mm. Inspired Trach.Air Alveolar oir(H-P) COg * 0.2 mm COs 1 36 7 mm Os ■ 143.6mm 0* • 102 3mm Ns • 542.1 mm Ns = 547 0mm HsO ■ 47 .0 mm HsO • 47 0 mm Arn - PCO, 38 7 0 89 tr»*p troch pOj— Alv.pOt 143 6 — 102 3 Cbort I-6-d-b PARTIAL PRESSURE OF INSPIRED. TRACHEAL AND ALVEOLAR AIR MAYO AERO MEDICAL UNIT INCREASED CIRCULATION RATE WITH ANOXIA ALVEOLAR AIR PRESSURE from M.AM.U. Chart X- 6b Millimeter mercury Percent Saturation ARTERIAL HEMOGLOBIN SATURATION VENOUS OXYGEN PRESSURE Millimeters mercury COMPENSATORY INCREASE CIRCULATION RATE Relative increase ALVEOLAR RESPIRATORY QUOTIENT I Alt. 1,000'sft.lO Chart XXI-7 Helmholz, Bateman and Booth by Aug 1944 NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development j COMMITTEE ON AVIATION MEDICINE Report No, 3fi4 September 1944 SUSCEPTIBILITY TO DECOMPRESSION SICKNESS* THE EFFECTS OF PROLONGED INHALATION OF CERTAIN NITROGEN-OXYGEN MIXTURES COMPARED WITH THOSE OF EXPOSURE TO PURE OXYGEN, From the Mayo Aero Medical Unit, Rochester, Minnesota , by J,B,Bateman, Responsible Investigator: Walter M, Boothby, SUMMARY The res,*, ujes i ' four experimental subjects to partial nitrogen elimination produced by prolonged inhalation of certain nitrogen-oxygen mixtures have been ooupared with tb effe bs of brief exposures to pure oxygen. The responses were measured by estimating nho susceptibilities of the subjects to decompression sickness during a 9° minute flig>t, with physical exercise, at 35,Of feet. The experiments Lad a practical and a re n tadenio aspects Problem (1): A person Xlies at 15,0(i() feet for about ix hoars, breathing air sufficiently enriched with oxygen to mak his Iveolar oxygen pressur- about qual to its usual value (a) at sea level and (b) at 5,0Cb feet. How much protection from "bend:" will these procedures afford if the s bjeot then ascends to 35,COO feet? Rult a s Inhalation of 40 per cent oxygen for six hours at 15,000 feet (sea level standard) protected three of the four subjects almost completely; in the most susceptible subject the amount of protect ion was hardly significant* In all cases the amount of protection was about the sane as that afforded by pre-oxygenation at ground for 100 minutes. Thirty per cent oxygen (5,000 feet standard) was somewhat less effective. Thus Individual differences are still apparent after six hours desaturation, so that the selection of personnel for mission? of the type simulated in the*e experiments should include measurement of the response to pre-oxygenation, It is also recommended that oxyge n regulators be set at a sen level rath* r than a 5,000 foot standard. Pro hi on (21 t Are Individual differences in response to pre-oxygenation duo pri- marily to differences In the rate of elimination of an "effective" or symptom-producing nitrogen fraction, or are they due to the operation of other factors which facilitate the production of symptoms In some individuals at Itwer nitrogen levels than in others? 2 Re suits t (i) "Calibration* by inhaling pure oxygen* the well-known effects wore shown. Three subjects were completely protected after inhd. ing oxygen for 110 ninutes; the fourth required more thin six hours. (li) "Equillbration" with a given partial pressure »f nitrogen required at least six hours for three subjects and at least twelve hours for the fourth. (iil) The Blew removal of a given fraction of the total dissolved nitrogen prsbably confers equal protection upon all subjects, regardless sf differences in their response to pure oxygen. Thus removal by true equilibration of ab#ut fO per cent provides nearly complete immunity* removal of 50 per ce t gives considerable protection; removal Sf only 30 per cent gives slight but measurable prelection. (iv) The threshold for production of symptoms at 35,iOO feet appears t be about 230 mm, Hg for dissolved nitrogen, Th theoretier‘j t . ,’jshol for growth of a bubble at thi altitude is about t 0 mm. (v) By comparing the "equilibration* data with the curves f.-“ protection by brief exposure to pure oxyg* n, it is possible to construe* elimination curves for the symptom-producing nitrogen with- out havig recourse to conventional nitrogen elimination me surements. The hall ’ess uraticn times rbtai ed in thii mner were c 100 minutes, r me re, for three subjec s, ar 1 260 ilnutt.s, or re, for the fourth. The former value is not gr itly fforant fr . the accepted overall values "nr nitr gen elimination, the i tter Is much greater. (vi) The data, although ;xtre rely United, ;hus appear to vindicate the hypothesis of a slowly el minated nitrogen racHon as the cause f symptoms of decompression sickness. NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH af the Office of Scientific Research and Development Ok COMMITTEE ON AVIATION MEDICINE Report No, 364 September 1944 SUSCEPTIBILITY TO DECOMPRESSION SICKNESSt THE EFFECTS OF PROLONGED INHALATION OF CERTAIN NITROGEN-OXYGEN MIXTURES COMPARED WITH THOSE OF EXPOSURE TO PURE OXYGEN. By J, B* Bateman from the Mayo Aero Medical Unit, Rochester, Minnesota, Responsible Investigator! Walter M, Boothby, 1. INTRODUCTION. The "well-esta'lished effect of oxygen in reducing or even abolishing the hazard of decompression sickness upon subsequent decompression provides the strongest available evidence for he supposition that the symptoms of this condition are closely correlated with the presence in the body of gas bubbles which owe heir ability to Increase in .ize, if not their actual existence, *-o the ©stab .ment of a physically definable . relationship between the uuncen* ratio of dissolv i nitrogen in the body and the prevailing barometric pressure. The det led picture s however far more obscure than this simple view would imply, and one of the sources of diffi- culty lie: in the existence of individual differences in the iegree of immunity developed by brief exposure to oxygen - differences fa* grea- r than might be expoc ec fmm th rather scanty available record, of the rates at which issolved nitrogen s eliminated during the breathing of oxygen. For this and oth,.r reasons it has been suggested that although the overall rates of nitrogen elimination do not exhlbi’ the required range of values, there may exist in the body slowly desaturating leer-1 1 H; for which time constants are not open to estimation by any existing dime’: method, however, our knowledge of the factors which influence the production and growth of bubbles, both in vitro and in the animal body, is sufficiently advanced to suggest that the hypothesis of slowly desaturating regions may be superfluous* Thjf ; s'ibility is amenable to direct experimental test, for if there are slowly desaturating regions, the response to elimination of a given fraction of the total t ly nltr.-gen should depend upon the time in which this is accomplished, the distribu- te n ■ - remaining nitrogs-n tending to become more uniform in course of time* It would follow that elimination of a given fraction of body nitrogen by brief expose to pv- oxygen "hould be less effective in preventing the symptoms of dacomprasalon si n .3 ‘ban elimination of the same amount of nitrogen by prolonged inhalation of a suitable nitrogen-oxygen mixture* It would further be expected that the latter procedure would tend to obliterate or diminish the individual differences that have beer ■* r kingly demonstrated by the usual procedure of pre-cxygonatlon* If, on Seer hand, the slowly eliminated nitrogen fraction is not of prime significance, one would expect the same response to elimination of a certain proportion of nitrogen whatevi: r 11 n course ia followed, and consequently the same individual different* i 2 In the present experiments we endeavoured to find a preliminary answer to the problem posed in the above terms. The first impetus to the project was provided by a practical question drawn to our attention by Dr, Code upon his return from a meeting of the Committee on Decompression Sickness, A person flies at 15,0#0 feet for about six