Acclimatization and tolerance to extreme altitude

Journal of Wilderness Medicine 4, 17-26 (1993) ORIGINAL ARTICLE Acclimatization and tolerance to extreme altitude JOHN B. WEST, M.D., PHD, DSc. Department ofmedicine, University of California, San Diego, La Jolla, CA 92093-0623, USA During the last ten years, two major experiments have elucidated the factors determining acclimatization and tolerance to extreme altitude (over 7000 m). These were the American Medical Research Expedition to Everest, and the low pressure chamber simulation, Operation Everest II. Extreme hyperventilation is one of the most important responses to extreme altitude. Its chief value is that it allows the climber to maintain an alveolar paz which keeps the arterial paz above dangerously low levels. Even so, there is evidence of residual impairment of central nervous system function after ascents to extreme altitude, and maximal oxygen consumption falls precipitously above 7000 m. The term 'acclimatization' is probably not appropriate for altitudes above 8000 m, because the body steadily deteriorates at these altitudes. Tolerance to extreme altitude is critically dependent on barometric pressure, and even seasonal changes in pressure probably affect climbing performance near the summit of Mt Everest. Supplementary oxygen always improves exercise tolerance at extreme altitudes, and rescue oxygen should be available on climbing expeditions to 8000 m peaks. Key words: hypoxia, hyperventilation, maximal oxygen uptake, respiratory alkalosis, exercise tolerance, barometric pressure, supplementary oxygen Introduction Two major experiments during the last ten years have clarified how human beings can tolerate the very severe hypoxia of extreme altitude. One was an ambitious field experiment, the 1981 American Medical Research Expedition to Mt Everest (AMREE) [1]. The other was the equally remarkable long-term low pressure chamber situation, Operation Everest II (OE II) [2]. Building upon experience gained by previous expeditions and using additional data from meteorological measurements of barometric pressure, we now have a much better idea of how human beings can tolerate the extreme oxygen deprivation of altitudes above 7000 m. Role of hyperventilation One of the most striking findings from both AMREE and OE II was the extreme degree of hyperventilation at great altitudes. This is best determined from alveolar PC0 2, the value of which is inversely proportional to alveolar ventilation for a given carbon dioxide production. Figure 1 shows that in AMREE the alveolar PC0 2 fell approximately linearly as barometric pressure was decreased at extreme altitudes, and in the only measurement ever made on the summit of Mt Everest, the extraordinarily low value of 0953-9859 1993 Chapman & Hall

18 West o 15 OJ a.. c: :5 a 10 LU ::;.. <[ 25r------...-------, 20 f-.. I OL.-...1-...J 250 300 350 BAROMETRIC PRESSURE (torr) Fig. 1. Relationship between alveolar ped 2 and barometric pressure at extreme altitudes. Note the extremely low value of about 7.5 mmhg on the summit of Mt Everest. From [3]. 7.5 mmhg was obtained [3]. This means that Dr Christopher Pizzo, who collected these data on the summit, increased his alveolar ventilation some five to six times. There was additional anecdotal confirmation of this extremely high ventilation. The audio tape recorded by Pizzo on the summit on which he recorded the first directly measured barometric pressure showed that he was so dyspneic that he had to pant for breath between every two or three words. Table 1. Alveolar gas and arterial blood values at altitudes of 8000 m and 8848 m (summit of Mt Everest) Altitude Study Barometric Inspired Alveolar Arterial (m) pressure P0 2 P0 2 PC0 2 P0 2 PC0 2 ph (mmhg) (mmhg) (mmhg) (mmhg) (mmhg) (mmhg) Sea level 760 149 100 40 95 40 7.40 8000 DEI 280* 49 30 17 DElI 280 49 38 12 37 12 7.53 AMREE 284 49 36 11 8848 DEI 250+ 43 25 15 DElI 240 43 32 11 30 11 7.56 AMREE 253 43 35 7.5 28~ 7.5 >7.7 OEI, OEII: Operation Everest I and II. Low pressure chamber simulated ascents [2,4, 19J AMREE: American Medical Research Expedition to Everest. Data collected at altitudes shown [1, 3J *Altitude incorrectly stated as 25000 feet; pressure was 280 mmhg because Standard Atmosphere was used [4J j Altitude incorrectly stated as 27700 feet; Standard Atmosphere used [4 J FI0 2 = 0.22 [2, 19] vblood values are calculated [3J