hours, breathing air sufficiently enriched with oxygen to make his alveolar oxygon pressure about equal to its usual value at sea level. How much protection from "bends" will this procedure afford if the aviator then asceaia 35.000 feet? A very good guess can be made, and was in fact made by Code and Robinson (1944); it could be confidently predicted that the preliminary flight at 15.000 feet would protect the fliers to a significant extent, but the amount of protection could not be expressed in terms of the equivalent period of breathing pure oxygen, nor could It be stated to what extent individual differences in behavior would be encountered. Thus solution of the operational problem would be a step toward clarification of the questions of mechanism just adumbrated, and it seemed opportune to extend the experiments somewhat beyond the stage demanded by practical considerations. We believe that the results presented below justify a fairly definite statement concerning the protective effect of long exposure to various gas mixtures prior to high altitude f ight, and that the implication# of tbe results in terms of the relative importance of "slow" nitrogen and other, undefined# factors a, o suggestive enough to w-rrant further work along the same liras. We have been hampered by our very limited access to experimental subjects; a limitation that is, however, significantly offset by the increased reproducibility of decompression sickness data that results from the use of a standardized exercise for accelerating the production of symptoms. On the other hand, it must be r' nembered tha an essen- tially different analysis might be necessary for the interpretation of d* •. obtained without this facilitating influence. 2, METHODS. Most ox the experiments were carried out with four subjects, wlisa pertinent physical characteristics are given in Table 1, The general procedure prior to the actual susceptibility test was always the same apart from differences in time, alaitude, and composition of Inhaled gas. The subjects breathed the ar< ropri »te gas mixture through a constant flow apparatus and i.8B mask in which the usual sponge rubber discs were usually replaced by expiratory valves, the flow being adjusted to a somewhat excessive value in order to avoid the possibility of inward leakage of air. During six-hour periods of breathing under these conditions Intervals were usually taken at the end of two and four hours, and at these times it was considered justifiable to allow the subjects to breathe air for 10 or 15 minutes. At the end of this preliminary period the subjects were taken rapidly (t ,C0( f e •: t per minute) to a simulated altitude of 35,-OOC feet. The standard ex180 - (pC + pO + 47) + 30 where N, G, 0 refer as before to nitrogen, carbon dioxide and oxygon, 9 N is identical numerically to the pressure pN of gaseous nitrogen which would be in equilibrium with the dissolved nitrogen of the tissues. Capillary pressure is taken as 30 mm, Hg, Putting (pC + pO) equal to about 73 mm,, the condition for growth bt-oomo s N > 90, The considerable difference between this limiting value and the experimental threshold for production of symptoms, 230 mm,, shows that considerable supcrsaturation - roughly fourfold - can bo tolerated in the body. Whether this fact arises from difficulties in the inception of bubbles, from the operation of capillary forces, or from physiological tolerance of bubbles below a certain size, it is beyond the scope of the present experiments to indicate# 5 . DISCUSSION,* The realization that many individuals aro only protected against decompression sickness by rather long periods of oxygen inhalation is comparatively * No effort has boon made to refer to all the available evidence nor to give a complete bibliography. Haldane and others among the early -workers on caisson sickness -were awaro that denitrogcnation rates of various tissues would be expected to depend both on the quantity of nitrogen dissolved and on the vascularity of the tissues concerned, but they considered that adoption of a half-saturation time of 75 minutes would give generous latitude for do saturation rates below those which had at that time boon detected by direct measurement. Improved measurements, notably by Behnko and his collaborators, have suggested that the greater part of the nitrogen elimination in man cours with a half-dosaturation time of 50 - 90 minutes. On the other hand, there are records of persons who required from four to six hours preoxygenation in order to acquire reasonable immunity to decompression sickness {of, Behnke, Wolham and Yarbrough, 1942; Ferris, Webb, Ryder, Engel, Romano and Blankenhorn, 1943), while for others - the majority - ono to two hours is Sufficient, It was natural to postulate differences in nitrogen elimination rates in order to account for this wide spread of susceptibilities, and further to attribute the production of symptoms to a relatively small fraction of the total dissolved nitrogen, the rate of elimina- tion of which cannot be determined by direct measurement. Such an extension of effective nitrogen elimination constants far beyond the limits warranted by direct measurement was in fact made the basis of a suggested routine for the denitrogcnation of aviators, with the additional assumption that "slow” nitrogen, once removed, would be reabsorbed equally slowly during the breathing of air prior to ascent (Bazctt, Thompson and Bateman 1941, 1942), The procedure now appears to bo warranted both by the original experimental data and that reviewed by Fraser, Stewart and Manning (1942), to which may perhaps be added recent results from the Yale laboratory (Fulton, 1944), We believe however that a direct proof of the importance .of a "slow" nitrogen fraction has been lacking. The method of true equilibration of the body with known partial pressures of nitrogen, used in the present series of experiments, permits, in principle, the measurement of the rate at which the "effective” or symptom-producing nitrogen can bo eliminated, and hence the establish- ment of its relationship to the overall rates as usually measured. Our first exploratory application of tho method fully justifies the "slow" nitrogen hypothesis for it shows conclusively that if the time of half-saturation for the "effective" nitrogen is around 100 minutes for the three relatively insusceptible subjects, it must be at least 260 minutes for the fourth and most susceptible; if the six hour period was not sufficient for virtually complete equilibration of the former group, then both of those half-losaturation times will have to be increased; in any event they cannot be smaller than the values given. This brings at least one figure far ovtsido the range of directly measured values. In view of the apparent importance of a slow desaturation process that cannot bo detected by the usual methods of measurement, it is of interest that signs have been observed of a correlation between susceptibility to decompression sickness and measured rates of inert gas exchange (Wu, 1942; H, B, Jones et al, 1942; Stov1; a ct al, 1943), The correlation is not particularly striking, but insofar as it is genuine it must imply the existence of a relationship between the various tion constants of the body. Such a relationship wouli be expected if differences in gross circulation rate are responsible for variations in susceptibility among individuals, but it would be hard to explain if susceptible persons owed their condition to local and perhaps pathologies 1 states unrelated to the efficiency cf their general circulation. The question evidently merits further experimental study. REFERENCES Bazett, K.C., Thompson, J.W, and Bateman, J,B, 1941, Report to National Research Council, Canada, November 1941, Bazett, H,C., Thompson, J.W, and Bateman, J,B, 1942, Proo, 15th Meeting of Associate Committee on Aviation Medical Research, National Research Council, Canada, June 1942. Behnke, A,R, and Willmon, T,L, 1941, Am,. J, Physiol,, 131, 619-626, Behnke, A.R., Welham, W, and Yarbrough, 0,D, 1942, O.S.R.D., C.A.M, Report No, 114, 15 December 1942, Boothby, W.M, 1944, Handbook (see below). Chart A-l, Table 1, Boothby, W.M,, Lovelace, W.R, and Benson, 0.0„ 1942, "Physiology of Flight," Aoro Medical Research Laboratory, Experimental Engineering Section, Materiel Center, Wright Field, Dayton, Ohio, Fig, 15, page 27, Code, C.F, and Robinson, F.J, 1944, Subcommittee on Decompression Sickness, Minutes, 8 February 1944, Ferris, E.B., Webb, J.P,, Ryder, H.W., Engel, G.L., Romano, J,.and Blarkonhorn, M,S, 1943. O.S.R.D., C.A.M. Report No, 121, 16 February 1943, Fraser, A.M., Stewart, C,B, and Manning, G.W, 1942, Report for Associate Committee on Aviation Medical Research, National Research Council, Canada, 17 Sept, 1942. Fulton, J,F, 1944, O.S.R.D,, Committee on Medical Research, Bi-monthly Progress Report No, 18 on Contract CEMomr-38, 1 June 1944, Handbook of Respiratory Data in Aviation, 1944, Subcommittee on Oxygen and Anoxia, National Research Council, Washington, D. C, Jones, H,B,, Smith, R,, Sears, N,, Wu, C,, Larkin, J,, French, R,, Hamilton, J. and Lawrence, J,H, 1942, 0,S.R,D,, C.A.M, Report No, 51, 29 May 1942, Stevens, C.D., Ryder, H.W,, Ferris, E,B,, Engel, G,L., Webb, J.P,, Senior, F, and Friedlander, J0K, 1943, O.S.R.D,, C.A.M, Report No, 237, December 1943, Wu, C, 1942, O.S.R.D,, C.A.M, Report No, 52, 29 May 1942, Yale Aero Medical Unit, 1944, Private communication from Dr, Nims. Table 1 DESCRIPTION OF SUBJECTS R.S. Ii,C, RaH, J c B, Sex F F F M Ago 38 37 19 36 Height, otn. 157 158 160 173 Weight, kg. 65,5 72 4 7 66,2 Table 2 DEGREES OF IMMUNITY TO DECOMPRESSION SICKNESS FOLLOWING VARIOUS FORMS OF DESATURATION Columns (3), (4) and (5) give details of gas inhaled, duration and altitude prior to decompression. Columns (6) to (9) give scores (see text) obtained by four subjects during a 90-minute flight, with exercise, at 35,000 feet. 1 | Procedure before Decompression Scores at 35,000 Feet (1) (2) (3) (4) (5) (6) (7) (8) (9) % 02 Duration Altitude . 1 No. Data inhaled (minutes) (feet) R. S, L,C, R.H„ J.B, 1 1 18 Feb. (20,9) Ground 63 50 86 62 7 23 Mar, (20.9) - If 147 64 121 68 2 22 Feb. 100 50 h 1 04+ 107 120+ 71 | 5 15 Mar, 100 50 N 297 162 360+ ' 121 1 3 5 M ar • ICO 80 n 170+ 284 (44) 111 | 4 10 Mar, 100 110 it 327 360++ 196 i 144 6 20 Mar, 100 110 ft 360+ 360+ 350+ i 101 10 « 5 May 100 240 M — — — 265 i 12 22 May 100 315 It — — fmmm 277 ( 14 31 May 100 390 a — — 360++ t 8 10 Ap r • 40 360 15,000 360+ 360+ 344+ 98 9 3 May 40 360 15,000 319+ 360+ 360+ 131 16 i 31 Aug, 30 360 15,000 — 348 — 106 11 11 May 40 360 Ground 143(+) 156 171 ! 90 13 25 May 57 360 it 137(+) 1 351 335 167 i 15 23 Aug, 40 720 Grtund 113 17 2 Sep *t, 59 720 — —— —— 271 Table 3 VALUE OF SIX HOURS AT 15,000 FEET BREATHING 30$ AND 40$ OXYGEN, EXPRESSED IN TERMS OF THE DURATION OF EXPOSURE TO PURE OXYGEN AT GROUND LEVEL REQUIRED TO GIVE THE SAME DEGREE OF IMMUNITY Per cent oxygon Sub jeot Mean score Mean score -without denitrogenation Equivalent time, minutes 40 Hrt S0 340+ 102 109+ L.C. 360+ 55 I 104+ - BjH, 350+ 100+ f (94) J.B. 115 65 89 30 ...L.C, 348 55 98 LtAt ~r" 106 65 81 Tabla 4 VALUE OF SIX HOURS INHALATION OF AND OXYGEN AT GROUND LEVEL, EXPRESSED IN TERMS OF THE DURATION OF EXPOSURE TO PURE OXYGEN AT GROUND LEVEL REQUIRED TO GIVE THE SAME DEGREE OF IMMUNITY Experiment Per cent oxygen Subject Score Equivalent time,(min.) 11 40 R.S. 143{+) 29U) L.C. . 156 54 R.H, 171 (41) J.B. 90 53 13 57 . R.-s. , 1?7(++) 25{++) L.C. r- 351 99 R.H. 335 . (89) J.B. 167 157 Table 5 CALCULATION OF ELIMINATION RATES OF SYMPTOM-PRODUCING NITROGEN Column (4) gives alveolar nitrogen pressures during inhalation of nitrogen + oxygen mixtures containing oxygen fractions given in Column (2), When such mixtures have been inhaled for suffic- iently long periods of time, the fractions of normal dissolved nitrogen remaining in the body are those given in Column (5), Subjects in this condition show the same degree of immunity to decompression sickness as they do after inhaling pure oxygen at ground level for the periods given in Column (6), These data are plotted in Fig, 4, (1) (2) (3) (4) (5) («) Exp, No. 1 fO Altitude (f oot) pN« » pN* */594 Equivalent times on pure oxygen (minutes) R.S. L, C« R.H, 1 J,B. ' 8 9 0,405 0,399 15.000 15.000 233,0 233.3 0,392 0,393 117+ 101 104+ 104+ 60-123 62-130 16 0.298 15,000 272.3 0.459 MW 93 — 11 15 0 * 398 it 1,000 i> 419,6 IV 0.706 n 29+ 54 29-54 94 13 17 0,577 0,585 - 1,000 tt 290.2 285,0 0,489 0,480 25(++; 99 59-120 I 271 ! Table 6 CALCULATION OF DISSOLVED NITROGEN FRACTION REMAINING AFTER SIX HOURS INHALATION OF VARIOUS NITROGEN-OXYGEN MIXTURES IN THE CASE OF A SUBJECT FOR WHOM SIX HOURS IS INSUFFICIENT FOR COMPLETE ECLUILIBRATION WITH THE INHALED PARTIAL PRESSURE OF NITROGEN. Assumptions are discussed in text. The data in Column (7) are plotted in Fig, 4. Subject? J.B, (1) (?) (3) (4 (5) (6) (7) N2 fraction left Ng fraction loft Equivalent time Sxp • at equilibrium after six hours on pure 02 No o fO I-ltitud© (from Table 5) (from Col. (4)) Score (minutes) 8 0,405 15,000 0,392 0,635 98 67 9 0,399 15,000 0,393 0,635 131 112 16 0.298 15,000 0«459 0,675 106 81 11 0.398 *1,000 0.706 0.824 90 53 13 0.577 1,000 0.489 0.693 167 157 LEGENDS TO FIGURES Fig. 1 * Bata illustrating tho degrees of immunity to decompression sickness acquired by four persons after breathing pure oxygon for various perieda. Abscissa* duration of oxygen breathing. Ordinate* degree of immunity, or recorded score in standard flight with exercise at 35,000 foot. The highest possible recorded score is 360 for a stay of 90 minutes at 35,000 feet without symptoms. The true score in such a case must be greater than 360, Arrows are used in the diagram to indicate this. Fig. 2; Indicating scores obtained by prolonged breathing of certain mixtures of nitrogen and oxygen, and the length of time for which pure oxygen must be breathed in order to obtain the same score. Abscissa* logarithm of duration of oxygon breathing. Ordinate* recorded score in standard flight, with exorcise, at feet. Smooth curves based on data given in Table 2 and Fig, 1, Horizontal linos indicate scores obtained in six-hour and twelve-hour experiments with 40/4 and 60% oxygon. The abscissae of thoir intersections vd. th tho smooth curves give tho equivalent period of oxygen breathing at ground level. Fig3 3: Decompression sickness scores obtained by four subjects after "equilibration" with various mixtures of hltrogon and oxygon. Abscissa* fraction of total dissolved nitrogen remaining in body after equilibrium level is reached. Ordinate* score obtained in 90 minute flight, with exorcise, at 35,000 feet, "Equilibration" periods wore six hours for subjects H.S., L.C,, R,H, and twelve hours for J.B, Fig. 4: Illustrating principle of method for determining elimination rate of symptom-producing nitrogen by measuring susceptibility at a known "equilibrium" level of dissolved nitrogen. Abscissa* duration of inhalation of gas mixture, arbitrary unit. Ordinates fraction of symptom-producing nitrogen remaining in body. Curve marked "oxygen + nitrogen" represents course of nitrogen elimination when a particular mixture is inhaled. Curve marked "oxygon" represents nitrogen elimination when pure oxygen is inhaled. Both curves are exponentials with the same half-saturation time • Horizontal dotted lino represents the equilibrium level after prolonged inhalation of the nitrogen-oxygen mixture. The abscissa *f its point of intersections with the "oxygen" curve gives the equivalent period of inhalation of pure oxygon. Inhalation of oxygen for this length of time will give the same protection against decompression sickness as prolonged inhalation of the nitrogen-oxygen mixture. Clearly the course of the oxygon curve, which in general is unknown, can be plotted experimentally by comparing "equilibrium" scores, obtained with