Acclimatization and tolerance to extreme altitude 19 ",;;' :s 20 o ~........., """ IlAHN AND ons (1949) 10 OL------'-_-'--------l._-L-_L.-----'-_-'--------l._--'-----' 20 30 40 50 60 70 80 90 100 110 120 ALVEOLAR Po, Itorr) Fig. 2. Oxygen-carbon dioxide diagram showing alveolar gas composition in acclimatized subjects from sea level (top right) to the Everest summit (bottom left). Above an altitude of about 7000 m, alveolar P0 2 remains almost constant at about 35 mmhg. From [3). The values for alveolar PC0 2 obtained in the low pressure chamber situation, OE II, were not quite so low, though the minimum value of about 11 mmhg was certainly impressive (Table 1). The slightly higher values can probably be explained by the fact that the OE II subjects were not so well acclimatized to extreme altitude. For example, the first simulated summit excursion occurred after 36 days from entering the low pressure chamber, whereas Pizzo had been exposed to high altitude for 77 days before he took his summit samples. The limited degree of acclimatization was also suggested by results obtained during a previous low pressure simulation, Operation Everest I (Table 1) [4], where the alveolar gas values for extreme altitude were almost the same as those for completely unacclimatized subjects, as pointed out by Rahn and Otis [5]. The rates of ascent for OE I and OE II were almost identical. The chief physiological value of hyperventilation is that it helps to maintain alveolar P0 2 at a viable level in the face of a falling inspired P0 2 as the climber ascends. This is shown in Fig. 2. Here, the alveolar gas values from AMREE are plotted on an oxygencarbon dioxide diagram, with sea level at top right and the summit of Mt Everest at bottom left [3]. It can be seen that as acclimatized subjects go to higher altitudes, both P0 2 and the PC0 2 fall. The P0 2 falls because of decreasing barometric pressure; the PC0 2 falls because of increasing hyperventilation. The interesting feature is that when the climber gets to an altitude of about 7000 m, there is no further fall in alveolar P0 2 and it settles out at a value of about 35 mmhg. In other words, the successful climber is defending the aveolar P0 2 at this value, and he can only do this by extreme hyperventilation. The critical role of hyperventilation at extreme altitude explains why many people who tolerate well great altitudes have a high hypoxic ventilatory response. Figure 3 shows some members of AMREE ranked by hypoxic ventilatory response as determined at sealevel [6,7]. This is simply a measure of the extent to which the subjects increased their breathing when they inhaled a low oxygen mixture. It turned out that the climber with the

20 West CK 0 CP 0 DJ PH D 0 i i FS SB Fig. 3. Members of AMREE, ranked by their hypoxic ventilatory response measured at sea level. The three subjects with the highest responses reached the summit, while those with the lowest responses generally did not tolerate extreme altitude well. From [7]; data from [6]. highest hypoxic ventilatory response reached the summit first, the one with the next highest response reached the summit second, and the one with the third highest response reached the summit third! This must be partly by chance, but nevertheless strongly suggests that a high hypoxic ventilatory response helps at extreme altitude. This point was made even clearer by considering the climbers who were unfortunate enough to be born with a low hypoxic ventilatory response. In general, they tolerated extreme altitudes poorly. It is not suggested that a high hypoxic ventilatory response is a necessary prerequisite for climbing to extreme altitudes. Indeed, there have been a number of studies pointing out that some very successful climbers have an average ventilatory response to hypoxia. Nevertheless, it is probably true that someone who has a low hypoxic ventilatory response is not likely to tolerate well extreme altitudes. This test might therefore be useful in assessing potential members of an expedition to extreme altitude, although the best predictor is probably whether a climber has tolerated well extreme altitude on a previous expedition. Figure 2 shows that the alveolar POz of Pizzo was near 35 mmhg on the summit of Mt Everest. However, there is evidence that the arterial POz was considerably lower because of diffusion limitation of oxygen transfer in the lung under these extraordinary conditions, especially during exercise. It turns out that the rate of rise of POz along the pulmonary capillary is extremely slow because man is working so low on the oxygen dissociation curve, and its slope is so steep. Calculations indicate that Pizzo's arterial POz was about 28 mmhg; direct measurements of arterial blood for the same inspired POz during the simulated ascent of OE II gave similar results (Table 1). Of course, an arterial POz of less than 30 indicates an extremely severe degree of arterial hypoxemia. There is also evidence that the arterial ph was extremely high. Base excess values measured on the venous blood of climbers Pizzo and Hackett during the morning after their successful summit climbs gave a mean value of about -7 mm. When combined with