mixtures containing various proportions of nitrogen and oxygen, with scores obtained after inhaling pure oxygon for various periods# Logends to Figures (oont,) Fige 5* Elimination rates of symptom-producing nitrogen. Abscissa* duration of oxygen breathing at ground level. Ordinate* fraction of initial dissolved nitrogen remaining after dosaturat ion. Points are based entirely upon measurement of susceptibility t* dooomprassi on sickness, They show the fractions of symptom-producing nitrogen remaining in the body after pure oxygen has boon breathed for various periods, the latter values having been obtained from "equilibration4 data in the manner indicated in Fig, 2 and Fig, 3. Smooth curves are measured nitrogen elimination curves obtained by Bohnke and Willmon, 1941 (lower curve) and Boothby, Lovelace and Benson, 1942 (upper curve). Broken curve is drawn through the experimental points for the most susceptible subject. It indicates a half-desaturation time of 270 minutes for the symptom-preducing nitrogen of this subject., compared with the values 60—90 minutes indicated by the smooth curve. EFFECTS OF PREOXYGENATION DEGREE OF IMMUNITY (ARBITRARY UNITS) DURATION OF OXYGEN INHALATION C.hnrt Nn. 'VTT - fln Mnvn Aern Mpriir.nl Unit ■ I R Rnfpmnn Ann IQ4d EFFECTS OF PROLONGED INHALATION OF GAS MIXTURES COMPARED WITH EFFECTS OF PREOXYGENATION FOR SIX HOURS FOR SIX HOURS FOR SIX HOURS FOR SIX HOURS DURATION OF INHALATION OF OXYGEN (MINUTES) Mayo Aero Medical Unit Chart No. 1ZIL- 8 c J.B. Bateman Aug. 1944 SCORES OBTAINED AFTER " EQUILIBRATION" WITH GAS MIXTURES SCORE IV FRACTION OF NORMAL DISSOLVED NITROGEN Mayo Aero Medical Unit Chart W — 8 g j.b Bateman Aug. 1944 PRINCIPLE OF EQUILIBRATION METHOD IN STUDY OF DECOMPRESSION SICKNESS J.BBofeman Sept. 1944 TIME Mayo Aero Medical Unit Chart TZH 8 d PERCENT INITIAL NITROGEN COURSE OF ELIMINATION OF SYMPTOM-PRODUCING NITROGEN FRACTION OF BODY'S INITIAL DISSOLVED NITROGEN REMAINING AFTER DESATURATION RS RH LC JB MAYO AERO MEDICAL UNIT Chart 3ZTL- 8f- 2 DURATION OF INHALATION OF OXYGEN J.B. Bateman Sept. 1944 NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH Office of Scientific Research and Development of the COMMITTEE ON AVIATION MEDICINE OPEN Report No, 381 September 1944 THE REDUCTION OF ALVEOLAR CARBON DIOXIDE PRESSURE DURING PRESSURE BREATHING AND ITS RELATION TO HYPERVENTILATION, TOGETHER WITH A NEW METHOD OF REPRESENTING THE EFFECTS OF HYPERVENTILATION. By J, B, Bateman, Mayo Aero Medical Unit, Rochester, Minnesota- Responsible Investigator! Walter M, Boothby, (C,S,R.D, oontraoti OEMomr-129) ABSTRACT The development of decreased alveslar carbon dioxide pressures during pressure breathing at ground level and the subsequent return to normal values occur significantly more rapidly than in the case of voluntary hyperventilation* This experimental fact is discussed with the aid of new diagrams, based on the blood nomogram, showing the relation between ventilation rate, alveolar carbon dioxide pressure, and venous carbon dioxide pressure. The facts concerning the kinetics of srdinary hyperventilation are accounted for fairly satisfactorily by such considerations. The contrasting behavior of the alveolar carbon dioxide pressure in pressure breathing leads us to conclude that in this case additional factors are Involved, including possible decrease in cardiac minute volume, decrease in peripheral circulation rate and expulsion of blood from the lung during pressure breathing, with a reflex compensatory diminution of ventilation rate after return to normal intra- pulmonary pressure. It is therefore possible that the decreased alveolar carbon dioxide pressures are not exclusively due to an increased ventilation rate associated with pressure and are hence not entirely reliable as an Index of develcping acapnia. NATIONAL RESEARCH COUNCIL, DIVISION OF MEDlCAh SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE OPEN Report No. 381 September 1944 THE REDUCTION OF ALVEOLAR CARBON DIOXIDE PRESSURE DURING PRESSURE BREATHING AND ITS RELATION TO HYPERVENTILATION, TOGETHER WITH A NEW METHOD OF REPRESENTING THE EFFECTS OF HYPERVENTILATION, By J, B, Bateman, Mayo Aero Medical Unit, Rochester, Minnesota, Responsible Investigator* Walter M, Boothby, (OtS.R,D, oontracti OEMomr-129) INTRODUCTION 1, Pressure breathing at ground level with constant pressure without the aid of a pressure vest has been found to be accompanied by a decrease of 4 to 8 mtn< Hg in the alveolar carbon dioxide pressures of several trained and experienced subjects studied in this laboratory. This decrease is of appreciable duration; it can still be observed after five minutes of pressure breathing. Although in sove.-a.l other subjects, on the contrary, the initial rapid decrease proved to be succeeded by a rapid return to normal values (compare ref, (4)), the nature of the m.'ro persistent effect In the former group was -worth examining. When the experiments were begun it was believed to be an experimental arifact associated with imperfect mixing in the lungs (compare ref, (l)), and it,seemed appropriate to test this possibility by comparing the rate at which the normal alveolar carbon dioxide level "s regained, after return to zero pressure, with rates of recovery from voluntary hyperventilation. Wo have in the meantime ceased to take the artifact hypothesis at all seriously in the present context. However, our results and their analysis may be applicable to current discussions of hyperventilation and pressure breathing* They have already been referred to by ourselves and by others, in discussions and reports unconnected with this laboratory. METHODS Pressure breathing* Oxygen# Bendix pressure demand regulator set at 8 i robes of water, Bulbulian Type 21B pressure mask provided with T-pieoe below mo sk for collection of alveolar samples. Pressure always released during alveolar expiration. Collection of samples through T-piaoe gave the same results as removing mask and using standard mouthpiece. Hyperventilationi "Moderate" hyperventilation is deep inspiration and normal expiration five times per minute, "Severe" hyperventilation is deep inspiration and deep expiration ten times per minute. 1. Course of change of alveolar carbon dioxide pressure, pC1’, during pressure breathing and hyperventilations The results of several individual runs on one subject are shown in Fig. 1, which gives some idea of the reproducibility of the data. One notes the following tendencies* (a) During pressure breathing an initial sharp decrease of pC* * of about 6 mm* Hg, which may be followed by a certain amount of recovery. (b) A much slower decrease during "moderate" hyperventilation, arriving after 5 minutes at a level, about 7 mm. Hg, below normal, not greatly different from that resulting from 5 minutes pressure breathing. (c) During "severe" hyperventilation, a sharp initial decrease of 8-11 mm. Hg. followed by a further slow decrease, reaching a value of 13-14 mm* Hg. below normal in 5 minutes. These differences become more clear-cut when the changes in pC'1 are ex- pressed as percentages of the change observed after five minutes and the individual values then averaged for all three subjects (Fig. 2), We note that the very rapid inital decrease is common to both pressure breathing and "severe" hyperventilation but is not observed with "moderate" hyperventilation* 2, Course of recovery from the effects of pressure breathing and hyper- rent ilati on: In Tig* 3 the gain in pG’* during recovery is expressed as percent- age of the preceding decrease. We note that (a) Recovery from two deep breaths and from two breaths of pressure breathing occurs very rapidly. It is complete in less than one minute, (b) Recovery from five minutes pressure breathing is just as rapid as recovery from two breaths pressure breathing or from two deep breaths. (o) Recovery from "moderate" hyperventilation requires four or five minutes, despite the fact that the actual decrease in pC11 resulting from this procedure is no greater than that resulting from pressure breathing* (d) Recovery from "severe" hyperventilation follows at first the same course as recovery from "moderate" hyperventilation* The curves then diverge, the later stages of recovery from "severe" hyperventil~ ation becoming gradually slower* The contrast between true hyperventilation and pressure breathing is brought out in Fig, 4, in which the tines of half recovery, taken from Fig. 3, are plotted against the maximum decreases in pC,f. method of presentation brings out the main point, but it is of course deceptive, in that it fails to show the differences between "moderate" and "severe" hyperventilation arising from the differ©- ences in the curves during the second half of the recovery process. DISCUSSION How oan these rather striking kinetic differences be accounted for? First we oan draw an obvious distinction between processes involving only the arterial blood and processes involving the tissues, A decrease in pC' at constant venous carbon dioxide pressure has the property that it is rapid in onset (time considerably less than circulation time) and is transitory. The observed effect of two deep breaths provides an illustration* rapid decrease in pC*', rapid recovery. If perpetuated by continued hyperventilation or by other means, the decrease in pC*' must involve a decrease in venous carton dioxide pressure. This, necessitating loss of some of the combined carbon dioxide in the tissues, occurs relatively slowly. Those processes may be followed more closely by reference to --a diagram such as Fig, 5, in which alveolar carbon dioxide pressure is plotted against venous carbon dioxide pressure. The diagram refers to the case of a subject whoso oxygen consumption, circulation rate, and metabolic respiratory quotient* (0,8) remain constant at all times, changes being brought about solely by changes in ventilation rate. Two sets of reference lines are included. Those for various values of the respiratory quotient* are implied in L, J, Henderson's nomograms representing the properties of human blood (?) and were derived from that source; those for various values of the alveolar ventilation rate,were, obtained by combining the same data vr'th a formula derived by Chadwick, Otis, Epstein and Fenn (5)» Y = 0.864 YQ/pC» » Y - alveolar ventilation rate, liters per minute moist gas at 37° C. Y = oxygen consumption = 300 oo./miri,dry, at 0* C. and 760 mm. where In Fig, 5 a single transverse line, for a respiratory quotient 0.8, which is taken as the metabolic respiratory quotient, represents all theoretically possible steady states for a person subject to the limitations defined above. Of all these possible states, that which is actually found under normal conditions is given by the single point 0, That it must be this point and not some other is of course determined by factors in physi'qiogioal’r‘coor dination for which no allowance has been made in current phy si co-chernicbl descriptions of blood. Starting from 0 as origin, d decreases in alveolar carbon dioxide pressure oan result in several ways I At the one extreme, a very slow progressive increase in ventilation rate could occur in such a way that the subject, although losing carbon dioxide and suf- fering a gradual decrease in venous carbon dioxide pressure, would always remain ta all intents and purposes in a steady state. This change, which would appear on Fig, 5 as a downward movement from 0 along the reference line for'Cl = 0,8, may be described as hyperventilation without change in respiratory quotient. At the other * In this paper the term "respiratory quotient," for which the symbol Cl is also used, is intended to specify the nature of the gas exchange oocuring in the lungs at any time. Unless the subject happens to be in a "steady state," the respiratory quotient so defined will not be equal to the "combustion quotient" or the "metabolic respiratory quotient." Thus our definition represents an extension of the term which is usually employed in literature on metabolism to denote the combustion quotient. This sugges- tion is made only, we hope, as a temporary expedient in absence of an accepted consistent terminology. extrema, a sudden increase in ventilation rate results in rapid decrease of pC*• without change in venous carbon dioxide pressure# The new instantaneous state is defined by a point lying on the appropriate reference line for alveolar ventilation rate and on the same horizontal level as 0, If the increased ventilation is continued, the initial rapid change must be succeeded by a further slow decrease in pC** and a simultaneous decrease in venous carbon dioxide pressure, the course of the change following the line of constant ventilation rate in the manner shown by the descending portions of the heavy lines C and A in Fig, 5, Since the reference lines for constant ventilation and constant respiratory quotient are not parallel, it is obvious that once the initial period of rapid adjustment is over, continued constant hyperventilation must be accompanied by a gradual decrease in respiratory quotient. There will be a trend, in other words, toward the original metabolic respiratory quotient. In general other factors will intervene before the change is completed, but it is possible in cases of mild hyperventilation that a new steady state may be attained in this manner and may be compatible with continued physical and mental well-being, return to the subject of immediate interest, we may say that the type of change described above and indicated in curves A and C accounts fairly well for the facts observed in the cases of "moderate" and "severe" hyper- ventilation respectively. The increasing slope of the reference lines with increasing ventilation indicates the increasing relative importance of the initial rapid fall in pC' ' , which may perhaps be accentuated by the fact that in the "moderate" hyper- ventialtion procedure used in our experiments there was opportunity during the slow normal expirations for partial recovery of the decreased pC'1 values during each breath. If this is correct, curve A should be replaced by a aig-zag, The case of pressure breathing, with an initial sharp decrease followed by a slow increase of pC’1, is illustrated by curve B, which shows the effect of a decrease «f ventilation rate in causing an increase of ,pC' without however preventing continued loss of tissue carbon dioxide in excess of. that corresponding to the metabolic respiratory quotient, It remains to consider the recovery curves. For "moderate" and "severe" hyperventilation, the initial stages of recovery are presumably represented by the horizontal portions of A’ and B', These involve in each case roughly the same prepor- tion of the total change in pC11, so that it is not surprising to find that the twe recovery curves have an initial stage in common, nor that they later separate, in accordance with different absolute amounts of carbon dioxide that have to be reacoumu- lated by the tissues. It is however surprising that the initial stage, although still comparable with the circulation time, is so much slower than the corresponding process in the response to increased ventilation. Contributing factors are undoubt- edly the rate of transfer of carbon dioxide from blood to alveolar air, which delays the establishment of an increased pC•1 after return to a normal ventilation rate, and also the convergence of the ventilation lines with decreasing venous carbon dioxide pressure, which considerably enhances the relative importance of the "slow recovery phase. Possibly other factors should also be considered. The recovery curves for pressure breathing require some additional postulate. In the first place, it is not possible within the framework of the present assumptions to account for the persistant fall in pC*' during pressure breathing without admitting * A closely similar recovery curve following two minutes forced