Acclimatization and tolerance to extreme altitude 21 the measured alveolar PC0 2 of 7.5 mmhg on the summit (and assuming that the arterial PC0 2 was the same), the calculated arterial ph was between 7.7 and 7.8, an extraordinary degree of respiratory alkalosis. The degree of respiratory alkalosis found in the simulated ascent, OE II, was not quite as great. This can be explained by the fact that the PC0 2 values were not quite so low as indicated above. In fact, the base excess values for the 'summiters' in the field expedition and simulated ascent were similar. The fact that the base excess values were not lower (which would reduce the degree of respiratory alkalosis) suggests that the kidneys were excreting bicarbonate very slowly at extreme altitude, possibly because of associated volume depletion. Consequences ofsevere hypoxia The severe hypoxia of great altitudes has a multitude of physiological effects on the body. It is not easy to think of a system that is not impaired by the hypoxia, although interestingly, the normal myocardium appears to be remarkably tolerant to extreme hypoxemia with preservation of left ventricular function up to extreme altitudes [8]. Nevertheless, abnormal cardiac rhythms are sometimes seen at high altitude, especially premature ventricular contractions, though even these are uncommon [9]. One of the most interesting effects of severe hypoxia is on the central nervous system (CNS). It is not surprising that CNS function is impaired at great altitudes because of its exquisite vulnerability to oxygen lack. What was surprising in AMREE, however, was evidence for residual CNS impairment when the expedition returned to sea level. The most significant abnormality was a reduction in finger tapping speed, a test of manipulative skill. Of 16 expedition members, 15 showed impairment after the expedition had returned to sea level and the abnormality was still present in 13 subjects one year later [10]. The impairment remained in some expedition members for several years. Motor impairment after prolonged periods at high altitude has been found by 60 ~ 50 E "- l 40 ;;;! ::;: 20 x<[ ::;: 10 Basal 0, Uptake 50 100 INSPIRED Po, (torr) 150 Fig. 4. Relationship between VOz max and inspired paz at extreme altitude. Note that that the paz corresponding to the Everest summit, VOz max was just over 11 min-i for a 70 kg man. The lowest two points were obtained by giving well-acclimatized subjects low oxygen concentrations to breathe at 6300 m altitude. From [13].

22 West others [11] and was also clearly demonstrated in the OE II study [12]. An interesting sidelight was that those individuals with the highest hypoxic ventilatory response, who generally tolerated high altitude best (Fig. 3), tended to have the most severe CNS abnormalities. This paradoxical result can presumably be explained by cerebral vasoconstriction caused by reduced arterial PCO z as a result of hyperventilation (Fig. 1). This vasoconstriction could explain the more severe cerebral hypoxia, in spite of a higher arterial PO z in these subjects. One of the most obvious consequences of severe hypoxia is the reduction in work capacity. Figure 4 shows the relationship between maximal oxygen consumption (VO z max) and inspired PO z as the altitude was increased in AMREE [13]. The point marked 'summit' was obtained by administering a low oxygen concentration to well-acclimatized subjects at the main laboratory, altitude 6300 m. At an inspired PO z equivalent to the summit, VO z max was reduced to about 1.11 min- t which is between 20-25% of its sea level value. An essentially identical figure was found at the 'summit' in the low pressure simulation, OE II. This is an extremely low maximal oxygen consumption, equivalent to that of a person walking slowly on the level, but is