breathing is to he found in an early paper by Boothby (3), that there must be some fall in venous carbon dioxide pressure. How then can recovery be so much more rapid than recovery from ordinary hyperventilation? The curves B’ and B'1 suggest that this may result from a temporary decrease of ventila- tion, of reflex origin, following the period of pressure breathing. It must bo however, that other relevant changes have been supposed to occur during pressure breathing, so that Fig, 5 may not be applicable to this case. There may sometimes be a decrease in cardiac output (2) (4} (6) (8) together with decreased peripheral circulation (2) (4) (6) and expulsion of stored blood from the lungs (4), Fig, 6 is meant to illustrate how a sudden decrease in cardiac minute volume could contribute both to rapid decrease in pC' when pressure breathing is started, and to rapid recovery; it further illustrates how readjustment to a new steady state can in such a case be effected rapidly and without any significant change in CO2 balance. It may be noted in passing, however, that a 30 per cent decrease in cardiac minute volume results, even at ground level when oxygen is being breathed, in a considerable and probably dangerous fall in venous oxygen pressure, and that this condition would not be significantly alleviated even if the response to decreased cardiac output were to follow the alternative but improbably course indicated by the vertical arrows in Fig, 6, Expulsion of stored blood from the lung, and its storage elsewhere, could also contribute to a rapid recovery of pC'' if relatively venous stored blood were suddenly thrown back into the pulmonary circulation upon release of the raised intrapulmonary pressure. Decreased peripheral circulation during pressure breathing if accompanied by an increase in the volume of blood in the limbs, could act in the same manner. Thus it appears probable that while the facts concerning the rates of decrease and increase of pC»• during and after ordinary hyperventilation can be broadly accounted for without postulating any physiological mechanism other than that arising directly from the altered ventilation rate, this is probably not true rf the decreased pC?* accompanying pressure breathing. The exceptional rapidity with which the original pC* * is regained after pressure breathing shows that other changes are operative, among which may be a reflex (?) decrease in ventilation rate after a period of pressure breathing, decrease in cardiac minute v*lume, and sudden return of stored bicod to the lungs. Insofar as a decrease in cardiac minute volume can bo held responsible, it is probable that the decreased pC» * in pressure breathing is no+ a reliable index of significant loss of carbon dioxide - cannot, in other words, bo regarded as a valid criterion of acapnia. REFERENCES (1) Bateman, J,B, and W,M. 1944, O.S.R.D., C.A.M. Report No, 341, (2) Bloom, W.L,, Kaufman, 3.S., Nims, L,F* and Nyboer, J, 1944, O.S.R.D,, C.A.M, Report No, 246, (3) Boothby, W,M, 1912, J, Physiol,, 45, 328, (4) Chadwick, L,E,, Fenn, W,0,, Otis, A,B, and Rahn, H, 1943, O.S.R.D, Second report on Contract OEMomr-147; summary in C.A.M, Report No, 249, (5) Chadwick, L,E,, Otis, A.B,, Rahn, H,, Epstein, M.A, and Fenn, W,©, 1944, 0,S.R,D., C.A.M, Report No, 304, (6) Fenn, W.O., Chadwick, L,E,, Mullins, L,J«, Bern, R.J., Otis, A.B., Blair, H,A., Rahn, H, and Gssselin, R.E, 1943, O.S.R.D,, C.a.M, Report No. 111. (?) Henderson, L,J, 1928, Blood, A study in general physiolsgy, Yale University Press, (8) Starr, I, and MoMiohael, M. 1943. O.S.R.D,, C.A.M, Repsrt No, 165^ J.B.BATEMAN SEPT. 1944 COURSE OF CHANGE OF ALVEOLAR COz PRESSURE DURING HYPERVENTILATION AND PRESSURE BREA T HIN G AT GROUND LEVEL Results of individual measurements on a single subject, L.C. "SEVERE HYPERVENTILATION": Deep inspiration, normal expiration, 10 times per minute "MODERATE HYPERVENTILATION": Deep inspiration, normal expiration, 5 times per minute PRESSURE BREATHING 8 INCHES WATER TIME (MINUTES) i i HYPERVENTILATION. SEVERE HYPER VENTIL ATION, MODERATE^ MAYO AERO MEDICAL UNIT CHART NO- SX - 4 (I) ALVEOLAR C02 (MM.MERCURY BELOW NORMAL) COURSE OF CHANGE OF ALVEOLAR CO2 PRESSURE DURING HYPERVENTILATION AND PRESSURE BREATHING AT GROUND LEVEL Average of all data for 3 subjects Abscissa: Time in minutes at which alveolar samples were taken Ordinate: Change of alveolar CO2 expressed as percentage of change produced in 5 minutes M o o or < _i o LlI < U_ O U o < o Moderote hyperventilation Severe hyperventilation Pressure breathing. 8 inches water TIME (MINUTES) J. B. BATEMAN SEPT. 1944 MAYO AERO MEDICAL UNIT CHART NO. XIX -4(2) RETURN OF ALVEOLAR C02 TO NORMAL VALUE FOLLOWING HYPERVENTILATION AND PRESSURE BREATHING AT GROUND LEVEL After 2 breaths pressure breathing, 8 inches water After 2 deep breaths After 5 minutes pressure breathing, 8 inches water After 5 minutes moderate hyperventilation: Deep inspiration normal expiration, 5 times per minute After 5 minutes hyperventilation: Deep inspiration, deep expiration, 10 times per minute PERCENT RECOVERY TIME (MINUTES) MAYO AERO MEDICAL UNIT CHART NO. TTT- 4 (3) a B. BATEMAN SEPT. 1944 TIMES FOR HALF RECOVERY FROM DECREASE IN ALVEOLAR C02 PRESSURE PRODUCED BY HYPERVENTILATION AND PRESSURE BREATHING AT GROUND LEVEL Recovery from 2 deep breaths Recovery from 2 breaths of pressure breathing (8inches water) Recovery from 5 minutes pressure breathing Recovery from 5 minutes .hyperventilation ; deep inspiration, normal inspiration, 5 times per minute Recovery from 5 minutes hyperventilation: deep inspiration,deep expiration, 10 times per minute TIME FOR HALF RECOVERY (MINUTES) INITIAL DECREASE OF ALVEOLAR COz PRESSURE mm. Hg. MAYO AERO MEDICAL UNIT CHART NO. XIX-4 ( 4) JIB. BATEMAN SEPT. 1944 CORRELATION OF CHANGES IN ALVEOLAR (ARTERIAL) AND VENOUS CARBON DIOXIDE PRESSURES DURING AND AFTER RECOVERY FROM HYPERVENTILATION Heavy transverse lines ore lines of equal respiratory quotient Light transverse lines are lines of equal alveolar ventilation Areas enclosed by heaviest lines represent course of change accompanying a period of hyperventilation followed by recoveryfthe normal state being represented by point 0 A A' : very moderate hypervent ilation BB‘, BB": suggested changes during and after pressure breathing CC* ! severe hyperventilation RESPIRATORY QUOTIENT 1.4 1.2 1.0 VENTILATION VENOUS CARBON DIOXIDE mm.** ALVEOLAR CARBON DIOXIDE mm. hq. MAYO AERO MEDICAL UNIT CHART NO. XIX -4(5) J.B. BATEMAN SEPT. 1944 POSSIBLE CHANGES IN ALVEOLAR AND VENOUS CARBON DIOXIDE AND VENOUS OXYGEN PRESSURES FOLLOWING A DECREASE IN CARDIAC MINUTE VOLUME VENOUS OXYGEN VENOUS CARBON DIOXIDE mm. Hg. ALVEOLAR CARBON DIOXIDE, mm Hg. MAYO AERO MEDICAL UNIT CHART NO. XIX - 4(6) J. B.BATEMAN SEPT. 1944 NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE Report No® 428 May 2, 1945 OPEN THE EFFECT OF PRESSURE BREATHING UPON THE SKIN TEMPERATURES OF THE EXTREMITIES* By J0 B, Bateman and Charles Sheard, Mayo Aero Medical Unit, Rochester, Minnesota, (OSRD Contracts OBMcmr-129) ABSTRACT The present studies of the variations in skin temperature of the ex- tremities resulting from pressure breathing, both in psychromatic rooms and in the decompression chamber, have provided evidence in support of an initial peripheral vasoconstriction following inception of constant positive pressure breathing* This is usually transient, however, being succeeded by a slow rise in temperature which usually exceeds the initial decrease. When conscious respiratory effort is abolished, as by the use of the Burns pneumatic resuscitator, the secondary rise in temperature appears to be absent*, NATIONAL RESEARCH COUNCIL, DIVISION OF MEDICAL SCIENCES acting for the COMMITTEE ON MEDICAL RESEARCH of the Office of Scientific Research and Development COMMITTEE ON AVIATION MEDICINE Report No0 428 May 2, 1945 ORBS THE EFFECT OF PRESSURE BREATHING UPON THE SKIN TEMPERATURES OF THE EXTREMITIES* By J* B* Batoman