apparently sufficient to account for the ability of climbers to just reach the summit of Mt Everest without supplementary oxygen. Acclimatization to high altitude The term 'acclimatization' should probably not be used for extreme altitudes above 8000 m, because all the evidence suggests that the body steadily deteriorates at such heights. Presumably, severe hypoxia is the most important factor in this deterioration. As indicated above, in spite of extreme hyperventilation near the summit of Mt Everest, alveolar PO z cannot be maintained above about 35 mmhg. A possible factor in acclimatization to more moderate altitudes is gradual renal compensation for respiratory alkalosis. By excreting bicarbonate, the kidneys tend to restore arterial ph to near its normal value of 7.4. As pointed out earlier, renal excretion of bicarbonate is apparently extremely slow at extreme altitudes. It is interesting to note that the respiratory alkalosis of extreme altitude apparently has some advantages for the climber, because it increases the affinity of hemoglobin for oxygen, enhancing uptake of oxygen by the pulmonary capillaries. At the same time, this alkalosis interferes with oxygen unloading by peripheral capillaries in exercising muscle. However, modeling studies show that the advantage of increasing loading in the lungs more than makes up for the disadvantage of reducing unloading in the peripheral muscles [14]. It is noteworthy that many animals placed in a hypoxic environment develop increased oxygen infinity of their hemoglobin by various mechanisms. The one used by the high altitude climber is extreme hyperventilation. When a climber remains at high altitude, gradual renal compensation for respiratory alkalosis reduces the oxygen infinity of hemoglobin and therefore makes it more difficult for pulmonary capillaries to take up oxygen. This suggests that a climber who plans to ascend an 8000 m peak without supplementary oxygen should acclimatize as well as possible to an altitude of 5500 to 6500 m, and then move as rapidly as possible to the summit and back [15]. Of course it is assumed that the climber will take advantage of the natural processes of acclimatization up to the base camp altitude of 5500 m, giving the.body plenty of time to respond to the increasing altitude.

A cclimatization and tolerance to extreme altitude 23 260,...--,-----,---.,.-----,----,----,-----, 250 t ~ MT EVEREST E -s J5 ~ 240 ~ «'"~... «'" => '" '" ~ u ~ 230 t ~ 220 ;;\ I LATITUDE OF MT McKINLEY 210 0 10 20 30 40 50 60 10 EQ LATITUDE Inonh. degrees) Fig. 5. Barometric pressure at an altitude of 8848 m (summit of Mt Everest) plotted against latitude. It is extremely unlikely that Everest could be climbed without supplementary oxygen if it were at the latitude of Mt McKinley [15]. Tolerance to extreme altitude It has been recently appreciated that tolerance to extreme altitude is critically dependent on barometric pressure. It has been known since the time of Paul Bert [16] that low inspired POz as a result of reduced barometric pressure is responsible for the deleterious effects of high altitude. However, it has only recently been appreciated how critical is the relationship between maximal oxygen uptake and barometric pressure at altitudes near those of the Everest summit. This can be seen in Fig. 4, where the steep slope of the line relating VOz max to barometric pressure is obvious. A remarkable feature of the world's 8000 m peaks is that they are relatively near the equator. This is fortunate, because it can be shown that barometric pressure for a given altitude is higher at equatorial latitudes than near the poles [17]. Thus, a climber on the summit of Mt Everest (altitude 8848 m) would be at considerable disadvantage if the mountain were at the latitude of Mt McKinley (63 0 N) rather than at its near equatorial latitude of 28 0 N (Fig. 5). Another important factor is the change of barometric pressure with season. As Fig. 6 shows, the barometric pressure on the summit of Mt Everest varies by 12 mmhg from mid-winter to mid-summer [18]. This change is certainly sufficient to alter VOz max and means that a climber is at a considerable disadvantage in the winter for an ascent without supplemental oxygen, quite apart from other factors such as cold and wind. It was therefore a remarkable achievement when Sherpa Ang Rita made the first winter ascent without supplementary oxygen on 22 December 1987. lowe him a bottle of champagne [15].

24 West Table 2. Barometric pressures at altitude 8848 m for some key ascents of Mt Everest Date Event Altitude (m) atpressure Pressure at 8848m 500mbar 300mbar 200mbar (mmhg) 8 May 1978 First ascent without 5850 9640 12390 251 supplementary 02 (by Messner and Habeler) 20 August 1980 First 'solo' ascent without 5830 9800 12630 256 supplementary O 2 (by Messner) 24 October 1981 First direct measurement of 5860 9650 12380 252 pressure on summit (by Pizzo) (Value was 253 mmhg) 22 December 1987 First winter ascent without 5810 9500 12170 247 supplementary 02 (by Sherpa Ang Rita) Data are from weather balloons released at 1200 hours UTe (5.40 P.M. Nepal time) from New Delhi or Lhasa. Altitude readings shown are heights of balloon for the given pressure (500 mbar, etc.). Pressures at 8848 m altitude are by interpolation (see Appendix and [18). Information about the barometric 'pressure at the altitude of the summit of Mt Everest can be obtained from charts prepared by the meteorological services. Table 2 lists some of the key ascents of Mt Everest without supplemental oxygen along with the summit pressures calculated in this way (see Appendix). Pizzo's first direct measurement of barometric pressure on the summit is also shown. Note the substantial difference in barometric pressure (9 mmhg) between the summer ascent of August 1980 and the winter ascent of December 1987. Using the relationship between VOz max and barometric pressure (Fig. 4), this 9 mmhg fall in barometric pressure during winter would reduce VOz max by 118 ml min-lor about 11%. This would presumably significantly increase the difficulties of reaching the summit. Note also that mid-winter barometric pressure can be as low as a mean of 243 mmhg (Fig. 6), which is a further fall of 4 mmhg below the value of 247 on 22 December 1987. An ascent without oxygen at this time would be even more demanding. Use of oxygen at extreme altitude The very low levels of maximal oxygen consumptionatgl'eat altitudes are directly related to the low inspired POz' Therefore, the use of supplemental oxygen will always improve physical and mental performance. The great gain in VOz max conferred by increasing inspired POz is shown in Fig. 4. Occasionally, climbers have stated that using oxygen does not seem to benefit them, but there is no physiological basis for this assertion at extreme altitude. It could be argued that rescue oxygen should always be carried by expeditions to extreme altitudes, because of the very substantial risks posed by bad weather and accidents. An expedition that attempts Mt Everest without rescue oxygen could reasonably be labeled as irresponsible.

Acclimatization and tolerance to extreme altitude 25 Ci ::c E5- ~ cr: 250 :::> (/) ~ cr: <>. u oc >- ~ 0 245 cr: «co 240 L.----'J~F -:':-M-A-'------:M':----'J---'J----7 A ---:!;-S~O~N~D:---7J---' Fig. 6. Variation of barometric pressure on the summit of Mt Everest with month of the year. A climber in the winter would be at a severe disadvantage because of the low pressure. Data from weather balloons. From [18]. Appendix: calculation of barometric pressure at the altitude of the Everest Summit using meteorological charts Northern Hemisphere constant pressure charts showing the altitudes of the surfaces for 200, 300 and 500 mbars pressure can be obtained from the US National Climatic Service, Environmental Data and Information Service, National Climatic Center, Federal Building, Asheville, NC 28801-2696; telephone: (704) 259-0682. The readings corresponding to the stations for New Delhi (28 0 35' N, 77" 12' E) and Lhasa (29 0 41' N, 90 0 10' E) were identified using the latitude and longitude markings on the charts. The data from the New Delhi station were used in the earlier analyses [18], but Lhasa is closer to Mt Everest (27 0 59' N, 86 0 56' E). The data from the two stations usually agree closely. The pressure at 8848 m altitude can then be found using the following expression: (y-8848)] antilog [ log 300 + (log 2.5) (z _ x) where x, y and z are the 500, 300 and 200 mbar altitudes, respectively. This is an interpolation procedure which calculates the pressure difference between the 8848 m altitude and the 300 mbar altitude according to the slope of the line joining the x and z points, using the fact that the relationship between log pressure and altitude is almost linear. The pressure in mmhg is found by multiplying the pressure in millibars by 0.7500. References 1. West, J.B. Human physiology at extreme altitudes on Mount Everest. Science 1984; 223, 784 8.