and Charles Sheard, Mayo Aero Medical Unit, Rochester, Minnesota* (OSR£ Contract* OBMcmr-129) REPORT Introduction An increase in intrapulmonary pressure of about 15 mm* of mercury is well tolerated by the majority of healthy persons. This fact implies the success- ful adaptation of the circulation to an entirely new set of mechanical condi- tions. Since the increase in pressure, which must, to a considerable extent, be transmitted to the entire thoracic cavity, is of the same order of magni- tude as the blood pressure, it is obvious that changes must occur in order that normal circulation can continue against an increased pulmonary resistance. Tho details of this circulatory adaptation are not fully voiderstood, but it is acknowledged that there must bo a rise in both arterial and venous blood pres- sure, a rise in intra-auricular pressure and an engorgement of veins in the systemic circulation* The fact that no untoward effects result from pressure breathing is in itself a proof that, in spite of these altered mechanical conditions, cardiac output is more or less maintained. It appears probable from measurements of performance of one kind or another that, as far as the cerebral circulation is concerned, no redistribution of blood occurs* It is known, however, that a considerable amount of blood-is passively displaced from the lung during pressure breathing and that this change compensates in part for the effects of venous congestion* A further adaptation would appear to consist in a redistribution of blood in tho systemic circulation whereby tho blood flow to the internal organs is maintained, despite a significant decrease in cardiac output, at tho expense of blood flow to the extremities* Plethysmographic measurements of various kinds appear to demonstrate this decrease in blood flow in the extremities during the first few minutes of pressure breath- ing, According to Form and his collaborators this effect is accompanied by a constriction of tho peripheral arteries. ♦Reference may bo made to tho admirable review by Barach, Fenn, Ferris and Schmidt* (l) The peripheral redistribution of blood apparently demonstrated in those measurements is not of any.subjective significance under ordinary conditions, but it has been suggested that under operational conditions vasoconstric- tion in the extremities might have deleterious effects upon the manual dexter- ity of aviators. Thus an effect which under laboratory conditions is com- paratively slight might be of some importance in the field* It seemed import- ant therefore to investigate as carefully as possible the vascular changes in the extremities which occur during pressure breathing* The present paper is the outcome of an attempt to do this by means of measurement of the skin temperature under carefully controlled conditions© Methods In the present work experiments were made both in a constant temperature room and also at simulated altitudes in a decompression chamber* The former series enabled us to establish the effects which occurred with some certainty, while the series in the decompression chamber, which were of necessity less satisfactory than those made in the constant temperature room wore .run in order to satisfy ourselves that the change in conditions did not fundamentally modify the observed reactions* The first series referred to was carried out in psychromatio rooms at one of the hospitals in Rochester, Minnesota, which the temperature was main- tained at any desired temperature with an accuracy of + 0o3° C0 within a range of 20° to 30° C,, and the humidity at 40 per centQ A uniform procedure was adopted throughout the series* The subject arrived, without having break- fasted, at 8*30 acm, and was then allowed to rest in bed,, lightly covered but with the extremities exposed, for an initial period of vascular adaptation During this period the surface thermocouples were placed in position — 3 on each hand and 3 on each foot — in the manner described in previous papers by Sheard and collaborators (vide article by Sheard^6))0 The recording of skin temperatures was commenced and at the end of the introductory period, which was usually marked by an approach of the temperatures of the fingers and toes to their respective levels, the experiment proper could be started* The pressure breathing mask was placed in position and the subject was allowed to breathe a gas mixture supplied through a standard pressure breathing regulator set at zero or at a positive pressure of 1 inch of water* The gas was either pure oxygen or a mixture containing 20% oxygen and nitrogenu In view of the fluctuations in pressure which occur during the use of these regulators at ground level a pressure setting of 1 inch of water corresponds roughly to breathing at ambient pressureQ The zero setting, on the other hand., involves an appreciable amount of suction during inspiration, and is a source of some discomfort© Since the adjustment of the mask frequently caused some vasomotor dis- turbance a considerable time usually was allowed before the commencement of pressure breathing, temperature measurements being taken at frequent inter- vals (at least every 10 minutes) during this period* When sufficient stability had been attained the regulator setting was changed in most cases to 8 inches water, and temperature measurements were continued at shorter inter- vals* Temperature measurements during pressure breathing were continued either until a satisfactory effect was established or until the discomfort of the subject necessitated an early conclusion* Then the pressure was turned off and whenever possible the temperature readings were continued for a further period of 30 or 40 minutes© The procedure followed in the decompression chamber tfas essentially the same* with some modification necessitated by the marked decrease in temperature which invariable occurs during decompression,. It is impossible to avoid such a decrease. It can be minimized by slow ascent and its ef- foots upon the subject to some extent reduced by the use of blankets0 How- over these procedures p.re never entirely satisfactory and only a certain proportion of experiments made in the decompression chamber can be con- sidered at all successful* The majority of experiments wore made with 3 female subjects with long experience in pressure breathing* Occasional runs were also possible with other subjects© Experimental results In presenting the results of these experiments, it may be said at once that the effects of pressure breathing upon the skin temperatures of the extremities-, although by no means negligible, are readily obscured in many subjects by the effects of accidental stimuli* The necessity for carefully controlled conditions and for complete tranquility throughout an experiment cannot be too heavily stressed® It was not uncommon with certain of our sub- jects to find comparatively violent reactions to the adjustment of the pressure breathing mask or merely to the announced intention of increasing the pressure* This extreme vasomotor sensitiveness was not connected with any inability to withstand pressure breathingj it was merely a disturbing factor which prevented the profitable use of such subjects in experiments of this kind* On the other hand* experiments also had to be abandoned on occasion when after a reasonable period of adjustment the subject had failed to roach a state of vasomotor reactivity, by reason of prolonged dilatation or excessive constriction of the peripheral blood vessels© The data obtained are illustrated by moons of a series of representative diagrams* In what we shall in the following discussion regard as typical behavior, because it is the one that wo have observed most frequently9 a steady increase in intra-pulmonary pressure at first brings about a slight de- crease in skin temperature of the extremities© This is transient and is succeeded by a slow rise in temperature which as a rule exceeds the initial decrease so that the peripheral vessels bee one more dilated than they were in the control period at zero pressure* When the pressure is turned off there may or may not be a tendency for vasoconstriction to occur$ in the majority of cases the dilatation brought about by pressure breathing persistsc This typical behavior is shown in Figure 1 where it will be noted that both fingers and toes respond in the manner described* Figure 2 illustrates a case in which the typical response is confined to the fingers, the toes re- maining constricted and somewhat below room temperature throughout the ex- periment. It will bo noted that the typical transient constriction followed by slow dilatation occurred in this experiment with each successive increase in intrapulmonary pressure and, further, that the return to ambient pressure caused a slight transient constriction. Figure 3 illustrates an experiment of similar nature on another subject in whom, however, a reflex dilatation of the toes followed the corresponding response of the fingers® Figure 4 repre- sents a case in which the fingers were unable to respond because of their initial state of dilatation and in which, therefore, the entire observed ef- fect occurred in the toes0 Figure 5 represents one of the less usual experi- ments, entirely satisfactory from a technical point of view, but in which no effect whatever was observed when the pressure was increased. This result was obtained with a subject in whom what we have described as the typical response has been repeatedly and consistently observed. In Figures 6 and 7 two experiments in the decompression chamber are presented., It has been mentioned already that such experiments are commonly less satisfactory than those made in a constant temperature room, but it seemed necessary to know, firsts whether an altitude of 40,000 feet., with the hazard of decompression sickness, would modify the result already obtained,, and, secondly, whether anoxia would produce any significant effect. Figure 6 gives the best answer that we were able to obtain to the first questionc It will be noted that despite the cooling of the decompression chamber during ascent, amounting to about 3° C®, it was possible to keep the fingers and toes at a reasonably steady temperature, and that the effect of pressure breathing on the temperature of the toes exceeded by far any effect produced by other circum- stances, resembling very closely the typical effect of transient constriction followed by dilatation* The same can be said of the experiment reported in Figure 7, in which the anoxia produced by breathing 18 per cent oxygon at 11,000 feet was a significant factor® Here again, although throughout the experimental period the arterial saturation was not much greater than 75 per cent, the behavior of the finger temperatures was fairly typical® We have not attempted in this investigation to study, in any detail the effects of intermittent pressure breathing, involving, as they do,* such secondary factors as hyperventilation, but two experiments were made, for reasons which will be given in the discussion, using the Burns pneumatic re- susoitator* This device (shown schematically in Figure 8) provides inter- mittent positive pressure respiration of an almost entirely passive kind; it works most satisfactorily when the subject is as completely relaxed as possible, and it has been shown '5; that there is no significant tendency for hyper- ventilation to occur* It will be seen in Figures 9 and 10 that although there was perhaps a tendency for on initial vasoconstriction to occur when the resusoitator was used, there is no sign of subsequent vasodilation® Discussion The data presented in this paper appear to indicate that the decrease in peripheral blood flow observed during pressure breathing by other workers is a transitory phenomenon, succeeded by a slow imcreaso in flow in the majority of normal subjects* Wo are at a loss for any completely satisfactory explanation of this course of events* The initial decrease in blood flow may signify a period during which the circulation is becoming adapted to the new mechanical conditions imposed by pressure breathing® That is to soy, there may be a period in which the volume of circulating bleed is reduced, in which the heart is unable to cope with increased pulmonary resistance and in which venous return is significantly impeded® Added to this, there may be a redistribution of blood in the systemic circulation, and indeed it has been shown (3)that constriction of the artcriolds occurs in the peripheral vascular bed® An obvious reason for the succeeding increase in blood flew would lie in an increase in metabolic rate resulting from the increased effort of pressure breathing® It is difficult to estimate in quantitative terms the amount of extra work involved. At first sight it would seem that breathing against constant pressure requires exactly the same amount of work as normal breaih ing. the difference consisting merely in the change from active inspiration to passive inspiration and from passive expiration to active expiration* As Fenn and colleagues v4) have pointed out, however, breathing against constant pressure causes a change in the level of respirationa and a change, therefore, in the effective mechanical properties of the lung. Fenn estimates that this change in level of breathing results in a threefold increase of muscular work; the absolute amounts of work involved, however, are not more them 0*1 per cent of the total resting metabolic rate® Thus it seems improbable on these grounds that the increased effort of breathing under positive pressure would be re- flected in any measurable change in skin temperature* It is possible, on the other hand, that the efficiency of muscular work in breathing under pressure is so reduced as to give rise to a significant increase in metabolic rateD Increases in oxygon consumption of subjects breathing under positive pressure as great as 10 per cent have been reported,'4' and measurements of our own, while technically unsatisfactory in most cases, likewise suggest that the increased metabolic rate may be considerably greater than the calculated value of Ool per cent© In any event, our data suggest strongly that fears as to the deleterious effects of pressure breathing upon temperature regulation are entirely unf oundedo References (1) Barach, A. L., Fenn, Wc Or, Ferris, B. B, and Schmidt, C. PcS The physiology of pressure breathing* A brief review of its present status. Prepared for the Committee on Medical Research, O.ScR.D. December 1944. (2) Bloom, W, L©, Kaufman, S© S», Nims, L. F. and Nyboer, Jaj The effect of positive pressure breathing at sea level on peripheral blood flow and cardiac output. C.AoM© Report No© 246, ■ January 1944, (3) Chadwick, L© Ec, Fenn, W. 0®, Hcgnauer, A© Ho, Otis, A, B© and Rahn, H0s Displacement of blood from the lungs during pressure breathing© OoS,RaD., C.M.R., C.A.M. Report No. 249, pages 76-93s December 19430 (4) Fenn, W. 0,, Chadwick, L« B., Mullins, L, J©* Dern, E© Jo, Otis, A. B©r Blair, H© A©, Rahn, H. and Gosselin, R© B.s Physiological effects of pressure breathing© 0*S«R.Da, C©M©R©, 0,A©M® Report No© llle January 194 30 (5) Olson, 0© Cc* Carbon dioxide elimination. A©A.Fe, Hq,, Air Technical Service Command, Engineering Division, Aero Medical Laboratory©, No. TSEAL«3-660-65-D, 15 December 1944. (6) Sheard, C,» Skin temperatures and thermal regulation of the body© Medical Physics, 1523~1555, The Year Book Publishers, Chicago, 1944® Effect of Increased Intrapulmonary Pressure on Dark Adaptation May, 1945 - C-Sheard Minutes FIG. 2 Minutes FIG. 1* Macular area Logarithm thresh old — mittrotni/Hlamher ts Logon thm threshold-/Tvcro/nlll/lamherts Threshold-micromlllilctmherts Threshold - m icromiHilar ,£> er ts Retinal area, 10° paramacular Minutes Minutes FIG. 1 FIG* 3 Macular area Chart XVII- 10 Logarithm threshold - micromillilamberts Logarithm threshold-mlcr>ornlllilami>erts