26 West 2. Houston, C.S., Sutton, J.R, Cymerman, A. and Reeves, J.T. Operation Everest II: man at extreme altitude. J Appl Physiol1987; 63, 877-82. 3. West, J.B., Hackett, P.H., Maret, K.H., Milledge, J.S., Peters Jr., RM., Pizzo, c.j. and Winslow, RM. Pulmonary gas exchange on the summit of Mt Everest. J ApplPhysiol: Respirat Environ Exercise Physiol1983; 55, 678-87. 4. Riley, R.L. and Houston, C.S. Composition of alveolar air and volume of pulmonary ventilation during long exposure to high altitude. J Appl Physiol1951; 3, 526-34. 5. Rahn, H. and Otis, AB. Man's respiratory response during and after acclimatization to high altitude. Am J Physiol1949; 157,445-9. 6. Schoene, RB., Lahiri, S., Hackett, P.H., Peters, Jr., R.M., Milledge, J.S., Pizzo, c.j., Sarnquist, F.H., Boyer, S.J., Graber, D.J., Maret, K.H. and West, J.B. Relationship of hypoxic ventilatory response to exercise performance on Mt Everest. J Appl Physiol1984; 56, 1478-83. 7. West, J.B. High living: lessons from extreme altitude. Am Rev Respir Dis 1984; 130, 917-23. 8. Reeves, J.T., Groves, B.M., Sutton, J.R., Wagner, P.D., Cymerman, A., Malconian, M.K., Rock, P.B., Young, P.M. and Houston, C.S. Operation Everest II: preservation of cardiac function at extreme altitude. J Appl Physiol1987; 63, 531-9. 9. Karliner, J.S., Sarnquist, F.H., Graber, D.H., Peters, RM. Jr, and West, J.B. The electrocardiogram at extreme altitude: experience on Mt Everest. Amer Heart J 1985; 109, 505-13. 10. Townes, B.D., Hornbein, T.F., Schoene, RB., Sarnquist, F.H. and Grant, I. Human cerebral function at extreme altitude. In: West, J.B. and Lahiri, S., eds. High Altitude and Man. Washington, DC: American Physiological Society, 1984: pp. 31-6. 11. Sharma, V.M., Malhotra, M.S. and Baskaran, AS. Variations in psychometer efficiency during prolonged stay at high altitude. Ergonomics 1975; 18, 511-16. 12. Hornbein, T.F., Townes, B.D., Schoene, RB., Sutton, J.R and Houston, C.S. The cost to the central nervous system of climbing to extremely high altitude. N Engl J Med 1989; 321, 1714 19. 13. West, J.B., Boyer, SJ., Graber, DJ., Hackett, P.H., Maret, K.H., Milledge, J.S., Peters Jr., RM., Pizzo, CJ., Samaja, M., Sarnquist, F.H., Schoene, R.B. and Winslow, RM. Maximal exercise at extreme altitudes on Mount Everest. J Appl Physiol: Respirat Environ Exercise Physiol1983; 55, 688-98. 14. Bencowitz, H.Z., Wagner, P.D. and West, J.B. Effect of change in P50 on exercise tolerance at high altitude: a theoretical study. J ApplPhysiol: Respirat Environ Exercise Physiol1982; 53, 1487-95. 15. West, J.B. 'Oxygenless' climbs and barometric pressure. Am Alpine J 1984; 58, 126-32. 16. Bert, P. La Pression Barometrique. Paris, Masson, 1878. English translation by Hitchcock, M.A and Hitchcock, F.A. Columbus, OH: College Book Co., 1943. 17. Brunt, D. Physical and Dynamical Meterology. 2nd ed. Cambridge: Cambridge University Press, 1952, p. 379 18. West, J.B., Lahiri, S., Maret, K.H., Peters Jr., RM. and Pizzo, CJ. Barometric pressures at extreme altitudes on Mt Everest: physiological significance. J Appl Physiol: Respirat Environ Exercise Physiol1983; 54, 1188-94. 19. Sutton, J.R, Reeves, J.T., Wagner, P.D., Groves, B.M., Cymerman, A, Malconian, M.K., Rock, P.B., Young, P.M., Walter, S.D. and Houston, C.S. Operation Everest II: oxygen transport during exercise at extreme simulated altitude. J Appl Physiol1988; 64, 1309-21.